From 76d08bb3f09054edc45326ce5c698a3f6c45f5d0 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Mon, 5 Jun 2006 13:54:14 -0700
Subject: [IA64] Add "model name" to /proc/cpuinfo

Linux ia64 port tried to decode the processor family number
to something human-readable, but Intel brandnames don't change
synchronously with updates to the family number.  Adopt a more
i386-like approach and just print the family number in decimal.
Add a new field "model name" that uses PAL_BRAND_INFO to find
the official name for the cpu, or on older systems, falls back
to using the well-known codenames (Merced, McKinley, Madison).

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/setup.c     | 41 +++++++++++++++++++++++++++++++----------
 include/asm-ia64/pal.h       | 10 ++++++++++
 include/asm-ia64/processor.h |  1 +
 3 files changed, 42 insertions(+), 10 deletions(-)

diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index e4dfda1eb7dd..3f7067c68b08 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -509,7 +509,7 @@ show_cpuinfo (struct seq_file *m, void *v)
 		{ 1UL << 1, "spontaneous deferral"},
 		{ 1UL << 2, "16-byte atomic ops" }
 	};
-	char family[32], features[128], *cp, sep;
+	char features[128], *cp, sep;
 	struct cpuinfo_ia64 *c = v;
 	unsigned long mask;
 	unsigned long proc_freq;
@@ -517,12 +517,6 @@ show_cpuinfo (struct seq_file *m, void *v)
 
 	mask = c->features;
 
-	switch (c->family) {
-	      case 0x07:	memcpy(family, "Itanium", 8); break;
-	      case 0x1f:	memcpy(family, "Itanium 2", 10); break;
-	      default:		sprintf(family, "%u", c->family); break;
-	}
-
 	/* build the feature string: */
 	memcpy(features, " standard", 10);
 	cp = features;
@@ -553,8 +547,9 @@ show_cpuinfo (struct seq_file *m, void *v)
 		   "processor  : %d\n"
 		   "vendor     : %s\n"
 		   "arch       : IA-64\n"
-		   "family     : %s\n"
+		   "family     : %u\n"
 		   "model      : %u\n"
+		   "model name : %s\n"
 		   "revision   : %u\n"
 		   "archrev    : %u\n"
 		   "features   :%s\n"	/* don't change this---it _is_ right! */
@@ -563,7 +558,8 @@ show_cpuinfo (struct seq_file *m, void *v)
 		   "cpu MHz    : %lu.%06lu\n"
 		   "itc MHz    : %lu.%06lu\n"
 		   "BogoMIPS   : %lu.%02lu\n",
-		   cpunum, c->vendor, family, c->model, c->revision, c->archrev,
+		   cpunum, c->vendor, c->family, c->model,
+		   c->model_name, c->revision, c->archrev,
 		   features, c->ppn, c->number,
 		   proc_freq / 1000, proc_freq % 1000,
 		   c->itc_freq / 1000000, c->itc_freq % 1000000,
@@ -611,6 +607,31 @@ struct seq_operations cpuinfo_op = {
 	.show =		show_cpuinfo
 };
 
+static char brandname[128];
+
+static char * __cpuinit
+get_model_name(__u8 family, __u8 model)
+{
+	char brand[128];
+
+	if (ia64_pal_get_brand_info(brand)) {
+		if (family == 0x7)
+			memcpy(brand, "Merced", 7);
+		else if (family == 0x1f) switch (model) {
+			case 0: memcpy(brand, "McKinley", 9); break;
+			case 1: memcpy(brand, "Madison", 8); break;
+			case 2: memcpy(brand, "Madison up to 9M cache", 23); break;
+		} else
+			memcpy(brand, "Unknown", 8);
+	}
+	if (brandname[0] == '\0')
+		return strcpy(brandname, brand);
+	else if (strcmp(brandname, brand) == 0)
+		return brandname;
+	else
+		return kstrdup(brand, GFP_KERNEL);
+}
+
 static void __cpuinit
 identify_cpu (struct cpuinfo_ia64 *c)
 {
@@ -640,7 +661,6 @@ identify_cpu (struct cpuinfo_ia64 *c)
 	pal_status_t status;
 	unsigned long impl_va_msb = 50, phys_addr_size = 44;	/* Itanium defaults */
 	int i;
-
 	for (i = 0; i < 5; ++i)
 		cpuid.bits[i] = ia64_get_cpuid(i);
 
@@ -663,6 +683,7 @@ identify_cpu (struct cpuinfo_ia64 *c)
 	c->family = cpuid.field.family;
 	c->archrev = cpuid.field.archrev;
 	c->features = cpuid.field.features;
+	c->model_name = get_model_name(c->family, c->model);
 
 	status = ia64_pal_vm_summary(&vm1, &vm2);
 	if (status == PAL_STATUS_SUCCESS) {
diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index 37e52a2836b0..312109d7be97 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -78,6 +78,7 @@
 #define PAL_VM_TR_READ		261	/* read contents of translation register */
 #define PAL_GET_PSTATE		262	/* get the current P-state */
 #define PAL_SET_PSTATE		263	/* set the P-state */
+#define PAL_BRAND_INFO		274	/* Processor branding information */
 
 #ifndef __ASSEMBLY__
 
@@ -1133,6 +1134,15 @@ ia64_pal_set_pstate (u64 pstate_index)
 	return iprv.status;
 }
 
+/* Processor branding information*/
+static inline s64
+ia64_pal_get_brand_info (char *brand_info)
+{
+	struct ia64_pal_retval iprv;
+	PAL_CALL_STK(iprv, PAL_BRAND_INFO, 0, (u64)brand_info, 0);
+	return iprv.status;
+}
+
 /* Cause the processor to enter LIGHT HALT state, where prefetching and execution are
  * suspended, but cache and TLB coherency is maintained.
  */
diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h
index b3bd58e80690..03100c4272c6 100644
--- a/include/asm-ia64/processor.h
+++ b/include/asm-ia64/processor.h
@@ -164,6 +164,7 @@ struct cpuinfo_ia64 {
 	__u8 family;
 	__u8 archrev;
 	char vendor[16];
+	char *model_name;
 
 #ifdef CONFIG_NUMA
 	struct ia64_node_data *node_data;
-- 
cgit v1.2.1


From 2ab561a116e16cdee3ae0e13d51910634c15aee9 Mon Sep 17 00:00:00 2001
From: David Mosberger-Tang <David.Mosberger@acm.org>
Date: Wed, 21 Jun 2006 11:19:22 -0700
Subject: [IA64] esi-support

Add support for making ESI calls [1].  ESI stands for "Extensible SAL
specification" and is basically a way for invoking firmware
subroutines which are identified by a GUID.  I don't know whether ESI
is used by vendors other than HP (if you do, please let me know) but
as firmware "backdoors" go, this seems one of the cleaner methods, so
it seems reasonable to support it, even though I'm not aware of any
publicly documented ESI calls.  I'd have liked to make the ESI module
completely stand-alone, but unfortunately that is not easily (or not
at all) possible because in order to make ESI calls in physical mode,
a small stub similar to the EFI stub is needed in the kernel proper.
I did try to create a stub that would work in user-level, but it
quickly got ugly beyond recognition (e.g., the stub had to make
assumptions about how the module-loader generated call-stubs work) and
I didn't even get it to work (that's probably fixable, but I didn't
bother because I concluded it was too ugly anyhow).  While it's not
terribly elegant to have kernel code which isn't actively used in the
kernel proper, I think it might be worth making an exception here for
two reasons: the code is trivially small (all that's really needed is
esi_stub.S) and by including it in the normal kernel distro, it might
encourage other OEMs to also use ESI, which I think would be far
better than each inventing their own firmware "backdoor".

The code was originally written by Alex.  I just massaged and packaged
it a bit (and perhaps messed up some things along the way...).

Changes since first version of patch that was posted to mailing list:
* Export ia64_esi_call and ia64_esi_call_phys() as GPL symbols.
* Disallow building esi.c as a module for now.  Building as a module
  would currently lead to an unresolved reference to "sal_lock" on SMP kernels
  because that symbol doesn't get exported.
* Export esi_call_phys() only if ESI is enabled.
* Remove internal stuff from esi.h and add a "proc_type" argument to
  ia64_esi_call() such that serialization-requirements can be expressed (ESI
  follows SAL here, where procedure calls may have to be serialized, are
  MP-safe, or MP-safe andr reentrant).

[1] h21007.www2.hp.com/dspp/tech/tech_TechDocumentDetailPage_IDX/1,1701,919,00.html

Signed-off-by: David Mosberger <David.Mosberger@acm.org>
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/Kconfig             |   8 ++
 arch/ia64/kernel/Makefile     |   5 ++
 arch/ia64/kernel/esi.c        | 205 ++++++++++++++++++++++++++++++++++++++++++
 arch/ia64/kernel/esi_stub.S   |  96 ++++++++++++++++++++
 arch/ia64/kernel/ia64_ksyms.c |   4 +
 include/asm-ia64/esi.h        |  30 +++++++
 6 files changed, 348 insertions(+)
 create mode 100644 arch/ia64/kernel/esi.c
 create mode 100644 arch/ia64/kernel/esi_stub.S
 create mode 100644 include/asm-ia64/esi.h

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 0f3076a820c3..bc106519482c 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -413,6 +413,14 @@ config IA64_PALINFO
 config SGI_SN
 	def_bool y if (IA64_SGI_SN2 || IA64_GENERIC)
 
+config IA64_ESI
+	bool "ESI (Extensible SAL Interface) support"
+	help
+	  If you say Y here, support is built into the kernel to
+	  make ESI calls.  ESI calls are used to support vendor-specific
+	  firmware extensions, such as the ability to inject memory-errors
+	  for test-purposes.  If you're unsure, say N.
+
 source "drivers/sn/Kconfig"
 
 source "drivers/firmware/Kconfig"
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 09a0dbc17fb6..66ac8e3dca9c 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -31,6 +31,11 @@ obj-$(CONFIG_KPROBES)		+= kprobes.o jprobes.o
 obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR)	+= uncached.o
 mca_recovery-y			+= mca_drv.o mca_drv_asm.o
 
+obj-$(CONFIG_IA64_ESI)		+= esi.o
+ifneq ($(CONFIG_IA64_ESI),)
+obj-y				+= esi_stub.o	# must be in kernel proper
+endif
+
 # The gate DSO image is built using a special linker script.
 targets += gate.so gate-syms.o
 
diff --git a/arch/ia64/kernel/esi.c b/arch/ia64/kernel/esi.c
new file mode 100644
index 000000000000..ebf4e988e78c
--- /dev/null
+++ b/arch/ia64/kernel/esi.c
@@ -0,0 +1,205 @@
+/*
+ * Extensible SAL Interface (ESI) support routines.
+ *
+ * Copyright (C) 2006 Hewlett-Packard Co
+ * 	Alex Williamson <alex.williamson@hp.com>
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/string.h>
+
+#include <asm/esi.h>
+#include <asm/sal.h>
+
+MODULE_AUTHOR("Alex Williamson <alex.williamson@hp.com>");
+MODULE_DESCRIPTION("Extensible SAL Interface (ESI) support");
+MODULE_LICENSE("GPL");
+
+#define MODULE_NAME	"esi"
+
+#define ESI_TABLE_GUID					\
+    EFI_GUID(0x43EA58DC, 0xCF28, 0x4b06, 0xB3,		\
+	     0x91, 0xB7, 0x50, 0x59, 0x34, 0x2B, 0xD4)
+
+enum esi_systab_entry_type {
+	ESI_DESC_ENTRY_POINT = 0
+};
+
+/*
+ * Entry type:	Size:
+ *	0	48
+ */
+#define ESI_DESC_SIZE(type)	"\060"[(unsigned) (type)]
+
+typedef struct ia64_esi_desc_entry_point {
+	u8 type;
+	u8 reserved1[15];
+	u64 esi_proc;
+	u64 gp;
+	efi_guid_t guid;
+} ia64_esi_desc_entry_point_t;
+
+struct pdesc {
+	void *addr;
+	void *gp;
+};
+
+static struct ia64_sal_systab *esi_systab;
+
+static int __init esi_init (void)
+{
+	efi_config_table_t *config_tables;
+	struct ia64_sal_systab *systab;
+	unsigned long esi = 0;
+	char *p;
+	int i;
+
+	config_tables = __va(efi.systab->tables);
+
+	for (i = 0; i < (int) efi.systab->nr_tables; ++i) {
+		if (efi_guidcmp(config_tables[i].guid, ESI_TABLE_GUID) == 0) {
+			esi = config_tables[i].table;
+			break;
+		}
+	}
+
+	if (!esi)
+		return -ENODEV;;
+
+	systab = __va(esi);
+
+	if (strncmp(systab->signature, "ESIT", 4) != 0) {
+		printk(KERN_ERR "bad signature in ESI system table!");
+		return -ENODEV;
+	}
+
+	p = (char *) (systab + 1);
+	for (i = 0; i < systab->entry_count; i++) {
+		/*
+		 * The first byte of each entry type contains the type
+		 * descriptor.
+		 */
+		switch (*p) {
+		      case ESI_DESC_ENTRY_POINT:
+			break;
+		      default:
+			printk(KERN_WARNING "Unkown table type %d found in "
+			       "ESI table, ignoring rest of table\n", *p);
+			return -ENODEV;
+		}
+
+		p += ESI_DESC_SIZE(*p);
+	}
+
+	esi_systab = systab;
+	return 0;
+}
+
+
+int ia64_esi_call (efi_guid_t guid, struct ia64_sal_retval *isrvp,
+		   enum esi_proc_type proc_type, u64 func,
+		   u64 arg1, u64 arg2, u64 arg3, u64 arg4, u64 arg5, u64 arg6,
+		   u64 arg7)
+{
+	struct ia64_fpreg fr[6];
+	unsigned long flags = 0;
+	int i;
+	char *p;
+
+	if (!esi_systab)
+		return -1;
+
+	p = (char *) (esi_systab + 1);
+	for (i = 0; i < esi_systab->entry_count; i++) {
+		if (*p == ESI_DESC_ENTRY_POINT) {
+			ia64_esi_desc_entry_point_t *esi = (void *)p;
+			if (!efi_guidcmp(guid, esi->guid)) {
+				ia64_sal_handler esi_proc;
+				struct pdesc pdesc;
+
+				pdesc.addr = __va(esi->esi_proc);
+				pdesc.gp = __va(esi->gp);
+
+				esi_proc = (ia64_sal_handler) &pdesc;
+
+				ia64_save_scratch_fpregs(fr);
+				if (proc_type == ESI_PROC_SERIALIZED)
+					spin_lock_irqsave(&sal_lock, flags);
+				else if (proc_type == ESI_PROC_MP_SAFE)
+					local_irq_save(flags);
+				else
+					preempt_disable();
+				*isrvp = (*esi_proc)(func, arg1, arg2, arg3,
+						     arg4, arg5, arg6, arg7);
+				if (proc_type == ESI_PROC_SERIALIZED)
+					spin_unlock_irqrestore(&sal_lock,
+							       flags);
+				else if (proc_type == ESI_PROC_MP_SAFE)
+					local_irq_restore(flags);
+				else
+					preempt_enable();
+				ia64_load_scratch_fpregs(fr);
+				return 0;
+			}
+		}
+		p += ESI_DESC_SIZE(*p);
+	}
+	return -1;
+}
+EXPORT_SYMBOL_GPL(ia64_esi_call);
+
+int ia64_esi_call_phys (efi_guid_t guid, struct ia64_sal_retval *isrvp,
+			u64 func, u64 arg1, u64 arg2, u64 arg3, u64 arg4,
+			u64 arg5, u64 arg6, u64 arg7)
+{
+	struct ia64_fpreg fr[6];
+	unsigned long flags;
+	u64 esi_params[8];
+	char *p;
+	int i;
+
+	if (!esi_systab)
+		return -1;
+
+	p = (char *) (esi_systab + 1);
+	for (i = 0; i < esi_systab->entry_count; i++) {
+		if (*p == ESI_DESC_ENTRY_POINT) {
+			ia64_esi_desc_entry_point_t *esi = (void *)p;
+			if (!efi_guidcmp(guid, esi->guid)) {
+				ia64_sal_handler esi_proc;
+				struct pdesc pdesc;
+
+				pdesc.addr = (void *)esi->esi_proc;
+				pdesc.gp = (void *)esi->gp;
+
+				esi_proc = (ia64_sal_handler) &pdesc;
+
+				esi_params[0] = func;
+				esi_params[1] = arg1;
+				esi_params[2] = arg2;
+				esi_params[3] = arg3;
+				esi_params[4] = arg4;
+				esi_params[5] = arg5;
+				esi_params[6] = arg6;
+				esi_params[7] = arg7;
+				ia64_save_scratch_fpregs(fr);
+				spin_lock_irqsave(&sal_lock, flags);
+				*isrvp = esi_call_phys(esi_proc, esi_params);
+				spin_unlock_irqrestore(&sal_lock, flags);
+				ia64_load_scratch_fpregs(fr);
+				return 0;
+			}
+		}
+		p += ESI_DESC_SIZE(*p);
+	}
+	return -1;
+}
+EXPORT_SYMBOL_GPL(ia64_esi_call_phys);
+
+static void __exit esi_exit (void)
+{
+}
+
+module_init(esi_init);
+module_exit(esi_exit);	/* makes module removable... */
diff --git a/arch/ia64/kernel/esi_stub.S b/arch/ia64/kernel/esi_stub.S
new file mode 100644
index 000000000000..6b3d6c1f99b6
--- /dev/null
+++ b/arch/ia64/kernel/esi_stub.S
@@ -0,0 +1,96 @@
+/*
+ * ESI call stub.
+ *
+ * Copyright (C) 2005 Hewlett-Packard Co
+ *	Alex Williamson <alex.williamson@hp.com>
+ *
+ * Based on EFI call stub by David Mosberger.  The stub is virtually
+ * identical to the one for EFI phys-mode calls, except that ESI
+ * calls may have up to 8 arguments, so they get passed to this routine
+ * through memory.
+ *
+ * This stub allows us to make ESI calls in physical mode with interrupts
+ * turned off.  ESI calls may not support calling from virtual mode.
+ *
+ * Google for "Extensible SAL specification" for a document describing the
+ * ESI standard.
+ */
+
+/*
+ * PSR settings as per SAL spec (Chapter 8 in the "IA-64 System
+ * Abstraction Layer Specification", revision 2.6e).  Note that
+ * psr.dfl and psr.dfh MUST be cleared, despite what this manual says.
+ * Otherwise, SAL dies whenever it's trying to do an IA-32 BIOS call
+ * (the br.ia instruction fails unless psr.dfl and psr.dfh are
+ * cleared).  Fortunately, SAL promises not to touch the floating
+ * point regs, so at least we don't have to save f2-f127.
+ */
+#define PSR_BITS_TO_CLEAR						\
+	(IA64_PSR_I | IA64_PSR_IT | IA64_PSR_DT | IA64_PSR_RT |		\
+	 IA64_PSR_DD | IA64_PSR_SS | IA64_PSR_RI | IA64_PSR_ED |	\
+	 IA64_PSR_DFL | IA64_PSR_DFH)
+
+#define PSR_BITS_TO_SET							\
+	(IA64_PSR_BN)
+
+#include <asm/processor.h>
+#include <asm/asmmacro.h>
+
+/*
+ * Inputs:
+ *	in0 = address of function descriptor of ESI routine to call
+ *	in1 = address of array of ESI parameters
+ *
+ * Outputs:
+ *	r8 = result returned by called function
+ */
+GLOBAL_ENTRY(esi_call_phys)
+	.prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(2)
+	alloc loc1=ar.pfs,2,7,8,0
+	ld8 r2=[in0],8			// load ESI function's entry point
+	mov loc0=rp
+	.body
+	;;
+	ld8 out0=[in1],8		// ESI params loaded from array
+	;;				// passing all as inputs doesn't work
+	ld8 out1=[in1],8
+	;;
+	ld8 out2=[in1],8
+	;;
+	ld8 out3=[in1],8
+	;;
+	ld8 out4=[in1],8
+	;;
+	ld8 out5=[in1],8
+	;;
+	ld8 out6=[in1],8
+	;;
+	ld8 out7=[in1]
+	mov loc2=gp			// save global pointer
+	mov loc4=ar.rsc			// save RSE configuration
+	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
+	;;
+	ld8 gp=[in0]			// load ESI function's global pointer
+	movl r16=PSR_BITS_TO_CLEAR
+	mov loc3=psr			// save processor status word
+	movl r17=PSR_BITS_TO_SET
+	;;
+	or loc3=loc3,r17
+	mov b6=r2
+	;;
+	andcm r16=loc3,r16	// get psr with IT, DT, and RT bits cleared
+	br.call.sptk.many rp=ia64_switch_mode_phys
+.ret0:	mov loc5=r19			// old ar.bsp
+	mov loc6=r20			// old sp
+	br.call.sptk.many rp=b6		// call the ESI function
+.ret1:	mov ar.rsc=0			// put RSE in enforced lazy, LE mode
+	mov r16=loc3			// save virtual mode psr
+	mov r19=loc5			// save virtual mode bspstore
+	mov r20=loc6			// save virtual mode sp
+	br.call.sptk.many rp=ia64_switch_mode_virt // return to virtual mode
+.ret2:	mov ar.rsc=loc4			// restore RSE configuration
+	mov ar.pfs=loc1
+	mov rp=loc0
+	mov gp=loc2
+	br.ret.sptk.many rp
+END(esi_call_phys)
diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index bbcfd08378a6..9583f310d486 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -106,5 +106,9 @@ EXPORT_SYMBOL(ia64_spinlock_contention);
 # endif
 #endif
 
+#if defined(CONFIG_IA64_ESI) || defined(CONFIG_IA64_ESI_MODULE)
+extern void esi_call_phys (void);
+EXPORT_SYMBOL_GPL(esi_call_phys);
+#endif
 extern char ia64_ivt[];
 EXPORT_SYMBOL(ia64_ivt);
diff --git a/include/asm-ia64/esi.h b/include/asm-ia64/esi.h
new file mode 100644
index 000000000000..84aac0e0b583
--- /dev/null
+++ b/include/asm-ia64/esi.h
@@ -0,0 +1,30 @@
+/*
+ * ESI service calls.
+ *
+ * Copyright (c) Copyright 2005-2006 Hewlett-Packard Development Company, L.P.
+ * 	Alex Williamson <alex.williamson@hp.com>
+ */
+#ifndef esi_h
+#define esi_h
+
+#include <linux/efi.h>
+
+#define ESI_QUERY			0x00000001
+#define ESI_OPEN_HANDLE			0x02000000
+#define ESI_CLOSE_HANDLE		0x02000001
+
+enum esi_proc_type {
+	ESI_PROC_SERIALIZED,	/* calls need to be serialized */
+	ESI_PROC_MP_SAFE,	/* MP-safe, but not reentrant */
+	ESI_PROC_REENTRANT	/* MP-safe and reentrant */
+};
+
+extern int ia64_esi_init (void);
+extern struct ia64_sal_retval esi_call_phys (void *, u64 *);
+extern int ia64_esi_call(efi_guid_t, struct ia64_sal_retval *,
+			 enum esi_proc_type,
+			 u64, u64, u64, u64, u64, u64, u64, u64);
+extern int ia64_esi_call_phys(efi_guid_t, struct ia64_sal_retval *, u64, u64,
+                              u64, u64, u64, u64, u64, u64);
+
+#endif /* esi_h */
-- 
cgit v1.2.1


From 00e4d116a7ef94eb910be037912b0b2fc09f608b Mon Sep 17 00:00:00 2001
From: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Date: Fri, 22 Sep 2006 09:33:58 +0100
Subject: [DCCP]: Allow default/fallback service code.

This has been discussed on dccp@vger and removes the necessity for applications
to supply service codes in each and every case.

If an application does not want to provide a service code, that's fine, it will
be given 0. Otherwise, service codes can be set via socket options as before.

This patch has been tested using various client/server configurations
(including listening on multiple service codes).

Signed-off-by: Gerrit Renker <gerrit@erg.abdn.ac.uk>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 Documentation/networking/dccp.txt |  8 +++++---
 include/linux/dccp.h              |  6 +-----
 net/dccp/ipv4.c                   |  3 ---
 net/dccp/proto.c                  | 11 +----------
 4 files changed, 7 insertions(+), 21 deletions(-)

diff --git a/Documentation/networking/dccp.txt b/Documentation/networking/dccp.txt
index c45daabd3bfe..74563b38ffd9 100644
--- a/Documentation/networking/dccp.txt
+++ b/Documentation/networking/dccp.txt
@@ -1,7 +1,6 @@
 DCCP protocol
 ============
 
-Last updated: 10 November 2005
 
 Contents
 ========
@@ -42,8 +41,11 @@ Socket options
 DCCP_SOCKOPT_PACKET_SIZE is used for CCID3 to set default packet size for
 calculations.
 
-DCCP_SOCKOPT_SERVICE sets the service. This is compulsory as per the
-specification. If you don't set it you will get EPROTO.
+DCCP_SOCKOPT_SERVICE sets the service. The specification mandates use of
+service codes (RFC 4340, sec. 8.1.2); if this socket option is not set,
+the socket will fall back to 0 (which means that no meaningful service code
+is present). Connecting sockets set at most one service option; for
+listening sockets, multiple service codes can be specified.
 
 Notes
 =====
diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 2d7671c92c0b..29f9291e45c0 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -404,6 +404,7 @@ struct dccp_service_list {
 };
 
 #define DCCP_SERVICE_INVALID_VALUE htonl((__u32)-1)
+#define DCCP_SERVICE_CODE_IS_ABSENT 		 0
 
 static inline int dccp_list_has_service(const struct dccp_service_list *sl,
 					const __be32 service)
@@ -484,11 +485,6 @@ static inline struct dccp_minisock *dccp_msk(const struct sock *sk)
 	return (struct dccp_minisock *)&dccp_sk(sk)->dccps_minisock;
 }
 
-static inline int dccp_service_not_initialized(const struct sock *sk)
-{
-	return dccp_sk(sk)->dccps_service == DCCP_SERVICE_INVALID_VALUE;
-}
-
 static inline const char *dccp_role(const struct sock *sk)
 {
 	switch (dccp_sk(sk)->dccps_role) {
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 9a1a76a7dc41..66be29b6f508 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -56,9 +56,6 @@ int dccp_v4_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
 
 	dp->dccps_role = DCCP_ROLE_CLIENT;
 
-	if (dccp_service_not_initialized(sk))
-		return -EPROTO;
-
 	if (addr_len < sizeof(struct sockaddr_in))
 		return -EINVAL;
 
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index 962df0ea31aa..72cbdcfc2c65 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -217,7 +217,7 @@ int dccp_init_sock(struct sock *sk, const __u8 ctl_sock_initialized)
 	icsk->icsk_sync_mss	= dccp_sync_mss;
 	dp->dccps_mss_cache	= 536;
 	dp->dccps_role		= DCCP_ROLE_UNDEFINED;
-	dp->dccps_service	= DCCP_SERVICE_INVALID_VALUE;
+	dp->dccps_service	= DCCP_SERVICE_CODE_IS_ABSENT;
 	dp->dccps_l_ack_ratio	= dp->dccps_r_ack_ratio = 1;
 
 	return 0;
@@ -267,12 +267,6 @@ static inline int dccp_listen_start(struct sock *sk)
 	struct dccp_sock *dp = dccp_sk(sk);
 
 	dp->dccps_role = DCCP_ROLE_LISTEN;
-	/*
-	 * Apps need to use setsockopt(DCCP_SOCKOPT_SERVICE)
-	 * before calling listen()
-	 */
-	if (dccp_service_not_initialized(sk))
-		return -EPROTO;
 	return inet_csk_listen_start(sk, TCP_SYNQ_HSIZE);
 }
 
@@ -540,9 +534,6 @@ static int dccp_getsockopt_service(struct sock *sk, int len,
 	int err = -ENOENT, slen = 0, total_len = sizeof(u32);
 
 	lock_sock(sk);
-	if (dccp_service_not_initialized(sk))
-		goto out;
-
 	if ((sl = dp->dccps_service_list) != NULL) {
 		slen = sl->dccpsl_nr * sizeof(u32);
 		total_len += slen;
-- 
cgit v1.2.1


From b83eff641ed39bd631535b9a8971e161b056f541 Mon Sep 17 00:00:00 2001
From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Fri, 22 Sep 2006 14:25:36 +1200
Subject: [DCCP]: Introduce constants for CCID numbers

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 include/linux/dccp.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/include/linux/dccp.h b/include/linux/dccp.h
index 29f9291e45c0..d6f4ec467a4b 100644
--- a/include/linux/dccp.h
+++ b/include/linux/dccp.h
@@ -169,6 +169,12 @@ enum {
 	DCCPO_MAX_CCID_SPECIFIC = 255,
 };
 
+/* DCCP CCIDS */
+enum {
+	DCCPC_CCID2 = 2,
+	DCCPC_CCID3 = 3,
+};
+
 /* DCCP features */
 enum {
 	DCCPF_RESERVED = 0,
@@ -320,7 +326,7 @@ static inline unsigned int dccp_hdr_len(const struct sk_buff *skb)
 /* initial values for each feature */
 #define DCCPF_INITIAL_SEQUENCE_WINDOW		100
 #define DCCPF_INITIAL_ACK_RATIO			2
-#define DCCPF_INITIAL_CCID			2
+#define DCCPF_INITIAL_CCID			DCCPC_CCID2
 #define DCCPF_INITIAL_SEND_ACK_VECTOR		1
 /* FIXME: for now we're default to 1 but it should really be 0 */
 #define DCCPF_INITIAL_SEND_NDP_COUNT		1
-- 
cgit v1.2.1


From 3dd9a7c3a155ee96160876cba92439fdc96d7e0b Mon Sep 17 00:00:00 2001
From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Fri, 22 Sep 2006 14:26:44 +1200
Subject: [DCCP]: Use constants for CCIDs

With constants for CCID numbers this now uses them in some places.

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 net/dccp/ccids/ccid2.c | 2 +-
 net/dccp/ccids/ccid3.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/dccp/ccids/ccid2.c b/net/dccp/ccids/ccid2.c
index 457dd3db7f41..2efb505aeb35 100644
--- a/net/dccp/ccids/ccid2.c
+++ b/net/dccp/ccids/ccid2.c
@@ -808,7 +808,7 @@ static void ccid2_hc_rx_packet_recv(struct sock *sk, struct sk_buff *skb)
 }
 
 static struct ccid_operations ccid2 = {
-	.ccid_id		= 2,
+	.ccid_id		= DCCPC_CCID2,
 	.ccid_name		= "ccid2",
 	.ccid_owner		= THIS_MODULE,
 	.ccid_hc_tx_obj_size	= sizeof(struct ccid2_hc_tx_sock),
diff --git a/net/dccp/ccids/ccid3.c b/net/dccp/ccids/ccid3.c
index 195aa9566228..67d2dc0e7c67 100644
--- a/net/dccp/ccids/ccid3.c
+++ b/net/dccp/ccids/ccid3.c
@@ -1240,7 +1240,7 @@ static int ccid3_hc_tx_getsockopt(struct sock *sk, const int optname, int len,
 }
 
 static struct ccid_operations ccid3 = {
-	.ccid_id		   = 3,
+	.ccid_id		   = DCCPC_CCID3,
 	.ccid_name		   = "ccid3",
 	.ccid_owner		   = THIS_MODULE,
 	.ccid_hc_tx_obj_size	   = sizeof(struct ccid3_hc_tx_sock),
-- 
cgit v1.2.1


From e41542f5167d6b506607f8dd111fa0a3e468ccb8 Mon Sep 17 00:00:00 2001
From: Ian McDonald <ian.mcdonald@jandi.co.nz>
Date: Fri, 22 Sep 2006 14:28:01 +1200
Subject: [DCCP]: Introduce dccp_probe

This adds DCCP probing shamelessly ripped off from TCP probes by Stephen
Hemminger.

I've put in here support for further CCID3 variables as well.
Andrea/Arnaldo might look to extend for CCID2.

Signed-off-by: Ian McDonald <ian.mcdonald@jandi.co.nz>
Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
---
 net/dccp/Kconfig  |  16 +++++
 net/dccp/Makefile |   2 +
 net/dccp/probe.c  | 198 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 216 insertions(+)
 create mode 100644 net/dccp/probe.c

diff --git a/net/dccp/Kconfig b/net/dccp/Kconfig
index 859e3359fcda..e2a095d0fd80 100644
--- a/net/dccp/Kconfig
+++ b/net/dccp/Kconfig
@@ -40,6 +40,22 @@ config IP_DCCP_DEBUG
 
 	  Just say N.
 
+config NET_DCCPPROBE
+	tristate "DCCP connection probing"
+	depends on PROC_FS && KPROBES
+	---help---
+	This module allows for capturing the changes to DCCP connection
+	state in response to incoming packets. It is used for debugging
+	DCCP congestion avoidance modules. If you don't understand
+	what was just said, you don't need it: say N.
+
+	Documentation on how to use the packet generator can be found
+	at http://linux-net.osdl.org/index.php/DccpProbe
+
+	To compile this code as a module, choose M here: the
+	module will be called dccp_probe.
+
+
 endmenu
 
 endmenu
diff --git a/net/dccp/Makefile b/net/dccp/Makefile
index 7696e219b05d..17ed99c46617 100644
--- a/net/dccp/Makefile
+++ b/net/dccp/Makefile
@@ -11,9 +11,11 @@ dccp_ipv4-y := ipv4.o
 dccp-$(CONFIG_IP_DCCP_ACKVEC) += ackvec.o
 
 obj-$(CONFIG_INET_DCCP_DIAG) += dccp_diag.o
+obj-$(CONFIG_NET_DCCPPROBE) += dccp_probe.o
 
 dccp-$(CONFIG_SYSCTL) += sysctl.o
 
 dccp_diag-y := diag.o
+dccp_probe-y := probe.o
 
 obj-y += ccids/
diff --git a/net/dccp/probe.c b/net/dccp/probe.c
new file mode 100644
index 000000000000..146496fce2e2
--- /dev/null
+++ b/net/dccp/probe.c
@@ -0,0 +1,198 @@
+/*
+ * dccp_probe - Observe the DCCP flow with kprobes.
+ *
+ * The idea for this came from Werner Almesberger's umlsim
+ * Copyright (C) 2004, Stephen Hemminger <shemminger@osdl.org>
+ *
+ * Modified for DCCP from Stephen Hemminger's code
+ * Copyright (C) 2006, Ian McDonald <ian.mcdonald@jandi.co.nz>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/socket.h>
+#include <linux/dccp.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
+#include <linux/kfifo.h>
+#include <linux/vmalloc.h>
+
+#include "dccp.h"
+#include "ccid.h"
+#include "ccids/ccid3.h"
+
+static int port;
+
+static int bufsize = 64 * 1024;
+
+static const char procname[] = "dccpprobe";
+
+struct {
+	struct kfifo	  *fifo;
+	spinlock_t	  lock;
+	wait_queue_head_t wait;
+	struct timeval	  tstart;
+} dccpw;
+
+static void printl(const char *fmt, ...)
+{
+	va_list args;
+	int len;
+	struct timeval now;
+	char tbuf[256];
+
+	va_start(args, fmt);
+	do_gettimeofday(&now);
+
+	now.tv_sec -= dccpw.tstart.tv_sec;
+	now.tv_usec -= dccpw.tstart.tv_usec;
+	if (now.tv_usec < 0) {
+		--now.tv_sec;
+		now.tv_usec += 1000000;
+	}
+
+	len = sprintf(tbuf, "%lu.%06lu ",
+		      (unsigned long) now.tv_sec,
+		      (unsigned long) now.tv_usec);
+	len += vscnprintf(tbuf+len, sizeof(tbuf)-len, fmt, args);
+	va_end(args);
+
+	kfifo_put(dccpw.fifo, tbuf, len);
+	wake_up(&dccpw.wait);
+}
+
+static int jdccp_sendmsg(struct kiocb *iocb, struct sock *sk,
+			 struct msghdr *msg, size_t size)
+{
+	const struct dccp_minisock *dmsk = dccp_msk(sk);
+	const struct inet_sock *inet = inet_sk(sk);
+	const struct ccid3_hc_tx_sock *hctx;
+
+	if (dmsk->dccpms_tx_ccid == DCCPC_CCID3)
+		hctx = ccid3_hc_tx_sk(sk);
+	else
+		hctx = NULL;
+
+	if (port == 0 || ntohs(inet->dport) == port ||
+	    ntohs(inet->sport) == port) {
+		if (hctx)
+			printl("%d.%d.%d.%d:%u %d.%d.%d.%d:%u %d %d %d %d %d\n",
+			   NIPQUAD(inet->saddr), ntohs(inet->sport),
+			   NIPQUAD(inet->daddr), ntohs(inet->dport), size,
+			   hctx->ccid3hctx_s, hctx->ccid3hctx_rtt,
+			   hctx->ccid3hctx_p, hctx->ccid3hctx_t_ipi);
+		else
+			printl("%d.%d.%d.%d:%u %d.%d.%d.%d:%u %d\n",
+			   NIPQUAD(inet->saddr), ntohs(inet->sport),
+			   NIPQUAD(inet->daddr), ntohs(inet->dport), size);
+	}
+
+	jprobe_return();
+	return 0;
+}
+
+static struct jprobe dccp_send_probe = {
+	.kp	= { .addr = (kprobe_opcode_t *)&dccp_sendmsg, },
+	.entry	= (kprobe_opcode_t *)&jdccp_sendmsg,
+};
+
+static int dccpprobe_open(struct inode *inode, struct file *file)
+{
+	kfifo_reset(dccpw.fifo);
+	do_gettimeofday(&dccpw.tstart);
+	return 0;
+}
+
+static ssize_t dccpprobe_read(struct file *file, char __user *buf,
+			      size_t len, loff_t *ppos)
+{
+	int error = 0, cnt = 0;
+	unsigned char *tbuf;
+
+	if (!buf || len < 0)
+		return -EINVAL;
+
+	if (len == 0)
+		return 0;
+
+	tbuf = vmalloc(len);
+	if (!tbuf)
+		return -ENOMEM;
+
+	error = wait_event_interruptible(dccpw.wait,
+					 __kfifo_len(dccpw.fifo) != 0);
+	if (error)
+		goto out_free;
+
+	cnt = kfifo_get(dccpw.fifo, tbuf, len);
+	error = copy_to_user(buf, tbuf, cnt);
+
+out_free:
+	vfree(tbuf);
+
+	return error ? error : cnt;
+}
+
+static struct file_operations dccpprobe_fops = {
+	.owner	 = THIS_MODULE,
+	.open	 = dccpprobe_open,
+	.read    = dccpprobe_read,
+};
+
+static __init int dccpprobe_init(void)
+{
+	int ret = -ENOMEM;
+
+	init_waitqueue_head(&dccpw.wait);
+	spin_lock_init(&dccpw.lock);
+	dccpw.fifo = kfifo_alloc(bufsize, GFP_KERNEL, &dccpw.lock);
+
+	if (!proc_net_fops_create(procname, S_IRUSR, &dccpprobe_fops))
+		goto err0;
+
+	ret = register_jprobe(&dccp_send_probe);
+	if (ret)
+		goto err1;
+
+	pr_info("DCCP watch registered (port=%d)\n", port);
+	return 0;
+err1:
+	proc_net_remove(procname);
+err0:
+	kfifo_free(dccpw.fifo);
+	return ret;
+}
+module_init(dccpprobe_init);
+
+static __exit void dccpprobe_exit(void)
+{
+	kfifo_free(dccpw.fifo);
+	proc_net_remove(procname);
+	unregister_jprobe(&dccp_send_probe);
+
+}
+module_exit(dccpprobe_exit);
+
+MODULE_PARM_DESC(port, "Port to match (0=all)");
+module_param(port, int, 0);
+
+MODULE_PARM_DESC(bufsize, "Log buffer size (default 64k)");
+module_param(bufsize, int, 0);
+
+MODULE_AUTHOR("Ian McDonald <ian.mcdonald@jandi.co.nz>");
+MODULE_DESCRIPTION("DCCP snooper");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 0d6c7ae22d4fadbc83677ea1a6144eea8911e319 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Sun, 24 Sep 2006 19:28:47 -0700
Subject: [NETFILTER]: Add dscp,DSCP headers to header-y

This patch adds xt_dscp.h and xt_DSCP.h to the kernel headers which are
exported via 'make headers_install'. These are necessary for userspace
to add rules using dscp match and DSCP target.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter/Kbuild | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/netfilter/Kbuild b/include/linux/netfilter/Kbuild
index 9a285cecf249..312bd2ffee33 100644
--- a/include/linux/netfilter/Kbuild
+++ b/include/linux/netfilter/Kbuild
@@ -10,6 +10,8 @@ header-y += xt_connmark.h
 header-y += xt_CONNMARK.h
 header-y += xt_conntrack.h
 header-y += xt_dccp.h
+header-y += xt_dscp.h
+header-y += xt_DSCP.h
 header-y += xt_esp.h
 header-y += xt_helper.h
 header-y += xt_length.h
-- 
cgit v1.2.1


From 5b7c714ec27584b18279b741b6043016f8adb9de Mon Sep 17 00:00:00 2001
From: Roland Dreier <roland@digitalvampire.org>
Date: Sun, 24 Sep 2006 20:09:33 -0700
Subject: [ATM] he: Fix __init/__devinit conflict

he_init_one() is declared __devinit, but calls lots of init functions
that are marked __init.  However, if CONFIG_HOTPLUG is enabled,
__devinit functions go into normal .text, which leads to

    WARNING: drivers/atm/he.o - Section mismatch: reference to .init.text: from .text between 'he_start' (at offset 0x2130) and 'he_service_tbrq'

Fix this by changing the __init functions to __devinit.

Signed-off-by: Roland Dreier <roland@digitalvampire.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/he.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/atm/he.c b/drivers/atm/he.c
index 41e052fecd7f..f2511b42dba2 100644
--- a/drivers/atm/he.c
+++ b/drivers/atm/he.c
@@ -454,7 +454,7 @@ rate_to_atmf(unsigned rate)		/* cps to atm forum format */
 	return (NONZERO | (exp << 9) | (rate & 0x1ff));
 }
 
-static void __init
+static void __devinit
 he_init_rx_lbfp0(struct he_dev *he_dev)
 {
 	unsigned i, lbm_offset, lbufd_index, lbuf_addr, lbuf_count;
@@ -485,7 +485,7 @@ he_init_rx_lbfp0(struct he_dev *he_dev)
 	he_writel(he_dev, he_dev->r0_numbuffs, RLBF0_C);
 }
 
-static void __init
+static void __devinit
 he_init_rx_lbfp1(struct he_dev *he_dev)
 {
 	unsigned i, lbm_offset, lbufd_index, lbuf_addr, lbuf_count;
@@ -516,7 +516,7 @@ he_init_rx_lbfp1(struct he_dev *he_dev)
 	he_writel(he_dev, he_dev->r1_numbuffs, RLBF1_C);
 }
 
-static void __init
+static void __devinit
 he_init_tx_lbfp(struct he_dev *he_dev)
 {
 	unsigned i, lbm_offset, lbufd_index, lbuf_addr, lbuf_count;
@@ -546,7 +546,7 @@ he_init_tx_lbfp(struct he_dev *he_dev)
 	he_writel(he_dev, lbufd_index - 1, TLBF_T);
 }
 
-static int __init
+static int __devinit
 he_init_tpdrq(struct he_dev *he_dev)
 {
 	he_dev->tpdrq_base = pci_alloc_consistent(he_dev->pci_dev,
@@ -568,7 +568,7 @@ he_init_tpdrq(struct he_dev *he_dev)
 	return 0;
 }
 
-static void __init
+static void __devinit
 he_init_cs_block(struct he_dev *he_dev)
 {
 	unsigned clock, rate, delta;
@@ -664,7 +664,7 @@ he_init_cs_block(struct he_dev *he_dev)
 
 }
 
-static int __init
+static int __devinit
 he_init_cs_block_rcm(struct he_dev *he_dev)
 {
 	unsigned (*rategrid)[16][16];
@@ -785,7 +785,7 @@ he_init_cs_block_rcm(struct he_dev *he_dev)
 	return 0;
 }
 
-static int __init
+static int __devinit
 he_init_group(struct he_dev *he_dev, int group)
 {
 	int i;
@@ -955,7 +955,7 @@ he_init_group(struct he_dev *he_dev, int group)
 	return 0;
 }
 
-static int __init
+static int __devinit
 he_init_irq(struct he_dev *he_dev)
 {
 	int i;
-- 
cgit v1.2.1


From 3d2573f7ebe507e372a23cdd3c8b03305d6e90aa Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Sun, 24 Sep 2006 20:11:58 -0700
Subject: [TCP]: default congestion control menu

Change how default TCP congestion control is chosen. Don't just use
last installed module, instead allow selection during configuration,
and make sure and use the default regardless of load order.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/Kconfig           | 45 ++++++++++++++++++++++++++++++++++++++++-----
 net/ipv4/sysctl_net_ipv4.c |  6 ++++++
 net/ipv4/tcp_cong.c        |  2 +-
 3 files changed, 47 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 1650b64415aa..1bbc3689cd8a 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -448,7 +448,7 @@ config INET_TCP_DIAG
 	depends on INET_DIAG
 	def_tristate INET_DIAG
 
-config TCP_CONG_ADVANCED
+menuconfig TCP_CONG_ADVANCED
 	bool "TCP: advanced congestion control"
 	---help---
 	  Support for selection of various TCP congestion control
@@ -459,9 +459,7 @@ config TCP_CONG_ADVANCED
 
 	  If unsure, say N.
 
-# TCP Reno is builtin (required as fallback)
-menu "TCP congestion control"
-	depends on TCP_CONG_ADVANCED
+if TCP_CONG_ADVANCED
 
 config TCP_CONG_BIC
 	tristate "Binary Increase Congestion (BIC) control"
@@ -574,12 +572,49 @@ config TCP_CONG_VENO
 	loss packets.
 	See http://www.ntu.edu.sg/home5/ZHOU0022/papers/CPFu03a.pdf
 
-endmenu
+choice
+	prompt "Default TCP congestion control"
+	default DEFAULT_BIC
+	help
+	  Select the TCP congestion control that will be used by default
+	  for all connections.
+
+	config DEFAULT_BIC
+		bool "Bic" if TCP_CONG_BIC=y
+
+	config DEFAULT_CUBIC
+		bool "Cubic" if TCP_CONG_CUBIC=y
+
+	config DEFAULT_HTCP
+		bool "Htcp" if TCP_CONG_HTCP=y
+
+	config DEFAULT_VEGAS
+		bool "Vegas" if TCP_CONG_VEGAS=y
+
+	config DEFAULT_WESTWOOD
+		bool "Westwood" if TCP_CONG_WESTWOOD=y
+
+	config DEFAULT_RENO
+		bool "Reno"
+
+endchoice
+
+endif
 
 config TCP_CONG_BIC
 	tristate
 	depends on !TCP_CONG_ADVANCED
 	default y
 
+config DEFAULT_TCP_CONG
+	string
+	default "bic" if DEFAULT_BIC
+	default "cubic" if DEFAULT_CUBIC
+	default "htcp" if DEFAULT_HTCP
+	default "vegas" if DEFAULT_VEGAS
+	default "westwood" if DEFAULT_WESTWOOD
+	default "reno" if DEFAULT_RENO
+	default "bic"
+
 source "net/ipv4/ipvs/Kconfig"
 
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 19b2071ff319..e82a5be894b5 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -129,6 +129,12 @@ static int sysctl_tcp_congestion_control(ctl_table *table, int __user *name,
 	return ret;
 }
 
+static int __init tcp_congestion_default(void)
+{
+	return tcp_set_default_congestion_control(CONFIG_DEFAULT_TCP_CONG);
+}
+
+late_initcall(tcp_congestion_default);
 
 ctl_table ipv4_table[] = {
         {
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 7ff2e4273a7c..af0aca1e6be6 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -48,7 +48,7 @@ int tcp_register_congestion_control(struct tcp_congestion_ops *ca)
 		printk(KERN_NOTICE "TCP %s already registered\n", ca->name);
 		ret = -EEXIST;
 	} else {
-		list_add_rcu(&ca->list, &tcp_cong_list);
+		list_add_tail_rcu(&ca->list, &tcp_cong_list);
 		printk(KERN_INFO "TCP %s registered\n", ca->name);
 	}
 	spin_unlock(&tcp_cong_list_lock);
-- 
cgit v1.2.1


From 597811ec167fa01c926a0957a91d9e39baa30e64 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Sun, 24 Sep 2006 20:13:03 -0700
Subject: [TCP]: make cubic the default

Change default congestion control used from BIC to the newer CUBIC
which it the successor to BIC but has better properties over long delay links.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/Kconfig | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 1bbc3689cd8a..30af4a4dfcc8 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -455,7 +455,7 @@ menuconfig TCP_CONG_ADVANCED
 	  modules.
 
 	  Nearly all users can safely say no here, and a safe default
-	  selection will be made (BIC-TCP with new Reno as a fallback).
+	  selection will be made (CUBIC with new Reno as a fallback).
 
 	  If unsure, say N.
 
@@ -463,7 +463,7 @@ if TCP_CONG_ADVANCED
 
 config TCP_CONG_BIC
 	tristate "Binary Increase Congestion (BIC) control"
-	default y
+	default m
 	---help---
 	BIC-TCP is a sender-side only change that ensures a linear RTT
 	fairness under large windows while offering both scalability and
@@ -477,7 +477,7 @@ config TCP_CONG_BIC
 
 config TCP_CONG_CUBIC
 	tristate "CUBIC TCP"
-	default m
+	default y
 	---help---
 	This is version 2.0 of BIC-TCP which uses a cubic growth function
 	among other techniques.
@@ -574,7 +574,7 @@ config TCP_CONG_VENO
 
 choice
 	prompt "Default TCP congestion control"
-	default DEFAULT_BIC
+	default DEFAULT_CUBIC
 	help
 	  Select the TCP congestion control that will be used by default
 	  for all connections.
@@ -601,7 +601,7 @@ endchoice
 
 endif
 
-config TCP_CONG_BIC
+config TCP_CONG_CUBIC
 	tristate
 	depends on !TCP_CONG_ADVANCED
 	default y
@@ -614,7 +614,7 @@ config DEFAULT_TCP_CONG
 	default "vegas" if DEFAULT_VEGAS
 	default "westwood" if DEFAULT_WESTWOOD
 	default "reno" if DEFAULT_RENO
-	default "bic"
+	default "cubic"
 
 source "net/ipv4/ipvs/Kconfig"
 
-- 
cgit v1.2.1


From a6d967a485c67ec8a1276261f39d81ace6a3e308 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 25 Sep 2006 15:33:09 -0400
Subject: [libata] No need for all those arch libata-portmap.h headers

They all contain the same thing.  Instead, have a single generic one in
include/asm-generic, and permit an arch to override as needed.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 include/asm-alpha/libata-portmap.h   | 1 -
 include/asm-frv/libata-portmap.h     | 1 -
 include/asm-i386/libata-portmap.h    | 1 -
 include/asm-ia64/libata-portmap.h    | 1 -
 include/asm-powerpc/libata-portmap.h | 1 -
 include/asm-sparc/libata-portmap.h   | 1 -
 include/asm-sparc64/libata-portmap.h | 1 -
 include/asm-x86_64/libata-portmap.h  | 1 -
 include/linux/libata.h               | 8 ++++++++
 9 files changed, 8 insertions(+), 8 deletions(-)
 delete mode 100644 include/asm-alpha/libata-portmap.h
 delete mode 100644 include/asm-frv/libata-portmap.h
 delete mode 100644 include/asm-i386/libata-portmap.h
 delete mode 100644 include/asm-ia64/libata-portmap.h
 delete mode 100644 include/asm-powerpc/libata-portmap.h
 delete mode 100644 include/asm-sparc/libata-portmap.h
 delete mode 100644 include/asm-sparc64/libata-portmap.h
 delete mode 100644 include/asm-x86_64/libata-portmap.h

diff --git a/include/asm-alpha/libata-portmap.h b/include/asm-alpha/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-alpha/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-frv/libata-portmap.h b/include/asm-frv/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-frv/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-i386/libata-portmap.h b/include/asm-i386/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-i386/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-ia64/libata-portmap.h b/include/asm-ia64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-ia64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-powerpc/libata-portmap.h b/include/asm-powerpc/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-powerpc/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc/libata-portmap.h b/include/asm-sparc/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-sparc/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-sparc64/libata-portmap.h b/include/asm-sparc64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-sparc64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/asm-x86_64/libata-portmap.h b/include/asm-x86_64/libata-portmap.h
deleted file mode 100644
index 75484ef0c743..000000000000
--- a/include/asm-x86_64/libata-portmap.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/libata-portmap.h>
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 1ef3d3901b47..d6a3d4b345fc 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -36,7 +36,15 @@
 #include <linux/workqueue.h>
 #include <scsi/scsi_host.h>
 
+/*
+ * Define if arch has non-standard setup.  This is a _PCI_ standard
+ * not a legacy or ISA standard.
+ */
+#ifdef CONFIG_ATA_NONSTANDARD
 #include <asm/libata-portmap.h>
+#else
+#include <asm-generic/libata-portmap.h>
+#endif
 
 /*
  * compile-time options: to be removed as soon as all the drivers are
-- 
cgit v1.2.1


From baef186519c69b11cf7e48c26e75feb1e6173baa Mon Sep 17 00:00:00 2001
From: "John W. Linville" <linville@tuxdriver.com>
Date: Fri, 8 Sep 2006 16:04:05 -0400
Subject: [PATCH] WE-21 support (core API)

This is version 21 of the Wireless Extensions. Changelog :
	o finishes migrating the ESSID API (remove the +1)
	o netdev->get_wireless_stats is no more
	o long/short retry

This is a redacted version of a patch originally submitted by Jean
Tourrilhes.  I removed most of the additions, in order to minimize
future support requirements for nl80211 (or other WE successor).

CC: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 include/linux/netdevice.h |  1 -
 include/linux/wireless.h  | 24 ++++++++++++++---
 net/core/net-sysfs.c      |  5 +---
 net/core/wireless.c       | 67 +++++++++++++++++++++++++++++------------------
 4 files changed, 62 insertions(+), 35 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 43289127b458..ac4f50213bc3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -334,7 +334,6 @@ struct net_device
 
 
 	struct net_device_stats* (*get_stats)(struct net_device *dev);
-	struct iw_statistics*	(*get_wireless_stats)(struct net_device *dev);
 
 	/* List of functions to handle Wireless Extensions (instead of ioctl).
 	 * See <net/iw_handler.h> for details. Jean II */
diff --git a/include/linux/wireless.h b/include/linux/wireless.h
index 13588564b42b..a50a0130fd9e 100644
--- a/include/linux/wireless.h
+++ b/include/linux/wireless.h
@@ -1,7 +1,7 @@
 /*
  * This file define a set of standard wireless extensions
  *
- * Version :	20	17.2.06
+ * Version :	21	14.3.06
  *
  * Authors :	Jean Tourrilhes - HPL - <jt@hpl.hp.com>
  * Copyright (c) 1997-2006 Jean Tourrilhes, All Rights Reserved.
@@ -69,9 +69,14 @@
 
 /***************************** INCLUDES *****************************/
 
+/* This header is used in user-space, therefore need to be sanitised
+ * for that purpose. Those includes are usually not compatible with glibc.
+ * To know which includes to use in user-space, check iwlib.h. */
+#ifdef __KERNEL__
 #include <linux/types.h>		/* for "caddr_t" et al		*/
 #include <linux/socket.h>		/* for "struct sockaddr" et al	*/
 #include <linux/if.h>			/* for IFNAMSIZ and co... */
+#endif	/* __KERNEL__ */
 
 /***************************** VERSION *****************************/
 /*
@@ -80,7 +85,7 @@
  * (there is some stuff that will be added in the future...)
  * I just plan to increment with each new version.
  */
-#define WIRELESS_EXT	20
+#define WIRELESS_EXT	21
 
 /*
  * Changes :
@@ -208,6 +213,14 @@
  * V19 to V20
  * ----------
  *	- RtNetlink requests support (SET/GET)
+ *
+ * V20 to V21
+ * ----------
+ *	- Remove (struct net_device *)->get_wireless_stats()
+ *	- Change length in ESSID and NICK to strlen() instead of strlen()+1
+ *	- Add IW_RETRY_SHORT/IW_RETRY_LONG retry modifiers
+ *	- Power/Retry relative values no longer * 100000
+ *	- Add explicit flag to tell stats are in 802.11k RCPI : IW_QUAL_RCPI
  */
 
 /**************************** CONSTANTS ****************************/
@@ -448,6 +461,7 @@
 #define IW_QUAL_QUAL_INVALID	0x10	/* Driver doesn't provide value */
 #define IW_QUAL_LEVEL_INVALID	0x20
 #define IW_QUAL_NOISE_INVALID	0x40
+#define IW_QUAL_RCPI		0x80	/* Level + Noise are 802.11k RCPI */
 #define IW_QUAL_ALL_INVALID	0x70
 
 /* Frequency flags */
@@ -500,10 +514,12 @@
 #define IW_RETRY_TYPE		0xF000	/* Type of parameter */
 #define IW_RETRY_LIMIT		0x1000	/* Maximum number of retries*/
 #define IW_RETRY_LIFETIME	0x2000	/* Maximum duration of retries in us */
-#define IW_RETRY_MODIFIER	0x000F	/* Modify a parameter */
+#define IW_RETRY_MODIFIER	0x00FF	/* Modify a parameter */
 #define IW_RETRY_MIN		0x0001	/* Value is a minimum  */
 #define IW_RETRY_MAX		0x0002	/* Value is a maximum */
 #define IW_RETRY_RELATIVE	0x0004	/* Value is not in seconds/ms/us */
+#define IW_RETRY_SHORT		0x0010	/* Value is for short packets  */
+#define IW_RETRY_LONG		0x0020	/* Value is for long packets */
 
 /* Scanning request flags */
 #define IW_SCAN_DEFAULT		0x0000	/* Default scan of the driver */
@@ -1017,7 +1033,7 @@ struct	iw_range
 	/* Note : this frequency list doesn't need to fit channel numbers,
 	 * because each entry contain its channel index */
 
-	__u32		enc_capa; /* IW_ENC_CAPA_* bit field */
+	__u32		enc_capa;	/* IW_ENC_CAPA_* bit field */
 };
 
 /*
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 13472762b18b..f47f319bb7dc 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -344,8 +344,6 @@ static ssize_t wireless_show(struct class_device *cd, char *buf,
 		if(dev->wireless_handlers &&
 		   dev->wireless_handlers->get_wireless_stats)
 			iw = dev->wireless_handlers->get_wireless_stats(dev);
-		else if (dev->get_wireless_stats)
-			iw = dev->get_wireless_stats(dev);
 		if (iw != NULL)
 			ret = (*format)(iw, buf);
 	}
@@ -465,8 +463,7 @@ int netdev_register_sysfs(struct net_device *net)
 		*groups++ = &netstat_group;
 
 #ifdef WIRELESS_EXT
-	if (net->get_wireless_stats
-	    || (net->wireless_handlers && net->wireless_handlers->get_wireless_stats))
+	if (net->wireless_handlers && net->wireless_handlers->get_wireless_stats)
 		*groups++ = &wireless_group;
 #endif
 
diff --git a/net/core/wireless.c b/net/core/wireless.c
index 3168fca312f7..ffff0da46c6e 100644
--- a/net/core/wireless.c
+++ b/net/core/wireless.c
@@ -68,6 +68,14 @@
  *
  * v8 - 17.02.06 - Jean II
  *	o RtNetlink requests support (SET/GET)
+ *
+ * v8b - 03.08.06 - Herbert Xu
+ *	o Fix Wireless Event locking issues.
+ *
+ * v9 - 14.3.06 - Jean II
+ *	o Change length in ESSID and NICK to strlen() instead of strlen()+1
+ *	o Make standard_ioctl_num and standard_event_num unsigned
+ *	o Remove (struct net_device *)->get_wireless_stats()
  */
 
 /***************************** INCLUDES *****************************/
@@ -234,24 +242,24 @@ static const struct iw_ioctl_description standard_ioctl[] = {
 	[SIOCSIWESSID	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 		.flags		= IW_DESCR_FLAG_EVENT,
 	},
 	[SIOCGIWESSID	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 		.flags		= IW_DESCR_FLAG_DUMP,
 	},
 	[SIOCSIWNICKN	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 	},
 	[SIOCGIWNICKN	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_POINT,
 		.token_size	= 1,
-		.max_tokens	= IW_ESSID_MAX_SIZE + 1,
+		.max_tokens	= IW_ESSID_MAX_SIZE,
 	},
 	[SIOCSIWRATE	- SIOCIWFIRST] = {
 		.header_type	= IW_HEADER_TYPE_PARAM,
@@ -338,8 +346,8 @@ static const struct iw_ioctl_description standard_ioctl[] = {
 		.max_tokens	= sizeof(struct iw_pmksa),
 	},
 };
-static const int standard_ioctl_num = (sizeof(standard_ioctl) /
-				       sizeof(struct iw_ioctl_description));
+static const unsigned standard_ioctl_num = (sizeof(standard_ioctl) /
+					    sizeof(struct iw_ioctl_description));
 
 /*
  * Meta-data about all the additional standard Wireless Extension events
@@ -389,8 +397,8 @@ static const struct iw_ioctl_description standard_event[] = {
 		.max_tokens	= sizeof(struct iw_pmkid_cand),
 	},
 };
-static const int standard_event_num = (sizeof(standard_event) /
-				       sizeof(struct iw_ioctl_description));
+static const unsigned standard_event_num = (sizeof(standard_event) /
+					    sizeof(struct iw_ioctl_description));
 
 /* Size (in bytes) of the various private data types */
 static const char iw_priv_type_size[] = {
@@ -465,17 +473,6 @@ static inline struct iw_statistics *get_wireless_stats(struct net_device *dev)
 	   (dev->wireless_handlers->get_wireless_stats != NULL))
 		return dev->wireless_handlers->get_wireless_stats(dev);
 
-	/* Old location, field to be removed in next WE */
-	if(dev->get_wireless_stats) {
-		static int printed_message;
-
-		if (!printed_message++)
-			printk(KERN_DEBUG "%s (WE) : Driver using old /proc/net/wireless support, please fix driver !\n",
-				dev->name);
-
-		return dev->get_wireless_stats(dev);
-	}
-
 	/* Not found */
 	return (struct iw_statistics *) NULL;
 }
@@ -1843,8 +1840,33 @@ int wireless_rtnetlink_set(struct net_device *	dev,
  */
 
 #ifdef WE_EVENT_RTNETLINK
+/* ---------------------------------------------------------------- */
+/*
+ * Locking...
+ * ----------
+ *
+ * Thanks to Herbert Xu <herbert@gondor.apana.org.au> for fixing
+ * the locking issue in here and implementing this code !
+ *
+ * The issue : wireless_send_event() is often called in interrupt context,
+ * while the Netlink layer can never be called in interrupt context.
+ * The fully formed RtNetlink events are queued, and then a tasklet is run
+ * to feed those to Netlink.
+ * The skb_queue is interrupt safe, and its lock is not held while calling
+ * Netlink, so there is no possibility of dealock.
+ * Jean II
+ */
+
 static struct sk_buff_head wireless_nlevent_queue;
 
+static int __init wireless_nlevent_init(void)
+{
+	skb_queue_head_init(&wireless_nlevent_queue);
+	return 0;
+}
+
+subsys_initcall(wireless_nlevent_init);
+
 static void wireless_nlevent_process(unsigned long data)
 {
 	struct sk_buff *skb;
@@ -1921,13 +1943,6 @@ static inline void rtmsg_iwinfo(struct net_device *	dev,
 	tasklet_schedule(&wireless_nlevent_tasklet);
 }
 
-static int __init wireless_nlevent_init(void)
-{
-	skb_queue_head_init(&wireless_nlevent_queue);
-	return 0;
-}
-
-subsys_initcall(wireless_nlevent_init);
 #endif	/* WE_EVENT_RTNETLINK */
 
 /* ---------------------------------------------------------------- */
-- 
cgit v1.2.1


From 7f8544cc95c7f521847fa760ce38d932e6ab4542 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 17:58:11 -0700
Subject: [PATCH] WE-21 for airo

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/airo.c | 19 ++++++++++---------
 1 file changed, 10 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/airo.c b/drivers/net/wireless/airo.c
index bff04cba3fed..ba737c6cebec 100644
--- a/drivers/net/wireless/airo.c
+++ b/drivers/net/wireless/airo.c
@@ -5868,7 +5868,7 @@ static int airo_set_essid(struct net_device *dev,
 		int	index = (dwrq->flags & IW_ENCODE_INDEX) - 1;
 
 		/* Check the size of the string */
-		if(dwrq->length > IW_ESSID_MAX_SIZE+1) {
+		if(dwrq->length > IW_ESSID_MAX_SIZE) {
 			return -E2BIG ;
 		}
 		/* Check if index is valid */
@@ -5880,7 +5880,7 @@ static int airo_set_essid(struct net_device *dev,
 		memset(SSID_rid.ssids[index].ssid, 0,
 		       sizeof(SSID_rid.ssids[index].ssid));
 		memcpy(SSID_rid.ssids[index].ssid, extra, dwrq->length);
-		SSID_rid.ssids[index].len = dwrq->length - 1;
+		SSID_rid.ssids[index].len = dwrq->length;
 	}
 	SSID_rid.len = sizeof(SSID_rid);
 	/* Write it to the card */
@@ -5990,7 +5990,7 @@ static int airo_set_nick(struct net_device *dev,
 	struct airo_info *local = dev->priv;
 
 	/* Check the size of the string */
-	if(dwrq->length > 16 + 1) {
+	if(dwrq->length > 16) {
 		return -E2BIG;
 	}
 	readConfigRid(local, 1);
@@ -6015,7 +6015,7 @@ static int airo_get_nick(struct net_device *dev,
 	readConfigRid(local, 1);
 	strncpy(extra, local->config.nodeName, 16);
 	extra[16] = '\0';
-	dwrq->length = strlen(extra) + 1;
+	dwrq->length = strlen(extra);
 
 	return 0;
 }
@@ -6767,9 +6767,9 @@ static int airo_set_retry(struct net_device *dev,
 	}
 	readConfigRid(local, 1);
 	if(vwrq->flags & IW_RETRY_LIMIT) {
-		if(vwrq->flags & IW_RETRY_MAX)
+		if(vwrq->flags & IW_RETRY_LONG)
 			local->config.longRetryLimit = vwrq->value;
-		else if (vwrq->flags & IW_RETRY_MIN)
+		else if (vwrq->flags & IW_RETRY_SHORT)
 			local->config.shortRetryLimit = vwrq->value;
 		else {
 			/* No modifier : set both */
@@ -6805,14 +6805,14 @@ static int airo_get_retry(struct net_device *dev,
 	if((vwrq->flags & IW_RETRY_TYPE) == IW_RETRY_LIFETIME) {
 		vwrq->flags = IW_RETRY_LIFETIME;
 		vwrq->value = (int)local->config.txLifetime * 1024;
-	} else if((vwrq->flags & IW_RETRY_MAX)) {
-		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+	} else if((vwrq->flags & IW_RETRY_LONG)) {
+		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 		vwrq->value = (int)local->config.longRetryLimit;
 	} else {
 		vwrq->flags = IW_RETRY_LIMIT;
 		vwrq->value = (int)local->config.shortRetryLimit;
 		if((int)local->config.shortRetryLimit != (int)local->config.longRetryLimit)
-			vwrq->flags |= IW_RETRY_MIN;
+			vwrq->flags |= IW_RETRY_SHORT;
 	}
 
 	return 0;
@@ -6990,6 +6990,7 @@ static int airo_set_power(struct net_device *dev,
 			local->config.rmode |= RXMODE_BC_MC_ADDR;
 			set_bit (FLAG_COMMIT, &local->flags);
 		case IW_POWER_ON:
+			/* This is broken, fixme ;-) */
 			break;
 		default:
 			return -EINVAL;
-- 
cgit v1.2.1


From 6a484db472e77218252025d31d4ef96dbc11ada9 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 17:59:03 -0700
Subject: [PATCH] WE-21 for atmel

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/atmel.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c
index 995c7bea5897..0fc267d626dc 100644
--- a/drivers/net/wireless/atmel.c
+++ b/drivers/net/wireless/atmel.c
@@ -1656,13 +1656,13 @@ static int atmel_set_essid(struct net_device *dev,
 		priv->connect_to_any_BSS = 0;
 
 		/* Check the size of the string */
-		if (dwrq->length > MAX_SSID_LENGTH + 1)
+		if (dwrq->length > MAX_SSID_LENGTH)
 			 return -E2BIG;
 		if (index != 0)
 			return -EINVAL;
 
-		memcpy(priv->new_SSID, extra, dwrq->length - 1);
-		priv->new_SSID_size = dwrq->length - 1;
+		memcpy(priv->new_SSID, extra, dwrq->length);
+		priv->new_SSID_size = dwrq->length;
 	}
 
 	return -EINPROGRESS;
@@ -2120,9 +2120,9 @@ static int atmel_set_retry(struct net_device *dev,
 	struct atmel_private *priv = netdev_priv(dev);
 
 	if (!vwrq->disabled && (vwrq->flags & IW_RETRY_LIMIT)) {
-		if (vwrq->flags & IW_RETRY_MAX)
+		if (vwrq->flags & IW_RETRY_LONG)
 			priv->long_retry = vwrq->value;
-		else if (vwrq->flags & IW_RETRY_MIN)
+		else if (vwrq->flags & IW_RETRY_SHORT)
 			priv->short_retry = vwrq->value;
 		else {
 			/* No modifier : set both */
@@ -2144,15 +2144,15 @@ static int atmel_get_retry(struct net_device *dev,
 
 	vwrq->disabled = 0;      /* Can't be disabled */
 
-	/* Note : by default, display the min retry number */
-	if (vwrq->flags & IW_RETRY_MAX) {
-		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+	/* Note : by default, display the short retry number */
+	if (vwrq->flags & IW_RETRY_LONG) {
+		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 		vwrq->value = priv->long_retry;
 	} else {
 		vwrq->flags = IW_RETRY_LIMIT;
 		vwrq->value = priv->short_retry;
 		if (priv->long_retry != priv->short_retry)
-			vwrq->flags |= IW_RETRY_MIN;
+			vwrq->flags |= IW_RETRY_SHORT;
 	}
 
 	return 0;
-- 
cgit v1.2.1


From 9fb08363f1f6d360dbaf6d7f51b9e7ca07c05ecd Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 17:59:47 -0700
Subject: [PATCH] WE-21 for hostap

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/hostap/hostap_ioctl.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/hostap/hostap_ioctl.c b/drivers/net/wireless/hostap/hostap_ioctl.c
index 7a4978516eac..d061fb3443ff 100644
--- a/drivers/net/wireless/hostap/hostap_ioctl.c
+++ b/drivers/net/wireless/hostap/hostap_ioctl.c
@@ -1412,9 +1412,9 @@ static int prism2_ioctl_siwretry(struct net_device *dev,
 	/* what could be done, if firmware would support this.. */
 
 	if (rrq->flags & IW_RETRY_LIMIT) {
-		if (rrq->flags & IW_RETRY_MAX)
+		if (rrq->flags & IW_RETRY_LONG)
 			HFA384X_RID_LONGRETRYLIMIT = rrq->value;
-		else if (rrq->flags & IW_RETRY_MIN)
+		else if (rrq->flags & IW_RETRY_SHORT)
 			HFA384X_RID_SHORTRETRYLIMIT = rrq->value;
 		else {
 			HFA384X_RID_LONGRETRYLIMIT = rrq->value;
@@ -1468,14 +1468,14 @@ static int prism2_ioctl_giwretry(struct net_device *dev,
 				rrq->value = le16_to_cpu(altretry);
 			else
 				rrq->value = local->manual_retry_count;
-		} else if ((rrq->flags & IW_RETRY_MAX)) {
-			rrq->flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+		} else if ((rrq->flags & IW_RETRY_LONG)) {
+			rrq->flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 			rrq->value = longretry;
 		} else {
 			rrq->flags = IW_RETRY_LIMIT;
 			rrq->value = shortretry;
 			if (shortretry != longretry)
-				rrq->flags |= IW_RETRY_MIN;
+				rrq->flags |= IW_RETRY_SHORT;
 		}
 	}
 	return 0;
-- 
cgit v1.2.1


From 5b63bae0ab750942e84bfb9b353e6222583457a2 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:00:48 -0700
Subject: [PATCH] WE-21 for ipw2100

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ipw2100.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/ipw2100.c b/drivers/net/wireless/ipw2100.c
index b4d81a04c895..6c5add701a6f 100644
--- a/drivers/net/wireless/ipw2100.c
+++ b/drivers/net/wireless/ipw2100.c
@@ -6958,7 +6958,7 @@ static int ipw2100_wx_set_essid(struct net_device *dev,
 	}
 
 	if (wrqu->essid.flags && wrqu->essid.length) {
-		length = wrqu->essid.length - 1;
+		length = wrqu->essid.length;
 		essid = extra;
 	}
 
@@ -7051,7 +7051,7 @@ static int ipw2100_wx_get_nick(struct net_device *dev,
 
 	struct ipw2100_priv *priv = ieee80211_priv(dev);
 
-	wrqu->data.length = strlen(priv->nick) + 1;
+	wrqu->data.length = strlen(priv->nick);
 	memcpy(extra, priv->nick, wrqu->data.length);
 	wrqu->data.flags = 1;	/* active */
 
@@ -7343,14 +7343,14 @@ static int ipw2100_wx_set_retry(struct net_device *dev,
 		goto done;
 	}
 
-	if (wrqu->retry.flags & IW_RETRY_MIN) {
+	if (wrqu->retry.flags & IW_RETRY_SHORT) {
 		err = ipw2100_set_short_retry(priv, wrqu->retry.value);
 		IPW_DEBUG_WX("SET Short Retry Limit -> %d \n",
 			     wrqu->retry.value);
 		goto done;
 	}
 
-	if (wrqu->retry.flags & IW_RETRY_MAX) {
+	if (wrqu->retry.flags & IW_RETRY_LONG) {
 		err = ipw2100_set_long_retry(priv, wrqu->retry.value);
 		IPW_DEBUG_WX("SET Long Retry Limit -> %d \n",
 			     wrqu->retry.value);
@@ -7383,14 +7383,14 @@ static int ipw2100_wx_get_retry(struct net_device *dev,
 	if ((wrqu->retry.flags & IW_RETRY_TYPE) == IW_RETRY_LIFETIME)
 		return -EINVAL;
 
-	if (wrqu->retry.flags & IW_RETRY_MAX) {
-		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+	if (wrqu->retry.flags & IW_RETRY_LONG) {
+		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 		wrqu->retry.value = priv->long_retry_limit;
 	} else {
 		wrqu->retry.flags =
 		    (priv->short_retry_limit !=
 		     priv->long_retry_limit) ?
-		    IW_RETRY_LIMIT | IW_RETRY_MIN : IW_RETRY_LIMIT;
+		    IW_RETRY_LIMIT | IW_RETRY_SHORT : IW_RETRY_LIMIT;
 
 		wrqu->retry.value = priv->short_retry_limit;
 	}
-- 
cgit v1.2.1


From eeec9f1a931262d69811135092c8447d6dccc3e6 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:02:31 -0700
Subject: [PATCH] WE-21 for orinoco

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/orinoco.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c
index 1840b69e3cb5..9e19a963febc 100644
--- a/drivers/net/wireless/orinoco.c
+++ b/drivers/net/wireless/orinoco.c
@@ -3037,7 +3037,7 @@ static int orinoco_ioctl_getessid(struct net_device *dev,
 	}
 
 	erq->flags = 1;
-	erq->length = strlen(essidbuf) + 1;
+	erq->length = strlen(essidbuf);
 
 	return 0;
 }
@@ -3078,7 +3078,7 @@ static int orinoco_ioctl_getnick(struct net_device *dev,
 	memcpy(nickbuf, priv->nick, IW_ESSID_MAX_SIZE+1);
 	orinoco_unlock(priv, &flags);
 
-	nrq->length = strlen(nickbuf)+1;
+	nrq->length = strlen(nickbuf);
 
 	return 0;
 }
@@ -3575,14 +3575,14 @@ static int orinoco_ioctl_getretry(struct net_device *dev,
 		rrq->value = lifetime * 1000;	/* ??? */
 	} else {
 		/* By default, display the min number */
-		if ((rrq->flags & IW_RETRY_MAX)) {
-			rrq->flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+		if ((rrq->flags & IW_RETRY_LONG)) {
+			rrq->flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 			rrq->value = long_limit;
 		} else {
 			rrq->flags = IW_RETRY_LIMIT;
 			rrq->value = short_limit;
 			if(short_limit != long_limit)
-				rrq->flags |= IW_RETRY_MIN;
+				rrq->flags |= IW_RETRY_SHORT;
 		}
 	}
 
-- 
cgit v1.2.1


From bad04f2ce00da5aaa6c8e97b1b80ad2ebd755b8b Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:03:23 -0700
Subject: [PATCH] WE-21 for Prism54

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/prism54/isl_ioctl.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/prism54/isl_ioctl.c b/drivers/net/wireless/prism54/isl_ioctl.c
index c09fbf733b3a..286325ca3293 100644
--- a/drivers/net/wireless/prism54/isl_ioctl.c
+++ b/drivers/net/wireless/prism54/isl_ioctl.c
@@ -742,9 +742,9 @@ prism54_set_essid(struct net_device *ndev, struct iw_request_info *info,
 
 	/* Check if we were asked for `any' */
 	if (dwrq->flags && dwrq->length) {
-		if (dwrq->length > min(33, IW_ESSID_MAX_SIZE + 1))
+		if (dwrq->length > 32)
 			return -E2BIG;
-		essid.length = dwrq->length - 1;
+		essid.length = dwrq->length;
 		memcpy(essid.octets, extra, dwrq->length);
 	} else
 		essid.length = 0;
@@ -814,7 +814,7 @@ prism54_get_nick(struct net_device *ndev, struct iw_request_info *info,
 	dwrq->length = 0;
 
 	down_read(&priv->mib_sem);
-	dwrq->length = strlen(priv->nickname) + 1;
+	dwrq->length = strlen(priv->nickname);
 	memcpy(extra, priv->nickname, dwrq->length);
 	up_read(&priv->mib_sem);
 
@@ -992,9 +992,9 @@ prism54_set_retry(struct net_device *ndev, struct iw_request_info *info,
 		return -EINVAL;
 
 	if (vwrq->flags & IW_RETRY_LIMIT) {
-		if (vwrq->flags & IW_RETRY_MIN)
+		if (vwrq->flags & IW_RETRY_SHORT)
 			slimit = vwrq->value;
-		else if (vwrq->flags & IW_RETRY_MAX)
+		else if (vwrq->flags & IW_RETRY_LONG)
 			llimit = vwrq->value;
 		else {
 			/* we are asked to set both */
@@ -1035,18 +1035,18 @@ prism54_get_retry(struct net_device *ndev, struct iw_request_info *info,
 		    mgt_get_request(priv, DOT11_OID_MAXTXLIFETIME, 0, NULL, &r);
 		vwrq->value = r.u * 1024;
 		vwrq->flags = IW_RETRY_LIFETIME;
-	} else if ((vwrq->flags & IW_RETRY_MAX)) {
+	} else if ((vwrq->flags & IW_RETRY_LONG)) {
 		/* we are asked for the long retry limit */
 		rvalue |=
 		    mgt_get_request(priv, DOT11_OID_LONGRETRIES, 0, NULL, &r);
 		vwrq->value = r.u;
-		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 	} else {
 		/* default. get the  short retry limit */
 		rvalue |=
 		    mgt_get_request(priv, DOT11_OID_SHORTRETRIES, 0, NULL, &r);
 		vwrq->value = r.u;
-		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_MIN;
+		vwrq->flags = IW_RETRY_LIMIT | IW_RETRY_SHORT;
 	}
 
 	return rvalue;
-- 
cgit v1.2.1


From 4ced38ac48137e1928dd40de3cdc074ec31c059d Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:04:09 -0700
Subject: [PATCH] WE-21 for ray_cs

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ray_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
index 4574290f971f..e82548ea609a 100644
--- a/drivers/net/wireless/ray_cs.c
+++ b/drivers/net/wireless/ray_cs.c
@@ -1173,7 +1173,7 @@ static int ray_set_essid(struct net_device *dev,
 		return -EOPNOTSUPP;
 	} else {
 		/* Check the size of the string */
-		if(dwrq->length > IW_ESSID_MAX_SIZE + 1) {
+		if(dwrq->length > IW_ESSID_MAX_SIZE) {
 			return -E2BIG;
 		}
 
-- 
cgit v1.2.1


From de9621bc5637be5f9e3e3ff2fc1475b3de41132e Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:05:37 -0700
Subject: [PATCH] WE-21 for wl3501

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/wl3501_cs.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/wl3501_cs.c b/drivers/net/wireless/wl3501_cs.c
index e0d294c12970..e3ae5f60d5be 100644
--- a/drivers/net/wireless/wl3501_cs.c
+++ b/drivers/net/wireless/wl3501_cs.c
@@ -1802,15 +1802,15 @@ static int wl3501_get_retry(struct net_device *dev,
 				      &retry, sizeof(retry));
 	if (rc)
 		goto out;
-	if (wrqu->retry.flags & IW_RETRY_MAX) {
-		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+	if (wrqu->retry.flags & IW_RETRY_LONG) {
+		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 		goto set_value;
 	}
 	rc = wl3501_get_mib_value(this, WL3501_MIB_ATTR_SHORT_RETRY_LIMIT,
 				  &retry, sizeof(retry));
 	if (rc)
 		goto out;
-	wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MIN;
+	wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_SHORT;
 set_value:
 	wrqu->retry.value = retry;
 	wrqu->retry.disabled = 0;
-- 
cgit v1.2.1


From 22b99262f59ddf5d283e19047e37dbc3be907e93 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:07:11 -0700
Subject: [PATCH] WE-21 for zd1201

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1201.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c
index c52e9bcf8d02..a44dda952c67 100644
--- a/drivers/net/wireless/zd1201.c
+++ b/drivers/net/wireless/zd1201.c
@@ -1218,7 +1218,7 @@ static int zd1201_set_essid(struct net_device *dev,
 		return -EINVAL;
 	if (data->length < 1)
 		data->length = 1;
-	zd->essidlen = data->length-1;
+	zd->essidlen = data->length;
 	memset(zd->essid, 0, IW_ESSID_MAX_SIZE+1);
 	memcpy(zd->essid, essid, data->length);
 	return zd1201_join(zd, zd->essid, zd->essidlen);
-- 
cgit v1.2.1


From a6082f4032a9667e844fecd974f17268249fb966 Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Wed, 30 Aug 2006 03:46:27 +0100
Subject: [PATCH] WE-21 for zd1211rw

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_netdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/zd1211rw/zd_netdev.c b/drivers/net/wireless/zd1211rw/zd_netdev.c
index 440ef24b5fd1..af3a7b36d078 100644
--- a/drivers/net/wireless/zd1211rw/zd_netdev.c
+++ b/drivers/net/wireless/zd1211rw/zd_netdev.c
@@ -82,7 +82,7 @@ static int iw_get_nick(struct net_device *netdev,
 		       union iwreq_data *req, char *extra)
 {
 	strcpy(extra, "zd1211");
-	req->data.length = strlen(extra) + 1;
+	req->data.length = strlen(extra);
 	req->data.flags = 1;
 	return 0;
 }
-- 
cgit v1.2.1


From b978d0278c3a4c41bda806743c6ef5dca86b4c61 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 31 Aug 2006 10:01:39 -0500
Subject: [PATCH] bcm43xx: WE-21 support

Patch to make bcm43xx-softmac be compatible with the revised SSID
length of WE-21.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/bcm43xx/bcm43xx_wx.c   | 2 +-
 net/ieee80211/softmac/ieee80211softmac_wx.c | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
index 888077fc14c4..9b7b15cf6561 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_wx.c
@@ -334,7 +334,7 @@ static int bcm43xx_wx_get_nick(struct net_device *net_dev,
 	size_t len;
 
 	mutex_lock(&bcm->mutex);
-	len = strlen(bcm->nick) + 1;
+	len = strlen(bcm->nick);
 	memcpy(extra, bcm->nick, len);
 	data->data.length = (__u16)len;
 	data->data.flags = 1;
diff --git a/net/ieee80211/softmac/ieee80211softmac_wx.c b/net/ieee80211/softmac/ieee80211softmac_wx.c
index 75320b6842ab..2aa779d18f38 100644
--- a/net/ieee80211/softmac/ieee80211softmac_wx.c
+++ b/net/ieee80211/softmac/ieee80211softmac_wx.c
@@ -80,10 +80,10 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev,
 	 * If it's our network, ignore the change, we're already doing it!
 	 */
 	if((sm->associnfo.associating || sm->associated) &&
-	   (data->essid.flags && data->essid.length && extra)) {
+	   (data->essid.flags && data->essid.length)) {
 		/* Get the associating network */
 		n = ieee80211softmac_get_network_by_bssid(sm, sm->associnfo.bssid);
-		if(n && n->essid.len == (data->essid.length - 1) &&
+		if(n && n->essid.len == data->essid.length &&
 		   !memcmp(n->essid.data, extra, n->essid.len)) {
 			dprintk(KERN_INFO PFX "Already associating or associated to "MAC_FMT"\n",
 				MAC_ARG(sm->associnfo.bssid));
@@ -109,8 +109,8 @@ ieee80211softmac_wx_set_essid(struct net_device *net_dev,
 	sm->associnfo.static_essid = 0;
 	sm->associnfo.assoc_wait = 0;
 
-	if (data->essid.flags && data->essid.length && extra /*required?*/) {
-		length = min(data->essid.length - 1, IW_ESSID_MAX_SIZE);
+	if (data->essid.flags && data->essid.length) {
+		length = min((int)data->essid.length, IW_ESSID_MAX_SIZE);
 		if (length) {
 			memcpy(sm->associnfo.req_essid.data, extra, length);
 			sm->associnfo.static_essid = 1;
-- 
cgit v1.2.1


From 919ee6ddcd3fcff09dee90c11af17a802196ad1f Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Tue, 29 Aug 2006 18:01:40 -0700
Subject: [PATCH] WE-21 for ipw2200

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/ipw2200.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c
index 7358664e0908..5685d7ba55bb 100644
--- a/drivers/net/wireless/ipw2200.c
+++ b/drivers/net/wireless/ipw2200.c
@@ -8875,8 +8875,6 @@ static int ipw_wx_set_essid(struct net_device *dev,
         }
 
 	length = min((int)wrqu->essid.length, IW_ESSID_MAX_SIZE);
-	if (!extra[length - 1])
-		length--;
 
 	priv->config |= CFG_STATIC_ESSID;
 
@@ -8953,7 +8951,7 @@ static int ipw_wx_get_nick(struct net_device *dev,
 	struct ipw_priv *priv = ieee80211_priv(dev);
 	IPW_DEBUG_WX("Getting nick\n");
 	mutex_lock(&priv->mutex);
-	wrqu->data.length = strlen(priv->nick) + 1;
+	wrqu->data.length = strlen(priv->nick);
 	memcpy(extra, priv->nick, wrqu->data.length);
 	wrqu->data.flags = 1;	/* active */
 	mutex_unlock(&priv->mutex);
@@ -9276,9 +9274,9 @@ static int ipw_wx_set_retry(struct net_device *dev,
 		return -EINVAL;
 
 	mutex_lock(&priv->mutex);
-	if (wrqu->retry.flags & IW_RETRY_MIN)
+	if (wrqu->retry.flags & IW_RETRY_SHORT)
 		priv->short_retry_limit = (u8) wrqu->retry.value;
-	else if (wrqu->retry.flags & IW_RETRY_MAX)
+	else if (wrqu->retry.flags & IW_RETRY_LONG)
 		priv->long_retry_limit = (u8) wrqu->retry.value;
 	else {
 		priv->short_retry_limit = (u8) wrqu->retry.value;
@@ -9307,11 +9305,11 @@ static int ipw_wx_get_retry(struct net_device *dev,
 		return -EINVAL;
 	}
 
-	if (wrqu->retry.flags & IW_RETRY_MAX) {
-		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MAX;
+	if (wrqu->retry.flags & IW_RETRY_LONG) {
+		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_LONG;
 		wrqu->retry.value = priv->long_retry_limit;
-	} else if (wrqu->retry.flags & IW_RETRY_MIN) {
-		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_MIN;
+	} else if (wrqu->retry.flags & IW_RETRY_SHORT) {
+		wrqu->retry.flags = IW_RETRY_LIMIT | IW_RETRY_SHORT;
 		wrqu->retry.value = priv->short_retry_limit;
 	} else {
 		wrqu->retry.flags = IW_RETRY_LIMIT;
-- 
cgit v1.2.1


From a271ca5bbb1df988806bead8910b603819f4190f Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Mon, 11 Sep 2006 21:50:56 -0500
Subject: [PATCH] bcm43xx-softmac: update PHY initialization

This patch updates the PHY initialization code for bcm43xx-softmac
to conform with recent changes in the clean-room specs at
http://bcm-specs.sipsolutions.net. Mostly, these changes implement
the sequence needed for chips with GPHY revision 8; however, the
patch also corrects a typo in one address, and some parts that were
missing from the spec when the initial coding was done.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/bcm43xx/bcm43xx_phy.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_phy.c b/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
index eafd0f662686..52ce2a9334fb 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
@@ -361,7 +361,7 @@ static void bcm43xx_phy_setupg(struct bcm43xx_private *bcm)
 	if (phy->rev <= 2)
 		for (i = 0; i < BCM43xx_ILT_NOISESCALEG_SIZE; i++)
 			bcm43xx_ilt_write(bcm, 0x1400 + i, bcm43xx_ilt_noisescaleg1[i]);
-	else if ((phy->rev == 7) && (bcm43xx_phy_read(bcm, 0x0449) & 0x0200))
+	else if ((phy->rev >= 7) && (bcm43xx_phy_read(bcm, 0x0449) & 0x0200))
 		for (i = 0; i < BCM43xx_ILT_NOISESCALEG_SIZE; i++)
 			bcm43xx_ilt_write(bcm, 0x1400 + i, bcm43xx_ilt_noisescaleg3[i]);
 	else
@@ -371,7 +371,7 @@ static void bcm43xx_phy_setupg(struct bcm43xx_private *bcm)
 	if (phy->rev == 2)
 		for (i = 0; i < BCM43xx_ILT_SIGMASQR_SIZE; i++)
 			bcm43xx_ilt_write(bcm, 0x5000 + i, bcm43xx_ilt_sigmasqr1[i]);
-	else if ((phy->rev > 2) && (phy->rev <= 7))
+	else if ((phy->rev > 2) && (phy->rev <= 8))
 		for (i = 0; i < BCM43xx_ILT_SIGMASQR_SIZE; i++)
 			bcm43xx_ilt_write(bcm, 0x5000 + i, bcm43xx_ilt_sigmasqr2[i]);
 	
@@ -1197,7 +1197,7 @@ static void bcm43xx_phy_initg(struct bcm43xx_private *bcm)
 
 	if (phy->rev == 1)
 		bcm43xx_phy_initb5(bcm);
-	else if (phy->rev >= 2 && phy->rev <= 7)
+	else
 		bcm43xx_phy_initb6(bcm);
 	if (phy->rev >= 2 || phy->connected)
 		bcm43xx_phy_inita(bcm);
@@ -1241,23 +1241,22 @@ static void bcm43xx_phy_initg(struct bcm43xx_private *bcm)
 		bcm43xx_phy_lo_g_measure(bcm);
 	} else {
 		if (radio->version == 0x2050 && radio->revision == 8) {
-			//FIXME
+			bcm43xx_radio_write16(bcm, 0x0052,
+					      (radio->txctl1 << 4) | radio->txctl2);
 		} else {
 			bcm43xx_radio_write16(bcm, 0x0052,
 					      (bcm43xx_radio_read16(bcm, 0x0052)
 					       & 0xFFF0) | radio->txctl1);
 		}
 		if (phy->rev >= 6) {
-			/*
 			bcm43xx_phy_write(bcm, 0x0036,
 					  (bcm43xx_phy_read(bcm, 0x0036)
-					   & 0xF000) | (FIXME << 12));
-			*/
+					   & 0xF000) | (radio->txctl2 << 12));
 		}
 		if (bcm->sprom.boardflags & BCM43xx_BFL_PACTRL)
 			bcm43xx_phy_write(bcm, 0x002E, 0x8075);
 		else
-			bcm43xx_phy_write(bcm, 0x003E, 0x807F);
+			bcm43xx_phy_write(bcm, 0x002E, 0x807F);
 		if (phy->rev < 2)
 			bcm43xx_phy_write(bcm, 0x002F, 0x0101);
 		else
-- 
cgit v1.2.1


From b2f206f4a711b0b1e6d3cdca52e919a986aceada Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Tue, 12 Sep 2006 12:40:32 -0500
Subject: [PATCH] bcm43xx-softmac: update noise handling

In bcm43xx-softmac, the bcm43xx_stats struct contains a variable that
is no longer used. In addition, two TODO entries related to noise
processing in bcm43xx_rx have been completed, and as unused one
is removed.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/bcm43xx/bcm43xx.h      | 1 -
 drivers/net/wireless/bcm43xx/bcm43xx_xmit.c | 5 ++---
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/wireless/bcm43xx/bcm43xx.h b/drivers/net/wireless/bcm43xx/bcm43xx.h
index 6d4ea36bc564..d6a8bf09878e 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx.h
+++ b/drivers/net/wireless/bcm43xx/bcm43xx.h
@@ -666,7 +666,6 @@ struct bcm43xx_noise_calculation {
 };
 
 struct bcm43xx_stats {
-	u8 link_quality;
 	u8 noise;
 	struct iw_statistics wstats;
 	/* Store the last TX/RX times here for updating the leds. */
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
index c0efbfe605a5..0159e4e93201 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_xmit.c
@@ -496,15 +496,14 @@ int bcm43xx_rx(struct bcm43xx_private *bcm,
 	stats.signal = bcm43xx_rssi_postprocess(bcm, rxhdr->rssi, is_ofdm,
 					      !!(rxflags1 & BCM43xx_RXHDR_FLAGS1_2053RSSIADJ),
 					      !!(rxflags3 & BCM43xx_RXHDR_FLAGS3_2050RSSIADJ));
-//TODO	stats.noise = 
+	stats.noise = bcm->stats.noise;
 	if (is_ofdm)
 		stats.rate = bcm43xx_plcp_get_bitrate_ofdm(plcp);
 	else
 		stats.rate = bcm43xx_plcp_get_bitrate_cck(plcp);
 	stats.received_channel = radio->channel;
-//TODO	stats.control = 
 	stats.mask = IEEE80211_STATMASK_SIGNAL |
-//TODO		     IEEE80211_STATMASK_NOISE |
+		     IEEE80211_STATMASK_NOISE |
 		     IEEE80211_STATMASK_RATE |
 		     IEEE80211_STATMASK_RSSI;
 	if (phy->type == BCM43xx_PHYTYPE_A)
-- 
cgit v1.2.1


From 8aeb9fc5699f1b3c64ba152910f7c5c075c3a93a Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Tue, 12 Sep 2006 15:29:04 -0500
Subject: [PATCH] bcm43xx-softmac: improve wrong firmware message

An error message is changed to a printk as the original dprintk
would be optimized away if debugging were not enabled. If the error
is triggered, a more meaningful message is returned.

Signed-off-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/bcm43xx/bcm43xx_main.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
index cb9a3ae8463a..b02e50d2ca32 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
@@ -2405,9 +2405,10 @@ static int bcm43xx_chip_init(struct bcm43xx_private *bcm)
 				   BCM43xx_UCODE_TIME) & 0x1f);
 
 	if ( value16 > 0x128 ) {
-		dprintk(KERN_ERR PFX
-			"Firmware: no support for microcode rev > 0x128\n");
-		err = -1;
+		printk(KERN_ERR PFX
+			"Firmware: no support for microcode extracted "
+			"from version 4.x binary drivers.\n");
+		err = -EOPNOTSUPP;
 		goto err_release_fw;
 	}
 
-- 
cgit v1.2.1


From 927db87fb1523101653eae44ac27223a30a18ada Mon Sep 17 00:00:00 2001
From: Daniel Drake <dsd@gentoo.org>
Date: Tue, 12 Sep 2006 23:37:25 +0100
Subject: [PATCH] softmac: Update MAINTAINERS entry

This mailing list has been deactivated.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 MAINTAINERS | 1 -
 1 file changed, 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 16af51ba44a5..bd446e251d5b 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2655,7 +2655,6 @@ M:	josejx@gentoo.org
 P:	Daniel Drake
 M:	dsd@gentoo.org
 W:	http://softmac.sipsolutions.net/
-L:	softmac-dev@sipsolutions.net
 L:	netdev@vger.kernel.org
 S:	Maintained
 
-- 
cgit v1.2.1


From cbb5e6bbb29a850dcb87d8efa30b457ce8014369 Mon Sep 17 00:00:00 2001
From: Ulrich Kunitz <kune@deine-taler.de>
Date: Wed, 13 Sep 2006 02:41:02 +0100
Subject: [PATCH] zd1211rw: 16-bit writes for physical control registers

Caused by the fact that physical control registers appear to have
only a width of 16 bit, 32-bit writes are not required.

Signed-off-by: Ulrich Kunitz <kune@deine-taler.de>
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_chip.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/zd1211rw/zd_chip.c b/drivers/net/wireless/zd1211rw/zd_chip.c
index 7c4e32cf0d47..8d3843d74098 100644
--- a/drivers/net/wireless/zd1211rw/zd_chip.c
+++ b/drivers/net/wireless/zd1211rw/zd_chip.c
@@ -1181,7 +1181,7 @@ static int update_pwr_int(struct zd_chip *chip, u8 channel)
 	u8 value = chip->pwr_int_values[channel - 1];
 	dev_dbg_f(zd_chip_dev(chip), "channel %d pwr_int %#04x\n",
 		 channel, value);
-	return zd_iowrite32_locked(chip, value, CR31);
+	return zd_iowrite16_locked(chip, value, CR31);
 }
 
 static int update_pwr_cal(struct zd_chip *chip, u8 channel)
@@ -1189,12 +1189,12 @@ static int update_pwr_cal(struct zd_chip *chip, u8 channel)
 	u8 value = chip->pwr_cal_values[channel-1];
 	dev_dbg_f(zd_chip_dev(chip), "channel %d pwr_cal %#04x\n",
 		 channel, value);
-	return zd_iowrite32_locked(chip, value, CR68);
+	return zd_iowrite16_locked(chip, value, CR68);
 }
 
 static int update_ofdm_cal(struct zd_chip *chip, u8 channel)
 {
-	struct zd_ioreq32 ioreqs[3];
+	struct zd_ioreq16 ioreqs[3];
 
 	ioreqs[0].addr = CR67;
 	ioreqs[0].value = chip->ofdm_cal_values[OFDM_36M_INDEX][channel-1];
@@ -1206,7 +1206,7 @@ static int update_ofdm_cal(struct zd_chip *chip, u8 channel)
 	dev_dbg_f(zd_chip_dev(chip),
 		"channel %d ofdm_cal 36M %#04x 48M %#04x 54M %#04x\n",
 		channel, ioreqs[0].value, ioreqs[1].value, ioreqs[2].value);
-	return zd_iowrite32a_locked(chip, ioreqs, ARRAY_SIZE(ioreqs));
+	return zd_iowrite16a_locked(chip, ioreqs, ARRAY_SIZE(ioreqs));
 }
 
 static int update_channel_integration_and_calibration(struct zd_chip *chip,
@@ -1218,7 +1218,7 @@ static int update_channel_integration_and_calibration(struct zd_chip *chip,
 	if (r)
 		return r;
 	if (chip->is_zd1211b) {
-		static const struct zd_ioreq32 ioreqs[] = {
+		static const struct zd_ioreq16 ioreqs[] = {
 			{ CR69, 0x28 },
 			{},
 			{ CR69, 0x2a },
@@ -1230,7 +1230,7 @@ static int update_channel_integration_and_calibration(struct zd_chip *chip,
 		r = update_pwr_cal(chip, channel);
 		if (r)
 			return r;
-		r = zd_iowrite32a_locked(chip, ioreqs, ARRAY_SIZE(ioreqs));
+		r = zd_iowrite16a_locked(chip, ioreqs, ARRAY_SIZE(ioreqs));
 		if (r)
 			return r;
 	}
@@ -1252,7 +1252,7 @@ static int patch_cck_gain(struct zd_chip *chip)
 	if (r)
 		return r;
 	dev_dbg_f(zd_chip_dev(chip), "patching value %x\n", value & 0xff);
-	return zd_iowrite32_locked(chip, value & 0xff, CR47);
+	return zd_iowrite16_locked(chip, value & 0xff, CR47);
 }
 
 int zd_chip_set_channel(struct zd_chip *chip, u8 channel)
-- 
cgit v1.2.1


From 44976c66bb2b44e4cf29a259a0cd6adadf8eb031 Mon Sep 17 00:00:00 2001
From: Ulrich Kunitz <kune@deine-taler.de>
Date: Wed, 13 Sep 2006 02:41:35 +0100
Subject: [PATCH] zd1211rw: Removes wrong assertions

Checking whether a mutex is not locked directly before
mutex_lock() is called, doesn't make sense. The whole point of
mutex_lock() is to block, if the mutex is locked.

Signed-off-by: Ulrich Kunitz <kune@deine-taler.de>
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_chip.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/net/wireless/zd1211rw/zd_chip.c b/drivers/net/wireless/zd1211rw/zd_chip.c
index 8d3843d74098..c213a2e56e55 100644
--- a/drivers/net/wireless/zd1211rw/zd_chip.c
+++ b/drivers/net/wireless/zd1211rw/zd_chip.c
@@ -249,7 +249,6 @@ int zd_ioread16(struct zd_chip *chip, zd_addr_t addr, u16 *value)
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_ioread16_locked(chip, value, addr);
 	mutex_unlock(&chip->mutex);
@@ -260,7 +259,6 @@ int zd_ioread32(struct zd_chip *chip, zd_addr_t addr, u32 *value)
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_ioread32_locked(chip, value, addr);
 	mutex_unlock(&chip->mutex);
@@ -271,7 +269,6 @@ int zd_iowrite16(struct zd_chip *chip, zd_addr_t addr, u16 value)
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_iowrite16_locked(chip, value, addr);
 	mutex_unlock(&chip->mutex);
@@ -282,7 +279,6 @@ int zd_iowrite32(struct zd_chip *chip, zd_addr_t addr, u32 value)
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_iowrite32_locked(chip, value, addr);
 	mutex_unlock(&chip->mutex);
@@ -294,7 +290,6 @@ int zd_ioread32v(struct zd_chip *chip, const zd_addr_t *addresses,
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_ioread32v_locked(chip, values, addresses, count);
 	mutex_unlock(&chip->mutex);
@@ -306,7 +301,6 @@ int zd_iowrite32a(struct zd_chip *chip, const struct zd_ioreq32 *ioreqs,
 {
 	int r;
 
-	ZD_ASSERT(!mutex_is_locked(&chip->mutex));
 	mutex_lock(&chip->mutex);
 	r = zd_iowrite32a_locked(chip, ioreqs, count);
 	mutex_unlock(&chip->mutex);
-- 
cgit v1.2.1


From bc5f06a8aaa29a79c9da2cedb5b9779b8081289c Mon Sep 17 00:00:00 2001
From: Ulrich Kunitz <kune@deine-taler.de>
Date: Wed, 13 Sep 2006 02:42:12 +0100
Subject: [PATCH] zd1211rw: Added workqueue

For housekeeping and watchdog tasks a workqueue is created. The
central workqueue is not used to prevent crashes creates by bugs.
It might be changed, when the housekeeping is stabilized.

Signed-off-by: Ulrich Kunitz <kune@deine-taler.de>
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_usb.c | 10 ++++++++++
 drivers/net/wireless/zd1211rw/zd_usb.h |  2 ++
 2 files changed, 12 insertions(+)

diff --git a/drivers/net/wireless/zd1211rw/zd_usb.c b/drivers/net/wireless/zd1211rw/zd_usb.c
index 31027e52b04b..5c265ad0485a 100644
--- a/drivers/net/wireless/zd1211rw/zd_usb.c
+++ b/drivers/net/wireless/zd1211rw/zd_usb.c
@@ -24,6 +24,7 @@
 #include <linux/errno.h>
 #include <linux/skbuff.h>
 #include <linux/usb.h>
+#include <linux/workqueue.h>
 #include <net/ieee80211.h>
 
 #include "zd_def.h"
@@ -1112,12 +1113,20 @@ static struct usb_driver driver = {
 	.disconnect	= disconnect,
 };
 
+struct workqueue_struct *zd_workqueue;
+
 static int __init usb_init(void)
 {
 	int r;
 
 	pr_debug("usb_init()\n");
 
+	zd_workqueue = create_singlethread_workqueue(driver.name);
+	if (zd_workqueue == NULL) {
+		printk(KERN_ERR "%s: couldn't create workqueue\n", driver.name);
+		return -ENOMEM;
+	}
+
 	r = usb_register(&driver);
 	if (r) {
 		printk(KERN_ERR "usb_register() failed. Error number %d\n", r);
@@ -1132,6 +1141,7 @@ static void __exit usb_exit(void)
 {
 	pr_debug("usb_exit()\n");
 	usb_deregister(&driver);
+	destroy_workqueue(zd_workqueue);
 }
 
 module_init(usb_init);
diff --git a/drivers/net/wireless/zd1211rw/zd_usb.h b/drivers/net/wireless/zd1211rw/zd_usb.h
index ded39de5f72d..e81a2d3cfffd 100644
--- a/drivers/net/wireless/zd1211rw/zd_usb.h
+++ b/drivers/net/wireless/zd1211rw/zd_usb.h
@@ -238,4 +238,6 @@ int zd_usb_iowrite16v(struct zd_usb *usb, const struct zd_ioreq16 *ioreqs,
 
 int zd_usb_rfwrite(struct zd_usb *usb, u32 value, u8 bits);
 
+extern struct workqueue_struct *zd_workqueue;
+
 #endif /* _ZD_USB_H */
-- 
cgit v1.2.1


From 583afd1e4f25c87000c85ad7d03f5299fd4155dc Mon Sep 17 00:00:00 2001
From: Ulrich Kunitz <kune@deine-taler.de>
Date: Wed, 13 Sep 2006 02:42:38 +0100
Subject: [PATCH] zd1211rw: Add LED support

This patch includes a big cleanup of the existing unused LED code,
and adds support for controlling the LED.

The link LED will blink if the device is not associated. The LED
switches between 2 seconds on and 1 second off. If the device is
associated the LED is switched on.

The link LED also indicates packet TX. I do a little bit more led
resetting than the vendor driver, but the device works now as
expected for single LED and double LED devices.

Signed-off-by: Ulrich Kunitz <kune@deine-taler.de>
Signed-off-by: Daniel Drake <dsd@gentoo.org>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/zd1211rw/zd_chip.c | 117 +++++++++++++-------------------
 drivers/net/wireless/zd1211rw/zd_chip.h |  23 ++++---
 drivers/net/wireless/zd1211rw/zd_mac.c  |  50 ++++++++++++++
 drivers/net/wireless/zd1211rw/zd_mac.h  |   5 ++
 4 files changed, 117 insertions(+), 78 deletions(-)

diff --git a/drivers/net/wireless/zd1211rw/zd_chip.c b/drivers/net/wireless/zd1211rw/zd_chip.c
index c213a2e56e55..aa661b2b76c7 100644
--- a/drivers/net/wireless/zd1211rw/zd_chip.c
+++ b/drivers/net/wireless/zd1211rw/zd_chip.c
@@ -325,13 +325,22 @@ static int read_pod(struct zd_chip *chip, u8 *rf_type)
 	chip->patch_cr157 = (value >> 13) & 0x1;
 	chip->patch_6m_band_edge = (value >> 21) & 0x1;
 	chip->new_phy_layout = (value >> 31) & 0x1;
+	chip->link_led = ((value >> 4) & 1) ? LED1 : LED2;
+	chip->supports_tx_led = 1;
+	if (value & (1 << 24)) { /* LED scenario */
+		if (value & (1 << 29))
+			chip->supports_tx_led = 0;
+	}
 
 	dev_dbg_f(zd_chip_dev(chip),
 		"RF %s %#01x PA type %#01x patch CCK %d patch CR157 %d "
-		"patch 6M %d new PHY %d\n",
+		"patch 6M %d new PHY %d link LED%d tx led %d\n",
 		zd_rf_name(*rf_type), *rf_type,
 		chip->pa_type, chip->patch_cck_gain,
-		chip->patch_cr157, chip->patch_6m_band_edge, chip->new_phy_layout);
+		chip->patch_cr157, chip->patch_6m_band_edge,
+		chip->new_phy_layout,
+		chip->link_led == LED1 ? 1 : 2,
+		chip->supports_tx_led);
 	return 0;
 error:
 	*rf_type = 0;
@@ -1289,89 +1298,60 @@ u8 zd_chip_get_channel(struct zd_chip *chip)
 	return channel;
 }
 
-static u16 led_mask(int led)
-{
-	switch (led) {
-	case 1:
-		return LED1;
-	case 2:
-		return LED2;
-	default:
-		return 0;
-	}
-}
-
-static int read_led_reg(struct zd_chip *chip, u16 *status)
-{
-	ZD_ASSERT(mutex_is_locked(&chip->mutex));
-	return zd_ioread16_locked(chip, status, CR_LED);
-}
-
-static int write_led_reg(struct zd_chip *chip, u16 status)
+int zd_chip_control_leds(struct zd_chip *chip, enum led_status status)
 {
-	ZD_ASSERT(mutex_is_locked(&chip->mutex));
-	return zd_iowrite16_locked(chip, status, CR_LED);
-}
+	static const zd_addr_t a[] = {
+		FW_LINK_STATUS,
+		CR_LED,
+	};
 
-int zd_chip_led_status(struct zd_chip *chip, int led, enum led_status status)
-{
-	int r, ret;
-	u16 mask = led_mask(led);
-	u16 reg;
+	int r;
+	u16 v[ARRAY_SIZE(a)];
+	struct zd_ioreq16 ioreqs[ARRAY_SIZE(a)] = {
+		[0] = { FW_LINK_STATUS },
+		[1] = { CR_LED },
+	};
+	u16 other_led;
 
-	if (!mask)
-		return -EINVAL;
 	mutex_lock(&chip->mutex);
-	r = read_led_reg(chip, &reg);
+	r = zd_ioread16v_locked(chip, v, (const zd_addr_t *)a, ARRAY_SIZE(a));
 	if (r)
-		return r;
+		goto out;
+
+	other_led = chip->link_led == LED1 ? LED2 : LED1;
+
 	switch (status) {
-	case LED_STATUS:
-		return (reg & mask) ? LED_ON : LED_OFF;
 	case LED_OFF:
-		reg &= ~mask;
-		ret = LED_OFF;
+		ioreqs[0].value = FW_LINK_OFF;
+		ioreqs[1].value = v[1] & ~(LED1|LED2);
 		break;
-	case LED_FLIP:
-		reg ^= mask;
-		ret = (reg&mask) ? LED_ON : LED_OFF;
+	case LED_SCANNING:
+		ioreqs[0].value = FW_LINK_OFF;
+		ioreqs[1].value = v[1] & ~other_led;
+		if (get_seconds() % 3 == 0) {
+			ioreqs[1].value &= ~chip->link_led;
+		} else {
+			ioreqs[1].value |= chip->link_led;
+		}
 		break;
-	case LED_ON:
-		reg |= mask;
-		ret = LED_ON;
+	case LED_ASSOCIATED:
+		ioreqs[0].value = FW_LINK_TX;
+		ioreqs[1].value = v[1] & ~other_led;
+		ioreqs[1].value |= chip->link_led;
 		break;
 	default:
-		return -EINVAL;
-	}
-	r = write_led_reg(chip, reg);
-	if (r) {
-		ret = r;
+		r = -EINVAL;
 		goto out;
 	}
-out:
-	mutex_unlock(&chip->mutex);
-	return r;
-}
 
-int zd_chip_led_flip(struct zd_chip *chip, int led,
-	const unsigned int *phases_msecs, unsigned int count)
-{
-	int i, r;
-	enum led_status status;
-
-	r = zd_chip_led_status(chip, led, LED_STATUS);
-	if (r)
-		return r;
-	status = r;
-	for (i = 0; i < count; i++) {
-		r = zd_chip_led_status(chip, led, LED_FLIP);
-		if (r < 0)
+	if (v[0] != ioreqs[0].value || v[1] != ioreqs[1].value) {
+		r = zd_iowrite16a_locked(chip, ioreqs, ARRAY_SIZE(ioreqs));
+		if (r)
 			goto out;
-		msleep(phases_msecs[i]);
 	}
-
+	r = 0;
 out:
-	zd_chip_led_status(chip, led, status);
+	mutex_unlock(&chip->mutex);
 	return r;
 }
 
@@ -1673,4 +1653,3 @@ int zd_rfwritev_cr_locked(struct zd_chip *chip,
 
 	return 0;
 }
-
diff --git a/drivers/net/wireless/zd1211rw/zd_chip.h b/drivers/net/wireless/zd1211rw/zd_chip.h
index 4b1250859897..ae59597ce4e1 100644
--- a/drivers/net/wireless/zd1211rw/zd_chip.h
+++ b/drivers/net/wireless/zd1211rw/zd_chip.h
@@ -428,6 +428,7 @@
 /* masks for controlling LEDs */
 #define LED1				0x0100
 #define LED2				0x0200
+#define LED_SW				0x0400
 
 /* Seems to indicate that the configuration is over.
  */
@@ -629,6 +630,10 @@
 #define FW_SOFT_RESET           FW_REG(4)
 #define FW_FLASH_CHK            FW_REG(5)
 
+#define FW_LINK_OFF		0x0
+#define FW_LINK_TX		0x1
+/* 0x2 - link led on? */
+
 enum {
 	CR_BASE_OFFSET			= 0x9000,
 	FW_START_OFFSET			= 0xee00,
@@ -663,8 +668,11 @@ struct zd_chip {
 	u8 pwr_int_values[E2P_CHANNEL_COUNT];
 	/* SetPointOFDM in the vendor driver */
 	u8 ofdm_cal_values[3][E2P_CHANNEL_COUNT];
-	u8 pa_type:4, patch_cck_gain:1, patch_cr157:1, patch_6m_band_edge:1,
-	   new_phy_layout:1, is_zd1211b:1;
+	u16 link_led;
+	unsigned int pa_type:4,
+		patch_cck_gain:1, patch_cr157:1, patch_6m_band_edge:1,
+		new_phy_layout:1,
+		is_zd1211b:1, supports_tx_led:1;
 };
 
 static inline struct zd_chip *zd_usb_to_chip(struct zd_usb *usb)
@@ -812,15 +820,12 @@ int zd_chip_lock_phy_regs(struct zd_chip *chip);
 int zd_chip_unlock_phy_regs(struct zd_chip *chip);
 
 enum led_status {
-	LED_OFF	   = 0,
-	LED_ON     = 1,
-	LED_FLIP   = 2,
-	LED_STATUS = 3,
+	LED_OFF = 0,
+	LED_SCANNING = 1,
+	LED_ASSOCIATED = 2,
 };
 
-int zd_chip_led_status(struct zd_chip *chip, int led, enum led_status status);
-int zd_chip_led_flip(struct zd_chip *chip, int led,
-	             const unsigned int *phases_msecs, unsigned int count);
+int zd_chip_control_leds(struct zd_chip *chip, enum led_status status);
 
 int zd_set_beacon_interval(struct zd_chip *chip, u32 interval);
 
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.c b/drivers/net/wireless/zd1211rw/zd_mac.c
index 1989f1c05fbe..2d12837052b0 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.c
+++ b/drivers/net/wireless/zd1211rw/zd_mac.c
@@ -33,6 +33,10 @@
 static void ieee_init(struct ieee80211_device *ieee);
 static void softmac_init(struct ieee80211softmac_device *sm);
 
+static void housekeeping_init(struct zd_mac *mac);
+static void housekeeping_enable(struct zd_mac *mac);
+static void housekeeping_disable(struct zd_mac *mac);
+
 int zd_mac_init(struct zd_mac *mac,
 	        struct net_device *netdev,
 	        struct usb_interface *intf)
@@ -46,6 +50,7 @@ int zd_mac_init(struct zd_mac *mac,
 	ieee_init(ieee);
 	softmac_init(ieee80211_priv(netdev));
 	zd_chip_init(&mac->chip, netdev, intf);
+	housekeeping_init(mac);
 	return 0;
 }
 
@@ -178,6 +183,7 @@ int zd_mac_open(struct net_device *netdev)
 	if (r < 0)
 		goto disable_rx;
 
+	housekeeping_enable(mac);
 	ieee80211softmac_start(netdev);
 	return 0;
 disable_rx:
@@ -204,6 +210,7 @@ int zd_mac_stop(struct net_device *netdev)
 	 */
 
 	zd_chip_disable_rx(chip);
+	housekeeping_disable(mac);
 	ieee80211softmac_stop(netdev);
 
 	zd_chip_disable_hwint(chip);
@@ -1080,3 +1087,46 @@ void zd_dump_rx_status(const struct rx_status *status)
 	}
 }
 #endif /* DEBUG */
+
+#define LINK_LED_WORK_DELAY HZ
+
+static void link_led_handler(void *p)
+{
+	struct zd_mac *mac = p;
+	struct zd_chip *chip = &mac->chip;
+	struct ieee80211softmac_device *sm = ieee80211_priv(mac->netdev);
+	int is_associated;
+	int r;
+
+	spin_lock_irq(&mac->lock);
+	is_associated = sm->associated != 0;
+	spin_unlock_irq(&mac->lock);
+
+	r = zd_chip_control_leds(chip,
+		                 is_associated ? LED_ASSOCIATED : LED_SCANNING);
+	if (r)
+		dev_err(zd_mac_dev(mac), "zd_chip_control_leds error %d\n", r);
+
+	queue_delayed_work(zd_workqueue, &mac->housekeeping.link_led_work,
+		           LINK_LED_WORK_DELAY);
+}
+
+static void housekeeping_init(struct zd_mac *mac)
+{
+	INIT_WORK(&mac->housekeeping.link_led_work, link_led_handler, mac);
+}
+
+static void housekeeping_enable(struct zd_mac *mac)
+{
+	dev_dbg_f(zd_mac_dev(mac), "\n");
+	queue_delayed_work(zd_workqueue, &mac->housekeeping.link_led_work,
+			   0);
+}
+
+static void housekeeping_disable(struct zd_mac *mac)
+{
+	dev_dbg_f(zd_mac_dev(mac), "\n");
+	cancel_rearming_delayed_workqueue(zd_workqueue,
+		&mac->housekeeping.link_led_work);
+	zd_chip_control_leds(&mac->chip, LED_OFF);
+}
diff --git a/drivers/net/wireless/zd1211rw/zd_mac.h b/drivers/net/wireless/zd1211rw/zd_mac.h
index 29b51fd7d4e5..b8ea3de7924a 100644
--- a/drivers/net/wireless/zd1211rw/zd_mac.h
+++ b/drivers/net/wireless/zd1211rw/zd_mac.h
@@ -120,6 +120,10 @@ enum mac_flags {
 	MAC_FIXED_CHANNEL = 0x01,
 };
 
+struct housekeeping {
+	struct work_struct link_led_work;
+};
+
 #define ZD_MAC_STATS_BUFFER_SIZE 16
 
 struct zd_mac {
@@ -128,6 +132,7 @@ struct zd_mac {
 	struct net_device *netdev;
 	/* Unlocked reading possible */
 	struct iw_statistics iw_stats;
+	struct housekeeping housekeeping;
 	unsigned int stats_count;
 	u8 qual_buffer[ZD_MAC_STATS_BUFFER_SIZE];
 	u8 rssi_buffer[ZD_MAC_STATS_BUFFER_SIZE];
-- 
cgit v1.2.1


From a67ab2bde752b26be75d4b68ecead9a14692eac5 Mon Sep 17 00:00:00 2001
From: Larry Finger <Larry.Finger@lwfinger.net>
Date: Thu, 14 Sep 2006 08:28:54 -0500
Subject: [PATCH] bcm43xx: fix netdev watchdog timeouts

The setup for running long periodic work has a bug that leads to
netdev watchdog tx timeouts. This change eliminates the timeouts.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Acked-by: Larry Finger <Larry.Finger@lwfinger.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/bcm43xx/bcm43xx_main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_main.c b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
index b02e50d2ca32..eb65db7393ba 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_main.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_main.c
@@ -3170,8 +3170,7 @@ static void bcm43xx_periodic_work_handler(void *d)
 		 * be preemtible.
 		 */
 		mutex_lock(&bcm->mutex);
-		netif_stop_queue(bcm->net_dev);
-		synchronize_net();
+		netif_tx_disable(bcm->net_dev);
 		spin_lock_irqsave(&bcm->irq_lock, flags);
 		bcm43xx_mac_suspend(bcm);
 		if (bcm43xx_using_pio(bcm))
-- 
cgit v1.2.1


From 14a72f53fb1bb5d5c2bdd8cf172219519664729a Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:52:01 -0700
Subject: [NetLabel]: correct improper handling of non-NetLabel peer contexts

Fix a problem where NetLabel would always set the value of
sk_security_struct->peer_sid in selinux_netlbl_sock_graft() to the context of
the socket, causing problems when users would query the context of the
connection.  This patch fixes this so that the value in
sk_security_struct->peer_sid is only set when the connection is NetLabel based,
otherwise the value is untouched.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h       |  7 ++++++
 include/net/netlabel.h         |  8 +++++++
 net/ipv4/cipso_ipv4.c          | 48 +++++++++++++++++++++++++++++-------------
 net/netlabel/netlabel_kapi.c   | 23 ++++++++++++++++++++
 security/selinux/ss/services.c | 12 ++++++++++-
 5 files changed, 82 insertions(+), 16 deletions(-)

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 59406e0dc5b2..6718452a5cd0 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -205,6 +205,7 @@ void cipso_v4_error(struct sk_buff *skb, int error, u32 gateway);
 int cipso_v4_socket_setattr(const struct socket *sock,
 			    const struct cipso_v4_doi *doi_def,
 			    const struct netlbl_lsm_secattr *secattr);
+int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr);
 int cipso_v4_socket_getattr(const struct socket *sock,
 			    struct netlbl_lsm_secattr *secattr);
 int cipso_v4_skbuff_getattr(const struct sk_buff *skb,
@@ -225,6 +226,12 @@ static inline int cipso_v4_socket_setattr(const struct socket *sock,
 	return -ENOSYS;
 }
 
+static inline int cipso_v4_sock_getattr(struct sock *sk,
+					struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
 static inline int cipso_v4_socket_getattr(const struct socket *sock,
 					  struct netlbl_lsm_secattr *secattr)
 {
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index dd5780b36919..bf7b564e3540 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -238,6 +238,8 @@ static inline void netlbl_secattr_free(struct netlbl_lsm_secattr *secattr,
 #ifdef CONFIG_NETLABEL
 int netlbl_socket_setattr(const struct socket *sock,
 			  const struct netlbl_lsm_secattr *secattr);
+int netlbl_sock_getattr(struct sock *sk,
+			struct netlbl_lsm_secattr *secattr);
 int netlbl_socket_getattr(const struct socket *sock,
 			  struct netlbl_lsm_secattr *secattr);
 int netlbl_skbuff_getattr(const struct sk_buff *skb,
@@ -250,6 +252,12 @@ static inline int netlbl_socket_setattr(const struct socket *sock,
 	return -ENOSYS;
 }
 
+static inline int netlbl_sock_getattr(struct sock *sk,
+				      struct netlbl_lsm_secattr *secattr)
+{
+	return -ENOSYS;
+}
+
 static inline int netlbl_socket_getattr(const struct socket *sock,
 					struct netlbl_lsm_secattr *secattr)
 {
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 80a2a0911b49..a3bae2ca8acc 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -1486,43 +1486,40 @@ socket_setattr_failure:
 }
 
 /**
- * cipso_v4_socket_getattr - Get the security attributes from a socket
- * @sock: the socket
+ * cipso_v4_sock_getattr - Get the security attributes from a sock
+ * @sk: the sock
  * @secattr: the security attributes
  *
  * Description:
- * Query @sock to see if there is a CIPSO option attached to the socket and if
- * there is return the CIPSO security attributes in @secattr.  Returns zero on
- * success and negative values on failure.
+ * Query @sk to see if there is a CIPSO option attached to the sock and if
+ * there is return the CIPSO security attributes in @secattr.  This function
+ * requires that @sk be locked, or privately held, but it does not do any
+ * locking itself.  Returns zero on success and negative values on failure.
  *
  */
-int cipso_v4_socket_getattr(const struct socket *sock,
-			    struct netlbl_lsm_secattr *secattr)
+int cipso_v4_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
 {
 	int ret_val = -ENOMSG;
-	struct sock *sk;
 	struct inet_sock *sk_inet;
 	unsigned char *cipso_ptr;
 	u32 doi;
 	struct cipso_v4_doi *doi_def;
 
-	sk = sock->sk;
-	lock_sock(sk);
 	sk_inet = inet_sk(sk);
 	if (sk_inet->opt == NULL || sk_inet->opt->cipso == 0)
-		goto socket_getattr_return;
+		return -ENOMSG;
 	cipso_ptr = sk_inet->opt->__data + sk_inet->opt->cipso -
 		sizeof(struct iphdr);
 	ret_val = cipso_v4_cache_check(cipso_ptr, cipso_ptr[1], secattr);
 	if (ret_val == 0)
-		goto socket_getattr_return;
+		return ret_val;
 
 	doi = ntohl(*(u32 *)&cipso_ptr[2]);
 	rcu_read_lock();
 	doi_def = cipso_v4_doi_getdef(doi);
 	if (doi_def == NULL) {
 		rcu_read_unlock();
-		goto socket_getattr_return;
+		return -ENOMSG;
 	}
 	switch (cipso_ptr[6]) {
 	case CIPSO_V4_TAG_RBITMAP:
@@ -1533,8 +1530,29 @@ int cipso_v4_socket_getattr(const struct socket *sock,
 	}
 	rcu_read_unlock();
 
-socket_getattr_return:
-	release_sock(sk);
+	return ret_val;
+}
+
+/**
+ * cipso_v4_socket_getattr - Get the security attributes from a socket
+ * @sock: the socket
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Query @sock to see if there is a CIPSO option attached to the socket and if
+ * there is return the CIPSO security attributes in @secattr.  Returns zero on
+ * success and negative values on failure.
+ *
+ */
+int cipso_v4_socket_getattr(const struct socket *sock,
+			    struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+
+	lock_sock(sock->sk);
+	ret_val = cipso_v4_sock_getattr(sock->sk, secattr);
+	release_sock(sock->sk);
+
 	return ret_val;
 }
 
diff --git a/net/netlabel/netlabel_kapi.c b/net/netlabel/netlabel_kapi.c
index 0fd8aaafe23f..54fb7de3c2b1 100644
--- a/net/netlabel/netlabel_kapi.c
+++ b/net/netlabel/netlabel_kapi.c
@@ -84,6 +84,29 @@ socket_setattr_return:
 	return ret_val;
 }
 
+/**
+ * netlbl_sock_getattr - Determine the security attributes of a sock
+ * @sk: the sock
+ * @secattr: the security attributes
+ *
+ * Description:
+ * Examines the given sock to see any NetLabel style labeling has been
+ * applied to the sock, if so it parses the socket label and returns the
+ * security attributes in @secattr.  Returns zero on success, negative values
+ * on failure.
+ *
+ */
+int netlbl_sock_getattr(struct sock *sk, struct netlbl_lsm_secattr *secattr)
+{
+	int ret_val;
+
+	ret_val = cipso_v4_sock_getattr(sk, secattr);
+	if (ret_val == 0)
+		return 0;
+
+	return netlbl_unlabel_getattr(secattr);
+}
+
 /**
  * netlbl_socket_getattr - Determine the security attributes of a socket
  * @sock: the socket
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 7eb69a602d8f..d67f7e658529 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2502,14 +2502,24 @@ void selinux_netlbl_sock_graft(struct sock *sk, struct socket *sock)
 {
 	struct inode_security_struct *isec = SOCK_INODE(sock)->i_security;
 	struct sk_security_struct *sksec = sk->sk_security;
+	struct netlbl_lsm_secattr secattr;
+	u32 nlbl_peer_sid;
 
 	sksec->sclass = isec->sclass;
 
 	if (sk->sk_family != PF_INET)
 		return;
 
+	netlbl_secattr_init(&secattr);
+	if (netlbl_sock_getattr(sk, &secattr) == 0 &&
+	    selinux_netlbl_secattr_to_sid(NULL,
+					  &secattr,
+					  sksec->sid,
+					  &nlbl_peer_sid) == 0)
+		sksec->peer_sid = nlbl_peer_sid;
+	netlbl_secattr_destroy(&secattr, 0);
+
 	sksec->nlbl_state = NLBL_REQUIRE;
-	sksec->peer_sid = sksec->sid;
 
 	/* Try to set the NetLabel on the socket to save time later, if we fail
 	 * here we will pick up the pieces in later calls to
-- 
cgit v1.2.1


From 609c92feea5652809319bb77f19d24a44615687d Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:52:37 -0700
Subject: [NetLabel]: make the CIPSOv4 cache spinlocks bottom half safe

The CIPSOv4 cache traversal routines are triggered both the userspace events
(cache invalidation due to DOI removal or updated SELinux policy) and network
packet processing events.  As a result there is a problem with the existing
CIPSOv4 cache spinlocks as they are not bottom-half/softirq safe.  This patch
converts the CIPSOv4 cache spin_[un]lock() calls into spin_[un]lock_bh() calls
to address this problem.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/cipso_ipv4.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index a3bae2ca8acc..87e71563335d 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -259,7 +259,7 @@ void cipso_v4_cache_invalidate(void)
 	u32 iter;
 
 	for (iter = 0; iter < CIPSO_V4_CACHE_BUCKETS; iter++) {
-		spin_lock(&cipso_v4_cache[iter].lock);
+		spin_lock_bh(&cipso_v4_cache[iter].lock);
 		list_for_each_entry_safe(entry,
 					 tmp_entry,
 					 &cipso_v4_cache[iter].list, list) {
@@ -267,7 +267,7 @@ void cipso_v4_cache_invalidate(void)
 			cipso_v4_cache_entry_free(entry);
 		}
 		cipso_v4_cache[iter].size = 0;
-		spin_unlock(&cipso_v4_cache[iter].lock);
+		spin_unlock_bh(&cipso_v4_cache[iter].lock);
 	}
 
 	return;
@@ -309,7 +309,7 @@ static int cipso_v4_cache_check(const unsigned char *key,
 
 	hash = cipso_v4_map_cache_hash(key, key_len);
 	bkt = hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
-	spin_lock(&cipso_v4_cache[bkt].lock);
+	spin_lock_bh(&cipso_v4_cache[bkt].lock);
 	list_for_each_entry(entry, &cipso_v4_cache[bkt].list, list) {
 		if (entry->hash == hash &&
 		    entry->key_len == key_len &&
@@ -318,7 +318,7 @@ static int cipso_v4_cache_check(const unsigned char *key,
 			secattr->cache.free = entry->lsm_data.free;
 			secattr->cache.data = entry->lsm_data.data;
 			if (prev_entry == NULL) {
-				spin_unlock(&cipso_v4_cache[bkt].lock);
+				spin_unlock_bh(&cipso_v4_cache[bkt].lock);
 				return 0;
 			}
 
@@ -333,12 +333,12 @@ static int cipso_v4_cache_check(const unsigned char *key,
 					   &prev_entry->list);
 			}
 
-			spin_unlock(&cipso_v4_cache[bkt].lock);
+			spin_unlock_bh(&cipso_v4_cache[bkt].lock);
 			return 0;
 		}
 		prev_entry = entry;
 	}
-	spin_unlock(&cipso_v4_cache[bkt].lock);
+	spin_unlock_bh(&cipso_v4_cache[bkt].lock);
 
 	return -ENOENT;
 }
@@ -387,7 +387,7 @@ int cipso_v4_cache_add(const struct sk_buff *skb,
 	entry->lsm_data.data = secattr->cache.data;
 
 	bkt = entry->hash & (CIPSO_V4_CACHE_BUCKETBITS - 1);
-	spin_lock(&cipso_v4_cache[bkt].lock);
+	spin_lock_bh(&cipso_v4_cache[bkt].lock);
 	if (cipso_v4_cache[bkt].size < cipso_v4_cache_bucketsize) {
 		list_add(&entry->list, &cipso_v4_cache[bkt].list);
 		cipso_v4_cache[bkt].size += 1;
@@ -398,7 +398,7 @@ int cipso_v4_cache_add(const struct sk_buff *skb,
 		list_add(&entry->list, &cipso_v4_cache[bkt].list);
 		cipso_v4_cache_entry_free(old_entry);
 	}
-	spin_unlock(&cipso_v4_cache[bkt].lock);
+	spin_unlock_bh(&cipso_v4_cache[bkt].lock);
 
 	return 0;
 
-- 
cgit v1.2.1


From df2115c3134d0d1a18c1f37f5192394e7f64d1e0 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:53:13 -0700
Subject: [NetLabel]: change the SELinux permissions

Change NetLabel to use the 'recvfrom' socket permission and the
SECINITSID_NETMSG SELinux SID as the NetLabel base SID for incoming packets.
This patch effectively makes the old, and currently unused, SELinux NETMSG
permissions NetLabel permissions.

Signed-of-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 security/selinux/ss/services.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index d67f7e658529..22ed17c17718 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2611,7 +2611,7 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 	u32 netlbl_sid;
 	u32 recv_perm;
 
-	rc = selinux_netlbl_skbuff_getsid(skb, sksec->sid, &netlbl_sid);
+	rc = selinux_netlbl_skbuff_getsid(skb, SECINITSID_NETMSG, &netlbl_sid);
 	if (rc != 0)
 		return rc;
 
@@ -2620,13 +2620,13 @@ int selinux_netlbl_sock_rcv_skb(struct sk_security_struct *sksec,
 
 	switch (sksec->sclass) {
 	case SECCLASS_UDP_SOCKET:
-		recv_perm = UDP_SOCKET__RECV_MSG;
+		recv_perm = UDP_SOCKET__RECVFROM;
 		break;
 	case SECCLASS_TCP_SOCKET:
-		recv_perm = TCP_SOCKET__RECV_MSG;
+		recv_perm = TCP_SOCKET__RECVFROM;
 		break;
 	default:
-		recv_perm = RAWIP_SOCKET__RECV_MSG;
+		recv_perm = RAWIP_SOCKET__RECVFROM;
 	}
 
 	rc = avc_has_perm(sksec->sid,
-- 
cgit v1.2.1


From 22acb19a91d2b551ea37647747972e5286284b22 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:53:37 -0700
Subject: [NETLINK]: add nla_for_each_nested() to the interface list

At the top of include/net/netlink.h is a list of Netlink interfaces, however,
the nla_for_each_nested() macro was not listed.  This patch adds this interface
to the list at the top of the header file.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 11dc2e7f679a..2897a73f81eb 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -151,6 +151,7 @@
  *   nla_parse()			parse and validate stream of attrs
  *   nla_parse_nested()			parse nested attribuets
  *   nla_for_each_attr()		loop over all attributes
+ *   nla_for_each_nested()		loop over the nested attributes
  *=========================================================================
  */
 
-- 
cgit v1.2.1


From 4fe5d5c07ab615a52fd1b0ceba5aeed7c612821a Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:54:03 -0700
Subject: [Netlink]: add nla_validate_nested()

Add a new function, nla_validate_nested(), to validate nested Netlink
attributes.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/netlink.h | 19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

diff --git a/include/net/netlink.h b/include/net/netlink.h
index 2897a73f81eb..4ab68a7a636a 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -146,6 +146,7 @@
  *   nla_ok(nla, remaining)		does nla fit into remaining bytes?
  *   nla_next(nla, remaining)		get next netlink attribute
  *   nla_validate()			validate a stream of attributes
+ *   nla_validate_nested()		validate a stream of nested attributes
  *   nla_find()				find attribute in stream of attributes
  *   nla_find_nested()			find attribute in nested attributes
  *   nla_parse()			parse and validate stream of attrs
@@ -950,6 +951,24 @@ static inline int nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
 	return nlmsg_trim(skb, start);
 }
 
+/**
+ * nla_validate_nested - Validate a stream of nested attributes
+ * @start: container attribute
+ * @maxtype: maximum attribute type to be expected
+ * @policy: validation policy
+ *
+ * Validates all attributes in the nested attribute stream against the
+ * specified policy. Attributes with a type exceeding maxtype will be
+ * ignored. See documenation of struct nla_policy for more details.
+ *
+ * Returns 0 on success or a negative error code.
+ */
+static inline int nla_validate_nested(struct nlattr *start, int maxtype,
+				      struct nla_policy *policy)
+{
+	return nla_validate(nla_data(start), nla_len(start), maxtype, policy);
+}
+
 /**
  * nla_for_each_attr - iterate over a stream of attributes
  * @pos: loop counter, set to current attribute
-- 
cgit v1.2.1


From fcd48280643e92ec6cb29a04e9079dd7b6b5bfef Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:56:09 -0700
Subject: [NetLabel]: rework the Netlink attribute handling (part 1)

At the suggestion of Thomas Graf, rewrite NetLabel's use of Netlink attributes
to better follow the common Netlink attribute usage.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/cipso_ipv4.h           |  16 ++-
 include/net/netlabel.h             |  49 +--------
 net/ipv4/cipso_ipv4.c              | 203 +++++--------------------------------
 net/netlabel/netlabel_domainhash.c | 183 +++++++--------------------------
 net/netlabel/netlabel_domainhash.h |   6 +-
 net/netlabel/netlabel_user.c       |  82 ---------------
 net/netlabel/netlabel_user.h       | 141 +-------------------------
 7 files changed, 76 insertions(+), 604 deletions(-)

diff --git a/include/net/cipso_ipv4.h b/include/net/cipso_ipv4.h
index 6718452a5cd0..2d72496c2029 100644
--- a/include/net/cipso_ipv4.h
+++ b/include/net/cipso_ipv4.h
@@ -130,8 +130,9 @@ extern int cipso_v4_rbm_strictvalid;
 int cipso_v4_doi_add(struct cipso_v4_doi *doi_def);
 int cipso_v4_doi_remove(u32 doi, void (*callback) (struct rcu_head * head));
 struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi);
-struct sk_buff *cipso_v4_doi_dump_all(size_t headroom);
-struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom);
+int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+	             void *cb_arg);
 int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def, const char *domain);
 int cipso_v4_doi_domhsh_remove(struct cipso_v4_doi *doi_def,
 			       const char *domain);
@@ -152,14 +153,11 @@ static inline struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
 	return NULL;
 }
 
-static inline struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+static inline int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+		     void *cb_arg)
 {
-	return NULL;
-}
-
-static inline struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
-{
-	return NULL;
+	return 0;
 }
 
 static inline int cipso_v4_doi_domhsh_add(struct cipso_v4_doi *doi_def,
diff --git a/include/net/netlabel.h b/include/net/netlabel.h
index bf7b564e3540..6692430063fd 100644
--- a/include/net/netlabel.h
+++ b/include/net/netlabel.h
@@ -57,9 +57,8 @@
  * The payload is dependent on the subsystem specified in the
  * 'nlmsghdr->nlmsg_type' and should be defined below, supporting functions
  * should be defined in the corresponding net/netlabel/netlabel_<subsys>.h|c
- * file.  All of the fields in the NetLabel payload are NETLINK attributes, the
- * length of each field is the length of the NETLINK attribute payload, see
- * include/net/netlink.h for more information on NETLINK attributes.
+ * file.  All of the fields in the NetLabel payload are NETLINK attributes, see
+ * the include/net/netlink.h file for more information on NETLINK attributes.
  *
  */
 
@@ -82,50 +81,6 @@
 #define NETLBL_NLTYPE_UNLABELED         5
 #define NETLBL_NLTYPE_UNLABELED_NAME    "NLBL_UNLBL"
 
-/* NetLabel return codes */
-#define NETLBL_E_OK                     0
-
-/*
- * Helper functions
- */
-
-#define NETLBL_LEN_U8                   nla_total_size(sizeof(u8))
-#define NETLBL_LEN_U16                  nla_total_size(sizeof(u16))
-#define NETLBL_LEN_U32                  nla_total_size(sizeof(u32))
-
-/**
- * netlbl_netlink_alloc_skb - Allocate a NETLINK message buffer
- * @head: the amount of headroom in bytes
- * @body: the desired size (minus headroom) in bytes
- * @gfp_flags: the alloc flags to pass to alloc_skb()
- *
- * Description:
- * Allocate a NETLINK message buffer based on the sizes given in @head and
- * @body.  If @head is greater than zero skb_reserve() is called to reserve
- * @head bytes at the start of the buffer.  Returns a valid sk_buff pointer on
- * success, NULL on failure.
- *
- */
-static inline struct sk_buff *netlbl_netlink_alloc_skb(size_t head,
-						       size_t body,
-						       gfp_t gfp_flags)
-{
-	struct sk_buff *skb;
-
-	skb = alloc_skb(NLMSG_ALIGN(head + body), gfp_flags);
-	if (skb == NULL)
-		return NULL;
-	if (head > 0) {
-		skb_reserve(skb, head);
-		if (skb_tailroom(skb) < body) {
-			kfree_skb(skb);
-			return NULL;
-		}
-	}
-
-	return skb;
-}
-
 /*
  * NetLabel - Kernel API for accessing the network packet label mappings.
  *
diff --git a/net/ipv4/cipso_ipv4.c b/net/ipv4/cipso_ipv4.c
index 87e71563335d..e6ce0b3ba62a 100644
--- a/net/ipv4/cipso_ipv4.c
+++ b/net/ipv4/cipso_ipv4.c
@@ -530,197 +530,42 @@ struct cipso_v4_doi *cipso_v4_doi_getdef(u32 doi)
 }
 
 /**
- * cipso_v4_doi_dump_all - Dump all the CIPSO DOI definitions into a sk_buff
- * @headroom: the amount of headroom to allocate for the sk_buff
+ * cipso_v4_doi_walk - Iterate through the DOI definitions
+ * @skip_cnt: skip past this number of DOI definitions, updated
+ * @callback: callback for each DOI definition
+ * @cb_arg: argument for the callback function
  *
  * Description:
- * Dump a list of all the configured DOI values into a sk_buff.  The returned
- * sk_buff has room at the front of the sk_buff for @headroom bytes.  See
- * net/netlabel/netlabel_cipso_v4.h for the LISTALL message format.  This
- * function may fail if another process is changing the DOI list at the same
- * time.  Returns a pointer to a sk_buff on success, NULL on error.
+ * Iterate over the DOI definition list, skipping the first @skip_cnt entries.
+ * For each entry call @callback, if @callback returns a negative value stop
+ * 'walking' through the list and return.  Updates the value in @skip_cnt upon
+ * return.  Returns zero on success, negative values on failure.
  *
  */
-struct sk_buff *cipso_v4_doi_dump_all(size_t headroom)
+int cipso_v4_doi_walk(u32 *skip_cnt,
+		     int (*callback) (struct cipso_v4_doi *doi_def, void *arg),
+		     void *cb_arg)
 {
-	struct sk_buff *skb = NULL;
-	struct cipso_v4_doi *iter;
+	int ret_val = -ENOENT;
 	u32 doi_cnt = 0;
-	ssize_t buf_len;
-
-	buf_len = NETLBL_LEN_U32;
-	rcu_read_lock();
-	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
-		if (iter->valid) {
-			doi_cnt += 1;
-			buf_len += 2 * NETLBL_LEN_U32;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto doi_dump_all_failure;
-
-	if (nla_put_u32(skb, NLA_U32, doi_cnt) != 0)
-		goto doi_dump_all_failure;
-	buf_len -= NETLBL_LEN_U32;
-	list_for_each_entry_rcu(iter, &cipso_v4_doi_list, list)
-		if (iter->valid) {
-			if (buf_len < 2 * NETLBL_LEN_U32)
-				goto doi_dump_all_failure;
-			if (nla_put_u32(skb, NLA_U32, iter->doi) != 0)
-				goto doi_dump_all_failure;
-			if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
-				goto doi_dump_all_failure;
-			buf_len -= 2 * NETLBL_LEN_U32;
-		}
-	rcu_read_unlock();
-
-	return skb;
-
-doi_dump_all_failure:
-	rcu_read_unlock();
-	kfree(skb);
-	return NULL;
-}
-
-/**
- * cipso_v4_doi_dump - Dump a CIPSO DOI definition into a sk_buff
- * @doi: the DOI value
- * @headroom: the amount of headroom to allocate for the sk_buff
- *
- * Description:
- * Lookup the DOI definition matching @doi and dump it's contents into a
- * sk_buff.  The returned sk_buff has room at the front of the sk_buff for
- * @headroom bytes.  See net/netlabel/netlabel_cipso_v4.h for the LIST message
- * format.  This function may fail if another process is changing the DOI list
- * at the same time.  Returns a pointer to a sk_buff on success, NULL on error.
- *
- */
-struct sk_buff *cipso_v4_doi_dump(u32 doi, size_t headroom)
-{
-	struct sk_buff *skb = NULL;
-	struct cipso_v4_doi *iter;
-	u32 tag_cnt = 0;
-	u32 lvl_cnt = 0;
-	u32 cat_cnt = 0;
-	ssize_t buf_len;
-	ssize_t tmp;
+	struct cipso_v4_doi *iter_doi;
 
 	rcu_read_lock();
-	iter = cipso_v4_doi_getdef(doi);
-	if (iter == NULL)
-		goto doi_dump_failure;
-	buf_len = NETLBL_LEN_U32;
-	switch (iter->type) {
-	case CIPSO_V4_MAP_PASS:
-		buf_len += NETLBL_LEN_U32;
-		while(tag_cnt < CIPSO_V4_TAG_MAXCNT &&
-		      iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
-			tag_cnt += 1;
-			buf_len += NETLBL_LEN_U8;
-		}
-		break;
-	case CIPSO_V4_MAP_STD:
-		buf_len += 3 * NETLBL_LEN_U32;
-		while (tag_cnt < CIPSO_V4_TAG_MAXCNT &&
-		       iter->tags[tag_cnt] != CIPSO_V4_TAG_INVALID) {
-			tag_cnt += 1;
-			buf_len += NETLBL_LEN_U8;
-		}
-		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
-			if (iter->map.std->lvl.local[tmp] !=
-			    CIPSO_V4_INV_LVL) {
-				lvl_cnt += 1;
-				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U8;
+	list_for_each_entry_rcu(iter_doi, &cipso_v4_doi_list, list)
+		if (iter_doi->valid) {
+			if (doi_cnt++ < *skip_cnt)
+				continue;
+			ret_val = callback(iter_doi, cb_arg);
+			if (ret_val < 0) {
+				doi_cnt--;
+				goto doi_walk_return;
 			}
-		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
-			if (iter->map.std->cat.local[tmp] !=
-			    CIPSO_V4_INV_CAT) {
-				cat_cnt += 1;
-				buf_len += NETLBL_LEN_U32 + NETLBL_LEN_U16;
-			}
-		break;
-	}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto doi_dump_failure;
-
-	if (nla_put_u32(skb, NLA_U32, iter->type) != 0)
-		goto doi_dump_failure;
-	buf_len -= NETLBL_LEN_U32;
-	if (iter != cipso_v4_doi_getdef(doi))
-		goto doi_dump_failure;
-	switch (iter->type) {
-	case CIPSO_V4_MAP_PASS:
-		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
-			goto doi_dump_failure;
-		buf_len -= NETLBL_LEN_U32;
-		for (tmp = 0;
-		     tmp < CIPSO_V4_TAG_MAXCNT &&
-			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
-		     tmp++) {
-			if (buf_len < NETLBL_LEN_U8)
-				goto doi_dump_failure;
-			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
-				goto doi_dump_failure;
-			buf_len -= NETLBL_LEN_U8;
 		}
-		break;
-	case CIPSO_V4_MAP_STD:
-		if (nla_put_u32(skb, NLA_U32, tag_cnt) != 0)
-			goto doi_dump_failure;
-		if (nla_put_u32(skb, NLA_U32, lvl_cnt) != 0)
-			goto doi_dump_failure;
-		if (nla_put_u32(skb, NLA_U32, cat_cnt) != 0)
-			goto doi_dump_failure;
-		buf_len -= 3 * NETLBL_LEN_U32;
-		for (tmp = 0;
-		     tmp < CIPSO_V4_TAG_MAXCNT &&
-			     iter->tags[tmp] != CIPSO_V4_TAG_INVALID;
-		     tmp++) {
-			if (buf_len < NETLBL_LEN_U8)
-				goto doi_dump_failure;
-			if (nla_put_u8(skb, NLA_U8, iter->tags[tmp]) != 0)
-				goto doi_dump_failure;
-			buf_len -= NETLBL_LEN_U8;
-		}
-		for (tmp = 0; tmp < iter->map.std->lvl.local_size; tmp++)
-			if (iter->map.std->lvl.local[tmp] !=
-			    CIPSO_V4_INV_LVL) {
-				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U8)
-					goto doi_dump_failure;
-				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
-					goto doi_dump_failure;
-				if (nla_put_u8(skb,
-					   NLA_U8,
-					   iter->map.std->lvl.local[tmp]) != 0)
-					goto doi_dump_failure;
-				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U8;
-			}
-		for (tmp = 0; tmp < iter->map.std->cat.local_size; tmp++)
-			if (iter->map.std->cat.local[tmp] !=
-			    CIPSO_V4_INV_CAT) {
-				if (buf_len < NETLBL_LEN_U32 + NETLBL_LEN_U16)
-					goto doi_dump_failure;
-				if (nla_put_u32(skb, NLA_U32, tmp) != 0)
-					goto doi_dump_failure;
-				if (nla_put_u16(skb,
-					   NLA_U16,
-					   iter->map.std->cat.local[tmp]) != 0)
-					goto doi_dump_failure;
-				buf_len -= NETLBL_LEN_U32 + NETLBL_LEN_U16;
-			}
-		break;
-	}
-	rcu_read_unlock();
-
-	return skb;
 
-doi_dump_failure:
+doi_walk_return:
 	rcu_read_unlock();
-	kfree(skb);
-	return NULL;
+	*skip_cnt = doi_cnt;
+	return ret_val;
 }
 
 /**
diff --git a/net/netlabel/netlabel_domainhash.c b/net/netlabel/netlabel_domainhash.c
index 0489a1378101..f56d7a8ac7b7 100644
--- a/net/netlabel/netlabel_domainhash.c
+++ b/net/netlabel/netlabel_domainhash.c
@@ -354,160 +354,51 @@ struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain)
 }
 
 /**
- * netlbl_domhsh_dump - Dump the domain hash table into a sk_buff
+ * netlbl_domhsh_walk - Iterate through the domain mapping hash table
+ * @skip_bkt: the number of buckets to skip at the start
+ * @skip_chain: the number of entries to skip in the first iterated bucket
+ * @callback: callback for each entry
+ * @cb_arg: argument for the callback function
  *
  * Description:
- * Dump the domain hash table into a buffer suitable for returning to an
- * application in response to a NetLabel management DOMAIN message.  This
- * function may fail if another process is growing the hash table at the same
- * time.  The returned sk_buff has room at the front of the sk_buff for
- * @headroom bytes.  See netlabel.h for the DOMAIN message format.  Returns a
- * pointer to a sk_buff on success, NULL on error.
+ * Interate over the domain mapping hash table, skipping the first @skip_bkt
+ * buckets and @skip_chain entries.  For each entry in the table call
+ * @callback, if @callback returns a negative value stop 'walking' through the
+ * table and return.  Updates the values in @skip_bkt and @skip_chain on
+ * return.  Returns zero on succcess, negative values on failure.
  *
  */
-struct sk_buff *netlbl_domhsh_dump(size_t headroom)
+int netlbl_domhsh_walk(u32 *skip_bkt,
+		     u32 *skip_chain,
+		     int (*callback) (struct netlbl_dom_map *entry, void *arg),
+		     void *cb_arg)
 {
-	struct sk_buff *skb = NULL;
-	ssize_t buf_len;
-	u32 bkt_iter;
-	u32 dom_cnt = 0;
-	struct netlbl_domhsh_tbl *hsh_tbl;
-	struct netlbl_dom_map *list_iter;
-	ssize_t tmp_len;
+	int ret_val = -ENOENT;
+	u32 iter_bkt;
+	struct netlbl_dom_map *iter_entry;
+	u32 chain_cnt = 0;
 
-	buf_len = NETLBL_LEN_U32;
 	rcu_read_lock();
-	hsh_tbl = rcu_dereference(netlbl_domhsh);
-	for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
-		list_for_each_entry_rcu(list_iter,
-					&hsh_tbl->tbl[bkt_iter], list) {
-			buf_len += NETLBL_LEN_U32 +
-				nla_total_size(strlen(list_iter->domain) + 1);
-			switch (list_iter->type) {
-			case NETLBL_NLTYPE_UNLABELED:
-				break;
-			case NETLBL_NLTYPE_CIPSOV4:
-				buf_len += 2 * NETLBL_LEN_U32;
-				break;
-			}
-			dom_cnt++;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto dump_failure;
-
-	if (nla_put_u32(skb, NLA_U32, dom_cnt) != 0)
-		goto dump_failure;
-	buf_len -= NETLBL_LEN_U32;
-	hsh_tbl = rcu_dereference(netlbl_domhsh);
-	for (bkt_iter = 0; bkt_iter < hsh_tbl->size; bkt_iter++)
-		list_for_each_entry_rcu(list_iter,
-					&hsh_tbl->tbl[bkt_iter], list) {
-			tmp_len = nla_total_size(strlen(list_iter->domain) +
-						 1);
-			if (buf_len < NETLBL_LEN_U32 + tmp_len)
-				goto dump_failure;
-			if (nla_put_string(skb,
-					   NLA_STRING,
-					   list_iter->domain) != 0)
-				goto dump_failure;
-			if (nla_put_u32(skb, NLA_U32, list_iter->type) != 0)
-				goto dump_failure;
-			buf_len -= NETLBL_LEN_U32 + tmp_len;
-			switch (list_iter->type) {
-			case NETLBL_NLTYPE_UNLABELED:
-				break;
-			case NETLBL_NLTYPE_CIPSOV4:
-				if (buf_len < 2 * NETLBL_LEN_U32)
-					goto dump_failure;
-				if (nla_put_u32(skb,
-				       NLA_U32,
-				       list_iter->type_def.cipsov4->type) != 0)
-					goto dump_failure;
-				if (nla_put_u32(skb,
-				       NLA_U32,
-				       list_iter->type_def.cipsov4->doi) != 0)
-					goto dump_failure;
-				buf_len -= 2 * NETLBL_LEN_U32;
-				break;
+	for (iter_bkt = *skip_bkt;
+	     iter_bkt < rcu_dereference(netlbl_domhsh)->size;
+	     iter_bkt++, chain_cnt = 0) {
+		list_for_each_entry_rcu(iter_entry,
+					&netlbl_domhsh->tbl[iter_bkt],
+					list)
+			if (iter_entry->valid) {
+				if (chain_cnt++ < *skip_chain)
+					continue;
+				ret_val = callback(iter_entry, cb_arg);
+				if (ret_val < 0) {
+					chain_cnt--;
+					goto walk_return;
+				}
 			}
-		}
-	rcu_read_unlock();
-
-	return skb;
-
-dump_failure:
-	rcu_read_unlock();
-	kfree_skb(skb);
-	return NULL;
-}
-
-/**
- * netlbl_domhsh_dump_default - Dump the default domain mapping into a sk_buff
- *
- * Description:
- * Dump the default domain mapping into a buffer suitable for returning to an
- * application in response to a NetLabel management DEFDOMAIN message.  This
- * function may fail if another process is changing the default domain mapping
- * at the same time.  The returned sk_buff has room at the front of the
- * skb_buff for @headroom bytes.  See netlabel.h for the DEFDOMAIN message
- * format.  Returns a pointer to a sk_buff on success, NULL on error.
- *
- */
-struct sk_buff *netlbl_domhsh_dump_default(size_t headroom)
-{
-	struct sk_buff *skb;
-	ssize_t buf_len;
-	struct netlbl_dom_map *entry;
-
-	buf_len = NETLBL_LEN_U32;
-	rcu_read_lock();
-	entry = rcu_dereference(netlbl_domhsh_def);
-	if (entry != NULL)
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			buf_len += 2 * NETLBL_LEN_U32;
-			break;
-		}
-
-	skb = netlbl_netlink_alloc_skb(headroom, buf_len, GFP_ATOMIC);
-	if (skb == NULL)
-		goto dump_default_failure;
-
-	if (entry != rcu_dereference(netlbl_domhsh_def))
-		goto dump_default_failure;
-	if (entry != NULL) {
-		if (nla_put_u32(skb, NLA_U32, entry->type) != 0)
-			goto dump_default_failure;
-		buf_len -= NETLBL_LEN_U32;
-		switch (entry->type) {
-		case NETLBL_NLTYPE_UNLABELED:
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			if (buf_len < 2 * NETLBL_LEN_U32)
-				goto dump_default_failure;
-			if (nla_put_u32(skb,
-					NLA_U32,
-					entry->type_def.cipsov4->type) != 0)
-				goto dump_default_failure;
-			if (nla_put_u32(skb,
-					NLA_U32,
-					entry->type_def.cipsov4->doi) != 0)
-				goto dump_default_failure;
-			buf_len -= 2 * NETLBL_LEN_U32;
-			break;
-		}
-	} else
-		nla_put_u32(skb, NLA_U32, NETLBL_NLTYPE_NONE);
-	rcu_read_unlock();
-
-	return skb;
+	}
 
-dump_default_failure:
+walk_return:
 	rcu_read_unlock();
-	kfree_skb(skb);
-	return NULL;
+	*skip_bkt = iter_bkt;
+	*skip_chain = chain_cnt;
+	return ret_val;
 }
diff --git a/net/netlabel/netlabel_domainhash.h b/net/netlabel/netlabel_domainhash.h
index 99a2287de246..02af72a7877c 100644
--- a/net/netlabel/netlabel_domainhash.h
+++ b/net/netlabel/netlabel_domainhash.h
@@ -61,7 +61,9 @@ int netlbl_domhsh_add(struct netlbl_dom_map *entry);
 int netlbl_domhsh_add_default(struct netlbl_dom_map *entry);
 int netlbl_domhsh_remove_default(void);
 struct netlbl_dom_map *netlbl_domhsh_getentry(const char *domain);
-struct sk_buff *netlbl_domhsh_dump(size_t headroom);
-struct sk_buff *netlbl_domhsh_dump_default(size_t headroom);
+int netlbl_domhsh_walk(u32 *skip_bkt,
+		     u32 *skip_chain,
+		     int (*callback) (struct netlbl_dom_map *entry, void *arg),
+		     void *cb_arg);
 
 #endif
diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
index 73cbe66e42ff..eeb7d768d2bb 100644
--- a/net/netlabel/netlabel_user.c
+++ b/net/netlabel/netlabel_user.c
@@ -74,85 +74,3 @@ int netlbl_netlink_init(void)
 
 	return 0;
 }
-
-/*
- * NetLabel Common Protocol Functions
- */
-
-/**
- * netlbl_netlink_send_ack - Send an ACK message
- * @info: the generic NETLINK information
- * @genl_family: the generic NETLINK family ID value
- * @ack_cmd: the generic NETLINK family ACK command value
- * @ret_code: return code to use
- *
- * Description:
- * This function sends an ACK message to the sender of the NETLINK message
- * specified by @info.
- *
- */
-void netlbl_netlink_send_ack(const struct genl_info *info,
-			     u32 genl_family,
-			     u8 ack_cmd,
-			     u32 ret_code)
-{
-	size_t data_size;
-	struct sk_buff *skb;
-
-	data_size = GENL_HDRLEN + 2 * NETLBL_LEN_U32;
-	skb = netlbl_netlink_alloc_skb(0, data_size, GFP_KERNEL);
-	if (skb == NULL)
-		return;
-
-	if (netlbl_netlink_hdr_put(skb,
-				   info->snd_pid,
-				   0,
-				   genl_family,
-				   ack_cmd) == NULL)
-		goto send_ack_failure;
-
-	if (nla_put_u32(skb, NLA_U32, info->snd_seq) != 0)
-		goto send_ack_failure;
-	if (nla_put_u32(skb, NLA_U32, ret_code) != 0)
-		goto send_ack_failure;
-
-	netlbl_netlink_snd(skb, info->snd_pid);
-	return;
-
-send_ack_failure:
-	kfree_skb(skb);
-}
-
-/*
- * NETLINK I/O Functions
- */
-
-/**
- * netlbl_netlink_snd - Send a NetLabel message
- * @skb: NetLabel message
- * @pid: destination PID
- *
- * Description:
- * Sends a unicast NetLabel message over the NETLINK socket.
- *
- */
-int netlbl_netlink_snd(struct sk_buff *skb, u32 pid)
-{
-	return genlmsg_unicast(skb, pid);
-}
-
-/**
- * netlbl_netlink_snd - Send a NetLabel message
- * @skb: NetLabel message
- * @pid: sending PID
- * @group: multicast group id
- *
- * Description:
- * Sends a multicast NetLabel message over the NETLINK socket to all members
- * of @group except @pid.
- *
- */
-int netlbl_netlink_snd_multicast(struct sk_buff *skb, u32 pid, u32 group)
-{
-	return genlmsg_multicast(skb, pid, group, GFP_KERNEL);
-}
diff --git a/net/netlabel/netlabel_user.h b/net/netlabel/netlabel_user.h
index 385a6c7488c6..3f9386b917df 100644
--- a/net/netlabel/netlabel_user.h
+++ b/net/netlabel/netlabel_user.h
@@ -40,72 +40,6 @@
 
 /* NetLabel NETLINK helper functions */
 
-/**
- * netlbl_netlink_cap_check - Check the NETLINK msg capabilities
- * @skb: the NETLINK buffer
- * @req_cap: the required capability
- *
- * Description:
- * Check the NETLINK buffer's capabilities against the required capabilities.
- * Returns zero on success, negative values on failure.
- *
- */
-static inline int netlbl_netlink_cap_check(const struct sk_buff *skb,
-					   kernel_cap_t req_cap)
-{
-	if (cap_raised(NETLINK_CB(skb).eff_cap, req_cap))
-		return 0;
-	return -EPERM;
-}
-
-/**
- * netlbl_getinc_u8 - Read a u8 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u8 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u8 netlbl_getinc_u8(struct nlattr **nla, int *rem_len)
-{
-	u8 val = nla_get_u8(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
-/**
- * netlbl_getinc_u16 - Read a u16 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u16 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u16 netlbl_getinc_u16(struct nlattr **nla, int *rem_len)
-{
-	u16 val = nla_get_u16(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
-/**
- * netlbl_getinc_u32 - Read a u32 value from a nlattr stream and move on
- * @nla: the attribute
- * @rem_len: remaining length
- *
- * Description:
- * Return a u32 value pointed to by @nla and advance it to the next attribute.
- *
- */
-static inline u32 netlbl_getinc_u32(struct nlattr **nla, int *rem_len)
-{
-	u32 val = nla_get_u32(*nla);
-	*nla = nla_next(*nla, rem_len);
-	return val;
-}
-
 /**
  * netlbl_netlink_hdr_put - Write the NETLINK buffers into a sk_buff
  * @skb: the packet
@@ -124,6 +58,7 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 					   u32 pid,
 					   u32 seq,
 					   int type,
+					   int flags,
 					   u8 cmd)
 {
 	return genlmsg_put(skb,
@@ -131,85 +66,13 @@ static inline void *netlbl_netlink_hdr_put(struct sk_buff *skb,
 			   seq,
 			   type,
 			   0,
-			   0,
+			   flags,
 			   cmd,
 			   NETLBL_PROTO_VERSION);
 }
 
-/**
- * netlbl_netlink_hdr_push - Write the NETLINK buffers into a sk_buff
- * @skb: the packet
- * @pid: the PID of the receipient
- * @seq: the sequence number
- * @type: the generic NETLINK message family type
- * @cmd: command
- *
- * Description:
- * Write both a NETLINK nlmsghdr structure and a Generic NETLINK genlmsghdr
- * struct to the packet.
- *
- */
-static inline void netlbl_netlink_hdr_push(struct sk_buff *skb,
-					   u32 pid,
-					   u32 seq,
-					   int type,
-					   u8 cmd)
-
-{
-	struct nlmsghdr *nlh;
-	struct genlmsghdr *hdr;
-
-	nlh = (struct nlmsghdr *)skb_push(skb, NLMSG_SPACE(GENL_HDRLEN));
-	nlh->nlmsg_type = type;
-	nlh->nlmsg_len = skb->len;
-	nlh->nlmsg_flags = 0;
-	nlh->nlmsg_pid = pid;
-	nlh->nlmsg_seq = seq;
-
-	hdr = nlmsg_data(nlh);
-	hdr->cmd = cmd;
-	hdr->version = NETLBL_PROTO_VERSION;
-	hdr->reserved = 0;
-}
-
-/**
- * netlbl_netlink_payload_len - Return the length of the payload
- * @skb: the NETLINK buffer
- *
- * Description:
- * This function returns the length of the NetLabel payload.
- *
- */
-static inline u32 netlbl_netlink_payload_len(const struct sk_buff *skb)
-{
-	return nlmsg_len((struct nlmsghdr *)skb->data) - GENL_HDRLEN;
-}
-
-/**
- * netlbl_netlink_payload_data - Returns a pointer to the start of the payload
- * @skb: the NETLINK buffer
- *
- * Description:
- * This function returns a pointer to the start of the NetLabel payload.
- *
- */
-static inline void *netlbl_netlink_payload_data(const struct sk_buff *skb)
-{
-  return (unsigned char *)nlmsg_data((struct nlmsghdr *)skb->data) +
-	  GENL_HDRLEN;
-}
-
-/* NetLabel common protocol functions */
-
-void netlbl_netlink_send_ack(const struct genl_info *info,
-			     u32 genl_family,
-			     u8 ack_cmd,
-			     u32 ret_code);
-
 /* NetLabel NETLINK I/O functions */
 
 int netlbl_netlink_init(void);
-int netlbl_netlink_snd(struct sk_buff *skb, u32 pid);
-int netlbl_netlink_snd_multicast(struct sk_buff *skb, u32 pid, u32 group);
 
 #endif
-- 
cgit v1.2.1


From fd3858554b62c3af6b7664b5c58ad864c87116c9 Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:56:37 -0700
Subject: [NetLabel]: rework the Netlink attribute handling (part 2)

At the suggestion of Thomas Graf, rewrite NetLabel's use of Netlink attributes
to better follow the common Netlink attribute usage.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netlabel/netlabel_cipso_v4.c  | 628 +++++++++++++++++++++++++-------------
 net/netlabel/netlabel_cipso_v4.h  | 225 ++++++--------
 net/netlabel/netlabel_mgmt.c      | 541 ++++++++++++++++----------------
 net/netlabel/netlabel_mgmt.h      | 211 +++++--------
 net/netlabel/netlabel_unlabeled.c |  79 +++--
 net/netlabel/netlabel_unlabeled.h |  41 +--
 6 files changed, 895 insertions(+), 830 deletions(-)

diff --git a/net/netlabel/netlabel_cipso_v4.c b/net/netlabel/netlabel_cipso_v4.c
index a4f40adc447b..4125a55f469f 100644
--- a/net/netlabel/netlabel_cipso_v4.c
+++ b/net/netlabel/netlabel_cipso_v4.c
@@ -41,15 +41,37 @@
 #include "netlabel_user.h"
 #include "netlabel_cipso_v4.h"
 
+/* Argument struct for cipso_v4_doi_walk() */
+struct netlbl_cipsov4_doiwalk_arg {
+	struct netlink_callback *nl_cb;
+	struct sk_buff *skb;
+	u32 seq;
+};
+
 /* NetLabel Generic NETLINK CIPSOv4 family */
 static struct genl_family netlbl_cipsov4_gnl_family = {
 	.id = GENL_ID_GENERATE,
 	.hdrsize = 0,
 	.name = NETLBL_NLTYPE_CIPSOV4_NAME,
 	.version = NETLBL_PROTO_VERSION,
-	.maxattr = 0,
+	.maxattr = NLBL_CIPSOV4_A_MAX,
 };
 
+/* NetLabel Netlink attribute policy */
+static struct nla_policy netlbl_cipsov4_genl_policy[NLBL_CIPSOV4_A_MAX + 1] = {
+	[NLBL_CIPSOV4_A_DOI] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_MTYPE] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_TAG] = { .type = NLA_U8 },
+	[NLBL_CIPSOV4_A_TAGLST] = { .type = NLA_NESTED },
+	[NLBL_CIPSOV4_A_MLSLVLLOC] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_MLSLVLREM] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_MLSLVL] = { .type = NLA_NESTED },
+	[NLBL_CIPSOV4_A_MLSLVLLST] = { .type = NLA_NESTED },
+	[NLBL_CIPSOV4_A_MLSCATLOC] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_MLSCATREM] = { .type = NLA_U32 },
+	[NLBL_CIPSOV4_A_MLSCAT] = { .type = NLA_NESTED },
+	[NLBL_CIPSOV4_A_MLSCATLST] = { .type = NLA_NESTED },
+};
 
 /*
  * Helper Functions
@@ -81,6 +103,41 @@ static void netlbl_cipsov4_doi_free(struct rcu_head *entry)
 	kfree(ptr);
 }
 
+/**
+ * netlbl_cipsov4_add_common - Parse the common sections of a ADD message
+ * @info: the Generic NETLINK info block
+ * @doi_def: the CIPSO V4 DOI definition
+ *
+ * Description:
+ * Parse the common sections of a ADD message and fill in the related values
+ * in @doi_def.  Returns zero on success, negative values on failure.
+ *
+ */
+static int netlbl_cipsov4_add_common(struct genl_info *info,
+				     struct cipso_v4_doi *doi_def)
+{
+	struct nlattr *nla;
+	int nla_rem;
+	u32 iter = 0;
+
+	doi_def->doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+
+	if (nla_validate_nested(info->attrs[NLBL_CIPSOV4_A_TAGLST],
+				NLBL_CIPSOV4_A_MAX,
+				netlbl_cipsov4_genl_policy) != 0)
+		return -EINVAL;
+
+	nla_for_each_nested(nla, info->attrs[NLBL_CIPSOV4_A_TAGLST], nla_rem)
+		if (nla->nla_type == NLBL_CIPSOV4_A_TAG) {
+			if (iter > CIPSO_V4_TAG_MAXCNT)
+				return -EINVAL;
+			doi_def->tags[iter++] = nla_get_u8(nla);
+		}
+	if (iter < CIPSO_V4_TAG_MAXCNT)
+		doi_def->tags[iter] = CIPSO_V4_TAG_INVALID;
+
+	return 0;
+}
 
 /*
  * NetLabel Command Handlers
@@ -88,9 +145,7 @@ static void netlbl_cipsov4_doi_free(struct rcu_head *entry)
 
 /**
  * netlbl_cipsov4_add_std - Adds a CIPSO V4 DOI definition
- * @doi: the DOI value
- * @msg: the ADD message data
- * @msg_size: the size of the ADD message buffer
+ * @info: the Generic NETLINK info block
  *
  * Description:
  * Create a new CIPSO_V4_MAP_STD DOI definition based on the given ADD message
@@ -98,29 +153,28 @@ static void netlbl_cipsov4_doi_free(struct rcu_head *entry)
  * error.
  *
  */
-static int netlbl_cipsov4_add_std(u32 doi, struct nlattr *msg, size_t msg_size)
+static int netlbl_cipsov4_add_std(struct genl_info *info)
 {
 	int ret_val = -EINVAL;
-	int msg_len = msg_size;
-	u32 num_tags;
-	u32 num_lvls;
-	u32 num_cats;
 	struct cipso_v4_doi *doi_def = NULL;
-	u32 iter;
-	u32 tmp_val_a;
-	u32 tmp_val_b;
+	struct nlattr *nla_a;
+	struct nlattr *nla_b;
+	int nla_a_rem;
+	int nla_b_rem;
 
-	if (msg_len < NETLBL_LEN_U32)
-		goto add_std_failure;
-	num_tags = netlbl_getinc_u32(&msg, &msg_len);
-	if (num_tags == 0 || num_tags > CIPSO_V4_TAG_MAXCNT)
-		goto add_std_failure;
+	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
+	    !info->attrs[NLBL_CIPSOV4_A_TAGLST] ||
+	    !info->attrs[NLBL_CIPSOV4_A_MLSLVLLST])
+		return -EINVAL;
+
+	if (nla_validate_nested(info->attrs[NLBL_CIPSOV4_A_MLSLVLLST],
+				NLBL_CIPSOV4_A_MAX,
+				netlbl_cipsov4_genl_policy) != 0)
+		return -EINVAL;
 
 	doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
-	if (doi_def == NULL) {
-		ret_val = -ENOMEM;
-		goto add_std_failure;
-	}
+	if (doi_def == NULL)
+		return -ENOMEM;
 	doi_def->map.std = kzalloc(sizeof(*doi_def->map.std), GFP_KERNEL);
 	if (doi_def->map.std == NULL) {
 		ret_val = -ENOMEM;
@@ -128,28 +182,32 @@ static int netlbl_cipsov4_add_std(u32 doi, struct nlattr *msg, size_t msg_size)
 	}
 	doi_def->type = CIPSO_V4_MAP_STD;
 
-	for (iter = 0; iter < num_tags; iter++) {
-		if (msg_len < NETLBL_LEN_U8)
-			goto add_std_failure;
-		doi_def->tags[iter] = netlbl_getinc_u8(&msg, &msg_len);
-		switch (doi_def->tags[iter]) {
-		case CIPSO_V4_TAG_RBITMAP:
-			break;
-		default:
-			goto add_std_failure;
-		}
-	}
-	if (iter < CIPSO_V4_TAG_MAXCNT)
-		doi_def->tags[iter] = CIPSO_V4_TAG_INVALID;
-
-	if (msg_len < 6 * NETLBL_LEN_U32)
+	ret_val = netlbl_cipsov4_add_common(info, doi_def);
+	if (ret_val != 0)
 		goto add_std_failure;
 
-	num_lvls = netlbl_getinc_u32(&msg, &msg_len);
-	if (num_lvls == 0)
-		goto add_std_failure;
-	doi_def->map.std->lvl.local_size = netlbl_getinc_u32(&msg, &msg_len);
-	if (doi_def->map.std->lvl.local_size > CIPSO_V4_MAX_LOC_LVLS)
+	nla_for_each_nested(nla_a,
+			    info->attrs[NLBL_CIPSOV4_A_MLSLVLLST],
+			    nla_a_rem)
+		if (nla_a->nla_type == NLBL_CIPSOV4_A_MLSLVL) {
+			nla_for_each_nested(nla_b, nla_a, nla_b_rem)
+				switch (nla_b->nla_type) {
+				case NLBL_CIPSOV4_A_MLSLVLLOC:
+					if (nla_get_u32(nla_b) >=
+					    doi_def->map.std->lvl.local_size)
+					     doi_def->map.std->lvl.local_size =
+						     nla_get_u32(nla_b) + 1;
+					break;
+				case NLBL_CIPSOV4_A_MLSLVLREM:
+					if (nla_get_u32(nla_b) >=
+					    doi_def->map.std->lvl.cipso_size)
+					     doi_def->map.std->lvl.cipso_size =
+						     nla_get_u32(nla_b) + 1;
+					break;
+				}
+		}
+	if (doi_def->map.std->lvl.local_size > CIPSO_V4_MAX_LOC_LVLS ||
+	    doi_def->map.std->lvl.cipso_size > CIPSO_V4_MAX_REM_LVLS)
 		goto add_std_failure;
 	doi_def->map.std->lvl.local = kcalloc(doi_def->map.std->lvl.local_size,
 					      sizeof(u32),
@@ -158,9 +216,6 @@ static int netlbl_cipsov4_add_std(u32 doi, struct nlattr *msg, size_t msg_size)
 		ret_val = -ENOMEM;
 		goto add_std_failure;
 	}
-	doi_def->map.std->lvl.cipso_size = netlbl_getinc_u8(&msg, &msg_len);
-	if (doi_def->map.std->lvl.cipso_size > CIPSO_V4_MAX_REM_LVLS)
-		goto add_std_failure;
 	doi_def->map.std->lvl.cipso = kcalloc(doi_def->map.std->lvl.cipso_size,
 					      sizeof(u32),
 					      GFP_KERNEL);
@@ -168,68 +223,101 @@ static int netlbl_cipsov4_add_std(u32 doi, struct nlattr *msg, size_t msg_size)
 		ret_val = -ENOMEM;
 		goto add_std_failure;
 	}
+	nla_for_each_nested(nla_a,
+			    info->attrs[NLBL_CIPSOV4_A_MLSLVLLST],
+			    nla_a_rem)
+		if (nla_a->nla_type == NLBL_CIPSOV4_A_MLSLVL) {
+			struct nlattr *lvl_loc;
+			struct nlattr *lvl_rem;
+
+			if (nla_validate_nested(nla_a,
+					      NLBL_CIPSOV4_A_MAX,
+					      netlbl_cipsov4_genl_policy) != 0)
+				goto add_std_failure;
+
+			lvl_loc = nla_find_nested(nla_a,
+						  NLBL_CIPSOV4_A_MLSLVLLOC);
+			lvl_rem = nla_find_nested(nla_a,
+						  NLBL_CIPSOV4_A_MLSLVLREM);
+			if (lvl_loc == NULL || lvl_rem == NULL)
+				goto add_std_failure;
+			doi_def->map.std->lvl.local[nla_get_u32(lvl_loc)] =
+				nla_get_u32(lvl_rem);
+			doi_def->map.std->lvl.cipso[nla_get_u32(lvl_rem)] =
+				nla_get_u32(lvl_loc);
+		}
 
-	num_cats = netlbl_getinc_u32(&msg, &msg_len);
-	doi_def->map.std->cat.local_size = netlbl_getinc_u32(&msg, &msg_len);
-	if (doi_def->map.std->cat.local_size > CIPSO_V4_MAX_LOC_CATS)
-		goto add_std_failure;
-	doi_def->map.std->cat.local = kcalloc(doi_def->map.std->cat.local_size,
+	if (info->attrs[NLBL_CIPSOV4_A_MLSCATLST]) {
+		if (nla_validate_nested(info->attrs[NLBL_CIPSOV4_A_MLSCATLST],
+					NLBL_CIPSOV4_A_MAX,
+					netlbl_cipsov4_genl_policy) != 0)
+			goto add_std_failure;
+
+		nla_for_each_nested(nla_a,
+				    info->attrs[NLBL_CIPSOV4_A_MLSCATLST],
+				    nla_a_rem)
+			if (nla_a->nla_type == NLBL_CIPSOV4_A_MLSCAT) {
+				if (nla_validate_nested(nla_a,
+					      NLBL_CIPSOV4_A_MAX,
+					      netlbl_cipsov4_genl_policy) != 0)
+					goto add_std_failure;
+				nla_for_each_nested(nla_b, nla_a, nla_b_rem)
+					switch (nla_b->nla_type) {
+					case NLBL_CIPSOV4_A_MLSCATLOC:
+						if (nla_get_u32(nla_b) >=
+					      doi_def->map.std->cat.local_size)
+					     doi_def->map.std->cat.local_size =
+						     nla_get_u32(nla_b) + 1;
+						break;
+					case NLBL_CIPSOV4_A_MLSCATREM:
+						if (nla_get_u32(nla_b) >=
+					      doi_def->map.std->cat.cipso_size)
+					     doi_def->map.std->cat.cipso_size =
+						     nla_get_u32(nla_b) + 1;
+						break;
+					}
+			}
+		if (doi_def->map.std->cat.local_size > CIPSO_V4_MAX_LOC_CATS ||
+		    doi_def->map.std->cat.cipso_size > CIPSO_V4_MAX_REM_CATS)
+			goto add_std_failure;
+		doi_def->map.std->cat.local = kcalloc(
+			                      doi_def->map.std->cat.local_size,
 					      sizeof(u32),
 					      GFP_KERNEL);
-	if (doi_def->map.std->cat.local == NULL) {
-		ret_val = -ENOMEM;
-		goto add_std_failure;
-	}
-	doi_def->map.std->cat.cipso_size = netlbl_getinc_u16(&msg, &msg_len);
-	if (doi_def->map.std->cat.cipso_size > CIPSO_V4_MAX_REM_CATS)
-		goto add_std_failure;
-	doi_def->map.std->cat.cipso = kcalloc(doi_def->map.std->cat.cipso_size,
+		if (doi_def->map.std->cat.local == NULL) {
+			ret_val = -ENOMEM;
+			goto add_std_failure;
+		}
+		doi_def->map.std->cat.cipso = kcalloc(
+			                      doi_def->map.std->cat.cipso_size,
 					      sizeof(u32),
 					      GFP_KERNEL);
-	if (doi_def->map.std->cat.cipso == NULL) {
-		ret_val = -ENOMEM;
-		goto add_std_failure;
-	}
-
-	if (msg_len <
-	    num_lvls * (NETLBL_LEN_U32 + NETLBL_LEN_U8) +
-	    num_cats * (NETLBL_LEN_U32 + NETLBL_LEN_U16))
-		goto add_std_failure;
-
-	for (iter = 0; iter < doi_def->map.std->lvl.cipso_size; iter++)
-		doi_def->map.std->lvl.cipso[iter] = CIPSO_V4_INV_LVL;
-	for (iter = 0; iter < doi_def->map.std->lvl.local_size; iter++)
-		doi_def->map.std->lvl.local[iter] = CIPSO_V4_INV_LVL;
-	for (iter = 0; iter < doi_def->map.std->cat.cipso_size; iter++)
-		doi_def->map.std->cat.cipso[iter] = CIPSO_V4_INV_CAT;
-	for (iter = 0; iter < doi_def->map.std->cat.local_size; iter++)
-		doi_def->map.std->cat.local[iter] = CIPSO_V4_INV_CAT;
-
-	for (iter = 0; iter < num_lvls; iter++) {
-		tmp_val_a = netlbl_getinc_u32(&msg, &msg_len);
-		tmp_val_b = netlbl_getinc_u8(&msg, &msg_len);
-
-		if (tmp_val_a >= doi_def->map.std->lvl.local_size ||
-		    tmp_val_b >= doi_def->map.std->lvl.cipso_size)
-			goto add_std_failure;
-
-		doi_def->map.std->lvl.cipso[tmp_val_b] = tmp_val_a;
-		doi_def->map.std->lvl.local[tmp_val_a] = tmp_val_b;
-	}
-
-	for (iter = 0; iter < num_cats; iter++) {
-		tmp_val_a = netlbl_getinc_u32(&msg, &msg_len);
-		tmp_val_b = netlbl_getinc_u16(&msg, &msg_len);
-
-		if (tmp_val_a >= doi_def->map.std->cat.local_size ||
-		    tmp_val_b >= doi_def->map.std->cat.cipso_size)
+		if (doi_def->map.std->cat.cipso == NULL) {
+			ret_val = -ENOMEM;
 			goto add_std_failure;
-
-		doi_def->map.std->cat.cipso[tmp_val_b] = tmp_val_a;
-		doi_def->map.std->cat.local[tmp_val_a] = tmp_val_b;
+		}
+		nla_for_each_nested(nla_a,
+				    info->attrs[NLBL_CIPSOV4_A_MLSCATLST],
+				    nla_a_rem)
+			if (nla_a->nla_type == NLBL_CIPSOV4_A_MLSCAT) {
+				struct nlattr *cat_loc;
+				struct nlattr *cat_rem;
+
+				cat_loc = nla_find_nested(nla_a,
+						     NLBL_CIPSOV4_A_MLSCATLOC);
+				cat_rem = nla_find_nested(nla_a,
+						     NLBL_CIPSOV4_A_MLSCATREM);
+				if (cat_loc == NULL || cat_rem == NULL)
+					goto add_std_failure;
+				doi_def->map.std->cat.local[
+				                        nla_get_u32(cat_loc)] =
+					nla_get_u32(cat_rem);
+				doi_def->map.std->cat.cipso[
+					                nla_get_u32(cat_rem)] =
+					nla_get_u32(cat_loc);
+			}
 	}
 
-	doi_def->doi = doi;
 	ret_val = cipso_v4_doi_add(doi_def);
 	if (ret_val != 0)
 		goto add_std_failure;
@@ -243,9 +331,7 @@ add_std_failure:
 
 /**
  * netlbl_cipsov4_add_pass - Adds a CIPSO V4 DOI definition
- * @doi: the DOI value
- * @msg: the ADD message data
- * @msg_size: the size of the ADD message buffer
+ * @info: the Generic NETLINK info block
  *
  * Description:
  * Create a new CIPSO_V4_MAP_PASS DOI definition based on the given ADD message
@@ -253,52 +339,31 @@ add_std_failure:
  * error.
  *
  */
-static int netlbl_cipsov4_add_pass(u32 doi,
-				   struct nlattr *msg,
-				   size_t msg_size)
+static int netlbl_cipsov4_add_pass(struct genl_info *info)
 {
-	int ret_val = -EINVAL;
-	int msg_len = msg_size;
-	u32 num_tags;
+	int ret_val;
 	struct cipso_v4_doi *doi_def = NULL;
-	u32 iter;
 
-	if (msg_len < NETLBL_LEN_U32)
-		goto add_pass_failure;
-	num_tags = netlbl_getinc_u32(&msg, &msg_len);
-	if (num_tags == 0 || num_tags > CIPSO_V4_TAG_MAXCNT)
-		goto add_pass_failure;
+	if (!info->attrs[NLBL_CIPSOV4_A_DOI] ||
+	    !info->attrs[NLBL_CIPSOV4_A_TAGLST])
+		return -EINVAL;
 
 	doi_def = kmalloc(sizeof(*doi_def), GFP_KERNEL);
-	if (doi_def == NULL) {
-		ret_val = -ENOMEM;
-		goto add_pass_failure;
-	}
+	if (doi_def == NULL)
+		return -ENOMEM;
 	doi_def->type = CIPSO_V4_MAP_PASS;
 
-	for (iter = 0; iter < num_tags; iter++) {
-		if (msg_len < NETLBL_LEN_U8)
-			goto add_pass_failure;
-		doi_def->tags[iter] = netlbl_getinc_u8(&msg, &msg_len);
-		switch (doi_def->tags[iter]) {
-		case CIPSO_V4_TAG_RBITMAP:
-			break;
-		default:
-			goto add_pass_failure;
-		}
-	}
-	if (iter < CIPSO_V4_TAG_MAXCNT)
-		doi_def->tags[iter] = CIPSO_V4_TAG_INVALID;
+	ret_val = netlbl_cipsov4_add_common(info, doi_def);
+	if (ret_val != 0)
+		goto add_pass_failure;
 
-	doi_def->doi = doi;
 	ret_val = cipso_v4_doi_add(doi_def);
 	if (ret_val != 0)
 		goto add_pass_failure;
 	return 0;
 
 add_pass_failure:
-	if (doi_def)
-		netlbl_cipsov4_doi_free(&doi_def->rcu);
+	netlbl_cipsov4_doi_free(&doi_def->rcu);
 	return ret_val;
 }
 
@@ -316,34 +381,21 @@ static int netlbl_cipsov4_add(struct sk_buff *skb, struct genl_info *info)
 
 {
 	int ret_val = -EINVAL;
-	u32 doi;
 	u32 map_type;
-	int msg_len = netlbl_netlink_payload_len(skb);
-	struct nlattr *msg = netlbl_netlink_payload_data(skb);
-
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		goto add_return;
 
-	if (msg_len < 2 * NETLBL_LEN_U32)
-		goto add_return;
+	if (!info->attrs[NLBL_CIPSOV4_A_MTYPE])
+		return -EINVAL;
 
-	doi = netlbl_getinc_u32(&msg, &msg_len);
-	map_type = netlbl_getinc_u32(&msg, &msg_len);
+	map_type = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_MTYPE]);
 	switch (map_type) {
 	case CIPSO_V4_MAP_STD:
-		ret_val = netlbl_cipsov4_add_std(doi, msg, msg_len);
+		ret_val = netlbl_cipsov4_add_std(info);
 		break;
 	case CIPSO_V4_MAP_PASS:
-		ret_val = netlbl_cipsov4_add_pass(doi, msg, msg_len);
+		ret_val = netlbl_cipsov4_add_pass(info);
 		break;
 	}
 
-add_return:
-	netlbl_netlink_send_ack(info,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_ACK,
-				-ret_val);
 	return ret_val;
 }
 
@@ -353,84 +405,239 @@ add_return:
  * @info: the Generic NETLINK info block
  *
  * Description:
- * Process a user generated LIST message and respond accordingly.  Returns
- * zero on success and negative values on error.
+ * Process a user generated LIST message and respond accordingly.  While the
+ * response message generated by the kernel is straightforward, determining
+ * before hand the size of the buffer to allocate is not (we have to generate
+ * the message to know the size).  In order to keep this function sane what we
+ * do is allocate a buffer of NLMSG_GOODSIZE and try to fit the response in
+ * that size, if we fail then we restart with a larger buffer and try again.
+ * We continue in this manner until we hit a limit of failed attempts then we
+ * give up and just send an error message.  Returns zero on success and
+ * negative values on error.
  *
  */
 static int netlbl_cipsov4_list(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val = -EINVAL;
+	int ret_val;
+	struct sk_buff *ans_skb = NULL;
+	u32 nlsze_mult = 1;
+	void *data;
 	u32 doi;
-	struct nlattr *msg = netlbl_netlink_payload_data(skb);
-	struct sk_buff *ans_skb;
+	struct nlattr *nla_a;
+	struct nlattr *nla_b;
+	struct cipso_v4_doi *doi_def;
+	u32 iter;
 
-	if (netlbl_netlink_payload_len(skb) != NETLBL_LEN_U32)
+	if (!info->attrs[NLBL_CIPSOV4_A_DOI]) {
+		ret_val = -EINVAL;
 		goto list_failure;
+	}
 
-	doi = nla_get_u32(msg);
-	ans_skb = cipso_v4_doi_dump(doi, NLMSG_SPACE(GENL_HDRLEN));
+list_start:
+	ans_skb = nlmsg_new(NLMSG_GOODSIZE * nlsze_mult, GFP_KERNEL);
 	if (ans_skb == NULL) {
 		ret_val = -ENOMEM;
 		goto list_failure;
 	}
-	netlbl_netlink_hdr_push(ans_skb,
-				info->snd_pid,
-				0,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_LIST);
+	data = netlbl_netlink_hdr_put(ans_skb,
+				      info->snd_pid,
+				      info->snd_seq,
+				      netlbl_cipsov4_gnl_family.id,
+				      0,
+				      NLBL_CIPSOV4_C_LIST);
+	if (data == NULL) {
+		ret_val = -ENOMEM;
+		goto list_failure;
+	}
+
+	doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+
+	rcu_read_lock();
+	doi_def = cipso_v4_doi_getdef(doi);
+	if (doi_def == NULL) {
+		ret_val = -EINVAL;
+		goto list_failure;
+	}
+
+	ret_val = nla_put_u32(ans_skb, NLBL_CIPSOV4_A_MTYPE, doi_def->type);
+	if (ret_val != 0)
+		goto list_failure_lock;
+
+	nla_a = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_TAGLST);
+	if (nla_a == NULL) {
+		ret_val = -ENOMEM;
+		goto list_failure_lock;
+	}
+	for (iter = 0;
+	     iter < CIPSO_V4_TAG_MAXCNT &&
+	       doi_def->tags[iter] != CIPSO_V4_TAG_INVALID;
+	     iter++) {
+		ret_val = nla_put_u8(ans_skb,
+				     NLBL_CIPSOV4_A_TAG,
+				     doi_def->tags[iter]);
+		if (ret_val != 0)
+			goto list_failure_lock;
+	}
+	nla_nest_end(ans_skb, nla_a);
+
+	switch (doi_def->type) {
+	case CIPSO_V4_MAP_STD:
+		nla_a = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSLVLLST);
+		if (nla_a == NULL) {
+			ret_val = -ENOMEM;
+			goto list_failure_lock;
+		}
+		for (iter = 0;
+		     iter < doi_def->map.std->lvl.local_size;
+		     iter++) {
+			if (doi_def->map.std->lvl.local[iter] ==
+			    CIPSO_V4_INV_LVL)
+				continue;
+
+			nla_b = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSLVL);
+			if (nla_b == NULL) {
+				ret_val = -ENOMEM;
+				goto list_retry;
+			}
+			ret_val = nla_put_u32(ans_skb,
+					      NLBL_CIPSOV4_A_MLSLVLLOC,
+					      iter);
+			if (ret_val != 0)
+				goto list_retry;
+			ret_val = nla_put_u32(ans_skb,
+					    NLBL_CIPSOV4_A_MLSLVLREM,
+					    doi_def->map.std->lvl.local[iter]);
+			if (ret_val != 0)
+				goto list_retry;
+			nla_nest_end(ans_skb, nla_b);
+		}
+		nla_nest_end(ans_skb, nla_a);
+
+		nla_a = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSCATLST);
+		if (nla_a == NULL) {
+			ret_val = -ENOMEM;
+			goto list_retry;
+		}
+		for (iter = 0;
+		     iter < doi_def->map.std->cat.local_size;
+		     iter++) {
+			if (doi_def->map.std->cat.local[iter] ==
+			    CIPSO_V4_INV_CAT)
+				continue;
+
+			nla_b = nla_nest_start(ans_skb, NLBL_CIPSOV4_A_MLSCAT);
+			if (nla_b == NULL) {
+				ret_val = -ENOMEM;
+				goto list_retry;
+			}
+			ret_val = nla_put_u32(ans_skb,
+					      NLBL_CIPSOV4_A_MLSCATLOC,
+					      iter);
+			if (ret_val != 0)
+				goto list_retry;
+			ret_val = nla_put_u32(ans_skb,
+					    NLBL_CIPSOV4_A_MLSCATREM,
+					    doi_def->map.std->cat.local[iter]);
+			if (ret_val != 0)
+				goto list_retry;
+			nla_nest_end(ans_skb, nla_b);
+		}
+		nla_nest_end(ans_skb, nla_a);
+
+		break;
+	}
+	rcu_read_unlock();
 
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+	genlmsg_end(ans_skb, data);
+
+	ret_val = genlmsg_unicast(ans_skb, info->snd_pid);
 	if (ret_val != 0)
 		goto list_failure;
 
 	return 0;
 
+list_retry:
+	/* XXX - this limit is a guesstimate */
+	if (nlsze_mult < 4) {
+		rcu_read_unlock();
+		kfree_skb(ans_skb);
+		nlsze_mult++;
+		goto list_start;
+	}
+list_failure_lock:
+	rcu_read_unlock();
 list_failure:
-	netlbl_netlink_send_ack(info,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_ACK,
-				-ret_val);
+	kfree_skb(ans_skb);
+	return ret_val;
+}
+
+/**
+ * netlbl_cipsov4_listall_cb - cipso_v4_doi_walk() callback for LISTALL
+ * @doi_def: the CIPSOv4 DOI definition
+ * @arg: the netlbl_cipsov4_doiwalk_arg structure
+ *
+ * Description:
+ * This function is designed to be used as a callback to the
+ * cipso_v4_doi_walk() function for use in generating a response for a LISTALL
+ * message.  Returns the size of the message on success, negative values on
+ * failure.
+ *
+ */
+static int netlbl_cipsov4_listall_cb(struct cipso_v4_doi *doi_def, void *arg)
+{
+	int ret_val = -ENOMEM;
+	struct netlbl_cipsov4_doiwalk_arg *cb_arg = arg;
+	void *data;
+
+	data = netlbl_netlink_hdr_put(cb_arg->skb,
+				      NETLINK_CB(cb_arg->nl_cb->skb).pid,
+				      cb_arg->seq,
+				      netlbl_cipsov4_gnl_family.id,
+				      NLM_F_MULTI,
+				      NLBL_CIPSOV4_C_LISTALL);
+	if (data == NULL)
+		goto listall_cb_failure;
+
+	ret_val = nla_put_u32(cb_arg->skb, NLBL_CIPSOV4_A_DOI, doi_def->doi);
+	if (ret_val != 0)
+		goto listall_cb_failure;
+	ret_val = nla_put_u32(cb_arg->skb,
+			      NLBL_CIPSOV4_A_MTYPE,
+			      doi_def->type);
+	if (ret_val != 0)
+		goto listall_cb_failure;
+
+	return genlmsg_end(cb_arg->skb, data);
+
+listall_cb_failure:
+	genlmsg_cancel(cb_arg->skb, data);
 	return ret_val;
 }
 
 /**
  * netlbl_cipsov4_listall - Handle a LISTALL message
  * @skb: the NETLINK buffer
- * @info: the Generic NETLINK info block
+ * @cb: the NETLINK callback
  *
  * Description:
  * Process a user generated LISTALL message and respond accordingly.  Returns
  * zero on success and negative values on error.
  *
  */
-static int netlbl_cipsov4_listall(struct sk_buff *skb, struct genl_info *info)
+static int netlbl_cipsov4_listall(struct sk_buff *skb,
+				  struct netlink_callback *cb)
 {
-	int ret_val = -EINVAL;
-	struct sk_buff *ans_skb;
+	struct netlbl_cipsov4_doiwalk_arg cb_arg;
+	int doi_skip = cb->args[0];
 
-	ans_skb = cipso_v4_doi_dump_all(NLMSG_SPACE(GENL_HDRLEN));
-	if (ans_skb == NULL) {
-		ret_val = -ENOMEM;
-		goto listall_failure;
-	}
-	netlbl_netlink_hdr_push(ans_skb,
-				info->snd_pid,
-				0,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_LISTALL);
+	cb_arg.nl_cb = cb;
+	cb_arg.skb = skb;
+	cb_arg.seq = cb->nlh->nlmsg_seq;
 
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
-	if (ret_val != 0)
-		goto listall_failure;
-
-	return 0;
+	cipso_v4_doi_walk(&doi_skip, netlbl_cipsov4_listall_cb, &cb_arg);
 
-listall_failure:
-	netlbl_netlink_send_ack(info,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_ACK,
-				-ret_val);
-	return ret_val;
+	cb->args[0] = doi_skip;
+	return skb->len;
 }
 
 /**
@@ -445,27 +652,14 @@ listall_failure:
  */
 static int netlbl_cipsov4_remove(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val;
+	int ret_val = -EINVAL;
 	u32 doi;
-	struct nlattr *msg = netlbl_netlink_payload_data(skb);
 
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		goto remove_return;
-
-	if (netlbl_netlink_payload_len(skb) != NETLBL_LEN_U32) {
-		ret_val = -EINVAL;
-		goto remove_return;
+	if (info->attrs[NLBL_CIPSOV4_A_DOI]) {
+		doi = nla_get_u32(info->attrs[NLBL_CIPSOV4_A_DOI]);
+		ret_val = cipso_v4_doi_remove(doi, netlbl_cipsov4_doi_free);
 	}
 
-	doi = nla_get_u32(msg);
-	ret_val = cipso_v4_doi_remove(doi, netlbl_cipsov4_doi_free);
-
-remove_return:
-	netlbl_netlink_send_ack(info,
-				netlbl_cipsov4_gnl_family.id,
-				NLBL_CIPSOV4_C_ACK,
-				-ret_val);
 	return ret_val;
 }
 
@@ -475,14 +669,16 @@ remove_return:
 
 static struct genl_ops netlbl_cipsov4_genl_c_add = {
 	.cmd = NLBL_CIPSOV4_C_ADD,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_cipsov4_genl_policy,
 	.doit = netlbl_cipsov4_add,
 	.dumpit = NULL,
 };
 
 static struct genl_ops netlbl_cipsov4_genl_c_remove = {
 	.cmd = NLBL_CIPSOV4_C_REMOVE,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_cipsov4_genl_policy,
 	.doit = netlbl_cipsov4_remove,
 	.dumpit = NULL,
 };
@@ -490,6 +686,7 @@ static struct genl_ops netlbl_cipsov4_genl_c_remove = {
 static struct genl_ops netlbl_cipsov4_genl_c_list = {
 	.cmd = NLBL_CIPSOV4_C_LIST,
 	.flags = 0,
+	.policy = netlbl_cipsov4_genl_policy,
 	.doit = netlbl_cipsov4_list,
 	.dumpit = NULL,
 };
@@ -497,8 +694,9 @@ static struct genl_ops netlbl_cipsov4_genl_c_list = {
 static struct genl_ops netlbl_cipsov4_genl_c_listall = {
 	.cmd = NLBL_CIPSOV4_C_LISTALL,
 	.flags = 0,
-	.doit = netlbl_cipsov4_listall,
-	.dumpit = NULL,
+	.policy = netlbl_cipsov4_genl_policy,
+	.doit = NULL,
+	.dumpit = netlbl_cipsov4_listall,
 };
 
 /*
diff --git a/net/netlabel/netlabel_cipso_v4.h b/net/netlabel/netlabel_cipso_v4.h
index 4c6ff4b93004..f03cf9b78286 100644
--- a/net/netlabel/netlabel_cipso_v4.h
+++ b/net/netlabel/netlabel_cipso_v4.h
@@ -34,175 +34,71 @@
 #include <net/netlabel.h>
 
 /*
- * The following NetLabel payloads are supported by the CIPSO subsystem, all
- * of which are preceeded by the nlmsghdr struct.
+ * The following NetLabel payloads are supported by the CIPSO subsystem.
  *
- * o ACK:
- *   Sent by the kernel in response to an applications message, applications
- *   should never send this message.
+ * o ADD:
+ *   Sent by an application to add a new DOI mapping table.
  *
- *   +----------------------+-----------------------+
- *   | seq number (32 bits) | return code (32 bits) |
- *   +----------------------+-----------------------+
+ *   Required attributes:
  *
- *     seq number:  the sequence number of the original message, taken from the
- *                  nlmsghdr structure
- *     return code: return value, based on errno values
+ *     NLBL_CIPSOV4_A_DOI
+ *     NLBL_CIPSOV4_A_MTYPE
+ *     NLBL_CIPSOV4_A_TAGLST
  *
- * o ADD:
- *   Sent by an application to add a new DOI mapping table, after completion
- *   of the task the kernel should ACK this message.
- *
- *   +---------------+--------------------+---------------------+
- *   | DOI (32 bits) | map type (32 bits) | tag count (32 bits) | ...
- *   +---------------+--------------------+---------------------+
- *
- *   +-----------------+
- *   | tag #X (8 bits) | ... repeated
- *   +-----------------+
- *
- *   +-------------- ---- --- -- -
- *   | mapping data
- *   +-------------- ---- --- -- -
- *
- *     DOI:          the DOI value
- *     map type:     the mapping table type (defined in the cipso_ipv4.h header
- *                   as CIPSO_V4_MAP_*)
- *     tag count:    the number of tags, must be greater than zero
- *     tag:          the CIPSO tag for the DOI, tags listed first are given
- *                   higher priorirty when sending packets
- *     mapping data: specific to the map type (see below)
- *
- *   CIPSO_V4_MAP_STD
- *
- *   +------------------+-----------------------+----------------------+
- *   | levels (32 bits) | max l level (32 bits) | max r level (8 bits) | ...
- *   +------------------+-----------------------+----------------------+
- *
- *   +----------------------+---------------------+---------------------+
- *   | categories (32 bits) | max l cat (32 bits) | max r cat (16 bits) | ...
- *   +----------------------+---------------------+---------------------+
- *
- *   +--------------------------+-------------------------+
- *   | local level #X (32 bits) | CIPSO level #X (8 bits) | ... repeated
- *   +--------------------------+-------------------------+
- *
- *   +-----------------------------+-----------------------------+
- *   | local category #X (32 bits) | CIPSO category #X (16 bits) | ... repeated
- *   +-----------------------------+-----------------------------+
- *
- *     levels:         the number of level mappings
- *     max l level:    the highest local level
- *     max r level:    the highest remote/CIPSO level
- *     categories:     the number of category mappings
- *     max l cat:      the highest local category
- *     max r cat:      the highest remote/CIPSO category
- *     local level:    the local part of a level mapping
- *     CIPSO level:    the remote/CIPSO part of a level mapping
- *     local category: the local part of a category mapping
- *     CIPSO category: the remote/CIPSO part of a category mapping
- *
- *   CIPSO_V4_MAP_PASS
- *
- *   No mapping data is needed for this map type.
+ *   If using CIPSO_V4_MAP_STD the following attributes are required:
+ *
+ *     NLBL_CIPSOV4_A_MLSLVLLST
+ *     NLBL_CIPSOV4_A_MLSCATLST
+ *
+ *   If using CIPSO_V4_MAP_PASS no additional attributes are required.
  *
  * o REMOVE:
  *   Sent by an application to remove a specific DOI mapping table from the
- *   CIPSO V4 system.  The kernel should ACK this message.
+ *   CIPSO V4 system.
  *
- *   +---------------+
- *   | DOI (32 bits) |
- *   +---------------+
+ *   Required attributes:
  *
- *     DOI:          the DOI value
+ *     NLBL_CIPSOV4_A_DOI
  *
  * o LIST:
- *   Sent by an application to list the details of a DOI definition.  The
- *   kernel should send an ACK on error or a response as indicated below.  The
- *   application generated message format is shown below.
+ *   Sent by an application to list the details of a DOI definition.  On
+ *   success the kernel should send a response using the following format.
  *
- *   +---------------+
- *   | DOI (32 bits) |
- *   +---------------+
+ *   Required attributes:
  *
- *     DOI:          the DOI value
+ *     NLBL_CIPSOV4_A_DOI
  *
  *   The valid response message format depends on the type of the DOI mapping,
- *   the known formats are shown below.
- *
- *   +--------------------+
- *   | map type (32 bits) | ...
- *   +--------------------+
- *
- *     map type:       the DOI mapping table type (defined in the cipso_ipv4.h
- *                     header as CIPSO_V4_MAP_*)
- *
- *   (map type == CIPSO_V4_MAP_STD)
- *
- *   +----------------+------------------+----------------------+
- *   | tags (32 bits) | levels (32 bits) | categories (32 bits) | ...
- *   +----------------+------------------+----------------------+
+ *   the defined formats are shown below.
  *
- *   +-----------------+
- *   | tag #X (8 bits) | ... repeated
- *   +-----------------+
+ *   Required attributes:
  *
- *   +--------------------------+-------------------------+
- *   | local level #X (32 bits) | CIPSO level #X (8 bits) | ... repeated
- *   +--------------------------+-------------------------+
+ *     NLBL_CIPSOV4_A_MTYPE
+ *     NLBL_CIPSOV4_A_TAGLST
  *
- *   +-----------------------------+-----------------------------+
- *   | local category #X (32 bits) | CIPSO category #X (16 bits) | ... repeated
- *   +-----------------------------+-----------------------------+
+ *   If using CIPSO_V4_MAP_STD the following attributes are required:
  *
- *     tags:           the number of CIPSO tag types
- *     levels:         the number of level mappings
- *     categories:     the number of category mappings
- *     tag:            the tag number, tags listed first are given higher
- *                     priority when sending packets
- *     local level:    the local part of a level mapping
- *     CIPSO level:    the remote/CIPSO part of a level mapping
- *     local category: the local part of a category mapping
- *     CIPSO category: the remote/CIPSO part of a category mapping
+ *     NLBL_CIPSOV4_A_MLSLVLLST
+ *     NLBL_CIPSOV4_A_MLSCATLST
  *
- *   (map type == CIPSO_V4_MAP_PASS)
- *
- *   +----------------+
- *   | tags (32 bits) | ...
- *   +----------------+
- *
- *   +-----------------+
- *   | tag #X (8 bits) | ... repeated
- *   +-----------------+
- *
- *     tags:           the number of CIPSO tag types
- *     tag:            the tag number, tags listed first are given higher
- *                     priority when sending packets
+ *   If using CIPSO_V4_MAP_PASS no additional attributes are required.
  *
  * o LISTALL:
  *   This message is sent by an application to list the valid DOIs on the
- *   system.  There is no payload and the kernel should respond with an ACK
- *   or the following message.
- *
- *   +---------------------+------------------+-----------------------+
- *   | DOI count (32 bits) | DOI #X (32 bits) | map type #X (32 bits) |
- *   +---------------------+------------------+-----------------------+
+ *   system.  When sent by an application there is no payload and the
+ *   NLM_F_DUMP flag should be set.  The kernel should respond with a series of
+ *   the following messages.
  *
- *   +-----------------------+
- *   | map type #X (32 bits) | ...
- *   +-----------------------+
+ *   Required attributes:
  *
- *     DOI count:      the number of DOIs
- *     DOI:            the DOI value
- *     map type:       the DOI mapping table type (defined in the cipso_ipv4.h
- *                     header as CIPSO_V4_MAP_*)
+ *    NLBL_CIPSOV4_A_DOI
+ *    NLBL_CIPSOV4_A_MTYPE
  *
  */
 
 /* NetLabel CIPSOv4 commands */
 enum {
 	NLBL_CIPSOV4_C_UNSPEC,
-	NLBL_CIPSOV4_C_ACK,
 	NLBL_CIPSOV4_C_ADD,
 	NLBL_CIPSOV4_C_REMOVE,
 	NLBL_CIPSOV4_C_LIST,
@@ -211,6 +107,59 @@ enum {
 };
 #define NLBL_CIPSOV4_C_MAX (__NLBL_CIPSOV4_C_MAX - 1)
 
+/* NetLabel CIPSOv4 attributes */
+enum {
+	NLBL_CIPSOV4_A_UNSPEC,
+	NLBL_CIPSOV4_A_DOI,
+	/* (NLA_U32)
+	 * the DOI value */
+	NLBL_CIPSOV4_A_MTYPE,
+	/* (NLA_U32)
+	 * the mapping table type (defined in the cipso_ipv4.h header as
+	 * CIPSO_V4_MAP_*) */
+	NLBL_CIPSOV4_A_TAG,
+	/* (NLA_U8)
+	 * a CIPSO tag type, meant to be used within a NLBL_CIPSOV4_A_TAGLST
+	 * attribute */
+	NLBL_CIPSOV4_A_TAGLST,
+	/* (NLA_NESTED)
+	 * the CIPSO tag list for the DOI, there must be at least one
+	 * NLBL_CIPSOV4_A_TAG attribute, tags listed first are given higher
+	 * priorirty when sending packets */
+	NLBL_CIPSOV4_A_MLSLVLLOC,
+	/* (NLA_U32)
+	 * the local MLS sensitivity level */
+	NLBL_CIPSOV4_A_MLSLVLREM,
+	/* (NLA_U32)
+	 * the remote MLS sensitivity level */
+	NLBL_CIPSOV4_A_MLSLVL,
+	/* (NLA_NESTED)
+	 * a MLS sensitivity level mapping, must contain only one attribute of
+	 * each of the following types: NLBL_CIPSOV4_A_MLSLVLLOC and
+	 * NLBL_CIPSOV4_A_MLSLVLREM */
+	NLBL_CIPSOV4_A_MLSLVLLST,
+	/* (NLA_NESTED)
+	 * the CIPSO level mappings, there must be at least one
+	 * NLBL_CIPSOV4_A_MLSLVL attribute */
+	NLBL_CIPSOV4_A_MLSCATLOC,
+	/* (NLA_U32)
+	 * the local MLS category */
+	NLBL_CIPSOV4_A_MLSCATREM,
+	/* (NLA_U32)
+	 * the remote MLS category */
+	NLBL_CIPSOV4_A_MLSCAT,
+	/* (NLA_NESTED)
+	 * a MLS category mapping, must contain only one attribute of each of
+	 * the following types: NLBL_CIPSOV4_A_MLSCATLOC and
+	 * NLBL_CIPSOV4_A_MLSCATREM */
+	NLBL_CIPSOV4_A_MLSCATLST,
+	/* (NLA_NESTED)
+	 * the CIPSO category mappings, there must be at least one
+	 * NLBL_CIPSOV4_A_MLSCAT attribute */
+	__NLBL_CIPSOV4_A_MAX,
+};
+#define NLBL_CIPSOV4_A_MAX (__NLBL_CIPSOV4_A_MAX - 1)
+
 /* NetLabel protocol functions */
 int netlbl_cipsov4_genl_init(void);
 
diff --git a/net/netlabel/netlabel_mgmt.c b/net/netlabel/netlabel_mgmt.c
index 85bc11a1fc46..8626c9f678eb 100644
--- a/net/netlabel/netlabel_mgmt.c
+++ b/net/netlabel/netlabel_mgmt.c
@@ -42,15 +42,29 @@
 #include "netlabel_user.h"
 #include "netlabel_mgmt.h"
 
+/* Argument struct for netlbl_domhsh_walk() */
+struct netlbl_domhsh_walk_arg {
+	struct netlink_callback *nl_cb;
+	struct sk_buff *skb;
+	u32 seq;
+};
+
 /* NetLabel Generic NETLINK CIPSOv4 family */
 static struct genl_family netlbl_mgmt_gnl_family = {
 	.id = GENL_ID_GENERATE,
 	.hdrsize = 0,
 	.name = NETLBL_NLTYPE_MGMT_NAME,
 	.version = NETLBL_PROTO_VERSION,
-	.maxattr = 0,
+	.maxattr = NLBL_MGMT_A_MAX,
 };
 
+/* NetLabel Netlink attribute policy */
+static struct nla_policy netlbl_mgmt_genl_policy[NLBL_MGMT_A_MAX + 1] = {
+	[NLBL_MGMT_A_DOMAIN] = { .type = NLA_NUL_STRING },
+	[NLBL_MGMT_A_PROTOCOL] = { .type = NLA_U32 },
+	[NLBL_MGMT_A_VERSION] = { .type = NLA_U32 },
+	[NLBL_MGMT_A_CV4DOI] = { .type = NLA_U32 },
+};
 
 /*
  * NetLabel Command Handlers
@@ -70,97 +84,62 @@ static struct genl_family netlbl_mgmt_gnl_family = {
 static int netlbl_mgmt_add(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -EINVAL;
-	struct nlattr *msg_ptr = netlbl_netlink_payload_data(skb);
-	int msg_len = netlbl_netlink_payload_len(skb);
-	u32 count;
 	struct netlbl_dom_map *entry = NULL;
-	u32 iter;
+	size_t tmp_size;
 	u32 tmp_val;
-	int tmp_size;
 
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
+	if (!info->attrs[NLBL_MGMT_A_DOMAIN] ||
+	    !info->attrs[NLBL_MGMT_A_PROTOCOL])
 		goto add_failure;
 
-	if (msg_len < NETLBL_LEN_U32)
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (entry == NULL) {
+		ret_val = -ENOMEM;
+		goto add_failure;
+	}
+	tmp_size = nla_len(info->attrs[NLBL_MGMT_A_DOMAIN]);
+	entry->domain = kmalloc(tmp_size, GFP_KERNEL);
+	if (entry->domain == NULL) {
+		ret_val = -ENOMEM;
 		goto add_failure;
-	count = netlbl_getinc_u32(&msg_ptr, &msg_len);
+	}
+	entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]);
+	nla_strlcpy(entry->domain, info->attrs[NLBL_MGMT_A_DOMAIN], tmp_size);
 
-	for (iter = 0; iter < count && msg_len > 0; iter++, entry = NULL) {
-		if (msg_len <= 0) {
-			ret_val = -EINVAL;
-			goto add_failure;
-		}
-		entry = kzalloc(sizeof(*entry), GFP_KERNEL);
-		if (entry == NULL) {
-			ret_val = -ENOMEM;
-			goto add_failure;
-		}
-		tmp_size = nla_len(msg_ptr);
-		if (tmp_size <= 0 || tmp_size > msg_len) {
-			ret_val = -EINVAL;
-			goto add_failure;
-		}
-		entry->domain = kmalloc(tmp_size, GFP_KERNEL);
-		if (entry->domain == NULL) {
-			ret_val = -ENOMEM;
+	switch (entry->type) {
+	case NETLBL_NLTYPE_UNLABELED:
+		ret_val = netlbl_domhsh_add(entry);
+		break;
+	case NETLBL_NLTYPE_CIPSOV4:
+		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
 			goto add_failure;
-		}
-		nla_strlcpy(entry->domain, msg_ptr, tmp_size);
-		entry->domain[tmp_size - 1] = '\0';
-		msg_ptr = nla_next(msg_ptr, &msg_len);
 
-		if (msg_len < NETLBL_LEN_U32) {
-			ret_val = -EINVAL;
-			goto add_failure;
-		}
-		tmp_val = netlbl_getinc_u32(&msg_ptr, &msg_len);
-		entry->type = tmp_val;
-		switch (tmp_val) {
-		case NETLBL_NLTYPE_UNLABELED:
-			ret_val = netlbl_domhsh_add(entry);
-			break;
-		case NETLBL_NLTYPE_CIPSOV4:
-			if (msg_len < NETLBL_LEN_U32) {
-				ret_val = -EINVAL;
-				goto add_failure;
-			}
-			tmp_val = netlbl_getinc_u32(&msg_ptr, &msg_len);
-			/* We should be holding a rcu_read_lock() here
-			 * while we hold the result but since the entry
-			 * will always be deleted when the CIPSO DOI
-			 * is deleted we aren't going to keep the lock. */
-			rcu_read_lock();
-			entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
-			if (entry->type_def.cipsov4 == NULL) {
-				rcu_read_unlock();
-				ret_val = -EINVAL;
-				goto add_failure;
-			}
-			ret_val = netlbl_domhsh_add(entry);
+		tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]);
+		/* We should be holding a rcu_read_lock() here while we hold
+		 * the result but since the entry will always be deleted when
+		 * the CIPSO DOI is deleted we aren't going to keep the
+		 * lock. */
+		rcu_read_lock();
+		entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
+		if (entry->type_def.cipsov4 == NULL) {
 			rcu_read_unlock();
-			break;
-		default:
-			ret_val = -EINVAL;
-		}
-		if (ret_val != 0)
 			goto add_failure;
+		}
+		ret_val = netlbl_domhsh_add(entry);
+		rcu_read_unlock();
+		break;
+	default:
+		goto add_failure;
 	}
+	if (ret_val != 0)
+		goto add_failure;
 
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				NETLBL_E_OK);
 	return 0;
 
 add_failure:
 	if (entry)
 		kfree(entry->domain);
 	kfree(entry);
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
 	return ret_val;
 }
 
@@ -176,87 +155,98 @@ add_failure:
  */
 static int netlbl_mgmt_remove(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val = -EINVAL;
-	struct nlattr *msg_ptr = netlbl_netlink_payload_data(skb);
-	int msg_len = netlbl_netlink_payload_len(skb);
-	u32 count;
-	u32 iter;
-	int tmp_size;
-	unsigned char *domain;
-
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		goto remove_return;
+	char *domain;
 
-	if (msg_len < NETLBL_LEN_U32)
-		goto remove_return;
-	count = netlbl_getinc_u32(&msg_ptr, &msg_len);
+	if (!info->attrs[NLBL_MGMT_A_DOMAIN])
+		return -EINVAL;
 
-	for (iter = 0; iter < count && msg_len > 0; iter++) {
-		if (msg_len <= 0) {
-			ret_val = -EINVAL;
-			goto remove_return;
-		}
-		tmp_size = nla_len(msg_ptr);
-		domain = nla_data(msg_ptr);
-		if (tmp_size <= 0 || tmp_size > msg_len ||
-		    domain[tmp_size - 1] != '\0') {
-			ret_val = -EINVAL;
-			goto remove_return;
-		}
-		ret_val = netlbl_domhsh_remove(domain);
+	domain = nla_data(info->attrs[NLBL_MGMT_A_DOMAIN]);
+	return netlbl_domhsh_remove(domain);
+}
+
+/**
+ * netlbl_mgmt_listall_cb - netlbl_domhsh_walk() callback for LISTALL
+ * @entry: the domain mapping hash table entry
+ * @arg: the netlbl_domhsh_walk_arg structure
+ *
+ * Description:
+ * This function is designed to be used as a callback to the
+ * netlbl_domhsh_walk() function for use in generating a response for a LISTALL
+ * message.  Returns the size of the message on success, negative values on
+ * failure.
+ *
+ */
+static int netlbl_mgmt_listall_cb(struct netlbl_dom_map *entry, void *arg)
+{
+	int ret_val = -ENOMEM;
+	struct netlbl_domhsh_walk_arg *cb_arg = arg;
+	void *data;
+
+	data = netlbl_netlink_hdr_put(cb_arg->skb,
+				      NETLINK_CB(cb_arg->nl_cb->skb).pid,
+				      cb_arg->seq,
+				      netlbl_mgmt_gnl_family.id,
+				      NLM_F_MULTI,
+				      NLBL_MGMT_C_LISTALL);
+	if (data == NULL)
+		goto listall_cb_failure;
+
+	ret_val = nla_put_string(cb_arg->skb,
+				 NLBL_MGMT_A_DOMAIN,
+				 entry->domain);
+	if (ret_val != 0)
+		goto listall_cb_failure;
+	ret_val = nla_put_u32(cb_arg->skb, NLBL_MGMT_A_PROTOCOL, entry->type);
+	if (ret_val != 0)
+		goto listall_cb_failure;
+	switch (entry->type) {
+	case NETLBL_NLTYPE_CIPSOV4:
+		ret_val = nla_put_u32(cb_arg->skb,
+				      NLBL_MGMT_A_CV4DOI,
+				      entry->type_def.cipsov4->doi);
 		if (ret_val != 0)
-			goto remove_return;
-		msg_ptr = nla_next(msg_ptr, &msg_len);
+			goto listall_cb_failure;
+		break;
 	}
 
-	ret_val = 0;
+	cb_arg->seq++;
+	return genlmsg_end(cb_arg->skb, data);
 
-remove_return:
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
+listall_cb_failure:
+	genlmsg_cancel(cb_arg->skb, data);
 	return ret_val;
 }
 
 /**
- * netlbl_mgmt_list - Handle a LIST message
+ * netlbl_mgmt_listall - Handle a LISTALL message
  * @skb: the NETLINK buffer
- * @info: the Generic NETLINK info block
+ * @cb: the NETLINK callback
  *
  * Description:
- * Process a user generated LIST message and dumps the domain hash table in a
- * form suitable for use in a kernel generated LIST message.  Returns zero on
- * success, negative values on failure.
+ * Process a user generated LISTALL message and dumps the domain hash table in
+ * a form suitable for use in a kernel generated LISTALL message.  Returns zero
+ * on success, negative values on failure.
  *
  */
-static int netlbl_mgmt_list(struct sk_buff *skb, struct genl_info *info)
+static int netlbl_mgmt_listall(struct sk_buff *skb,
+			       struct netlink_callback *cb)
 {
-	int ret_val = -ENOMEM;
-	struct sk_buff *ans_skb;
-
-	ans_skb = netlbl_domhsh_dump(NLMSG_SPACE(GENL_HDRLEN));
-	if (ans_skb == NULL)
-		goto list_failure;
-	netlbl_netlink_hdr_push(ans_skb,
-				info->snd_pid,
-				0,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_LIST);
-
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
-	if (ret_val != 0)
-		goto list_failure;
-
-	return 0;
-
-list_failure:
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
-	return ret_val;
+	struct netlbl_domhsh_walk_arg cb_arg;
+	u32 skip_bkt = cb->args[0];
+	u32 skip_chain = cb->args[1];
+
+	cb_arg.nl_cb = cb;
+	cb_arg.skb = skb;
+	cb_arg.seq = cb->nlh->nlmsg_seq;
+
+	netlbl_domhsh_walk(&skip_bkt,
+			   &skip_chain,
+			   netlbl_mgmt_listall_cb,
+			   &cb_arg);
+
+	cb->args[0] = skip_bkt;
+	cb->args[1] = skip_chain;
+	return skb->len;
 }
 
 /**
@@ -272,68 +262,51 @@ list_failure:
 static int netlbl_mgmt_adddef(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -EINVAL;
-	struct nlattr *msg_ptr = netlbl_netlink_payload_data(skb);
-	int msg_len = netlbl_netlink_payload_len(skb);
 	struct netlbl_dom_map *entry = NULL;
 	u32 tmp_val;
 
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		goto adddef_failure;
-
-	if (msg_len < NETLBL_LEN_U32)
+	if (!info->attrs[NLBL_MGMT_A_PROTOCOL])
 		goto adddef_failure;
-	tmp_val = netlbl_getinc_u32(&msg_ptr, &msg_len);
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (entry == NULL) {
 		ret_val = -ENOMEM;
 		goto adddef_failure;
 	}
+	entry->type = nla_get_u32(info->attrs[NLBL_MGMT_A_PROTOCOL]);
 
-	entry->type = tmp_val;
 	switch (entry->type) {
 	case NETLBL_NLTYPE_UNLABELED:
 		ret_val = netlbl_domhsh_add_default(entry);
 		break;
 	case NETLBL_NLTYPE_CIPSOV4:
-		if (msg_len < NETLBL_LEN_U32) {
-			ret_val = -EINVAL;
+		if (!info->attrs[NLBL_MGMT_A_CV4DOI])
 			goto adddef_failure;
-		}
-		tmp_val = netlbl_getinc_u32(&msg_ptr, &msg_len);
-		/* We should be holding a rcu_read_lock here while we
-		 * hold the result but since the entry will always be
-		 * deleted when the CIPSO DOI is deleted we are going
-		 * to skip the lock. */
+
+		tmp_val = nla_get_u32(info->attrs[NLBL_MGMT_A_CV4DOI]);
+		/* We should be holding a rcu_read_lock() here while we hold
+		 * the result but since the entry will always be deleted when
+		 * the CIPSO DOI is deleted we aren't going to keep the
+		 * lock. */
 		rcu_read_lock();
 		entry->type_def.cipsov4 = cipso_v4_doi_getdef(tmp_val);
 		if (entry->type_def.cipsov4 == NULL) {
 			rcu_read_unlock();
-			ret_val = -EINVAL;
 			goto adddef_failure;
 		}
 		ret_val = netlbl_domhsh_add_default(entry);
 		rcu_read_unlock();
 		break;
 	default:
-		ret_val = -EINVAL;
+		goto adddef_failure;
 	}
 	if (ret_val != 0)
 		goto adddef_failure;
 
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				NETLBL_E_OK);
 	return 0;
 
 adddef_failure:
 	kfree(entry);
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
 	return ret_val;
 }
 
@@ -349,20 +322,7 @@ adddef_failure:
  */
 static int netlbl_mgmt_removedef(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val;
-
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		goto removedef_return;
-
-	ret_val = netlbl_domhsh_remove_default();
-
-removedef_return:
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
-	return ret_val;
+	return netlbl_domhsh_remove_default();
 }
 
 /**
@@ -379,87 +339,130 @@ removedef_return:
 static int netlbl_mgmt_listdef(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -ENOMEM;
-	struct sk_buff *ans_skb;
+	struct sk_buff *ans_skb = NULL;
+	void *data;
+	struct netlbl_dom_map *entry;
 
-	ans_skb = netlbl_domhsh_dump_default(NLMSG_SPACE(GENL_HDRLEN));
+	ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (ans_skb == NULL)
+		return -ENOMEM;
+	data = netlbl_netlink_hdr_put(ans_skb,
+				      info->snd_pid,
+				      info->snd_seq,
+				      netlbl_mgmt_gnl_family.id,
+				      0,
+				      NLBL_MGMT_C_LISTDEF);
+	if (data == NULL)
 		goto listdef_failure;
-	netlbl_netlink_hdr_push(ans_skb,
-				info->snd_pid,
-				0,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_LISTDEF);
 
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+	rcu_read_lock();
+	entry = netlbl_domhsh_getentry(NULL);
+	if (entry == NULL) {
+		ret_val = -ENOENT;
+		goto listdef_failure_lock;
+	}
+	ret_val = nla_put_u32(ans_skb, NLBL_MGMT_A_PROTOCOL, entry->type);
 	if (ret_val != 0)
-		goto listdef_failure;
+		goto listdef_failure_lock;
+	switch (entry->type) {
+	case NETLBL_NLTYPE_CIPSOV4:
+		ret_val = nla_put_u32(ans_skb,
+				      NLBL_MGMT_A_CV4DOI,
+				      entry->type_def.cipsov4->doi);
+		if (ret_val != 0)
+			goto listdef_failure_lock;
+		break;
+	}
+	rcu_read_unlock();
 
+	genlmsg_end(ans_skb, data);
+
+	ret_val = genlmsg_unicast(ans_skb, info->snd_pid);
+	if (ret_val != 0)
+		goto listdef_failure;
 	return 0;
 
+listdef_failure_lock:
+	rcu_read_unlock();
 listdef_failure:
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
+	kfree_skb(ans_skb);
 	return ret_val;
 }
 
 /**
- * netlbl_mgmt_modules - Handle a MODULES message
- * @skb: the NETLINK buffer
- * @info: the Generic NETLINK info block
+ * netlbl_mgmt_protocols_cb - Write an individual PROTOCOL message response
+ * @skb: the skb to write to
+ * @seq: the NETLINK sequence number
+ * @cb: the NETLINK callback
+ * @protocol: the NetLabel protocol to use in the message
  *
  * Description:
- * Process a user generated MODULES message and respond accordingly.
+ * This function is to be used in conjunction with netlbl_mgmt_protocols() to
+ * answer a application's PROTOCOLS message.  Returns the size of the message
+ * on success, negative values on failure.
  *
  */
-static int netlbl_mgmt_modules(struct sk_buff *skb, struct genl_info *info)
+static int netlbl_mgmt_protocols_cb(struct sk_buff *skb,
+				    struct netlink_callback *cb,
+				    u32 protocol)
 {
 	int ret_val = -ENOMEM;
-	size_t data_size;
-	u32 mod_count;
-	struct sk_buff *ans_skb = NULL;
-
-	/* unlabeled + cipsov4 */
-	mod_count = 2;
-
-	data_size = GENL_HDRLEN + NETLBL_LEN_U32 + mod_count * NETLBL_LEN_U32;
-	ans_skb = netlbl_netlink_alloc_skb(0, data_size, GFP_KERNEL);
-	if (ans_skb == NULL)
-		goto modules_failure;
-
-	if (netlbl_netlink_hdr_put(ans_skb,
-				   info->snd_pid,
-				   0,
-				   netlbl_mgmt_gnl_family.id,
-				   NLBL_MGMT_C_MODULES) == NULL)
-		goto modules_failure;
-
-	ret_val = nla_put_u32(ans_skb, NLA_U32, mod_count);
-	if (ret_val != 0)
-		goto modules_failure;
-	ret_val = nla_put_u32(ans_skb, NLA_U32, NETLBL_NLTYPE_UNLABELED);
+	void *data;
+
+	data = netlbl_netlink_hdr_put(skb,
+				      NETLINK_CB(cb->skb).pid,
+				      cb->nlh->nlmsg_seq,
+				      netlbl_mgmt_gnl_family.id,
+				      NLM_F_MULTI,
+				      NLBL_MGMT_C_PROTOCOLS);
+	if (data == NULL)
+		goto protocols_cb_failure;
+
+	ret_val = nla_put_u32(skb, NLBL_MGMT_A_PROTOCOL, protocol);
 	if (ret_val != 0)
-		goto modules_failure;
-	ret_val = nla_put_u32(ans_skb, NLA_U32, NETLBL_NLTYPE_CIPSOV4);
-	if (ret_val != 0)
-		goto modules_failure;
-
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
-	if (ret_val != 0)
-		goto modules_failure;
+		goto protocols_cb_failure;
 
-	return 0;
+	return genlmsg_end(skb, data);
 
-modules_failure:
-	kfree_skb(ans_skb);
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
+protocols_cb_failure:
+	genlmsg_cancel(skb, data);
 	return ret_val;
 }
 
+/**
+ * netlbl_mgmt_protocols - Handle a PROTOCOLS message
+ * @skb: the NETLINK buffer
+ * @cb: the NETLINK callback
+ *
+ * Description:
+ * Process a user generated PROTOCOLS message and respond accordingly.
+ *
+ */
+static int netlbl_mgmt_protocols(struct sk_buff *skb,
+				 struct netlink_callback *cb)
+{
+	u32 protos_sent = cb->args[0];
+
+	if (protos_sent == 0) {
+		if (netlbl_mgmt_protocols_cb(skb,
+					     cb,
+					     NETLBL_NLTYPE_UNLABELED) < 0)
+			goto protocols_return;
+		protos_sent++;
+	}
+	if (protos_sent == 1) {
+		if (netlbl_mgmt_protocols_cb(skb,
+					     cb,
+					     NETLBL_NLTYPE_CIPSOV4) < 0)
+			goto protocols_return;
+		protos_sent++;
+	}
+
+protocols_return:
+	cb->args[0] = protos_sent;
+	return skb->len;
+}
+
 /**
  * netlbl_mgmt_version - Handle a VERSION message
  * @skb: the NETLINK buffer
@@ -474,35 +477,35 @@ static int netlbl_mgmt_version(struct sk_buff *skb, struct genl_info *info)
 {
 	int ret_val = -ENOMEM;
 	struct sk_buff *ans_skb = NULL;
+	void *data;
 
-	ans_skb = netlbl_netlink_alloc_skb(0,
-					   GENL_HDRLEN + NETLBL_LEN_U32,
-					   GFP_KERNEL);
+	ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (ans_skb == NULL)
-		goto version_failure;
-	if (netlbl_netlink_hdr_put(ans_skb,
-				   info->snd_pid,
-				   0,
-				   netlbl_mgmt_gnl_family.id,
-				   NLBL_MGMT_C_VERSION) == NULL)
+		return -ENOMEM;
+	data = netlbl_netlink_hdr_put(ans_skb,
+				      info->snd_pid,
+				      info->snd_seq,
+				      netlbl_mgmt_gnl_family.id,
+				      0,
+				      NLBL_MGMT_C_VERSION);
+	if (data == NULL)
 		goto version_failure;
 
-	ret_val = nla_put_u32(ans_skb, NLA_U32, NETLBL_PROTO_VERSION);
+	ret_val = nla_put_u32(ans_skb,
+			      NLBL_MGMT_A_VERSION,
+			      NETLBL_PROTO_VERSION);
 	if (ret_val != 0)
 		goto version_failure;
 
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+	genlmsg_end(ans_skb, data);
+
+	ret_val = genlmsg_unicast(ans_skb, info->snd_pid);
 	if (ret_val != 0)
 		goto version_failure;
-
 	return 0;
 
 version_failure:
 	kfree_skb(ans_skb);
-	netlbl_netlink_send_ack(info,
-				netlbl_mgmt_gnl_family.id,
-				NLBL_MGMT_C_ACK,
-				-ret_val);
 	return ret_val;
 }
 
@@ -513,35 +516,40 @@ version_failure:
 
 static struct genl_ops netlbl_mgmt_genl_c_add = {
 	.cmd = NLBL_MGMT_C_ADD,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_add,
 	.dumpit = NULL,
 };
 
 static struct genl_ops netlbl_mgmt_genl_c_remove = {
 	.cmd = NLBL_MGMT_C_REMOVE,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_remove,
 	.dumpit = NULL,
 };
 
-static struct genl_ops netlbl_mgmt_genl_c_list = {
-	.cmd = NLBL_MGMT_C_LIST,
+static struct genl_ops netlbl_mgmt_genl_c_listall = {
+	.cmd = NLBL_MGMT_C_LISTALL,
 	.flags = 0,
-	.doit = netlbl_mgmt_list,
-	.dumpit = NULL,
+	.policy = netlbl_mgmt_genl_policy,
+	.doit = NULL,
+	.dumpit = netlbl_mgmt_listall,
 };
 
 static struct genl_ops netlbl_mgmt_genl_c_adddef = {
 	.cmd = NLBL_MGMT_C_ADDDEF,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_adddef,
 	.dumpit = NULL,
 };
 
 static struct genl_ops netlbl_mgmt_genl_c_removedef = {
 	.cmd = NLBL_MGMT_C_REMOVEDEF,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_removedef,
 	.dumpit = NULL,
 };
@@ -549,20 +557,23 @@ static struct genl_ops netlbl_mgmt_genl_c_removedef = {
 static struct genl_ops netlbl_mgmt_genl_c_listdef = {
 	.cmd = NLBL_MGMT_C_LISTDEF,
 	.flags = 0,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_listdef,
 	.dumpit = NULL,
 };
 
-static struct genl_ops netlbl_mgmt_genl_c_modules = {
-	.cmd = NLBL_MGMT_C_MODULES,
+static struct genl_ops netlbl_mgmt_genl_c_protocols = {
+	.cmd = NLBL_MGMT_C_PROTOCOLS,
 	.flags = 0,
-	.doit = netlbl_mgmt_modules,
-	.dumpit = NULL,
+	.policy = netlbl_mgmt_genl_policy,
+	.doit = NULL,
+	.dumpit = netlbl_mgmt_protocols,
 };
 
 static struct genl_ops netlbl_mgmt_genl_c_version = {
 	.cmd = NLBL_MGMT_C_VERSION,
 	.flags = 0,
+	.policy = netlbl_mgmt_genl_policy,
 	.doit = netlbl_mgmt_version,
 	.dumpit = NULL,
 };
@@ -596,7 +607,7 @@ int netlbl_mgmt_genl_init(void)
 	if (ret_val != 0)
 		return ret_val;
 	ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
-				    &netlbl_mgmt_genl_c_list);
+				    &netlbl_mgmt_genl_c_listall);
 	if (ret_val != 0)
 		return ret_val;
 	ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
@@ -612,7 +623,7 @@ int netlbl_mgmt_genl_init(void)
 	if (ret_val != 0)
 		return ret_val;
 	ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
-				    &netlbl_mgmt_genl_c_modules);
+				    &netlbl_mgmt_genl_c_protocols);
 	if (ret_val != 0)
 		return ret_val;
 	ret_val = genl_register_ops(&netlbl_mgmt_gnl_family,
diff --git a/net/netlabel/netlabel_mgmt.h b/net/netlabel/netlabel_mgmt.h
index fd6c6acbfa08..3642d3bfc8eb 100644
--- a/net/netlabel/netlabel_mgmt.h
+++ b/net/netlabel/netlabel_mgmt.h
@@ -34,212 +34,137 @@
 #include <net/netlabel.h>
 
 /*
- * The following NetLabel payloads are supported by the management interface,
- * all of which are preceeded by the nlmsghdr struct.
- *
- * o ACK:
- *   Sent by the kernel in response to an applications message, applications
- *   should never send this message.
- *
- *   +----------------------+-----------------------+
- *   | seq number (32 bits) | return code (32 bits) |
- *   +----------------------+-----------------------+
- *
- *     seq number:  the sequence number of the original message, taken from the
- *                  nlmsghdr structure
- *     return code: return value, based on errno values
+ * The following NetLabel payloads are supported by the management interface.
  *
  * o ADD:
  *   Sent by an application to add a domain mapping to the NetLabel system.
- *   The kernel should respond with an ACK.
- *
- *   +-------------------+
- *   | domains (32 bits) | ...
- *   +-------------------+
- *
- *     domains: the number of domains in the message
- *
- *   +--------------------------+-------------------------+
- *   | domain string (variable) | protocol type (32 bits) | ...
- *   +--------------------------+-------------------------+
  *
- *   +-------------- ---- --- -- -
- *   | mapping data                ... repeated
- *   +-------------- ---- --- -- -
+ *   Required attributes:
  *
- *     domain string: the domain string, NULL terminated
- *     protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
- *     mapping data:  specific to the map type (see below)
+ *     NLBL_MGMT_A_DOMAIN
+ *     NLBL_MGMT_A_PROTOCOL
  *
- *   NETLBL_NLTYPE_UNLABELED
+ *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
  *
- *     No mapping data for this protocol type.
+ *     NLBL_MGMT_A_CV4DOI
  *
- *   NETLBL_NLTYPE_CIPSOV4
- *
- *   +---------------+
- *   | doi (32 bits) |
- *   +---------------+
- *
- *     doi:  the CIPSO DOI value
+ *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
  *
  * o REMOVE:
  *   Sent by an application to remove a domain mapping from the NetLabel
- *   system.  The kernel should ACK this message.
- *
- *   +-------------------+
- *   | domains (32 bits) | ...
- *   +-------------------+
+ *   system.
  *
- *     domains: the number of domains in the message
+ *   Required attributes:
  *
- *   +--------------------------+
- *   | domain string (variable) | ...
- *   +--------------------------+
+ *     NLBL_MGMT_A_DOMAIN
  *
- *     domain string: the domain string, NULL terminated
- *
- * o LIST:
+ * o LISTALL:
  *   This message can be sent either from an application or by the kernel in
- *   response to an application generated LIST message.  When sent by an
- *   application there is no payload.  The kernel should respond to a LIST
- *   message either with a LIST message on success or an ACK message on
- *   failure.
- *
- *   +-------------------+
- *   | domains (32 bits) | ...
- *   +-------------------+
- *
- *     domains: the number of domains in the message
+ *   response to an application generated LISTALL message.  When sent by an
+ *   application there is no payload and the NLM_F_DUMP flag should be set.
+ *   The kernel should respond with a series of the following messages.
  *
- *   +--------------------------+
- *   | domain string (variable) | ...
- *   +--------------------------+
+ *   Required attributes:
  *
- *   +-------------------------+-------------- ---- --- -- -
- *   | protocol type (32 bits) | mapping data                ... repeated
- *   +-------------------------+-------------- ---- --- -- -
+ *     NLBL_MGMT_A_DOMAIN
+ *     NLBL_MGMT_A_PROTOCOL
  *
- *     domain string: the domain string, NULL terminated
- *     protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
- *     mapping data:  specific to the map type (see below)
+ *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
  *
- *   NETLBL_NLTYPE_UNLABELED
+ *     NLBL_MGMT_A_CV4DOI
  *
- *     No mapping data for this protocol type.
- *
- *   NETLBL_NLTYPE_CIPSOV4
- *
- *   +----------------+---------------+
- *   | type (32 bits) | doi (32 bits) |
- *   +----------------+---------------+
- *
- *     type: the CIPSO mapping table type (defined in the cipso_ipv4.h header
- *           as CIPSO_V4_MAP_*)
- *     doi:  the CIPSO DOI value
+ *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
  *
  * o ADDDEF:
  *   Sent by an application to set the default domain mapping for the NetLabel
- *   system.  The kernel should respond with an ACK.
+ *   system.
  *
- *   +-------------------------+-------------- ---- --- -- -
- *   | protocol type (32 bits) | mapping data                ... repeated
- *   +-------------------------+-------------- ---- --- -- -
+ *   Required attributes:
  *
- *     protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
- *     mapping data:  specific to the map type (see below)
+ *     NLBL_MGMT_A_PROTOCOL
  *
- *   NETLBL_NLTYPE_UNLABELED
+ *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
  *
- *     No mapping data for this protocol type.
+ *     NLBL_MGMT_A_CV4DOI
  *
- *   NETLBL_NLTYPE_CIPSOV4
- *
- *   +---------------+
- *   | doi (32 bits) |
- *   +---------------+
- *
- *     doi:  the CIPSO DOI value
+ *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
  *
  * o REMOVEDEF:
  *   Sent by an application to remove the default domain mapping from the
- *   NetLabel system, there is no payload.  The kernel should ACK this message.
+ *   NetLabel system, there is no payload.
  *
  * o LISTDEF:
  *   This message can be sent either from an application or by the kernel in
  *   response to an application generated LISTDEF message.  When sent by an
- *   application there is no payload.  The kernel should respond to a
- *   LISTDEF message either with a LISTDEF message on success or an ACK message
- *   on failure.
- *
- *   +-------------------------+-------------- ---- --- -- -
- *   | protocol type (32 bits) | mapping data                ... repeated
- *   +-------------------------+-------------- ---- --- -- -
+ *   application there is no payload.  On success the kernel should send a
+ *   response using the following format.
  *
- *     protocol type: the protocol type (defined by NETLBL_NLTYPE_*)
- *     mapping data:  specific to the map type (see below)
+ *   Required attributes:
  *
- *   NETLBL_NLTYPE_UNLABELED
+ *     NLBL_MGMT_A_PROTOCOL
  *
- *     No mapping data for this protocol type.
+ *   If using NETLBL_NLTYPE_CIPSOV4 the following attributes are required:
  *
- *   NETLBL_NLTYPE_CIPSOV4
+ *     NLBL_MGMT_A_CV4DOI
  *
- *   +----------------+---------------+
- *   | type (32 bits) | doi (32 bits) |
- *   +----------------+---------------+
+ *   If using NETLBL_NLTYPE_UNLABELED no other attributes are required.
  *
- *     type: the CIPSO mapping table type (defined in the cipso_ipv4.h header
- *           as CIPSO_V4_MAP_*)
- *     doi:  the CIPSO DOI value
+ * o PROTOCOLS:
+ *   Sent by an application to request a list of configured NetLabel protocols
+ *   in the kernel.  When sent by an application there is no payload and the
+ *   NLM_F_DUMP flag should be set.  The kernel should respond with a series of
+ *   the following messages.
  *
- * o MODULES:
- *   Sent by an application to request a list of configured NetLabel modules
- *   in the kernel.  When sent by an application there is no payload.
+ *   Required attributes:
  *
- *   +-------------------+
- *   | modules (32 bits) | ...
- *   +-------------------+
- *
- *     modules: the number of modules in the message, if this is an application
- *              generated message and the value is zero then return a list of
- *              the configured modules
- *
- *   +------------------+
- *   | module (32 bits) | ... repeated
- *   +------------------+
- *
- *     module: the module number as defined by NETLBL_NLTYPE_*
+ *     NLBL_MGMT_A_PROTOCOL
  *
  * o VERSION:
- *   Sent by an application to request the NetLabel version string.  When sent
- *   by an application there is no payload.  This message type is also used by
- *   the kernel to respond to an VERSION request.
+ *   Sent by an application to request the NetLabel version.  When sent by an
+ *   application there is no payload.  This message type is also used by the
+ *   kernel to respond to an VERSION request.
  *
- *   +-------------------+
- *   | version (32 bits) |
- *   +-------------------+
+ *   Required attributes:
  *
- *     version: the protocol version number
+ *     NLBL_MGMT_A_VERSION
  *
  */
 
 /* NetLabel Management commands */
 enum {
 	NLBL_MGMT_C_UNSPEC,
-	NLBL_MGMT_C_ACK,
 	NLBL_MGMT_C_ADD,
 	NLBL_MGMT_C_REMOVE,
-	NLBL_MGMT_C_LIST,
+	NLBL_MGMT_C_LISTALL,
 	NLBL_MGMT_C_ADDDEF,
 	NLBL_MGMT_C_REMOVEDEF,
 	NLBL_MGMT_C_LISTDEF,
-	NLBL_MGMT_C_MODULES,
+	NLBL_MGMT_C_PROTOCOLS,
 	NLBL_MGMT_C_VERSION,
 	__NLBL_MGMT_C_MAX,
 };
 #define NLBL_MGMT_C_MAX (__NLBL_MGMT_C_MAX - 1)
 
+/* NetLabel Management attributes */
+enum {
+	NLBL_MGMT_A_UNSPEC,
+	NLBL_MGMT_A_DOMAIN,
+	/* (NLA_NUL_STRING)
+	 * the NULL terminated LSM domain string */
+	NLBL_MGMT_A_PROTOCOL,
+	/* (NLA_U32)
+	 * the NetLabel protocol type (defined by NETLBL_NLTYPE_*) */
+	NLBL_MGMT_A_VERSION,
+	/* (NLA_U32)
+	 * the NetLabel protocol version number (defined by
+	 * NETLBL_PROTO_VERSION) */
+	NLBL_MGMT_A_CV4DOI,
+	/* (NLA_U32)
+	 * the CIPSOv4 DOI value */
+	__NLBL_MGMT_A_MAX,
+};
+#define NLBL_MGMT_A_MAX (__NLBL_MGMT_A_MAX - 1)
+
 /* NetLabel protocol functions */
 int netlbl_mgmt_genl_init(void);
 
diff --git a/net/netlabel/netlabel_unlabeled.c b/net/netlabel/netlabel_unlabeled.c
index 785f4960e0d3..440f5c4e1e2d 100644
--- a/net/netlabel/netlabel_unlabeled.c
+++ b/net/netlabel/netlabel_unlabeled.c
@@ -55,9 +55,13 @@ static struct genl_family netlbl_unlabel_gnl_family = {
 	.hdrsize = 0,
 	.name = NETLBL_NLTYPE_UNLABELED_NAME,
 	.version = NETLBL_PROTO_VERSION,
-	.maxattr = 0,
+	.maxattr = NLBL_UNLABEL_A_MAX,
 };
 
+/* NetLabel Netlink attribute policy */
+static struct nla_policy netlbl_unlabel_genl_policy[NLBL_UNLABEL_A_MAX + 1] = {
+	[NLBL_UNLABEL_A_ACPTFLG] = { .type = NLA_U8 },
+};
 
 /*
  * NetLabel Command Handlers
@@ -75,31 +79,18 @@ static struct genl_family netlbl_unlabel_gnl_family = {
  */
 static int netlbl_unlabel_accept(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val;
-	struct nlattr *data = netlbl_netlink_payload_data(skb);
-	u32 value;
-
-	ret_val = netlbl_netlink_cap_check(skb, CAP_NET_ADMIN);
-	if (ret_val != 0)
-		return ret_val;
+	int ret_val = -EINVAL;
+	u8 value;
 
-	if (netlbl_netlink_payload_len(skb) == NETLBL_LEN_U32) {
-		value = nla_get_u32(data);
+	if (info->attrs[NLBL_UNLABEL_A_ACPTFLG]) {
+		value = nla_get_u8(info->attrs[NLBL_UNLABEL_A_ACPTFLG]);
 		if (value == 1 || value == 0) {
 			atomic_set(&netlabel_unlabel_accept_flg, value);
-			netlbl_netlink_send_ack(info,
-						netlbl_unlabel_gnl_family.id,
-						NLBL_UNLABEL_C_ACK,
-						NETLBL_E_OK);
-			return 0;
+			ret_val = 0;
 		}
 	}
 
-	netlbl_netlink_send_ack(info,
-				netlbl_unlabel_gnl_family.id,
-				NLBL_UNLABEL_C_ACK,
-				EINVAL);
-	return -EINVAL;
+	return ret_val;
 }
 
 /**
@@ -114,39 +105,39 @@ static int netlbl_unlabel_accept(struct sk_buff *skb, struct genl_info *info)
  */
 static int netlbl_unlabel_list(struct sk_buff *skb, struct genl_info *info)
 {
-	int ret_val = -ENOMEM;
+	int ret_val = -EINVAL;
 	struct sk_buff *ans_skb;
+	void *data;
 
-	ans_skb = netlbl_netlink_alloc_skb(0,
-					   GENL_HDRLEN + NETLBL_LEN_U32,
-					   GFP_KERNEL);
+	ans_skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL);
 	if (ans_skb == NULL)
 		goto list_failure;
-
-	if (netlbl_netlink_hdr_put(ans_skb,
-				   info->snd_pid,
-				   0,
-				   netlbl_unlabel_gnl_family.id,
-				   NLBL_UNLABEL_C_LIST) == NULL)
+	data = netlbl_netlink_hdr_put(ans_skb,
+				      info->snd_pid,
+				      info->snd_seq,
+				      netlbl_unlabel_gnl_family.id,
+				      0,
+				      NLBL_UNLABEL_C_LIST);
+	if (data == NULL) {
+		ret_val = -ENOMEM;
 		goto list_failure;
+	}
 
-	ret_val = nla_put_u32(ans_skb,
-			      NLA_U32,
-			      atomic_read(&netlabel_unlabel_accept_flg));
+	ret_val = nla_put_u8(ans_skb,
+			     NLBL_UNLABEL_A_ACPTFLG,
+			     atomic_read(&netlabel_unlabel_accept_flg));
 	if (ret_val != 0)
 		goto list_failure;
 
-	ret_val = netlbl_netlink_snd(ans_skb, info->snd_pid);
+	genlmsg_end(ans_skb, data);
+
+	ret_val = genlmsg_unicast(ans_skb, info->snd_pid);
 	if (ret_val != 0)
 		goto list_failure;
-
 	return 0;
 
 list_failure:
-	netlbl_netlink_send_ack(info,
-				netlbl_unlabel_gnl_family.id,
-				NLBL_UNLABEL_C_ACK,
-				-ret_val);
+	kfree(ans_skb);
 	return ret_val;
 }
 
@@ -157,7 +148,8 @@ list_failure:
 
 static struct genl_ops netlbl_unlabel_genl_c_accept = {
 	.cmd = NLBL_UNLABEL_C_ACCEPT,
-	.flags = 0,
+	.flags = GENL_ADMIN_PERM,
+	.policy = netlbl_unlabel_genl_policy,
 	.doit = netlbl_unlabel_accept,
 	.dumpit = NULL,
 };
@@ -165,6 +157,7 @@ static struct genl_ops netlbl_unlabel_genl_c_accept = {
 static struct genl_ops netlbl_unlabel_genl_c_list = {
 	.cmd = NLBL_UNLABEL_C_LIST,
 	.flags = 0,
+	.policy = netlbl_unlabel_genl_policy,
 	.doit = netlbl_unlabel_list,
 	.dumpit = NULL,
 };
@@ -218,10 +211,8 @@ int netlbl_unlabel_genl_init(void)
  */
 int netlbl_unlabel_getattr(struct netlbl_lsm_secattr *secattr)
 {
-	if (atomic_read(&netlabel_unlabel_accept_flg) == 1) {
-		memset(secattr, 0, sizeof(*secattr));
-		return 0;
-	}
+	if (atomic_read(&netlabel_unlabel_accept_flg) == 1)
+		return netlbl_secattr_init(secattr);
 
 	return -ENOMSG;
 }
diff --git a/net/netlabel/netlabel_unlabeled.h b/net/netlabel/netlabel_unlabeled.h
index f300e54e14b6..c2917fbb42cf 100644
--- a/net/netlabel/netlabel_unlabeled.h
+++ b/net/netlabel/netlabel_unlabeled.h
@@ -36,56 +36,47 @@
 /*
  * The following NetLabel payloads are supported by the Unlabeled subsystem.
  *
- * o ACK:
- *   Sent by the kernel in response to an applications message, applications
- *   should never send this message.
- *
- *   +----------------------+-----------------------+
- *   | seq number (32 bits) | return code (32 bits) |
- *   +----------------------+-----------------------+
- *
- *     seq number:  the sequence number of the original message, taken from the
- *                  nlmsghdr structure
- *     return code: return value, based on errno values
- *
  * o ACCEPT
  *   This message is sent from an application to specify if the kernel should
  *   allow unlabled packets to pass if they do not match any of the static
  *   mappings defined in the unlabeled module.
  *
- *   +-----------------+
- *   | allow (32 bits) |
- *   +-----------------+
+ *   Required attributes:
  *
- *     allow: if true (1) then allow the packets to pass, if false (0) then
- *            reject the packets
+ *     NLBL_UNLABEL_A_ACPTFLG
  *
  * o LIST
  *   This message can be sent either from an application or by the kernel in
  *   response to an application generated LIST message.  When sent by an
  *   application there is no payload.  The kernel should respond to a LIST
- *   message either with a LIST message on success or an ACK message on
- *   failure.
+ *   message with a LIST message on success.
  *
- *   +-----------------------+
- *   | accept flag (32 bits) |
- *   +-----------------------+
+ *   Required attributes:
  *
- *     accept flag: if true (1) then unlabeled packets are allowed to pass,
- *                  if false (0) then unlabeled packets are rejected
+ *     NLBL_UNLABEL_A_ACPTFLG
  *
  */
 
 /* NetLabel Unlabeled commands */
 enum {
 	NLBL_UNLABEL_C_UNSPEC,
-	NLBL_UNLABEL_C_ACK,
 	NLBL_UNLABEL_C_ACCEPT,
 	NLBL_UNLABEL_C_LIST,
 	__NLBL_UNLABEL_C_MAX,
 };
 #define NLBL_UNLABEL_C_MAX (__NLBL_UNLABEL_C_MAX - 1)
 
+/* NetLabel Unlabeled attributes */
+enum {
+	NLBL_UNLABEL_A_UNSPEC,
+	NLBL_UNLABEL_A_ACPTFLG,
+	/* (NLA_U8)
+	 * if true then unlabeled packets are allowed to pass, else unlabeled
+	 * packets are rejected */
+	__NLBL_UNLABEL_A_MAX,
+};
+#define NLBL_UNLABEL_A_MAX (__NLBL_UNLABEL_A_MAX - 1)
+
 /* NetLabel protocol functions */
 int netlbl_unlabel_genl_init(void);
 
-- 
cgit v1.2.1


From 4cc6773508299377099aa30cf30e6a2196c5872d Mon Sep 17 00:00:00 2001
From: Paul Moore <paul.moore@hp.com>
Date: Mon, 25 Sep 2006 15:57:13 -0700
Subject: [NetLabel]: update docs with website information

Now that all of the supporting pieces of NetLabel have a home at SourceForge
update the Kconfig help text and add an entry to the MAINTAINERS file.

Signed-off-by: Paul Moore <paul.moore@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 MAINTAINERS          | 7 +++++++
 net/netlabel/Kconfig | 5 ++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 16af51ba44a5..23348c0d37bc 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2031,6 +2031,13 @@ L:	netfilter@lists.netfilter.org
 L:	netfilter-devel@lists.netfilter.org
 S:	Supported
 
+NETLABEL
+P:	Paul Moore
+M:	paul.moore@hp.com
+W:	http://netlabel.sf.net
+L:	netdev@vger.kernel.org
+S:	Supported
+
 NETROM NETWORK LAYER
 P:	Ralf Baechle
 M:	ralf@linux-mips.org
diff --git a/net/netlabel/Kconfig b/net/netlabel/Kconfig
index fe23cb7f1e87..9f7121ae13e9 100644
--- a/net/netlabel/Kconfig
+++ b/net/netlabel/Kconfig
@@ -9,6 +9,9 @@ config NETLABEL
 	---help---
 	  NetLabel provides support for explicit network packet labeling
 	  protocols such as CIPSO and RIPSO.  For more information see
-	  Documentation/netlabel.
+	  Documentation/netlabel as well as the NetLabel SourceForge project
+	  for configuration tools and additional documentation.
+
+	   * http://netlabel.sf.net
 
 	  If you are unsure, say N.
-- 
cgit v1.2.1


From a4c0291aa942dceddabe23bf2b74addb958d0964 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 25 Sep 2006 14:00:45 -0700
Subject: [SPARC64]: Fix section-mismatch errors in solaris emul module.

init_socksys() was marked __init but invoked from a
non-__init function.

Use the correct module_{init,exit}() faciltiies while we're
here and eliminate some seriously bogus ifdefs.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/solaris/misc.c    | 20 ++++----------------
 arch/sparc64/solaris/socksys.c |  6 ++----
 2 files changed, 6 insertions(+), 20 deletions(-)

diff --git a/arch/sparc64/solaris/misc.c b/arch/sparc64/solaris/misc.c
index 8135ec322c9c..642541769a17 100644
--- a/arch/sparc64/solaris/misc.c
+++ b/arch/sparc64/solaris/misc.c
@@ -736,20 +736,15 @@ struct exec_domain solaris_exec_domain = {
 
 extern int init_socksys(void);
 
-#ifdef MODULE
-
 MODULE_AUTHOR("Jakub Jelinek (jj@ultra.linux.cz), Patrik Rak (prak3264@ss1000.ms.mff.cuni.cz)");
 MODULE_DESCRIPTION("Solaris binary emulation module");
 MODULE_LICENSE("GPL");
 
-#ifdef __sparc_v9__
 extern u32 tl0_solaris[8];
 #define update_ttable(x) 										\
 	tl0_solaris[3] = (((long)(x) - (long)tl0_solaris - 3) >> 2) | 0x40000000;			\
 	wmb();		\
 	__asm__ __volatile__ ("flush %0" : : "r" (&tl0_solaris[3]))
-#else
-#endif	
 
 extern u32 solaris_sparc_syscall[];
 extern u32 solaris_syscall[];
@@ -757,7 +752,7 @@ extern void cleanup_socksys(void);
 
 extern u32 entry64_personality_patch;
 
-int init_module(void)
+static int __init solaris_init(void)
 {
 	int ret;
 
@@ -777,19 +772,12 @@ int init_module(void)
 	return 0;
 }
 
-void cleanup_module(void)
+static void __exit solaris_exit(void)
 {
 	update_ttable(solaris_syscall);
 	cleanup_socksys();
 	unregister_exec_domain(&solaris_exec_domain);
 }
 
-#else
-int init_solaris_emul(void)
-{
-	register_exec_domain(&solaris_exec_domain);
-	init_socksys();
-	return 0;
-}
-#endif
-
+module_init(solaris_init);
+module_exit(solaris_exit);
diff --git a/arch/sparc64/solaris/socksys.c b/arch/sparc64/solaris/socksys.c
index bc3df95bc057..7c90e41fd3be 100644
--- a/arch/sparc64/solaris/socksys.c
+++ b/arch/sparc64/solaris/socksys.c
@@ -168,8 +168,7 @@ static struct file_operations socksys_fops = {
 	.release =	socksys_release,
 };
 
-int __init
-init_socksys(void)
+int __init init_socksys(void)
 {
 	int ret;
 	struct file * file;
@@ -199,8 +198,7 @@ init_socksys(void)
 	return 0;
 }
 
-void
-cleanup_socksys(void)
+void __exit cleanup_socksys(void)
 {
 	if (unregister_chrdev(30, "socksys"))
 		printk ("Couldn't unregister socksys character device\n");
-- 
cgit v1.2.1


From efdbc1a7caf770b1b312000a42c630597f06973d Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 25 Sep 2006 14:04:49 -0700
Subject: [SUNLANCE]: Mark sparc_lance_probe_one as __devinit.

Fixes section mismatch warnings when built as a module.

Also, mark find_ledma and sun4 init function as __devinit
too.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/sunlance.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/sunlance.c b/drivers/net/sunlance.c
index 77670741e101..feb42db10ee1 100644
--- a/drivers/net/sunlance.c
+++ b/drivers/net/sunlance.c
@@ -1323,9 +1323,9 @@ static const struct ethtool_ops sparc_lance_ethtool_ops = {
 	.get_link		= sparc_lance_get_link,
 };
 
-static int __init sparc_lance_probe_one(struct sbus_dev *sdev,
-					struct sbus_dma *ledma,
-					struct sbus_dev *lebuffer)
+static int __devinit sparc_lance_probe_one(struct sbus_dev *sdev,
+					   struct sbus_dma *ledma,
+					   struct sbus_dev *lebuffer)
 {
 	static unsigned version_printed;
 	struct net_device *dev;
@@ -1515,7 +1515,7 @@ fail:
 }
 
 /* On 4m, find the associated dma for the lance chip */
-static inline struct sbus_dma *find_ledma(struct sbus_dev *sdev)
+static struct sbus_dma * __devinit find_ledma(struct sbus_dev *sdev)
 {
 	struct sbus_dma *p;
 
@@ -1533,7 +1533,7 @@ static inline struct sbus_dma *find_ledma(struct sbus_dev *sdev)
 
 /* Find all the lance cards on the system and initialize them */
 static struct sbus_dev sun4_sdev;
-static int __init sparc_lance_init(void)
+static int __devinit sparc_lance_init(void)
 {
 	if ((idprom->id_machtype == (SM_SUN4|SM_4_330)) ||
 	    (idprom->id_machtype == (SM_SUN4|SM_4_470))) {
-- 
cgit v1.2.1


From be5b6d3d6cb7311893c9fbeebf094591d5f760a8 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 25 Sep 2006 14:08:37 -0700
Subject: [SOUND] sparc/amd7930: Use __devinit and __devinitdata as needed.

Fixes section-mismatch errors.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 sound/sparc/amd7930.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/sound/sparc/amd7930.c b/sound/sparc/amd7930.c
index 2bd8e40b8541..be0bd503f013 100644
--- a/sound/sparc/amd7930.c
+++ b/sound/sparc/amd7930.c
@@ -755,7 +755,7 @@ static struct snd_pcm_ops snd_amd7930_capture_ops = {
 	.pointer	=	snd_amd7930_capture_pointer,
 };
 
-static int __init snd_amd7930_pcm(struct snd_amd7930 *amd)
+static int __devinit snd_amd7930_pcm(struct snd_amd7930 *amd)
 {
 	struct snd_pcm *pcm;
 	int err;
@@ -870,7 +870,7 @@ static int snd_amd7930_put_volume(struct snd_kcontrol *kctl, struct snd_ctl_elem
 	return change;
 }
 
-static struct snd_kcontrol_new amd7930_controls[] __initdata = {
+static struct snd_kcontrol_new amd7930_controls[] __devinitdata = {
 	{
 		.iface		=	SNDRV_CTL_ELEM_IFACE_MIXER,
 		.name		=	"Monitor Volume",
@@ -900,7 +900,7 @@ static struct snd_kcontrol_new amd7930_controls[] __initdata = {
 	},
 };
 
-static int __init snd_amd7930_mixer(struct snd_amd7930 *amd)
+static int __devinit snd_amd7930_mixer(struct snd_amd7930 *amd)
 {
 	struct snd_card *card;
 	int idx, err;
@@ -945,11 +945,11 @@ static struct snd_device_ops snd_amd7930_dev_ops = {
 	.dev_free	=	snd_amd7930_dev_free,
 };
 
-static int __init snd_amd7930_create(struct snd_card *card,
-				     struct resource *rp,
-				     unsigned int reg_size,
-				     int irq, int dev,
-				     struct snd_amd7930 **ramd)
+static int __devinit snd_amd7930_create(struct snd_card *card,
+					struct resource *rp,
+					unsigned int reg_size,
+					int irq, int dev,
+					struct snd_amd7930 **ramd)
 {
 	unsigned long flags;
 	struct snd_amd7930 *amd;
@@ -1013,7 +1013,7 @@ static int __init snd_amd7930_create(struct snd_card *card,
 	return 0;
 }
 
-static int __init amd7930_attach_common(struct resource *rp, int irq)
+static int __devinit amd7930_attach_common(struct resource *rp, int irq)
 {
 	static int dev_num;
 	struct snd_card *card;
@@ -1065,7 +1065,7 @@ out_err:
 	return err;
 }
 
-static int __init amd7930_obio_attach(struct device_node *dp)
+static int __devinit amd7930_obio_attach(struct device_node *dp)
 {
 	struct linux_prom_registers *regs;
 	struct linux_prom_irqs *irqp;
-- 
cgit v1.2.1


From 6432dc1f44ff3f02a304db26717c4f76e2e57be9 Mon Sep 17 00:00:00 2001
From: Deepak Saxena <dsaxena@plexity.net>
Date: Mon, 25 Sep 2006 16:39:18 -0700
Subject: [PATCH] Update smc91x driver with ARM Versatile board info

We need to specify a Versatile-specific SMC_IRQ_FLAGS value or the new
generic IRQ layer will complain thusly:

No IRQF_TRIGGER set_type function for IRQ 25 (<NULL>)

Signed-off-by: Deepak Saxena <dsaxena@plexity.net>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Nicolas Pitre <nico@cam.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/smc91x.h | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/net/smc91x.h b/drivers/net/smc91x.h
index 7aa7fbac8224..c660e33f43a2 100644
--- a/drivers/net/smc91x.h
+++ b/drivers/net/smc91x.h
@@ -379,6 +379,24 @@ static inline void LPD7_SMC_outsw (unsigned char* a, int r,
 
 #define SMC_IRQ_FLAGS		(0)
 
+#elif	defined(CONFIG_ARCH_VERSATILE)
+
+#define SMC_CAN_USE_8BIT	1
+#define SMC_CAN_USE_16BIT	1
+#define SMC_CAN_USE_32BIT	1
+#define SMC_NOWAIT		1
+
+#define SMC_inb(a, r)		readb((a) + (r))
+#define SMC_inw(a, r)		readw((a) + (r))
+#define SMC_inl(a, r)		readl((a) + (r))
+#define SMC_outb(v, a, r)	writeb(v, (a) + (r))
+#define SMC_outw(v, a, r)	writew(v, (a) + (r))
+#define SMC_outl(v, a, r)	writel(v, (a) + (r))
+#define SMC_insl(a, r, p, l)	readsl((a) + (r), p, l)
+#define SMC_outsl(a, r, p, l)	writesl((a) + (r), p, l)
+
+#define SMC_IRQ_FLAGS		(0)
+
 #else
 
 #define SMC_CAN_USE_8BIT	1
-- 
cgit v1.2.1


From bef363a1e07a913508cf10ab2d7e2b11e6313257 Mon Sep 17 00:00:00 2001
From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Date: Mon, 25 Sep 2006 16:39:19 -0700
Subject: [PATCH] drivers/net/acenic.c Removal of old code

Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Jes Sorensen <jes@trained-monkey.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/acenic.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/acenic.c b/drivers/net/acenic.c
index a075246f6f43..71a4f60f7325 100644
--- a/drivers/net/acenic.c
+++ b/drivers/net/acenic.c
@@ -163,11 +163,7 @@ MODULE_DEVICE_TABLE(pci, acenic_pci_tbl);
 #define SET_NETDEV_DEV(net, pdev)	do{} while(0)
 #endif
 
-#if LINUX_VERSION_CODE >= 0x2051c
 #define ace_sync_irq(irq)	synchronize_irq(irq)
-#else
-#define ace_sync_irq(irq)	synchronize_irq()
-#endif
 
 #ifndef offset_in_page
 #define offset_in_page(ptr)	((unsigned long)(ptr) & ~PAGE_MASK)
-- 
cgit v1.2.1


From 9098f5ba6947944d9b0d3e989535ab55ae6af3f6 Mon Sep 17 00:00:00 2001
From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Date: Mon, 25 Sep 2006 16:39:20 -0700
Subject: [PATCH] drivers/net/tokenring/lanstreamer.c Removal of old code

Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/tokenring/lanstreamer.c | 59 -------------------------------------
 1 file changed, 59 deletions(-)

diff --git a/drivers/net/tokenring/lanstreamer.c b/drivers/net/tokenring/lanstreamer.c
index 0d66700c6ced..bfc8c3eae9a1 100644
--- a/drivers/net/tokenring/lanstreamer.c
+++ b/drivers/net/tokenring/lanstreamer.c
@@ -1876,7 +1876,6 @@ static int sprintf_info(char *buffer, struct net_device *dev)
 		datap[size+1]=io_word & 0xff;
 	}
 
-
 	size = sprintf(buffer, "\n%6s: Adapter Address   : Node Address      : Functional Addr\n", dev->name);
 
 	size += sprintf(buffer + size,
@@ -1932,64 +1931,6 @@ static int sprintf_info(char *buffer, struct net_device *dev)
 #endif
 #endif
 
-#if STREAMER_IOCTL && (LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0))
-static int streamer_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-        int i;
-	struct streamer_private *streamer_priv = (struct streamer_private *) dev->priv;
-	u8 __iomem *streamer_mmio = streamer_priv->streamer_mmio;
-
-	switch(cmd) {
-	case IOCTL_SISR_MASK:
-		writew(SISR_MI, streamer_mmio + SISR_MASK_SUM);
-		break;
-	case IOCTL_SPIN_LOCK_TEST:
-	        printk(KERN_INFO "spin_lock() called.\n");
-		spin_lock(&streamer_priv->streamer_lock);
-		spin_unlock(&streamer_priv->streamer_lock);
-		printk(KERN_INFO "spin_unlock() finished.\n");
-		break;
-	case IOCTL_PRINT_BDAS:
-	        printk(KERN_INFO "bdas: RXBDA: %x RXLBDA: %x TX2FDA: %x TX2LFDA: %x\n",
-		       readw(streamer_mmio + RXBDA),
-		       readw(streamer_mmio + RXLBDA),
-		       readw(streamer_mmio + TX2FDA),
-		       readw(streamer_mmio + TX2LFDA));
-		break;
-	case IOCTL_PRINT_REGISTERS:
-	        printk(KERN_INFO "registers:\n");
-		printk(KERN_INFO "SISR: %04x MISR: %04x LISR: %04x BCTL: %04x BMCTL: %04x\nmask  %04x mask  %04x\n", 
-		       readw(streamer_mmio + SISR),
-		       readw(streamer_mmio + MISR_RUM),
-		       readw(streamer_mmio + LISR),
-		       readw(streamer_mmio + BCTL),
-		       readw(streamer_mmio + BMCTL_SUM),
-		       readw(streamer_mmio + SISR_MASK),
-		       readw(streamer_mmio + MISR_MASK));
-		break;
-	case IOCTL_PRINT_RX_BUFS:
-	        printk(KERN_INFO "Print rx bufs:\n");
-		for(i=0; i<STREAMER_RX_RING_SIZE; i++)
-		        printk(KERN_INFO "rx_ring %d status: 0x%x\n", i, 
-			       streamer_priv->streamer_rx_ring[i].status);
-		break;
-	case IOCTL_PRINT_TX_BUFS:
-	        printk(KERN_INFO "Print tx bufs:\n");
-		for(i=0; i<STREAMER_TX_RING_SIZE; i++)
-		        printk(KERN_INFO "tx_ring %d status: 0x%x\n", i, 
-			       streamer_priv->streamer_tx_ring[i].status);
-		break;
-	case IOCTL_RX_CMD:
-	        streamer_rx(dev);
-		printk(KERN_INFO "Sent rx command.\n");
-		break;
-	default:
-	        printk(KERN_INFO "Bad ioctl!\n");
-	}
-	return 0;
-}
-#endif
-
 static struct pci_driver streamer_pci_driver = {
   .name     = "lanstreamer",
   .id_table = streamer_pci_tbl,
-- 
cgit v1.2.1


From f8d9c9c876d3a2a109caa2c06b2761af0337c2f4 Mon Sep 17 00:00:00 2001
From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Date: Mon, 25 Sep 2006 16:39:21 -0700
Subject: [PATCH] drivers/net/tokenring/lanstreamer.h Removal of old code

Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/tokenring/lanstreamer.h | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/net/tokenring/lanstreamer.h b/drivers/net/tokenring/lanstreamer.h
index 5557d8e1e22d..e7bb3494afc7 100644
--- a/drivers/net/tokenring/lanstreamer.h
+++ b/drivers/net/tokenring/lanstreamer.h
@@ -62,18 +62,6 @@
 
 #include <linux/version.h>
 
-#if STREAMER_IOCTL && (LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0))
-#include <asm/ioctl.h>
-#define IOCTL_PRINT_RX_BUFS   SIOCDEVPRIVATE
-#define IOCTL_PRINT_TX_BUFS   SIOCDEVPRIVATE+1
-#define IOCTL_RX_CMD          SIOCDEVPRIVATE+2
-#define IOCTL_TX_CMD          SIOCDEVPRIVATE+3
-#define IOCTL_PRINT_REGISTERS SIOCDEVPRIVATE+4
-#define IOCTL_PRINT_BDAS      SIOCDEVPRIVATE+5
-#define IOCTL_SPIN_LOCK_TEST  SIOCDEVPRIVATE+6
-#define IOCTL_SISR_MASK       SIOCDEVPRIVATE+7
-#endif
-
 /* MAX_INTR - the maximum number of times we can loop
  * inside the interrupt function before returning
  * control to the OS (maximum value is 256)
-- 
cgit v1.2.1


From 0fe2480aded9504434e24815c410cbebb4f4efad Mon Sep 17 00:00:00 2001
From: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Date: Mon, 25 Sep 2006 16:39:21 -0700
Subject: [PATCH] drivers/net/typhoon.c Removal of old code

Signed-off-by: Michal Piotrowski <michal.k.k.piotrowski@gmail.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/typhoon.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c
index 8f6f6fd8b87d..d5c32e9caa97 100644
--- a/drivers/net/typhoon.c
+++ b/drivers/net/typhoon.c
@@ -333,11 +333,7 @@ enum state_values {
 #define TYPHOON_RESET_TIMEOUT_NOSLEEP	((6 * 1000000) / TYPHOON_UDELAY)
 #define TYPHOON_WAIT_TIMEOUT		((1000000 / 2) / TYPHOON_UDELAY)
 
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 5, 28)
-#define typhoon_synchronize_irq(x) synchronize_irq()
-#else
 #define typhoon_synchronize_irq(x) synchronize_irq(x)
-#endif
 
 #if defined(NETIF_F_TSO)
 #define skb_tso_size(x)		(skb_shinfo(x)->gso_size)
-- 
cgit v1.2.1


From 84c22d7901f793bd267b5f79270080964b252826 Mon Sep 17 00:00:00 2001
From: Eric Sesterhenn <snakebyte@gmx.de>
Date: Mon, 25 Sep 2006 16:39:22 -0700
Subject: [PATCH] Signedness issue in drivers/net/phy/phy_device.c

While checking gcc 4.1 -Wextra warnings, I stumbled across the following
two warnings:

drivers/net/phy/phy_device.c:528: warning: comparison of unsigned expression < 0 is always false
drivers/net/phy/phy_device.c:546: warning: comparison of unsigned expression < 0 is always false

Since phy_read() returns an integer and can return negative values, it seems
to me the best way to get proper error handling working again is to make val
an int.  Currently it is an u32, so the < 0 check always fails.

Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/phy/phy_device.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 2d1ecfdc80db..ecd3da151e2d 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -522,7 +522,7 @@ EXPORT_SYMBOL(genphy_read_status);
 
 static int genphy_config_init(struct phy_device *phydev)
 {
-	u32 val;
+	int val;
 	u32 features;
 
 	/* For now, I'll claim that the generic driver supports
-- 
cgit v1.2.1


From 46798c897e235e71e1e9c46a5e6e9adfffd8b85d Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Mon, 25 Sep 2006 16:39:24 -0700
Subject: [PATCH] fix possible NULL ptr deref in forcedeth

There seems to be a possible NULL pointer deref bug in
drivers/net/forcedeth.c::nv_loopback_test().  If dev_alloc_skb() fails, the
next line will call skb_put() with a NULL first argument which it'll then
try to deref - kaboom: a NULL pointer deref.  Found by coverity (#1337).

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Cc: Ayaz Abdulla <aabdulla@nvidia.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Stephen Hemminger <shemminger@osdl.org>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/forcedeth.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c
index 97db910fbc8c..eea1d66c530e 100644
--- a/drivers/net/forcedeth.c
+++ b/drivers/net/forcedeth.c
@@ -3789,6 +3789,12 @@ static int nv_loopback_test(struct net_device *dev)
 	/* setup packet for tx */
 	pkt_len = ETH_DATA_LEN;
 	tx_skb = dev_alloc_skb(pkt_len);
+	if (!tx_skb) {
+		printk(KERN_ERR "dev_alloc_skb() failed during loopback test"
+			 " of %s\n", dev->name);
+		ret = 0;
+		goto out;
+	}
 	pkt_data = skb_put(tx_skb, pkt_len);
 	for (i = 0; i < pkt_len; i++)
 		pkt_data[i] = (u8)(i & 0xff);
@@ -3853,7 +3859,7 @@ static int nv_loopback_test(struct net_device *dev)
 		       tx_skb->end-tx_skb->data,
 		       PCI_DMA_TODEVICE);
 	dev_kfree_skb_any(tx_skb);
-
+ out:
 	/* stop engines */
 	nv_stop_rx(dev);
 	nv_stop_tx(dev);
-- 
cgit v1.2.1


From 64f6b64dfb8cdda21652f24a0fb0a68e2f0b0022 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Sat, 23 Sep 2006 21:25:28 -0700
Subject: [PATCH] skge: fiber support

Add support for older fiber versions of the SysKonnect board. These chipsets
use an internal PHY so they require special handling. The older sk98lin
driver already supported these

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/skge.c | 357 ++++++++++++++++++++++++++++++++++++++++-------------
 drivers/net/skge.h |  37 ++++--
 2 files changed, 301 insertions(+), 93 deletions(-)

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index 9142d91355bc..705e9a8fa30f 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -58,6 +58,7 @@
 #define TX_WATCHDOG		(5 * HZ)
 #define NAPI_WEIGHT		64
 #define BLINK_MS		250
+#define LINK_HZ			(HZ/2)
 
 MODULE_DESCRIPTION("SysKonnect Gigabit Ethernet driver");
 MODULE_AUTHOR("Stephen Hemminger <shemminger@osdl.org>");
@@ -605,7 +606,12 @@ static void skge_led(struct skge_port *skge, enum led_mode mode)
 	if (hw->chip_id == CHIP_ID_GENESIS) {
 		switch (mode) {
 		case LED_MODE_OFF:
-			xm_phy_write(hw, port, PHY_BCOM_P_EXT_CTRL, PHY_B_PEC_LED_OFF);
+			if (hw->phy_type == SK_PHY_BCOM)
+				xm_phy_write(hw, port, PHY_BCOM_P_EXT_CTRL, PHY_B_PEC_LED_OFF);
+			else {
+				skge_write32(hw, SK_REG(port, TX_LED_VAL), 0);
+				skge_write8(hw, SK_REG(port, TX_LED_CTRL), LED_T_OFF);
+			}
 			skge_write8(hw, SK_REG(port, LNK_LED_REG), LINKLED_OFF);
 			skge_write32(hw, SK_REG(port, RX_LED_VAL), 0);
 			skge_write8(hw, SK_REG(port, RX_LED_CTRL), LED_T_OFF);
@@ -625,8 +631,14 @@ static void skge_led(struct skge_port *skge, enum led_mode mode)
 			skge_write32(hw, SK_REG(port, RX_LED_VAL), 100);
 			skge_write8(hw, SK_REG(port, RX_LED_CTRL), LED_START);
 
-			xm_phy_write(hw, port, PHY_BCOM_P_EXT_CTRL, PHY_B_PEC_LED_ON);
-			break;
+			if (hw->phy_type == SK_PHY_BCOM)
+				xm_phy_write(hw, port, PHY_BCOM_P_EXT_CTRL, PHY_B_PEC_LED_ON);
+			else {
+				skge_write8(hw, SK_REG(port, TX_LED_TST), LED_T_ON);
+				skge_write32(hw, SK_REG(port, TX_LED_VAL), 100);
+				skge_write8(hw, SK_REG(port, TX_LED_CTRL), LED_START);
+			}
+
 		}
 	} else {
 		switch (mode) {
@@ -879,6 +891,9 @@ static int __xm_phy_read(struct skge_hw *hw, int port, u16 reg, u16 *val)
 	xm_write16(hw, port, XM_PHY_ADDR, reg | hw->phy_addr);
 	*val = xm_read16(hw, port, XM_PHY_DATA);
 
+	if (hw->phy_type == SK_PHY_XMAC)
+		goto ready;
+
 	for (i = 0; i < PHY_RETRIES; i++) {
 		if (xm_read16(hw, port, XM_MMU_CMD) & XM_MMU_PHY_RDY)
 			goto ready;
@@ -965,7 +980,8 @@ static void genesis_reset(struct skge_hw *hw, int port)
 	xm_write16(hw, port, XM_RX_CMD, 0);	/* reset RX CMD Reg */
 
 	/* disable Broadcom PHY IRQ */
-	xm_write16(hw, port, PHY_BCOM_INT_MASK, 0xffff);
+	if (hw->phy_type == SK_PHY_BCOM)
+		xm_write16(hw, port, PHY_BCOM_INT_MASK, 0xffff);
 
 	xm_outhash(hw, port, XM_HSM, zero);
 }
@@ -1000,60 +1016,64 @@ static void bcom_check_link(struct skge_hw *hw, int port)
 
 		if (netif_carrier_ok(dev))
 			skge_link_down(skge);
-	} else {
-		if (skge->autoneg == AUTONEG_ENABLE &&
-		    (status & PHY_ST_AN_OVER)) {
-			u16 lpa = xm_phy_read(hw, port, PHY_BCOM_AUNE_LP);
-			u16 aux = xm_phy_read(hw, port, PHY_BCOM_AUX_STAT);
-
-			if (lpa & PHY_B_AN_RF) {
-				printk(KERN_NOTICE PFX "%s: remote fault\n",
-				       dev->name);
-				return;
-			}
+		return;
+	}
 
-			/* Check Duplex mismatch */
-			switch (aux & PHY_B_AS_AN_RES_MSK) {
-			case PHY_B_RES_1000FD:
-				skge->duplex = DUPLEX_FULL;
-				break;
-			case PHY_B_RES_1000HD:
-				skge->duplex = DUPLEX_HALF;
-				break;
-			default:
-				printk(KERN_NOTICE PFX "%s: duplex mismatch\n",
-				       dev->name);
-				return;
-			}
+	if (skge->autoneg == AUTONEG_ENABLE) {
+		u16 lpa, aux;
 
+		if (!(status & PHY_ST_AN_OVER))
+			return;
 
-			/* We are using IEEE 802.3z/D5.0 Table 37-4 */
-			switch (aux & PHY_B_AS_PAUSE_MSK) {
-			case PHY_B_AS_PAUSE_MSK:
-				skge->flow_control = FLOW_MODE_SYMMETRIC;
-				break;
-			case PHY_B_AS_PRR:
-				skge->flow_control = FLOW_MODE_REM_SEND;
-				break;
-			case PHY_B_AS_PRT:
-				skge->flow_control = FLOW_MODE_LOC_SEND;
-				break;
-			default:
-				skge->flow_control = FLOW_MODE_NONE;
-			}
+		lpa = xm_phy_read(hw, port, PHY_XMAC_AUNE_LP);
+		if (lpa & PHY_B_AN_RF) {
+			printk(KERN_NOTICE PFX "%s: remote fault\n",
+			       dev->name);
+			return;
+		}
 
-			skge->speed = SPEED_1000;
+		aux = xm_phy_read(hw, port, PHY_BCOM_AUX_STAT);
+
+		/* Check Duplex mismatch */
+		switch (aux & PHY_B_AS_AN_RES_MSK) {
+		case PHY_B_RES_1000FD:
+			skge->duplex = DUPLEX_FULL;
+			break;
+		case PHY_B_RES_1000HD:
+			skge->duplex = DUPLEX_HALF;
+			break;
+		default:
+			printk(KERN_NOTICE PFX "%s: duplex mismatch\n",
+			       dev->name);
+			return;
 		}
 
-		if (!netif_carrier_ok(dev))
-			genesis_link_up(skge);
+
+		/* We are using IEEE 802.3z/D5.0 Table 37-4 */
+		switch (aux & PHY_B_AS_PAUSE_MSK) {
+		case PHY_B_AS_PAUSE_MSK:
+			skge->flow_control = FLOW_MODE_SYMMETRIC;
+			break;
+		case PHY_B_AS_PRR:
+			skge->flow_control = FLOW_MODE_REM_SEND;
+			break;
+		case PHY_B_AS_PRT:
+			skge->flow_control = FLOW_MODE_LOC_SEND;
+			break;
+		default:
+			skge->flow_control = FLOW_MODE_NONE;
+		}
+		skge->speed = SPEED_1000;
 	}
+
+	if (!netif_carrier_ok(dev))
+		genesis_link_up(skge);
 }
 
 /* Broadcom 5400 only supports giagabit! SysKonnect did not put an additional
  * Phy on for 100 or 10Mbit operation
  */
-static void bcom_phy_init(struct skge_port *skge, int jumbo)
+static void bcom_phy_init(struct skge_port *skge)
 {
 	struct skge_hw *hw = skge->hw;
 	int port = skge->port;
@@ -1144,7 +1164,7 @@ static void bcom_phy_init(struct skge_port *skge, int jumbo)
 		     phy_pause_map[skge->flow_control] | PHY_AN_CSMA);
 
 	/* Handle Jumbo frames */
-	if (jumbo) {
+	if (hw->dev[port]->mtu > ETH_DATA_LEN) {
 		xm_phy_write(hw, port, PHY_BCOM_AUX_CTRL,
 			     PHY_B_AC_TX_TST | PHY_B_AC_LONG_PACK);
 
@@ -1157,8 +1177,154 @@ static void bcom_phy_init(struct skge_port *skge, int jumbo)
 
 	/* Use link status change interrupt */
 	xm_phy_write(hw, port, PHY_BCOM_INT_MASK, PHY_B_DEF_MSK);
+}
 
-	bcom_check_link(hw, port);
+static void xm_phy_init(struct skge_port *skge)
+{
+	struct skge_hw *hw = skge->hw;
+	int port = skge->port;
+	u16 ctrl = 0;
+
+	if (skge->autoneg == AUTONEG_ENABLE) {
+		if (skge->advertising & ADVERTISED_1000baseT_Half)
+			ctrl |= PHY_X_AN_HD;
+		if (skge->advertising & ADVERTISED_1000baseT_Full)
+			ctrl |= PHY_X_AN_FD;
+
+		switch(skge->flow_control) {
+		case FLOW_MODE_NONE:
+			ctrl |= PHY_X_P_NO_PAUSE;
+			break;
+		case FLOW_MODE_LOC_SEND:
+			ctrl |= PHY_X_P_ASYM_MD;
+			break;
+		case FLOW_MODE_SYMMETRIC:
+			ctrl |= PHY_X_P_BOTH_MD;
+			break;
+		}
+
+		xm_phy_write(hw, port, PHY_XMAC_AUNE_ADV, ctrl);
+
+		/* Restart Auto-negotiation */
+		ctrl = PHY_CT_ANE | PHY_CT_RE_CFG;
+	} else {
+		/* Set DuplexMode in Config register */
+		if (skge->duplex == DUPLEX_FULL)
+			ctrl |= PHY_CT_DUP_MD;
+		/*
+		 * Do NOT enable Auto-negotiation here. This would hold
+		 * the link down because no IDLEs are transmitted
+		 */
+	}
+
+	xm_phy_write(hw, port, PHY_XMAC_CTRL, ctrl);
+
+	/* Poll PHY for status changes */
+	schedule_delayed_work(&skge->link_thread, LINK_HZ);
+}
+
+static void xm_check_link(struct net_device *dev)
+{
+	struct skge_port *skge = netdev_priv(dev);
+	struct skge_hw *hw = skge->hw;
+	int port = skge->port;
+	u16 status;
+
+	/* read twice because of latch */
+	(void) xm_phy_read(hw, port, PHY_XMAC_STAT);
+	status = xm_phy_read(hw, port, PHY_XMAC_STAT);
+
+	if ((status & PHY_ST_LSYNC) == 0) {
+		u16 cmd = xm_read16(hw, port, XM_MMU_CMD);
+		cmd &= ~(XM_MMU_ENA_RX | XM_MMU_ENA_TX);
+		xm_write16(hw, port, XM_MMU_CMD, cmd);
+		/* dummy read to ensure writing */
+		(void) xm_read16(hw, port, XM_MMU_CMD);
+
+		if (netif_carrier_ok(dev))
+			skge_link_down(skge);
+		return;
+	}
+
+	if (skge->autoneg == AUTONEG_ENABLE) {
+		u16 lpa, res;
+
+		if (!(status & PHY_ST_AN_OVER))
+			return;
+
+		lpa = xm_phy_read(hw, port, PHY_XMAC_AUNE_LP);
+		if (lpa & PHY_B_AN_RF) {
+			printk(KERN_NOTICE PFX "%s: remote fault\n",
+			       dev->name);
+			return;
+		}
+
+		res = xm_phy_read(hw, port, PHY_XMAC_RES_ABI);
+
+		/* Check Duplex mismatch */
+		switch (res & (PHY_X_RS_HD | PHY_X_RS_FD)) {
+		case PHY_X_RS_FD:
+			skge->duplex = DUPLEX_FULL;
+			break;
+		case PHY_X_RS_HD:
+			skge->duplex = DUPLEX_HALF;
+			break;
+		default:
+			printk(KERN_NOTICE PFX "%s: duplex mismatch\n",
+			       dev->name);
+			return;
+		}
+
+		/* We are using IEEE 802.3z/D5.0 Table 37-4 */
+		if (lpa & PHY_X_P_SYM_MD)
+			skge->flow_control = FLOW_MODE_SYMMETRIC;
+		else if ((lpa & PHY_X_RS_PAUSE) == PHY_X_P_ASYM_MD)
+			skge->flow_control = FLOW_MODE_REM_SEND;
+		else if ((lpa & PHY_X_RS_PAUSE) == PHY_X_P_BOTH_MD)
+			skge->flow_control = FLOW_MODE_LOC_SEND;
+		else
+			skge->flow_control = FLOW_MODE_NONE;
+
+
+		skge->speed = SPEED_1000;
+	}
+
+	if (!netif_carrier_ok(dev))
+		genesis_link_up(skge);
+}
+
+/* Poll to check for link coming up.
+ * Since internal PHY is wired to a level triggered pin, can't
+ * get an interrupt when carrier is detected.
+ */
+static void xm_link_timer(void *arg)
+{
+	struct net_device *dev = arg;
+	struct skge_port *skge = netdev_priv(arg);
+ 	struct skge_hw *hw = skge->hw;
+	int port = skge->port;
+
+	if (!netif_running(dev))
+		return;
+
+	if (netif_carrier_ok(dev)) {
+		xm_read16(hw, port, XM_ISRC);
+		if (!(xm_read16(hw, port, XM_ISRC) & XM_IS_INP_ASS))
+			goto nochange;
+	} else {
+		if (xm_read32(hw, port, XM_GP_PORT) & XM_GP_INP_ASS)
+			goto nochange;
+		xm_read16(hw, port, XM_ISRC);
+		if (xm_read16(hw, port, XM_ISRC) & XM_IS_INP_ASS)
+			goto nochange;
+	}
+
+	mutex_lock(&hw->phy_mutex);
+	xm_check_link(dev);
+	mutex_unlock(&hw->phy_mutex);
+
+nochange:
+	schedule_delayed_work(&skge->link_thread, LINK_HZ);
 }
 
 static void genesis_mac_init(struct skge_hw *hw, int port)
@@ -1189,20 +1355,29 @@ static void genesis_mac_init(struct skge_hw *hw, int port)
 	 * namely for the 1000baseTX cards that use the XMAC's
 	 * GMII mode.
 	 */
-	/* Take external Phy out of reset */
-	r = skge_read32(hw, B2_GP_IO);
-	if (port == 0)
-		r |= GP_DIR_0|GP_IO_0;
-	else
-		r |= GP_DIR_2|GP_IO_2;
+	if (hw->phy_type != SK_PHY_XMAC) {
+		/* Take external Phy out of reset */
+		r = skge_read32(hw, B2_GP_IO);
+		if (port == 0)
+			r |= GP_DIR_0|GP_IO_0;
+		else
+			r |= GP_DIR_2|GP_IO_2;
 
-	skge_write32(hw, B2_GP_IO, r);
+		skge_write32(hw, B2_GP_IO, r);
 
+		/* Enable GMII interface */
+		xm_write16(hw, port, XM_HW_CFG, XM_HW_GMII_MD);
+	}
 
-	/* Enable GMII interface */
-	xm_write16(hw, port, XM_HW_CFG, XM_HW_GMII_MD);
 
-	bcom_phy_init(skge, jumbo);
+	switch(hw->phy_type) {
+	case SK_PHY_XMAC:
+		xm_phy_init(skge);
+		break;
+	case SK_PHY_BCOM:
+		bcom_phy_init(skge);
+		bcom_check_link(hw, port);
+	}
 
 	/* Set Station Address */
 	xm_outaddr(hw, port, XM_SA, dev->dev_addr);
@@ -1335,16 +1510,18 @@ static void genesis_stop(struct skge_port *skge)
 	skge_write16(hw, SK_REG(port, TX_MFF_CTRL1), MFF_SET_MAC_RST);
 
 	/* For external PHYs there must be special handling */
-	reg = skge_read32(hw, B2_GP_IO);
-	if (port == 0) {
-		reg |= GP_DIR_0;
-		reg &= ~GP_IO_0;
-	} else {
-		reg |= GP_DIR_2;
-		reg &= ~GP_IO_2;
+	if (hw->phy_type != SK_PHY_XMAC) {
+		reg = skge_read32(hw, B2_GP_IO);
+		if (port == 0) {
+			reg |= GP_DIR_0;
+			reg &= ~GP_IO_0;
+		} else {
+			reg |= GP_DIR_2;
+			reg &= ~GP_IO_2;
+		}
+		skge_write32(hw, B2_GP_IO, reg);
+		skge_read32(hw, B2_GP_IO);
 	}
-	skge_write32(hw, B2_GP_IO, reg);
-	skge_read32(hw, B2_GP_IO);
 
 	xm_write16(hw, port, XM_MMU_CMD,
 			xm_read16(hw, port, XM_MMU_CMD)
@@ -1406,7 +1583,7 @@ static void genesis_link_up(struct skge_port *skge)
 	struct skge_hw *hw = skge->hw;
 	int port = skge->port;
 	u16 cmd;
-	u32 mode, msk;
+	u32 mode;
 
 	cmd = xm_read16(hw, port, XM_MMU_CMD);
 
@@ -1454,27 +1631,24 @@ static void genesis_link_up(struct skge_port *skge)
 	}
 
 	xm_write32(hw, port, XM_MODE, mode);
-
-	msk = XM_DEF_MSK;
-	/* disable GP0 interrupt bit for external Phy */
-	msk |= XM_IS_INP_ASS;
-
-	xm_write16(hw, port, XM_IMSK, msk);
+	xm_write16(hw, port, XM_IMSK, XM_DEF_MSK);
 	xm_read16(hw, port, XM_ISRC);
 
 	/* get MMU Command Reg. */
 	cmd = xm_read16(hw, port, XM_MMU_CMD);
-	if (skge->duplex == DUPLEX_FULL)
+	if (hw->phy_type != SK_PHY_XMAC && skge->duplex == DUPLEX_FULL)
 		cmd |= XM_MMU_GMII_FD;
 
 	/*
 	 * Workaround BCOM Errata (#10523) for all BCom Phys
 	 * Enable Power Management after link up
 	 */
-	xm_phy_write(hw, port, PHY_BCOM_AUX_CTRL,
-		     xm_phy_read(hw, port, PHY_BCOM_AUX_CTRL)
-		     & ~PHY_B_AC_DIS_PM);
-	xm_phy_write(hw, port, PHY_BCOM_INT_MASK, PHY_B_DEF_MSK);
+	if (hw->phy_type == SK_PHY_BCOM) {
+		xm_phy_write(hw, port, PHY_BCOM_AUX_CTRL,
+			     xm_phy_read(hw, port, PHY_BCOM_AUX_CTRL)
+			     & ~PHY_B_AC_DIS_PM);
+		xm_phy_write(hw, port, PHY_BCOM_INT_MASK, PHY_B_DEF_MSK);
+	}
 
 	/* enable Rx/Tx */
 	xm_write16(hw, port, XM_MMU_CMD,
@@ -2240,6 +2414,8 @@ static int skge_down(struct net_device *dev)
 		printk(KERN_INFO PFX "%s: disabling interface\n", dev->name);
 
 	netif_stop_queue(dev);
+	if (hw->chip_id == CHIP_ID_GENESIS && hw->phy_type == SK_PHY_XMAC)
+		cancel_rearming_delayed_work(&skge->link_thread);
 
 	skge_write8(skge->hw, SK_REG(skge->port, LNK_LED_REG), LED_OFF);
 	if (hw->chip_id == CHIP_ID_GENESIS)
@@ -2862,7 +3038,7 @@ static void skge_extirq(void *arg)
 		if (netif_running(dev)) {
 			if (hw->chip_id != CHIP_ID_GENESIS)
 				yukon_phy_intr(skge);
-			else
+			else if (hw->phy_type == SK_PHY_BCOM)
 				bcom_phy_intr(skge);
 		}
 	}
@@ -3014,7 +3190,7 @@ static int skge_reset(struct skge_hw *hw)
 {
 	u32 reg;
 	u16 ctst, pci_status;
-	u8 t8, mac_cfg, pmd_type, phy_type;
+	u8 t8, mac_cfg, pmd_type;
 	int i;
 
 	ctst = skge_read16(hw, B0_CTST);
@@ -3038,19 +3214,22 @@ static int skge_reset(struct skge_hw *hw)
 		     ctst & (CS_CLK_RUN_HOT|CS_CLK_RUN_RST|CS_CLK_RUN_ENA));
 
 	hw->chip_id = skge_read8(hw, B2_CHIP_ID);
-	phy_type = skge_read8(hw, B2_E_1) & 0xf;
+	hw->phy_type = skge_read8(hw, B2_E_1) & 0xf;
 	pmd_type = skge_read8(hw, B2_PMD_TYP);
 	hw->copper = (pmd_type == 'T' || pmd_type == '1');
 
 	switch (hw->chip_id) {
 	case CHIP_ID_GENESIS:
-		switch (phy_type) {
+		switch (hw->phy_type) {
+		case SK_PHY_XMAC:
+			hw->phy_addr = PHY_ADDR_XMAC;
+			break;
 		case SK_PHY_BCOM:
 			hw->phy_addr = PHY_ADDR_BCOM;
 			break;
 		default:
 			printk(KERN_ERR PFX "%s: unsupported phy type 0x%x\n",
-			       pci_name(hw->pdev), phy_type);
+			       pci_name(hw->pdev), hw->phy_type);
 			return -EOPNOTSUPP;
 		}
 		break;
@@ -3058,7 +3237,7 @@ static int skge_reset(struct skge_hw *hw)
 	case CHIP_ID_YUKON:
 	case CHIP_ID_YUKON_LITE:
 	case CHIP_ID_YUKON_LP:
-		if (phy_type < SK_PHY_MARV_COPPER && pmd_type != 'S')
+		if (hw->phy_type < SK_PHY_MARV_COPPER && pmd_type != 'S')
 			hw->copper = 1;
 
 		hw->phy_addr = PHY_ADDR_MARV;
@@ -3089,10 +3268,13 @@ static int skge_reset(struct skge_hw *hw)
 	else
 		hw->ram_size = t8 * 4096;
 
-	hw->intr_mask = IS_HW_ERR | IS_EXT_REG | IS_PORT_1;
+	hw->intr_mask = IS_HW_ERR | IS_PORT_1;
 	if (hw->ports > 1)
 		hw->intr_mask |= IS_PORT_2;
 
+	if (!(hw->chip_id == CHIP_ID_GENESIS && hw->phy_type == SK_PHY_XMAC))
+		hw->intr_mask |= IS_EXT_REG;
+
 	if (hw->chip_id == CHIP_ID_GENESIS)
 		genesis_init(hw);
 	else {
@@ -3226,6 +3408,9 @@ static struct net_device *skge_devinit(struct skge_hw *hw, int port,
 
 	skge->port = port;
 
+	/* Only used for Genesis XMAC */
+	INIT_WORK(&skge->link_thread, xm_link_timer, dev);
+
 	if (hw->chip_id != CHIP_ID_GENESIS) {
 		dev->features |= NETIF_F_IP_CSUM | NETIF_F_SG;
 		skge->rx_csum = 1;
diff --git a/drivers/net/skge.h b/drivers/net/skge.h
index 79e09271bcf9..d0b47d46cf9d 100644
--- a/drivers/net/skge.h
+++ b/drivers/net/skge.h
@@ -934,7 +934,7 @@ enum {
 	PHY_XMAC_AUNE_ADV	= 0x04,/* 16 bit r/w	Auto-Neg. Advertisement */
 	PHY_XMAC_AUNE_LP	= 0x05,/* 16 bit r/o	Link Partner Abi Reg */
 	PHY_XMAC_AUNE_EXP	= 0x06,/* 16 bit r/o	Auto-Neg. Expansion Reg */
-	PHY_XMAC_NEPG	= 0x07,/* 16 bit r/w	Next Page Register */
+	PHY_XMAC_NEPG		= 0x07,/* 16 bit r/w	Next Page Register */
 	PHY_XMAC_NEPG_LP	= 0x08,/* 16 bit r/o	Next Page Link Partner */
 
 	PHY_XMAC_EXT_STAT	= 0x0f,/* 16 bit r/o	Ext Status Register */
@@ -1097,13 +1097,36 @@ enum {
 
 /* Pause Bits (PHY_X_AN_PAUSE and PHY_X_RS_PAUSE) encoding */
 enum {
-	PHY_X_P_NO_PAUSE	= 0<<7,/* Bit  8..7:	no Pause Mode */
+	PHY_X_P_NO_PAUSE= 0<<7,/* Bit  8..7:	no Pause Mode */
 	PHY_X_P_SYM_MD	= 1<<7, /* Bit  8..7:	symmetric Pause Mode */
 	PHY_X_P_ASYM_MD	= 2<<7,/* Bit  8..7:	asymmetric Pause Mode */
 	PHY_X_P_BOTH_MD	= 3<<7,/* Bit  8..7:	both Pause Mode */
 };
 
 
+/*****  PHY_XMAC_EXT_STAT	16 bit r/w	Extended Status Register *****/
+enum {
+	PHY_X_EX_FD	= 1<<15, /* Bit 15:	Device Supports Full Duplex */
+	PHY_X_EX_HD	= 1<<14, /* Bit 14:	Device Supports Half Duplex */
+};
+
+/*****  PHY_XMAC_RES_ABI	16 bit r/o	PHY Resolved Ability *****/
+enum {
+	PHY_X_RS_PAUSE	= 3<<7,	/* Bit  8..7:	selected Pause Mode */
+	PHY_X_RS_HD	= 1<<6,	/* Bit  6:	Half Duplex Mode selected */
+	PHY_X_RS_FD	= 1<<5,	/* Bit  5:	Full Duplex Mode selected */
+	PHY_X_RS_ABLMIS = 1<<4,	/* Bit  4:	duplex or pause cap mismatch */
+	PHY_X_RS_PAUMIS = 1<<3,	/* Bit  3:	pause capability mismatch */
+};
+
+/* Remote Fault Bits (PHY_X_AN_RFB) encoding */
+enum {
+	X_RFB_OK	= 0<<12,/* Bit 13..12	No errors, Link OK */
+	X_RFB_LF	= 1<<12,/* Bit 13..12	Link Failure */
+	X_RFB_OFF	= 2<<12,/* Bit 13..12	Offline */
+	X_RFB_AN_ERR	= 3<<12,/* Bit 13..12	Auto-Negotiation Error */
+};
+
 /* Broadcom-Specific */
 /*****  PHY_BCOM_1000T_CTRL	16 bit r/w	1000Base-T Control Reg *****/
 enum {
@@ -2158,8 +2181,8 @@ enum {
 	XM_IS_LNK_AE	= 1<<14, /* Bit 14:	Link Asynchronous Event */
 	XM_IS_TX_ABORT	= 1<<13, /* Bit 13:	Transmit Abort, late Col. etc */
 	XM_IS_FRC_INT	= 1<<12, /* Bit 12:	Force INT bit set in GP */
-	XM_IS_INP_ASS	= 1<<11,	/* Bit 11:	Input Asserted, GP bit 0 set */
-	XM_IS_LIPA_RC	= 1<<10,	/* Bit 10:	Link Partner requests config */
+	XM_IS_INP_ASS	= 1<<11, /* Bit 11:	Input Asserted, GP bit 0 set */
+	XM_IS_LIPA_RC	= 1<<10, /* Bit 10:	Link Partner requests config */
 	XM_IS_RX_PAGE	= 1<<9,	/* Bit  9:	Page Received */
 	XM_IS_TX_PAGE	= 1<<8,	/* Bit  8:	Next Page Loaded for Transmit */
 	XM_IS_AND	= 1<<7,	/* Bit  7:	Auto-Negotiation Done */
@@ -2172,9 +2195,7 @@ enum {
 	XM_IS_RX_COMP	= 1<<0,	/* Bit  0:	Frame Rx Complete */
 };
 
-#define XM_DEF_MSK	(~(XM_IS_INP_ASS | XM_IS_LIPA_RC | XM_IS_RX_PAGE | \
-			   XM_IS_AND | XM_IS_RXC_OV | XM_IS_TXC_OV | \
-			   XM_IS_RXF_OV | XM_IS_TXF_UR))
+#define XM_DEF_MSK	(~(XM_IS_RXC_OV | XM_IS_TXC_OV | XM_IS_RXF_OV | XM_IS_TXF_UR))
 
 
 /*	XM_HW_CFG	16 bit r/w	Hardware Config Register */
@@ -2396,6 +2417,7 @@ struct skge_hw {
 	u8		     chip_rev;
 	u8		     copper;
 	u8		     ports;
+	u8		     phy_type;
 
 	u32	     	     ram_size;
 	u32	     	     ram_offset;
@@ -2422,6 +2444,7 @@ struct skge_port {
 
 	struct net_device_stats net_stats;
 
+	struct work_struct   link_thread;
 	u8		     rx_csum;
 	u8		     blink_on;
 	u8		     flow_control;
-- 
cgit v1.2.1


From e8126c82ccbfd39ccf3d22645207fbcc798021a3 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 25 Sep 2006 20:06:24 -0400
Subject: Delete unused drivers/net/gt64240eth.h

Noticed by Yoichi Yuasa.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/gt64240eth.h | 402 -----------------------------------------------
 1 file changed, 402 deletions(-)
 delete mode 100644 drivers/net/gt64240eth.h

diff --git a/drivers/net/gt64240eth.h b/drivers/net/gt64240eth.h
deleted file mode 100644
index 0d6f486e4579..000000000000
--- a/drivers/net/gt64240eth.h
+++ /dev/null
@@ -1,402 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 Patton Electronics Company
- * Copyright (C) 2002 Momentum Computer
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	stevel@mvista.com or support@mvista.com
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- * Ethernet driver definitions for the MIPS GT96100 Advanced
- * Communication Controller.
- *
- * Modified for the Marvellous GT64240 Retarded Communication Controller.
- */
-#ifndef _GT64240ETH_H
-#define _GT64240ETH_H
-
-#include <asm/gt64240.h>
-
-#define ETHERNET_PORTS_DIFFERENCE_OFFSETS	0x400
-
-/* Translate those weanie names from Galileo/VxWorks header files: */
-
-#define GT64240_MRR                    MAIN_ROUTING_REGISTER
-#define GT64240_CIU_ARBITER_CONFIG     COMM_UNIT_ARBITER_CONFIGURATION_REGISTER
-#define GT64240_CIU_ARBITER_CONTROL    COMM_UNIT_ARBITER_CONTROL
-#define GT64240_MAIN_LOW_CAUSE         LOW_INTERRUPT_CAUSE_REGISTER
-#define GT64240_MAIN_HIGH_CAUSE        HIGH_INTERRUPT_CAUSE_REGISTER
-#define GT64240_CPU_LOW_MASK           CPU_INTERRUPT_MASK_REGISTER_LOW
-#define GT64240_CPU_HIGH_MASK          CPU_INTERRUPT_MASK_REGISTER_HIGH
-#define GT64240_CPU_SELECT_CAUSE       CPU_SELECT_CAUSE_REGISTER
-
-#define GT64240_ETH_PHY_ADDR_REG       ETHERNET_PHY_ADDRESS_REGISTER
-#define GT64240_ETH_PORT_CONFIG        ETHERNET0_PORT_CONFIGURATION_REGISTER
-#define GT64240_ETH_PORT_CONFIG_EXT    ETHERNET0_PORT_CONFIGURATION_EXTEND_REGISTER
-#define GT64240_ETH_PORT_COMMAND       ETHERNET0_PORT_COMMAND_REGISTER
-#define GT64240_ETH_PORT_STATUS        ETHERNET0_PORT_STATUS_REGISTER
-#define GT64240_ETH_IO_SIZE            ETHERNET_PORTS_DIFFERENCE_OFFSETS
-#define GT64240_ETH_SMI_REG            ETHERNET_SMI_REGISTER
-#define GT64240_ETH_MIB_COUNT_BASE     ETHERNET0_MIB_COUNTER_BASE
-#define GT64240_ETH_SDMA_CONFIG        ETHERNET0_SDMA_CONFIGURATION_REGISTER
-#define GT64240_ETH_SDMA_COMM          ETHERNET0_SDMA_COMMAND_REGISTER
-#define GT64240_ETH_INT_MASK           ETHERNET0_INTERRUPT_MASK_REGISTER
-#define GT64240_ETH_INT_CAUSE          ETHERNET0_INTERRUPT_CAUSE_REGISTER
-#define GT64240_ETH_CURR_TX_DESC_PTR0  ETHERNET0_CURRENT_TX_DESCRIPTOR_POINTER0
-#define GT64240_ETH_CURR_TX_DESC_PTR1  ETHERNET0_CURRENT_TX_DESCRIPTOR_POINTER1
-#define GT64240_ETH_1ST_RX_DESC_PTR0   ETHERNET0_FIRST_RX_DESCRIPTOR_POINTER0
-#define GT64240_ETH_CURR_RX_DESC_PTR0  ETHERNET0_CURRENT_RX_DESCRIPTOR_POINTER0
-#define GT64240_ETH_HASH_TBL_PTR       ETHERNET0_HASH_TABLE_POINTER_REGISTER
-
-/* Turn on NAPI by default */
-
-#define	GT64240_NAPI			1
-
-/* Some 64240 settings that SHOULD eventually be setup in PROM monitor: */
-/* (Board-specific to the DSL3224 Rev A board ONLY!)                    */
-#define D3224_MPP_CTRL0_SETTING		0x66669900
-#define D3224_MPP_CTRL1_SETTING		0x00000000
-#define D3224_MPP_CTRL2_SETTING		0x00887700
-#define D3224_MPP_CTRL3_SETTING		0x00000044
-#define D3224_GPP_IO_CTRL_SETTING	0x0000e800
-#define D3224_GPP_LEVEL_CTRL_SETTING	0xf001f703
-#define D3224_GPP_VALUE_SETTING		0x00000000
-
-/* Keep the ring sizes a power of two for efficiency. */
-//-#define TX_RING_SIZE 16
-#define TX_RING_SIZE	64	/* TESTING !!! */
-#define RX_RING_SIZE	32
-#define PKT_BUF_SZ	1536	/* Size of each temporary Rx buffer. */
-
-#define RX_HASH_TABLE_SIZE 16384
-#define HASH_HOP_NUMBER 12
-
-#define NUM_INTERFACES 3
-
-#define GT64240ETH_TX_TIMEOUT HZ/4
-
-#define MIPS_GT64240_BASE 0xf4000000
-#define GT64240_ETH0_BASE (MIPS_GT64240_BASE + GT64240_ETH_PORT_CONFIG)
-#define GT64240_ETH1_BASE (GT64240_ETH0_BASE + GT64240_ETH_IO_SIZE)
-#define GT64240_ETH2_BASE (GT64240_ETH1_BASE + GT64240_ETH_IO_SIZE)
-
-#if defined(CONFIG_MIPS_DSL3224)
-#define GT64240_ETHER0_IRQ 4
-#define GT64240_ETHER1_IRQ 4
-#else
-#define GT64240_ETHER0_IRQ -1
-#define GT64240_ETHER1_IRQ -1
-#endif
-
-#define REV_GT64240  0x1
-#define REV_GT64240A 0x10
-
-#define GT64240ETH_READ(gp, offset)					\
-	GT_READ((gp)->port_offset + (offset))
-
-#define GT64240ETH_WRITE(gp, offset, data)				\
-	GT_WRITE((gp)->port_offset + (offset), (data))
-
-#define GT64240ETH_SETBIT(gp, offset, bits)				\
-	GT64240ETH_WRITE((gp), (offset),				\
-	                 GT64240ETH_READ((gp), (offset)) | (bits))
-
-#define GT64240ETH_CLRBIT(gp, offset, bits)				\
-	GT64240ETH_WRITE((gp), (offset),				\
-	                 GT64240ETH_READ((gp), (offset)) & ~(bits))
-
-#define GT64240_READ(ofs)		GT_READ(ofs)
-#define GT64240_WRITE(ofs, data)	GT_WRITE((ofs), (data))
-
-/* Bit definitions of the SMI Reg */
-enum {
-	smirDataMask = 0xffff,
-	smirPhyAdMask = 0x1f << 16,
-	smirPhyAdBit = 16,
-	smirRegAdMask = 0x1f << 21,
-	smirRegAdBit = 21,
-	smirOpCode = 1 << 26,
-	smirReadValid = 1 << 27,
-	smirBusy = 1 << 28
-};
-
-/* Bit definitions of the Port Config Reg */
-enum pcr_bits {
-	pcrPM = 1 << 0,
-	pcrRBM = 1 << 1,
-	pcrPBF = 1 << 2,
-	pcrEN = 1 << 7,
-	pcrLPBKMask = 0x3 << 8,
-	pcrLPBKBit = 1 << 8,
-	pcrFC = 1 << 10,
-	pcrHS = 1 << 12,
-	pcrHM = 1 << 13,
-	pcrHDM = 1 << 14,
-	pcrHD = 1 << 15,
-	pcrISLMask = 0x7 << 28,
-	pcrISLBit = 28,
-	pcrACCS = 1 << 31
-};
-
-/* Bit definitions of the Port Config Extend Reg */
-enum pcxr_bits {
-	pcxrIGMP = 1,
-	pcxrSPAN = 2,
-	pcxrPAR = 4,
-	pcxrPRIOtxMask = 0x7 << 3,
-	pcxrPRIOtxBit = 3,
-	pcxrPRIOrxMask = 0x3 << 6,
-	pcxrPRIOrxBit = 6,
-	pcxrPRIOrxOverride = 1 << 8,
-	pcxrDPLXen = 1 << 9,
-	pcxrFCTLen = 1 << 10,
-	pcxrFLP = 1 << 11,
-	pcxrFCTL = 1 << 12,
-	pcxrMFLMask = 0x3 << 14,
-	pcxrMFLBit = 14,
-	pcxrMIBclrMode = 1 << 16,
-	pcxrSpeed = 1 << 18,
-	pcxrSpeeden = 1 << 19,
-	pcxrRMIIen = 1 << 20,
-	pcxrDSCPen = 1 << 21
-};
-
-/* Bit definitions of the Port Command Reg */
-enum pcmr_bits {
-	pcmrFJ = 1 << 15
-};
-
-
-/* Bit definitions of the Port Status Reg */
-enum psr_bits {
-	psrSpeed = 1,
-	psrDuplex = 2,
-	psrFctl = 4,
-	psrLink = 8,
-	psrPause = 1 << 4,
-	psrTxLow = 1 << 5,
-	psrTxHigh = 1 << 6,
-	psrTxInProg = 1 << 7
-};
-
-/* Bit definitions of the SDMA Config Reg */
-enum sdcr_bits {
-	sdcrRCMask = 0xf << 2,
-	sdcrRCBit = 2,
-	sdcrBLMR = 1 << 6,
-	sdcrBLMT = 1 << 7,
-	sdcrPOVR = 1 << 8,
-	sdcrRIFB = 1 << 9,
-	sdcrBSZMask = 0x3 << 12,
-	sdcrBSZBit = 12
-};
-
-/* Bit definitions of the SDMA Command Reg */
-enum sdcmr_bits {
-	sdcmrERD = 1 << 7,
-	sdcmrAR = 1 << 15,
-	sdcmrSTDH = 1 << 16,
-	sdcmrSTDL = 1 << 17,
-	sdcmrTXDH = 1 << 23,
-	sdcmrTXDL = 1 << 24,
-	sdcmrAT = 1 << 31
-};
-
-/* Bit definitions of the Interrupt Cause Reg */
-enum icr_bits {
-	icrRxBuffer = 1,
-	icrTxBufferHigh = 1 << 2,
-	icrTxBufferLow = 1 << 3,
-	icrTxEndHigh = 1 << 6,
-	icrTxEndLow = 1 << 7,
-	icrRxError = 1 << 8,
-	icrTxErrorHigh = 1 << 10,
-	icrTxErrorLow = 1 << 11,
-	icrRxOVR = 1 << 12,
-	icrTxUdr = 1 << 13,
-	icrRxBufferQ0 = 1 << 16,
-	icrRxBufferQ1 = 1 << 17,
-	icrRxBufferQ2 = 1 << 18,
-	icrRxBufferQ3 = 1 << 19,
-	icrRxErrorQ0 = 1 << 20,
-	icrRxErrorQ1 = 1 << 21,
-	icrRxErrorQ2 = 1 << 22,
-	icrRxErrorQ3 = 1 << 23,
-	icrMIIPhySTC = 1 << 28,
-	icrSMIdone = 1 << 29,
-	icrEtherIntSum = 1 << 31
-};
-
-
-/* The Rx and Tx descriptor lists. */
-#ifdef __LITTLE_ENDIAN
-typedef struct {
-	u32 cmdstat;
-	u16 reserved;		//-prk21aug01    u32 reserved:16;
-	u16 byte_cnt;		//-prk21aug01    u32 byte_cnt:16;
-	u32 buff_ptr;
-	u32 next;
-} gt64240_td_t;
-
-typedef struct {
-	u32 cmdstat;
-	u16 byte_cnt;		//-prk21aug01    u32 byte_cnt:16;
-	u16 buff_sz;		//-prk21aug01    u32 buff_sz:16;
-	u32 buff_ptr;
-	u32 next;
-} gt64240_rd_t;
-#elif defined(__BIG_ENDIAN)
-typedef struct {
-	u16 byte_cnt;		//-prk21aug01    u32 byte_cnt:16;
-	u16 reserved;		//-prk21aug01    u32 reserved:16;
-	u32 cmdstat;
-	u32 next;
-	u32 buff_ptr;
-} gt64240_td_t;
-
-typedef struct {
-	u16 buff_sz;		//-prk21aug01    u32 buff_sz:16;
-	u16 byte_cnt;		//-prk21aug01    u32 byte_cnt:16;
-	u32 cmdstat;
-	u32 next;
-	u32 buff_ptr;
-} gt64240_rd_t;
-#else
-#error Either __BIG_ENDIAN or __LITTLE_ENDIAN must be defined!
-#endif
-
-
-/* Values for the Tx command-status descriptor entry. */
-enum td_cmdstat {
-	txOwn = 1 << 31,
-	txAutoMode = 1 << 30,
-	txEI = 1 << 23,
-	txGenCRC = 1 << 22,
-	txPad = 1 << 18,
-	txFirst = 1 << 17,
-	txLast = 1 << 16,
-	txErrorSummary = 1 << 15,
-	txReTxCntMask = 0x0f << 10,
-	txReTxCntBit = 10,
-	txCollision = 1 << 9,
-	txReTxLimit = 1 << 8,
-	txUnderrun = 1 << 6,
-	txLateCollision = 1 << 5
-};
-
-
-/* Values for the Rx command-status descriptor entry. */
-enum rd_cmdstat {
-	rxOwn = 1 << 31,
-	rxAutoMode = 1 << 30,
-	rxEI = 1 << 23,
-	rxFirst = 1 << 17,
-	rxLast = 1 << 16,
-	rxErrorSummary = 1 << 15,
-	rxIGMP = 1 << 14,
-	rxHashExpired = 1 << 13,
-	rxMissedFrame = 1 << 12,
-	rxFrameType = 1 << 11,
-	rxShortFrame = 1 << 8,
-	rxMaxFrameLen = 1 << 7,
-	rxOverrun = 1 << 6,
-	rxCollision = 1 << 4,
-	rxCRCError = 1
-};
-
-/* Bit fields of a Hash Table Entry */
-enum hash_table_entry {
-	hteValid = 1,
-	hteSkip = 2,
-	hteRD = 4
-};
-
-// The MIB counters
-typedef struct {
-	u32 byteReceived;
-	u32 byteSent;
-	u32 framesReceived;
-	u32 framesSent;
-	u32 totalByteReceived;
-	u32 totalFramesReceived;
-	u32 broadcastFramesReceived;
-	u32 multicastFramesReceived;
-	u32 cRCError;
-	u32 oversizeFrames;
-	u32 fragments;
-	u32 jabber;
-	u32 collision;
-	u32 lateCollision;
-	u32 frames64;
-	u32 frames65_127;
-	u32 frames128_255;
-	u32 frames256_511;
-	u32 frames512_1023;
-	u32 frames1024_MaxSize;
-	u32 macRxError;
-	u32 droppedFrames;
-	u32 outMulticastFrames;
-	u32 outBroadcastFrames;
-	u32 undersizeFrames;
-} mib_counters_t;
-
-
-struct gt64240_private {
-	gt64240_rd_t *rx_ring;
-	gt64240_td_t *tx_ring;
-	// The Rx and Tx rings must be 16-byte aligned
-	dma_addr_t rx_ring_dma;
-	dma_addr_t tx_ring_dma;
-	char *hash_table;
-	// The Hash Table must be 8-byte aligned
-	dma_addr_t hash_table_dma;
-	int hash_mode;
-
-	// The Rx buffers must be 8-byte aligned
-	char *rx_buff;
-	dma_addr_t rx_buff_dma;
-	// Tx buffers (tx_skbuff[i]->data) with less than 8 bytes
-	// of payload must be 8-byte aligned
-	struct sk_buff *tx_skbuff[TX_RING_SIZE];
-	int rx_next_out;	/* The next free ring entry to receive */
-	int tx_next_in;		/* The next free ring entry to send */
-	int tx_next_out;	/* The last ring entry the ISR processed */
-	int tx_count;		/* current # of pkts waiting to be sent in Tx ring */
-	int intr_work_done;	/* number of Rx and Tx pkts processed in the isr */
-	int tx_full;		/* Tx ring is full */
-
-	mib_counters_t mib;
-	struct net_device_stats stats;
-
-	int io_size;
-	int port_num;		// 0 or 1
-	u32 port_offset;
-
-	int phy_addr;		// PHY address
-	u32 last_psr;		// last value of the port status register
-
-	int options;		/* User-settable misc. driver options. */
-	int drv_flags;
-	spinlock_t lock;	/* Serialise access to device */
-	struct mii_if_info mii_if;
-
-	u32 msg_enable;
-};
-
-#endif /* _GT64240ETH_H */
-- 
cgit v1.2.1


From 94dbffd540eea601aecad07e2df5bfd8a46672f3 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:52:15 -0700
Subject: [PATCH] bonding: Allow bonding to enslave a 10 Gig adapter

Allow channel bonding to enslave a 10 Gig adapter without errors.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Acked-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_3ad.c  | 11 ++++++++++-
 drivers/net/bonding/bond_main.c |  1 +
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 6a407070c2e8..c24b20aaf57f 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -85,6 +85,7 @@
 #define     AD_LINK_SPEED_BITMASK_10MBPS      0x2
 #define     AD_LINK_SPEED_BITMASK_100MBPS     0x4
 #define     AD_LINK_SPEED_BITMASK_1000MBPS    0x8
+#define     AD_LINK_SPEED_BITMASK_10000MBPS   0x10
 //endalloun
 
 // compare MAC addresses
@@ -330,7 +331,8 @@ static inline void __release_rx_machine_lock(struct port *port)
  *     0,
  *     %AD_LINK_SPEED_BITMASK_10MBPS,
  *     %AD_LINK_SPEED_BITMASK_100MBPS,
- *     %AD_LINK_SPEED_BITMASK_1000MBPS
+ *     %AD_LINK_SPEED_BITMASK_1000MBPS,
+ *     %AD_LINK_SPEED_BITMASK_10000MBPS
  */
 static u16 __get_link_speed(struct port *port)
 {
@@ -357,6 +359,10 @@ static u16 __get_link_speed(struct port *port)
 			speed = AD_LINK_SPEED_BITMASK_1000MBPS;
 			break;
 
+		case SPEED_10000:
+			speed = AD_LINK_SPEED_BITMASK_10000MBPS;
+			break;
+
 		default:
 			speed = 0; // unknown speed value from ethtool. shouldn't happen
 			break;
@@ -775,6 +781,9 @@ static u32 __get_agg_bandwidth(struct aggregator *aggregator)
 		case AD_LINK_SPEED_BITMASK_1000MBPS:
 			bandwidth = aggregator->num_of_ports * 1000;
 			break;
+		case AD_LINK_SPEED_BITMASK_10000MBPS:
+			bandwidth = aggregator->num_of_ports * 10000;
+			break;
 		default:
 			bandwidth=0; // to silent the compilor ....
 		}
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 850aae21a2fe..13b434220ff6 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -638,6 +638,7 @@ verify:
 	case SPEED_10:
 	case SPEED_100:
 	case SPEED_1000:
+	case SPEED_10000:
 		break;
 	default:
 		return -1;
-- 
cgit v1.2.1


From 8bb5f96b0c1f430e6be56edd6c7032bcedd86ff0 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:52:51 -0700
Subject: [PATCH] bonding: Convert delay value from s16 to int

The value of "downdelay/miimon" and "updelay/miimon" are stored in
slave->delay. The type of downdelay, updelay, and miimon are all int.
However, slave->delay is type short, and it is not possible to store the
value of "downdelay/miimon" or "updelay/miimon" in some cases. (For example,
miimon=1 downdelay=32768)

The attached patch fixes this problem.

Signed-off-by: Kenzo Iwami <k-iwami@cj.jp.nec.com>
Acked-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bonding.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 0bdfe2c71453..17caafe58247 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -149,7 +149,7 @@ struct slave {
 	struct net_device *dev; /* first - useful for panic debug */
 	struct slave *next;
 	struct slave *prev;
-	s16    delay;
+	int    delay;
 	u32    jiffies;
 	s8     link;    /* one of BOND_LINK_XXXX */
 	s8     state;   /* one of BOND_STATE_XXXX */
-- 
cgit v1.2.1


From 65509645ae05886eccc81b8a453afea07f0eabb6 Mon Sep 17 00:00:00 2001
From: Kenzo Iwami <k-iwami@cj.jp.nec.com>
Date: Fri, 22 Sep 2006 21:53:08 -0700
Subject: [PATCH] bonding: Format fix in seq_printf call

Though link_failure_count is type unsigned int, this value is outputted to
/proc/net/bonding/bondX file using "%d" instead of "%u".

The attached patch fixes this problem.

Signed-off-by: Kenzo Iwami <k-iwami@cj.jp.nec.com>
Acked-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 13b434220ff6..3d7693d1c512 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2951,7 +2951,7 @@ static void bond_info_show_slave(struct seq_file *seq, const struct slave *slave
 	seq_printf(seq, "\nSlave Interface: %s\n", slave->dev->name);
 	seq_printf(seq, "MII Status: %s\n",
 		   (slave->link == BOND_LINK_UP) ?  "up" : "down");
-	seq_printf(seq, "Link Failure Count: %d\n",
+	seq_printf(seq, "Link Failure Count: %u\n",
 		   slave->link_failure_count);
 
 	seq_printf(seq,
-- 
cgit v1.2.1


From a50d8de2cc872818b61e60c20c75be3f19aa6887 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:53:25 -0700
Subject: [PATCH] bonding: Remove unneeded NULL test

Remove unneeded test for NULL.  Reported by Thomas Dillig
<tdillig@stanford.edu> and Isil Dillig <isil@stanford.edu> via Stephen
Hemminger <shemminger@osdl.org>.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_sysfs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index cfe4dc3a93a3..15b6a29bb4d4 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -1093,7 +1093,7 @@ static ssize_t bonding_store_active_slave(struct class_device *cd, const char *b
 			     strlen(slave->dev->name)) == 0) {
         			old_active = bond->curr_active_slave;
         			new_active = slave;
-        			if (new_active && (new_active == old_active)) {
+        			if (new_active == old_active) {
 					/* do nothing */
 					printk(KERN_INFO DRV_NAME
 				       	       ": %s: %s is already the current active slave.\n",
-- 
cgit v1.2.1


From 54ef313714070b397d3857289f0fd099b7643631 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:53:39 -0700
Subject: [PATCH] bonding: Handle large hard_header_len

The bonding driver fails to adjust its hard_header_len when enslaving
interfaces.  Whenever an interface with a hard_header_len greater than the
ETH_HLEN default is enslaved, the potential for an oops exists, and if the
oops happens while responding to an arp request, for example, the system
panics.  GIANFAR devices may use an extended hard_header for VLAN or
hardware checksumming.  Enslaving such a device and then transmitting over
it causes a kernel panic.

Patch modified from submitter's original, but submitter agreed with this
patch in private email.

Signed-off-by: Mark Huth <mhuth@mvista.com>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 3d7693d1c512..d2f460b0dbab 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1211,10 +1211,14 @@ static int bond_compute_features(struct bonding *bond)
 	unsigned long features = BOND_INTERSECT_FEATURES;
 	struct slave *slave;
 	struct net_device *bond_dev = bond->dev;
+	unsigned short max_hard_header_len = ETH_HLEN;
 	int i;
 
-	bond_for_each_slave(bond, slave, i)
+	bond_for_each_slave(bond, slave, i) {
 		features &= (slave->dev->features & BOND_INTERSECT_FEATURES);
+		if (slave->dev->hard_header_len > max_hard_header_len)
+			max_hard_header_len = slave->dev->hard_header_len;
+	}
 
 	if ((features & NETIF_F_SG) && 
 	    !(features & NETIF_F_ALL_CSUM))
@@ -1232,6 +1236,7 @@ static int bond_compute_features(struct bonding *bond)
 
 	features |= (bond_dev->features & ~BOND_INTERSECT_FEATURES);
 	bond_dev->features = features;
+	bond_dev->hard_header_len = max_hard_header_len;
 
 	return 0;
 }
-- 
cgit v1.2.1


From 0b680e753724d31a9c45f059d1aad29df54584a1 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:54:10 -0700
Subject: [PATCH] bonding: Add priv_flag to avoid event mishandling

Add priv_flag to specifically identify bonding-involved devices.  Needed
because IFF_MASTER is an unreliable identifier (vlan interfaces above bonding
will inherit IFF_MASTER).  Misidentification of devices would cause
notifier events for other devices to be erroneously processed by bonding,
causing various havoc.

Bug discovered by Martin Papik <martin.papik@ipsec.info>; this patch is
modified from his original.

Signed-off-by: Martin Papik <martin.papik@ipsec.info>
Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 7 ++++++-
 include/linux/if.h              | 1 +
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d2f460b0dbab..9e5a533a1622 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1371,6 +1371,7 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 	}
 
 	new_slave->dev = slave_dev;
+	slave_dev->priv_flags |= IFF_BONDING;
 
 	if ((bond->params.mode == BOND_MODE_TLB) ||
 	    (bond->params.mode == BOND_MODE_ALB)) {
@@ -1784,7 +1785,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 	dev_set_mac_address(slave_dev, &addr);
 
 	slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
-				   IFF_SLAVE_INACTIVE);
+				   IFF_SLAVE_INACTIVE | IFF_BONDING);
 
 	kfree(slave);
 
@@ -3216,6 +3217,9 @@ static int bond_netdev_event(struct notifier_block *this, unsigned long event, v
 		(event_dev ? event_dev->name : "None"),
 		event);
 
+	if (!(event_dev->priv_flags & IFF_BONDING))
+		return NOTIFY_DONE;
+
 	if (event_dev->flags & IFF_MASTER) {
 		dprintk("IFF_MASTER\n");
 		return bond_master_netdev_event(event, event_dev);
@@ -4185,6 +4189,7 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params)
 	/* Initialize the device options */
 	bond_dev->tx_queue_len = 0;
 	bond_dev->flags |= IFF_MASTER|IFF_MULTICAST;
+	bond_dev->priv_flags |= IFF_BONDING;
 
 	/* At first, we block adding VLANs. That's the only way to
 	 * prevent problems that occur when adding VLANs over an
diff --git a/include/linux/if.h b/include/linux/if.h
index cd080d765324..a023ec1274fe 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -59,6 +59,7 @@
 #define IFF_SLAVE_INACTIVE	0x4	/* bonding slave not the curr. active */
 #define IFF_MASTER_8023AD	0x8	/* bonding master, 802.3ad. 	*/
 #define IFF_MASTER_ALB	0x10		/* bonding master, balance-alb.	*/
+#define IFF_BONDING	0x20		/* bonding master or slave	*/
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
-- 
cgit v1.2.1


From 70298705bb29fb7982b85089adf17cd37b94baa7 Mon Sep 17 00:00:00 2001
From: jamal <hadi@cyberus.ca>
Date: Fri, 22 Sep 2006 21:54:37 -0700
Subject: [PATCH] bonding: Don't release slaves when master is admin down

When a bonding netdevice is admin-ed down it loses the slaves
attributes (set via ifenslave). This is not consistent with other
behavior of netdevices (example a qdisc attached to a netdevice doesnt
disappear or an attached IP address etc).
The included patch fixes this. Ive tested by ifenslaving, downing the
bond, checking /proc and making sure it still has the slaves, up-ing the
bond and making sure things continue to work.

Jay/Bonding folks if you are ok with it, just ACK it or include it in
your tree etc. Otherwise we can discuss.

Acked-by: Jay Vosburgh <fubar@us.ibm.com>

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 9e5a533a1622..bafe62f7c9b7 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -3430,7 +3430,6 @@ static int bond_close(struct net_device *bond_dev)
 
 	write_lock_bh(&bond->lock);
 
-	bond_mc_list_destroy(bond);
 
 	/* signal timers not to re-arm */
 	bond->kill_timers = 1;
@@ -3461,8 +3460,6 @@ static int bond_close(struct net_device *bond_dev)
 		break;
 	}
 
-	/* Release the bonded slaves */
-	bond_release_all(bond_dev);
 
 	if ((bond->params.mode == BOND_MODE_TLB) ||
 	    (bond->params.mode == BOND_MODE_ALB)) {
@@ -4248,6 +4245,9 @@ static void bond_free_all(void)
 	list_for_each_entry_safe(bond, nxt, &bond_dev_list, bond_list) {
 		struct net_device *bond_dev = bond->dev;
 
+		bond_mc_list_destroy(bond);
+		/* Release the bonded slaves */
+		bond_release_all(bond_dev);
 		unregister_netdevice(bond_dev);
 		bond_deinit(bond_dev);
 	}
-- 
cgit v1.2.1


From f5b2b966f032f22d3a289045a5afd4afa09f09c6 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:54:53 -0700
Subject: [PATCH] bonding: Validate probe replies in ARP monitor

	Add logic to check ARP request / reply packets used for ARP
monitor link integrity checking.

	The current method simply examines the slave device to see if it
has sent and received traffic; this can be fooled by extraneous traffic.
For example, if multiple hosts running bonding are behind a common
switch, the probe traffic from the multiple instances of bonding will
update the tx/rx times on each other's slave devices.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 Documentation/networking/bonding.txt |  59 ++++++++++++
 drivers/net/bonding/bond_main.c      | 182 +++++++++++++++++++++++++++++++++--
 drivers/net/bonding/bond_sysfs.c     |  54 +++++++++++
 drivers/net/bonding/bonding.h        |  32 +++++-
 include/linux/if.h                   |   1 +
 include/linux/netdevice.h            |   7 +-
 6 files changed, 325 insertions(+), 10 deletions(-)

diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index afac780445cd..dc942eaf490f 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -192,6 +192,17 @@ or, for backwards compatibility, the option value.  E.g.,
 arp_interval
 
 	Specifies the ARP link monitoring frequency in milliseconds.
+
+	The ARP monitor works by periodically checking the slave
+	devices to determine whether they have sent or received
+	traffic recently (the precise criteria depends upon the
+	bonding mode, and the state of the slave).  Regular traffic is
+	generated via ARP probes issued for the addresses specified by
+	the arp_ip_target option.
+
+	This behavior can be modified by the arp_validate option,
+	below.
+
 	If ARP monitoring is used in an etherchannel compatible mode
 	(modes 0 and 2), the switch should be configured in a mode
 	that evenly distributes packets across all links. If the
@@ -213,6 +224,54 @@ arp_ip_target
 	maximum number of targets that can be specified is 16.  The
 	default value is no IP addresses.
 
+arp_validate
+
+	Specifies whether or not ARP probes and replies should be
+	validated in the active-backup mode.  This causes the ARP
+	monitor to examine the incoming ARP requests and replies, and
+	only consider a slave to be up if it is receiving the
+	appropriate ARP traffic.
+
+	Possible values are:
+
+	none or 0
+
+		No validation is performed.  This is the default.
+
+	active or 1
+
+		Validation is performed only for the active slave.
+
+	backup or 2
+
+		Validation is performed only for backup slaves.
+
+	all or 3
+
+		Validation is performed for all slaves.
+
+	For the active slave, the validation checks ARP replies to
+	confirm that they were generated by an arp_ip_target.  Since
+	backup slaves do not typically receive these replies, the
+	validation performed for backup slaves is on the ARP request
+	sent out via the active slave.  It is possible that some
+	switch or network configurations may result in situations
+	wherein the backup slaves do not receive the ARP requests; in
+	such a situation, validation of backup slaves must be
+	disabled.
+
+	This option is useful in network configurations in which
+	multiple bonding hosts are concurrently issuing ARPs to one or
+	more targets beyond a common switch.  Should the link between
+	the switch and target fail (but not the switch itself), the
+	probe traffic generated by the multiple bonding instances will
+	fool the standard ARP monitor into considering the links as
+	still up.  Use of the arp_validate option can resolve this, as
+	the ARP monitor will only consider ARP requests and replies
+	associated with its own instance of bonding.
+
+	This option was added in bonding version 3.1.0.
+
 downdelay
 
 	Specifies the time, in milliseconds, to wait before disabling
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index bafe62f7c9b7..fd521b05db83 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -96,6 +96,7 @@ static char *lacp_rate	= NULL;
 static char *xmit_hash_policy = NULL;
 static int arp_interval = BOND_LINK_ARP_INTERV;
 static char *arp_ip_target[BOND_MAX_ARP_TARGETS] = { NULL, };
+static char *arp_validate = NULL;
 struct bond_params bonding_defaults;
 
 module_param(max_bonds, int, 0);
@@ -127,6 +128,8 @@ module_param(arp_interval, int, 0);
 MODULE_PARM_DESC(arp_interval, "arp interval in milliseconds");
 module_param_array(arp_ip_target, charp, NULL, 0);
 MODULE_PARM_DESC(arp_ip_target, "arp targets in n.n.n.n form");
+module_param(arp_validate, charp, 0);
+MODULE_PARM_DESC(arp_validate, "validate src/dst of ARP probes: none (default), active, backup or all");
 
 /*----------------------------- Global variables ----------------------------*/
 
@@ -170,6 +173,14 @@ struct bond_parm_tbl xmit_hashtype_tbl[] = {
 {	NULL,			-1},
 };
 
+struct bond_parm_tbl arp_validate_tbl[] = {
+{	"none",			BOND_ARP_VALIDATE_NONE},
+{	"active",		BOND_ARP_VALIDATE_ACTIVE},
+{	"backup",		BOND_ARP_VALIDATE_BACKUP},
+{	"all",			BOND_ARP_VALIDATE_ALL},
+{	NULL,			-1},
+};
+
 /*-------------------------- Forward declarations ---------------------------*/
 
 static void bond_send_gratuitous_arp(struct bonding *bond);
@@ -1424,6 +1435,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 
 	bond_compute_features(bond);
 
+	new_slave->last_arp_rx = jiffies;
+
 	if (bond->params.miimon && !bond->params.use_carrier) {
 		link_reporting = bond_check_dev_link(bond, slave_dev, 1);
 
@@ -1785,7 +1798,8 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 	dev_set_mac_address(slave_dev, &addr);
 
 	slave_dev->priv_flags &= ~(IFF_MASTER_8023AD | IFF_MASTER_ALB |
-				   IFF_SLAVE_INACTIVE | IFF_BONDING);
+				   IFF_SLAVE_INACTIVE | IFF_BONDING |
+				   IFF_SLAVE_NEEDARP);
 
 	kfree(slave);
 
@@ -2298,6 +2312,25 @@ static int bond_has_ip(struct bonding *bond)
 	return 0;
 }
 
+static int bond_has_this_ip(struct bonding *bond, u32 ip)
+{
+	struct vlan_entry *vlan, *vlan_next;
+
+	if (ip == bond->master_ip)
+		return 1;
+
+	if (list_empty(&bond->vlan_list))
+		return 0;
+
+	list_for_each_entry_safe(vlan, vlan_next, &bond->vlan_list,
+				 vlan_list) {
+		if (ip == vlan->vlan_ip)
+			return 1;
+	}
+
+	return 0;
+}
+
 /*
  * We go to the (large) trouble of VLAN tagging ARP frames because
  * switches in VLAN mode (especially if ports are configured as
@@ -2436,6 +2469,93 @@ static void bond_send_gratuitous_arp(struct bonding *bond)
 	}
 }
 
+static void bond_validate_arp(struct bonding *bond, struct slave *slave, u32 sip, u32 tip)
+{
+	int i;
+	u32 *targets = bond->params.arp_targets;
+
+	targets = bond->params.arp_targets;
+	for (i = 0; (i < BOND_MAX_ARP_TARGETS) && targets[i]; i++) {
+		dprintk("bva: sip %u.%u.%u.%u tip %u.%u.%u.%u t[%d] "
+			"%u.%u.%u.%u bhti(tip) %d\n",
+		       NIPQUAD(sip), NIPQUAD(tip), i, NIPQUAD(targets[i]),
+		       bond_has_this_ip(bond, tip));
+		if (sip == targets[i]) {
+			if (bond_has_this_ip(bond, tip))
+				slave->last_arp_rx = jiffies;
+			return;
+		}
+	}
+}
+
+static int bond_arp_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev)
+{
+	struct arphdr *arp;
+	struct slave *slave;
+	struct bonding *bond;
+	unsigned char *arp_ptr;
+	u32 sip, tip;
+
+	if (!(dev->priv_flags & IFF_BONDING) || !(dev->flags & IFF_MASTER))
+		goto out;
+
+	bond = dev->priv;
+	read_lock(&bond->lock);
+
+	dprintk("bond_arp_rcv: bond %s skb->dev %s orig_dev %s\n",
+		bond->dev->name, skb->dev ? skb->dev->name : "NULL",
+		orig_dev ? orig_dev->name : "NULL");
+
+	slave = bond_get_slave_by_dev(bond, orig_dev);
+	if (!slave || !slave_do_arp_validate(bond, slave))
+		goto out_unlock;
+
+	/* ARP header, plus 2 device addresses, plus 2 IP addresses.  */
+	if (!pskb_may_pull(skb, (sizeof(struct arphdr) +
+				 (2 * dev->addr_len) +
+				 (2 * sizeof(u32)))))
+		goto out_unlock;
+
+	arp = skb->nh.arph;
+	if (arp->ar_hln != dev->addr_len ||
+	    skb->pkt_type == PACKET_OTHERHOST ||
+	    skb->pkt_type == PACKET_LOOPBACK ||
+	    arp->ar_hrd != htons(ARPHRD_ETHER) ||
+	    arp->ar_pro != htons(ETH_P_IP) ||
+	    arp->ar_pln != 4)
+		goto out_unlock;
+
+	arp_ptr = (unsigned char *)(arp + 1);
+	arp_ptr += dev->addr_len;
+	memcpy(&sip, arp_ptr, 4);
+	arp_ptr += 4 + dev->addr_len;
+	memcpy(&tip, arp_ptr, 4);
+
+	dprintk("bond_arp_rcv: %s %s/%d av %d sv %d sip %u.%u.%u.%u"
+		" tip %u.%u.%u.%u\n", bond->dev->name, slave->dev->name,
+		slave->state, bond->params.arp_validate,
+		slave_do_arp_validate(bond, slave), NIPQUAD(sip), NIPQUAD(tip));
+
+	/*
+	 * Backup slaves won't see the ARP reply, but do come through
+	 * here for each ARP probe (so we swap the sip/tip to validate
+	 * the probe).  In a "redundant switch, common router" type of
+	 * configuration, the ARP probe will (hopefully) travel from
+	 * the active, through one switch, the router, then the other
+	 * switch before reaching the backup.
+	 */
+	if (slave->state == BOND_STATE_ACTIVE)
+		bond_validate_arp(bond, slave, sip, tip);
+	else
+		bond_validate_arp(bond, slave, tip, sip);
+
+out_unlock:
+	read_unlock(&bond->lock);
+out:
+	dev_kfree_skb(skb);
+	return NET_RX_SUCCESS;
+}
+
 /*
  * this function is called regularly to monitor each slave's link
  * ensuring that traffic is being sent and received when arp monitoring
@@ -2600,7 +2720,8 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 	 */
 	bond_for_each_slave(bond, slave, i) {
 		if (slave->link != BOND_LINK_UP) {
-			if ((jiffies - slave->dev->last_rx) <= delta_in_ticks) {
+			if ((jiffies - slave_last_rx(bond, slave)) <=
+			     delta_in_ticks) {
 
 				slave->link = BOND_LINK_UP;
 
@@ -2645,7 +2766,7 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 
 			if ((slave != bond->curr_active_slave) &&
 			    (!bond->current_arp_slave) &&
-			    (((jiffies - slave->dev->last_rx) >= 3*delta_in_ticks) &&
+			    (((jiffies - slave_last_rx(bond, slave)) >= 3*delta_in_ticks) &&
 			     bond_has_ip(bond))) {
 				/* a backup slave has gone down; three times
 				 * the delta allows the current slave to be
@@ -2692,7 +2813,7 @@ void bond_activebackup_arp_mon(struct net_device *bond_dev)
 		 * if it is up and needs to take over as the curr_active_slave
 		 */
 		if ((((jiffies - slave->dev->trans_start) >= (2*delta_in_ticks)) ||
-	    (((jiffies - slave->dev->last_rx) >= (2*delta_in_ticks)) &&
+	    (((jiffies - slave_last_rx(bond, slave)) >= (2*delta_in_ticks)) &&
 	     bond_has_ip(bond))) &&
 		    ((jiffies - slave->jiffies) >= 2*delta_in_ticks)) {
 
@@ -3315,6 +3436,21 @@ static void bond_unregister_lacpdu(struct bonding *bond)
 	dev_remove_pack(&(BOND_AD_INFO(bond).ad_pkt_type));
 }
 
+void bond_register_arp(struct bonding *bond)
+{
+	struct packet_type *pt = &bond->arp_mon_pt;
+
+	pt->type = htons(ETH_P_ARP);
+	pt->dev = NULL; /*bond->dev;XXX*/
+	pt->func = bond_arp_rcv;
+	dev_add_pack(pt);
+}
+
+void bond_unregister_arp(struct bonding *bond)
+{
+	dev_remove_pack(&bond->arp_mon_pt);
+}
+
 /*---------------------------- Hashing Policies -----------------------------*/
 
 /*
@@ -3401,6 +3537,9 @@ static int bond_open(struct net_device *bond_dev)
 		} else {
 			arp_timer->function = (void *)&bond_loadbalance_arp_mon;
 		}
+		if (bond->params.arp_validate)
+			bond_register_arp(bond);
+
 		add_timer(arp_timer);
 	}
 
@@ -3428,6 +3567,9 @@ static int bond_close(struct net_device *bond_dev)
 		bond_unregister_lacpdu(bond);
 	}
 
+	if (bond->params.arp_validate)
+		bond_unregister_arp(bond);
+
 	write_lock_bh(&bond->lock);
 
 
@@ -4281,6 +4423,8 @@ int bond_parse_parm(char *mode_arg, struct bond_parm_tbl *tbl)
 
 static int bond_check_params(struct bond_params *params)
 {
+	int arp_validate_value;
+
 	/*
 	 * Convert string parameters.
 	 */
@@ -4484,6 +4628,29 @@ static int bond_check_params(struct bond_params *params)
 		arp_interval = 0;
 	}
 
+	if (arp_validate) {
+		if (bond_mode != BOND_MODE_ACTIVEBACKUP) {
+			printk(KERN_ERR DRV_NAME
+	       ": arp_validate only supported in active-backup mode\n");
+			return -EINVAL;
+		}
+		if (!arp_interval) {
+			printk(KERN_ERR DRV_NAME
+			       ": arp_validate requires arp_interval\n");
+			return -EINVAL;
+		}
+
+		arp_validate_value = bond_parse_parm(arp_validate,
+						     arp_validate_tbl);
+		if (arp_validate_value == -1) {
+			printk(KERN_ERR DRV_NAME
+			       ": Error: invalid arp_validate \"%s\"\n",
+			       arp_validate == NULL ? "NULL" : arp_validate);
+			return -EINVAL;
+		}
+	} else
+		arp_validate_value = 0;
+
 	if (miimon) {
 		printk(KERN_INFO DRV_NAME
 		       ": MII link monitoring set to %d ms\n",
@@ -4492,8 +4659,10 @@ static int bond_check_params(struct bond_params *params)
 		int i;
 
 		printk(KERN_INFO DRV_NAME
-		       ": ARP monitoring set to %d ms with %d target(s):",
-		       arp_interval, arp_ip_count);
+		       ": ARP monitoring set to %d ms, validate %s, with %d target(s):",
+		       arp_interval,
+		       arp_validate_tbl[arp_validate_value].modename,
+		       arp_ip_count);
 
 		for (i = 0; i < arp_ip_count; i++)
 			printk (" %s", arp_ip_target[i]);
@@ -4527,6 +4696,7 @@ static int bond_check_params(struct bond_params *params)
 	params->xmit_policy = xmit_hashtype;
 	params->miimon = miimon;
 	params->arp_interval = arp_interval;
+	params->arp_validate = arp_validate_value;
 	params->updelay = updelay;
 	params->downdelay = downdelay;
 	params->use_carrier = use_carrier;
diff --git a/drivers/net/bonding/bond_sysfs.c b/drivers/net/bonding/bond_sysfs.c
index 15b6a29bb4d4..ced9ed8f995a 100644
--- a/drivers/net/bonding/bond_sysfs.c
+++ b/drivers/net/bonding/bond_sysfs.c
@@ -51,6 +51,7 @@ extern struct bond_params bonding_defaults;
 extern struct bond_parm_tbl bond_mode_tbl[];
 extern struct bond_parm_tbl bond_lacp_tbl[];
 extern struct bond_parm_tbl xmit_hashtype_tbl[];
+extern struct bond_parm_tbl arp_validate_tbl[];
 
 static int expected_refcount = -1;
 static struct class *netdev_class;
@@ -502,6 +503,53 @@ out:
 }
 static CLASS_DEVICE_ATTR(xmit_hash_policy, S_IRUGO | S_IWUSR, bonding_show_xmit_hash, bonding_store_xmit_hash);
 
+/*
+ * Show and set arp_validate.
+ */
+static ssize_t bonding_show_arp_validate(struct class_device *cd, char *buf)
+{
+	struct bonding *bond = to_bond(cd);
+
+	return sprintf(buf, "%s %d\n",
+		       arp_validate_tbl[bond->params.arp_validate].modename,
+		       bond->params.arp_validate) + 1;
+}
+
+static ssize_t bonding_store_arp_validate(struct class_device *cd, const char *buf, size_t count)
+{
+	int new_value;
+	struct bonding *bond = to_bond(cd);
+
+	new_value = bond_parse_parm((char *)buf, arp_validate_tbl);
+	if (new_value < 0) {
+		printk(KERN_ERR DRV_NAME
+		       ": %s: Ignoring invalid arp_validate value %s\n",
+		       bond->dev->name, buf);
+		return -EINVAL;
+	}
+	if (new_value && (bond->params.mode != BOND_MODE_ACTIVEBACKUP)) {
+		printk(KERN_ERR DRV_NAME
+		       ": %s: arp_validate only supported in active-backup mode.\n",
+		       bond->dev->name);
+		return -EINVAL;
+	}
+	printk(KERN_INFO DRV_NAME ": %s: setting arp_validate to %s (%d).\n",
+	       bond->dev->name, arp_validate_tbl[new_value].modename,
+	       new_value);
+
+	if (!bond->params.arp_validate && new_value) {
+		bond_register_arp(bond);
+	} else if (bond->params.arp_validate && !new_value) {
+		bond_unregister_arp(bond);
+	}
+
+	bond->params.arp_validate = new_value;
+
+	return count;
+}
+
+static CLASS_DEVICE_ATTR(arp_validate, S_IRUGO | S_IWUSR, bonding_show_arp_validate, bonding_store_arp_validate);
+
 /*
  * Show and set the arp timer interval.  There are two tricky bits
  * here.  First, if ARP monitoring is activated, then we must disable
@@ -914,6 +962,11 @@ static ssize_t bonding_store_miimon(struct class_device *cd, const char *buf, si
 			       "ARP monitoring. Disabling ARP monitoring...\n",
 			       bond->dev->name);
 			bond->params.arp_interval = 0;
+			if (bond->params.arp_validate) {
+				bond_unregister_arp(bond);
+				bond->params.arp_validate =
+					BOND_ARP_VALIDATE_NONE;
+			}
 			/* Kill ARP timer, else it brings bond's link down */
 			if (bond->mii_timer.function) {
 				printk(KERN_INFO DRV_NAME
@@ -1273,6 +1326,7 @@ static CLASS_DEVICE_ATTR(ad_partner_mac, S_IRUGO, bonding_show_ad_partner_mac, N
 static struct attribute *per_bond_attrs[] = {
 	&class_device_attr_slaves.attr,
 	&class_device_attr_mode.attr,
+	&class_device_attr_arp_validate.attr,
 	&class_device_attr_arp_interval.attr,
 	&class_device_attr_arp_ip_target.attr,
 	&class_device_attr_downdelay.attr,
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 17caafe58247..db16fee40a5f 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -22,8 +22,8 @@
 #include "bond_3ad.h"
 #include "bond_alb.h"
 
-#define DRV_VERSION	"3.0.3"
-#define DRV_RELDATE	"March 23, 2006"
+#define DRV_VERSION	"3.1.0-test"
+#define DRV_RELDATE	"September 9, 2006"
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"
 
@@ -126,6 +126,7 @@ struct bond_params {
 	int xmit_policy;
 	int miimon;
 	int arp_interval;
+	int arp_validate;
 	int use_carrier;
 	int updelay;
 	int downdelay;
@@ -151,6 +152,7 @@ struct slave {
 	struct slave *prev;
 	int    delay;
 	u32    jiffies;
+	u32    last_arp_rx;
 	s8     link;    /* one of BOND_LINK_XXXX */
 	s8     state;   /* one of BOND_STATE_XXXX */
 	u32    original_flags;
@@ -198,6 +200,7 @@ struct bonding {
 	struct   bond_params params;
 	struct   list_head vlan_list;
 	struct   vlan_group *vlgrp;
+	struct   packet_type arp_mon_pt;
 };
 
 /**
@@ -228,6 +231,25 @@ static inline struct bonding *bond_get_bond_by_slave(struct slave *slave)
 	return (struct bonding *)slave->dev->master->priv;
 }
 
+#define BOND_ARP_VALIDATE_NONE		0
+#define BOND_ARP_VALIDATE_ACTIVE	(1 << BOND_STATE_ACTIVE)
+#define BOND_ARP_VALIDATE_BACKUP	(1 << BOND_STATE_BACKUP)
+#define BOND_ARP_VALIDATE_ALL		(BOND_ARP_VALIDATE_ACTIVE | \
+					 BOND_ARP_VALIDATE_BACKUP)
+
+extern inline int slave_do_arp_validate(struct bonding *bond, struct slave *slave)
+{
+	return bond->params.arp_validate & (1 << slave->state);
+}
+
+extern inline u32 slave_last_rx(struct bonding *bond, struct slave *slave)
+{
+	if (slave_do_arp_validate(bond, slave))
+		return slave->last_arp_rx;
+
+	return slave->dev->last_rx;
+}
+
 static inline void bond_set_slave_inactive_flags(struct slave *slave)
 {
 	struct bonding *bond = slave->dev->master->priv;
@@ -235,12 +257,14 @@ static inline void bond_set_slave_inactive_flags(struct slave *slave)
 	    bond->params.mode != BOND_MODE_ALB)
 		slave->state = BOND_STATE_BACKUP;
 	slave->dev->priv_flags |= IFF_SLAVE_INACTIVE;
+	if (slave_do_arp_validate(bond, slave))
+		slave->dev->priv_flags |= IFF_SLAVE_NEEDARP;
 }
 
 static inline void bond_set_slave_active_flags(struct slave *slave)
 {
 	slave->state = BOND_STATE_ACTIVE;
-	slave->dev->priv_flags &= ~IFF_SLAVE_INACTIVE;
+	slave->dev->priv_flags &= ~(IFF_SLAVE_INACTIVE | IFF_SLAVE_NEEDARP);
 }
 
 static inline void bond_set_master_3ad_flags(struct bonding *bond)
@@ -284,6 +308,8 @@ int bond_parse_parm(char *mode_arg, struct bond_parm_tbl *tbl);
 const char *bond_mode_name(int mode);
 void bond_select_active_slave(struct bonding *bond);
 void bond_change_active_slave(struct bonding *bond, struct slave *new_active);
+void bond_register_arp(struct bonding *);
+void bond_unregister_arp(struct bonding *);
 
 #endif /* _LINUX_BONDING_H */
 
diff --git a/include/linux/if.h b/include/linux/if.h
index a023ec1274fe..8018c2e22c0c 100644
--- a/include/linux/if.h
+++ b/include/linux/if.h
@@ -60,6 +60,7 @@
 #define IFF_MASTER_8023AD	0x8	/* bonding master, 802.3ad. 	*/
 #define IFF_MASTER_ALB	0x10		/* bonding master, balance-alb.	*/
 #define IFF_BONDING	0x20		/* bonding master or slave	*/
+#define IFF_SLAVE_NEEDARP 0x40		/* need ARPs for validation	*/
 
 #define IF_GET_IFACE	0x0001		/* for querying only */
 #define IF_GET_PROTO	0x0002
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 43289127b458..afd80eff2725 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1016,7 +1016,8 @@ static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
 }
 
 /* On bonding slaves other than the currently active slave, suppress
- * duplicates except for 802.3ad ETH_P_SLOW and alb non-mcast/bcast.
+ * duplicates except for 802.3ad ETH_P_SLOW, alb non-mcast/bcast, and
+ * ARP on active-backup slaves with arp_validate enabled.
  */
 static inline int skb_bond_should_drop(struct sk_buff *skb)
 {
@@ -1025,6 +1026,10 @@ static inline int skb_bond_should_drop(struct sk_buff *skb)
 
 	if (master &&
 	    (dev->priv_flags & IFF_SLAVE_INACTIVE)) {
+		if ((dev->priv_flags & IFF_SLAVE_NEEDARP) &&
+		    skb->protocol == __constant_htons(ETH_P_ARP))
+			return 0;
+
 		if (master->priv_flags & IFF_MASTER_ALB) {
 			if (skb->pkt_type != PACKET_BROADCAST &&
 			    skb->pkt_type != PACKET_MULTICAST)
-- 
cgit v1.2.1


From 89cc76f95af3608d83a1d70b3c76b71ffe66e1f7 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:55:32 -0700
Subject: [PATCH] bonding: Don't mangle LACPDUs

	Fixed handling of 802.3ad LACPDUs.  Do not byte swap data in
place in the packet.  Updated nomenclature of "__ntohs_lacpdu" to be
"htons"; it was previously used for both ntohs and htons operations, but
only called ntohs functions.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_3ad.c | 59 +++++++++++++++++++++---------------------
 1 file changed, 29 insertions(+), 30 deletions(-)

diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index c24b20aaf57f..3fb354d9c515 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -100,7 +100,7 @@ static u16 __get_link_speed(struct port *port);
 static u8 __get_duplex(struct port *port);
 static inline void __initialize_port_locks(struct port *port);
 //conversions
-static void __ntohs_lacpdu(struct lacpdu *lacpdu);
+static void __htons_lacpdu(struct lacpdu *lacpdu);
 static u16 __ad_timer_to_ticks(u16 timer_type, u16 Par);
 
 
@@ -420,23 +420,23 @@ static inline void __initialize_port_locks(struct port *port)
 
 //conversions
 /**
- * __ntohs_lacpdu - convert the contents of a LACPDU to host byte order
+ * __htons_lacpdu - convert the contents of a LACPDU to network byte order
  * @lacpdu: the speicifed lacpdu
  *
  * For each multi-byte field in the lacpdu, convert its content
  */
-static void __ntohs_lacpdu(struct lacpdu *lacpdu)
+static void __htons_lacpdu(struct lacpdu *lacpdu)
 {
 	if (lacpdu) {
-		lacpdu->actor_system_priority =   ntohs(lacpdu->actor_system_priority);
-		lacpdu->actor_key =               ntohs(lacpdu->actor_key);
-		lacpdu->actor_port_priority =     ntohs(lacpdu->actor_port_priority);
-		lacpdu->actor_port =              ntohs(lacpdu->actor_port);
-		lacpdu->partner_system_priority = ntohs(lacpdu->partner_system_priority);
-		lacpdu->partner_key =             ntohs(lacpdu->partner_key);
-		lacpdu->partner_port_priority =   ntohs(lacpdu->partner_port_priority);
-		lacpdu->partner_port =            ntohs(lacpdu->partner_port);
-		lacpdu->collector_max_delay =     ntohs(lacpdu->collector_max_delay);
+		lacpdu->actor_system_priority =   htons(lacpdu->actor_system_priority);
+		lacpdu->actor_key =               htons(lacpdu->actor_key);
+		lacpdu->actor_port_priority =     htons(lacpdu->actor_port_priority);
+		lacpdu->actor_port =              htons(lacpdu->actor_port);
+		lacpdu->partner_system_priority = htons(lacpdu->partner_system_priority);
+		lacpdu->partner_key =             htons(lacpdu->partner_key);
+		lacpdu->partner_port_priority =   htons(lacpdu->partner_port_priority);
+		lacpdu->partner_port =            htons(lacpdu->partner_port);
+		lacpdu->collector_max_delay =     htons(lacpdu->collector_max_delay);
 	}
 }
 
@@ -496,11 +496,11 @@ static void __record_pdu(struct lacpdu *lacpdu, struct port *port)
 	// validate lacpdu and port
 	if (lacpdu && port) {
 		// record the new parameter values for the partner operational
-		port->partner_oper_port_number = lacpdu->actor_port;
-		port->partner_oper_port_priority = lacpdu->actor_port_priority;
+		port->partner_oper_port_number = ntohs(lacpdu->actor_port);
+		port->partner_oper_port_priority = ntohs(lacpdu->actor_port_priority);
 		port->partner_oper_system = lacpdu->actor_system;
-		port->partner_oper_system_priority = lacpdu->actor_system_priority;
-		port->partner_oper_key = lacpdu->actor_key;
+		port->partner_oper_system_priority = ntohs(lacpdu->actor_system_priority);
+		port->partner_oper_key = ntohs(lacpdu->actor_key);
 		// zero partener's lase states
 		port->partner_oper_port_state = 0;
 		port->partner_oper_port_state |= (lacpdu->actor_state & AD_STATE_LACP_ACTIVITY);
@@ -567,11 +567,11 @@ static void __update_selected(struct lacpdu *lacpdu, struct port *port)
 	// validate lacpdu and port
 	if (lacpdu && port) {
 		// check if any parameter is different
-		if ((lacpdu->actor_port != port->partner_oper_port_number) ||
-		    (lacpdu->actor_port_priority != port->partner_oper_port_priority) ||
+		if ((ntohs(lacpdu->actor_port) != port->partner_oper_port_number) ||
+		    (ntohs(lacpdu->actor_port_priority) != port->partner_oper_port_priority) ||
 		    MAC_ADDRESS_COMPARE(&(lacpdu->actor_system), &(port->partner_oper_system)) ||
-		    (lacpdu->actor_system_priority != port->partner_oper_system_priority) ||
-		    (lacpdu->actor_key != port->partner_oper_key) ||
+		    (ntohs(lacpdu->actor_system_priority) != port->partner_oper_system_priority) ||
+		    (ntohs(lacpdu->actor_key) != port->partner_oper_key) ||
 		    ((lacpdu->actor_state & AD_STATE_AGGREGATION) != (port->partner_oper_port_state & AD_STATE_AGGREGATION))
 		   ) {
 			// update the state machine Selected variable
@@ -634,11 +634,11 @@ static void __choose_matched(struct lacpdu *lacpdu, struct port *port)
 	// validate lacpdu and port
 	if (lacpdu && port) {
 		// check if all parameters are alike
-		if (((lacpdu->partner_port == port->actor_port_number) &&
-		     (lacpdu->partner_port_priority == port->actor_port_priority) &&
+		if (((ntohs(lacpdu->partner_port) == port->actor_port_number) &&
+		     (ntohs(lacpdu->partner_port_priority) == port->actor_port_priority) &&
 		     !MAC_ADDRESS_COMPARE(&(lacpdu->partner_system), &(port->actor_system)) &&
-		     (lacpdu->partner_system_priority == port->actor_system_priority) &&
-		     (lacpdu->partner_key == port->actor_oper_port_key) &&
+		     (ntohs(lacpdu->partner_system_priority) == port->actor_system_priority) &&
+		     (ntohs(lacpdu->partner_key) == port->actor_oper_port_key) &&
 		     ((lacpdu->partner_state & AD_STATE_AGGREGATION) == (port->actor_oper_port_state & AD_STATE_AGGREGATION))) ||
 		    // or this is individual link(aggregation == FALSE)
 		    ((lacpdu->actor_state & AD_STATE_AGGREGATION) == 0)
@@ -668,11 +668,11 @@ static void __update_ntt(struct lacpdu *lacpdu, struct port *port)
 	// validate lacpdu and port
 	if (lacpdu && port) {
 		// check if any parameter is different
-		if ((lacpdu->partner_port != port->actor_port_number) ||
-		    (lacpdu->partner_port_priority != port->actor_port_priority) ||
+		if ((ntohs(lacpdu->partner_port) != port->actor_port_number) ||
+		    (ntohs(lacpdu->partner_port_priority) != port->actor_port_priority) ||
 		    MAC_ADDRESS_COMPARE(&(lacpdu->partner_system), &(port->actor_system)) ||
-		    (lacpdu->partner_system_priority != port->actor_system_priority) ||
-		    (lacpdu->partner_key != port->actor_oper_port_key) ||
+		    (ntohs(lacpdu->partner_system_priority) != port->actor_system_priority) ||
+		    (ntohs(lacpdu->partner_key) != port->actor_oper_port_key) ||
 		    ((lacpdu->partner_state & AD_STATE_LACP_ACTIVITY) != (port->actor_oper_port_state & AD_STATE_LACP_ACTIVITY)) ||
 		    ((lacpdu->partner_state & AD_STATE_LACP_TIMEOUT) != (port->actor_oper_port_state & AD_STATE_LACP_TIMEOUT)) ||
 		    ((lacpdu->partner_state & AD_STATE_SYNCHRONIZATION) != (port->actor_oper_port_state & AD_STATE_SYNCHRONIZATION)) ||
@@ -856,7 +856,7 @@ static inline void __update_lacpdu_from_port(struct port *port)
 	 */
 
 	/* Convert all non u8 parameters to Big Endian for transmit */
-	__ntohs_lacpdu(lacpdu);
+	__htons_lacpdu(lacpdu);
 }
 
 //////////////////////////////////////////////////////////////////////////////////////
@@ -2180,7 +2180,6 @@ static void bond_3ad_rx_indication(struct lacpdu *lacpdu, struct slave *slave, u
 
 		switch (lacpdu->subtype) {
 		case AD_TYPE_LACPDU:
-			__ntohs_lacpdu(lacpdu);
 			dprintk("Received LACPDU on port %d\n", port->actor_port_number);
 			ad_rx_machine(lacpdu, port);
 			break;
-- 
cgit v1.2.1


From 8a8e447b2aa1f9139d0bfc94a2a3426be9c8d40a Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Fri, 22 Sep 2006 21:56:15 -0700
Subject: [PATCH] bonding: Fix primary selection error at enslavement time

	At enslavement time, the primary slave might not be activated if
there is already an active slave and the new slave is the primary.
Replaced complicated logic with a call to bond_select_active_slave(),
which does the right thing.

	Fixes http://bugzilla.kernel.org/show_bug.cgi?id=6378

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bond_main.c | 25 ++-----------------------
 1 file changed, 2 insertions(+), 23 deletions(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index fd521b05db83..0fb5f653d3ce 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1513,29 +1513,8 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
 
 	switch (bond->params.mode) {
 	case BOND_MODE_ACTIVEBACKUP:
-		/* if we're in active-backup mode, we need one and
-		 * only one active interface. The backup interfaces
-		 * will have their SLAVE_INACTIVE flag set because we
-		 * need them to be drop all packets. Thus, since we
-		 * guarantee that curr_active_slave always point to
-		 * the last usable interface, we just have to verify
-		 * this interface's flag.
-		 */
-		if (((!bond->curr_active_slave) ||
-		     (bond->curr_active_slave->dev->priv_flags & IFF_SLAVE_INACTIVE)) &&
-		    (new_slave->link != BOND_LINK_DOWN)) {
-			/* first slave or no active slave yet, and this link
-			   is OK, so make this interface the active one */
-			bond_change_active_slave(bond, new_slave);
-			printk(KERN_INFO DRV_NAME
-			       ": %s: first active interface up!\n",
-			       bond->dev->name);
-			netif_carrier_on(bond->dev);
-
-		} else {
-			dprintk("This is just a backup slave\n");
-			bond_set_slave_inactive_flags(new_slave);
-		}
+		bond_set_slave_inactive_flags(new_slave);
+		bond_select_active_slave(bond);
 		break;
 	case BOND_MODE_8023AD:
 		/* in 802.3ad mode, the internal mechanism
-- 
cgit v1.2.1


From ee705dba75c2f7daae5403ad4599b6738e2da7a0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 23 Sep 2006 01:28:17 +0100
Subject: [PATCH] 64bit bugs in s2io

le32_to_cpu() on 64bit values

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/s2io.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index f5dbeb27b6f0..c50f4bf34202 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -4303,11 +4303,11 @@ static struct net_device_stats *s2io_get_stats(struct net_device *dev)
 	sp->stats.tx_errors =
 		le32_to_cpu(mac_control->stats_info->tmac_any_err_frms);
 	sp->stats.rx_errors =
-		le32_to_cpu(mac_control->stats_info->rmac_drop_frms);
+		le64_to_cpu(mac_control->stats_info->rmac_drop_frms);
 	sp->stats.multicast =
 		le32_to_cpu(mac_control->stats_info->rmac_vld_mcst_frms);
 	sp->stats.rx_length_errors =
-		le32_to_cpu(mac_control->stats_info->rmac_long_frms);
+		le64_to_cpu(mac_control->stats_info->rmac_long_frms);
 
 	return (&sp->stats);
 }
-- 
cgit v1.2.1


From 0c5649af3accc7f2941c5eebcc613e1a81b03448 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 23 Sep 2006 01:32:40 +0100
Subject: [PATCH] restore __iomem annotations in e1000

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/e1000/e1000_hw.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/e1000/e1000_hw.h b/drivers/net/e1000/e1000_hw.h
index a170e96251f6..4020acb55005 100644
--- a/drivers/net/e1000/e1000_hw.h
+++ b/drivers/net/e1000/e1000_hw.h
@@ -1367,8 +1367,8 @@ struct e1000_hw_stats {
 
 /* Structure containing variables used by the shared code (e1000_hw.c) */
 struct e1000_hw {
-    uint8_t *hw_addr;
-    uint8_t *flash_address;
+    uint8_t __iomem *hw_addr;
+    uint8_t __iomem *flash_address;
     e1000_mac_type mac_type;
     e1000_phy_type phy_type;
     uint32_t phy_init_script;
-- 
cgit v1.2.1


From cc3afe6f856054a3752ef2b3ccc5eebf33bd5024 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sat, 23 Sep 2006 01:33:40 +0100
Subject: [PATCH] more s2io __iomem annotations

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/s2io.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index c50f4bf34202..1bf23e41f580 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -2904,7 +2904,7 @@ static void s2io_mdio_write(u32 mmd_type, u64 addr, u16 value, struct net_device
 {
 	u64 val64 = 0x0;
 	nic_t *sp = dev->priv;
-	XENA_dev_config_t *bar0 = (XENA_dev_config_t *)sp->bar0;
+	XENA_dev_config_t __iomem *bar0 = sp->bar0;
 
 	//address transaction
 	val64 = val64 | MDIO_MMD_INDX_ADDR(addr)
@@ -2953,7 +2953,7 @@ static u64 s2io_mdio_read(u32 mmd_type, u64 addr, struct net_device *dev)
 	u64 val64 = 0x0;
 	u64 rval64 = 0x0;
 	nic_t *sp = dev->priv;
-	XENA_dev_config_t *bar0 = (XENA_dev_config_t *)sp->bar0;
+	XENA_dev_config_t __iomem *bar0 = sp->bar0;
 
 	/* address transaction */
 	val64 = val64 | MDIO_MMD_INDX_ADDR(addr)
@@ -3276,7 +3276,7 @@ static void alarm_intr_handler(struct s2io_nic *nic)
  *   SUCCESS on success and FAILURE on failure.
  */
 
-static int wait_for_cmd_complete(void *addr, u64 busy_bit)
+static int wait_for_cmd_complete(void __iomem *addr, u64 busy_bit)
 {
 	int ret = FAILURE, cnt = 0;
 	u64 val64;
-- 
cgit v1.2.1


From c32a8fd7cb33f30bcd855188dc8e243b144c5cee Mon Sep 17 00:00:00 2001
From: Henne <henne@nachtwindheim.de>
Date: Mon, 25 Sep 2006 22:00:46 +0200
Subject: [PATCH] ata-piix: fixes kerneldoc error

Fixes an error in kerneldoc of ata_piix.c.
Signed-off-by: Henrik Kretzschmar <henne@nachtwindheim.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ata_piix.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index ab2ecccf7798..ffa111eea9da 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -851,7 +851,7 @@ static void piix_set_piomode (struct ata_port *ap, struct ata_device *adev)
  *	@ap: Port whose timings we are configuring
  *	@adev: Drive in question
  *	@udma: udma mode, 0 - 6
- *	@is_ich: set if the chip is an ICH device
+ *	@isich: set if the chip is an ICH device
  *
  *	Set UDMA mode for device, in host controller PCI config space.
  *
-- 
cgit v1.2.1


From c0ba7e5147829eaa607351997bccd06200a8db12 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Mon, 25 Sep 2006 16:24:16 -0700
Subject: [PATCH] autofs4: zero timeout prevents shutdown

If the timeout of an autofs mount is set to zero then umounts are disabled.
 This works fine, however the kernel module checks the expire timeout and
goes no further if it is zero.  This is not the right thing to do at
shutdown as the module is passed an option to expire mounts regardless of
their timeout setting.

This patch allows autofs to honor the force expire option.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/autofs4/expire.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 8dbd44f10e9d..d96e5c14a9ca 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -32,7 +32,7 @@ static inline int autofs4_can_expire(struct dentry *dentry,
 
 	if (!do_now) {
 		/* Too young to die */
-		if (time_after(ino->last_used + timeout, now))
+		if (!timeout || time_after(ino->last_used + timeout, now))
 			return 0;
 
 		/* update last_used here :-
@@ -253,7 +253,7 @@ static struct dentry *autofs4_expire_direct(struct super_block *sb,
 	struct dentry *root = dget(sb->s_root);
 	int do_now = how & AUTOFS_EXP_IMMEDIATE;
 
-	if (!sbi->exp_timeout || !root)
+	if (!root)
 		return NULL;
 
 	now = jiffies;
@@ -293,7 +293,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 	int do_now = how & AUTOFS_EXP_IMMEDIATE;
 	int exp_leaves = how & AUTOFS_EXP_LEAVES;
 
-	if ( !sbi->exp_timeout || !root )
+	if (!root)
 		return NULL;
 
 	now = jiffies;
-- 
cgit v1.2.1


From 0b16f21f144010aa627c58b3a33be49ebfd685dc Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 16:24:23 -0700
Subject: [PATCH] rtc: lockdep fix/workaround

BUG: warning at kernel/lockdep.c:1816/trace_hardirqs_on() (Not tainted)
 [<c04051ee>] show_trace_log_lvl+0x58/0x171
 [<c0405802>] show_trace+0xd/0x10
 [<c040591b>] dump_stack+0x19/0x1b
 [<c043abee>] trace_hardirqs_on+0xa2/0x11e
 [<c06143c3>] _spin_unlock_irq+0x22/0x26
 [<c0541540>] rtc_get_rtc_time+0x32/0x176
 [<c0419ba4>] hpet_rtc_interrupt+0x92/0x14d
 [<c0450f94>] handle_IRQ_event+0x20/0x4d
 [<c0451055>] __do_IRQ+0x94/0xef
 [<c040678d>] do_IRQ+0x9e/0xbd
 [<c0404a49>] common_interrupt+0x25/0x2c
DWARF2 unwinder stuck at common_interrupt+0x25/0x2c

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/rtc.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index 6e6a7c7a7eff..ab6429b4a84e 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -209,11 +209,12 @@ static const unsigned char days_in_mo[] =
  */
 static inline unsigned char rtc_is_updating(void)
 {
+	unsigned long flags;
 	unsigned char uip;
 
-	spin_lock_irq(&rtc_lock);
+	spin_lock_irqsave(&rtc_lock, flags);
 	uip = (CMOS_READ(RTC_FREQ_SELECT) & RTC_UIP);
-	spin_unlock_irq(&rtc_lock);
+	spin_unlock_irqrestore(&rtc_lock, flags);
 	return uip;
 }
 
-- 
cgit v1.2.1


From 24fd425edd53ea580cad917e825c1f6715e9b939 Mon Sep 17 00:00:00 2001
From: keith mannthey <kmannth@us.ibm.com>
Date: Mon, 25 Sep 2006 16:24:39 -0700
Subject: [PATCH] i386 bootioremap / kexec fix

With CONFIG_PHYSICAL_START set to a non default values the i386
boot_ioremap code calculated its pte index wrong and users of boot_ioremap
have their areas incorrectly mapped (for me SRAT table not mapped during
early boot).  This patch removes the addr < BOOT_PTE_PTRS constraint.

[ Keith says this is applicable to 2.6.16 and 2.6.17 as well ]

Signed-off-by: Keith Mannthey<kmannth@us.ibm.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: <stable@kernel.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/boot_ioremap.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/i386/mm/boot_ioremap.c b/arch/i386/mm/boot_ioremap.c
index 5d44f4f5ff59..4de11f508c3a 100644
--- a/arch/i386/mm/boot_ioremap.c
+++ b/arch/i386/mm/boot_ioremap.c
@@ -29,8 +29,11 @@
  */
 
 #define BOOT_PTE_PTRS (PTRS_PER_PTE*2)
-#define boot_pte_index(address) \
-	     (((address) >> PAGE_SHIFT) & (BOOT_PTE_PTRS - 1))
+
+static unsigned long boot_pte_index(unsigned long vaddr) 
+{
+	return __pa(vaddr) >> PAGE_SHIFT;
+}
 
 static inline boot_pte_t* boot_vaddr_to_pte(void *address)
 {
-- 
cgit v1.2.1


From f3ef9ead31ae995251b420ac98398bd7545bf4e1 Mon Sep 17 00:00:00 2001
From: David Rientjes <rientjes@cs.washington.edu>
Date: Mon, 25 Sep 2006 16:24:57 -0700
Subject: [PATCH] do not free non slab allocated per_cpu_pageset

Stops panic associated with attempting to free a non slab-allocated
per_cpu_pageset.

Signed-off-by: David Rientjes <rientjes@cs.washington.edu>
Acked-by: Christoph Lameter <clameter@sgi.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 3b5358a0561f..8a52ba9fe693 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1845,8 +1845,10 @@ static inline void free_zone_pagesets(int cpu)
 	for_each_zone(zone) {
 		struct per_cpu_pageset *pset = zone_pcp(zone, cpu);
 
+		/* Free per_cpu_pageset if it is slab allocated */
+		if (pset != &boot_pageset[cpu])
+			kfree(pset);
 		zone_pcp(zone, cpu) = NULL;
-		kfree(pset);
 	}
 }
 
-- 
cgit v1.2.1


From 25981de5b836581364612a4b1fe27db3b5d1f861 Mon Sep 17 00:00:00 2001
From: Michael Hanselmann <linux-kernel@hansmi.ch>
Date: Mon, 25 Sep 2006 16:25:07 -0700
Subject: [PATCH] backlight: fix oops in __mutex_lock_slowpath during head
 /sys/class/graphics/fb0/*

Seems like not all drivers use the framebuffer_alloc() function and won't
have an initialized mutex.  But those don't have a backlight, anyway.

Signed-off-by: Michael Hanselmann <linux-kernel@hansmi.ch>
Cc: Olaf Hering <olaf@aepfle.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Daniel R Thompson <daniel.thompson@st.com>
Cc: Jon Smirl <jonsmirl@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/fbsysfs.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/video/fbsysfs.c b/drivers/video/fbsysfs.c
index 4f78f234473d..c151dcf68786 100644
--- a/drivers/video/fbsysfs.c
+++ b/drivers/video/fbsysfs.c
@@ -397,6 +397,12 @@ static ssize_t store_bl_curve(struct class_device *class_device,
 	u8 tmp_curve[FB_BACKLIGHT_LEVELS];
 	unsigned int i;
 
+	/* Some drivers don't use framebuffer_alloc(), but those also
+	 * don't have backlights.
+	 */
+	if (!fb_info || !fb_info->bl_dev)
+		return -ENODEV;
+
 	if (count != (FB_BACKLIGHT_LEVELS / 8 * 24))
 		return -EINVAL;
 
@@ -430,6 +436,12 @@ static ssize_t show_bl_curve(struct class_device *class_device, char *buf)
 	ssize_t len = 0;
 	unsigned int i;
 
+	/* Some drivers don't use framebuffer_alloc(), but those also
+	 * don't have backlights.
+	 */
+	if (!fb_info || !fb_info->bl_dev)
+		return -ENODEV;
+
 	mutex_lock(&fb_info->bl_mutex);
 	for (i = 0; i < FB_BACKLIGHT_LEVELS; i += 8)
 		len += snprintf(&buf[len], PAGE_SIZE,
-- 
cgit v1.2.1


From 08992986497471ce575f23796268fb1b50b5c2ab Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Mon, 25 Sep 2006 16:25:21 -0700
Subject: [PATCH] cpu to node relationship fixup: acpi_map_cpu2node

Problem description:

  We have additional_cpus= option for allocating possible_cpus.  But nid
  for possible cpus are not fixed at boot time.  cpus which is offlined at
  boot or cpus which is not on SRAT is not tied to its node.  This will
  cause panic at cpu onlining.

Usually, pxm_to_nid() mapping is fixed at boot time by SRAT.

But, unfortunately, some system (my system!) do not include
full SRAT table for possible cpus.  (Then, I use
additiona_cpus= option.)

For such possible cpus, pxm<->nid should be fixed at
hot-add.  We now have acpi_map_pxm_to_node() which is also
used at boot.  It's suitable here.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/acpi.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 0176556aeecc..32c3abededc6 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -771,16 +771,19 @@ int acpi_map_cpu2node(acpi_handle handle, int cpu, long physid)
 {
 #ifdef CONFIG_ACPI_NUMA
 	int pxm_id;
+	int nid;
 
 	pxm_id = acpi_get_pxm(handle);
-
 	/*
-	 * Assuming that the container driver would have set the proximity
-	 * domain and would have initialized pxm_to_node(pxm_id) && pxm_flag
+	 * We don't have cpu-only-node hotadd. But if the system equips
+	 * SRAT table, pxm is already found and node is ready.
+  	 * So, just pxm_to_nid(pxm) is OK.
+	 * This code here is for the system which doesn't have full SRAT
+  	 * table for possible cpus.
 	 */
-	node_cpuid[cpu].nid = (pxm_id < 0) ? 0 : pxm_to_node(pxm_id);
-
+	nid = acpi_map_pxm_to_node(pxm_id);
 	node_cpuid[cpu].phys_id = physid;
+	node_cpuid[cpu].nid = nid;
 #endif
 	return (0);
 }
-- 
cgit v1.2.1


From 3212fe1594e577463bc8601d28aa008f520c3377 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Mon, 25 Sep 2006 16:25:31 -0700
Subject: [PATCH] cpu to node relationship fixup: map cpu to node

Assume that a cpu is *physically* offlined at boot time...

Because smpboot.c::smp_boot_cpu_map() canoot find cpu's sapicid,
numa.c::build_cpu_to_node_map() cannot build cpu<->node map for
offlined cpu.

For such cpus, cpu_to_node map should be fixed at cpu-hot-add.
This mapping should be done before cpu onlining.

This patch also handles cpu hotremove case.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/numa.c     | 34 +++++++++++++++++++++++++++++++---
 arch/ia64/kernel/topology.c |  6 +++++-
 include/asm-ia64/numa.h     |  4 ++++
 3 files changed, 40 insertions(+), 4 deletions(-)

diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c
index 1cc360c83e7a..20340631179f 100644
--- a/arch/ia64/kernel/numa.c
+++ b/arch/ia64/kernel/numa.c
@@ -29,6 +29,36 @@ EXPORT_SYMBOL(cpu_to_node_map);
 
 cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned;
 
+void __cpuinit map_cpu_to_node(int cpu, int nid)
+{
+	int oldnid;
+	if (nid < 0) { /* just initialize by zero */
+		cpu_to_node_map[cpu] = 0;
+		return;
+	}
+	/* sanity check first */
+	oldnid = cpu_to_node_map[cpu];
+	if (cpu_isset(cpu, node_to_cpu_mask[oldnid])) {
+		return; /* nothing to do */
+	}
+	/* we don't have cpu-driven node hot add yet...
+	   In usual case, node is created from SRAT at boot time. */
+	if (!node_online(nid))
+		nid = first_online_node;
+	cpu_to_node_map[cpu] = nid;
+	cpu_set(cpu, node_to_cpu_mask[nid]);
+	return;
+}
+
+void __cpuinit unmap_cpu_from_node(int cpu, int nid)
+{
+	WARN_ON(!cpu_isset(cpu, node_to_cpu_mask[nid]));
+	WARN_ON(cpu_to_node_map[cpu] != nid);
+	cpu_to_node_map[cpu] = 0;
+	cpu_clear(cpu, node_to_cpu_mask[nid]);
+}
+
+
 /**
  * build_cpu_to_node_map - setup cpu to node and node to cpumask arrays
  *
@@ -49,8 +79,6 @@ void __init build_cpu_to_node_map(void)
 				node = node_cpuid[i].nid;
 				break;
 			}
-		cpu_to_node_map[cpu] = (node >= 0) ? node : 0;
-		if (node >= 0)
-			cpu_set(cpu, node_to_cpu_mask[node]);
+		map_cpu_to_node(cpu, node);
 	}
 }
diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index f648c610b10c..05bdf7affb43 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -36,6 +36,9 @@ int arch_register_cpu(int num)
 	 */
 	if (!can_cpei_retarget() && is_cpu_cpei_target(num))
 		sysfs_cpus[num].cpu.no_control = 1;
+#ifdef CONFIG_NUMA
+	map_cpu_to_node(num, node_cpuid[num].nid);
+#endif
 #endif
 
 	return register_cpu(&sysfs_cpus[num].cpu, num);
@@ -45,7 +48,8 @@ int arch_register_cpu(int num)
 
 void arch_unregister_cpu(int num)
 {
-	return unregister_cpu(&sysfs_cpus[num].cpu);
+	unregister_cpu(&sysfs_cpus[num].cpu);
+	unmap_cpu_from_node(num, cpu_to_node(num));
 }
 EXPORT_SYMBOL(arch_register_cpu);
 EXPORT_SYMBOL(arch_unregister_cpu);
diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h
index e5a8260593a5..e0a1d173e42d 100644
--- a/include/asm-ia64/numa.h
+++ b/include/asm-ia64/numa.h
@@ -64,6 +64,10 @@ extern int paddr_to_nid(unsigned long paddr);
 
 #define local_nodeid (cpu_to_node_map[smp_processor_id()])
 
+extern void map_cpu_to_node(int cpu, int nid);
+extern void unmap_cpu_from_node(int cpu, int nid);
+
+
 #else /* !CONFIG_NUMA */
 
 #define paddr_to_nid(addr)	0
-- 
cgit v1.2.1


From bfa0e9a07cd31f3858239dbc93011b82780acf4b Mon Sep 17 00:00:00 2001
From: keith mannthey <kmannth@us.ibm.com>
Date: Mon, 25 Sep 2006 16:25:35 -0700
Subject: [PATCH] i386: fix flat mode numa on a real numa system

If there is only 1 node in the system cpus should think they are apart of
some other node.

If cases where a real numa system boots the Flat numa option make sure the
cpus don't claim to be apart on a non-existent node.

Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smpboot.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index f948419c888a..efe07990e7fc 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -642,9 +642,13 @@ static void map_cpu_to_logical_apicid(void)
 {
 	int cpu = smp_processor_id();
 	int apicid = logical_smp_processor_id();
+	int node = apicid_to_node(apicid);
+
+	if (!node_online(node))
+		node = first_online_node;
 
 	cpu_2_logical_apicid[cpu] = apicid;
-	map_cpu_to_node(cpu, apicid_to_node(apicid));
+	map_cpu_to_node(cpu, node);
 }
 
 static void unmap_cpu_to_logical_apicid(int cpu)
-- 
cgit v1.2.1


From 1cc5f7142eca352109895fe20b1fc6405dd17727 Mon Sep 17 00:00:00 2001
From: Ed Swierk <eswierk@arastra.com>
Date: Mon, 25 Sep 2006 16:25:36 -0700
Subject: [PATCH] load_module: no BUG if module_subsys uninitialized

Invoking load_module() before param_sysfs_init() is called crashes in
mod_sysfs_setup(), since the kset in module_subsys is not initialized yet.

In my case, net-pf-1 is getting modprobed as a result of hotplug trying to
create a UNIX socket.  Calls to hotplug begin after the topology_init
initcall.

Another patch for the same symptom (module_subsys-initialize-earlier.patch)
moves param_sysfs_init() to the subsys initcalls, but this is still not
early enough in the boot process in some cases.  In particular,
topology_init() causes /sbin/hotplug to run, which requests net-pf-1 (the
UNIX socket protocol) which can be compiled as a module.  Moving
param_sysfs_init() to the postcore initcalls fixes this particular race,
but there might well be other cases where a usermodehelper causes a module
to load earlier still.

The patch makes load_module() return an error rather than crashing the
kernel if invoked before module_subsys is initialized.

Cc: Mark Huang <mlhuang@cs.princeton.edu>
Cc: Greg KH <greg@kroah.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/module.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/kernel/module.c b/kernel/module.c
index 2a19cd47c046..b7fe6e840963 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -1054,6 +1054,12 @@ static int mod_sysfs_setup(struct module *mod,
 {
 	int err;
 
+	if (!module_subsys.kset.subsys) {
+		printk(KERN_ERR "%s: module_subsys not initialized\n",
+		       mod->name);
+		err = -EINVAL;
+		goto out;
+	}
 	memset(&mod->mkobj.kobj, 0, sizeof(mod->mkobj.kobj));
 	err = kobject_set_name(&mod->mkobj.kobj, "%s", mod->name);
 	if (err)
-- 
cgit v1.2.1


From b7de567bf3187ccf776e2fe0e241593cdcba5459 Mon Sep 17 00:00:00 2001
From: Jonathan Corbet <corbet-v4l@lwn.net>
Date: Mon, 25 Sep 2006 16:25:37 -0700
Subject: [PATCH] VIDIOC_ENUMSTD bug

The v4l2 API documentation for VIDIOC_ENUMSTD says:

	To enumerate all standards applications shall begin at index
	zero, incrementing by one until the driver returns EINVAL.

The actual code, however, tests the index this way:

               if (index<=0 || index >= vfd->tvnormsize) {
                        ret=-EINVAL;

So any application which passes in index=0 gets EINVAL right off the bat
- and, in fact, this is what happens to mplayer.  So I think the
following patch is called for, and maybe even appropriate for a 2.6.18.x
stable release.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/media/video/videodev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/media/video/videodev.c b/drivers/media/video/videodev.c
index 88bf2af2a0e7..edd7b83c3464 100644
--- a/drivers/media/video/videodev.c
+++ b/drivers/media/video/videodev.c
@@ -836,7 +836,7 @@ static int __video_do_ioctl(struct inode *inode, struct file *file,
 			break;
 		}
 
-		if (index<=0 || index >= vfd->tvnormsize) {
+		if (index < 0 || index >= vfd->tvnormsize) {
 			ret=-EINVAL;
 			break;
 		}
-- 
cgit v1.2.1


From 29da9f6d9cc3685ae7f6c8b817f6ed8864c78a4c Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 25 Sep 2006 21:56:33 -0400
Subject: [libata] Fix oops introduced in non-uniform port handling fix

Noticed by several people.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/sata_nv.c  | 6 +++---
 drivers/ata/sata_sis.c | 6 +++---
 drivers/ata/sata_uli.c | 6 +++---
 drivers/ata/sata_via.c | 7 ++++---
 4 files changed, 13 insertions(+), 12 deletions(-)

diff --git a/drivers/ata/sata_nv.c b/drivers/ata/sata_nv.c
index 27c22feebf30..8cd730fe5dd3 100644
--- a/drivers/ata/sata_nv.c
+++ b/drivers/ata/sata_nv.c
@@ -484,7 +484,7 @@ static void nv_error_handler(struct ata_port *ap)
 static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version = 0;
-	struct ata_port_info *ppi;
+	struct ata_port_info *ppi[2];
 	struct ata_probe_ent *probe_ent;
 	int pci_dev_busy = 0;
 	int rc;
@@ -520,8 +520,8 @@ static int nv_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	rc = -ENOMEM;
 
-	ppi = &nv_port_info[ent->driver_data];
-	probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
+	ppi[0] = ppi[1] = &nv_port_info[ent->driver_data];
+	probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
 	if (!probe_ent)
 		goto err_out_regions;
 
diff --git a/drivers/ata/sata_sis.c b/drivers/ata/sata_sis.c
index 9b17375d8056..18d49fff8dc4 100644
--- a/drivers/ata/sata_sis.c
+++ b/drivers/ata/sata_sis.c
@@ -240,7 +240,7 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	struct ata_probe_ent *probe_ent = NULL;
 	int rc;
 	u32 genctl;
-	struct ata_port_info *ppi;
+	struct ata_port_info *ppi[2];
 	int pci_dev_busy = 0;
 	u8 pmr;
 	u8 port2_start;
@@ -265,8 +265,8 @@ static int sis_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc)
 		goto err_out_regions;
 
-	ppi = &sis_port_info;
-	probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
+	ppi[0] = ppi[1] = &sis_port_info;
+	probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
 	if (!probe_ent) {
 		rc = -ENOMEM;
 		goto err_out_regions;
diff --git a/drivers/ata/sata_uli.c b/drivers/ata/sata_uli.c
index 8fc6e800011a..dd76f37be182 100644
--- a/drivers/ata/sata_uli.c
+++ b/drivers/ata/sata_uli.c
@@ -185,7 +185,7 @@ static int uli_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	static int printed_version;
 	struct ata_probe_ent *probe_ent;
-	struct ata_port_info *ppi;
+	struct ata_port_info *ppi[2];
 	int rc;
 	unsigned int board_idx = (unsigned int) ent->driver_data;
 	int pci_dev_busy = 0;
@@ -211,8 +211,8 @@ static int uli_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (rc)
 		goto err_out_regions;
 
-	ppi = &uli_port_info;
-	probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
+	ppi[0] = ppi[1] = &uli_port_info;
+	probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
 	if (!probe_ent) {
 		rc = -ENOMEM;
 		goto err_out_regions;
diff --git a/drivers/ata/sata_via.c b/drivers/ata/sata_via.c
index 7f087aef99de..a72a2389a11c 100644
--- a/drivers/ata/sata_via.c
+++ b/drivers/ata/sata_via.c
@@ -318,9 +318,10 @@ static void vt6421_init_addrs(struct ata_probe_ent *probe_ent,
 static struct ata_probe_ent *vt6420_init_probe_ent(struct pci_dev *pdev)
 {
 	struct ata_probe_ent *probe_ent;
-	struct ata_port_info *ppi = &vt6420_port_info;
-
-	probe_ent = ata_pci_init_native_mode(pdev, &ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
+	struct ata_port_info *ppi[2];
+	
+	ppi[0] = ppi[1] = &vt6420_port_info;
+	probe_ent = ata_pci_init_native_mode(pdev, ppi, ATA_PORT_PRIMARY | ATA_PORT_SECONDARY);
 	if (!probe_ent)
 		return NULL;
 
-- 
cgit v1.2.1


From 6d99eb33f55879bc71ddba077e09cf8a61129707 Mon Sep 17 00:00:00 2001
From: "jens m. noedler" <noedler@web.de>
Date: Sun, 10 Sep 2006 18:46:33 +0200
Subject: Documentation/ABI: devfs is not obsolete, but removed!

Signed-off-by: Jens M. Noedler <noedler@web.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/ABI/obsolete/devfs | 13 -------------
 Documentation/ABI/removed/devfs  | 12 ++++++++++++
 2 files changed, 12 insertions(+), 13 deletions(-)
 delete mode 100644 Documentation/ABI/obsolete/devfs
 create mode 100644 Documentation/ABI/removed/devfs

diff --git a/Documentation/ABI/obsolete/devfs b/Documentation/ABI/obsolete/devfs
deleted file mode 100644
index b8b87399bc8f..000000000000
--- a/Documentation/ABI/obsolete/devfs
+++ /dev/null
@@ -1,13 +0,0 @@
-What:		devfs
-Date:		July 2005
-Contact:	Greg Kroah-Hartman <gregkh@suse.de>
-Description:
-	devfs has been unmaintained for a number of years, has unfixable
-	races, contains a naming policy within the kernel that is
-	against the LSB, and can be replaced by using udev.
-	The files fs/devfs/*, include/linux/devfs_fs*.h will be removed,
-	along with the the assorted devfs function calls throughout the
-	kernel tree.
-
-Users:
-
diff --git a/Documentation/ABI/removed/devfs b/Documentation/ABI/removed/devfs
new file mode 100644
index 000000000000..8195c4e0d0a1
--- /dev/null
+++ b/Documentation/ABI/removed/devfs
@@ -0,0 +1,12 @@
+What:		devfs
+Date:		July 2005 (scheduled), finally removed in kernel v2.6.18
+Contact:	Greg Kroah-Hartman <gregkh@suse.de>
+Description:
+	devfs has been unmaintained for a number of years, has unfixable
+	races, contains a naming policy within the kernel that is
+	against the LSB, and can be replaced by using udev.
+	The files fs/devfs/*, include/linux/devfs_fs*.h were removed,
+	along with the the assorted devfs function calls throughout the
+	kernel tree.
+
+Users:
-- 
cgit v1.2.1


From d81d9d6b9f7d0af7d4341951d48d296681878f7a Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@suse.de>
Date: Sun, 13 Aug 2006 06:17:09 +0200
Subject: deprecate PHYSDEV* keys

deprecate PHYSDEV* values in the uevent environment

These values are no longer needed and inconsistent with the
stacking of class devices. The event environment should not
carry properties of a parent device. The key PHYSDEVDRIVER is
available as DRIVER, PHYDEVBUS is indentical SUBSYSTEM. Class
devices should not carry any of these values.

Signed-off-by: Kay Sievers <kay.sievers@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/feature-removal-schedule.txt | 12 ++++++++++++
 drivers/base/class.c                       |  2 +-
 drivers/base/core.c                        | 10 +++++++---
 3 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 552507fe9a7e..a89a1b7f7cf7 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -294,3 +294,15 @@ Why:	The frame diverter is included in most distribution kernels, but is
 	It is not clear if anyone is still using it.
 Who:	Stephen Hemminger <shemminger@osdl.org>
 
+---------------------------
+
+
+What:	PHYSDEVPATH, PHYSDEVBUS, PHYSDEVDRIVER in the uevent environment
+When:	Oktober 2008
+Why:	The stacking of class devices makes these values misleading and
+	inconsistent.
+	Class devices should not carry any of these properties, and bus
+	devices have SUBSYTEM and DRIVER as a replacement.
+Who:	Kay Sievers <kay.sievers@suse.de>
+
+---------------------------
diff --git a/drivers/base/class.c b/drivers/base/class.c
index de8908320f23..46336f1b93b6 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -361,7 +361,7 @@ static int class_uevent(struct kset *kset, struct kobject *kobj, char **envp,
 	pr_debug("%s - name = %s\n", __FUNCTION__, class_dev->class_id);
 
 	if (class_dev->dev) {
-		/* add physical device, backing this device  */
+		/* add device, backing this class device (deprecated) */
 		struct device *dev = class_dev->dev;
 		char *path = kobject_get_path(&dev->kobj, GFP_KERNEL);
 
diff --git a/drivers/base/core.c b/drivers/base/core.c
index be6b5bc0677d..04d089fd4f76 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -149,17 +149,21 @@ static int dev_uevent(struct kset *kset, struct kobject *kobj, char **envp,
 			       "MINOR=%u", MINOR(dev->devt));
 	}
 
-	/* add bus name of physical device */
+	/* add bus name (same as SUBSYSTEM, deprecated) */
 	if (dev->bus)
 		add_uevent_var(envp, num_envp, &i,
 			       buffer, buffer_size, &length,
 			       "PHYSDEVBUS=%s", dev->bus->name);
 
-	/* add driver name of physical device */
-	if (dev->driver)
+	/* add driver name (PHYSDEV* values are deprecated)*/
+	if (dev->driver) {
+		add_uevent_var(envp, num_envp, &i,
+			       buffer, buffer_size, &length,
+			       "DRIVER=%s", dev->driver->name);
 		add_uevent_var(envp, num_envp, &i,
 			       buffer, buffer_size, &length,
 			       "PHYSDEVDRIVER=%s", dev->driver->name);
+	}
 
 	/* terminate, set to next free slot, shrink available space */
 	envp[i] = NULL;
-- 
cgit v1.2.1


From ddd5d35a8f7c1924049e8b6877b3177c1787e6a3 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor@insightbb.com>
Date: Thu, 10 Aug 2006 01:18:18 -0400
Subject: class_device_create(): make fmt argument 'const char *'

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 3 ++-
 include/linux/device.h | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 46336f1b93b6..75057aaab809 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -679,7 +679,8 @@ int class_device_register(struct class_device *class_dev)
 struct class_device *class_device_create(struct class *cls,
 					 struct class_device *parent,
 					 dev_t devt,
-					 struct device *device, char *fmt, ...)
+					 struct device *device,
+					 const char *fmt, ...)
 {
 	va_list args;
 	struct class_device *class_dev = NULL;
diff --git a/include/linux/device.h b/include/linux/device.h
index 1e5f30da98bc..1fec28546525 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -277,7 +277,7 @@ extern struct class_device *class_device_create(struct class *cls,
 						struct class_device *parent,
 						dev_t devt,
 						struct device *device,
-						char *fmt, ...)
+						const char *fmt, ...)
 					__attribute__((format(printf,5,6)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
-- 
cgit v1.2.1


From 5cbe5f8a5897470698222ac9924429056e57d84c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Thu, 10 Aug 2006 01:19:19 -0400
Subject: device_create(): make fmt argument 'const char *'

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 2 +-
 include/linux/device.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 04d089fd4f76..5d4b7e04471e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -583,7 +583,7 @@ static void device_create_release(struct device *dev)
  * been created with a call to class_create().
  */
 struct device *device_create(struct class *class, struct device *parent,
-			     dev_t devt, char *fmt, ...)
+			     dev_t devt, const char *fmt, ...)
 {
 	va_list args;
 	struct device *dev = NULL;
diff --git a/include/linux/device.h b/include/linux/device.h
index 1fec28546525..8d92013f9ce1 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -384,7 +384,7 @@ extern void device_reprobe(struct device *dev);
  * Easy functions for dynamically creating devices on the fly
  */
 extern struct device *device_create(struct class *cls, struct device *parent,
-				    dev_t devt, char *fmt, ...)
+				    dev_t devt, const char *fmt, ...)
 				    __attribute__((format(printf,4,5)));
 extern void device_destroy(struct class *cls, dev_t devt);
 
-- 
cgit v1.2.1


From ab7d7371acc68fa9130b079a9ba879191202035f Mon Sep 17 00:00:00 2001
From: Miguel Ojeda Sandonis <maxextreme@gmail.com>
Date: Wed, 13 Sep 2006 15:34:05 +0200
Subject: Driver core: add const to class_create

Adds const to class_create second parameter, because:

struct class {
	const char * name;

	/*...*/
}

Signed-off-by: Miguel Ojeda Sandonis <maxextreme@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 2 +-
 include/linux/device.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/base/class.c b/drivers/base/class.c
index 75057aaab809..e078bc21d52e 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -197,7 +197,7 @@ static int class_device_create_uevent(struct class_device *class_dev,
  * Note, the pointer created here is to be destroyed when finished by
  * making a call to class_destroy().
  */
-struct class *class_create(struct module *owner, char *name)
+struct class *class_create(struct module *owner, const char *name)
 {
 	struct class *cls;
 	int retval;
diff --git a/include/linux/device.h b/include/linux/device.h
index 8d92013f9ce1..8a648cd94fa9 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -271,7 +271,7 @@ struct class_interface {
 extern int class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
-extern struct class *class_create(struct module *owner, char *name);
+extern struct class *class_create(struct module *owner, const char *name);
 extern void class_destroy(struct class *cls);
 extern struct class_device *class_device_create(struct class *cls,
 						struct class_device *parent,
-- 
cgit v1.2.1


From eea3f8911f34001d2450a272c2d02f175279b53c Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Juha=20Yrj=F6l=E4?= <juha.yrjola@solidboot.com>
Date: Thu, 3 Aug 2006 19:06:25 +0300
Subject: sysfs: Make poll behaviour consistent

When no events have been reported by sysfs_notify(), sd->s_events
was previously set to zero.  The initial value for new readers is
also zero, so poll was blocking, regardless of whether the attribute
was read by the process or not.

Make poll behave consistently by setting the initial value of
sd->s_events to non-zero.

Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/sysfs/dir.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 61c42430cba3..5f3d725d1125 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -43,7 +43,7 @@ static struct sysfs_dirent * sysfs_new_dirent(struct sysfs_dirent * parent_sd,
 
 	memset(sd, 0, sizeof(*sd));
 	atomic_set(&sd->s_count, 1);
-	atomic_set(&sd->s_event, 0);
+	atomic_set(&sd->s_event, 1);
 	INIT_LIST_HEAD(&sd->s_children);
 	list_add(&sd->s_sibling, &parent_sd->s_children);
 	sd->s_element = element;
-- 
cgit v1.2.1


From 6468b3afa7bcd9b0abc5997e5330d78f0bb6626a Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Thu, 20 Jul 2006 08:16:42 -0700
Subject: Debugfs: kernel-doc fixes for debugfs

Fix kernel-doc and typos/spellos in fs/debugfs/.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/debugfs/file.c  | 56 ++++++++++++++++++++++++------------------------------
 fs/debugfs/inode.c | 15 ++++++---------
 2 files changed, 31 insertions(+), 40 deletions(-)

diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index 39640fd03458..e4b430552c88 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -55,12 +55,11 @@ static u64 debugfs_u8_get(void *data)
 DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n");
 
 /**
- * debugfs_create_u8 - create a file in the debugfs filesystem that is used to read and write an unsigned 8 bit value.
- *
+ * debugfs_create_u8 - create a debugfs file that is used to read and write an unsigned 8-bit value
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
- *          directory dentry if set.  If this paramater is NULL, then the
+ *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @value: a pointer to the variable that the file should read to and write
  *         from.
@@ -72,11 +71,11 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u8, debugfs_u8_get, debugfs_u8_set, "%llu\n");
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_u8(const char *name, mode_t mode,
@@ -97,12 +96,11 @@ static u64 debugfs_u16_get(void *data)
 DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n");
 
 /**
- * debugfs_create_u16 - create a file in the debugfs filesystem that is used to read and write an unsigned 16 bit value.
- *
+ * debugfs_create_u16 - create a debugfs file that is used to read and write an unsigned 16-bit value
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
- *          directory dentry if set.  If this paramater is NULL, then the
+ *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @value: a pointer to the variable that the file should read to and write
  *         from.
@@ -114,11 +112,11 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u16, debugfs_u16_get, debugfs_u16_set, "%llu\n");
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_u16(const char *name, mode_t mode,
@@ -139,12 +137,11 @@ static u64 debugfs_u32_get(void *data)
 DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n");
 
 /**
- * debugfs_create_u32 - create a file in the debugfs filesystem that is used to read and write an unsigned 32 bit value.
- *
+ * debugfs_create_u32 - create a debugfs file that is used to read and write an unsigned 32-bit value
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
- *          directory dentry if set.  If this paramater is NULL, then the
+ *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @value: a pointer to the variable that the file should read to and write
  *         from.
@@ -156,11 +153,11 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_u32, debugfs_u32_get, debugfs_u32_set, "%llu\n");
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_u32(const char *name, mode_t mode,
@@ -219,12 +216,11 @@ static const struct file_operations fops_bool = {
 };
 
 /**
- * debugfs_create_bool - create a file in the debugfs filesystem that is used to read and write a boolean value.
- *
+ * debugfs_create_bool - create a debugfs file that is used to read and write a boolean value
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
- *          directory dentry if set.  If this paramater is NULL, then the
+ *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @value: a pointer to the variable that the file should read to and write
  *         from.
@@ -236,11 +232,11 @@ static const struct file_operations fops_bool = {
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_bool(const char *name, mode_t mode,
@@ -264,13 +260,11 @@ static struct file_operations fops_blob = {
 };
 
 /**
- * debugfs_create_blob - create a file in the debugfs filesystem that is
- * used to read and write a binary blob.
- *
+ * debugfs_create_blob - create a debugfs file that is used to read and write a binary blob
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
- *          directory dentry if set.  If this paramater is NULL, then the
+ *          directory dentry if set.  If this parameter is %NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @blob: a pointer to a struct debugfs_blob_wrapper which contains a pointer
  *        to the blob data and the size of the data.
@@ -282,11 +276,11 @@ static struct file_operations fops_blob = {
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_blob(const char *name, mode_t mode,
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index e8ae3042b806..3ca268d2e5a2 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -162,7 +162,6 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
 
 /**
  * debugfs_create_file - create a file in the debugfs filesystem
- *
  * @name: a pointer to a string containing the name of the file to create.
  * @mode: the permission that the file should have
  * @parent: a pointer to the parent dentry for this file.  This should be a
@@ -182,11 +181,11 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_file(const char *name, mode_t mode,
@@ -221,7 +220,6 @@ EXPORT_SYMBOL_GPL(debugfs_create_file);
 
 /**
  * debugfs_create_dir - create a directory in the debugfs filesystem
- *
  * @name: a pointer to a string containing the name of the directory to
  *        create.
  * @parent: a pointer to the parent dentry for this file.  This should be a
@@ -233,11 +231,11 @@ EXPORT_SYMBOL_GPL(debugfs_create_file);
  * This function will return a pointer to a dentry if it succeeds.  This
  * pointer must be passed to the debugfs_remove() function when the file is
  * to be removed (no automatic cleanup happens if your module is unloaded,
- * you are responsible here.)  If an error occurs, NULL will be returned.
+ * you are responsible here.)  If an error occurs, %NULL will be returned.
  *
- * If debugfs is not enabled in the kernel, the value -ENODEV will be
+ * If debugfs is not enabled in the kernel, the value -%ENODEV will be
  * returned.  It is not wise to check for this value, but rather, check for
- * NULL or !NULL instead as to eliminate the need for #ifdef in the calling
+ * %NULL or !%NULL instead as to eliminate the need for #ifdef in the calling
  * code.
  */
 struct dentry *debugfs_create_dir(const char *name, struct dentry *parent)
@@ -250,7 +248,6 @@ EXPORT_SYMBOL_GPL(debugfs_create_dir);
 
 /**
  * debugfs_remove - removes a file or directory from the debugfs filesystem
- *
  * @dentry: a pointer to a the dentry of the file or directory to be
  *          removed.
  *
-- 
cgit v1.2.1


From ceeee1fb2897651b434547eb26d93e6d2ff5a1a5 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 9 Apr 2002 12:14:34 -0700
Subject: SYSFS: allow sysfs_create_link to create symlinks in the root of
 sysfs

This is needed to make the compatible link for /sys/block in the future.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/sysfs/symlink.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c
index d2eac3ceed5f..f50e3cc2ded8 100644
--- a/fs/sysfs/symlink.c
+++ b/fs/sysfs/symlink.c
@@ -3,6 +3,7 @@
  */
 
 #include <linux/fs.h>
+#include <linux/mount.h>
 #include <linux/module.h>
 #include <linux/kobject.h>
 #include <linux/namei.h>
@@ -82,10 +83,19 @@ exit1:
  */
 int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name)
 {
-	struct dentry * dentry = kobj->dentry;
+	struct dentry *dentry = NULL;
 	int error = -EEXIST;
 
-	BUG_ON(!kobj || !kobj->dentry || !name);
+	BUG_ON(!name);
+
+	if (!kobj) {
+		if (sysfs_mount && sysfs_mount->mnt_sb)
+			dentry = sysfs_mount->mnt_sb->s_root;
+	} else
+		dentry = kobj->dentry;
+
+	if (!dentry)
+		return -EFAULT;
 
 	mutex_lock(&dentry->d_inode->i_mutex);
 	if (!sysfs_dirent_exist(dentry->d_fsdata, name))
-- 
cgit v1.2.1


From 7c8265f51073bc8632a99de78d5fd19117ed78b7 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@osdl.org>
Date: Sat, 24 Jun 2006 14:50:29 -0700
Subject: Suspend infrastructure cleanup and extension

Allow devices to participate in the suspend process more intimately,
in particular, allow the final phase (with interrupts disabled) to
also be open to normal devices, not just system devices.

Also, allow classes to participate in device suspend.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/resume.c  |  28 ++++++++--
 drivers/base/power/suspend.c | 122 ++++++++++++++++++++++++++++++++-----------
 include/linux/device.h       |  11 +++-
 include/linux/pm.h           |   1 +
 kernel/power/main.c          |   4 ++
 5 files changed, 130 insertions(+), 36 deletions(-)

diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c
index 826093ef4c7e..48e3d49d7b65 100644
--- a/drivers/base/power/resume.c
+++ b/drivers/base/power/resume.c
@@ -38,13 +38,35 @@ int resume_device(struct device * dev)
 		dev_dbg(dev,"resuming\n");
 		error = dev->bus->resume(dev);
 	}
+	if (dev->class && dev->class->resume) {
+		dev_dbg(dev,"class resume\n");
+		error = dev->class->resume(dev);
+	}
 	up(&dev->sem);
 	TRACE_RESUME(error);
 	return error;
 }
 
 
+static int resume_device_early(struct device * dev)
+{
+	int error = 0;
 
+	TRACE_DEVICE(dev);
+	TRACE_RESUME(0);
+	if (dev->bus && dev->bus->resume_early) {
+		dev_dbg(dev,"EARLY resume\n");
+		error = dev->bus->resume_early(dev);
+	}
+	TRACE_RESUME(error);
+	return error;
+}
+
+/*
+ * Resume the devices that have either not gone through
+ * the late suspend, or that did go through it but also
+ * went through the early resume
+ */
 void dpm_resume(void)
 {
 	down(&dpm_list_sem);
@@ -99,10 +121,8 @@ void dpm_power_up(void)
 		struct list_head * entry = dpm_off_irq.next;
 		struct device * dev = to_device(entry);
 
-		get_device(dev);
-		list_move_tail(entry, &dpm_active);
-		resume_device(dev);
-		put_device(dev);
+		list_move_tail(entry, &dpm_off);
+		resume_device_early(dev);
 	}
 }
 
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 69509e02f703..10e8032aee1a 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -65,7 +65,19 @@ int suspend_device(struct device * dev, pm_message_t state)
 
 	dev->power.prev_state = dev->power.power_state;
 
-	if (dev->bus && dev->bus->suspend && !dev->power.power_state.event) {
+	if (dev->class && dev->class->suspend && !dev->power.power_state.event) {
+		dev_dbg(dev, "class %s%s\n",
+			suspend_verb(state.event),
+			((state.event == PM_EVENT_SUSPEND)
+					&& device_may_wakeup(dev))
+				? ", may wakeup"
+				: ""
+			);
+		error = dev->class->suspend(dev, state);
+		suspend_report_result(dev->class->suspend, error);
+	}
+
+	if (!error && dev->bus && dev->bus->suspend && !dev->power.power_state.event) {
 		dev_dbg(dev, "%s%s\n",
 			suspend_verb(state.event),
 			((state.event == PM_EVENT_SUSPEND)
@@ -81,15 +93,74 @@ int suspend_device(struct device * dev, pm_message_t state)
 }
 
 
+/*
+ * This is called with interrupts off, only a single CPU
+ * running. We can't do down() on a semaphore (and we don't
+ * need the protection)
+ */
+static int suspend_device_late(struct device *dev, pm_message_t state)
+{
+	int error = 0;
+
+	if (dev->power.power_state.event) {
+		dev_dbg(dev, "PM: suspend_late %d-->%d\n",
+			dev->power.power_state.event, state.event);
+	}
+
+	if (dev->bus && dev->bus->suspend_late && !dev->power.power_state.event) {
+		dev_dbg(dev, "LATE %s%s\n",
+			suspend_verb(state.event),
+			((state.event == PM_EVENT_SUSPEND)
+					&& device_may_wakeup(dev))
+				? ", may wakeup"
+				: ""
+			);
+		error = dev->bus->suspend_late(dev, state);
+		suspend_report_result(dev->bus->suspend_late, error);
+	}
+	return error;
+}
+
+/**
+ *	device_prepare_suspend - save state and prepare to suspend
+ *
+ *	NOTE! Devices cannot detach at this point - not only do we
+ *	hold the device list semaphores over the whole prepare, but
+ *	the whole point is to do non-invasive preparatory work, not
+ *	the actual suspend.
+ */
+int device_prepare_suspend(pm_message_t state)
+{
+	int error = 0;
+	struct device * dev;
+
+	down(&dpm_sem);
+	down(&dpm_list_sem);
+	list_for_each_entry_reverse(dev, &dpm_active, power.entry) {
+		if (!dev->bus || !dev->bus->suspend_prepare)
+			continue;
+		error = dev->bus->suspend_prepare(dev, state);
+		if (error)
+			break;
+	}
+	up(&dpm_list_sem);
+	up(&dpm_sem);
+	return error;
+}
+
 /**
  *	device_suspend - Save state and stop all devices in system.
  *	@state:		Power state to put each device in.
  *
  *	Walk the dpm_active list, call ->suspend() for each device, and move
- *	it to dpm_off.
- *	Check the return value for each. If it returns 0, then we move the
- *	the device to the dpm_off list. If it returns -EAGAIN, we move it to
- *	the dpm_off_irq list. If we get a different error, try and back out.
+ *	it to the dpm_off list.
+ *
+ *	(For historical reasons, if it returns -EAGAIN, that used to mean
+ *	that the device would be called again with interrupts disabled.
+ *	These days, we use the "suspend_late()" callback for that, so we
+ *	print a warning and consider it an error).
+ *
+ *	If we get a different error, try and back out.
  *
  *	If we hit a failure with any of the devices, call device_resume()
  *	above to bring the suspended devices back to life.
@@ -115,39 +186,27 @@ int device_suspend(pm_message_t state)
 
 		/* Check if the device got removed */
 		if (!list_empty(&dev->power.entry)) {
-			/* Move it to the dpm_off or dpm_off_irq list */
+			/* Move it to the dpm_off list */
 			if (!error)
 				list_move(&dev->power.entry, &dpm_off);
-			else if (error == -EAGAIN) {
-				list_move(&dev->power.entry, &dpm_off_irq);
-				error = 0;
-			}
 		}
 		if (error)
 			printk(KERN_ERR "Could not suspend device %s: "
-				"error %d\n", kobject_name(&dev->kobj), error);
+				"error %d%s\n",
+				kobject_name(&dev->kobj), error,
+				error == -EAGAIN ? " (please convert to suspend_late)" : "");
 		put_device(dev);
 	}
 	up(&dpm_list_sem);
-	if (error) {
-		/* we failed... before resuming, bring back devices from
-		 * dpm_off_irq list back to main dpm_off list, we do want
-		 * to call resume() on them, in case they partially suspended
-		 * despite returning -EAGAIN
-		 */
-		while (!list_empty(&dpm_off_irq)) {
-			struct list_head * entry = dpm_off_irq.next;
-			list_move(entry, &dpm_off);
-		}
+	if (error)
 		dpm_resume();
-	}
+
 	up(&dpm_sem);
 	return error;
 }
 
 EXPORT_SYMBOL_GPL(device_suspend);
 
-
 /**
  *	device_power_down - Shut down special devices.
  *	@state:		Power state to enter.
@@ -162,14 +221,17 @@ int device_power_down(pm_message_t state)
 	int error = 0;
 	struct device * dev;
 
-	list_for_each_entry_reverse(dev, &dpm_off_irq, power.entry) {
-		if ((error = suspend_device(dev, state)))
-			break;
+	while (!list_empty(&dpm_off)) {
+		struct list_head * entry = dpm_off.prev;
+
+		dev = to_device(entry);
+		error = suspend_device_late(dev, state);
+		if (error)
+			goto Error;
+		list_move(&dev->power.entry, &dpm_off_irq);
 	}
-	if (error)
-		goto Error;
-	if ((error = sysdev_suspend(state)))
-		goto Error;
+
+	error = sysdev_suspend(state);
  Done:
 	return error;
  Error:
diff --git a/include/linux/device.h b/include/linux/device.h
index 8a648cd94fa9..b40be6fca6fa 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -51,8 +51,12 @@ struct bus_type {
 	int		(*probe)(struct device * dev);
 	int		(*remove)(struct device * dev);
 	void		(*shutdown)(struct device * dev);
-	int		(*suspend)(struct device * dev, pm_message_t state);
-	int		(*resume)(struct device * dev);
+
+	int (*suspend_prepare)(struct device * dev, pm_message_t state);
+	int (*suspend)(struct device * dev, pm_message_t state);
+	int (*suspend_late)(struct device * dev, pm_message_t state);
+	int (*resume_early)(struct device * dev);
+	int (*resume)(struct device * dev);
 };
 
 extern int bus_register(struct bus_type * bus);
@@ -154,6 +158,9 @@ struct class {
 
 	void	(*release)(struct class_device *dev);
 	void	(*class_release)(struct class *class);
+
+	int	(*suspend)(struct device *, pm_message_t state);
+	int	(*resume)(struct device *);
 };
 
 extern int class_register(struct class *);
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 658c1b93d5bb..096fb6f754cf 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -190,6 +190,7 @@ extern void device_resume(void);
 extern suspend_disk_method_t pm_disk_mode;
 
 extern int device_suspend(pm_message_t state);
+extern int device_prepare_suspend(pm_message_t state);
 
 #define device_set_wakeup_enable(dev,val) \
 	((dev)->power.should_wakeup = !!(val))
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6d295c776794..0c3ed6ac938e 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -57,6 +57,10 @@ static int suspend_prepare(suspend_state_t state)
 	if (!pm_ops || !pm_ops->enter)
 		return -EPERM;
 
+	error = device_prepare_suspend(PMSG_SUSPEND);
+	if (error)
+		return error;
+
 	pm_prepare_console();
 
 	disable_nonboot_cpus();
-- 
cgit v1.2.1


From cbd69dbbf1adfce6e048f15afc8629901ca9dae5 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@osdl.org>
Date: Sat, 24 Jun 2006 14:50:29 -0700
Subject: Suspend changes for PCI core

Changes the PCI core to use the new suspend infrastructure changes.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci-driver.c | 41 ++++++++++++++++++++++++++++++++++++++++-
 include/linux/pci.h      |  3 +++
 2 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 474e9cd0e9e4..9e7d6ceb3805 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -264,6 +264,19 @@ static int pci_device_remove(struct device * dev)
 	return 0;
 }
 
+static int pci_device_suspend_prepare(struct device * dev, pm_message_t state)
+{
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+	int i = 0;
+
+	if (drv && drv->suspend_prepare) {
+		i = drv->suspend_prepare(pci_dev, state);
+		suspend_report_result(drv->suspend_prepare, i);
+	}
+	return i;
+}
+
 static int pci_device_suspend(struct device * dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
@@ -279,6 +292,18 @@ static int pci_device_suspend(struct device * dev, pm_message_t state)
 	return i;
 }
 
+static int pci_device_suspend_late(struct device * dev, pm_message_t state)
+{
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+	int i = 0;
+
+	if (drv && drv->suspend_late) {
+		i = drv->suspend_late(pci_dev, state);
+		suspend_report_result(drv->suspend_late, i);
+	}
+	return i;
+}
 
 /*
  * Default resume method for devices that have no driver provided resume,
@@ -313,6 +338,17 @@ static int pci_device_resume(struct device * dev)
 	return error;
 }
 
+static int pci_device_resume_early(struct device * dev)
+{
+	int error = 0;
+	struct pci_dev * pci_dev = to_pci_dev(dev);
+	struct pci_driver * drv = pci_dev->driver;
+
+	if (drv && drv->resume_early)
+		error = drv->resume_early(pci_dev);
+	return error;
+}
+
 static void pci_device_shutdown(struct device *dev)
 {
 	struct pci_dev *pci_dev = to_pci_dev(dev);
@@ -508,9 +544,12 @@ struct bus_type pci_bus_type = {
 	.uevent		= pci_uevent,
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
+	.suspend_prepare= pci_device_suspend_prepare,
 	.suspend	= pci_device_suspend,
-	.shutdown	= pci_device_shutdown,
+	.suspend_late	= pci_device_suspend_late,
+	.resume_early	= pci_device_resume_early,
 	.resume		= pci_device_resume,
+	.shutdown	= pci_device_shutdown,
 	.dev_attrs	= pci_dev_attrs,
 };
 
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 8565b81d7fbc..4b2e629467c7 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -345,7 +345,10 @@ struct pci_driver {
 	const struct pci_device_id *id_table;	/* must be non-NULL for probe to be called */
 	int  (*probe)  (struct pci_dev *dev, const struct pci_device_id *id);	/* New device inserted */
 	void (*remove) (struct pci_dev *dev);	/* Device removed (NULL if not a hot-plug capable driver) */
+	int  (*suspend_prepare) (struct pci_dev *dev, pm_message_t state);
 	int  (*suspend) (struct pci_dev *dev, pm_message_t state);	/* Device suspended */
+	int  (*suspend_late) (struct pci_dev *dev, pm_message_t state);
+	int  (*resume_early) (struct pci_dev *dev);
 	int  (*resume) (struct pci_dev *dev);	                /* Device woken up */
 	int  (*enable_wake) (struct pci_dev *dev, pci_power_t state, int enable);   /* Enable wake event */
 	void (*shutdown) (struct pci_dev *dev);
-- 
cgit v1.2.1


From 8f4bcc20ee39f9c087f3532672e3e5f086e92281 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 9 Jul 2006 16:28:28 -0700
Subject: make suspend quieter

Fix a goof in Linus' recent PM API updates:  don't emit any messages in the
typical NOP "already suspended it" late suspend case.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/suspend.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 10e8032aee1a..0bda4a7f2042 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -102,11 +102,6 @@ static int suspend_device_late(struct device *dev, pm_message_t state)
 {
 	int error = 0;
 
-	if (dev->power.power_state.event) {
-		dev_dbg(dev, "PM: suspend_late %d-->%d\n",
-			dev->power.power_state.event, state.event);
-	}
-
 	if (dev->bus && dev->bus->suspend_late && !dev->power.power_state.event) {
 		dev_dbg(dev, "LATE %s%s\n",
 			suspend_verb(state.event),
-- 
cgit v1.2.1


From 8b4b8a24e4e49dc9fe36d4d079f6d2c23f942b03 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:03 -0700
Subject: fix broken/dubious driver suspend() methods

Small driver suspend() fixes in preparation for the PRETHAW events:

 - Only compare message events for equality against PM_EVENT_* codes;
   not against integers, or using greater/less-than comparisons.
   (PM_EVENT_* should really become a __bitwise thing.)

 - Explicitly test for SUSPEND events (rather than not-something-else)
   before suspending devices.

 - Removes more of the confusion between a pm_message_t (wraps event code)
   and a "state" ... suspend() originally took a target system state.

These updates are correct and appropriate even without new PM_EVENT codes.

benh: "I think in the Mesh case, we should handle the freeze case as well or
we might get wild DMA."

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: Greg KH <greg@kroah.com>
Cc: Paul Mackerras <paulus@samba.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/ide/ppc/pmac.c                  | 14 ++++++++------
 drivers/media/dvb/cinergyT2/cinergyT2.c |  2 +-
 drivers/scsi/mesh.c                     | 15 +++++++++++----
 3 files changed, 20 insertions(+), 11 deletions(-)

diff --git a/drivers/ide/ppc/pmac.c b/drivers/ide/ppc/pmac.c
index 996c694341bc..31ad79f52df7 100644
--- a/drivers/ide/ppc/pmac.c
+++ b/drivers/ide/ppc/pmac.c
@@ -1369,15 +1369,16 @@ pmac_ide_macio_attach(struct macio_dev *mdev, const struct of_device_id *match)
 }
 
 static int
-pmac_ide_macio_suspend(struct macio_dev *mdev, pm_message_t state)
+pmac_ide_macio_suspend(struct macio_dev *mdev, pm_message_t mesg)
 {
 	ide_hwif_t	*hwif = (ide_hwif_t *)dev_get_drvdata(&mdev->ofdev.dev);
 	int		rc = 0;
 
-	if (state.event != mdev->ofdev.dev.power.power_state.event && state.event >= PM_EVENT_SUSPEND) {
+	if (mesg.event != mdev->ofdev.dev.power.power_state.event
+			&& mesg.event == PM_EVENT_SUSPEND) {
 		rc = pmac_ide_do_suspend(hwif);
 		if (rc == 0)
-			mdev->ofdev.dev.power.power_state = state;
+			mdev->ofdev.dev.power.power_state = mesg;
 	}
 
 	return rc;
@@ -1473,15 +1474,16 @@ pmac_ide_pci_attach(struct pci_dev *pdev, const struct pci_device_id *id)
 }
 
 static int
-pmac_ide_pci_suspend(struct pci_dev *pdev, pm_message_t state)
+pmac_ide_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
 	ide_hwif_t	*hwif = (ide_hwif_t *)pci_get_drvdata(pdev);
 	int		rc = 0;
 	
-	if (state.event != pdev->dev.power.power_state.event && state.event >= 2) {
+	if (mesg.event != pdev->dev.power.power_state.event
+			&& mesg.event == PM_EVENT_SUSPEND) {
 		rc = pmac_ide_do_suspend(hwif);
 		if (rc == 0)
-			pdev->dev.power.power_state = state;
+			pdev->dev.power.power_state = mesg;
 	}
 
 	return rc;
diff --git a/drivers/media/dvb/cinergyT2/cinergyT2.c b/drivers/media/dvb/cinergyT2/cinergyT2.c
index 001c71b6be61..410fa6d620ff 100644
--- a/drivers/media/dvb/cinergyT2/cinergyT2.c
+++ b/drivers/media/dvb/cinergyT2/cinergyT2.c
@@ -981,7 +981,7 @@ static int cinergyt2_suspend (struct usb_interface *intf, pm_message_t state)
 	if (cinergyt2->disconnect_pending || mutex_lock_interruptible(&cinergyt2->sem))
 		return -ERESTARTSYS;
 
-	if (state.event > PM_EVENT_ON) {
+	if (1) {
 		struct cinergyt2 *cinergyt2 = usb_get_intfdata (intf);
 
 		cinergyt2_suspend_rc(cinergyt2);
diff --git a/drivers/scsi/mesh.c b/drivers/scsi/mesh.c
index 592b52afe658..683fc7ae4b8f 100644
--- a/drivers/scsi/mesh.c
+++ b/drivers/scsi/mesh.c
@@ -1756,16 +1756,23 @@ static void set_mesh_power(struct mesh_state *ms, int state)
 		pmac_call_feature(PMAC_FTR_MESH_ENABLE, macio_get_of_node(ms->mdev), 0, 0);
 		msleep(10);
 	}
-}			
+}
 
 
 #ifdef CONFIG_PM
-static int mesh_suspend(struct macio_dev *mdev, pm_message_t state)
+static int mesh_suspend(struct macio_dev *mdev, pm_message_t mesg)
 {
 	struct mesh_state *ms = (struct mesh_state *)macio_get_drvdata(mdev);
 	unsigned long flags;
 
-	if (state.event == mdev->ofdev.dev.power.power_state.event || state.event < 2)
+	switch (mesg.event) {
+	case PM_EVENT_SUSPEND:
+	case PM_EVENT_FREEZE:
+		break;
+	default:
+		return 0;
+	}
+	if (mesg.event == mdev->ofdev.dev.power.power_state.event)
 		return 0;
 
 	scsi_block_requests(ms->host);
@@ -1780,7 +1787,7 @@ static int mesh_suspend(struct macio_dev *mdev, pm_message_t state)
 	disable_irq(ms->meshintr);
 	set_mesh_power(ms, 0);
 
-	mdev->ofdev.dev.power.power_state = state;
+	mdev->ofdev.dev.power.power_state = mesg;
 
 	return 0;
 }
-- 
cgit v1.2.1


From 82bb67f2c1f9ef438c56ac24e7dca027fe7289b5 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:04 -0700
Subject: PM: define PM_EVENT_PRETHAW

This adds a new pm_message_t event type to use when preparing to restore a
swsusp snapshot.  Devices that have been initialized by Linux after resume
(rather than left in power-up-reset state) may need to be reset; this new
event type give drivers the chance to do that.

The drivers that will care about this are those which understand more hardware
states than just "on" and "reset", relying on hardware state during resume()
methods to be either the state left by the preceding suspend(), or a
power-lost reset.  The best current example of this class of drivers are USB
host controller drivers, which currently do not work through swsusp when
they're statically linked.

When the swsusp freeze/thaw mechanism kicks in, a troublesome third state
could exist: one state set up by a different kernel instance, before a
snapshot image is resumed.  This mechanism lets drivers prevent that state.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/pm.h | 62 +++++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 47 insertions(+), 15 deletions(-)

diff --git a/include/linux/pm.h b/include/linux/pm.h
index 096fb6f754cf..6b27e07aef19 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -142,29 +142,61 @@ typedef struct pm_message {
 } pm_message_t;
 
 /*
- * There are 4 important states driver can be in:
- * ON     -- driver is working
- * FREEZE -- stop operations and apply whatever policy is applicable to a
- *           suspended driver of that class, freeze queues for block like IDE
- *           does, drop packets for ethernet, etc... stop DMA engine too etc...
- *           so a consistent image can be saved; but do not power any hardware
- *           down.
- * SUSPEND - like FREEZE, but hardware is doing as much powersaving as
- *           possible. Roughly pci D3.
+ * Several driver power state transitions are externally visible, affecting
+ * the state of pending I/O queues and (for drivers that touch hardware)
+ * interrupts, wakeups, DMA, and other hardware state.  There may also be
+ * internal transitions to various low power modes, which are transparent
+ * to the rest of the driver stack (such as a driver that's ON gating off
+ * clocks which are not in active use).
  *
- * Unfortunately, current drivers only recognize numeric values 0 (ON) and 3
- * (SUSPEND).  We'll need to fix the drivers. So yes, putting 3 to all different
- * defines is intentional, and will go away as soon as drivers are fixed.  Also
- * note that typedef is neccessary, we'll probably want to switch to
- *   typedef struct pm_message_t { int event; int flags; } pm_message_t
- * or something similar soon.
+ * One transition is triggered by resume(), after a suspend() call; the
+ * message is implicit:
+ *
+ * ON		Driver starts working again, responding to hardware events
+ * 		and software requests.  The hardware may have gone through
+ * 		a power-off reset, or it may have maintained state from the
+ * 		previous suspend() which the driver will rely on while
+ * 		resuming.  On most platforms, there are no restrictions on
+ * 		availability of resources like clocks during resume().
+ *
+ * Other transitions are triggered by messages sent using suspend().  All
+ * these transitions quiesce the driver, so that I/O queues are inactive.
+ * That commonly entails turning off IRQs and DMA; there may be rules
+ * about how to quiesce that are specific to the bus or the device's type.
+ * (For example, network drivers mark the link state.)  Other details may
+ * differ according to the message:
+ *
+ * SUSPEND	Quiesce, enter a low power device state appropriate for
+ * 		the upcoming system state (such as PCI_D3hot), and enable
+ * 		wakeup events as appropriate.
+ *
+ * FREEZE	Quiesce operations so that a consistent image can be saved;
+ * 		but do NOT otherwise enter a low power device state, and do
+ * 		NOT emit system wakeup events.
+ *
+ * PRETHAW	Quiesce as if for FREEZE; additionally, prepare for restoring
+ * 		the system from a snapshot taken after an earlier FREEZE.
+ * 		Some drivers will need to reset their hardware state instead
+ * 		of preserving it, to ensure that it's never mistaken for the
+ * 		state which that earlier snapshot had set up.
+ *
+ * A minimally power-aware driver treats all messages as SUSPEND, fully
+ * reinitializes its device during resume() -- whether or not it was reset
+ * during the suspend/resume cycle -- and can't issue wakeup events.
+ *
+ * More power-aware drivers may also use low power states at runtime as
+ * well as during system sleep states like PM_SUSPEND_STANDBY.  They may
+ * be able to use wakeup events to exit from runtime low-power states,
+ * or from system low-power states such as standby or suspend-to-RAM.
  */
 
 #define PM_EVENT_ON 0
 #define PM_EVENT_FREEZE 1
 #define PM_EVENT_SUSPEND 2
+#define PM_EVENT_PRETHAW 3
 
 #define PMSG_FREEZE	((struct pm_message){ .event = PM_EVENT_FREEZE, })
+#define PMSG_PRETHAW	((struct pm_message){ .event = PM_EVENT_PRETHAW, })
 #define PMSG_SUSPEND	((struct pm_message){ .event = PM_EVENT_SUSPEND, })
 #define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
 
-- 
cgit v1.2.1


From b887d2e63c8857149ef59eb6e05adfaa018b8ebf Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:05 -0700
Subject: PM: PCI and IDE handle PM_EVENT_PRETHAW

Convert some framework code to handle the new PRETHAW message.

  - IDE just treats it like a FREEZE.

  - The pci_choose_state() thingie still doesn't use PCI_D0 when it gets a
    FREEZE (and now PRETHAW) event, which seems rather buglike but wasn't
    something to change with this patch.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/ide/ide.c | 6 ++++--
 drivers/pci/pci.c | 4 +++-
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index defd4b4bd374..9c8468de1a75 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -1207,7 +1207,7 @@ int system_bus_clock (void)
 
 EXPORT_SYMBOL(system_bus_clock);
 
-static int generic_ide_suspend(struct device *dev, pm_message_t state)
+static int generic_ide_suspend(struct device *dev, pm_message_t mesg)
 {
 	ide_drive_t *drive = dev->driver_data;
 	struct request rq;
@@ -1221,7 +1221,9 @@ static int generic_ide_suspend(struct device *dev, pm_message_t state)
 	rq.special = &args;
 	rq.end_io_data = &rqpm;
 	rqpm.pm_step = ide_pm_state_start_suspend;
-	rqpm.pm_state = state.event;
+	if (mesg.event == PM_EVENT_PRETHAW)
+		mesg.event = PM_EVENT_FREEZE;
+	rqpm.pm_state = mesg.event;
 
 	return ide_do_drive_cmd(drive, &rq, ide_wait);
 }
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 9f79dd6d51ab..8ab027886034 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -432,10 +432,12 @@ pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state)
 	case PM_EVENT_ON:
 		return PCI_D0;
 	case PM_EVENT_FREEZE:
+	case PM_EVENT_PRETHAW:
+		/* REVISIT both freeze and pre-thaw "should" use D0 */
 	case PM_EVENT_SUSPEND:
 		return PCI_D3hot;
 	default:
-		printk("They asked me for state %d\n", state.event);
+		printk("Unrecognized suspend event %d\n", state.event);
 		BUG();
 	}
 	return PCI_D0;
-- 
cgit v1.2.1


From c78a7c2dd913e68ce853d43edaba14eac91b2fd1 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:06 -0700
Subject: PM: video drivers and PM_EVENT_PRETHAW

Video drivers which explicitly test for messages reporting PM_EVENT_FREEZE
will now handle PM_EVENT_PRETHAW the same way.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/video/aty/radeon_pm.c          | 15 +++++++++------
 drivers/video/i810/i810_main.c         | 12 +++++++-----
 drivers/video/nvidia/nvidia.c          | 13 +++++++------
 drivers/video/savage/savagefb_driver.c | 14 +++++++-------
 4 files changed, 30 insertions(+), 24 deletions(-)

diff --git a/drivers/video/aty/radeon_pm.c b/drivers/video/aty/radeon_pm.c
index e308ed2d249a..365de5dcc888 100644
--- a/drivers/video/aty/radeon_pm.c
+++ b/drivers/video/aty/radeon_pm.c
@@ -2621,25 +2621,28 @@ static int radeon_restore_pci_cfg(struct radeonfb_info *rinfo)
 }
 
 
-int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t state)
+int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t mesg)
 {
         struct fb_info *info = pci_get_drvdata(pdev);
         struct radeonfb_info *rinfo = info->par;
 	int i;
 
-	if (state.event == pdev->dev.power.power_state.event)
+	if (mesg.event == pdev->dev.power.power_state.event)
 		return 0;
 
-	printk(KERN_DEBUG "radeonfb (%s): suspending to state: %d...\n",
-	       pci_name(pdev), state.event);
+	printk(KERN_DEBUG "radeonfb (%s): suspending for event: %d...\n",
+	       pci_name(pdev), mesg.event);
 
 	/* For suspend-to-disk, we cheat here. We don't suspend anything and
 	 * let fbcon continue drawing until we are all set. That shouldn't
 	 * really cause any problem at this point, provided that the wakeup
 	 * code knows that any state in memory may not match the HW
 	 */
-	if (state.event == PM_EVENT_FREEZE)
+	switch (mesg.event) {
+	case PM_EVENT_FREEZE:		/* about to take snapshot */
+	case PM_EVENT_PRETHAW:		/* before restoring snapshot */
 		goto done;
+	}
 
 	acquire_console_sem();
 
@@ -2706,7 +2709,7 @@ int radeonfb_pci_suspend(struct pci_dev *pdev, pm_message_t state)
 	release_console_sem();
 
  done:
-	pdev->dev.power.power_state = state;
+	pdev->dev.power.power_state = mesg;
 
 	return 0;
 }
diff --git a/drivers/video/i810/i810_main.c b/drivers/video/i810/i810_main.c
index a6ca02f2156a..d42edaccb84c 100644
--- a/drivers/video/i810/i810_main.c
+++ b/drivers/video/i810/i810_main.c
@@ -1554,15 +1554,17 @@ static struct fb_ops i810fb_ops __devinitdata = {
 /***********************************************************************
  *                         Power Management                            *
  ***********************************************************************/
-static int i810fb_suspend(struct pci_dev *dev, pm_message_t state)
+static int i810fb_suspend(struct pci_dev *dev, pm_message_t mesg)
 {
 	struct fb_info *info = pci_get_drvdata(dev);
 	struct i810fb_par *par = info->par;
 
-	par->cur_state = state.event;
+	par->cur_state = mesg.event;
 
-	if (state.event == PM_EVENT_FREEZE) {
-		dev->dev.power.power_state = state;
+	switch (mesg.event) {
+	case PM_EVENT_FREEZE:
+	case PM_EVENT_PRETHAW:
+		dev->dev.power.power_state = mesg;
 		return 0;
 	}
 
@@ -1578,7 +1580,7 @@ static int i810fb_suspend(struct pci_dev *dev, pm_message_t state)
 
 	pci_save_state(dev);
 	pci_disable_device(dev);
-	pci_set_power_state(dev, pci_choose_state(dev, state));
+	pci_set_power_state(dev, pci_choose_state(dev, mesg));
 	release_console_sem();
 
 	return 0;
diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c
index d4f850117874..f8cd4c519aeb 100644
--- a/drivers/video/nvidia/nvidia.c
+++ b/drivers/video/nvidia/nvidia.c
@@ -950,24 +950,25 @@ static struct fb_ops nvidia_fb_ops = {
 };
 
 #ifdef CONFIG_PM
-static int nvidiafb_suspend(struct pci_dev *dev, pm_message_t state)
+static int nvidiafb_suspend(struct pci_dev *dev, pm_message_t mesg)
 {
 	struct fb_info *info = pci_get_drvdata(dev);
 	struct nvidia_par *par = info->par;
 
+	if (mesg.event == PM_EVENT_PRETHAW)
+		mesg.event = PM_EVENT_FREEZE;
 	acquire_console_sem();
-	par->pm_state = state.event;
+	par->pm_state = mesg.event;
 
-	if (state.event == PM_EVENT_FREEZE) {
-		dev->dev.power.power_state = state;
-	} else {
+	if (mesg.event == PM_EVENT_SUSPEND) {
 		fb_set_suspend(info, 1);
 		nvidiafb_blank(FB_BLANK_POWERDOWN, info);
 		nvidia_write_regs(par, &par->SavedReg);
 		pci_save_state(dev);
 		pci_disable_device(dev);
-		pci_set_power_state(dev, pci_choose_state(dev, state));
+		pci_set_power_state(dev, pci_choose_state(dev, mesg));
 	}
+	dev->dev.power.power_state = mesg;
 
 	release_console_sem();
 	return 0;
diff --git a/drivers/video/savage/savagefb_driver.c b/drivers/video/savage/savagefb_driver.c
index 461e094e7b45..82b3deaae02d 100644
--- a/drivers/video/savage/savagefb_driver.c
+++ b/drivers/video/savage/savagefb_driver.c
@@ -2323,24 +2323,24 @@ static void __devexit savagefb_remove(struct pci_dev *dev)
 	}
 }
 
-static int savagefb_suspend(struct pci_dev* dev, pm_message_t state)
+static int savagefb_suspend(struct pci_dev *dev, pm_message_t mesg)
 {
 	struct fb_info *info = pci_get_drvdata(dev);
 	struct savagefb_par *par = info->par;
 
 	DBG("savagefb_suspend");
 
-
-	par->pm_state = state.event;
+	if (mesg.event == PM_EVENT_PRETHAW)
+		mesg.event = PM_EVENT_FREEZE;
+	par->pm_state = mesg.event;
+	dev->dev.power.power_state = mesg;
 
 	/*
 	 * For PM_EVENT_FREEZE, do not power down so the console
 	 * can remain active.
 	 */
-	if (state.event == PM_EVENT_FREEZE) {
-		dev->dev.power.power_state = state;
+	if (mesg.event == PM_EVENT_FREEZE)
 		return 0;
-	}
 
 	acquire_console_sem();
 	fb_set_suspend(info, 1);
@@ -2353,7 +2353,7 @@ static int savagefb_suspend(struct pci_dev* dev, pm_message_t state)
 	savage_disable_mmio(par);
 	pci_save_state(dev);
 	pci_disable_device(dev);
-	pci_set_power_state(dev, pci_choose_state(dev, state));
+	pci_set_power_state(dev, pci_choose_state(dev, mesg));
 	release_console_sem();
 
 	return 0;
-- 
cgit v1.2.1


From 185849991d592497e43bcd264c6152af1261ffe2 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:06 -0700
Subject: PM: USB HCDs use PM_EVENT_PRETHAW

This teaches several USB host controller drivers to treat PRETHAW as a chip
reset since the controller, and all devices connected to it, are no longer in
states compatible with how the snapshotted suspend() left them.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd-pci.c   | 2 +-
 drivers/usb/host/ehci-pci.c  | 6 ++++++
 drivers/usb/host/ohci-pci.c  | 5 +++++
 drivers/usb/host/sl811-hcd.c | 9 +++++++--
 drivers/usb/host/uhci-hcd.c  | 4 ++++
 5 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c
index 5078fb3375e3..fa36391fedd3 100644
--- a/drivers/usb/core/hcd-pci.c
+++ b/drivers/usb/core/hcd-pci.c
@@ -281,7 +281,7 @@ int usb_hcd_pci_suspend (struct pci_dev *dev, pm_message_t message)
 			(void) usb_hcd_pci_resume (dev);
 		}
 
-	} else {
+	} else if (hcd->state != HC_STATE_HALT) {
 		dev_dbg (hcd->self.controller, "hcd state %d; not suspended\n",
 			hcd->state);
 		WARN_ON(1);
diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c
index cadffacd945b..6967ab71e282 100644
--- a/drivers/usb/host/ehci-pci.c
+++ b/drivers/usb/host/ehci-pci.c
@@ -238,6 +238,12 @@ static int ehci_pci_suspend(struct usb_hcd *hcd, pm_message_t message)
 	writel (0, &ehci->regs->intr_enable);
 	(void)readl(&ehci->regs->intr_enable);
 
+	/* make sure snapshot being resumed re-enumerates everything */
+	if (message.event == PM_EVENT_PRETHAW) {
+		ehci_halt(ehci);
+		ehci_reset(ehci);
+	}
+
 	clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
  bail:
 	spin_unlock_irqrestore (&ehci->lock, flags);
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index b268537e389e..37e122812b67 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -135,6 +135,11 @@ static int ohci_pci_suspend (struct usb_hcd *hcd, pm_message_t message)
 	}
 	ohci_writel(ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
 	(void)ohci_readl(ohci, &ohci->regs->intrdisable);
+
+	/* make sure snapshot being resumed re-enumerates everything */
+	if (message.event == PM_EVENT_PRETHAW)
+		ohci_usb_reset(ohci);
+
 	clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
  bail:
 	spin_unlock_irqrestore (&ohci->lock, flags);
diff --git a/drivers/usb/host/sl811-hcd.c b/drivers/usb/host/sl811-hcd.c
index fa34092bbcde..9de115d9db27 100644
--- a/drivers/usb/host/sl811-hcd.c
+++ b/drivers/usb/host/sl811-hcd.c
@@ -1783,10 +1783,15 @@ sl811h_suspend(struct platform_device *dev, pm_message_t state)
 	struct sl811	*sl811 = hcd_to_sl811(hcd);
 	int		retval = 0;
 
-	if (state.event == PM_EVENT_FREEZE)
+	switch (state.event) {
+	case PM_EVENT_FREEZE:
 		retval = sl811h_bus_suspend(hcd);
-	else if (state.event == PM_EVENT_SUSPEND)
+		break;
+	case PM_EVENT_SUSPEND:
+	case PM_EVENT_PRETHAW:		/* explicitly discard hw state */
 		port_power(sl811, 0);
+		break;
+	}
 	if (retval == 0)
 		dev->dev.power.power_state = state;
 	return retval;
diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c
index 4151f618602d..b7402ceb3e93 100644
--- a/drivers/usb/host/uhci-hcd.c
+++ b/drivers/usb/host/uhci-hcd.c
@@ -734,6 +734,10 @@ static int uhci_suspend(struct usb_hcd *hcd, pm_message_t message)
 
 	/* FIXME: Enable non-PME# remote wakeup? */
 
+	/* make sure snapshot being resumed re-enumerates everything */
+	if (message.event == PM_EVENT_PRETHAW)
+		uhci_hc_died(uhci);
+
 done_okay:
 	clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
 done:
-- 
cgit v1.2.1


From f1cc0a894c963923b766eb2d455747495e6e982d Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 14 Aug 2006 23:11:08 -0700
Subject: PM: issue PM_EVENT_PRETHAW

This patch is the first of this series that should actually change any
behavior ...  by issuing the new event, now tha the rest of the kernel is
prepared to receive it.

This converts the PM core to issue the new PRETHAW message, which the rest of
the kernel is now ready to receive.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/suspend.c | 1 +
 kernel/power/disk.c          | 4 ++--
 kernel/power/swsusp.c        | 9 ++++++++-
 kernel/power/user.c          | 2 +-
 4 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 0bda4a7f2042..e86db83746ac 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -34,6 +34,7 @@ static inline char *suspend_verb(u32 event)
 	switch (event) {
 	case PM_EVENT_SUSPEND:	return "suspend";
 	case PM_EVENT_FREEZE:	return "freeze";
+	case PM_EVENT_PRETHAW:	return "prethaw";
 	default:		return "(unknown suspend event)";
 	}
 }
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index e13e74067845..a3c34fb14321 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -98,7 +98,7 @@ static void unprepare_processes(void)
 }
 
 /**
- *	pm_suspend_disk - The granpappy of power management.
+ *	pm_suspend_disk - The granpappy of hibernation power management.
  *
  *	If we're going through the firmware, then get it over with quickly.
  *
@@ -207,7 +207,7 @@ static int software_resume(void)
 
 	pr_debug("PM: Preparing devices for restore.\n");
 
-	if ((error = device_suspend(PMSG_FREEZE))) {
+	if ((error = device_suspend(PMSG_PRETHAW))) {
 		printk("Some devices failed to suspend\n");
 		swsusp_free();
 		goto Thaw;
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 17f669c83012..62752899b1a1 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -248,6 +248,9 @@ int swsusp_suspend(void)
 	restore_processor_state();
 Restore_highmem:
 	restore_highmem();
+	/* NOTE:  device_power_up() is just a resume() for devices
+	 * that suspended with irqs off ... no overall powerup.
+	 */
 	device_power_up();
 Enable_irqs:
 	local_irq_enable();
@@ -257,8 +260,12 @@ Enable_irqs:
 int swsusp_resume(void)
 {
 	int error;
+
 	local_irq_disable();
-	if (device_power_down(PMSG_FREEZE))
+	/* NOTE:  device_power_down() is just a suspend() with irqs off;
+	 * it has no special "power things down" semantics
+	 */
+	if (device_power_down(PMSG_PRETHAW))
 		printk(KERN_ERR "Some devices failed to power down, very bad\n");
 	/* We'll ignore saved state, but this gets preempt count (etc) right */
 	save_processor_state();
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 3f1539fbe48a..5a8d060d7909 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -191,7 +191,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		}
 		down(&pm_sem);
 		pm_prepare_console();
-		error = device_suspend(PMSG_FREEZE);
+		error = device_suspend(PMSG_PRETHAW);
 		if (!error) {
 			error = swsusp_resume();
 			device_resume();
-- 
cgit v1.2.1


From 4fc0840006bf3df69dcf92397fdcaf85a542f939 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 10 Aug 2006 16:38:28 -0700
Subject: updated Documentation/power/devices.txt

This turned into a rewrite of Documentation/power/devices.txt:

 - Provide more of the "big picture"

 - Fixup some of the horribly ancient/obsolete description of device suspend()
   semantics; lots of text just got deleted.

 - Add a decent description of PM_EVENT_* codes, including the new PRETHAW code
   needed in some swsusp scenarios.

 - Describe the new PM factorization from Linus:
     * class suspend, current suspend, then suspend_late
     * NOT suspend_prepare, it wasn't really usable
     * resume_early, current resume, class resume.

 - Updates power/state docs to be correct, and deprecate its usage except for
   driver testing.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/power/devices.txt | 725 +++++++++++++++++++++++++++++-----------
 1 file changed, 535 insertions(+), 190 deletions(-)

diff --git a/Documentation/power/devices.txt b/Documentation/power/devices.txt
index fba1e05c47c7..d0e79d5820a5 100644
--- a/Documentation/power/devices.txt
+++ b/Documentation/power/devices.txt
@@ -1,208 +1,553 @@
+Most of the code in Linux is device drivers, so most of the Linux power
+management code is also driver-specific.  Most drivers will do very little;
+others, especially for platforms with small batteries (like cell phones),
+will do a lot.
+
+This writeup gives an overview of how drivers interact with system-wide
+power management goals, emphasizing the models and interfaces that are
+shared by everything that hooks up to the driver model core.  Read it as
+background for the domain-specific work you'd do with any specific driver.
+
+
+Two Models for Device Power Management
+======================================
+Drivers will use one or both of these models to put devices into low-power
+states:
+
+    System Sleep model:
+	Drivers can enter low power states as part of entering system-wide
+	low-power states like "suspend-to-ram", or (mostly for systems with
+	disks) "hibernate" (suspend-to-disk).
+
+	This is something that device, bus, and class drivers collaborate on
+	by implementing various role-specific suspend and resume methods to
+	cleanly power down hardware and software subsystems, then reactivate
+	them without loss of data.
+
+	Some drivers can manage hardware wakeup events, which make the system
+	leave that low-power state.  This feature may be disabled using the
+	relevant /sys/devices/.../power/wakeup file; enabling it may cost some
+	power usage, but let the whole system enter low power states more often.
+
+    Runtime Power Management model:
+	Drivers may also enter low power states while the system is running,
+	independently of other power management activity.  Upstream drivers
+	will normally not know (or care) if the device is in some low power
+	state when issuing requests; the driver will auto-resume anything
+	that's needed when it gets a request.
+
+	This doesn't have, or need much infrastructure; it's just something you
+	should do when writing your drivers.  For example, clk_disable() unused
+	clocks as part of minimizing power drain for currently-unused hardware.
+	Of course, sometimes clusters of drivers will collaborate with each
+	other, which could involve task-specific power management.
+
+There's not a lot to be said about those low power states except that they
+are very system-specific, and often device-specific.  Also, that if enough
+drivers put themselves into low power states (at "runtime"), the effect may be
+the same as entering some system-wide low-power state (system sleep) ... and
+that synergies exist, so that several drivers using runtime pm might put the
+system into a state where even deeper power saving options are available.
+
+Most suspended devices will have quiesced all I/O:  no more DMA or irqs, no
+more data read or written, and requests from upstream drivers are no longer
+accepted.  A given bus or platform may have different requirements though.
+
+Examples of hardware wakeup events include an alarm from a real time clock,
+network wake-on-LAN packets, keyboard or mouse activity, and media insertion
+or removal (for PCMCIA, MMC/SD, USB, and so on).
+
+
+Interfaces for Entering System Sleep States
+===========================================
+Most of the programming interfaces a device driver needs to know about
+relate to that first model:  entering a system-wide low power state,
+rather than just minimizing power consumption by one device.
+
+
+Bus Driver Methods
+------------------
+The core methods to suspend and resume devices reside in struct bus_type.
+These are mostly of interest to people writing infrastructure for busses
+like PCI or USB, or because they define the primitives that device drivers
+may need to apply in domain-specific ways to their devices:
 
-Device Power Management
+struct bus_type {
+	...
+	int  (*suspend)(struct device *dev, pm_message_t state);
+	int  (*suspend_late)(struct device *dev, pm_message_t state);
 
+	int  (*resume_early)(struct device *dev);
+	int  (*resume)(struct device *dev);
+};
 
-Device power management encompasses two areas - the ability to save
-state and transition a device to a low-power state when the system is
-entering a low-power state; and the ability to transition a device to
-a low-power state while the system is running (and independently of
-any other power management activity). 
+Bus drivers implement those methods as appropriate for the hardware and
+the drivers using it; PCI works differently from USB, and so on.  Not many
+people write bus drivers; most driver code is a "device driver" that
+builds on top of bus-specific framework code.
+
+For more information on these driver calls, see the description later;
+they are called in phases for every device, respecting the parent-child
+sequencing in the driver model tree.  Note that as this is being written,
+only the suspend() and resume() are widely available; not many bus drivers
+leverage all of those phases, or pass them down to lower driver levels.
+
+
+/sys/devices/.../power/wakeup files
+-----------------------------------
+All devices in the driver model have two flags to control handling of
+wakeup events, which are hardware signals that can force the device and/or
+system out of a low power state.  These are initialized by bus or device
+driver code using device_init_wakeup(dev,can_wakeup).
+
+The "can_wakeup" flag just records whether the device (and its driver) can
+physically support wakeup events.  When that flag is clear, the sysfs
+"wakeup" file is empty, and device_may_wakeup() returns false.
+
+For devices that can issue wakeup events, a separate flag controls whether
+that device should try to use its wakeup mechanism.  The initial value of
+device_may_wakeup() will be true, so that the device's "wakeup" file holds
+the value "enabled".  Userspace can change that to "disabled" so that
+device_may_wakeup() returns false; or change it back to "enabled" (so that
+it returns true again).
+
+
+EXAMPLE:  PCI Device Driver Methods
+-----------------------------------
+PCI framework software calls these methods when the PCI device driver bound
+to a device device has provided them:
+
+struct pci_driver {
+	...
+	int  (*suspend)(struct pci_device *pdev, pm_message_t state);
+	int  (*suspend_late)(struct pci_device *pdev, pm_message_t state);
+
+	int  (*resume_early)(struct pci_device *pdev);
+	int  (*resume)(struct pci_device *pdev);
+};
 
+Drivers will implement those methods, and call PCI-specific procedures
+like pci_set_power_state(), pci_enable_wake(), pci_save_state(), and
+pci_restore_state() to manage PCI-specific mechanisms.  (PCI config space
+could be saved during driver probe, if it weren't for the fact that some
+systems rely on userspace tweaking using setpci.)  Devices are suspended
+before their bridges enter low power states, and likewise bridges resume
+before their devices.
+
+
+Upper Layers of Driver Stacks
+-----------------------------
+Device drivers generally have at least two interfaces, and the methods
+sketched above are the ones which apply to the lower level (nearer PCI, USB,
+or other bus hardware).  The network and block layers are examples of upper
+level interfaces, as is a character device talking to userspace.
+
+Power management requests normally need to flow through those upper levels,
+which often use domain-oriented requests like "blank that screen".  In
+some cases those upper levels will have power management intelligence that
+relates to end-user activity, or other devices that work in cooperation.
+
+When those interfaces are structured using class interfaces, there is a
+standard way to have the upper layer stop issuing requests to a given
+class device (and restart later):
+
+struct class {
+	...
+	int  (*suspend)(struct device *dev, pm_message_t state);
+	int  (*resume)(struct device *dev);
+};
 
-Methods
+Those calls are issued in specific phases of the process by which the
+system enters a low power "suspend" state, or resumes from it.
+
+
+Calling Drivers to Enter System Sleep States
+============================================
+When the system enters a low power state, each device's driver is asked
+to suspend the device by putting it into state compatible with the target
+system state.  That's usually some version of "off", but the details are
+system-specific.  Also, wakeup-enabled devices will usually stay partly
+functional in order to wake the system.
+
+When the system leaves that low power state, the device's driver is asked
+to resume it.  The suspend and resume operations always go together, and
+both are multi-phase operations.
+
+For simple drivers, suspend might quiesce the device using the class code
+and then turn its hardware as "off" as possible with late_suspend.  The
+matching resume calls would then completely reinitialize the hardware
+before reactivating its class I/O queues.
+
+More power-aware drivers drivers will use more than one device low power
+state, either at runtime or during system sleep states, and might trigger
+system wakeup events.
+
+
+Call Sequence Guarantees
+------------------------
+To ensure that bridges and similar links needed to talk to a device are
+available when the device is suspended or resumed, the device tree is
+walked in a bottom-up order to suspend devices.  A top-down order is
+used to resume those devices.
+
+The ordering of the device tree is defined by the order in which devices
+get registered:  a child can never be registered, probed or resumed before
+its parent; and can't be removed or suspended after that parent.
+
+The policy is that the device tree should match hardware bus topology.
+(Or at least the control bus, for devices which use multiple busses.)
+
+
+Suspending Devices
+------------------
+Suspending a given device is done in several phases.  Suspending the
+system always includes every phase, executing calls for every device
+before the next phase begins.  Not all busses or classes support all
+these callbacks; and not all drivers use all the callbacks.
+
+The phases are seen by driver notifications issued in this order:
+
+   1	class.suspend(dev, message) is called after tasks are frozen, for
+	devices associated with a class that has such a method.  This
+	method may sleep.
+
+	Since I/O activity usually comes from such higher layers, this is
+	a good place to quiesce all drivers of a given type (and keep such
+	code out of those drivers).
+
+   2	bus.suspend(dev, message) is called next.  This method may sleep,
+	and is often morphed into a device driver call with bus-specific
+	parameters and/or rules.
+
+	This call should handle parts of device suspend logic that require
+	sleeping.  It probably does work to quiesce the device which hasn't
+	been abstracted into class.suspend() or bus.suspend_late().
+
+   3	bus.suspend_late(dev, message) is called with IRQs disabled, and
+	with only one CPU active.  Until the bus.resume_early() phase
+	completes (see later), IRQs are not enabled again.  This method
+	won't be exposed by all busses; for message based busses like USB,
+	I2C, or SPI, device interactions normally require IRQs.  This bus
+	call may be morphed into a driver call with bus-specific parameters.
+
+	This call might save low level hardware state that might otherwise
+	be lost in the upcoming low power state, and actually put the
+	device into a low power state ... so that in some cases the device
+	may stay partly usable until this late.  This "late" call may also
+	help when coping with hardware that behaves badly.
+
+The pm_message_t parameter is currently used to refine those semantics
+(described later).
+
+At the end of those phases, drivers should normally have stopped all I/O
+transactions (DMA, IRQs), saved enough state that they can re-initialize
+or restore previous state (as needed by the hardware), and placed the
+device into a low-power state.  On many platforms they will also use
+clk_disable() to gate off one or more clock sources; sometimes they will
+also switch off power supplies, or reduce voltages.  Drivers which have
+runtime PM support may already have performed some or all of the steps
+needed to prepare for the upcoming system sleep state.
+
+When any driver sees that its device_can_wakeup(dev), it should make sure
+to use the relevant hardware signals to trigger a system wakeup event.
+For example, enable_irq_wake() might identify GPIO signals hooked up to
+a switch or other external hardware, and pci_enable_wake() does something
+similar for PCI's PME# signal.
+
+If a driver (or bus, or class) fails it suspend method, the system won't
+enter the desired low power state; it will resume all the devices it's
+suspended so far.
+
+Note that drivers may need to perform different actions based on the target
+system lowpower/sleep state.  At this writing, there are only platform
+specific APIs through which drivers could determine those target states.
+
+
+Device Low Power (suspend) States
+---------------------------------
+Device low-power states aren't very standard.  One device might only handle
+"on" and "off, while another might support a dozen different versions of
+"on" (how many engines are active?), plus a state that gets back to "on"
+faster than from a full "off".
+
+Some busses define rules about what different suspend states mean.  PCI
+gives one example:  after the suspend sequence completes, a non-legacy
+PCI device may not perform DMA or issue IRQs, and any wakeup events it
+issues would be issued through the PME# bus signal.  Plus, there are
+several PCI-standard device states, some of which are optional.
+
+In contrast, integrated system-on-chip processors often use irqs as the
+wakeup event sources (so drivers would call enable_irq_wake) and might
+be able to treat DMA completion as a wakeup event (sometimes DMA can stay
+active too, it'd only be the CPU and some peripherals that sleep).
+
+Some details here may be platform-specific.  Systems may have devices that
+can be fully active in certain sleep states, such as an LCD display that's
+refreshed using DMA while most of the system is sleeping lightly ... and
+its frame buffer might even be updated by a DSP or other non-Linux CPU while
+the Linux control processor stays idle.
+
+Moreover, the specific actions taken may depend on the target system state.
+One target system state might allow a given device to be very operational;
+another might require a hard shut down with re-initialization on resume.
+And two different target systems might use the same device in different
+ways; the aforementioned LCD might be active in one product's "standby",
+but a different product using the same SOC might work differently.
+
+
+Meaning of pm_message_t.event
+-----------------------------
+Parameters to suspend calls include the device affected and a message of
+type pm_message_t, which has one field:  the event.  If driver does not
+recognize the event code, suspend calls may abort the request and return
+a negative errno.  However, most drivers will be fine if they implement
+PM_EVENT_SUSPEND semantics for all messages.
+
+The event codes are used to refine the goal of suspending the device, and
+mostly matter when creating or resuming system memory image snapshots, as
+used with suspend-to-disk:
+
+    PM_EVENT_SUSPEND -- quiesce the driver and put hardware into a low-power
+	state.  When used with system sleep states like "suspend-to-RAM" or
+	"standby", the upcoming resume() call will often be able to rely on
+	state kept in hardware, or issue system wakeup events.  When used
+	instead with suspend-to-disk, few devices support this capability;
+	most are completely powered off.
+
+    PM_EVENT_FREEZE -- quiesce the driver, but don't necessarily change into
+	any low power mode.  A system snapshot is about to be taken, often
+	followed by a call to the driver's resume() method.  Neither wakeup
+	events nor DMA are allowed.
+
+    PM_EVENT_PRETHAW -- quiesce the driver, knowing that the upcoming resume()
+	will restore a suspend-to-disk snapshot from a different kernel image.
+	Drivers that are smart enough to look at their hardware state during
+	resume() processing need that state to be correct ... a PRETHAW could
+	be used to invalidate that state (by resetting the device), like a
+	shutdown() invocation would before a kexec() or system halt.  Other
+	drivers might handle this the same way as PM_EVENT_FREEZE.  Neither
+	wakeup events nor DMA are allowed.
+
+To enter "standby" (ACPI S1) or "Suspend to RAM" (STR, ACPI S3) states, or
+the similarly named APM states, only PM_EVENT_SUSPEND is used; for "Suspend
+to Disk" (STD, hibernate, ACPI S4), all of those event codes are used.
+
+There's also PM_EVENT_ON, a value which never appears as a suspend event
+but is sometimes used to record the "not suspended" device state.
+
+
+Resuming Devices
+----------------
+Resuming is done in multiple phases, much like suspending, with all
+devices processing each phase's calls before the next phase begins.
+
+The phases are seen by driver notifications issued in this order:
+
+   1	bus.resume_early(dev) is called with IRQs disabled, and with
+   	only one CPU active.  As with bus.suspend_late(), this method
+	won't be supported on busses that require IRQs in order to
+	interact with devices.
+
+	This reverses the effects of bus.suspend_late().
+
+   2	bus.resume(dev) is called next.  This may be morphed into a device
+   	driver call with bus-specific parameters; implementations may sleep.
+
+	This reverses the effects of bus.suspend().
+
+   3	class.resume(dev) is called for devices associated with a class
+	that has such a method.  Implementations may sleep.
+
+	This reverses the effects of class.suspend(), and would usually
+	reactivate the device's I/O queue.
+
+At the end of those phases, drivers should normally be as functional as
+they were before suspending:  I/O can be performed using DMA and IRQs, and
+the relevant clocks are gated on.  The device need not be "fully on"; it
+might be in a runtime lowpower/suspend state that acts as if it were.
+
+However, the details here may again be platform-specific.  For example,
+some systems support multiple "run" states, and the mode in effect at
+the end of resume() might not be the one which preceded suspension.
+That means availability of certain clocks or power supplies changed,
+which could easily affect how a driver works.
+
+
+Drivers need to be able to handle hardware which has been reset since the
+suspend methods were called, for example by complete reinitialization.
+This may be the hardest part, and the one most protected by NDA'd documents
+and chip errata.  It's simplest if the hardware state hasn't changed since
+the suspend() was called, but that can't always be guaranteed.
+
+Drivers must also be prepared to notice that the device has been removed
+while the system was powered off, whenever that's physically possible.
+PCMCIA, MMC, USB, Firewire, SCSI, and even IDE are common examples of busses
+where common Linux platforms will see such removal.  Details of how drivers
+will notice and handle such removals are currently bus-specific, and often
+involve a separate thread.
 
-The methods to suspend and resume devices reside in struct bus_type: 
 
-struct bus_type {
-       ...
-       int             (*suspend)(struct device * dev, pm_message_t state);
-       int             (*resume)(struct device * dev);
-};
+Note that the bus-specific runtime PM wakeup mechanism can exist, and might
+be defined to share some of the same driver code as for system wakeup.  For
+example, a bus-specific device driver's resume() method might be used there,
+so it wouldn't only be called from bus.resume() during system-wide wakeup.
+See bus-specific information about how runtime wakeup events are handled.
 
-Each bus driver is responsible implementing these methods, translating
-the call into a bus-specific request and forwarding the call to the
-bus-specific drivers. For example, PCI drivers implement suspend() and
-resume() methods in struct pci_driver. The PCI core is simply
-responsible for translating the pointers to PCI-specific ones and
-calling the low-level driver.
-
-This is done to a) ease transition to the new power management methods
-and leverage the existing PM code in various bus drivers; b) allow
-buses to implement generic and default PM routines for devices, and c)
-make the flow of execution obvious to the reader. 
-
-
-System Power Management
-
-When the system enters a low-power state, the device tree is walked in
-a depth-first fashion to transition each device into a low-power
-state. The ordering of the device tree is guaranteed by the order in
-which devices get registered - children are never registered before
-their ancestors, and devices are placed at the back of the list when
-registered. By walking the list in reverse order, we are guaranteed to
-suspend devices in the proper order. 
-
-Devices are suspended once with interrupts enabled. Drivers are
-expected to stop I/O transactions, save device state, and place the
-device into a low-power state. Drivers may sleep, allocate memory,
-etc. at will. 
-
-Some devices are broken and will inevitably have problems powering
-down or disabling themselves with interrupts enabled. For these
-special cases, they may return -EAGAIN. This will put the device on a
-list to be taken care of later. When interrupts are disabled, before
-we enter the low-power state, their drivers are called again to put
-their device to sleep. 
-
-On resume, the devices that returned -EAGAIN will be called to power
-themselves back on with interrupts disabled. Once interrupts have been
-re-enabled, the rest of the drivers will be called to resume their
-devices. On resume, a driver is responsible for powering back on each
-device, restoring state, and re-enabling I/O transactions for that
-device. 
 
+System Devices
+--------------
 System devices follow a slightly different API, which can be found in
 
 	include/linux/sysdev.h
 	drivers/base/sys.c
 
-System devices will only be suspended with interrupts disabled, and
-after all other devices have been suspended. On resume, they will be
-resumed before any other devices, and also with interrupts disabled.
+System devices will only be suspended with interrupts disabled, and after
+all other devices have been suspended.  On resume, they will be resumed
+before any other devices, and also with interrupts disabled.
 
+That is, IRQs are disabled, the suspend_late() phase begins, then the
+sysdev_driver.suspend() phase, and the system enters a sleep state.  Then
+the sysdev_driver.resume() phase begins, followed by the resume_early()
+phase, after which IRQs are enabled.
 
-Runtime Power Management
-
-Many devices are able to dynamically power down while the system is
-still running. This feature is useful for devices that are not being
-used, and can offer significant power savings on a running system. 
-
-In each device's directory, there is a 'power' directory, which
-contains at least a 'state' file. Reading from this file displays what
-power state the device is currently in. Writing to this file initiates
-a transition to the specified power state, which must be a decimal in
-the range 1-3, inclusive; or 0 for 'On'.
+Code to actually enter and exit the system-wide low power state sometimes
+involves hardware details that are only known to the boot firmware, and
+may leave a CPU running software (from SRAM or flash memory) that monitors
+the system and manages its wakeup sequence.
 
-The PM core will call the ->suspend() method in the bus_type object
-that the device belongs to if the specified state is not 0, or
-->resume() if it is. 
 
-Nothing will happen if the specified state is the same state the
-device is currently in. 
-
-If the device is already in a low-power state, and the specified state
-is another, but different, low-power state, the ->resume() method will
-first be called to power the device back on, then ->suspend() will be
-called again with the new state. 
-
-The driver is responsible for saving the working state of the device
-and putting it into the low-power state specified. If this was
-successful, it returns 0, and the device's power_state field is
-updated. 
-
-The driver must take care to know whether or not it is able to
-properly resume the device, including all step of reinitialization
-necessary. (This is the hardest part, and the one most protected by
-NDA'd documents). 
-
-The driver must also take care not to suspend a device that is
-currently in use. It is their responsibility to provide their own
-exclusion mechanisms.
-
-The runtime power transition happens with interrupts enabled. If a
-device cannot support being powered down with interrupts, it may
-return -EAGAIN (as it would during a system power management
-transition),  but it will _not_ be called again, and the transaction
-will fail.
-
-There is currently no way to know what states a device or driver
-supports a priori. This will change in the future. 
-
-pm_message_t meaning
-
-pm_message_t has two fields. event ("major"), and flags.  If driver
-does not know event code, it aborts the request, returning error. Some
-drivers may need to deal with special cases based on the actual type
-of suspend operation being done at the system level. This is why
-there are flags.
-
-Event codes are:
-
-ON -- no need to do anything except special cases like broken
-HW.
-
-# NOTIFICATION -- pretty much same as ON?
-
-FREEZE -- stop DMA and interrupts, and be prepared to reinit HW from
-scratch. That probably means stop accepting upstream requests, the
-actual policy of what to do with them being specific to a given
-driver. It's acceptable for a network driver to just drop packets
-while a block driver is expected to block the queue so no request is
-lost. (Use IDE as an example on how to do that). FREEZE requires no
-power state change, and it's expected for drivers to be able to
-quickly transition back to operating state.
-
-SUSPEND -- like FREEZE, but also put hardware into low-power state. If
-there's need to distinguish several levels of sleep, additional flag
-is probably best way to do that.
-
-Transitions are only from a resumed state to a suspended state, never
-between 2 suspended states. (ON -> FREEZE or ON -> SUSPEND can happen,
-FREEZE -> SUSPEND or SUSPEND -> FREEZE can not).
-
-All events are:
-
-[NOTE NOTE NOTE: If you are driver author, you should not care; you
-should only look at event, and ignore flags.]
-
-#Prepare for suspend -- userland is still running but we are going to
-#enter suspend state. This gives drivers chance to load firmware from
-#disk and store it in memory, or do other activities taht require
-#operating userland, ability to kmalloc GFP_KERNEL, etc... All of these
-#are forbiden once the suspend dance is started.. event = ON, flags =
-#PREPARE_TO_SUSPEND
-
-Apm standby -- prepare for APM event. Quiesce devices to make life
-easier for APM BIOS. event = FREEZE, flags = APM_STANDBY
-
-Apm suspend -- same as APM_STANDBY, but it we should probably avoid
-spinning down disks. event = FREEZE, flags = APM_SUSPEND
-
-System halt, reboot -- quiesce devices to make life easier for BIOS. event
-= FREEZE, flags = SYSTEM_HALT or SYSTEM_REBOOT
-
-System shutdown -- at least disks need to be spun down, or data may be
-lost. Quiesce devices, just to make life easier for BIOS. event =
-FREEZE, flags = SYSTEM_SHUTDOWN
-
-Kexec    -- turn off DMAs and put hardware into some state where new
-kernel can take over. event = FREEZE, flags = KEXEC
-
-Powerdown at end of swsusp -- very similar to SYSTEM_SHUTDOWN, except wake
-may need to be enabled on some devices. This actually has at least 3
-subtypes, system can reboot, enter S4 and enter S5 at the end of
-swsusp. event = FREEZE, flags = SWSUSP and one of SYSTEM_REBOOT,
-SYSTEM_SHUTDOWN, SYSTEM_S4
-
-Suspend to ram  -- put devices into low power state. event = SUSPEND,
-flags = SUSPEND_TO_RAM
-
-Freeze for swsusp snapshot -- stop DMA and interrupts. No need to put
-devices into low power mode, but you must be able to reinitialize
-device from scratch in resume method. This has two flavors, its done
-once on suspending kernel, once on resuming kernel. event = FREEZE,
-flags = DURING_SUSPEND or DURING_RESUME
-
-Device detach requested from /sys -- deinitialize device; proably same as
-SYSTEM_SHUTDOWN, I do not understand this one too much. probably event
-= FREEZE, flags = DEV_DETACH.
-
-#These are not really events sent:
-#
-#System fully on -- device is working normally; this is probably never
-#passed to suspend() method... event = ON, flags = 0
-#
-#Ready after resume -- userland is now running, again. Time to free any
-#memory you ate during prepare to suspend... event = ON, flags =
-#READY_AFTER_RESUME
-#
+Runtime Power Management
+========================
+Many devices are able to dynamically power down while the system is still
+running. This feature is useful for devices that are not being used, and
+can offer significant power savings on a running system.  These devices
+often support a range of runtime power states, which might use names such
+as "off", "sleep", "idle", "active", and so on.  Those states will in some
+cases (like PCI) be partially constrained by a bus the device uses, and will
+usually include hardware states that are also used in system sleep states.
+
+However, note that if a driver puts a device into a runtime low power state
+and the system then goes into a system-wide sleep state, it normally ought
+to resume into that runtime low power state rather than "full on".  Such
+distinctions would be part of the driver-internal state machine for that
+hardware; the whole point of runtime power management is to be sure that
+drivers are decoupled in that way from the state machine governing phases
+of the system-wide power/sleep state transitions.
+
+
+Power Saving Techniques
+-----------------------
+Normally runtime power management is handled by the drivers without specific
+userspace or kernel intervention, by device-aware use of techniques like:
+
+    Using information provided by other system layers
+	- stay deeply "off" except between open() and close()
+	- if transceiver/PHY indicates "nobody connected", stay "off"
+	- application protocols may include power commands or hints
+
+    Using fewer CPU cycles
+	- using DMA instead of PIO
+	- removing timers, or making them lower frequency
+	- shortening "hot" code paths
+	- eliminating cache misses
+	- (sometimes) offloading work to device firmware
+
+    Reducing other resource costs
+	- gating off unused clocks in software (or hardware)
+	- switching off unused power supplies
+	- eliminating (or delaying/merging) IRQs
+	- tuning DMA to use word and/or burst modes
+
+    Using device-specific low power states
+	- using lower voltages
+	- avoiding needless DMA transfers
+
+Read your hardware documentation carefully to see the opportunities that
+may be available.  If you can, measure the actual power usage and check
+it against the budget established for your project.
+
+
+Examples:  USB hosts, system timer, system CPU
+----------------------------------------------
+USB host controllers make interesting, if complex, examples.  In many cases
+these have no work to do:  no USB devices are connected, or all of them are
+in the USB "suspend" state.  Linux host controller drivers can then disable
+periodic DMA transfers that would otherwise be a constant power drain on the
+memory subsystem, and enter a suspend state.  In power-aware controllers,
+entering that suspend state may disable the clock used with USB signaling,
+saving a certain amount of power.
+
+The controller will be woken from that state (with an IRQ) by changes to the
+signal state on the data lines of a given port, for example by an existing
+peripheral requesting "remote wakeup" or by plugging a new peripheral.  The
+same wakeup mechanism usually works from "standby" sleep states, and on some
+systems also from "suspend to RAM" (or even "suspend to disk") states.
+(Except that ACPI may be involved instead of normal IRQs, on some hardware.)
+
+System devices like timers and CPUs may have special roles in the platform
+power management scheme.  For example, system timers using a "dynamic tick"
+approach don't just save CPU cycles (by eliminating needless timer IRQs),
+but they may also open the door to using lower power CPU "idle" states that
+cost more than a jiffie to enter and exit.  On x86 systems these are states
+like "C3"; note that periodic DMA transfers from a USB host controller will
+also prevent entry to a C3 state, much like a periodic timer IRQ.
+
+That kind of runtime mechanism interaction is common.  "System On Chip" (SOC)
+processors often have low power idle modes that can't be entered unless
+certain medium-speed clocks (often 12 or 48 MHz) are gated off.  When the
+drivers gate those clocks effectively, then the system idle task may be able
+to use the lower power idle modes and thereby increase battery life.
+
+If the CPU can have a "cpufreq" driver, there also may be opportunities
+to shift to lower voltage settings and reduce the power cost of executing
+a given number of instructions.  (Without voltage adjustment, it's rare
+for cpufreq to save much power; the cost-per-instruction must go down.)
+
+
+/sys/devices/.../power/state files
+==================================
+For now you can also test some of this functionality using sysfs.
+
+	DEPRECATED:  USE "power/state" ONLY FOR DRIVER TESTING, AND
+	AVOID USING dev->power.power_state IN DRIVERS.
+
+	THESE WILL BE REMOVED.  IF THE "power/state" FILE GETS REPLACED,
+	IT WILL BECOME SOMETHING COUPLED TO THE BUS OR DRIVER.
+
+In each device's directory, there is a 'power' directory, which contains
+at least a 'state' file.  The value of this field is effectively boolean,
+PM_EVENT_ON or PM_EVENT_SUSPEND.
+
+   *	Reading from this file displays a value corresponding to
+	the power.power_state.event field.  All nonzero values are
+	displayed as "2", corresponding to a low power state; zero
+	is displayed as "0", corresponding to normal operation.
+
+   *	Writing to this file initiates a transition using the
+   	specified event code number; only '0', '2', and '3' are
+	accepted (without a newline); '2' and '3' are both
+	mapped to PM_EVENT_SUSPEND.
+
+On writes, the PM core relies on that recorded event code and the device/bus
+capabilities to determine whether it uses a partial suspend() or resume()
+sequence to change things so that the recorded event corresponds to the
+numeric parameter.
+
+   -	If the bus requires the irqs-disabled suspend_late()/resume_early()
+	phases, writes fail because those operations are not supported here.
+
+   -	If the recorded value is the expected value, nothing is done.
+
+   -	If the recorded value is nonzero, the device is partially resumed,
+	using the bus.resume() and/or class.resume() methods.
+
+   -	If the target value is nonzero, the device is partially suspended,
+	using the class.suspend() and/or bus.suspend() methods and the
+	PM_EVENT_SUSPEND message.
+
+Drivers have no way to tell whether their suspend() and resume() calls
+have come through the sysfs power/state file or as part of entering a
+system sleep state, except that when accessed through sysfs the normal
+parent/child sequencing rules are ignored.  Drivers (such as bus, bridge,
+or hub drivers) which expose child devices may need to enforce those rules
+on their own.
-- 
cgit v1.2.1


From 047bda36150d11422b2c7bacca1df324c909c0b3 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 14:12:48 -0700
Subject: PM: update docs for writing .../power/state

Updates to match current code:

 - Make writes to the /sys/devices/.../power/state files fail cleanly
   if the device requires the irqs-off call variants.

 - Fix comments describing the /sys/devices/.../power/state file writes
   to match the code; the last several releases have invalidated the
   previous text.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/sysfs.c | 29 +++++++++++++++++++----------
 1 file changed, 19 insertions(+), 10 deletions(-)

diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index 40d7242a07c1..e55b3c2779e9 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -11,18 +11,23 @@
  *	state - Control current power state of device
  *
  *	show() returns the current power state of the device. '0' indicates
- *	the device is on. Other values (1-3) indicate the device is in a low
+ *	the device is on. Other values (2) indicate the device is in some low
  *	power state.
  *
- *	store() sets the current power state, which is an integer value
- *	between 0-3. If the device is on ('0'), and the value written is
- *	greater than 0, then the device is placed directly into the low-power
- *	state (via its driver's ->suspend() method).
- *	If the device is currently in a low-power state, and the value is 0,
- *	the device is powered back on (via the ->resume() method).
- *	If the device is in a low-power state, and a different low-power state
- *	is requested, the device is first resumed, then suspended into the new
- *	low-power state.
+ *	store() sets the current power state, which is an integer valued
+ *	0, 2, or 3.  Devices with bus.suspend_late(), or bus.resume_early()
+ *	methods fail this operation; those methods couldn't be called.
+ *	Otherwise,
+ *
+ *	- If the recorded dev->power.power_state.event matches the
+ *	  target value, nothing is done.
+ *	- If the recorded event code is nonzero, the device is reactivated
+ *	  by calling bus.resume() and/or class.resume().
+ *	- If the target value is nonzero, the device is suspended by
+ *	  calling class.suspend() and/or bus.suspend() with event code
+ *	  PM_EVENT_SUSPEND.
+ *
+ *	This mechanism is DEPRECATED and should only be used for testing.
  */
 
 static ssize_t state_show(struct device * dev, struct device_attribute *attr, char * buf)
@@ -38,6 +43,10 @@ static ssize_t state_store(struct device * dev, struct device_attribute *attr, c
 	pm_message_t state;
 	int error = -EINVAL;
 
+	/* disallow incomplete suspend sequences */
+	if (dev->bus && (dev->bus->suspend_late || dev->bus->resume_early))
+		return error;
+
 	state.event = PM_EVENT_SUSPEND;
 	/* Older apps expected to write "3" here - confused with PCI D3 */
 	if ((n == 1) && !strcmp(buf, "3"))
-- 
cgit v1.2.1


From 2bca293e56b6a8cd16bb6e70a09b2adac9c723b5 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 13:54:36 -0700
Subject: PM: add kconfig option for deprecated .../power/state files

Add a new PM_SYSFS_DEPRECATED config option to control whether or
not the /sys/devices/.../power/state files are provided.  This will
make it easier to get rid of that mechanism when the time comes,
and to verify that userspace tools work right without it.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/sysfs.c |  6 ++++++
 kernel/power/Kconfig       | 11 +++++++++++
 2 files changed, 17 insertions(+)

diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c
index e55b3c2779e9..2d47517dbe32 100644
--- a/drivers/base/power/sysfs.c
+++ b/drivers/base/power/sysfs.c
@@ -7,6 +7,8 @@
 #include "power.h"
 
 
+#ifdef	CONFIG_PM_SYSFS_DEPRECATED
+
 /**
  *	state - Control current power state of device
  *
@@ -66,6 +68,8 @@ static ssize_t state_store(struct device * dev, struct device_attribute *attr, c
 static DEVICE_ATTR(state, 0644, state_show, state_store);
 
 
+#endif	/* CONFIG_PM_SYSFS_DEPRECATED */
+
 /*
  *	wakeup - Report/change current wakeup option for device
  *
@@ -139,7 +143,9 @@ static DEVICE_ATTR(wakeup, 0644, wake_show, wake_store);
 
 
 static struct attribute * power_attrs[] = {
+#ifdef	CONFIG_PM_SYSFS_DEPRECATED
 	&dev_attr_state.attr,
+#endif
 	&dev_attr_wakeup.attr,
 	NULL,
 };
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 619ecabf7c58..1ed972070d19 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -53,6 +53,17 @@ config PM_TRACE
 	CAUTION: this option will cause your machine's real-time clock to be
 	set to an invalid time after a resume.
 
+config PM_SYSFS_DEPRECATED
+	bool "Driver model /sys/devices/.../power/state files (DEPRECATED)"
+	depends on PM && SYSFS
+	default n
+	help
+	  The driver model started out with a sysfs file intended to provide
+	  a userspace hook for device power management.  This feature has never
+	  worked very well, except for limited testing purposes, and so it will
+	  be removed.   It's not clear that a generic mechanism could really
+	  handle the wide variability of device power states; any replacements
+	  are likely to be bus or driver specific.
 
 config SOFTWARE_SUSPEND
 	bool "Software Suspend"
-- 
cgit v1.2.1


From 1ebfd79eb46a89eacf3cbdf4a463d5d572d3557b Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Wed, 30 Aug 2006 13:50:27 -0700
Subject: PM: schedule /sys/devices/.../power/state for removal

This lists the /sys/devices/.../power/state file, and its internal support,
as due for removal next year.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/feature-removal-schedule.txt | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index a89a1b7f7cf7..611acc32fdf5 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -6,6 +6,21 @@ be removed from this file.
 
 ---------------------------
 
+What:	/sys/devices/.../power/state
+	dev->power.power_state
+	dpm_runtime_{suspend,resume)()
+When:	July 2007
+Why:	Broken design for runtime control over driver power states, confusing
+	driver-internal runtime power management with:  mechanisms to support
+	system-wide sleep state transitions; event codes that distinguish
+	different phases of swsusp "sleep" transitions; and userspace policy
+	inputs.  This framework was never widely used, and most attempts to
+	use it were broken.  Drivers should instead be exposing domain-specific
+	interfaces either to kernel or to userspace.
+Who:	Pavel Machek <pavel@suse.cz>
+
+---------------------------
+
 What:	RAW driver (CONFIG_RAW_DRIVER)
 When:	December 2005
 Why:	declared obsolete since kernel 2.6.3
-- 
cgit v1.2.1


From 1d3a82af45428c5e8deaa119cdeb79611ae46371 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 14:09:47 -0700
Subject: PM: no suspend_prepare() phase

Remove the new suspend_prepare() phase.  It doesn't seem very usable,
has never been tested, doesn't address fault cleanup, and would need
a sibling resume_complete(); plus there are no real use cases.  It
could be restored later if those issues get resolved.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/suspend.c | 27 ---------------------------
 drivers/pci/pci-driver.c     | 14 --------------
 include/linux/device.h       |  1 -
 include/linux/pci.h          |  1 -
 kernel/power/main.c          |  4 ----
 5 files changed, 47 deletions(-)

diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index e86db83746ac..6453c25103d2 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -117,33 +117,6 @@ static int suspend_device_late(struct device *dev, pm_message_t state)
 	return error;
 }
 
-/**
- *	device_prepare_suspend - save state and prepare to suspend
- *
- *	NOTE! Devices cannot detach at this point - not only do we
- *	hold the device list semaphores over the whole prepare, but
- *	the whole point is to do non-invasive preparatory work, not
- *	the actual suspend.
- */
-int device_prepare_suspend(pm_message_t state)
-{
-	int error = 0;
-	struct device * dev;
-
-	down(&dpm_sem);
-	down(&dpm_list_sem);
-	list_for_each_entry_reverse(dev, &dpm_active, power.entry) {
-		if (!dev->bus || !dev->bus->suspend_prepare)
-			continue;
-		error = dev->bus->suspend_prepare(dev, state);
-		if (error)
-			break;
-	}
-	up(&dpm_list_sem);
-	up(&dpm_sem);
-	return error;
-}
-
 /**
  *	device_suspend - Save state and stop all devices in system.
  *	@state:		Power state to put each device in.
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 9e7d6ceb3805..8948ac9ab681 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -264,19 +264,6 @@ static int pci_device_remove(struct device * dev)
 	return 0;
 }
 
-static int pci_device_suspend_prepare(struct device * dev, pm_message_t state)
-{
-	struct pci_dev * pci_dev = to_pci_dev(dev);
-	struct pci_driver * drv = pci_dev->driver;
-	int i = 0;
-
-	if (drv && drv->suspend_prepare) {
-		i = drv->suspend_prepare(pci_dev, state);
-		suspend_report_result(drv->suspend_prepare, i);
-	}
-	return i;
-}
-
 static int pci_device_suspend(struct device * dev, pm_message_t state)
 {
 	struct pci_dev * pci_dev = to_pci_dev(dev);
@@ -544,7 +531,6 @@ struct bus_type pci_bus_type = {
 	.uevent		= pci_uevent,
 	.probe		= pci_device_probe,
 	.remove		= pci_device_remove,
-	.suspend_prepare= pci_device_suspend_prepare,
 	.suspend	= pci_device_suspend,
 	.suspend_late	= pci_device_suspend_late,
 	.resume_early	= pci_device_resume_early,
diff --git a/include/linux/device.h b/include/linux/device.h
index b40be6fca6fa..b3646462d6dc 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -52,7 +52,6 @@ struct bus_type {
 	int		(*remove)(struct device * dev);
 	void		(*shutdown)(struct device * dev);
 
-	int (*suspend_prepare)(struct device * dev, pm_message_t state);
 	int (*suspend)(struct device * dev, pm_message_t state);
 	int (*suspend_late)(struct device * dev, pm_message_t state);
 	int (*resume_early)(struct device * dev);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 4b2e629467c7..9514bbfe96e2 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -345,7 +345,6 @@ struct pci_driver {
 	const struct pci_device_id *id_table;	/* must be non-NULL for probe to be called */
 	int  (*probe)  (struct pci_dev *dev, const struct pci_device_id *id);	/* New device inserted */
 	void (*remove) (struct pci_dev *dev);	/* Device removed (NULL if not a hot-plug capable driver) */
-	int  (*suspend_prepare) (struct pci_dev *dev, pm_message_t state);
 	int  (*suspend) (struct pci_dev *dev, pm_message_t state);	/* Device suspended */
 	int  (*suspend_late) (struct pci_dev *dev, pm_message_t state);
 	int  (*resume_early) (struct pci_dev *dev);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 0c3ed6ac938e..6d295c776794 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -57,10 +57,6 @@ static int suspend_prepare(suspend_state_t state)
 	if (!pm_ops || !pm_ops->enter)
 		return -EPERM;
 
-	error = device_prepare_suspend(PMSG_SUSPEND);
-	if (error)
-		return error;
-
 	pm_prepare_console();
 
 	disable_nonboot_cpus();
-- 
cgit v1.2.1


From 84ed64ee8f7dfd89f59857124dbeb8a350c6e03d Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 30 Aug 2006 23:38:06 +0200
Subject: PM: add /sys/power documentation to Documentation/ABI

The file sysfs-power that documents the interface in the /sys/power/ directory
is added to Documentation/ABI/testing.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/ABI/testing/sysfs-power | 88 +++++++++++++++++++++++++++++++++++
 1 file changed, 88 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-power

diff --git a/Documentation/ABI/testing/sysfs-power b/Documentation/ABI/testing/sysfs-power
new file mode 100644
index 000000000000..d882f8093871
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-power
@@ -0,0 +1,88 @@
+What:		/sys/power/
+Date:		August 2006
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/power directory will contain files that will
+		provide a unified interface to the power management
+		subsystem.
+
+What:		/sys/power/state
+Date:		August 2006
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/power/state file controls the system power state.
+		Reading from this file returns what states are supported,
+		which is hard-coded to 'standby' (Power-On Suspend), 'mem'
+		(Suspend-to-RAM), and 'disk' (Suspend-to-Disk).
+
+		Writing to this file one of these strings causes the system to
+		transition into that state. Please see the file
+		Documentation/power/states.txt for a description of each of
+		these states.
+
+What:		/sys/power/disk
+Date:		August 2006
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/power/disk file controls the operating mode of the
+		suspend-to-disk mechanism.  Reading from this file returns
+		the name of the method by which the system will be put to
+		sleep on the next suspend.  There are four methods supported:
+		'firmware' - means that the memory image will be saved to disk
+		by some firmware, in which case we also assume that the
+		firmware will handle the system suspend.
+		'platform' - the memory image will be saved by the kernel and
+		the system will be put to sleep by the platform driver (e.g.
+		ACPI or other PM registers).
+		'shutdown' - the memory image will be saved by the kernel and
+		the system will be powered off.
+		'reboot' - the memory image will be saved by the kernel and
+		the system will be rebooted.
+
+		The suspend-to-disk method may be chosen by writing to this
+		file one of the accepted strings:
+
+		'firmware'
+		'platform'
+		'shutdown'
+		'reboot'
+
+		It will only change to 'firmware' or 'platform' if the system
+		supports that.
+
+What:		/sys/power/image_size
+Date:		August 2006
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/power/image_size file controls the size of the image
+		created by the suspend-to-disk mechanism.  It can be written a
+		string representing a non-negative integer that will be used
+		as an upper limit of the image size, in bytes.  The kernel's
+		suspend-to-disk code will do its best to ensure the image size
+		will not exceed this number.  However, if it turns out to be
+		impossible, the kernel will try to suspend anyway using the
+		smallest image possible.  In particular, if "0" is written to
+		this file, the suspend image will be as small as possible.
+
+		Reading from this file will display the current image size
+		limit, which is set to 500 MB by default.
+
+What:		/sys/power/pm_trace
+Date:		August 2006
+Contact:	Rafael J. Wysocki <rjw@sisk.pl>
+Description:
+		The /sys/power/pm_trace file controls the code which saves the
+		last PM event point in the RTC across reboots, so that you can
+		debug a machine that just hangs during suspend (or more
+		commonly, during resume).  Namely, the RTC is only used to save
+		the last PM event point if this file contains '1'.  Initially
+		it contains '0' which may be changed to '1' by writing a
+		string representing a nonzero integer into it.
+
+		To use this debugging feature you should attempt to suspend
+		the machine, then reboot it and run
+
+		dmesg -s 1000000 | grep 'hash matches'
+
+		CAUTION: Using it will cause your machine's real-time (CMOS)
+		clock to be set to a random invalid time after a resume.
-- 
cgit v1.2.1


From bb84c89f94851161f387285d0a449b4a3f29f4df Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Thu, 31 Aug 2006 22:02:11 -0700
Subject: PM: device_suspend/resume may sleep

This adds warning when someone tries them from atomic context.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/resume.c  | 1 +
 drivers/base/power/suspend.c | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c
index 48e3d49d7b65..7cb62d62c958 100644
--- a/drivers/base/power/resume.c
+++ b/drivers/base/power/resume.c
@@ -96,6 +96,7 @@ void dpm_resume(void)
 
 void device_resume(void)
 {
+	might_sleep();
 	down(&dpm_sem);
 	dpm_resume();
 	up(&dpm_sem);
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 6453c25103d2..ece136bf97e3 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -140,6 +140,7 @@ int device_suspend(pm_message_t state)
 {
 	int error = 0;
 
+	might_sleep();
 	down(&dpm_sem);
 	down(&dpm_list_sem);
 	while (!list_empty(&dpm_active) && error == 0) {
-- 
cgit v1.2.1


From 386415d88b1ae50304f9c61aa3e0db082fa90428 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 3 Sep 2006 13:16:45 -0700
Subject: PM: platform_bus and late_suspend/early_resume

Teach platform_bus about the new suspend_late/resume_early PM calls,
issued with IRQs off.  Do we really need sysdev and friends any more,
or can janitors start switching its users over to platform_device so
we can do a minor code-ectomy?

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/platform.c         | 30 ++++++++++++++++++++++++++++--
 include/linux/platform_device.h |  2 ++
 2 files changed, 30 insertions(+), 2 deletions(-)

diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 2b8755db76c6..940ce41f1887 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -505,12 +505,36 @@ static int platform_match(struct device * dev, struct device_driver * drv)
 	return (strncmp(pdev->name, drv->name, BUS_ID_SIZE) == 0);
 }
 
-static int platform_suspend(struct device * dev, pm_message_t state)
+static int platform_suspend(struct device *dev, pm_message_t mesg)
 {
 	int ret = 0;
 
 	if (dev->driver && dev->driver->suspend)
-		ret = dev->driver->suspend(dev, state);
+		ret = dev->driver->suspend(dev, mesg);
+
+	return ret;
+}
+
+static int platform_suspend_late(struct device *dev, pm_message_t mesg)
+{
+	struct platform_driver *drv = to_platform_driver(dev->driver);
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+	int ret = 0;
+
+	if (dev->driver && drv->suspend_late)
+		ret = drv->suspend_late(pdev, mesg);
+
+	return ret;
+}
+
+static int platform_resume_early(struct device *dev)
+{
+	struct platform_driver *drv = to_platform_driver(dev->driver);
+	struct platform_device *pdev = container_of(dev, struct platform_device, dev);
+	int ret = 0;
+
+	if (dev->driver && drv->resume_early)
+		ret = drv->resume_early(pdev);
 
 	return ret;
 }
@@ -531,6 +555,8 @@ struct bus_type platform_bus_type = {
 	.match		= platform_match,
 	.uevent		= platform_uevent,
 	.suspend	= platform_suspend,
+	.suspend_late	= platform_suspend_late,
+	.resume_early	= platform_resume_early,
 	.resume		= platform_resume,
 };
 EXPORT_SYMBOL_GPL(platform_bus_type);
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index 782090c68932..29cd6dee13db 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -49,6 +49,8 @@ struct platform_driver {
 	int (*remove)(struct platform_device *);
 	void (*shutdown)(struct platform_device *);
 	int (*suspend)(struct platform_device *, pm_message_t state);
+	int (*suspend_late)(struct platform_device *, pm_message_t state);
+	int (*resume_early)(struct platform_device *);
 	int (*resume)(struct platform_device *);
 	struct device_driver driver;
 };
-- 
cgit v1.2.1


From de0ff00d723fd821d372496e2c084805644aa5e1 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 27 Jun 2006 00:06:09 -0700
Subject: Driver core: add groups support to struct device

This is needed for the network class devices in order to be able to
convert over to use struct device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 34 ++++++++++++++++++++++++++++++++++
 include/linux/device.h |  1 +
 2 files changed, 35 insertions(+)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 5d4b7e04471e..641c0c42bb48 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -197,6 +197,35 @@ static ssize_t store_uevent(struct device *dev, struct device_attribute *attr,
 	return count;
 }
 
+static int device_add_groups(struct device *dev)
+{
+	int i;
+	int error = 0;
+
+	if (dev->groups) {
+		for (i = 0; dev->groups[i]; i++) {
+			error = sysfs_create_group(&dev->kobj, dev->groups[i]);
+			if (error) {
+				while (--i >= 0)
+					sysfs_remove_group(&dev->kobj, dev->groups[i]);
+				goto out;
+			}
+		}
+	}
+out:
+	return error;
+}
+
+static void device_remove_groups(struct device *dev)
+{
+	int i;
+	if (dev->groups) {
+		for (i = 0; dev->groups[i]; i++) {
+			sysfs_remove_group(&dev->kobj, dev->groups[i]);
+		}
+	}
+}
+
 static ssize_t show_dev(struct device *dev, struct device_attribute *attr,
 			char *buf)
 {
@@ -350,6 +379,8 @@ int device_add(struct device *dev)
 		sysfs_create_link(&dev->parent->kobj, &dev->kobj, class_name);
 	}
 
+	if ((error = device_add_groups(dev)))
+		goto GroupError;
 	if ((error = device_pm_add(dev)))
 		goto PMError;
 	if ((error = bus_add_device(dev)))
@@ -376,6 +407,8 @@ int device_add(struct device *dev)
  BusError:
 	device_pm_remove(dev);
  PMError:
+	device_remove_groups(dev);
+ GroupError:
 	if (dev->devt_attr) {
 		device_remove_file(dev, dev->devt_attr);
 		kfree(dev->devt_attr);
@@ -470,6 +503,7 @@ void device_del(struct device * dev)
 		up(&dev->class->sem);
 	}
 	device_remove_file(dev, &dev->uevent_attr);
+	device_remove_groups(dev);
 
 	/* Notify the platform of the removal, in case they
 	 * need to do anything...
diff --git a/include/linux/device.h b/include/linux/device.h
index b3646462d6dc..994d3ebd53f4 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -344,6 +344,7 @@ struct device {
 	struct list_head	node;
 	struct class		*class;		/* optional*/
 	dev_t			devt;		/* dev_t, creates the sysfs "dev" */
+	struct attribute_group	**groups;	/* optional groups */
 
 	void	(*release)(struct device * dev);
 };
-- 
cgit v1.2.1


From 64bb5d2c116478dba7501d2acf078ed74ba30c1f Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 28 Jun 2006 16:19:58 -0700
Subject: Driver core: allow devices in classes to have no parent

This fixes an oops when a device is attached to a class, yet has no
"parent" device.  An example of this would be the "lo" device in the
network core.

We should create a "virtual" subdirectory under /sys/devices/ for these,
but no one seems to agree on a proper name for it yet...

Oh, and update my copyright on the driver core.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 641c0c42bb48..5c91d0d81e78 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -3,6 +3,8 @@
  *
  * Copyright (c) 2002-3 Patrick Mochel
  * Copyright (c) 2002-3 Open Source Development Labs
+ * Copyright (c) 2006 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (c) 2006 Novell, Inc.
  *
  * This file is released under the GPLv2
  *
@@ -373,10 +375,11 @@ int device_add(struct device *dev)
 				  "subsystem");
 		sysfs_create_link(&dev->class->subsys.kset.kobj, &dev->kobj,
 				  dev->bus_id);
-
-		sysfs_create_link(&dev->kobj, &dev->parent->kobj, "device");
-		class_name = make_class_name(dev->class->name, &dev->kobj);
-		sysfs_create_link(&dev->parent->kobj, &dev->kobj, class_name);
+		if (parent) {
+			sysfs_create_link(&dev->kobj, &dev->parent->kobj, "device");
+			class_name = make_class_name(dev->class->name, &dev->kobj);
+			sysfs_create_link(&dev->parent->kobj, &dev->kobj, class_name);
+		}
 	}
 
 	if ((error = device_add_groups(dev)))
@@ -495,8 +498,10 @@ void device_del(struct device * dev)
 		sysfs_remove_link(&dev->kobj, "subsystem");
 		sysfs_remove_link(&dev->class->subsys.kset.kobj, dev->bus_id);
 		class_name = make_class_name(dev->class->name, &dev->kobj);
-		sysfs_remove_link(&dev->kobj, "device");
-		sysfs_remove_link(&dev->parent->kobj, class_name);
+		if (parent) {
+			sysfs_remove_link(&dev->kobj, "device");
+			sysfs_remove_link(&dev->parent->kobj, class_name);
+		}
 		kfree(class_name);
 		down(&dev->class->sem);
 		list_del_init(&dev->node);
@@ -625,10 +630,6 @@ struct device *device_create(struct class *class, struct device *parent,
 
 	if (class == NULL || IS_ERR(class))
 		goto error;
-	if (parent == NULL) {
-		printk(KERN_WARNING "%s does not work yet for NULL parents\n", __FUNCTION__);
-		goto error;
-	}
 
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
 	if (!dev) {
-- 
cgit v1.2.1


From 2620efef7029bb040430f50f0fc148f2d5e002ad Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 28 Jun 2006 16:19:58 -0700
Subject: Driver core: add ability for classes to handle devices properly

This adds two new callbacks to the class structure:
	int	(*dev_uevent)(struct device *dev, char **envp, int num_envp,
			char *buffer, int buffer_size);
	void	(*dev_release)(struct device *dev);

And one pointer:
	struct device_attribute		* dev_attrs;

which all corrispond with the same thing as the "normal" class devices
do, yet this is for when a struct device is bound to a class.

Someday soon, struct class_device will go away, and then the other
fields in this structure can be removed too.  But this is necessary in
order to get the transition to work properly.

Tested out on a network core patch that converted it to use struct
device instead of struct class_device.


Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 53 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h |  4 ++++
 2 files changed, 57 insertions(+)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 5c91d0d81e78..f1228f25efe0 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -94,6 +94,8 @@ static void device_release(struct kobject * kobj)
 
 	if (dev->release)
 		dev->release(dev);
+	else if (dev->class && dev->class->dev_release)
+		dev->class->dev_release(dev);
 	else {
 		printk(KERN_ERR "Device '%s' does not have a release() function, "
 			"it is broken and must be fixed.\n",
@@ -183,6 +185,15 @@ static int dev_uevent(struct kset *kset, struct kobject *kobj, char **envp,
 		}
 	}
 
+	if (dev->class && dev->class->dev_uevent) {
+		/* have the class specific function add its stuff */
+		retval = dev->class->dev_uevent(dev, envp, num_envp, buffer, buffer_size);
+			if (retval) {
+				pr_debug("%s - dev_uevent() returned %d\n",
+					 __FUNCTION__, retval);
+		}
+	}
+
 	return retval;
 }
 
@@ -228,6 +239,43 @@ static void device_remove_groups(struct device *dev)
 	}
 }
 
+static int device_add_attrs(struct device *dev)
+{
+	struct class *class = dev->class;
+	int error = 0;
+	int i;
+
+	if (!class)
+		return 0;
+
+	if (class->dev_attrs) {
+		for (i = 0; attr_name(class->dev_attrs[i]); i++) {
+			error = device_create_file(dev, &class->dev_attrs[i]);
+			if (error)
+				break;
+		}
+	}
+	if (error)
+		while (--i >= 0)
+			device_remove_file(dev, &class->dev_attrs[i]);
+	return error;
+}
+
+static void device_remove_attrs(struct device *dev)
+{
+	struct class *class = dev->class;
+	int i;
+
+	if (!class)
+		return;
+
+	if (class->dev_attrs) {
+		for (i = 0; attr_name(class->dev_attrs[i]); i++)
+			device_remove_file(dev, &class->dev_attrs[i]);
+	}
+}
+
+
 static ssize_t show_dev(struct device *dev, struct device_attribute *attr,
 			char *buf)
 {
@@ -382,6 +430,8 @@ int device_add(struct device *dev)
 		}
 	}
 
+	if ((error = device_add_attrs(dev)))
+		goto AttrsError;
 	if ((error = device_add_groups(dev)))
 		goto GroupError;
 	if ((error = device_pm_add(dev)))
@@ -412,6 +462,8 @@ int device_add(struct device *dev)
  PMError:
 	device_remove_groups(dev);
  GroupError:
+ 	device_remove_attrs(dev);
+ AttrsError:
 	if (dev->devt_attr) {
 		device_remove_file(dev, dev->devt_attr);
 		kfree(dev->devt_attr);
@@ -509,6 +561,7 @@ void device_del(struct device * dev)
 	}
 	device_remove_file(dev, &dev->uevent_attr);
 	device_remove_groups(dev);
+	device_remove_attrs(dev);
 
 	/* Notify the platform of the removal, in case they
 	 * need to do anything...
diff --git a/include/linux/device.h b/include/linux/device.h
index 994d3ebd53f4..3122bd25ce42 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -151,12 +151,16 @@ struct class {
 
 	struct class_attribute		* class_attrs;
 	struct class_device_attribute	* class_dev_attrs;
+	struct device_attribute		* dev_attrs;
 
 	int	(*uevent)(struct class_device *dev, char **envp,
 			   int num_envp, char *buffer, int buffer_size);
+	int	(*dev_uevent)(struct device *dev, char **envp, int num_envp,
+				char *buffer, int buffer_size);
 
 	void	(*release)(struct class_device *dev);
 	void	(*class_release)(struct class *class);
+	void	(*dev_release)(struct device *dev);
 
 	int	(*suspend)(struct device *, pm_message_t state);
 	int	(*resume)(struct device *);
-- 
cgit v1.2.1


From a2de48cace5d0993da6cfa28b276ae724dc3569b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 3 Jul 2006 14:31:12 -0700
Subject: Driver core: add device_rename function

The network layer needs this to convert to using struct device instead
of a struct class_device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/device.h |  1 +
 2 files changed, 56 insertions(+)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index f1228f25efe0..0db47561331e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -736,3 +736,58 @@ void device_destroy(struct class *class, dev_t devt)
 		device_unregister(dev);
 }
 EXPORT_SYMBOL_GPL(device_destroy);
+
+/**
+ * device_rename - renames a device
+ * @dev: the pointer to the struct device to be renamed
+ * @new_name: the new name of the device
+ */
+int device_rename(struct device *dev, char *new_name)
+{
+	char *old_class_name = NULL;
+	char *new_class_name = NULL;
+	char *old_symlink_name = NULL;
+	int error;
+
+	dev = get_device(dev);
+	if (!dev)
+		return -EINVAL;
+
+	pr_debug("DEVICE: renaming '%s' to '%s'\n", dev->bus_id, new_name);
+
+	if ((dev->class) && (dev->parent))
+		old_class_name = make_class_name(dev->class->name, &dev->kobj);
+
+	if (dev->class) {
+		old_symlink_name = kmalloc(BUS_ID_SIZE, GFP_KERNEL);
+		if (!old_symlink_name)
+			return -ENOMEM;
+		strlcpy(old_symlink_name, dev->bus_id, BUS_ID_SIZE);
+	}
+
+	strlcpy(dev->bus_id, new_name, BUS_ID_SIZE);
+
+	error = kobject_rename(&dev->kobj, new_name);
+
+	if (old_class_name) {
+		new_class_name = make_class_name(dev->class->name, &dev->kobj);
+		if (new_class_name) {
+			sysfs_create_link(&dev->parent->kobj, &dev->kobj,
+					  new_class_name);
+			sysfs_remove_link(&dev->parent->kobj, old_class_name);
+		}
+	}
+	if (dev->class) {
+		sysfs_remove_link(&dev->class->subsys.kset.kobj,
+				  old_symlink_name);
+		sysfs_create_link(&dev->class->subsys.kset.kobj, &dev->kobj,
+				  dev->bus_id);
+	}
+	put_device(dev);
+
+	kfree(old_class_name);
+	kfree(new_class_name);
+	kfree(old_symlink_name);
+
+	return error;
+}
diff --git a/include/linux/device.h b/include/linux/device.h
index 3122bd25ce42..3400e09bf458 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -380,6 +380,7 @@ extern int device_add(struct device * dev);
 extern void device_del(struct device * dev);
 extern int device_for_each_child(struct device *, void *,
 		     int (*fn)(struct device *, void *));
+extern int device_rename(struct device *dev, char *new_name);
 
 /*
  * Manual binding of a device to driver. See drivers/base/bus.c
-- 
cgit v1.2.1


From c205ef4880273d2de4ee5388d4e52227ff688cc4 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 7 Aug 2006 22:19:37 -0700
Subject: Driver core: create devices/virtual/ tree

This change creates a devices/virtual/CLASS_NAME tree for struct devices
that belong to a class, yet do not have a "real" struct device for a
parent.  It automatically creates the directories on the fly as needed.


Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 17 +++++++++++++++++
 drivers/base/core.c    |  7 +++++++
 include/linux/device.h |  5 ++++-
 3 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/drivers/base/class.c b/drivers/base/class.c
index e078bc21d52e..cbdf47c0c60d 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -19,6 +19,8 @@
 #include <linux/slab.h>
 #include "base.h"
 
+extern struct subsystem devices_subsys;
+
 #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr)
 #define to_class(obj) container_of(obj, struct class, subsys.kset.kobj)
 
@@ -878,7 +880,22 @@ void class_interface_unregister(struct class_interface *class_intf)
 	class_put(parent);
 }
 
+int virtual_device_parent(struct device *dev)
+{
+	if (!dev->class)
+		return -ENODEV;
+
+	if (!dev->class->virtual_dir) {
+		static struct kobject *virtual_dir = NULL;
 
+		if (!virtual_dir)
+			virtual_dir = kobject_add_dir(&devices_subsys.kset.kobj, "virtual");
+		dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name);
+	}
+
+	dev->kobj.parent = dev->class->virtual_dir;
+	return 0;
+}
 
 int __init classes_init(void)
 {
diff --git a/drivers/base/core.c b/drivers/base/core.c
index 0db47561331e..e21a65fc043e 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -378,6 +378,13 @@ int device_add(struct device *dev)
 	if (!dev || !strlen(dev->bus_id))
 		goto Error;
 
+	/* if this is a class device, and has no parent, create one */
+	if ((dev->class) && (dev->parent == NULL)) {
+		error = virtual_device_parent(dev);
+		if (error)
+			goto Error;
+	}
+
 	parent = get_device(dev->parent);
 
 	pr_debug("DEV: registering device: ID = '%s'\n", dev->bus_id);
diff --git a/include/linux/device.h b/include/linux/device.h
index 3400e09bf458..bbb0d6b5d232 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -149,6 +149,8 @@ struct class {
 	struct list_head	interfaces;
 	struct semaphore	sem;	/* locks both the children and interfaces lists */
 
+	struct kobject		*virtual_dir;
+
 	struct class_attribute		* class_attrs;
 	struct class_device_attribute	* class_dev_attrs;
 	struct device_attribute		* dev_attrs;
@@ -291,7 +293,6 @@ extern struct class_device *class_device_create(struct class *cls,
 					__attribute__((format(printf,5,6)));
 extern void class_device_destroy(struct class *cls, dev_t devt);
 
-
 /* interface for exporting device attributes */
 struct device_attribute {
 	struct attribute	attr;
@@ -400,6 +401,8 @@ extern struct device *device_create(struct class *cls, struct device *parent,
 				    __attribute__((format(printf,4,5)));
 extern void device_destroy(struct class *cls, dev_t devt);
 
+extern int virtual_device_parent(struct device *dev);
+
 /*
  * Platform "fixup" functions - allow the platform to have their say
  * about devices and actions that the general device layer doesn't
-- 
cgit v1.2.1


From c47ed219ba81632595e9f02e27318151fec16c9e Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 13 Sep 2006 15:34:05 +0200
Subject: Class: add support for class interfaces for devices

When moving class_device usage over to device, we need to handle
class_interfaces properly with devices.  This patch adds that support.


Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/class.c   | 10 ++++++++++
 drivers/base/core.c    | 14 +++++++++++++-
 include/linux/device.h |  2 ++
 3 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/drivers/base/class.c b/drivers/base/class.c
index cbdf47c0c60d..b06b0e2b9c62 100644
--- a/drivers/base/class.c
+++ b/drivers/base/class.c
@@ -842,6 +842,7 @@ int class_interface_register(struct class_interface *class_intf)
 {
 	struct class *parent;
 	struct class_device *class_dev;
+	struct device *dev;
 
 	if (!class_intf || !class_intf->class)
 		return -ENODEV;
@@ -856,6 +857,10 @@ int class_interface_register(struct class_interface *class_intf)
 		list_for_each_entry(class_dev, &parent->children, node)
 			class_intf->add(class_dev, class_intf);
 	}
+	if (class_intf->add_dev) {
+		list_for_each_entry(dev, &parent->devices, node)
+			class_intf->add_dev(dev, class_intf);
+	}
 	up(&parent->sem);
 
 	return 0;
@@ -865,6 +870,7 @@ void class_interface_unregister(struct class_interface *class_intf)
 {
 	struct class * parent = class_intf->class;
 	struct class_device *class_dev;
+	struct device *dev;
 
 	if (!parent)
 		return;
@@ -875,6 +881,10 @@ void class_interface_unregister(struct class_interface *class_intf)
 		list_for_each_entry(class_dev, &parent->children, node)
 			class_intf->remove(class_dev, class_intf);
 	}
+	if (class_intf->remove_dev) {
+		list_for_each_entry(dev, &parent->devices, node)
+			class_intf->remove_dev(dev, class_intf);
+	}
 	up(&parent->sem);
 
 	class_put(parent);
diff --git a/drivers/base/core.c b/drivers/base/core.c
index e21a65fc043e..1d3d3582fcca 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -372,6 +372,7 @@ int device_add(struct device *dev)
 {
 	struct device *parent = NULL;
 	char *class_name = NULL;
+	struct class_interface *class_intf;
 	int error = -EINVAL;
 
 	dev = get_device(dev);
@@ -451,9 +452,14 @@ int device_add(struct device *dev)
 		klist_add_tail(&dev->knode_parent, &parent->klist_children);
 
 	if (dev->class) {
-		/* tie the class to the device */
 		down(&dev->class->sem);
+		/* tie the class to the device */
 		list_add_tail(&dev->node, &dev->class->devices);
+
+		/* notify any interfaces that the device is here */
+		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+			if (class_intf->add_dev)
+				class_intf->add_dev(dev, class_intf);
 		up(&dev->class->sem);
 	}
 
@@ -548,6 +554,7 @@ void device_del(struct device * dev)
 {
 	struct device * parent = dev->parent;
 	char *class_name = NULL;
+	struct class_interface *class_intf;
 
 	if (parent)
 		klist_del(&dev->knode_parent);
@@ -563,6 +570,11 @@ void device_del(struct device * dev)
 		}
 		kfree(class_name);
 		down(&dev->class->sem);
+		/* notify any interfaces that the device is now gone */
+		list_for_each_entry(class_intf, &dev->class->interfaces, node)
+			if (class_intf->remove_dev)
+				class_intf->remove_dev(dev, class_intf);
+		/* remove the device from the class list */
 		list_del_init(&dev->node);
 		up(&dev->class->sem);
 	}
diff --git a/include/linux/device.h b/include/linux/device.h
index bbb0d6b5d232..e0fae0e76fa9 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -278,6 +278,8 @@ struct class_interface {
 
 	int (*add)	(struct class_device *, struct class_interface *);
 	void (*remove)	(struct class_device *, struct class_interface *);
+	int (*add_dev)		(struct device *, struct class_interface *);
+	void (*remove_dev)	(struct device *, struct class_interface *);
 };
 
 extern int class_interface_register(struct class_interface *);
-- 
cgit v1.2.1


From 2589f1887b0bf9f08ec3d7f3c5705ccb7c628076 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 19 Sep 2006 09:39:19 -0700
Subject: Driver core: add ability for devices to create and remove bin files

Makes it easier for devices to create and remove binary attribute files
so they don't have to call directly into sysfs.  This is needed to help
with the conversion from struct class_device to struct device.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c    | 26 ++++++++++++++++++++++++++
 include/linux/device.h |  4 ++++
 2 files changed, 30 insertions(+)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 1d3d3582fcca..bc9f35c81691 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -319,6 +319,32 @@ void device_remove_file(struct device * dev, struct device_attribute * attr)
 	}
 }
 
+/**
+ * device_create_bin_file - create sysfs binary attribute file for device.
+ * @dev: device.
+ * @attr: device binary attribute descriptor.
+ */
+int device_create_bin_file(struct device *dev, struct bin_attribute *attr)
+{
+	int error = -EINVAL;
+	if (dev)
+		error = sysfs_create_bin_file(&dev->kobj, attr);
+	return error;
+}
+EXPORT_SYMBOL_GPL(device_create_bin_file);
+
+/**
+ * device_remove_bin_file - remove sysfs binary attribute file
+ * @dev: device.
+ * @attr: device binary attribute descriptor.
+ */
+void device_remove_bin_file(struct device *dev, struct bin_attribute *attr)
+{
+	if (dev)
+		sysfs_remove_bin_file(&dev->kobj, attr);
+}
+EXPORT_SYMBOL_GPL(device_remove_bin_file);
+
 static void klist_children_get(struct klist_node *n)
 {
 	struct device *dev = container_of(n, struct device, knode_parent);
diff --git a/include/linux/device.h b/include/linux/device.h
index e0fae0e76fa9..7d447d7271c5 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -309,6 +309,10 @@ struct device_attribute dev_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
 extern int device_create_file(struct device *device, struct device_attribute * entry);
 extern void device_remove_file(struct device * dev, struct device_attribute * attr);
+extern int __must_check device_create_bin_file(struct device *dev,
+					       struct bin_attribute *attr);
+extern void device_remove_bin_file(struct device *dev,
+				   struct bin_attribute *attr);
 struct device {
 	struct klist		klist_children;
 	struct klist_node	knode_parent;		/* node in sibling list */
-- 
cgit v1.2.1


From 10188012daa586ae7fcbef272e4db4f404741adf Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Tue, 11 Jul 2006 20:49:41 -0700
Subject: kobject: must_check fixes

Check all __must_check warnings in lib/kobject.c

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 lib/kobject.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/lib/kobject.c b/lib/kobject.c
index 8e7c71993487..1699eb9161f3 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -407,6 +407,7 @@ static struct kobj_type dir_ktype = {
 struct kobject *kobject_add_dir(struct kobject *parent, const char *name)
 {
 	struct kobject *k;
+	int ret;
 
 	if (!parent)
 		return NULL;
@@ -418,7 +419,13 @@ struct kobject *kobject_add_dir(struct kobject *parent, const char *name)
 	k->parent = parent;
 	k->ktype = &dir_ktype;
 	kobject_set_name(k, name);
-	kobject_register(k);
+	ret = kobject_register(k);
+	if (ret < 0) {
+		printk(KERN_WARNING "kobject_add_dir: "
+			"kobject_register error: %d\n", ret);
+		kobject_del(k);
+		return NULL;
+	}
 
 	return k;
 }
-- 
cgit v1.2.1


From 995982ca79d9262869513948ec7c540f32035491 Mon Sep 17 00:00:00 2001
From: "Randy.Dunlap" <rdunlap@xenotime.net>
Date: Mon, 10 Jul 2006 23:05:25 -0700
Subject: sysfs_remove_bin_file: no return value, dump_stack on error

Make sysfs_remove_bin_file() void.  If it detects an error,
printk the file name and call dump_stack().

sysfs_hash_and_remove() now returns an error code indicating
its success or failure so that sysfs_remove_bin_file() can
know success/failure.

Convert the only driver that checked the return value of
sysfs_remove_bin_file().

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_ibm.c |  4 +---
 fs/sysfs/bin.c                    | 13 ++++++++-----
 fs/sysfs/inode.c                  | 11 ++++++++---
 fs/sysfs/sysfs.h                  |  2 +-
 include/linux/sysfs.h             |  2 +-
 5 files changed, 19 insertions(+), 13 deletions(-)

diff --git a/drivers/pci/hotplug/acpiphp_ibm.c b/drivers/pci/hotplug/acpiphp_ibm.c
index 317457dd4014..d0a07d9ab30c 100644
--- a/drivers/pci/hotplug/acpiphp_ibm.c
+++ b/drivers/pci/hotplug/acpiphp_ibm.c
@@ -487,9 +487,7 @@ static void __exit ibm_acpiphp_exit(void)
 	if (ACPI_FAILURE(status))
 		err("%s: Notification handler removal failed\n", __FUNCTION__);
 	/* remove the /sys entries */
-	if (sysfs_remove_bin_file(sysdir, &ibm_apci_table_attr))
-		err("%s: removal of sysfs file apci_table failed\n",
-				__FUNCTION__);
+	sysfs_remove_bin_file(sysdir, &ibm_apci_table_attr);
 }
 
 module_init(ibm_acpiphp_init);
diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c
index c16a93c353c0..98022e41cda1 100644
--- a/fs/sysfs/bin.c
+++ b/fs/sysfs/bin.c
@@ -10,6 +10,7 @@
 
 #include <linux/errno.h>
 #include <linux/fs.h>
+#include <linux/kernel.h>
 #include <linux/kobject.h>
 #include <linux/module.h>
 #include <linux/slab.h>
@@ -176,7 +177,6 @@ const struct file_operations bin_fops = {
  *	sysfs_create_bin_file - create binary file for object.
  *	@kobj:	object.
  *	@attr:	attribute descriptor.
- *
  */
 
 int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
@@ -191,13 +191,16 @@ int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr)
  *	sysfs_remove_bin_file - remove binary file for object.
  *	@kobj:	object.
  *	@attr:	attribute descriptor.
- *
  */
 
-int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
+void sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr)
 {
-	sysfs_hash_and_remove(kobj->dentry,attr->attr.name);
-	return 0;
+	if (sysfs_hash_and_remove(kobj->dentry, attr->attr.name) < 0) {
+		printk(KERN_ERR "%s: "
+			"bad dentry or inode or no such file: \"%s\"\n",
+			__FUNCTION__, attr->attr.name);
+		dump_stack();
+	}
 }
 
 EXPORT_SYMBOL_GPL(sysfs_create_bin_file);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 9889e54e1f13..fd7cd5f843d2 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -12,6 +12,7 @@
 #include <linux/namei.h>
 #include <linux/backing-dev.h>
 #include <linux/capability.h>
+#include <linux/errno.h>
 #include "sysfs.h"
 
 extern struct super_block * sysfs_sb;
@@ -234,17 +235,18 @@ void sysfs_drop_dentry(struct sysfs_dirent * sd, struct dentry * parent)
 	}
 }
 
-void sysfs_hash_and_remove(struct dentry * dir, const char * name)
+int sysfs_hash_and_remove(struct dentry * dir, const char * name)
 {
 	struct sysfs_dirent * sd;
 	struct sysfs_dirent * parent_sd;
+	int found = 0;
 
 	if (!dir)
-		return;
+		return -ENOENT;
 
 	if (dir->d_inode == NULL)
 		/* no inode means this hasn't been made visible yet */
-		return;
+		return -ENOENT;
 
 	parent_sd = dir->d_fsdata;
 	mutex_lock(&dir->d_inode->i_mutex);
@@ -255,8 +257,11 @@ void sysfs_hash_and_remove(struct dentry * dir, const char * name)
 			list_del_init(&sd->s_sibling);
 			sysfs_drop_dentry(sd, dir);
 			sysfs_put(sd);
+			found = 1;
 			break;
 		}
 	}
 	mutex_unlock(&dir->d_inode->i_mutex);
+
+	return found ? 0 : -ENOENT;
 }
diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h
index 3651ffb5ec09..6f3d6bd52887 100644
--- a/fs/sysfs/sysfs.h
+++ b/fs/sysfs/sysfs.h
@@ -10,7 +10,7 @@ extern int sysfs_make_dirent(struct sysfs_dirent *, struct dentry *, void *,
 				umode_t, int);
 
 extern int sysfs_add_file(struct dentry *, const struct attribute *, int);
-extern void sysfs_hash_and_remove(struct dentry * dir, const char * name);
+extern int sysfs_hash_and_remove(struct dentry * dir, const char * name);
 extern struct sysfs_dirent *sysfs_find(struct sysfs_dirent *dir, const char * name);
 
 extern int sysfs_create_subdir(struct kobject *, const char *, struct dentry **);
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 1ea5d3cda6ae..95f6db54d8d2 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -114,7 +114,7 @@ extern void
 sysfs_remove_link(struct kobject *, const char * name);
 
 int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
-int sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr);
+void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr);
 
 int sysfs_create_group(struct kobject *, const struct attribute_group *);
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
-- 
cgit v1.2.1


From 9de72ee59029087fc8300633113c75a5fe73a7b8 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor@insightbb.com>
Date: Sat, 15 Jul 2006 00:31:54 -0400
Subject: Driver core: fix comments in drivers/base/power/resume.c

Driver core: fix comments in drivers/base/power/resume.c

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/power/resume.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c
index 7cb62d62c958..020be36705a6 100644
--- a/drivers/base/power/resume.c
+++ b/drivers/base/power/resume.c
@@ -106,12 +106,12 @@ EXPORT_SYMBOL_GPL(device_resume);
 
 
 /**
- *	device_power_up_irq - Power on some devices.
+ *	dpm_power_up - Power on some devices.
  *
  *	Walk the dpm_off_irq list and power each device up. This
  *	is used for devices that required they be powered down with
- *	interrupts disabled. As devices are powered on, they are moved to
- *	the dpm_suspended list.
+ *	interrupts disabled. As devices are powered on, they are moved
+ *	to the dpm_active list.
  *
  *	Interrupts must be disabled when calling this.
  */
@@ -129,7 +129,7 @@ void dpm_power_up(void)
 
 
 /**
- *	device_pm_power_up - Turn on all devices that need special attention.
+ *	device_power_up - Turn on all devices that need special attention.
  *
  *	Power on system devices then devices that required we shut them down
  *	with interrupts disabled.
-- 
cgit v1.2.1


From 35acfdd7253025e8441883fd8f879f5240844f95 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Sat, 15 Jul 2006 01:30:11 +0900
Subject: Driver core: fixed add_bind_files() definition

When CONFIG_HOTPLUG is n, add_bind_files() definition is wrong.
This patch has fixed it.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 2e954d07175a..4d22a1d10a1c 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -467,7 +467,7 @@ static void remove_bind_files(struct device_driver *drv)
 	driver_remove_file(drv, &driver_attr_unbind);
 }
 #else
-static inline void add_bind_files(struct device_driver *drv) {}
+static inline int add_bind_files(struct device_driver *drv) { return 0; }
 static inline void remove_bind_files(struct device_driver *drv) {}
 #endif
 
-- 
cgit v1.2.1


From 4a7fb6363f2d1a6c09a10253937f672f3c7929e1 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:17 -0700
Subject: add __must_check to device management code

We're getting a lot of crashes in the sysfs/kobject/device/bus/class code and
they're very hard to diagnose.

I'm suspecting that in some cases this is because drivers aren't checking
return values and aren't handling errors correctly.  So the code blithely
blunders on and crashes later in very obscure ways.

There's just no reason to ignore errors which can and do occur.  So the patch
sprinkles __must_check all over these APIs.

Causes 1,513 new warnings.  Heh.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/device.h  | 52 ++++++++++++++++++++++++++-----------------------
 include/linux/kobject.h | 16 ++++++++-------
 include/linux/pci.h     | 34 +++++++++++++++++---------------
 include/linux/sysfs.h   | 19 ++++++++++--------
 4 files changed, 66 insertions(+), 55 deletions(-)

diff --git a/include/linux/device.h b/include/linux/device.h
index 7d447d7271c5..ad4db72f5733 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -15,6 +15,7 @@
 #include <linux/kobject.h>
 #include <linux/klist.h>
 #include <linux/list.h>
+#include <linux/compiler.h>
 #include <linux/types.h>
 #include <linux/module.h>
 #include <linux/pm.h>
@@ -58,7 +59,7 @@ struct bus_type {
 	int (*resume)(struct device * dev);
 };
 
-extern int bus_register(struct bus_type * bus);
+extern int __must_check bus_register(struct bus_type * bus);
 extern void bus_unregister(struct bus_type * bus);
 
 extern void bus_rescan_devices(struct bus_type * bus);
@@ -70,9 +71,9 @@ int bus_for_each_dev(struct bus_type * bus, struct device * start, void * data,
 struct device * bus_find_device(struct bus_type *bus, struct device *start,
 				void *data, int (*match)(struct device *, void *));
 
-int bus_for_each_drv(struct bus_type * bus, struct device_driver * start, 
-		     void * data, int (*fn)(struct device_driver *, void *));
-
+int __must_check bus_for_each_drv(struct bus_type *bus,
+		struct device_driver *start, void *data,
+		int (*fn)(struct device_driver *, void *));
 
 /* driverfs interface for exporting bus attributes */
 
@@ -85,7 +86,8 @@ struct bus_attribute {
 #define BUS_ATTR(_name,_mode,_show,_store)	\
 struct bus_attribute bus_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int bus_create_file(struct bus_type *, struct bus_attribute *);
+extern int __must_check bus_create_file(struct bus_type *,
+					struct bus_attribute *);
 extern void bus_remove_file(struct bus_type *, struct bus_attribute *);
 
 struct device_driver {
@@ -107,14 +109,13 @@ struct device_driver {
 };
 
 
-extern int driver_register(struct device_driver * drv);
+extern int __must_check driver_register(struct device_driver * drv);
 extern void driver_unregister(struct device_driver * drv);
 
 extern struct device_driver * get_driver(struct device_driver * drv);
 extern void put_driver(struct device_driver * drv);
 extern struct device_driver *driver_find(const char *name, struct bus_type *bus);
 
-
 /* driverfs interface for exporting driver attributes */
 
 struct driver_attribute {
@@ -126,16 +127,17 @@ struct driver_attribute {
 #define DRIVER_ATTR(_name,_mode,_show,_store)	\
 struct driver_attribute driver_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int driver_create_file(struct device_driver *, struct driver_attribute *);
+extern int __must_check driver_create_file(struct device_driver *,
+					struct driver_attribute *);
 extern void driver_remove_file(struct device_driver *, struct driver_attribute *);
 
-extern int driver_for_each_device(struct device_driver * drv, struct device * start,
-				  void * data, int (*fn)(struct device *, void *));
+extern int __must_check driver_for_each_device(struct device_driver * drv,
+		struct device *start, void *data,
+		int (*fn)(struct device *, void *));
 struct device * driver_find_device(struct device_driver *drv,
 				   struct device *start, void *data,
 				   int (*match)(struct device *, void *));
 
-
 /*
  * device classes
  */
@@ -168,7 +170,7 @@ struct class {
 	int	(*resume)(struct device *);
 };
 
-extern int class_register(struct class *);
+extern int __must_check class_register(struct class *);
 extern void class_unregister(struct class *);
 
 
@@ -181,7 +183,8 @@ struct class_attribute {
 #define CLASS_ATTR(_name,_mode,_show,_store)			\
 struct class_attribute class_attr_##_name = __ATTR(_name,_mode,_show,_store) 
 
-extern int class_create_file(struct class *, const struct class_attribute *);
+extern int __must_check class_create_file(struct class *,
+					const struct class_attribute *);
 extern void class_remove_file(struct class *, const struct class_attribute *);
 
 struct class_device_attribute {
@@ -194,7 +197,7 @@ struct class_device_attribute {
 struct class_device_attribute class_device_attr_##_name = 	\
 	__ATTR(_name,_mode,_show,_store)
 
-extern int class_device_create_file(struct class_device *,
+extern int __must_check class_device_create_file(struct class_device *,
 				    const struct class_device_attribute *);
 
 /**
@@ -254,10 +257,10 @@ class_set_devdata (struct class_device *dev, void *data)
 }
 
 
-extern int class_device_register(struct class_device *);
+extern int __must_check class_device_register(struct class_device *);
 extern void class_device_unregister(struct class_device *);
 extern void class_device_initialize(struct class_device *);
-extern int class_device_add(struct class_device *);
+extern int __must_check class_device_add(struct class_device *);
 extern void class_device_del(struct class_device *);
 
 extern int class_device_rename(struct class_device *, char *);
@@ -267,7 +270,7 @@ extern void class_device_put(struct class_device *);
 
 extern void class_device_remove_file(struct class_device *, 
 				     const struct class_device_attribute *);
-extern int class_device_create_bin_file(struct class_device *,
+extern int __must_check class_device_create_bin_file(struct class_device *,
 					struct bin_attribute *);
 extern void class_device_remove_bin_file(struct class_device *,
 					 struct bin_attribute *);
@@ -282,7 +285,7 @@ struct class_interface {
 	void (*remove_dev)	(struct device *, struct class_interface *);
 };
 
-extern int class_interface_register(struct class_interface *);
+extern int __must_check class_interface_register(struct class_interface *);
 extern void class_interface_unregister(struct class_interface *);
 
 extern struct class *class_create(struct module *owner, const char *name);
@@ -307,7 +310,8 @@ struct device_attribute {
 #define DEVICE_ATTR(_name,_mode,_show,_store) \
 struct device_attribute dev_attr_##_name = __ATTR(_name,_mode,_show,_store)
 
-extern int device_create_file(struct device *device, struct device_attribute * entry);
+extern int __must_check device_create_file(struct device *device,
+					struct device_attribute * entry);
 extern void device_remove_file(struct device * dev, struct device_attribute * attr);
 extern int __must_check device_create_bin_file(struct device *dev,
 					       struct bin_attribute *attr);
@@ -380,12 +384,12 @@ static inline int device_is_registered(struct device *dev)
 /*
  * High level routines for use by the bus drivers
  */
-extern int device_register(struct device * dev);
+extern int __must_check device_register(struct device * dev);
 extern void device_unregister(struct device * dev);
 extern void device_initialize(struct device * dev);
-extern int device_add(struct device * dev);
+extern int __must_check device_add(struct device * dev);
 extern void device_del(struct device * dev);
-extern int device_for_each_child(struct device *, void *,
+extern int __must_check device_for_each_child(struct device *, void *,
 		     int (*fn)(struct device *, void *));
 extern int device_rename(struct device *dev, char *new_name);
 
@@ -395,7 +399,7 @@ extern int device_rename(struct device *dev, char *new_name);
  */
 extern void device_bind_driver(struct device * dev);
 extern void device_release_driver(struct device * dev);
-extern int  device_attach(struct device * dev);
+extern int  __must_check device_attach(struct device * dev);
 extern void driver_attach(struct device_driver * drv);
 extern void device_reprobe(struct device *dev);
 
@@ -433,7 +437,7 @@ extern void device_shutdown(void);
 
 
 /* drivers/base/firmware.c */
-extern int firmware_register(struct subsystem *);
+extern int __must_check firmware_register(struct subsystem *);
 extern void firmware_unregister(struct subsystem *);
 
 /* debugging and troubleshooting/diagnostic helpers. */
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 2d229327959e..bcd9cd173c2c 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -20,6 +20,7 @@
 #include <linux/types.h>
 #include <linux/list.h>
 #include <linux/sysfs.h>
+#include <linux/compiler.h>
 #include <linux/spinlock.h>
 #include <linux/rwsem.h>
 #include <linux/kref.h>
@@ -71,12 +72,12 @@ static inline const char * kobject_name(const struct kobject * kobj)
 extern void kobject_init(struct kobject *);
 extern void kobject_cleanup(struct kobject *);
 
-extern int kobject_add(struct kobject *);
+extern int __must_check kobject_add(struct kobject *);
 extern void kobject_del(struct kobject *);
 
-extern int kobject_rename(struct kobject *, const char *new_name);
+extern int __must_check kobject_rename(struct kobject *, const char *new_name);
 
-extern int kobject_register(struct kobject *);
+extern int __must_check kobject_register(struct kobject *);
 extern void kobject_unregister(struct kobject *);
 
 extern struct kobject * kobject_get(struct kobject *);
@@ -128,8 +129,8 @@ struct kset {
 
 
 extern void kset_init(struct kset * k);
-extern int kset_add(struct kset * k);
-extern int kset_register(struct kset * k);
+extern int __must_check kset_add(struct kset * k);
+extern int __must_check kset_register(struct kset * k);
 extern void kset_unregister(struct kset * k);
 
 static inline struct kset * to_kset(struct kobject * kobj)
@@ -239,7 +240,7 @@ extern struct subsystem hypervisor_subsys;
 	(obj)->subsys.kset.kobj.kset = &(_subsys).kset
 
 extern void subsystem_init(struct subsystem *);
-extern int subsystem_register(struct subsystem *);
+extern int __must_check subsystem_register(struct subsystem *);
 extern void subsystem_unregister(struct subsystem *);
 
 static inline struct subsystem * subsys_get(struct subsystem * s)
@@ -258,7 +259,8 @@ struct subsys_attribute {
 	ssize_t (*store)(struct subsystem *, const char *, size_t); 
 };
 
-extern int subsys_create_file(struct subsystem * , struct subsys_attribute *);
+extern int __must_check subsys_create_file(struct subsystem * ,
+					struct subsys_attribute *);
 
 #if defined(CONFIG_HOTPLUG)
 void kobject_uevent(struct kobject *kobj, enum kobject_action action);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 9514bbfe96e2..3ec72551ac31 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -49,6 +49,7 @@
 #include <linux/types.h>
 #include <linux/ioport.h>
 #include <linux/list.h>
+#include <linux/compiler.h>
 #include <linux/errno.h>
 #include <linux/device.h>
 
@@ -403,7 +404,7 @@ extern struct list_head pci_root_buses;	/* list of all known PCI buses */
 extern struct list_head pci_devices;	/* list of all devices */
 
 void pcibios_fixup_bus(struct pci_bus *);
-int pcibios_enable_device(struct pci_dev *, int mask);
+int __must_check pcibios_enable_device(struct pci_dev *, int mask);
 char *pcibios_setup (char *str);
 
 /* Used only when drivers/pci/setup.c is used */
@@ -490,19 +491,19 @@ static inline int pci_write_config_dword(struct pci_dev *dev, int where, u32 val
 	return pci_bus_write_config_dword (dev->bus, dev->devfn, where, val);
 }
 
-int pci_enable_device(struct pci_dev *dev);
-int pci_enable_device_bars(struct pci_dev *dev, int mask);
+int __must_check pci_enable_device(struct pci_dev *dev);
+int __must_check pci_enable_device_bars(struct pci_dev *dev, int mask);
 void pci_disable_device(struct pci_dev *dev);
 void pci_set_master(struct pci_dev *dev);
 #define HAVE_PCI_SET_MWI
-int pci_set_mwi(struct pci_dev *dev);
+int __must_check pci_set_mwi(struct pci_dev *dev);
 void pci_clear_mwi(struct pci_dev *dev);
 void pci_intx(struct pci_dev *dev, int enable);
 int pci_set_dma_mask(struct pci_dev *dev, u64 mask);
 int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask);
 void pci_update_resource(struct pci_dev *dev, struct resource *res, int resno);
-int pci_assign_resource(struct pci_dev *dev, int i);
-int pci_assign_resource_fixed(struct pci_dev *dev, int i);
+int __must_check pci_assign_resource(struct pci_dev *dev, int i);
+int __must_check pci_assign_resource_fixed(struct pci_dev *dev, int i);
 void pci_restore_bars(struct pci_dev *dev);
 
 /* ROM control related routines */
@@ -528,23 +529,24 @@ void pdev_sort_resources(struct pci_dev *, struct resource_list *);
 void pci_fixup_irqs(u8 (*)(struct pci_dev *, u8 *),
 		    int (*)(struct pci_dev *, u8, u8));
 #define HAVE_PCI_REQ_REGIONS	2
-int pci_request_regions(struct pci_dev *, const char *);
+int __must_check pci_request_regions(struct pci_dev *, const char *);
 void pci_release_regions(struct pci_dev *);
-int pci_request_region(struct pci_dev *, int, const char *);
+int __must_check pci_request_region(struct pci_dev *, int, const char *);
 void pci_release_region(struct pci_dev *, int);
 
 /* drivers/pci/bus.c */
-int pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
-			   resource_size_t size, resource_size_t align,
-			   resource_size_t min, unsigned int type_mask,
-			   void (*alignf)(void *, struct resource *,
-					  resource_size_t, resource_size_t),
-			   void *alignf_data);
+int __must_check pci_bus_alloc_resource(struct pci_bus *bus,
+			struct resource *res, resource_size_t size,
+			resource_size_t align, resource_size_t min,
+			unsigned int type_mask,
+			void (*alignf)(void *, struct resource *,
+				resource_size_t, resource_size_t),
+			void *alignf_data);
 void pci_enable_bridges(struct pci_bus *bus);
 
 /* Proper probing supporting hot-pluggable devices */
-int __pci_register_driver(struct pci_driver *, struct module *);
-static inline int pci_register_driver(struct pci_driver *driver)
+int __must_check __pci_register_driver(struct pci_driver *, struct module *);
+static inline int __must_check pci_register_driver(struct pci_driver *driver)
 {
 	return __pci_register_driver(driver, THIS_MODULE);
 }
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 95f6db54d8d2..02ad790921fd 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -10,6 +10,7 @@
 #ifndef _SYSFS_H_
 #define _SYSFS_H_
 
+#include <linux/compiler.h>
 #include <asm/atomic.h>
 
 struct kobject;
@@ -86,37 +87,39 @@ struct sysfs_dirent {
 
 #ifdef CONFIG_SYSFS
 
-extern int
+extern int __must_check
 sysfs_create_dir(struct kobject *);
 
 extern void
 sysfs_remove_dir(struct kobject *);
 
-extern int
+extern int __must_check
 sysfs_rename_dir(struct kobject *, const char *new_name);
 
-extern int
+extern int __must_check
 sysfs_create_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_update_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_chmod_file(struct kobject *kobj, struct attribute *attr, mode_t mode);
 
 extern void
 sysfs_remove_file(struct kobject *, const struct attribute *);
 
-extern int
+extern int __must_check
 sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name);
 
 extern void
 sysfs_remove_link(struct kobject *, const char * name);
 
-int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr);
+int __must_check sysfs_create_bin_file(struct kobject *kobj,
+					struct bin_attribute *attr);
 void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr);
 
-int sysfs_create_group(struct kobject *, const struct attribute_group *);
+int __must_check sysfs_create_group(struct kobject *,
+					const struct attribute_group *);
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
 void sysfs_notify(struct kobject * k, char *dir, char *attr);
 
-- 
cgit v1.2.1


From cebc04ba9aeb3a646cc746300421fc0e5aa4f253 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:18 -0700
Subject: add CONFIG_ENABLE_MUST_CHECK

Those 1500 warnings can be a bit of a pain.  Add a config option to shut them
up.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/compiler.h | 5 +++++
 lib/Kconfig.debug        | 7 +++++++
 2 files changed, 12 insertions(+)

diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 9b4f11094937..060b96112ec6 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -99,6 +99,11 @@ extern void __chk_io_ptr(void __iomem *);
 #define __must_check
 #endif
 
+#ifndef CONFIG_ENABLE_MUST_CHECK
+#undef __must_check
+#define __must_check
+#endif
+
 /*
  * Allow us to avoid 'defined but not used' warnings on functions and data,
  * as well as force them to be emitted to the assembly file.
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 554ee688a9f8..5c114c3f03c5 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -8,6 +8,13 @@ config PRINTK_TIME
 	  operations.  This is useful for identifying long delays
 	  in kernel startup.
 
+config ENABLE_MUST_CHECK
+	bool "Enable __must_check logic"
+	default y
+	help
+	  Enable the __must_check logic in the kernel build.  Disable this to
+	  suppress the "warning: ignoring return value of 'foo', declared with
+	  attribute warn_unused_result" messages.
 
 config MAGIC_SYSRQ
 	bool "Magic SysRq key"
-- 
cgit v1.2.1


From 8a6914ab7c95d471c23b42268aa8e1f55b3d2fdb Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:19 -0700
Subject: v4l-dev2: handle __must_check

We get hundreds of these:

include/media/v4l2-dev.h:348: warning: ignoring return value of 'class_device_create_file', declared with attribute warn_unused_result

Handle it, and propagate the __must_check back a level.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/media/v4l2-dev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/media/v4l2-dev.h b/include/media/v4l2-dev.h
index 810462f8a374..bb495b7f4680 100644
--- a/include/media/v4l2-dev.h
+++ b/include/media/v4l2-dev.h
@@ -341,7 +341,7 @@ extern int video_usercopy(struct inode *inode, struct file *file,
 extern struct video_device* video_devdata(struct file*);
 
 #define to_video_device(cd) container_of(cd, struct video_device, class_dev)
-static inline int
+static inline int __must_check
 video_device_create_file(struct video_device *vfd,
 			 struct class_device_attribute *attr)
 {
-- 
cgit v1.2.1


From 370226449ced358e52d198081120826ef52c166b Mon Sep 17 00:00:00 2001
From: Brian Walsh <brian@walsh.ws>
Date: Mon, 14 Aug 2006 22:43:19 -0700
Subject: drivers/base: Platform notify needs to occur before drivers attach to
 the device

The platform_notify call for Arm and PPC architectures needs to be called
before the driver attaches to the device.  The problem only presents itself
when hotplugging certain devices while the driver is already loaded.

Signed-off-by: Brian Walsh <brian@walsh.ws>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/core.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index bc9f35c81691..b224bb43ff63 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -424,6 +424,10 @@ int device_add(struct device *dev)
 	if ((error = kobject_add(&dev->kobj)))
 		goto Error;
 
+	/* notify platform of device entry */
+	if (platform_notify)
+		platform_notify(dev);
+
 	dev->uevent_attr.attr.name = "uevent";
 	dev->uevent_attr.attr.mode = S_IWUSR;
 	if (dev->driver)
@@ -488,10 +492,6 @@ int device_add(struct device *dev)
 				class_intf->add_dev(dev, class_intf);
 		up(&dev->class->sem);
 	}
-
-	/* notify platform of device entry */
-	if (platform_notify)
-		platform_notify(dev);
  Done:
  	kfree(class_name);
 	put_device(dev);
-- 
cgit v1.2.1


From f86db396ff455ed586751d21816a1ebd431264e5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:20 -0700
Subject: drivers/base: check errors

Add lots of return-value checking.

<pcornelia.huck@de.ibm.com>: fix bus_rescan_devices()]
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/base.h    |   2 +-
 drivers/base/bus.c     | 107 +++++++++++++++++++++++++++++++++----------------
 drivers/base/dd.c      |  37 ++++++++++++-----
 include/linux/device.h |   8 ++--
 4 files changed, 104 insertions(+), 50 deletions(-)

diff --git a/drivers/base/base.h b/drivers/base/base.h
index c3b8dc98b8a7..d26644a59537 100644
--- a/drivers/base/base.h
+++ b/drivers/base/base.h
@@ -16,7 +16,7 @@ extern int cpu_dev_init(void);
 extern int attribute_container_init(void);
 
 extern int bus_add_device(struct device * dev);
-extern void bus_attach_device(struct device * dev);
+extern int bus_attach_device(struct device * dev);
 extern void bus_remove_device(struct device * dev);
 extern struct bus_type *get_bus(struct bus_type * bus);
 extern void put_bus(struct bus_type * bus);
diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 4d22a1d10a1c..aa685a20b649 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -371,12 +371,20 @@ int bus_add_device(struct device * dev)
 	if (bus) {
 		pr_debug("bus %s: add device %s\n", bus->name, dev->bus_id);
 		error = device_add_attrs(bus, dev);
-		if (!error) {
-			sysfs_create_link(&bus->devices.kobj, &dev->kobj, dev->bus_id);
-			sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "subsystem");
-			sysfs_create_link(&dev->kobj, &dev->bus->subsys.kset.kobj, "bus");
-		}
+		if (error)
+			goto out;
+		error = sysfs_create_link(&bus->devices.kobj,
+						&dev->kobj, dev->bus_id);
+		if (error)
+			goto out;
+		error = sysfs_create_link(&dev->kobj,
+				&dev->bus->subsys.kset.kobj, "subsystem");
+		if (error)
+			goto out;
+		error = sysfs_create_link(&dev->kobj,
+				&dev->bus->subsys.kset.kobj, "bus");
 	}
+out:
 	return error;
 }
 
@@ -386,14 +394,19 @@ int bus_add_device(struct device * dev)
  *
  *	- Try to attach to driver.
  */
-void bus_attach_device(struct device * dev)
+int bus_attach_device(struct device * dev)
 {
-	struct bus_type * bus = dev->bus;
+	struct bus_type *bus = dev->bus;
+	int ret = 0;
 
 	if (bus) {
-		device_attach(dev);
-		klist_add_tail(&dev->knode_bus, &bus->klist_devices);
+		ret = device_attach(dev);
+		if (ret >= 0) {
+			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
+			ret = 0;
+		}
 	}
+	return ret;
 }
 
 /**
@@ -455,10 +468,17 @@ static void driver_remove_attrs(struct bus_type * bus, struct device_driver * dr
  * Thanks to drivers making their tables __devinit, we can't allow manual
  * bind and unbind from userspace unless CONFIG_HOTPLUG is enabled.
  */
-static void add_bind_files(struct device_driver *drv)
+static int __must_check add_bind_files(struct device_driver *drv)
 {
-	driver_create_file(drv, &driver_attr_unbind);
-	driver_create_file(drv, &driver_attr_bind);
+	int ret;
+
+	ret = driver_create_file(drv, &driver_attr_unbind);
+	if (ret == 0) {
+		ret = driver_create_file(drv, &driver_attr_bind);
+		if (ret)
+			driver_remove_file(drv, &driver_attr_unbind);
+	}
+	return ret;
 }
 
 static void remove_bind_files(struct device_driver *drv)
@@ -476,7 +496,7 @@ static inline void remove_bind_files(struct device_driver *drv) {}
  *	@drv:	driver.
  *
  */
-int bus_add_driver(struct device_driver * drv)
+int bus_add_driver(struct device_driver *drv)
 {
 	struct bus_type * bus = get_bus(drv->bus);
 	int error = 0;
@@ -484,27 +504,39 @@ int bus_add_driver(struct device_driver * drv)
 	if (bus) {
 		pr_debug("bus %s: add driver %s\n", bus->name, drv->name);
 		error = kobject_set_name(&drv->kobj, "%s", drv->name);
-		if (error) {
-			put_bus(bus);
-			return error;
-		}
+		if (error)
+			goto out_put_bus;
 		drv->kobj.kset = &bus->drivers;
-		if ((error = kobject_register(&drv->kobj))) {
-			put_bus(bus);
-			return error;
-		}
+		if ((error = kobject_register(&drv->kobj)))
+			goto out_put_bus;
 
-		driver_attach(drv);
+		error = driver_attach(drv);
+		if (error)
+			goto out_unregister;
 		klist_add_tail(&drv->knode_bus, &bus->klist_drivers);
 		module_add_driver(drv->owner, drv);
 
-		driver_add_attrs(bus, drv);
-		add_bind_files(drv);
+		error = driver_add_attrs(bus, drv);
+		if (error) {
+			/* How the hell do we get out of this pickle? Give up */
+			printk(KERN_ERR "%s: driver_add_attrs(%s) failed\n",
+				__FUNCTION__, drv->name);
+		}
+		error = add_bind_files(drv);
+		if (error) {
+			/* Ditto */
+			printk(KERN_ERR "%s: add_bind_files(%s) failed\n",
+				__FUNCTION__, drv->name);
+		}
 	}
 	return error;
+out_unregister:
+	kobject_unregister(&drv->kobj);
+out_put_bus:
+	put_bus(bus);
+	return error;
 }
 
-
 /**
  *	bus_remove_driver - delete driver from bus's knowledge.
  *	@drv:	driver.
@@ -530,16 +562,21 @@ void bus_remove_driver(struct device_driver * drv)
 
 
 /* Helper for bus_rescan_devices's iter */
-static int bus_rescan_devices_helper(struct device *dev, void *data)
+static int __must_check bus_rescan_devices_helper(struct device *dev,
+						void *data)
 {
+	int ret = 0;
+
 	if (!dev->driver) {
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
-		device_attach(dev);
+		ret = device_attach(dev);
 		if (dev->parent)
 			up(&dev->parent->sem);
+		if (ret > 0)
+			ret = 0;
 	}
-	return 0;
+	return ret < 0 ? ret : 0;
 }
 
 /**
@@ -550,9 +587,9 @@ static int bus_rescan_devices_helper(struct device *dev, void *data)
  * attached and rescan it against existing drivers to see if it matches
  * any by calling device_attach() for the unbound devices.
  */
-void bus_rescan_devices(struct bus_type * bus)
+int bus_rescan_devices(struct bus_type * bus)
 {
-	bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
+	return bus_for_each_dev(bus, NULL, NULL, bus_rescan_devices_helper);
 }
 
 /**
@@ -564,7 +601,7 @@ void bus_rescan_devices(struct bus_type * bus)
  * to use if probing criteria changed during a devices lifetime and
  * driver attachment should change accordingly.
  */
-void device_reprobe(struct device *dev)
+int device_reprobe(struct device *dev)
 {
 	if (dev->driver) {
 		if (dev->parent)        /* Needed for USB */
@@ -573,14 +610,14 @@ void device_reprobe(struct device *dev)
 		if (dev->parent)
 			up(&dev->parent->sem);
 	}
-
-	bus_rescan_devices_helper(dev, NULL);
+	return bus_rescan_devices_helper(dev, NULL);
 }
 EXPORT_SYMBOL_GPL(device_reprobe);
 
-struct bus_type * get_bus(struct bus_type * bus)
+struct bus_type *get_bus(struct bus_type *bus)
 {
-	return bus ? container_of(subsys_get(&bus->subsys), struct bus_type, subsys) : NULL;
+	return bus ? container_of(subsys_get(&bus->subsys),
+				struct bus_type, subsys) : NULL;
 }
 
 void put_bus(struct bus_type * bus)
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 889c71111239..9f6f11ca0ab6 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -38,17 +38,29 @@
  *
  *	This function must be called with @dev->sem held.
  */
-void device_bind_driver(struct device * dev)
+int device_bind_driver(struct device *dev)
 {
-	if (klist_node_attached(&dev->knode_driver))
-		return;
+	int ret;
+
+	if (klist_node_attached(&dev->knode_driver)) {
+		printk(KERN_WARNING "%s: device %s already bound\n",
+			__FUNCTION__, kobject_name(&dev->kobj));
+		return 0;
+	}
 
 	pr_debug("bound device '%s' to driver '%s'\n",
 		 dev->bus_id, dev->driver->name);
 	klist_add_tail(&dev->knode_driver, &dev->driver->klist_devices);
-	sysfs_create_link(&dev->driver->kobj, &dev->kobj,
+	ret = sysfs_create_link(&dev->driver->kobj, &dev->kobj,
 			  kobject_name(&dev->kobj));
-	sysfs_create_link(&dev->kobj, &dev->driver->kobj, "driver");
+	if (ret == 0) {
+		ret = sysfs_create_link(&dev->kobj, &dev->driver->kobj,
+					"driver");
+		if (ret)
+			sysfs_remove_link(&dev->driver->kobj,
+					kobject_name(&dev->kobj));
+	}
+	return ret;
 }
 
 /**
@@ -91,7 +103,11 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 			goto ProbeFailed;
 		}
 	}
-	device_bind_driver(dev);
+	if (device_bind_driver(dev)) {
+		printk(KERN_ERR "%s: device_bind_driver(%s) failed\n",
+			__FUNCTION__, dev->bus_id);
+		/* How does undo a ->probe?  We're screwed. */
+	}
 	ret = 1;
 	pr_debug("%s: Bound Device %s to Driver %s\n",
 		 drv->bus->name, dev->bus_id, drv->name);
@@ -139,8 +155,9 @@ int device_attach(struct device * dev)
 
 	down(&dev->sem);
 	if (dev->driver) {
-		device_bind_driver(dev);
-		ret = 1;
+		ret = device_bind_driver(dev);
+		if (ret == 0)
+			ret = 1;
 	} else
 		ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach);
 	up(&dev->sem);
@@ -182,9 +199,9 @@ static int __driver_attach(struct device * dev, void * data)
  *	returns 0 and the @dev->driver is set, we've found a
  *	compatible pair.
  */
-void driver_attach(struct device_driver * drv)
+int driver_attach(struct device_driver * drv)
 {
-	bus_for_each_dev(drv->bus, NULL, drv, __driver_attach);
+	return bus_for_each_dev(drv->bus, NULL, drv, __driver_attach);
 }
 
 /**
diff --git a/include/linux/device.h b/include/linux/device.h
index ad4db72f5733..b3da9a870cfc 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -62,7 +62,7 @@ struct bus_type {
 extern int __must_check bus_register(struct bus_type * bus);
 extern void bus_unregister(struct bus_type * bus);
 
-extern void bus_rescan_devices(struct bus_type * bus);
+extern int __must_check bus_rescan_devices(struct bus_type * bus);
 
 /* iterator helpers for buses */
 
@@ -397,11 +397,11 @@ extern int device_rename(struct device *dev, char *new_name);
  * Manual binding of a device to driver. See drivers/base/bus.c
  * for information on use.
  */
-extern void device_bind_driver(struct device * dev);
+extern int __must_check device_bind_driver(struct device *dev);
 extern void device_release_driver(struct device * dev);
 extern int  __must_check device_attach(struct device * dev);
-extern void driver_attach(struct device_driver * drv);
-extern void device_reprobe(struct device *dev);
+extern int __must_check driver_attach(struct device_driver *drv);
+extern int __must_check device_reprobe(struct device *dev);
 
 /*
  * Easy functions for dynamically creating devices on the fly
-- 
cgit v1.2.1


From f20a9ead0d005fbeeae3fc21a96f9bf197ac1c1c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 22:43:23 -0700
Subject: sysfs: add proper sysfs_init() prototype

Don't be crufty.  Mark it __must_check too.

Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 fs/namespace.c        | 10 +---------
 include/linux/sysfs.h |  7 +++++++
 2 files changed, 8 insertions(+), 9 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index fa7ed6a9fc2d..36d180858136 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -17,6 +17,7 @@
 #include <linux/acct.h>
 #include <linux/capability.h>
 #include <linux/module.h>
+#include <linux/sysfs.h>
 #include <linux/seq_file.h>
 #include <linux/namespace.h>
 #include <linux/namei.h>
@@ -28,15 +29,6 @@
 
 extern int __init init_rootfs(void);
 
-#ifdef CONFIG_SYSFS
-extern int __init sysfs_init(void);
-#else
-static inline int sysfs_init(void)
-{
-	return 0;
-}
-#endif
-
 /* spinlock for vfsmount related operations, inplace of dcache_lock */
 __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock);
 
diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
index 02ad790921fd..6d5c43d31dec 100644
--- a/include/linux/sysfs.h
+++ b/include/linux/sysfs.h
@@ -123,6 +123,8 @@ int __must_check sysfs_create_group(struct kobject *,
 void sysfs_remove_group(struct kobject *, const struct attribute_group *);
 void sysfs_notify(struct kobject * k, char *dir, char *attr);
 
+extern int __must_check sysfs_init(void);
+
 #else /* CONFIG_SYSFS */
 
 static inline int sysfs_create_dir(struct kobject * k)
@@ -194,6 +196,11 @@ static inline void sysfs_notify(struct kobject * k, char *dir, char *attr)
 {
 }
 
+static inline int __must_check sysfs_init(void)
+{
+	return 0;
+}
+
 #endif /* CONFIG_SYSFS */
 
 #endif /* _SYSFS_H_ */
-- 
cgit v1.2.1


From d779249ed4cb3b50690de6de8448829d65a1cd08 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 18 Jul 2006 10:59:59 -0700
Subject: Driver Core: add ability for drivers to do a threaded probe

This adds the infrastructure for drivers to do a threaded probe, and
waits at init time for all currently outstanding probes to complete.

A new kernel thread will be created when the probe() function for the
driver is called, if the multithread_probe bit is set in the driver
saying it can support this kind of operation.

I have tested this with USB and PCI, and it works, and shaves off a lot
of time in the boot process, but there are issues with finding root boot
disks, and some USB drivers assume that this can never happen, so it is
currently not enabled for any bus type.  Individual drivers can enable
this right now if they wish, and bus authors can selectivly turn it on
as well, once they determine that their subsystem will work properly
with it.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/dd.c      | 108 ++++++++++++++++++++++++++++++++++++-------------
 include/linux/device.h |   3 ++
 init/do_mounts.c       |   5 +++
 3 files changed, 89 insertions(+), 27 deletions(-)

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 9f6f11ca0ab6..319a73be4180 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -17,6 +17,7 @@
 
 #include <linux/device.h>
 #include <linux/module.h>
+#include <linux/kthread.h>
 
 #include "base.h"
 #include "power/power.h"
@@ -63,44 +64,35 @@ int device_bind_driver(struct device *dev)
 	return ret;
 }
 
-/**
- *	driver_probe_device - attempt to bind device & driver.
- *	@drv:	driver.
- *	@dev:	device.
- *
- *	First, we call the bus's match function, if one present, which
- *	should compare the device IDs the driver supports with the
- *	device IDs of the device. Note we don't do this ourselves
- *	because we don't know the format of the ID structures, nor what
- *	is to be considered a match and what is not.
- *
- *	This function returns 1 if a match is found, an error if one
- *	occurs (that is not -ENODEV or -ENXIO), and 0 otherwise.
- *
- *	This function must be called with @dev->sem held.  When called
- *	for a USB interface, @dev->parent->sem must be held as well.
- */
-int driver_probe_device(struct device_driver * drv, struct device * dev)
+struct stupid_thread_structure {
+	struct device_driver *drv;
+	struct device *dev;
+};
+
+static atomic_t probe_count = ATOMIC_INIT(0);
+static int really_probe(void *void_data)
 {
+	struct stupid_thread_structure *data = void_data;
+	struct device_driver *drv = data->drv;
+	struct device *dev = data->dev;
 	int ret = 0;
 
-	if (drv->bus->match && !drv->bus->match(dev, drv))
-		goto Done;
+	atomic_inc(&probe_count);
+	pr_debug("%s: Probing driver %s with device %s\n",
+		 drv->bus->name, drv->name, dev->bus_id);
 
-	pr_debug("%s: Matched Device %s with Driver %s\n",
-		 drv->bus->name, dev->bus_id, drv->name);
 	dev->driver = drv;
 	if (dev->bus->probe) {
 		ret = dev->bus->probe(dev);
 		if (ret) {
 			dev->driver = NULL;
-			goto ProbeFailed;
+			goto probe_failed;
 		}
 	} else if (drv->probe) {
 		ret = drv->probe(dev);
 		if (ret) {
 			dev->driver = NULL;
-			goto ProbeFailed;
+			goto probe_failed;
 		}
 	}
 	if (device_bind_driver(dev)) {
@@ -111,9 +103,9 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 	ret = 1;
 	pr_debug("%s: Bound Device %s to Driver %s\n",
 		 drv->bus->name, dev->bus_id, drv->name);
-	goto Done;
+	goto done;
 
- ProbeFailed:
+probe_failed:
 	if (ret == -ENODEV || ret == -ENXIO) {
 		/* Driver matched, but didn't support device
 		 * or device not found.
@@ -126,7 +118,69 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 		       "%s: probe of %s failed with error %d\n",
 		       drv->name, dev->bus_id, ret);
 	}
- Done:
+done:
+	kfree(data);
+	atomic_dec(&probe_count);
+	return ret;
+}
+
+/**
+ * driver_probe_done
+ * Determine if the probe sequence is finished or not.
+ *
+ * Should somehow figure out how to use a semaphore, not an atomic variable...
+ */
+int driver_probe_done(void)
+{
+	pr_debug("%s: probe_count = %d\n", __FUNCTION__,
+		 atomic_read(&probe_count));
+	if (atomic_read(&probe_count))
+		return -EBUSY;
+	return 0;
+}
+
+/**
+ * driver_probe_device - attempt to bind device & driver together
+ * @drv: driver to bind a device to
+ * @dev: device to try to bind to the driver
+ *
+ * First, we call the bus's match function, if one present, which should
+ * compare the device IDs the driver supports with the device IDs of the
+ * device. Note we don't do this ourselves because we don't know the
+ * format of the ID structures, nor what is to be considered a match and
+ * what is not.
+ *
+ * This function returns 1 if a match is found, an error if one occurs
+ * (that is not -ENODEV or -ENXIO), and 0 otherwise.
+ *
+ * This function must be called with @dev->sem held.  When called for a
+ * USB interface, @dev->parent->sem must be held as well.
+ */
+int driver_probe_device(struct device_driver * drv, struct device * dev)
+{
+	struct stupid_thread_structure *data;
+	struct task_struct *probe_task;
+	int ret = 0;
+
+	if (drv->bus->match && !drv->bus->match(dev, drv))
+		goto done;
+
+	pr_debug("%s: Matched Device %s with Driver %s\n",
+		 drv->bus->name, dev->bus_id, drv->name);
+
+	data = kmalloc(sizeof(*data), GFP_KERNEL);
+	data->drv = drv;
+	data->dev = dev;
+
+	if (drv->multithread_probe) {
+		probe_task = kthread_run(really_probe, data,
+					 "probe-%s", dev->bus_id);
+		if (IS_ERR(probe_task))
+			ret = PTR_ERR(probe_task);
+	} else
+		ret = really_probe(data);
+
+done:
 	return ret;
 }
 
diff --git a/include/linux/device.h b/include/linux/device.h
index b3da9a870cfc..74246efba931 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -106,6 +106,8 @@ struct device_driver {
 	void	(*shutdown)	(struct device * dev);
 	int	(*suspend)	(struct device * dev, pm_message_t state);
 	int	(*resume)	(struct device * dev);
+
+	unsigned int multithread_probe:1;
 };
 
 
@@ -115,6 +117,7 @@ extern void driver_unregister(struct device_driver * drv);
 extern struct device_driver * get_driver(struct device_driver * drv);
 extern void put_driver(struct device_driver * drv);
 extern struct device_driver *driver_find(const char *name, struct bus_type *bus);
+extern int driver_probe_done(void);
 
 /* driverfs interface for exporting driver attributes */
 
diff --git a/init/do_mounts.c b/init/do_mounts.c
index 94aeec7aa917..b290aadb1d3f 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -8,6 +8,7 @@
 #include <linux/security.h>
 #include <linux/delay.h>
 #include <linux/mount.h>
+#include <linux/device.h>
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_fs_sb.h>
@@ -403,6 +404,10 @@ void __init prepare_namespace(void)
 		ssleep(root_delay);
 	}
 
+	/* wait for the known devices to complete their probing */
+	while (driver_probe_done() != 0)
+		msleep(100);
+
 	md_run_setup();
 
 	if (saved_root_name[0]) {
-- 
cgit v1.2.1


From 0f397f865076e3471ec884ee73ad5e34165fac2a Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Tue, 18 Jul 2006 10:59:59 -0700
Subject: PCI: enable driver multi-threaded probe

This provides a build and run-time option to turn on multhreaded probe
for all PCI drivers.  It can cause bad problems on multi-processor
machines that take a while to find their root disks, and play havoc on
machines that don't use persistant device names for block or network
devices.

But it can cause speedups on some machines, my tiny laptop's boot goes
up by 0.4 seconds, and my desktop boots up several seconds faster.

Use at your own risk!!!

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/Kconfig      | 25 +++++++++++++++++++++++++
 drivers/pci/pci-driver.c | 11 +++++++++++
 2 files changed, 36 insertions(+)

diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig
index 4d762fc4878c..c27e782e6df9 100644
--- a/drivers/pci/Kconfig
+++ b/drivers/pci/Kconfig
@@ -17,6 +17,31 @@ config PCI_MSI
 
 	   If you don't know what to do here, say N.
 
+config PCI_MULTITHREAD_PROBE
+	bool "PCI Multi-threaded probe (EXPERIMENTAL)"
+	depends on PCI && EXPERIMENTAL
+	help
+	  Say Y here if you want the PCI core to spawn a new thread for
+	  every PCI device that is probed.  This can cause a huge
+	  speedup in boot times on multiprocessor machines, and even a
+	  smaller speedup on single processor machines.
+
+	  But it can also cause lots of bad things to happen.  A number
+	  of PCI drivers can not properly handle running in this way,
+	  some will just not work properly at all, while others might
+	  decide to blow up power supplies with a huge load all at once,
+	  so use this option at your own risk.
+
+	  It is very unwise to use this option if you are not using a
+	  boot process that can handle devices being created in any
+	  order.  A program that can create persistant block and network
+	  device names (like udev) is a good idea if you wish to use
+	  this option.
+
+	  Again, use this option at your own risk, you have been warned!
+
+	  When in doubt, say N.
+
 config PCI_DEBUG
 	bool "PCI Debugging"
 	depends on PCI && DEBUG_KERNEL
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 8948ac9ab681..d8ace1f90dd2 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -17,6 +17,16 @@
  *  Registration of PCI drivers and handling of hot-pluggable devices.
  */
 
+/* multithreaded probe logic */
+static int pci_multithread_probe =
+#ifdef CONFIG_PCI_MULTITHREAD_PROBE
+	1;
+#else
+	0;
+#endif
+__module_param_call("", pci_multithread_probe, param_set_bool, param_get_bool, &pci_multithread_probe, 0644);
+
+
 /*
  * Dynamic device IDs are disabled for !CONFIG_HOTPLUG
  */
@@ -408,6 +418,7 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner)
 	drv->driver.bus = &pci_bus_type;
 	drv->driver.owner = owner;
 	drv->driver.kobj.ktype = &pci_driver_kobj_type;
+	drv->driver.multithread_probe = pci_multithread_probe;
 
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
-- 
cgit v1.2.1


From f2eaae197f4590c4d96f31b09b0ee9067421a95c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 18 Sep 2006 16:22:34 -0400
Subject: Driver core: Fix potential deadlock in driver core

There is a potential deadlock in the driver core.  It boils down to
the fact that bus_remove_device() calls klist_remove() instead of
klist_del(), thereby waiting until the reference count of the
klist_node in the bus's klist of devices drops to 0.  The refcount
can't reach 0 so long as a modprobe process is trying to bind a new
driver to the device being removed, by calling __driver_attach().  The
problem is that __driver_attach() tries to acquire the device's
parent's semaphore, but the caller of bus_remove_device() is quite
likely to own that semaphore already.

It isn't sufficient just to replace klist_remove() with klist_del().
Doing so runs the risk that the device would remain on the bus's klist
of devices for some time, and so could be bound to another driver even
after it was unregistered.  What's needed is a new way to distinguish
whether or not a device is registered, based on a criterion other than
whether its klist_node is linked into the bus's klist of devices.  That
way driver binding can fail when the device is unregistered, even if
it is still linked into the klist.

This patch (as782) implements the solution, by adding a new bitflag to
indiate when a struct device is registered, by testing the flag before
allowing a driver to bind a device, and by changing the definition of
the device_is_registered() inline.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c     | 8 ++++++--
 drivers/base/dd.c      | 2 ++
 include/linux/device.h | 3 ++-
 3 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index aa685a20b649..636af538a2b5 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -392,6 +392,7 @@ out:
  *	bus_attach_device - add device to bus
  *	@dev:	device tried to attach to a driver
  *
+ *	- Add device to bus's list of devices.
  *	- Try to attach to driver.
  */
 int bus_attach_device(struct device * dev)
@@ -400,11 +401,13 @@ int bus_attach_device(struct device * dev)
 	int ret = 0;
 
 	if (bus) {
+		dev->is_registered = 1;
 		ret = device_attach(dev);
 		if (ret >= 0) {
 			klist_add_tail(&dev->knode_bus, &bus->klist_devices);
 			ret = 0;
-		}
+		} else
+			dev->is_registered = 0;
 	}
 	return ret;
 }
@@ -425,7 +428,8 @@ void bus_remove_device(struct device * dev)
 		sysfs_remove_link(&dev->kobj, "bus");
 		sysfs_remove_link(&dev->bus->devices.kobj, dev->bus_id);
 		device_remove_attrs(dev->bus, dev);
-		klist_remove(&dev->knode_bus);
+		dev->is_registered = 0;
+		klist_del(&dev->knode_bus);
 		pr_debug("bus %s: remove device %s\n", dev->bus->name, dev->bus_id);
 		device_release_driver(dev);
 		put_bus(dev->bus);
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index 319a73be4180..b5f43c3e44fa 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -162,6 +162,8 @@ int driver_probe_device(struct device_driver * drv, struct device * dev)
 	struct task_struct *probe_task;
 	int ret = 0;
 
+	if (!device_is_registered(dev))
+		return -ENODEV;
 	if (drv->bus->match && !drv->bus->match(dev, drv))
 		goto done;
 
diff --git a/include/linux/device.h b/include/linux/device.h
index 74246efba931..662e6a10144e 100644
--- a/include/linux/device.h
+++ b/include/linux/device.h
@@ -329,6 +329,7 @@ struct device {
 
 	struct kobject kobj;
 	char	bus_id[BUS_ID_SIZE];	/* position on parent bus */
+	unsigned		is_registered:1;
 	struct device_attribute uevent_attr;
 	struct device_attribute *devt_attr;
 
@@ -381,7 +382,7 @@ dev_set_drvdata (struct device *dev, void *data)
 
 static inline int device_is_registered(struct device *dev)
 {
-	return klist_node_attached(&dev->knode_bus);
+	return dev->is_registered;
 }
 
 /*
-- 
cgit v1.2.1


From 81107bf531d2524afbcd61f3b4ad57a71295d591 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 18 Sep 2006 16:24:28 -0400
Subject: Driver core: Remove unneeded routines from driver core

This patch (as783) simplifies the driver core slightly by removing four
unnecessary _get and _put methods.

It is vital that when a driver is removed from its bus's klist of
registered drivers, or when a device is removed from a driver's klist
of bound devices, that the klist updates complete synchronously.
Otherwise the kernel might try binding an unregistered driver to a
newly-registered device, or adding a device to the klist for a new
driver before it has been removed from the old driver's klist.

Since the removals must be synchronous, they don't need to update any
reference counts.  Hence the _get and _put methods can be dispensed
with.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/base/bus.c    | 18 +-----------------
 drivers/base/driver.c | 16 +---------------
 2 files changed, 2 insertions(+), 32 deletions(-)

diff --git a/drivers/base/bus.c b/drivers/base/bus.c
index 636af538a2b5..12173d16bea7 100644
--- a/drivers/base/bus.c
+++ b/drivers/base/bus.c
@@ -696,22 +696,6 @@ static void klist_devices_put(struct klist_node *n)
 	put_device(dev);
 }
 
-static void klist_drivers_get(struct klist_node *n)
-{
-	struct device_driver *drv = container_of(n, struct device_driver,
-						 knode_bus);
-
-	get_driver(drv);
-}
-
-static void klist_drivers_put(struct klist_node *n)
-{
-	struct device_driver *drv = container_of(n, struct device_driver,
-						 knode_bus);
-
-	put_driver(drv);
-}
-
 /**
  *	bus_register - register a bus with the system.
  *	@bus:	bus.
@@ -747,7 +731,7 @@ int bus_register(struct bus_type * bus)
 		goto bus_drivers_fail;
 
 	klist_init(&bus->klist_devices, klist_devices_get, klist_devices_put);
-	klist_init(&bus->klist_drivers, klist_drivers_get, klist_drivers_put);
+	klist_init(&bus->klist_drivers, NULL, NULL);
 	bus_add_attrs(bus);
 
 	pr_debug("bus type '%s' registered\n", bus->name);
diff --git a/drivers/base/driver.c b/drivers/base/driver.c
index 562600dd540a..1214cbd17d86 100644
--- a/drivers/base/driver.c
+++ b/drivers/base/driver.c
@@ -142,20 +142,6 @@ void put_driver(struct device_driver * drv)
 	kobject_put(&drv->kobj);
 }
 
-static void klist_devices_get(struct klist_node *n)
-{
-	struct device *dev = container_of(n, struct device, knode_driver);
-
-	get_device(dev);
-}
-
-static void klist_devices_put(struct klist_node *n)
-{
-	struct device *dev = container_of(n, struct device, knode_driver);
-
-	put_device(dev);
-}
-
 /**
  *	driver_register - register driver with bus
  *	@drv:	driver to register
@@ -175,7 +161,7 @@ int driver_register(struct device_driver * drv)
 	    (drv->bus->shutdown && drv->shutdown)) {
 		printk(KERN_WARNING "Driver '%s' needs updating - please use bus_type methods\n", drv->name);
 	}
-	klist_init(&drv->klist_devices, klist_devices_get, klist_devices_put);
+	klist_init(&drv->klist_devices, NULL, NULL);
 	init_completion(&drv->unloaded);
 	return bus_add_driver(drv);
 }
-- 
cgit v1.2.1


From 7e9f4b2d3e21e87c26025810413ef1592834e63b Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 18 Sep 2006 16:28:06 -0400
Subject: Driver core: Don't call put methods while holding a spinlock

The klist utility routines currently call _put methods while holding a
spinlock.  This is of course illegal; a put routine could try to
unregister a device and hence need to sleep.

No problems have arisen until now because in many cases klist removals
were done synchronously, so the _put methods were never actually used.
In other cases we may simply have been lucky.

This patch (as784) reworks the klist routines so that _put methods are
called only _after_ the klist's spinlock has been released.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 lib/klist.c | 26 +++++++++++++++-----------
 1 file changed, 15 insertions(+), 11 deletions(-)

diff --git a/lib/klist.c b/lib/klist.c
index 9c94f0b163a1..120bd175aa78 100644
--- a/lib/klist.c
+++ b/lib/klist.c
@@ -123,12 +123,10 @@ EXPORT_SYMBOL_GPL(klist_add_tail);
 static void klist_release(struct kref * kref)
 {
 	struct klist_node * n = container_of(kref, struct klist_node, n_ref);
-	void (*put)(struct klist_node *) = n->n_klist->put;
+
 	list_del(&n->n_node);
 	complete(&n->n_removed);
 	n->n_klist = NULL;
-	if (put)
-		put(n);
 }
 
 static int klist_dec_and_del(struct klist_node * n)
@@ -145,10 +143,14 @@ static int klist_dec_and_del(struct klist_node * n)
 void klist_del(struct klist_node * n)
 {
 	struct klist * k = n->n_klist;
+	void (*put)(struct klist_node *) = k->put;
 
 	spin_lock(&k->k_lock);
-	klist_dec_and_del(n);
+	if (!klist_dec_and_del(n))
+		put = NULL;
 	spin_unlock(&k->k_lock);
+	if (put)
+		put(n);
 }
 
 EXPORT_SYMBOL_GPL(klist_del);
@@ -161,10 +163,7 @@ EXPORT_SYMBOL_GPL(klist_del);
 
 void klist_remove(struct klist_node * n)
 {
-	struct klist * k = n->n_klist;
-	spin_lock(&k->k_lock);
-	klist_dec_and_del(n);
-	spin_unlock(&k->k_lock);
+	klist_del(n);
 	wait_for_completion(&n->n_removed);
 }
 
@@ -260,12 +259,15 @@ static struct klist_node * to_klist_node(struct list_head * n)
 struct klist_node * klist_next(struct klist_iter * i)
 {
 	struct list_head * next;
+	struct klist_node * lnode = i->i_cur;
 	struct klist_node * knode = NULL;
+	void (*put)(struct klist_node *) = i->i_klist->put;
 
 	spin_lock(&i->i_klist->k_lock);
-	if (i->i_cur) {
-		next = i->i_cur->n_node.next;
-		klist_dec_and_del(i->i_cur);
+	if (lnode) {
+		next = lnode->n_node.next;
+		if (!klist_dec_and_del(lnode))
+			put = NULL;
 	} else
 		next = i->i_head->next;
 
@@ -275,6 +277,8 @@ struct klist_node * klist_next(struct klist_iter * i)
 	}
 	i->i_cur = knode;
 	spin_unlock(&i->i_klist->k_lock);
+	if (put && lnode)
+		put(lnode);
 	return knode;
 }
 
-- 
cgit v1.2.1


From 265baba316ea258ca015aa79bc6f107cd9fce2b3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] Update defconfig

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/defconfig | 109 ++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 78 insertions(+), 31 deletions(-)

diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index 5fb970715941..647610ecb580 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc4
-# Thu Aug 24 21:05:55 2006
+# Linux kernel version: 2.6.18-git5
+# Tue Sep 26 09:30:47 2006
 #
 CONFIG_X86_64=y
 CONFIG_64BIT=y
@@ -19,6 +19,7 @@ CONFIG_GENERIC_ISA_DMA=y
 CONFIG_GENERIC_IOMAP=y
 CONFIG_ARCH_MAY_HAVE_PC_FDC=y
 CONFIG_DMI=y
+CONFIG_AUDIT_ARCH=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
@@ -38,16 +39,16 @@ CONFIG_SYSVIPC=y
 CONFIG_POSIX_MQUEUE=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_TASKSTATS is not set
-CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 # CONFIG_CPUSETS is not set
 # CONFIG_RELAY is not set
 CONFIG_INITRAMFS_SOURCE=""
-CONFIG_UID16=y
 CONFIG_CC_OPTIMIZE_FOR_SIZE=y
 # CONFIG_EMBEDDED is not set
+CONFIG_UID16=y
+CONFIG_SYSCTL=y
 CONFIG_KALLSYMS=y
 CONFIG_KALLSYMS_ALL=y
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
@@ -56,12 +57,12 @@ CONFIG_PRINTK=y
 CONFIG_BUG=y
 CONFIG_ELF_CORE=y
 CONFIG_BASE_FULL=y
-CONFIG_RT_MUTEXES=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
 CONFIG_SHMEM=y
 CONFIG_SLAB=y
 CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_RT_MUTEXES=y
 # CONFIG_TINY_SHMEM is not set
 CONFIG_BASE_SMALL=0
 # CONFIG_SLOB is not set
@@ -160,6 +161,7 @@ CONFIG_X86_MCE_AMD=y
 # CONFIG_CRASH_DUMP is not set
 CONFIG_PHYSICAL_START=0x200000
 CONFIG_SECCOMP=y
+# CONFIG_CC_STACKPROTECTOR is not set
 # CONFIG_HZ_100 is not set
 CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
@@ -307,18 +309,23 @@ CONFIG_IP_PNP_DHCP=y
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_BIC=y
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
 CONFIG_IPV6=y
 # CONFIG_IPV6_PRIVACY is not set
 # CONFIG_IPV6_ROUTER_PREF is not set
 # CONFIG_INET6_AH is not set
 # CONFIG_INET6_ESP is not set
 # CONFIG_INET6_IPCOMP is not set
+# CONFIG_IPV6_MIP6 is not set
 # CONFIG_INET6_XFRM_TUNNEL is not set
 # CONFIG_INET6_TUNNEL is not set
 # CONFIG_INET6_XFRM_MODE_TRANSPORT is not set
 # CONFIG_INET6_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set
 # CONFIG_IPV6_TUNNEL is not set
+# CONFIG_IPV6_SUBTREES is not set
+# CONFIG_IPV6_MULTIPLE_TABLES is not set
 # CONFIG_NETWORK_SECMARK is not set
 # CONFIG_NETFILTER is not set
 
@@ -345,7 +352,6 @@ CONFIG_IPV6=y
 # CONFIG_ATALK is not set
 # CONFIG_X25 is not set
 # CONFIG_LAPB is not set
-# CONFIG_NET_DIVERT is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
 
@@ -487,6 +493,7 @@ CONFIG_IDEDMA_AUTO=y
 #
 # CONFIG_RAID_ATTRS is not set
 CONFIG_SCSI=y
+CONFIG_SCSI_NETLINK=y
 # CONFIG_SCSI_PROC_FS is not set
 
 #
@@ -508,12 +515,13 @@ CONFIG_SCSI_CONSTANTS=y
 # CONFIG_SCSI_LOGGING is not set
 
 #
-# SCSI Transport Attributes
+# SCSI Transports
 #
 CONFIG_SCSI_SPI_ATTRS=y
 CONFIG_SCSI_FC_ATTRS=y
 # CONFIG_SCSI_ISCSI_ATTRS is not set
 CONFIG_SCSI_SAS_ATTRS=y
+# CONFIG_SCSI_SAS_LIBSAS is not set
 
 #
 # SCSI low-level drivers
@@ -532,29 +540,14 @@ CONFIG_AIC79XX_RESET_DELAY_MS=4000
 # CONFIG_AIC79XX_DEBUG_ENABLE is not set
 CONFIG_AIC79XX_DEBUG_MASK=0
 # CONFIG_AIC79XX_REG_PRETTY_PRINT is not set
+# CONFIG_SCSI_AIC94XX is not set
+# CONFIG_SCSI_ARCMSR is not set
 CONFIG_MEGARAID_NEWGEN=y
 CONFIG_MEGARAID_MM=y
 CONFIG_MEGARAID_MAILBOX=y
 # CONFIG_MEGARAID_LEGACY is not set
 CONFIG_MEGARAID_SAS=y
-CONFIG_SCSI_SATA=y
-CONFIG_SCSI_SATA_AHCI=y
-CONFIG_SCSI_SATA_SVW=y
-CONFIG_SCSI_ATA_PIIX=y
-# CONFIG_SCSI_SATA_MV is not set
-CONFIG_SCSI_SATA_NV=y
-# CONFIG_SCSI_PDC_ADMA is not set
 # CONFIG_SCSI_HPTIOP is not set
-# CONFIG_SCSI_SATA_QSTOR is not set
-# CONFIG_SCSI_SATA_PROMISE is not set
-# CONFIG_SCSI_SATA_SX4 is not set
-CONFIG_SCSI_SATA_SIL=y
-# CONFIG_SCSI_SATA_SIL24 is not set
-# CONFIG_SCSI_SATA_SIS is not set
-# CONFIG_SCSI_SATA_ULI is not set
-CONFIG_SCSI_SATA_VIA=y
-# CONFIG_SCSI_SATA_VITESSE is not set
-CONFIG_SCSI_SATA_INTEL_COMBINED=y
 # CONFIG_SCSI_BUSLOGIC is not set
 # CONFIG_SCSI_DMX3191D is not set
 # CONFIG_SCSI_EATA is not set
@@ -563,6 +556,7 @@ CONFIG_SCSI_SATA_INTEL_COMBINED=y
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_STEX is not set
 # CONFIG_SCSI_SYM53C8XX_2 is not set
 # CONFIG_SCSI_IPR is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
@@ -572,6 +566,62 @@ CONFIG_SCSI_SATA_INTEL_COMBINED=y
 # CONFIG_SCSI_DC390T is not set
 # CONFIG_SCSI_DEBUG is not set
 
+#
+# Serial ATA (prod) and Parallel ATA (experimental) drivers
+#
+CONFIG_ATA=y
+CONFIG_SATA_AHCI=y
+CONFIG_SATA_SVW=y
+CONFIG_ATA_PIIX=y
+# CONFIG_SATA_MV is not set
+CONFIG_SATA_NV=y
+# CONFIG_PDC_ADMA is not set
+# CONFIG_SATA_QSTOR is not set
+# CONFIG_SATA_PROMISE is not set
+# CONFIG_SATA_SX4 is not set
+CONFIG_SATA_SIL=y
+# CONFIG_SATA_SIL24 is not set
+# CONFIG_SATA_SIS is not set
+# CONFIG_SATA_ULI is not set
+CONFIG_SATA_VIA=y
+# CONFIG_SATA_VITESSE is not set
+CONFIG_SATA_INTEL_COMBINED=y
+# CONFIG_PATA_ALI is not set
+# CONFIG_PATA_AMD is not set
+# CONFIG_PATA_ARTOP is not set
+# CONFIG_PATA_ATIIXP is not set
+# CONFIG_PATA_CMD64X is not set
+# CONFIG_PATA_CS5520 is not set
+# CONFIG_PATA_CS5530 is not set
+# CONFIG_PATA_CYPRESS is not set
+# CONFIG_PATA_EFAR is not set
+# CONFIG_ATA_GENERIC is not set
+# CONFIG_PATA_HPT366 is not set
+# CONFIG_PATA_HPT37X is not set
+# CONFIG_PATA_HPT3X2N is not set
+# CONFIG_PATA_HPT3X3 is not set
+# CONFIG_PATA_IT821X is not set
+# CONFIG_PATA_JMICRON is not set
+# CONFIG_PATA_LEGACY is not set
+# CONFIG_PATA_TRIFLEX is not set
+# CONFIG_PATA_MPIIX is not set
+# CONFIG_PATA_OLDPIIX is not set
+# CONFIG_PATA_NETCELL is not set
+# CONFIG_PATA_NS87410 is not set
+# CONFIG_PATA_OPTI is not set
+# CONFIG_PATA_OPTIDMA is not set
+# CONFIG_PATA_PDC_OLD is not set
+# CONFIG_PATA_QDI is not set
+# CONFIG_PATA_RADISYS is not set
+# CONFIG_PATA_RZ1000 is not set
+# CONFIG_PATA_SC1200 is not set
+# CONFIG_PATA_SERVERWORKS is not set
+# CONFIG_PATA_PDC2027X is not set
+# CONFIG_PATA_SIL680 is not set
+# CONFIG_PATA_SIS is not set
+# CONFIG_PATA_VIA is not set
+# CONFIG_PATA_WINBOND is not set
+
 #
 # Multi-device support (RAID and LVM)
 #
@@ -678,6 +728,7 @@ CONFIG_NET_PCI=y
 # CONFIG_ADAPTEC_STARFIRE is not set
 CONFIG_B44=y
 CONFIG_FORCEDETH=y
+# CONFIG_FORCEDETH_NAPI is not set
 # CONFIG_DGRS is not set
 # CONFIG_EEPRO100 is not set
 CONFIG_E100=y
@@ -714,6 +765,7 @@ CONFIG_E1000=y
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
 CONFIG_BNX2=y
+# CONFIG_QLA3XXX is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -1036,6 +1088,7 @@ CONFIG_SOUND=y
 # Open Sound System
 #
 CONFIG_SOUND_PRIME=y
+CONFIG_OSS_OBSOLETE_DRIVER=y
 # CONFIG_SOUND_BT878 is not set
 # CONFIG_SOUND_EMU10K1 is not set
 # CONFIG_SOUND_FUSION is not set
@@ -1046,7 +1099,6 @@ CONFIG_SOUND_ICH=y
 # CONFIG_SOUND_MSNDPIN is not set
 # CONFIG_SOUND_VIA82CXXX is not set
 # CONFIG_SOUND_OSS is not set
-# CONFIG_SOUND_TVMIXER is not set
 
 #
 # USB support
@@ -1203,7 +1255,6 @@ CONFIG_USB_MON=y
 # InfiniBand support
 #
 # CONFIG_INFINIBAND is not set
-# CONFIG_IPATH_CORE is not set
 
 #
 # EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
@@ -1448,10 +1499,6 @@ CONFIG_DEBUG_STACKOVERFLOW=y
 #
 # CONFIG_CRYPTO is not set
 
-#
-# Hardware crypto devices
-#
-
 #
 # Library routines
 #
-- 
cgit v1.2.1


From 9142e0c8396b25ed4cb549b5efa636065768ebe0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Update defconfig

This is based on the x86-64 defconfig which works on a wide range of systems.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/defconfig | 1063 ++++++++++++++++++++++++---------------------------
 1 file changed, 501 insertions(+), 562 deletions(-)

diff --git a/arch/i386/defconfig b/arch/i386/defconfig
index 89ebb7a316ab..1a29bfa26d0c 100644
--- a/arch/i386/defconfig
+++ b/arch/i386/defconfig
@@ -1,41 +1,51 @@
 #
 # Automatically generated make config: don't edit
+# Linux kernel version: 2.6.18-git5
+# Tue Sep 26 09:30:47 2006
 #
 CONFIG_X86_32=y
+CONFIG_GENERIC_TIME=y
+CONFIG_LOCKDEP_SUPPORT=y
+CONFIG_STACKTRACE_SUPPORT=y
 CONFIG_SEMAPHORE_SLEEPERS=y
 CONFIG_X86=y
 CONFIG_MMU=y
 CONFIG_GENERIC_ISA_DMA=y
 CONFIG_GENERIC_IOMAP=y
+CONFIG_GENERIC_HWEIGHT=y
 CONFIG_ARCH_MAY_HAVE_PC_FDC=y
 CONFIG_DMI=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
 #
 # General setup
 #
 CONFIG_LOCALVERSION=""
-# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_LOCALVERSION_AUTO=y
 CONFIG_SWAP=y
 CONFIG_SYSVIPC=y
-# CONFIG_POSIX_MQUEUE is not set
+CONFIG_POSIX_MQUEUE=y
 # CONFIG_BSD_PROCESS_ACCT is not set
-CONFIG_SYSCTL=y
+# CONFIG_TASKSTATS is not set
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
+# CONFIG_CPUSETS is not set
+# CONFIG_RELAY is not set
 CONFIG_INITRAMFS_SOURCE=""
-CONFIG_UID16=y
-CONFIG_VM86=y
 CONFIG_CC_OPTIMIZE_FOR_SIZE=y
 # CONFIG_EMBEDDED is not set
+CONFIG_UID16=y
+CONFIG_SYSCTL=y
 CONFIG_KALLSYMS=y
+CONFIG_KALLSYMS_ALL=y
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
 CONFIG_HOTPLUG=y
 CONFIG_PRINTK=y
@@ -45,11 +55,9 @@ CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
 CONFIG_SHMEM=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
 CONFIG_SLAB=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_RT_MUTEXES=y
 # CONFIG_TINY_SHMEM is not set
 CONFIG_BASE_SMALL=0
 # CONFIG_SLOB is not set
@@ -60,41 +68,45 @@ CONFIG_BASE_SMALL=0
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
 CONFIG_MODULE_FORCE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 # CONFIG_KMOD is not set
+CONFIG_STOP_MACHINE=y
 
 #
 # Block layer
 #
-# CONFIG_LBD is not set
+CONFIG_LBD=y
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
 
 #
 # IO Schedulers
 #
 CONFIG_IOSCHED_NOOP=y
-# CONFIG_IOSCHED_AS is not set
-# CONFIG_IOSCHED_DEADLINE is not set
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
 CONFIG_IOSCHED_CFQ=y
-# CONFIG_DEFAULT_AS is not set
+CONFIG_DEFAULT_AS=y
 # CONFIG_DEFAULT_DEADLINE is not set
-CONFIG_DEFAULT_CFQ=y
+# CONFIG_DEFAULT_CFQ is not set
 # CONFIG_DEFAULT_NOOP is not set
-CONFIG_DEFAULT_IOSCHED="cfq"
+CONFIG_DEFAULT_IOSCHED="anticipatory"
 
 #
 # Processor type and features
 #
-CONFIG_X86_PC=y
+CONFIG_SMP=y
+# CONFIG_X86_PC is not set
 # CONFIG_X86_ELAN is not set
 # CONFIG_X86_VOYAGER is not set
 # CONFIG_X86_NUMAQ is not set
 # CONFIG_X86_SUMMIT is not set
 # CONFIG_X86_BIGSMP is not set
 # CONFIG_X86_VISWS is not set
-# CONFIG_X86_GENERICARCH is not set
+CONFIG_X86_GENERICARCH=y
 # CONFIG_X86_ES7000 is not set
+CONFIG_X86_CYCLONE_TIMER=y
 # CONFIG_M386 is not set
 # CONFIG_M486 is not set
 # CONFIG_M586 is not set
@@ -102,11 +114,11 @@ CONFIG_X86_PC=y
 # CONFIG_M586MMX is not set
 # CONFIG_M686 is not set
 # CONFIG_MPENTIUMII is not set
-# CONFIG_MPENTIUMIII is not set
+CONFIG_MPENTIUMIII=y
 # CONFIG_MPENTIUMM is not set
 # CONFIG_MPENTIUM4 is not set
 # CONFIG_MK6 is not set
-CONFIG_MK7=y
+# CONFIG_MK7 is not set
 # CONFIG_MK8 is not set
 # CONFIG_MCRUSOE is not set
 # CONFIG_MEFFICEON is not set
@@ -117,10 +129,10 @@ CONFIG_MK7=y
 # CONFIG_MGEODE_LX is not set
 # CONFIG_MCYRIXIII is not set
 # CONFIG_MVIAC3_2 is not set
-# CONFIG_X86_GENERIC is not set
+CONFIG_X86_GENERIC=y
 CONFIG_X86_CMPXCHG=y
 CONFIG_X86_XADD=y
-CONFIG_X86_L1_CACHE_SHIFT=6
+CONFIG_X86_L1_CACHE_SHIFT=7
 CONFIG_RWSEM_XCHGADD_ALGORITHM=y
 CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_X86_WP_WORKS_OK=y
@@ -131,26 +143,28 @@ CONFIG_X86_CMPXCHG64=y
 CONFIG_X86_GOOD_APIC=y
 CONFIG_X86_INTEL_USERCOPY=y
 CONFIG_X86_USE_PPRO_CHECKSUM=y
-CONFIG_X86_USE_3DNOW=y
 CONFIG_X86_TSC=y
-# CONFIG_HPET_TIMER is not set
-# CONFIG_SMP is not set
-CONFIG_PREEMPT_NONE=y
-# CONFIG_PREEMPT_VOLUNTARY is not set
+CONFIG_HPET_TIMER=y
+CONFIG_HPET_EMULATE_RTC=y
+CONFIG_NR_CPUS=32
+CONFIG_SCHED_SMT=y
+CONFIG_SCHED_MC=y
+# CONFIG_PREEMPT_NONE is not set
+CONFIG_PREEMPT_VOLUNTARY=y
 # CONFIG_PREEMPT is not set
-CONFIG_X86_UP_APIC=y
-CONFIG_X86_UP_IOAPIC=y
+CONFIG_PREEMPT_BKL=y
 CONFIG_X86_LOCAL_APIC=y
 CONFIG_X86_IO_APIC=y
 CONFIG_X86_MCE=y
 CONFIG_X86_MCE_NONFATAL=y
-# CONFIG_X86_MCE_P4THERMAL is not set
+CONFIG_X86_MCE_P4THERMAL=y
+CONFIG_VM86=y
 # CONFIG_TOSHIBA is not set
 # CONFIG_I8K is not set
 # CONFIG_X86_REBOOTFIXUPS is not set
-# CONFIG_MICROCODE is not set
-# CONFIG_X86_MSR is not set
-# CONFIG_X86_CPUID is not set
+CONFIG_MICROCODE=y
+CONFIG_X86_MSR=y
+CONFIG_X86_CPUID=y
 
 #
 # Firmware Drivers
@@ -158,68 +172,67 @@ CONFIG_X86_MCE_NONFATAL=y
 # CONFIG_EDD is not set
 # CONFIG_DELL_RBU is not set
 # CONFIG_DCDBAS is not set
-CONFIG_NOHIGHMEM=y
-# CONFIG_HIGHMEM4G is not set
+# CONFIG_NOHIGHMEM is not set
+CONFIG_HIGHMEM4G=y
 # CONFIG_HIGHMEM64G is not set
-CONFIG_VMSPLIT_3G=y
-# CONFIG_VMSPLIT_3G_OPT is not set
-# CONFIG_VMSPLIT_2G is not set
-# CONFIG_VMSPLIT_1G is not set
 CONFIG_PAGE_OFFSET=0xC0000000
-CONFIG_ARCH_FLATMEM_ENABLE=y
-CONFIG_ARCH_SPARSEMEM_ENABLE=y
-CONFIG_ARCH_SELECT_MEMORY_MODEL=y
+CONFIG_HIGHMEM=y
 CONFIG_SELECT_MEMORY_MODEL=y
 CONFIG_FLATMEM_MANUAL=y
 # CONFIG_DISCONTIGMEM_MANUAL is not set
 # CONFIG_SPARSEMEM_MANUAL is not set
 CONFIG_FLATMEM=y
 CONFIG_FLAT_NODE_MEM_MAP=y
-CONFIG_SPARSEMEM_STATIC=y
+# CONFIG_SPARSEMEM_STATIC is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
+CONFIG_RESOURCES_64BIT=y
+# CONFIG_HIGHPTE is not set
 # CONFIG_MATH_EMULATION is not set
 CONFIG_MTRR=y
 # CONFIG_EFI is not set
+# CONFIG_IRQBALANCE is not set
 CONFIG_REGPARM=y
-# CONFIG_SECCOMP is not set
-CONFIG_HZ_100=y
-# CONFIG_HZ_250 is not set
+CONFIG_SECCOMP=y
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
 # CONFIG_HZ_1000 is not set
-CONFIG_HZ=100
+CONFIG_HZ=250
 # CONFIG_KEXEC is not set
+# CONFIG_CRASH_DUMP is not set
 CONFIG_PHYSICAL_START=0x100000
-CONFIG_DOUBLEFAULT=y
+# CONFIG_HOTPLUG_CPU is not set
+CONFIG_COMPAT_VDSO=y
+CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y
 
 #
 # Power management options (ACPI, APM)
 #
 CONFIG_PM=y
-# CONFIG_PM_LEGACY is not set
+CONFIG_PM_LEGACY=y
 # CONFIG_PM_DEBUG is not set
-CONFIG_SOFTWARE_SUSPEND=y
-CONFIG_PM_STD_PARTITION=""
 
 #
 # ACPI (Advanced Configuration and Power Interface) Support
 #
 CONFIG_ACPI=y
-# CONFIG_ACPI_SLEEP is not set
-# CONFIG_ACPI_AC is not set
-# CONFIG_ACPI_BATTERY is not set
-# CONFIG_ACPI_BUTTON is not set
+CONFIG_ACPI_AC=y
+CONFIG_ACPI_BATTERY=y
+CONFIG_ACPI_BUTTON=y
 # CONFIG_ACPI_VIDEO is not set
 # CONFIG_ACPI_HOTKEY is not set
-# CONFIG_ACPI_FAN is not set
-# CONFIG_ACPI_PROCESSOR is not set
+CONFIG_ACPI_FAN=y
+# CONFIG_ACPI_DOCK is not set
+CONFIG_ACPI_PROCESSOR=y
+CONFIG_ACPI_THERMAL=y
 # CONFIG_ACPI_ASUS is not set
 # CONFIG_ACPI_IBM is not set
 # CONFIG_ACPI_TOSHIBA is not set
-CONFIG_ACPI_BLACKLIST_YEAR=0
-# CONFIG_ACPI_DEBUG is not set
+CONFIG_ACPI_BLACKLIST_YEAR=2001
+CONFIG_ACPI_DEBUG=y
 CONFIG_ACPI_EC=y
 CONFIG_ACPI_POWER=y
 CONFIG_ACPI_SYSTEM=y
-# CONFIG_X86_PM_TIMER is not set
+CONFIG_X86_PM_TIMER=y
 # CONFIG_ACPI_CONTAINER is not set
 
 #
@@ -230,7 +243,41 @@ CONFIG_ACPI_SYSTEM=y
 #
 # CPU Frequency scaling
 #
-# CONFIG_CPU_FREQ is not set
+CONFIG_CPU_FREQ=y
+CONFIG_CPU_FREQ_TABLE=y
+CONFIG_CPU_FREQ_DEBUG=y
+CONFIG_CPU_FREQ_STAT=y
+# CONFIG_CPU_FREQ_STAT_DETAILS is not set
+CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE=y
+# CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE is not set
+CONFIG_CPU_FREQ_GOV_PERFORMANCE=y
+# CONFIG_CPU_FREQ_GOV_POWERSAVE is not set
+CONFIG_CPU_FREQ_GOV_USERSPACE=y
+CONFIG_CPU_FREQ_GOV_ONDEMAND=y
+# CONFIG_CPU_FREQ_GOV_CONSERVATIVE is not set
+
+#
+# CPUFreq processor drivers
+#
+CONFIG_X86_ACPI_CPUFREQ=y
+# CONFIG_X86_POWERNOW_K6 is not set
+# CONFIG_X86_POWERNOW_K7 is not set
+CONFIG_X86_POWERNOW_K8=y
+CONFIG_X86_POWERNOW_K8_ACPI=y
+# CONFIG_X86_GX_SUSPMOD is not set
+# CONFIG_X86_SPEEDSTEP_CENTRINO is not set
+# CONFIG_X86_SPEEDSTEP_ICH is not set
+# CONFIG_X86_SPEEDSTEP_SMI is not set
+# CONFIG_X86_P4_CLOCKMOD is not set
+# CONFIG_X86_CPUFREQ_NFORCE2 is not set
+# CONFIG_X86_LONGRUN is not set
+# CONFIG_X86_LONGHAUL is not set
+
+#
+# shared options
+#
+CONFIG_X86_ACPI_CPUFREQ_PROC_INTF=y
+# CONFIG_X86_SPEEDSTEP_LIB is not set
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -244,12 +291,13 @@ CONFIG_PCI_BIOS=y
 CONFIG_PCI_DIRECT=y
 CONFIG_PCI_MMCONFIG=y
 # CONFIG_PCIEPORTBUS is not set
-# CONFIG_PCI_MSI is not set
-# CONFIG_PCI_LEGACY_PROC is not set
+CONFIG_PCI_MSI=y
+# CONFIG_PCI_DEBUG is not set
 CONFIG_ISA_DMA_API=y
 # CONFIG_ISA is not set
 # CONFIG_MCA is not set
 # CONFIG_SCx200 is not set
+CONFIG_K8_NB=y
 
 #
 # PCCARD (PCMCIA/CardBus) support
@@ -278,93 +326,54 @@ CONFIG_NET=y
 #
 # CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
-CONFIG_PACKET_MMAP=y
+# CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
+CONFIG_XFRM=y
+# CONFIG_XFRM_USER is not set
+# CONFIG_XFRM_SUB_POLICY is not set
 # CONFIG_NET_KEY is not set
 CONFIG_INET=y
-# CONFIG_IP_MULTICAST is not set
+CONFIG_IP_MULTICAST=y
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_FIB_HASH=y
-# CONFIG_IP_PNP is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
+# CONFIG_IP_MROUTE is not set
 # CONFIG_ARPD is not set
 # CONFIG_SYN_COOKIES is not set
 # CONFIG_INET_AH is not set
 # CONFIG_INET_ESP is not set
 # CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
 # CONFIG_INET_TUNNEL is not set
-# CONFIG_INET_DIAG is not set
+CONFIG_INET_XFRM_MODE_TRANSPORT=y
+CONFIG_INET_XFRM_MODE_TUNNEL=y
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_BIC=y
-
-#
-# IP: Virtual Server Configuration
-#
-# CONFIG_IP_VS is not set
-# CONFIG_IPV6 is not set
-CONFIG_NETFILTER=y
-# CONFIG_NETFILTER_DEBUG is not set
-
-#
-# Core Netfilter Configuration
-#
-# CONFIG_NETFILTER_NETLINK is not set
-CONFIG_NETFILTER_XTABLES=y
-# CONFIG_NETFILTER_XT_TARGET_CLASSIFY is not set
-# CONFIG_NETFILTER_XT_TARGET_MARK is not set
-# CONFIG_NETFILTER_XT_TARGET_NFQUEUE is not set
-# CONFIG_NETFILTER_XT_MATCH_COMMENT is not set
-# CONFIG_NETFILTER_XT_MATCH_CONNTRACK is not set
-# CONFIG_NETFILTER_XT_MATCH_DCCP is not set
-# CONFIG_NETFILTER_XT_MATCH_HELPER is not set
-# CONFIG_NETFILTER_XT_MATCH_LENGTH is not set
-CONFIG_NETFILTER_XT_MATCH_LIMIT=y
-CONFIG_NETFILTER_XT_MATCH_MAC=y
-# CONFIG_NETFILTER_XT_MATCH_MARK is not set
-# CONFIG_NETFILTER_XT_MATCH_PKTTYPE is not set
-# CONFIG_NETFILTER_XT_MATCH_REALM is not set
-# CONFIG_NETFILTER_XT_MATCH_SCTP is not set
-CONFIG_NETFILTER_XT_MATCH_STATE=y
-# CONFIG_NETFILTER_XT_MATCH_STRING is not set
-# CONFIG_NETFILTER_XT_MATCH_TCPMSS is not set
-
-#
-# IP: Netfilter Configuration
-#
-CONFIG_IP_NF_CONNTRACK=y
-# CONFIG_IP_NF_CT_ACCT is not set
-# CONFIG_IP_NF_CONNTRACK_MARK is not set
-# CONFIG_IP_NF_CONNTRACK_EVENTS is not set
-# CONFIG_IP_NF_CT_PROTO_SCTP is not set
-CONFIG_IP_NF_FTP=y
-# CONFIG_IP_NF_IRC is not set
-# CONFIG_IP_NF_NETBIOS_NS is not set
-# CONFIG_IP_NF_TFTP is not set
-# CONFIG_IP_NF_AMANDA is not set
-# CONFIG_IP_NF_PPTP is not set
-# CONFIG_IP_NF_QUEUE is not set
-CONFIG_IP_NF_IPTABLES=y
-# CONFIG_IP_NF_MATCH_IPRANGE is not set
-# CONFIG_IP_NF_MATCH_MULTIPORT is not set
-# CONFIG_IP_NF_MATCH_TOS is not set
-# CONFIG_IP_NF_MATCH_RECENT is not set
-# CONFIG_IP_NF_MATCH_ECN is not set
-# CONFIG_IP_NF_MATCH_DSCP is not set
-# CONFIG_IP_NF_MATCH_AH_ESP is not set
-# CONFIG_IP_NF_MATCH_TTL is not set
-# CONFIG_IP_NF_MATCH_OWNER is not set
-# CONFIG_IP_NF_MATCH_ADDRTYPE is not set
-# CONFIG_IP_NF_MATCH_HASHLIMIT is not set
-CONFIG_IP_NF_FILTER=y
-# CONFIG_IP_NF_TARGET_REJECT is not set
-CONFIG_IP_NF_TARGET_LOG=y
-# CONFIG_IP_NF_TARGET_ULOG is not set
-# CONFIG_IP_NF_TARGET_TCPMSS is not set
-# CONFIG_IP_NF_NAT is not set
-# CONFIG_IP_NF_MANGLE is not set
-# CONFIG_IP_NF_RAW is not set
-# CONFIG_IP_NF_ARPTABLES is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+CONFIG_IPV6=y
+# CONFIG_IPV6_PRIVACY is not set
+# CONFIG_IPV6_ROUTER_PREF is not set
+# CONFIG_INET6_AH is not set
+# CONFIG_INET6_ESP is not set
+# CONFIG_INET6_IPCOMP is not set
+# CONFIG_IPV6_MIP6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+CONFIG_INET6_XFRM_MODE_TRANSPORT=y
+CONFIG_INET6_XFRM_MODE_TUNNEL=y
+# CONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION is not set
+# CONFIG_IPV6_TUNNEL is not set
+# CONFIG_IPV6_SUBTREES is not set
+# CONFIG_IPV6_MULTIPLE_TABLES is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
 
 #
 # DCCP Configuration (EXPERIMENTAL)
@@ -389,7 +398,6 @@ CONFIG_IP_NF_TARGET_LOG=y
 # CONFIG_ATALK is not set
 # CONFIG_X25 is not set
 # CONFIG_LAPB is not set
-# CONFIG_NET_DIVERT is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
 
@@ -402,6 +410,7 @@ CONFIG_IP_NF_TARGET_LOG=y
 # Network testing
 #
 # CONFIG_NET_PKTGEN is not set
+# CONFIG_NET_TCPPROBE is not set
 # CONFIG_HAMRADIO is not set
 # CONFIG_IRDA is not set
 # CONFIG_BT is not set
@@ -416,7 +425,9 @@ CONFIG_IP_NF_TARGET_LOG=y
 #
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
+CONFIG_FW_LOADER=y
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_SYS_HYPERVISOR is not set
 
 #
 # Connector - unified userspace <-> kernelspace linker
@@ -431,13 +442,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 #
 # Parallel port support
 #
-CONFIG_PARPORT=y
-CONFIG_PARPORT_PC=y
-# CONFIG_PARPORT_SERIAL is not set
-# CONFIG_PARPORT_PC_FIFO is not set
-# CONFIG_PARPORT_PC_SUPERIO is not set
-# CONFIG_PARPORT_GSC is not set
-CONFIG_PARPORT_1284=y
+# CONFIG_PARPORT is not set
 
 #
 # Plug and Play support
@@ -447,8 +452,7 @@ CONFIG_PARPORT_1284=y
 #
 # Block devices
 #
-# CONFIG_BLK_DEV_FD is not set
-# CONFIG_PARIDE is not set
+CONFIG_BLK_DEV_FD=y
 # CONFIG_BLK_CPQ_DA is not set
 # CONFIG_BLK_CPQ_CISS_DA is not set
 # CONFIG_BLK_DEV_DAC960 is not set
@@ -459,8 +463,11 @@ CONFIG_BLK_DEV_LOOP=y
 # CONFIG_BLK_DEV_NBD is not set
 # CONFIG_BLK_DEV_SX8 is not set
 # CONFIG_BLK_DEV_UB is not set
-# CONFIG_BLK_DEV_RAM is not set
+CONFIG_BLK_DEV_RAM=y
 CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
+CONFIG_BLK_DEV_INITRD=y
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
 
@@ -476,7 +483,7 @@ CONFIG_BLK_DEV_IDE=y
 # CONFIG_BLK_DEV_IDE_SATA is not set
 # CONFIG_BLK_DEV_HD_IDE is not set
 CONFIG_BLK_DEV_IDEDISK=y
-# CONFIG_IDEDISK_MULTI_MODE is not set
+CONFIG_IDEDISK_MULTI_MODE=y
 CONFIG_BLK_DEV_IDECD=y
 # CONFIG_BLK_DEV_IDETAPE is not set
 # CONFIG_BLK_DEV_IDEFLOPPY is not set
@@ -486,10 +493,10 @@ CONFIG_BLK_DEV_IDECD=y
 #
 # IDE chipset support/bugfixes
 #
-# CONFIG_IDE_GENERIC is not set
+CONFIG_IDE_GENERIC=y
 # CONFIG_BLK_DEV_CMD640 is not set
 CONFIG_BLK_DEV_IDEPCI=y
-CONFIG_IDEPCI_SHARE_IRQ=y
+# CONFIG_IDEPCI_SHARE_IRQ is not set
 # CONFIG_BLK_DEV_OFFBOARD is not set
 # CONFIG_BLK_DEV_GENERIC is not set
 # CONFIG_BLK_DEV_OPTI621 is not set
@@ -500,7 +507,7 @@ CONFIG_IDEDMA_PCI_AUTO=y
 # CONFIG_IDEDMA_ONLYDISK is not set
 # CONFIG_BLK_DEV_AEC62XX is not set
 # CONFIG_BLK_DEV_ALI15X3 is not set
-# CONFIG_BLK_DEV_AMD74XX is not set
+CONFIG_BLK_DEV_AMD74XX=y
 # CONFIG_BLK_DEV_ATIIXP is not set
 # CONFIG_BLK_DEV_CMD64X is not set
 # CONFIG_BLK_DEV_TRIFLEX is not set
@@ -511,7 +518,7 @@ CONFIG_IDEDMA_PCI_AUTO=y
 # CONFIG_BLK_DEV_HPT34X is not set
 # CONFIG_BLK_DEV_HPT366 is not set
 # CONFIG_BLK_DEV_SC1200 is not set
-# CONFIG_BLK_DEV_PIIX is not set
+CONFIG_BLK_DEV_PIIX=y
 # CONFIG_BLK_DEV_IT821X is not set
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
@@ -521,7 +528,7 @@ CONFIG_IDEDMA_PCI_AUTO=y
 # CONFIG_BLK_DEV_SIS5513 is not set
 # CONFIG_BLK_DEV_SLC90E66 is not set
 # CONFIG_BLK_DEV_TRM290 is not set
-CONFIG_BLK_DEV_VIA82CXXX=y
+# CONFIG_BLK_DEV_VIA82CXXX is not set
 # CONFIG_IDE_ARM is not set
 CONFIG_BLK_DEV_IDEDMA=y
 # CONFIG_IDEDMA_IVB is not set
@@ -533,6 +540,7 @@ CONFIG_IDEDMA_AUTO=y
 #
 # CONFIG_RAID_ATTRS is not set
 CONFIG_SCSI=y
+CONFIG_SCSI_NETLINK=y
 # CONFIG_SCSI_PROC_FS is not set
 
 #
@@ -541,8 +549,9 @@ CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 # CONFIG_CHR_DEV_ST is not set
 # CONFIG_CHR_DEV_OSST is not set
-# CONFIG_BLK_DEV_SR is not set
-# CONFIG_CHR_DEV_SG is not set
+CONFIG_BLK_DEV_SR=y
+# CONFIG_BLK_DEV_SR_VENDOR is not set
+CONFIG_CHR_DEV_SG=y
 # CONFIG_CHR_DEV_SCH is not set
 
 #
@@ -553,29 +562,44 @@ CONFIG_BLK_DEV_SD=y
 # CONFIG_SCSI_LOGGING is not set
 
 #
-# SCSI Transport Attributes
+# SCSI Transports
 #
-# CONFIG_SCSI_SPI_ATTRS is not set
-# CONFIG_SCSI_FC_ATTRS is not set
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_FC_ATTRS=y
 # CONFIG_SCSI_ISCSI_ATTRS is not set
 # CONFIG_SCSI_SAS_ATTRS is not set
+# CONFIG_SCSI_SAS_LIBSAS is not set
 
 #
 # SCSI low-level drivers
 #
 # CONFIG_ISCSI_TCP is not set
-# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+CONFIG_BLK_DEV_3W_XXXX_RAID=y
 # CONFIG_SCSI_3W_9XXX is not set
 # CONFIG_SCSI_ACARD is not set
 # CONFIG_SCSI_AACRAID is not set
-# CONFIG_SCSI_AIC7XXX is not set
+CONFIG_SCSI_AIC7XXX=y
+CONFIG_AIC7XXX_CMDS_PER_DEVICE=32
+CONFIG_AIC7XXX_RESET_DELAY_MS=5000
+CONFIG_AIC7XXX_DEBUG_ENABLE=y
+CONFIG_AIC7XXX_DEBUG_MASK=0
+CONFIG_AIC7XXX_REG_PRETTY_PRINT=y
 # CONFIG_SCSI_AIC7XXX_OLD is not set
-# CONFIG_SCSI_AIC79XX is not set
+CONFIG_SCSI_AIC79XX=y
+CONFIG_AIC79XX_CMDS_PER_DEVICE=32
+CONFIG_AIC79XX_RESET_DELAY_MS=4000
+# CONFIG_AIC79XX_ENABLE_RD_STRM is not set
+# CONFIG_AIC79XX_DEBUG_ENABLE is not set
+CONFIG_AIC79XX_DEBUG_MASK=0
+# CONFIG_AIC79XX_REG_PRETTY_PRINT is not set
+# CONFIG_SCSI_AIC94XX is not set
 # CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_SCSI_ADVANSYS is not set
+# CONFIG_SCSI_ARCMSR is not set
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
 # CONFIG_MEGARAID_SAS is not set
-# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_HPTIOP is not set
 # CONFIG_SCSI_BUSLOGIC is not set
 # CONFIG_SCSI_DMX3191D is not set
 # CONFIG_SCSI_EATA is not set
@@ -584,11 +608,9 @@ CONFIG_BLK_DEV_SD=y
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
-# CONFIG_SCSI_PPA is not set
-# CONFIG_SCSI_IMM is not set
+# CONFIG_SCSI_STEX is not set
 # CONFIG_SCSI_SYM53C8XX_2 is not set
 # CONFIG_SCSI_IPR is not set
-# CONFIG_SCSI_QLOGIC_FC is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
 # CONFIG_SCSI_QLA_FC is not set
 # CONFIG_SCSI_LPFC is not set
@@ -597,23 +619,115 @@ CONFIG_BLK_DEV_SD=y
 # CONFIG_SCSI_NSP32 is not set
 # CONFIG_SCSI_DEBUG is not set
 
+#
+# Serial ATA (prod) and Parallel ATA (experimental) drivers
+#
+CONFIG_ATA=y
+CONFIG_SATA_AHCI=y
+CONFIG_SATA_SVW=y
+CONFIG_ATA_PIIX=y
+# CONFIG_SATA_MV is not set
+CONFIG_SATA_NV=y
+# CONFIG_PDC_ADMA is not set
+# CONFIG_SATA_QSTOR is not set
+# CONFIG_SATA_PROMISE is not set
+# CONFIG_SATA_SX4 is not set
+CONFIG_SATA_SIL=y
+# CONFIG_SATA_SIL24 is not set
+# CONFIG_SATA_SIS is not set
+# CONFIG_SATA_ULI is not set
+CONFIG_SATA_VIA=y
+# CONFIG_SATA_VITESSE is not set
+CONFIG_SATA_INTEL_COMBINED=y
+# CONFIG_PATA_ALI is not set
+# CONFIG_PATA_AMD is not set
+# CONFIG_PATA_ARTOP is not set
+# CONFIG_PATA_ATIIXP is not set
+# CONFIG_PATA_CMD64X is not set
+# CONFIG_PATA_CS5520 is not set
+# CONFIG_PATA_CS5530 is not set
+# CONFIG_PATA_CS5535 is not set
+# CONFIG_PATA_CYPRESS is not set
+# CONFIG_PATA_EFAR is not set
+# CONFIG_ATA_GENERIC is not set
+# CONFIG_PATA_HPT366 is not set
+# CONFIG_PATA_HPT37X is not set
+# CONFIG_PATA_HPT3X2N is not set
+# CONFIG_PATA_HPT3X3 is not set
+# CONFIG_PATA_IT821X is not set
+# CONFIG_PATA_JMICRON is not set
+# CONFIG_PATA_LEGACY is not set
+# CONFIG_PATA_TRIFLEX is not set
+# CONFIG_PATA_MPIIX is not set
+# CONFIG_PATA_OLDPIIX is not set
+# CONFIG_PATA_NETCELL is not set
+# CONFIG_PATA_NS87410 is not set
+# CONFIG_PATA_OPTI is not set
+# CONFIG_PATA_OPTIDMA is not set
+# CONFIG_PATA_PDC_OLD is not set
+# CONFIG_PATA_QDI is not set
+# CONFIG_PATA_RADISYS is not set
+# CONFIG_PATA_RZ1000 is not set
+# CONFIG_PATA_SC1200 is not set
+# CONFIG_PATA_SERVERWORKS is not set
+# CONFIG_PATA_PDC2027X is not set
+# CONFIG_PATA_SIL680 is not set
+# CONFIG_PATA_SIS is not set
+# CONFIG_PATA_VIA is not set
+# CONFIG_PATA_WINBOND is not set
+
 #
 # Multi-device support (RAID and LVM)
 #
-# CONFIG_MD is not set
+CONFIG_MD=y
+# CONFIG_BLK_DEV_MD is not set
+CONFIG_BLK_DEV_DM=y
+# CONFIG_DM_CRYPT is not set
+# CONFIG_DM_SNAPSHOT is not set
+# CONFIG_DM_MIRROR is not set
+# CONFIG_DM_ZERO is not set
+# CONFIG_DM_MULTIPATH is not set
 
 #
 # Fusion MPT device support
 #
-# CONFIG_FUSION is not set
-# CONFIG_FUSION_SPI is not set
+CONFIG_FUSION=y
+CONFIG_FUSION_SPI=y
 # CONFIG_FUSION_FC is not set
 # CONFIG_FUSION_SAS is not set
+CONFIG_FUSION_MAX_SGE=128
+# CONFIG_FUSION_CTL is not set
 
 #
 # IEEE 1394 (FireWire) support
 #
-# CONFIG_IEEE1394 is not set
+CONFIG_IEEE1394=y
+
+#
+# Subsystem Options
+#
+# CONFIG_IEEE1394_VERBOSEDEBUG is not set
+# CONFIG_IEEE1394_OUI_DB is not set
+# CONFIG_IEEE1394_EXTRA_CONFIG_ROMS is not set
+# CONFIG_IEEE1394_EXPORT_FULL_API is not set
+
+#
+# Device Drivers
+#
+
+#
+# Texas Instruments PCILynx requires I2C
+#
+CONFIG_IEEE1394_OHCI1394=y
+
+#
+# Protocol Drivers
+#
+# CONFIG_IEEE1394_VIDEO1394 is not set
+# CONFIG_IEEE1394_SBP2 is not set
+# CONFIG_IEEE1394_ETH1394 is not set
+# CONFIG_IEEE1394_DV1394 is not set
+CONFIG_IEEE1394_RAWIO=y
 
 #
 # I2O device support
@@ -652,46 +766,63 @@ CONFIG_MII=y
 #
 # Tulip family network device support
 #
-# CONFIG_NET_TULIP is not set
+CONFIG_NET_TULIP=y
+# CONFIG_DE2104X is not set
+CONFIG_TULIP=y
+# CONFIG_TULIP_MWI is not set
+# CONFIG_TULIP_MMIO is not set
+# CONFIG_TULIP_NAPI is not set
+# CONFIG_DE4X5 is not set
+# CONFIG_WINBOND_840 is not set
+# CONFIG_DM9102 is not set
+# CONFIG_ULI526X is not set
 # CONFIG_HP100 is not set
 CONFIG_NET_PCI=y
 # CONFIG_PCNET32 is not set
 # CONFIG_AMD8111_ETH is not set
 # CONFIG_ADAPTEC_STARFIRE is not set
-# CONFIG_B44 is not set
-# CONFIG_FORCEDETH is not set
+CONFIG_B44=y
+CONFIG_FORCEDETH=y
+# CONFIG_FORCEDETH_NAPI is not set
 # CONFIG_DGRS is not set
 # CONFIG_EEPRO100 is not set
 CONFIG_E100=y
 # CONFIG_FEALNX is not set
 # CONFIG_NATSEMI is not set
 # CONFIG_NE2K_PCI is not set
-# CONFIG_8139CP is not set
-# CONFIG_8139TOO is not set
+CONFIG_8139CP=y
+CONFIG_8139TOO=y
+# CONFIG_8139TOO_PIO is not set
+# CONFIG_8139TOO_TUNE_TWISTER is not set
+# CONFIG_8139TOO_8129 is not set
+# CONFIG_8139_OLD_RX_RESET is not set
 # CONFIG_SIS900 is not set
 # CONFIG_EPIC100 is not set
 # CONFIG_SUNDANCE is not set
 # CONFIG_TLAN is not set
 # CONFIG_VIA_RHINE is not set
-# CONFIG_NET_POCKET is not set
 
 #
 # Ethernet (1000 Mbit)
 #
 # CONFIG_ACENIC is not set
 # CONFIG_DL2K is not set
-# CONFIG_E1000 is not set
+CONFIG_E1000=y
+# CONFIG_E1000_NAPI is not set
+# CONFIG_E1000_DISABLE_PACKET_SPLIT is not set
 # CONFIG_NS83820 is not set
 # CONFIG_HAMACHI is not set
 # CONFIG_YELLOWFIN is not set
-# CONFIG_R8169 is not set
+CONFIG_R8169=y
+# CONFIG_R8169_NAPI is not set
 # CONFIG_SIS190 is not set
 # CONFIG_SKGE is not set
-# CONFIG_SKY2 is not set
+CONFIG_SKY2=y
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
-# CONFIG_TIGON3 is not set
-# CONFIG_BNX2 is not set
+CONFIG_TIGON3=y
+CONFIG_BNX2=y
+# CONFIG_QLA3XXX is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -699,6 +830,7 @@ CONFIG_E100=y
 # CONFIG_CHELSIO_T1 is not set
 # CONFIG_IXGB is not set
 # CONFIG_S2IO is not set
+# CONFIG_MYRI10GE is not set
 
 #
 # Token Ring devices
@@ -716,14 +848,15 @@ CONFIG_E100=y
 # CONFIG_WAN is not set
 # CONFIG_FDDI is not set
 # CONFIG_HIPPI is not set
-# CONFIG_PLIP is not set
 # CONFIG_PPP is not set
 # CONFIG_SLIP is not set
 # CONFIG_NET_FC is not set
 # CONFIG_SHAPER is not set
-# CONFIG_NETCONSOLE is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
+CONFIG_NETCONSOLE=y
+CONFIG_NETPOLL=y
+# CONFIG_NETPOLL_RX is not set
+# CONFIG_NETPOLL_TRAP is not set
+CONFIG_NET_POLL_CONTROLLER=y
 
 #
 # ISDN subsystem
@@ -745,8 +878,8 @@ CONFIG_INPUT=y
 #
 CONFIG_INPUT_MOUSEDEV=y
 CONFIG_INPUT_MOUSEDEV_PSAUX=y
-CONFIG_INPUT_MOUSEDEV_SCREEN_X=1280
-CONFIG_INPUT_MOUSEDEV_SCREEN_Y=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
 # CONFIG_INPUT_JOYDEV is not set
 # CONFIG_INPUT_TSDEV is not set
 CONFIG_INPUT_EVDEV=y
@@ -776,7 +909,6 @@ CONFIG_SERIO=y
 CONFIG_SERIO_I8042=y
 # CONFIG_SERIO_SERPORT is not set
 # CONFIG_SERIO_CT82C710 is not set
-# CONFIG_SERIO_PARKBD is not set
 # CONFIG_SERIO_PCIPS2 is not set
 CONFIG_SERIO_LIBPS2=y
 # CONFIG_SERIO_RAW is not set
@@ -788,14 +920,15 @@ CONFIG_SERIO_LIBPS2=y
 CONFIG_VT=y
 CONFIG_VT_CONSOLE=y
 CONFIG_HW_CONSOLE=y
+# CONFIG_VT_HW_CONSOLE_BINDING is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
 #
 # Serial drivers
 #
 CONFIG_SERIAL_8250=y
-# CONFIG_SERIAL_8250_CONSOLE is not set
-# CONFIG_SERIAL_8250_ACPI is not set
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_SERIAL_8250_PCI=y
 CONFIG_SERIAL_8250_NR_UARTS=4
 CONFIG_SERIAL_8250_RUNTIME_UARTS=4
 # CONFIG_SERIAL_8250_EXTENDED is not set
@@ -804,14 +937,11 @@ CONFIG_SERIAL_8250_RUNTIME_UARTS=4
 # Non-8250 serial port support
 #
 CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
 # CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
-CONFIG_PRINTER=y
-# CONFIG_LP_CONSOLE is not set
-# CONFIG_PPDEV is not set
-# CONFIG_TIPAR is not set
 
 #
 # IPMI
@@ -822,8 +952,12 @@ CONFIG_PRINTER=y
 # Watchdog Cards
 #
 # CONFIG_WATCHDOG is not set
-# CONFIG_HW_RANDOM is not set
-CONFIG_NVRAM=y
+CONFIG_HW_RANDOM=y
+CONFIG_HW_RANDOM_INTEL=y
+CONFIG_HW_RANDOM_AMD=y
+CONFIG_HW_RANDOM_GEODE=y
+CONFIG_HW_RANDOM_VIA=y
+# CONFIG_NVRAM is not set
 CONFIG_RTC=y
 # CONFIG_DTLK is not set
 # CONFIG_R3964 is not set
@@ -833,31 +967,28 @@ CONFIG_RTC=y
 #
 # Ftape, the floppy tape device driver
 #
-# CONFIG_FTAPE is not set
 CONFIG_AGP=y
 # CONFIG_AGP_ALI is not set
 # CONFIG_AGP_ATI is not set
 # CONFIG_AGP_AMD is not set
-# CONFIG_AGP_AMD64 is not set
-# CONFIG_AGP_INTEL is not set
+CONFIG_AGP_AMD64=y
+CONFIG_AGP_INTEL=y
 # CONFIG_AGP_NVIDIA is not set
 # CONFIG_AGP_SIS is not set
 # CONFIG_AGP_SWORKS is not set
-CONFIG_AGP_VIA=y
+# CONFIG_AGP_VIA is not set
 # CONFIG_AGP_EFFICEON is not set
-CONFIG_DRM=y
-# CONFIG_DRM_TDFX is not set
-# CONFIG_DRM_R128 is not set
-CONFIG_DRM_RADEON=y
-# CONFIG_DRM_MGA is not set
-# CONFIG_DRM_SIS is not set
-# CONFIG_DRM_VIA is not set
-# CONFIG_DRM_SAVAGE is not set
+# CONFIG_DRM is not set
 # CONFIG_MWAVE is not set
+# CONFIG_PC8736x_GPIO is not set
+# CONFIG_NSC_GPIO is not set
 # CONFIG_CS5535_GPIO is not set
-# CONFIG_RAW_DRIVER is not set
-# CONFIG_HPET is not set
-# CONFIG_HANGCHECK_TIMER is not set
+CONFIG_RAW_DRIVER=y
+CONFIG_MAX_RAW_DEVS=256
+CONFIG_HPET=y
+# CONFIG_HPET_RTC_IRQ is not set
+CONFIG_HPET_MMAP=y
+CONFIG_HANGCHECK_TIMER=y
 
 #
 # TPM devices
@@ -868,59 +999,7 @@ CONFIG_DRM_RADEON=y
 #
 # I2C support
 #
-CONFIG_I2C=y
-CONFIG_I2C_CHARDEV=y
-
-#
-# I2C Algorithms
-#
-CONFIG_I2C_ALGOBIT=y
-# CONFIG_I2C_ALGOPCF is not set
-# CONFIG_I2C_ALGOPCA is not set
-
-#
-# I2C Hardware Bus support
-#
-# CONFIG_I2C_ALI1535 is not set
-# CONFIG_I2C_ALI1563 is not set
-# CONFIG_I2C_ALI15X3 is not set
-# CONFIG_I2C_AMD756 is not set
-# CONFIG_I2C_AMD8111 is not set
-# CONFIG_I2C_I801 is not set
-# CONFIG_I2C_I810 is not set
-# CONFIG_I2C_PIIX4 is not set
-CONFIG_I2C_ISA=y
-# CONFIG_I2C_NFORCE2 is not set
-# CONFIG_I2C_PARPORT is not set
-# CONFIG_I2C_PARPORT_LIGHT is not set
-# CONFIG_I2C_PROSAVAGE is not set
-# CONFIG_I2C_SAVAGE4 is not set
-# CONFIG_SCx200_ACB is not set
-# CONFIG_I2C_SIS5595 is not set
-# CONFIG_I2C_SIS630 is not set
-# CONFIG_I2C_SIS96X is not set
-# CONFIG_I2C_STUB is not set
-# CONFIG_I2C_VIA is not set
-CONFIG_I2C_VIAPRO=y
-# CONFIG_I2C_VOODOO3 is not set
-# CONFIG_I2C_PCA_ISA is not set
-
-#
-# Miscellaneous I2C Chip support
-#
-# CONFIG_SENSORS_DS1337 is not set
-# CONFIG_SENSORS_DS1374 is not set
-# CONFIG_SENSORS_EEPROM is not set
-# CONFIG_SENSORS_PCF8574 is not set
-# CONFIG_SENSORS_PCA9539 is not set
-# CONFIG_SENSORS_PCF8591 is not set
-# CONFIG_SENSORS_RTC8564 is not set
-# CONFIG_SENSORS_MAX6875 is not set
-# CONFIG_RTC_X1205_I2C is not set
-# CONFIG_I2C_DEBUG_CORE is not set
-# CONFIG_I2C_DEBUG_ALGO is not set
-# CONFIG_I2C_DEBUG_BUS is not set
-# CONFIG_I2C_DEBUG_CHIP is not set
+# CONFIG_I2C is not set
 
 #
 # SPI support
@@ -931,169 +1010,44 @@ CONFIG_I2C_VIAPRO=y
 #
 # Dallas's 1-wire bus
 #
-# CONFIG_W1 is not set
 
 #
 # Hardware Monitoring support
 #
-CONFIG_HWMON=y
-CONFIG_HWMON_VID=y
-# CONFIG_SENSORS_ADM1021 is not set
-# CONFIG_SENSORS_ADM1025 is not set
-# CONFIG_SENSORS_ADM1026 is not set
-# CONFIG_SENSORS_ADM1031 is not set
-# CONFIG_SENSORS_ADM9240 is not set
-# CONFIG_SENSORS_ASB100 is not set
-# CONFIG_SENSORS_ATXP1 is not set
-# CONFIG_SENSORS_DS1621 is not set
-# CONFIG_SENSORS_F71805F is not set
-# CONFIG_SENSORS_FSCHER is not set
-# CONFIG_SENSORS_FSCPOS is not set
-# CONFIG_SENSORS_GL518SM is not set
-# CONFIG_SENSORS_GL520SM is not set
-CONFIG_SENSORS_IT87=y
-# CONFIG_SENSORS_LM63 is not set
-# CONFIG_SENSORS_LM75 is not set
-# CONFIG_SENSORS_LM77 is not set
-# CONFIG_SENSORS_LM78 is not set
-# CONFIG_SENSORS_LM80 is not set
-# CONFIG_SENSORS_LM83 is not set
-# CONFIG_SENSORS_LM85 is not set
-# CONFIG_SENSORS_LM87 is not set
-# CONFIG_SENSORS_LM90 is not set
-# CONFIG_SENSORS_LM92 is not set
-# CONFIG_SENSORS_MAX1619 is not set
-# CONFIG_SENSORS_PC87360 is not set
-# CONFIG_SENSORS_SIS5595 is not set
-# CONFIG_SENSORS_SMSC47M1 is not set
-# CONFIG_SENSORS_SMSC47B397 is not set
-# CONFIG_SENSORS_VIA686A is not set
-# CONFIG_SENSORS_VT8231 is not set
-# CONFIG_SENSORS_W83781D is not set
-# CONFIG_SENSORS_W83792D is not set
-# CONFIG_SENSORS_W83L785TS is not set
-# CONFIG_SENSORS_W83627HF is not set
-# CONFIG_SENSORS_W83627EHF is not set
-# CONFIG_SENSORS_HDAPS is not set
-# CONFIG_HWMON_DEBUG_CHIP is not set
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
 
 #
 # Misc devices
 #
 # CONFIG_IBM_ASM is not set
 
-#
-# Multimedia Capabilities Port drivers
-#
-
 #
 # Multimedia devices
 #
-CONFIG_VIDEO_DEV=y
-
-#
-# Video For Linux
-#
-
-#
-# Video Adapters
-#
-# CONFIG_VIDEO_ADV_DEBUG is not set
-# CONFIG_VIDEO_BT848 is not set
-# CONFIG_VIDEO_BWQCAM is not set
-# CONFIG_VIDEO_CQCAM is not set
-# CONFIG_VIDEO_W9966 is not set
-# CONFIG_VIDEO_CPIA is not set
-# CONFIG_VIDEO_SAA5246A is not set
-# CONFIG_VIDEO_SAA5249 is not set
-# CONFIG_TUNER_3036 is not set
-# CONFIG_VIDEO_STRADIS is not set
-# CONFIG_VIDEO_ZORAN is not set
-CONFIG_VIDEO_SAA7134=y
-# CONFIG_VIDEO_SAA7134_ALSA is not set
-# CONFIG_VIDEO_MXB is not set
-# CONFIG_VIDEO_DPC is not set
-# CONFIG_VIDEO_HEXIUM_ORION is not set
-# CONFIG_VIDEO_HEXIUM_GEMINI is not set
-# CONFIG_VIDEO_CX88 is not set
-# CONFIG_VIDEO_EM28XX is not set
-# CONFIG_VIDEO_OVCAMCHIP is not set
-# CONFIG_VIDEO_AUDIO_DECODER is not set
-# CONFIG_VIDEO_DECODER is not set
-
-#
-# Radio Adapters
-#
-# CONFIG_RADIO_GEMTEK_PCI is not set
-# CONFIG_RADIO_MAXIRADIO is not set
-# CONFIG_RADIO_MAESTRO is not set
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
 
 #
 # Digital Video Broadcasting Devices
 #
 # CONFIG_DVB is not set
-CONFIG_VIDEO_TUNER=y
-CONFIG_VIDEO_BUF=y
-CONFIG_VIDEO_IR=y
+# CONFIG_USB_DABUSB is not set
 
 #
 # Graphics support
 #
-CONFIG_FB=y
-CONFIG_FB_CFB_FILLRECT=y
-CONFIG_FB_CFB_COPYAREA=y
-CONFIG_FB_CFB_IMAGEBLIT=y
-# CONFIG_FB_MACMODES is not set
-CONFIG_FB_MODE_HELPERS=y
-# CONFIG_FB_TILEBLITTING is not set
-# CONFIG_FB_CIRRUS is not set
-# CONFIG_FB_PM2 is not set
-# CONFIG_FB_CYBER2000 is not set
-# CONFIG_FB_ARC is not set
-# CONFIG_FB_ASILIANT is not set
-# CONFIG_FB_IMSTT is not set
-# CONFIG_FB_VGA16 is not set
-# CONFIG_FB_VESA is not set
-CONFIG_VIDEO_SELECT=y
-# CONFIG_FB_HGA is not set
-# CONFIG_FB_S1D13XXX is not set
-# CONFIG_FB_NVIDIA is not set
-# CONFIG_FB_RIVA is not set
-# CONFIG_FB_I810 is not set
-# CONFIG_FB_INTEL is not set
-# CONFIG_FB_MATROX is not set
-# CONFIG_FB_RADEON_OLD is not set
-CONFIG_FB_RADEON=y
-CONFIG_FB_RADEON_I2C=y
-# CONFIG_FB_RADEON_DEBUG is not set
-# CONFIG_FB_ATY128 is not set
-# CONFIG_FB_ATY is not set
-# CONFIG_FB_SAVAGE is not set
-# CONFIG_FB_SIS is not set
-# CONFIG_FB_NEOMAGIC is not set
-# CONFIG_FB_KYRO is not set
-# CONFIG_FB_3DFX is not set
-# CONFIG_FB_VOODOO1 is not set
-# CONFIG_FB_CYBLA is not set
-# CONFIG_FB_TRIDENT is not set
-# CONFIG_FB_GEODE is not set
-# CONFIG_FB_VIRTUAL is not set
+CONFIG_FIRMWARE_EDID=y
+# CONFIG_FB is not set
 
 #
 # Console display driver support
 #
 CONFIG_VGA_CONSOLE=y
+CONFIG_VGACON_SOFT_SCROLLBACK=y
+CONFIG_VGACON_SOFT_SCROLLBACK_SIZE=128
+CONFIG_VIDEO_SELECT=y
 CONFIG_DUMMY_CONSOLE=y
-CONFIG_FRAMEBUFFER_CONSOLE=y
-# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set
-# CONFIG_FONTS is not set
-CONFIG_FONT_8x8=y
-CONFIG_FONT_8x16=y
-
-#
-# Logo configuration
-#
-# CONFIG_LOGO is not set
 # CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
@@ -1104,97 +1058,30 @@ CONFIG_SOUND=y
 #
 # Advanced Linux Sound Architecture
 #
-CONFIG_SND=y
-CONFIG_SND_TIMER=y
-CONFIG_SND_PCM=y
-CONFIG_SND_RAWMIDI=y
-CONFIG_SND_SEQUENCER=y
-# CONFIG_SND_SEQ_DUMMY is not set
-# CONFIG_SND_MIXER_OSS is not set
-# CONFIG_SND_PCM_OSS is not set
-# CONFIG_SND_SEQUENCER_OSS is not set
-CONFIG_SND_RTCTIMER=y
-CONFIG_SND_SEQ_RTCTIMER_DEFAULT=y
-# CONFIG_SND_DYNAMIC_MINORS is not set
-# CONFIG_SND_SUPPORT_OLD_API is not set
-# CONFIG_SND_VERBOSE_PRINTK is not set
-# CONFIG_SND_DEBUG is not set
-
-#
-# Generic devices
-#
-CONFIG_SND_MPU401_UART=y
-CONFIG_SND_AC97_CODEC=y
-CONFIG_SND_AC97_BUS=y
-# CONFIG_SND_DUMMY is not set
-# CONFIG_SND_VIRMIDI is not set
-# CONFIG_SND_MTPAV is not set
-# CONFIG_SND_SERIAL_U16550 is not set
-# CONFIG_SND_MPU401 is not set
-
-#
-# PCI devices
-#
-# CONFIG_SND_AD1889 is not set
-# CONFIG_SND_ALS4000 is not set
-# CONFIG_SND_ALI5451 is not set
-# CONFIG_SND_ATIIXP is not set
-# CONFIG_SND_ATIIXP_MODEM is not set
-# CONFIG_SND_AU8810 is not set
-# CONFIG_SND_AU8820 is not set
-# CONFIG_SND_AU8830 is not set
-# CONFIG_SND_AZT3328 is not set
-# CONFIG_SND_BT87X is not set
-# CONFIG_SND_CA0106 is not set
-# CONFIG_SND_CMIPCI is not set
-# CONFIG_SND_CS4281 is not set
-# CONFIG_SND_CS46XX is not set
-# CONFIG_SND_CS5535AUDIO is not set
-# CONFIG_SND_EMU10K1 is not set
-# CONFIG_SND_EMU10K1X is not set
-# CONFIG_SND_ENS1370 is not set
-# CONFIG_SND_ENS1371 is not set
-# CONFIG_SND_ES1938 is not set
-# CONFIG_SND_ES1968 is not set
-# CONFIG_SND_FM801 is not set
-# CONFIG_SND_HDA_INTEL is not set
-# CONFIG_SND_HDSP is not set
-# CONFIG_SND_HDSPM is not set
-# CONFIG_SND_ICE1712 is not set
-# CONFIG_SND_ICE1724 is not set
-# CONFIG_SND_INTEL8X0 is not set
-# CONFIG_SND_INTEL8X0M is not set
-# CONFIG_SND_KORG1212 is not set
-# CONFIG_SND_MAESTRO3 is not set
-# CONFIG_SND_MIXART is not set
-# CONFIG_SND_NM256 is not set
-# CONFIG_SND_PCXHR is not set
-# CONFIG_SND_RME32 is not set
-# CONFIG_SND_RME96 is not set
-# CONFIG_SND_RME9652 is not set
-# CONFIG_SND_SONICVIBES is not set
-# CONFIG_SND_TRIDENT is not set
-CONFIG_SND_VIA82XX=y
-# CONFIG_SND_VIA82XX_MODEM is not set
-# CONFIG_SND_VX222 is not set
-# CONFIG_SND_YMFPCI is not set
-
-#
-# USB devices
-#
-# CONFIG_SND_USB_AUDIO is not set
-# CONFIG_SND_USB_USX2Y is not set
+# CONFIG_SND is not set
 
 #
 # Open Sound System
 #
-# CONFIG_SOUND_PRIME is not set
+CONFIG_SOUND_PRIME=y
+CONFIG_OSS_OBSOLETE_DRIVER=y
+# CONFIG_SOUND_BT878 is not set
+# CONFIG_SOUND_EMU10K1 is not set
+# CONFIG_SOUND_FUSION is not set
+# CONFIG_SOUND_ES1371 is not set
+CONFIG_SOUND_ICH=y
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+# CONFIG_SOUND_VIA82CXXX is not set
+# CONFIG_SOUND_OSS is not set
 
 #
 # USB support
 #
 CONFIG_USB_ARCH_HAS_HCD=y
 CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB_ARCH_HAS_EHCI=y
 CONFIG_USB=y
 # CONFIG_USB_DEBUG is not set
 
@@ -1213,17 +1100,19 @@ CONFIG_USB_DEVICEFS=y
 CONFIG_USB_EHCI_HCD=y
 # CONFIG_USB_EHCI_SPLIT_ISO is not set
 # CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_EHCI_TT_NEWSCHED is not set
 # CONFIG_USB_ISP116X_HCD is not set
-# CONFIG_USB_OHCI_HCD is not set
+CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
 CONFIG_USB_UHCI_HCD=y
 # CONFIG_USB_SL811_HCD is not set
 
 #
 # USB Device Class drivers
 #
-# CONFIG_OBSOLETE_OSS_USB_DRIVER is not set
 # CONFIG_USB_ACM is not set
-# CONFIG_USB_PRINTER is not set
+CONFIG_USB_PRINTER=y
 
 #
 # NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
@@ -1248,21 +1137,17 @@ CONFIG_USB_STORAGE=y
 #
 # USB Input Devices
 #
-# CONFIG_USB_HID is not set
-
-#
-# USB HID Boot Protocol drivers
-#
-# CONFIG_USB_KBD is not set
-# CONFIG_USB_MOUSE is not set
+CONFIG_USB_HID=y
+CONFIG_USB_HIDINPUT=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
+# CONFIG_HID_FF is not set
+# CONFIG_USB_HIDDEV is not set
 # CONFIG_USB_AIPTEK is not set
 # CONFIG_USB_WACOM is not set
 # CONFIG_USB_ACECAD is not set
 # CONFIG_USB_KBTAB is not set
 # CONFIG_USB_POWERMATE is not set
-# CONFIG_USB_MTOUCH is not set
-# CONFIG_USB_ITMTOUCH is not set
-# CONFIG_USB_EGALAX is not set
+# CONFIG_USB_TOUCHSCREEN is not set
 # CONFIG_USB_YEALINK is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
@@ -1276,21 +1161,6 @@ CONFIG_USB_STORAGE=y
 # CONFIG_USB_MDC800 is not set
 # CONFIG_USB_MICROTEK is not set
 
-#
-# USB Multimedia devices
-#
-# CONFIG_USB_DABUSB is not set
-# CONFIG_USB_VICAM is not set
-# CONFIG_USB_DSBR is not set
-# CONFIG_USB_ET61X251 is not set
-# CONFIG_USB_IBMCAM is not set
-# CONFIG_USB_KONICAWC is not set
-# CONFIG_USB_OV511 is not set
-# CONFIG_USB_SE401 is not set
-# CONFIG_USB_SN9C102 is not set
-# CONFIG_USB_STV680 is not set
-# CONFIG_USB_PWC is not set
-
 #
 # USB Network Adapters
 #
@@ -1299,12 +1169,11 @@ CONFIG_USB_STORAGE=y
 # CONFIG_USB_PEGASUS is not set
 # CONFIG_USB_RTL8150 is not set
 # CONFIG_USB_USBNET is not set
-# CONFIG_USB_MON is not set
+CONFIG_USB_MON=y
 
 #
 # USB port drivers
 #
-# CONFIG_USB_USS720 is not set
 
 #
 # USB Serial Converter support
@@ -1321,10 +1190,12 @@ CONFIG_USB_STORAGE=y
 # CONFIG_USB_LEGOTOWER is not set
 # CONFIG_USB_LCD is not set
 # CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
 # CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_PHIDGETKIT is not set
 # CONFIG_USB_PHIDGETSERVO is not set
 # CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_APPLEDISPLAY is not set
 # CONFIG_USB_SISUSBVGA is not set
 # CONFIG_USB_LD is not set
 # CONFIG_USB_TEST is not set
@@ -1343,57 +1214,97 @@ CONFIG_USB_STORAGE=y
 #
 # CONFIG_MMC is not set
 
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
 #
 # InfiniBand support
 #
 # CONFIG_INFINIBAND is not set
 
 #
-# SN Devices
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+# CONFIG_EDAC is not set
+
+#
+# Real Time Clock
 #
+# CONFIG_RTC_CLASS is not set
 
 #
-# EDAC - error detection and reporting (RAS)
+# DMA Engine support
+#
+# CONFIG_DMA_ENGINE is not set
+
+#
+# DMA Clients
+#
+
+#
+# DMA Devices
 #
-# CONFIG_EDAC is not set
 
 #
 # File systems
 #
 CONFIG_EXT2_FS=y
-# CONFIG_EXT2_FS_XATTR is not set
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT2_FS_POSIX_ACL=y
+# CONFIG_EXT2_FS_SECURITY is not set
 # CONFIG_EXT2_FS_XIP is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_REISERFS_FS is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+CONFIG_EXT3_FS_POSIX_ACL=y
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+CONFIG_REISERFS_FS=y
+# CONFIG_REISERFS_CHECK is not set
+# CONFIG_REISERFS_PROC_INFO is not set
+CONFIG_REISERFS_FS_XATTR=y
+CONFIG_REISERFS_FS_POSIX_ACL=y
+# CONFIG_REISERFS_FS_SECURITY is not set
 # CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
+CONFIG_FS_POSIX_ACL=y
 # CONFIG_XFS_FS is not set
 # CONFIG_OCFS2_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_ROMFS_FS is not set
-# CONFIG_INOTIFY is not set
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
 CONFIG_DNOTIFY=y
 # CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
+CONFIG_AUTOFS4_FS=y
 # CONFIG_FUSE_FS is not set
 
 #
 # CD-ROM/DVD Filesystems
 #
 CONFIG_ISO9660_FS=y
-CONFIG_JOLIET=y
-CONFIG_ZISOFS=y
-CONFIG_ZISOFS_FS=y
+# CONFIG_JOLIET is not set
+# CONFIG_ZISOFS is not set
 # CONFIG_UDF_FS is not set
 
 #
 # DOS/FAT/NT Filesystems
 #
 CONFIG_FAT_FS=y
-# CONFIG_MSDOS_FS is not set
+CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
-CONFIG_FAT_DEFAULT_CODEPAGE=850
+CONFIG_FAT_DEFAULT_CODEPAGE=437
 CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 # CONFIG_NTFS_FS is not set
 
@@ -1404,10 +1315,9 @@ CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
 CONFIG_TMPFS=y
-# CONFIG_HUGETLBFS is not set
-# CONFIG_HUGETLB_PAGE is not set
+CONFIG_HUGETLBFS=y
+CONFIG_HUGETLB_PAGE=y
 CONFIG_RAMFS=y
-# CONFIG_RELAYFS_FS is not set
 # CONFIG_CONFIGFS_FS is not set
 
 #
@@ -1430,13 +1340,26 @@ CONFIG_RAMFS=y
 #
 # Network File Systems
 #
-# CONFIG_NFS_FS is not set
-# CONFIG_NFSD is not set
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
 # CONFIG_SMB_FS is not set
-CONFIG_CIFS=y
-# CONFIG_CIFS_STATS is not set
-# CONFIG_CIFS_XATTR is not set
-# CONFIG_CIFS_EXPERIMENTAL is not set
+# CONFIG_CIFS is not set
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
@@ -1445,33 +1368,18 @@ CONFIG_CIFS=y
 #
 # Partition Types
 #
-CONFIG_PARTITION_ADVANCED=y
-# CONFIG_ACORN_PARTITION is not set
-# CONFIG_OSF_PARTITION is not set
-# CONFIG_AMIGA_PARTITION is not set
-# CONFIG_ATARI_PARTITION is not set
-# CONFIG_MAC_PARTITION is not set
+# CONFIG_PARTITION_ADVANCED is not set
 CONFIG_MSDOS_PARTITION=y
-# CONFIG_BSD_DISKLABEL is not set
-# CONFIG_MINIX_SUBPARTITION is not set
-# CONFIG_SOLARIS_X86_PARTITION is not set
-# CONFIG_UNIXWARE_DISKLABEL is not set
-# CONFIG_LDM_PARTITION is not set
-# CONFIG_SGI_PARTITION is not set
-# CONFIG_ULTRIX_PARTITION is not set
-# CONFIG_SUN_PARTITION is not set
-# CONFIG_KARMA_PARTITION is not set
-# CONFIG_EFI_PARTITION is not set
 
 #
 # Native Language Support
 #
 CONFIG_NLS=y
-CONFIG_NLS_DEFAULT="iso8859-15"
-# CONFIG_NLS_CODEPAGE_437 is not set
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
 # CONFIG_NLS_CODEPAGE_737 is not set
 # CONFIG_NLS_CODEPAGE_775 is not set
-CONFIG_NLS_CODEPAGE_850=y
+# CONFIG_NLS_CODEPAGE_850 is not set
 # CONFIG_NLS_CODEPAGE_852 is not set
 # CONFIG_NLS_CODEPAGE_855 is not set
 # CONFIG_NLS_CODEPAGE_857 is not set
@@ -1491,7 +1399,7 @@ CONFIG_NLS_CODEPAGE_850=y
 # CONFIG_NLS_ISO8859_8 is not set
 # CONFIG_NLS_CODEPAGE_1250 is not set
 # CONFIG_NLS_CODEPAGE_1251 is not set
-# CONFIG_NLS_ASCII is not set
+CONFIG_NLS_ASCII=y
 CONFIG_NLS_ISO8859_1=y
 # CONFIG_NLS_ISO8859_2 is not set
 # CONFIG_NLS_ISO8859_3 is not set
@@ -1510,20 +1418,50 @@ CONFIG_NLS_UTF8=y
 #
 # Instrumentation Support
 #
-# CONFIG_PROFILING is not set
-# CONFIG_KPROBES is not set
+CONFIG_PROFILING=y
+CONFIG_OPROFILE=y
+CONFIG_KPROBES=y
 
 #
 # Kernel hacking
 #
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
 CONFIG_MAGIC_SYSRQ=y
-# CONFIG_DEBUG_KERNEL is not set
-CONFIG_LOG_BUF_SHIFT=14
+CONFIG_UNUSED_SYMBOLS=y
+CONFIG_DEBUG_KERNEL=y
+CONFIG_LOG_BUF_SHIFT=18
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_DEBUG_SLAB is not set
+# CONFIG_DEBUG_RT_MUTEXES is not set
+# CONFIG_RT_MUTEX_TESTER is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_RWSEMS is not set
+# CONFIG_DEBUG_LOCK_ALLOC is not set
+# CONFIG_PROVE_LOCKING is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_HIGHMEM is not set
 CONFIG_DEBUG_BUGVERBOSE=y
+# CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_DEBUG_VM is not set
+# CONFIG_FRAME_POINTER is not set
+CONFIG_UNWIND_INFO=y
+CONFIG_STACK_UNWIND=y
+# CONFIG_FORCED_INLINING is not set
+# CONFIG_RCU_TORTURE_TEST is not set
 CONFIG_EARLY_PRINTK=y
+CONFIG_DEBUG_STACKOVERFLOW=y
+# CONFIG_DEBUG_STACK_USAGE is not set
+# CONFIG_DEBUG_RODATA is not set
+# CONFIG_4KSTACKS is not set
 CONFIG_X86_FIND_SMP_CONFIG=y
 CONFIG_X86_MPPARSE=y
+CONFIG_DOUBLEFAULT=y
 
 #
 # Security options
@@ -1536,10 +1474,6 @@ CONFIG_X86_MPPARSE=y
 #
 # CONFIG_CRYPTO is not set
 
-#
-# Hardware crypto devices
-#
-
 #
 # Library routines
 #
@@ -1548,7 +1482,12 @@ CONFIG_X86_MPPARSE=y
 CONFIG_CRC32=y
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
+CONFIG_PLIST=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_PENDING_IRQ=y
+CONFIG_X86_SMP=y
+CONFIG_X86_HT=y
 CONFIG_X86_BIOS_REBOOT=y
+CONFIG_X86_TRAMPOLINE=y
 CONFIG_KTIME_SCALAR=y
-- 
cgit v1.2.1


From 874c4fe389d1358f82c96dc9b5092fc5c7690604 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Allow to use GENERICARCH for UP kernels

There are some machines around (large xSeries or Unisys ES7000) that
need physical IO-APIC destination mode to access all of their IO
devices. This currently doesn't work in UP kernels as used in
distribution installers.

This patch allows to compile even UP kernels as GENERICARCH which
allows to use physical or clustered APIC mode.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Kconfig                        |  7 ++--
 arch/i386/kernel/io_apic.c               |  1 +
 arch/i386/kernel/mpparse.c               |  1 +
 arch/i386/mach-generic/bigsmp.c          |  1 +
 arch/i386/mach-generic/es7000.c          |  1 +
 arch/i386/mach-generic/probe.c           |  2 +
 arch/i386/mach-generic/summit.c          |  1 +
 include/asm-i386/genapic.h               | 69 +++++++++++++++++++-------------
 include/asm-i386/mach-es7000/mach_apic.h |  4 ++
 include/asm-i386/mach-summit/mach_apic.h | 11 ++++-
 include/asm-i386/smp.h                   | 19 +++++----
 11 files changed, 76 insertions(+), 41 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index b2751eadbc56..c134a545b37e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -166,7 +166,6 @@ config X86_VISWS
 
 config X86_GENERICARCH
        bool "Generic architecture (Summit, bigsmp, ES7000, default)"
-       depends on SMP
        help
           This option compiles in the Summit, bigsmp, ES7000, default subarchitectures.
 	  It is intended for a generic binary kernel.
@@ -263,7 +262,7 @@ source "kernel/Kconfig.preempt"
 
 config X86_UP_APIC
 	bool "Local APIC support on uniprocessors"
-	depends on !SMP && !(X86_VISWS || X86_VOYAGER)
+	depends on !SMP && !(X86_VISWS || X86_VOYAGER || X86_GENERICARCH)
 	help
 	  A local APIC (Advanced Programmable Interrupt Controller) is an
 	  integrated interrupt controller in the CPU. If you have a single-CPU
@@ -288,12 +287,12 @@ config X86_UP_IOAPIC
 
 config X86_LOCAL_APIC
 	bool
-	depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER)
+	depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER) || X86_GENERICARCH
 	default y
 
 config X86_IO_APIC
 	bool
-	depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER))
+	depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER)) || X86_GENERICARCH
 	default y
 
 config X86_VISWS_APIC
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 4fb32c551fe0..4eacd52eeb74 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -40,6 +40,7 @@
 #include <asm/nmi.h>
 
 #include <mach_apic.h>
+#include <mach_apicdef.h>
 
 #include "io_ports.h"
 
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index a70b5fa0ef06..827569688562 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -30,6 +30,7 @@
 #include <asm/io_apic.h>
 
 #include <mach_apic.h>
+#include <mach_apicdef.h>
 #include <mach_mpparse.h>
 #include <bios_ebda.h>
 
diff --git a/arch/i386/mach-generic/bigsmp.c b/arch/i386/mach-generic/bigsmp.c
index ef7a6e6fcb9f..33d9f93557ba 100644
--- a/arch/i386/mach-generic/bigsmp.c
+++ b/arch/i386/mach-generic/bigsmp.c
@@ -5,6 +5,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/arch/i386/mach-generic/es7000.c b/arch/i386/mach-generic/es7000.c
index 845cdd0b3593..aa144d82334d 100644
--- a/arch/i386/mach-generic/es7000.c
+++ b/arch/i386/mach-generic/es7000.c
@@ -4,6 +4,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/arch/i386/mach-generic/probe.c b/arch/i386/mach-generic/probe.c
index bcd1bcfaa723..793d1b473251 100644
--- a/arch/i386/mach-generic/probe.c
+++ b/arch/i386/mach-generic/probe.c
@@ -119,7 +119,9 @@ int __init acpi_madt_oem_check(char *oem_id, char *oem_table_id)
 	return 0;	
 }
 
+#ifdef CONFIG_SMP
 int hard_smp_processor_id(void)
 {
 	return genapic->get_apic_id(*(unsigned long *)(APIC_BASE+APIC_ID));
 }
+#endif
diff --git a/arch/i386/mach-generic/summit.c b/arch/i386/mach-generic/summit.c
index b73501ddd653..f7e5d66648dc 100644
--- a/arch/i386/mach-generic/summit.c
+++ b/arch/i386/mach-generic/summit.c
@@ -4,6 +4,7 @@
 #define APIC_DEFINITION 1
 #include <linux/threads.h>
 #include <linux/cpumask.h>
+#include <asm/smp.h>
 #include <asm/mpspec.h>
 #include <asm/genapic.h>
 #include <asm/fixmap.h>
diff --git a/include/asm-i386/genapic.h b/include/asm-i386/genapic.h
index b3783a32abee..8ffbb0f07457 100644
--- a/include/asm-i386/genapic.h
+++ b/include/asm-i386/genapic.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_GENAPIC_H
 #define _ASM_GENAPIC_H 1
 
+#include <asm/mpspec.h>
+
 /*
  * Generic APIC driver interface.
  *
@@ -63,14 +65,25 @@ struct genapic {
 	unsigned (*get_apic_id)(unsigned long x);
 	unsigned long apic_id_mask;
 	unsigned int (*cpu_mask_to_apicid)(cpumask_t cpumask);
-	
+
+#ifdef CONFIG_SMP
 	/* ipi */
 	void (*send_IPI_mask)(cpumask_t mask, int vector);
 	void (*send_IPI_allbutself)(int vector);
 	void (*send_IPI_all)(int vector);
+#endif
 }; 
 
-#define APICFUNC(x) .x = x
+#define APICFUNC(x) .x = x,
+
+/* More functions could be probably marked IPIFUNC and save some space
+   in UP GENERICARCH kernels, but I don't have the nerve right now
+   to untangle this mess. -AK  */
+#ifdef CONFIG_SMP
+#define IPIFUNC(x) APICFUNC(x)
+#else
+#define IPIFUNC(x)
+#endif
 
 #define APIC_INIT(aname, aprobe) { \
 	.name = aname, \
@@ -80,33 +93,33 @@ struct genapic {
 	.no_balance_irq = NO_BALANCE_IRQ, \
 	.ESR_DISABLE = esr_disable, \
 	.apic_destination_logical = APIC_DEST_LOGICAL, \
-	APICFUNC(apic_id_registered), \
-	APICFUNC(target_cpus), \
-	APICFUNC(check_apicid_used), \
-	APICFUNC(check_apicid_present), \
-	APICFUNC(init_apic_ldr), \
-	APICFUNC(ioapic_phys_id_map), \
-	APICFUNC(clustered_apic_check), \
-	APICFUNC(multi_timer_check), \
-	APICFUNC(apicid_to_node), \
-	APICFUNC(cpu_to_logical_apicid), \
-	APICFUNC(cpu_present_to_apicid), \
-	APICFUNC(apicid_to_cpu_present), \
-	APICFUNC(mpc_apic_id), \
-	APICFUNC(setup_portio_remap), \
-	APICFUNC(check_phys_apicid_present), \
-	APICFUNC(mpc_oem_bus_info), \
-	APICFUNC(mpc_oem_pci_bus), \
-	APICFUNC(mps_oem_check), \
-	APICFUNC(get_apic_id), \
+	APICFUNC(apic_id_registered) \
+	APICFUNC(target_cpus) \
+	APICFUNC(check_apicid_used) \
+	APICFUNC(check_apicid_present) \
+	APICFUNC(init_apic_ldr) \
+	APICFUNC(ioapic_phys_id_map) \
+	APICFUNC(clustered_apic_check) \
+	APICFUNC(multi_timer_check) \
+	APICFUNC(apicid_to_node) \
+	APICFUNC(cpu_to_logical_apicid) \
+	APICFUNC(cpu_present_to_apicid) \
+	APICFUNC(apicid_to_cpu_present) \
+	APICFUNC(mpc_apic_id) \
+	APICFUNC(setup_portio_remap) \
+	APICFUNC(check_phys_apicid_present) \
+	APICFUNC(mpc_oem_bus_info) \
+	APICFUNC(mpc_oem_pci_bus) \
+	APICFUNC(mps_oem_check) \
+	APICFUNC(get_apic_id) \
 	.apic_id_mask = APIC_ID_MASK, \
-	APICFUNC(cpu_mask_to_apicid), \
-	APICFUNC(acpi_madt_oem_check), \
-	APICFUNC(send_IPI_mask), \
-	APICFUNC(send_IPI_allbutself), \
-	APICFUNC(send_IPI_all), \
-	APICFUNC(enable_apic_mode), \
-	APICFUNC(phys_pkg_id), \
+	APICFUNC(cpu_mask_to_apicid) \
+	APICFUNC(acpi_madt_oem_check) \
+	IPIFUNC(send_IPI_mask) \
+	IPIFUNC(send_IPI_allbutself) \
+	IPIFUNC(send_IPI_all) \
+	APICFUNC(enable_apic_mode) \
+	APICFUNC(phys_pkg_id) \
 	}
 
 extern struct genapic *genapic;
diff --git a/include/asm-i386/mach-es7000/mach_apic.h b/include/asm-i386/mach-es7000/mach_apic.h
index b5f3f0d0b2bc..26333685a7fb 100644
--- a/include/asm-i386/mach-es7000/mach_apic.h
+++ b/include/asm-i386/mach-es7000/mach_apic.h
@@ -123,9 +123,13 @@ extern u8 cpu_2_logical_apicid[];
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
 {
+#ifdef CONFIG_SMP
        if (cpu >= NR_CPUS)
 	       return BAD_APICID;
        return (int)cpu_2_logical_apicid[cpu];
+#else
+	return logical_smp_processor_id();
+#endif
 }
 
 static inline int mpc_apic_id(struct mpc_config_processor *m, struct mpc_config_translation *unused)
diff --git a/include/asm-i386/mach-summit/mach_apic.h b/include/asm-i386/mach-summit/mach_apic.h
index 9fd073286289..a81b05961595 100644
--- a/include/asm-i386/mach-summit/mach_apic.h
+++ b/include/asm-i386/mach-summit/mach_apic.h
@@ -46,10 +46,12 @@ extern u8 cpu_2_logical_apicid[];
 static inline void init_apic_ldr(void)
 {
 	unsigned long val, id;
-	int i, count;
-	u8 lid;
+	int count = 0;
 	u8 my_id = (u8)hard_smp_processor_id();
 	u8 my_cluster = (u8)apicid_cluster(my_id);
+#ifdef CONFIG_SMP
+	u8 lid;
+	int i;
 
 	/* Create logical APIC IDs by counting CPUs already in cluster. */
 	for (count = 0, i = NR_CPUS; --i >= 0; ) {
@@ -57,6 +59,7 @@ static inline void init_apic_ldr(void)
 		if (lid != BAD_APICID && apicid_cluster(lid) == my_cluster)
 			++count;
 	}
+#endif
 	/* We only have a 4 wide bitmap in cluster mode.  If a deranged
 	 * BIOS puts 5 CPUs in one APIC cluster, we're hosed. */
 	BUG_ON(count >= XAPIC_DEST_CPUS_SHIFT);
@@ -91,9 +94,13 @@ static inline int apicid_to_node(int logical_apicid)
 /* Mapping from cpu number to logical apicid */
 static inline int cpu_to_logical_apicid(int cpu)
 {
+#ifdef CONFIG_SMP
        if (cpu >= NR_CPUS)
 	       return BAD_APICID;
 	return (int)cpu_2_logical_apicid[cpu];
+#else
+	return logical_smp_processor_id();
+#endif
 }
 
 static inline int cpu_present_to_apicid(int mps_cpu)
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 142d10e34ade..f87826039a51 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -80,17 +80,11 @@ static inline int hard_smp_processor_id(void)
 	return GET_APIC_ID(*(unsigned long *)(APIC_BASE+APIC_ID));
 }
 #endif
-
-static __inline int logical_smp_processor_id(void)
-{
-	/* we don't want to mark this access volatile - bad code generation */
-	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
-}
-
 #endif
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
+
 #endif /* !__ASSEMBLY__ */
 
 #else /* CONFIG_SMP */
@@ -100,4 +94,15 @@ extern void __cpu_die(unsigned int cpu);
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 #endif
+
+#ifndef __ASSEMBLY__
+#ifdef CONFIG_X86_LOCAL_APIC
+static __inline int logical_smp_processor_id(void)
+{
+	/* we don't want to mark this access volatile - bad code generation */
+	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
+}
+#endif
+#endif
+
 #endif
-- 
cgit v1.2.1


From b07f8915cda3fcd73b8b68075ba1e6cd0673365d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Temporarily revert parts of the Core 2 nmi nmi watchdog
 support

This makes merging easier.  They are readded a few patches later.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c                  | 65 +-------------------------
 arch/x86_64/kernel/nmi.c                | 81 ++-------------------------------
 include/asm-i386/intel_arch_perfmon.h   | 19 --------
 include/asm-x86_64/intel_arch_perfmon.h | 19 --------
 4 files changed, 6 insertions(+), 178 deletions(-)
 delete mode 100644 include/asm-i386/intel_arch_perfmon.h
 delete mode 100644 include/asm-x86_64/intel_arch_perfmon.h

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index acb351478e42..1282d70ff971 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -24,7 +24,6 @@
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
-#include <asm/intel_arch_perfmon.h>
 
 #include "mach_traps.h"
 
@@ -96,9 +95,6 @@ int nmi_active;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
-#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
-#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
-
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -211,8 +207,6 @@ static int __init setup_nmi_watchdog(char *str)
 
 __setup("nmi_watchdog=", setup_nmi_watchdog);
 
-static void disable_intel_arch_watchdog(void);
-
 static void disable_lapic_nmi_watchdog(void)
 {
 	if (nmi_active <= 0)
@@ -222,10 +216,6 @@ static void disable_lapic_nmi_watchdog(void)
 		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			disable_intel_arch_watchdog();
-			break;
-		}
 		switch (boot_cpu_data.x86) {
 		case 6:
 			if (boot_cpu_data.x86_model > 0xd)
@@ -454,53 +444,6 @@ static int setup_p4_watchdog(void)
 	return 1;
 }
 
-static void disable_intel_arch_watchdog(void)
-{
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if (!(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, 0, 0);
-}
-
-static int setup_intel_arch_watchdog(void)
-{
-	unsigned int evntsel;
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if ((ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		return 0;
-
-	nmi_perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
-
-	clear_msr_range(MSR_ARCH_PERFMON_EVENTSEL0, 2);
-	clear_msr_range(MSR_ARCH_PERFMON_PERFCTR0, 2);
-
-	evntsel = ARCH_PERFMON_EVENTSEL_INT
-		| ARCH_PERFMON_EVENTSEL_OS
-		| ARCH_PERFMON_EVENTSEL_USR
-		| ARCH_PERFMON_NMI_EVENT_SEL
-		| ARCH_PERFMON_NMI_EVENT_UMASK;
-
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	write_watchdog_counter("INTEL_ARCH_PERFCTR0");
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	return 1;
-}
-
 void setup_apic_nmi_watchdog (void)
 {
 	switch (boot_cpu_data.x86_vendor) {
@@ -510,11 +453,6 @@ void setup_apic_nmi_watchdog (void)
 		setup_k7_watchdog();
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			if (!setup_intel_arch_watchdog())
-				return;
-			break;
-		}
 		switch (boot_cpu_data.x86) {
 		case 6:
 			if (boot_cpu_data.x86_model > 0xd)
@@ -619,8 +557,7 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 			apic_write(APIC_LVTPC, APIC_DM_NMI);
 		}
-		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0 ||
-		         nmi_perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) {
 			/* Only P6 based Pentium M need to re-unmask
 			 * the apic vector but it doesn't hurt
 			 * other P6 variant */
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5baa0c726e97..42c05d6907b9 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -26,7 +26,6 @@
 #include <asm/proto.h>
 #include <asm/kdebug.h>
 #include <asm/mce.h>
-#include <asm/intel_arch_perfmon.h>
 
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
@@ -66,9 +65,6 @@ static unsigned int nmi_p4_cccr_val;
 #define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
 #define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
 
-#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
-#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
-
 #define MSR_P4_MISC_ENABLE	0x1A0
 #define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
 #define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
@@ -100,10 +96,7 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return boot_cpu_data.x86 == 15;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
-			return 1;
-		else
-			return (boot_cpu_data.x86 == 15);
+		return boot_cpu_data.x86 == 15;
 	}
 	return 0;
 }
@@ -209,8 +202,6 @@ int __init setup_nmi_watchdog(char *str)
 
 __setup("nmi_watchdog=", setup_nmi_watchdog);
 
-static void disable_intel_arch_watchdog(void);
-
 static void disable_lapic_nmi_watchdog(void)
 {
 	if (nmi_active <= 0)
@@ -223,8 +214,6 @@ static void disable_lapic_nmi_watchdog(void)
 		if (boot_cpu_data.x86 == 15) {
 			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
 			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-		} else if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			disable_intel_arch_watchdog();
 		}
 		break;
 	}
@@ -377,53 +366,6 @@ static void setup_k7_watchdog(void)
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
 }
 
-static void disable_intel_arch_watchdog(void)
-{
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if (!(ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, 0, 0);
-}
-
-static int setup_intel_arch_watchdog(void)
-{
-	unsigned int evntsel;
-	unsigned ebx;
-
-	/*
-	 * Check whether the Architectural PerfMon supports
-	 * Unhalted Core Cycles Event or not.
-	 * NOTE: Corresponding bit = 0 in ebp indicates event present.
-	 */
-	ebx = cpuid_ebx(10);
-	if ((ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
-		return 0;
-
-	nmi_perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
-
-	clear_msr_range(MSR_ARCH_PERFMON_EVENTSEL0, 2);
-	clear_msr_range(MSR_ARCH_PERFMON_PERFCTR0, 2);
-
-	evntsel = ARCH_PERFMON_EVENTSEL_INT
-		| ARCH_PERFMON_EVENTSEL_OS
-		| ARCH_PERFMON_EVENTSEL_USR
-		| ARCH_PERFMON_NMI_EVENT_SEL
-		| ARCH_PERFMON_NMI_EVENT_UMASK;
-
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	wrmsrl(MSR_ARCH_PERFMON_PERFCTR0, -((u64)cpu_khz * 1000 / nmi_hz));
-	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
-	wrmsr(MSR_ARCH_PERFMON_EVENTSEL0, evntsel, 0);
-	return 1;
-}
-
 
 static int setup_p4_watchdog(void)
 {
@@ -477,16 +419,10 @@ void setup_apic_nmi_watchdog(void)
 		setup_k7_watchdog();
 		break;
 	case X86_VENDOR_INTEL:
-		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
-			if (!setup_intel_arch_watchdog())
-				return;
-		} else if (boot_cpu_data.x86 == 15) {
-			if (!setup_p4_watchdog())
-				return;
-		} else {
+		if (boot_cpu_data.x86 != 15)
+			return;
+		if (!setup_p4_watchdog())
 			return;
-		}
-
 		break;
 
 	default:
@@ -571,14 +507,7 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
  			 */
  			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
  			apic_write(APIC_LVTPC, APIC_DM_NMI);
- 		} else if (nmi_perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
-			/*
-			 * For Intel based architectural perfmon
-			 * - LVTPC is masked on interrupt and must be
-			 *   unmasked by the LVTPC handler.
-			 */
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
-		}
+ 		}
 		wrmsrl(nmi_perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	}
 }
diff --git a/include/asm-i386/intel_arch_perfmon.h b/include/asm-i386/intel_arch_perfmon.h
deleted file mode 100644
index 134ea9cc5283..000000000000
--- a/include/asm-i386/intel_arch_perfmon.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef X86_INTEL_ARCH_PERFMON_H
-#define X86_INTEL_ARCH_PERFMON_H 1
-
-#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
-#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
-
-#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
-#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
-
-#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
-#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
-#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
-#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
-
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT (1 << 0)
-
-#endif	/* X86_INTEL_ARCH_PERFMON_H */
diff --git a/include/asm-x86_64/intel_arch_perfmon.h b/include/asm-x86_64/intel_arch_perfmon.h
deleted file mode 100644
index 59c396431569..000000000000
--- a/include/asm-x86_64/intel_arch_perfmon.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef X86_64_INTEL_ARCH_PERFMON_H
-#define X86_64_INTEL_ARCH_PERFMON_H 1
-
-#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
-#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
-
-#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
-#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
-
-#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
-#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
-#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
-#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
-
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
-#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT (1 << 0)
-
-#endif	/* X86_64_INTEL_ARCH_PERFMON_H */
-- 
cgit v1.2.1


From 828f0afda123a96ff4e8078f057a302f4b4232ae Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Add performance counter reservation framework for UP
 kernels

Adds basic infrastructure to allow subsystems to reserve performance
counters on the x86 chips.  Only UP kernels are supported in this patch to
make reviewing easier.  The SMP portion makes a lot more changes.

Think of this as a locking mechanism where each bit represents a different
counter.  In addition, each subsystem should also reserve an appropriate
event selection register that will correspond to the performance counter it
will be using (this is mainly neccessary for the Pentium 4 chips as they
break the 1:1 relationship to performance counters).

This will help prevent subsystems like oprofile from interfering with the
nmi watchdog.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c   | 188 +++++++++++++++++++++++++++++++++++++++--------
 arch/x86_64/kernel/nmi.c | 178 ++++++++++++++++++++++++++++++++++----------
 include/asm-i386/nmi.h   |   7 ++
 include/asm-x86_64/nmi.h |   8 +-
 4 files changed, 308 insertions(+), 73 deletions(-)

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 1282d70ff971..5d58dfeacd59 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -34,6 +34,20 @@ static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
 static unsigned int nmi_p4_cccr_val;
 extern void show_registers(struct pt_regs *regs);
 
+/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
+ * evtsel_nmi_owner tracks the ownership of the event selection
+ * - different performance counters/ event selection may be reserved for
+ *   different subsystems this reservation system just tries to coordinate
+ *   things a little
+ */
+static DEFINE_PER_CPU(unsigned long, perfctr_nmi_owner);
+static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]);
+
+/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
+ * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
+ */
+#define NMI_MAX_COUNTER_BITS 66
+
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
  * - it may be reserved by some other driver, or not
@@ -95,6 +109,105 @@ int nmi_active;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the performance counter register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_PERFCTR0);
+	case X86_VENDOR_INTEL:
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_PERFCTR0);
+		case 15:
+			return (msr - MSR_P4_BPU_PERFCTR0);
+		}
+	}
+	return 0;
+}
+
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the event selection register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_EVNTSEL0);
+	case X86_VENDOR_INTEL:
+		switch (boot_cpu_data.x86) {
+		case 6:
+			return (msr - MSR_P6_EVNTSEL0);
+		case 15:
+			return (msr - MSR_P4_BSU_ESCR0);
+		}
+	}
+	return 0;
+}
+
+/* checks for a bit availability (hack for oprofile) */
+int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
+{
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+/* checks the an msr for availability */
+int avail_to_resrv_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(perfctr_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(perfctr_nmi_owner));
+}
+
+int reserve_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]))
+		return 1;
+	return 0;
+}
+
+void release_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]);
+}
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -344,14 +457,6 @@ late_initcall(init_lapic_nmi_sysfs);
  * Original code written by Keith Owens.
  */
 
-static void clear_msr_range(unsigned int base, unsigned int n)
-{
-	unsigned int i;
-
-	for(i = 0; i < n; ++i)
-		wrmsr(base+i, 0, 0);
-}
-
 static void write_watchdog_counter(const char *descr)
 {
 	u64 count = (u64)cpu_khz * 1000;
@@ -362,14 +467,19 @@ static void write_watchdog_counter(const char *descr)
 	wrmsrl(nmi_perfctr_msr, 0 - count);
 }
 
-static void setup_k7_watchdog(void)
+static int setup_k7_watchdog(void)
 {
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_K7_PERFCTR0;
 
-	clear_msr_range(MSR_K7_EVNTSEL0, 4);
-	clear_msr_range(MSR_K7_PERFCTR0, 4);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+		goto fail1;
+
+	wrmsrl(MSR_K7_PERFCTR0, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
@@ -381,16 +491,24 @@ static void setup_k7_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
-static void setup_p6_watchdog(void)
+static int setup_p6_watchdog(void)
 {
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_P6_PERFCTR0;
 
-	clear_msr_range(MSR_P6_EVNTSEL0, 2);
-	clear_msr_range(MSR_P6_PERFCTR0, 2);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P6_EVNTSEL0))
+		goto fail1;
 
 	evntsel = P6_EVNTSEL_INT
 		| P6_EVNTSEL_OS
@@ -402,6 +520,11 @@ static void setup_p6_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= P6_EVNTSEL0_ENABLE;
 	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
+	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 static int setup_p4_watchdog(void)
@@ -419,22 +542,11 @@ static int setup_p4_watchdog(void)
 		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
 #endif
 
-	if (!(misc_enable & MSR_P4_MISC_ENABLE_PEBS_UNAVAIL))
-		clear_msr_range(0x3F1, 2);
-	/* MSR 0x3F0 seems to have a default value of 0xFC00, but current
-	   docs doesn't fully define it, so leave it alone for now. */
-	if (boot_cpu_data.x86_model >= 0x3) {
-		/* MSR_P4_IQ_ESCR0/1 (0x3ba/0x3bb) removed */
-		clear_msr_range(0x3A0, 26);
-		clear_msr_range(0x3BC, 3);
-	} else {
-		clear_msr_range(0x3A0, 31);
-	}
-	clear_msr_range(0x3C0, 6);
-	clear_msr_range(0x3C8, 6);
-	clear_msr_range(0x3E0, 2);
-	clear_msr_range(MSR_P4_CCCR0, 18);
-	clear_msr_range(MSR_P4_PERFCTR0, 18);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+		goto fail1;
 
 	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
 	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
@@ -442,6 +554,10 @@ static int setup_p4_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 void setup_apic_nmi_watchdog (void)
@@ -450,7 +566,8 @@ void setup_apic_nmi_watchdog (void)
 	case X86_VENDOR_AMD:
 		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
 			return;
-		setup_k7_watchdog();
+		if (!setup_k7_watchdog())
+			return;
 		break;
 	case X86_VENDOR_INTEL:
 		switch (boot_cpu_data.x86) {
@@ -458,7 +575,8 @@ void setup_apic_nmi_watchdog (void)
 			if (boot_cpu_data.x86_model > 0xd)
 				return;
 
-			setup_p6_watchdog();
+			if(!setup_p6_watchdog())
+				return;
 			break;
 		case 15:
 			if (boot_cpu_data.x86_model > 0x4)
@@ -612,6 +730,12 @@ int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file,
 
 EXPORT_SYMBOL(nmi_active);
 EXPORT_SYMBOL(nmi_watchdog);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(release_perfctr_nmi);
+EXPORT_SYMBOL(reserve_evntsel_nmi);
+EXPORT_SYMBOL(release_evntsel_nmi);
 EXPORT_SYMBOL(reserve_lapic_nmi);
 EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 42c05d6907b9..b7a7c9973849 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -27,6 +27,20 @@
 #include <asm/kdebug.h>
 #include <asm/mce.h>
 
+/* perfctr_nmi_owner tracks the ownership of the perfctr registers:
+ * evtsel_nmi_owner tracks the ownership of the event selection
+ * - different performance counters/ event selection may be reserved for
+ *   different subsystems this reservation system just tries to coordinate
+ *   things a little
+ */
+static DEFINE_PER_CPU(unsigned, perfctr_nmi_owner);
+static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]);
+
+/* this number is calculated from Intel's MSR_P4_CRU_ESCR5 register and it's
+ * offset from MSR_P4_BSU_ESCR0.  It will be the max for all platforms (for now)
+ */
+#define NMI_MAX_COUNTER_BITS 66
+
 /*
  * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
  * - it may be reserved by some other driver, or not
@@ -90,6 +104,95 @@ static unsigned int nmi_p4_cccr_val;
 	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
 	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
 
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the performance counter register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_PERFCTR0);
+	case X86_VENDOR_INTEL:
+		return (msr - MSR_P4_BPU_PERFCTR0);
+	}
+	return 0;
+}
+
+/* converts an msr to an appropriate reservation bit */
+static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
+{
+	/* returns the bit offset of the event selection register */
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return (msr - MSR_K7_EVNTSEL0);
+	case X86_VENDOR_INTEL:
+		return (msr - MSR_P4_BSU_ESCR0);
+	}
+	return 0;
+}
+
+/* checks for a bit availability (hack for oprofile) */
+int avail_to_resrv_perfctr_nmi_bit(unsigned int counter)
+{
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+/* checks the an msr for availability */
+int avail_to_resrv_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	return (!test_bit(counter, &__get_cpu_var(perfctr_nmi_owner)));
+}
+
+int reserve_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(perfctr_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_perfctr_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_perfctr_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(perfctr_nmi_owner));
+}
+
+int reserve_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	if (!test_and_set_bit(counter, &__get_cpu_var(evntsel_nmi_owner)))
+		return 1;
+	return 0;
+}
+
+void release_evntsel_nmi(unsigned int msr)
+{
+	unsigned int counter;
+
+	counter = nmi_evntsel_msr_to_bit(msr);
+	BUG_ON(counter > NMI_MAX_COUNTER_BITS);
+
+	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner));
+}
+
 static __cpuinit inline int nmi_known_cpu(void)
 {
 	switch (boot_cpu_data.x86_vendor) {
@@ -325,34 +428,22 @@ late_initcall(init_lapic_nmi_sysfs);
 
 #endif	/* CONFIG_PM */
 
-/*
- * Activate the NMI watchdog via the local APIC.
- * Original code written by Keith Owens.
- */
-
-static void clear_msr_range(unsigned int base, unsigned int n)
+static int setup_k7_watchdog(void)
 {
-	unsigned int i;
-
-	for(i = 0; i < n; ++i)
-		wrmsr(base+i, 0, 0);
-}
-
-static void setup_k7_watchdog(void)
-{
-	int i;
 	unsigned int evntsel;
 
 	nmi_perfctr_msr = MSR_K7_PERFCTR0;
 
-	for(i = 0; i < 4; ++i) {
-		/* Simulator may not support it */
-		if (checking_wrmsrl(MSR_K7_EVNTSEL0+i, 0UL)) {
-			nmi_perfctr_msr = 0;
-			return;
-		}
-		wrmsrl(MSR_K7_PERFCTR0+i, 0UL);
-	}
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+		goto fail1;
+
+	/* Simulator may not support it */
+	if (checking_wrmsrl(MSR_K7_EVNTSEL0, 0UL))
+		goto fail2;
+	wrmsrl(MSR_K7_PERFCTR0, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
@@ -364,6 +455,13 @@ static void setup_k7_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
 	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	return 1;
+fail2:
+	release_evntsel_nmi(MSR_K7_EVNTSEL0);
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 
@@ -382,22 +480,11 @@ static int setup_p4_watchdog(void)
 		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
 #endif
 
-	if (!(misc_enable & MSR_P4_MISC_ENABLE_PEBS_UNAVAIL))
-		clear_msr_range(0x3F1, 2);
-	/* MSR 0x3F0 seems to have a default value of 0xFC00, but current
-	   docs doesn't fully define it, so leave it alone for now. */
-	if (boot_cpu_data.x86_model >= 0x3) {
-		/* MSR_P4_IQ_ESCR0/1 (0x3ba/0x3bb) removed */
-		clear_msr_range(0x3A0, 26);
-		clear_msr_range(0x3BC, 3);
-	} else {
-		clear_msr_range(0x3A0, 31);
-	}
-	clear_msr_range(0x3C0, 6);
-	clear_msr_range(0x3C8, 6);
-	clear_msr_range(0x3E0, 2);
-	clear_msr_range(MSR_P4_CCCR0, 18);
-	clear_msr_range(MSR_P4_PERFCTR0, 18);
+	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+		goto fail1;
 
 	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
 	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
@@ -406,6 +493,10 @@ static int setup_p4_watchdog(void)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
 	return 1;
+fail1:
+	release_perfctr_nmi(nmi_perfctr_msr);
+fail:
+	return 0;
 }
 
 void setup_apic_nmi_watchdog(void)
@@ -416,7 +507,8 @@ void setup_apic_nmi_watchdog(void)
 			return;
 		if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
 			return;
-		setup_k7_watchdog();
+		if (!setup_k7_watchdog())
+			return;
 		break;
 	case X86_VENDOR_INTEL:
 		if (boot_cpu_data.x86 != 15)
@@ -588,6 +680,12 @@ int proc_unknown_nmi_panic(struct ctl_table *table, int write, struct file *file
 
 EXPORT_SYMBOL(nmi_active);
 EXPORT_SYMBOL(nmi_watchdog);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi);
+EXPORT_SYMBOL(avail_to_resrv_perfctr_nmi_bit);
+EXPORT_SYMBOL(reserve_perfctr_nmi);
+EXPORT_SYMBOL(release_perfctr_nmi);
+EXPORT_SYMBOL(reserve_evntsel_nmi);
+EXPORT_SYMBOL(release_evntsel_nmi);
 EXPORT_SYMBOL(reserve_lapic_nmi);
 EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 67d994799999..27fc9e6f630e 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -25,6 +25,13 @@ void set_nmi_callback(nmi_callback_t callback);
  */
 void unset_nmi_callback(void);
 
+extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int avail_to_resrv_perfctr_nmi(unsigned int);
+extern int reserve_perfctr_nmi(unsigned int);
+extern void release_perfctr_nmi(unsigned int);
+extern int reserve_evntsel_nmi(unsigned int);
+extern void release_evntsel_nmi(unsigned int);
+
 extern void setup_apic_nmi_watchdog (void);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index efb45c894d76..62a784cb8f0c 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -56,7 +56,13 @@ extern int panic_on_timeout;
 extern int unknown_nmi_panic;
 
 extern int check_nmi_watchdog(void);
- 
+extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
+extern int avail_to_resrv_perfctr_nmi(unsigned int);
+extern int reserve_perfctr_nmi(unsigned int);
+extern void release_perfctr_nmi(unsigned int);
+extern int reserve_evntsel_nmi(unsigned int);
+extern void release_evntsel_nmi(unsigned int);
+
 extern void setup_apic_nmi_watchdog (void);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
-- 
cgit v1.2.1


From cb9c448c661d40ce2efbce8e9c19cc4d420d8ccc Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Utilize performance counter reservation framework in
 oprofile

Incorporates the new performance counter reservation system in oprofile.
Also cleans up a lot of the initialization code.  The code original zero'd
out every register associated with performance counters regardless if those
registers were used or not.  This causes issues with the nmi watchdog.
Now oprofile tries to reserve registers and gives up if it can't get them.

Cc: levon@movementarian.org
Cc: oprofile-list@lists.sf.net

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/oprofile/nmi_int.c         |  41 +++++++---
 arch/i386/oprofile/op_model_athlon.c |  54 ++++++++++---
 arch/i386/oprofile/op_model_p4.c     | 152 +++++++++++++++++------------------
 arch/i386/oprofile/op_model_ppro.c   |  65 ++++++++++++---
 arch/i386/oprofile/op_x86_model.h    |   1 +
 5 files changed, 199 insertions(+), 114 deletions(-)

diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index 5f8dc8a21bd7..8710ca081b1e 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -98,15 +98,19 @@ static void nmi_cpu_save_registers(struct op_msrs * msrs)
 	unsigned int i;
 
 	for (i = 0; i < nr_ctrs; ++i) {
-		rdmsr(counters[i].addr,
-			counters[i].saved.low,
-			counters[i].saved.high);
+		if (counters[i].addr){
+			rdmsr(counters[i].addr,
+				counters[i].saved.low,
+				counters[i].saved.high);
+		}
 	}
  
 	for (i = 0; i < nr_ctrls; ++i) {
-		rdmsr(controls[i].addr,
-			controls[i].saved.low,
-			controls[i].saved.high);
+		if (controls[i].addr){
+			rdmsr(controls[i].addr,
+				controls[i].saved.low,
+				controls[i].saved.high);
+		}
 	}
 }
 
@@ -205,15 +209,19 @@ static void nmi_restore_registers(struct op_msrs * msrs)
 	unsigned int i;
 
 	for (i = 0; i < nr_ctrls; ++i) {
-		wrmsr(controls[i].addr,
-			controls[i].saved.low,
-			controls[i].saved.high);
+		if (controls[i].addr){
+			wrmsr(controls[i].addr,
+				controls[i].saved.low,
+				controls[i].saved.high);
+		}
 	}
  
 	for (i = 0; i < nr_ctrs; ++i) {
-		wrmsr(counters[i].addr,
-			counters[i].saved.low,
-			counters[i].saved.high);
+		if (counters[i].addr){
+			wrmsr(counters[i].addr,
+				counters[i].saved.low,
+				counters[i].saved.high);
+		}
 	}
 }
  
@@ -234,6 +242,7 @@ static void nmi_cpu_shutdown(void * dummy)
 	apic_write(APIC_LVTPC, saved_lvtpc[cpu]);
 	apic_write(APIC_LVTERR, v);
 	nmi_restore_registers(msrs);
+	model->shutdown(msrs);
 }
 
  
@@ -284,6 +293,14 @@ static int nmi_create_files(struct super_block * sb, struct dentry * root)
 		struct dentry * dir;
 		char buf[4];
  
+ 		/* quick little hack to _not_ expose a counter if it is not
+		 * available for use.  This should protect userspace app.
+		 * NOTE:  assumes 1:1 mapping here (that counters are organized
+		 *        sequentially in their struct assignment).
+		 */
+		if (unlikely(!avail_to_resrv_perfctr_nmi_bit(i)))
+			continue;
+
 		snprintf(buf,  sizeof(buf), "%d", i);
 		dir = oprofilefs_mkdir(sb, root, buf);
 		oprofilefs_create_ulong(sb, dir, "enabled", &counter_config[i].enabled); 
diff --git a/arch/i386/oprofile/op_model_athlon.c b/arch/i386/oprofile/op_model_athlon.c
index 693bdea4a52b..3057a19e4641 100644
--- a/arch/i386/oprofile/op_model_athlon.c
+++ b/arch/i386/oprofile/op_model_athlon.c
@@ -21,10 +21,12 @@
 #define NUM_COUNTERS 4
 #define NUM_CONTROLS 4
 
+#define CTR_IS_RESERVED(msrs,c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_READ(l,h,msrs,c) do {rdmsr(msrs->counters[(c)].addr, (l), (h));} while (0)
 #define CTR_WRITE(l,msrs,c) do {wrmsr(msrs->counters[(c)].addr, -(unsigned int)(l), -1);} while (0)
 #define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
 
+#define CTRL_IS_RESERVED(msrs,c) (msrs->controls[(c)].addr ? 1 : 0)
 #define CTRL_READ(l,h,msrs,c) do {rdmsr(msrs->controls[(c)].addr, (l), (h));} while (0)
 #define CTRL_WRITE(l,h,msrs,c) do {wrmsr(msrs->controls[(c)].addr, (l), (h));} while (0)
 #define CTRL_SET_ACTIVE(n) (n |= (1<<22))
@@ -40,15 +42,21 @@ static unsigned long reset_value[NUM_COUNTERS];
  
 static void athlon_fill_in_addresses(struct op_msrs * const msrs)
 {
-	msrs->counters[0].addr = MSR_K7_PERFCTR0;
-	msrs->counters[1].addr = MSR_K7_PERFCTR1;
-	msrs->counters[2].addr = MSR_K7_PERFCTR2;
-	msrs->counters[3].addr = MSR_K7_PERFCTR3;
-
-	msrs->controls[0].addr = MSR_K7_EVNTSEL0;
-	msrs->controls[1].addr = MSR_K7_EVNTSEL1;
-	msrs->controls[2].addr = MSR_K7_EVNTSEL2;
-	msrs->controls[3].addr = MSR_K7_EVNTSEL3;
+	int i;
+
+	for (i=0; i < NUM_COUNTERS; i++) {
+		if (reserve_perfctr_nmi(MSR_K7_PERFCTR0 + i))
+			msrs->counters[i].addr = MSR_K7_PERFCTR0 + i;
+		else
+			msrs->counters[i].addr = 0;
+	}
+
+	for (i=0; i < NUM_CONTROLS; i++) {
+		if (reserve_evntsel_nmi(MSR_K7_EVNTSEL0 + i))
+			msrs->controls[i].addr = MSR_K7_EVNTSEL0 + i;
+		else
+			msrs->controls[i].addr = 0;
+	}
 }
 
  
@@ -59,19 +67,23 @@ static void athlon_setup_ctrs(struct op_msrs const * const msrs)
  
 	/* clear all counters */
 	for (i = 0 ; i < NUM_CONTROLS; ++i) {
+		if (unlikely(!CTRL_IS_RESERVED(msrs,i)))
+			continue;
 		CTRL_READ(low, high, msrs, i);
 		CTRL_CLEAR(low);
 		CTRL_WRITE(low, high, msrs, i);
 	}
-	
+
 	/* avoid a false detection of ctr overflows in NMI handler */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
+		if (unlikely(!CTR_IS_RESERVED(msrs,i)))
+			continue;
 		CTR_WRITE(1, msrs, i);
 	}
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (counter_config[i].enabled) {
+		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs,i))) {
 			reset_value[i] = counter_config[i].count;
 
 			CTR_WRITE(counter_config[i].count, msrs, i);
@@ -98,6 +110,8 @@ static int athlon_check_ctrs(struct pt_regs * const regs,
 	int i;
 
 	for (i = 0 ; i < NUM_COUNTERS; ++i) {
+		if (!reset_value[i])
+			continue;
 		CTR_READ(low, high, msrs, i);
 		if (CTR_OVERFLOWED(low)) {
 			oprofile_add_sample(regs, i);
@@ -132,12 +146,27 @@ static void athlon_stop(struct op_msrs const * const msrs)
 	/* Subtle: stop on all counters to avoid race with
 	 * setting our pm callback */
 	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (!reset_value[i])
+			continue;
 		CTRL_READ(low, high, msrs, i);
 		CTRL_SET_INACTIVE(low);
 		CTRL_WRITE(low, high, msrs, i);
 	}
 }
 
+static void athlon_shutdown(struct op_msrs const * const msrs)
+{
+	int i;
+
+	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (CTR_IS_RESERVED(msrs,i))
+			release_perfctr_nmi(MSR_K7_PERFCTR0 + i);
+	}
+	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
+		if (CTRL_IS_RESERVED(msrs,i))
+			release_evntsel_nmi(MSR_K7_EVNTSEL0 + i);
+	}
+}
 
 struct op_x86_model_spec const op_athlon_spec = {
 	.num_counters = NUM_COUNTERS,
@@ -146,5 +175,6 @@ struct op_x86_model_spec const op_athlon_spec = {
 	.setup_ctrs = &athlon_setup_ctrs,
 	.check_ctrs = &athlon_check_ctrs,
 	.start = &athlon_start,
-	.stop = &athlon_stop
+	.stop = &athlon_stop,
+	.shutdown = &athlon_shutdown
 };
diff --git a/arch/i386/oprofile/op_model_p4.c b/arch/i386/oprofile/op_model_p4.c
index 7c61d357b82b..47925927b12f 100644
--- a/arch/i386/oprofile/op_model_p4.c
+++ b/arch/i386/oprofile/op_model_p4.c
@@ -32,7 +32,7 @@
 #define NUM_CONTROLS_HT2 (NUM_ESCRS_HT2 + NUM_CCCRS_HT2)
 
 static unsigned int num_counters = NUM_COUNTERS_NON_HT;
-
+static unsigned int num_controls = NUM_CONTROLS_NON_HT;
 
 /* this has to be checked dynamically since the
    hyper-threadedness of a chip is discovered at
@@ -40,8 +40,10 @@ static unsigned int num_counters = NUM_COUNTERS_NON_HT;
 static inline void setup_num_counters(void)
 {
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
+	if (smp_num_siblings == 2){
 		num_counters = NUM_COUNTERS_HT2;
+		num_controls = NUM_CONTROLS_HT2;
+	}
 #endif
 }
 
@@ -97,15 +99,6 @@ static struct p4_counter_binding p4_counters [NUM_COUNTERS_NON_HT] = {
 
 #define NUM_UNUSED_CCCRS	NUM_CCCRS_NON_HT - NUM_COUNTERS_NON_HT
 
-/* All cccr we don't use. */
-static int p4_unused_cccr[NUM_UNUSED_CCCRS] = {
-	MSR_P4_BPU_CCCR1,	MSR_P4_BPU_CCCR3,
-	MSR_P4_MS_CCCR1,	MSR_P4_MS_CCCR3,
-	MSR_P4_FLAME_CCCR1,	MSR_P4_FLAME_CCCR3,
-	MSR_P4_IQ_CCCR0,	MSR_P4_IQ_CCCR1,
-	MSR_P4_IQ_CCCR2,	MSR_P4_IQ_CCCR3
-};
-
 /* p4 event codes in libop/op_event.h are indices into this table. */
 
 static struct p4_event_binding p4_events[NUM_EVENTS] = {
@@ -372,6 +365,8 @@ static struct p4_event_binding p4_events[NUM_EVENTS] = {
 #define CCCR_OVF_P(cccr) ((cccr) & (1U<<31))
 #define CCCR_CLEAR_OVF(cccr) ((cccr) &= (~(1U<<31)))
 
+#define CTRL_IS_RESERVED(msrs,c) (msrs->controls[(c)].addr ? 1 : 0)
+#define CTR_IS_RESERVED(msrs,c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_READ(l,h,i) do {rdmsr(p4_counters[(i)].counter_address, (l), (h));} while (0)
 #define CTR_WRITE(l,i) do {wrmsr(p4_counters[(i)].counter_address, -(u32)(l), -1);} while (0)
 #define CTR_OVERFLOW_P(ctr) (!((ctr) & 0x80000000))
@@ -401,29 +396,34 @@ static unsigned long reset_value[NUM_COUNTERS_NON_HT];
 static void p4_fill_in_addresses(struct op_msrs * const msrs)
 {
 	unsigned int i; 
-	unsigned int addr, stag;
+	unsigned int addr, cccraddr, stag;
 
 	setup_num_counters();
 	stag = get_stagger();
 
-	/* the counter registers we pay attention to */
+	/* initialize some registers */
 	for (i = 0; i < num_counters; ++i) {
-		msrs->counters[i].addr = 
-			p4_counters[VIRT_CTR(stag, i)].counter_address;
+		msrs->counters[i].addr = 0;
 	}
-
-	/* FIXME: bad feeling, we don't save the 10 counters we don't use. */
-
-	/* 18 CCCR registers */
-	for (i = 0, addr = MSR_P4_BPU_CCCR0 + stag;
-	     addr <= MSR_P4_IQ_CCCR5; ++i, addr += addr_increment()) {
-		msrs->controls[i].addr = addr;
+	for (i = 0; i < num_controls; ++i) {
+		msrs->controls[i].addr = 0;
 	}
 	
+	/* the counter & cccr registers we pay attention to */
+	for (i = 0; i < num_counters; ++i) {
+		addr = p4_counters[VIRT_CTR(stag, i)].counter_address;
+		cccraddr = p4_counters[VIRT_CTR(stag, i)].cccr_address;
+		if (reserve_perfctr_nmi(addr)){
+			msrs->counters[i].addr = addr;
+			msrs->controls[i].addr = cccraddr;
+		}
+	}
+
 	/* 43 ESCR registers in three or four discontiguous group */
 	for (addr = MSR_P4_BSU_ESCR0 + stag;
 	     addr < MSR_P4_IQ_ESCR0; ++i, addr += addr_increment()) {
-		msrs->controls[i].addr = addr;
+		if (reserve_evntsel_nmi(addr))
+			msrs->controls[i].addr = addr;
 	}
 
 	/* no IQ_ESCR0/1 on some models, we save a seconde time BSU_ESCR0/1
@@ -431,47 +431,57 @@ static void p4_fill_in_addresses(struct op_msrs * const msrs)
 	if (boot_cpu_data.x86_model >= 0x3) {
 		for (addr = MSR_P4_BSU_ESCR0 + stag;
 		     addr <= MSR_P4_BSU_ESCR1; ++i, addr += addr_increment()) {
-			msrs->controls[i].addr = addr;
+			if (reserve_evntsel_nmi(addr))
+				msrs->controls[i].addr = addr;
 		}
 	} else {
 		for (addr = MSR_P4_IQ_ESCR0 + stag;
 		     addr <= MSR_P4_IQ_ESCR1; ++i, addr += addr_increment()) {
-			msrs->controls[i].addr = addr;
+			if (reserve_evntsel_nmi(addr))
+				msrs->controls[i].addr = addr;
 		}
 	}
 
 	for (addr = MSR_P4_RAT_ESCR0 + stag;
 	     addr <= MSR_P4_SSU_ESCR0; ++i, addr += addr_increment()) {
-		msrs->controls[i].addr = addr;
+		if (reserve_evntsel_nmi(addr))
+			msrs->controls[i].addr = addr;
 	}
 	
 	for (addr = MSR_P4_MS_ESCR0 + stag;
 	     addr <= MSR_P4_TC_ESCR1; ++i, addr += addr_increment()) { 
-		msrs->controls[i].addr = addr;
+		if (reserve_evntsel_nmi(addr))
+			msrs->controls[i].addr = addr;
 	}
 	
 	for (addr = MSR_P4_IX_ESCR0 + stag;
 	     addr <= MSR_P4_CRU_ESCR3; ++i, addr += addr_increment()) { 
-		msrs->controls[i].addr = addr;
+		if (reserve_evntsel_nmi(addr))
+			msrs->controls[i].addr = addr;
 	}
 
 	/* there are 2 remaining non-contiguously located ESCRs */
 
 	if (num_counters == NUM_COUNTERS_NON_HT) {		
 		/* standard non-HT CPUs handle both remaining ESCRs*/
-		msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
-		msrs->controls[i++].addr = MSR_P4_CRU_ESCR4;
+		if (reserve_evntsel_nmi(MSR_P4_CRU_ESCR5))
+			msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
+		if (reserve_evntsel_nmi(MSR_P4_CRU_ESCR4))
+			msrs->controls[i++].addr = MSR_P4_CRU_ESCR4;
 
 	} else if (stag == 0) {
 		/* HT CPUs give the first remainder to the even thread, as
 		   the 32nd control register */
-		msrs->controls[i++].addr = MSR_P4_CRU_ESCR4;
+		if (reserve_evntsel_nmi(MSR_P4_CRU_ESCR4))
+			msrs->controls[i++].addr = MSR_P4_CRU_ESCR4;
 
 	} else {
 		/* and two copies of the second to the odd thread,
 		   for the 22st and 23nd control registers */
-		msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
-		msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
+		if (reserve_evntsel_nmi(MSR_P4_CRU_ESCR5)) {
+			msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
+			msrs->controls[i++].addr = MSR_P4_CRU_ESCR5;
+		}
 	}
 }
 
@@ -544,7 +554,6 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs)
 {
 	unsigned int i;
 	unsigned int low, high;
-	unsigned int addr;
 	unsigned int stag;
 
 	stag = get_stagger();
@@ -557,59 +566,24 @@ static void p4_setup_ctrs(struct op_msrs const * const msrs)
 
 	/* clear the cccrs we will use */
 	for (i = 0 ; i < num_counters ; i++) {
+		if (unlikely(!CTRL_IS_RESERVED(msrs,i)))
+			continue;
 		rdmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 		CCCR_CLEAR(low);
 		CCCR_SET_REQUIRED_BITS(low);
 		wrmsr(p4_counters[VIRT_CTR(stag, i)].cccr_address, low, high);
 	}
 
-	/* clear cccrs outside our concern */
-	for (i = stag ; i < NUM_UNUSED_CCCRS ; i += addr_increment()) {
-		rdmsr(p4_unused_cccr[i], low, high);
-		CCCR_CLEAR(low);
-		CCCR_SET_REQUIRED_BITS(low);
-		wrmsr(p4_unused_cccr[i], low, high);
-	}
-
 	/* clear all escrs (including those outside our concern) */
-	for (addr = MSR_P4_BSU_ESCR0 + stag;
-	     addr <  MSR_P4_IQ_ESCR0; addr += addr_increment()) {
-		wrmsr(addr, 0, 0);
-	}
-
-	/* On older models clear also MSR_P4_IQ_ESCR0/1 */
-	if (boot_cpu_data.x86_model < 0x3) {
-		wrmsr(MSR_P4_IQ_ESCR0, 0, 0);
-		wrmsr(MSR_P4_IQ_ESCR1, 0, 0);
-	}
-
-	for (addr = MSR_P4_RAT_ESCR0 + stag;
-	     addr <= MSR_P4_SSU_ESCR0; ++i, addr += addr_increment()) {
-		wrmsr(addr, 0, 0);
-	}
-	
-	for (addr = MSR_P4_MS_ESCR0 + stag;
-	     addr <= MSR_P4_TC_ESCR1; addr += addr_increment()){ 
-		wrmsr(addr, 0, 0);
-	}
-	
-	for (addr = MSR_P4_IX_ESCR0 + stag;
-	     addr <= MSR_P4_CRU_ESCR3; addr += addr_increment()){ 
-		wrmsr(addr, 0, 0);
+	for (i = num_counters; i < num_controls; i++) {
+		if (unlikely(!CTRL_IS_RESERVED(msrs,i)))
+			continue;
+		wrmsr(msrs->controls[i].addr, 0, 0);
 	}
 
-	if (num_counters == NUM_COUNTERS_NON_HT) {		
-		wrmsr(MSR_P4_CRU_ESCR4, 0, 0);
-		wrmsr(MSR_P4_CRU_ESCR5, 0, 0);
-	} else if (stag == 0) {
-		wrmsr(MSR_P4_CRU_ESCR4, 0, 0);
-	} else {
-		wrmsr(MSR_P4_CRU_ESCR5, 0, 0);
-	}		
-	
 	/* setup all counters */
 	for (i = 0 ; i < num_counters ; ++i) {
-		if (counter_config[i].enabled) {
+		if ((counter_config[i].enabled) && (CTRL_IS_RESERVED(msrs,i))) {
 			reset_value[i] = counter_config[i].count;
 			pmc_setup_one_p4_counter(i);
 			CTR_WRITE(counter_config[i].count, VIRT_CTR(stag, i));
@@ -696,12 +670,32 @@ static void p4_stop(struct op_msrs const * const msrs)
 	stag = get_stagger();
 
 	for (i = 0; i < num_counters; ++i) {
+		if (!reset_value[i])
+			continue;
 		CCCR_READ(low, high, VIRT_CTR(stag, i));
 		CCCR_SET_DISABLE(low);
 		CCCR_WRITE(low, high, VIRT_CTR(stag, i));
 	}
 }
 
+static void p4_shutdown(struct op_msrs const * const msrs)
+{
+	int i;
+
+	for (i = 0 ; i < num_counters ; ++i) {
+		if (CTR_IS_RESERVED(msrs,i))
+			release_perfctr_nmi(msrs->counters[i].addr);
+	}
+	/* some of the control registers are specially reserved in
+	 * conjunction with the counter registers (hence the starting offset).
+	 * This saves a few bits.
+	 */
+	for (i = num_counters ; i < num_controls ; ++i) {
+		if (CTRL_IS_RESERVED(msrs,i))
+			release_evntsel_nmi(msrs->controls[i].addr);
+	}
+}
+
 
 #ifdef CONFIG_SMP
 struct op_x86_model_spec const op_p4_ht2_spec = {
@@ -711,7 +705,8 @@ struct op_x86_model_spec const op_p4_ht2_spec = {
 	.setup_ctrs = &p4_setup_ctrs,
 	.check_ctrs = &p4_check_ctrs,
 	.start = &p4_start,
-	.stop = &p4_stop
+	.stop = &p4_stop,
+	.shutdown = &p4_shutdown
 };
 #endif
 
@@ -722,5 +717,6 @@ struct op_x86_model_spec const op_p4_spec = {
 	.setup_ctrs = &p4_setup_ctrs,
 	.check_ctrs = &p4_check_ctrs,
 	.start = &p4_start,
-	.stop = &p4_stop
+	.stop = &p4_stop,
+	.shutdown = &p4_shutdown
 };
diff --git a/arch/i386/oprofile/op_model_ppro.c b/arch/i386/oprofile/op_model_ppro.c
index 5c3ab4b027ad..f88e05ba8eb3 100644
--- a/arch/i386/oprofile/op_model_ppro.c
+++ b/arch/i386/oprofile/op_model_ppro.c
@@ -22,10 +22,12 @@
 #define NUM_COUNTERS 2
 #define NUM_CONTROLS 2
 
+#define CTR_IS_RESERVED(msrs,c) (msrs->counters[(c)].addr ? 1 : 0)
 #define CTR_READ(l,h,msrs,c) do {rdmsr(msrs->counters[(c)].addr, (l), (h));} while (0)
 #define CTR_WRITE(l,msrs,c) do {wrmsr(msrs->counters[(c)].addr, -(u32)(l), -1);} while (0)
 #define CTR_OVERFLOWED(n) (!((n) & (1U<<31)))
 
+#define CTRL_IS_RESERVED(msrs,c) (msrs->controls[(c)].addr ? 1 : 0)
 #define CTRL_READ(l,h,msrs,c) do {rdmsr((msrs->controls[(c)].addr), (l), (h));} while (0)
 #define CTRL_WRITE(l,h,msrs,c) do {wrmsr((msrs->controls[(c)].addr), (l), (h));} while (0)
 #define CTRL_SET_ACTIVE(n) (n |= (1<<22))
@@ -41,11 +43,21 @@ static unsigned long reset_value[NUM_COUNTERS];
  
 static void ppro_fill_in_addresses(struct op_msrs * const msrs)
 {
-	msrs->counters[0].addr = MSR_P6_PERFCTR0;
-	msrs->counters[1].addr = MSR_P6_PERFCTR1;
+	int i;
+
+	for (i=0; i < NUM_COUNTERS; i++) {
+		if (reserve_perfctr_nmi(MSR_P6_PERFCTR0 + i))
+			msrs->counters[i].addr = MSR_P6_PERFCTR0 + i;
+		else
+			msrs->counters[i].addr = 0;
+	}
 	
-	msrs->controls[0].addr = MSR_P6_EVNTSEL0;
-	msrs->controls[1].addr = MSR_P6_EVNTSEL1;
+	for (i=0; i < NUM_CONTROLS; i++) {
+		if (reserve_evntsel_nmi(MSR_P6_EVNTSEL0 + i))
+			msrs->controls[i].addr = MSR_P6_EVNTSEL0 + i;
+		else
+			msrs->controls[i].addr = 0;
+	}
 }
 
 
@@ -56,6 +68,8 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
 
 	/* clear all counters */
 	for (i = 0 ; i < NUM_CONTROLS; ++i) {
+		if (unlikely(!CTRL_IS_RESERVED(msrs,i)))
+			continue;
 		CTRL_READ(low, high, msrs, i);
 		CTRL_CLEAR(low);
 		CTRL_WRITE(low, high, msrs, i);
@@ -63,12 +77,14 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
 	
 	/* avoid a false detection of ctr overflows in NMI handler */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
+		if (unlikely(!CTR_IS_RESERVED(msrs,i)))
+			continue;
 		CTR_WRITE(1, msrs, i);
 	}
 
 	/* enable active counters */
 	for (i = 0; i < NUM_COUNTERS; ++i) {
-		if (counter_config[i].enabled) {
+		if ((counter_config[i].enabled) && (CTR_IS_RESERVED(msrs,i))) {
 			reset_value[i] = counter_config[i].count;
 
 			CTR_WRITE(counter_config[i].count, msrs, i);
@@ -81,6 +97,8 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs)
 			CTRL_SET_UM(low, counter_config[i].unit_mask);
 			CTRL_SET_EVENT(low, counter_config[i].event);
 			CTRL_WRITE(low, high, msrs, i);
+		} else {
+			reset_value[i] = 0;
 		}
 	}
 }
@@ -93,6 +111,8 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
 	int i;
  
 	for (i = 0 ; i < NUM_COUNTERS; ++i) {
+		if (!reset_value[i])
+			continue;
 		CTR_READ(low, high, msrs, i);
 		if (CTR_OVERFLOWED(low)) {
 			oprofile_add_sample(regs, i);
@@ -118,18 +138,38 @@ static int ppro_check_ctrs(struct pt_regs * const regs,
 static void ppro_start(struct op_msrs const * const msrs)
 {
 	unsigned int low,high;
-	CTRL_READ(low, high, msrs, 0);
-	CTRL_SET_ACTIVE(low);
-	CTRL_WRITE(low, high, msrs, 0);
+
+	if (reset_value[0]) {
+		CTRL_READ(low, high, msrs, 0);
+		CTRL_SET_ACTIVE(low);
+		CTRL_WRITE(low, high, msrs, 0);
+	}
 }
 
 
 static void ppro_stop(struct op_msrs const * const msrs)
 {
 	unsigned int low,high;
-	CTRL_READ(low, high, msrs, 0);
-	CTRL_SET_INACTIVE(low);
-	CTRL_WRITE(low, high, msrs, 0);
+
+	if (reset_value[0]) {
+		CTRL_READ(low, high, msrs, 0);
+		CTRL_SET_INACTIVE(low);
+		CTRL_WRITE(low, high, msrs, 0);
+	}
+}
+
+static void ppro_shutdown(struct op_msrs const * const msrs)
+{
+	int i;
+
+	for (i = 0 ; i < NUM_COUNTERS ; ++i) {
+		if (CTR_IS_RESERVED(msrs,i))
+			release_perfctr_nmi(MSR_P6_PERFCTR0 + i);
+	}
+	for (i = 0 ; i < NUM_CONTROLS ; ++i) {
+		if (CTRL_IS_RESERVED(msrs,i))
+			release_evntsel_nmi(MSR_P6_EVNTSEL0 + i);
+	}
 }
 
 
@@ -140,5 +180,6 @@ struct op_x86_model_spec const op_ppro_spec = {
 	.setup_ctrs = &ppro_setup_ctrs,
 	.check_ctrs = &ppro_check_ctrs,
 	.start = &ppro_start,
-	.stop = &ppro_stop
+	.stop = &ppro_stop,
+	.shutdown = &ppro_shutdown
 };
diff --git a/arch/i386/oprofile/op_x86_model.h b/arch/i386/oprofile/op_x86_model.h
index 123b7e90a9ee..abb1aa95b979 100644
--- a/arch/i386/oprofile/op_x86_model.h
+++ b/arch/i386/oprofile/op_x86_model.h
@@ -40,6 +40,7 @@ struct op_x86_model_spec {
 		struct op_msrs const * const msrs);
 	void (*start)(struct op_msrs const * const msrs);
 	void (*stop)(struct op_msrs const * const msrs);
+	void (*shutdown)(struct op_msrs const * const msrs);
 };
 
 extern struct op_x86_model_spec const op_ppro_spec;
-- 
cgit v1.2.1


From f2802e7f571c05f9a901b1f5bd144aa730ccc88e Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] Add SMP support on x86_64 to reservation framework

This patch includes the changes to make the nmi watchdog on x86_64 SMP
aware.  A bunch of code was moved around to make it simpler to read.  In
addition, it is now possible to determine if a particular NMI was the result
of the watchdog or not.  This feature allows the kernel to filter out
unknown NMIs easier.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c |   3 +-
 arch/x86_64/kernel/nmi.c  | 426 +++++++++++++++++++++++++++++++---------------
 include/asm-x86_64/nmi.h  |   3 +-
 3 files changed, 290 insertions(+), 142 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 2b8cef037a65..692e9b579743 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -479,8 +479,7 @@ void __cpuinit setup_local_APIC (void)
 	}
 
 	nmi_watchdog_default();
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		setup_apic_nmi_watchdog();
+	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
 
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index b7a7c9973849..d42374a952d7 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -56,53 +56,29 @@ static unsigned int lapic_nmi_owner;
 #define LAPIC_NMI_RESERVED	(1<<1)
 
 /* nmi_active:
- * +1: the lapic NMI watchdog is active, but can be disabled
- *  0: the lapic NMI watchdog has not been set up, and cannot
+ * >0: the lapic NMI watchdog is active, but can be disabled
+ * <0: the lapic NMI watchdog has not been set up, and cannot
  *     be enabled
- * -1: the lapic NMI watchdog is disabled, but can be enabled
+ *  0: the lapic NMI watchdog is disabled, but can be enabled
  */
-int nmi_active;		/* oprofile uses this */
+atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 int panic_on_timeout;
 
 unsigned int nmi_watchdog = NMI_DEFAULT;
 static unsigned int nmi_hz = HZ;
-static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
-static unsigned int nmi_p4_cccr_val;
 
-/* Note that these events don't tick when the CPU idles. This means
-   the frequency varies with CPU load. */
-
-#define K7_EVNTSEL_ENABLE	(1 << 22)
-#define K7_EVNTSEL_INT		(1 << 20)
-#define K7_EVNTSEL_OS		(1 << 17)
-#define K7_EVNTSEL_USR		(1 << 16)
-#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
-#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+struct nmi_watchdog_ctlblk {
+	int enabled;
+	u64 check_bit;
+	unsigned int cccr_msr;
+	unsigned int perfctr_msr;  /* the MSR to reset in NMI handler */
+	unsigned int evntsel_msr;  /* the MSR to select the events to handle */
+};
+static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
-#define MSR_P4_MISC_ENABLE	0x1A0
-#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
-#define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
-#define MSR_P4_PERFCTR0		0x300
-#define MSR_P4_CCCR0		0x360
-#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
-#define P4_ESCR_OS		(1<<3)
-#define P4_ESCR_USR		(1<<2)
-#define P4_CCCR_OVF_PMI0	(1<<26)
-#define P4_CCCR_OVF_PMI1	(1<<27)
-#define P4_CCCR_THRESHOLD(N)	((N)<<20)
-#define P4_CCCR_COMPLEMENT	(1<<19)
-#define P4_CCCR_COMPARE		(1<<18)
-#define P4_CCCR_REQUIRED	(3<<16)
-#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
-#define P4_CCCR_ENABLE		(1<<12)
-/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
-   CRU_ESCR0 (with any non-null event selector) through a complemented
-   max threshold. [IA32-Vol3, Section 14.9.9] */
-#define MSR_P4_IQ_COUNTER0	0x30C
-#define P4_NMI_CRU_ESCR0	(P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
-#define P4_NMI_IQ_CCCR0	\
-	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
-	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
+/* local prototypes */
+static void stop_apic_nmi_watchdog(void *unused);
+static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 /* converts an msr to an appropriate reservation bit */
 static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
@@ -241,6 +217,12 @@ int __init check_nmi_watchdog (void)
 	int *counts;
 	int cpu;
 
+	if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT))
+		return 0;
+
+	if (!atomic_read(&nmi_active))
+		return 0;
+
 	counts = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
 	if (!counts)
 		return -1;
@@ -258,19 +240,22 @@ int __init check_nmi_watchdog (void)
 	mdelay((10*1000)/nmi_hz); // wait 10 ticks
 
 	for_each_online_cpu(cpu) {
+		if (!per_cpu(nmi_watchdog_ctlblk, cpu).enabled)
+			continue;
 		if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
-			endflag = 1;
 			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
 			       cpu,
 			       counts[cpu],
 			       cpu_pda(cpu)->__nmi_count);
-			nmi_active = 0;
-			lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
-			nmi_perfctr_msr = 0;
-			kfree(counts);
-			return -1;
+			per_cpu(nmi_watchdog_ctlblk, cpu).enabled = 0;
+			atomic_dec(&nmi_active);
 		}
 	}
+	if (!atomic_read(&nmi_active)) {
+		kfree(counts);
+		atomic_set(&nmi_active, -1);
+		return -1;
+	}
 	endflag = 1;
 	printk("OK.\n");
 
@@ -297,8 +282,11 @@ int __init setup_nmi_watchdog(char *str)
 
 	get_option(&str, &nmi);
 
-	if (nmi >= NMI_INVALID)
+	if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE))
 		return 0;
+
+	if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0))
+		return 0;  /* no lapic support */
 	nmi_watchdog = nmi;
 	return 1;
 }
@@ -307,31 +295,30 @@ __setup("nmi_watchdog=", setup_nmi_watchdog);
 
 static void disable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active <= 0)
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
-		break;
-	case X86_VENDOR_INTEL:
-		if (boot_cpu_data.x86 == 15) {
-			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
-			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-		}
-		break;
-	}
-	nmi_active = -1;
-	/* tell do_nmi() and others that we're not active any more */
-	nmi_watchdog = 0;
+
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 static void enable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_LOCAL_APIC;
-		touch_nmi_watchdog();
-		setup_apic_nmi_watchdog();
-	}
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	/* are we already enabled */
+	if (atomic_read(&nmi_active) != 0)
+		return;
+
+	/* are we lapic aware */
+	if (nmi_known_cpu() <= 0)
+		return;
+
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	touch_nmi_watchdog();
 }
 
 int reserve_lapic_nmi(void)
@@ -363,21 +350,24 @@ void release_lapic_nmi(void)
 
 void disable_timer_nmi_watchdog(void)
 {
-	if ((nmi_watchdog != NMI_IO_APIC) || (nmi_active <= 0))
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
 
 	disable_irq(0);
-	unset_nmi_callback();
-	nmi_active = -1;
-	nmi_watchdog = NMI_NONE;
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 void enable_timer_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_IO_APIC;
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) == 0) {
 		touch_nmi_watchdog();
-		nmi_active = 1;
+		on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
 		enable_irq(0);
 	}
 }
@@ -388,7 +378,7 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
-	nmi_pm_active = nmi_active;
+	nmi_pm_active = atomic_read(&nmi_active);
 	disable_lapic_nmi_watchdog();
 	return 0;
 }
@@ -396,7 +386,7 @@ static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 static int lapic_nmi_resume(struct sys_device *dev)
 {
 	if (nmi_pm_active > 0)
-	enable_lapic_nmi_watchdog();
+		enable_lapic_nmi_watchdog();
 	return 0;
 }
 
@@ -415,7 +405,13 @@ static int __init init_lapic_nmi_sysfs(void)
 {
 	int error;
 
-	if (nmi_active == 0 || nmi_watchdog != NMI_LOCAL_APIC)
+	/* should really be a BUG_ON but b/c this is an
+	 * init call, it just doesn't work.  -dcz
+	 */
+	if (nmi_watchdog != NMI_LOCAL_APIC)
+		return 0;
+
+	if ( atomic_read(&nmi_active) < 0 )
 		return 0;
 
 	error = sysdev_class_register(&nmi_sysclass);
@@ -428,100 +424,232 @@ late_initcall(init_lapic_nmi_sysfs);
 
 #endif	/* CONFIG_PM */
 
+/*
+ * Activate the NMI watchdog via the local APIC.
+ * Original code written by Keith Owens.
+ */
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define K7_EVNTSEL_ENABLE	(1 << 22)
+#define K7_EVNTSEL_INT		(1 << 20)
+#define K7_EVNTSEL_OS		(1 << 17)
+#define K7_EVNTSEL_USR		(1 << 16)
+#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
+#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+
 static int setup_k7_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_K7_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_K7_PERFCTR0;
+	evntsel_msr = MSR_K7_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
 	/* Simulator may not support it */
-	if (checking_wrmsrl(MSR_K7_EVNTSEL0, 0UL))
+	if (checking_wrmsrl(evntsel_msr, 0UL))
 		goto fail2;
-	wrmsrl(MSR_K7_PERFCTR0, 0UL);
+	wrmsrl(perfctr_msr, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
 		| K7_EVNTSEL_USR
 		| K7_NMI_EVENT;
 
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	wrmsrl(MSR_K7_PERFCTR0, -((u64)cpu_khz * 1000 / nmi_hz));
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<63;
 	return 1;
 fail2:
-	release_evntsel_nmi(MSR_K7_EVNTSEL0);
+	release_evntsel_nmi(evntsel_msr);
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_k7_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
+#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
+#define P4_ESCR_OS		(1<<3)
+#define P4_ESCR_USR		(1<<2)
+#define P4_CCCR_OVF_PMI0	(1<<26)
+#define P4_CCCR_OVF_PMI1	(1<<27)
+#define P4_CCCR_THRESHOLD(N)	((N)<<20)
+#define P4_CCCR_COMPLEMENT	(1<<19)
+#define P4_CCCR_COMPARE		(1<<18)
+#define P4_CCCR_REQUIRED	(3<<16)
+#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
+#define P4_CCCR_ENABLE		(1<<12)
+#define P4_CCCR_OVF 		(1<<31)
+/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
+   CRU_ESCR0 (with any non-null event selector) through a complemented
+   max threshold. [IA32-Vol3, Section 14.9.9] */
 
 static int setup_p4_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr, cccr_msr;
+	unsigned int evntsel, cccr_val;
 	unsigned int misc_enable, dummy;
+	unsigned int ht_num;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	rdmsr(MSR_P4_MISC_ENABLE, misc_enable, dummy);
+	rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy);
 	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
 		return 0;
 
-	nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
-	nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
-		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
+	/* detect which hyperthread we are on */
+	if (smp_num_siblings == 2) {
+		unsigned int ebx, apicid;
+
+        	ebx = cpuid_ebx(1);
+	        apicid = (ebx >> 24) & 0xff;
+        	ht_num = apicid & 1;
+	} else
 #endif
+		ht_num = 0;
+
+	/* performance counters are shared resources
+	 * assign each hyperthread its own set
+	 * (re-use the ESCR0 register, seems safe
+	 * and keeps the cccr_val the same)
+	 */
+	if (!ht_num) {
+		/* logical cpu 0 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR0;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR0;
+		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+	} else {
+		/* logical cpu 1 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR1;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR1;
+		cccr_val = P4_CCCR_OVF_PMI1 | P4_CCCR_ESCR_SELECT(4);
+	}
 
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
-	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
-	Dprintk("setting P4_IQ_COUNTER0 to 0x%08lx\n", -(cpu_khz * 1000UL / nmi_hz));
-	wrmsrl(MSR_P4_IQ_COUNTER0, -((u64)cpu_khz * 1000 / nmi_hz));
+	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
+	 	| P4_ESCR_OS
+		| P4_ESCR_USR;
+
+	cccr_val |= P4_CCCR_THRESHOLD(15)
+		 | P4_CCCR_COMPLEMENT
+		 | P4_CCCR_COMPARE
+		 | P4_CCCR_REQUIRED;
+
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsr(cccr_msr, cccr_val, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = cccr_msr;
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
-void setup_apic_nmi_watchdog(void)
+static void stop_p4_watchdog(void)
 {
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		if (boot_cpu_data.x86 != 15)
-			return;
-		if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
-			return;
-		if (!setup_k7_watchdog())
-			return;
-		break;
-	case X86_VENDOR_INTEL:
-		if (boot_cpu_data.x86 != 15)
-			return;
-		if (!setup_p4_watchdog())
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->cccr_msr, 0, 0);
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+void setup_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
+				return;
+			if (!setup_k7_watchdog())
+				return;
+			break;
+		case X86_VENDOR_INTEL:
+			if (!setup_p4_watchdog())
+				return;
+			break;
+		default:
 			return;
-		break;
+		}
+	}
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	atomic_inc(&nmi_active);
+}
 
-	default:
-		return;
+static void stop_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (strstr(boot_cpu_data.x86_model_id, "Screwdriver"))
+				return;
+			stop_k7_watchdog();
+			break;
+		case X86_VENDOR_INTEL:
+			stop_p4_watchdog();
+			break;
+		default:
+			return;
+		}
 	}
-	lapic_nmi_owner = LAPIC_NMI_WATCHDOG;
-	nmi_active = 1;
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	atomic_dec(&nmi_active);
 }
 
 /*
@@ -558,50 +686,70 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	u64 dummy;
+
+	/* check for other users first */
+	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
+			== NOTIFY_STOP) {
+		touched = 1;
+	}
 
 	sum = read_pda(apic_timer_irqs);
 	if (__get_cpu_var(nmi_touch)) {
 		__get_cpu_var(nmi_touch) = 0;
 		touched = 1;
 	}
+
 #ifdef CONFIG_X86_MCE
 	/* Could check oops_in_progress here too, but it's safer
 	   not too */
 	if (atomic_read(&mce_entry) > 0)
 		touched = 1;
 #endif
+	/* if the apic timer isn't firing, this cpu isn't doing much */
 	if (!touched && __get_cpu_var(last_irq_sum) == sum) {
 		/*
 		 * Ayiee, looks like this CPU is stuck ...
 		 * wait a few IRQs (5 seconds) before doing the oops ...
 		 */
 		local_inc(&__get_cpu_var(alert_counter));
-		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz) {
-			if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
-							== NOTIFY_STOP) {
-				local_set(&__get_cpu_var(alert_counter), 0);
-				return;
-			}
+		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz)
 			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs);
-		}
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
 		local_set(&__get_cpu_var(alert_counter), 0);
 	}
-	if (nmi_perfctr_msr) {
- 		if (nmi_perfctr_msr == MSR_P4_IQ_COUNTER0) {
- 			/*
- 			 * P4 quirks:
- 			 * - An overflown perfctr will assert its interrupt
- 			 *   until the OVF flag in its CCCR is cleared.
- 			 * - LVTPC is masked on interrupt and must be
- 			 *   unmasked by the LVTPC handler.
- 			 */
- 			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
- 			apic_write(APIC_LVTPC, APIC_DM_NMI);
- 		}
-		wrmsrl(nmi_perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+
+	/* see if the nmi watchdog went off */
+	if (wd->enabled) {
+		if (nmi_watchdog == NMI_LOCAL_APIC) {
+			rdmsrl(wd->perfctr_msr, dummy);
+			if (dummy & wd->check_bit){
+				/* this wasn't a watchdog timer interrupt */
+				goto done;
+			}
+
+			/* only Intel uses the cccr msr */
+	 		if (wd->cccr_msr != 0) {
+	 			/*
+	 			 * P4 quirks:
+	 			 * - An overflown perfctr will assert its interrupt
+	 			 *   until the OVF flag in its CCCR is cleared.
+	 			 * - LVTPC is masked on interrupt and must be
+	 			 *   unmasked by the LVTPC handler.
+	 			 */
+				rdmsrl(wd->cccr_msr, dummy);
+				dummy &= ~P4_CCCR_OVF;
+	 			wrmsrl(wd->cccr_msr, dummy);
+	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	 		}
+			/* start the cycle over again */
+			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+		}
 	}
+done:
+	return;
 }
 
 static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 62a784cb8f0c..5918136fd853 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -63,7 +63,7 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-extern void setup_apic_nmi_watchdog (void);
+extern void setup_apic_nmi_watchdog (void *);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
@@ -73,6 +73,7 @@ extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 extern void nmi_watchdog_default(void);
 extern int setup_nmi_watchdog(char *);
 
+extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
 #define NMI_DEFAULT	-1
 #define NMI_NONE	0
-- 
cgit v1.2.1


From b7471c6da94d30d3deadc55986cc38d1ff57f9ca Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] i386: Add SMP support on i386 to reservation framework

This patch includes the changes to make the nmi watchdog on i386 SMP aware.
A bunch of code was moved around to make it simpler to read.  In addition,
it is now possible to determine if a particular NMI was the result of the
watchdog or not.  This feature allows the kernel to filter out unknown NMIs
easier.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/apic.c            |   3 +-
 arch/i386/kernel/nmi.c             | 537 ++++++++++++++++++++++++-------------
 arch/i386/kernel/traps.c           |   2 +-
 arch/i386/oprofile/nmi_timer_int.c |   4 +-
 include/asm-i386/nmi.h             |   5 +-
 5 files changed, 360 insertions(+), 191 deletions(-)

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 8c844d07862f..1a34fc57800b 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -586,8 +586,7 @@ void __devinit setup_local_APIC(void)
 			printk("No ESR for 82489DX.\n");
 	}
 
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-		setup_apic_nmi_watchdog();
+	setup_apic_nmi_watchdog(NULL);
 	apic_pm_activate();
 }
 
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 5d58dfeacd59..d88004343034 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -24,16 +24,10 @@
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
+#include <asm/kdebug.h>
 
 #include "mach_traps.h"
 
-unsigned int nmi_watchdog = NMI_NONE;
-extern int unknown_nmi_panic;
-static unsigned int nmi_hz = HZ;
-static unsigned int nmi_perfctr_msr;	/* the MSR to reset in NMI handler */
-static unsigned int nmi_p4_cccr_val;
-extern void show_registers(struct pt_regs *regs);
-
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
  * - different performance counters/ event selection may be reserved for
@@ -63,51 +57,31 @@ static unsigned int lapic_nmi_owner;
 #define LAPIC_NMI_RESERVED	(1<<1)
 
 /* nmi_active:
- * +1: the lapic NMI watchdog is active, but can be disabled
- *  0: the lapic NMI watchdog has not been set up, and cannot
+ * >0: the lapic NMI watchdog is active, but can be disabled
+ * <0: the lapic NMI watchdog has not been set up, and cannot
  *     be enabled
- * -1: the lapic NMI watchdog is disabled, but can be enabled
+ *  0: the lapic NMI watchdog is disabled, but can be enabled
  */
-int nmi_active;
+atomic_t nmi_active = ATOMIC_INIT(0);		/* oprofile uses this */
 
-#define K7_EVNTSEL_ENABLE	(1 << 22)
-#define K7_EVNTSEL_INT		(1 << 20)
-#define K7_EVNTSEL_OS		(1 << 17)
-#define K7_EVNTSEL_USR		(1 << 16)
-#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
-#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+unsigned int nmi_watchdog = NMI_DEFAULT;
+static unsigned int nmi_hz = HZ;
 
-#define P6_EVNTSEL0_ENABLE	(1 << 22)
-#define P6_EVNTSEL_INT		(1 << 20)
-#define P6_EVNTSEL_OS		(1 << 17)
-#define P6_EVNTSEL_USR		(1 << 16)
-#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
-#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED
+struct nmi_watchdog_ctlblk {
+	int enabled;
+	u64 check_bit;
+	unsigned int cccr_msr;
+	unsigned int perfctr_msr;  /* the MSR to reset in NMI handler */
+	unsigned int evntsel_msr;  /* the MSR to select the events to handle */
+};
+static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
-#define MSR_P4_MISC_ENABLE	0x1A0
-#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
-#define MSR_P4_MISC_ENABLE_PEBS_UNAVAIL	(1<<12)
-#define MSR_P4_PERFCTR0		0x300
-#define MSR_P4_CCCR0		0x360
-#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
-#define P4_ESCR_OS		(1<<3)
-#define P4_ESCR_USR		(1<<2)
-#define P4_CCCR_OVF_PMI0	(1<<26)
-#define P4_CCCR_OVF_PMI1	(1<<27)
-#define P4_CCCR_THRESHOLD(N)	((N)<<20)
-#define P4_CCCR_COMPLEMENT	(1<<19)
-#define P4_CCCR_COMPARE		(1<<18)
-#define P4_CCCR_REQUIRED	(3<<16)
-#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
-#define P4_CCCR_ENABLE		(1<<12)
-/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
-   CRU_ESCR0 (with any non-null event selector) through a complemented
-   max threshold. [IA32-Vol3, Section 14.9.9] */
-#define MSR_P4_IQ_COUNTER0	0x30C
-#define P4_NMI_CRU_ESCR0	(P4_ESCR_EVENT_SELECT(0x3F)|P4_ESCR_OS|P4_ESCR_USR)
-#define P4_NMI_IQ_CCCR0	\
-	(P4_CCCR_OVF_PMI0|P4_CCCR_THRESHOLD(15)|P4_CCCR_COMPLEMENT|	\
-	 P4_CCCR_COMPARE|P4_CCCR_REQUIRED|P4_CCCR_ESCR_SELECT(4)|P4_CCCR_ENABLE)
+/* local prototypes */
+static void stop_apic_nmi_watchdog(void *unused);
+static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
+
+extern void show_registers(struct pt_regs *regs);
+extern int unknown_nmi_panic;
 
 /* converts an msr to an appropriate reservation bit */
 static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
@@ -208,6 +182,17 @@ void release_evntsel_nmi(unsigned int msr)
 	clear_bit(counter, &__get_cpu_var(evntsel_nmi_owner)[0]);
 }
 
+static __cpuinit inline int nmi_known_cpu(void)
+{
+	switch (boot_cpu_data.x86_vendor) {
+	case X86_VENDOR_AMD:
+		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+	case X86_VENDOR_INTEL:
+		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+	}
+	return 0;
+}
+
 #ifdef CONFIG_SMP
 /* The performance counters used by NMI_LOCAL_APIC don't trigger when
  * the CPU is idle. To make sure the NMI watchdog really ticks on all
@@ -234,7 +219,10 @@ static int __init check_nmi_watchdog(void)
 	unsigned int *prev_nmi_count;
 	int cpu;
 
-	if (nmi_watchdog == NMI_NONE)
+	if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT))
+		return 0;
+
+	if (!atomic_read(&nmi_active))
 		return 0;
 
 	prev_nmi_count = kmalloc(NR_CPUS * sizeof(int), GFP_KERNEL);
@@ -258,18 +246,22 @@ static int __init check_nmi_watchdog(void)
 		if (!cpu_isset(cpu, cpu_callin_map))
 			continue;
 #endif
+		if (!per_cpu(nmi_watchdog_ctlblk, cpu).enabled)
+			continue;
 		if (nmi_count(cpu) - prev_nmi_count[cpu] <= 5) {
-			endflag = 1;
 			printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
 				cpu,
 				prev_nmi_count[cpu],
 				nmi_count(cpu));
-			nmi_active = 0;
-			lapic_nmi_owner &= ~LAPIC_NMI_WATCHDOG;
-			kfree(prev_nmi_count);
-			return -1;
+			per_cpu(nmi_watchdog_ctlblk, cpu).enabled = 0;
+			atomic_dec(&nmi_active);
 		}
 	}
+	if (!atomic_read(&nmi_active)) {
+		kfree(prev_nmi_count);
+		atomic_set(&nmi_active, -1);
+		return -1;
+	}
 	endflag = 1;
 	printk("OK.\n");
 
@@ -290,31 +282,16 @@ static int __init setup_nmi_watchdog(char *str)
 
 	get_option(&str, &nmi);
 
-	if (nmi >= NMI_INVALID)
+	if ((nmi >= NMI_INVALID) || (nmi < NMI_NONE))
 		return 0;
-	if (nmi == NMI_NONE)
-		nmi_watchdog = nmi;
 	/*
 	 * If any other x86 CPU has a local APIC, then
 	 * please test the NMI stuff there and send me the
 	 * missing bits. Right now Intel P6/P4 and AMD K7 only.
 	 */
-	if ((nmi == NMI_LOCAL_APIC) &&
-			(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
-			(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
-		nmi_watchdog = nmi;
-	if ((nmi == NMI_LOCAL_APIC) &&
-			(boot_cpu_data.x86_vendor == X86_VENDOR_AMD) &&
-	  		(boot_cpu_data.x86 == 6 || boot_cpu_data.x86 == 15))
-		nmi_watchdog = nmi;
-	/*
-	 * We can enable the IO-APIC watchdog
-	 * unconditionally.
-	 */
-	if (nmi == NMI_IO_APIC) {
-		nmi_active = 1;
-		nmi_watchdog = nmi;
-	}
+	if ((nmi == NMI_LOCAL_APIC) && (nmi_known_cpu() == 0))
+		return 0;  /* no lapic support */
+	nmi_watchdog = nmi;
 	return 1;
 }
 
@@ -322,41 +299,30 @@ __setup("nmi_watchdog=", setup_nmi_watchdog);
 
 static void disable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active <= 0)
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		wrmsr(MSR_K7_EVNTSEL0, 0, 0);
-		break;
-	case X86_VENDOR_INTEL:
-		switch (boot_cpu_data.x86) {
-		case 6:
-			if (boot_cpu_data.x86_model > 0xd)
-				break;
 
-			wrmsr(MSR_P6_EVNTSEL0, 0, 0);
-			break;
-		case 15:
-			if (boot_cpu_data.x86_model > 0x4)
-				break;
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
 
-			wrmsr(MSR_P4_IQ_CCCR0, 0, 0);
-			wrmsr(MSR_P4_CRU_ESCR0, 0, 0);
-			break;
-		}
-		break;
-	}
-	nmi_active = -1;
-	/* tell do_nmi() and others that we're not active any more */
-	nmi_watchdog = 0;
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 static void enable_lapic_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_LOCAL_APIC;
-		setup_apic_nmi_watchdog();
-	}
+	BUG_ON(nmi_watchdog != NMI_LOCAL_APIC);
+
+	/* are we already enabled */
+	if (atomic_read(&nmi_active) != 0)
+		return;
+
+	/* are we lapic aware */
+	if (nmi_known_cpu() <= 0)
+		return;
+
+	on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+	touch_nmi_watchdog();
 }
 
 int reserve_lapic_nmi(void)
@@ -388,20 +354,25 @@ void release_lapic_nmi(void)
 
 void disable_timer_nmi_watchdog(void)
 {
-	if ((nmi_watchdog != NMI_IO_APIC) || (nmi_active <= 0))
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) <= 0)
 		return;
 
-	unset_nmi_callback();
-	nmi_active = -1;
-	nmi_watchdog = NMI_NONE;
+	disable_irq(0);
+	on_each_cpu(stop_apic_nmi_watchdog, NULL, 0, 1);
+
+	BUG_ON(atomic_read(&nmi_active) != 0);
 }
 
 void enable_timer_nmi_watchdog(void)
 {
-	if (nmi_active < 0) {
-		nmi_watchdog = NMI_IO_APIC;
+	BUG_ON(nmi_watchdog != NMI_IO_APIC);
+
+	if (atomic_read(&nmi_active) == 0) {
 		touch_nmi_watchdog();
-		nmi_active = 1;
+		on_each_cpu(setup_apic_nmi_watchdog, NULL, 0, 1);
+		enable_irq(0);
 	}
 }
 
@@ -411,7 +382,7 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
-	nmi_pm_active = nmi_active;
+	nmi_pm_active = atomic_read(&nmi_active);
 	disable_lapic_nmi_watchdog();
 	return 0;
 }
@@ -439,7 +410,13 @@ static int __init init_lapic_nmi_sysfs(void)
 {
 	int error;
 
-	if (nmi_active == 0 || nmi_watchdog != NMI_LOCAL_APIC)
+	/* should really be a BUG_ON but b/c this is an
+	 * init call, it just doesn't work.  -dcz
+	 */
+	if (nmi_watchdog != NMI_LOCAL_APIC)
+		return 0;
+
+	if ( atomic_read(&nmi_active) < 0 )
 		return 0;
 
 	error = sysdev_class_register(&nmi_sysclass);
@@ -457,143 +434,312 @@ late_initcall(init_lapic_nmi_sysfs);
  * Original code written by Keith Owens.
  */
 
-static void write_watchdog_counter(const char *descr)
+static void write_watchdog_counter(unsigned int perfctr_msr, const char *descr)
 {
 	u64 count = (u64)cpu_khz * 1000;
 
 	do_div(count, nmi_hz);
 	if(descr)
 		Dprintk("setting %s to -0x%08Lx\n", descr, count);
-	wrmsrl(nmi_perfctr_msr, 0 - count);
+	wrmsrl(perfctr_msr, 0 - count);
 }
 
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define K7_EVNTSEL_ENABLE	(1 << 22)
+#define K7_EVNTSEL_INT		(1 << 20)
+#define K7_EVNTSEL_OS		(1 << 17)
+#define K7_EVNTSEL_USR		(1 << 16)
+#define K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING	0x76
+#define K7_NMI_EVENT		K7_EVENT_CYCLES_PROCESSOR_IS_RUNNING
+
 static int setup_k7_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_K7_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_K7_PERFCTR0;
+	evntsel_msr = MSR_K7_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_K7_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsrl(MSR_K7_PERFCTR0, 0UL);
+	wrmsrl(perfctr_msr, 0UL);
 
 	evntsel = K7_EVNTSEL_INT
 		| K7_EVNTSEL_OS
 		| K7_EVNTSEL_USR
 		| K7_NMI_EVENT;
 
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
-	write_watchdog_counter("K7_PERFCTR0");
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "K7_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= K7_EVNTSEL_ENABLE;
-	wrmsr(MSR_K7_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<63;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_k7_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+#define P6_EVNTSEL0_ENABLE	(1 << 22)
+#define P6_EVNTSEL_INT		(1 << 20)
+#define P6_EVNTSEL_OS		(1 << 17)
+#define P6_EVNTSEL_USR		(1 << 16)
+#define P6_EVENT_CPU_CLOCKS_NOT_HALTED	0x79
+#define P6_NMI_EVENT		P6_EVENT_CPU_CLOCKS_NOT_HALTED
+
 static int setup_p6_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr;
 	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	nmi_perfctr_msr = MSR_P6_PERFCTR0;
-
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	perfctr_msr = MSR_P6_PERFCTR0;
+	evntsel_msr = MSR_P6_EVNTSEL0;
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P6_EVNTSEL0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
+	wrmsrl(perfctr_msr, 0UL);
+
 	evntsel = P6_EVNTSEL_INT
 		| P6_EVNTSEL_OS
 		| P6_EVNTSEL_USR
 		| P6_NMI_EVENT;
 
-	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
-	write_watchdog_counter("P6_PERFCTR0");
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "P6_PERFCTR0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 	evntsel |= P6_EVNTSEL0_ENABLE;
-	wrmsr(MSR_P6_EVNTSEL0, evntsel, 0);
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
+static void stop_p6_watchdog(void)
+{
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+/* Note that these events don't tick when the CPU idles. This means
+   the frequency varies with CPU load. */
+
+#define MSR_P4_MISC_ENABLE_PERF_AVAIL	(1<<7)
+#define P4_ESCR_EVENT_SELECT(N)	((N)<<25)
+#define P4_ESCR_OS		(1<<3)
+#define P4_ESCR_USR		(1<<2)
+#define P4_CCCR_OVF_PMI0	(1<<26)
+#define P4_CCCR_OVF_PMI1	(1<<27)
+#define P4_CCCR_THRESHOLD(N)	((N)<<20)
+#define P4_CCCR_COMPLEMENT	(1<<19)
+#define P4_CCCR_COMPARE		(1<<18)
+#define P4_CCCR_REQUIRED	(3<<16)
+#define P4_CCCR_ESCR_SELECT(N)	((N)<<13)
+#define P4_CCCR_ENABLE		(1<<12)
+#define P4_CCCR_OVF 		(1<<31)
+/* Set up IQ_COUNTER0 to behave like a clock, by having IQ_CCCR0 filter
+   CRU_ESCR0 (with any non-null event selector) through a complemented
+   max threshold. [IA32-Vol3, Section 14.9.9] */
+
 static int setup_p4_watchdog(void)
 {
+	unsigned int perfctr_msr, evntsel_msr, cccr_msr;
+	unsigned int evntsel, cccr_val;
 	unsigned int misc_enable, dummy;
+	unsigned int ht_num;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 
-	rdmsr(MSR_P4_MISC_ENABLE, misc_enable, dummy);
+	rdmsr(MSR_IA32_MISC_ENABLE, misc_enable, dummy);
 	if (!(misc_enable & MSR_P4_MISC_ENABLE_PERF_AVAIL))
 		return 0;
 
-	nmi_perfctr_msr = MSR_P4_IQ_COUNTER0;
-	nmi_p4_cccr_val = P4_NMI_IQ_CCCR0;
 #ifdef CONFIG_SMP
-	if (smp_num_siblings == 2)
-		nmi_p4_cccr_val |= P4_CCCR_OVF_PMI1;
+	/* detect which hyperthread we are on */
+	if (smp_num_siblings == 2) {
+		unsigned int ebx, apicid;
+
+        	ebx = cpuid_ebx(1);
+	        apicid = (ebx >> 24) & 0xff;
+        	ht_num = apicid & 1;
+	} else
 #endif
+		ht_num = 0;
+
+	/* performance counters are shared resources
+	 * assign each hyperthread its own set
+	 * (re-use the ESCR0 register, seems safe
+	 * and keeps the cccr_val the same)
+	 */
+	if (!ht_num) {
+		/* logical cpu 0 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR0;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR0;
+		cccr_val = P4_CCCR_OVF_PMI0 | P4_CCCR_ESCR_SELECT(4);
+	} else {
+		/* logical cpu 1 */
+		perfctr_msr = MSR_P4_IQ_PERFCTR1;
+		evntsel_msr = MSR_P4_CRU_ESCR0;
+		cccr_msr = MSR_P4_IQ_CCCR1;
+		cccr_val = P4_CCCR_OVF_PMI1 | P4_CCCR_ESCR_SELECT(4);
+	}
 
-	if (!reserve_perfctr_nmi(nmi_perfctr_msr))
+	if (!reserve_perfctr_nmi(perfctr_msr))
 		goto fail;
 
-	if (!reserve_evntsel_nmi(MSR_P4_CRU_ESCR0))
+	if (!reserve_evntsel_nmi(evntsel_msr))
 		goto fail1;
 
-	wrmsr(MSR_P4_CRU_ESCR0, P4_NMI_CRU_ESCR0, 0);
-	wrmsr(MSR_P4_IQ_CCCR0, P4_NMI_IQ_CCCR0 & ~P4_CCCR_ENABLE, 0);
-	write_watchdog_counter("P4_IQ_COUNTER0");
+	evntsel = P4_ESCR_EVENT_SELECT(0x3F)
+	 	| P4_ESCR_OS
+		| P4_ESCR_USR;
+
+	cccr_val |= P4_CCCR_THRESHOLD(15)
+		 | P4_CCCR_COMPLEMENT
+		 | P4_CCCR_COMPARE
+		 | P4_CCCR_REQUIRED;
+
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsr(cccr_msr, cccr_val, 0);
+	write_watchdog_counter(perfctr_msr, "P4_IQ_COUNTER0");
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
-	wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
+	cccr_val |= P4_CCCR_ENABLE;
+	wrmsr(cccr_msr, cccr_val, 0);
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = cccr_msr;
+	wd->check_bit = 1ULL<<39;
 	return 1;
 fail1:
-	release_perfctr_nmi(nmi_perfctr_msr);
+	release_perfctr_nmi(perfctr_msr);
 fail:
 	return 0;
 }
 
-void setup_apic_nmi_watchdog (void)
+static void stop_p4_watchdog(void)
 {
-	switch (boot_cpu_data.x86_vendor) {
-	case X86_VENDOR_AMD:
-		if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
-			return;
-		if (!setup_k7_watchdog())
-			return;
-		break;
-	case X86_VENDOR_INTEL:
-		switch (boot_cpu_data.x86) {
-		case 6:
-			if (boot_cpu_data.x86_model > 0xd)
-				return;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	wrmsr(wd->cccr_msr, 0, 0);
+	wrmsr(wd->evntsel_msr, 0, 0);
 
-			if(!setup_p6_watchdog())
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
+void setup_apic_nmi_watchdog (void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			if (boot_cpu_data.x86 != 6 && boot_cpu_data.x86 != 15)
 				return;
-			break;
-		case 15:
-			if (boot_cpu_data.x86_model > 0x4)
+			if (!setup_k7_watchdog())
 				return;
+			break;
+		case X86_VENDOR_INTEL:
+			switch (boot_cpu_data.x86) {
+			case 6:
+				if (boot_cpu_data.x86_model > 0xd)
+					return;
+
+				if (!setup_p6_watchdog())
+					return;
+				break;
+			case 15:
+				if (boot_cpu_data.x86_model > 0x4)
+					return;
 
-			if (!setup_p4_watchdog())
+				if (!setup_p4_watchdog())
+					return;
+				break;
+			default:
 				return;
+			}
+			break;
+		default:
+			return;
+		}
+	}
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	atomic_inc(&nmi_active);
+}
+
+static void stop_apic_nmi_watchdog(void *unused)
+{
+	/* only support LOCAL and IO APICs for now */
+	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
+	    (nmi_watchdog != NMI_IO_APIC))
+	    	return;
+
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		switch (boot_cpu_data.x86_vendor) {
+		case X86_VENDOR_AMD:
+			stop_k7_watchdog();
+			break;
+		case X86_VENDOR_INTEL:
+			switch (boot_cpu_data.x86) {
+			case 6:
+				if (boot_cpu_data.x86_model > 0xd)
+					break;
+				stop_p6_watchdog();
+				break;
+			case 15:
+				if (boot_cpu_data.x86_model > 0x4)
+					break;
+				stop_p4_watchdog();
+				break;
+			}
 			break;
 		default:
 			return;
 		}
-		break;
-	default:
-		return;
 	}
-	lapic_nmi_owner = LAPIC_NMI_WATCHDOG;
-	nmi_active = 1;
+	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	atomic_dec(&nmi_active);
 }
 
 /*
@@ -635,7 +781,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-void nmi_watchdog_tick (struct pt_regs * regs)
+void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 {
 
 	/*
@@ -644,11 +790,21 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 	 * smp_processor_id().
 	 */
 	unsigned int sum;
+	int touched = 0;
 	int cpu = smp_processor_id();
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+	u64 dummy;
+
+	/* check for other users first */
+	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
+			== NOTIFY_STOP) {
+		touched = 1;
+	}
 
 	sum = per_cpu(irq_stat, cpu).apic_timer_irqs;
 
-	if (last_irq_sums[cpu] == sum) {
+	/* if the apic timer isn't firing, this cpu isn't doing much */
+	if (!touched && last_irq_sums[cpu] == sum) {
 		/*
 		 * Ayiee, looks like this CPU is stuck ...
 		 * wait a few IRQs (5 seconds) before doing the oops ...
@@ -663,26 +819,41 @@ void nmi_watchdog_tick (struct pt_regs * regs)
 		last_irq_sums[cpu] = sum;
 		alert_counter[cpu] = 0;
 	}
-	if (nmi_perfctr_msr) {
-		if (nmi_perfctr_msr == MSR_P4_IQ_COUNTER0) {
-			/*
-			 * P4 quirks:
-			 * - An overflown perfctr will assert its interrupt
-			 *   until the OVF flag in its CCCR is cleared.
-			 * - LVTPC is masked on interrupt and must be
-			 *   unmasked by the LVTPC handler.
-			 */
-			wrmsr(MSR_P4_IQ_CCCR0, nmi_p4_cccr_val, 0);
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
-		}
-		else if (nmi_perfctr_msr == MSR_P6_PERFCTR0) {
-			/* Only P6 based Pentium M need to re-unmask
-			 * the apic vector but it doesn't hurt
-			 * other P6 variant */
-			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	/* see if the nmi watchdog went off */
+	if (wd->enabled) {
+		if (nmi_watchdog == NMI_LOCAL_APIC) {
+			rdmsrl(wd->perfctr_msr, dummy);
+			if (dummy & wd->check_bit){
+				/* this wasn't a watchdog timer interrupt */
+				goto done;
+			}
+
+			/* only Intel P4 uses the cccr msr */
+	 		if (wd->cccr_msr != 0) {
+	 			/*
+	 			 * P4 quirks:
+	 			 * - An overflown perfctr will assert its interrupt
+	 			 *   until the OVF flag in its CCCR is cleared.
+	 			 * - LVTPC is masked on interrupt and must be
+	 			 *   unmasked by the LVTPC handler.
+	 			 */
+				rdmsrl(wd->cccr_msr, dummy);
+				dummy &= ~P4_CCCR_OVF;
+	 			wrmsrl(wd->cccr_msr, dummy);
+	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
+	 		}
+			else if (wd->perfctr_msr == MSR_P6_PERFCTR0) {
+				/* Only P6 based Pentium M need to re-unmask
+				 * the apic vector but it doesn't hurt
+				 * other P6 variant */
+				apic_write(APIC_LVTPC, APIC_DM_NMI);
+			}
+			/* start the cycle over again */
+			write_watchdog_counter(wd->perfctr_msr, NULL);
 		}
-		write_watchdog_counter(NULL);
 	}
+done:
+	return;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7e9edafffd8a..3a07b2677e2a 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -724,7 +724,7 @@ static void default_do_nmi(struct pt_regs * regs)
 		 * so it must be the NMI watchdog.
 		 */
 		if (nmi_watchdog) {
-			nmi_watchdog_tick(regs);
+			nmi_watchdog_tick(regs, reason);
 			return;
 		}
 #endif
diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c
index 930a1127bb30..a33a73bb502d 100644
--- a/arch/i386/oprofile/nmi_timer_int.c
+++ b/arch/i386/oprofile/nmi_timer_int.c
@@ -42,9 +42,7 @@ static void timer_stop(void)
 
 int __init op_nmi_timer_init(struct oprofile_operations * ops)
 {
-	extern int nmi_active;
-
-	if (nmi_active <= 0)
+	if (atomic_read(&nmi_active) <= 0)
 		return -ENODEV;
 
 	ops->start = timer_start;
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 27fc9e6f630e..4cda6801ecb8 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -32,13 +32,14 @@ extern void release_perfctr_nmi(unsigned int);
 extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
-extern void setup_apic_nmi_watchdog (void);
+extern void setup_apic_nmi_watchdog (void *);
 extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs);
+extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
+extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
 #define NMI_DEFAULT     -1
 #define NMI_NONE	0
-- 
cgit v1.2.1


From 3adbbcce9a49b900d4cc118cdccfdefa78bf1afb Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] x86: Cleanup NMI interrupt path

This patch cleans up the NMI interrupt path.  Instead of being gated by if
the 'nmi callback' is set, the interrupt handler now calls everyone who is
registered on the die_chain and additionally checks the nmi watchdog,
reseting it if enabled.  This allows more subsystems to hook into the NMI if
they need to (without being block by set_nmi_callback).

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c     | 16 +++++++++++++---
 arch/i386/kernel/traps.c   | 24 +++++++++++-------------
 arch/x86_64/kernel/nmi.c   | 26 +++++++++++++++++++-------
 arch/x86_64/kernel/traps.c |  8 ++++----
 include/asm-i386/nmi.h     |  2 +-
 include/asm-x86_64/nmi.h   | 10 +++++++++-
 6 files changed, 57 insertions(+), 29 deletions(-)

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index d88004343034..bd96ea4f2942 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -781,7 +781,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
+int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 {
 
 	/*
@@ -794,10 +794,12 @@ void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 	int cpu = smp_processor_id();
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 	u64 dummy;
+	int rc=0;
 
 	/* check for other users first */
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
 			== NOTIFY_STOP) {
+		rc = 1;
 		touched = 1;
 	}
 
@@ -850,10 +852,18 @@ void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 			}
 			/* start the cycle over again */
 			write_watchdog_counter(wd->perfctr_msr, NULL);
-		}
+			rc = 1;
+		} else if (nmi_watchdog == NMI_IO_APIC) {
+			/* don't know how to accurately check for this.
+			 * just assume it was a watchdog timer interrupt
+			 * This matches the old behaviour.
+			 */
+			rc = 1;
+		} else
+			printk(KERN_WARNING "Unknown enabled NMI hardware?!\n");
 	}
 done:
-	return;
+	return rc;
 }
 
 #ifdef CONFIG_SYSCTL
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 3a07b2677e2a..282f0bd40dfd 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -706,6 +706,13 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
+static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
+{
+	return 0;
+}
+
+static nmi_callback_t nmi_callback = dummy_nmi_callback;
+
 static void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = 0;
@@ -723,12 +730,11 @@ static void default_do_nmi(struct pt_regs * regs)
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
 		 */
-		if (nmi_watchdog) {
-			nmi_watchdog_tick(regs, reason);
+		if (nmi_watchdog_tick(regs, reason))
 			return;
-		}
 #endif
-		unknown_nmi_error(reason, regs);
+		if (!rcu_dereference(nmi_callback)(regs, smp_processor_id()))
+			unknown_nmi_error(reason, regs);
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -744,13 +750,6 @@ static void default_do_nmi(struct pt_regs * regs)
 	reassert_nmi();
 }
 
-static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
- 
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
- 
 fastcall void do_nmi(struct pt_regs * regs, long error_code)
 {
 	int cpu;
@@ -761,8 +760,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 
 	++nmi_count(cpu);
 
-	if (!rcu_dereference(nmi_callback)(regs, cpu))
-		default_do_nmi(regs);
+	default_do_nmi(regs);
 
 	nmi_exit();
 }
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index d42374a952d7..f6b881b23a70 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -682,16 +682,18 @@ void touch_nmi_watchdog (void)
  	touch_softlockup_watchdog();
 }
 
-void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
+int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 	int sum;
 	int touched = 0;
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
 	u64 dummy;
+	int rc=0;
 
 	/* check for other users first */
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT)
 			== NOTIFY_STOP) {
+		rc = 1;
 		touched = 1;
 	}
 
@@ -746,10 +748,18 @@ void __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 	 		}
 			/* start the cycle over again */
 			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
-		}
+			rc = 1;
+		} else 	if (nmi_watchdog == NMI_IO_APIC) {
+			/* don't know how to accurately check for this.
+			 * just assume it was a watchdog timer interrupt
+			 * This matches the old behaviour.
+			 */
+			rc = 1;
+		} else
+			printk(KERN_WARNING "Unknown enabled NMI hardware?!\n");
 	}
 done:
-	return;
+	return rc;
 }
 
 static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
@@ -761,15 +771,17 @@ static nmi_callback_t nmi_callback = dummy_nmi_callback;
  
 asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
-	int cpu = safe_smp_processor_id();
-
 	nmi_enter();
 	add_pda(__nmi_count,1);
-	if (!rcu_dereference(nmi_callback)(regs, cpu))
-		default_do_nmi(regs);
+	default_do_nmi(regs);
 	nmi_exit();
 }
 
+int do_nmi_callback(struct pt_regs * regs, int cpu)
+{
+	return rcu_dereference(nmi_callback)(regs, cpu);
+}
+
 void set_nmi_callback(nmi_callback_t callback)
 {
 	vmalloc_sync_all();
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index b1249774d1e8..42bc070fdf11 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -781,12 +781,12 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
 		 */
-		if (nmi_watchdog > 0) {
-			nmi_watchdog_tick(regs,reason);
+		if (nmi_watchdog_tick(regs,reason))
 			return;
-		}
+		if (!do_nmi_callback(regs,cpu))
 #endif
-		unknown_nmi_error(reason, regs);
+			unknown_nmi_error(reason, regs);
+
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 4cda6801ecb8..da0e0b4e9139 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -37,7 +37,7 @@ extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
+extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
 extern atomic_t nmi_active;
 extern unsigned int nmi_watchdog;
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 5918136fd853..8f02a2a416e6 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -26,6 +26,14 @@ void set_nmi_callback(nmi_callback_t callback);
  */
 void unset_nmi_callback(void);
 
+/**
+ * do_nmi_callback
+ *
+ * Check to see if a callback exists and execute it.  Return 1
+ * if the handler exists and was handled successfully.
+ */
+int do_nmi_callback(struct pt_regs *regs, int cpu);
+
 #ifdef CONFIG_PM
  
 /** Replace the PM callback routine for NMI. */
@@ -68,7 +76,7 @@ extern int reserve_lapic_nmi(void);
 extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
-extern void nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
+extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
 
 extern void nmi_watchdog_default(void);
 extern int setup_nmi_watchdog(char *);
-- 
cgit v1.2.1


From 1d001df19d5323e642ba8ac821c713675ebccd82 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:26 +0200
Subject: [PATCH] Add TIF_RESTORE_SIGMASK

We need TIF_RESTORE_SIGMASK in order to support ppoll() and pselect()
system calls. This patch originally came from Andi, and was based
heavily on David Howells' implementation of same on i386. I fixed a typo
which was causing do_signal() to use the wrong signal mask.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_signal.c   | 28 ++++++--------
 arch/x86_64/kernel/signal.c      | 82 +++++++++++++++++-----------------------
 include/asm-x86_64/signal.h      |  4 --
 include/asm-x86_64/thread_info.h |  2 +
 include/asm-x86_64/unistd.h      |  1 +
 5 files changed, 49 insertions(+), 68 deletions(-)

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index 25e5ca22204c..549de439fb2d 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -113,25 +113,19 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
 }
 
 asmlinkage long
-sys32_sigsuspend(int history0, int history1, old_sigset_t mask,
-		 struct pt_regs *regs)
+sys32_sigsuspend(int history0, int history1, old_sigset_t mask)
 {
-	sigset_t saveset;
-
 	mask &= _BLOCKABLE;
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	siginitset(&current->blocked, mask);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs->rax = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage long
@@ -508,11 +502,11 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
 
 int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
@@ -595,7 +589,7 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	regs->ss = __USER32_DS; 
 
 	set_fs(USER_DS);
-    regs->eflags &= ~TF_MASK;
+	regs->eflags &= ~TF_MASK;
     if (test_thread_flag(TIF_SINGLESTEP))
         ptrace_notify(SIGTRAP);
 
@@ -604,9 +598,9 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 28161170fb0a..7f58bc9a056d 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -37,37 +37,6 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 int ia32_setup_frame(int sig, struct k_sigaction *ka,
             sigset_t *set, struct pt_regs * regs); 
 
-asmlinkage long
-sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize, struct pt_regs *regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-#ifdef DEBUG_SIG
-	printk("rt_sigsuspend savset(%lx) newset(%lx) regs(%p) rip(%lx)\n",
-		saveset, newset, regs, regs->rip);
-#endif 
-	regs->rax = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
-}
-
 asmlinkage long
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		struct pt_regs *regs)
@@ -341,11 +310,11 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		current->comm, current->pid, frame, regs->rip, frame->pretcode);
 #endif
 
-	return 1;
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
-	return 0;
+	return -EFAULT;
 }
 
 /*
@@ -408,7 +377,7 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
 #endif
 	ret = setup_rt_frame(sig, ka, info, oldset, regs);
 
-	if (ret) {
+	if (ret == 0) {
 		spin_lock_irq(&current->sighand->siglock);
 		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
 		if (!(ka->sa.sa_flags & SA_NODEFER))
@@ -425,11 +394,12 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
  * want to handle. Thus you cannot kill init even with a SIGKILL even by
  * mistake.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
+static void do_signal(struct pt_regs *regs)
 {
 	struct k_sigaction ka;
 	siginfo_t info;
 	int signr;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which
@@ -438,9 +408,11 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
@@ -454,30 +426,46 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 			set_debugreg(current->thread.debugreg7, 7);
 
 		/* Whee!  Actually deliver the signal.  */
-		return handle_signal(signr, &info, &ka, oldset, regs);
+		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
+		return;
 	}
 
 	/* Did we come from a system call? */
 	if ((long)regs->orig_rax >= 0) {
 		/* Restart the system call - no handlers present */
 		long res = regs->rax;
-		if (res == -ERESTARTNOHAND ||
-		    res == -ERESTARTSYS ||
-		    res == -ERESTARTNOINTR) {
+		switch (res) {
+		case -ERESTARTNOHAND:
+		case -ERESTARTSYS:
+		case -ERESTARTNOINTR:
 			regs->rax = regs->orig_rax;
 			regs->rip -= 2;
-		}
-		if (regs->rax == (unsigned long)-ERESTART_RESTARTBLOCK) {
+			break;
+		case -ERESTART_RESTARTBLOCK:
 			regs->rax = test_thread_flag(TIF_IA32) ?
 					__NR_ia32_restart_syscall :
 					__NR_restart_syscall;
 			regs->rip -= 2;
+			break;
 		}
 	}
-	return 0;
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	   back. */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
 }
 
-void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, __u32 thread_info_flags)
+void
+do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags)
 {
 #ifdef DEBUG_SIG
 	printk("do_notify_resume flags:%x rip:%lx rsp:%lx caller:%lx pending:%lx\n",
@@ -491,8 +479,8 @@ void do_notify_resume(struct pt_regs *regs, sigset_t *oldset, __u32 thread_info_
 	}
 
 	/* deal with pending signal delivery */
-	if (thread_info_flags & _TIF_SIGPENDING)
-		do_signal(regs,oldset);
+	if (thread_info_flags & (_TIF_SIGPENDING|_TIF_RESTORE_SIGMASK))
+		do_signal(regs);
 }
 
 void signal_fault(struct pt_regs *regs, void __user *frame, char *where)
diff --git a/include/asm-x86_64/signal.h b/include/asm-x86_64/signal.h
index 3ede2a61973a..4581f978b299 100644
--- a/include/asm-x86_64/signal.h
+++ b/include/asm-x86_64/signal.h
@@ -24,10 +24,6 @@ typedef struct {
 } sigset_t;
 
 
-struct pt_regs; 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
-
-
 #else
 /* Here we must cater to libcs that poke about in kernel headers.  */
 
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index 2029b00351f3..790c512a4369 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -114,6 +114,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_IRET		5	/* force IRET */
 #define TIF_SYSCALL_AUDIT	7	/* syscall auditing active */
 #define TIF_SECCOMP		8	/* secure computing */
+#define TIF_RESTORE_SIGMASK	9	/* restore signal mask in do_signal */
 /* 16 free */
 #define TIF_IA32		17	/* 32bit process */ 
 #define TIF_FORK		18	/* ret_from_fork */
@@ -128,6 +129,7 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_IRET		(1<<TIF_IRET)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1<<TIF_SECCOMP)
+#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_IA32		(1<<TIF_IA32)
 #define _TIF_FORK		(1<<TIF_FORK)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index 80fd48e84bbb..f266de294003 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -658,6 +658,7 @@ do { \
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
 #define __ARCH_WANT_SYS_TIME
 #define __ARCH_WANT_COMPAT_SYS_TIME
 
-- 
cgit v1.2.1


From 957dc87c1bd849440f0eef27e2ade67387001e13 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] Add ppoll/pselect syscalls

Needed TIF_RESTORE_SIGMASK first

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32entry.S | 4 ++--
 include/asm-x86_64/unistd.h  | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 5d4a7d125ed0..30ed0f6f4a2b 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -703,8 +703,8 @@ ia32_sys_call_table:
 	.quad sys_readlinkat		/* 305 */
 	.quad sys_fchmodat
 	.quad sys_faccessat
-	.quad quiet_ni_syscall		/* pselect6 for now */
-	.quad quiet_ni_syscall		/* ppoll for now */
+	.quad compat_sys_pselect6
+	.quad compat_sys_ppoll
 	.quad sys_unshare		/* 310 */
 	.quad compat_sys_set_robust_list
 	.quad compat_sys_get_robust_list
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index f266de294003..eeb98c168e98 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -600,9 +600,9 @@ __SYSCALL(__NR_fchmodat, sys_fchmodat)
 #define __NR_faccessat		269
 __SYSCALL(__NR_faccessat, sys_faccessat)
 #define __NR_pselect6		270
-__SYSCALL(__NR_pselect6, sys_ni_syscall)	/* for now */
+__SYSCALL(__NR_pselect6, sys_pselect6)
 #define __NR_ppoll		271
-__SYSCALL(__NR_ppoll,	sys_ni_syscall)		/* for now */
+__SYSCALL(__NR_ppoll,	sys_ppoll)
 #define __NR_unshare		272
 __SYSCALL(__NR_unshare,	sys_unshare)
 #define __NR_set_robust_list	273
-- 
cgit v1.2.1


From 2fbe7b25c8edaf2d10e6c1a4cc9f8afe714c4764 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386/x86-64: Remove un/set_nmi_callback and
 reserve/release_lapic_nmi functions

Removes the un/set_nmi_callback and reserve/release_lapic_nmi functions as
they are no longer needed.  The various subsystems are modified to register
with the die_notifier instead.

Also includes compile fixes by Andrew Morton.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/crash.c           |  20 +++++++-
 arch/i386/kernel/nmi.c             |  85 ++++---------------------------
 arch/i386/kernel/traps.c           |  23 +--------
 arch/i386/oprofile/nmi_int.c       |  47 ++++++++++-------
 arch/i386/oprofile/nmi_timer_int.c |  33 +++++++++---
 arch/x86_64/kernel/crash.c         |  20 +++++++-
 arch/x86_64/kernel/nmi.c           | 102 +++----------------------------------
 include/asm-i386/nmi.h             |  21 ++------
 include/asm-x86_64/nmi.h           |  21 --------
 kernel/sysctl.c                    |   4 +-
 10 files changed, 116 insertions(+), 260 deletions(-)

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 5b96f038367f..736c76d6b31d 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -22,6 +22,8 @@
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
 #include <asm/apic.h>
+#include <asm/kdebug.h>
+
 #include <mach_ipi.h>
 
 
@@ -93,9 +95,18 @@ static void crash_save_self(struct pt_regs *regs)
 #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
 static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct pt_regs *regs, int cpu)
+static int crash_nmi_callback(struct notifier_block *self,
+			unsigned long val, void *data)
 {
+	struct pt_regs *regs;
 	struct pt_regs fixed_regs;
+	int cpu;
+
+	if (val != DIE_NMI)
+		return NOTIFY_OK;
+
+	regs = ((struct die_args *)data)->regs;
+	cpu = raw_smp_processor_id();
 
 	/* Don't do anything if this handler is invoked on crashing cpu.
 	 * Otherwise, system will completely hang. Crashing cpu can get
@@ -125,13 +136,18 @@ static void smp_send_nmi_allbutself(void)
 	send_IPI_allbutself(NMI_VECTOR);
 }
 
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
 static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
 
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
 	/* Would it be better to replace the trap vector here? */
-	set_nmi_callback(crash_nmi_callback);
+	if (register_die_notifier(&crash_nmi_nb))
+		return;		/* return what? */
 	/* Ensure the new callback function is set before sending
 	 * out the NMI
 	 */
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index bd96ea4f2942..acd3fdea2a21 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -42,20 +42,6 @@ static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]);
  */
 #define NMI_MAX_COUNTER_BITS 66
 
-/*
- * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
- * - it may be reserved by some other driver, or not
- * - when not reserved by some other driver, it may be used for
- *   the NMI watchdog, or not
- *
- * This is maintained separately from nmi_active because the NMI
- * watchdog may also be driven from the I/O APIC timer.
- */
-static DEFINE_SPINLOCK(lapic_nmi_owner_lock);
-static unsigned int lapic_nmi_owner;
-#define LAPIC_NMI_WATCHDOG	(1<<0)
-#define LAPIC_NMI_RESERVED	(1<<1)
-
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
  * <0: the lapic NMI watchdog has not been set up, and cannot
@@ -325,33 +311,6 @@ static void enable_lapic_nmi_watchdog(void)
 	touch_nmi_watchdog();
 }
 
-int reserve_lapic_nmi(void)
-{
-	unsigned int old_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	old_owner = lapic_nmi_owner;
-	lapic_nmi_owner |= LAPIC_NMI_RESERVED;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (old_owner & LAPIC_NMI_RESERVED)
-		return -EBUSY;
-	if (old_owner & LAPIC_NMI_WATCHDOG)
-		disable_lapic_nmi_watchdog();
-	return 0;
-}
-
-void release_lapic_nmi(void)
-{
-	unsigned int new_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED;
-	lapic_nmi_owner = new_owner;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (new_owner & LAPIC_NMI_WATCHDOG)
-		enable_lapic_nmi_watchdog();
-}
-
 void disable_timer_nmi_watchdog(void)
 {
 	BUG_ON(nmi_watchdog != NMI_IO_APIC);
@@ -866,6 +825,15 @@ done:
 	return rc;
 }
 
+int do_nmi_callback(struct pt_regs * regs, int cpu)
+{
+#ifdef CONFIG_SYSCTL
+	if (unknown_nmi_panic)
+		return unknown_nmi_panic_callback(regs, cpu);
+#endif
+	return 0;
+}
+
 #ifdef CONFIG_SYSCTL
 
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
@@ -873,37 +841,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	unsigned char reason = get_nmi_reason();
 	char buf[64];
 
-	if (!(reason & 0xc0)) {
-		sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-		die_nmi(regs, buf);
-	}
-	return 0;
-}
-
-/*
- * proc handler for /proc/sys/kernel/unknown_nmi_panic
- */
-int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file,
-			void __user *buffer, size_t *length, loff_t *ppos)
-{
-	int old_state;
-
-	old_state = unknown_nmi_panic;
-	proc_dointvec(table, write, file, buffer, length, ppos);
-	if (!!old_state == !!unknown_nmi_panic)
-		return 0;
-
-	if (unknown_nmi_panic) {
-		if (reserve_lapic_nmi() < 0) {
-			unknown_nmi_panic = 0;
-			return -EBUSY;
-		} else {
-			set_nmi_callback(unknown_nmi_panic_callback);
-		}
-	} else {
-		release_lapic_nmi();
-		unset_nmi_callback();
-	}
+	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
+	die_nmi(regs, buf);
 	return 0;
 }
 
@@ -917,7 +856,5 @@ EXPORT_SYMBOL(reserve_perfctr_nmi);
 EXPORT_SYMBOL(release_perfctr_nmi);
 EXPORT_SYMBOL(reserve_evntsel_nmi);
 EXPORT_SYMBOL(release_evntsel_nmi);
-EXPORT_SYMBOL(reserve_lapic_nmi);
-EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
 EXPORT_SYMBOL(enable_timer_nmi_watchdog);
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 282f0bd40dfd..7db664d0b25c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -706,13 +706,6 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
-static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
-
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
-
 static void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = 0;
@@ -732,9 +725,10 @@ static void default_do_nmi(struct pt_regs * regs)
 		 */
 		if (nmi_watchdog_tick(regs, reason))
 			return;
+		if (!do_nmi_callback(regs, smp_processor_id()))
 #endif
-		if (!rcu_dereference(nmi_callback)(regs, smp_processor_id()))
 			unknown_nmi_error(reason, regs);
+
 		return;
 	}
 	if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
@@ -765,19 +759,6 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
 	nmi_exit();
 }
 
-void set_nmi_callback(nmi_callback_t callback)
-{
-	vmalloc_sync_all();
-	rcu_assign_pointer(nmi_callback, callback);
-}
-EXPORT_SYMBOL_GPL(set_nmi_callback);
-
-void unset_nmi_callback(void)
-{
-	nmi_callback = dummy_nmi_callback;
-}
-EXPORT_SYMBOL_GPL(unset_nmi_callback);
-
 #ifdef CONFIG_KPROBES
 fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
 {
diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index 8710ca081b1e..b3610188bcf0 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -17,14 +17,15 @@
 #include <asm/nmi.h>
 #include <asm/msr.h>
 #include <asm/apic.h>
+#include <asm/kdebug.h>
  
 #include "op_counter.h"
 #include "op_x86_model.h"
- 
+
 static struct op_x86_model_spec const * model;
 static struct op_msrs cpu_msrs[NR_CPUS];
 static unsigned long saved_lvtpc[NR_CPUS];
- 
+
 static int nmi_start(void);
 static void nmi_stop(void);
 
@@ -82,13 +83,24 @@ static void exit_driverfs(void)
 #define exit_driverfs() do { } while (0)
 #endif /* CONFIG_PM */
 
-
-static int nmi_callback(struct pt_regs * regs, int cpu)
+int profile_exceptions_notify(struct notifier_block *self,
+				unsigned long val, void *data)
 {
-	return model->check_ctrs(regs, &cpu_msrs[cpu]);
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+	int cpu = smp_processor_id();
+
+	switch(val) {
+	case DIE_NMI:
+		if (model->check_ctrs(args->regs, &cpu_msrs[cpu]))
+			ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+	return ret;
 }
- 
- 
+
 static void nmi_cpu_save_registers(struct op_msrs * msrs)
 {
 	unsigned int const nr_ctrs = model->num_counters;
@@ -174,27 +186,29 @@ static void nmi_cpu_setup(void * dummy)
 	apic_write(APIC_LVTPC, APIC_DM_NMI);
 }
 
+static struct notifier_block profile_exceptions_nb = {
+	.notifier_call = profile_exceptions_notify,
+	.next = NULL,
+	.priority = 0
+};
 
 static int nmi_setup(void)
 {
+	int err=0;
+
 	if (!allocate_msrs())
 		return -ENOMEM;
 
-	/* We walk a thin line between law and rape here.
-	 * We need to be careful to install our NMI handler
-	 * without actually triggering any NMIs as this will
-	 * break the core code horrifically.
-	 */
-	if (reserve_lapic_nmi() < 0) {
+	if ((err = register_die_notifier(&profile_exceptions_nb))){
 		free_msrs();
-		return -EBUSY;
+		return err;
 	}
+
 	/* We need to serialize save and setup for HT because the subset
 	 * of msrs are distinct for save and setup operations
 	 */
 	on_each_cpu(nmi_save_registers, NULL, 0, 1);
 	on_each_cpu(nmi_cpu_setup, NULL, 0, 1);
-	set_nmi_callback(nmi_callback);
 	nmi_enabled = 1;
 	return 0;
 }
@@ -250,8 +264,7 @@ static void nmi_shutdown(void)
 {
 	nmi_enabled = 0;
 	on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1);
-	unset_nmi_callback();
-	release_lapic_nmi();
+	unregister_die_notifier(&profile_exceptions_nb);
 	free_msrs();
 }
 
diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c
index a33a73bb502d..93ca48f804ac 100644
--- a/arch/i386/oprofile/nmi_timer_int.c
+++ b/arch/i386/oprofile/nmi_timer_int.c
@@ -17,32 +17,49 @@
 #include <asm/nmi.h>
 #include <asm/apic.h>
 #include <asm/ptrace.h>
+#include <asm/kdebug.h>
  
-static int nmi_timer_callback(struct pt_regs * regs, int cpu)
+int profile_timer_exceptions_notify(struct notifier_block *self,
+				unsigned long val, void *data)
 {
-	oprofile_add_sample(regs, 0);
-	return 1;
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	switch(val) {
+	case DIE_NMI:
+		oprofile_add_sample(args->regs, 0);
+		ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+	return ret;
 }
 
+static struct notifier_block profile_timer_exceptions_nb = {
+	.notifier_call = profile_timer_exceptions_notify,
+	.next = NULL,
+	.priority = 0
+};
+
 static int timer_start(void)
 {
-	disable_timer_nmi_watchdog();
-	set_nmi_callback(nmi_timer_callback);
+	if (register_die_notifier(&profile_timer_exceptions_nb))
+		return 1;
 	return 0;
 }
 
 
 static void timer_stop(void)
 {
-	enable_timer_nmi_watchdog();
-	unset_nmi_callback();
+	unregister_die_notifier(&profile_timer_exceptions_nb);
 	synchronize_sched();  /* Allow already-started NMIs to complete. */
 }
 
 
 int __init op_nmi_timer_init(struct oprofile_operations * ops)
 {
-	if (atomic_read(&nmi_active) <= 0)
+	if ((nmi_watchdog != NMI_IO_APIC) || (atomic_read(&nmi_active) <= 0))
 		return -ENODEV;
 
 	ops->start = timer_start;
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index d8d5750d6106..44c8af65325e 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -23,6 +23,7 @@
 #include <asm/nmi.h>
 #include <asm/hw_irq.h>
 #include <asm/mach_apic.h>
+#include <asm/kdebug.h>
 
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
@@ -95,8 +96,18 @@ static void crash_save_self(struct pt_regs *regs)
 #ifdef CONFIG_SMP
 static atomic_t waiting_for_crash_ipi;
 
-static int crash_nmi_callback(struct pt_regs *regs, int cpu)
+static int crash_nmi_callback(struct notifier_block *self,
+				unsigned long val, void *data)
 {
+	struct pt_regs *regs;
+	int cpu;
+
+	if (val != DIE_NMI)
+		return NOTIFY_OK;
+
+	regs = ((struct die_args *)data)->regs;
+	cpu = raw_smp_processor_id();
+
 	/*
 	 * Don't do anything if this handler is invoked on crashing cpu.
 	 * Otherwise, system will completely hang. Crashing cpu can get
@@ -127,12 +138,17 @@ static void smp_send_nmi_allbutself(void)
  * cpu hotplug shouldn't matter.
  */
 
+static struct notifier_block crash_nmi_nb = {
+	.notifier_call = crash_nmi_callback,
+};
+
 static void nmi_shootdown_cpus(void)
 {
 	unsigned long msecs;
 
 	atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1);
-	set_nmi_callback(crash_nmi_callback);
+	if (register_die_notifier(&crash_nmi_nb))
+		return;         /* return what? */
 
 	/*
 	 * Ensure the new callback function is set before sending
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index f6b881b23a70..9d175dcf3a2d 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -41,20 +41,6 @@ static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]);
  */
 #define NMI_MAX_COUNTER_BITS 66
 
-/*
- * lapic_nmi_owner tracks the ownership of the lapic NMI hardware:
- * - it may be reserved by some other driver, or not
- * - when not reserved by some other driver, it may be used for
- *   the NMI watchdog, or not
- *
- * This is maintained separately from nmi_active because the NMI
- * watchdog may also be driven from the I/O APIC timer.
- */
-static DEFINE_SPINLOCK(lapic_nmi_owner_lock);
-static unsigned int lapic_nmi_owner;
-#define LAPIC_NMI_WATCHDOG	(1<<0)
-#define LAPIC_NMI_RESERVED	(1<<1)
-
 /* nmi_active:
  * >0: the lapic NMI watchdog is active, but can be disabled
  * <0: the lapic NMI watchdog has not been set up, and cannot
@@ -321,33 +307,6 @@ static void enable_lapic_nmi_watchdog(void)
 	touch_nmi_watchdog();
 }
 
-int reserve_lapic_nmi(void)
-{
-	unsigned int old_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	old_owner = lapic_nmi_owner;
-	lapic_nmi_owner |= LAPIC_NMI_RESERVED;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (old_owner & LAPIC_NMI_RESERVED)
-		return -EBUSY;
-	if (old_owner & LAPIC_NMI_WATCHDOG)
-		disable_lapic_nmi_watchdog();
-	return 0;
-}
-
-void release_lapic_nmi(void)
-{
-	unsigned int new_owner;
-
-	spin_lock(&lapic_nmi_owner_lock);
-	new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED;
-	lapic_nmi_owner = new_owner;
-	spin_unlock(&lapic_nmi_owner_lock);
-	if (new_owner & LAPIC_NMI_WATCHDOG)
-		enable_lapic_nmi_watchdog();
-}
-
 void disable_timer_nmi_watchdog(void)
 {
 	BUG_ON(nmi_watchdog != NMI_IO_APIC);
@@ -762,13 +721,6 @@ done:
 	return rc;
 }
 
-static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu)
-{
-	return 0;
-}
- 
-static nmi_callback_t nmi_callback = dummy_nmi_callback;
- 
 asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
 	nmi_enter();
@@ -779,21 +731,12 @@ asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 
 int do_nmi_callback(struct pt_regs * regs, int cpu)
 {
-	return rcu_dereference(nmi_callback)(regs, cpu);
-}
-
-void set_nmi_callback(nmi_callback_t callback)
-{
-	vmalloc_sync_all();
-	rcu_assign_pointer(nmi_callback, callback);
-}
-EXPORT_SYMBOL_GPL(set_nmi_callback);
-
-void unset_nmi_callback(void)
-{
-	nmi_callback = dummy_nmi_callback;
+#ifdef CONFIG_SYSCTL
+	if (unknown_nmi_panic)
+		return unknown_nmi_panic_callback(regs, cpu);
+#endif
+	return 0;
 }
-EXPORT_SYMBOL_GPL(unset_nmi_callback);
 
 #ifdef CONFIG_SYSCTL
 
@@ -802,37 +745,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	unsigned char reason = get_nmi_reason();
 	char buf[64];
 
-	if (!(reason & 0xc0)) {
-		sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-		die_nmi(buf,regs);
-	}
-	return 0;
-}
-
-/*
- * proc handler for /proc/sys/kernel/unknown_nmi_panic
- */
-int proc_unknown_nmi_panic(struct ctl_table *table, int write, struct file *file,
-			void __user *buffer, size_t *length, loff_t *ppos)
-{
-	int old_state;
-
-	old_state = unknown_nmi_panic;
-	proc_dointvec(table, write, file, buffer, length, ppos);
-	if (!!old_state == !!unknown_nmi_panic)
-		return 0;
-
-	if (unknown_nmi_panic) {
-		if (reserve_lapic_nmi() < 0) {
-			unknown_nmi_panic = 0;
-			return -EBUSY;
-		} else {
-			set_nmi_callback(unknown_nmi_panic_callback);
-		}
-	} else {
-		release_lapic_nmi();
-		unset_nmi_callback();
-	}
+	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
+	die_nmi(buf,regs);
 	return 0;
 }
 
@@ -846,8 +760,6 @@ EXPORT_SYMBOL(reserve_perfctr_nmi);
 EXPORT_SYMBOL(release_perfctr_nmi);
 EXPORT_SYMBOL(reserve_evntsel_nmi);
 EXPORT_SYMBOL(release_evntsel_nmi);
-EXPORT_SYMBOL(reserve_lapic_nmi);
-EXPORT_SYMBOL(release_lapic_nmi);
 EXPORT_SYMBOL(disable_timer_nmi_watchdog);
 EXPORT_SYMBOL(enable_timer_nmi_watchdog);
 EXPORT_SYMBOL(touch_nmi_watchdog);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index da0e0b4e9139..34d6bf063b6e 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -6,24 +6,13 @@
 
 #include <linux/pm.h>
 
-struct pt_regs;
-
-typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
-
-/**
- * set_nmi_callback
- *
- * Set a handler for an NMI. Only one handler may be
- * set. Return 1 if the NMI was handled.
- */
-void set_nmi_callback(nmi_callback_t callback);
-
 /**
- * unset_nmi_callback
+ * do_nmi_callback
  *
- * Remove the handler previously set.
+ * Check to see if a callback exists and execute it.  Return 1
+ * if the handler exists and was handled successfully.
  */
-void unset_nmi_callback(void);
+int do_nmi_callback(struct pt_regs *regs, int cpu);
 
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
 extern int avail_to_resrv_perfctr_nmi(unsigned int);
@@ -33,8 +22,6 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
-extern int reserve_lapic_nmi(void);
-extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 8f02a2a416e6..8818c39d34e0 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -7,25 +7,6 @@
 #include <linux/pm.h>
 #include <asm/io.h>
  
-struct pt_regs;
-
-typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
-
-/**
- * set_nmi_callback
- *
- * Set a handler for an NMI. Only one handler may be
- * set. Return 1 if the NMI was handled.
- */
-void set_nmi_callback(nmi_callback_t callback);
-
-/**
- * unset_nmi_callback
- *
- * Remove the handler previously set.
- */
-void unset_nmi_callback(void);
-
 /**
  * do_nmi_callback
  *
@@ -72,8 +53,6 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
-extern int reserve_lapic_nmi(void);
-extern void release_lapic_nmi(void);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 362a0cc37138..83f168361624 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -76,8 +76,6 @@ extern int compat_log;
 
 #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
 int unknown_nmi_panic;
-extern int proc_unknown_nmi_panic(ctl_table *, int, struct file *,
-				  void __user *, size_t *, loff_t *);
 #endif
 
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
@@ -628,7 +626,7 @@ static ctl_table kern_table[] = {
 		.data           = &unknown_nmi_panic,
 		.maxlen         = sizeof (int),
 		.mode           = 0644,
-		.proc_handler   = &proc_unknown_nmi_panic,
+		.proc_handler   = &proc_dointvec,
 	},
 #endif
 #if defined(CONFIG_X86)
-- 
cgit v1.2.1


From 407984f1af259b31957c7c05075a454a751bb801 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: Add abilty to enable/disable nmi watchdog with sysctl

Adds a new /proc/sys/kernel/nmi call that will enable/disable the nmi
watchdog.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c   | 52 ++++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/nmi.c | 48 ++++++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/nmi.h   |  1 +
 include/asm-x86_64/nmi.h |  1 +
 include/linux/sysctl.h   |  1 +
 kernel/sysctl.c          | 11 ++++++++++
 6 files changed, 114 insertions(+)

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index acd3fdea2a21..28065d0b71a9 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -846,6 +846,58 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	return 0;
 }
 
+/*
+ * proc handler for /proc/sys/kernel/nmi_watchdog
+ */
+int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
+			void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int old_state;
+
+	nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0;
+	old_state = nmi_watchdog_enabled;
+	proc_dointvec(table, write, file, buffer, length, ppos);
+	if (!!old_state == !!nmi_watchdog_enabled)
+		return 0;
+
+	if (atomic_read(&nmi_active) < 0) {
+		printk(KERN_WARNING "NMI watchdog is permanently disabled\n");
+		return -EINVAL;
+	}
+
+	if (nmi_watchdog == NMI_DEFAULT) {
+		if (nmi_known_cpu() > 0)
+			nmi_watchdog = NMI_LOCAL_APIC;
+		else
+			nmi_watchdog = NMI_IO_APIC;
+	}
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+	{
+		if (nmi_watchdog_enabled)
+			enable_lapic_nmi_watchdog();
+		else
+			disable_lapic_nmi_watchdog();
+	} else if (nmi_watchdog == NMI_IO_APIC) {
+		/* FIXME
+		 * for some reason these functions don't work
+		 */
+		printk("Can not enable/disable NMI on IO APIC\n");
+		return -EINVAL;
+#if 0
+		if (nmi_watchdog_enabled)
+			enable_timer_nmi_watchdog();
+		else
+			disable_timer_nmi_watchdog();
+#endif
+	} else {
+		printk( KERN_WARNING
+			"NMI watchdog doesn't know what hardware to touch\n");
+		return -EIO;
+	}
+	return 0;
+}
+
 #endif
 
 EXPORT_SYMBOL(nmi_active);
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 9d175dcf3a2d..3a17411a9a19 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -750,6 +750,54 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	return 0;
 }
 
+/*
+ * proc handler for /proc/sys/kernel/nmi
+ */
+int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
+			void __user *buffer, size_t *length, loff_t *ppos)
+{
+	int old_state;
+
+	nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0;
+	old_state = nmi_watchdog_enabled;
+	proc_dointvec(table, write, file, buffer, length, ppos);
+	if (!!old_state == !!nmi_watchdog_enabled)
+		return 0;
+
+	if (atomic_read(&nmi_active) < 0) {
+		printk( KERN_WARNING "NMI watchdog is permanently disabled\n");
+		return -EINVAL;
+	}
+
+	/* if nmi_watchdog is not set yet, then set it */
+	nmi_watchdog_default();
+
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+	{
+		if (nmi_watchdog_enabled)
+			enable_lapic_nmi_watchdog();
+		else
+			disable_lapic_nmi_watchdog();
+	} else if (nmi_watchdog == NMI_IO_APIC) {
+		/* FIXME
+		 * for some reason these functions don't work
+		 */
+		printk("Can not enable/disable NMI on IO APIC\n");
+		return -EIO;
+#if 0
+		if (nmi_watchdog_enabled)
+			enable_timer_nmi_watchdog();
+		else
+			disable_timer_nmi_watchdog();
+#endif
+	} else {
+		printk(KERN_WARNING
+			"NMI watchdog doesn't know what hardware to touch\n");
+		return -EIO;
+	}
+	return 0;
+}
+
 #endif
 
 EXPORT_SYMBOL(nmi_active);
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 34d6bf063b6e..13b5d8311bf7 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -14,6 +14,7 @@
  */
 int do_nmi_callback(struct pt_regs *regs, int cpu);
 
+extern int nmi_watchdog_enabled;
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
 extern int avail_to_resrv_perfctr_nmi(unsigned int);
 extern int reserve_perfctr_nmi(unsigned int);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 8818c39d34e0..2c23b0df87d2 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -43,6 +43,7 @@ extern void die_nmi(char *str, struct pt_regs *regs);
 
 extern int panic_on_timeout;
 extern int unknown_nmi_panic;
+extern int nmi_watchdog_enabled;
 
 extern int check_nmi_watchdog(void);
 extern int avail_to_resrv_perfctr_nmi_bit(unsigned int);
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 736ed917a4f8..ecb79ba52ae1 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -150,6 +150,7 @@ enum
 	KERN_IA64_UNALIGNED=72, /* int: ia64 unaligned userland trap enable */
 	KERN_COMPAT_LOG=73,	/* int: print compat layer  messages */
 	KERN_MAX_LOCK_DEPTH=74,
+	KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */
 };
 
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 83f168361624..040de6bd74dd 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -76,6 +76,9 @@ extern int compat_log;
 
 #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86)
 int unknown_nmi_panic;
+int nmi_watchdog_enabled;
+extern int proc_nmi_enabled(struct ctl_table *, int , struct file *,
+			void __user *, size_t *, loff_t *);
 #endif
 
 /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */
@@ -628,6 +631,14 @@ static ctl_table kern_table[] = {
 		.mode           = 0644,
 		.proc_handler   = &proc_dointvec,
 	},
+	{
+		.ctl_name       = KERN_NMI_WATCHDOG,
+		.procname       = "nmi_watchdog",
+		.data           = &nmi_watchdog_enabled,
+		.maxlen         = sizeof (int),
+		.mode           = 0644,
+		.proc_handler   = &proc_nmi_enabled,
+	},
 #endif
 #if defined(CONFIG_X86)
 	{
-- 
cgit v1.2.1


From e33e89ab1a8d295de0500b697f4f31c3ceee9aa2 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: Add abilty to enable/disable nmi watchdog from procfs
 (update)

Adds a new /proc/sys/kernel/nmi_watchdog call that will enable/disable the
nmi watchdog.

By entering a non-zero value here, a user can enable the nmi watchdog to
monitor the online cpus in the system.  By entering a zero value here, a
user can disable the nmi watchdog and free up a performance counter which
could then be utilized by the oprofile subsystem, otherwise oprofile may be
short a counter when in use.

Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 Documentation/filesystems/proc.txt | 14 +++++++++-----
 arch/i386/kernel/nmi.c             | 21 ++++-----------------
 arch/x86_64/kernel/nmi.c           | 21 ++++-----------------
 3 files changed, 17 insertions(+), 39 deletions(-)

diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 99902ae6804e..7db71d6fba82 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -1124,11 +1124,15 @@ debugging information is displayed on console.
 NMI switch that most IA32 servers have fires unknown NMI up, for example.
 If a system hangs up, try pressing the NMI switch.
 
-[NOTE]
-   This function and oprofile share a NMI callback. Therefore this function
-   cannot be enabled when oprofile is activated.
-   And NMI watchdog will be disabled when the value in this file is set to
-   non-zero.
+nmi_watchdog
+------------
+
+Enables/Disables the NMI watchdog on x86 systems.  When the value is non-zero
+the NMI watchdog is enabled and will continuously test all online cpus to
+determine whether or not they are still functioning properly.
+
+Because the NMI watchdog shares registers with oprofile, by disabling the NMI
+watchdog, oprofile may have more registers to utilize.
 
 
 2.4 /proc/sys/vm - The virtual memory subsystem
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 28065d0b71a9..6241e4448cab 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -847,7 +847,7 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 }
 
 /*
- * proc handler for /proc/sys/kernel/nmi_watchdog
+ * proc handler for /proc/sys/kernel/nmi
  */
 int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
 			void __user *buffer, size_t *length, loff_t *ppos)
@@ -861,8 +861,8 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
 		return 0;
 
 	if (atomic_read(&nmi_active) < 0) {
-		printk(KERN_WARNING "NMI watchdog is permanently disabled\n");
-		return -EINVAL;
+		printk( KERN_WARNING "NMI watchdog is permanently disabled\n");
+		return -EIO;
 	}
 
 	if (nmi_watchdog == NMI_DEFAULT) {
@@ -872,24 +872,11 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
 			nmi_watchdog = NMI_IO_APIC;
 	}
 
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-	{
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		if (nmi_watchdog_enabled)
 			enable_lapic_nmi_watchdog();
 		else
 			disable_lapic_nmi_watchdog();
-	} else if (nmi_watchdog == NMI_IO_APIC) {
-		/* FIXME
-		 * for some reason these functions don't work
-		 */
-		printk("Can not enable/disable NMI on IO APIC\n");
-		return -EINVAL;
-#if 0
-		if (nmi_watchdog_enabled)
-			enable_timer_nmi_watchdog();
-		else
-			disable_timer_nmi_watchdog();
-#endif
 	} else {
 		printk( KERN_WARNING
 			"NMI watchdog doesn't know what hardware to touch\n");
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 3a17411a9a19..dd57410dad51 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -167,7 +167,7 @@ static __cpuinit inline int nmi_known_cpu(void)
 }
 
 /* Run after command line and cpu_init init, but before all other checks */
-void __cpuinit nmi_watchdog_default(void)
+void nmi_watchdog_default(void)
 {
 	if (nmi_watchdog != NMI_DEFAULT)
 		return;
@@ -766,32 +766,19 @@ int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file,
 
 	if (atomic_read(&nmi_active) < 0) {
 		printk( KERN_WARNING "NMI watchdog is permanently disabled\n");
-		return -EINVAL;
+		return -EIO;
 	}
 
 	/* if nmi_watchdog is not set yet, then set it */
 	nmi_watchdog_default();
 
-	if (nmi_watchdog == NMI_LOCAL_APIC)
-	{
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		if (nmi_watchdog_enabled)
 			enable_lapic_nmi_watchdog();
 		else
 			disable_lapic_nmi_watchdog();
-	} else if (nmi_watchdog == NMI_IO_APIC) {
-		/* FIXME
-		 * for some reason these functions don't work
-		 */
-		printk("Can not enable/disable NMI on IO APIC\n");
-		return -EIO;
-#if 0
-		if (nmi_watchdog_enabled)
-			enable_timer_nmi_watchdog();
-		else
-			disable_timer_nmi_watchdog();
-#endif
 	} else {
-		printk(KERN_WARNING
+		printk( KERN_WARNING
 			"NMI watchdog doesn't know what hardware to touch\n");
 		return -EIO;
 	}
-- 
cgit v1.2.1


From 8da5adda91df3d2fcc5300e68da491694c9af019 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: Allow users to force a panic on NMI

To quote Alan Cox:

The default Linux behaviour on an NMI of either memory or unknown is to
continue operation. For many environments such as scientific computing
it is preferable that the box is taken out and the error dealt with than
an uncorrected parity/ECC error get propogated.

A small number of systems do generate NMI's for bizarre random reasons
such as power management so the default is unchanged. In other respects
the new proc/sys entry works like the existing panic controls already in
that directory.

This is separate to the edac support - EDAC allows supported chipsets to
handle ECC errors well, this change allows unsupported cases to at least
panic rather than cause problems further down the line.

Signed-off-by: Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c   | 6 ++++++
 arch/x86_64/kernel/traps.c | 6 ++++++
 include/linux/kernel.h     | 1 +
 include/linux/sysctl.h     | 1 +
 kernel/panic.c             | 1 +
 kernel/sysctl.c            | 8 ++++++++
 6 files changed, 23 insertions(+)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7db664d0b25c..2f6cb8276480 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -635,6 +635,8 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 			"to continue\n");
 	printk(KERN_EMERG "You probably have a hardware problem with your RAM "
 			"chips\n");
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
 
 	/* Clear and disable the memory parity error line. */
 	clear_mem_error(reason);
@@ -670,6 +672,10 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 		reason, smp_processor_id());
 	printk("Dazed and confused, but trying to continue\n");
 	printk("Do you have a strange power saving mode enabled?\n");
+
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
+
 }
 
 static DEFINE_SPINLOCK(nmi_print_lock);
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 42bc070fdf11..b18829db2a6a 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -732,6 +732,8 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
 	printk("You probably have a hardware problem with your RAM chips\n");
+	if (panic_on_unrecovered_nmi)
+               panic("NMI: Not continuing");
 
 	/* Clear and disable the memory parity error line. */
 	reason = (reason & 0xf) | 4;
@@ -757,6 +759,10 @@ unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {	printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
 	printk("Dazed and confused, but trying to continue\n");
 	printk("Do you have a strange power saving mode enabled?\n");
+
+	if (panic_on_unrecovered_nmi)
+                panic("NMI: Not continuing");
+
 }
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b2ae4fdce8b..1ff9609300b4 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -186,6 +186,7 @@ extern void bust_spinlocks(int yes);
 extern int oops_in_progress;		/* If set, an oops, panic(), BUG() or die() is in progress */
 extern int panic_timeout;
 extern int panic_on_oops;
+extern int panic_on_unrecovered_nmi;
 extern int tainted;
 extern const char *print_tainted(void);
 extern void add_taint(unsigned);
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index ecb79ba52ae1..432778446ad2 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -151,6 +151,7 @@ enum
 	KERN_COMPAT_LOG=73,	/* int: print compat layer  messages */
 	KERN_MAX_LOCK_DEPTH=74,
 	KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */
+	KERN_PANIC_ON_NMI=76, /* int: whether we will panic on an unrecovered */
 };
 
 
diff --git a/kernel/panic.c b/kernel/panic.c
index 8010b9b17aca..d2db3e2209e0 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -21,6 +21,7 @@
 #include <linux/debug_locks.h>
 
 int panic_on_oops;
+int panic_on_unrecovered_nmi;
 int tainted;
 static int pause_on_oops;
 static int pause_on_oops_flag;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 040de6bd74dd..220e20564124 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -641,6 +641,14 @@ static ctl_table kern_table[] = {
 	},
 #endif
 #if defined(CONFIG_X86)
+	{
+		.ctl_name	= KERN_PANIC_ON_NMI,
+		.procname	= "panic_on_unrecovered_nmi",
+		.data		= &panic_on_unrecovered_nmi,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 	{
 		.ctl_name	= KERN_BOOTLOADER_TYPE,
 		.procname	= "bootloader_type",
-- 
cgit v1.2.1


From c41c5cd3b20a2d81c30498f13b1527847a8fdf69 Mon Sep 17 00:00:00 2001
From: Don Zickus <dzickus@redhat.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: x86 clean up nmi panic messages

Clean up some of the output messages on the nmi error paths to make more
sense when they are displayed.  This is mainly a cosmetic fix and
shouldn't impact any normal code path.

Signed-off-by:  Don Zickus <dzickus@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c   | 15 ++++++++-------
 arch/x86_64/kernel/traps.c | 21 ++++++++++++++-------
 2 files changed, 22 insertions(+), 14 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 2f6cb8276480..3c85c89f68d8 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -631,13 +631,15 @@ gp_in_kernel:
 
 static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
-	printk(KERN_EMERG "Uhhuh. NMI received. Dazed and confused, but trying "
-			"to continue\n");
+	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on "
+		"CPU %d.\n", reason, smp_processor_id());
 	printk(KERN_EMERG "You probably have a hardware problem with your RAM "
 			"chips\n");
 	if (panic_on_unrecovered_nmi)
                 panic("NMI: Not continuing");
 
+	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
+
 	/* Clear and disable the memory parity error line. */
 	clear_mem_error(reason);
 }
@@ -668,14 +670,13 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 		return;
 	}
 #endif
-	printk("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
-		reason, smp_processor_id());
-	printk("Dazed and confused, but trying to continue\n");
-	printk("Do you have a strange power saving mode enabled?\n");
-
+	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on "
+		"CPU %d.\n", reason, smp_processor_id());
+	printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n");
 	if (panic_on_unrecovered_nmi)
                 panic("NMI: Not continuing");
 
+	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
 }
 
 static DEFINE_SPINLOCK(nmi_print_lock);
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index b18829db2a6a..dae10df60926 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -730,10 +730,15 @@ asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
 static __kprobes void
 mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
-	printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
-	printk("You probably have a hardware problem with your RAM chips\n");
+	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
+		reason);
+	printk(KERN_EMERG "You probably have a hardware problem with your "
+		"RAM chips\n");
+
 	if (panic_on_unrecovered_nmi)
-               panic("NMI: Not continuing");
+		panic("NMI: Not continuing");
+
+	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
 
 	/* Clear and disable the memory parity error line. */
 	reason = (reason & 0xf) | 4;
@@ -756,13 +761,15 @@ io_check_error(unsigned char reason, struct pt_regs * regs)
 
 static __kprobes void
 unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
-{	printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
-	printk("Dazed and confused, but trying to continue\n");
-	printk("Do you have a strange power saving mode enabled?\n");
+{
+	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x.\n",
+		reason);
+	printk(KERN_EMERG "Do you have a strange power saving mode enabled?\n");
 
 	if (panic_on_unrecovered_nmi)
-                panic("NMI: Not continuing");
+		panic("NMI: Not continuing");
 
+	printk(KERN_EMERG "Dazed and confused, but trying to continue\n");
 }
 
 /* Runs on IST stack. This code must keep interrupts off all the time.
-- 
cgit v1.2.1


From 4038f901cf102a40715b900984ed7540a9fa637f Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386/x86-64: Fix NMI watchdog suspend/resume

Making NMI suspend/resume work with SMP. We use CPU hotplug to offline
APs in SMP suspend/resume. Only BSP executes sysdev's .suspend/.resume
method. APs should follow CPU hotplug code path.

And:

+From: Don Zickus <dzickus@redhat.com>

Makes the start/stop paths of nmi watchdog more robust to handle the
suspend/resume cases more gracefully.

AK: I merged the two patches together

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/nmi.c       | 33 ++++++++++++++++++++++++++-------
 arch/i386/kernel/smpboot.c   |  3 ++-
 arch/x86_64/kernel/nmi.c     | 33 ++++++++++++++++++++++++++-------
 arch/x86_64/kernel/smpboot.c |  2 ++
 include/asm-i386/nmi.h       |  1 +
 include/asm-x86_64/nmi.h     |  1 +
 6 files changed, 58 insertions(+), 15 deletions(-)

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 6241e4448cab..8e4ed930ce6b 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk {
 static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
 /* local prototypes */
-static void stop_apic_nmi_watchdog(void *unused);
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 extern void show_registers(struct pt_regs *regs);
@@ -341,15 +340,20 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
+	/* only CPU0 goes here, other CPUs should be offline */
 	nmi_pm_active = atomic_read(&nmi_active);
-	disable_lapic_nmi_watchdog();
+	stop_apic_nmi_watchdog(NULL);
+	BUG_ON(atomic_read(&nmi_active) != 0);
 	return 0;
 }
 
 static int lapic_nmi_resume(struct sys_device *dev)
 {
-	if (nmi_pm_active > 0)
-		enable_lapic_nmi_watchdog();
+	/* only CPU0 goes here, other CPUs should be offline */
+	if (nmi_pm_active > 0) {
+		setup_apic_nmi_watchdog(NULL);
+		touch_nmi_watchdog();
+	}
 	return 0;
 }
 
@@ -626,11 +630,21 @@ static void stop_p4_watchdog(void)
 
 void setup_apic_nmi_watchdog (void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 1)
+		return;
+
+	/* cheap hack to support suspend/resume */
+	/* if cpu0 is not active neither should the other cpus */
+	if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0))
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -663,17 +677,22 @@ void setup_apic_nmi_watchdog (void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	wd->enabled = 1;
 	atomic_inc(&nmi_active);
 }
 
-static void stop_apic_nmi_watchdog(void *unused)
+void stop_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 0)
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -697,7 +716,7 @@ static void stop_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	wd->enabled = 0;
 	atomic_dec(&nmi_active);
 }
 
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index efe07990e7fc..9367af76ce37 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -1376,7 +1376,8 @@ int __cpu_disable(void)
 	 */
 	if (cpu == 0)
 		return -EBUSY;
-
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
 	clear_local_APIC();
 	/* Allow any queued timer interrupts to get serviced */
 	local_irq_enable();
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index dd57410dad51..5a35975e5763 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -63,7 +63,6 @@ struct nmi_watchdog_ctlblk {
 static DEFINE_PER_CPU(struct nmi_watchdog_ctlblk, nmi_watchdog_ctlblk);
 
 /* local prototypes */
-static void stop_apic_nmi_watchdog(void *unused);
 static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu);
 
 /* converts an msr to an appropriate reservation bit */
@@ -337,15 +336,20 @@ static int nmi_pm_active; /* nmi_active before suspend */
 
 static int lapic_nmi_suspend(struct sys_device *dev, pm_message_t state)
 {
+	/* only CPU0 goes here, other CPUs should be offline */
 	nmi_pm_active = atomic_read(&nmi_active);
-	disable_lapic_nmi_watchdog();
+	stop_apic_nmi_watchdog(NULL);
+	BUG_ON(atomic_read(&nmi_active) != 0);
 	return 0;
 }
 
 static int lapic_nmi_resume(struct sys_device *dev)
 {
-	if (nmi_pm_active > 0)
-		enable_lapic_nmi_watchdog();
+	/* only CPU0 goes here, other CPUs should be offline */
+	if (nmi_pm_active > 0) {
+		setup_apic_nmi_watchdog(NULL);
+		touch_nmi_watchdog();
+	}
 	return 0;
 }
 
@@ -561,11 +565,21 @@ static void stop_p4_watchdog(void)
 
 void setup_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 1)
+		return;
+
+	/* cheap hack to support suspend/resume */
+	/* if cpu0 is not active neither should the other cpus */
+	if ((smp_processor_id() != 0) && (atomic_read(&nmi_active) <= 0))
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -582,17 +596,22 @@ void setup_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 1;
+	wd->enabled = 1;
 	atomic_inc(&nmi_active);
 }
 
-static void stop_apic_nmi_watchdog(void *unused)
+void stop_apic_nmi_watchdog(void *unused)
 {
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 	/* only support LOCAL and IO APICs for now */
 	if ((nmi_watchdog != NMI_LOCAL_APIC) &&
 	    (nmi_watchdog != NMI_IO_APIC))
 	    	return;
 
+	if (wd->enabled == 0)
+		return;
+
 	if (nmi_watchdog == NMI_LOCAL_APIC) {
 		switch (boot_cpu_data.x86_vendor) {
 		case X86_VENDOR_AMD:
@@ -607,7 +626,7 @@ static void stop_apic_nmi_watchdog(void *unused)
 			return;
 		}
 	}
-	__get_cpu_var(nmi_watchdog_ctlblk.enabled) = 0;
+	wd->enabled = 0;
 	atomic_dec(&nmi_active);
 }
 
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 975380207b46..15555879ce95 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1233,6 +1233,8 @@ int __cpu_disable(void)
 	if (cpu == 0)
 		return -EBUSY;
 
+	if (nmi_watchdog == NMI_LOCAL_APIC)
+		stop_apic_nmi_watchdog(NULL);
 	clear_local_APIC();
 
 	/*
diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h
index 13b5d8311bf7..303bcd4592bb 100644
--- a/include/asm-i386/nmi.h
+++ b/include/asm-i386/nmi.h
@@ -23,6 +23,7 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
+extern void stop_apic_nmi_watchdog (void *);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 2c23b0df87d2..578596494275 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -54,6 +54,7 @@ extern int reserve_evntsel_nmi(unsigned int);
 extern void release_evntsel_nmi(unsigned int);
 
 extern void setup_apic_nmi_watchdog (void *);
+extern void stop_apic_nmi_watchdog (void *);
 extern void disable_timer_nmi_watchdog(void);
 extern void enable_timer_nmi_watchdog(void);
 extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason);
-- 
cgit v1.2.1


From fac58550e80c307bf17cfa0dd544fca4eff120a5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] Fix up panic messages for different NMI panics

When a unknown NMI happened the panic would claim a NMI watchdog timeout.
Also it would check the variable set by nmi_watchdog=panic and panic then.

Fix up the panic message to be generic
Unconditionally panic on unknown NMI when panic on unknown nmi is enabled.

Noticed by Jan Beulich

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/nmi.c   | 5 +++--
 arch/x86_64/kernel/traps.c | 7 +++----
 include/asm-x86_64/nmi.h   | 2 +-
 3 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5a35975e5763..1b76d1574529 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -695,7 +695,8 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 		 */
 		local_inc(&__get_cpu_var(alert_counter));
 		if (local_read(&__get_cpu_var(alert_counter)) == 5*nmi_hz)
-			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs);
+			die_nmi("NMI Watchdog detected LOCKUP on CPU %d\n", regs,
+				panic_on_timeout);
 	} else {
 		__get_cpu_var(last_irq_sum) = sum;
 		local_set(&__get_cpu_var(alert_counter), 0);
@@ -765,7 +766,7 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu)
 	char buf[64];
 
 	sprintf(buf, "NMI received for unknown reason %02x\n", reason);
-	die_nmi(buf,regs);
+	die_nmi(buf, regs, 1);	/* Always panic here */
 	return 0;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index dae10df60926..96f62a033242 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -570,7 +570,7 @@ void die(const char * str, struct pt_regs * regs, long err)
 	do_exit(SIGSEGV); 
 }
 
-void __kprobes die_nmi(char *str, struct pt_regs *regs)
+void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 {
 	unsigned long flags = oops_begin();
 
@@ -582,9 +582,8 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs)
 	show_registers(regs);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
-	if (panic_on_timeout || panic_on_oops)
-		panic("nmi watchdog");
-	printk("console shuts up ...\n");
+	if (do_panic || panic_on_oops)
+		panic("Non maskable interrupt");
 	oops_end(flags);
 	nmi_exit();
 	local_irq_enable();
diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h
index 578596494275..cbf2669bca71 100644
--- a/include/asm-x86_64/nmi.h
+++ b/include/asm-x86_64/nmi.h
@@ -37,7 +37,7 @@ static inline void unset_nmi_pm_callback(struct pm_dev * dev)
 #endif /* CONFIG_PM */
  
 extern void default_do_nmi(struct pt_regs *);
-extern void die_nmi(char *str, struct pt_regs *regs);
+extern void die_nmi(char *str, struct pt_regs *regs, int do_panic);
 
 #define get_nmi_reason() inb(0x61)
 
-- 
cgit v1.2.1


From c7c19f8e5e564fb1a354a065befc8a88a9bb08fd Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386: make functions static

This patch makes the following needlessly global functions static:
- nmi_int.c: profile_exceptions_notify()
- nmi_timer_int.c: profile_timer_exceptions_notify()

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/oprofile/nmi_int.c       | 4 ++--
 arch/i386/oprofile/nmi_timer_int.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index b3610188bcf0..3700eef78743 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -83,8 +83,8 @@ static void exit_driverfs(void)
 #define exit_driverfs() do { } while (0)
 #endif /* CONFIG_PM */
 
-int profile_exceptions_notify(struct notifier_block *self,
-				unsigned long val, void *data)
+static int profile_exceptions_notify(struct notifier_block *self,
+				     unsigned long val, void *data)
 {
 	struct die_args *args = (struct die_args *)data;
 	int ret = NOTIFY_DONE;
diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c
index 93ca48f804ac..abf0ba52a635 100644
--- a/arch/i386/oprofile/nmi_timer_int.c
+++ b/arch/i386/oprofile/nmi_timer_int.c
@@ -19,8 +19,8 @@
 #include <asm/ptrace.h>
 #include <asm/kdebug.h>
  
-int profile_timer_exceptions_notify(struct notifier_block *self,
-				unsigned long val, void *data)
+static int profile_timer_exceptions_notify(struct notifier_block *self,
+					   unsigned long val, void *data)
 {
 	struct die_args *args = (struct die_args *)data;
 	int ret = NOTIFY_DONE;
-- 
cgit v1.2.1


From 3f22c5789eb76fd9aabe4be37ba609c793f046f9 Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] kdump x86_64 nmi event notification fix

After a crash we should wait for NMI IPI event and not for external NMI or
NMI watchdog tick.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/x86_64/kernel/crash.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index 44c8af65325e..fc57a139123e 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -102,7 +102,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 	struct pt_regs *regs;
 	int cpu;
 
-	if (val != DIE_NMI)
+	if (val != DIE_NMI_IPI)
 		return NOTIFY_OK;
 
 	regs = ((struct die_args *)data)->regs;
@@ -114,7 +114,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 	 * an NMI if system was initially booted with nmi_watchdog parameter.
 	 */
 	if (cpu == crashing_cpu)
-		return 1;
+		return NOTIFY_STOP;
 	local_irq_disable();
 
 	crash_save_this_cpu(regs, cpu);
-- 
cgit v1.2.1


From 260d6790b6a2a0a048b7f96d154c2b49f1e6515a Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386: Kdump i386 nmi event notification fix

After a crash we should wait for NMI IPI event and not for external NMI or
NMI watchdog tick.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/crash.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index 736c76d6b31d..67d297dc1003 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -102,7 +102,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 	struct pt_regs fixed_regs;
 	int cpu;
 
-	if (val != DIE_NMI)
+	if (val != DIE_NMI_IPI)
 		return NOTIFY_OK;
 
 	regs = ((struct die_args *)data)->regs;
@@ -113,7 +113,7 @@ static int crash_nmi_callback(struct notifier_block *self,
 	 * an NMI if system was initially booted with nmi_watchdog parameter.
 	 */
 	if (cpu == crashing_cpu)
-		return 1;
+		return NOTIFY_STOP;
 	local_irq_disable();
 
 	if (!user_mode_vm(regs)) {
-- 
cgit v1.2.1


From 1de84979dfc527c422abf63f27beabe43892989b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] i386: Enable NMI watchdog by default

I've had good experiences with having this on by default on x86-64.
It turns nasty hangs into easier to debug oopses.

Enable the local APIC wdog by default for systems newer than 2004.

This comes from a strange compromise: according to arjan the reason
it was off by default was some old IBM systems that corrupted
registered when NMI happened in SMI. Can't remember more specific,
but >= 2004 should avoid these. It's probably overly broad
because most older systems should be ok (and the really old systems
won't be supported by the local apic watchdog anyways)

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 8e4ed930ce6b..6e5085d5d2f6 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -21,6 +21,7 @@
 #include <linux/sysdev.h>
 #include <linux/sysctl.h>
 #include <linux/percpu.h>
+#include <linux/dmi.h>
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
@@ -204,6 +205,14 @@ static int __init check_nmi_watchdog(void)
 	unsigned int *prev_nmi_count;
 	int cpu;
 
+	/* Enable NMI watchdog for newer systems.
+           Actually it should be safe for most systems before 2004 too except
+	   for some IBM systems that corrupt registers when NMI happens
+	   during SMM. Unfortunately we don't have more exact information
+ 	   on these and use this coarse check. */
+	if (nmi_watchdog == NMI_DEFAULT && dmi_get_year(DMI_BIOS_DATE) >= 2004)
+		nmi_watchdog = NMI_LOCAL_APIC;
+
 	if ((nmi_watchdog == NMI_NONE) || (nmi_watchdog == NMI_DEFAULT))
 		return 0;
 
-- 
cgit v1.2.1


From 248dcb2ffffe8f3e4a369556a68988788c208111 Mon Sep 17 00:00:00 2001
From: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Date: Tue, 26 Sep 2006 10:52:27 +0200
Subject: [PATCH] x86: i386/x86-64 Add nmi watchdog support for new Intel CPUs

AK: This redoes the changes I temporarily reverted.

Intel now has support for Architectural Performance Monitoring Counters
( Refer to IA-32 Intel Architecture Software Developer's Manual
http://www.intel.com/design/pentium4/manuals/253669.htm ). This
feature is present starting from Intel Core Duo and Intel Core Solo processors.

What this means is, the performance monitoring counters and some performance
monitoring events are now defined in an architectural way (using cpuid).
And there will be no need to check for family/model etc for these architectural
events.

Below is the patch to use this performance counters in nmi watchdog driver.
Patch handles both i386 and x86-64 kernels.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/nmi.c                  | 126 +++++++++++++++++++++++++++++--
 arch/x86_64/kernel/nmi.c                | 130 ++++++++++++++++++++++++++++++--
 include/asm-i386/intel_arch_perfmon.h   |  31 ++++++++
 include/asm-x86_64/intel_arch_perfmon.h |  31 ++++++++
 4 files changed, 308 insertions(+), 10 deletions(-)
 create mode 100644 include/asm-i386/intel_arch_perfmon.h
 create mode 100644 include/asm-x86_64/intel_arch_perfmon.h

diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 6e5085d5d2f6..7b9a053effa3 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -26,6 +26,7 @@
 #include <asm/smp.h>
 #include <asm/nmi.h>
 #include <asm/kdebug.h>
+#include <asm/intel_arch_perfmon.h>
 
 #include "mach_traps.h"
 
@@ -77,6 +78,9 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_PERFCTR0);
 	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return (msr - MSR_P6_PERFCTR0);
@@ -95,6 +99,9 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_EVNTSEL0);
 	case X86_VENDOR_INTEL:
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+
 		switch (boot_cpu_data.x86) {
 		case 6:
 			return (msr - MSR_P6_EVNTSEL0);
@@ -174,7 +181,10 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
 	case X86_VENDOR_INTEL:
-		return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return 1;
+		else
+			return ((boot_cpu_data.x86 == 15) || (boot_cpu_data.x86 == 6));
 	}
 	return 0;
 }
@@ -261,8 +271,24 @@ static int __init check_nmi_watchdog(void)
 
 	/* now that we know it works we can reduce NMI frequency to
 	   something more reasonable; makes a difference in some configs */
-	if (nmi_watchdog == NMI_LOCAL_APIC)
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 		nmi_hz = 1;
+		/*
+		 * On Intel CPUs with ARCH_PERFMON only 32 bits in the counter
+		 * are writable, with higher bits sign extending from bit 31.
+		 * So, we can only program the counter with 31 bit values and
+		 * 32nd bit should be 1, for 33.. to be 1.
+		 * Find the appropriate nmi_hz
+		 */
+	 	if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0 &&
+			((u64)cpu_khz * 1000) > 0x7fffffffULL) {
+			u64 count = (u64)cpu_khz * 1000;
+			do_div(count, 0x7fffffffUL);
+			nmi_hz = count + 1;
+		}
+	}
 
 	kfree(prev_nmi_count);
 	return 0;
@@ -637,6 +663,85 @@ static void stop_p4_watchdog(void)
 	release_perfctr_nmi(wd->perfctr_msr);
 }
 
+#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
+#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
+
+static int setup_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		goto fail;
+
+	perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
+	evntsel_msr = MSR_ARCH_PERFMON_EVENTSEL0;
+
+	if (!reserve_perfctr_nmi(perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(evntsel_msr))
+		goto fail1;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = ARCH_PERFMON_EVENTSEL_INT
+		| ARCH_PERFMON_EVENTSEL_OS
+		| ARCH_PERFMON_EVENTSEL_USR
+		| ARCH_PERFMON_NMI_EVENT_SEL
+		| ARCH_PERFMON_NMI_EVENT_UMASK;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	write_watchdog_counter(perfctr_msr, "INTEL_ARCH_PERFCTR0");
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL << (eax.split.bit_width - 1);
+	return 1;
+fail1:
+	release_perfctr_nmi(perfctr_msr);
+fail:
+	return 0;
+}
+
+static void stop_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		return;
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
 void setup_apic_nmi_watchdog (void *unused)
 {
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
@@ -663,6 +768,11 @@ void setup_apic_nmi_watchdog (void *unused)
 				return;
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				if (!setup_intel_arch_watchdog())
+					return;
+				break;
+			}
 			switch (boot_cpu_data.x86) {
 			case 6:
 				if (boot_cpu_data.x86_model > 0xd)
@@ -708,6 +818,10 @@ void stop_apic_nmi_watchdog(void *unused)
 			stop_k7_watchdog();
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				stop_intel_arch_watchdog();
+				break;
+			}
 			switch (boot_cpu_data.x86) {
 			case 6:
 				if (boot_cpu_data.x86_model > 0xd)
@@ -831,10 +945,12 @@ int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 	 			wrmsrl(wd->cccr_msr, dummy);
 	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
 	 		}
-			else if (wd->perfctr_msr == MSR_P6_PERFCTR0) {
-				/* Only P6 based Pentium M need to re-unmask
+			else if (wd->perfctr_msr == MSR_P6_PERFCTR0 ||
+				 wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+				/* P6 based Pentium M need to re-unmask
 				 * the apic vector but it doesn't hurt
-				 * other P6 variant */
+				 * other P6 variant.
+				 * ArchPerfom/Core Duo also needs this */
 				apic_write(APIC_LVTPC, APIC_DM_NMI);
 			}
 			/* start the cycle over again */
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 1b76d1574529..4d6fb047952e 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -26,6 +26,7 @@
 #include <asm/proto.h>
 #include <asm/kdebug.h>
 #include <asm/mce.h>
+#include <asm/intel_arch_perfmon.h>
 
 /* perfctr_nmi_owner tracks the ownership of the perfctr registers:
  * evtsel_nmi_owner tracks the ownership of the event selection
@@ -73,7 +74,10 @@ static inline unsigned int nmi_perfctr_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_PERFCTR0);
 	case X86_VENDOR_INTEL:
-		return (msr - MSR_P4_BPU_PERFCTR0);
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_PERFCTR0);
+		else
+			return (msr - MSR_P4_BPU_PERFCTR0);
 	}
 	return 0;
 }
@@ -86,7 +90,10 @@ static inline unsigned int nmi_evntsel_msr_to_bit(unsigned int msr)
 	case X86_VENDOR_AMD:
 		return (msr - MSR_K7_EVNTSEL0);
 	case X86_VENDOR_INTEL:
-		return (msr - MSR_P4_BSU_ESCR0);
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return (msr - MSR_ARCH_PERFMON_EVENTSEL0);
+		else
+			return (msr - MSR_P4_BSU_ESCR0);
 	}
 	return 0;
 }
@@ -160,7 +167,10 @@ static __cpuinit inline int nmi_known_cpu(void)
 	case X86_VENDOR_AMD:
 		return boot_cpu_data.x86 == 15;
 	case X86_VENDOR_INTEL:
-		return boot_cpu_data.x86 == 15;
+		if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON))
+			return 1;
+		else
+			return (boot_cpu_data.x86 == 15);
 	}
 	return 0;
 }
@@ -246,8 +256,22 @@ int __init check_nmi_watchdog (void)
 
 	/* now that we know it works we can reduce NMI frequency to
 	   something more reasonable; makes a difference in some configs */
-	if (nmi_watchdog == NMI_LOCAL_APIC)
+	if (nmi_watchdog == NMI_LOCAL_APIC) {
+		struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
 		nmi_hz = 1;
+		/*
+		 * On Intel CPUs with ARCH_PERFMON only 32 bits in the counter
+		 * are writable, with higher bits sign extending from bit 31.
+		 * So, we can only program the counter with 31 bit values and
+		 * 32nd bit should be 1, for 33.. to be 1.
+		 * Find the appropriate nmi_hz
+		 */
+	 	if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0 &&
+			((u64)cpu_khz * 1000) > 0x7fffffffULL) {
+			nmi_hz = ((u64)cpu_khz * 1000) / 0x7fffffffUL + 1;
+		}
+	}
 
 	kfree(counts);
 	return 0;
@@ -563,6 +587,87 @@ static void stop_p4_watchdog(void)
 	release_perfctr_nmi(wd->perfctr_msr);
 }
 
+#define ARCH_PERFMON_NMI_EVENT_SEL	ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL
+#define ARCH_PERFMON_NMI_EVENT_UMASK	ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK
+
+static int setup_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	unsigned int perfctr_msr, evntsel_msr;
+	unsigned int evntsel;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		goto fail;
+
+	perfctr_msr = MSR_ARCH_PERFMON_PERFCTR0;
+	evntsel_msr = MSR_ARCH_PERFMON_EVENTSEL0;
+
+	if (!reserve_perfctr_nmi(perfctr_msr))
+		goto fail;
+
+	if (!reserve_evntsel_nmi(evntsel_msr))
+		goto fail1;
+
+	wrmsrl(perfctr_msr, 0UL);
+
+	evntsel = ARCH_PERFMON_EVENTSEL_INT
+		| ARCH_PERFMON_EVENTSEL_OS
+		| ARCH_PERFMON_EVENTSEL_USR
+		| ARCH_PERFMON_NMI_EVENT_SEL
+		| ARCH_PERFMON_NMI_EVENT_UMASK;
+
+	/* setup the timer */
+	wrmsr(evntsel_msr, evntsel, 0);
+	wrmsrl(perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
+
+	apic_write(APIC_LVTPC, APIC_DM_NMI);
+	evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE;
+	wrmsr(evntsel_msr, evntsel, 0);
+
+	wd->perfctr_msr = perfctr_msr;
+	wd->evntsel_msr = evntsel_msr;
+	wd->cccr_msr = 0;  //unused
+	wd->check_bit = 1ULL << (eax.split.bit_width - 1);
+	return 1;
+fail1:
+	release_perfctr_nmi(perfctr_msr);
+fail:
+	return 0;
+}
+
+static void stop_intel_arch_watchdog(void)
+{
+	unsigned int ebx;
+	union cpuid10_eax eax;
+	unsigned int unused;
+	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
+
+	/*
+	 * Check whether the Architectural PerfMon supports
+	 * Unhalted Core Cycles Event or not.
+	 * NOTE: Corresponding bit = 0 in ebx indicates event present.
+	 */
+	cpuid(10, &(eax.full), &ebx, &unused, &unused);
+	if ((eax.split.mask_length < (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
+	    (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
+		return;
+
+	wrmsr(wd->evntsel_msr, 0, 0);
+
+	release_evntsel_nmi(wd->evntsel_msr);
+	release_perfctr_nmi(wd->perfctr_msr);
+}
+
 void setup_apic_nmi_watchdog(void *unused)
 {
 	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
@@ -589,6 +694,11 @@ void setup_apic_nmi_watchdog(void *unused)
 				return;
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				if (!setup_intel_arch_watchdog())
+					return;
+				break;
+			}
 			if (!setup_p4_watchdog())
 				return;
 			break;
@@ -620,6 +730,10 @@ void stop_apic_nmi_watchdog(void *unused)
 			stop_k7_watchdog();
 			break;
 		case X86_VENDOR_INTEL:
+			if (cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) {
+				stop_intel_arch_watchdog();
+				break;
+			}
 			stop_p4_watchdog();
 			break;
 		default:
@@ -724,7 +838,13 @@ int __kprobes nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 				dummy &= ~P4_CCCR_OVF;
 	 			wrmsrl(wd->cccr_msr, dummy);
 	 			apic_write(APIC_LVTPC, APIC_DM_NMI);
-	 		}
+	 		} else if (wd->perfctr_msr == MSR_ARCH_PERFMON_PERFCTR0) {
+				/*
+				 * ArchPerfom/Core Duo needs to re-unmask
+				 * the apic vector
+				 */
+				apic_write(APIC_LVTPC, APIC_DM_NMI);
+			}
 			/* start the cycle over again */
 			wrmsrl(wd->perfctr_msr, -((u64)cpu_khz * 1000 / nmi_hz));
 			rc = 1;
diff --git a/include/asm-i386/intel_arch_perfmon.h b/include/asm-i386/intel_arch_perfmon.h
new file mode 100644
index 000000000000..b52cd60a075b
--- /dev/null
+++ b/include/asm-i386/intel_arch_perfmon.h
@@ -0,0 +1,31 @@
+#ifndef X86_INTEL_ARCH_PERFMON_H
+#define X86_INTEL_ARCH_PERFMON_H 1
+
+#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
+#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
+
+#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
+#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
+
+#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
+#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
+#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
+#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
+
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX (0)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT \
+				(1 << (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX))
+
+union cpuid10_eax {
+	struct {
+		unsigned int version_id:8;
+		unsigned int num_counters:8;
+		unsigned int bit_width:8;
+		unsigned int mask_length:8;
+	} split;
+	unsigned int full;
+};
+
+#endif	/* X86_INTEL_ARCH_PERFMON_H */
diff --git a/include/asm-x86_64/intel_arch_perfmon.h b/include/asm-x86_64/intel_arch_perfmon.h
new file mode 100644
index 000000000000..8633331420ec
--- /dev/null
+++ b/include/asm-x86_64/intel_arch_perfmon.h
@@ -0,0 +1,31 @@
+#ifndef X86_64_INTEL_ARCH_PERFMON_H
+#define X86_64_INTEL_ARCH_PERFMON_H 1
+
+#define MSR_ARCH_PERFMON_PERFCTR0		0xc1
+#define MSR_ARCH_PERFMON_PERFCTR1		0xc2
+
+#define MSR_ARCH_PERFMON_EVENTSEL0		0x186
+#define MSR_ARCH_PERFMON_EVENTSEL1		0x187
+
+#define ARCH_PERFMON_EVENTSEL0_ENABLE      (1 << 22)
+#define ARCH_PERFMON_EVENTSEL_INT          (1 << 20)
+#define ARCH_PERFMON_EVENTSEL_OS           (1 << 17)
+#define ARCH_PERFMON_EVENTSEL_USR          (1 << 16)
+
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL	(0x3c)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK	(0x00 << 8)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX (0)
+#define ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT \
+				(1 << (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX))
+
+union cpuid10_eax {
+	struct {
+		unsigned int version_id:8;
+		unsigned int num_counters:8;
+		unsigned int bit_width:8;
+		unsigned int mask_length:8;
+	} split;
+	unsigned int full;
+};
+
+#endif	/* X86_64_INTEL_ARCH_PERFMON_H */
-- 
cgit v1.2.1


From 81af4449af9c9b686a4eeeb00112614621655704 Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add macros for rdtscp

This patch adds macros for reading tsc via the RDTSCP instruction, as well
as writing the auxilliary MSR read by RDTSCP to msr.h

[AK: changed rdtscp definition for old binutils]

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/msr.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/include/asm-x86_64/msr.h b/include/asm-x86_64/msr.h
index 10f8b51cec8b..37e194169fac 100644
--- a/include/asm-x86_64/msr.h
+++ b/include/asm-x86_64/msr.h
@@ -66,14 +66,25 @@
 #define rdtscl(low) \
      __asm__ __volatile__ ("rdtsc" : "=a" (low) : : "edx")
 
+#define rdtscp(low,high,aux) \
+     asm volatile (".byte 0x0f,0x01,0xf9" : "=a" (low), "=d" (high), "=c" (aux))
+
 #define rdtscll(val) do { \
      unsigned int __a,__d; \
      asm volatile("rdtsc" : "=a" (__a), "=d" (__d)); \
      (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
 } while(0)
 
+#define rdtscpll(val, aux) do { \
+     unsigned long __a, __d; \
+     asm volatile (".byte 0x0f,0x01,0xf9" : "=a" (__a), "=d" (__d), "=c" (aux)); \
+     (val) = (__d << 32) | __a; \
+} while (0)
+
 #define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
 
+#define write_rdtscp_aux(val) wrmsr(0xc0000103, val, 0)
+
 #define rdpmc(counter,low,high) \
      __asm__ __volatile__("rdpmc" \
 			  : "=a" (low), "=d" (high) \
-- 
cgit v1.2.1


From a670fad0adb1cc6202a607d250f10bd380593905 Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add initalization of the RDTSCP auxilliary values

This patch adds initalization of the RDTSCP auxilliary values to CPU numbers
to time.c. If RDTSCP is available, the MSRs are written with the respective
values. It can be later used to initalize per-cpu timekeeping variables.

AK: Some cleanups. Move externs into headers and fix CPU hotplug.

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/smpboot.c |  2 ++
 arch/x86_64/kernel/time.c    | 47 +++++++++++++++++++++++++++++++++-----------
 include/asm-x86_64/proto.h   |  1 +
 3 files changed, 38 insertions(+), 12 deletions(-)

diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 15555879ce95..582896f7d420 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1181,6 +1181,8 @@ void __init smp_cpus_done(unsigned int max_cpus)
 #endif
 
 	check_nmi_watchdog();
+
+	time_init_gtod();
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 7a9b18224182..97b9e46d1992 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -24,6 +24,8 @@
 #include <linux/device.h>
 #include <linux/sysdev.h>
 #include <linux/bcd.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
 #include <linux/kallsyms.h>
 #include <linux/acpi.h>
 #ifdef CONFIG_ACPI
@@ -49,7 +51,7 @@ static void cpufreq_delayed_get(void);
 extern void i8254_timer_resume(void);
 extern int using_apic_timer;
 
-static char *time_init_gtod(void);
+static char *timename = NULL;
 
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
@@ -893,11 +895,21 @@ static struct irqaction irq0 = {
 	timer_interrupt, IRQF_DISABLED, CPU_MASK_NONE, "timer", NULL, NULL
 };
 
-void __init time_init(void)
+static int __cpuinit
+time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu)
 {
-	char *timename;
-	char *gtod;
+	unsigned cpu = (unsigned long) hcpu;
+	if (action == CPU_ONLINE &&
+		cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
+		unsigned p;
+		p = smp_processor_id() | (cpu_to_node(smp_processor_id())<<12);
+		write_rdtscp_aux(p);
+	}
+	return NOTIFY_DONE;
+}
 
+void __init time_init(void)
+{
 	if (nohpet)
 		vxtime.hpet_address = 0;
 
@@ -931,18 +943,19 @@ void __init time_init(void)
 	}
 
 	vxtime.mode = VXTIME_TSC;
-	gtod = time_init_gtod();
-
-	printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
-	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename, gtod);
-	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
-		cpu_khz / 1000, cpu_khz % 1000);
 	vxtime.quot = (USEC_PER_SEC << US_SCALE) / vxtime_hz;
 	vxtime.tsc_quot = (USEC_PER_MSEC << US_SCALE) / cpu_khz;
 	vxtime.last_tsc = get_cycles_sync();
 	setup_irq(0, &irq0);
 
 	set_cyc2ns_scale(cpu_khz);
+
+	hotcpu_notifier(time_cpu_notifier, 0);
+	time_cpu_notifier(NULL, CPU_ONLINE, (void *)(long)smp_processor_id());
+
+#ifndef CONFIG_SMP
+	time_init_gtod();
+#endif
 }
 
 /*
@@ -973,12 +986,13 @@ __cpuinit int unsynchronized_tsc(void)
 /*
  * Decide what mode gettimeofday should use.
  */
-__init static char *time_init_gtod(void)
+void time_init_gtod(void)
 {
 	char *timetype;
 
 	if (unsynchronized_tsc())
 		notsc = 1;
+
 	if (vxtime.hpet_address && notsc) {
 		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
 		if (hpet_use_timer)
@@ -1001,7 +1015,16 @@ __init static char *time_init_gtod(void)
 		timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC";
 		vxtime.mode = VXTIME_TSC;
 	}
-	return timetype;
+
+	printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
+	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename, timetype);
+	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
+		cpu_khz / 1000, cpu_khz % 1000);
+	vxtime.quot = (USEC_PER_SEC << US_SCALE) / vxtime_hz;
+	vxtime.tsc_quot = (USEC_PER_MSEC << US_SCALE) / cpu_khz;
+	vxtime.last_tsc = get_cycles_sync();
+
+	set_cyc2ns_scale(cpu_khz);
 }
 
 __setup("report_lost_ticks", time_setup);
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 038fe1f47e6f..3b1c60247902 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -51,6 +51,7 @@ extern unsigned long long monotonic_base;
 extern int sysctl_vsyscall;
 extern int nohpet;
 extern unsigned long vxtime_hz;
+extern void time_init_gtod(void);
 
 extern int numa_setup(char *opt);
 
-- 
cgit v1.2.1


From c08c820508233b424deab3302bc404bbecc6493a Mon Sep 17 00:00:00 2001
From: Vojtech Pavlik <vojtech@suse.cz>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Add the vgetcpu vsyscall

This patch adds a vgetcpu vsyscall, which depending on the CPU RDTSCP
capability uses either the RDTSCP or CPUID to obtain a CPU and node
numbers and pass them to the program.

AK: Lots of changes over Vojtech's original code:
Better prototype for vgetcpu()
It's better to pass the cpu / node numbers as separate arguments
to avoid mistakes when going from SMP to NUMA.
Also add a fast time stamp based cache using a user supplied
argument to speed things more up.
Use fast method from Chuck Ebbert to retrieve node/cpu from
GDT limit instead of CPUID
Made sure RDTSCP init is always executed after node is known.
Drop printk

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/head.S        |  2 +-
 arch/x86_64/kernel/time.c        | 13 ++++---
 arch/x86_64/kernel/vmlinux.lds.S |  3 ++
 arch/x86_64/kernel/vsyscall.c    | 84 +++++++++++++++++++++++++++++++++++++++-
 include/asm-x86_64/segment.h     |  5 ++-
 include/asm-x86_64/smp.h         | 12 ++++--
 include/asm-x86_64/vsyscall.h    |  9 +++++
 include/linux/getcpu.h           | 16 ++++++++
 8 files changed, 130 insertions(+), 14 deletions(-)
 create mode 100644 include/linux/getcpu.h

diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index c9739ca81d06..505ec4a57506 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -371,7 +371,7 @@ ENTRY(cpu_gdt_table)
 	.quad	0,0			/* TSS */
 	.quad	0,0			/* LDT */
 	.quad   0,0,0			/* three TLS descriptors */ 
-	.quad	0			/* unused */
+	.quad	0x0000f40000000000	/* node/CPU stored in limit */
 gdt_end:	
 	/* asm/segment.h:GDT_ENTRIES must match this */	
 	/* This should be a multiple of the cache line size */
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 97b9e46d1992..560ed944dc0e 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -899,12 +899,8 @@ static int __cpuinit
 time_cpu_notifier(struct notifier_block *nb, unsigned long action, void *hcpu)
 {
 	unsigned cpu = (unsigned long) hcpu;
-	if (action == CPU_ONLINE &&
-		cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
-		unsigned p;
-		p = smp_processor_id() | (cpu_to_node(smp_processor_id())<<12);
-		write_rdtscp_aux(p);
-	}
+	if (action == CPU_ONLINE)
+		vsyscall_set_cpu(cpu);
 	return NOTIFY_DONE;
 }
 
@@ -993,6 +989,11 @@ void time_init_gtod(void)
 	if (unsynchronized_tsc())
 		notsc = 1;
 
+ 	if (cpu_has(&boot_cpu_data, X86_FEATURE_RDTSCP))
+		vgetcpu_mode = VGETCPU_RDTSCP;
+	else
+		vgetcpu_mode = VGETCPU_LSL;
+
 	if (vxtime.hpet_address && notsc) {
 		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
 		if (hpet_use_timer)
diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 7c4de31471d4..8d5a5149bb3a 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -99,6 +99,9 @@ SECTIONS
   .vxtime : AT(VLOAD(.vxtime)) { *(.vxtime) }
   vxtime = VVIRT(.vxtime);
 
+  .vgetcpu_mode : AT(VLOAD(.vgetcpu_mode)) { *(.vgetcpu_mode) }
+  vgetcpu_mode = VVIRT(.vgetcpu_mode);
+
   .wall_jiffies : AT(VLOAD(.wall_jiffies)) { *(.wall_jiffies) }
   wall_jiffies = VVIRT(.wall_jiffies);
 
diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index f603037df162..902783bc4d53 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -26,6 +26,7 @@
 #include <linux/seqlock.h>
 #include <linux/jiffies.h>
 #include <linux/sysctl.h>
+#include <linux/getcpu.h>
 
 #include <asm/vsyscall.h>
 #include <asm/pgtable.h>
@@ -33,11 +34,15 @@
 #include <asm/fixmap.h>
 #include <asm/errno.h>
 #include <asm/io.h>
+#include <asm/segment.h>
+#include <asm/desc.h>
+#include <asm/topology.h>
 
 #define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
 
 int __sysctl_vsyscall __section_sysctl_vsyscall = 1;
 seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED;
+int __vgetcpu_mode __section_vgetcpu_mode;
 
 #include <asm/unistd.h>
 
@@ -127,9 +132,46 @@ time_t __vsyscall(1) vtime(time_t *t)
 	return __xtime.tv_sec;
 }
 
-long __vsyscall(2) venosys_0(void)
+/* Fast way to get current CPU and node.
+   This helps to do per node and per CPU caches in user space.
+   The result is not guaranteed without CPU affinity, but usually
+   works out because the scheduler tries to keep a thread on the same
+   CPU.
+
+   tcache must point to a two element sized long array.
+   All arguments can be NULL. */
+long __vsyscall(2)
+vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
 {
-	return -ENOSYS;
+	unsigned int dummy, p;
+	unsigned long j = 0;
+
+	/* Fast cache - only recompute value once per jiffies and avoid
+	   relatively costly rdtscp/cpuid otherwise.
+	   This works because the scheduler usually keeps the process
+	   on the same CPU and this syscall doesn't guarantee its
+	   results anyways.
+	   We do this here because otherwise user space would do it on
+	   its own in a likely inferior way (no access to jiffies).
+	   If you don't like it pass NULL. */
+	if (tcache && tcache->t0 == (j = __jiffies)) {
+		p = tcache->t1;
+	} else if (__vgetcpu_mode == VGETCPU_RDTSCP) {
+		/* Load per CPU data from RDTSCP */
+		rdtscp(dummy, dummy, p);
+	} else {
+		/* Load per CPU data from GDT */
+		asm("lsl %1,%0" : "=r" (p) : "r" (__PER_CPU_SEG));
+	}
+	if (tcache) {
+		tcache->t0 = j;
+		tcache->t1 = p;
+	}
+	if (cpu)
+		*cpu = p & 0xfff;
+	if (node)
+		*node = p >> 12;
+	return 0;
 }
 
 long __vsyscall(3) venosys_1(void)
@@ -200,6 +242,43 @@ static ctl_table kernel_root_table2[] = {
 
 #endif
 
+static void __cpuinit write_rdtscp_cb(void *info)
+{
+	write_rdtscp_aux((unsigned long)info);
+}
+
+void __cpuinit vsyscall_set_cpu(int cpu)
+{
+	unsigned long *d;
+	unsigned long node = 0;
+#ifdef CONFIG_NUMA
+	node = cpu_to_node[cpu];
+#endif
+	if (cpu_has(&cpu_data[cpu], X86_FEATURE_RDTSCP)) {
+		void *info = (void *)((node << 12) | cpu);
+		/* Can happen on preemptive kernel */
+		if (get_cpu() == cpu)
+			write_rdtscp_cb(info);
+#ifdef CONFIG_SMP
+		else {
+			/* the notifier is unfortunately not executed on the
+			   target CPU */
+			smp_call_function_single(cpu,write_rdtscp_cb,info,0,1);
+		}
+#endif
+		put_cpu();
+	}
+
+	/* Store cpu number in limit so that it can be loaded quickly
+	   in user space in vgetcpu.
+	   12 bits for the CPU and 8 bits for the node. */
+	d = (unsigned long *)(cpu_gdt(cpu) + GDT_ENTRY_PER_CPU);
+	*d = 0x0f40000000000ULL;
+	*d |= cpu;
+	*d |= (node & 0xf) << 12;
+	*d |= (node >> 4) << 48;
+}
+
 static void __init map_vsyscall(void)
 {
 	extern char __vsyscall_0;
@@ -214,6 +293,7 @@ static int __init vsyscall_init(void)
 			VSYSCALL_ADDR(__NR_vgettimeofday)));
 	BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
 	BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
+	BUG_ON((unsigned long) &vgetcpu != VSYSCALL_ADDR(__NR_vgetcpu));
 	map_vsyscall();
 #ifdef CONFIG_SYSCTL
 	register_sysctl_table(kernel_root_table2, 0);
diff --git a/include/asm-x86_64/segment.h b/include/asm-x86_64/segment.h
index d4bed33fb32c..334ddcdd8f92 100644
--- a/include/asm-x86_64/segment.h
+++ b/include/asm-x86_64/segment.h
@@ -20,15 +20,16 @@
 #define __USER_CS     0x33   /* 6*8+3 */ 
 #define __USER32_DS	__USER_DS 
 
-#define GDT_ENTRY_TLS 1
 #define GDT_ENTRY_TSS 8	/* needs two entries */
 #define GDT_ENTRY_LDT 10 /* needs two entries */
 #define GDT_ENTRY_TLS_MIN 12
 #define GDT_ENTRY_TLS_MAX 14
-/* 15 free */
 
 #define GDT_ENTRY_TLS_ENTRIES 3
 
+#define GDT_ENTRY_PER_CPU 15	/* Abused to load per CPU data from limit */
+#define __PER_CPU_SEG	(GDT_ENTRY_PER_CPU * 8 + 3)
+
 /* TLS indexes for 64bit - hardcoded in arch_prctl */
 #define FS_TLS 0	
 #define GS_TLS 1	
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6805e1feb300..d61547fd833b 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -133,13 +133,19 @@ static __inline int logical_smp_processor_id(void)
 	/* we don't want to mark this access volatile - bad code generation */
 	return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
 }
-#endif
 
 #ifdef CONFIG_SMP
 #define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
 #define cpu_physical_id(cpu)		boot_cpu_id
-#endif
-
+static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
+				void *info, int retry, int wait)
+{
+	/* Disable interrupts here? */
+	func(info);
+	return 0;
+}
+#endif /* !CONFIG_SMP */
+#endif /* !__ASSEMBLY */
 #endif
 
diff --git a/include/asm-x86_64/vsyscall.h b/include/asm-x86_64/vsyscall.h
index 146b24402a5f..2281e9399b96 100644
--- a/include/asm-x86_64/vsyscall.h
+++ b/include/asm-x86_64/vsyscall.h
@@ -4,6 +4,7 @@
 enum vsyscall_num {
 	__NR_vgettimeofday,
 	__NR_vtime,
+	__NR_vgetcpu,
 };
 
 #define VSYSCALL_START (-10UL << 20)
@@ -15,6 +16,7 @@ enum vsyscall_num {
 #include <linux/seqlock.h>
 
 #define __section_vxtime __attribute__ ((unused, __section__ (".vxtime"), aligned(16)))
+#define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16)))
 #define __section_wall_jiffies __attribute__ ((unused, __section__ (".wall_jiffies"), aligned(16)))
 #define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16)))
 #define __section_sys_tz __attribute__ ((unused, __section__ (".sys_tz"), aligned(16)))
@@ -26,6 +28,9 @@ enum vsyscall_num {
 #define VXTIME_HPET	2
 #define VXTIME_PMTMR	3
 
+#define VGETCPU_RDTSCP	1
+#define VGETCPU_LSL	2
+
 struct vxtime_data {
 	long hpet_address;	/* HPET base address */
 	int last;
@@ -40,6 +45,7 @@ struct vxtime_data {
 
 /* vsyscall space (readonly) */
 extern struct vxtime_data __vxtime;
+extern int __vgetcpu_mode;
 extern struct timespec __xtime;
 extern volatile unsigned long __jiffies;
 extern unsigned long __wall_jiffies;
@@ -48,6 +54,7 @@ extern seqlock_t __xtime_lock;
 
 /* kernel space (writeable) */
 extern struct vxtime_data vxtime;
+extern int vgetcpu_mode;
 extern unsigned long wall_jiffies;
 extern struct timezone sys_tz;
 extern int sysctl_vsyscall;
@@ -55,6 +62,8 @@ extern seqlock_t xtime_lock;
 
 extern int sysctl_vsyscall;
 
+extern void vsyscall_set_cpu(int cpu);
+
 #define ARCH_HAVE_XTIME_LOCK 1
 
 #endif /* __KERNEL__ */
diff --git a/include/linux/getcpu.h b/include/linux/getcpu.h
new file mode 100644
index 000000000000..031ed3780e45
--- /dev/null
+++ b/include/linux/getcpu.h
@@ -0,0 +1,16 @@
+#ifndef _LINUX_GETCPU_H
+#define _LINUX_GETCPU_H 1
+
+/* Cache for getcpu() to speed it up. Results might be upto a jiffie
+   out of date, but will be faster.
+   User programs should not refer to the contents of this structure.
+   It is only a cache for vgetcpu(). It might change in future kernels.
+   The user program must store this information per thread (__thread)
+   If you want 100% accurate information pass NULL instead. */
+struct getcpu_cache {
+	unsigned long t0;
+	unsigned long t1;
+	unsigned long res[4];
+};
+
+#endif
-- 
cgit v1.2.1


From 3cfc348bf90ffaa777c188652aa297f04eb94de8 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] x86: Add portable getcpu call

For NUMA optimization and some other algorithms it is useful to have a fast
to get the current CPU and node numbers in user space.

x86-64 added a fast way to do this in a vsyscall. This adds a generic
syscall for other architectures to make it a generic portable facility.

I expect some of them will also implement it as a faster vsyscall.

The cache is an optimization for the x86-64 vsyscall optimization. Since
what the syscall returns is an approximation anyways and user space
often wants very fast results it can be cached for some time.  The norma
methods to get this information in user space are relatively slow

The vsyscall is in a better position to manage the cache because it has direct
access to a fast time stamp (jiffies). For the generic syscall optimization
it doesn't help much, but enforce a valid argument to keep programs
portable

I only added an i386 syscall entry for now. Other architectures can follow
as needed.

AK: Also added some cleanups from Andrew Morton

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/syscall_table.S |  1 +
 arch/x86_64/ia32/ia32entry.S     |  1 +
 include/asm-i386/unistd.h        |  3 ++-
 include/linux/syscalls.h         |  2 ++
 kernel/sys.c                     | 31 +++++++++++++++++++++++++++++++
 5 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index dd63d4775398..7e639f78b0b9 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -317,3 +317,4 @@ ENTRY(sys_call_table)
 	.long sys_tee			/* 315 */
 	.long sys_vmsplice
 	.long sys_move_pages
+	.long sys_getcpu
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 30ed0f6f4a2b..32fd32bea07c 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -713,4 +713,5 @@ ia32_sys_call_table:
 	.quad sys_tee
 	.quad compat_sys_vmsplice
 	.quad compat_sys_move_pages
+	.quad sys_getcpu
 ia32_syscall_end:		
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index fc1c8ddae149..565d0897b205 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -323,10 +323,11 @@
 #define __NR_tee		315
 #define __NR_vmsplice		316
 #define __NR_move_pages		317
+#define __NR_getcpu		318
 
 #ifdef __KERNEL__
 
-#define NR_syscalls 318
+#define NR_syscalls 319
 
 /*
  * user-visible error numbers are in the range -1 - -128: see
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 008f04c56737..3f0f716225ec 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -53,6 +53,7 @@ struct mq_attr;
 struct compat_stat;
 struct compat_timeval;
 struct robust_list_head;
+struct getcpu_cache;
 
 #include <linux/types.h>
 #include <linux/aio_abi.h>
@@ -596,5 +597,6 @@ asmlinkage long sys_get_robust_list(int pid,
 				    size_t __user *len_ptr);
 asmlinkage long sys_set_robust_list(struct robust_list_head __user *head,
 				    size_t len);
+asmlinkage long sys_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *cache);
 
 #endif
diff --git a/kernel/sys.c b/kernel/sys.c
index e236f98f7ec5..3f894775488d 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -28,6 +28,7 @@
 #include <linux/tty.h>
 #include <linux/signal.h>
 #include <linux/cn_proc.h>
+#include <linux/getcpu.h>
 
 #include <linux/compat.h>
 #include <linux/syscalls.h>
@@ -2062,3 +2063,33 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3,
 	}
 	return error;
 }
+
+asmlinkage long sys_getcpu(unsigned __user *cpup, unsigned __user *nodep,
+	   		   struct getcpu_cache __user *cache)
+{
+	int err = 0;
+	int cpu = raw_smp_processor_id();
+	if (cpup)
+		err |= put_user(cpu, cpup);
+	if (nodep)
+		err |= put_user(cpu_to_node(cpu), nodep);
+	if (cache) {
+		/*
+		 * The cache is not needed for this implementation,
+		 * but make sure user programs pass something
+		 * valid. vsyscall implementations can instead make
+		 * good use of the cache. Only use t0 and t1 because
+		 * these are available in both 32bit and 64bit ABI (no
+		 * need for a compat_getcpu). 32bit has enough
+		 * padding
+		 */
+		unsigned long t0, t1;
+		get_user(t0, &cache->t0);
+		get_user(t1, &cache->t1);
+		t0++;
+		t1++;
+		put_user(t0, &cache->t0);
+		put_user(t1, &cache->t1);
+	}
+	return err ? -EFAULT : 0;
+}
-- 
cgit v1.2.1


From 2f766d16062d0147edff91be15de4a950667ca42 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Clean up asm/smp.h includes

No need to include it from entry.S
Drop all the #ifdef __ASSEMBLY__

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S |  2 --
 include/asm-x86_64/smp.h   | 12 ------------
 2 files changed, 14 deletions(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index aa8d8939abc1..7eb1678e098d 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -27,10 +27,8 @@
  * - schedule it carefully for the final hardware.
  */
 
-#define ASSEMBLY 1
 #include <linux/linkage.h>
 #include <asm/segment.h>
-#include <asm/smp.h>
 #include <asm/cache.h>
 #include <asm/errno.h>
 #include <asm/dwarf2.h>
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index d61547fd833b..612d208961a2 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -4,15 +4,12 @@
 /*
  * We need the APIC definitions automatically as part of 'smp.h'
  */
-#ifndef __ASSEMBLY__
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 #include <linux/bitops.h>
 extern int disable_apic;
-#endif
 
 #ifdef CONFIG_X86_LOCAL_APIC
-#ifndef __ASSEMBLY__
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
 #ifdef CONFIG_X86_IO_APIC
@@ -21,10 +18,8 @@ extern int disable_apic;
 #include <asm/apic.h>
 #include <asm/thread_info.h>
 #endif
-#endif
 
 #ifdef CONFIG_SMP
-#ifndef ASSEMBLY
 
 #include <asm/pda.h>
 
@@ -83,13 +78,10 @@ extern void prefill_possible_map(void);
 extern unsigned num_processors;
 extern unsigned disabled_cpus;
 
-#endif /* !ASSEMBLY */
-
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
 
 #endif
 
-#ifndef ASSEMBLY
 /*
  * Some lowlevel functions might want to know about
  * the real APIC ID <-> CPU # mapping.
@@ -111,8 +103,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 		return BAD_APICID;
 }
 
-#endif /* !ASSEMBLY */
-
 #ifndef CONFIG_SMP
 #define stack_smp_processor_id() 0
 #define safe_smp_processor_id() 0
@@ -127,7 +117,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 })
 #endif
 
-#ifndef __ASSEMBLY__
 static __inline int logical_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
@@ -146,6 +135,5 @@ static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
 	return 0;
 }
 #endif /* !CONFIG_SMP */
-#endif /* !__ASSEMBLY */
 #endif
 
-- 
cgit v1.2.1


From d3a4f48d4866b8623ca9adde8ce4e5fde979c132 Mon Sep 17 00:00:00 2001
From: Stephane Eranian <eranian@hpl.hp.com>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] x86-64 TIF flags for debug regs and io bitmap in ctxsw

Hello,

Following my discussion with Andi. Here is a patch that introduces
two new TIF flags to simplify the context switch code in __switch_to().
The idea is to minimize the number of cache lines accessed in the common
case, i.e., when neither the debug registers nor the I/O bitmap are used.

This patch covers the x86-64 modifications. A patch for i386 follows.

Changelog:
	- add TIF_DEBUG to track when debug registers are active
	- add TIF_IO_BITMAP to track when I/O bitmap is used
	- modify __switch_to() to use the new TIF flags

<signed-off-by>: eranian@hpl.hp.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ptrace32.c      |  4 +++
 arch/x86_64/kernel/ioport.c      |  1 +
 arch/x86_64/kernel/process.c     | 73 +++++++++++++++++++++++-----------------
 arch/x86_64/kernel/ptrace.c      |  8 +++--
 include/asm-x86_64/thread_info.h |  7 ++++
 5 files changed, 60 insertions(+), 33 deletions(-)

diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c
index 659c0722f6b8..72bf92a9d375 100644
--- a/arch/x86_64/ia32/ptrace32.c
+++ b/arch/x86_64/ia32/ptrace32.c
@@ -117,6 +117,10 @@ static int putreg32(struct task_struct *child, unsigned regno, u32 val)
 			if ((0x5454 >> ((val >> (16 + 4*i)) & 0xf)) & 1)
 			       return -EIO;
 		child->thread.debugreg7 = val; 
+		if (val)
+			set_tsk_thread_flag(child, TIF_DEBUG);
+		else
+			clear_tsk_thread_flag(child, TIF_DEBUG);
 		break; 
 		    
 	default:
diff --git a/arch/x86_64/kernel/ioport.c b/arch/x86_64/kernel/ioport.c
index b81614970ecc..fe063d3cfe42 100644
--- a/arch/x86_64/kernel/ioport.c
+++ b/arch/x86_64/kernel/ioport.c
@@ -56,6 +56,7 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
 
 		memset(bitmap, 0xff, IO_BITMAP_BYTES);
 		t->io_bitmap_ptr = bitmap;
+		set_thread_flag(TIF_IO_BITMAP);
 	}
 
 	/*
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index bb6745d13b8f..6e0527635b4c 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -350,6 +350,7 @@ void exit_thread(void)
 
 		kfree(t->io_bitmap_ptr);
 		t->io_bitmap_ptr = NULL;
+		clear_thread_flag(TIF_IO_BITMAP);
 		/*
 		 * Careful, clear this in the TSS too:
 		 */
@@ -369,6 +370,7 @@ void flush_thread(void)
 		if (t->flags & _TIF_IA32)
 			current_thread_info()->status |= TS_COMPAT;
 	}
+	t->flags &= ~_TIF_DEBUG;
 
 	tsk->thread.debugreg0 = 0;
 	tsk->thread.debugreg1 = 0;
@@ -461,7 +463,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 	asm("mov %%es,%0" : "=m" (p->thread.es));
 	asm("mov %%ds,%0" : "=m" (p->thread.ds));
 
-	if (unlikely(me->thread.io_bitmap_ptr != NULL)) { 
+	if (unlikely(test_tsk_thread_flag(me, TIF_IO_BITMAP))) {
 		p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
 		if (!p->thread.io_bitmap_ptr) {
 			p->thread.io_bitmap_max = 0;
@@ -469,6 +471,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
 		}
 		memcpy(p->thread.io_bitmap_ptr, me->thread.io_bitmap_ptr,
 				IO_BITMAP_BYTES);
+		set_tsk_thread_flag(p, TIF_IO_BITMAP);
 	} 
 
 	/*
@@ -498,6 +501,40 @@ out:
  */
 #define loaddebug(thread,r) set_debugreg(thread->debugreg ## r, r)
 
+static inline void __switch_to_xtra(struct task_struct *prev_p,
+			     	    struct task_struct *next_p,
+			     	    struct tss_struct *tss)
+{
+	struct thread_struct *prev, *next;
+
+	prev = &prev_p->thread,
+	next = &next_p->thread;
+
+	if (test_tsk_thread_flag(next_p, TIF_DEBUG)) {
+		loaddebug(next, 0);
+		loaddebug(next, 1);
+		loaddebug(next, 2);
+		loaddebug(next, 3);
+		/* no 4 and 5 */
+		loaddebug(next, 6);
+		loaddebug(next, 7);
+	}
+
+	if (test_tsk_thread_flag(next_p, TIF_IO_BITMAP)) {
+		/*
+		 * Copy the relevant range of the IO bitmap.
+		 * Normally this is 128 bytes or less:
+		 */
+		memcpy(tss->io_bitmap, next->io_bitmap_ptr,
+		       max(prev->io_bitmap_max, next->io_bitmap_max));
+	} else if (test_tsk_thread_flag(prev_p, TIF_IO_BITMAP)) {
+		/*
+		 * Clear any possible leftover bits:
+		 */
+		memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
+	}
+}
+
 /*
  *	switch_to(x,y) should switch tasks from x to y.
  *
@@ -586,37 +623,11 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 		  task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
 
 	/*
-	 * Now maybe reload the debug registers
+	 * Now maybe reload the debug registers and handle I/O bitmaps
 	 */
-	if (unlikely(next->debugreg7)) {
-		loaddebug(next, 0);
-		loaddebug(next, 1);
-		loaddebug(next, 2);
-		loaddebug(next, 3);
-		/* no 4 and 5 */
-		loaddebug(next, 6);
-		loaddebug(next, 7);
-	}
-
-
-	/* 
-	 * Handle the IO bitmap 
-	 */ 
-	if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
-		if (next->io_bitmap_ptr)
-			/*
-			 * Copy the relevant range of the IO bitmap.
-			 * Normally this is 128 bytes or less:
- 			 */
-			memcpy(tss->io_bitmap, next->io_bitmap_ptr,
-				max(prev->io_bitmap_max, next->io_bitmap_max));
-		else {
-			/*
-			 * Clear any possible leftover bits:
-			 */
-			memset(tss->io_bitmap, 0xff, prev->io_bitmap_max);
-		}
-	}
+	if (unlikely((task_thread_info(next_p)->flags & _TIF_WORK_CTXSW))
+	    || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
+		__switch_to_xtra(prev_p, next_p, tss);
 
 	return prev_p;
 }
diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index 2d50024c9f30..d35ec1bc696a 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -420,9 +420,13 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 				if ((0x5554 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
 					break;
 			if (i == 4) {
-				child->thread.debugreg7 = data;
+			  child->thread.debugreg7 = data;
+			  if (data)
+			  	set_tsk_thread_flag(child, TIF_DEBUG);
+			  else
+			  	clear_tsk_thread_flag(child, TIF_DEBUG);
 			  ret = 0;
-		  }
+		  	}
 		  break;
 		}
 		break;
diff --git a/include/asm-x86_64/thread_info.h b/include/asm-x86_64/thread_info.h
index 790c512a4369..787a08114b48 100644
--- a/include/asm-x86_64/thread_info.h
+++ b/include/asm-x86_64/thread_info.h
@@ -120,6 +120,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define TIF_FORK		18	/* ret_from_fork */
 #define TIF_ABI_PENDING		19
 #define TIF_MEMDIE		20
+#define TIF_DEBUG		21	/* uses debug registers */
+#define TIF_IO_BITMAP		22	/* uses I/O bitmap */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
@@ -133,6 +135,8 @@ static inline struct thread_info *stack_thread_info(void)
 #define _TIF_IA32		(1<<TIF_IA32)
 #define _TIF_FORK		(1<<TIF_FORK)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
+#define _TIF_DEBUG		(1<<TIF_DEBUG)
+#define _TIF_IO_BITMAP		(1<<TIF_IO_BITMAP)
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
@@ -140,6 +144,9 @@ static inline struct thread_info *stack_thread_info(void)
 /* work to do on any return to user space */
 #define _TIF_ALLWORK_MASK (0x0000FFFF & ~_TIF_SECCOMP)
 
+/* flags to check in __switch_to() */
+#define _TIF_WORK_CTXSW (_TIF_DEBUG|_TIF_IO_BITMAP)
+
 #define PREEMPT_ACTIVE     0x10000000
 
 /*
-- 
cgit v1.2.1


From 3034d11c930f795d61321c9244c4ffaaabf0c282 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Don't print virtual address in HPET initialization

virtual addresses don't belong into kernel logs for non debugging

Cc: clemens@ladisch.de
Signed-off-by: Andi Kleen <ak@suse.de>
---
 drivers/char/hpet.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 8afba339f05a..58b0eb581114 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -868,8 +868,8 @@ int hpet_alloc(struct hpet_data *hdp)
 	do_div(temp, period);
 	hpetp->hp_tick_freq = temp; /* ticks per second */
 
-	printk(KERN_INFO "hpet%d: at MMIO 0x%lx (virtual 0x%p), IRQ%s",
-		hpetp->hp_which, hdp->hd_phys_address, hdp->hd_address,
+	printk(KERN_INFO "hpet%d: at MMIO 0x%lx, IRQ%s",
+		hpetp->hp_which, hdp->hd_phys_address,
 		hpetp->hp_ntimer > 1 ? "s" : "");
 	for (i = 0; i < hpetp->hp_ntimer; i++)
 		printk("%s %d", i > 0 ? "," : "", hdp->hd_irq[i]);
-- 
cgit v1.2.1


From d5d9ca6d882f7c8d47ef91a701fc042cbebbc334 Mon Sep 17 00:00:00 2001
From: Adam Henley <adamazing@gmail.com>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] A few trivial spelling and grammar fixes

A few trivial spelling and grammar mistakes picked up in
"arch/x86_64/aperture.c", "arch/x86_64/crash.c" and
"arch/x86_64/apic.c". I think all are correct fixes but am ever aware
of my fallibility :o) This is my first patch submission so all
feedback is appreciated, esp. WRT CCing to Linus, Andi and
trivial@kernel.org, is this correct? And which is the most appropriate
kernel version to diff against? If any.

Should apply cleanly to 2.6.18-rc1

Signed-off-by: Adam Henley <adamazing@gmail.com>
Signed-off-by: Andi Kleen <ak@suse.de>

-  adam
---
 arch/x86_64/kernel/aperture.c | 2 +-
 arch/x86_64/kernel/apic.c     | 4 ++--
 arch/x86_64/kernel/crash.c    | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index 58af8e73738b..04dbf1662523 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -48,7 +48,7 @@ static u32 __init allocate_aperture(void)
 
 	/* 
 	 * Aperture has to be naturally aligned. This means an 2GB aperture won't
-	 * have much chances to find a place in the lower 4GB of memory.
+	 * have much chance of finding a place in the lower 4GB of memory.
 	 * Unfortunately we cannot move it up because that would make the
 	 * IOMMU useless.
 	 */
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 692e9b579743..63d9b037afc6 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -400,7 +400,7 @@ void __cpuinit setup_local_APIC (void)
 	value |= APIC_SPIV_APIC_ENABLED;
 
 	/*
-	 * Some unknown Intel IO/APIC (or APIC) errata is biting us with
+	 * Some unknown Intel IO/APIC (or APIC) errata are biting us with
 	 * certain networking cards. If high frequency interrupts are
 	 * happening on a particular IOAPIC pin, plus the IOAPIC routing
 	 * entry is masked/unmasked at a high rate as well then sooner or
@@ -950,7 +950,7 @@ void smp_local_timer_interrupt(struct pt_regs *regs)
 	 * We take the 'long' return path, and there every subsystem
 	 * grabs the appropriate locks (kernel lock/ irq lock).
 	 *
-	 * we might want to decouple profiling from the 'long path',
+	 * We might want to decouple profiling from the 'long path',
 	 * and do the profiling totally in assembly.
 	 *
 	 * Currently this isn't too much of an issue (performance wise),
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index fc57a139123e..7d7a9e75f70c 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -69,7 +69,7 @@ static void crash_save_this_cpu(struct pt_regs *regs, int cpu)
 	 * for the data I pass, and I need tags
 	 * on the data to indicate what information I have
 	 * squirrelled away.  ELF notes happen to provide
-	 * all of that that no need to invent something new.
+	 * all of that, no need to invent something new.
 	 */
 
 	buf = (u32*)per_cpu_ptr(crash_notes, cpu);
-- 
cgit v1.2.1


From c16b63e09d9d03158e0a92e961234e94c4862620 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] i386/x86-64: Don't randomize stack top when no randomization
 personality is set

Based on patch from Frank van Maarseveen <frankvm@frankvm.com>, but
extended.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/process.c   | 3 ++-
 arch/x86_64/kernel/process.c | 2 +-
 fs/binfmt_elf.c              | 3 ++-
 3 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 8657c739656a..b741c3e1a5eb 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -37,6 +37,7 @@
 #include <linux/kallsyms.h>
 #include <linux/ptrace.h>
 #include <linux/random.h>
+#include <linux/personality.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -905,7 +906,7 @@ asmlinkage int sys_get_thread_area(struct user_desc __user *u_info)
 
 unsigned long arch_align_stack(unsigned long sp)
 {
-	if (randomize_va_space)
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
 		sp -= get_random_int() % 8192;
 	return sp & ~0xf;
 }
diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 6e0527635b4c..6fbd19564e4e 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -845,7 +845,7 @@ int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
 
 unsigned long arch_align_stack(unsigned long sp)
 {
-	if (randomize_va_space)
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
 		sp -= get_random_int() % 8192;
 	return sp & ~0xf;
 }
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 672a3b90bc55..5109dbff93bf 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -515,7 +515,8 @@ static unsigned long randomize_stack_top(unsigned long stack_top)
 {
 	unsigned int random_variable = 0;
 
-	if (current->flags & PF_RANDOMIZE) {
+	if ((current->flags & PF_RANDOMIZE) &&
+		!(current->personality & ADDR_NO_RANDOMIZE)) {
 		random_variable = get_random_int() & STACK_RND_MASK;
 		random_variable <<= PAGE_SHIFT;
 	}
-- 
cgit v1.2.1


From 0cb91a2293648507886563ccb91979cfc94d6a4b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] i386: Account spinlocks to the caller during profiling for
 !FP kernels

This ports the algorithm from x86-64 (with improvements) to i386.
Previously this only worked for frame pointer enabled kernels.
But spinlocks have a very simple stack frame that can be manually
analyzed. Do this.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/time.c   | 23 +++++++++++++++++++----
 include/asm-i386/ptrace.h |  4 ----
 kernel/spinlock.c         |  5 +++++
 3 files changed, 24 insertions(+), 8 deletions(-)

diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index edd00f6cee37..5af802ef00b2 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -130,18 +130,33 @@ static int set_rtc_mmss(unsigned long nowtime)
 
 int timer_ack;
 
-#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 unsigned long profile_pc(struct pt_regs *regs)
 {
 	unsigned long pc = instruction_pointer(regs);
 
-	if (!user_mode_vm(regs) && in_lock_functions(pc))
+#ifdef CONFIG_SMP
+	if (!user_mode_vm(regs) && in_lock_functions(pc)) {
+#ifdef CONFIG_FRAME_POINTER
 		return *(unsigned long *)(regs->ebp + 4);
-
+#else
+		unsigned long *sp;
+		if ((regs->xcs & 3) == 0)
+			sp = (unsigned long *)&regs->esp;
+		else
+			sp = (unsigned long *)regs->esp;
+		/* Return address is either directly at stack pointer
+		   or above a saved eflags. Eflags has bits 22-31 zero,
+		   kernel addresses don't. */
+ 		if (sp[0] >> 22)
+			return sp[0];
+		if (sp[1] >> 22)
+			return sp[1];
+#endif
+	}
+#endif
 	return pc;
 }
 EXPORT_SYMBOL(profile_pc);
-#endif
 
 /*
  * This is the same as the above, except we _also_ save the current
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index f324c53b6f9a..30a442ec2059 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -80,11 +80,7 @@ static inline int user_mode_vm(struct pt_regs *regs)
 	return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0;
 }
 #define instruction_pointer(regs) ((regs)->eip)
-#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
 extern unsigned long profile_pc(struct pt_regs *regs);
-#else
-#define profile_pc(regs) instruction_pointer(regs)
-#endif
 #endif /* __KERNEL__ */
 
 #endif
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index fb524b009eef..9644a41e0bef 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -7,6 +7,11 @@
  *
  * This file contains the spinlock/rwlock implementations for the
  * SMP and the DEBUG_SPINLOCK cases. (UP-nondebug inlines them)
+ *
+ * Note that some architectures have special knowledge about the
+ * stack frames of these functions in their profile_pc. If you
+ * change anything significant here that could change the stack
+ * frame contact the architecture maintainers.
  */
 
 #include <linux/linkage.h>
-- 
cgit v1.2.1


From 31679f38d88696ed032d59e457f1605c97e7d719 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Simplify profile_pc on x86-64

Use knowledge about EFLAGS layout (bits 22:63 are always 0) to distingush
EFLAGS word and kernel address in the spin lock stack frame.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/time.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)

diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 560ed944dc0e..ea00915d393a 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -189,20 +189,15 @@ unsigned long profile_pc(struct pt_regs *regs)
 {
 	unsigned long pc = instruction_pointer(regs);
 
-	/* Assume the lock function has either no stack frame or only a single 
-	   word.  This checks if the address on the stack looks like a kernel 
-	   text address.
-	   There is a small window for false hits, but in that case the tick
-	   is just accounted to the spinlock function.
-	   Better would be to write these functions in assembler again
-	   and check exactly. */
+	/* Assume the lock function has either no stack frame or a copy
+	   of eflags from PUSHF
+	   Eflags always has bits 22 and up cleared unlike kernel addresses. */
 	if (!user_mode(regs) && in_lock_functions(pc)) {
-		char *v = *(char **)regs->rsp;
-		if ((v >= _stext && v <= _etext) ||
-			(v >= _sinittext && v <= _einittext) ||
-			(v >= (char *)MODULES_VADDR  && v <= (char *)MODULES_END))
-			return (unsigned long)v;
-		return ((unsigned long *)regs->rsp)[1];
+		unsigned long *sp = (unsigned long *)regs->rsp;
+		if (sp[0] >> 22)
+			return sp[0];
+		if (sp[1] >> 22)
+			return sp[1];
 	}
 	return pc;
 }
-- 
cgit v1.2.1


From 538b5b419c7ae39a4b2deb15278da36102e42346 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Document backtracer selection options

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/x86_64/boot-options.txt | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 6da24e7a56cb..4303e0c12476 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -245,6 +245,13 @@ Debugging
 		newfallback: use new unwinder but fall back to old if it gets
 			stuck (default)
 
+  call_trace=[old|both|newfallback|new]
+		old: use old inexact backtracer
+		new: use new exact dwarf2 unwinder
+ 		both: print entries from both
+		newfallback: use new unwinder but fall back to old if it gets
+			stuck (default)
+
 Misc
 
   noreplacement  Don't replace instructions with more appropriate ones
-- 
cgit v1.2.1


From b4062b16094038334d9bbadac0397a3fc9e981b0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Support patchable lock prefix for pure assembly files

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/alternative-asm.i | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/asm-x86_64/alternative-asm.i

diff --git a/include/asm-x86_64/alternative-asm.i b/include/asm-x86_64/alternative-asm.i
new file mode 100644
index 000000000000..e4041f4fa4dc
--- /dev/null
+++ b/include/asm-x86_64/alternative-asm.i
@@ -0,0 +1,14 @@
+#include <linux/config.h>
+
+#ifdef CONFIG_SMP
+	.macro LOCK_PREFIX
+1:	lock
+	.section .smp_locks,"a"
+	.align 8
+	.quad 1b
+	.previous
+	.endm
+#else
+	.macro LOCK_PREFIX
+	.endm
+#endif
-- 
cgit v1.2.1


From 9a0b26e6bc4ae1979d9bcc6194e57a71b2b5cac6 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:28 +0200
Subject: [PATCH] Clean up read write lock assembly

- Move the slow path fallbacks to their own assembly files
This makes them much easier to read and is needed for the next change.
- Add CFI annotations for unwinding (XXX need review)
- Remove constant case which can never happen with out of line spinlocks
- Use patchable LOCK prefixes
- Don't use lock sections anymore for inline code because they can't
be expressed by the unwinder (this adds one taken jump to the lock
fast path)

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/lib/Makefile      |  2 +-
 arch/x86_64/lib/rwlock.S      | 38 ++++++++++++++++++++++++
 arch/x86_64/lib/thunk.S       | 30 -------------------
 include/asm-x86_64/rwlock.h   | 68 +++++++------------------------------------
 include/asm-x86_64/spinlock.h | 11 ++-----
 5 files changed, 51 insertions(+), 98 deletions(-)
 create mode 100644 arch/x86_64/lib/rwlock.S

diff --git a/arch/x86_64/lib/Makefile b/arch/x86_64/lib/Makefile
index ccef6ae747a3..b78d4170fce2 100644
--- a/arch/x86_64/lib/Makefile
+++ b/arch/x86_64/lib/Makefile
@@ -9,4 +9,4 @@ obj-y := io.o iomap_copy.o
 lib-y := csum-partial.o csum-copy.o csum-wrappers.o delay.o \
 	usercopy.o getuser.o putuser.o  \
 	thunk.o clear_page.o copy_page.o bitstr.o bitops.o
-lib-y += memcpy.o memmove.o memset.o copy_user.o
+lib-y += memcpy.o memmove.o memset.o copy_user.o rwlock.o
diff --git a/arch/x86_64/lib/rwlock.S b/arch/x86_64/lib/rwlock.S
new file mode 100644
index 000000000000..0cde1f807314
--- /dev/null
+++ b/arch/x86_64/lib/rwlock.S
@@ -0,0 +1,38 @@
+/* Slow paths of read/write spinlocks. */
+
+#include <linux/linkage.h>
+#include <asm/rwlock.h>
+#include <asm/alternative-asm.i>
+#include <asm/dwarf2.h>
+
+/* rdi:	pointer to rwlock_t */
+ENTRY(__write_lock_failed)
+	CFI_STARTPROC
+	LOCK_PREFIX
+	addl $RW_LOCK_BIAS,(%rdi)
+1:	rep
+	nop
+	cmpl $RW_LOCK_BIAS,(%rdi)
+	jne 1b
+	LOCK_PREFIX
+	subl $RW_LOCK_BIAS,(%rdi)
+	jnz  __write_lock_failed
+	ret
+	CFI_ENDPROC
+END(__write_lock_failed)
+
+/* rdi:	pointer to rwlock_t */
+ENTRY(__read_lock_failed)
+	CFI_STARTPROC
+	LOCK_PREFIX
+	incl (%rdi)
+1:	rep
+	nop
+	cmpl $1,(%rdi)
+	js 1b
+	LOCK_PREFIX
+	decl (%rdi)
+	js __read_lock_failed
+	ret
+	CFI_ENDPROC
+END(__read_lock_failed)
diff --git a/arch/x86_64/lib/thunk.S b/arch/x86_64/lib/thunk.S
index 332ea5dff916..6cff27c775ae 100644
--- a/arch/x86_64/lib/thunk.S
+++ b/arch/x86_64/lib/thunk.S
@@ -67,33 +67,3 @@ restore_norax:
 	RESTORE_ARGS 1
 	ret
 	CFI_ENDPROC
-
-#ifdef CONFIG_SMP
-/* Support for read/write spinlocks. */
-	.text
-/* rax:	pointer to rwlock_t */	
-ENTRY(__write_lock_failed)
-	lock
-	addl $RW_LOCK_BIAS,(%rax)
-1:	rep
-	nop
-	cmpl $RW_LOCK_BIAS,(%rax)
-	jne 1b
-	lock 
-	subl $RW_LOCK_BIAS,(%rax)
-	jnz  __write_lock_failed
-	ret
-
-/* rax:	pointer to rwlock_t */	
-ENTRY(__read_lock_failed)
-	lock
-	incl (%rax)
-1:	rep
-	nop
-	cmpl $1,(%rax)
-	js 1b
-	lock
-	decl (%rax)
-	js __read_lock_failed
-	ret
-#endif
diff --git a/include/asm-x86_64/rwlock.h b/include/asm-x86_64/rwlock.h
index dea0e9459264..28a080d23119 100644
--- a/include/asm-x86_64/rwlock.h
+++ b/include/asm-x86_64/rwlock.h
@@ -18,69 +18,21 @@
 #ifndef _ASM_X86_64_RWLOCK_H
 #define _ASM_X86_64_RWLOCK_H
 
-#include <linux/stringify.h>
-
 #define RW_LOCK_BIAS		 0x01000000
-#define RW_LOCK_BIAS_STR	"0x01000000"
+#define RW_LOCK_BIAS_STR	 "0x01000000"
 
-#define __build_read_lock_ptr(rw, helper)   \
+#define __build_read_lock(rw)   \
 	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t" \
-		     "js 2f\n" \
-		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tcall " helper "\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     ::"a" (rw) : "memory")
-
-#define __build_read_lock_const(rw, helper)   \
-	asm volatile(LOCK_PREFIX "subl $1,%0\n\t" \
-		     "js 2f\n" \
+		     "jns 1f\n" \
+		     "call __read_lock_failed\n" \
 		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tpushq %%rax\n\t" \
-		     "leaq %0,%%rax\n\t" \
-		     "call " helper "\n\t" \
-		     "popq %%rax\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     :"=m" (*((volatile int *)rw))::"memory")
-
-#define __build_read_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_read_lock_const(rw, helper); \
-						else \
-							__build_read_lock_ptr(rw, helper); \
-					} while (0)
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
 
-#define __build_write_lock_ptr(rw, helper) \
-	asm volatile(LOCK_PREFIX "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
-		     "jnz 2f\n" \
+#define __build_write_lock(rw) \
+	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t" \
+		     "jz 1f\n" \
+		     "\tcall __write_lock_failed\n\t" \
 		     "1:\n" \
-		     LOCK_SECTION_START("") \
-		     "2:\tcall " helper "\n\t" \
-		     "jmp 1b\n" \
-		     LOCK_SECTION_END \
-		     ::"a" (rw) : "memory")
-
-#define __build_write_lock_const(rw, helper) \
-	asm volatile(LOCK_PREFIX "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
-		     "jnz 2f\n" \
-		     "1:\n" \
-		    LOCK_SECTION_START("") \
-		     "2:\tpushq %%rax\n\t" \
-		     "leaq %0,%%rax\n\t" \
-		     "call " helper "\n\t" \
-		     "popq %%rax\n\t" \
-		     "jmp 1b\n" \
-		    LOCK_SECTION_END \
-		     :"=m" (*((volatile long *)rw))::"memory")
-
-#define __build_write_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_write_lock_const(rw, helper); \
-						else \
-							__build_write_lock_ptr(rw, helper); \
-					} while (0)
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
 
 #endif
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index 248a79f0eaff..a8e3d89f591f 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -79,13 +79,6 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  *
  * On x86, we implement read-write locks as a 32-bit counter
  * with the high bit (sign) being the "contended" bit.
- *
- * The inline assembly is non-obvious. Think about it.
- *
- * Changed to use the same technique as rw semaphores.  See
- * semaphore.h for details.  -ben
- *
- * the helpers are in arch/i386/kernel/semaphore.c
  */
 
 #define __raw_read_can_lock(x)		((int)(x)->lock > 0)
@@ -93,12 +86,12 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw, "__read_lock_failed");
+	__build_read_lock(rw);
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw, "__write_lock_failed");
+	__build_write_lock(rw);
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.1


From 1a015b5644ec6df0a2c4cbeff1f8a3d24ba0478e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Remove const case for rwlocks

rwlocks are now out of line, so it near never triggers.  Also it was
incompatible with the new dwarf2 unwinder because it had unannotiatable
push/pops.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/rwlock.h | 38 ++------------------------------------
 1 file changed, 2 insertions(+), 36 deletions(-)

diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index 87c069ccba08..f40ccbd8cb7f 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -20,52 +20,18 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	"0x01000000"
 
-#define __build_read_lock_ptr(rw, helper)   \
+#define __build_read_lock(rw, helper)   \
 	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" \
 			"jns 1f\n" \
 			"call " helper "\n\t" \
 			"1:\n" \
 			::"a" (rw) : "memory")
 
-#define __build_read_lock_const(rw, helper)   \
-	asm volatile(LOCK_PREFIX " subl $1,%0\n\t" \
-			"jns 1f\n" \
-			"pushl %%eax\n\t" \
-			"leal %0,%%eax\n\t" \
-			"call " helper "\n\t" \
-			"popl %%eax\n\t" \
-			"1:\n" \
-			:"+m" (*(volatile int *)rw) : : "memory")
-
-#define __build_read_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_read_lock_const(rw, helper); \
-						else \
-							__build_read_lock_ptr(rw, helper); \
-					} while (0)
-
-#define __build_write_lock_ptr(rw, helper) \
+#define __build_write_lock(rw, helper) \
 	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
 			"jz 1f\n" \
 			"call " helper "\n\t" \
 			"1:\n" \
 			::"a" (rw) : "memory")
 
-#define __build_write_lock_const(rw, helper) \
-	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
-			"jz 1f\n" \
-			"pushl %%eax\n\t" \
-			"leal %0,%%eax\n\t" \
-			"call " helper "\n\t" \
-			"popl %%eax\n\t" \
-			"1:\n" \
-			:"+m" (*(volatile int *)rw) : : "memory")
-
-#define __build_write_lock(rw, helper)	do { \
-						if (__builtin_constant_p(rw)) \
-							__build_write_lock_const(rw, helper); \
-						else \
-							__build_write_lock_ptr(rw, helper); \
-					} while (0)
-
 #endif
-- 
cgit v1.2.1


From b06babac45e1546dfb504f1f25eb0495632bfc41 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Add proper alignment to ENTRY

Previously it didn't align. Use the same one as the C compiler
in blended mode, which is good for K8 and Core2 and doesn't hurt
on P4.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S   | 3 +--
 include/asm-x86_64/linkage.h | 2 +-
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 7eb1678e098d..2dc5c01f754d 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -617,8 +617,7 @@ retint_signal:
 #ifdef CONFIG_PREEMPT
 	/* Returning to kernel space. Check if we need preemption */
 	/* rcx:	 threadinfo. interrupts off. */
-	.p2align
-retint_kernel:	
+ENTRY(retint_kernel)
 	cmpl $0,threadinfo_preempt_count(%rcx)
 	jnz  retint_restore_args
 	bt  $TIF_NEED_RESCHED,threadinfo_flags(%rcx)
diff --git a/include/asm-x86_64/linkage.h b/include/asm-x86_64/linkage.h
index 291c2d01c44f..b5f39d0189ce 100644
--- a/include/asm-x86_64/linkage.h
+++ b/include/asm-x86_64/linkage.h
@@ -1,6 +1,6 @@
 #ifndef __ASM_LINKAGE_H
 #define __ASM_LINKAGE_H
 
-/* Nothing to see here... */
+#define __ALIGN .p2align 4,,15
 
 #endif
-- 
cgit v1.2.1


From 07c9819b31eda7954feddc83f2fae035f31c11e1 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: add alternative-asm.h to allow LOCK_PREFIX replacement
 in .S files

LOCK_PREFIX is replaced by nops on UP systems, so it has to be a special
macro.  Previously this was only possible from C. Allow it for pure
assembly files too. Similar to earlier x86-64 patch.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/alternative-asm.i | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 include/asm-i386/alternative-asm.i

diff --git a/include/asm-i386/alternative-asm.i b/include/asm-i386/alternative-asm.i
new file mode 100644
index 000000000000..6c47e3b9484b
--- /dev/null
+++ b/include/asm-i386/alternative-asm.i
@@ -0,0 +1,14 @@
+#include <linux/config.h>
+
+#ifdef CONFIG_SMP
+	.macro LOCK_PREFIX
+1:	lock
+	.section .smp_locks,"a"
+	.align 4
+	.long 1b
+	.previous
+	.endm
+#else
+	.macro LOCK_PREFIX
+	.endm
+#endif
-- 
cgit v1.2.1


From ecaf45ee5ce60afe7cc46e91d82c1b0cbda09387 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Redo semaphore and rwlock assembly helpers

- Move them to a pure assembly file. Previously they were in
a C file that only consisted of inline assembly. Doing it in pure
assembler is much nicer.
- Add a frame.i include with FRAME/ENDFRAME macros to easily
add frame pointers to assembly functions
- Add dwarf2 annotation to them so that the new dwarf2 unwinder
doesn't get stuck on them
- Random cleanups

Includes feedback from Jan Beulich and a UML build fix from Andrew
Morton.

Cc: jbeulich@novell.com
Cc: jdike@addtoit.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/Makefile        |   2 +-
 arch/i386/kernel/semaphore.c     | 134 ----------------------------------
 arch/i386/lib/Makefile           |   2 +-
 arch/i386/lib/semaphore.S        | 154 +++++++++++++++++++++++++++++++++++++++
 arch/um/sys-i386/Makefile        |   2 +-
 include/asm-i386/frame.i         |  24 ++++++
 include/asm-um/alternative-asm.i |   6 ++
 include/asm-um/frame.i           |   6 ++
 8 files changed, 193 insertions(+), 137 deletions(-)
 delete mode 100644 arch/i386/kernel/semaphore.c
 create mode 100644 arch/i386/lib/semaphore.S
 create mode 100644 include/asm-i386/frame.i
 create mode 100644 include/asm-um/alternative-asm.i
 create mode 100644 include/asm-um/frame.i

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 5427a842e841..dab497472deb 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -4,7 +4,7 @@
 
 extra-y := head.o init_task.o vmlinux.lds
 
-obj-y	:= process.o semaphore.o signal.o entry.o traps.o irq.o \
+obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \
 		pci-dma.o i386_ksyms.o i387.o bootflag.o \
 		quirks.o i8237.o topology.o alternative.o i8253.o tsc.o
diff --git a/arch/i386/kernel/semaphore.c b/arch/i386/kernel/semaphore.c
deleted file mode 100644
index 98352c374c76..000000000000
--- a/arch/i386/kernel/semaphore.c
+++ /dev/null
@@ -1,134 +0,0 @@
-/*
- * i386 semaphore implementation.
- *
- * (C) Copyright 1999 Linus Torvalds
- *
- * Portions Copyright 1999 Red Hat, Inc.
- *
- *	This program is free software; you can redistribute it and/or
- *	modify it under the terms of the GNU General Public License
- *	as published by the Free Software Foundation; either version
- *	2 of the License, or (at your option) any later version.
- *
- * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
- */
-#include <asm/semaphore.h>
-
-/*
- * The semaphore operations have a special calling sequence that
- * allow us to do a simpler in-line version of them. These routines
- * need to convert that sequence back into the C sequence when
- * there is contention on the semaphore.
- *
- * %eax contains the semaphore pointer on entry. Save the C-clobbered
- * registers (%eax, %edx and %ecx) except %eax whish is either a return
- * value or just clobbered..
- */
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed\n"
-"__down_failed:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed_interruptible\n"
-"__down_failed_interruptible:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down_interruptible\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __down_failed_trylock\n"
-"__down_failed_trylock:\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"pushl %ebp\n\t"
-	"movl  %esp,%ebp\n\t"
-#endif
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __down_trylock\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-#if defined(CONFIG_FRAME_POINTER)
-	"movl %ebp,%esp\n\t"
-	"popl %ebp\n\t"
-#endif
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align 4\n"
-".globl __up_wakeup\n"
-"__up_wakeup:\n\t"
-	"pushl %edx\n\t"
-	"pushl %ecx\n\t"
-	"call __up\n\t"
-	"popl %ecx\n\t"
-	"popl %edx\n\t"
-	"ret"
-);
-
-/*
- * rw spinlock fallbacks
- */
-#if defined(CONFIG_SMP)
-asm(
-".section .sched.text\n"
-".align	4\n"
-".globl	__write_lock_failed\n"
-"__write_lock_failed:\n\t"
-	LOCK_PREFIX "addl	$" RW_LOCK_BIAS_STR ",(%eax)\n"
-"1:	rep; nop\n\t"
-	"cmpl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
-	"jne	1b\n\t"
-	LOCK_PREFIX "subl	$" RW_LOCK_BIAS_STR ",(%eax)\n\t"
-	"jnz	__write_lock_failed\n\t"
-	"ret"
-);
-
-asm(
-".section .sched.text\n"
-".align	4\n"
-".globl	__read_lock_failed\n"
-"__read_lock_failed:\n\t"
-	LOCK_PREFIX "incl	(%eax)\n"
-"1:	rep; nop\n\t"
-	"cmpl	$1,(%eax)\n\t"
-	"js	1b\n\t"
-	LOCK_PREFIX "decl	(%eax)\n\t"
-	"js	__read_lock_failed\n\t"
-	"ret"
-);
-#endif
diff --git a/arch/i386/lib/Makefile b/arch/i386/lib/Makefile
index 914933e9ec3d..d86a548b8d54 100644
--- a/arch/i386/lib/Makefile
+++ b/arch/i386/lib/Makefile
@@ -4,6 +4,6 @@
 
 
 lib-y = checksum.o delay.o usercopy.o getuser.o putuser.o memcpy.o strstr.o \
-	bitops.o
+	bitops.o semaphore.o
 
 lib-$(CONFIG_X86_USE_3DNOW) += mmx.o
diff --git a/arch/i386/lib/semaphore.S b/arch/i386/lib/semaphore.S
new file mode 100644
index 000000000000..e16ff1696390
--- /dev/null
+++ b/arch/i386/lib/semaphore.S
@@ -0,0 +1,154 @@
+/*
+ * i386 semaphore implementation.
+ *
+ * (C) Copyright 1999 Linus Torvalds
+ *
+ * Portions Copyright 1999 Red Hat, Inc.
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ *
+ * rw semaphores implemented November 1999 by Benjamin LaHaise <bcrl@kvack.org>
+ */
+
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/rwlock.h>
+#include <asm/alternative-asm.i>
+#include <asm/frame.i>
+#include <asm/dwarf2.h>
+
+/*
+ * The semaphore operations have a special calling sequence that
+ * allow us to do a simpler in-line version of them. These routines
+ * need to convert that sequence back into the C sequence when
+ * there is contention on the semaphore.
+ *
+ * %eax contains the semaphore pointer on entry. Save the C-clobbered
+ * registers (%eax, %edx and %ecx) except %eax whish is either a return
+ * value or just clobbered..
+ */
+	.section .sched.text
+ENTRY(__down_failed)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed)
+
+ENTRY(__down_failed_interruptible)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down_interruptible
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed_interruptible)
+
+ENTRY(__down_failed_trylock)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __down_trylock
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__down_failed_trylock)
+
+ENTRY(__up_wakeup)
+	CFI_STARTPROC
+	FRAME
+	pushl %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	pushl %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call __up
+	popl %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ecx
+	popl %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE edx
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__up_wakeup)
+
+/*
+ * rw spinlock fallbacks
+ */
+#ifdef CONFIG_SMP
+ENTRY(__write_lock_failed)
+	CFI_STARTPROC simple
+	FRAME
+2: 	LOCK_PREFIX
+	addl	$ RW_LOCK_BIAS,(%eax)
+1:	rep; nop
+	cmpl	$ RW_LOCK_BIAS,(%eax)
+	jne	1b
+	LOCK_PREFIX
+	subl	$ RW_LOCK_BIAS,(%eax)
+	jnz	2b
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__write_lock_failed)
+
+ENTRY(__read_lock_failed)
+	CFI_STARTPROC
+	FRAME
+2: 	LOCK_PREFIX
+	incl	(%eax)
+1:	rep; nop
+	cmpl	$1,(%eax)
+	js	1b
+	LOCK_PREFIX
+	decl	(%eax)
+	js	2b
+	ENDFRAME
+	ret
+	CFI_ENDPROC
+	END(__read_lock_failed)
+
+#endif
diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile
index 374d61a19439..6289779ca67c 100644
--- a/arch/um/sys-i386/Makefile
+++ b/arch/um/sys-i386/Makefile
@@ -4,7 +4,7 @@ obj-y = bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \
 
 obj-$(CONFIG_MODE_SKAS) += stub.o stub_segv.o
 
-subarch-obj-y = lib/bitops.o kernel/semaphore.o
+subarch-obj-y = lib/bitops.o lib/semaphore.o
 subarch-obj-$(CONFIG_HIGHMEM) += mm/highmem.o
 subarch-obj-$(CONFIG_MODULES) += kernel/module.o
 
diff --git a/include/asm-i386/frame.i b/include/asm-i386/frame.i
new file mode 100644
index 000000000000..4d68ddce18b6
--- /dev/null
+++ b/include/asm-i386/frame.i
@@ -0,0 +1,24 @@
+#include <linux/config.h>
+#include <asm/dwarf2.h>
+
+/* The annotation hides the frame from the unwinder and makes it look
+   like a ordinary ebp save/restore. This avoids some special cases for
+   frame pointer later */
+#ifdef CONFIG_FRAME_POINTER
+	.macro FRAME
+	pushl %ebp
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ebp,0
+	movl %esp,%ebp
+	.endm
+	.macro ENDFRAME
+	popl %ebp
+	CFI_ADJUST_CFA_OFFSET -4
+	CFI_RESTORE ebp
+	.endm
+#else
+	.macro FRAME
+	.endm
+	.macro ENDFRAME
+	.endm
+#endif
diff --git a/include/asm-um/alternative-asm.i b/include/asm-um/alternative-asm.i
new file mode 100644
index 000000000000..cae9faca132f
--- /dev/null
+++ b/include/asm-um/alternative-asm.i
@@ -0,0 +1,6 @@
+#ifndef __UM_ALTERNATIVE_ASM_I
+#define __UM_ALTERNATIVE_ASM_I
+
+#include "asm/arch/alternative-asm.i"
+
+#endif
diff --git a/include/asm-um/frame.i b/include/asm-um/frame.i
new file mode 100644
index 000000000000..09d5dca5d928
--- /dev/null
+++ b/include/asm-um/frame.i
@@ -0,0 +1,6 @@
+#ifndef __UM_FRAME_I
+#define __UM_FRAME_I
+
+#include "asm/arch/frame.i"
+
+#endif
-- 
cgit v1.2.1


From cc1e684a9f1cc069b2ac249aa11486613369dd69 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove leftover CVS Id in thunk.S

And move the comment to a proper place.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/lib/thunk.S | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/arch/x86_64/lib/thunk.S b/arch/x86_64/lib/thunk.S
index 6cff27c775ae..0025535cac8d 100644
--- a/arch/x86_64/lib/thunk.S
+++ b/arch/x86_64/lib/thunk.S
@@ -1,10 +1,9 @@
-	/*
-	 * Save registers before calling assembly functions. This avoids
-	 * disturbance of register allocation in some inline assembly constructs.
-	 * Copyright 2001,2002 by Andi Kleen, SuSE Labs.
-	 * Subject to the GNU public license, v.2. No warranty of any kind.
-	 * $Id: thunk.S,v 1.2 2002/03/13 20:06:58 ak Exp $
-	 */
+/*
+ * Save registers before calling assembly functions. This avoids
+ * disturbance of register allocation in some inline assembly constructs.
+ * Copyright 2001,2002 by Andi Kleen, SuSE Labs.
+ * Subject to the GNU public license, v.2. No warranty of any kind.
+ */
 
 	#include <linux/config.h>
 	#include <linux/linkage.h>
-- 
cgit v1.2.1


From 5ba5891d44a6acade44887a0f3195489d46c12de Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Add some comments what tce.c actually does

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/tce.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86_64/kernel/tce.c b/arch/x86_64/kernel/tce.c
index 5530dda3f27a..0905ce6b9b32 100644
--- a/arch/x86_64/kernel/tce.c
+++ b/arch/x86_64/kernel/tce.c
@@ -1,4 +1,6 @@
 /*
+ * This file manages the translation entries for the IBM Calgary IOMMU.
+ *
  * Derived from arch/powerpc/platforms/pseries/iommu.c
  *
  * Copyright (C) IBM Corporation, 2006
-- 
cgit v1.2.1


From 7f11d8a5efd625ffa41cde1d8472f85e885478ec Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove all ifdefs for local/io apic

IO-APIC or local APIC can only be disabled at runtime anyways and
Kconfig has forced these options on for a long time now.

The Kconfigs are kept only now for the benefit of the shared acpi
boot.c code.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/Makefile  | 4 ++--
 arch/x86_64/kernel/apic.c    | 6 +-----
 arch/x86_64/kernel/crash.c   | 2 --
 arch/x86_64/kernel/entry.S   | 2 --
 arch/x86_64/kernel/head64.c  | 2 --
 arch/x86_64/kernel/i8259.c   | 9 ---------
 arch/x86_64/kernel/irq.c     | 6 ------
 arch/x86_64/kernel/mpparse.c | 9 ---------
 arch/x86_64/kernel/setup.c   | 6 ------
 arch/x86_64/kernel/smpboot.c | 5 -----
 arch/x86_64/kernel/time.c    | 8 --------
 arch/x86_64/kernel/traps.c   | 2 --
 include/asm-x86_64/apic.h    | 4 ----
 include/asm-x86_64/fixmap.h  | 4 ----
 include/asm-x86_64/io_apic.h | 6 ------
 include/asm-x86_64/irq.h     | 2 --
 include/asm-x86_64/mpspec.h  | 2 --
 include/asm-x86_64/smp.h     | 4 ----
 18 files changed, 3 insertions(+), 80 deletions(-)

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index b5aaeafc1cd3..0ef39553c0f6 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -20,8 +20,8 @@ obj-$(CONFIG_X86_MSR)		+= msr.o
 obj-$(CONFIG_MICROCODE)		+= microcode.o
 obj-$(CONFIG_X86_CPUID)		+= cpuid.o
 obj-$(CONFIG_SMP)		+= smp.o smpboot.o trampoline.o
-obj-$(CONFIG_X86_LOCAL_APIC)	+= apic.o  nmi.o
-obj-$(CONFIG_X86_IO_APIC)	+= io_apic.o mpparse.o \
+obj-y				+= apic.o  nmi.o
+obj-y				+= io_apic.o mpparse.o \
 		genapic.o genapic_cluster.o genapic_flat.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o crash.o
 obj-$(CONFIG_CRASH_DUMP)	+= crash_dump.o
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 63d9b037afc6..c6ca100a7dea 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -661,7 +661,6 @@ void __init init_apic_mappings(void)
 	 */
 	boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
 
-#ifdef CONFIG_X86_IO_APIC
 	{
 		unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
 		int i;
@@ -679,7 +678,6 @@ void __init init_apic_mappings(void)
 			idx++;
 		}
 	}
-#endif
 }
 
 /*
@@ -1129,12 +1127,10 @@ int __init APIC_init_uniprocessor (void)
 
 	setup_local_APIC();
 
-#ifdef CONFIG_X86_IO_APIC
 	if (smp_found_config && !skip_ioapic_setup && nr_ioapics)
-			setup_IO_APIC();
+		setup_IO_APIC();
 	else
 		nr_ioapics = 0;
-#endif
 	setup_boot_APIC_clock();
 	check_nmi_watchdog();
 	return 0;
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index 7d7a9e75f70c..3525f884af82 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -194,9 +194,7 @@ void machine_crash_shutdown(struct pt_regs *regs)
 	if(cpu_has_apic)
 		 disable_local_APIC();
 
-#if defined(CONFIG_X86_IO_APIC)
 	disable_IO_APIC();
-#endif
 
 	crash_save_self(regs);
 }
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 2dc5c01f754d..a0cf36ba7c56 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -676,7 +676,6 @@ ENTRY(call_function_interrupt)
 END(call_function_interrupt)
 #endif
 
-#ifdef CONFIG_X86_LOCAL_APIC	
 ENTRY(apic_timer_interrupt)
 	apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
 END(apic_timer_interrupt)
@@ -688,7 +687,6 @@ END(error_interrupt)
 ENTRY(spurious_interrupt)
 	apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
 END(spurious_interrupt)
-#endif
 				
 /*
  * Exception entry points.
diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index 36647ce6aecb..bacbd75c63ad 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -111,10 +111,8 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	if (s != NULL)
 		numa_setup(s+5);
 #endif
-#ifdef CONFIG_X86_IO_APIC
 	if (strstr(saved_command_line, "disableapic"))
 		disable_apic = 1;
-#endif
 	/* You need early console to see that */
 	if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
 		panic("Kernel too big for kernel mapping\n");
diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c
index 0434b1f8e3dd..d2f5acde5749 100644
--- a/arch/x86_64/kernel/i8259.c
+++ b/arch/x86_64/kernel/i8259.c
@@ -55,7 +55,6 @@
  */
 BUILD_16_IRQS(0x0)
 
-#ifdef CONFIG_X86_LOCAL_APIC
 /*
  * The IO-APIC gives us many more interrupt sources. Most of these 
  * are unused but an SMP system is supposed to have enough memory ...
@@ -75,8 +74,6 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd)
 	BUILD_15_IRQS(0xe)
 #endif
 
-#endif
-
 #undef BUILD_16_IRQS
 #undef BUILD_15_IRQS
 #undef BI
@@ -100,7 +97,6 @@ BUILD_16_IRQS(0xc) BUILD_16_IRQS(0xd)
 void (*interrupt[NR_IRQS])(void) = {
 	IRQLIST_16(0x0),
 
-#ifdef CONFIG_X86_IO_APIC
 			 IRQLIST_16(0x1), IRQLIST_16(0x2), IRQLIST_16(0x3),
 	IRQLIST_16(0x4), IRQLIST_16(0x5), IRQLIST_16(0x6), IRQLIST_16(0x7),
 	IRQLIST_16(0x8), IRQLIST_16(0x9), IRQLIST_16(0xa), IRQLIST_16(0xb),
@@ -110,7 +106,6 @@ void (*interrupt[NR_IRQS])(void) = {
 	, IRQLIST_15(0xe)
 #endif
 
-#endif
 };
 
 #undef IRQ
@@ -453,9 +448,7 @@ void __init init_ISA_irqs (void)
 {
 	int i;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	init_bsp_APIC();
-#endif
 	init_8259A(0);
 
 	for (i = 0; i < NR_IRQS; i++) {
@@ -581,14 +574,12 @@ void __init init_IRQ(void)
 	set_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt);
 	set_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt);
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	/* self generated IPI for local APIC timer */
 	set_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);
 
 	/* IPI vectors for APIC spurious and error interrupts */
 	set_intr_gate(SPURIOUS_APIC_VECTOR, spurious_interrupt);
 	set_intr_gate(ERROR_APIC_VECTOR, error_interrupt);
-#endif
 
 	/*
 	 * Set the clock to HZ Hz, we already have a valid
diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index 5221a53e90c1..bfbcf92cfa4a 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -20,11 +20,9 @@
 #include <asm/idle.h>
 
 atomic_t irq_err_count;
-#ifdef CONFIG_X86_IO_APIC
 #ifdef APIC_MISMATCH_DEBUG
 atomic_t irq_mis_count;
 #endif
-#endif
 
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 /*
@@ -92,17 +90,13 @@ skip:
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
 		seq_putc(p, '\n');
-#ifdef CONFIG_X86_LOCAL_APIC
 		seq_printf(p, "LOC: ");
 		for_each_online_cpu(j)
 			seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
 		seq_putc(p, '\n');
-#endif
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
-#ifdef CONFIG_X86_IO_APIC
 #ifdef APIC_MISMATCH_DEBUG
 		seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
-#endif
 #endif
 	}
 	return 0;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index a1ab4197f8a1..c73dd1f41e65 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -74,14 +74,10 @@ physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
 /* ACPI MADT entry parsing functions */
 #ifdef CONFIG_ACPI
 extern struct acpi_boot_flags acpi_boot;
-#ifdef CONFIG_X86_LOCAL_APIC
 extern int acpi_parse_lapic (acpi_table_entry_header *header);
 extern int acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header);
 extern int acpi_parse_lapic_nmi (acpi_table_entry_header *header);
-#endif /*CONFIG_X86_LOCAL_APIC*/
-#ifdef CONFIG_X86_IO_APIC
 extern int acpi_parse_ioapic (acpi_table_entry_header *header);
-#endif /*CONFIG_X86_IO_APIC*/
 #endif /*CONFIG_ACPI*/
 
 u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
@@ -661,9 +657,7 @@ void __init find_intel_smp (void)
  */
 void __init find_smp_config (void)
 {
-#ifdef CONFIG_X86_LOCAL_APIC
 	find_intel_smp();
-#endif
 }
 
 
@@ -717,8 +711,6 @@ void __cpuinit mp_register_lapic (
 	MP_processor_info(&processor);
 }
 
-#ifdef CONFIG_X86_IO_APIC
-
 #define MP_ISA_BUS		0
 #define MP_MAX_IOAPIC_PIN	127
 
@@ -997,5 +989,4 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity)
 	return gsi;
 }
 
-#endif /*CONFIG_X86_IO_APIC*/
 #endif /*CONFIG_ACPI*/
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 34afad704824..fbe9f7faa2db 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -340,10 +340,8 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 		else if (fullarg(from, "acpi=strict")) {
 			acpi_strict = 1;
 		}
-#ifdef CONFIG_X86_IO_APIC
 		else if (fullarg(from, "acpi_skip_timer_override"))
 			acpi_skip_timer_override = 1;
-#endif
 #endif
 
 		if (fullarg(from, "disable_timer_pin_1"))
@@ -625,12 +623,10 @@ void __init setup_arch(char **cmdline_p)
         */
        acpi_reserve_bootmem();
 #endif
-#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * Find and reserve possible boot-time SMP configuration:
 	 */
 	find_smp_config();
-#endif
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (LOADER_TYPE && INITRD_START) {
 		if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
@@ -674,14 +670,12 @@ void __init setup_arch(char **cmdline_p)
 
 	init_cpu_to_node();
 
-#ifdef CONFIG_X86_LOCAL_APIC
 	/*
 	 * get boot-time SMP configuration:
 	 */
 	if (smp_found_config)
 		get_smp_config();
 	init_apic_mappings();
-#endif
 
 	/*
 	 * Request address space for all standard RAM and ROM resources
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 582896f7d420..d29571e24953 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1175,13 +1175,8 @@ int __cpuinit __cpu_up(unsigned int cpu)
 void __init smp_cpus_done(unsigned int max_cpus)
 {
 	smp_cleanup_boot();
-
-#ifdef CONFIG_X86_IO_APIC
 	setup_ioapic_dest();
-#endif
-
 	check_nmi_watchdog();
-
 	time_init_gtod();
 }
 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index ea00915d393a..d66c7f750e75 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -41,9 +41,7 @@
 #include <asm/sections.h>
 #include <linux/cpufreq.h>
 #include <linux/hpet.h>
-#ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/apic.h>
-#endif
 
 #ifdef CONFIG_CPU_FREQ
 static void cpufreq_delayed_get(void);
@@ -438,12 +436,8 @@ void main_timer_handler(struct pt_regs *regs)
  * have to call the local interrupt handler.
  */
 
-#ifndef CONFIG_X86_LOCAL_APIC
-	profile_tick(CPU_PROFILING, regs);
-#else
 	if (!using_apic_timer)
 		smp_local_timer_interrupt(regs);
-#endif
 
 /*
  * If we have an externally synchronized Linux clock, then update CMOS clock
@@ -467,10 +461,8 @@ static irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	if (apic_runs_main_timer > 1)
 		return IRQ_HANDLED;
 	main_timer_handler(regs);
-#ifdef CONFIG_X86_LOCAL_APIC
 	if (using_apic_timer)
 		smp_send_timer_broadcast_ipi();
-#endif
 	return IRQ_HANDLED;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 96f62a033242..56d7ff0c894c 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -788,7 +788,6 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
 								== NOTIFY_STOP)
 			return;
-#ifdef CONFIG_X86_LOCAL_APIC
 		/*
 		 * Ok, so this is none of the documented NMI sources,
 		 * so it must be the NMI watchdog.
@@ -796,7 +795,6 @@ asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
 		if (nmi_watchdog_tick(regs,reason))
 			return;
 		if (!do_nmi_callback(regs,cpu))
-#endif
 			unknown_nmi_error(reason, regs);
 
 		return;
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 9c96a0a8d1bd..8ed0f4d67b8d 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -29,8 +29,6 @@ extern int apic_runs_main_timer;
 			printk(s, ##a);    \
 	} while (0)
 
-#ifdef CONFIG_X86_LOCAL_APIC
-
 struct pt_regs;
 
 /*
@@ -104,8 +102,6 @@ void switch_ipi_to_APIC_timer(void *cpumask);
 
 #define ARCH_APICTIMER_STOPS_ON_C3	1
 
-#endif /* CONFIG_X86_LOCAL_APIC */
-
 extern unsigned boot_cpu_id;
 
 #endif /* __ASM_APIC_H */
diff --git a/include/asm-x86_64/fixmap.h b/include/asm-x86_64/fixmap.h
index 0b4ffbd1a125..1b620db5b9e3 100644
--- a/include/asm-x86_64/fixmap.h
+++ b/include/asm-x86_64/fixmap.h
@@ -37,13 +37,9 @@ enum fixed_addresses {
 	VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
 	VSYSCALL_HPET,
 	FIX_HPET_BASE,
-#ifdef CONFIG_X86_LOCAL_APIC
 	FIX_APIC_BASE,	/* local (CPU) APIC) -- required for SMP or not */
-#endif
-#ifdef CONFIG_X86_IO_APIC
 	FIX_IO_APIC_BASE_0,
 	FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
-#endif
 	__end_of_fixed_addresses
 };
 
diff --git a/include/asm-x86_64/io_apic.h b/include/asm-x86_64/io_apic.h
index fb7a0909a174..5d1b5c68e36e 100644
--- a/include/asm-x86_64/io_apic.h
+++ b/include/asm-x86_64/io_apic.h
@@ -10,8 +10,6 @@
  * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar
  */
 
-#ifdef CONFIG_X86_IO_APIC
-
 #ifdef CONFIG_PCI_MSI
 static inline int use_pci_vector(void)	{return 1;}
 static inline void disable_edge_ioapic_vector(unsigned int vector) { }
@@ -209,10 +207,6 @@ extern int timer_uses_ioapic_pin_0;
 
 extern int sis_apic_bug; /* dummy */ 
 
-#else  /* !CONFIG_X86_IO_APIC */
-#define io_apic_assign_pci_irqs 0
-#endif
-
 extern int assign_irq_vector(int irq);
 
 void enable_NMI_through_LVT0 (void * dummy);
diff --git a/include/asm-x86_64/irq.h b/include/asm-x86_64/irq.h
index 9db5a1b4f7b1..43469d8ab71a 100644
--- a/include/asm-x86_64/irq.h
+++ b/include/asm-x86_64/irq.h
@@ -44,9 +44,7 @@ static __inline__ int irq_canonicalize(int irq)
 	return ((irq == 2) ? 9 : irq);
 }
 
-#ifdef CONFIG_X86_LOCAL_APIC
 #define ARCH_HAS_NMI_WATCHDOG		/* See include/linux/nmi.h */
-#endif
 
 #ifdef CONFIG_HOTPLUG_CPU
 #include <linux/cpumask.h>
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 14fc3ddd9031..3c7aae9fd157 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -184,12 +184,10 @@ extern int pic_mode;
 extern void mp_register_lapic (u8 id, u8 enabled);
 extern void mp_register_lapic_address (u64 address);
 
-#ifdef CONFIG_X86_IO_APIC
 extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
 extern void mp_override_legacy_irq (u8 bus_irq, u8 polarity, u8 trigger, u32 gsi);
 extern void mp_config_acpi_legacy_irqs (void);
 extern int mp_register_gsi (u32 gsi, int triggering, int polarity);
-#endif /*CONFIG_X86_IO_APIC*/
 #endif
 
 extern int using_apic_timer;
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 612d208961a2..6353fa41bebe 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -9,15 +9,11 @@
 #include <linux/bitops.h>
 extern int disable_apic;
 
-#ifdef CONFIG_X86_LOCAL_APIC
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
-#ifdef CONFIG_X86_IO_APIC
 #include <asm/io_apic.h>
-#endif
 #include <asm/apic.h>
 #include <asm/thread_info.h>
-#endif
 
 #ifdef CONFIG_SMP
 
-- 
cgit v1.2.1


From 6c96a29f20762e8ce40b674f906055d009c302ee Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove apic mismatch counter

Nobody has been setting the mismatch counter and the ifdef was never
set so remove it.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/irq.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index bfbcf92cfa4a..b3677e6ccc6e 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -20,9 +20,6 @@
 #include <asm/idle.h>
 
 atomic_t irq_err_count;
-#ifdef APIC_MISMATCH_DEBUG
-atomic_t irq_mis_count;
-#endif
 
 #ifdef CONFIG_DEBUG_STACKOVERFLOW
 /*
@@ -95,9 +92,6 @@ skip:
 			seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
 		seq_putc(p, '\n');
 		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
-#ifdef APIC_MISMATCH_DEBUG
-		seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
-#endif
 	}
 	return 0;
 }
-- 
cgit v1.2.1


From 3f14c746a61ec932c204aca820c02c293118c5df Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove old "focus disabled" chipset errata workaround

The new systems already use focus disabled and the comment was
completely outdated.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c | 28 ++--------------------------
 1 file changed, 2 insertions(+), 26 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index c6ca100a7dea..78c43b1a3064 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -399,32 +399,8 @@ void __cpuinit setup_local_APIC (void)
 	 */
 	value |= APIC_SPIV_APIC_ENABLED;
 
-	/*
-	 * Some unknown Intel IO/APIC (or APIC) errata are biting us with
-	 * certain networking cards. If high frequency interrupts are
-	 * happening on a particular IOAPIC pin, plus the IOAPIC routing
-	 * entry is masked/unmasked at a high rate as well then sooner or
-	 * later IOAPIC line gets 'stuck', no more interrupts are received
-	 * from the device. If focus CPU is disabled then the hang goes
-	 * away, oh well :-(
-	 *
-	 * [ This bug can be reproduced easily with a level-triggered
-	 *   PCI Ne2000 networking cards and PII/PIII processors, dual
-	 *   BX chipset. ]
-	 */
-	/*
-	 * Actually disabling the focus CPU check just makes the hang less
-	 * frequent as it makes the interrupt distributon model be more
-	 * like LRU than MRU (the short-term load is more even across CPUs).
-	 * See also the comment in end_level_ioapic_irq().  --macro
-	 */
-#if 1
-	/* Enable focus processor (bit==0) */
-	value &= ~APIC_SPIV_FOCUS_DISABLED;
-#else
-	/* Disable focus processor (bit==1) */
-	value |= APIC_SPIV_FOCUS_DISABLED;
-#endif
+	/* We always use processor focus */
+
 	/*
 	 * Set spurious IRQ vector
 	 */
-- 
cgit v1.2.1


From b1c78c0fcc29097567e1afc39701012e6d89adb7 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Clean up and minor fixes to TLB flush

- Convert CR* accesses to dedicated inline functions and rewrite
the rest as C inlines
- Don't do a double flush for global flushes (pointed out by Zach Amsden)
This was a bug workaround for old CPUs that don't do 64bit and is obsolete.
- Add a proper memory clobber to invlpg
- Remove an unused extern

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pgtable.h  |  2 --
 include/asm-x86_64/tlbflush.h | 70 +++++++++++++++++++++----------------------
 2 files changed, 35 insertions(+), 37 deletions(-)

diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index a31ab4e68a9b..0c1e2422400a 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -25,8 +25,6 @@ extern int nonx_setup(char *str);
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
-extern unsigned long pgkern_mask;
-
 /*
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..
diff --git a/include/asm-x86_64/tlbflush.h b/include/asm-x86_64/tlbflush.h
index d16d5b60f419..983bd296c81a 100644
--- a/include/asm-x86_64/tlbflush.h
+++ b/include/asm-x86_64/tlbflush.h
@@ -4,44 +4,44 @@
 #include <linux/mm.h>
 #include <asm/processor.h>
 
-#define __flush_tlb()							\
-	do {								\
-		unsigned long tmpreg;					\
-									\
-		__asm__ __volatile__(					\
-			"movq %%cr3, %0;  # flush TLB \n"		\
-			"movq %0, %%cr3;              \n"		\
-			: "=r" (tmpreg)					\
-			:: "memory");					\
-	} while (0)
+static inline unsigned long get_cr3(void)
+{
+	unsigned long cr3;
+	asm volatile("mov %%cr3,%0" : "=r" (cr3));
+	return cr3;
+}
 
-/*
- * Global pages have to be flushed a bit differently. Not a real
- * performance problem because this does not happen often.
- */
-#define __flush_tlb_global()						\
-	do {								\
-		unsigned long tmpreg, cr4, cr4_orig;			\
-									\
-		__asm__ __volatile__(					\
-			"movq %%cr4, %2;  # turn off PGE     \n"	\
-			"movq %2, %1;                        \n"	\
-			"andq %3, %1;                        \n"	\
-			"movq %1, %%cr4;                     \n"	\
-			"movq %%cr3, %0;  # flush TLB        \n"	\
-			"movq %0, %%cr3;                     \n"	\
-			"movq %2, %%cr4;  # turn PGE back on \n"	\
-			: "=&r" (tmpreg), "=&r" (cr4), "=&r" (cr4_orig)	\
-			: "i" (~X86_CR4_PGE)				\
-			: "memory");					\
-	} while (0)
-
-extern unsigned long pgkern_mask;
-
-#define __flush_tlb_all() __flush_tlb_global()
+static inline void set_cr3(unsigned long cr3)
+{
+	asm volatile("mov %0,%%cr3" :: "r" (cr3) : "memory");
+}
+
+static inline void __flush_tlb(void)
+{
+	set_cr3(get_cr3());
+}
+
+static inline unsigned long get_cr4(void)
+{
+	unsigned long cr4;
+	asm volatile("mov %%cr4,%0" : "=r" (cr4));
+	return cr4;
+}
+
+static inline void set_cr4(unsigned long cr4)
+{
+	asm volatile("mov %0,%%cr4" :: "r" (cr4) : "memory");
+}
+
+static inline void __flush_tlb_all(void)
+{
+	unsigned long cr4 = get_cr4();
+	set_cr4(cr4 & ~X86_CR4_PGE);	/* clear PGE */
+	set_cr4(cr4);			/* write old PGE again and flush TLBs */
+}
 
 #define __flush_tlb_one(addr) \
-	__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
+	__asm__ __volatile__("invlpg (%0)" :: "r" (addr) : "memory")
 
 
 /*
-- 
cgit v1.2.1


From 107878bb14b1fb6bddc646f4d5e72e8beaa2f6a2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] i386: Minor fixes & cleanup to tlb flush

(based on x86-64 changes)
- Add a proper memory clobber to invlpg
- Remove an unused extern

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/tlbflush.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/include/asm-i386/tlbflush.h b/include/asm-i386/tlbflush.h
index d57ca5c540b6..360648b0f2b3 100644
--- a/include/asm-i386/tlbflush.h
+++ b/include/asm-i386/tlbflush.h
@@ -36,8 +36,6 @@
 			: "memory");					\
 	} while (0)
 
-extern unsigned long pgkern_mask;
-
 # define __flush_tlb_all()						\
 	do {								\
 		if (cpu_has_pge)					\
@@ -49,7 +47,7 @@ extern unsigned long pgkern_mask;
 #define cpu_has_invlpg	(boot_cpu_data.x86 > 3)
 
 #define __flush_tlb_single(addr) \
-	__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
+	__asm__ __volatile__("invlpg (%0)" ::"r" (addr) : "memory")
 
 #ifdef CONFIG_X86_INVLPG
 # define __flush_tlb_one(addr) __flush_tlb_single(addr)
-- 
cgit v1.2.1


From 2e91a17b35116885373e04af142b1d08cf1b47bf Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Add some comments to entry.S

And remove some old obsolete ones.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index a0cf36ba7c56..a04fc4108ff2 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -22,9 +22,21 @@
  * at the top of the kernel process stack.	
  * - partial stack frame: partially saved registers upto R11.
  * - full stack frame: Like partial stack frame, but all register saved. 
- *	
- * TODO:	 
- * - schedule it carefully for the final hardware.
+ *
+ * Some macro usage:
+ * - CFI macros are used to generate dwarf2 unwind information for better
+ * backtraces. They don't change any code.
+ * - SAVE_ALL/RESTORE_ALL - Save/restore all registers
+ * - SAVE_ARGS/RESTORE_ARGS - Save/restore registers that C functions modify.
+ * There are unfortunately lots of special cases where some registers
+ * not touched. The macro is a big mess that should be cleaned up.
+ * - SAVE_REST/RESTORE_REST - Handle the registers not saved by SAVE_ARGS.
+ * Gives a full stack frame.
+ * - ENTRY/END Define functions in the symbol table.
+ * - FIXUP_TOP_OF_STACK/RESTORE_TOP_OF_STACK - Fix up the hardware stack
+ * frame that is otherwise undefined after a SYSCALL
+ * - TRACE_IRQ_* - Trace hard interrupt state for lock debugging.
+ * - errorentry/paranoidentry/zeroentry - Define exception entry points.
  */
 
 #include <linux/linkage.h>
-- 
cgit v1.2.1


From 5cb6b99928a8f1a1bbdac7a16b88aef8cb64d432 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove pirq overwrite support

This was an old workaround for broken MP-BIOS. The user could
specify overwrites on the command line.

I've never seen it being used for anything on 64bit. So get
rid of it for now.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 55 --------------------------------------------
 1 file changed, 55 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 924a4a332954..86cc07caa0e5 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -225,14 +225,6 @@ static void clear_IO_APIC (void)
 			clear_IO_APIC_pin(apic, pin);
 }
 
-/*
- * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to
- * specific CPU-side IRQs.
- */
-
-#define MAX_PIRQS 8
-static int pirq_entries [MAX_PIRQS];
-static int pirqs_enabled;
 int skip_ioapic_setup;
 int ioapic_force;
 
@@ -370,34 +362,6 @@ void __init check_ioapic(void)
 	}
 } 
 
-static int __init ioapic_pirq_setup(char *str)
-{
-	int i, max;
-	int ints[MAX_PIRQS+1];
-
-	get_options(str, ARRAY_SIZE(ints), ints);
-
-	for (i = 0; i < MAX_PIRQS; i++)
-		pirq_entries[i] = -1;
-
-	pirqs_enabled = 1;
-	apic_printk(APIC_VERBOSE, "PIRQ redirection, working around broken MP-BIOS.\n");
-	max = MAX_PIRQS;
-	if (ints[0] < MAX_PIRQS)
-		max = ints[0];
-
-	for (i = 0; i < max; i++) {
-		apic_printk(APIC_VERBOSE, "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
-		/*
-		 * PIRQs are mapped upside down, usually.
-		 */
-		pirq_entries[MAX_PIRQS-i-1] = ints[i+1];
-	}
-	return 1;
-}
-
-__setup("pirq=", ioapic_pirq_setup);
-
 /*
  * Find the IRQ entry number of a certain pin.
  */
@@ -793,22 +757,6 @@ static int pin_2_irq(int idx, int apic, int pin)
 		}
 	}
 	BUG_ON(irq >= NR_IRQS);
-
-	/*
-	 * PCI IRQ command line redirection. Yes, limits are hardcoded.
-	 */
-	if ((pin >= 16) && (pin <= 23)) {
-		if (pirq_entries[pin-16] != -1) {
-			if (!pirq_entries[pin-16]) {
-				apic_printk(APIC_VERBOSE, "disabling PIRQ%d\n", pin-16);
-			} else {
-				irq = pirq_entries[pin-16];
-				apic_printk(APIC_VERBOSE, "using PIRQ%d -> IRQ %d\n",
-						pin-16, irq);
-			}
-		}
-	}
-	BUG_ON(irq >= NR_IRQS);
 	return irq;
 }
 
@@ -1281,9 +1229,6 @@ static void __init enable_IO_APIC(void)
 		irq_2_pin[i].pin = -1;
 		irq_2_pin[i].next = 0;
 	}
-	if (!pirqs_enabled)
-		for (i = 0; i < MAX_PIRQS; i++)
-			pirq_entries[i] = -1;
 
 	/*
 	 * The number of IO-APIC IRQ registers (== #pins):
-- 
cgit v1.2.1


From e50991343488edf25d58820be1684322808c763c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:29 +0200
Subject: [PATCH] Remove leftover MCE/EISA support

No 64bit EISA or Microchannel systems ever. Remove the left over code
in the IO-APIC driver and the mptable parser

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 57 ++------------------------------------------
 arch/x86_64/kernel/mpparse.c | 16 +------------
 2 files changed, 3 insertions(+), 70 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 86cc07caa0e5..97e90f6abcc4 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -389,9 +389,7 @@ static int __init find_isa_irq_pin(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
-		     mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
-		     mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
+		if (mp_bus_id_to_type[lbus] == MP_BUS_ISA &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 
@@ -407,9 +405,7 @@ static int __init find_isa_irq_apic(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
-		     mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
-		     mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
+		if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA) &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 			break;
@@ -472,27 +468,6 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
 	return best_guess;
 }
 
-/*
- * EISA Edge/Level control register, ELCR
- */
-static int EISA_ELCR(unsigned int irq)
-{
-	if (irq < 16) {
-		unsigned int port = 0x4d0 + (irq >> 3);
-		return (inb(port) >> (irq & 7)) & 1;
-	}
-	apic_printk(APIC_VERBOSE, "Broken MPtable reports ISA irq %d\n", irq);
-	return 0;
-}
-
-/* EISA interrupts are always polarity zero and can be edge or level
- * trigger depending on the ELCR value.  If an interrupt is listed as
- * EISA conforming in the MP table, that means its trigger type must
- * be read in from the ELCR */
-
-#define default_EISA_trigger(idx)	(EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
-#define default_EISA_polarity(idx)	(0)
-
 /* ISA interrupts are always polarity zero edge triggered,
  * when listed as conforming in the MP table. */
 
@@ -505,12 +480,6 @@ static int EISA_ELCR(unsigned int irq)
 #define default_PCI_trigger(idx)	(1)
 #define default_PCI_polarity(idx)	(1)
 
-/* MCA interrupts are always polarity zero level triggered,
- * when listed as conforming in the MP table. */
-
-#define default_MCA_trigger(idx)	(1)
-#define default_MCA_polarity(idx)	(0)
-
 static int __init MPBIOS_polarity(int idx)
 {
 	int bus = mp_irqs[idx].mpc_srcbus;
@@ -530,21 +499,11 @@ static int __init MPBIOS_polarity(int idx)
 					polarity = default_ISA_polarity(idx);
 					break;
 				}
-				case MP_BUS_EISA: /* EISA pin */
-				{
-					polarity = default_EISA_polarity(idx);
-					break;
-				}
 				case MP_BUS_PCI: /* PCI pin */
 				{
 					polarity = default_PCI_polarity(idx);
 					break;
 				}
-				case MP_BUS_MCA: /* MCA pin */
-				{
-					polarity = default_MCA_polarity(idx);
-					break;
-				}
 				default:
 				{
 					printk(KERN_WARNING "broken BIOS!!\n");
@@ -599,21 +558,11 @@ static int MPBIOS_trigger(int idx)
 					trigger = default_ISA_trigger(idx);
 					break;
 				}
-				case MP_BUS_EISA: /* EISA pin */
-				{
-					trigger = default_EISA_trigger(idx);
-					break;
-				}
 				case MP_BUS_PCI: /* PCI pin */
 				{
 					trigger = default_PCI_trigger(idx);
 					break;
 				}
-				case MP_BUS_MCA: /* MCA pin */
-				{
-					trigger = default_MCA_trigger(idx);
-					break;
-				}
 				default:
 				{
 					printk(KERN_WARNING "broken BIOS!!\n");
@@ -731,8 +680,6 @@ static int pin_2_irq(int idx, int apic, int pin)
 	switch (mp_bus_id_to_type[bus])
 	{
 		case MP_BUS_ISA: /* ISA pin */
-		case MP_BUS_EISA:
-		case MP_BUS_MCA:
 		{
 			irq = mp_irqs[idx].mpc_srcbusirq;
 			break;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index c73dd1f41e65..90e99cf27aa5 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -175,14 +175,10 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 
 	if (strncmp(str, "ISA", 3) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
-	} else if (strncmp(str, "EISA", 4) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
 	} else if (strncmp(str, "PCI", 3) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
-	} else if (strncmp(str, "MCA", 3) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
 	} else {
 		printk(KERN_ERR "Unknown bustype %s\n", str);
 	}
@@ -465,14 +461,6 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 		case 5:
 			memcpy(bus.mpc_bustype, "ISA   ", 6);
 			break;
-		case 2:
-		case 6:
-		case 3:
-			memcpy(bus.mpc_bustype, "EISA  ", 6);
-			break;
-		case 4:
-		case 7:
-			memcpy(bus.mpc_bustype, "MCA   ", 6);
 	}
 	MP_bus_info(&bus);
 	if (mpc_default_type > 4) {
@@ -629,9 +617,7 @@ void __init find_intel_smp (void)
 			smp_scan_config(0xF0000,0x10000))
 		return;
 	/*
-	 * If it is an SMP machine we should know now, unless the
-	 * configuration is in an EISA/MCA bus machine with an
-	 * extended bios data area.
+	 * If it is an SMP machine we should know now.
 	 *
 	 * there is a real-mode segmented pointer pointing to the
 	 * 4K EBDA area at 0x40E, calculate and scan it here.
-- 
cgit v1.2.1


From a8fcf1a24a16e1c735c795d99773b5dcefb71518 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Remove obsolete PIC mode

PIC mode is an outdated way to drive the APICs that was used on
some early MP boards. It is not supported in the ACPI model.

It is unlikely to be ever configured by any x86-64 system

Remove it thus.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c    | 92 ++++++++++++++------------------------------
 arch/x86_64/kernel/mpparse.c |  8 ----
 arch/x86_64/kernel/smpboot.c |  1 -
 include/asm-x86_64/mpspec.h  |  1 -
 include/asm-x86_64/smp.h     |  1 -
 5 files changed, 29 insertions(+), 74 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 78c43b1a3064..20a76f209629 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -136,72 +136,40 @@ void clear_local_APIC(void)
 	apic_read(APIC_ESR);
 }
 
-void __init connect_bsp_APIC(void)
-{
-	if (pic_mode) {
-		/*
-		 * Do not trust the local APIC being empty at bootup.
-		 */
-		clear_local_APIC();
-		/*
-		 * PIC mode, enable APIC mode in the IMCR, i.e.
-		 * connect BSP's local APIC to INT and NMI lines.
-		 */
-		apic_printk(APIC_VERBOSE, "leaving PIC mode, enabling APIC mode.\n");
-		outb(0x70, 0x22);
-		outb(0x01, 0x23);
-	}
-}
-
 void disconnect_bsp_APIC(int virt_wire_setup)
 {
-	if (pic_mode) {
-		/*
-		 * Put the board back into PIC mode (has an effect
-		 * only on certain older boards).  Note that APIC
-		 * interrupts, including IPIs, won't work beyond
-		 * this point!  The only exception are INIT IPIs.
-		 */
-		apic_printk(APIC_QUIET, "disabling APIC mode, entering PIC mode.\n");
-		outb(0x70, 0x22);
-		outb(0x00, 0x23);
-	}
-	else {
-		/* Go back to Virtual Wire compatibility mode */
-		unsigned long value;
-
-		/* For the spurious interrupt use vector F, and enable it */
-		value = apic_read(APIC_SPIV);
-		value &= ~APIC_VECTOR_MASK;
-		value |= APIC_SPIV_APIC_ENABLED;
-		value |= 0xf;
-		apic_write(APIC_SPIV, value);
-
-		if (!virt_wire_setup) {
-			/* For LVT0 make it edge triggered, active high, external and enabled */
-			value = apic_read(APIC_LVT0);
-			value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
-				APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-				APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
-			value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-			value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
-			apic_write(APIC_LVT0, value);
-		}
-		else {
-			/* Disable LVT0 */
-			apic_write(APIC_LVT0, APIC_LVT_MASKED);
-		}
+	/* Go back to Virtual Wire compatibility mode */
+	unsigned long value;
 
-		/* For LVT1 make it edge triggered, active high, nmi and enabled */
-		value = apic_read(APIC_LVT1);
-		value &= ~(
-			APIC_MODE_MASK | APIC_SEND_PENDING |
+	/* For the spurious interrupt use vector F, and enable it */
+	value = apic_read(APIC_SPIV);
+	value &= ~APIC_VECTOR_MASK;
+	value |= APIC_SPIV_APIC_ENABLED;
+	value |= 0xf;
+	apic_write(APIC_SPIV, value);
+
+	if (!virt_wire_setup) {
+		/* For LVT0 make it edge triggered, active high, external and enabled */
+		value = apic_read(APIC_LVT0);
+		value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
 			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
-			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED );
 		value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
-		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
-		apic_write(APIC_LVT1, value);
+		value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_EXTINT);
+		apic_write(APIC_LVT0, value);
+	} else {
+		/* Disable LVT0 */
+		apic_write(APIC_LVT0, APIC_LVT_MASKED);
 	}
+
+	/* For LVT1 make it edge triggered, active high, nmi and enabled */
+	value = apic_read(APIC_LVT1);
+	value &= ~(APIC_MODE_MASK | APIC_SEND_PENDING |
+			APIC_INPUT_POLARITY | APIC_LVT_REMOTE_IRR |
+			APIC_LVT_LEVEL_TRIGGER | APIC_LVT_MASKED);
+	value |= APIC_LVT_REMOTE_IRR | APIC_SEND_PENDING;
+	value = SET_APIC_DELIVERY_MODE(value, APIC_MODE_NMI);
+	apic_write(APIC_LVT1, value);
 }
 
 void disable_local_APIC(void)
@@ -418,7 +386,7 @@ void __cpuinit setup_local_APIC (void)
 	 * TODO: set up through-local-APIC from through-I/O-APIC? --macro
 	 */
 	value = apic_read(APIC_LVT0) & APIC_LVT_MASKED;
-	if (!smp_processor_id() && (pic_mode || !value)) {
+	if (!smp_processor_id() && !value) {
 		value = APIC_DM_EXTINT;
 		apic_printk(APIC_VERBOSE, "enabled ExtINT on CPU#%d\n", smp_processor_id());
 	} else {
@@ -1096,8 +1064,6 @@ int __init APIC_init_uniprocessor (void)
 
 	verify_local_APIC();
 
-	connect_bsp_APIC();
-
 	phys_cpu_present_map = physid_mask_of_physid(boot_cpu_id);
 	apic_write(APIC_ID, SET_APIC_ID(boot_cpu_id));
 
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 90e99cf27aa5..022624503552 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -56,7 +56,6 @@ struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
 int mp_irq_entries;
 
 int nr_ioapics;
-int pic_mode;
 unsigned long mp_lapic_addr = 0;
 
 
@@ -514,13 +513,6 @@ void __init get_smp_config (void)
  		printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
 
 	printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
-	if (mpf->mpf_feature2 & (1<<7)) {
-		printk(KERN_INFO "    IMCR and PIC compatibility mode.\n");
-		pic_mode = 1;
-	} else {
-		printk(KERN_INFO "    Virtual Wire compatibility mode.\n");
-		pic_mode = 0;
-	}
 
 	/*
 	 * Now see if we need to read further.
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index d29571e24953..ae7d6d84e352 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1090,7 +1090,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	/*
 	 * Switch from PIC to APIC mode.
 	 */
-	connect_bsp_APIC();
 	setup_local_APIC();
 
 	if (GET_APIC_ID(apic_read(APIC_ID)) != boot_cpu_id) {
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 3c7aae9fd157..1dc83436cc91 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -178,7 +178,6 @@ extern int mp_irq_entries;
 extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
 extern int mpc_default_type;
 extern unsigned long mp_lapic_addr;
-extern int pic_mode;
 
 #ifdef CONFIG_ACPI
 extern void mp_register_lapic (u8 id, u8 enabled);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6353fa41bebe..498fbc1fc179 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -33,7 +33,6 @@ extern cpumask_t cpu_initialized;
  
 extern void smp_alloc_memory(void);
 extern volatile unsigned long smp_invalidate_needed;
-extern int pic_mode;
 extern void lock_ipi_call_lock(void);
 extern void unlock_ipi_call_lock(void);
 extern int smp_num_siblings;
-- 
cgit v1.2.1


From c1a58b42b428e717afbbb298356e041cea54ad17 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] i386/x86-64: Remove obsolete sanity check in mptable parsing

It apparently has never triggered in many years.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/mpparse.c   | 13 -------------
 arch/x86_64/kernel/mpparse.c | 13 -------------
 2 files changed, 26 deletions(-)

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 827569688562..a2b126fe9a54 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -294,19 +294,6 @@ static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
 			m->mpc_irqtype, m->mpc_irqflag & 3,
 			(m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
 			m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
-	/*
-	 * Well it seems all SMP boards in existence
-	 * use ExtINT/LVT1 == LINT0 and
-	 * NMI/LVT2 == LINT1 - the following check
-	 * will show us if this assumptions is false.
-	 * Until then we do not have to add baggage.
-	 */
-	if ((m->mpc_irqtype == mp_ExtINT) &&
-		(m->mpc_destapiclint != 0))
-			BUG();
-	if ((m->mpc_irqtype == mp_NMI) &&
-		(m->mpc_destapiclint != 1))
-			BUG();
 }
 
 #ifdef CONFIG_X86_NUMAQ
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 022624503552..c12acc3b5552 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -223,19 +223,6 @@ static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
 			m->mpc_irqtype, m->mpc_irqflag & 3,
 			(m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
 			m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
-	/*
-	 * Well it seems all SMP boards in existence
-	 * use ExtINT/LVT1 == LINT0 and
-	 * NMI/LVT2 == LINT1 - the following check
-	 * will show us if this assumptions is false.
-	 * Until then we do not have to add baggage.
-	 */
-	if ((m->mpc_irqtype == mp_ExtINT) &&
-		(m->mpc_destapiclint != 0))
-			BUG();
-	if ((m->mpc_irqtype == mp_NMI) &&
-		(m->mpc_destapiclint != 1))
-			BUG();
 }
 
 /*
-- 
cgit v1.2.1


From eea0e11c1f0d6ef89e64182b2f1223a4ca2b74a2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Factor out common io apic routing entry access

The IO APIC code had lots of duplicated code to read/write 64bit
routing entries into the IO-APIC. Factor this out int common read/write
functions

In a few cases the IO APIC lock is taken more often now, but this
isn't a problem because it's all initialization/shutdown only
slow path code.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 82 ++++++++++++++++++++++----------------------
 1 file changed, 41 insertions(+), 41 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 97e90f6abcc4..1ccf1b41c3ed 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -111,6 +111,33 @@ int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
 	FINAL;								\
 }
 
+union entry_union {
+	struct { u32 w1, w2; };
+	struct IO_APIC_route_entry entry;
+};
+
+static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin)
+{
+	union entry_union eu;
+	unsigned long flags;
+	spin_lock_irqsave(&ioapic_lock, flags);
+	eu.w1 = io_apic_read(apic, 0x10 + 2 * pin);
+	eu.w2 = io_apic_read(apic, 0x11 + 2 * pin);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+	return eu.entry;
+}
+
+static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
+{
+	unsigned long flags;
+	union entry_union eu;
+	eu.entry = e;
+	spin_lock_irqsave(&ioapic_lock, flags);
+	io_apic_write(apic, 0x10 + 2*pin, eu.w1);
+	io_apic_write(apic, 0x11 + 2*pin, eu.w2);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
 #ifdef CONFIG_SMP
 static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
 {
@@ -196,13 +223,9 @@ static void unmask_IO_APIC_irq (unsigned int irq)
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 {
 	struct IO_APIC_route_entry entry;
-	unsigned long flags;
 
 	/* Check delivery_mode to be sure we're not clearing an SMI pin */
-	spin_lock_irqsave(&ioapic_lock, flags);
-	*(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
-	*(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	entry = ioapic_read_entry(apic, pin);
 	if (entry.delivery_mode == dest_SMI)
 		return;
 	/*
@@ -210,10 +233,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 	 */
 	memset(&entry, 0, sizeof(entry));
 	entry.mask = 1;
-	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
-	io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	ioapic_write_entry(apic, pin, entry);
 }
 
 static void clear_IO_APIC (void)
@@ -838,9 +858,9 @@ static void __init setup_IO_APIC_irqs(void)
 			if (!apic && (irq < 16))
 				disable_8259A_irq(irq);
 		}
+		ioapic_write_entry(apic, pin, entry);
+
 		spin_lock_irqsave(&ioapic_lock, flags);
-		io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
-		io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
 		set_native_irq_info(irq, TARGET_CPUS);
 		spin_unlock_irqrestore(&ioapic_lock, flags);
 	}
@@ -978,10 +998,7 @@ void __apicdebuginit print_IO_APIC(void)
 	for (i = 0; i <= reg_01.bits.entries; i++) {
 		struct IO_APIC_route_entry entry;
 
-		spin_lock_irqsave(&ioapic_lock, flags);
-		*(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
-		*(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
+		entry = ioapic_read_entry(apic, i);
 
 		printk(KERN_DEBUG " %02x %03X %02X  ",
 			i,
@@ -1191,11 +1208,7 @@ static void __init enable_IO_APIC(void)
 		/* See if any of the pins is in ExtINT mode */
 		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
 			struct IO_APIC_route_entry entry;
-			spin_lock_irqsave(&ioapic_lock, flags);
-			*(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
-			*(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
-			spin_unlock_irqrestore(&ioapic_lock, flags);
-
+			entry = ioapic_read_entry(apic, pin);
 
 			/* If the interrupt line is enabled and in ExtInt mode
 			 * I have found the pin where the i8259 is connected.
@@ -1247,7 +1260,6 @@ void disable_IO_APIC(void)
 	 */
 	if (ioapic_i8259.pin != -1) {
 		struct IO_APIC_route_entry entry;
-		unsigned long flags;
 
 		memset(&entry, 0, sizeof(entry));
 		entry.mask            = 0; /* Enabled */
@@ -1264,12 +1276,7 @@ void disable_IO_APIC(void)
 		/*
 		 * Add it to the IO-APIC irq-routing table:
 		 */
-		spin_lock_irqsave(&ioapic_lock, flags);
-		io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
-			*(((int *)&entry)+1));
-		io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
-			*(((int *)&entry)+0));
-		spin_unlock_irqrestore(&ioapic_lock, flags);
+		ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
 	}
 
 	disconnect_bsp_APIC(ioapic_i8259.pin != -1);
@@ -1879,17 +1886,12 @@ static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
 {
 	struct IO_APIC_route_entry *entry;
 	struct sysfs_ioapic_data *data;
-	unsigned long flags;
 	int i;
 
 	data = container_of(dev, struct sysfs_ioapic_data, dev);
 	entry = data->entry;
-	spin_lock_irqsave(&ioapic_lock, flags);
-	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
-		*(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
-		*(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
-	}
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ )
+		*entry = ioapic_read_entry(dev->id, i);
 
 	return 0;
 }
@@ -1911,11 +1913,9 @@ static int ioapic_resume(struct sys_device *dev)
 		reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
 		io_apic_write(dev->id, 0, reg_00.raw);
 	}
-	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
-		io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
-		io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
-	}
 	spin_unlock_irqrestore(&ioapic_lock, flags);
+	for (i = 0; i < nr_ioapic_registers[dev->id]; i++)
+		ioapic_write_entry(dev->id, i, entry[i]);
 
 	return 0;
 }
@@ -2040,10 +2040,10 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int triggering, int p
 	if (!ioapic && (irq < 16))
 		disable_8259A_irq(irq);
 
+	ioapic_write_entry(ioapic, pin, entry);
+
 	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
-	io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
-	set_native_irq_info(use_pci_vector() ?  entry.vector : irq, TARGET_CPUS);
+	set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 
 	return 0;
-- 
cgit v1.2.1


From cf4c6a2f27f5db810b69dcb1da7f194489e8ff88 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] i386: Factor out common io apic routing entry access

The IO APIC code had lots of duplicated code to read/write 64bit
routing entries into the IO-APIC. Factor this out int common read/write
functions

In a few cases the IO APIC lock is taken more often now, but this
isn't a problem because it's all initialization/shutdown only
slow path code.

Similar to earlier x86-64 patch.

Includes a fix by Jiri Slaby for a mistake that broke resume

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/io_apic.c | 100 +++++++++++++++++++--------------------------
 1 file changed, 43 insertions(+), 57 deletions(-)

diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 4eacd52eeb74..b713f27d7e86 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -94,6 +94,34 @@ int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
 #define vector_to_irq(vector)	(vector)
 #endif
 
+
+union entry_union {
+	struct { u32 w1, w2; };
+	struct IO_APIC_route_entry entry;
+};
+
+static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin)
+{
+	union entry_union eu;
+	unsigned long flags;
+	spin_lock_irqsave(&ioapic_lock, flags);
+	eu.w1 = io_apic_read(apic, 0x10 + 2 * pin);
+	eu.w2 = io_apic_read(apic, 0x11 + 2 * pin);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+	return eu.entry;
+}
+
+static void ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e)
+{
+	unsigned long flags;
+	union entry_union eu;
+	eu.entry = e;
+	spin_lock_irqsave(&ioapic_lock, flags);
+	io_apic_write(apic, 0x10 + 2*pin, eu.w1);
+	io_apic_write(apic, 0x11 + 2*pin, eu.w2);
+	spin_unlock_irqrestore(&ioapic_lock, flags);
+}
+
 /*
  * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
  * shared ISA-space IRQs, so we have to support them. We are super
@@ -201,13 +229,9 @@ static void unmask_IO_APIC_irq (unsigned int irq)
 static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 {
 	struct IO_APIC_route_entry entry;
-	unsigned long flags;
 	
 	/* Check delivery_mode to be sure we're not clearing an SMI pin */
-	spin_lock_irqsave(&ioapic_lock, flags);
-	*(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
-	*(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	entry = ioapic_read_entry(apic, pin);
 	if (entry.delivery_mode == dest_SMI)
 		return;
 
@@ -216,10 +240,7 @@ static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
 	 */
 	memset(&entry, 0, sizeof(entry));
 	entry.mask = 1;
-	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
-	io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	ioapic_write_entry(apic, pin, entry);
 }
 
 static void clear_IO_APIC (void)
@@ -1284,9 +1305,8 @@ static void __init setup_IO_APIC_irqs(void)
 			if (!apic && (irq < 16))
 				disable_8259A_irq(irq);
 		}
+		ioapic_write_entry(apic, pin, entry);
 		spin_lock_irqsave(&ioapic_lock, flags);
-		io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
-		io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
 		set_native_irq_info(irq, TARGET_CPUS);
 		spin_unlock_irqrestore(&ioapic_lock, flags);
 	}
@@ -1302,7 +1322,6 @@ static void __init setup_IO_APIC_irqs(void)
 static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
 {
 	struct IO_APIC_route_entry entry;
-	unsigned long flags;
 
 	memset(&entry,0,sizeof(entry));
 
@@ -1332,10 +1351,7 @@ static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, in
 	/*
 	 * Add it to the IO-APIC irq-routing table:
 	 */
-	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
-	io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	ioapic_write_entry(apic, pin, entry);
 
 	enable_8259A_irq(0);
 }
@@ -1445,10 +1461,7 @@ void __init print_IO_APIC(void)
 	for (i = 0; i <= reg_01.bits.entries; i++) {
 		struct IO_APIC_route_entry entry;
 
-		spin_lock_irqsave(&ioapic_lock, flags);
-		*(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
-		*(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
+		entry = ioapic_read_entry(apic, i);
 
 		printk(KERN_DEBUG " %02x %03X %02X  ",
 			i,
@@ -1667,10 +1680,7 @@ static void __init enable_IO_APIC(void)
 		/* See if any of the pins is in ExtINT mode */
 		for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
 			struct IO_APIC_route_entry entry;
-			spin_lock_irqsave(&ioapic_lock, flags);
-			*(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
-			*(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
-			spin_unlock_irqrestore(&ioapic_lock, flags);
+			entry = ioapic_read_entry(apic, pin);
 
 
 			/* If the interrupt line is enabled and in ExtInt mode
@@ -1727,7 +1737,6 @@ void disable_IO_APIC(void)
 	 */
 	if (ioapic_i8259.pin != -1) {
 		struct IO_APIC_route_entry entry;
-		unsigned long flags;
 
 		memset(&entry, 0, sizeof(entry));
 		entry.mask            = 0; /* Enabled */
@@ -1744,12 +1753,7 @@ void disable_IO_APIC(void)
 		/*
 		 * Add it to the IO-APIC irq-routing table:
 		 */
-		spin_lock_irqsave(&ioapic_lock, flags);
-		io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
-			*(((int *)&entry)+1));
-		io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
-			*(((int *)&entry)+0));
-		spin_unlock_irqrestore(&ioapic_lock, flags);
+		ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
 	}
 	disconnect_bsp_APIC(ioapic_i8259.pin != -1);
 }
@@ -2214,17 +2218,13 @@ static inline void unlock_ExtINT_logic(void)
 	int apic, pin, i;
 	struct IO_APIC_route_entry entry0, entry1;
 	unsigned char save_control, save_freq_select;
-	unsigned long flags;
 
 	pin  = find_isa_irq_pin(8, mp_INT);
 	apic = find_isa_irq_apic(8, mp_INT);
 	if (pin == -1)
 		return;
 
-	spin_lock_irqsave(&ioapic_lock, flags);
-	*(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
-	*(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	entry0 = ioapic_read_entry(apic, pin);
 	clear_IO_APIC_pin(apic, pin);
 
 	memset(&entry1, 0, sizeof(entry1));
@@ -2237,10 +2237,7 @@ static inline void unlock_ExtINT_logic(void)
 	entry1.trigger = 0;
 	entry1.vector = 0;
 
-	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
-	io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	ioapic_write_entry(apic, pin, entry1);
 
 	save_control = CMOS_READ(RTC_CONTROL);
 	save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
@@ -2259,10 +2256,7 @@ static inline void unlock_ExtINT_logic(void)
 	CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
 	clear_IO_APIC_pin(apic, pin);
 
-	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
-	io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	ioapic_write_entry(apic, pin, entry0);
 }
 
 int timer_uses_ioapic_pin_0;
@@ -2462,17 +2456,12 @@ static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
 {
 	struct IO_APIC_route_entry *entry;
 	struct sysfs_ioapic_data *data;
-	unsigned long flags;
 	int i;
 	
 	data = container_of(dev, struct sysfs_ioapic_data, dev);
 	entry = data->entry;
-	spin_lock_irqsave(&ioapic_lock, flags);
-	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
-		*(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
-		*(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
-	}
-	spin_unlock_irqrestore(&ioapic_lock, flags);
+	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++)
+		entry[i] = ioapic_read_entry(dev->id, i);
 
 	return 0;
 }
@@ -2494,11 +2483,9 @@ static int ioapic_resume(struct sys_device *dev)
 		reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
 		io_apic_write(dev->id, 0, reg_00.raw);
 	}
-	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
-		io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
-		io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
-	}
 	spin_unlock_irqrestore(&ioapic_lock, flags);
+	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++)
+		ioapic_write_entry(dev->id, i, entry[i]);
 
 	return 0;
 }
@@ -2695,9 +2682,8 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a
 	if (!ioapic && (irq < 16))
 		disable_8259A_irq(irq);
 
+	ioapic_write_entry(ioapic, pin, entry);
 	spin_lock_irqsave(&ioapic_lock, flags);
-	io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
-	io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
 	set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
 	spin_unlock_irqrestore(&ioapic_lock, flags);
 
-- 
cgit v1.2.1


From 606bd58de6542e847c51b1b6d83a4cd70a632fe7 Mon Sep 17 00:00:00 2001
From: Diego Calleja <diegocg@gmail.com>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] x86: AUX_DEVICE_INFO is one byte long, use 'movb'

Bugzilla #6552 says:

"In arch/i386/boot/setup.S, movw is used instead of movb for PS/2 mouse
information, although it is unsigned char. This does not harm, because
the jmp instruction overwritten by movw is used before executing movw,
and never be used again"

I've no idea if this is a real bug or how it gets fixed, so I'm submitting
it for review instead of letting it die of boredom in bugzilla. Aditionally
to i386, I've changed x86-64, which mirrors the same code.

Credits to Yoshinori K. Okuji, who found the problem and suggested a fix.

Signed-off-by: Diego Calleja <diegocg@gmail.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/boot/setup.S   | 4 ++--
 arch/x86_64/boot/setup.S | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/i386/boot/setup.S b/arch/i386/boot/setup.S
index d2b684cd620a..3aec4538a113 100644
--- a/arch/i386/boot/setup.S
+++ b/arch/i386/boot/setup.S
@@ -494,12 +494,12 @@ no_voyager:
 	movw	%cs, %ax			# aka SETUPSEG
 	subw	$DELTA_INITSEG, %ax		# aka INITSEG
 	movw	%ax, %ds
-	movw	$0, (0x1ff)			# default is no pointing device
+	movb	$0, (0x1ff)			# default is no pointing device
 	int	$0x11				# int 0x11: equipment list
 	testb	$0x04, %al			# check if mouse installed
 	jz	no_psmouse
 
-	movw	$0xAA, (0x1ff)			# device present
+	movb	$0xAA, (0x1ff)			# device present
 no_psmouse:
 
 #if defined(CONFIG_X86_SPEEDSTEP_SMI) || defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
diff --git a/arch/x86_64/boot/setup.S b/arch/x86_64/boot/setup.S
index a50b631f4d2b..c3bfd223ab49 100644
--- a/arch/x86_64/boot/setup.S
+++ b/arch/x86_64/boot/setup.S
@@ -526,12 +526,12 @@ is_disk1:
 	movw	%cs, %ax			# aka SETUPSEG
 	subw	$DELTA_INITSEG, %ax		# aka INITSEG
 	movw	%ax, %ds
-	movw	$0, (0x1ff)			# default is no pointing device
+	movb	$0, (0x1ff)			# default is no pointing device
 	int	$0x11				# int 0x11: equipment list
 	testb	$0x04, %al			# check if mouse installed
 	jz	no_psmouse
 
-	movw	$0xAA, (0x1ff)			# device present
+	movb	$0xAA, (0x1ff)			# device present
 no_psmouse:
 
 #include "../../i386/boot/edd.S"
-- 
cgit v1.2.1


From edd965229669f8f8dfddec8c863250440fb65ab3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Remove MPS table APIC renumbering

The MPS table specification says that the operating system should
renumber the IO-APICs following the table as needed.  However in
ACPI this is not allowed or neeeded and all x86-64 systems are ACPI
compliant.

The code was already disabled on some systems because it caused
problems there. Remove it completely now.

CC: mdomsch@dell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 71 --------------------------------------------
 1 file changed, 71 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 1ccf1b41c3ed..a02ed17d05dc 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -1282,72 +1282,6 @@ void disable_IO_APIC(void)
 	disconnect_bsp_APIC(ioapic_i8259.pin != -1);
 }
 
-/*
- * function to set the IO-APIC physical IDs based on the
- * values stored in the MPC table.
- *
- * by Matt Domsch <Matt_Domsch@dell.com>  Tue Dec 21 12:25:05 CST 1999
- */
-
-static void __init setup_ioapic_ids_from_mpc (void)
-{
-	union IO_APIC_reg_00 reg_00;
-	int apic;
-	int i;
-	unsigned char old_id;
-	unsigned long flags;
-
-	/*
-	 * Set the IOAPIC ID to the value stored in the MPC table.
-	 */
-	for (apic = 0; apic < nr_ioapics; apic++) {
-
-		/* Read the register 0 value */
-		spin_lock_irqsave(&ioapic_lock, flags);
-		reg_00.raw = io_apic_read(apic, 0);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
-		
-		old_id = mp_ioapics[apic].mpc_apicid;
-
-
-		printk(KERN_INFO "Using IO-APIC %d\n", mp_ioapics[apic].mpc_apicid);
-
-
-		/*
-		 * We need to adjust the IRQ routing table
-		 * if the ID changed.
-		 */
-		if (old_id != mp_ioapics[apic].mpc_apicid)
-			for (i = 0; i < mp_irq_entries; i++)
-				if (mp_irqs[i].mpc_dstapic == old_id)
-					mp_irqs[i].mpc_dstapic
-						= mp_ioapics[apic].mpc_apicid;
-
-		/*
-		 * Read the right value from the MPC table and
-		 * write it into the ID register.
-	 	 */
-		apic_printk(APIC_VERBOSE,KERN_INFO "...changing IO-APIC physical APIC ID to %d ...",
-				mp_ioapics[apic].mpc_apicid);
-
-		reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
-		spin_lock_irqsave(&ioapic_lock, flags);
-		io_apic_write(apic, 0, reg_00.raw);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
-
-		/*
-		 * Sanity check
-		 */
-		spin_lock_irqsave(&ioapic_lock, flags);
-		reg_00.raw = io_apic_read(apic, 0);
-		spin_unlock_irqrestore(&ioapic_lock, flags);
-		if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
-			printk("could not set ID!\n");
-		else
-			apic_printk(APIC_VERBOSE," ok.\n");
-	}
-}
-
 /*
  * There is a nasty bug in some older SMP boards, their mptable lies
  * about the timer IRQ. We do the following to work around the situation:
@@ -1863,11 +1797,6 @@ void __init setup_IO_APIC(void)
 
 	apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
 
-	/*
-	 * Set up the IO-APIC IRQ routing table.
-	 */
-	if (!acpi_ioapic)
-		setup_ioapic_ids_from_mpc();
 	sync_Arb_IDs();
 	setup_IO_APIC_irqs();
 	init_IO_APIC_traps();
-- 
cgit v1.2.1


From dfa4698c50bf85b7927214b0e4a3dc4bc3b3c4a9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Move early chipset quirks out to new file

They did not really belong into io_apic.c. Move them into a new file
and clean it up a bit.

Also remove outdated ATI quirk that was obsolete,

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/Makefile       |   2 +-
 arch/x86_64/kernel/early-quirks.c | 118 ++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/io_apic.c      | 101 --------------------------------
 arch/x86_64/kernel/setup.c        |   2 +-
 include/asm-x86_64/proto.h        |   2 +-
 5 files changed, 121 insertions(+), 104 deletions(-)
 create mode 100644 arch/x86_64/kernel/early-quirks.c

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 0ef39553c0f6..000e67e8f028 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -8,7 +8,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_x86_64.o \
 		x8664_ksyms.o i387.o syscall.o vsyscall.o \
 		setup64.o bootflag.o e820.o reboot.o quirks.o i8237.o \
-		pci-dma.o pci-nommu.o alternative.o
+		pci-dma.o pci-nommu.o alternative.o early-quirks.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-$(CONFIG_X86_MCE)         += mce.o
diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c
new file mode 100644
index 000000000000..d637cff1c4b1
--- /dev/null
+++ b/arch/x86_64/kernel/early-quirks.c
@@ -0,0 +1,118 @@
+/* Various workarounds for chipset bugs.
+   This code runs very early and can't use the regular PCI subsystem
+   The entries are keyed to PCI bridges which usually identify chipsets
+   uniquely.
+   This is only for whole classes of chipsets with specific problems which
+   need early invasive action (e.g. before the timers are initialized).
+   Most PCI device specific workarounds can be done later and should be
+   in standard PCI quirks
+   Mainboard specific bugs should be handled by DMI entries.
+   CPU specific bugs in setup.c */
+
+#include <linux/pci.h>
+#include <linux/acpi.h>
+#include <linux/pci_ids.h>
+#include <asm/pci-direct.h>
+#include <asm/proto.h>
+#include <asm/dma.h>
+
+static void via_bugs(void)
+{
+#ifdef CONFIG_IOMMU
+	if ((end_pfn > MAX_DMA32_PFN ||  force_iommu) &&
+	    !iommu_aperture_allowed) {
+		printk(KERN_INFO
+  "Looks like a VIA chipset. Disabling IOMMU. Override with iommu=allowed\n");
+		iommu_aperture_disabled = 1;
+	}
+#endif
+}
+
+#ifdef CONFIG_ACPI
+
+static int nvidia_hpet_detected __initdata;
+
+static int __init nvidia_hpet_check(unsigned long phys, unsigned long size)
+{
+	nvidia_hpet_detected = 1;
+	return 0;
+}
+#endif
+
+static void nvidia_bugs(void)
+{
+#ifdef CONFIG_ACPI
+	/*
+	 * All timer overrides on Nvidia are
+	 * wrong unless HPET is enabled.
+	 */
+	nvidia_hpet_detected = 0;
+	acpi_table_parse(ACPI_HPET, nvidia_hpet_check);
+	if (nvidia_hpet_detected == 0) {
+		acpi_skip_timer_override = 1;
+		printk(KERN_INFO "Nvidia board "
+		       "detected. Ignoring ACPI "
+		       "timer override.\n");
+	}
+#endif
+	/* RED-PEN skip them on mptables too? */
+
+}
+
+static void ati_bugs(void)
+{
+#if 1 /* for testing */
+	printk("ATI board detected\n");
+#endif
+	/* No bugs right now */
+}
+
+struct chipset {
+	u16 vendor;
+	void (*f)(void);
+};
+
+static struct chipset early_qrk[] = {
+	{ PCI_VENDOR_ID_NVIDIA, nvidia_bugs },
+	{ PCI_VENDOR_ID_VIA, via_bugs },
+	{ PCI_VENDOR_ID_ATI, ati_bugs },
+	{}
+};
+
+void __init early_quirks(void)
+{
+	int num, slot, func;
+	/* Poor man's PCI discovery */
+	for (num = 0; num < 32; num++) {
+		for (slot = 0; slot < 32; slot++) {
+			for (func = 0; func < 8; func++) {
+				u32 class;
+				u32 vendor;
+				u8 type;
+				int i;
+				class = read_pci_config(num,slot,func,
+							PCI_CLASS_REVISION);
+				if (class == 0xffffffff)
+					break;
+
+		       		if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
+					continue;
+
+				vendor = read_pci_config(num, slot, func,
+							 PCI_VENDOR_ID);
+				vendor &= 0xffff;
+
+				for (i = 0; early_qrk[i].f; i++)
+					if (early_qrk[i].vendor == vendor) {
+						early_qrk[i].f();
+						return;
+					}
+
+				type = read_pci_config_byte(num, slot, func,
+							    PCI_HEADER_TYPE);
+				if (!(type & 0x80))
+					break;
+			}
+		}
+	}
+}
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index a02ed17d05dc..8bbff64c20d7 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -280,107 +280,6 @@ static int __init setup_enable_8254_timer(char *s)
 __setup("disable_8254_timer", setup_disable_8254_timer);
 __setup("enable_8254_timer", setup_enable_8254_timer);
 
-#include <asm/pci-direct.h>
-#include <linux/pci_ids.h>
-#include <linux/pci.h>
-
-
-#ifdef CONFIG_ACPI
-
-static int nvidia_hpet_detected __initdata;
-
-static int __init nvidia_hpet_check(unsigned long phys, unsigned long size)
-{
-	nvidia_hpet_detected = 1;
-	return 0;
-}
-#endif
-
-/* Temporary Hack. Nvidia and VIA boards currently only work with IO-APIC
-   off. Check for an Nvidia or VIA PCI bridge and turn it off.
-   Use pci direct infrastructure because this runs before the PCI subsystem. 
-
-   Can be overwritten with "apic"
-
-   And another hack to disable the IOMMU on VIA chipsets.
-
-   ... and others. Really should move this somewhere else.
-
-   Kludge-O-Rama. */
-void __init check_ioapic(void) 
-{ 
-	int num,slot,func; 
-	/* Poor man's PCI discovery */
-	for (num = 0; num < 32; num++) { 
-		for (slot = 0; slot < 32; slot++) { 
-			for (func = 0; func < 8; func++) { 
-				u32 class;
-				u32 vendor;
-				u8 type;
-				class = read_pci_config(num,slot,func,
-							PCI_CLASS_REVISION);
-				if (class == 0xffffffff)
-					break; 
-
-		       		if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
-					continue; 
-
-				vendor = read_pci_config(num, slot, func, 
-							 PCI_VENDOR_ID);
-				vendor &= 0xffff;
-				switch (vendor) { 
-				case PCI_VENDOR_ID_VIA:
-#ifdef CONFIG_IOMMU
-					if ((end_pfn > MAX_DMA32_PFN ||
-					     force_iommu) &&
-					    !iommu_aperture_allowed) {
-						printk(KERN_INFO
-    "Looks like a VIA chipset. Disabling IOMMU. Override with \"iommu=allowed\"\n");
-						iommu_aperture_disabled = 1;
-					}
-#endif
-					return;
-				case PCI_VENDOR_ID_NVIDIA:
-#ifdef CONFIG_ACPI
-					/*
-					 * All timer overrides on Nvidia are
-					 * wrong unless HPET is enabled.
-					 */
-					nvidia_hpet_detected = 0;
-					acpi_table_parse(ACPI_HPET,
-							nvidia_hpet_check);
-					if (nvidia_hpet_detected == 0) {
-						acpi_skip_timer_override = 1;
-						printk(KERN_INFO "Nvidia board "
-						    "detected. Ignoring ACPI "
-						    "timer override.\n");
-					}
-#endif
-					/* RED-PEN skip them on mptables too? */
-					return;
-
-				/* This should be actually default, but
-				   for 2.6.16 let's do it for ATI only where
-				   it's really needed. */
-				case PCI_VENDOR_ID_ATI:
-					if (timer_over_8254 == 1) {	
-						timer_over_8254 = 0;	
-					printk(KERN_INFO
-		"ATI board detected. Disabling timer routing over 8254.\n");
-					}	
-					return;
-				} 
-
-
-				/* No multi-function device? */
-				type = read_pci_config_byte(num,slot,func,
-							    PCI_HEADER_TYPE);
-				if (!(type & 0x80))
-					break;
-			} 
-		}
-	}
-} 
 
 /*
  * Find the IRQ entry number of a certain pin.
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index fbe9f7faa2db..aab4bd66aa0d 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -653,7 +653,7 @@ void __init setup_arch(char **cmdline_p)
 
 	paging_init();
 
-	check_ioapic();
+	early_quirks();
 
 	/*
 	 * set this early, so we dont allocate cpu0
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 3b1c60247902..58fec91318e4 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -92,7 +92,7 @@ extern void syscall32_cpu_init(void);
 
 extern void setup_node_bootmem(int nodeid, unsigned long start, unsigned long end);
 
-extern void check_ioapic(void);
+extern void early_quirks(void);
 extern void check_efer(void);
 
 extern int unhandled_signal(struct task_struct *tsk, int sig);
-- 
cgit v1.2.1


From 55f05ffaa788e039df2f1ebe0d7bfbcb6f39d0b4 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Replace mp bus array with bitmap for bus not pci

Since we only support PCI and ISA legacy busses now there is no need to
have an full array with checking.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 92 +++++++++++---------------------------------
 arch/x86_64/kernel/mpparse.c |  9 ++---
 include/asm-x86_64/mpspec.h  |  8 +---
 3 files changed, 27 insertions(+), 82 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 8bbff64c20d7..a1412042b918 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -308,7 +308,7 @@ static int __init find_isa_irq_pin(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if (mp_bus_id_to_type[lbus] == MP_BUS_ISA &&
+		if (test_bit(lbus, mp_bus_not_pci) &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 
@@ -324,7 +324,7 @@ static int __init find_isa_irq_apic(int irq, int type)
 	for (i = 0; i < mp_irq_entries; i++) {
 		int lbus = mp_irqs[i].mpc_srcbus;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA) &&
+		if (test_bit(lbus, mp_bus_not_pci) &&
 		    (mp_irqs[i].mpc_irqtype == type) &&
 		    (mp_irqs[i].mpc_srcbusirq == irq))
 			break;
@@ -364,7 +364,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
 			    mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
 				break;
 
-		if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
+		if (!test_bit(lbus, mp_bus_not_pci) &&
 		    !mp_irqs[i].mpc_irqtype &&
 		    (bus == lbus) &&
 		    (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
@@ -410,28 +410,11 @@ static int __init MPBIOS_polarity(int idx)
 	switch (mp_irqs[idx].mpc_irqflag & 3)
 	{
 		case 0: /* conforms, ie. bus-type dependent polarity */
-		{
-			switch (mp_bus_id_to_type[bus])
-			{
-				case MP_BUS_ISA: /* ISA pin */
-				{
-					polarity = default_ISA_polarity(idx);
-					break;
-				}
-				case MP_BUS_PCI: /* PCI pin */
-				{
-					polarity = default_PCI_polarity(idx);
-					break;
-				}
-				default:
-				{
-					printk(KERN_WARNING "broken BIOS!!\n");
-					polarity = 1;
-					break;
-				}
-			}
+			if (test_bit(bus, mp_bus_not_pci))
+				polarity = default_ISA_polarity(idx);
+			else
+				polarity = default_PCI_polarity(idx);
 			break;
-		}
 		case 1: /* high active */
 		{
 			polarity = 0;
@@ -469,28 +452,11 @@ static int MPBIOS_trigger(int idx)
 	switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
 	{
 		case 0: /* conforms, ie. bus-type dependent */
-		{
-			switch (mp_bus_id_to_type[bus])
-			{
-				case MP_BUS_ISA: /* ISA pin */
-				{
-					trigger = default_ISA_trigger(idx);
-					break;
-				}
-				case MP_BUS_PCI: /* PCI pin */
-				{
-					trigger = default_PCI_trigger(idx);
-					break;
-				}
-				default:
-				{
-					printk(KERN_WARNING "broken BIOS!!\n");
-					trigger = 1;
-					break;
-				}
-			}
+			if (test_bit(bus, mp_bus_not_pci))
+				trigger = default_ISA_trigger(idx);
+			else
+				trigger = default_PCI_trigger(idx);
 			break;
-		}
 		case 1: /* edge */
 		{
 			trigger = 0;
@@ -596,31 +562,17 @@ static int pin_2_irq(int idx, int apic, int pin)
 	if (mp_irqs[idx].mpc_dstirq != pin)
 		printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
 
-	switch (mp_bus_id_to_type[bus])
-	{
-		case MP_BUS_ISA: /* ISA pin */
-		{
-			irq = mp_irqs[idx].mpc_srcbusirq;
-			break;
-		}
-		case MP_BUS_PCI: /* PCI pin */
-		{
-			/*
-			 * PCI IRQs are mapped in order
-			 */
-			i = irq = 0;
-			while (i < apic)
-				irq += nr_ioapic_registers[i++];
-			irq += pin;
-			irq = gsi_irq_sharing(irq);
-			break;
-		}
-		default:
-		{
-			printk(KERN_ERR "unknown bus type %d.\n",bus); 
-			irq = 0;
-			break;
-		}
+	if (test_bit(bus, mp_bus_not_pci)) {
+		irq = mp_irqs[idx].mpc_srcbusirq;
+	} else {
+		/*
+		 * PCI IRQs are mapped in order
+		 */
+		i = irq = 0;
+		while (i < apic)
+			irq += nr_ioapic_registers[i++];
+		irq += pin;
+		irq = gsi_irq_sharing(irq);
 	}
 	BUG_ON(irq >= NR_IRQS);
 	return irq;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index c12acc3b5552..a8d38c0bb449 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -42,7 +42,7 @@ int acpi_found_madt;
  * MP-table.
  */
 unsigned char apic_version [MAX_APICS];
-unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
+DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 
 static int mp_current_pci_id = 0;
@@ -173,9 +173,9 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 	Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
 
 	if (strncmp(str, "ISA", 3) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
+		set_bit(m->mpc_busid, mp_bus_not_pci);
 	} else if (strncmp(str, "PCI", 3) == 0) {
-		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
+		clear_bit(m->mpc_busid, mp_bus_not_pci);
 		mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
 		mp_current_pci_id++;
 	} else {
@@ -808,8 +808,7 @@ void __init mp_config_acpi_legacy_irqs (void)
 	/* 
 	 * Fabricate the legacy ISA bus (bus #31).
 	 */
-	mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
-	Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
+	set_bit(MP_ISA_BUS, mp_bus_not_pci);
 
 	/* 
 	 * Locate the IOAPIC that manages the ISA IRQs (0-15). 
diff --git a/include/asm-x86_64/mpspec.h b/include/asm-x86_64/mpspec.h
index 1dc83436cc91..017fddb61dc5 100644
--- a/include/asm-x86_64/mpspec.h
+++ b/include/asm-x86_64/mpspec.h
@@ -159,13 +159,7 @@ struct mpc_config_lintsrc
 #define MAX_MP_BUSSES 256
 /* Each PCI slot may be a combo card with its own bus.  4 IRQ pins per slot. */
 #define MAX_IRQ_SOURCES (MAX_MP_BUSSES * 4)
-enum mp_bustype {
-	MP_BUS_ISA = 1,
-	MP_BUS_EISA,
-	MP_BUS_PCI,
-	MP_BUS_MCA
-};
-extern unsigned char mp_bus_id_to_type [MAX_MP_BUSSES];
+extern DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
 
 extern unsigned int boot_cpu_physical_apicid;
-- 
cgit v1.2.1


From a01fd3baff5e2e744f9d1af0e5d3e6b6082e4dcf Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Remove useless wrapper in mpparse.c code

It used to contain support code for NUMAQ, but that is long gone already
on 64bit.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index a8d38c0bb449..79423d1bfcac 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -579,7 +579,7 @@ static int __init smp_scan_config (unsigned long base, unsigned long length)
 	return 0;
 }
 
-void __init find_intel_smp (void)
+void __init find_smp_config(void)
 {
 	unsigned int address;
 
@@ -617,15 +617,6 @@ void __init find_intel_smp (void)
 	 printk(KERN_INFO "No mptable found.\n");
 }
 
-/*
- * - Intel MP Configuration Table
- */
-void __init find_smp_config (void)
-{
-	find_intel_smp();
-}
-
-
 /* --------------------------------------------------------------------------
                             ACPI-based MP Configuration
    -------------------------------------------------------------------------- */
-- 
cgit v1.2.1


From 276ec1a76ad204d47947894a914e10e798400ffb Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Remove some unneeded ACPI externs in mpparse.c

They are not used in this file so remove them. i386 didn't have them either.

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 79423d1bfcac..0cc88f7f288b 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -70,15 +70,6 @@ unsigned disabled_cpus __initdata;
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
 
-/* ACPI MADT entry parsing functions */
-#ifdef CONFIG_ACPI
-extern struct acpi_boot_flags acpi_boot;
-extern int acpi_parse_lapic (acpi_table_entry_header *header);
-extern int acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header);
-extern int acpi_parse_lapic_nmi (acpi_table_entry_header *header);
-extern int acpi_parse_ioapic (acpi_table_entry_header *header);
-#endif /*CONFIG_ACPI*/
-
 u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
 
 
-- 
cgit v1.2.1


From efec3b9a3282714c8441b9bf476f8358bed9ecaa Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Fix up some non linuxy style in ACPI functions in mpparse.c

No functional changes.

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 57 +++++++++++++-------------------------------
 1 file changed, 16 insertions(+), 41 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 0cc88f7f288b..d63f849aea2c 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -614,23 +614,17 @@ void __init find_smp_config(void)
 
 #ifdef CONFIG_ACPI
 
-void __init mp_register_lapic_address (
-	u64			address)
+void __init mp_register_lapic_address(u64 address)
 {
 	mp_lapic_addr = (unsigned long) address;
-
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
-
 	if (boot_cpu_id == -1U)
 		boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
 
 	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
-
-void __cpuinit mp_register_lapic (
-	u8			id, 
-	u8			enabled)
+void __cpuinit mp_register_lapic (u8 id, u8 enabled)
 {
 	struct mpc_config_processor processor;
 	int			boot_cpu = 0;
@@ -668,11 +662,9 @@ static struct mp_ioapic_routing {
 	u32			pin_programmed[4];
 } mp_ioapic_routing[MAX_IO_APICS];
 
-
-static int mp_find_ioapic (
-	int			gsi)
+static int mp_find_ioapic(int gsi)
 {
-	int			i = 0;
+	int i = 0;
 
 	/* Find the IOAPIC that manages this GSI. */
 	for (i = 0; i < nr_ioapics; i++) {
@@ -682,17 +674,12 @@ static int mp_find_ioapic (
 	}
 
 	printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
-
 	return -1;
 }
-	
 
-void __init mp_register_ioapic (
-	u8			id, 
-	u32			address,
-	u32			gsi_base)
+void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 {
-	int			idx = 0;
+	int idx = 0;
 
 	if (nr_ioapics >= MAX_IO_APICS) {
 		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
@@ -729,16 +716,10 @@ void __init mp_register_ioapic (
 		mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
 		mp_ioapic_routing[idx].gsi_start,
 		mp_ioapic_routing[idx].gsi_end);
-
-	return;
 }
 
-
-void __init mp_override_legacy_irq (
-	u8			bus_irq,
-	u8			polarity, 
-	u8			trigger, 
-	u32			gsi)
+void __init
+mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32	gsi)
 {
 	struct mpc_config_intsrc intsrc;
 	int			ioapic = -1;
@@ -776,16 +757,13 @@ void __init mp_override_legacy_irq (
 	mp_irqs[mp_irq_entries] = intsrc;
 	if (++mp_irq_entries == MAX_IRQ_SOURCES)
 		panic("Max # of irq sources exceeded!\n");
-
-	return;
 }
 
-
-void __init mp_config_acpi_legacy_irqs (void)
+void __init mp_config_acpi_legacy_irqs(void)
 {
 	struct mpc_config_intsrc intsrc;
-	int			i = 0;
-	int			ioapic = -1;
+	int i = 0;
+	int ioapic = -1;
 
 	/* 
 	 * Fabricate the legacy ISA bus (bus #31).
@@ -843,24 +821,22 @@ void __init mp_config_acpi_legacy_irqs (void)
 		if (++mp_irq_entries == MAX_IRQ_SOURCES)
 			panic("Max # of irq sources exceeded!\n");
 	}
-
-	return;
 }
 
 #define MAX_GSI_NUM	4096
 
 int mp_register_gsi(u32 gsi, int triggering, int polarity)
 {
-	int			ioapic = -1;
-	int			ioapic_pin = 0;
-	int			idx, bit = 0;
-	static int		pci_irq = 16;
+	int ioapic = -1;
+	int ioapic_pin = 0;
+	int idx, bit = 0;
+	static int pci_irq = 16;
 	/*
 	 * Mapping between Global System Interrupts, which
 	 * represent all possible interrupts, to the IRQs
 	 * assigned to actual devices.
 	 */
-	static int		gsi_to_irq[MAX_GSI_NUM];
+	static int gsi_to_irq[MAX_GSI_NUM];
 
 	if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
 		return gsi;
@@ -934,5 +910,4 @@ int mp_register_gsi(u32 gsi, int triggering, int polarity)
 		polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
 	return gsi;
 }
-
 #endif /*CONFIG_ACPI*/
-- 
cgit v1.2.1


From 19f03ffecdb599c1f64113b6dda0a1f143d2bab9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] i386: Clean up code style in mpparse.c ACPI code

Remove some unlinuxy ways to write function parameter definitions.
Remove some stray "return;"s

No functional change.

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/mpparse.c | 52 +++++++++++++++-------------------------------
 1 file changed, 17 insertions(+), 35 deletions(-)

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index a2b126fe9a54..bdaa75a5bc1c 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -810,8 +810,7 @@ int es7000_plat;
 
 #ifdef CONFIG_ACPI
 
-void __init mp_register_lapic_address (
-	u64			address)
+void __init mp_register_lapic_address(u64 address)
 {
 	mp_lapic_addr = (unsigned long) address;
 
@@ -823,13 +822,10 @@ void __init mp_register_lapic_address (
 	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
-
-void __devinit mp_register_lapic (
-	u8			id, 
-	u8			enabled)
+void __devinit mp_register_lapic (u8 id, u8 enabled)
 {
 	struct mpc_config_processor processor;
-	int			boot_cpu = 0;
+	int boot_cpu = 0;
 	
 	if (MAX_APICS - id <= 0) {
 		printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
@@ -866,11 +862,9 @@ static struct mp_ioapic_routing {
 	u32			pin_programmed[4];
 } mp_ioapic_routing[MAX_IO_APICS];
 
-
-static int mp_find_ioapic (
-	int			gsi)
+static int mp_find_ioapic (int gsi)
 {
-	int			i = 0;
+	int i = 0;
 
 	/* Find the IOAPIC that manages this GSI. */
 	for (i = 0; i < nr_ioapics; i++) {
@@ -883,15 +877,11 @@ static int mp_find_ioapic (
 
 	return -1;
 }
-	
 
-void __init mp_register_ioapic (
-	u8			id, 
-	u32			address,
-	u32			gsi_base)
+void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 {
-	int			idx = 0;
-	int			tmpid;
+	int idx = 0;
+	int tmpid;
 
 	if (nr_ioapics >= MAX_IO_APICS) {
 		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
@@ -937,16 +927,10 @@ void __init mp_register_ioapic (
 		mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
 		mp_ioapic_routing[idx].gsi_base,
 		mp_ioapic_routing[idx].gsi_end);
-
-	return;
 }
 
-
-void __init mp_override_legacy_irq (
-	u8			bus_irq,
-	u8			polarity, 
-	u8			trigger, 
-	u32			gsi)
+void __init
+mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi)
 {
 	struct mpc_config_intsrc intsrc;
 	int			ioapic = -1;
@@ -984,15 +968,13 @@ void __init mp_override_legacy_irq (
 	mp_irqs[mp_irq_entries] = intsrc;
 	if (++mp_irq_entries == MAX_IRQ_SOURCES)
 		panic("Max # of irq sources exceeded!\n");
-
-	return;
 }
 
 void __init mp_config_acpi_legacy_irqs (void)
 {
 	struct mpc_config_intsrc intsrc;
-	int			i = 0;
-	int			ioapic = -1;
+	int i = 0;
+	int ioapic = -1;
 
 	/* 
 	 * Fabricate the legacy ISA bus (bus #31).
@@ -1061,12 +1043,12 @@ void __init mp_config_acpi_legacy_irqs (void)
 
 #define MAX_GSI_NUM	4096
 
-int mp_register_gsi (u32 gsi, int triggering, int polarity)
+int mp_register_gsi(u32 gsi, int triggering, int polarity)
 {
-	int			ioapic = -1;
-	int			ioapic_pin = 0;
-	int			idx, bit = 0;
-	static int		pci_irq = 16;
+	int ioapic = -1;
+	int ioapic_pin = 0;
+	int idx, bit = 0;
+	static int pci_irq = 16;
 	/*
 	 * Mapping between Global System Interrups, which
 	 * represent all possible interrupts, and IRQs
-- 
cgit v1.2.1


From fe7414a2882c953788af13a7f2c9d570ed8f71c5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] Use BUILD_BUG_ON in apic.c build sanity checking

Makes code a little shorter.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 20a76f209629..496cd1bd2ae5 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -265,8 +265,6 @@ void __init sync_Arb_IDs(void)
 				| APIC_DM_INIT);
 }
 
-extern void __error_in_apic_c (void);
-
 /*
  * An initial setup of the virtual wire mode.
  */
@@ -313,8 +311,7 @@ void __cpuinit setup_local_APIC (void)
 
 	value = apic_read(APIC_LVR);
 
-	if ((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f)
-		__error_in_apic_c();
+	BUILD_BUG_ON((SPURIOUS_APIC_VECTOR & 0x0f) != 0x0f);
 
 	/*
 	 * Double-check whether this APIC is really registered.
-- 
cgit v1.2.1


From e2414910f212c52d9d7c64c99a22863488ac5b48 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:30 +0200
Subject: [PATCH] x86: Detect CFI support in the assembler at runtime

... instead of using a CONFIG option. The config option still controls
if the resulting executable actually has unwind information.

This is useful to prevent compilation errors when users select
CONFIG_STACK_UNWIND on old binutils and also allows to use
CFI in the future for non kernel debugging applications.

Cc: jbeulich@novell.com
Cc: sam@ravnborg.org

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/kbuild/makefiles.txt | 5 +++++
 arch/i386/Makefile                 | 4 ++++
 arch/x86_64/Makefile               | 3 +++
 include/asm-x86_64/dwarf2.h        | 2 +-
 scripts/Kbuild.include             | 7 +++++++
 5 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/Documentation/kbuild/makefiles.txt b/Documentation/kbuild/makefiles.txt
index b7d6abb501a6..e2cbd59cf2d0 100644
--- a/Documentation/kbuild/makefiles.txt
+++ b/Documentation/kbuild/makefiles.txt
@@ -421,6 +421,11 @@ more details, with real examples.
 	The second argument is optional, and if supplied will be used
 	if first argument is not supported.
 
+    as-instr
+	as-instr checks if the assembler reports a specific instruction
+	and then outputs either option1 or option2
+	C escapes are supported in the test instruction
+
     cc-option
 	cc-option is used to check if $(CC) supports a given option, and not
 	supported to use an optional second option.
diff --git a/arch/i386/Makefile b/arch/i386/Makefile
index 3e4adb1e2244..508cdbeb3a09 100644
--- a/arch/i386/Makefile
+++ b/arch/i386/Makefile
@@ -46,6 +46,10 @@ cflags-y += -ffreestanding
 # a lot more stack due to the lack of sharing of stacklots:
 CFLAGS				+= $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-option,-fno-unit-at-a-time); fi ;)
 
+# do binutils support CFI?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+
 CFLAGS += $(cflags-y)
 
 # Default subarch .c files
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 431bb4bc36cd..d6472ddf5f6e 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -54,6 +54,9 @@ endif
 cflags-y += $(call cc-option,-funit-at-a-time)
 # prevent gcc from generating any FP code by mistake
 cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
+# do binutils support CFI?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
 CFLAGS += $(cflags-y)
 CFLAGS_KERNEL += $(cflags-kernel-y)
diff --git a/include/asm-x86_64/dwarf2.h b/include/asm-x86_64/dwarf2.h
index 0744db777676..2b9368365fad 100644
--- a/include/asm-x86_64/dwarf2.h
+++ b/include/asm-x86_64/dwarf2.h
@@ -13,7 +13,7 @@
    away for older version. 
  */
 
-#ifdef CONFIG_UNWIND_INFO
+#ifdef CONFIG_AS_CFI
 
 #define CFI_STARTPROC .cfi_startproc
 #define CFI_ENDPROC .cfi_endproc
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index 3d523899fdc0..7adef12a0c26 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -63,6 +63,13 @@ as-option = $(shell if $(CC) $(CFLAGS) $(1) -Wa,-Z -c -o /dev/null \
 	     -xassembler /dev/null > /dev/null 2>&1; then echo "$(1)"; \
 	     else echo "$(2)"; fi ;)
 
+# as-instr
+# Usage: cflags-y += $(call as-instr, instr, option1, option2)
+
+as-instr = $(shell if echo -e "$(1)" | $(AS) -Z -o astest$$$$.out \
+		   2>&1 >/dev/null ; then echo "$(2)"; else echo "$(3)"; fi; \
+	           rm -f astest$$$$.out)
+
 # cc-option
 # Usage: cflags-y += $(call cc-option, -march=winchip-c6, -march=i586)
 
-- 
cgit v1.2.1


From 44cc45267bbe7c64f7d85b074bd670b48b5abdfb Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Remove obsolete CVS $Id$ from assembler files in
 arch/x86_64/kernel/*

CVS hasn't been used for a long time for them.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S | 2 --
 arch/x86_64/kernel/head.S  | 2 --
 2 files changed, 4 deletions(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index a04fc4108ff2..2092f565aa87 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -4,8 +4,6 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *  Copyright (C) 2000, 2001, 2002  Andi Kleen SuSE Labs
  *  Copyright (C) 2000  Pavel Machek <pavel@suse.cz>
- * 
- *  $Id$
  */
 
 /*
diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index 505ec4a57506..4a3326ce709c 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -5,8 +5,6 @@
  *  Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
  *  Copyright (C) 2000 Karsten Keil <kkeil@suse.de>
  *  Copyright (C) 2001,2002 Andi Kleen <ak@suse.de>
- *
- *  $Id: head.S,v 1.49 2002/03/19 17:39:25 ak Exp $
  */
 
 
-- 
cgit v1.2.1


From caff0710ebf6f2c44cbd2b8b31fd6329148bed2e Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] initialize end of memory variables as early as possible

While an earlier patch already did a small step into that direction,
this patch moves initialization of all memory end variables to as
early as possible, so that dependent code doesn't need to check
whether these variables have already been set.

Also, remove a misleading (perhaps just outdated) comment, and make
static a variable only used in a single file.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/e820.c  | 2 +-
 arch/x86_64/kernel/setup.c | 7 ++++++-
 arch/x86_64/mm/init.c      | 6 ------
 3 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index d6d7f731f6f0..1362aad4f3aa 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -40,7 +40,7 @@ unsigned long end_pfn_map;
 /* 
  * Last pfn which the user wants to use.
  */
-unsigned long end_user_pfn = MAXMEM>>PAGE_SHIFT;  
+static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
 
 extern struct resource code_resource, data_resource;
 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index aab4bd66aa0d..ddc3b2d9a165 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -554,7 +554,7 @@ void __init setup_arch(char **cmdline_p)
 	 * we are rounding upwards:
 	 */
 	end_pfn = e820_end_of_ram();
-	num_physpages = end_pfn;		/* for pfn_valid */
+	num_physpages = end_pfn;
 
 	check_efer();
 
@@ -574,6 +574,11 @@ void __init setup_arch(char **cmdline_p)
 	acpi_boot_table_init();
 #endif
 
+	/* How many end-of-memory variables you have, grandma! */
+	max_low_pfn = end_pfn;
+	max_pfn = end_pfn;
+	high_memory = (void *)__va(end_pfn * PAGE_SIZE - 1) + 1;
+
 #ifdef CONFIG_ACPI_NUMA
 	/*
 	 * Parse SRAT to discover nodes.
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index d14fb2dfbfc4..d40134bd6399 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -597,12 +597,6 @@ void __init mem_init(void)
 
 	pci_iommu_alloc();
 
-	/* How many end-of-memory variables you have, grandma! */
-	max_low_pfn = end_pfn;
-	max_pfn = end_pfn;
-	num_physpages = end_pfn;
-	high_memory = (void *) __va(end_pfn * PAGE_SIZE);
-
 	/* clear the zero-page */
 	memset(empty_zero_page, 0, PAGE_SIZE);
 
-- 
cgit v1.2.1


From 3b94355c47e2b025a7ececa0a14180e8fce6b0f1 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] remove int_delivery_dest

The genapic field and the accessor macro weren't used anywhere.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/genapic_cluster.c | 1 -
 arch/x86_64/kernel/genapic_flat.c    | 2 --
 include/asm-x86_64/genapic.h         | 1 -
 include/asm-x86_64/mach_apic.h       | 1 -
 4 files changed, 5 deletions(-)

diff --git a/arch/x86_64/kernel/genapic_cluster.c b/arch/x86_64/kernel/genapic_cluster.c
index 3020917546de..cdb90e671b88 100644
--- a/arch/x86_64/kernel/genapic_cluster.c
+++ b/arch/x86_64/kernel/genapic_cluster.c
@@ -118,7 +118,6 @@ struct genapic apic_cluster = {
 	.name = "clustered",
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
-	.int_delivery_dest = APIC_DEST_PHYSICAL | APIC_DM_FIXED,
 	.target_cpus = cluster_target_cpus,
 	.apic_id_registered = cluster_apic_id_registered,
 	.init_apic_ldr = cluster_init_apic_ldr,
diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c
index eb86d374813a..2ceb8b872b57 100644
--- a/arch/x86_64/kernel/genapic_flat.c
+++ b/arch/x86_64/kernel/genapic_flat.c
@@ -121,7 +121,6 @@ struct genapic apic_flat =  {
 	.name = "flat",
 	.int_delivery_mode = dest_LowestPrio,
 	.int_dest_mode = (APIC_DEST_LOGICAL != 0),
-	.int_delivery_dest = APIC_DEST_LOGICAL | APIC_DM_LOWEST,
 	.target_cpus = flat_target_cpus,
 	.apic_id_registered = flat_apic_id_registered,
 	.init_apic_ldr = flat_init_apic_ldr,
@@ -180,7 +179,6 @@ struct genapic apic_physflat =  {
 	.name = "physical flat",
 	.int_delivery_mode = dest_Fixed,
 	.int_dest_mode = (APIC_DEST_PHYSICAL != 0),
-	.int_delivery_dest = APIC_DEST_PHYSICAL | APIC_DM_FIXED,
 	.target_cpus = physflat_target_cpus,
 	.apic_id_registered = flat_apic_id_registered,
 	.init_apic_ldr = flat_init_apic_ldr,/*not needed, but shouldn't hurt*/
diff --git a/include/asm-x86_64/genapic.h b/include/asm-x86_64/genapic.h
index 50b38e7c58e4..81e714665344 100644
--- a/include/asm-x86_64/genapic.h
+++ b/include/asm-x86_64/genapic.h
@@ -16,7 +16,6 @@ struct genapic {
 	char *name;
 	u32 int_delivery_mode;
 	u32 int_dest_mode;
-	u32 int_delivery_dest;	/* for quick IPIs */
 	int (*apic_id_registered)(void);
 	cpumask_t (*target_cpus)(void);
 	void (*init_apic_ldr)(void);
diff --git a/include/asm-x86_64/mach_apic.h b/include/asm-x86_64/mach_apic.h
index 0acea44c9377..d33422450c00 100644
--- a/include/asm-x86_64/mach_apic.h
+++ b/include/asm-x86_64/mach_apic.h
@@ -16,7 +16,6 @@
 
 #define INT_DELIVERY_MODE (genapic->int_delivery_mode)
 #define INT_DEST_MODE (genapic->int_dest_mode)
-#define INT_DELIVERY_DEST (genapic->int_delivery_dest)
 #define TARGET_CPUS	  (genapic->target_cpus())
 #define apic_id_registered (genapic->apic_id_registered)
 #define init_apic_ldr (genapic->init_apic_ldr)
-- 
cgit v1.2.1


From ba9c231f7499ff6918c069c72ff5fd836c76b963 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] i386: initialize end-of-memory variables as early as possible

Move initialization of all memory end variables to as early as
possible, so that dependent code doesn't need to check whether these
variables have already been set.

Change the range check in kunmap_atomic to actually make use of this
so that the no-mapping-estabished path (under CONFIG_DEBUG_HIGHMEM)
gets used only when the address is inside the lowmem area (and BUG()
otherwise).

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/setup.c |  8 ++++++++
 arch/i386/mm/discontig.c |  5 +++++
 arch/i386/mm/highmem.c   |  2 +-
 arch/i386/mm/init.c      | 20 --------------------
 4 files changed, 14 insertions(+), 21 deletions(-)

diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index f1682206d304..71a540362b78 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1170,6 +1170,14 @@ static unsigned long __init setup_memory(void)
 	}
 	printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
 		pages_to_mb(highend_pfn - highstart_pfn));
+	num_physpages = highend_pfn;
+	high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
+#else
+	num_physpages = max_low_pfn;
+	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
+#endif
+#ifdef CONFIG_FLATMEM
+	max_mapnr = num_physpages;
 #endif
 	printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
 			pages_to_mb(max_low_pfn));
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 7c392dc553b8..f0c10b3cd158 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -313,6 +313,11 @@ unsigned long __init setup_memory(void)
 		highstart_pfn = system_max_low_pfn;
 	printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
 	       pages_to_mb(highend_pfn - highstart_pfn));
+	num_physpages = highend_pfn;
+	high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
+#else
+	num_physpages = system_max_low_pfn;
+	high_memory = (void *) __va(system_max_low_pfn * PAGE_SIZE - 1) + 1;
 #endif
 	printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
 			pages_to_mb(system_max_low_pfn));
diff --git a/arch/i386/mm/highmem.c b/arch/i386/mm/highmem.c
index b6eb4dcb8777..ba44000b9069 100644
--- a/arch/i386/mm/highmem.c
+++ b/arch/i386/mm/highmem.c
@@ -54,7 +54,7 @@ void kunmap_atomic(void *kvaddr, enum km_type type)
 	unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
 	enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
 
-	if (vaddr < FIXADDR_START) { // FIXME
+	if (vaddr >= PAGE_OFFSET && vaddr < (unsigned long)high_memory) {
 		dec_preempt_count();
 		preempt_check_resched();
 		return;
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 89e8486aac34..d0c2fdf6a4d7 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -552,18 +552,6 @@ static void __init test_wp_bit(void)
 	}
 }
 
-static void __init set_max_mapnr_init(void)
-{
-#ifdef CONFIG_HIGHMEM
-	num_physpages = highend_pfn;
-#else
-	num_physpages = max_low_pfn;
-#endif
-#ifdef CONFIG_FLATMEM
-	max_mapnr = num_physpages;
-#endif
-}
-
 static struct kcore_list kcore_mem, kcore_vmalloc; 
 
 void __init mem_init(void)
@@ -590,14 +578,6 @@ void __init mem_init(void)
 	}
 #endif
  
-	set_max_mapnr_init();
-
-#ifdef CONFIG_HIGHMEM
-	high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
-#else
-	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
-#endif
-
 	/* this will put all low memory onto the freelists */
 	totalram_pages += free_all_bootmem();
 
-- 
cgit v1.2.1


From 352f7bae81faa2befa2a3c02b84478dce16b8fd6 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Add stack documentation document from Keith Owens

Describes the stack organization on x86-64.

I changed it a bit and removed some obsolete information and the
questions.

Cc: kaos@sgi.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/x86_64/kernel-stacks | 99 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 99 insertions(+)
 create mode 100644 Documentation/x86_64/kernel-stacks

diff --git a/Documentation/x86_64/kernel-stacks b/Documentation/x86_64/kernel-stacks
new file mode 100644
index 000000000000..bddfddd466ab
--- /dev/null
+++ b/Documentation/x86_64/kernel-stacks
@@ -0,0 +1,99 @@
+Most of the text from Keith Owens, hacked by AK
+
+x86_64 page size (PAGE_SIZE) is 4K.
+
+Like all other architectures, x86_64 has a kernel stack for every
+active thread.  These thread stacks are THREAD_SIZE (2*PAGE_SIZE) big.
+These stacks contain useful data as long as a thread is alive or a
+zombie. While the thread is in user space the kernel stack is empty
+except for the thread_info structure at the bottom.
+
+In addition to the per thread stacks, there are specialized stacks
+associated with each cpu.  These stacks are only used while the kernel
+is in control on that cpu, when a cpu returns to user space the
+specialized stacks contain no useful data.  The main cpu stacks is
+
+* Interrupt stack.  IRQSTACKSIZE
+
+  Used for external hardware interrupts.  If this is the first external
+  hardware interrupt (i.e. not a nested hardware interrupt) then the
+  kernel switches from the current task to the interrupt stack.  Like
+  the split thread and interrupt stacks on i386 (with CONFIG_4KSTACKS),
+  this gives more room for kernel interrupt processing without having
+  to increase the size of every per thread stack.
+
+  The interrupt stack is also used when processing a softirq.
+
+Switching to the kernel interrupt stack is done by software based on a
+per CPU interrupt nest counter. This is needed because x86-64 "IST"
+hardware stacks cannot nest without races.
+
+x86_64 also has a feature which is not available on i386, the ability
+to automatically switch to a new stack for designated events such as
+double fault or NMI, which makes it easier to handle these unusual
+events on x86_64.  This feature is called the Interrupt Stack Table
+(IST).  There can be up to 7 IST entries per cpu. The IST code is an
+index into the Task State Segment (TSS), the IST entries in the TSS
+point to dedicated stacks, each stack can be a different size.
+
+An IST is selected by an non-zero value in the IST field of an
+interrupt-gate descriptor.  When an interrupt occurs and the hardware
+loads such a descriptor, the hardware automatically sets the new stack
+pointer based on the IST value, then invokes the interrupt handler.  If
+software wants to allow nested IST interrupts then the handler must
+adjust the IST values on entry to and exit from the interrupt handler.
+(this is occasionally done, e.g. for debug exceptions)
+
+Events with different IST codes (i.e. with different stacks) can be
+nested.  For example, a debug interrupt can safely be interrupted by an
+NMI.  arch/x86_64/kernel/entry.S::paranoidentry adjusts the stack
+pointers on entry to and exit from all IST events, in theory allowing
+IST events with the same code to be nested.  However in most cases, the
+stack size allocated to an IST assumes no nesting for the same code.
+If that assumption is ever broken then the stacks will become corrupt.
+
+The currently assigned IST stacks are :-
+
+* STACKFAULT_STACK.  EXCEPTION_STKSZ (PAGE_SIZE).
+
+  Used for interrupt 12 - Stack Fault Exception (#SS).
+
+  This allows to recover from invalid stack segments. Rarely
+  happens.
+
+* DOUBLEFAULT_STACK.  EXCEPTION_STKSZ (PAGE_SIZE).
+
+  Used for interrupt 8 - Double Fault Exception (#DF).
+
+  Invoked when handling a exception causes another exception. Happens
+  when the kernel is very confused (e.g. kernel stack pointer corrupt)
+  Using a separate stack allows to recover from it well enough in many
+  cases to still output an oops.
+
+* NMI_STACK.  EXCEPTION_STKSZ (PAGE_SIZE).
+
+  Used for non-maskable interrupts (NMI).
+
+  NMI can be delivered at any time, including when the kernel is in the
+  middle of switching stacks.  Using IST for NMI events avoids making
+  assumptions about the previous state of the kernel stack.
+
+* DEBUG_STACK.  DEBUG_STKSZ
+
+  Used for hardware debug interrupts (interrupt 1) and for software
+  debug interrupts (INT3).
+
+  When debugging a kernel, debug interrupts (both hardware and
+  software) can occur at any time.  Using IST for these interrupts
+  avoids making assumptions about the previous state of the kernel
+  stack.
+
+* MCE_STACK.  EXCEPTION_STKSZ (PAGE_SIZE).
+
+  Used for interrupt 18 - Machine Check Exception (#MC).
+
+  MCE can be delivered at any time, including when the kernel is in the
+  middle of switching stacks.  Using IST for MCE events avoids making
+  assumptions about the previous state of the kernel stack.
+
+For more details see the Intel IA32 or AMD AMD64 architecture manuals.
-- 
cgit v1.2.1


From 5f4a7a93886ce1a4327f6028cc05d423f39eebf0 Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: rearrange 'struct iommu_table' members

Rearrange struct members loosely based on size for improved alignment
and to save a few bytes.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/calgary.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-x86_64/calgary.h b/include/asm-x86_64/calgary.h
index 4e3919524240..0a03bda94d03 100644
--- a/include/asm-x86_64/calgary.h
+++ b/include/asm-x86_64/calgary.h
@@ -34,12 +34,12 @@ struct iommu_table {
 	unsigned long  it_base;      /* mapped address of tce table */
 	unsigned long  it_hint;      /* Hint for next alloc */
 	unsigned long *it_map;       /* A simple allocation bitmap for now */
+	void __iomem  *bbar;         /* Bridge BAR */
+	u64	       tar_val;      /* Table Address Register */
+	struct timer_list watchdog_timer;
 	spinlock_t     it_lock;      /* Protects it_map */
 	unsigned int   it_size;      /* Size of iommu table in entries */
 	unsigned char  it_busno;     /* Bus number this table belongs to */
-	void __iomem  *bbar;
-	u64	       tar_val;
-	struct timer_list watchdog_timer;
 };
 
 #define TCE_TABLE_SIZE_UNSPECIFIED	~0
-- 
cgit v1.2.1


From f38db651d5da5e10235fd7dd31095969fb7ef6fb Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: consolidate per bus data structures

Move the tce_table_kva array, disabled bitmap and bus_to_phb array
into a new per bus 'struct calgary_bus_info'. Also slightly reorganize
build_tce_table and tce_table_setparms to avoid exporting bus_info to
tce.c.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 53 +++++++++++++++++++++-------------------
 arch/x86_64/kernel/tce.c         | 10 --------
 include/asm-x86_64/tce.h         |  1 -
 3 files changed, 28 insertions(+), 36 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 146924ba5df5..96f6a866afad 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -111,17 +111,17 @@ static const unsigned long phb_offsets[] = {
 	0xB000 /* PHB3 */
 };
 
-static char bus_to_phb[MAX_PHB_BUS_NUM];
-void* tce_table_kva[MAX_PHB_BUS_NUM];
 unsigned int specified_table_size = TCE_TABLE_SIZE_UNSPECIFIED;
 static int translate_empty_slots __read_mostly = 0;
 static int calgary_detected __read_mostly = 0;
 
-/*
- * the bitmap of PHBs the user requested that we disable
- * translation on.
- */
-static DECLARE_BITMAP(translation_disabled, MAX_PHB_BUS_NUM);
+struct calgary_bus_info {
+	void *tce_space;
+	int translation_disabled;
+	signed char phbid;
+};
+
+static struct calgary_bus_info bus_info[MAX_PHB_BUS_NUM] = { { NULL, 0, 0 }, };
 
 static void tce_cache_blast(struct iommu_table *tbl);
 
@@ -149,7 +149,7 @@ static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen)
 
 static inline int translate_phb(struct pci_dev* dev)
 {
-	int disabled = test_bit(dev->bus->number, translation_disabled);
+	int disabled = bus_info[dev->bus->number].translation_disabled;
 	return !disabled;
 }
 
@@ -454,7 +454,7 @@ static struct dma_mapping_ops calgary_dma_ops = {
 
 static inline int busno_to_phbid(unsigned char num)
 {
-	return bus_to_phb[num];
+	return bus_info[num].phbid;
 }
 
 static inline unsigned long split_queue_offset(unsigned char num)
@@ -631,6 +631,10 @@ static int __init calgary_setup_tar(struct pci_dev *dev, void __iomem *bbar)
 	if (ret)
 		return ret;
 
+	tbl = dev->sysdata;
+	tbl->it_base = (unsigned long)bus_info[dev->bus->number].tce_space;
+	tce_free(tbl, 0, tbl->it_size);
+
 	calgary_reserve_regions(dev);
 
 	/* set TARs for each PHB */
@@ -824,7 +828,7 @@ static int __init calgary_init(void)
 			calgary_init_one_nontraslated(dev);
 			continue;
 		}
-		if (!tce_table_kva[dev->bus->number] && !translate_empty_slots) {
+		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots) {
 			pci_dev_put(dev);
 			continue;
 		}
@@ -844,7 +848,7 @@ error:
 			pci_dev_put(dev);
 			continue;
 		}
-		if (!tce_table_kva[dev->bus->number] && !translate_empty_slots)
+		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots)
 			continue;
 		calgary_disable_translation(dev);
 		calgary_free_tar(dev);
@@ -894,9 +898,8 @@ void __init detect_calgary(void)
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		int dev;
-
-		tce_table_kva[bus] = NULL;
-		bus_to_phb[bus] = -1;
+		struct calgary_bus_info *info = &bus_info[bus];
+		info->phbid = -1;
 
 		if (read_pci_config(bus, 0, 0, 0) != PCI_VENDOR_DEVICE_ID_CALGARY)
 			continue;
@@ -907,12 +910,9 @@ void __init detect_calgary(void)
 		 */
 		phb = (phb + 1) % PHBS_PER_CALGARY;
 
-		if (test_bit(bus, translation_disabled)) {
-			printk(KERN_INFO "Calgary: translation is disabled for "
-			       "PHB 0x%x\n", bus);
-			/* skip this phb, don't allocate a tbl for it */
+		if (info->translation_disabled)
 			continue;
-		}
+
 		/*
 		 * Scan the slots of the PCI bus to see if there is a device present.
 		 * The parent bus will be the zero-ith device, so start at 1.
@@ -923,8 +923,8 @@ void __init detect_calgary(void)
 				tbl = alloc_tce_table();
 				if (!tbl)
 					goto cleanup;
-				tce_table_kva[bus] = tbl;
-				bus_to_phb[bus] = phb;
+				info->tce_space = tbl;
+				info->phbid = phb;
 				calgary_found = 1;
 				break;
 			}
@@ -940,9 +940,12 @@ void __init detect_calgary(void)
 	return;
 
 cleanup:
-	for (--bus; bus >= 0; --bus)
-		if (tce_table_kva[bus])
-			free_tce_table(tce_table_kva[bus]);
+	for (--bus; bus >= 0; --bus) {
+		struct calgary_bus_info *info = &bus_info[bus];
+
+		if (info->tce_space)
+			free_tce_table(info->tce_space);
+	}
 }
 
 int __init calgary_iommu_init(void)
@@ -1016,7 +1019,7 @@ static int __init calgary_parse_options(char *p)
 			if (bridge < MAX_PHB_BUS_NUM) {
 				printk(KERN_INFO "Calgary: disabling "
 				       "translation for PHB 0x%x\n", bridge);
-				set_bit(bridge, translation_disabled);
+				bus_info[bridge].translation_disabled = 1;
 			}
 		}
 
diff --git a/arch/x86_64/kernel/tce.c b/arch/x86_64/kernel/tce.c
index 0905ce6b9b32..cbabfdf78e06 100644
--- a/arch/x86_64/kernel/tce.c
+++ b/arch/x86_64/kernel/tce.c
@@ -106,14 +106,6 @@ static int tce_table_setparms(struct pci_dev *dev, struct iommu_table *tbl)
 	/* set the tce table size - measured in entries */
 	tbl->it_size = table_size_to_number_of_entries(specified_table_size);
 
-	tbl->it_base = (unsigned long)tce_table_kva[dev->bus->number];
-	if (!tbl->it_base) {
-		printk(KERN_ERR "Calgary: iommu_table_setparms: "
-		       "no table allocated?!\n");
-		ret = -ENOMEM;
-		goto done;
-	}
-
 	/*
 	 * number of bytes needed for the bitmap size in number of
 	 * entries; we need one bit per entry
@@ -162,8 +154,6 @@ int build_tce_table(struct pci_dev *dev, void __iomem *bbar)
 	if (ret)
 		goto free_tbl;
 
-	tce_free(tbl, 0, tbl->it_size);
-
 	tbl->bbar = bbar;
 
 	/*
diff --git a/include/asm-x86_64/tce.h b/include/asm-x86_64/tce.h
index 53e9a68b3336..dbb047febc5e 100644
--- a/include/asm-x86_64/tce.h
+++ b/include/asm-x86_64/tce.h
@@ -24,7 +24,6 @@
 #ifndef _ASM_X86_64_TCE_H
 #define _ASM_X86_64_TCE_H
 
-extern void* tce_table_kva[];
 extern unsigned int specified_table_size;
 struct iommu_table;
 
-- 
cgit v1.2.1


From 9f2dc46d5ec6fd7787182d2232a1003af11879f1 Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: break out of pci_find_device_reverse if dev
 not found

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 96f6a866afad..b2182c936305 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -844,6 +844,8 @@ error:
 		dev = pci_find_device_reverse(PCI_VENDOR_ID_IBM,
 					      PCI_DEVICE_ID_IBM_CALGARY,
 					      dev);
+		if (!dev)
+			break;
 		if (!translate_phb(dev)) {
 			pci_dev_put(dev);
 			continue;
-- 
cgit v1.2.1


From b8f4fe66a560b5ccbb4d4d0d2df356a5d9b98b83 Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: fix error path memleak in calgary_free_tar

We were freeing the iommu_table and leaking the bitmap pages. Also
rename it to calgary_free_bus, which is more accurate.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index b2182c936305..6c5da1d813f3 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -658,11 +658,12 @@ static int __init calgary_setup_tar(struct pci_dev *dev, void __iomem *bbar)
 	return 0;
 }
 
-static void __init calgary_free_tar(struct pci_dev *dev)
+static void __init calgary_free_bus(struct pci_dev *dev)
 {
 	u64 val64;
 	struct iommu_table *tbl = dev->sysdata;
 	void __iomem *target;
+	unsigned int bitmapsz;
 
 	target = calgary_reg(tbl->bbar, tar_offset(dev->bus->number));
 	val64 = be64_to_cpu(readq(target));
@@ -670,8 +671,15 @@ static void __init calgary_free_tar(struct pci_dev *dev)
 	writeq(cpu_to_be64(val64), target);
 	readq(target); /* flush */
 
+	bitmapsz = tbl->it_size / BITS_PER_BYTE;
+	free_pages((unsigned long)tbl->it_map, get_order(bitmapsz));
+	tbl->it_map = NULL;
+
 	kfree(tbl);
 	dev->sysdata = NULL;
+
+	/* Can't free bootmem allocated memory after system is up :-( */
+	bus_info[dev->bus->number].tce_space = NULL;
 }
 
 static void calgary_watchdog(unsigned long data)
@@ -853,7 +861,7 @@ error:
 		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots)
 			continue;
 		calgary_disable_translation(dev);
-		calgary_free_tar(dev);
+		calgary_free_bus(dev);
 		pci_dev_put(dev);
 	}
 
-- 
cgit v1.2.1


From 871b17008e93d7e96f96829ce5f0393c9902d25b Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: fix reference counting of Calgary PCI devices

The pci_get_device() API decrements the reference count on the 'from'
parameter when it continues searching. Therefore, take a ref count on
Calgary bus when we initialize them in either translated or
non-translated mode.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 6c5da1d813f3..7302834718c8 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -786,6 +786,7 @@ static inline unsigned int __init locate_register_space(struct pci_dev *dev)
 
 static int __init calgary_init_one_nontraslated(struct pci_dev *dev)
 {
+	pci_dev_get(dev);
 	dev->sysdata = NULL;
 	dev->bus->self = dev;
 
@@ -810,6 +811,7 @@ static int __init calgary_init_one(struct pci_dev *dev)
 	if (ret)
 		goto iounmap;
 
+	pci_dev_get(dev);
 	dev->bus->self = dev;
 	calgary_enable_translation(dev);
 
@@ -836,10 +838,9 @@ static int __init calgary_init(void)
 			calgary_init_one_nontraslated(dev);
 			continue;
 		}
-		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots) {
-			pci_dev_put(dev);
+		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots)
 			continue;
-		}
+
 		ret = calgary_init_one(dev);
 		if (ret)
 			goto error;
@@ -860,9 +861,10 @@ error:
 		}
 		if (!bus_info[dev->bus->number].tce_space && !translate_empty_slots)
 			continue;
+
 		calgary_disable_translation(dev);
 		calgary_free_bus(dev);
-		pci_dev_put(dev);
+		pci_dev_put(dev); /* Undo calgary_init_one()'s pci_dev_get() */
 	}
 
 	return ret;
-- 
cgit v1.2.1


From a4fc520a0ff92810eea46d74bf742ef73849302e Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: calgary_init_one_nontraslated() can return
 void

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 7302834718c8..f541fb564d92 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -784,13 +784,11 @@ static inline unsigned int __init locate_register_space(struct pci_dev *dev)
 	return address;
 }
 
-static int __init calgary_init_one_nontraslated(struct pci_dev *dev)
+static void __init calgary_init_one_nontraslated(struct pci_dev *dev)
 {
 	pci_dev_get(dev);
 	dev->sysdata = NULL;
 	dev->bus->self = dev;
-
-	return 0;
 }
 
 static int __init calgary_init_one(struct pci_dev *dev)
-- 
cgit v1.2.1


From 0577f148b5e9a773020e3da1e6332a7c6df9d601 Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] Calgary IOMMU: save a bit of space in bus_info

Make translation_disabled a uchar rather than an int

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index f541fb564d92..caf84e4e4ca9 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -117,7 +117,7 @@ static int calgary_detected __read_mostly = 0;
 
 struct calgary_bus_info {
 	void *tce_space;
-	int translation_disabled;
+	unsigned char translation_disabled;
 	signed char phbid;
 };
 
-- 
cgit v1.2.1


From 01215ad8d83e18321d99e9b5750a6f21cac243a2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] i386: Remove lock section support in mutex.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/mutex.h | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/include/asm-i386/mutex.h b/include/asm-i386/mutex.h
index 05a538531229..7a17d9e58ad6 100644
--- a/include/asm-i386/mutex.h
+++ b/include/asm-i386/mutex.h
@@ -30,14 +30,10 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   decl (%%eax)	\n"			\
-			"   js 2f		\n"			\
+			"   jns 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
 			"1:			\n"			\
 									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
-									\
 		:"=a" (dummy)						\
 		: "a" (count)						\
 		: "memory", "ecx", "edx");				\
@@ -86,14 +82,10 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   incl (%%eax)	\n"			\
-			"   jle 2f		\n"			\
+			"   jg	1f		\n"			\
+			"   call "#fail_fn"	\n"			\
 			"1:			\n"			\
 									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
-									\
 		:"=a" (dummy)						\
 		: "a" (count)						\
 		: "memory", "ecx", "edx");				\
-- 
cgit v1.2.1


From add659bf8aa92f8b3f01a8c0220557c959507fb1 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:31 +0200
Subject: [PATCH] i386: Remove lock section support in rwsem.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Also move the trampolines into semaphore.S and add proper CFI
annotations.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/lib/semaphore.S | 63 +++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/rwsem.h  | 62 +++++++++-------------------------------------
 2 files changed, 75 insertions(+), 50 deletions(-)

diff --git a/arch/i386/lib/semaphore.S b/arch/i386/lib/semaphore.S
index e16ff1696390..01f80b5c45d2 100644
--- a/arch/i386/lib/semaphore.S
+++ b/arch/i386/lib/semaphore.S
@@ -152,3 +152,66 @@ ENTRY(__read_lock_failed)
 	END(__read_lock_failed)
 
 #endif
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_down_read_failed)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	push %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	call rwsem_down_read_failed
+	pop %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_down_read_failed)
+
+ENTRY(call_rwsem_down_write_failed)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	calll rwsem_down_write_failed
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_down_write_failed)
+
+ENTRY(call_rwsem_wake)
+	CFI_STARTPROC
+	decw %dx    /* do nothing if still outstanding active readers */
+	jnz 1f
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	call rwsem_wake
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+1:	ret
+	CFI_ENDPROC
+	END(call_rwsem_wake)
+
+/* Fix up special calling conventions */
+ENTRY(call_rwsem_downgrade_wake)
+	CFI_STARTPROC
+	push %ecx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET ecx,0
+	push %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	CFI_REL_OFFSET edx,0
+	call rwsem_downgrade_wake
+	pop %edx
+	CFI_ADJUST_CFA_OFFSET -4
+	pop %ecx
+	CFI_ADJUST_CFA_OFFSET -4
+	ret
+	CFI_ENDPROC
+	END(call_rwsem_downgrade_wake)
+
diff --git a/include/asm-i386/rwsem.h b/include/asm-i386/rwsem.h
index 43113f5608eb..bc598d6388e3 100644
--- a/include/asm-i386/rwsem.h
+++ b/include/asm-i386/rwsem.h
@@ -99,17 +99,9 @@ static inline void __down_read(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning down_read\n\t"
 LOCK_PREFIX	"  incl      (%%eax)\n\t" /* adds 0x00000001, returns the old value */
-		"  js        2f\n\t" /* jump if we weren't granted the lock */
+		"  jns        1f\n"
+		"  call call_rwsem_down_read_failed\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  pushl     %%edx\n\t"
-		"  call      rwsem_down_read_failed\n\t"
-		"  popl      %%edx\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending down_read\n\t"
 		: "+m" (sem->count)
 		: "a" (sem)
@@ -151,15 +143,9 @@ static inline void __down_write_nested(struct rw_semaphore *sem, int subclass)
 		"# beginning down_write\n\t"
 LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtract 0x0000ffff, returns the old value */
 		"  testl     %%edx,%%edx\n\t" /* was the count 0 before? */
-		"  jnz       2f\n\t" /* jump if we weren't granted the lock */
-		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_down_write_failed\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
+		"  jz        1f\n"
+		"  call call_rwsem_down_write_failed\n"
+		"1:\n"
 		"# ending down_write"
 		: "+m" (sem->count), "=d" (tmp)
 		: "a" (sem), "1" (tmp)
@@ -193,17 +179,9 @@ static inline void __up_read(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning __up_read\n\t"
 LOCK_PREFIX	"  xadd      %%edx,(%%eax)\n\t" /* subtracts 1, returns the old value */
-		"  js        2f\n\t" /* jump if the lock is being waited upon */
-		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  decw      %%dx\n\t" /* do nothing if still outstanding active readers */
-		"  jnz       1b\n\t"
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_wake\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
+		"  jns        1f\n\t"
+		"  call call_rwsem_wake\n"
+		"1:\n"
 		"# ending __up_read\n"
 		: "+m" (sem->count), "=d" (tmp)
 		: "a" (sem), "1" (tmp)
@@ -219,17 +197,9 @@ static inline void __up_write(struct rw_semaphore *sem)
 		"# beginning __up_write\n\t"
 		"  movl      %2,%%edx\n\t"
 LOCK_PREFIX	"  xaddl     %%edx,(%%eax)\n\t" /* tries to transition 0xffff0001 -> 0x00000000 */
-		"  jnz       2f\n\t" /* jump if the lock is being waited upon */
+		"  jz       1f\n"
+		"  call call_rwsem_wake\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  decw      %%dx\n\t" /* did the active count reduce to 0? */
-		"  jnz       1b\n\t" /* jump back if not */
-		"  pushl     %%ecx\n\t"
-		"  call      rwsem_wake\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending __up_write\n"
 		: "+m" (sem->count)
 		: "a" (sem), "i" (-RWSEM_ACTIVE_WRITE_BIAS)
@@ -244,17 +214,9 @@ static inline void __downgrade_write(struct rw_semaphore *sem)
 	__asm__ __volatile__(
 		"# beginning __downgrade_write\n\t"
 LOCK_PREFIX	"  addl      %2,(%%eax)\n\t" /* transitions 0xZZZZ0001 -> 0xYYYY0001 */
-		"  js        2f\n\t" /* jump if the lock is being waited upon */
+		"  jns       1f\n\t"
+		"  call call_rwsem_downgrade_wake\n"
 		"1:\n\t"
-		LOCK_SECTION_START("")
-		"2:\n\t"
-		"  pushl     %%ecx\n\t"
-		"  pushl     %%edx\n\t"
-		"  call      rwsem_downgrade_wake\n\t"
-		"  popl      %%edx\n\t"
-		"  popl      %%ecx\n\t"
-		"  jmp       1b\n"
-		LOCK_SECTION_END
 		"# ending __downgrade_write\n"
 		: "+m" (sem->count)
 		: "a" (sem), "i" (-RWSEM_WAITING_BIAS)
-- 
cgit v1.2.1


From 7ca2b49b06a6d26e89e3535653889f1d7892b085 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Remove lock section support in semaphore.h

Lock sections don't work the new dwarf2 unwinder
This generates slightly smaller code. It adds one more taken
jump to the fast path.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/semaphore.h | 49 +++++++++++++++-----------------------------
 1 file changed, 17 insertions(+), 32 deletions(-)

diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index d51e800acf29..e63b6a68f04c 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -100,13 +100,10 @@ static inline void down(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic down operation\n\t"
 		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"js 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %0,%%eax\n\t"
-		"call __down_failed\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n"
+		"\tlea %0,%%eax\n\t"
+		"call __down_failed\n"
+		"2:"
 		:"+m" (sem->count)
 		:
 		:"memory","ax");
@@ -123,15 +120,12 @@ static inline int down_interruptible(struct semaphore * sem)
 	might_sleep();
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %1,%%eax\n\t"
-		"call __down_failed_interruptible\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"lea %1,%%eax\n\t"
+		"call __down_failed_interruptible\n"
+		"2:"
 		:"=a" (result), "+m" (sem->count)
 		:
 		:"memory");
@@ -148,15 +142,12 @@ static inline int down_trylock(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %1,%%eax\n\t"
+		"jns 2f\n\t"
+		"lea %1,%%eax\n\t"
 		"call __down_failed_trylock\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"2:\n"
 		:"=a" (result), "+m" (sem->count)
 		:
 		:"memory");
@@ -166,22 +157,16 @@ static inline int down_trylock(struct semaphore * sem)
 /*
  * Note! This is subtle. We jump to wake people up only if
  * the semaphore was negative (== somebody was waiting on it).
- * The default case (no contention) will result in NO
- * jumps for both down() and up().
  */
 static inline void up(struct semaphore * sem)
 {
 	__asm__ __volatile__(
 		"# atomic up operation\n\t"
 		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jle 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tlea %0,%%eax\n\t"
-		"call __up_wakeup\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
-		".subsection 0\n"
+		"jg 1f\n\t"
+		"lea %0,%%eax\n\t"
+		"call __up_wakeup\n"
+		"1:"
 		:"+m" (sem->count)
 		:
 		:"memory","ax");
-- 
cgit v1.2.1


From 8578bdf835d6d6ece6718858d351456367b8a5cf Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Don't use lock section for mutexes and semaphores

Lock sections cannot be handled by the dwarf2 unwinder.

Disadvantage is a taken branch in the hot path.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/mutex.h     | 20 ++++++--------------
 include/asm-x86_64/semaphore.h | 40 ++++++++++++++--------------------------
 include/asm-x86_64/spinlock.h  |  9 ++++-----
 3 files changed, 24 insertions(+), 45 deletions(-)

diff --git a/include/asm-x86_64/mutex.h b/include/asm-x86_64/mutex.h
index 06fab6de2a88..16396b1de3e4 100644
--- a/include/asm-x86_64/mutex.h
+++ b/include/asm-x86_64/mutex.h
@@ -25,13 +25,9 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   decl (%%rdi)	\n"			\
-			"   js 2f		\n"			\
-			"1:			\n"			\
-									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
+			"   jns 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
+			"1:"						\
 									\
 		:"=D" (dummy)						\
 		: "D" (v)						\
@@ -75,13 +71,9 @@ do {									\
 									\
 	__asm__ __volatile__(						\
 		LOCK_PREFIX "   incl (%%rdi)	\n"			\
-			"   jle 2f		\n"			\
-			"1:			\n"			\
-									\
-		LOCK_SECTION_START("")					\
-			"2: call "#fail_fn"	\n"			\
-			"   jmp 1b		\n"			\
-		LOCK_SECTION_END					\
+			"   jg 1f		\n"			\
+			"   call "#fail_fn"	\n"			\
+			"1:			  "			\
 									\
 		:"=D" (dummy)						\
 		: "D" (v)						\
diff --git a/include/asm-x86_64/semaphore.h b/include/asm-x86_64/semaphore.h
index 064df08b9a0f..107bd90429e8 100644
--- a/include/asm-x86_64/semaphore.h
+++ b/include/asm-x86_64/semaphore.h
@@ -107,12 +107,9 @@ static inline void down(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic down operation\n\t"
 		LOCK_PREFIX "decl %0\n\t"     /* --sem->count */
-		"js 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 1f\n\t"
+		"call __down_failed\n"
+		"1:"
 		:"=m" (sem->count)
 		:"D" (sem)
 		:"memory");
@@ -130,14 +127,11 @@ static inline int down_interruptible(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed_interruptible\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"call __down_failed_interruptible\n"
+		"2:\n"
 		:"=a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory");
@@ -154,14 +148,11 @@ static inline int down_trylock(struct semaphore * sem)
 
 	__asm__ __volatile__(
 		"# atomic interruptible down operation\n\t"
+		"xorl %0,%0\n\t"
 		LOCK_PREFIX "decl %1\n\t"     /* --sem->count */
-		"js 2f\n\t"
-		"xorl %0,%0\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __down_failed_trylock\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jns 2f\n\t"
+		"call __down_failed_trylock\n\t"
+		"2:\n"
 		:"=a" (result), "=m" (sem->count)
 		:"D" (sem)
 		:"memory","cc");
@@ -179,12 +170,9 @@ static inline void up(struct semaphore * sem)
 	__asm__ __volatile__(
 		"# atomic up operation\n\t"
 		LOCK_PREFIX "incl %0\n\t"     /* ++sem->count */
-		"jle 2f\n"
-		"1:\n"
-		LOCK_SECTION_START("")
-		"2:\tcall __up_wakeup\n\t"
-		"jmp 1b\n"
-		LOCK_SECTION_END
+		"jg 1f\n\t"
+		"call __up_wakeup\n"
+		"1:"
 		:"=m" (sem->count)
 		:"D" (sem)
 		:"memory");
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index a8e3d89f591f..1d0733b38a4b 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -22,14 +22,13 @@
 #define __raw_spin_lock_string \
 	"\n1:\t" \
 	LOCK_PREFIX " ; decl %0\n\t" \
-	"js 2f\n" \
-	LOCK_SECTION_START("") \
-	"2:\t" \
+	"jns 2f\n" \
+	"3:\n" \
 	"rep;nop\n\t" \
 	"cmpl $0,%0\n\t" \
-	"jle 2b\n\t" \
+	"jle 3b\n\t" \
 	"jmp 1b\n" \
-	LOCK_SECTION_END
+	"2:\t" \
 
 #define __raw_spin_lock_string_up \
 	"\n\tdecl %0"
-- 
cgit v1.2.1


From a752d7194c4fb5a3e767c95542d04fc5decb1d52 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] fix is_at_popf() for compat tasks

When testing for the REX instruction prefix, first check
for 32-bit mode because in compat mode the REX prefix is an
increment instruction.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/ptrace.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index d35ec1bc696a..3a52c7bcfd12 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -141,8 +141,11 @@ static int is_at_popf(struct task_struct *child, struct pt_regs *regs)
 		case 0xf0: case 0xf2: case 0xf3:
 			continue;
 
-		/* REX prefixes */
 		case 0x40 ... 0x4f:
+			if (regs->cs != __USER_CS)
+				/* 32-bit mode: register increment */
+				return 0;
+			/* 64-bit mode: REX prefix */
 			continue;
 
 			/* CHECKME: f0, f2, f3 */
-- 
cgit v1.2.1


From 8b059d2373c16b6d32787a49daf8ccf72dc61b71 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Clean up spin/rwlocks

- Inline spinlock strings into their inline functions
- Convert macros to typesafe inlines
- Replace some leftover __asm__ __volatile__s with asm volatile

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/rwlock.h   | 14 +--------
 include/asm-x86_64/spinlock.h | 71 ++++++++++++++++++++++++-------------------
 2 files changed, 41 insertions(+), 44 deletions(-)

diff --git a/include/asm-x86_64/rwlock.h b/include/asm-x86_64/rwlock.h
index 28a080d23119..72aeebed920b 100644
--- a/include/asm-x86_64/rwlock.h
+++ b/include/asm-x86_64/rwlock.h
@@ -21,18 +21,6 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	 "0x01000000"
 
-#define __build_read_lock(rw)   \
-	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t" \
-		     "jns 1f\n" \
-		     "call __read_lock_failed\n" \
-		     "1:\n" \
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
-
-#define __build_write_lock(rw) \
-	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t" \
-		     "jz 1f\n" \
-		     "\tcall __write_lock_failed\n\t" \
-		     "1:\n" \
-		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory")
+/* Actual code is in asm/spinlock.h or in arch/x86_64/lib/rwlock.S */
 
 #endif
diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index 1d0733b38a4b..be7a9e629fb2 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -16,30 +16,23 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#define __raw_spin_is_locked(x) \
-		(*(volatile signed int *)(&(x)->slock) <= 0)
-
-#define __raw_spin_lock_string \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decl %0\n\t" \
-	"jns 2f\n" \
-	"3:\n" \
-	"rep;nop\n\t" \
-	"cmpl $0,%0\n\t" \
-	"jle 3b\n\t" \
-	"jmp 1b\n" \
-	"2:\t" \
-
-#define __raw_spin_lock_string_up \
-	"\n\tdecl %0"
-
-#define __raw_spin_unlock_string \
-	"movl $1,%0" \
-		:"=m" (lock->slock) : : "memory"
+static inline int __raw_spin_is_locked(raw_spinlock_t *lock)
+{
+	return *(volatile signed int *)(&(lock)->slock) <= 0;
+}
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm volatile(__raw_spin_lock_string : "=m" (lock->slock) : : "memory");
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; decl %0\n\t"
+		"jns 2f\n"
+		"3:\n"
+		"rep;nop\n\t"
+		"cmpl $0,%0\n\t"
+		"jle 3b\n\t"
+		"jmp 1b\n"
+		"2:\t" : "=m" (lock->slock) : : "memory");
 }
 
 #define __raw_spin_lock_flags(lock, flags) __raw_spin_lock(lock)
@@ -48,7 +41,7 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	int oldval;
 
-	__asm__ __volatile__(
+	asm volatile(
 		"xchgl %0,%1"
 		:"=q" (oldval), "=m" (lock->slock)
 		:"0" (0) : "memory");
@@ -58,13 +51,14 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("movl $1,%0" :"=m" (lock->slock) :: "memory");
 }
 
-#define __raw_spin_unlock_wait(lock) \
-	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
+{
+	while (__raw_spin_is_locked(lock))
+		cpu_relax();
+}
 
 /*
  * Read-write spinlocks, allowing multiple readers
@@ -80,17 +74,32 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  * with the high bit (sign) being the "contended" bit.
  */
 
-#define __raw_read_can_lock(x)		((int)(x)->lock > 0)
-#define __raw_write_can_lock(x)		((x)->lock == RW_LOCK_BIAS)
+static inline int __raw_read_can_lock(raw_rwlock_t *lock)
+{
+	return (int)(lock)->lock > 0;
+}
+
+static inline int __raw_write_can_lock(raw_rwlock_t *lock)
+{
+	return (lock)->lock == RW_LOCK_BIAS;
+}
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw);
+	asm volatile(LOCK_PREFIX "subl $1,(%0)\n\t"
+		     "jns 1f\n"
+		     "call __read_lock_failed\n"
+		     "1:\n"
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw);
+	asm volatile(LOCK_PREFIX "subl %1,(%0)\n\t"
+		     "jz 1f\n"
+		     "\tcall __write_lock_failed\n\t"
+		     "1:\n"
+		     ::"D" (rw), "i" (RW_LOCK_BIAS) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.1


From fb2e28485679418e459583605f9b19807a72ceca Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Clean up spin/rwlocks

- Inline spinlock strings into their inline functions
- Convert macros to typesafe inlines
- Replace some leftover __asm__ __volatile__s with asm volatile

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/rwlock.h   |  14 +----
 include/asm-i386/spinlock.h | 131 ++++++++++++++++++++++----------------------
 2 files changed, 68 insertions(+), 77 deletions(-)

diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index f40ccbd8cb7f..c3e5db32fa48 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -20,18 +20,6 @@
 #define RW_LOCK_BIAS		 0x01000000
 #define RW_LOCK_BIAS_STR	"0x01000000"
 
-#define __build_read_lock(rw, helper)   \
-	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" \
-			"jns 1f\n" \
-			"call " helper "\n\t" \
-			"1:\n" \
-			::"a" (rw) : "memory")
-
-#define __build_write_lock(rw, helper) \
-	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
-			"jz 1f\n" \
-			"call " helper "\n\t" \
-			"1:\n" \
-			::"a" (rw) : "memory")
+/* Code is in asm-i386/spinlock.h */
 
 #endif
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index d1020363c41a..324329313af8 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -4,6 +4,7 @@
 #include <asm/atomic.h>
 #include <asm/rwlock.h>
 #include <asm/page.h>
+#include <asm/processor.h>
 #include <linux/compiler.h>
 
 /*
@@ -17,67 +18,64 @@
  * (the type definitions are in asm/spinlock_types.h)
  */
 
-#define __raw_spin_is_locked(x) \
-		(*(volatile signed char *)(&(x)->slock) <= 0)
-
-#define __raw_spin_lock_string \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decb %0\n\t" \
-	"jns 3f\n" \
-	"2:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0,%0\n\t" \
-	"jle 2b\n\t" \
-	"jmp 1b\n" \
-	"3:\n\t"
-
-/*
- * NOTE: there's an irqs-on section here, which normally would have to be
- * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use
- * __raw_spin_lock_string_flags().
- */
-#define __raw_spin_lock_string_flags \
-	"\n1:\t" \
-	LOCK_PREFIX " ; decb %0\n\t" \
-	"jns 5f\n" \
-	"2:\t" \
-	"testl $0x200, %1\n\t" \
-	"jz 4f\n\t" \
-	"sti\n" \
-	"3:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0, %0\n\t" \
-	"jle 3b\n\t" \
-	"cli\n\t" \
-	"jmp 1b\n" \
-	"4:\t" \
-	"rep;nop\n\t" \
-	"cmpb $0, %0\n\t" \
-	"jg 1b\n\t" \
-	"jmp 4b\n" \
-	"5:\n\t"
+static inline int __raw_spin_is_locked(raw_spinlock_t *x)
+{
+	return *(volatile signed char *)(&(x)->slock) <= 0;
+}
 
 static inline void __raw_spin_lock(raw_spinlock_t *lock)
 {
-	asm(__raw_spin_lock_string : "+m" (lock->slock) : : "memory");
+	asm volatile("\n1:\t"
+		     LOCK_PREFIX " ; decb %0\n\t"
+		     "jns 3f\n"
+		     "2:\t"
+		     "rep;nop\n\t"
+		     "cmpb $0,%0\n\t"
+		     "jle 2b\n\t"
+		     "jmp 1b\n"
+		     "3:\n\t"
+		     : "+m" (lock->slock) : : "memory");
 }
 
 /*
  * It is easier for the lock validator if interrupts are not re-enabled
  * in the middle of a lock-acquire. This is a performance feature anyway
  * so we turn it off:
+ *
+ * NOTE: there's an irqs-on section here, which normally would have to be
+ * irq-traced, but on CONFIG_TRACE_IRQFLAGS we never use this variant.
  */
 #ifndef CONFIG_PROVE_LOCKING
 static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
 {
-	asm(__raw_spin_lock_string_flags : "+m" (lock->slock) : "r" (flags) : "memory");
+	asm volatile(
+		"\n1:\t"
+		LOCK_PREFIX " ; decb %0\n\t"
+		"jns 5f\n"
+		"2:\t"
+		"testl $0x200, %1\n\t"
+		"jz 4f\n\t"
+		"sti\n"
+		"3:\t"
+		"rep;nop\n\t"
+		"cmpb $0, %0\n\t"
+		"jle 3b\n\t"
+		"cli\n\t"
+		"jmp 1b\n"
+		"4:\t"
+		"rep;nop\n\t"
+		"cmpb $0, %0\n\t"
+		"jg 1b\n\t"
+		"jmp 4b\n"
+		"5:\n\t"
+		: "+m" (lock->slock) : "r" (flags) : "memory");
 }
 #endif
 
 static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 {
 	char oldval;
-	__asm__ __volatile__(
+	asm volatile(
 		"xchgb %b0,%1"
 		:"=q" (oldval), "+m" (lock->slock)
 		:"0" (0) : "memory");
@@ -93,38 +91,29 @@ static inline int __raw_spin_trylock(raw_spinlock_t *lock)
 
 #if !defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE)
 
-#define __raw_spin_unlock_string \
-	"movb $1,%0" \
-		:"+m" (lock->slock) : : "memory"
-
-
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("movb $1,%0" : "+m" (lock->slock) :: "memory");
 }
 
 #else
 
-#define __raw_spin_unlock_string \
-	"xchgb %b0, %1" \
-		:"=q" (oldval), "+m" (lock->slock) \
-		:"0" (oldval) : "memory"
-
 static inline void __raw_spin_unlock(raw_spinlock_t *lock)
 {
 	char oldval = 1;
 
-	__asm__ __volatile__(
-		__raw_spin_unlock_string
-	);
+	asm volatile("xchgb %b0, %1"
+		     : "=q" (oldval), "+m" (lock->slock)
+		     : "0" (oldval) : "memory");
 }
 
 #endif
 
-#define __raw_spin_unlock_wait(lock) \
-	do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
+static inline void __raw_spin_unlock_wait(raw_spinlock_t *lock)
+{
+	while (__raw_spin_is_locked(lock))
+		cpu_relax();
+}
 
 /*
  * Read-write spinlocks, allowing multiple readers
@@ -151,22 +140,36 @@ static inline void __raw_spin_unlock(raw_spinlock_t *lock)
  * read_can_lock - would read_trylock() succeed?
  * @lock: the rwlock in question.
  */
-#define __raw_read_can_lock(x)		((int)(x)->lock > 0)
+static inline int __raw_read_can_lock(raw_rwlock_t *x)
+{
+	return (int)(x)->lock > 0;
+}
 
 /**
  * write_can_lock - would write_trylock() succeed?
  * @lock: the rwlock in question.
  */
-#define __raw_write_can_lock(x)		((x)->lock == RW_LOCK_BIAS)
+static inline int __raw_write_can_lock(raw_rwlock_t *x)
+{
+	return (x)->lock == RW_LOCK_BIAS;
+}
 
 static inline void __raw_read_lock(raw_rwlock_t *rw)
 {
-	__build_read_lock(rw, "__read_lock_failed");
+	asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t"
+		     "jns 1f\n"
+		     "call __read_lock_failed\n\t"
+		     "1:\n"
+		     ::"a" (rw) : "memory");
 }
 
 static inline void __raw_write_lock(raw_rwlock_t *rw)
 {
-	__build_write_lock(rw, "__write_lock_failed");
+	asm volatile(LOCK_PREFIX " subl $" RW_LOCK_BIAS_STR ",(%0)\n\t"
+		     "jz 1f\n"
+		     "call __write_lock_failed\n\t"
+		     "1:\n"
+		     ::"a" (rw) : "memory");
 }
 
 static inline int __raw_read_trylock(raw_rwlock_t *lock)
-- 
cgit v1.2.1


From 8d379dad8f1670d233ac67b76b1c5a42ad3714a3 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] annotate arch/x86_64/lib/*.S

Add unwind annotations to arch/x86_64/lib/*.S, and also use the macros
provided by linux/linkage.h where-ever possible.

Some of the alternative instructions handling needed to be adjusted so
that the replacement code would also have valid unwind information.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/lib/clear_page.S | 47 +++++++++++++++-----------
 arch/x86_64/lib/copy_page.S  | 53 +++++++++++++++++++----------
 arch/x86_64/lib/copy_user.S  | 39 ++++++++++++++++------
 arch/x86_64/lib/csum-copy.S  | 26 ++++++++++++---
 arch/x86_64/lib/getuser.S    | 32 +++++++++++-------
 arch/x86_64/lib/iomap_copy.S | 10 ++++--
 arch/x86_64/lib/memcpy.S     | 69 +++++++++++++++++++++-----------------
 arch/x86_64/lib/memset.S     | 79 ++++++++++++++++++++++++--------------------
 arch/x86_64/lib/putuser.S    | 32 +++++++++++-------
 9 files changed, 244 insertions(+), 143 deletions(-)

diff --git a/arch/x86_64/lib/clear_page.S b/arch/x86_64/lib/clear_page.S
index 1f81b79b796c..9a10a78bb4a4 100644
--- a/arch/x86_64/lib/clear_page.S
+++ b/arch/x86_64/lib/clear_page.S
@@ -1,10 +1,22 @@
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+
 /*
  * Zero a page. 	
  * rdi	page
  */			
-	.globl clear_page
-	.p2align 4
-clear_page:
+	ALIGN
+clear_page_c:
+	CFI_STARTPROC
+	movl $4096/8,%ecx
+	xorl %eax,%eax
+	rep stosq
+	ret
+	CFI_ENDPROC
+ENDPROC(clear_page)
+
+ENTRY(clear_page)
+	CFI_STARTPROC
 	xorl   %eax,%eax
 	movl   $4096/64,%ecx
 	.p2align 4
@@ -23,28 +35,25 @@ clear_page:
 	jnz	.Lloop
 	nop
 	ret
-clear_page_end:
+	CFI_ENDPROC
+.Lclear_page_end:
+ENDPROC(clear_page)
 
 	/* Some CPUs run faster using the string instructions.
 	   It is also a lot simpler. Use this when possible */
 
 #include <asm/cpufeature.h>
 
+	.section .altinstr_replacement,"ax"
+1:	.byte 0xeb					/* jmp <disp8> */
+	.byte (clear_page_c - clear_page) - (2f - 1b)	/* offset */
+2:
+	.previous
 	.section .altinstructions,"a"
 	.align 8
-	.quad  clear_page
-	.quad  clear_page_c
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  clear_page_end-clear_page
-	.byte  clear_page_c_end-clear_page_c
-	.previous
-
-	.section .altinstr_replacement,"ax"
-clear_page_c:
-	movl $4096/8,%ecx
-	xorl %eax,%eax
-	rep 
-	stosq
-	ret
-clear_page_c_end:
+	.quad clear_page
+	.quad 1b
+	.byte X86_FEATURE_REP_GOOD
+	.byte .Lclear_page_end - clear_page
+	.byte 2b - 1b
 	.previous
diff --git a/arch/x86_64/lib/copy_page.S b/arch/x86_64/lib/copy_page.S
index 8fa19d96a7ee..0ebb03b60e79 100644
--- a/arch/x86_64/lib/copy_page.S
+++ b/arch/x86_64/lib/copy_page.S
@@ -1,17 +1,33 @@
 /* Written 2003 by Andi Kleen, based on a kernel by Evandro Menezes */
 	
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+
+	ALIGN
+copy_page_c:
+	CFI_STARTPROC
+	movl $4096/8,%ecx
+	rep movsq
+	ret
+	CFI_ENDPROC
+ENDPROC(copy_page_c)
+
 /* Don't use streaming store because it's better when the target
    ends up in cache. */
 	    
 /* Could vary the prefetch distance based on SMP/UP */
 
-	.globl copy_page
-	.p2align 4
-copy_page:
+ENTRY(copy_page)
+	CFI_STARTPROC
 	subq	$3*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 3*8
 	movq	%rbx,(%rsp)
+	CFI_REL_OFFSET rbx, 0
 	movq	%r12,1*8(%rsp)
+	CFI_REL_OFFSET r12, 1*8
 	movq	%r13,2*8(%rsp)
+	CFI_REL_OFFSET r13, 2*8
 
 	movl	$(4096/64)-5,%ecx
 	.p2align 4
@@ -72,30 +88,33 @@ copy_page:
 	jnz	.Loop2
 
 	movq	(%rsp),%rbx
+	CFI_RESTORE rbx
 	movq	1*8(%rsp),%r12
+	CFI_RESTORE r12
 	movq	2*8(%rsp),%r13
+	CFI_RESTORE r13
 	addq	$3*8,%rsp
+	CFI_ADJUST_CFA_OFFSET -3*8
 	ret
+.Lcopy_page_end:
+	CFI_ENDPROC
+ENDPROC(copy_page)
 
 	/* Some CPUs run faster using the string copy instructions.
 	   It is also a lot simpler. Use this when possible */
 
 #include <asm/cpufeature.h>
 
+	.section .altinstr_replacement,"ax"
+1:	.byte 0xeb					/* jmp <disp8> */
+	.byte (copy_page_c - copy_page) - (2f - 1b)	/* offset */
+2:
+	.previous
 	.section .altinstructions,"a"
 	.align 8
-	.quad  copy_page
-	.quad  copy_page_c
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  copy_page_c_end-copy_page_c
-	.byte  copy_page_c_end-copy_page_c
-	.previous
-
-	.section .altinstr_replacement,"ax"
-copy_page_c:
-	movl $4096/8,%ecx
-	rep 
-	movsq 
-	ret
-copy_page_c_end:
+	.quad copy_page
+	.quad 1b
+	.byte X86_FEATURE_REP_GOOD
+	.byte .Lcopy_page_end - copy_page
+	.byte 2b - 1b
 	.previous
diff --git a/arch/x86_64/lib/copy_user.S b/arch/x86_64/lib/copy_user.S
index f64569b83b54..962f3a693c5e 100644
--- a/arch/x86_64/lib/copy_user.S
+++ b/arch/x86_64/lib/copy_user.S
@@ -4,6 +4,9 @@
  * Functions to copy from and to user space.		
  */		 
 
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+
 #define FIX_ALIGNMENT 1
 
 	#include <asm/current.h>
@@ -12,9 +15,8 @@
 	#include <asm/cpufeature.h>
 
 /* Standard copy_to_user with segment limit checking */		
-	.globl copy_to_user
-	.p2align 4	
-copy_to_user:
+ENTRY(copy_to_user)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%rax)
 	movq %rdi,%rcx
 	addq %rdx,%rcx
@@ -25,9 +27,11 @@ copy_to_user:
 	.byte 0xe9	/* 32bit jump */
 	.long .Lcug-1f
 1:
+	CFI_ENDPROC
+ENDPROC(copy_to_user)
 
 	.section .altinstr_replacement,"ax"
-3:	.byte 0xe9			/* replacement jmp with 8 bit immediate */
+3:	.byte 0xe9			/* replacement jmp with 32 bit immediate */
 	.long copy_user_generic_c-1b	/* offset */
 	.previous
 	.section .altinstructions,"a"
@@ -40,9 +44,8 @@ copy_to_user:
 	.previous
 
 /* Standard copy_from_user with segment limit checking */	
-	.globl copy_from_user
-	.p2align 4	
-copy_from_user:
+ENTRY(copy_from_user)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%rax)
 	movq %rsi,%rcx
 	addq %rdx,%rcx
@@ -50,10 +53,13 @@ copy_from_user:
 	cmpq threadinfo_addr_limit(%rax),%rcx
 	jae  bad_from_user
 	/* FALL THROUGH to copy_user_generic */
+	CFI_ENDPROC
+ENDPROC(copy_from_user)
 	
 	.section .fixup,"ax"
 	/* must zero dest */
 bad_from_user:
+	CFI_STARTPROC
 	movl %edx,%ecx
 	xorl %eax,%eax
 	rep
@@ -61,6 +67,8 @@ bad_from_user:
 bad_to_user:
 	movl	%edx,%eax
 	ret
+	CFI_ENDPROC
+END(bad_from_user)
 	.previous
 	
 		
@@ -75,9 +83,8 @@ bad_to_user:
  * Output:		
  * eax uncopied bytes or 0 if successful.
  */
-	.globl copy_user_generic
-	.p2align 4
-copy_user_generic:
+ENTRY(copy_user_generic)
+	CFI_STARTPROC
 	.byte 0x66,0x66,0x90	/* 5 byte nop for replacement jump */
 	.byte 0x66,0x90
 1:
@@ -95,6 +102,8 @@ copy_user_generic:
 	.previous
 .Lcug:
 	pushq %rbx
+	CFI_ADJUST_CFA_OFFSET 8
+	CFI_REL_OFFSET rbx, 0
 	xorl %eax,%eax		/*zero for the exception handler */
 
 #ifdef FIX_ALIGNMENT
@@ -168,9 +177,13 @@ copy_user_generic:
 	decl %ecx
 	jnz .Lloop_1
 
+	CFI_REMEMBER_STATE
 .Lende:
 	popq %rbx
+	CFI_ADJUST_CFA_OFFSET -8
+	CFI_RESTORE rbx
 	ret
+	CFI_RESTORE_STATE
 
 #ifdef FIX_ALIGNMENT
 	/* align destination */
@@ -261,6 +274,9 @@ copy_user_generic:
 .Le_zero:
 	movq %rdx,%rax
 	jmp .Lende
+	CFI_ENDPROC
+ENDPROC(copy_user_generic)
+
 
 	/* Some CPUs run faster using the string copy instructions.
 	   This is also a lot simpler. Use them when possible.
@@ -282,6 +298,7 @@ copy_user_generic:
   * this please consider this.
    */
 copy_user_generic_c:
+	CFI_STARTPROC
 	movl %edx,%ecx
 	shrl $3,%ecx
 	andl $7,%edx	
@@ -294,6 +311,8 @@ copy_user_generic_c:
 	ret
 3:	lea (%rdx,%rcx,8),%rax
 	ret
+	CFI_ENDPROC
+END(copy_user_generic_c)
 
 	.section __ex_table,"a"
 	.quad 1b,3b
diff --git a/arch/x86_64/lib/csum-copy.S b/arch/x86_64/lib/csum-copy.S
index 72fd55ee896e..f0dba36578ea 100644
--- a/arch/x86_64/lib/csum-copy.S
+++ b/arch/x86_64/lib/csum-copy.S
@@ -5,8 +5,9 @@
  * License.  See the file COPYING in the main directory of this archive
  * for more details. No warranty for anything given at all.
  */
- 	#include <linux/linkage.h>
-	#include <asm/errno.h>
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+#include <asm/errno.h>
 
 /*
  * Checksum copy with exception handling.
@@ -53,19 +54,24 @@
 	.endm
 	
 				
-	.globl csum_partial_copy_generic
-	.p2align 4
-csum_partial_copy_generic:
+ENTRY(csum_partial_copy_generic)
+	CFI_STARTPROC
 	cmpl	 $3*64,%edx
 	jle	 .Lignore
 
 .Lignore:		
 	subq  $7*8,%rsp
+	CFI_ADJUST_CFA_OFFSET 7*8
 	movq  %rbx,2*8(%rsp)
+	CFI_REL_OFFSET rbx, 2*8
 	movq  %r12,3*8(%rsp)
+	CFI_REL_OFFSET r12, 3*8
 	movq  %r14,4*8(%rsp)
+	CFI_REL_OFFSET r14, 4*8
 	movq  %r13,5*8(%rsp)
+	CFI_REL_OFFSET r13, 5*8
 	movq  %rbp,6*8(%rsp)
+	CFI_REL_OFFSET rbp, 6*8
 
 	movq  %r8,(%rsp)
 	movq  %r9,1*8(%rsp)
@@ -208,14 +214,22 @@ csum_partial_copy_generic:
 	addl %ebx,%eax
 	adcl %r9d,%eax		/* carry */
 			
+	CFI_REMEMBER_STATE
 .Lende:
 	movq 2*8(%rsp),%rbx
+	CFI_RESTORE rbx
 	movq 3*8(%rsp),%r12
+	CFI_RESTORE r12
 	movq 4*8(%rsp),%r14
+	CFI_RESTORE r14
 	movq 5*8(%rsp),%r13
+	CFI_RESTORE r13
 	movq 6*8(%rsp),%rbp
+	CFI_RESTORE rbp
 	addq $7*8,%rsp
+	CFI_ADJUST_CFA_OFFSET -7*8
 	ret
+	CFI_RESTORE_STATE
 
 	/* Exception handlers. Very simple, zeroing is done in the wrappers */
 .Lbad_source:
@@ -231,3 +245,5 @@ csum_partial_copy_generic:
 	jz   .Lende	
 	movl $-EFAULT,(%rax)
 	jmp .Lende
+	CFI_ENDPROC
+ENDPROC(csum_partial_copy_generic)
diff --git a/arch/x86_64/lib/getuser.S b/arch/x86_64/lib/getuser.S
index 3844d5e885a4..5448876261f8 100644
--- a/arch/x86_64/lib/getuser.S
+++ b/arch/x86_64/lib/getuser.S
@@ -27,25 +27,26 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/dwarf2.h>
 #include <asm/page.h>
 #include <asm/errno.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
 
 	.text
-	.p2align 4
-.globl __get_user_1
-__get_user_1:	
+ENTRY(__get_user_1)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	cmpq threadinfo_addr_limit(%r8),%rcx
 	jae bad_get_user
 1:	movzb (%rcx),%edx
 	xorl %eax,%eax
 	ret
+	CFI_ENDPROC
+ENDPROC(__get_user_1)
 
-	.p2align 4
-.globl __get_user_2
-__get_user_2:
+ENTRY(__get_user_2)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $1,%rcx
 	jc 20f
@@ -57,10 +58,11 @@ __get_user_2:
 	ret
 20:	decq    %rcx
 	jmp	bad_get_user
+	CFI_ENDPROC
+ENDPROC(__get_user_2)
 
-	.p2align 4
-.globl __get_user_4
-__get_user_4:
+ENTRY(__get_user_4)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $3,%rcx
 	jc 30f
@@ -72,10 +74,11 @@ __get_user_4:
 	ret
 30:	subq $3,%rcx
 	jmp bad_get_user
+	CFI_ENDPROC
+ENDPROC(__get_user_4)
 
-	.p2align 4
-.globl __get_user_8
-__get_user_8:
+ENTRY(__get_user_8)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $7,%rcx
 	jc 40f
@@ -87,11 +90,16 @@ __get_user_8:
 	ret
 40:	subq $7,%rcx
 	jmp bad_get_user
+	CFI_ENDPROC
+ENDPROC(__get_user_8)
 
 bad_get_user:
+	CFI_STARTPROC
 	xorl %edx,%edx
 	movq $(-EFAULT),%rax
 	ret
+	CFI_ENDPROC
+END(bad_get_user)
 
 .section __ex_table,"a"
 	.quad 1b,bad_get_user
diff --git a/arch/x86_64/lib/iomap_copy.S b/arch/x86_64/lib/iomap_copy.S
index 8bbade5fea05..05a95e713da8 100644
--- a/arch/x86_64/lib/iomap_copy.S
+++ b/arch/x86_64/lib/iomap_copy.S
@@ -15,12 +15,16 @@
  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.
  */
 
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+
 /*
  * override generic version in lib/iomap_copy.c
  */
- 	.globl __iowrite32_copy
-	.p2align 4
-__iowrite32_copy:
+ENTRY(__iowrite32_copy)
+	CFI_STARTPROC
 	movl %edx,%ecx
 	rep movsd
 	ret
+	CFI_ENDPROC
+ENDPROC(__iowrite32_copy)
diff --git a/arch/x86_64/lib/memcpy.S b/arch/x86_64/lib/memcpy.S
index 5554948b5554..967b22fa7d07 100644
--- a/arch/x86_64/lib/memcpy.S
+++ b/arch/x86_64/lib/memcpy.S
@@ -1,6 +1,10 @@
 /* Copyright 2002 Andi Kleen */
 	
-	#include <asm/cpufeature.h>		
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+#include <asm/cpufeature.h>
+
 /*
  * memcpy - Copy a memory block.
  *
@@ -13,12 +17,26 @@
  * rax original destination
  */	
 
- 	.globl __memcpy
-	.globl memcpy
-	.p2align 4
-__memcpy:
-memcpy:		
+	ALIGN
+memcpy_c:
+	CFI_STARTPROC
+	movq %rdi,%rax
+	movl %edx,%ecx
+	shrl $3,%ecx
+	andl $7,%edx
+	rep movsq
+	movl %edx,%ecx
+	rep movsb
+	ret
+	CFI_ENDPROC
+ENDPROC(memcpy_c)
+
+ENTRY(__memcpy)
+ENTRY(memcpy)
+	CFI_STARTPROC
 	pushq %rbx
+	CFI_ADJUST_CFA_OFFSET 8
+	CFI_REL_OFFSET rbx, 0
 	movq %rdi,%rax
 
 	movl %edx,%ecx
@@ -86,36 +104,27 @@ memcpy:
 
 .Lende:
 	popq %rbx
+	CFI_ADJUST_CFA_OFFSET -8
+	CFI_RESTORE rbx
 	ret
 .Lfinal:
+	CFI_ENDPROC
+ENDPROC(memcpy)
+ENDPROC(__memcpy)
 
 	/* Some CPUs run faster using the string copy instructions.
 	   It is also a lot simpler. Use this when possible */
 
+	.section .altinstr_replacement,"ax"
+1:	.byte 0xeb				/* jmp <disp8> */
+	.byte (memcpy_c - memcpy) - (2f - 1b)	/* offset */
+2:
+	.previous
 	.section .altinstructions,"a"
 	.align 8
-	.quad  memcpy
-	.quad  memcpy_c
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  .Lfinal-memcpy
-	.byte  memcpy_c_end-memcpy_c
-	.previous
-
-	.section .altinstr_replacement,"ax"
- /* rdi	destination
-  * rsi source
-  * rdx count
-  */
-memcpy_c:
-	movq %rdi,%rax
-	movl %edx,%ecx
-	shrl $3,%ecx
-	andl $7,%edx	
-	rep 
-	movsq 
-	movl %edx,%ecx
-	rep
-	movsb
-	ret
-memcpy_c_end:
+	.quad memcpy
+	.quad 1b
+	.byte X86_FEATURE_REP_GOOD
+	.byte .Lfinal - memcpy
+	.byte 2b - 1b
 	.previous
diff --git a/arch/x86_64/lib/memset.S b/arch/x86_64/lib/memset.S
index ad397f2c7de8..09ed1f6b0eaa 100644
--- a/arch/x86_64/lib/memset.S
+++ b/arch/x86_64/lib/memset.S
@@ -1,4 +1,9 @@
 /* Copyright 2002 Andi Kleen, SuSE Labs */
+
+#include <linux/config.h>
+#include <linux/linkage.h>
+#include <asm/dwarf2.h>
+
 /*
  * ISO C memset - set a memory block to a byte value.
  *	
@@ -8,11 +13,29 @@
  * 
  * rax   original destination
  */	
- 	.globl __memset
-	.globl memset
-	.p2align 4
-memset:	
-__memset:
+	ALIGN
+memset_c:
+	CFI_STARTPROC
+	movq %rdi,%r9
+	movl %edx,%r8d
+	andl $7,%r8d
+	movl %edx,%ecx
+	shrl $3,%ecx
+	/* expand byte value  */
+	movzbl %sil,%esi
+	movabs $0x0101010101010101,%rax
+	mulq %rsi		/* with rax, clobbers rdx */
+	rep stosq
+	movl %r8d,%ecx
+	rep stosb
+	movq %r9,%rax
+	ret
+	CFI_ENDPROC
+ENDPROC(memset_c)
+
+ENTRY(memset)
+ENTRY(__memset)
+	CFI_STARTPROC
 	movq %rdi,%r10
 	movq %rdx,%r11
 
@@ -25,6 +48,7 @@ __memset:
 	movl  %edi,%r9d
 	andl  $7,%r9d
 	jnz  .Lbad_alignment
+	CFI_REMEMBER_STATE
 .Lafter_bad_alignment:
 
 	movl %r11d,%ecx
@@ -75,6 +99,7 @@ __memset:
 	movq	%r10,%rax
 	ret
 
+	CFI_RESTORE_STATE
 .Lbad_alignment:
 	cmpq $7,%r11
 	jbe	.Lhandle_7
@@ -84,42 +109,26 @@ __memset:
 	addq %r8,%rdi
 	subq %r8,%r11
 	jmp .Lafter_bad_alignment
+.Lfinal:
+	CFI_ENDPROC
+ENDPROC(memset)
+ENDPROC(__memset)
 
 	/* Some CPUs run faster using the string instructions.
 	   It is also a lot simpler. Use this when possible */
 
 #include <asm/cpufeature.h>
 
+	.section .altinstr_replacement,"ax"
+1:	.byte 0xeb				/* jmp <disp8> */
+	.byte (memset_c - memset) - (2f - 1b)	/* offset */
+2:
+	.previous
 	.section .altinstructions,"a"
 	.align 8
-	.quad  memset
-	.quad  memset_c
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  memset_c_end-memset_c
-	.byte  memset_c_end-memset_c
-	.previous
-
-	.section .altinstr_replacement,"ax"
- /* rdi	destination
-  * rsi value
-  * rdx count
-  */
-memset_c:
-	movq %rdi,%r9
-	movl %edx,%r8d
-	andl $7,%r8d		
-	movl %edx,%ecx
-	shrl $3,%ecx		
-	/* expand byte value  */
-	movzbl %sil,%esi
-	movabs $0x0101010101010101,%rax
-	mulq   %rsi		/* with rax, clobbers rdx */
-	rep
-	stosq	
-	movl %r8d,%ecx
-	rep
-	stosb
-	movq %r9,%rax
-	ret
-memset_c_end:
+	.quad memset
+	.quad 1b
+	.byte X86_FEATURE_REP_GOOD
+	.byte .Lfinal - memset
+	.byte 2b - 1b
 	.previous
diff --git a/arch/x86_64/lib/putuser.S b/arch/x86_64/lib/putuser.S
index 7f5593974e2d..4989f5a8fa9b 100644
--- a/arch/x86_64/lib/putuser.S
+++ b/arch/x86_64/lib/putuser.S
@@ -25,25 +25,26 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/dwarf2.h>
 #include <asm/page.h>
 #include <asm/errno.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
 
 	.text
-	.p2align 4
-.globl __put_user_1
-__put_user_1:
+ENTRY(__put_user_1)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	cmpq threadinfo_addr_limit(%r8),%rcx
 	jae bad_put_user
 1:	movb %dl,(%rcx)
 	xorl %eax,%eax
 	ret
+	CFI_ENDPROC
+ENDPROC(__put_user_1)
 
-	.p2align 4
-.globl __put_user_2
-__put_user_2:
+ENTRY(__put_user_2)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $1,%rcx
 	jc 20f
@@ -55,10 +56,11 @@ __put_user_2:
 	ret
 20:	decq %rcx
 	jmp bad_put_user
+	CFI_ENDPROC
+ENDPROC(__put_user_2)
 
-	.p2align 4
-.globl __put_user_4
-__put_user_4:
+ENTRY(__put_user_4)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $3,%rcx
 	jc 30f
@@ -70,10 +72,11 @@ __put_user_4:
 	ret
 30:	subq $3,%rcx
 	jmp bad_put_user
+	CFI_ENDPROC
+ENDPROC(__put_user_4)
 
-	.p2align 4
-.globl __put_user_8
-__put_user_8:
+ENTRY(__put_user_8)
+	CFI_STARTPROC
 	GET_THREAD_INFO(%r8)
 	addq $7,%rcx
 	jc 40f
@@ -85,10 +88,15 @@ __put_user_8:
 	ret
 40:	subq $7,%rcx
 	jmp bad_put_user
+	CFI_ENDPROC
+ENDPROC(__put_user_8)
 
 bad_put_user:
+	CFI_STARTPROC
 	movq $(-EFAULT),%rax
 	ret
+	CFI_ENDPROC
+END(bad_put_user)
 
 .section __ex_table,"a"
 	.quad 1b,bad_put_user
-- 
cgit v1.2.1


From 29a6c25bd63dbd52e73d25268402d065ad84c0fe Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Fix gdt table size in trampoline.S

Allows easier extension of the GDT by using the proper C symbol
for the size in the descriptor.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/trampoline.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/trampoline.S b/arch/x86_64/kernel/trampoline.S
index 23a03eb91fc7..c79b99a9e2f6 100644
--- a/arch/x86_64/kernel/trampoline.S
+++ b/arch/x86_64/kernel/trampoline.S
@@ -64,7 +64,7 @@ idt_48:
 	.word	0, 0			# idt base = 0L
 
 gdt_48:
-	.short	__KERNEL32_CS + 7	# gdt limit
+	.short	GDT_ENTRIES*8 - 1	# gdt limit
 	.long	cpu_gdt_table-__START_KERNEL_map
 
 .globl trampoline_end
-- 
cgit v1.2.1


From 145106e81090fcdfb57b7f2e90b98c6445b6f33d Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] remove superflous BUG_ON's in nommu and gart

There's no need to check for invalid DMA data direction in nommu and
gart since we do it in dma-mapping.h anyway before calling the
individual dma-ops.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-gart.c  | 3 ---
 arch/x86_64/kernel/pci-nommu.c | 1 -
 2 files changed, 4 deletions(-)

diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index 6d3e61baf7a0..16261a8a3303 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -239,8 +239,6 @@ dma_addr_t gart_map_single(struct device *dev, void *addr, size_t size, int dir)
 {
 	unsigned long phys_mem, bus;
 
-	BUG_ON(dir == DMA_NONE);
-
 	if (!dev)
 		dev = &fallback_dev;
 
@@ -383,7 +381,6 @@ int gart_map_sg(struct device *dev, struct scatterlist *sg, int nents, int dir)
 	unsigned long pages = 0;
 	int need = 0, nextneed;
 
-	BUG_ON(dir == DMA_NONE);
 	if (nents == 0) 
 		return 0;
 
diff --git a/arch/x86_64/kernel/pci-nommu.c b/arch/x86_64/kernel/pci-nommu.c
index aad7609d8e92..df09ab05a1bd 100644
--- a/arch/x86_64/kernel/pci-nommu.c
+++ b/arch/x86_64/kernel/pci-nommu.c
@@ -59,7 +59,6 @@ int nommu_map_sg(struct device *hwdev, struct scatterlist *sg,
 {
 	int i;
 
-	BUG_ON(direction == DMA_NONE);
  	for (i = 0; i < nents; i++ ) {
 		struct scatterlist *s = &sg[i];
 		BUG_ON(!s->page);
-- 
cgit v1.2.1


From d4d35854a1cf7fb170c8e133a5e40a9d19e7fc17 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] remove lock prefix from is_at_popf() tests

The lock prefix will cause an exception when used with the
popf instruction, so no need to continue searching after it's
found.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/ptrace.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index 3a52c7bcfd12..412c6a8bf906 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -138,7 +138,7 @@ static int is_at_popf(struct task_struct *child, struct pt_regs *regs)
 		case 0x26: case 0x2e:
 		case 0x36: case 0x3e:
 		case 0x64: case 0x65:
-		case 0xf0: case 0xf2: case 0xf3:
+		case 0xf2: case 0xf3:
 			continue;
 
 		case 0x40 ... 0x4f:
@@ -148,7 +148,7 @@ static int is_at_popf(struct task_struct *child, struct pt_regs *regs)
 			/* 64-bit mode: REX prefix */
 			continue;
 
-			/* CHECKME: f0, f2, f3 */
+			/* CHECKME: f2, f3 */
 
 		/*
 		 * pushf: NOTE! We should probably not let
-- 
cgit v1.2.1


From 9ca33eb6981549c0d1b7aea7f99f1ba602161356 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Use early CPU identify before early command line parsing

This makes it possible to modify CPU flags in command line
options without hacks.

And remove another copy in head64.c

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/head64.c | 19 -------------------
 arch/x86_64/kernel/setup.c  |  4 ++--
 2 files changed, 2 insertions(+), 21 deletions(-)

diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index bacbd75c63ad..f2461fde9f8f 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -56,24 +56,6 @@ static void __init copy_bootdata(char *real_mode_data)
 	printk("Bootdata ok (command line is %s)\n", saved_command_line);	
 }
 
-static void __init setup_boot_cpu_data(void)
-{
-	unsigned int dummy, eax;
-
-	/* get vendor info */
-	cpuid(0, (unsigned int *)&boot_cpu_data.cpuid_level,
-	      (unsigned int *)&boot_cpu_data.x86_vendor_id[0],
-	      (unsigned int *)&boot_cpu_data.x86_vendor_id[8],
-	      (unsigned int *)&boot_cpu_data.x86_vendor_id[4]);
-
-	/* get cpu type */
-	cpuid(1, &eax, &dummy, &dummy,
-		(unsigned int *) &boot_cpu_data.x86_capability);
-	boot_cpu_data.x86 = (eax >> 8) & 0xf;
-	boot_cpu_data.x86_model = (eax >> 4) & 0xf;
-	boot_cpu_data.x86_mask = eax & 0xf;
-}
-
 void __init x86_64_start_kernel(char * real_mode_data)
 {
 	char *s;
@@ -117,6 +99,5 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
 		panic("Kernel too big for kernel mapping\n");
 
-	setup_boot_cpu_data();
 	start_kernel();
 }
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index ddc3b2d9a165..73f1cdd140fe 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -545,10 +545,10 @@ void __init setup_arch(char **cmdline_p)
 	data_resource.start = virt_to_phys(&_etext);
 	data_resource.end = virt_to_phys(&_edata)-1;
 
-	parse_cmdline_early(cmdline_p);
-
 	early_identify_cpu(&boot_cpu_data);
 
+	parse_cmdline_early(cmdline_p);
+
 	/*
 	 * partially used pages are not usable - thus
 	 * we are rounding upwards:
-- 
cgit v1.2.1


From 33df0d19ea425d28bd5afb48898af32237fe81af Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Allow early_param and identical __setup to exist

We currently assume that boot parameters which are handled by
early_param() will not overlap boot parameters handled by __setup: if
they do, behaviour is dependent on link order, usually meaning __setup
will not get called.

ACPI wants to use early_param("pci"), and pci uses __setup("pci="), so
we modify the core to let them coexist: "pci=noacpi" will now get
passed to both.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 init/main.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/init/main.c b/init/main.c
index 8651a720a092..6b69564738c1 100644
--- a/init/main.c
+++ b/init/main.c
@@ -162,16 +162,19 @@ extern struct obs_kernel_param __setup_start[], __setup_end[];
 static int __init obsolete_checksetup(char *line)
 {
 	struct obs_kernel_param *p;
+	int had_early_param = 0;
 
 	p = __setup_start;
 	do {
 		int n = strlen(p->str);
 		if (!strncmp(line, p->str, n)) {
 			if (p->early) {
-				/* Already done in parse_early_param?  (Needs
-				 * exact match on param part) */
+				/* Already done in parse_early_param?
+				 * (Needs exact match on param part).
+				 * Keep iterating, as we can have early
+				 * params and __setups of same names 8( */
 				if (line[n] == '\0' || line[n] == '=')
-					return 1;
+					had_early_param = 1;
 			} else if (!p->setup_func) {
 				printk(KERN_WARNING "Parameter %s is obsolete,"
 				       " ignored\n", p->str);
@@ -181,7 +184,8 @@ static int __init obsolete_checksetup(char *line)
 		}
 		p++;
 	} while (p < __setup_end);
-	return 0;
+
+	return had_early_param;
 }
 
 /*
-- 
cgit v1.2.1


From 1a3f239ddf9208f2e52d36fef1c1c4518cbbbabe Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] i386: Replace i386 open-coded cmdline parsing with

This patch replaces the open-coded early commandline parsing
throughout the i386 boot code with the generic mechanism (already used
by ppc, powerpc, ia64 and s390).  The code was inconsistent with
whether it deletes the option from the cmdline or not, meaning some of
these will get passed through the environment into init.

This transformation is mainly mechanical, but there are some notable
parts:

1) Grammar: s/linux never set's it up/linux never sets it up/

2) Remove hacked-in earlyprintk= option scanning.  When someone
   actually implements CONFIG_EARLY_PRINTK, then they can use
   early_param().
[AK: actually it is implemented, but I'm adding the early_param it in the next
x86-64 patch]

3) Move declaration of generic_apic_probe() from setup.c into asm/apic.h

4) Various parameters now moved into their appropriate files (thanks Andi).

5) All parse functions which examine arg need to check for NULL,
   except one where it has subtle humor value.

AK: readded acpi_sci handling which was completely dropped
AK: moved some more variables into acpi/boot.c

Cc: len.brown@intel.com

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c     |  76 ++++++++-
 arch/i386/kernel/apic.c          |  15 ++
 arch/i386/kernel/io_apic.c       |  24 ++-
 arch/i386/kernel/machine_kexec.c |  23 +++
 arch/i386/kernel/setup.c         | 350 +++++++++++++--------------------------
 arch/i386/kernel/smpboot.c       |  13 ++
 arch/i386/mach-generic/probe.c   |  58 ++++---
 arch/i386/mm/init.c              |  18 +-
 include/asm-i386/acpi.h          |  14 --
 include/asm-i386/apic.h          |   4 +-
 include/asm-i386/io_apic.h       |  11 ++
 include/asm-i386/pgtable.h       |   2 -
 12 files changed, 319 insertions(+), 289 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index ee003bc0e8b1..87e2ab50b694 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -36,6 +36,8 @@
 #include <asm/io.h>
 #include <asm/mpspec.h>
 
+int __initdata acpi_force = 0;
+
 #ifdef	CONFIG_X86_64
 
 extern void __init clustered_apic_check(void);
@@ -860,8 +862,6 @@ static void __init acpi_process_madt(void)
 	return;
 }
 
-extern int acpi_force;
-
 #ifdef __i386__
 
 static int __init disable_acpi_irq(struct dmi_system_id *d)
@@ -1163,3 +1163,75 @@ int __init acpi_boot_init(void)
 
 	return 0;
 }
+
+static int __init parse_acpi(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
+
+	/* "acpi=off" disables both ACPI table parsing and interpreter */
+	if (strcmp(arg, "off") == 0) {
+		disable_acpi();
+	}
+	/* acpi=force to over-ride black-list */
+	else if (strcmp(arg, "force") == 0) {
+		acpi_force = 1;
+		acpi_ht = 1;
+		acpi_disabled = 0;
+	}
+	/* acpi=strict disables out-of-spec workarounds */
+	else if (strcmp(arg, "strict") == 0) {
+		acpi_strict = 1;
+	}
+	/* Limit ACPI just to boot-time to enable HT */
+	else if (strcmp(arg, "ht") == 0) {
+		if (!acpi_force)
+			disable_acpi();
+		acpi_ht = 1;
+	}
+	/* "acpi=noirq" disables ACPI interrupt routing */
+	else if (strcmp(arg, "noirq") == 0) {
+		acpi_noirq_set();
+	} else {
+		/* Core will printk when we return error. */
+		return -EINVAL;
+	}
+	return 0;
+}
+early_param("acpi", parse_acpi);
+
+/* FIXME: Using pci= for an ACPI parameter is a travesty. */
+static int __init parse_pci(char *arg)
+{
+	if (arg && strcmp(arg, "noacpi") == 0)
+		acpi_disable_pci();
+	return 0;
+}
+early_param("pci", parse_pci);
+
+#ifdef CONFIG_X86_IO_APIC
+static int __init parse_acpi_skip_timer_override(char *arg)
+{
+	acpi_skip_timer_override = 1;
+	return 0;
+}
+early_param("acpi_skip_timer_override", parse_acpi_skip_timer_override);
+#endif /* CONFIG_X86_IO_APIC */
+
+static int __init setup_acpi_sci(char *s)
+{
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "edge"))
+		acpi_sci_flags.trigger = 1;
+	else if (!strcmp(s, "level"))
+		acpi_sci_flags.trigger = 3;
+	else if (!strcmp(s, "high"))
+		acpi_sci_flags.polarity = 1;
+	else if (!strcmp(s, "low"))
+		acpi_sci_flags.polarity = 3;
+	else
+		return -EINVAL;
+	return 0;
+}
+early_param("acpi_sci", setup_acpi_sci);
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 1a34fc57800b..ce75e71b694f 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -1372,3 +1372,18 @@ int __init APIC_init_uniprocessor (void)
 
 	return 0;
 }
+
+static int __init parse_lapic(char *arg)
+{
+	lapic_enable();
+	return 0;
+}
+early_param("lapic", parse_lapic);
+
+static int __init parse_nolapic(char *arg)
+{
+	lapic_disable();
+	return 0;
+}
+early_param("nolapic", parse_nolapic);
+
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index b713f27d7e86..fd0df75cfbda 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -66,7 +66,7 @@ int sis_apic_bug = -1;
  */
 int nr_ioapic_registers[MAX_IO_APICS];
 
-int disable_timer_pin_1 __initdata;
+static int disable_timer_pin_1 __initdata;
 
 /*
  * Rough estimation of how many shared IRQs there are, can
@@ -2691,3 +2691,25 @@ int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int a
 }
 
 #endif /* CONFIG_ACPI */
+
+static int __init parse_disable_timer_pin_1(char *arg)
+{
+	disable_timer_pin_1 = 1;
+	return 0;
+}
+early_param("disable_timer_pin_1", parse_disable_timer_pin_1);
+
+static int __init parse_enable_timer_pin_1(char *arg)
+{
+	disable_timer_pin_1 = -1;
+	return 0;
+}
+early_param("enable_timer_pin_1", parse_enable_timer_pin_1);
+
+static int __init parse_noapic(char *arg)
+{
+	/* disable IO-APIC */
+	disable_ioapic_setup();
+	return 0;
+}
+early_param("noapic", parse_noapic);
diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
index 6b1ae6ba76f0..66c3dc99a655 100644
--- a/arch/i386/kernel/machine_kexec.c
+++ b/arch/i386/kernel/machine_kexec.c
@@ -9,6 +9,7 @@
 #include <linux/mm.h>
 #include <linux/kexec.h>
 #include <linux/delay.h>
+#include <linux/init.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/tlbflush.h>
@@ -209,3 +210,25 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	rnk = (relocate_new_kernel_t) reboot_code_buffer;
 	(*rnk)(page_list, reboot_code_buffer, image->start, cpu_has_pae);
 }
+
+/* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel.  By reserving this memory we guarantee
+ * that linux never sets it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+static int __init parse_crashkernel(char *arg)
+{
+	unsigned long size, base;
+	size = memparse(arg, &arg);
+	if (*arg == '@') {
+		base = memparse(arg+1, &arg);
+		/* FIXME: Do I want a sanity check
+		 * to validate the memory range?
+		 */
+		crashk_res.start = base;
+		crashk_res.end   = base + size - 1;
+	}
+	return 0;
+}
+early_param("crashkernel", parse_crashkernel);
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 71a540362b78..c6e31ed386f5 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -96,11 +96,6 @@ unsigned long mmu_cr4_features;
 #endif
 EXPORT_SYMBOL(acpi_disabled);
 
-#ifdef	CONFIG_ACPI
-int __initdata acpi_force = 0;
-extern acpi_interrupt_flags	acpi_sci_flags;
-#endif
-
 /* for MCA, but anyone else can use it if they want */
 unsigned int machine_id;
 #ifdef CONFIG_MCA
@@ -148,7 +143,6 @@ EXPORT_SYMBOL(ist_info);
 struct e820map e820;
 
 extern void early_cpu_init(void);
-extern void generic_apic_probe(char *);
 extern int root_mountflags;
 
 unsigned long saved_videomode;
@@ -700,238 +694,132 @@ static inline void copy_edd(void)
 }
 #endif
 
-static void __init parse_cmdline_early (char ** cmdline_p)
-{
-	char c = ' ', *to = command_line, *from = saved_command_line;
-	int len = 0;
-	int userdef = 0;
+static int __initdata user_defined_memmap = 0;
 
-	/* Save unparsed command line copy for /proc/cmdline */
-	saved_command_line[COMMAND_LINE_SIZE-1] = '\0';
+/*
+ * "mem=nopentium" disables the 4MB page tables.
+ * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
+ * to <mem>, overriding the bios size.
+ * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
+ * <start> to <start>+<mem>, overriding the bios size.
+ *
+ * HPA tells me bootloaders need to parse mem=, so no new
+ * option should be mem=  [also see Documentation/i386/boot.txt]
+ */
+static int __init parse_mem(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-	for (;;) {
-		if (c != ' ')
-			goto next_char;
-		/*
-		 * "mem=nopentium" disables the 4MB page tables.
-		 * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
-		 * to <mem>, overriding the bios size.
-		 * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
-		 * <start> to <start>+<mem>, overriding the bios size.
-		 *
-		 * HPA tells me bootloaders need to parse mem=, so no new
-		 * option should be mem=  [also see Documentation/i386/boot.txt]
+	if (strcmp(arg, "nopentium") == 0) {
+		clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
+		disable_pse = 1;
+	} else {
+		/* If the user specifies memory size, we
+		 * limit the BIOS-provided memory map to
+		 * that size. exactmap can be used to specify
+		 * the exact map. mem=number can be used to
+		 * trim the existing memory map.
 		 */
-		if (!memcmp(from, "mem=", 4)) {
-			if (to != command_line)
-				to--;
-			if (!memcmp(from+4, "nopentium", 9)) {
-				from += 9+4;
-				clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
-				disable_pse = 1;
-			} else {
-				/* If the user specifies memory size, we
-				 * limit the BIOS-provided memory map to
-				 * that size. exactmap can be used to specify
-				 * the exact map. mem=number can be used to
-				 * trim the existing memory map.
-				 */
-				unsigned long long mem_size;
+		unsigned long long mem_size;
  
-				mem_size = memparse(from+4, &from);
-				limit_regions(mem_size);
-				userdef=1;
-			}
-		}
-
-		else if (!memcmp(from, "memmap=", 7)) {
-			if (to != command_line)
-				to--;
-			if (!memcmp(from+7, "exactmap", 8)) {
-#ifdef CONFIG_CRASH_DUMP
-				/* If we are doing a crash dump, we
-				 * still need to know the real mem
-				 * size before original memory map is
-				 * reset.
-				 */
-				find_max_pfn();
-				saved_max_pfn = max_pfn;
-#endif
-				from += 8+7;
-				e820.nr_map = 0;
-				userdef = 1;
-			} else {
-				/* If the user specifies memory size, we
-				 * limit the BIOS-provided memory map to
-				 * that size. exactmap can be used to specify
-				 * the exact map. mem=number can be used to
-				 * trim the existing memory map.
-				 */
-				unsigned long long start_at, mem_size;
- 
-				mem_size = memparse(from+7, &from);
-				if (*from == '@') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_RAM);
-				} else if (*from == '#') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_ACPI);
-				} else if (*from == '$') {
-					start_at = memparse(from+1, &from);
-					add_memory_region(start_at, mem_size, E820_RESERVED);
-				} else {
-					limit_regions(mem_size);
-					userdef=1;
-				}
-			}
-		}
-
-		else if (!memcmp(from, "noexec=", 7))
-			noexec_setup(from + 7);
+		mem_size = memparse(arg, &arg);
+		limit_regions(mem_size);
+		user_defined_memmap = 1;
+	}
+	return 0;
+}
+early_param("mem", parse_mem);
 
+static int __init parse_memmap(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-#ifdef  CONFIG_X86_SMP
-		/*
-		 * If the BIOS enumerates physical processors before logical,
-		 * maxcpus=N at enumeration-time can be used to disable HT.
+	if (strcmp(arg, "exactmap") == 0) {
+#ifdef CONFIG_CRASH_DUMP
+		/* If we are doing a crash dump, we
+		 * still need to know the real mem
+		 * size before original memory map is
+		 * reset.
 		 */
-		else if (!memcmp(from, "maxcpus=", 8)) {
-			extern unsigned int maxcpus;
-
-			maxcpus = simple_strtoul(from + 8, NULL, 0);
-		}
+		find_max_pfn();
+		saved_max_pfn = max_pfn;
 #endif
-
-#ifdef CONFIG_ACPI
-		/* "acpi=off" disables both ACPI table parsing and interpreter */
-		else if (!memcmp(from, "acpi=off", 8)) {
-			disable_acpi();
-		}
-
-		/* acpi=force to over-ride black-list */
-		else if (!memcmp(from, "acpi=force", 10)) {
-			acpi_force = 1;
-			acpi_ht = 1;
-			acpi_disabled = 0;
-		}
-
-		/* acpi=strict disables out-of-spec workarounds */
-		else if (!memcmp(from, "acpi=strict", 11)) {
-			acpi_strict = 1;
-		}
-
-		/* Limit ACPI just to boot-time to enable HT */
-		else if (!memcmp(from, "acpi=ht", 7)) {
-			if (!acpi_force)
-				disable_acpi();
-			acpi_ht = 1;
-		}
-		
-		/* "pci=noacpi" disable ACPI IRQ routing and PCI scan */
-		else if (!memcmp(from, "pci=noacpi", 10)) {
-			acpi_disable_pci();
-		}
-		/* "acpi=noirq" disables ACPI interrupt routing */
-		else if (!memcmp(from, "acpi=noirq", 10)) {
-			acpi_noirq_set();
+		e820.nr_map = 0;
+		user_defined_memmap = 1;
+	} else {
+		/* If the user specifies memory size, we
+		 * limit the BIOS-provided memory map to
+		 * that size. exactmap can be used to specify
+		 * the exact map. mem=number can be used to
+		 * trim the existing memory map.
+		 */
+		unsigned long long start_at, mem_size;
+
+		mem_size = memparse(arg, &arg);
+		if (*arg == '@') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_RAM);
+		} else if (*arg == '#') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_ACPI);
+		} else if (*arg == '$') {
+			start_at = memparse(arg+1, &arg);
+			add_memory_region(start_at, mem_size, E820_RESERVED);
+		} else {
+			limit_regions(mem_size);
+			user_defined_memmap = 1;
 		}
+	}
+	return 0;
+}
+early_param("memmap", parse_memmap);
 
-		else if (!memcmp(from, "acpi_sci=edge", 13))
-			acpi_sci_flags.trigger =  1;
-
-		else if (!memcmp(from, "acpi_sci=level", 14))
-			acpi_sci_flags.trigger = 3;
-
-		else if (!memcmp(from, "acpi_sci=high", 13))
-			acpi_sci_flags.polarity = 1;
-
-		else if (!memcmp(from, "acpi_sci=low", 12))
-			acpi_sci_flags.polarity = 3;
-
-#ifdef CONFIG_X86_IO_APIC
-		else if (!memcmp(from, "acpi_skip_timer_override", 24))
-			acpi_skip_timer_override = 1;
-
-		if (!memcmp(from, "disable_timer_pin_1", 19))
-			disable_timer_pin_1 = 1;
-		if (!memcmp(from, "enable_timer_pin_1", 18))
-			disable_timer_pin_1 = -1;
+#ifdef CONFIG_PROC_VMCORE
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel.
+ */
+static int __init parse_elfcorehdr(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/* disable IO-APIC */
-		else if (!memcmp(from, "noapic", 6))
-			disable_ioapic_setup();
-#endif /* CONFIG_X86_IO_APIC */
-#endif /* CONFIG_ACPI */
+	elfcorehdr_addr = memparse(arg, &arg);
+	return 0;
+}
+early_param("elfcorehdr", parse_elfcorehdr);
+#endif /* CONFIG_PROC_VMCORE */
 
-#ifdef CONFIG_X86_LOCAL_APIC
-		/* enable local APIC */
-		else if (!memcmp(from, "lapic", 5))
-			lapic_enable();
+/*
+ * highmem=size forces highmem to be exactly 'size' bytes.
+ * This works even on boxes that have no highmem otherwise.
+ * This also works to reduce highmem size on bigger boxes.
+ */
+static int __init parse_highmem(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/* disable local APIC */
-		else if (!memcmp(from, "nolapic", 6))
-			lapic_disable();
-#endif /* CONFIG_X86_LOCAL_APIC */
+	highmem_pages = memparse(arg, &arg) >> PAGE_SHIFT;
+	return 0;
+}
+early_param("highmem", parse_highmem);
 
-#ifdef CONFIG_KEXEC
-		/* crashkernel=size@addr specifies the location to reserve for
-		 * a crash kernel.  By reserving this memory we guarantee
-		 * that linux never set's it up as a DMA target.
-		 * Useful for holding code to do something appropriate
-		 * after a kernel panic.
-		 */
-		else if (!memcmp(from, "crashkernel=", 12)) {
-			unsigned long size, base;
-			size = memparse(from+12, &from);
-			if (*from == '@') {
-				base = memparse(from+1, &from);
-				/* FIXME: Do I want a sanity check
-				 * to validate the memory range?
-				 */
-				crashk_res.start = base;
-				crashk_res.end   = base + size - 1;
-			}
-		}
-#endif
-#ifdef CONFIG_PROC_VMCORE
-		/* elfcorehdr= specifies the location of elf core header
-		 * stored by the crashed kernel.
-		 */
-		else if (!memcmp(from, "elfcorehdr=", 11))
-			elfcorehdr_addr = memparse(from+11, &from);
-#endif
+/*
+ * vmalloc=size forces the vmalloc area to be exactly 'size'
+ * bytes. This can be used to increase (or decrease) the
+ * vmalloc area - the default is 128m.
+ */
+static int __init parse_vmalloc(char *arg)
+{
+	if (!arg)
+		return -EINVAL;
 
-		/*
-		 * highmem=size forces highmem to be exactly 'size' bytes.
-		 * This works even on boxes that have no highmem otherwise.
-		 * This also works to reduce highmem size on bigger boxes.
-		 */
-		else if (!memcmp(from, "highmem=", 8))
-			highmem_pages = memparse(from+8, &from) >> PAGE_SHIFT;
-	
-		/*
-		 * vmalloc=size forces the vmalloc area to be exactly 'size'
-		 * bytes. This can be used to increase (or decrease) the
-		 * vmalloc area - the default is 128m.
-		 */
-		else if (!memcmp(from, "vmalloc=", 8))
-			__VMALLOC_RESERVE = memparse(from+8, &from);
-
-	next_char:
-		c = *(from++);
-		if (!c)
-			break;
-		if (COMMAND_LINE_SIZE <= ++len)
-			break;
-		*(to++) = c;
-	}
-	*to = '\0';
-	*cmdline_p = command_line;
-	if (userdef) {
-		printk(KERN_INFO "user-defined physical RAM map:\n");
-		print_memory_map("user");
-	}
+	__VMALLOC_RESERVE = memparse(arg, &arg);
+	return 0;
 }
+early_param("vmalloc", parse_vmalloc);
 
 /*
  * Callback for efi_memory_walk.
@@ -1507,17 +1395,15 @@ void __init setup_arch(char **cmdline_p)
 	data_resource.start = virt_to_phys(_etext);
 	data_resource.end = virt_to_phys(_edata)-1;
 
-	parse_cmdline_early(cmdline_p);
+	parse_early_param();
 
-#ifdef CONFIG_EARLY_PRINTK
-	{
-		char *s = strstr(*cmdline_p, "earlyprintk=");
-		if (s) {
-			setup_early_printk(strchr(s, '=') + 1);
-			printk("early console enabled\n");
-		}
+	if (user_defined_memmap) {
+		printk(KERN_INFO "user-defined physical RAM map:\n");
+		print_memory_map("user");
 	}
-#endif
+
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
 
 	max_low_pfn = setup_memory();
 
@@ -1546,7 +1432,7 @@ void __init setup_arch(char **cmdline_p)
 	dmi_scan_machine();
 
 #ifdef CONFIG_X86_GENERICARCH
-	generic_apic_probe(*cmdline_p);
+	generic_apic_probe();
 #endif	
 	if (efi_enabled)
 		efi_map_memmap();
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 9367af76ce37..517eb3874550 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -1491,3 +1491,16 @@ void __init smp_intr_init(void)
 	/* IPI for generic function call */
 	set_intr_gate(CALL_FUNCTION_VECTOR, call_function_interrupt);
 }
+
+/*
+ * If the BIOS enumerates physical processors before logical,
+ * maxcpus=N at enumeration-time can be used to disable HT.
+ */
+static int __init parse_maxcpus(char *arg)
+{
+	extern unsigned int maxcpus;
+
+	maxcpus = simple_strtoul(arg, NULL, 0);
+	return 0;
+}
+early_param("maxcpus", parse_maxcpus);
diff --git a/arch/i386/mach-generic/probe.c b/arch/i386/mach-generic/probe.c
index 793d1b473251..94b1fd9cbe3c 100644
--- a/arch/i386/mach-generic/probe.c
+++ b/arch/i386/mach-generic/probe.c
@@ -9,6 +9,7 @@
 #include <linux/kernel.h>
 #include <linux/ctype.h>
 #include <linux/init.h>
+#include <linux/errno.h>
 #include <asm/fixmap.h>
 #include <asm/mpspec.h>
 #include <asm/apicdef.h>
@@ -29,7 +30,24 @@ struct genapic *apic_probe[] __initdata = {
 	NULL,
 };
 
-static int cmdline_apic;
+static int cmdline_apic __initdata;
+static int __init parse_apic(char *arg)
+{
+	int i;
+
+	if (!arg)
+		return -EINVAL;
+
+	for (i = 0; apic_probe[i]; i++) {
+		if (!strcmp(apic_probe[i]->name, arg)) {
+			genapic = apic_probe[i];
+			cmdline_apic = 1;
+			return 0;
+		}
+	}
+	return -ENOENT;
+}
+early_param("apic", parse_apic);
 
 void __init generic_bigsmp_probe(void)
 {
@@ -48,40 +66,20 @@ void __init generic_bigsmp_probe(void)
 		}
 }
 
-void __init generic_apic_probe(char *command_line) 
+void __init generic_apic_probe(void)
 { 
-	char *s;
-	int i;
-	int changed = 0;
-
-	s = strstr(command_line, "apic=");
-	if (s && (s == command_line || isspace(s[-1]))) { 
-		char *p = strchr(s, ' '), old; 
-		if (!p)
-			p = strchr(s, '\0'); 
-		old = *p; 
-		*p = 0; 
-		for (i = 0; !changed && apic_probe[i]; i++) {
-			if (!strcmp(apic_probe[i]->name, s+5)) { 
-				changed = 1;
+	if (!cmdline_apic) {
+		int i;
+		for (i = 0; apic_probe[i]; i++) {
+			if (apic_probe[i]->probe()) {
 				genapic = apic_probe[i];
+				break;
 			}
 		}
-		if (!changed)
-			printk(KERN_ERR "Unknown genapic `%s' specified.\n", s);
-		*p = old;
-		cmdline_apic = changed;
-	} 
-	for (i = 0; !changed && apic_probe[i]; i++) { 
-		if (apic_probe[i]->probe()) {
-			changed = 1;
-			genapic = apic_probe[i]; 
-		} 
+		/* Not visible without early console */
+		if (!apic_probe[i])
+			panic("Didn't find an APIC driver");
 	}
-	/* Not visible without early console */ 
-	if (!changed) 
-		panic("Didn't find an APIC driver"); 
-
 	printk(KERN_INFO "Using APIC driver %s\n", genapic->name);
 } 
 
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index d0c2fdf6a4d7..951386606d09 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -435,16 +435,22 @@ u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
  * on      Enable
  * off     Disable
  */
-void __init noexec_setup(const char *str)
+static int __init noexec_setup(char *str)
 {
-	if (!strncmp(str, "on",2) && cpu_has_nx) {
-		__supported_pte_mask |= _PAGE_NX;
-		disable_nx = 0;
-	} else if (!strncmp(str,"off",3)) {
+	if (!str || !strcmp(str, "on")) {
+		if (cpu_has_nx) {
+			__supported_pte_mask |= _PAGE_NX;
+			disable_nx = 0;
+		}
+	} else if (!strcmp(str,"off")) {
 		disable_nx = 1;
 		__supported_pte_mask &= ~_PAGE_NX;
-	}
+	} else
+		return -EINVAL;
+
+	return 0;
 }
+early_param("noexec", noexec_setup);
 
 int nx_enabled = 0;
 #ifdef CONFIG_X86_PAE
diff --git a/include/asm-i386/acpi.h b/include/asm-i386/acpi.h
index 20f523954218..6016632d032f 100644
--- a/include/asm-i386/acpi.h
+++ b/include/asm-i386/acpi.h
@@ -131,21 +131,7 @@ static inline void disable_acpi(void)
 extern int acpi_gsi_to_irq(u32 gsi, unsigned int *irq);
 
 #ifdef CONFIG_X86_IO_APIC
-extern int skip_ioapic_setup;
 extern int acpi_skip_timer_override;
-
-static inline void disable_ioapic_setup(void)
-{
-	skip_ioapic_setup = 1;
-}
-
-static inline int ioapic_setup_disabled(void)
-{
-	return skip_ioapic_setup;
-}
-
-#else
-static inline void disable_ioapic_setup(void) { }
 #endif
 
 static inline void acpi_noirq_set(void) { acpi_noirq = 1; }
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index 2c1e371cebb6..8c5331ceca7b 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -42,6 +42,8 @@ static inline void lapic_enable(void)
 	} while (0)
 
 
+extern void generic_apic_probe(void);
+
 #ifdef CONFIG_X86_LOCAL_APIC
 
 /*
@@ -117,8 +119,6 @@ extern void enable_APIC_timer(void);
 
 extern void enable_NMI_through_LVT0 (void * dummy);
 
-extern int disable_timer_pin_1;
-
 void smp_send_timer_broadcast_ipi(struct pt_regs *regs);
 void switch_APIC_timer_to_ipi(void *cpumask);
 void switch_ipi_to_APIC_timer(void *cpumask);
diff --git a/include/asm-i386/io_apic.h b/include/asm-i386/io_apic.h
index 5092e819b8a2..5d309275a1dc 100644
--- a/include/asm-i386/io_apic.h
+++ b/include/asm-i386/io_apic.h
@@ -188,6 +188,16 @@ static inline void io_apic_modify(unsigned int apic, unsigned int reg, unsigned
 /* 1 if "noapic" boot option passed */
 extern int skip_ioapic_setup;
 
+static inline void disable_ioapic_setup(void)
+{
+	skip_ioapic_setup = 1;
+}
+
+static inline int ioapic_setup_disabled(void)
+{
+	return skip_ioapic_setup;
+}
+
 /*
  * If we use the IO-APIC for IRQ routing, disable automatic
  * assignment of PCI IRQ's.
@@ -206,6 +216,7 @@ extern int (*ioapic_renumber_irq)(int ioapic, int irq);
 
 #else  /* !CONFIG_X86_IO_APIC */
 #define io_apic_assign_pci_irqs 0
+static inline void disable_ioapic_setup(void) { }
 #endif
 
 extern int assign_irq_vector(int irq);
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 09697fec3d2b..64140f2f1b95 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -391,8 +391,6 @@ extern pte_t *lookup_address(unsigned long address);
  static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
 #endif
 
-extern void noexec_setup(const char *str);
-
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
 	((pte_t *)kmap_atomic(pmd_page(*(dir)),KM_PTE0) + pte_index(address))
-- 
cgit v1.2.1


From 2c8c0e6b8d7700a990da8d24eff767f9ca223b96 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Convert x86-64 to early param

Instead of hackish manual parsing

Requires earlier i386 patchkit, but also fixes i386 early_printk again.

I removed some obsolete really early parameters which didn't do anything useful.
Also made a few parameters that needed it early (mostly oops printing setup)

Also removed one panic check that wasn't visible without
early console anyways (the early console is now initialized after that
panic)

This cleans up a lot of code.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c          |  34 ++++---
 arch/x86_64/kernel/e820.c          |  53 ++++++++--
 arch/x86_64/kernel/early_printk.c  |  20 ++--
 arch/x86_64/kernel/head64.c        |  15 ---
 arch/x86_64/kernel/io_apic.c       |  15 ++-
 arch/x86_64/kernel/machine_kexec.c |  28 ++++++
 arch/x86_64/kernel/pci-dma.c       |   7 +-
 arch/x86_64/kernel/setup.c         | 197 ++++---------------------------------
 arch/x86_64/kernel/setup64.c       |   8 +-
 arch/x86_64/kernel/smpboot.c       |   6 +-
 arch/x86_64/kernel/traps.c         |  24 +++--
 arch/x86_64/mm/numa.c              |   9 +-
 include/asm-x86_64/apic.h          |   4 +-
 include/asm-x86_64/e820.h          |   3 +-
 include/asm-x86_64/pgtable.h       |   1 -
 include/asm-x86_64/proto.h         |  10 --
 16 files changed, 159 insertions(+), 275 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 496cd1bd2ae5..660602c51f51 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -36,6 +36,7 @@
 #include <asm/idle.h>
 #include <asm/proto.h>
 #include <asm/timex.h>
+#include <asm/apic.h>
 
 int apic_verbosity;
 int apic_runs_main_timer;
@@ -546,18 +547,24 @@ static void apic_pm_activate(void) { }
 
 static int __init apic_set_verbosity(char *str)
 {
+	if (str == NULL)  {
+		skip_ioapic_setup = 0;
+		ioapic_force = 1;
+		return 0;
+	}
 	if (strcmp("debug", str) == 0)
 		apic_verbosity = APIC_DEBUG;
 	else if (strcmp("verbose", str) == 0)
 		apic_verbosity = APIC_VERBOSE;
-	else
+	else {
 		printk(KERN_WARNING "APIC Verbosity level %s not recognised"
-				" use apic=verbose or apic=debug", str);
+				" use apic=verbose or apic=debug\n", str);
+		return -EINVAL;
+	}
 
-	return 1;
+	return 0;
 }
-
-__setup("apic=", apic_set_verbosity);
+early_param("apic", apic_set_verbosity);
 
 /*
  * Detect and enable local APICs on non-SMP boards.
@@ -1078,14 +1085,17 @@ int __init APIC_init_uniprocessor (void)
 static __init int setup_disableapic(char *str) 
 { 
 	disable_apic = 1;
-	return 1;
-} 
+	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+	return 0;
+}
+early_param("disableapic", setup_disableapic);
 
+/* same as disableapic, for compatibility */
 static __init int setup_nolapic(char *str) 
 { 
-	disable_apic = 1;
-	return 1;
+	return setup_disableapic(str);
 } 
+early_param("nolapic", setup_nolapic);
 
 static __init int setup_noapictimer(char *str) 
 { 
@@ -1118,11 +1128,5 @@ static __init int setup_apicpmtimer(char *s)
 }
 __setup("apicpmtimer", setup_apicpmtimer);
 
-/* dummy parsing: see setup.c */
-
-__setup("disableapic", setup_disableapic); 
-__setup("nolapic", setup_nolapic);  /* same as disableapic, for compatibility */
-
 __setup("noapictimer", setup_noapictimer); 
 
-/* no "lapic" flag - we only use the lapic when the BIOS tells us so. */
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 1362aad4f3aa..a4739dc82937 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -596,31 +596,64 @@ void __init setup_memory_region(void)
 	e820_print_map(who);
 }
 
-void __init parse_memopt(char *p, char **from) 
-{ 
-	end_user_pfn = memparse(p, from);
+static int __init parse_memopt(char *p)
+{
+	if (!p)
+		return -EINVAL;
+	end_user_pfn = memparse(p, &p);
 	end_user_pfn >>= PAGE_SHIFT;	
+	return 0;
 } 
+early_param("mem", parse_memopt);
+
+static int userdef __initdata;
 
-void __init parse_memmapopt(char *p, char **from)
+static int __init parse_memmap_opt(char *p)
 {
+	char *oldp;
 	unsigned long long start_at, mem_size;
 
-	mem_size = memparse(p, from);
-	p = *from;
+	if (!strcmp(p, "exactmap")) {
+#ifdef CONFIG_CRASH_DUMP
+		/* If we are doing a crash dump, we
+		 * still need to know the real mem
+		 * size before original memory map is
+		 * reset.
+		 */
+		saved_max_pfn = e820_end_of_ram();
+#endif
+		end_pfn_map = 0;
+		e820.nr_map = 0;
+		userdef = 1;
+		return 0;
+	}
+
+	oldp = p;
+	mem_size = memparse(p, &p);
+	if (p == oldp)
+		return -EINVAL;
 	if (*p == '@') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_RAM);
 	} else if (*p == '#') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_ACPI);
 	} else if (*p == '$') {
-		start_at = memparse(p+1, from);
+		start_at = memparse(p+1, &p);
 		add_memory_region(start_at, mem_size, E820_RESERVED);
 	} else {
 		end_user_pfn = (mem_size >> PAGE_SHIFT);
 	}
-	p = *from;
+	return *p == '\0' ? 0 : -EINVAL;
+}
+early_param("memmap", parse_memmap_opt);
+
+void finish_e820_parsing(void)
+{
+	if (userdef) {
+		printk(KERN_INFO "user-defined physical RAM map:\n");
+		e820_print_map("user");
+	}
 }
 
 unsigned long pci_mem_start = 0xaeedbabe;
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index 140051e07fa6..e22ecd54870d 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -215,20 +215,16 @@ void early_printk(const char *fmt, ...)
 
 static int __initdata keep_early;
 
-int __init setup_early_printk(char *opt)
+static int __init setup_early_printk(char *buf)
 {
-	char *space;
-	char buf[256];
+	if (!buf)
+		return 0;
 
 	if (early_console_initialized)
-		return 1;
-
-	strlcpy(buf,opt,sizeof(buf));
-	space = strchr(buf, ' ');
-	if (space)
-		*space = 0;
+		return 0;
+	early_console_initialized = 1;
 
-	if (strstr(buf,"keep"))
+	if (!strcmp(buf,"keep"))
 		keep_early = 1;
 
 	if (!strncmp(buf, "serial", 6)) {
@@ -248,11 +244,12 @@ int __init setup_early_printk(char *opt)
  		early_console = &simnow_console;
  		keep_early = 1;
 	}
-	early_console_initialized = 1;
 	register_console(early_console);
 	return 0;
 }
 
+early_param("earlyprintk", setup_early_printk);
+
 void __init disable_early_printk(void)
 {
 	if (!early_console_initialized || !early_console)
@@ -266,4 +263,3 @@ void __init disable_early_printk(void)
 	}
 }
 
-__setup("earlyprintk=", setup_early_printk);
diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index f2461fde9f8f..6716cbfc34ac 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -58,7 +58,6 @@ static void __init copy_bootdata(char *real_mode_data)
 
 void __init x86_64_start_kernel(char * real_mode_data)
 {
-	char *s;
 	int i;
 
 	for (i = 0; i < 256; i++)
@@ -85,19 +84,5 @@ void __init x86_64_start_kernel(char * real_mode_data)
 #ifdef CONFIG_SMP
 	cpu_set(0, cpu_online_map);
 #endif
-	s = strstr(saved_command_line, "earlyprintk=");
-	if (s != NULL)
-		setup_early_printk(strchr(s, '=') + 1);
-#ifdef CONFIG_NUMA
-	s = strstr(saved_command_line, "numa=");
-	if (s != NULL)
-		numa_setup(s+5);
-#endif
-	if (strstr(saved_command_line, "disableapic"))
-		disable_apic = 1;
-	/* You need early console to see that */
-	if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
-		panic("Kernel too big for kernel mapping\n");
-
 	start_kernel();
 }
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index a1412042b918..afac3dbb3729 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -48,7 +48,7 @@ int sis_apic_bug; /* not actually supported, dummy for compile */
 
 static int no_timer_check;
 
-int disable_timer_pin_1 __initdata;
+static int disable_timer_pin_1 __initdata;
 
 int timer_over_8254 __initdata = 0;
 
@@ -253,18 +253,17 @@ int ioapic_force;
 static int __init disable_ioapic_setup(char *str)
 {
 	skip_ioapic_setup = 1;
-	return 1;
+	return 0;
 }
+early_param("noapic", disable_ioapic_setup);
 
-static int __init enable_ioapic_setup(char *str)
+/* Actually the next is obsolete, but keep it for paranoid reasons -AK */
+static int __init disable_timer_pin_setup(char *arg)
 {
-	ioapic_force = 1;
-	skip_ioapic_setup = 0;
+	disable_timer_pin_1 = 1;
 	return 1;
 }
-
-__setup("noapic", disable_ioapic_setup);
-__setup("apic", enable_ioapic_setup);
+__setup("disable_timer_pin_1", disable_timer_pin_setup);
 
 static int __init setup_disable_8254_timer(char *s)
 {
diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 106076b370fc..2e94c072d84a 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -226,3 +226,31 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	rnk = (relocate_new_kernel_t) control_code_buffer;
 	(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
 }
+
+/* crashkernel=size@addr specifies the location to reserve for
+ * a crash kernel.  By reserving this memory we guarantee
+ * that linux never set's it up as a DMA target.
+ * Useful for holding code to do something appropriate
+ * after a kernel panic.
+ */
+static int __init setup_crashkernel(char *arg)
+{
+	unsigned long size, base;
+	char *p;
+	if (!arg)
+		return -EINVAL;
+	size = memparse(arg, &p);
+	if (arg == p)
+		return -EINVAL;
+	if (*p == '@') {
+		base = memparse(p+1, &p);
+		/* FIXME: Do I want a sanity check to validate the
+		 * memory range?  Yes you do, but it's too early for
+		 * e820 -AK */
+		crashk_res.start = base;
+		crashk_res.end   = base + size - 1;
+	}
+	return 0;
+}
+early_param("crashkernel", setup_crashkernel);
+
diff --git a/arch/x86_64/kernel/pci-dma.c b/arch/x86_64/kernel/pci-dma.c
index 9c44f4f2433d..4dcb671bd19f 100644
--- a/arch/x86_64/kernel/pci-dma.c
+++ b/arch/x86_64/kernel/pci-dma.c
@@ -236,6 +236,9 @@ __init int iommu_setup(char *p)
 {
     iommu_merge = 1;
 
+	if (!p)
+		return -EINVAL;
+
     while (*p) {
 	    if (!strncmp(p,"off",3))
 		    no_iommu = 1;
@@ -278,9 +281,9 @@ __init int iommu_setup(char *p)
 	    if (*p == ',')
 		    ++p;
     }
-    return 1;
+    return 0;
 }
-__setup("iommu=", iommu_setup);
+early_param("iommu", iommu_setup);
 
 void __init pci_iommu_alloc(void)
 {
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 73f1cdd140fe..f55540a6084e 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -76,11 +76,6 @@ unsigned long mmu_cr4_features;
 
 int acpi_disabled;
 EXPORT_SYMBOL(acpi_disabled);
-#ifdef	CONFIG_ACPI
-extern int __initdata acpi_ht;
-extern acpi_interrupt_flags	acpi_sci_flags;
-int __initdata acpi_force = 0;
-#endif
 
 int acpi_numa __initdata;
 
@@ -276,183 +271,22 @@ static void __init probe_roms(void)
 	}
 }
 
-/* Check for full argument with no trailing characters */
-static int fullarg(char *p, char *arg)
+#ifdef CONFIG_PROC_VMCORE
+/* elfcorehdr= specifies the location of elf core header
+ * stored by the crashed kernel. This option will be passed
+ * by kexec loader to the capture kernel.
+ */
+static int __init setup_elfcorehdr(char *arg)
 {
-	int l = strlen(arg);
-	return !memcmp(p, arg, l) && (p[l] == 0 || isspace(p[l]));
+	char *end;
+	if (!arg)
+		return -EINVAL;
+	elfcorehdr_addr = memparse(arg, &end);
+	return end > arg ? 0 : -EINVAL;
 }
-
-static __init void parse_cmdline_early (char ** cmdline_p)
-{
-	char c = ' ', *to = command_line, *from = COMMAND_LINE;
-	int len = 0;
-	int userdef = 0;
-
-	for (;;) {
-		if (c != ' ') 
-			goto next_char; 
-
-#ifdef  CONFIG_SMP
-		/*
-		 * If the BIOS enumerates physical processors before logical,
-		 * maxcpus=N at enumeration-time can be used to disable HT.
-		 */
-		else if (!memcmp(from, "maxcpus=", 8)) {
-			extern unsigned int maxcpus;
-
-			maxcpus = simple_strtoul(from + 8, NULL, 0);
-		}
-#endif
-#ifdef CONFIG_ACPI
-		/* "acpi=off" disables both ACPI table parsing and interpreter init */
-		if (fullarg(from,"acpi=off"))
-			disable_acpi();
-
-		if (fullarg(from, "acpi=force")) { 
-			/* add later when we do DMI horrors: */
-			acpi_force = 1;
-			acpi_disabled = 0;
-		}
-
-		/* acpi=ht just means: do ACPI MADT parsing 
-		   at bootup, but don't enable the full ACPI interpreter */
-		if (fullarg(from, "acpi=ht")) { 
-			if (!acpi_force)
-				disable_acpi();
-			acpi_ht = 1; 
-		}
-                else if (fullarg(from, "pci=noacpi")) 
-			acpi_disable_pci();
-		else if (fullarg(from, "acpi=noirq"))
-			acpi_noirq_set();
-
-		else if (fullarg(from, "acpi_sci=edge"))
-			acpi_sci_flags.trigger =  1;
-		else if (fullarg(from, "acpi_sci=level"))
-			acpi_sci_flags.trigger = 3;
-		else if (fullarg(from, "acpi_sci=high"))
-			acpi_sci_flags.polarity = 1;
-		else if (fullarg(from, "acpi_sci=low"))
-			acpi_sci_flags.polarity = 3;
-
-		/* acpi=strict disables out-of-spec workarounds */
-		else if (fullarg(from, "acpi=strict")) {
-			acpi_strict = 1;
-		}
-		else if (fullarg(from, "acpi_skip_timer_override"))
-			acpi_skip_timer_override = 1;
+early_param("elfcorehdr", setup_elfcorehdr);
 #endif
 
-		if (fullarg(from, "disable_timer_pin_1"))
-			disable_timer_pin_1 = 1;
-		if (fullarg(from, "enable_timer_pin_1"))
-			disable_timer_pin_1 = -1;
-
-		if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
-			clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
-			disable_apic = 1;
-		}
-
-		if (fullarg(from, "noapic"))
-			skip_ioapic_setup = 1;
-
-		if (fullarg(from,"apic")) {
-			skip_ioapic_setup = 0;
-			ioapic_force = 1;
-		}
-			
-		if (!memcmp(from, "mem=", 4))
-			parse_memopt(from+4, &from); 
-
-		if (!memcmp(from, "memmap=", 7)) {
-			/* exactmap option is for used defined memory */
-			if (!memcmp(from+7, "exactmap", 8)) {
-#ifdef CONFIG_CRASH_DUMP
-				/* If we are doing a crash dump, we
-				 * still need to know the real mem
-				 * size before original memory map is
-				 * reset.
-				 */
-				saved_max_pfn = e820_end_of_ram();
-#endif
-				from += 8+7;
-				end_pfn_map = 0;
-				e820.nr_map = 0;
-				userdef = 1;
-			}
-			else {
-				parse_memmapopt(from+7, &from);
-				userdef = 1;
-			}
-		}
-
-#ifdef CONFIG_NUMA
-		if (!memcmp(from, "numa=", 5))
-			numa_setup(from+5); 
-#endif
-
-		if (!memcmp(from,"iommu=",6)) { 
-			iommu_setup(from+6); 
-		}
-
-		if (fullarg(from,"oops=panic"))
-			panic_on_oops = 1;
-
-		if (!memcmp(from, "noexec=", 7))
-			nonx_setup(from + 7);
-
-#ifdef CONFIG_KEXEC
-		/* crashkernel=size@addr specifies the location to reserve for
-		 * a crash kernel.  By reserving this memory we guarantee
-		 * that linux never set's it up as a DMA target.
-		 * Useful for holding code to do something appropriate
-		 * after a kernel panic.
-		 */
-		else if (!memcmp(from, "crashkernel=", 12)) {
-			unsigned long size, base;
-			size = memparse(from+12, &from);
-			if (*from == '@') {
-				base = memparse(from+1, &from);
-				/* FIXME: Do I want a sanity check
-				 * to validate the memory range?
-				 */
-				crashk_res.start = base;
-				crashk_res.end   = base + size - 1;
-			}
-		}
-#endif
-
-#ifdef CONFIG_PROC_VMCORE
-		/* elfcorehdr= specifies the location of elf core header
-		 * stored by the crashed kernel. This option will be passed
-		 * by kexec loader to the capture kernel.
-		 */
-		else if(!memcmp(from, "elfcorehdr=", 11))
-			elfcorehdr_addr = memparse(from+11, &from);
-#endif
-
-#ifdef CONFIG_HOTPLUG_CPU
-		else if (!memcmp(from, "additional_cpus=", 16))
-			setup_additional_cpus(from+16);
-#endif
-
-	next_char:
-		c = *(from++);
-		if (!c)
-			break;
-		if (COMMAND_LINE_SIZE <= ++len)
-			break;
-		*(to++) = c;
-	}
-	if (userdef) {
-		printk(KERN_INFO "user-defined physical RAM map:\n");
-		e820_print_map("user");
-	}
-	*to = '\0';
-	*cmdline_p = command_line;
-}
-
 #ifndef CONFIG_NUMA
 static void __init
 contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
@@ -547,7 +381,12 @@ void __init setup_arch(char **cmdline_p)
 
 	early_identify_cpu(&boot_cpu_data);
 
-	parse_cmdline_early(cmdline_p);
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
+
+	parse_early_param();
+
+	finish_e820_parsing();
 
 	/*
 	 * partially used pages are not usable - thus
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 417de564456e..646caa0a20bb 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -46,8 +46,10 @@ Control non executable mappings for 64bit processes.
 on	Enable(default)
 off	Disable
 */ 
-int __init nonx_setup(char *str)
+static int __init nonx_setup(char *str)
 {
+	if (!str)
+		return -EINVAL;
 	if (!strncmp(str, "on", 2)) {
                 __supported_pte_mask |= _PAGE_NX; 
  		do_not_nx = 0; 
@@ -55,9 +57,9 @@ int __init nonx_setup(char *str)
 		do_not_nx = 1;
 		__supported_pte_mask &= ~_PAGE_NX;
         }
-	return 1;
+	return 0;
 } 
-__setup("noexec=", nonx_setup);	/* parsed early actually */
+early_param("noexec", nonx_setup);
 
 int force_personality32 = 0; 
 
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index ae7d6d84e352..0b9f28b05e7d 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1270,11 +1270,11 @@ void __cpu_die(unsigned int cpu)
  	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
 
-__init int setup_additional_cpus(char *s)
+static __init int setup_additional_cpus(char *s)
 {
-	return get_option(&s, &additional_cpus);
+	return s && get_option(&s, &additional_cpus) ? 0 : -EINVAL;
 }
-__setup("additional_cpus=", setup_additional_cpus);
+early_param("additional_cpus", setup_additional_cpus);
 
 #else /* ... !CONFIG_HOTPLUG_CPU */
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 56d7ff0c894c..9ec2b1d5893d 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -1119,24 +1119,30 @@ void __init trap_init(void)
 }
 
 
-/* Actual parsing is done early in setup.c. */
-static int __init oops_dummy(char *s)
+static int __init oops_setup(char *s)
 { 
-	panic_on_oops = 1;
-	return 1;
+	if (!s)
+		return -EINVAL;
+	if (!strcmp(s, "panic"))
+		panic_on_oops = 1;
+	return 0;
 } 
-__setup("oops=", oops_dummy); 
+early_param("oops", oops_setup);
 
 static int __init kstack_setup(char *s)
 {
+	if (!s)
+		return -EINVAL;
 	kstack_depth_to_print = simple_strtoul(s,NULL,0);
-	return 1;
+	return 0;
 }
-__setup("kstack=", kstack_setup);
+early_param("kstack", kstack_setup);
 
 #ifdef CONFIG_STACK_UNWIND
 static int __init call_trace_setup(char *s)
 {
+	if (!s)
+		return -EINVAL;
 	if (strcmp(s, "old") == 0)
 		call_trace = -1;
 	else if (strcmp(s, "both") == 0)
@@ -1145,7 +1151,7 @@ static int __init call_trace_setup(char *s)
 		call_trace = 1;
 	else if (strcmp(s, "new") == 0)
 		call_trace = 2;
-	return 1;
+	return 0;
 }
-__setup("call_trace=", call_trace_setup);
+early_param("call_trace", call_trace_setup);
 #endif
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index b2fac14baac0..d64d6d93d04b 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -348,9 +348,10 @@ void __init paging_init(void)
 	}
 } 
 
-/* [numa=off] */
-__init int numa_setup(char *opt) 
+static __init int numa_setup(char *opt)
 { 
+	if (!opt)
+		return -EINVAL;
 	if (!strncmp(opt,"off",3))
 		numa_off = 1;
 #ifdef CONFIG_NUMA_EMU
@@ -366,9 +367,11 @@ __init int numa_setup(char *opt)
 	if (!strncmp(opt,"hotadd=", 7))
 		hotadd_percent = simple_strtoul(opt+7, NULL, 10);
 #endif
-	return 1;
+	return 0;
 } 
 
+early_param("numa", numa_setup);
+
 /*
  * Setup early cpu_to_node.
  *
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 8ed0f4d67b8d..29ee735278f2 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -17,6 +17,7 @@
 
 extern int apic_verbosity;
 extern int apic_runs_main_timer;
+extern int ioapic_force;
 
 /*
  * Define the default level of output to be very little
@@ -93,9 +94,6 @@ extern void setup_APIC_extened_lvt(unsigned char lvt_off, unsigned char vector,
 #define K8_APIC_EXT_INT_MSG_EXT 0x7
 #define K8_APIC_EXT_LVT_ENTRY_THRESHOLD    0
 
-extern int disable_timer_pin_1;
-
-
 void smp_send_timer_broadcast_ipi(void);
 void switch_APIC_timer_to_ipi(void *cpumask);
 void switch_ipi_to_APIC_timer(void *cpumask);
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 670a3388e70a..dba012151856 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -55,8 +55,7 @@ extern void e820_setup_gap(void);
 extern unsigned long e820_hole_size(unsigned long start_pfn,
 				    unsigned long end_pfn);
 
-extern void __init parse_memopt(char *p, char **end);
-extern void __init parse_memmapopt(char *p, char **end);
+extern void finish_e820_parsing(void);
 
 extern struct e820map e820;
 
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 0c1e2422400a..d79e7441b513 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -21,7 +21,6 @@ extern unsigned long __supported_pte_mask;
 
 #define swapper_pg_dir init_level4_pgt
 
-extern int nonx_setup(char *str);
 extern void paging_init(void);
 extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
 
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 58fec91318e4..151826f8ce69 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -53,9 +53,6 @@ extern int nohpet;
 extern unsigned long vxtime_hz;
 extern void time_init_gtod(void);
 
-extern int numa_setup(char *opt);
-
-extern int setup_early_printk(char *); 
 extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
 
 extern void early_identify_cpu(struct cpuinfo_x86 *c);
@@ -104,13 +101,7 @@ extern void select_idle_routine(const struct cpuinfo_x86 *c);
 extern unsigned long table_start, table_end;
 
 extern int exception_trace;
-extern int using_apic_timer;
-extern int disable_apic;
 extern unsigned cpu_khz;
-extern int ioapic_force;
-extern int skip_ioapic_setup;
-extern int acpi_ht;
-extern int acpi_disabled;
 
 extern void no_iommu_init(void);
 extern int force_iommu, no_iommu;
@@ -132,7 +123,6 @@ extern int fix_aperture;
 
 extern int reboot_force;
 extern int notsc_setup(char *);
-extern int setup_additional_cpus(char *);
 
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 
-- 
cgit v1.2.1


From 43c85c9c5dff76efc1e411d3302840027ea92004 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:32 +0200
Subject: [PATCH] Remove need for early lockdep init

I think it was only needed for the printks and we can do them later.

I put in a single early_printk so that we know the kernel is alive
(early_printk doesn't need any locks)

This makes some things easier for initialization of unwind for
lockdep, which is needed by later patches.

cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/head64.c | 8 +-------
 arch/x86_64/kernel/setup.c  | 2 ++
 2 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/arch/x86_64/kernel/head64.c b/arch/x86_64/kernel/head64.c
index 6716cbfc34ac..9561eb3c5b5c 100644
--- a/arch/x86_64/kernel/head64.c
+++ b/arch/x86_64/kernel/head64.c
@@ -45,15 +45,12 @@ static void __init copy_bootdata(char *real_mode_data)
 	new_data = *(int *) (x86_boot_params + NEW_CL_POINTER);
 	if (!new_data) {
 		if (OLD_CL_MAGIC != * (u16 *) OLD_CL_MAGIC_ADDR) {
-			printk("so old bootloader that it does not support commandline?!\n");
 			return;
 		}
 		new_data = OLD_CL_BASE_ADDR + * (u16 *) OLD_CL_OFFSET;
-		printk("old bootloader convention, maybe loadlin?\n");
 	}
 	command_line = (char *) ((u64)(new_data));
 	memcpy(saved_command_line, command_line, COMMAND_LINE_SIZE);
-	printk("Bootdata ok (command line is %s)\n", saved_command_line);	
 }
 
 void __init x86_64_start_kernel(char * real_mode_data)
@@ -65,10 +62,7 @@ void __init x86_64_start_kernel(char * real_mode_data)
 	asm volatile("lidt %0" :: "m" (idt_descr));
 	clear_bss();
 
-	/*
-	 * This must be called really, really early:
-	 */
-	lockdep_init();
+	early_printk("Kernel alive\n");
 
 	/*
 	 * switch to init_level4_pgt from boot_level4_pgt
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index f55540a6084e..9c4d93c53c06 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -353,6 +353,8 @@ static void discover_ebda(void)
 
 void __init setup_arch(char **cmdline_p)
 {
+	printk(KERN_INFO "Command line: %s\n", saved_command_line);
+
  	ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
  	screen_info = SCREEN_INFO;
 	edid_info = EDID_INFO;
-- 
cgit v1.2.1


From df3bb57d2c0160ccd1ee51322f50aa295c3b0858 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] i386/x86-64: Move acpi_disabled variables into acpi/boot.c

Removes code duplication between i386/x86-64.

Not needed anymore in setup.c since early_param cleanup

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 7 +++++++
 arch/i386/kernel/setup.c     | 7 -------
 arch/x86_64/kernel/setup.c   | 3 ---
 3 files changed, 7 insertions(+), 10 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 87e2ab50b694..f3761b98c8dd 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -38,6 +38,13 @@
 
 int __initdata acpi_force = 0;
 
+#ifdef	CONFIG_ACPI
+int acpi_disabled = 0;
+#else
+int acpi_disabled = 1;
+#endif
+EXPORT_SYMBOL(acpi_disabled);
+
 #ifdef	CONFIG_X86_64
 
 extern void __init clustered_apic_check(void);
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index c6e31ed386f5..ea17567dbe72 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -89,13 +89,6 @@ EXPORT_SYMBOL(boot_cpu_data);
 
 unsigned long mmu_cr4_features;
 
-#ifdef	CONFIG_ACPI
-	int acpi_disabled = 0;
-#else
-	int acpi_disabled = 1;
-#endif
-EXPORT_SYMBOL(acpi_disabled);
-
 /* for MCA, but anyone else can use it if they want */
 unsigned int machine_id;
 #ifdef CONFIG_MCA
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 9c4d93c53c06..0f86976811d6 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -74,9 +74,6 @@ EXPORT_SYMBOL(boot_cpu_data);
 
 unsigned long mmu_cr4_features;
 
-int acpi_disabled;
-EXPORT_SYMBOL(acpi_disabled);
-
 int acpi_numa __initdata;
 
 /* Boot loader ID as an integer, for the benefit of proc_dointvec */
-- 
cgit v1.2.1


From c31fbb1ad890b11f037c16496e53f28877c12722 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Clean up acpi_numa variable

Move it into srat.c No need to clutter up setup.c for it

And remove use in setup.c completely - it only guarded a printk
which can be done unconditionally.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup.c | 5 +----
 arch/x86_64/mm/srat.c      | 2 ++
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 0f86976811d6..b85528597aa8 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -74,8 +74,6 @@ EXPORT_SYMBOL(boot_cpu_data);
 
 unsigned long mmu_cr4_features;
 
-int acpi_numa __initdata;
-
 /* Boot loader ID as an integer, for the benefit of proc_dointvec */
 int bootloader_type;
 
@@ -812,8 +810,7 @@ static void srat_detect_node(void)
 		node = first_node(node_online_map);
 	numa_set_node(cpu, node);
 
-	if (acpi_numa > 0)
-		printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
+	printk(KERN_INFO "CPU %d/%x -> Node %d\n", cpu, apicid, node);
 #endif
 }
 
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 502fce65e96a..ca10701e7a90 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -21,6 +21,8 @@
 #include <asm/numa.h>
 #include <asm/e820.h>
 
+int acpi_numa __initdata;
+
 #if (defined(CONFIG_ACPI_HOTPLUG_MEMORY) || \
 	defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)) \
 		&& !defined(CONFIG_MEMORY_HOTPLUG)
-- 
cgit v1.2.1


From 3bd4d18cbab622c504f131f3c0029c3aa29c05be Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Move e820 map into e820.c

Minor cleanup. Keep setup.c free from unrelated clutter.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/e820.c  | 2 ++
 arch/x86_64/kernel/setup.c | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index a4739dc82937..164d0b83dc92 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -24,6 +24,8 @@
 #include <asm/bootsetup.h>
 #include <asm/sections.h>
 
+struct e820map e820 __initdata;
+
 /* 
  * PFN of last memory page.
  */
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index b85528597aa8..4be29832e3dc 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -97,7 +97,6 @@ struct sys_desc_table_struct {
 
 struct edid_info edid_info;
 EXPORT_SYMBOL_GPL(edid_info);
-struct e820map e820;
 
 extern int root_mountflags;
 
-- 
cgit v1.2.1


From 131cfd7bd54767ec8959e013f83839442a54d546 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Add sparse annotation to vsyscall.c

Fixes

linux/arch/x86_64/kernel/vsyscall.c:276:7: warning: constant 0x0f40000000000 is so big it is long
linux/arch/x86_64/kernel/vsyscall.c:80:14: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/kernel/vsyscall.c:80:14:    expected void const volatile [noderef] *addr<asn:2>
linux/arch/x86_64/kernel/vsyscall.c:80:14:    got void *<noident>
linux/arch/x86_64/kernel/vsyscall.c:200:7: warning: incorrect type in assignment (different address spaces)
linux/arch/x86_64/kernel/vsyscall.c:200:7:    expected unsigned short [usertype] *map1
linux/arch/x86_64/kernel/vsyscall.c:200:7:    got void [noderef] *<asn:2>
linux/arch/x86_64/kernel/vsyscall.c:203:7: warning: incorrect type in assignment (different address spaces)
linux/arch/x86_64/kernel/vsyscall.c:203:7:    expected unsigned short [usertype] *map2
linux/arch/x86_64/kernel/vsyscall.c:203:7:    got void [noderef] *<asn:2>
linux/arch/x86_64/kernel/vsyscall.c:215:10: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/kernel/vsyscall.c:215:10:    expected void volatile [noderef] *addr<asn:2>
linux/arch/x86_64/kernel/vsyscall.c:215:10:    got unsigned short [usertype] *map2
linux/arch/x86_64/kernel/vsyscall.c:217:10: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/kernel/vsyscall.c:217:10:    expected void volatile [noderef] *addr<asn:2>
linux/arch/x86_64/kernel/vsyscall.c:217:10:    got unsigned short [usertype] *map1

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/vsyscall.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/arch/x86_64/kernel/vsyscall.c b/arch/x86_64/kernel/vsyscall.c
index 902783bc4d53..ac48c3857ddb 100644
--- a/arch/x86_64/kernel/vsyscall.c
+++ b/arch/x86_64/kernel/vsyscall.c
@@ -77,7 +77,8 @@ static __always_inline void do_vgettimeofday(struct timeval * tv)
 				 __vxtime.tsc_quot) >> 32;
 			/* See comment in x86_64 do_gettimeofday. */
 		} else {
-			usec += ((readl((void *)fix_to_virt(VSYSCALL_HPET) + 0xf0) -
+			usec += ((readl((void __iomem *)
+				   fix_to_virt(VSYSCALL_HPET) + 0xf0) -
 				  __vxtime.last) * __vxtime.quot) >> 32;
 		}
 	} while (read_seqretry(&__xtime_lock, sequence));
@@ -191,7 +192,8 @@ static int vsyscall_sysctl_change(ctl_table *ctl, int write, struct file * filp,
                         void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	extern u16 vsysc1, vsysc2;
-	u16 *map1, *map2;
+	u16 __iomem *map1;
+	u16 __iomem *map2;
 	int ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
 	if (!write)
 		return ret;
@@ -206,11 +208,11 @@ static int vsyscall_sysctl_change(ctl_table *ctl, int write, struct file * filp,
 		goto out;
 	}
 	if (!sysctl_vsyscall) {
-		*map1 = SYSCALL;
-		*map2 = SYSCALL;
+		writew(SYSCALL, map1);
+		writew(SYSCALL, map2);
 	} else {
-		*map1 = NOP2;
-		*map2 = NOP2;
+		writew(NOP2, map1);
+		writew(NOP2, map2);
 	}
 	iounmap(map2);
 out:
-- 
cgit v1.2.1


From dd2994f619752fb731f21c89ad16536dd6673948 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Add sparse annotations to quiet sparse in
 arch/x86_64/mm/fault.c

Fixes

linux/arch/x86_64/mm/fault.c:125:7: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/mm/fault.c:125:7:    expected void [noderef] *<noident><asn:1>
linux/arch/x86_64/mm/fault.c:125:7:    got unsigned char *[assigned] instr
linux/arch/x86_64/mm/fault.c:163:8: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/mm/fault.c:163:8:    expected void [noderef] *<noident><asn:1>
linux/arch/x86_64/mm/fault.c:163:8:    got unsigned char *[assigned] instr
linux/arch/x86_64/mm/fault.c:179:9: warning: incorrect type in argument 1 (different address spaces)
linux/arch/x86_64/mm/fault.c:179:9:    expected void [noderef] *<noident><asn:1>
linux/arch/x86_64/mm/fault.c:179:9:    got unsigned long *<noident>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/fault.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index ac8ea66ccb94..449a437f5d70 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -102,7 +102,7 @@ void bust_spinlocks(int yes)
 static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 				unsigned long error_code)
 { 
-	unsigned char *instr;
+	unsigned char __user *instr;
 	int scan_more = 1;
 	int prefetch = 0; 
 	unsigned char *max_instr;
@@ -111,7 +111,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 	if (error_code & PF_INSTR)
 		return 0;
 	
-	instr = (unsigned char *)convert_rip_to_linear(current, regs);
+	instr = (unsigned char __user *)convert_rip_to_linear(current, regs);
 	max_instr = instr + 15;
 
 	if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
@@ -122,7 +122,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 		unsigned char instr_hi;
 		unsigned char instr_lo;
 
-		if (__get_user(opcode, instr))
+		if (__get_user(opcode, (char __user *)instr))
 			break; 
 
 		instr_hi = opcode & 0xf0; 
@@ -160,7 +160,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 		case 0x00:
 			/* Prefetch instruction is 0x0F0D or 0x0F18 */
 			scan_more = 0;
-			if (__get_user(opcode, instr)) 
+			if (__get_user(opcode, (char __user *)instr))
 				break;
 			prefetch = (instr_lo == 0xF) &&
 				(opcode == 0x0D || opcode == 0x18);
@@ -176,7 +176,7 @@ static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
 static int bad_address(void *p) 
 { 
 	unsigned long dummy;
-	return __get_user(dummy, (unsigned long *)p);
+	return __get_user(dummy, (unsigned long __user *)p);
 } 
 
 void dump_pagetable(unsigned long address)
-- 
cgit v1.2.1


From ddb15ec130d38cb8076a9926040c7435b126db65 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Fix most sparse warnings in sys_ia32.c

Mostly by adding casts.

I didn't touch the "invalid access past ..." which are caused
by the sigset conversion.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/sys_ia32.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c
index 9c130993380d..33cd0a4fe388 100644
--- a/arch/x86_64/ia32/sys_ia32.c
+++ b/arch/x86_64/ia32/sys_ia32.c
@@ -60,6 +60,7 @@
 #include <linux/highuid.h>
 #include <linux/vmalloc.h>
 #include <linux/fsnotify.h>
+#include <linux/sysctl.h>
 #include <asm/mman.h>
 #include <asm/types.h>
 #include <asm/uaccess.h>
@@ -389,7 +390,9 @@ sys32_rt_sigprocmask(int how, compat_sigset_t __user *set,
 		}
 	}
 	set_fs (KERNEL_DS);
-	ret = sys_rt_sigprocmask(how, set ? &s : NULL, oset ? &s : NULL,
+	ret = sys_rt_sigprocmask(how,
+				 set ? (sigset_t __user *)&s : NULL,
+				 oset ? (sigset_t __user *)&s : NULL,
 				 sigsetsize); 
 	set_fs (old_fs);
 	if (ret) return ret;
@@ -541,7 +544,7 @@ sys32_sysinfo(struct sysinfo32 __user *info)
 	int bitcount = 0;
 	
 	set_fs (KERNEL_DS);
-	ret = sys_sysinfo(&s);
+	ret = sys_sysinfo((struct sysinfo __user *)&s);
 	set_fs (old_fs);
 
         /* Check to see if any memory value is too large for 32-bit and scale
@@ -589,7 +592,7 @@ sys32_sched_rr_get_interval(compat_pid_t pid, struct compat_timespec __user *int
 	mm_segment_t old_fs = get_fs ();
 	
 	set_fs (KERNEL_DS);
-	ret = sys_sched_rr_get_interval(pid, &t);
+	ret = sys_sched_rr_get_interval(pid, (struct timespec __user *)&t);
 	set_fs (old_fs);
 	if (put_compat_timespec(&t, interval))
 		return -EFAULT;
@@ -605,7 +608,7 @@ sys32_rt_sigpending(compat_sigset_t __user *set, compat_size_t sigsetsize)
 	mm_segment_t old_fs = get_fs();
 		
 	set_fs (KERNEL_DS);
-	ret = sys_rt_sigpending(&s, sigsetsize);
+	ret = sys_rt_sigpending((sigset_t __user *)&s, sigsetsize);
 	set_fs (old_fs);
 	if (!ret) {
 		switch (_NSIG_WORDS) {
@@ -630,7 +633,7 @@ sys32_rt_sigqueueinfo(int pid, int sig, compat_siginfo_t __user *uinfo)
 	if (copy_siginfo_from_user32(&info, uinfo))
 		return -EFAULT;
 	set_fs (KERNEL_DS);
-	ret = sys_rt_sigqueueinfo(pid, sig, &info);
+	ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __user *)&info);
 	set_fs (old_fs);
 	return ret;
 }
@@ -666,9 +669,6 @@ sys32_sysctl(struct sysctl_ia32 __user *args32)
 	size_t oldlen;
 	int __user *namep;
 	long ret;
-	extern int do_sysctl(int *name, int nlen, void *oldval, size_t *oldlenp,
-		     void *newval, size_t newlen);
-
 
 	if (copy_from_user(&a32, args32, sizeof (a32)))
 		return -EFAULT;
@@ -692,7 +692,8 @@ sys32_sysctl(struct sysctl_ia32 __user *args32)
 
 	set_fs(KERNEL_DS);
 	lock_kernel();
-	ret = do_sysctl(namep, a32.nlen, oldvalp, &oldlen, newvalp, (size_t) a32.newlen);
+	ret = do_sysctl(namep, a32.nlen, oldvalp, (size_t __user *)&oldlen,
+			newvalp, (size_t) a32.newlen);
 	unlock_kernel();
 	set_fs(old_fs);
 
@@ -743,7 +744,8 @@ sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, s32 count)
 		return -EFAULT;
 		
 	set_fs(KERNEL_DS);
-	ret = sys_sendfile(out_fd, in_fd, offset ? &of : NULL, count);
+	ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __user *)&of : NULL,
+			   count);
 	set_fs(old_fs);
 	
 	if (offset && put_user(of, offset))
@@ -831,7 +833,7 @@ long sys32_ustat(unsigned dev, struct ustat32 __user *u32p)
 	
 	seg = get_fs(); 
 	set_fs(KERNEL_DS); 
-	ret = sys_ustat(dev,&u); 
+	ret = sys_ustat(dev, (struct ustat __user *)&u);
 	set_fs(seg);
 	if (ret >= 0) { 
 		if (!access_ok(VERIFY_WRITE,u32p,sizeof(struct ustat32)) || 
-- 
cgit v1.2.1


From 52d522f53f137c7903db22f9196a48ad8658fb2b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Fix sparse warnings in compat aout code

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_aout.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/x86_64/ia32/ia32_aout.c b/arch/x86_64/ia32/ia32_aout.c
index 3bf58af98936..396d3c100011 100644
--- a/arch/x86_64/ia32/ia32_aout.c
+++ b/arch/x86_64/ia32/ia32_aout.c
@@ -333,7 +333,8 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 			return error;
 		}
 
-		error = bprm->file->f_op->read(bprm->file, (char *)text_addr,
+		error = bprm->file->f_op->read(bprm->file,
+			 (char __user *)text_addr,
 			  ex.a_text+ex.a_data, &pos);
 		if ((signed long)error < 0) {
 			send_sig(SIGKILL, current, 0);
@@ -366,7 +367,8 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 			down_write(&current->mm->mmap_sem);
 			do_brk(N_TXTADDR(ex), ex.a_text+ex.a_data);
 			up_write(&current->mm->mmap_sem);
-			bprm->file->f_op->read(bprm->file,(char *)N_TXTADDR(ex),
+			bprm->file->f_op->read(bprm->file,
+					(char __user *)N_TXTADDR(ex),
 					ex.a_text+ex.a_data, &pos);
 			flush_icache_range((unsigned long) N_TXTADDR(ex),
 					   (unsigned long) N_TXTADDR(ex) +
@@ -477,7 +479,7 @@ static int load_aout_library(struct file *file)
 		do_brk(start_addr, ex.a_text + ex.a_data + ex.a_bss);
 		up_write(&current->mm->mmap_sem);
 		
-		file->f_op->read(file, (char *)start_addr,
+		file->f_op->read(file, (char __user *)start_addr,
 			ex.a_text + ex.a_data, &pos);
 		flush_icache_range((unsigned long) start_addr,
 				   (unsigned long) start_addr + ex.a_text + ex.a_data);
-- 
cgit v1.2.1


From 2b94ab2fd55768030e177a6ec224dedd031ad06b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fernando=20Luis=20V=E1zquez=20Cao?= <fernando@oss.ntt.co.jp>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Replace local_save_flags+local_irq_disable with

The combination of "local_save_flags" and "local_irq_disable" seems to be
equivalent to "local_irq_save" (see code snips below). Consequently, replace
occurrences of local_save_flags+local_irq_disable with local_irq_save.

* local_irq_save
#define raw_local_irq_save(flags) \
                do { (flags) = __raw_local_irq_save(); } while (0)

static inline unsigned long __raw_local_irq_save(void)
{
        unsigned long flags = __raw_local_save_flags();

        raw_local_irq_disable();

        return flags;
}

* local_save_flags
#define raw_local_save_flags(flags) \
                do { (flags) = __raw_local_save_flags(); } while (0)

Signed-off-by: Fernando Vazquez <fernando@intellilink.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c         | 3 +--
 arch/x86_64/kernel/genapic_flat.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 660602c51f51..b95d1e378b96 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -468,8 +468,7 @@ static int lapic_suspend(struct sys_device *dev, pm_message_t state)
 	apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
 	apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
 	apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
-	local_save_flags(flags);
-	local_irq_disable();
+	local_irq_save(flags);
 	disable_local_APIC();
 	local_irq_restore(flags);
 	return 0;
diff --git a/arch/x86_64/kernel/genapic_flat.c b/arch/x86_64/kernel/genapic_flat.c
index 2ceb8b872b57..50ad153eaac4 100644
--- a/arch/x86_64/kernel/genapic_flat.c
+++ b/arch/x86_64/kernel/genapic_flat.c
@@ -49,8 +49,7 @@ static void flat_send_IPI_mask(cpumask_t cpumask, int vector)
 	unsigned long cfg;
 	unsigned long flags;
 
-	local_save_flags(flags);
-	local_irq_disable();
+	local_irq_save(flags);
 
 	/*
 	 * Wait for idle.
-- 
cgit v1.2.1


From 91cd444e56ebe0c2acd9576a045d77490b26f607 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] x86: Remove unneeded externs in acpi/boot.c

And move one into proto.h

Cc: len.brown@intel.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 3 ---
 include/asm-x86_64/proto.h   | 2 ++
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index f3761b98c8dd..c809a3f4c0bf 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -47,9 +47,6 @@ EXPORT_SYMBOL(acpi_disabled);
 
 #ifdef	CONFIG_X86_64
 
-extern void __init clustered_apic_check(void);
-
-extern int gsi_irq_sharing(int gsi);
 #include <asm/proto.h>
 
 static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 151826f8ce69..b73d0c76613c 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -124,6 +124,8 @@ extern int fix_aperture;
 extern int reboot_force;
 extern int notsc_setup(char *);
 
+extern int gsi_irq_sharing(int gsi);
+
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 
 long do_arch_prctl(struct task_struct *task, int code, unsigned long addr);
-- 
cgit v1.2.1


From 2ade2920dcefdf5595c6380ebed131c964190855 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] i386/x86-64: rename is_at_popf(), add iret to tests and fix

is_at_popf() needs to test for the iret instruction as well as
popf.  So add that test and rename it to is_setting_trap_flag().

Also change max insn length from 16 to 15 to match reality.

LAHF / SAHF can't affect TF, so the comment in x86_64 is removed.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/ptrace.c   | 10 +++++-----
 arch/x86_64/kernel/ptrace.c | 12 +++++-------
 2 files changed, 10 insertions(+), 12 deletions(-)

diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c
index d3db03f4085d..775f50e9395b 100644
--- a/arch/i386/kernel/ptrace.c
+++ b/arch/i386/kernel/ptrace.c
@@ -185,17 +185,17 @@ static unsigned long convert_eip_to_linear(struct task_struct *child, struct pt_
 	return addr;
 }
 
-static inline int is_at_popf(struct task_struct *child, struct pt_regs *regs)
+static inline int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
 {
 	int i, copied;
-	unsigned char opcode[16];
+	unsigned char opcode[15];
 	unsigned long addr = convert_eip_to_linear(child, regs);
 
 	copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0);
 	for (i = 0; i < copied; i++) {
 		switch (opcode[i]) {
-		/* popf */
-		case 0x9d:
+		/* popf and iret */
+		case 0x9d: case 0xcf:
 			return 1;
 		/* opcode and address size prefixes */
 		case 0x66: case 0x67:
@@ -247,7 +247,7 @@ static void set_singlestep(struct task_struct *child)
 	 * don't mark it as being "us" that set it, so that we
 	 * won't clear it by hand later.
 	 */
-	if (is_at_popf(child, regs))
+	if (is_setting_trap_flag(child, regs))
 		return;
 	
 	child->ptrace |= PT_DTRACE;
diff --git a/arch/x86_64/kernel/ptrace.c b/arch/x86_64/kernel/ptrace.c
index 412c6a8bf906..addc14af0c56 100644
--- a/arch/x86_64/kernel/ptrace.c
+++ b/arch/x86_64/kernel/ptrace.c
@@ -116,17 +116,17 @@ unsigned long convert_rip_to_linear(struct task_struct *child, struct pt_regs *r
 	return addr;
 }
 
-static int is_at_popf(struct task_struct *child, struct pt_regs *regs)
+static int is_setting_trap_flag(struct task_struct *child, struct pt_regs *regs)
 {
 	int i, copied;
-	unsigned char opcode[16];
+	unsigned char opcode[15];
 	unsigned long addr = convert_rip_to_linear(child, regs);
 
 	copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0);
 	for (i = 0; i < copied; i++) {
 		switch (opcode[i]) {
-		/* popf */
-		case 0x9d:
+		/* popf and iret */
+		case 0x9d: case 0xcf:
 			return 1;
 
 			/* CHECKME: 64 65 */
@@ -189,10 +189,8 @@ static void set_singlestep(struct task_struct *child)
 	 * ..but if TF is changed by the instruction we will trace,
 	 * don't mark it as being "us" that set it, so that we
 	 * won't clear it by hand later.
-	 *
-	 * AK: this is not enough, LAHF and IRET can change TF in user space too.
 	 */
-	if (is_at_popf(child, regs))
+	if (is_setting_trap_flag(child, regs))
 		return;
 
 	child->ptrace |= PT_DTRACE;
-- 
cgit v1.2.1


From de684652f34f57cb60d4d78d09139a0e0c5e7b1b Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] print whether CONFIG_IOMMU_DEBUG is enabled

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index caf84e4e4ca9..ebe4e930b64d 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -127,15 +127,19 @@ static void tce_cache_blast(struct iommu_table *tbl);
 
 /* enable this to stress test the chip's TCE cache */
 #ifdef CONFIG_IOMMU_DEBUG
+int debugging __read_mostly = 1;
+
 static inline void tce_cache_blast_stress(struct iommu_table *tbl)
 {
 	tce_cache_blast(tbl);
 }
-#else
+#else /* debugging is disabled */
+int debugging __read_mostly = 0;
+
 static inline void tce_cache_blast_stress(struct iommu_table *tbl)
 {
 }
-#endif /* BLAST_TCE_CACHE_ON_UNMAP */
+#endif /* CONFIG_IOMMU_DEBUG */
 
 static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen)
 {
@@ -944,8 +948,10 @@ void __init detect_calgary(void)
 	if (calgary_found) {
 		iommu_detected = 1;
 		calgary_detected = 1;
-		printk(KERN_INFO "PCI-DMA: Calgary IOMMU detected. "
-		       "TCE table spec is %d.\n", specified_table_size);
+		printk(KERN_INFO "PCI-DMA: Calgary IOMMU detected.\n");
+		printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d, "
+		       "CONFIG_IOMMU_DEBUG is %s.\n", specified_table_size,
+		       debugging ? "enabled" : "disabled");
 	}
 	return;
 
-- 
cgit v1.2.1


From 796e4390e0378e1e57c033349610cfc741696a3d Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] only verify the allocation bitmap if CONFIG_IOMMU_DEBUG is on

Introduce new function verify_bit_range(). Define two versions, one
for CONFIG_IOMMU_DEBUG enabled and one for disabled. Previously we
were checking that the bitmap was consistent every time we allocated
or freed an entry in the TCE table, which is good for debugging but
incurs an unnecessary penalty on non debug builds.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 44 ++++++++++++++++++++++++++++++++--------
 1 file changed, 35 insertions(+), 9 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index ebe4e930b64d..7c43cb0f71a3 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -133,12 +133,35 @@ static inline void tce_cache_blast_stress(struct iommu_table *tbl)
 {
 	tce_cache_blast(tbl);
 }
+
+static inline unsigned long verify_bit_range(unsigned long* bitmap,
+	int expected, unsigned long start, unsigned long end)
+{
+	unsigned long idx = start;
+
+	BUG_ON(start >= end);
+
+	while (idx < end) {
+		if (!!test_bit(idx, bitmap) != expected)
+			return idx;
+		++idx;
+	}
+
+	/* all bits have the expected value */
+	return ~0UL;
+}
 #else /* debugging is disabled */
 int debugging __read_mostly = 0;
 
 static inline void tce_cache_blast_stress(struct iommu_table *tbl)
 {
 }
+
+static inline unsigned long verify_bit_range(unsigned long* bitmap,
+	int expected, unsigned long start, unsigned long end)
+{
+	return ~0UL;
+}
 #endif /* CONFIG_IOMMU_DEBUG */
 
 static inline unsigned int num_dma_pages(unsigned long dma, unsigned int dmalen)
@@ -162,6 +185,7 @@ static void iommu_range_reserve(struct iommu_table *tbl,
 {
 	unsigned long index;
 	unsigned long end;
+	unsigned long badbit;
 
 	index = start_addr >> PAGE_SHIFT;
 
@@ -173,14 +197,15 @@ static void iommu_range_reserve(struct iommu_table *tbl,
 	if (end > tbl->it_size) /* don't go off the table */
 		end = tbl->it_size;
 
-	while (index < end) {
-		if (test_bit(index, tbl->it_map))
+	badbit = verify_bit_range(tbl->it_map, 0, index, end);
+	if (badbit != ~0UL) {
+		if (printk_ratelimit())
 			printk(KERN_ERR "Calgary: entry already allocated at "
 			       "0x%lx tbl %p dma 0x%lx npages %u\n",
-			       index, tbl, start_addr, npages);
-		++index;
+			       badbit, tbl, start_addr, npages);
 	}
-	set_bit_string(tbl->it_map, start_addr >> PAGE_SHIFT, npages);
+
+	set_bit_string(tbl->it_map, index, npages);
 }
 
 static unsigned long iommu_range_alloc(struct iommu_table *tbl,
@@ -247,7 +272,7 @@ static void __iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr,
 	unsigned int npages)
 {
 	unsigned long entry;
-	unsigned long i;
+	unsigned long badbit;
 
 	entry = dma_addr >> PAGE_SHIFT;
 
@@ -255,11 +280,12 @@ static void __iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr,
 
 	tce_free(tbl, entry, npages);
 
-	for (i = 0; i < npages; ++i) {
-		if (!test_bit(entry + i, tbl->it_map))
+	badbit = verify_bit_range(tbl->it_map, 1, entry, entry + npages);
+	if (badbit != ~0UL) {
+		if (printk_ratelimit())
 			printk(KERN_ERR "Calgary: bit is off at 0x%lx "
 			       "tbl %p dma 0x%Lx entry 0x%lx npages %u\n",
-			       entry + i, tbl, dma_addr, entry, npages);
+			       badbit, tbl, dma_addr, entry, npages);
 	}
 
 	__clear_bit_string(tbl->it_map, entry, npages);
-- 
cgit v1.2.1


From 4ccf4ae3144360ab9c00d0b53427f43369287bfb Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] remove tce_cache_blast_stress()

tce_cache_blast_stress was useful during bringup to stress the IOMMU's
cache flushing. Now that we quiesce DMAs on every cache flush, using
_stress() brings the machine down to its knees once you put it under
load. Remove this debug / bringup code that isn't useful anymore
completely.

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 11 -----------
 1 file changed, 11 deletions(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 7c43cb0f71a3..cd866135e4ef 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -129,11 +129,6 @@ static void tce_cache_blast(struct iommu_table *tbl);
 #ifdef CONFIG_IOMMU_DEBUG
 int debugging __read_mostly = 1;
 
-static inline void tce_cache_blast_stress(struct iommu_table *tbl)
-{
-	tce_cache_blast(tbl);
-}
-
 static inline unsigned long verify_bit_range(unsigned long* bitmap,
 	int expected, unsigned long start, unsigned long end)
 {
@@ -153,10 +148,6 @@ static inline unsigned long verify_bit_range(unsigned long* bitmap,
 #else /* debugging is disabled */
 int debugging __read_mostly = 0;
 
-static inline void tce_cache_blast_stress(struct iommu_table *tbl)
-{
-}
-
 static inline unsigned long verify_bit_range(unsigned long* bitmap,
 	int expected, unsigned long start, unsigned long end)
 {
@@ -289,8 +280,6 @@ static void __iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr,
 	}
 
 	__clear_bit_string(tbl->it_map, entry, npages);
-
-	tce_cache_blast_stress(tbl);
 }
 
 static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr,
-- 
cgit v1.2.1


From 4ea8a5d8b57cd504b4b2de1212523848e7ab50cf Mon Sep 17 00:00:00 2001
From: Muli Ben-Yehuda <muli@il.ibm.com>
Date: Tue, 26 Sep 2006 10:52:33 +0200
Subject: [PATCH] Calgary IOMMU: eradicate sole remaining 80 chars per line
 offender

Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/pci-calgary.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index cd866135e4ef..466588f95601 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -86,7 +86,8 @@
 
 #define MAX_NUM_OF_PHBS		8 /* how many PHBs in total? */
 #define MAX_NUM_CHASSIS		8 /* max number of chassis */
-#define MAX_PHB_BUS_NUM		(MAX_NUM_OF_PHBS * MAX_NUM_CHASSIS * 2) /* max dev->bus->number */
+/* MAX_PHB_BUS_NUM is the maximal possible dev->bus->number */
+#define MAX_PHB_BUS_NUM		(MAX_NUM_OF_PHBS * MAX_NUM_CHASSIS * 2)
 #define PHBS_PER_CALGARY	4
 
 /* register offsets in Calgary's internal register space */
-- 
cgit v1.2.1


From 5a1b3999d6cb7ab87f1f3b1700bc91839fd6fa29 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] x86: Some preparationary cleanup for stack trace

- Remove unused all_contexts parameter
No caller used it
- Move skip argument into the structure (needed for
followon patches)

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/stacktrace.c   | 11 +++--------
 arch/s390/kernel/stacktrace.c   | 17 ++++++++---------
 arch/x86_64/kernel/stacktrace.c | 14 +++++---------
 include/linux/stacktrace.h      |  7 ++++---
 kernel/lockdep.c                |  5 ++++-
 5 files changed, 24 insertions(+), 30 deletions(-)

diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c
index e62a037ab399..ae3c32a87add 100644
--- a/arch/i386/kernel/stacktrace.c
+++ b/arch/i386/kernel/stacktrace.c
@@ -61,12 +61,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip,
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
- * If all_contexts is set, all contexts (hardirq, softirq and process)
- * are saved. If not set then only the current context is saved.
  */
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	unsigned long ebp;
 	unsigned long *stack = &ebp;
@@ -85,10 +81,9 @@ void save_stack_trace(struct stack_trace *trace,
 		struct thread_info *context = (struct thread_info *)
 				((unsigned long)stack & (~(THREAD_SIZE - 1)));
 
-		ebp = save_context_stack(trace, skip, context, stack, ebp);
+		ebp = save_context_stack(trace, trace->skip, context, stack, ebp);
 		stack = (unsigned long *)context->previous_esp;
-		if (!all_contexts || !stack ||
-				trace->nr_entries >= trace->max_entries)
+		if (!stack || trace->nr_entries >= trace->max_entries)
 			break;
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 		if (trace->nr_entries >= trace->max_entries)
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index de83f38288d0..d9428a0fc8fb 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -59,9 +59,7 @@ static inline unsigned long save_context_stack(struct stack_trace *trace,
 	}
 }
 
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	register unsigned long sp asm ("15");
 	unsigned long orig_sp;
@@ -69,22 +67,23 @@ void save_stack_trace(struct stack_trace *trace,
 	sp &= PSW_ADDR_INSN;
 	orig_sp = sp;
 
-	sp = save_context_stack(trace, &skip, sp,
+	sp = save_context_stack(trace, &trace->skip, sp,
 				S390_lowcore.panic_stack - PAGE_SIZE,
 				S390_lowcore.panic_stack);
-	if ((sp != orig_sp) && !all_contexts)
+	if ((sp != orig_sp) && !trace->all_contexts)
 		return;
-	sp = save_context_stack(trace, &skip, sp,
+	sp = save_context_stack(trace, &trace->skip, sp,
 				S390_lowcore.async_stack - ASYNC_SIZE,
 				S390_lowcore.async_stack);
-	if ((sp != orig_sp) && !all_contexts)
+	if ((sp != orig_sp) && !trace->all_contexts)
 		return;
 	if (task)
-		save_context_stack(trace, &skip, sp,
+		save_context_stack(trace, &trace->skip, sp,
 				   (unsigned long) task_stack_page(task),
 				   (unsigned long) task_stack_page(task) + THREAD_SIZE);
 	else
-		save_context_stack(trace, &skip, sp, S390_lowcore.thread_info,
+		save_context_stack(trace, &trace->skip, sp,
+				   S390_lowcore.thread_info,
 				   S390_lowcore.thread_info + THREAD_SIZE);
 	return;
 }
diff --git a/arch/x86_64/kernel/stacktrace.c b/arch/x86_64/kernel/stacktrace.c
index 32cf55eb9af8..1c022af8fe1e 100644
--- a/arch/x86_64/kernel/stacktrace.c
+++ b/arch/x86_64/kernel/stacktrace.c
@@ -109,9 +109,10 @@ out_restore:
  * Save stack-backtrace addresses into a stack_trace buffer:
  */
 static inline unsigned long
-save_context_stack(struct stack_trace *trace, unsigned int skip,
+save_context_stack(struct stack_trace *trace,
 		   unsigned long stack, unsigned long stack_end)
 {
+	int skip = trace->skip;
 	unsigned long addr;
 
 #ifdef CONFIG_FRAME_POINTER
@@ -159,12 +160,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip,
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
- * If all_contexts is set, all contexts (hardirq, softirq and process)
- * are saved. If not set then only the current context is saved.
  */
-void save_stack_trace(struct stack_trace *trace,
-		      struct task_struct *task, int all_contexts,
-		      unsigned int skip)
+void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
 	unsigned long stack = (unsigned long)&stack;
 	int i, nr_stacks = 0, stacks_done[MAX_STACKS];
@@ -207,9 +204,8 @@ void save_stack_trace(struct stack_trace *trace,
 				return;
 		stacks_done[nr_stacks] = stack_end;
 
-		stack = save_context_stack(trace, skip, stack, stack_end);
-		if (!all_contexts || !stack ||
-				trace->nr_entries >= trace->max_entries)
+		stack = save_context_stack(trace, stack, stack_end);
+		if (!stack || trace->nr_entries >= trace->max_entries)
 			return;
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 		if (trace->nr_entries >= trace->max_entries)
diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h
index 9cc81e572224..50e2b01e517c 100644
--- a/include/linux/stacktrace.h
+++ b/include/linux/stacktrace.h
@@ -5,15 +5,16 @@
 struct stack_trace {
 	unsigned int nr_entries, max_entries;
 	unsigned long *entries;
+	int skip;	/* input argument: How many entries to skip */
+	int all_contexts; /* input argument: if true do than one stack */
 };
 
 extern void save_stack_trace(struct stack_trace *trace,
-			     struct task_struct *task, int all_contexts,
-			     unsigned int skip);
+			     struct task_struct *task);
 
 extern void print_stack_trace(struct stack_trace *trace, int spaces);
 #else
-# define save_stack_trace(trace, task, all, skip)	do { } while (0)
+# define save_stack_trace(trace, task)			do { } while (0)
 # define print_stack_trace(trace)			do { } while (0)
 #endif
 
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 9bad17884513..900b4cb1a024 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -224,7 +224,10 @@ static int save_trace(struct stack_trace *trace)
 	trace->max_entries = MAX_STACK_TRACE_ENTRIES - nr_stack_trace_entries;
 	trace->entries = stack_trace + nr_stack_trace_entries;
 
-	save_stack_trace(trace, NULL, 0, 3);
+	trace->skip = 3;
+	trace->all_contexts = 0;
+
+	save_stack_trace(trace, NULL);
 
 	trace->max_entries = trace->nr_entries;
 
-- 
cgit v1.2.1


From 3fa7c794fe4dc127f7fac3fad4d13628e68f89ce Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Avoid recursion in lockdep when stack tracer takes locks

The new dwarf2 unwinder needs to take locks to do backtraces
inside modules. This patch makes sure lockdep which calls
stacktrace is not reentered.

Thanks to Ingo for suggesting this simpler approach.

Cc: mingo@elte.hu
Signed-off-by: Andi Kleen <ak@suse.de>
---
 kernel/lockdep.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 900b4cb1a024..c088e5542e84 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -227,7 +227,11 @@ static int save_trace(struct stack_trace *trace)
 	trace->skip = 3;
 	trace->all_contexts = 0;
 
+	/* Make sure to not recurse in case the the unwinder needs to tak
+e	   locks. */
+	lockdep_off();
 	save_stack_trace(trace, NULL);
+	lockdep_on();
 
 	trace->max_entries = trace->nr_entries;
 
-- 
cgit v1.2.1


From b7f5e3c7742d5332b78b831131f43fc3630e6322 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Don't access the APIC in safe_smp_processor_id when it is not
 mapped yet

Lockdep can call the dwarf2 unwinder early, and the dwarf2 code
uses safe_smp_processor_id which tries to access the local APIC page.
But that doesn't work before the APIC code has set up its fixmap.

Check for this case and always return boot cpu then.

Cc: jbeulich@novell.com
Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c | 2 ++
 arch/x86_64/kernel/smp.c  | 2 +-
 include/asm-x86_64/apic.h | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index b95d1e378b96..6472e321cad7 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -38,6 +38,7 @@
 #include <asm/timex.h>
 #include <asm/apic.h>
 
+int apic_mapped;
 int apic_verbosity;
 int apic_runs_main_timer;
 int apic_calibrate_pmtmr __initdata;
@@ -600,6 +601,7 @@ void __init init_apic_mappings(void)
 		apic_phys = mp_lapic_addr;
 
 	set_fixmap_nocache(FIX_APIC_BASE, apic_phys);
+	apic_mapped = 1;
 	apic_printk(APIC_VERBOSE,"mapped APIC to %16lx (%16lx)\n", APIC_BASE, apic_phys);
 
 	/*
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 06af6ca60129..2df988bbf312 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -526,7 +526,7 @@ int safe_smp_processor_id(void)
 {
 	unsigned apicid, i;
 
-	if (disable_apic)
+	if (disable_apic || !apic_mapped)
 		return 0;
 
 	apicid = hard_smp_processor_id();
diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h
index 29ee735278f2..9e66d32330c9 100644
--- a/include/asm-x86_64/apic.h
+++ b/include/asm-x86_64/apic.h
@@ -18,6 +18,7 @@
 extern int apic_verbosity;
 extern int apic_runs_main_timer;
 extern int ioapic_force;
+extern int apic_mapped;
 
 /*
  * Define the default level of output to be very little
-- 
cgit v1.2.1


From c9538ed49272fb244ac06ba643ff076a68a77e12 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Move unwind_init earlier

Needed for use of the unwinder in lockdep, because lockdep runs really
early too.

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 init/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/init/main.c b/init/main.c
index 6b69564738c1..913e48d658ee 100644
--- a/init/main.c
+++ b/init/main.c
@@ -468,6 +468,7 @@ asmlinkage void __init start_kernel(void)
 	 * Need to run as early as possible, to initialize the
 	 * lockdep hash:
 	 */
+	unwind_init();
 	lockdep_init();
 
 	local_irq_disable();
@@ -506,7 +507,6 @@ asmlinkage void __init start_kernel(void)
 		   __stop___param - __start___param,
 		   &unknown_bootoption);
 	sort_main_extable();
-	unwind_init();
 	trap_init();
 	rcu_init();
 	init_IRQ();
-- 
cgit v1.2.1


From c0b766f13d8e1189ce4d00e54700c9d96b543b9a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Merge stacktrace and show_trace

This unifies the standard backtracer and the new stacktrace
in memory backtracer. The standard one is converted to use callbacks
and then reimplement stacktrace using new callbacks.

The main advantage is that stacktrace can now use the new dwarf2 unwinder
and avoid false positives in many cases.

I kept it simple to make sure the standard backtracer stays reliable.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/stacktrace.c | 214 +++++-----------------------------------
 arch/x86_64/kernel/traps.c      |  99 ++++++++++++++-----
 include/asm-x86_64/stacktrace.h |  18 ++++
 3 files changed, 120 insertions(+), 211 deletions(-)
 create mode 100644 include/asm-x86_64/stacktrace.h

diff --git a/arch/x86_64/kernel/stacktrace.c b/arch/x86_64/kernel/stacktrace.c
index 1c022af8fe1e..6026b31d037e 100644
--- a/arch/x86_64/kernel/stacktrace.c
+++ b/arch/x86_64/kernel/stacktrace.c
@@ -7,211 +7,49 @@
  */
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
+#include <linux/module.h>
+#include <asm/stacktrace.h>
 
-#include <asm/smp.h>
-
-static inline int
-in_range(unsigned long start, unsigned long addr, unsigned long end)
+static void save_stack_warning(void *data, char *msg)
 {
-	return addr >= start && addr <= end;
 }
 
-static unsigned long
-get_stack_end(struct task_struct *task, unsigned long stack)
+static void
+save_stack_warning_symbol(void *data, char *msg, unsigned long symbol)
 {
-	unsigned long stack_start, stack_end, flags;
-	int i, cpu;
-
-	/*
-	 * The most common case is that we are in the task stack:
-	 */
-	stack_start = (unsigned long)task->thread_info;
-	stack_end = stack_start + THREAD_SIZE;
-
-	if (in_range(stack_start, stack, stack_end))
-		return stack_end;
-
-	/*
-	 * We are in an interrupt if irqstackptr is set:
-	 */
-	raw_local_irq_save(flags);
-	cpu = safe_smp_processor_id();
-	stack_end = (unsigned long)cpu_pda(cpu)->irqstackptr;
-
-	if (stack_end) {
-		stack_start = stack_end & ~(IRQSTACKSIZE-1);
-		if (in_range(stack_start, stack, stack_end))
-			goto out_restore;
-		/*
-		 * We get here if we are in an IRQ context but we
-		 * are also in an exception stack.
-		 */
-	}
-
-	/*
-	 * Iterate over all exception stacks, and figure out whether
-	 * 'stack' is in one of them:
-	 */
-	for (i = 0; i < N_EXCEPTION_STACKS; i++) {
-		/*
-		 * set 'end' to the end of the exception stack.
-		 */
-		stack_end = per_cpu(init_tss, cpu).ist[i];
-		stack_start = stack_end - EXCEPTION_STKSZ;
-
-		/*
-		 * Is 'stack' above this exception frame's end?
-		 * If yes then skip to the next frame.
-		 */
-		if (stack >= stack_end)
-			continue;
-		/*
-		 * Is 'stack' above this exception frame's start address?
-		 * If yes then we found the right frame.
-		 */
-		if (stack >= stack_start)
-			goto out_restore;
-
-		/*
-		 * If this is a debug stack, and if it has a larger size than
-		 * the usual exception stacks, then 'stack' might still
-		 * be within the lower portion of the debug stack:
-		 */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		if (i == DEBUG_STACK - 1 && stack >= stack_end - DEBUG_STKSZ) {
-			/*
-			 * Black magic. A large debug stack is composed of
-			 * multiple exception stack entries, which we
-			 * iterate through now. Dont look:
-			 */
-			do {
-				stack_end -= EXCEPTION_STKSZ;
-				stack_start -= EXCEPTION_STKSZ;
-			} while (stack < stack_start);
-
-			goto out_restore;
-		}
-#endif
-	}
-	/*
-	 * Ok, 'stack' is not pointing to any of the system stacks.
-	 */
-	stack_end = 0;
-
-out_restore:
-	raw_local_irq_restore(flags);
-
-	return stack_end;
 }
 
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer:
- */
-static inline unsigned long
-save_context_stack(struct stack_trace *trace,
-		   unsigned long stack, unsigned long stack_end)
+static int save_stack_stack(void *data, char *name)
 {
-	int skip = trace->skip;
-	unsigned long addr;
-
-#ifdef CONFIG_FRAME_POINTER
-	unsigned long prev_stack = 0;
+	struct stack_trace *trace = (struct stack_trace *)data;
+	return trace->all_contexts ? 0 : -1;
+}
 
-	while (in_range(prev_stack, stack, stack_end)) {
-		pr_debug("stack:          %p\n", (void *)stack);
-		addr = (unsigned long)(((unsigned long *)stack)[1]);
-		pr_debug("addr:           %p\n", (void *)addr);
-		if (!skip)
-			trace->entries[trace->nr_entries++] = addr-1;
-		else
-			skip--;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-		if (!addr)
-			return 0;
-		/*
-		 * Stack frames must go forwards (otherwise a loop could
-		 * happen if the stackframe is corrupted), so we move
-		 * prev_stack forwards:
-		 */
-		prev_stack = stack;
-		stack = (unsigned long)(((unsigned long *)stack)[0]);
-	}
-	pr_debug("invalid:        %p\n", (void *)stack);
-#else
-	while (stack < stack_end) {
-		addr = ((unsigned long *)stack)[0];
-		stack += sizeof(long);
-		if (__kernel_text_address(addr)) {
-			if (!skip)
-				trace->entries[trace->nr_entries++] = addr-1;
-			else
-				skip--;
-			if (trace->nr_entries >= trace->max_entries)
-				break;
-		}
+static void save_stack_address(void *data, unsigned long addr)
+{
+	struct stack_trace *trace = (struct stack_trace *)data;
+	if (trace->skip > 0) {
+		trace->skip--;
+		return;
 	}
-#endif
-	return stack;
+	if (trace->nr_entries < trace->max_entries - 1)
+		trace->entries[trace->nr_entries++] = addr;
 }
 
-#define MAX_STACKS 10
+static struct stacktrace_ops save_stack_ops = {
+	.warning = save_stack_warning,
+	.warning_symbol = save_stack_warning_symbol,
+	.stack = save_stack_stack,
+	.address = save_stack_address,
+};
 
 /*
  * Save stack-backtrace addresses into a stack_trace buffer.
  */
 void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
 {
-	unsigned long stack = (unsigned long)&stack;
-	int i, nr_stacks = 0, stacks_done[MAX_STACKS];
-
-	WARN_ON(trace->nr_entries || !trace->max_entries);
-
-	if (!task)
-		task = current;
-
-	pr_debug("task: %p, ti: %p\n", task, task->thread_info);
-
-	if (!task || task == current) {
-		/* Grab rbp right from our regs: */
-		asm ("mov %%rbp, %0" : "=r" (stack));
-		pr_debug("rbp:            %p\n", (void *)stack);
-	} else {
-		/* rbp is the last reg pushed by switch_to(): */
-		stack = task->thread.rsp;
-		pr_debug("other task rsp: %p\n", (void *)stack);
-		stack = (unsigned long)(((unsigned long *)stack)[0]);
-		pr_debug("other task rbp: %p\n", (void *)stack);
-	}
-
-	while (1) {
-		unsigned long stack_end = get_stack_end(task, stack);
-
-		pr_debug("stack:          %p\n", (void *)stack);
-		pr_debug("stack end:      %p\n", (void *)stack_end);
-
-		/*
-		 * Invalid stack addres?
-		 */
-		if (!stack_end)
-			return;
-		/*
-		 * Were we in this stack already? (recursion)
-		 */
-		for (i = 0; i < nr_stacks; i++)
-			if (stacks_done[i] == stack_end)
-				return;
-		stacks_done[nr_stacks] = stack_end;
-
-		stack = save_context_stack(trace, stack, stack_end);
-		if (!stack || trace->nr_entries >= trace->max_entries)
-			return;
-		trace->entries[trace->nr_entries++] = ULONG_MAX;
-		if (trace->nr_entries >= trace->max_entries)
-			return;
-		if (++nr_stacks >= MAX_STACKS)
-			return;
-	}
+	dump_trace(task, NULL, NULL, &save_stack_ops, trace);
+	trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
+EXPORT_SYMBOL(save_stack_trace);
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 9ec2b1d5893d..4ac18b02eada 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -45,6 +45,7 @@
 #include <asm/pda.h>
 #include <asm/proto.h>
 #include <asm/nmi.h>
+#include <asm/stacktrace.h>
 
 asmlinkage void divide_error(void);
 asmlinkage void debug(void);
@@ -142,7 +143,7 @@ void printk_address(unsigned long address)
 #endif
 
 static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
-					unsigned *usedp, const char **idp)
+					unsigned *usedp, char **idp)
 {
 	static char ids[][8] = {
 		[DEBUG_STACK - 1] = "#DB",
@@ -234,13 +235,19 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 	return NULL;
 }
 
-static int show_trace_unwind(struct unwind_frame_info *info, void *context)
+struct ops_and_data {
+	struct stacktrace_ops *ops;
+	void *data;
+};
+
+static int dump_trace_unwind(struct unwind_frame_info *info, void *context)
 {
+	struct ops_and_data *oad = (struct ops_and_data *)context;
 	int n = 0;
 
 	while (unwind(info) == 0 && UNW_PC(info)) {
 		n++;
-		printk_address(UNW_PC(info));
+		oad->ops->address(oad->data, UNW_PC(info));
 		if (arch_unw_user_mode(info))
 			break;
 	}
@@ -254,45 +261,51 @@ static int show_trace_unwind(struct unwind_frame_info *info, void *context)
  * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
  */
 
-void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack)
+void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack,
+		struct stacktrace_ops *ops, void *data)
 {
 	const unsigned cpu = safe_smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 
-	printk("\nCall Trace:\n");
-
 	if (!tsk)
 		tsk = current;
 
 	if (call_trace >= 0) {
 		int unw_ret = 0;
 		struct unwind_frame_info info;
+		struct ops_and_data oad = { .ops = ops, .data = data };
 
 		if (regs) {
 			if (unwind_init_frame_info(&info, tsk, regs) == 0)
-				unw_ret = show_trace_unwind(&info, NULL);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		} else if (tsk == current)
-			unw_ret = unwind_init_running(&info, show_trace_unwind, NULL);
+			unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad);
 		else {
 			if (unwind_init_blocked(&info, tsk) == 0)
-				unw_ret = show_trace_unwind(&info, NULL);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		}
 		if (unw_ret > 0) {
 			if (call_trace == 1 && !arch_unw_user_mode(&info)) {
-				print_symbol("DWARF2 unwinder stuck at %s\n",
+				ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n",
 					     UNW_PC(&info));
 				if ((long)UNW_SP(&info) < 0) {
-					printk("Leftover inexact backtrace:\n");
+					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (unsigned long *)UNW_SP(&info);
 				} else
-					printk("Full inexact backtrace again:\n");
+					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
 				return;
 			else
-				printk("Full inexact backtrace again:\n");
+				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
-			printk("Inexact backtrace:\n");
+			ops->warning(data, "Inexact backtrace:\n");
+	}
+	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (tsk && tsk != current)
+			stack = (unsigned long *)tsk->thread.rsp;
 	}
 
 	/*
@@ -312,7 +325,7 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 			 * down the cause of the crash will be able to figure \
 			 * out the call path that was taken. \
 			 */ \
-			printk_address(addr); \
+			ops->address(data, addr);   \
 		} \
 	} while (0)
 
@@ -321,16 +334,17 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 	 * current stack address. If the stacks consist of nested
 	 * exceptions
 	 */
-	for ( ; ; ) {
-		const char *id;
+	for (;;) {
+		char *id;
 		unsigned long *estack_end;
 		estack_end = in_exception_stack(cpu, (unsigned long)stack,
 						&used, &id);
 
 		if (estack_end) {
-			printk(" <%s>", id);
+			if (ops->stack(data, id) < 0)
+				break;
 			HANDLE_STACK (stack < estack_end);
-			printk(" <EOE>");
+			ops->stack(data, "<EOE>");
 			/*
 			 * We link to the next stack via the
 			 * second-to-last pointer (index -2 to end) in the
@@ -345,7 +359,8 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 				(IRQSTACKSIZE - 64) / sizeof(*irqstack);
 
 			if (stack >= irqstack && stack < irqstack_end) {
-				printk(" <IRQ>");
+				if (ops->stack(data, "IRQ") < 0)
+					break;
 				HANDLE_STACK (stack < irqstack_end);
 				/*
 				 * We link to the next stack (which would be
@@ -354,7 +369,7 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 				 */
 				stack = (unsigned long *) (irqstack_end[-1]);
 				irqstack_end = NULL;
-				printk(" <EOI>");
+				ops->stack(data, "EOI");
 				continue;
 			}
 		}
@@ -362,15 +377,53 @@ void show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 	}
 
 	/*
-	 * This prints the process stack:
+	 * This handles the process stack:
 	 */
 	HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0);
 #undef HANDLE_STACK
+}
+EXPORT_SYMBOL(dump_trace);
 
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s\n", msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	printk(" <%s> ", name);
+	return 0;
+}
+
+static void print_trace_address(void *data, unsigned long addr)
+{
+	printk_address(addr);
+}
+
+static struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+void
+show_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack)
+{
+	printk("\nCall Trace:\n");
+	dump_trace(tsk, regs, stack, &print_trace_ops, NULL);
 	printk("\n");
 }
 
-static void _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long * rsp)
+static void
+_show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 {
 	unsigned long *stack;
 	int i;
diff --git a/include/asm-x86_64/stacktrace.h b/include/asm-x86_64/stacktrace.h
new file mode 100644
index 000000000000..5eb9799bef76
--- /dev/null
+++ b/include/asm-x86_64/stacktrace.h
@@ -0,0 +1,18 @@
+#ifndef _ASM_STACKTRACE_H
+#define _ASM_STACKTRACE_H 1
+
+/* Generic stack tracer with callbacks */
+
+struct stacktrace_ops {
+	void (*warning)(void *data, char *msg);
+	/* msg must contain %s for the symbol */
+	void (*warning_symbol)(void *data, char *msg, unsigned long symbol);
+	void (*address)(void *data, unsigned long address);
+	/* On negative return stop dumping */
+	int (*stack)(void *data, char *name);
+};
+
+void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long *stack,
+		struct stacktrace_ops *ops, void *data);
+
+#endif
-- 
cgit v1.2.1


From be7a91709b90825990e571b2f20cea937d5eef6c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Check for end of stack trace before falling back

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/traps.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 4ac18b02eada..28e53342f294 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -292,6 +292,8 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 				if ((long)UNW_SP(&info) < 0) {
 					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (unsigned long *)UNW_SP(&info);
+					if (!stack)
+						return;
 				} else
 					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
-- 
cgit v1.2.1


From 2b14a78cd07a52001b8c3865ed615d8b9b905b78 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: Do stacktracer conversion too

Following x86-64 patches. Reuses code from them in fact.

Convert the standard backtracer to do all output using
callbacks.   Use the x86-64 stack tracer implementation
that uses these callbacks to implement the stacktrace interface.

This allows to use the new dwarf2 unwinder for stacktrace
and get better backtraces.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/Makefile     |   1 +
 arch/i386/kernel/stacktrace.c |  93 ------------------------------------
 arch/i386/kernel/traps.c      | 108 +++++++++++++++++++++++++++++++-----------
 include/asm-i386/stacktrace.h |   1 +
 4 files changed, 82 insertions(+), 121 deletions(-)
 delete mode 100644 arch/i386/kernel/stacktrace.c
 create mode 100644 include/asm-i386/stacktrace.h

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index dab497472deb..1a884b6e6e5c 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -81,4 +81,5 @@ $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
 	$(call if_changed,syscall)
 
 k8-y                      += ../../x86_64/kernel/k8.o
+stacktrace-y		  += ../../x86_64/kernel/stacktrace.o
 
diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c
deleted file mode 100644
index ae3c32a87add..000000000000
--- a/arch/i386/kernel/stacktrace.c
+++ /dev/null
@@ -1,93 +0,0 @@
-/*
- * arch/i386/kernel/stacktrace.c
- *
- * Stack trace management functions
- *
- *  Copyright (C) 2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
- */
-#include <linux/sched.h>
-#include <linux/stacktrace.h>
-
-static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
-{
-	return	p > (void *)tinfo &&
-		p < (void *)tinfo + THREAD_SIZE - 3;
-}
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer:
- */
-static inline unsigned long
-save_context_stack(struct stack_trace *trace, unsigned int skip,
-		   struct thread_info *tinfo, unsigned long *stack,
-		   unsigned long ebp)
-{
-	unsigned long addr;
-
-#ifdef CONFIG_FRAME_POINTER
-	while (valid_stack_ptr(tinfo, (void *)ebp)) {
-		addr = *(unsigned long *)(ebp + 4);
-		if (!skip)
-			trace->entries[trace->nr_entries++] = addr;
-		else
-			skip--;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-		/*
-		 * break out of recursive entries (such as
-		 * end_of_stack_stop_unwind_function):
-	 	 */
-		if (ebp == *(unsigned long *)ebp)
-			break;
-
-		ebp = *(unsigned long *)ebp;
-	}
-#else
-	while (valid_stack_ptr(tinfo, stack)) {
-		addr = *stack++;
-		if (__kernel_text_address(addr)) {
-			if (!skip)
-				trace->entries[trace->nr_entries++] = addr;
-			else
-				skip--;
-			if (trace->nr_entries >= trace->max_entries)
-				break;
-		}
-	}
-#endif
-
-	return ebp;
-}
-
-/*
- * Save stack-backtrace addresses into a stack_trace buffer.
- */
-void save_stack_trace(struct stack_trace *trace, struct task_struct *task)
-{
-	unsigned long ebp;
-	unsigned long *stack = &ebp;
-
-	WARN_ON(trace->nr_entries || !trace->max_entries);
-
-	if (!task || task == current) {
-		/* Grab ebp right from our regs: */
-		asm ("movl %%ebp, %0" : "=r" (ebp));
-	} else {
-		/* ebp is the last reg pushed by switch_to(): */
-		ebp = *(unsigned long *) task->thread.esp;
-	}
-
-	while (1) {
-		struct thread_info *context = (struct thread_info *)
-				((unsigned long)stack & (~(THREAD_SIZE - 1)));
-
-		ebp = save_context_stack(trace, trace->skip, context, stack, ebp);
-		stack = (unsigned long *)context->previous_esp;
-		if (!stack || trace->nr_entries >= trace->max_entries)
-			break;
-		trace->entries[trace->nr_entries++] = ULONG_MAX;
-		if (trace->nr_entries >= trace->max_entries)
-			break;
-	}
-}
-
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 3c85c89f68d8..4ced4285163b 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -51,6 +51,7 @@
 #include <asm/smp.h>
 #include <asm/arch_hooks.h>
 #include <asm/kdebug.h>
+#include <asm/stacktrace.h>
 
 #include <linux/module.h>
 
@@ -118,26 +119,16 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
 		p < (void *)tinfo + THREAD_SIZE - 3;
 }
 
-/*
- * Print one address/symbol entries per line.
- */
-static inline void print_addr_and_symbol(unsigned long addr, char *log_lvl)
-{
-	printk(" [<%08lx>] ", addr);
-
-	print_symbol("%s\n", addr);
-}
-
 static inline unsigned long print_context_stack(struct thread_info *tinfo,
 				unsigned long *stack, unsigned long ebp,
-				char *log_lvl)
+				struct stacktrace_ops *ops, void *data)
 {
 	unsigned long addr;
 
 #ifdef	CONFIG_FRAME_POINTER
 	while (valid_stack_ptr(tinfo, (void *)ebp)) {
 		addr = *(unsigned long *)(ebp + 4);
-		print_addr_and_symbol(addr, log_lvl);
+		ops->address(data, addr);
 		/*
 		 * break out of recursive entries (such as
 		 * end_of_stack_stop_unwind_function):
@@ -150,28 +141,35 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
 	while (valid_stack_ptr(tinfo, stack)) {
 		addr = *stack++;
 		if (__kernel_text_address(addr))
-			print_addr_and_symbol(addr, log_lvl);
+			ops->address(data, addr);
 	}
 #endif
 	return ebp;
 }
 
+struct ops_and_data {
+	struct stacktrace_ops *ops;
+	void *data;
+};
+
 static asmlinkage int
-show_trace_unwind(struct unwind_frame_info *info, void *log_lvl)
+dump_trace_unwind(struct unwind_frame_info *info, void *data)
 {
+	struct ops_and_data *oad = (struct ops_and_data *)data;
 	int n = 0;
 
 	while (unwind(info) == 0 && UNW_PC(info)) {
 		n++;
-		print_addr_and_symbol(UNW_PC(info), log_lvl);
+		oad->ops->address(oad->data, UNW_PC(info));
 		if (arch_unw_user_mode(info))
 			break;
 	}
 	return n;
 }
 
-static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
-			       unsigned long *stack, char *log_lvl)
+void dump_trace(struct task_struct *task, struct pt_regs *regs,
+	        unsigned long *stack,
+		struct stacktrace_ops *ops, void *data)
 {
 	unsigned long ebp;
 
@@ -181,31 +179,37 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 	if (call_trace >= 0) {
 		int unw_ret = 0;
 		struct unwind_frame_info info;
+		struct ops_and_data oad = { .ops = ops, .data = data };
 
 		if (regs) {
 			if (unwind_init_frame_info(&info, task, regs) == 0)
-				unw_ret = show_trace_unwind(&info, log_lvl);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		} else if (task == current)
-			unw_ret = unwind_init_running(&info, show_trace_unwind, log_lvl);
+			unw_ret = unwind_init_running(&info, dump_trace_unwind, &oad);
 		else {
 			if (unwind_init_blocked(&info, task) == 0)
-				unw_ret = show_trace_unwind(&info, log_lvl);
+				unw_ret = dump_trace_unwind(&info, &oad);
 		}
 		if (unw_ret > 0) {
 			if (call_trace == 1 && !arch_unw_user_mode(&info)) {
-				print_symbol("DWARF2 unwinder stuck at %s\n",
+				ops->warning_symbol(data, "DWARF2 unwinder stuck at %s\n",
 					     UNW_PC(&info));
 				if (UNW_SP(&info) >= PAGE_OFFSET) {
-					printk("Leftover inexact backtrace:\n");
+					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (void *)UNW_SP(&info);
 				} else
-					printk("Full inexact backtrace again:\n");
+					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
 				return;
 			else
-				printk("Full inexact backtrace again:\n");
+				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
-			printk("Inexact backtrace:\n");
+			ops->warning(data, "Inexact backtrace:\n");
+	} else 	if (!stack) {
+		unsigned long dummy;
+		stack = &dummy;
+		if (task && task != current)
+			stack = (unsigned long *)task->thread.esp;
 	}
 
 	if (task == current) {
@@ -220,15 +224,63 @@ static void show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
 		struct thread_info *context;
 		context = (struct thread_info *)
 			((unsigned long)stack & (~(THREAD_SIZE - 1)));
-		ebp = print_context_stack(context, stack, ebp, log_lvl);
+		ebp = print_context_stack(context, stack, ebp, ops, data);
+		/* Should be after the line below, but somewhere
+		   in early boot context comes out corrupted and we
+		   can't reference it -AK */
+		if (ops->stack(data, "IRQ") < 0)
+			break;
 		stack = (unsigned long*)context->previous_esp;
 		if (!stack)
 			break;
-		printk("%s =======================\n", log_lvl);
 	}
 }
+EXPORT_SYMBOL(dump_trace);
+
+static void
+print_trace_warning_symbol(void *data, char *msg, unsigned long symbol)
+{
+	printk(data);
+	print_symbol(msg, symbol);
+	printk("\n");
+}
+
+static void print_trace_warning(void *data, char *msg)
+{
+	printk("%s%s\n", (char *)data, msg);
+}
+
+static int print_trace_stack(void *data, char *name)
+{
+	return 0;
+}
+
+/*
+ * Print one address/symbol entries per line.
+ */
+static void print_trace_address(void *data, unsigned long addr)
+{
+	printk("%s [<%08lx>] ", (char *)data, addr);
+	print_symbol("%s\n", addr);
+}
+
+static struct stacktrace_ops print_trace_ops = {
+	.warning = print_trace_warning,
+	.warning_symbol = print_trace_warning_symbol,
+	.stack = print_trace_stack,
+	.address = print_trace_address,
+};
+
+static void
+show_trace_log_lvl(struct task_struct *task, struct pt_regs *regs,
+		   unsigned long * stack, char *log_lvl)
+{
+	dump_trace(task, regs, stack, &print_trace_ops, log_lvl);
+	printk("%s =======================\n", log_lvl);
+}
 
-void show_trace(struct task_struct *task, struct pt_regs *regs, unsigned long * stack)
+void show_trace(struct task_struct *task, struct pt_regs *regs,
+		unsigned long * stack)
 {
 	show_trace_log_lvl(task, regs, stack, "");
 }
diff --git a/include/asm-i386/stacktrace.h b/include/asm-i386/stacktrace.h
new file mode 100644
index 000000000000..7d1f6a5cbfca
--- /dev/null
+++ b/include/asm-i386/stacktrace.h
@@ -0,0 +1 @@
+#include <asm-x86_64/stacktrace.h>
-- 
cgit v1.2.1


From 950fee84557416a3427dd404a13addc4be7b3e6c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: Terminate backtrace fallback early if unwinder stack
 pointer is zero

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 4ced4285163b..86fa7e47f301 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -197,6 +197,8 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 				if (UNW_SP(&info) >= PAGE_OFFSET) {
 					ops->warning(data, "Leftover inexact backtrace:\n");
 					stack = (void *)UNW_SP(&info);
+					if (!stack)
+						return;
 				} else
 					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
-- 
cgit v1.2.1


From a32cf3975bed3b84491f8ffeb24abe8c45d86ab0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: Get ebp from unwinder state when continuing fallback
 backtrace

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c  | 22 ++++++++++++++--------
 include/asm-i386/unwind.h |  3 +++
 2 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 86fa7e47f301..bdf949c30c7c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -171,7 +171,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 	        unsigned long *stack,
 		struct stacktrace_ops *ops, void *data)
 {
-	unsigned long ebp;
+	unsigned long ebp = 0;
 
 	if (!task)
 		task = current;
@@ -199,6 +199,7 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 					stack = (void *)UNW_SP(&info);
 					if (!stack)
 						return;
+					ebp = UNW_FP(&info);
 				} else
 					ops->warning(data, "Full inexact backtrace again:\n");
 			} else if (call_trace >= 1)
@@ -207,20 +208,25 @@ void dump_trace(struct task_struct *task, struct pt_regs *regs,
 				ops->warning(data, "Full inexact backtrace again:\n");
 		} else
 			ops->warning(data, "Inexact backtrace:\n");
-	} else 	if (!stack) {
+	}
+	if (!stack) {
 		unsigned long dummy;
 		stack = &dummy;
 		if (task && task != current)
 			stack = (unsigned long *)task->thread.esp;
 	}
 
-	if (task == current) {
-		/* Grab ebp right from our regs */
-		asm ("movl %%ebp, %0" : "=r" (ebp) : );
-	} else {
-		/* ebp is the last reg pushed by switch_to */
-		ebp = *(unsigned long *) task->thread.esp;
+#ifdef CONFIG_FRAME_POINTER
+	if (!ebp) {
+		if (task == current) {
+			/* Grab ebp right from our regs */
+			asm ("movl %%ebp, %0" : "=r" (ebp) : );
+		} else {
+			/* ebp is the last reg pushed by switch_to */
+			ebp = *(unsigned long *) task->thread.esp;
+		}
 	}
+#endif
 
 	while (1) {
 		struct thread_info *context;
diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h
index 4c1a0b968569..f0ac399bae3c 100644
--- a/include/asm-i386/unwind.h
+++ b/include/asm-i386/unwind.h
@@ -28,6 +28,8 @@ struct unwind_frame_info
 #define FRAME_LINK_OFFSET    0
 #define STACK_BOTTOM(tsk)    STACK_LIMIT((tsk)->thread.esp0)
 #define STACK_TOP(tsk)       ((tsk)->thread.esp0)
+#else
+#define UNW_FP(frame) ((void)(frame), 0)
 #endif
 #define STACK_LIMIT(ptr)     (((ptr) - 1) & ~(THREAD_SIZE - 1))
 
@@ -88,6 +90,7 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info)
 
 #define UNW_PC(frame) ((void)(frame), 0)
 #define UNW_SP(frame) ((void)(frame), 0)
+#define UNW_FP(frame) ((void)(frame), 0)
 
 static inline int arch_unw_user_mode(const void *info)
 {
-- 
cgit v1.2.1


From 3b8d1fe0f28202ce1592d9fbc216959b49b72c95 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] Don't force frame pointers for lockdep

Now that stacktrace supports dwarf2 don't force frame pointers for lockdep anymore

Cc: mingo@elte.hu
Signed-off-by: Andi Kleen <ak@suse.de>
---
 lib/Kconfig.debug | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 554ee688a9f8..c795a17a7043 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -218,7 +218,7 @@ config LOCKDEP
 	bool
 	depends on DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
 	select STACKTRACE
-	select FRAME_POINTER
+	select FRAME_POINTER if !X86
 	select KALLSYMS
 	select KALLSYMS_ALL
 
-- 
cgit v1.2.1


From 5758d5dfef1c514200bda3f29ba700f1c3e3bc99 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: fix dubious segment register clear in cpu_init()

Fix a very dubious piece of code in
arch/i386/kernel/cpu/common.c:cpu_init().  This clears out %fs and
%gs, but clobbers %eax in the process without telling gcc.  It turns
out that gcc happens to be not using %eax at that point anyway so it
doesn't matter much, but it looks like a bomb waiting to go off.

This does end up saving an instruction, because gcc wants %eax==0 for
the set_debugreg()s below.

Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 70c87de582c7..730a7ab75009 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -675,7 +675,7 @@ old_gdt:
 #endif
 
 	/* Clear %fs and %gs. */
-	asm volatile ("xorl %eax, %eax; movl %eax, %fs; movl %eax, %gs");
+	asm volatile ("movl %0, %%fs; movl %0, %%gs" : : "r" (0));
 
 	/* Clear all 6 debug registers: */
 	set_debugreg(0, 0);
-- 
cgit v1.2.1


From 3ca113ea74836a80645c79adba24caaa7a74120c Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] i386: don't taint UP K7's running SMP kernels.

We have a test that looks for invalid pairings of certain athlon/durons
that weren't designed for SMP, and taint accordingly (with 'S') if we find
such a configuration.  However, this test shouldn't fire if there's only
a single CPU present. It's perfectly valid for an SMP kernel to boot on UP
hardware for example.

AK: changed to num_possible_cpus()

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/smpboot.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 517eb3874550..020d873b7d21 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -177,6 +177,9 @@ static void __devinit smp_store_cpu_info(int id)
 	 */
 	if ((c->x86_vendor == X86_VENDOR_AMD) && (c->x86 == 6)) {
 
+		if (num_possible_cpus() == 1)
+			goto valid_k7;
+
 		/* Athlon 660/661 is valid. */	
 		if ((c->x86_model==6) && ((c->x86_mask==0) || (c->x86_mask==1)))
 			goto valid_k7;
-- 
cgit v1.2.1


From d28c4393a7bf558538e9def269c1caeab6ec056f Mon Sep 17 00:00:00 2001
From: "Prasanna S.P" <prasanna@in.ibm.com>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] x86: error_code is not safe for kprobes

This patch moves the entry.S:error_entry to .kprobes.text section,
since code marked unsafe for kprobes jumps directly to entry.S::error_entry,
that must be marked unsafe as well.
This patch also moves all the ".previous.text" asm directives to ".previous"
for kprobes section.

AK: Following a similar i386 patch from Chuck Ebbert
AK: Also merged Jeremy's fix in.

+From: Jeremy Fitzhardinge <jeremy@goop.org>

KPROBE_ENTRY does a .section .kprobes.text, and expects its users to
do a .previous at the end of the function.

Unfortunately, if any code within the function switches sections, for
example .fixup, then the .previous ends up putting all subsequent code
into .fixup.  Worse, any subsequent .fixup code gets intermingled with
the code its supposed to be fixing (which is also in .fixup).  It's
surprising this didn't cause more havok.

The fix is to use .pushsection/.popsection, so this stuff nests
properly.  A further cleanup would be to get rid of all
.section/.previous pairs, since they're inherently fragile.

+From: Chuck Ebbert <76306.1226@compuserve.com>

Because code marked unsafe for kprobes jumps directly to
entry.S::error_code, that must be marked unsafe as well.
The easiest way to do that is to move the page fault entry
point to just before error_code and let it inherit the same
section.

Also moved all the ".previous" asm directives for kprobes
sections to column 1 and removed ".text" from them.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S   | 25 +++++++++++++------------
 arch/x86_64/kernel/entry.S | 19 +++++++------------
 include/linux/linkage.h    |  6 +++++-
 3 files changed, 25 insertions(+), 25 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 87f9f60b803b..ba22ec8fab54 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -591,11 +591,9 @@ ENTRY(name)				\
 /* The include is where all of the SMP etc. interrupts come from */
 #include "entry_arch.h"
 
-ENTRY(divide_error)
-	RING0_INT_FRAME
-	pushl $0			# no error code
-	CFI_ADJUST_CFA_OFFSET 4
-	pushl $do_divide_error
+KPROBE_ENTRY(page_fault)
+	RING0_EC_FRAME
+	pushl $do_page_fault
 	CFI_ADJUST_CFA_OFFSET 4
 	ALIGN
 error_code:
@@ -645,6 +643,7 @@ error_code:
 	call *%edi
 	jmp ret_from_exception
 	CFI_ENDPROC
+KPROBE_END(page_fault)
 
 ENTRY(coprocessor_error)
 	RING0_INT_FRAME
@@ -720,7 +719,8 @@ debug_stack_correct:
 	call do_debug
 	jmp ret_from_exception
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(debug)
+
 /*
  * NMI is doubly nasty. It can happen _while_ we're handling
  * a debug fault, and the debug fault hasn't yet been able to
@@ -816,7 +816,7 @@ KPROBE_ENTRY(int3)
 	call do_int3
 	jmp ret_from_exception
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(int3)
 
 ENTRY(overflow)
 	RING0_INT_FRAME
@@ -881,7 +881,7 @@ KPROBE_ENTRY(general_protection)
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
-	.previous .text
+KPROBE_END(general_protection)
 
 ENTRY(alignment_check)
 	RING0_EC_FRAME
@@ -890,13 +890,14 @@ ENTRY(alignment_check)
 	jmp error_code
 	CFI_ENDPROC
 
-KPROBE_ENTRY(page_fault)
-	RING0_EC_FRAME
-	pushl $do_page_fault
+ENTRY(divide_error)
+	RING0_INT_FRAME
+	pushl $0			# no error code
+	CFI_ADJUST_CFA_OFFSET 4
+	pushl $do_divide_error
 	CFI_ADJUST_CFA_OFFSET 4
 	jmp error_code
 	CFI_ENDPROC
-	.previous .text
 
 #ifdef CONFIG_X86_MCE
 ENTRY(machine_check)
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 2092f565aa87..780f9b26169f 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -819,7 +819,7 @@ paranoid_schedule\trace:
  * Exception entry point. This expects an error code/orig_rax on the stack
  * and the exception handler in %rax.	
  */ 		  				
-ENTRY(error_entry)
+KPROBE_ENTRY(error_entry)
 	_frame RDI
 	/* rdi slot contains rax, oldrax contains error code */
 	cld	
@@ -903,7 +903,7 @@ error_kernelspace:
 	cmpq $gs_change,RIP(%rsp)
         je   error_swapgs
 	jmp  error_sti
-END(error_entry)
+KPROBE_END(error_entry)
 	
        /* Reload gs selector with exception handling */
        /* edi:  new selector */ 
@@ -1025,8 +1025,7 @@ ENDPROC(execve)
 
 KPROBE_ENTRY(page_fault)
 	errorentry do_page_fault
-END(page_fault)
-	.previous .text
+KPROBE_END(page_fault)
 
 ENTRY(coprocessor_error)
 	zeroentry do_coprocessor_error
@@ -1047,8 +1046,7 @@ KPROBE_ENTRY(debug)
 	CFI_ADJUST_CFA_OFFSET 8		
 	paranoidentry do_debug, DEBUG_STACK
 	paranoidexit
-END(debug)
-	.previous .text
+KPROBE_END(debug)
 
 	/* runs on exception stack */	
 KPROBE_ENTRY(nmi)
@@ -1062,8 +1060,7 @@ KPROBE_ENTRY(nmi)
 	jmp paranoid_exit1
  	CFI_ENDPROC
 #endif
-END(nmi)
-	.previous .text
+KPROBE_END(nmi)
 
 KPROBE_ENTRY(int3)
  	INTR_FRAME
@@ -1072,8 +1069,7 @@ KPROBE_ENTRY(int3)
  	paranoidentry do_int3, DEBUG_STACK
  	jmp paranoid_exit1
  	CFI_ENDPROC
-END(int3)
-	.previous .text
+KPROBE_END(int3)
 
 ENTRY(overflow)
 	zeroentry do_overflow
@@ -1121,8 +1117,7 @@ END(stack_segment)
 
 KPROBE_ENTRY(general_protection)
 	errorentry do_general_protection
-END(general_protection)
-	.previous .text
+KPROBE_END(general_protection)
 
 ENTRY(alignment_check)
 	errorentry do_alignment_check
diff --git a/include/linux/linkage.h b/include/linux/linkage.h
index 932021f872d5..6c9873f88287 100644
--- a/include/linux/linkage.h
+++ b/include/linux/linkage.h
@@ -35,9 +35,13 @@
 #endif
 
 #define KPROBE_ENTRY(name) \
-  .section .kprobes.text, "ax"; \
+  .pushsection .kprobes.text, "ax"; \
   ENTRY(name)
 
+#define KPROBE_END(name) \
+  END(name);		 \
+  .popsection
+
 #ifndef END
 #define END(name) \
   .size name, .-name
-- 
cgit v1.2.1


From cbf9b4bb76c9ce53b7fdde0dffcd000951b5f0d4 Mon Sep 17 00:00:00 2001
From: Dimitri Sivanich <sivanich@sgi.com>
Date: Tue, 26 Sep 2006 10:52:34 +0200
Subject: [PATCH] X86_64 monotonic_clock goes backwards

I've noticed some erratic behavior while testing the X86_64 version
of monotonic_clock().

While spinning in a loop reading monotonic clock values (pinned to a
single cpu) I noticed that the difference between subsequent values
occasionally went negative (time going backwards).

I found that in the following code:
                this_offset = get_cycles_sync();
                /* FIXME: 1000 or 1000000? */
-->             offset = (this_offset - last_offset)*1000 / cpu_khz;
        }
        return base + offset;

the offset sometimes turns out to be 0, even though
this_offset > last_offset.

+Added fix From: Toyo Abe <toyoa@mvista.com>

The x86_64-mm-monotonic-clock.patch in 2.6.18-rc4-mm2 made a change to
the updating of monotonic_base. It now uses cycles_2_ns().

I suggest that a set_cyc2ns_scale() should be done prior to the setup_irq().
Because cycles_2_ns() can be called from the timer ISR right after the irq0
is enabled.

Signed-off-by: Toyo Abe <toyoa@mvista.com>
Signed-off-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/time.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index d66c7f750e75..97115e608ed8 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -276,6 +276,7 @@ static void set_rtc_mmss(unsigned long nowtime)
  *		Note: This function is required to return accurate
  *		time even in the absence of multiple timer ticks.
  */
+static inline unsigned long long cycles_2_ns(unsigned long long cyc);
 unsigned long long monotonic_clock(void)
 {
 	unsigned long seq;
@@ -300,8 +301,7 @@ unsigned long long monotonic_clock(void)
 			base = monotonic_base;
 		} while (read_seqretry(&xtime_lock, seq));
 		this_offset = get_cycles_sync();
-		/* FIXME: 1000 or 1000000? */
-		offset = (this_offset - last_offset)*1000 / cpu_khz;
+		offset = cycles_2_ns(this_offset - last_offset);
 	}
 	return base + offset;
 }
@@ -405,8 +405,7 @@ void main_timer_handler(struct pt_regs *regs)
 			offset %= USEC_PER_TICK;
 		}
 
-		/* FIXME: 1000 or 1000000? */
-		monotonic_base += (tsc - vxtime.last_tsc) * 1000000 / cpu_khz;
+		monotonic_base += cycles_2_ns(tsc - vxtime.last_tsc);
 
 		vxtime.last_tsc = tsc - vxtime.quot * delay / vxtime.tsc_quot;
 
@@ -929,10 +928,8 @@ void __init time_init(void)
 	vxtime.quot = (USEC_PER_SEC << US_SCALE) / vxtime_hz;
 	vxtime.tsc_quot = (USEC_PER_MSEC << US_SCALE) / cpu_khz;
 	vxtime.last_tsc = get_cycles_sync();
-	setup_irq(0, &irq0);
-
 	set_cyc2ns_scale(cpu_khz);
-
+	setup_irq(0, &irq0);
 	hotcpu_notifier(time_cpu_notifier, 0);
 	time_cpu_notifier(NULL, CPU_ONLINE, (void *)(long)smp_processor_id());
 
-- 
cgit v1.2.1


From 1edf777803bdd2aeeb04cf44508fd9b88737fba8 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386/x86-64: Improve Kconfig description of CRASH_DUMP

Improve Kconfig description of CONFIG_CRASH_DUMP. Previously
it was too brief to be useful.

Cc: vgoyal@in.ibm.com
Cc: ebiederm@xmission.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Kconfig   | 7 +++++++
 arch/x86_64/Kconfig | 9 ++++++++-
 2 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index c134a545b37e..e34a4dbd6b92 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -761,6 +761,13 @@ config CRASH_DUMP
 	depends on HIGHMEM
 	help
 	  Generate crash dump after being started by kexec.
+          This should be normally only set in special crash dump kernels
+	  which are loaded in the main kernel with kexec-tools into
+	  a specially reserved region and then later executed after
+	  a crash by kdump/kexec. The crash dump kernel must be compiled
+          to a memory address not used by the main kernel or BIOS using
+          PHYSICAL_START.
+	  For more details see Documentation/kdump/kdump.txt
 
 config PHYSICAL_START
 	hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 6cd4878625f1..927b392a4aa0 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -488,7 +488,14 @@ config CRASH_DUMP
 	bool "kernel crash dumps (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
 	help
-		Generate crash dump after being started by kexec.
+          Generate crash dump after being started by kexec.
+          This should be normally only set in special crash dump kernels
+          which are loaded in the main kernel with kexec-tools into
+          a specially reserved region and then later executed after
+          a crash by kdump/kexec. The crash dump kernel must be compiled
+	  to a memory address not used by the main kernel or BIOS using
+	  PHYSICAL_START.
+          For more details see Documentation/kdump/kdump.txt
 
 config PHYSICAL_START
 	hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
-- 
cgit v1.2.1


From 2717941c6a1d8fa543ddca337d450ab30ef91543 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] Make boot_param_data pure BSS

Since it's all zero.

Actually I think gcc 4+ will do that automatically, but earlier compilers won't
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 646caa0a20bb..b09e60fa96b4 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -24,7 +24,7 @@
 #include <asm/proto.h>
 #include <asm/sections.h>
 
-char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,};
+char x86_boot_params[BOOT_PARAM_SIZE] __initdata;
 
 cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
 
-- 
cgit v1.2.1


From 5e4edbb711417be40f350a319db39888a4edd450 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Fix warning in mpparse.c

Fix

linux/arch/i386/kernel/mpparse.c: In function #MP_bus_info#:
linux/arch/i386/kernel/mpparse.c:232: warning: comparison is always false due to limited range of data type

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/mpparse.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index bdaa75a5bc1c..772a4233bfe6 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -229,12 +229,14 @@ static void __init MP_bus_info (struct mpc_config_bus *m)
 
 	mpc_oem_bus_info(m, str, translation_table[mpc_record]);
 
+#if MAX_MP_BUSSES < 256
 	if (m->mpc_busid >= MAX_MP_BUSSES) {
 		printk(KERN_WARNING "MP table busid value (%d) for bustype %s "
 			" is too large, max. supported is %d\n",
 			m->mpc_busid, str, MAX_MP_BUSSES - 1);
 		return;
 	}
+#endif
 
 	if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) {
 		mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
-- 
cgit v1.2.1


From 273819a2d982faace30e587b86a0683882251fe7 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] make fault notifier unconditional and export it

It's needed for external debuggers and overhead is very small.

Also make the actual notifier chain they use static

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/fault.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index 449a437f5d70..3ae3e7bb2fce 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -40,8 +40,7 @@
 #define PF_RSVD	(1<<3)
 #define PF_INSTR	(1<<4)
 
-#ifdef CONFIG_KPROBES
-ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+static ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
 
 /* Hook to register for page fault notifications */
 int register_page_fault_notifier(struct notifier_block *nb)
@@ -49,11 +48,13 @@ int register_page_fault_notifier(struct notifier_block *nb)
 	vmalloc_sync_all();
 	return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
 }
+EXPORT_SYMBOL_GPL(register_page_fault_notifier);
 
 int unregister_page_fault_notifier(struct notifier_block *nb)
 {
 	return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
 }
+EXPORT_SYMBOL_GPL(unregister_page_fault_notifier);
 
 static inline int notify_page_fault(enum die_val val, const char *str,
 			struct pt_regs *regs, long err, int trap, int sig)
@@ -67,13 +68,6 @@ static inline int notify_page_fault(enum die_val val, const char *str,
 	};
 	return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
 }
-#else
-static inline int notify_page_fault(enum die_val val, const char *str,
-			struct pt_regs *regs, long err, int trap, int sig)
-{
-	return NOTIFY_DONE;
-}
-#endif
 
 void bust_spinlocks(int yes)
 {
-- 
cgit v1.2.1


From 474c256841074b913e76e392082373e12103a75d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: make fault notifier unconditional and export it

It's needed for external debuggers and overhead is very small.

Also make the actual notifier chain they use static

Cc: jbeulich@novell.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/mm/fault.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index f7279468323a..0ce86168a0b1 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -30,18 +30,20 @@
 
 extern void die(const char *,struct pt_regs *,long);
 
-#ifdef CONFIG_KPROBES
-ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+static ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+
 int register_page_fault_notifier(struct notifier_block *nb)
 {
 	vmalloc_sync_all();
 	return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
 }
+EXPORT_SYMBOL_GPL(register_page_fault_notifier);
 
 int unregister_page_fault_notifier(struct notifier_block *nb)
 {
 	return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
 }
+EXPORT_SYMBOL_GPL(unregister_page_fault_notifier);
 
 static inline int notify_page_fault(enum die_val val, const char *str,
 			struct pt_regs *regs, long err, int trap, int sig)
@@ -55,14 +57,6 @@ static inline int notify_page_fault(enum die_val val, const char *str,
 	};
 	return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
 }
-#else
-static inline int notify_page_fault(enum die_val val, const char *str,
-			struct pt_regs *regs, long err, int trap, int sig)
-{
-	return NOTIFY_DONE;
-}
-#endif
-
 
 /*
  * Unlock any spinlocks which will prevent us from getting the
-- 
cgit v1.2.1


From e8924acb2ef46b96c93f97025815ef3843cb67a2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Make acpi_force static

acpi_force can become static.

Cc: len.brown@intel.com

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index c809a3f4c0bf..0fba420d668a 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -36,7 +36,7 @@
 #include <asm/io.h>
 #include <asm/mpspec.h>
 
-int __initdata acpi_force = 0;
+static int __initdata acpi_force = 0;
 
 #ifdef	CONFIG_ACPI
 int acpi_disabled = 0;
-- 
cgit v1.2.1


From 3d08a256da8aed5300bd0752200ece426f49b050 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Make enable_local_apic static

enable_local_apic can now become static.

Cc: len.brown@intel.com

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/apic.c | 13 ++++++++++++-
 include/asm-i386/apic.h | 12 ------------
 2 files changed, 12 insertions(+), 13 deletions(-)

diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index ce75e71b694f..90faae5c5d30 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -52,7 +52,18 @@ static cpumask_t timer_bcast_ipi;
 /*
  * Knob to control our willingness to enable the local APIC.
  */
-int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+static int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
+
+static inline void lapic_disable(void)
+{
+	enable_local_apic = -1;
+	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
+}
+
+static inline void lapic_enable(void)
+{
+	enable_local_apic = 1;
+}
 
 /*
  * Debug level
diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h
index 8c5331ceca7b..3a42b7d6fc92 100644
--- a/include/asm-i386/apic.h
+++ b/include/asm-i386/apic.h
@@ -16,20 +16,8 @@
 #define APIC_VERBOSE 1
 #define APIC_DEBUG   2
 
-extern int enable_local_apic;
 extern int apic_verbosity;
 
-static inline void lapic_disable(void)
-{
-	enable_local_apic = -1;
-	clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
-}
-
-static inline void lapic_enable(void)
-{
-	enable_local_apic = 1;
-}
-
 /*
  * Define the default level of output to be very little
  * This can be turned up by using apic=verbose for more
-- 
cgit v1.2.1


From 02ba1a32dbd3d406530a17a2643a8f0f8cbf3acc Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: move kernel_thread_helper into entry.S

And add proper CFI annotation to it which was previously
impossible. This prevents "stuck" messages by the dwarf2 unwinder
when reaching the top of a kernel stack.

Includes feedback from Jan Beulich

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S   | 13 +++++++++++++
 arch/i386/kernel/process.c |  9 ---------
 2 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index ba22ec8fab54..dede506e5bd0 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -950,6 +950,19 @@ ENTRY(arch_unwind_init_running)
 ENDPROC(arch_unwind_init_running)
 #endif
 
+ENTRY(kernel_thread_helper)
+	pushl $0		# fake return address for unwinder
+	CFI_STARTPROC
+	movl %edx,%eax
+	push %edx
+	CFI_ADJUST_CFA_OFFSET 4
+	call *%ebx
+	push %eax
+	CFI_ADJUST_CFA_OFFSET 4
+	call do_exit
+	CFI_ENDPROC
+ENDPROC(kernel_thread_helper)
+
 .section .rodata,"a"
 #include "syscall_table.S"
 
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index b741c3e1a5eb..220aeca59c3a 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -321,15 +321,6 @@ void show_regs(struct pt_regs * regs)
  * the "args".
  */
 extern void kernel_thread_helper(void);
-__asm__(".section .text\n"
-	".align 4\n"
-	"kernel_thread_helper:\n\t"
-	"movl %edx,%eax\n\t"
-	"pushl %edx\n\t"
-	"call *%ebx\n\t"
-	"pushl %eax\n\t"
-	"call do_exit\n"
-	".previous");
 
 /*
  * Create a kernel thread
-- 
cgit v1.2.1


From 522e93e3fcdbf00ba85c72fde6df28cfc0486a65 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Descriptor and trap table cleanups.

The implementation comes from Zach's [RFC, PATCH 10/24] i386 Vmi
descriptor changes:

Descriptor and trap table cleanups.  Add cleanly written accessors for
IDT and GDT gates so the subarch may override them.  Note that this
allows the hypervisor to transparently tweak the DPL of the descriptors
as well as the RPL of segments in those descriptors, with no unnecessary
kernel code modification.  It also allows the hypervisor implementation
of the VMI to tweak the gates, allowing for custom exception frames or
extra layers of indirection above the guest fault / IRQ handlers.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c |  24 ++--------
 include/asm-i386/desc.h  | 121 +++++++++++++++++++++++++++++------------------
 2 files changed, 81 insertions(+), 64 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index bdf949c30c7c..00d643f3de41 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -1165,20 +1165,6 @@ void __init trap_init_f00f_bug(void)
 }
 #endif
 
-#define _set_gate(gate_addr,type,dpl,addr,seg) \
-do { \
-  int __d0, __d1; \
-  __asm__ __volatile__ ("movw %%dx,%%ax\n\t" \
-	"movw %4,%%dx\n\t" \
-	"movl %%eax,%0\n\t" \
-	"movl %%edx,%1" \
-	:"=m" (*((long *) (gate_addr))), \
-	 "=m" (*(1+(long *) (gate_addr))), "=&a" (__d0), "=&d" (__d1) \
-	:"i" ((short) (0x8000+(dpl<<13)+(type<<8))), \
-	 "3" ((char *) (addr)),"2" ((seg) << 16)); \
-} while (0)
-
-
 /*
  * This needs to use 'idt_table' rather than 'idt', and
  * thus use the _nonmapped_ version of the IDT, as the
@@ -1187,7 +1173,7 @@ do { \
  */
 void set_intr_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,14,0,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_INT, addr, __KERNEL_CS);
 }
 
 /*
@@ -1195,22 +1181,22 @@ void set_intr_gate(unsigned int n, void *addr)
  */
 static inline void set_system_intr_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n, 14, 3, addr, __KERNEL_CS);
+	_set_gate(n, DESCTYPE_INT | DESCTYPE_DPL3, addr, __KERNEL_CS);
 }
 
 static void __init set_trap_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,15,0,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_TRAP, addr, __KERNEL_CS);
 }
 
 static void __init set_system_gate(unsigned int n, void *addr)
 {
-	_set_gate(idt_table+n,15,3,addr,__KERNEL_CS);
+	_set_gate(n, DESCTYPE_TRAP | DESCTYPE_DPL3, addr, __KERNEL_CS);
 }
 
 static void __init set_task_gate(unsigned int n, unsigned int gdt_entry)
 {
-	_set_gate(idt_table+n,5,0,0,(gdt_entry<<3));
+	_set_gate(n, DESCTYPE_TASK, (void *)0, (gdt_entry<<3));
 }
 
 
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 89b8b82c82b3..5db9e96e8dc1 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -33,50 +33,99 @@ static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 	return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
 }
 
+/*
+ * This is the ldt that every process will get unless we need
+ * something other than this.
+ */
+extern struct desc_struct default_ldt[];
+extern struct desc_struct idt_table[];
+extern void set_intr_gate(unsigned int irq, void * addr);
+
+static inline void pack_descriptor(__u32 *a, __u32 *b,
+	unsigned long base, unsigned long limit, unsigned char type, unsigned char flags)
+{
+	*a = ((base & 0xffff) << 16) | (limit & 0xffff);
+	*b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
+	     ((type & 0xff) << 8) | ((flags & 0xf) << 12);
+}
+
+static inline void pack_gate(__u32 *a, __u32 *b,
+	unsigned long base, unsigned short seg, unsigned char type, unsigned char flags)
+{
+	*a = (seg << 16) | (base & 0xffff);
+	*b = (base & 0xffff0000) | ((type & 0xff) << 8) | (flags & 0xff);
+}
+
+#define DESCTYPE_LDT 	0x82	/* present, system, DPL-0, LDT */
+#define DESCTYPE_TSS 	0x89	/* present, system, DPL-0, 32-bit TSS */
+#define DESCTYPE_TASK	0x85	/* present, system, DPL-0, task gate */
+#define DESCTYPE_INT	0x8e	/* present, system, DPL-0, interrupt gate */
+#define DESCTYPE_TRAP	0x8f	/* present, system, DPL-0, trap gate */
+#define DESCTYPE_DPL3	0x60	/* DPL-3 */
+#define DESCTYPE_S	0x10	/* !system */
+
 #define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8))
 #define load_LDT_desc() __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8))
 
 #define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr))
 #define load_idt(dtr) __asm__ __volatile("lidt %0"::"m" (*dtr))
-#define load_tr(tr) __asm__ __volatile("ltr %0"::"mr" (tr))
-#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"mr" (ldt))
+#define load_tr(tr) __asm__ __volatile("ltr %0"::"m" (tr))
+#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"m" (ldt))
 
 #define store_gdt(dtr) __asm__ ("sgdt %0":"=m" (*dtr))
 #define store_idt(dtr) __asm__ ("sidt %0":"=m" (*dtr))
-#define store_tr(tr) __asm__ ("str %0":"=mr" (tr))
-#define store_ldt(ldt) __asm__ ("sldt %0":"=mr" (ldt))
+#define store_tr(tr) __asm__ ("str %0":"=m" (tr))
+#define store_ldt(ldt) __asm__ ("sldt %0":"=m" (ldt))
 
-/*
- * This is the ldt that every process will get unless we need
- * something other than this.
- */
-extern struct desc_struct default_ldt[];
-extern void set_intr_gate(unsigned int irq, void * addr);
+#if TLS_SIZE != 24
+# error update this code.
+#endif
 
-#define _set_tssldt_desc(n,addr,limit,type) \
-__asm__ __volatile__ ("movw %w3,0(%2)\n\t" \
-	"movw %w1,2(%2)\n\t" \
-	"rorl $16,%1\n\t" \
-	"movb %b1,4(%2)\n\t" \
-	"movb %4,5(%2)\n\t" \
-	"movb $0,6(%2)\n\t" \
-	"movb %h1,7(%2)\n\t" \
-	"rorl $16,%1" \
-	: "=m"(*(n)) : "q" (addr), "r"(n), "ir"(limit), "i"(type))
-
-static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, void *addr)
+static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
-	_set_tssldt_desc(&get_cpu_gdt_table(cpu)[entry], (int)addr,
-		offsetof(struct tss_struct, __cacheline_filler) - 1, 0x89);
+#define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]
+	C(0); C(1); C(2);
+#undef C
 }
 
-#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+static inline void write_dt_entry(void *dt, int entry, __u32 entry_a, __u32 entry_b)
+{
+	__u32 *lp = (__u32 *)((char *)dt + entry*8);
+	*lp = entry_a;
+	*(lp+1) = entry_b;
+}
+
+#define write_ldt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_gdt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+#define write_idt_entry(dt, entry, a, b) write_dt_entry(dt, entry, a, b)
+
+static inline void _set_gate(int gate, unsigned int type, void *addr, unsigned short seg)
+{
+	__u32 a, b;
+	pack_gate(&a, &b, (unsigned long)addr, seg, type, 0);
+	write_idt_entry(idt_table, gate, a, b);
+}
 
-static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
+static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, const void *addr)
 {
-	_set_tssldt_desc(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT], (int)addr, ((size << 3)-1), 0x82);
+	__u32 a, b;
+	pack_descriptor(&a, &b, (unsigned long)addr,
+			offsetof(struct tss_struct, __cacheline_filler) - 1,
+			DESCTYPE_TSS, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), entry, a, b);
 }
 
+static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int entries)
+{
+	__u32 a, b;
+	pack_descriptor(&a, &b, (unsigned long)addr,
+			entries * sizeof(struct desc_struct) - 1,
+			DESCTYPE_LDT, 0);
+	write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_LDT, a, b);
+}
+
+#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
+
 #define LDT_entry_a(info) \
 	((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
 
@@ -102,24 +151,6 @@ static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
 	(info)->seg_not_present	== 1	&& \
 	(info)->useable		== 0	)
 
-static inline void write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b)
-{
-	__u32 *lp = (__u32 *)((char *)ldt + entry*8);
-	*lp = entry_a;
-	*(lp+1) = entry_b;
-}
-
-#if TLS_SIZE != 24
-# error update this code.
-#endif
-
-static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
-{
-#define C(i) get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]
-	C(0); C(1); C(2);
-#undef C
-}
-
 static inline void clear_LDT(void)
 {
 	int cpu = get_cpu();
-- 
cgit v1.2.1


From ba4d40bb5c465f0a4dcc30d02dab80c2cb7e1ff3 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] Auto size the per cpu area.

Now for a completely different but trivial approach.
I just boot tested it with 255 CPUS and everything worked.

Currently everything (except module data) we place in
the per cpu area we know about at compile time.  So
instead of allocating a fixed size for the per_cpu area
allocate the number of bytes we need plus a fixed constant
for to be used for modules.

It isn't perfect but it is much less of a pain to
work with than what we are doing now.

AK: fixed warning

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c |  7 ++-----
 include/asm-x86_64/percpu.h  | 10 ++++++++++
 2 files changed, 12 insertions(+), 5 deletions(-)

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index b09e60fa96b4..e85cfbb49b63 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -95,12 +95,9 @@ void __init setup_per_cpu_areas(void)
 #endif
 
 	/* Copy section for each CPU (we discard the original) */
-	size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
-#ifdef CONFIG_MODULES
-	if (size < PERCPU_ENOUGH_ROOM)
-		size = PERCPU_ENOUGH_ROOM;
-#endif
+	size = PERCPU_ENOUGH_ROOM;
 
+	printk(KERN_INFO "PERCPU: Allocating %lu bytes of per cpu data\n", size);
 	for_each_cpu_mask (i, cpu_possible_map) {
 		char *ptr;
 
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 08dd9f9dda81..39d2bab9b520 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -11,6 +11,16 @@
 
 #include <asm/pda.h>
 
+#ifdef CONFIG_MODULES
+# define PERCPU_MODULE_RESERVE 8192
+#else
+# define PERCPU_MODULE_RESERVE 0
+#endif
+
+#define PERCPU_ENOUGH_ROOM \
+	(ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES) + \
+	 PERCPU_MODULE_RESERVE)
+
 #define __per_cpu_offset(cpu) (cpu_pda(cpu)->data_offset)
 #define __my_cpu_offset() read_pda(data_offset)
 
-- 
cgit v1.2.1


From aada06c9b7f4cdedbeb2bc905893cf1923a0abad Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: clean up topology.c

There is no need to duplicate the topology_init() function.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/topology.c | 21 +++------------------
 1 file changed, 3 insertions(+), 18 deletions(-)

diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c
index e2e281d4bcc8..07d6da36a825 100644
--- a/arch/i386/kernel/topology.c
+++ b/arch/i386/kernel/topology.c
@@ -28,6 +28,7 @@
 #include <linux/init.h>
 #include <linux/smp.h>
 #include <linux/nodemask.h>
+#include <linux/mmzone.h>
 #include <asm/cpu.h>
 
 static struct i386_cpu cpu_devices[NR_CPUS];
@@ -55,34 +56,18 @@ EXPORT_SYMBOL(arch_register_cpu);
 EXPORT_SYMBOL(arch_unregister_cpu);
 #endif /*CONFIG_HOTPLUG_CPU*/
 
-
-
-#ifdef CONFIG_NUMA
-#include <linux/mmzone.h>
-
 static int __init topology_init(void)
 {
 	int i;
 
+#ifdef CONFIG_NUMA
 	for_each_online_node(i)
 		register_one_node(i);
+#endif /* CONFIG_NUMA */
 
 	for_each_present_cpu(i)
 		arch_register_cpu(i);
 	return 0;
 }
 
-#else /* !CONFIG_NUMA */
-
-static int __init topology_init(void)
-{
-	int i;
-
-	for_each_present_cpu(i)
-		arch_register_cpu(i);
-	return 0;
-}
-
-#endif /* CONFIG_NUMA */
-
 subsys_initcall(topology_init);
-- 
cgit v1.2.1


From c0d83745cc67ed71a08c14739a0b286d0239b1e2 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: mark two more functions as __init

cyrix_identify() should be __init because transmeta_identify() is.
tsc_init() is only called from setup_arch() which is marked as __init.

These two section mismatches have been detected using running modpost on
a vmlinux image compiled with CONFIG_RELOCATABLE=y.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/tsc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/i386/kernel/tsc.c b/arch/i386/kernel/tsc.c
index 7e0d8dab2075..b8fa0a8b2e47 100644
--- a/arch/i386/kernel/tsc.c
+++ b/arch/i386/kernel/tsc.c
@@ -192,7 +192,7 @@ int recalibrate_cpu_khz(void)
 
 EXPORT_SYMBOL(recalibrate_cpu_khz);
 
-void tsc_init(void)
+void __init tsc_init(void)
 {
 	if (!cpu_has_tsc || tsc_disable)
 		return;
-- 
cgit v1.2.1


From 40bee2ee73c745922e9b2d5595c46f19d1cf1b6f Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] fix bus numbering format in mmconfig warning

Make an mmconfig warning print the bus id with a regular format.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/x86_64/pci/mmconfig.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index 3c55c76c6fd5..dd78f4d18df1 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -156,9 +156,8 @@ static __init void unreachable_devices(void)
 			addr = pci_dev_base(0, k, PCI_DEVFN(i, 0));
 			if (addr == NULL|| readl(addr) != val1) {
 				set_bit(i + 32*k, fallback_slots);
-				printk(KERN_NOTICE
-				"PCI: No mmconfig possible on device %x:%x\n",
-					k, i);
+				printk(KERN_NOTICE "PCI: No mmconfig possible"
+				       " on device %02x:%02x\n", k, i);
 			}
 		}
 	}
-- 
cgit v1.2.1


From 73fea175303926055440c06bc8894f0c5c58afc8 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Tue, 26 Sep 2006 10:52:35 +0200
Subject: [PATCH] i386: Support physical cpu hotplug for x86_64

This patch enables ACPI based physical CPU hotplug support for x86_64.
Implements acpi_map_lsapic() and acpi_unmap_lsapic() to support physical cpu
hotplug.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/acpi/boot.c | 69 +++++++++++++++++++++++++++++++++++++++++---
 arch/i386/kernel/mpparse.c   |  2 +-
 include/asm-i386/smp.h       |  1 +
 3 files changed, 67 insertions(+), 5 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 0fba420d668a..8a6c5b412348 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -26,6 +26,7 @@
 #include <linux/init.h>
 #include <linux/acpi.h>
 #include <linux/efi.h>
+#include <linux/cpumask.h>
 #include <linux/module.h>
 #include <linux/dmi.h>
 #include <linux/irq.h>
@@ -512,16 +513,76 @@ EXPORT_SYMBOL(acpi_register_gsi);
 #ifdef CONFIG_ACPI_HOTPLUG_CPU
 int acpi_map_lsapic(acpi_handle handle, int *pcpu)
 {
-	/* TBD */
-	return -EINVAL;
+	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+	union acpi_object *obj;
+	struct acpi_table_lapic *lapic;
+	cpumask_t tmp_map, new_map;
+	u8 physid;
+	int cpu;
+
+	if (ACPI_FAILURE(acpi_evaluate_object(handle, "_MAT", NULL, &buffer)))
+		return -EINVAL;
+
+	if (!buffer.length || !buffer.pointer)
+		return -EINVAL;
+
+	obj = buffer.pointer;
+	if (obj->type != ACPI_TYPE_BUFFER ||
+	    obj->buffer.length < sizeof(*lapic)) {
+		kfree(buffer.pointer);
+		return -EINVAL;
+	}
+
+	lapic = (struct acpi_table_lapic *)obj->buffer.pointer;
+
+	if ((lapic->header.type != ACPI_MADT_LAPIC) ||
+	    (!lapic->flags.enabled)) {
+		kfree(buffer.pointer);
+		return -EINVAL;
+	}
+
+	physid = lapic->id;
+
+	kfree(buffer.pointer);
+	buffer.length = ACPI_ALLOCATE_BUFFER;
+	buffer.pointer = NULL;
+
+	tmp_map = cpu_present_map;
+	mp_register_lapic(physid, lapic->flags.enabled);
+
+	/*
+	 * If mp_register_lapic successfully generates a new logical cpu
+	 * number, then the following will get us exactly what was mapped
+	 */
+	cpus_andnot(new_map, cpu_present_map, tmp_map);
+	if (cpus_empty(new_map)) {
+		printk ("Unable to map lapic to logical cpu number\n");
+		return -EINVAL;
+	}
+
+	cpu = first_cpu(new_map);
+
+	*pcpu = cpu;
+	return 0;
 }
 
 EXPORT_SYMBOL(acpi_map_lsapic);
 
 int acpi_unmap_lsapic(int cpu)
 {
-	/* TBD */
-	return -EINVAL;
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (x86_acpiid_to_apicid[i] == x86_cpu_to_apicid[cpu]) {
+			x86_acpiid_to_apicid[i] = -1;
+			break;
+		}
+	}
+	x86_cpu_to_apicid[cpu] = -1;
+	cpu_clear(cpu, cpu_present_map);
+	num_processors--;
+
+	return (0);
 }
 
 EXPORT_SYMBOL(acpi_unmap_lsapic);
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 772a4233bfe6..442aaf8c77eb 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -69,7 +69,7 @@ unsigned int def_to_bigsmp = 0;
 /* Processor that is doing the boot up */
 unsigned int boot_cpu_physical_apicid = -1U;
 /* Internal processor count */
-static unsigned int __devinitdata num_processors;
+unsigned int __cpuinitdata num_processors;
 
 /* Bitmask of physically existing CPUs */
 physid_mask_t phys_cpu_present_map;
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index f87826039a51..32ac8c91d5c5 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -84,6 +84,7 @@ static inline int hard_smp_processor_id(void)
 
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
+extern unsigned int num_processors;
 
 #endif /* !__ASSEMBLY__ */
 
-- 
cgit v1.2.1


From e07e23e1fd3000289fc7ccc6c71879070d3b19e0 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] non lazy "sleazy" fpu implementation

Right now the kernel on x86-64 has a 100% lazy fpu behavior: after *every*
context switch a trap is taken for the first FPU use to restore the FPU
context lazily.  This is of course great for applications that have very
sporadic or no FPU use (since then you avoid doing the expensive
save/restore all the time).  However for very frequent FPU users...  you
take an extra trap every context switch.

The patch below adds a simple heuristic to this code: After 5 consecutive
context switches of FPU use, the lazy behavior is disabled and the context
gets restored every context switch.  If the app indeed uses the FPU, the
trap is avoided.  (the chance of the 6th time slice using FPU after the
previous 5 having done so are quite high obviously).

After 256 switches, this is reset and lazy behavior is returned (until
there are 5 consecutive ones again).  The reason for this is to give apps
that do longer bursts of FPU use still the lazy behavior back after some
time.

[akpm@osdl.org: place new task_struct field next to jit_keyring to save space]
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/x86_64/kernel/process.c | 10 ++++++++++
 arch/x86_64/kernel/traps.c   |  1 +
 include/asm-x86_64/i387.h    |  5 ++++-
 include/linux/sched.h        |  9 +++++++++
 4 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 6fbd19564e4e..9e9a70e50c72 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -552,6 +552,10 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	int cpu = smp_processor_id();  
 	struct tss_struct *tss = &per_cpu(init_tss, cpu);
 
+	/* we're going to use this soon, after a few expensive things */
+	if (next_p->fpu_counter>5)
+		prefetch(&next->i387.fxsave);
+
 	/*
 	 * Reload esp0, LDT and the page table pointer:
 	 */
@@ -629,6 +633,12 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	    || test_tsk_thread_flag(prev_p, TIF_IO_BITMAP))
 		__switch_to_xtra(prev_p, next_p, tss);
 
+	/* If the task has used fpu the last 5 timeslices, just do a full
+	 * restore of the math state immediately to avoid the trap; the
+	 * chances of needing FPU soon are obviously high now
+	 */
+	if (next_p->fpu_counter>5)
+		math_state_restore();
 	return prev_p;
 }
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 28e53342f294..ffc40cff1e07 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -1136,6 +1136,7 @@ asmlinkage void math_state_restore(void)
 		init_fpu(me);
 	restore_fpu_checking(&me->thread.i387.fxsave);
 	task_thread_info(me)->status |= TS_USEDFPU;
+	me->fpu_counter++;
 }
 
 void __init trap_init(void)
diff --git a/include/asm-x86_64/i387.h b/include/asm-x86_64/i387.h
index cba8a3b0cded..60c0f4853fdb 100644
--- a/include/asm-x86_64/i387.h
+++ b/include/asm-x86_64/i387.h
@@ -24,6 +24,7 @@ extern unsigned int mxcsr_feature_mask;
 extern void mxcsr_feature_mask_init(void);
 extern void init_fpu(struct task_struct *child);
 extern int save_i387(struct _fpstate __user *buf);
+extern asmlinkage void math_state_restore(void);
 
 /*
  * FPU lazy state save handling...
@@ -31,7 +32,9 @@ extern int save_i387(struct _fpstate __user *buf);
 
 #define unlazy_fpu(tsk) do { \
 	if (task_thread_info(tsk)->status & TS_USEDFPU) \
-		save_init_fpu(tsk); \
+		save_init_fpu(tsk); 			\
+	else						\
+		tsk->fpu_counter = 0;			\
 } while (0)
 
 /* Ignore delayed exceptions from user space */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 34ed0d99b1bd..807556c5bcd2 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -865,6 +865,15 @@ struct task_struct {
 	struct key *thread_keyring;	/* keyring private to this thread */
 	unsigned char jit_keyring;	/* default keyring to attach requested keys to */
 #endif
+	/*
+	 * fpu_counter contains the number of consecutive context switches
+	 * that the FPU is used. If this is over a threshold, the lazy fpu
+	 * saving becomes unlazy to save the trap. This is an unsigned char
+	 * so that after 256 times the counter wraps and the behavior turns
+	 * lazy again; this to deal with bursty apps that only use FPU for
+	 * a short time
+	 */
+	unsigned char fpu_counter;
 	int oomkilladj; /* OOM kill score adjustment (bit shift). */
 	char comm[TASK_COMM_LEN]; /* executable name excluding path
 				     - access with [gs]et_task_comm (which lock
-- 
cgit v1.2.1


From abf0f10948b316b577851ef21c728341f1046552 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] wire up oops_enter()/oops_exit()

Implement pause_on_oops() on x86_64.

AK: I redid the patch to do the oops_enter/exit in the existing
oops_begin()/end(). This makes it much shorter.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/traps.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index ffc40cff1e07..34660b1e2720 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -561,6 +561,8 @@ unsigned __kprobes long oops_begin(void)
 	int cpu = safe_smp_processor_id();
 	unsigned long flags;
 
+	oops_enter();
+
 	/* racy, but better than risking deadlock. */
 	local_irq_save(flags);
 	if (!spin_trylock(&die_lock)) { 
@@ -589,6 +591,7 @@ void __kprobes oops_end(unsigned long flags)
 		spin_unlock_irqrestore(&die_lock, flags);
 	if (panic_on_oops)
 		panic("Fatal exception");
+	oops_exit();
 }
 
 void __kprobes __die(const char * str, struct pt_regs * regs, long err)
-- 
cgit v1.2.1


From 6ad916581181a105d7832a7dec9e1eb58f7a1621 Mon Sep 17 00:00:00 2001
From: Keith Mannthey <kmannth@us.ibm.com>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] x86_64 kernel mapping fix

Fix for the x86_64 kernel mapping code.  Without this patch the update path
only inits one pmd_page worth of memory and tramples any entries on it.  now
the calling convention to phys_pmd_init and phys_init is to always pass a
[pmd/pud] page not an offset within a page.

Signed-off-by: Keith Mannthey<kmannth@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/x86_64/mm/init.c | 51 ++++++++++++++++++++++++++-------------------------
 1 file changed, 26 insertions(+), 25 deletions(-)

diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index d40134bd6399..984155b75e4c 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -250,12 +250,13 @@ __init void early_iounmap(void *addr, unsigned long size)
 }
 
 static void __meminit
-phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
+phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
 {
-	int i;
+	int i = pmd_index(address);
 
-	for (i = 0; i < PTRS_PER_PMD; pmd++, i++, address += PMD_SIZE) {
+	for (; i < PTRS_PER_PMD; i++, address += PMD_SIZE) {
 		unsigned long entry;
+		pmd_t *pmd = pmd_page + pmd_index(address);
 
 		if (address >= end) {
 			if (!after_bootmem)
@@ -263,6 +264,10 @@ phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
 					set_pmd(pmd, __pmd(0));
 			break;
 		}
+
+		if (pmd_val(*pmd))
+			continue;
+
 		entry = _PAGE_NX|_PAGE_PSE|_KERNPG_TABLE|_PAGE_GLOBAL|address;
 		entry &= __supported_pte_mask;
 		set_pmd(pmd, __pmd(entry));
@@ -272,45 +277,41 @@ phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
 static void __meminit
 phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
 {
-	pmd_t *pmd = pmd_offset(pud, (unsigned long)__va(address));
-
-	if (pmd_none(*pmd)) {
-		spin_lock(&init_mm.page_table_lock);
-		phys_pmd_init(pmd, address, end);
-		spin_unlock(&init_mm.page_table_lock);
-		__flush_tlb_all();
-	}
+	pmd_t *pmd = pmd_offset(pud,0);
+	spin_lock(&init_mm.page_table_lock);
+	phys_pmd_init(pmd, address, end);
+	spin_unlock(&init_mm.page_table_lock);
+	__flush_tlb_all();
 }
 
-static void __meminit phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
+static void __meminit phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
 { 
-	long i = pud_index(address);
-
-	pud = pud + i;
+	int i = pud_index(addr);
 
-	if (after_bootmem && pud_val(*pud)) {
-		phys_pmd_update(pud, address, end);
-		return;
-	}
 
-	for (; i < PTRS_PER_PUD; pud++, i++) {
+	for (; i < PTRS_PER_PUD; i++, addr = (addr & PUD_MASK) + PUD_SIZE ) {
 		int map; 
-		unsigned long paddr, pmd_phys;
+		unsigned long pmd_phys;
+		pud_t *pud = pud_page + pud_index(addr);
 		pmd_t *pmd;
 
-		paddr = (address & PGDIR_MASK) + i*PUD_SIZE;
-		if (paddr >= end)
+		if (addr >= end)
 			break;
 
-		if (!after_bootmem && !e820_any_mapped(paddr, paddr+PUD_SIZE, 0)) {
+		if (!after_bootmem && !e820_any_mapped(addr,addr+PUD_SIZE,0)) {
 			set_pud(pud, __pud(0)); 
 			continue;
 		} 
 
+		if (pud_val(*pud)) {
+			phys_pmd_update(pud, addr, end);
+			continue;
+		}
+
 		pmd = alloc_low_page(&map, &pmd_phys);
 		spin_lock(&init_mm.page_table_lock);
 		set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
-		phys_pmd_init(pmd, paddr, end);
+		phys_pmd_init(pmd, addr, end);
 		spin_unlock(&init_mm.page_table_lock);
 		unmap_low_page(map);
 	}
-- 
cgit v1.2.1


From 68bbc172cd1b0ee01814304b8a7bef8922d5fdca Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: remove redundant generic_identify() calls when
 identifying cpus

cpu_dev->c_identify is only called from arch/i386/common.c:identify_cpu(), and
this after generic_identify() already has been called. There is no need to call
this function twice and hook it in c_identify - but I may be wrong, please
double check before applying.

This patch also removes generic_identify() from cpu.h to avoid unnecessary
future nesting.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/amd.c       | 1 -
 arch/i386/kernel/cpu/common.c    | 2 +-
 arch/i386/kernel/cpu/cpu.h       | 2 --
 arch/i386/kernel/cpu/cyrix.c     | 2 --
 arch/i386/kernel/cpu/intel.c     | 1 -
 arch/i386/kernel/cpu/nexgen.c    | 1 -
 arch/i386/kernel/cpu/transmeta.c | 1 -
 7 files changed, 1 insertion(+), 9 deletions(-)

diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c
index e6a2d6b80cda..d6e7778bac70 100644
--- a/arch/i386/kernel/cpu/amd.c
+++ b/arch/i386/kernel/cpu/amd.c
@@ -275,7 +275,6 @@ static struct cpu_dev amd_cpu_dev __initdata = {
 		},
 	},
 	.c_init		= init_amd,
-	.c_identify	= generic_identify,
 	.c_size_cache	= amd_size_cache,
 };
 
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 730a7ab75009..24ac51a02378 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -265,7 +265,7 @@ static void __init early_cpu_detect(void)
 	}
 }
 
-void __cpuinit generic_identify(struct cpuinfo_x86 * c)
+static void __cpuinit generic_identify(struct cpuinfo_x86 * c)
 {
 	u32 tfms, xlvl;
 	int ebx;
diff --git a/arch/i386/kernel/cpu/cpu.h b/arch/i386/kernel/cpu/cpu.h
index 5a1d4f163e84..2f6432cef6ff 100644
--- a/arch/i386/kernel/cpu/cpu.h
+++ b/arch/i386/kernel/cpu/cpu.h
@@ -24,7 +24,5 @@ extern struct cpu_dev * cpu_devs [X86_VENDOR_NUM];
 extern int get_model_name(struct cpuinfo_x86 *c);
 extern void display_cacheinfo(struct cpuinfo_x86 *c);
 
-extern void generic_identify(struct cpuinfo_x86 * c);
-
 extern void early_intel_workaround(struct cpuinfo_x86 *c);
 
diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c
index f03b7f94c304..bb02ed1fe560 100644
--- a/arch/i386/kernel/cpu/cyrix.c
+++ b/arch/i386/kernel/cpu/cyrix.c
@@ -427,7 +427,6 @@ static void cyrix_identify(struct cpuinfo_x86 * c)
 			local_irq_restore(flags);
 		}
 	}
-	generic_identify(c);
 }
 
 static struct cpu_dev cyrix_cpu_dev __initdata = {
@@ -457,7 +456,6 @@ static struct cpu_dev nsc_cpu_dev __initdata = {
 	.c_vendor	= "NSC",
 	.c_ident 	= { "Geode by NSC" },
 	.c_init		= init_nsc,
-	.c_identify	= generic_identify,
 };
 
 int __init nsc_init_cpu(void)
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 5a2e270924b1..26243e019d11 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -263,7 +263,6 @@ static struct cpu_dev intel_cpu_dev __cpuinitdata = {
 		},
 	},
 	.c_init		= init_intel,
-	.c_identify	= generic_identify,
 	.c_size_cache	= intel_size_cache,
 };
 
diff --git a/arch/i386/kernel/cpu/nexgen.c b/arch/i386/kernel/cpu/nexgen.c
index ad87fa58058d..df188bfd759b 100644
--- a/arch/i386/kernel/cpu/nexgen.c
+++ b/arch/i386/kernel/cpu/nexgen.c
@@ -38,7 +38,6 @@ static void __init nexgen_identify(struct cpuinfo_x86 * c)
 	if ( deep_magic_nexgen_probe() ) {
 		strcpy(c->x86_vendor_id, "NexGenDriven");
 	}
-	generic_identify(c);
 }
 
 static struct cpu_dev nexgen_cpu_dev __initdata = {
diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c
index 7214c9b577ab..4027c369bc90 100644
--- a/arch/i386/kernel/cpu/transmeta.c
+++ b/arch/i386/kernel/cpu/transmeta.c
@@ -88,7 +88,6 @@ static void __init init_transmeta(struct cpuinfo_x86 *c)
 static void __init transmeta_identify(struct cpuinfo_x86 * c)
 {
 	u32 xlvl;
-	generic_identify(c);
 
 	/* Transmeta-defined flags: level 0x80860001 */
 	xlvl = cpuid_eax(0x80860000);
-- 
cgit v1.2.1


From ed77504b2007ff7ce56841227467ac3ead52df62 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] mark init_amd() as __cpuinit

The init_amd() function is only called from identify_cpu() which is already
marked as __cpuinit. So let's mark it as __cpuinit.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 4be29832e3dc..e328e3eb8cbe 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -672,7 +672,7 @@ static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
 #endif
 }
 
-static void __init init_amd(struct cpuinfo_x86 *c)
+static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 {
 	unsigned level;
 
-- 
cgit v1.2.1


From 95414930548871c6c92a5b0e607b12b81f3d84d8 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: mark cpu_dev structures as __cpuinitdata

The different cpu_dev structures are all used from __cpuinit callers what
I can tell. So mark them as __cpuinitdata instead of __initdata. I am a
little bit unsure about arch/i386/common.c:default_cpu, especially when it
comes to the purpose of this_cpu.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/amd.c       | 2 +-
 arch/i386/kernel/cpu/centaur.c   | 2 +-
 arch/i386/kernel/cpu/common.c    | 2 +-
 arch/i386/kernel/cpu/cyrix.c     | 4 ++--
 arch/i386/kernel/cpu/nexgen.c    | 2 +-
 arch/i386/kernel/cpu/rise.c      | 2 +-
 arch/i386/kernel/cpu/transmeta.c | 2 +-
 arch/i386/kernel/cpu/umc.c       | 2 +-
 8 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c
index d6e7778bac70..eb134a20a7b8 100644
--- a/arch/i386/kernel/cpu/amd.c
+++ b/arch/i386/kernel/cpu/amd.c
@@ -259,7 +259,7 @@ static unsigned int amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 	return size;
 }
 
-static struct cpu_dev amd_cpu_dev __initdata = {
+static struct cpu_dev amd_cpu_dev __cpuinitdata = {
 	.c_vendor	= "AMD",
 	.c_ident 	= { "AuthenticAMD" },
 	.c_models = {
diff --git a/arch/i386/kernel/cpu/centaur.c b/arch/i386/kernel/cpu/centaur.c
index bd75629dd262..b4d70dcea9f8 100644
--- a/arch/i386/kernel/cpu/centaur.c
+++ b/arch/i386/kernel/cpu/centaur.c
@@ -457,7 +457,7 @@ static unsigned int centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size
 	return size;
 }
 
-static struct cpu_dev centaur_cpu_dev __initdata = {
+static struct cpu_dev centaur_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Centaur",
 	.c_ident	= { "CentaurHauls" },
 	.c_init		= init_centaur,
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 24ac51a02378..99144449b9a1 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -49,7 +49,7 @@ static void default_init(struct cpuinfo_x86 * c)
 	}
 }
 
-static struct cpu_dev default_cpu = {
+static struct cpu_dev __cpuinitdata default_cpu = {
 	.c_init	= default_init,
 	.c_vendor = "Unknown",
 };
diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c
index bb02ed1fe560..c2a0da9e0b55 100644
--- a/arch/i386/kernel/cpu/cyrix.c
+++ b/arch/i386/kernel/cpu/cyrix.c
@@ -429,7 +429,7 @@ static void cyrix_identify(struct cpuinfo_x86 * c)
 	}
 }
 
-static struct cpu_dev cyrix_cpu_dev __initdata = {
+static struct cpu_dev cyrix_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Cyrix",
 	.c_ident 	= { "CyrixInstead" },
 	.c_init		= init_cyrix,
@@ -452,7 +452,7 @@ static int __init cyrix_exit_cpu(void)
 
 late_initcall(cyrix_exit_cpu);
 
-static struct cpu_dev nsc_cpu_dev __initdata = {
+static struct cpu_dev nsc_cpu_dev __cpuinitdata = {
 	.c_vendor	= "NSC",
 	.c_ident 	= { "Geode by NSC" },
 	.c_init		= init_nsc,
diff --git a/arch/i386/kernel/cpu/nexgen.c b/arch/i386/kernel/cpu/nexgen.c
index df188bfd759b..9d622598c055 100644
--- a/arch/i386/kernel/cpu/nexgen.c
+++ b/arch/i386/kernel/cpu/nexgen.c
@@ -40,7 +40,7 @@ static void __init nexgen_identify(struct cpuinfo_x86 * c)
 	}
 }
 
-static struct cpu_dev nexgen_cpu_dev __initdata = {
+static struct cpu_dev nexgen_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Nexgen",
 	.c_ident	= { "NexGenDriven" },
 	.c_models = {
diff --git a/arch/i386/kernel/cpu/rise.c b/arch/i386/kernel/cpu/rise.c
index d08d5a2811c8..b597e435f6c1 100644
--- a/arch/i386/kernel/cpu/rise.c
+++ b/arch/i386/kernel/cpu/rise.c
@@ -28,7 +28,7 @@ static void __init init_rise(struct cpuinfo_x86 *c)
 	set_bit(X86_FEATURE_CX8, c->x86_capability);
 }
 
-static struct cpu_dev rise_cpu_dev __initdata = {
+static struct cpu_dev rise_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Rise",
 	.c_ident	= { "RiseRiseRise" },
 	.c_models = {
diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c
index 4027c369bc90..0b6e8392ed33 100644
--- a/arch/i386/kernel/cpu/transmeta.c
+++ b/arch/i386/kernel/cpu/transmeta.c
@@ -97,7 +97,7 @@ static void __init transmeta_identify(struct cpuinfo_x86 * c)
 	}
 }
 
-static struct cpu_dev transmeta_cpu_dev __initdata = {
+static struct cpu_dev transmeta_cpu_dev __cpuinitdata = {
 	.c_vendor	= "Transmeta",
 	.c_ident	= { "GenuineTMx86", "TransmetaCPU" },
 	.c_init		= init_transmeta,
diff --git a/arch/i386/kernel/cpu/umc.c b/arch/i386/kernel/cpu/umc.c
index 2cd988f6dc55..392b195948d6 100644
--- a/arch/i386/kernel/cpu/umc.c
+++ b/arch/i386/kernel/cpu/umc.c
@@ -10,7 +10,7 @@ static void __init init_umc(struct cpuinfo_x86 * c)
 
 }
 
-static struct cpu_dev umc_cpu_dev __initdata = {
+static struct cpu_dev umc_cpu_dev __cpuinitdata = {
 	.c_vendor	= "UMC",
 	.c_ident 	= { "UMC UMC UMC" },
 	.c_models = {
-- 
cgit v1.2.1


From b4af3f7cf11e6b5904af08a652d4a2429af17c74 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: mark cpu init functions as __cpuinit, data as
 __cpuinitdata

Mark i386-specific cpu init functions as __cpuinit. They are all
only called from arch/i386/common.c:identify_cpu() that already is marked as
__cpuinit. This patch also removes the empty function init_umc().

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/amd.c       |  2 +-
 arch/i386/kernel/cpu/centaur.c   | 20 ++++++++++----------
 arch/i386/kernel/cpu/common.c    |  2 +-
 arch/i386/kernel/cpu/cyrix.c     | 32 ++++++++++++++++----------------
 arch/i386/kernel/cpu/nexgen.c    |  2 +-
 arch/i386/kernel/cpu/rise.c      |  2 +-
 arch/i386/kernel/cpu/transmeta.c |  2 +-
 arch/i386/kernel/cpu/umc.c       |  5 -----
 8 files changed, 31 insertions(+), 36 deletions(-)

diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c
index eb134a20a7b8..856cd08f2f3d 100644
--- a/arch/i386/kernel/cpu/amd.c
+++ b/arch/i386/kernel/cpu/amd.c
@@ -22,7 +22,7 @@
 extern void vide(void);
 __asm__(".align 4\nvide: ret");
 
-static void __init init_amd(struct cpuinfo_x86 *c)
+static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 {
 	u32 l, h;
 	int mbytes = num_physpages >> (20-PAGE_SHIFT);
diff --git a/arch/i386/kernel/cpu/centaur.c b/arch/i386/kernel/cpu/centaur.c
index b4d70dcea9f8..34a81ede7892 100644
--- a/arch/i386/kernel/cpu/centaur.c
+++ b/arch/i386/kernel/cpu/centaur.c
@@ -9,7 +9,7 @@
 
 #ifdef CONFIG_X86_OOSTORE
 
-static u32 __init power2(u32 x)
+static u32 __cpuinit power2(u32 x)
 {
 	u32 s=1;
 	while(s<=x)
@@ -22,7 +22,7 @@ static u32 __init power2(u32 x)
  *	Set up an actual MCR
  */
  
-static void __init centaur_mcr_insert(int reg, u32 base, u32 size, int key)
+static void __cpuinit centaur_mcr_insert(int reg, u32 base, u32 size, int key)
 {
 	u32 lo, hi;
 	
@@ -40,7 +40,7 @@ static void __init centaur_mcr_insert(int reg, u32 base, u32 size, int key)
  *	Shortcut: We know you can't put 4Gig of RAM on a winchip
  */
 
-static u32 __init ramtop(void)		/* 16388 */
+static u32 __cpuinit ramtop(void)		/* 16388 */
 {
 	int i;
 	u32 top = 0;
@@ -91,7 +91,7 @@ static u32 __init ramtop(void)		/* 16388 */
  *	Compute a set of MCR's to give maximum coverage
  */
 
-static int __init centaur_mcr_compute(int nr, int key)
+static int __cpuinit centaur_mcr_compute(int nr, int key)
 {
 	u32 mem = ramtop();
 	u32 root = power2(mem);
@@ -166,7 +166,7 @@ static int __init centaur_mcr_compute(int nr, int key)
 	return ct;
 }
 
-static void __init centaur_create_optimal_mcr(void)
+static void __cpuinit centaur_create_optimal_mcr(void)
 {
 	int i;
 	/*
@@ -189,7 +189,7 @@ static void __init centaur_create_optimal_mcr(void)
 		wrmsr(MSR_IDT_MCR0+i, 0, 0);
 }
 
-static void __init winchip2_create_optimal_mcr(void)
+static void __cpuinit winchip2_create_optimal_mcr(void)
 {
 	u32 lo, hi;
 	int i;
@@ -227,7 +227,7 @@ static void __init winchip2_create_optimal_mcr(void)
  *	Handle the MCR key on the Winchip 2.
  */
 
-static void __init winchip2_unprotect_mcr(void)
+static void __cpuinit winchip2_unprotect_mcr(void)
 {
 	u32 lo, hi;
 	u32 key;
@@ -239,7 +239,7 @@ static void __init winchip2_unprotect_mcr(void)
 	wrmsr(MSR_IDT_MCR_CTRL, lo, hi);
 }
 
-static void __init winchip2_protect_mcr(void)
+static void __cpuinit winchip2_protect_mcr(void)
 {
 	u32 lo, hi;
 	
@@ -257,7 +257,7 @@ static void __init winchip2_protect_mcr(void)
 #define RNG_ENABLED	(1 << 3)
 #define RNG_ENABLE	(1 << 6)	/* MSR_VIA_RNG */
 
-static void __init init_c3(struct cpuinfo_x86 *c)
+static void __cpuinit init_c3(struct cpuinfo_x86 *c)
 {
 	u32  lo, hi;
 
@@ -303,7 +303,7 @@ static void __init init_c3(struct cpuinfo_x86 *c)
 	display_cacheinfo(c);
 }
 
-static void __init init_centaur(struct cpuinfo_x86 *c)
+static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 {
 	enum {
 		ECX8=1<<1,
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 99144449b9a1..2799baaadf45 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -36,7 +36,7 @@ struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
 
 extern int disable_pse;
 
-static void default_init(struct cpuinfo_x86 * c)
+static void __cpuinit default_init(struct cpuinfo_x86 * c)
 {
 	/* Not much we can do here... */
 	/* Check if at least it has cpuid */
diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c
index c2a0da9e0b55..d34b1bdb15fd 100644
--- a/arch/i386/kernel/cpu/cyrix.c
+++ b/arch/i386/kernel/cpu/cyrix.c
@@ -52,25 +52,25 @@ static void __init do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
  * Actually since bugs.h doesn't even reference this perhaps someone should
  * fix the documentation ???
  */
-static unsigned char Cx86_dir0_msb __initdata = 0;
+static unsigned char Cx86_dir0_msb __cpuinitdata = 0;
 
-static char Cx86_model[][9] __initdata = {
+static char Cx86_model[][9] __cpuinitdata = {
 	"Cx486", "Cx486", "5x86 ", "6x86", "MediaGX ", "6x86MX ",
 	"M II ", "Unknown"
 };
-static char Cx486_name[][5] __initdata = {
+static char Cx486_name[][5] __cpuinitdata = {
 	"SLC", "DLC", "SLC2", "DLC2", "SRx", "DRx",
 	"SRx2", "DRx2"
 };
-static char Cx486S_name[][4] __initdata = {
+static char Cx486S_name[][4] __cpuinitdata = {
 	"S", "S2", "Se", "S2e"
 };
-static char Cx486D_name[][4] __initdata = {
+static char Cx486D_name[][4] __cpuinitdata = {
 	"DX", "DX2", "?", "?", "?", "DX4"
 };
-static char Cx86_cb[] __initdata = "?.5x Core/Bus Clock";
-static char cyrix_model_mult1[] __initdata = "12??43";
-static char cyrix_model_mult2[] __initdata = "12233445";
+static char Cx86_cb[] __cpuinitdata = "?.5x Core/Bus Clock";
+static char cyrix_model_mult1[] __cpuinitdata = "12??43";
+static char cyrix_model_mult2[] __cpuinitdata = "12233445";
 
 /*
  * Reset the slow-loop (SLOP) bit on the 686(L) which is set by some old
@@ -82,7 +82,7 @@ static char cyrix_model_mult2[] __initdata = "12233445";
 
 extern void calibrate_delay(void) __init;
 
-static void __init check_cx686_slop(struct cpuinfo_x86 *c)
+static void __cpuinit check_cx686_slop(struct cpuinfo_x86 *c)
 {
 	unsigned long flags;
 	
@@ -107,7 +107,7 @@ static void __init check_cx686_slop(struct cpuinfo_x86 *c)
 }
 
 
-static void __init set_cx86_reorder(void)
+static void __cpuinit set_cx86_reorder(void)
 {
 	u8 ccr3;
 
@@ -122,7 +122,7 @@ static void __init set_cx86_reorder(void)
 	setCx86(CX86_CCR3, ccr3);
 }
 
-static void __init set_cx86_memwb(void)
+static void __cpuinit set_cx86_memwb(void)
 {
 	u32 cr0;
 
@@ -137,7 +137,7 @@ static void __init set_cx86_memwb(void)
 	setCx86(CX86_CCR2, getCx86(CX86_CCR2) | 0x14 );
 }
 
-static void __init set_cx86_inc(void)
+static void __cpuinit set_cx86_inc(void)
 {
 	unsigned char ccr3;
 
@@ -158,7 +158,7 @@ static void __init set_cx86_inc(void)
  *	Configure later MediaGX and/or Geode processor.
  */
 
-static void __init geode_configure(void)
+static void __cpuinit geode_configure(void)
 {
 	unsigned long flags;
 	u8 ccr3, ccr4;
@@ -184,14 +184,14 @@ static void __init geode_configure(void)
 
 
 #ifdef CONFIG_PCI
-static struct pci_device_id __initdata cyrix_55x0[] = {
+static struct pci_device_id __cpuinitdata cyrix_55x0[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5510) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5520) },
 	{ },
 };
 #endif
 
-static void __init init_cyrix(struct cpuinfo_x86 *c)
+static void __cpuinit init_cyrix(struct cpuinfo_x86 *c)
 {
 	unsigned char dir0, dir0_msn, dir0_lsn, dir1 = 0;
 	char *buf = c->x86_model_id;
@@ -346,7 +346,7 @@ static void __init init_cyrix(struct cpuinfo_x86 *c)
 /*
  * Handle National Semiconductor branded processors
  */
-static void __init init_nsc(struct cpuinfo_x86 *c)
+static void __cpuinit init_nsc(struct cpuinfo_x86 *c)
 {
 	/* There may be GX1 processors in the wild that are branded
 	 * NSC and not Cyrix.
diff --git a/arch/i386/kernel/cpu/nexgen.c b/arch/i386/kernel/cpu/nexgen.c
index 9d622598c055..d8cc88224d17 100644
--- a/arch/i386/kernel/cpu/nexgen.c
+++ b/arch/i386/kernel/cpu/nexgen.c
@@ -27,7 +27,7 @@ static int __init deep_magic_nexgen_probe(void)
 	return  ret;
 }
 
-static void __init init_nexgen(struct cpuinfo_x86 * c)
+static void __cpuinit init_nexgen(struct cpuinfo_x86 * c)
 {
 	c->x86_cache_size = 256; /* A few had 1 MB... */
 }
diff --git a/arch/i386/kernel/cpu/rise.c b/arch/i386/kernel/cpu/rise.c
index b597e435f6c1..9317f7414989 100644
--- a/arch/i386/kernel/cpu/rise.c
+++ b/arch/i386/kernel/cpu/rise.c
@@ -5,7 +5,7 @@
 
 #include "cpu.h"
 
-static void __init init_rise(struct cpuinfo_x86 *c)
+static void __cpuinit init_rise(struct cpuinfo_x86 *c)
 {
 	printk("CPU: Rise iDragon");
 	if (c->x86_model > 2)
diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c
index 0b6e8392ed33..725404cd77f7 100644
--- a/arch/i386/kernel/cpu/transmeta.c
+++ b/arch/i386/kernel/cpu/transmeta.c
@@ -5,7 +5,7 @@
 #include <asm/msr.h>
 #include "cpu.h"
 
-static void __init init_transmeta(struct cpuinfo_x86 *c)
+static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 {
 	unsigned int cap_mask, uk, max, dummy;
 	unsigned int cms_rev1, cms_rev2;
diff --git a/arch/i386/kernel/cpu/umc.c b/arch/i386/kernel/cpu/umc.c
index 392b195948d6..1bf3f87e9c5b 100644
--- a/arch/i386/kernel/cpu/umc.c
+++ b/arch/i386/kernel/cpu/umc.c
@@ -5,10 +5,6 @@
 
 /* UMC chips appear to be only either 386 or 486, so no special init takes place.
  */
-static void __init init_umc(struct cpuinfo_x86 * c)
-{
-
-}
 
 static struct cpu_dev umc_cpu_dev __cpuinitdata = {
 	.c_vendor	= "UMC",
@@ -21,7 +17,6 @@ static struct cpu_dev umc_cpu_dev __cpuinitdata = {
 		  }
 		},
 	},
-	.c_init		= init_umc,
 };
 
 int __init umc_init_cpu(void)
-- 
cgit v1.2.1


From 5f0f1c166647860bb2c2a206338e7d9af3834753 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: mark cpu identify functions as __cpuinit

Mark i386-specific cpu identification functions as __cpuinit. They are all
only called from arch/i386/common.c:identify_cpu() that already is marked as
__cpuinit.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/cyrix.c     | 4 ++--
 arch/i386/kernel/cpu/nexgen.c    | 4 ++--
 arch/i386/kernel/cpu/transmeta.c | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/i386/kernel/cpu/cyrix.c b/arch/i386/kernel/cpu/cyrix.c
index d34b1bdb15fd..c0c3b59de32c 100644
--- a/arch/i386/kernel/cpu/cyrix.c
+++ b/arch/i386/kernel/cpu/cyrix.c
@@ -12,7 +12,7 @@
 /*
  * Read NSC/Cyrix DEVID registers (DIR) to get more detailed info. about the CPU
  */
-static void __init do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
+static void __cpuinit do_cyrix_devid(unsigned char *dir0, unsigned char *dir1)
 {
 	unsigned char ccr2, ccr3;
 	unsigned long flags;
@@ -394,7 +394,7 @@ static inline int test_cyrix_52div(void)
 	return (unsigned char) (test >> 8) == 0x02;
 }
 
-static void cyrix_identify(struct cpuinfo_x86 * c)
+static void __cpuinit cyrix_identify(struct cpuinfo_x86 * c)
 {
 	/* Detect Cyrix with disabled CPUID */
 	if ( c->x86 == 4 && test_cyrix_52div() ) {
diff --git a/arch/i386/kernel/cpu/nexgen.c b/arch/i386/kernel/cpu/nexgen.c
index d8cc88224d17..8bf23cc80c63 100644
--- a/arch/i386/kernel/cpu/nexgen.c
+++ b/arch/i386/kernel/cpu/nexgen.c
@@ -10,7 +10,7 @@
  *	to have CPUID. (Thanks to Herbert Oppmann)
  */
  
-static int __init deep_magic_nexgen_probe(void)
+static int __cpuinit deep_magic_nexgen_probe(void)
 {
 	int ret;
 	
@@ -32,7 +32,7 @@ static void __cpuinit init_nexgen(struct cpuinfo_x86 * c)
 	c->x86_cache_size = 256; /* A few had 1 MB... */
 }
 
-static void __init nexgen_identify(struct cpuinfo_x86 * c)
+static void __cpuinit nexgen_identify(struct cpuinfo_x86 * c)
 {
 	/* Detect NexGen with old hypercode */
 	if ( deep_magic_nexgen_probe() ) {
diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c
index 725404cd77f7..4056fb7d2cdf 100644
--- a/arch/i386/kernel/cpu/transmeta.c
+++ b/arch/i386/kernel/cpu/transmeta.c
@@ -85,7 +85,7 @@ static void __cpuinit init_transmeta(struct cpuinfo_x86 *c)
 #endif
 }
 
-static void __init transmeta_identify(struct cpuinfo_x86 * c)
+static void __cpuinit transmeta_identify(struct cpuinfo_x86 * c)
 {
 	u32 xlvl;
 
-- 
cgit v1.2.1


From e9dff0ee6694b2edd40b1b448cb786f6a7b02336 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: mark cpu cache functions as __cpuinit

Mark i386-specific cpu cache functions as __cpuinit. They are all
only called from arch/i386/common.c:display_cache_info() that already is
marked as __cpuinit.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/amd.c     | 2 +-
 arch/i386/kernel/cpu/centaur.c | 2 +-
 arch/i386/kernel/cpu/intel.c   | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/cpu/amd.c b/arch/i386/kernel/cpu/amd.c
index 856cd08f2f3d..e4758095d87a 100644
--- a/arch/i386/kernel/cpu/amd.c
+++ b/arch/i386/kernel/cpu/amd.c
@@ -246,7 +246,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 		num_cache_leaves = 3;
 }
 
-static unsigned int amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+static unsigned int __cpuinit amd_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 {
 	/* AMD errata T13 (order #21922) */
 	if ((c->x86 == 6)) {
diff --git a/arch/i386/kernel/cpu/centaur.c b/arch/i386/kernel/cpu/centaur.c
index 34a81ede7892..8c25047975c0 100644
--- a/arch/i386/kernel/cpu/centaur.c
+++ b/arch/i386/kernel/cpu/centaur.c
@@ -442,7 +442,7 @@ static void __cpuinit init_centaur(struct cpuinfo_x86 *c)
 	}
 }
 
-static unsigned int centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+static unsigned int __cpuinit centaur_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 {
 	/* VIA C3 CPUs (670-68F) need further shifting. */
 	if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8)))
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 26243e019d11..94a95aa5227e 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -198,7 +198,7 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 }
 
 
-static unsigned int intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
+static unsigned int __cpuinit intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
 {
 	/* Intel PIII Tualatin. This comes in two flavours.
 	 * One has 256kb of cache, the other 512. We have no way
-- 
cgit v1.2.1


From 6f6b1e0477ccb2f25a9b045e38440347d2ce21c8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fernando=20Luis=20V=E1zquez=20Cao?= <fernando@oss.ntt.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: Disallow kprobes on NMI handlers

A kprobe executes IRET early and that could cause NMI recursion and stack
corruption.

Note: This problem was originally spotted by Andi Kleen. This patch
      adds fixes not included in his original patch.
[AK: Jan Beulich originally discovered these classes of bugs]

Signed-off-by: Fernando Vazquez <fernando@intellilink.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/mca.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/mca.c b/arch/i386/kernel/mca.c
index cd5456f14af4..eb57a851789d 100644
--- a/arch/i386/kernel/mca.c
+++ b/arch/i386/kernel/mca.c
@@ -42,6 +42,7 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/mca.h>
+#include <linux/kprobes.h>
 #include <asm/system.h>
 #include <asm/io.h>
 #include <linux/proc_fs.h>
@@ -414,7 +415,8 @@ subsys_initcall(mca_init);
 
 /*--------------------------------------------------------------------*/
 
-static void mca_handle_nmi_device(struct mca_device *mca_dev, int check_flag)
+static __kprobes void
+mca_handle_nmi_device(struct mca_device *mca_dev, int check_flag)
 {
 	int slot = mca_dev->slot;
 
@@ -444,7 +446,7 @@ static void mca_handle_nmi_device(struct mca_device *mca_dev, int check_flag)
 
 /*--------------------------------------------------------------------*/
 
-static int mca_handle_nmi_callback(struct device *dev, void *data)
+static int __kprobes mca_handle_nmi_callback(struct device *dev, void *data)
 {
 	struct mca_device *mca_dev = to_mca_device(dev);
 	unsigned char pos5;
@@ -462,7 +464,7 @@ static int mca_handle_nmi_callback(struct device *dev, void *data)
 	return 0;
 }
 
-void mca_handle_nmi(void)
+void __kprobes mca_handle_nmi(void)
 {
 	/* First try - scan the various adapters and see if a specific
 	 * adapter was responsible for the error.
-- 
cgit v1.2.1


From 06039754d775d3e48e4a292e4f353321205eff53 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Fernando=20Luis=20V=E1zquez=20Cao?= <fernando@oss.ntt.co.jp>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] i386: Disallow kprobes on NMI handlers

A kprobe executes IRET early and that could cause NMI recursion and stack
corruption.

Note: This problem was originally spotted and solved by Andi Kleen in the
x86_64 architecture. This patch is an adaption of his patch for i386.

AK: Merged with current code which was a bit different.
AK: Removed printk in nmi handler that shouldn't be there in the first time
AK: Added missing include.
AK: added KPROBES_END

Signed-off-by: Fernando Vazquez <fernando@intellilink.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S |  3 ++-
 arch/i386/kernel/nmi.c   |  6 +++---
 arch/i386/kernel/traps.c | 15 +++++++++------
 3 files changed, 14 insertions(+), 10 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index dede506e5bd0..0928f70639aa 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -729,7 +729,7 @@ KPROBE_END(debug)
  * check whether we got an NMI on the debug path where the debug
  * fault happened on the sysenter path.
  */
-ENTRY(nmi)
+KPROBE_ENTRY(nmi)
 	RING0_INT_FRAME
 	pushl %eax
 	CFI_ADJUST_CFA_OFFSET 4
@@ -805,6 +805,7 @@ nmi_16bit_stack:
 	.align 4
 	.long 1b,iret_exc
 .previous
+KPROBE_END(nmi)
 
 KPROBE_ENTRY(int3)
 	RING0_INT_FRAME
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index 7b9a053effa3..dbda706fdd14 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -22,6 +22,7 @@
 #include <linux/sysctl.h>
 #include <linux/percpu.h>
 #include <linux/dmi.h>
+#include <linux/kprobes.h>
 
 #include <asm/smp.h>
 #include <asm/nmi.h>
@@ -882,7 +883,7 @@ EXPORT_SYMBOL(touch_nmi_watchdog);
 
 extern void die_nmi(struct pt_regs *, const char *msg);
 
-int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
+__kprobes int nmi_watchdog_tick(struct pt_regs * regs, unsigned reason)
 {
 
 	/*
@@ -962,8 +963,7 @@ int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason)
 			 * This matches the old behaviour.
 			 */
 			rc = 1;
-		} else
-			printk(KERN_WARNING "Unknown enabled NMI hardware?!\n");
+		}
 	}
 done:
 	return rc;
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 00d643f3de41..5c0f4960c67d 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -689,7 +689,8 @@ gp_in_kernel:
 	}
 }
 
-static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+mem_parity_error(unsigned char reason, struct pt_regs * regs)
 {
 	printk(KERN_EMERG "Uhhuh. NMI received for unknown reason %02x on "
 		"CPU %d.\n", reason, smp_processor_id());
@@ -704,7 +705,8 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
 	clear_mem_error(reason);
 }
 
-static void io_check_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+io_check_error(unsigned char reason, struct pt_regs * regs)
 {
 	unsigned long i;
 
@@ -720,7 +722,8 @@ static void io_check_error(unsigned char reason, struct pt_regs * regs)
 	outb(reason, 0x61);
 }
 
-static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+static __kprobes void
+unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 {
 #ifdef CONFIG_MCA
 	/* Might actually be able to figure out what the guilty party
@@ -741,7 +744,7 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
 
 static DEFINE_SPINLOCK(nmi_print_lock);
 
-void die_nmi (struct pt_regs *regs, const char *msg)
+void __kprobes die_nmi(struct pt_regs *regs, const char *msg)
 {
 	if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) ==
 	    NOTIFY_STOP)
@@ -773,7 +776,7 @@ void die_nmi (struct pt_regs *regs, const char *msg)
 	do_exit(SIGSEGV);
 }
 
-static void default_do_nmi(struct pt_regs * regs)
+static __kprobes void default_do_nmi(struct pt_regs * regs)
 {
 	unsigned char reason = 0;
 
@@ -811,7 +814,7 @@ static void default_do_nmi(struct pt_regs * regs)
 	reassert_nmi();
 }
 
-fastcall void do_nmi(struct pt_regs * regs, long error_code)
+fastcall __kprobes void do_nmi(struct pt_regs * regs, long error_code)
 {
 	int cpu;
 
-- 
cgit v1.2.1


From f704cb935006580db0495e54d3c82631f6e2a984 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] x86: remove config.h includes from asm-i386 & asm-x86_64

This is now automatically included by kbuild.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/dwarf2.h    | 2 --
 include/asm-i386/tsc.h       | 1 -
 include/asm-x86_64/calgary.h | 1 -
 3 files changed, 4 deletions(-)

diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index 2280f6272f80..fe2392f361ee 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -1,8 +1,6 @@
 #ifndef _DWARF2_H
 #define _DWARF2_H
 
-#include <linux/config.h>
-
 #ifndef __ASSEMBLY__
 #warning "asm/dwarf2.h should be only included in pure assembly files"
 #endif
diff --git a/include/asm-i386/tsc.h b/include/asm-i386/tsc.h
index 97b828ce31e0..c13933185c1c 100644
--- a/include/asm-i386/tsc.h
+++ b/include/asm-i386/tsc.h
@@ -6,7 +6,6 @@
 #ifndef _ASM_i386_TSC_H
 #define _ASM_i386_TSC_H
 
-#include <linux/config.h>
 #include <asm/processor.h>
 
 /*
diff --git a/include/asm-x86_64/calgary.h b/include/asm-x86_64/calgary.h
index 0a03bda94d03..6b93f5a3a5c8 100644
--- a/include/asm-x86_64/calgary.h
+++ b/include/asm-x86_64/calgary.h
@@ -24,7 +24,6 @@
 #ifndef _ASM_X86_64_CALGARY_H
 #define _ASM_X86_64_CALGARY_H
 
-#include <linux/config.h>
 #include <linux/spinlock.h>
 #include <linux/device.h>
 #include <linux/dma-mapping.h>
-- 
cgit v1.2.1


From dbf9272e863bf4b17ee8e3c66c26682b2061d40d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:36 +0200
Subject: [PATCH] Don't force reserve the 640k-1MB range

From i386 x86-64 inherited code to force reserve the 640k-1MB area.
That was needed on some old systems.

But we generally trust the e820 map to be correct on 64bit systems
and mark all areas that are not memory correctly.

This patch will allow to use the real memory in there.

Or rather the only way to find out if it's still needed is to
try. So far I'm optimistic.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/e820.c | 33 +--------------------------------
 1 file changed, 1 insertion(+), 32 deletions(-)

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index 164d0b83dc92..e06c2714ecf3 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -71,12 +71,7 @@ static inline int bad_addr(unsigned long *addrp, unsigned long size)
 		return 1;
 	} 
 #endif
-	/* kernel code + 640k memory hole (later should not be needed, but 
-	   be paranoid for now) */
-	if (last >= 640*1024 && addr < 1024*1024) {
-		*addrp = 1024*1024;
-		return 1;
-	}
+	/* kernel code */
 	if (last >= __pa_symbol(&_text) && last < __pa_symbol(&_end)) {
 		*addrp = __pa_symbol(&_end);
 		return 1;
@@ -519,13 +514,6 @@ static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
  * If we're lucky and live on a modern system, the setup code
  * will have given us a memory map that we can use to properly
  * set up memory.  If we aren't, we'll fake a memory map.
- *
- * We check to see that the memory map contains at least 2 elements
- * before we'll use it, because the detection code in setup.S may
- * not be perfect and most every PC known to man has two memory
- * regions: one from 0 to 640k, and one from 1mb up.  (The IBM
- * thinkpad 560x, for example, does not cooperate with the memory
- * detection code.)
  */
 static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
 {
@@ -543,25 +531,6 @@ static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
 		if (start > end)
 			return -1;
 
-		/*
-		 * Some BIOSes claim RAM in the 640k - 1M region.
-		 * Not right. Fix it up.
-		 * 
-		 * This should be removed on Hammer which is supposed to not
-		 * have non e820 covered ISA mappings there, but I had some strange
-		 * problems so it stays for now.  -AK
-		 */
-		if (type == E820_RAM) {
-			if (start < 0x100000ULL && end > 0xA0000ULL) {
-				if (start < 0xA0000ULL)
-					add_memory_region(start, 0xA0000ULL-start, type);
-				if (end <= 0x100000ULL)
-					continue;
-				start = 0x100000ULL;
-				size = end - start;
-			}
-		}
-
 		add_memory_region(start, size, type);
 	} while (biosmap++,--nr_map);
 	return 0;
-- 
cgit v1.2.1


From 1bb4996bcebca1cde49d964b4e012699ce180e61 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Move compiler check for modules to ia64 only

Apparently IA64 needs it, but i386/x86-64 don't anymore
since gcc 2.95 support was dropped.  Nobody else on linux-arch
requested keeping it generically

Cc: tony.luck@intel.com
Cc: kaos@sgi.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-ia64/module.h | 3 ++-
 include/linux/vermagic.h  | 4 ++--
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/asm-ia64/module.h b/include/asm-ia64/module.h
index 85c82bd819f2..d2da61e4c49b 100644
--- a/include/asm-ia64/module.h
+++ b/include/asm-ia64/module.h
@@ -28,7 +28,8 @@ struct mod_arch_specific {
 #define Elf_Ehdr	Elf64_Ehdr
 
 #define MODULE_PROC_FAMILY	"ia64"
-#define MODULE_ARCH_VERMAGIC	MODULE_PROC_FAMILY
+#define MODULE_ARCH_VERMAGIC	MODULE_PROC_FAMILY \
+	"gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
 
 #define ARCH_SHF_SMALL	SHF_IA_64_SHORT
 
diff --git a/include/linux/vermagic.h b/include/linux/vermagic.h
index 46919f9f5eb3..4d0909e53595 100644
--- a/include/linux/vermagic.h
+++ b/include/linux/vermagic.h
@@ -24,5 +24,5 @@
 #define VERMAGIC_STRING 						\
 	UTS_RELEASE " "							\
 	MODULE_VERMAGIC_SMP MODULE_VERMAGIC_PREEMPT 			\
-	MODULE_VERMAGIC_MODULE_UNLOAD MODULE_ARCH_VERMAGIC 		\
-	"gcc-" __stringify(__GNUC__) "." __stringify(__GNUC_MINOR__)
+	MODULE_VERMAGIC_MODULE_UNLOAD MODULE_ARCH_VERMAGIC
+
-- 
cgit v1.2.1


From 1164c9994fe37d5b7035a5cf9328c98dd38af7b1 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] make numa_emulation() __init

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/numa.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index d64d6d93d04b..322bf45fc36a 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -225,7 +225,7 @@ void __init numa_init_array(void)
 int numa_fake __initdata = 0;
 
 /* Numa emulation */
-static int numa_emulation(unsigned long start_pfn, unsigned long end_pfn)
+static int __init numa_emulation(unsigned long start_pfn, unsigned long end_pfn)
 {
  	int i;
  	struct bootnode nodes[MAX_NUMNODES];
-- 
cgit v1.2.1


From a549b86dd0f3cbffcd5f9343f4ae7fcd59f7e756 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] i386: annotate FIX_STACK() and the rest of nmi()

In i386's entry.S, FIX_STACK() needs annotation because it
replaces the stack pointer.  And the rest of nmi() needs
annotation in order to compile with these new annotations.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S  | 18 +++++++++++++++---
 include/asm-i386/dwarf2.h |  2 ++
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 0928f70639aa..4b0845249222 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -701,9 +701,15 @@ device_not_available_emulate:
 	jne ok;					\
 label:						\
 	movl TSS_sysenter_esp0+offset(%esp),%esp;	\
+	CFI_DEF_CFA esp, 0;			\
+	CFI_UNDEFINED eip;			\
 	pushfl;					\
+	CFI_ADJUST_CFA_OFFSET 4;		\
 	pushl $__KERNEL_CS;			\
-	pushl $sysenter_past_esp
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	pushl $sysenter_past_esp;		\
+	CFI_ADJUST_CFA_OFFSET 4;		\
+	CFI_REL_OFFSET eip, 0
 
 KPROBE_ENTRY(debug)
 	RING0_INT_FRAME
@@ -754,6 +760,7 @@ KPROBE_ENTRY(nmi)
 	cmpl $sysenter_entry,12(%esp)
 	je nmi_debug_stack_check
 nmi_stack_correct:
+	/* We have a RING0_INT_FRAME here */
 	pushl %eax
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
@@ -764,9 +771,12 @@ nmi_stack_correct:
 	CFI_ENDPROC
 
 nmi_stack_fixup:
+	RING0_INT_FRAME
 	FIX_STACK(12,nmi_stack_correct, 1)
 	jmp nmi_stack_correct
+
 nmi_debug_stack_check:
+	/* We have a RING0_INT_FRAME here */
 	cmpw $__KERNEL_CS,16(%esp)
 	jne nmi_stack_correct
 	cmpl $debug,(%esp)
@@ -777,8 +787,10 @@ nmi_debug_stack_check:
 	jmp nmi_stack_correct
 
 nmi_16bit_stack:
-	RING0_INT_FRAME
-	/* create the pointer to lss back */
+	/* We have a RING0_INT_FRAME here.
+	 *
+	 * create the pointer to lss back
+	 */
 	pushl %ss
 	CFI_ADJUST_CFA_OFFSET 4
 	pushl %esp
diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index fe2392f361ee..5d1a8db5a9b0 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -26,6 +26,7 @@
 #define CFI_RESTORE .cfi_restore
 #define CFI_REMEMBER_STATE .cfi_remember_state
 #define CFI_RESTORE_STATE .cfi_restore_state
+#define CFI_UNDEFINED .cfi_undefined
 
 #else
 
@@ -46,6 +47,7 @@
 #define CFI_RESTORE	ignore
 #define CFI_REMEMBER_STATE ignore
 #define CFI_RESTORE_STATE ignore
+#define CFI_UNDEFINED ignore
 
 #endif
 
-- 
cgit v1.2.1


From 34464a5b8937b79801776dfb6970c1b949fed4be Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Detect clock skew during suspend

Detect the situations in which the time after a resume from disk would
be earlier than the time before the suspend and prevent them from
happening on x86_64.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/time.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 97115e608ed8..9dd15d12b659 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -1039,8 +1039,16 @@ static int timer_resume(struct sys_device *dev)
 	unsigned long flags;
 	unsigned long sec;
 	unsigned long ctime = get_cmos_time();
-	unsigned long sleep_length = (ctime - sleep_start) * HZ;
+	long sleep_length = (ctime - sleep_start) * HZ;
 
+	if (sleep_length < 0) {
+		printk(KERN_WARNING "Time skew detected in timer resume!\n");
+		/* The time after the resume must not be earlier than the time
+		 * before the suspend or some nasty things will happen
+		 */
+		sleep_length = 0;
+		ctime = sleep_start;
+	}
 	if (vxtime.hpet_address)
 		hpet_reenable();
 	else
-- 
cgit v1.2.1


From 151f8cc1169f9052095b2be36183ab132d75c6c2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove safe_smp_processor_id()

And replace all users with ordinary smp_processor_id.  The function
was originally added to get some basic oops information out even
if the GS register was corrupted. However that didn't
work for some anymore because printk is needed to print the oops
and it uses smp_processor_id() already. Also GS register corruptions
are not particularly common anymore.

This also helps the Xen port which would otherwise need to
do this in a special way because it can't access the local APIC.

Cc: Chris Wright <chrisw@sous-sol.org>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mce.c   |  2 +-
 arch/x86_64/kernel/smp.c   | 23 -----------------------
 arch/x86_64/kernel/traps.c | 10 +++++-----
 include/asm-x86_64/smp.h   |  2 --
 4 files changed, 6 insertions(+), 31 deletions(-)

diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 4e017fb30fb3..1a93c3738404 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -182,7 +182,7 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 		goto out2;
 
 	memset(&m, 0, sizeof(struct mce));
-	m.cpu = safe_smp_processor_id();
+	m.cpu = smp_processor_id();
 	rdmsrl(MSR_IA32_MCG_STATUS, m.mcgstatus);
 	if (!(m.mcgstatus & MCG_STATUS_RIPV))
 		kill_it = 1;
diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c
index 2df988bbf312..4f67697f5036 100644
--- a/arch/x86_64/kernel/smp.c
+++ b/arch/x86_64/kernel/smp.c
@@ -522,26 +522,3 @@ asmlinkage void smp_call_function_interrupt(void)
 	}
 }
 
-int safe_smp_processor_id(void)
-{
-	unsigned apicid, i;
-
-	if (disable_apic || !apic_mapped)
-		return 0;
-
-	apicid = hard_smp_processor_id();
-	if (apicid < NR_CPUS && x86_cpu_to_apicid[apicid] == apicid)
-		return apicid;
-
-	for (i = 0; i < NR_CPUS; ++i) {
-		if (x86_cpu_to_apicid[i] == apicid)
-			return i;
-	}
-
-	/* No entries in x86_cpu_to_apicid?  Either no MPS|ACPI,
-	 * or called too early.  Either way, we must be CPU 0. */
-      	if (x86_cpu_to_apicid[0] == BAD_APICID)
-		return 0;
-
-	return 0; /* Should not happen */
-}
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 34660b1e2720..38bc821e457b 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -264,7 +264,7 @@ static int dump_trace_unwind(struct unwind_frame_info *info, void *context)
 void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * stack,
 		struct stacktrace_ops *ops, void *data)
 {
-	const unsigned cpu = safe_smp_processor_id();
+	const unsigned cpu = smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
 	unsigned used = 0;
 
@@ -429,7 +429,7 @@ _show_stack(struct task_struct *tsk, struct pt_regs *regs, unsigned long *rsp)
 {
 	unsigned long *stack;
 	int i;
-	const int cpu = safe_smp_processor_id();
+	const int cpu = smp_processor_id();
 	unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr);
 	unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
 
@@ -483,7 +483,7 @@ void show_registers(struct pt_regs *regs)
 	int i;
 	int in_kernel = !user_mode(regs);
 	unsigned long rsp;
-	const int cpu = safe_smp_processor_id(); 
+	const int cpu = smp_processor_id();
 	struct task_struct *cur = cpu_pda(cpu)->pcurrent;
 
 		rsp = regs->rsp;
@@ -558,7 +558,7 @@ static unsigned int die_nest_count;
 
 unsigned __kprobes long oops_begin(void)
 {
-	int cpu = safe_smp_processor_id();
+	int cpu = smp_processor_id();
 	unsigned long flags;
 
 	oops_enter();
@@ -636,7 +636,7 @@ void __kprobes die_nmi(char *str, struct pt_regs *regs, int do_panic)
 	 * We are in trouble anyway, lets at least try
 	 * to get a message out.
 	 */
-	printk(str, safe_smp_processor_id());
+	printk(str, smp_processor_id());
 	show_registers(regs);
 	if (kexec_should_crash(current))
 		crash_kexec(regs);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 498fbc1fc179..58b5d6149a42 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -66,7 +66,6 @@ static inline int hard_smp_processor_id(void)
 	return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
 }
 
-extern int safe_smp_processor_id(void);
 extern int __cpu_disable(void);
 extern void __cpu_die(unsigned int cpu);
 extern void prefill_possible_map(void);
@@ -100,7 +99,6 @@ static inline int cpu_present_to_apicid(int mps_cpu)
 
 #ifndef CONFIG_SMP
 #define stack_smp_processor_id() 0
-#define safe_smp_processor_id() 0
 #define cpu_logical_map(x) (x)
 #else
 #include <asm/thread_info.h>
-- 
cgit v1.2.1


From d3cf7f061521c78ad62e275eb6fbdc8f43fc75a7 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove bogus warning from early_ioremap

It is correct for its only caller right now, but not for possible
future others.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/init.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 984155b75e4c..53b49c86d7e0 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -229,7 +229,6 @@ __init void *early_ioremap(unsigned long addr, unsigned long size)
 
 	/* actually usually some more */
 	if (size >= LARGE_PAGE_SIZE) { 
-		printk("SMBIOS area too long %lu\n", size);
 		return NULL;
 	}
 	set_pmd(temp_mappings[0].pmd,  __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
-- 
cgit v1.2.1


From df992848f5aa803fcacd2c5e7d67034bb89e3fa3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Fix pte_exec/mkexec and use it in change_page_attr()

Fix the pte_exec/mkexec page table accessor functions to really
use the NX bit. Previously they only checked the USER bit, but
weren't actually used for anything.

Then use them in change_page_attr() to manipulate the NX bit
properly.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/pageattr.c    | 8 +++++---
 include/asm-x86_64/pgtable.h | 4 ++--
 2 files changed, 7 insertions(+), 5 deletions(-)

diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c
index 2685b1f3671c..01591b649f4e 100644
--- a/arch/x86_64/mm/pageattr.c
+++ b/arch/x86_64/mm/pageattr.c
@@ -190,10 +190,12 @@ int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
 		 * lowmem */
 		if (__pa(address) < KERNEL_TEXT_SIZE) {
 			unsigned long addr2;
-			pgprot_t prot2 = prot;
+			pgprot_t prot2;
 			addr2 = __START_KERNEL_map + __pa(address);
- 			pgprot_val(prot2) &= ~_PAGE_NX;
-			err = __change_page_attr(addr2, pfn, prot2, PAGE_KERNEL_EXEC);
+			/* Make sure the kernel mappings stay executable */
+			prot2 = pte_pgprot(pte_mkexec(pfn_pte(0, prot)));
+			err = __change_page_attr(addr2, pfn, prot2,
+						 PAGE_KERNEL_EXEC);
 		} 
 	} 	
 	up_write(&init_mm.mmap_sem); 
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index d79e7441b513..f7614670c655 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -265,7 +265,7 @@ static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
 #define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
 static inline int pte_user(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
 static inline int pte_read(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
-static inline int pte_exec(pte_t pte)		{ return pte_val(pte) & _PAGE_USER; }
+static inline int pte_exec(pte_t pte)		{ return !(pte_val(pte) & _PAGE_NX); }
 static inline int pte_dirty(pte_t pte)		{ return pte_val(pte) & _PAGE_DIRTY; }
 static inline int pte_young(pte_t pte)		{ return pte_val(pte) & _PAGE_ACCESSED; }
 static inline int pte_write(pte_t pte)		{ return pte_val(pte) & _PAGE_RW; }
@@ -278,7 +278,7 @@ static inline pte_t pte_mkclean(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) &
 static inline pte_t pte_mkold(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_wrprotect(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_RW)); return pte; }
 static inline pte_t pte_mkread(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER)); return pte; }
-static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER)); return pte; }
+static inline pte_t pte_mkexec(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_NX)); return pte; }
 static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY)); return pte; }
 static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
-- 
cgit v1.2.1


From 5e6b0bfe5b452957b7be4b6ef181cd41880f8359 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Use proper accessors to change PSE bits in change_page_attr()

Use normal pte accessors in change_page_attr() to access the PSE
bits.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/mm/pageattr.c    | 16 ++++++----------
 include/asm-x86_64/pgtable.h |  1 +
 2 files changed, 7 insertions(+), 10 deletions(-)

diff --git a/arch/x86_64/mm/pageattr.c b/arch/x86_64/mm/pageattr.c
index 01591b649f4e..3e231d762aaa 100644
--- a/arch/x86_64/mm/pageattr.c
+++ b/arch/x86_64/mm/pageattr.c
@@ -108,8 +108,8 @@ static void revert_page(unsigned long address, pgprot_t ref_prot)
 	BUG_ON(pud_none(*pud));
 	pmd = pmd_offset(pud, address);
 	BUG_ON(pmd_val(*pmd) & _PAGE_PSE);
-	pgprot_val(ref_prot) |= _PAGE_PSE;
 	large_pte = mk_pte_phys(__pa(address) & LARGE_PAGE_MASK, ref_prot);
+	large_pte = pte_mkhuge(large_pte);
 	set_pte((pte_t *)pmd, large_pte);
 }      
 
@@ -119,32 +119,28 @@ __change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
 { 
 	pte_t *kpte; 
 	struct page *kpte_page;
-	unsigned kpte_flags;
 	pgprot_t ref_prot2;
 	kpte = lookup_address(address);
 	if (!kpte) return 0;
 	kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
-	kpte_flags = pte_val(*kpte); 
 	if (pgprot_val(prot) != pgprot_val(ref_prot)) { 
-		if ((kpte_flags & _PAGE_PSE) == 0) { 
+		if (!pte_huge(*kpte)) {
 			set_pte(kpte, pfn_pte(pfn, prot));
 		} else {
  			/*
 			 * split_large_page will take the reference for this
 			 * change_page_attr on the split page.
  			 */
-
 			struct page *split;
-			ref_prot2 = __pgprot(pgprot_val(pte_pgprot(*lookup_address(address))) & ~(1<<_PAGE_BIT_PSE));
-
+			ref_prot2 = pte_pgprot(pte_clrhuge(*kpte));
 			split = split_large_page(address, prot, ref_prot2);
 			if (!split)
 				return -ENOMEM;
-			set_pte(kpte,mk_pte(split, ref_prot2));
+			set_pte(kpte, mk_pte(split, ref_prot2));
 			kpte_page = split;
-		}	
+		}
 		page_private(kpte_page)++;
-	} else if ((kpte_flags & _PAGE_PSE) == 0) { 
+	} else if (!pte_huge(*kpte)) {
 		set_pte(kpte, pfn_pte(pfn, ref_prot));
 		BUG_ON(page_private(kpte_page) == 0);
 		page_private(kpte_page)--;
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index f7614670c655..b34f43acdef1 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -283,6 +283,7 @@ static inline pte_t pte_mkdirty(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) |
 static inline pte_t pte_mkyoung(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED)); return pte; }
 static inline pte_t pte_mkwrite(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW)); return pte; }
 static inline pte_t pte_mkhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) | _PAGE_PSE)); return pte; }
+static inline pte_t pte_clrhuge(pte_t pte)	{ set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_PSE)); return pte; }
 
 struct vm_area_struct;
 
-- 
cgit v1.2.1


From f2c2cca3acef8b253a36381d9b469ad4fb08563a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove APIC version/cpu capability mpparse checking/printing

ACPI went to great trouble to get the APIC version and CPU capabilities
of different CPUs before passing them to the mpparser. But all
that data was used was to print it out.  Actually it even faked some data
based on the boot cpu, not on the actual CPU being booted.

Remove all this code because it's not needed.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/io_apic.c | 13 -----------
 arch/x86_64/kernel/mpparse.c | 52 +++++++++++++++-----------------------------
 2 files changed, 17 insertions(+), 48 deletions(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index afac3dbb3729..0491019d4c8d 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -1748,19 +1748,6 @@ device_initcall(ioapic_init_sysfs);
 
 #define IO_APIC_MAX_ID		0xFE
 
-int __init io_apic_get_version (int ioapic)
-{
-	union IO_APIC_reg_01	reg_01;
-	unsigned long flags;
-
-	spin_lock_irqsave(&ioapic_lock, flags);
-	reg_01.raw = io_apic_read(ioapic, 1);
-	spin_unlock_irqrestore(&ioapic_lock, flags);
-
-	return reg_01.bits.version;
-}
-
-
 int __init io_apic_get_redir_entries (int ioapic)
 {
 	union IO_APIC_reg_01	reg_01;
diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index d63f849aea2c..32b6977f80c0 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -41,7 +41,6 @@ int acpi_found_madt;
  * Various Linux-internal data structures created from the
  * MP-table.
  */
-unsigned char apic_version [MAX_APICS];
 DECLARE_BITMAP(mp_bus_not_pci, MAX_MP_BUSSES);
 int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
 
@@ -94,24 +93,21 @@ static int __init mpf_checksum(unsigned char *mp, int len)
 static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 {
 	int cpu;
-	unsigned char ver;
 	cpumask_t tmp_map;
+	char *bootup_cpu = "";
 
 	if (!(m->mpc_cpuflag & CPU_ENABLED)) {
 		disabled_cpus++;
 		return;
 	}
 
-	printk(KERN_INFO "Processor #%d %d:%d APIC version %d\n",
-		m->mpc_apicid,
-	       (m->mpc_cpufeature & CPU_FAMILY_MASK)>>8,
-	       (m->mpc_cpufeature & CPU_MODEL_MASK)>>4,
-		m->mpc_apicver);
-
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
-		Dprintk("    Bootup CPU\n");
+		bootup_cpu = " (Bootup-CPU)";
 		boot_cpu_id = m->mpc_apicid;
 	}
+
+	printk(KERN_INFO "Processor #%d%s\n", m->mpc_apicid, bootup_cpu);
+
 	if (num_processors >= NR_CPUS) {
 		printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
 			" Processor ignored.\n", NR_CPUS);
@@ -129,17 +125,7 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 		return;
 	}
 #endif
-	ver = m->mpc_apicver;
-
 	physid_set(m->mpc_apicid, phys_cpu_present_map);
-	/*
-	 * Validate version
-	 */
-	if (ver == 0x0) {
-		printk(KERN_ERR "BIOS bug, APIC version is 0 for CPU#%d! fixing up to 0x10. (tell your hw vendor)\n", m->mpc_apicid);
-		ver = 0x10;
-	}
-	apic_version[m->mpc_apicid] = ver;
  	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
  		/*
  		 * bios_cpu_apicid is required to have processors listed
@@ -179,8 +165,8 @@ static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
 	if (!(m->mpc_flags & MPC_APIC_USABLE))
 		return;
 
-	printk("I/O APIC #%d Version %d at 0x%X.\n",
-		m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
+	printk("I/O APIC #%d at 0x%X.\n",
+		m->mpc_apicid, m->mpc_apicaddr);
 	if (nr_ioapics >= MAX_IO_APICS) {
 		printk(KERN_ERR "Max # of I/O APICs (%d) exceeded (found %d).\n",
 			MAX_IO_APICS, nr_ioapics);
@@ -413,13 +399,10 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 	 * 2 CPUs, numbered 0 & 1.
 	 */
 	processor.mpc_type = MP_PROCESSOR;
-	/* Either an integrated APIC or a discrete 82489DX. */
-	processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+	processor.mpc_apicver = 0;
 	processor.mpc_cpuflag = CPU_ENABLED;
-	processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
-				   (boot_cpu_data.x86_model << 4) |
-				   boot_cpu_data.x86_mask;
-	processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+	processor.mpc_cpufeature = 0;
+	processor.mpc_featureflag = 0;
 	processor.mpc_reserved[0] = 0;
 	processor.mpc_reserved[1] = 0;
 	for (i = 0; i < 2; i++) {
@@ -448,7 +431,7 @@ static inline void __init construct_default_ISA_mptable(int mpc_default_type)
 
 	ioapic.mpc_type = MP_IOAPIC;
 	ioapic.mpc_apicid = 2;
-	ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
+	ioapic.mpc_apicver = 0;
 	ioapic.mpc_flags = MPC_APIC_USABLE;
 	ioapic.mpc_apicaddr = 0xFEC00000;
 	MP_ioapic_info(&ioapic);
@@ -640,12 +623,11 @@ void __cpuinit mp_register_lapic (u8 id, u8 enabled)
 
 	processor.mpc_type = MP_PROCESSOR;
 	processor.mpc_apicid = id;
-	processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR));
+	processor.mpc_apicver = 0;
 	processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0);
 	processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0);
-	processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) | 
-		(boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
-	processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
+	processor.mpc_cpufeature = 0;
+	processor.mpc_featureflag = 0;
 	processor.mpc_reserved[0] = 0;
 	processor.mpc_reserved[1] = 0;
 
@@ -700,7 +682,7 @@ void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 
 	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
 	mp_ioapics[idx].mpc_apicid = id;
-	mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
+	mp_ioapics[idx].mpc_apicver = 0;
 	
 	/* 
 	 * Build basic IRQ lookup table to facilitate gsi->io_apic lookups
@@ -711,9 +693,9 @@ void __init mp_register_ioapic(u8 id, u32 address, u32 gsi_base)
 	mp_ioapic_routing[idx].gsi_end = gsi_base + 
 		io_apic_get_redir_entries(idx);
 
-	printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
+	printk(KERN_INFO "IOAPIC[%d]: apic_id %d, address 0x%x, "
 		"GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid, 
-		mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
+		mp_ioapics[idx].mpc_apicaddr,
 		mp_ioapic_routing[idx].gsi_start,
 		mp_ioapic_routing[idx].gsi_end);
 }
-- 
cgit v1.2.1


From e4251e130deef9de5226cc36faa70a1c6671d3c5 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove some cruft in apic id checking during processor setup

- Remove a define that was used only once
- Remove the too large APIC ID check because we always support
the full 8bit range of APICs.
- Restructure code a bit to be simpler.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 18 +-----------------
 include/asm-x86_64/acpi.h    |  2 --
 2 files changed, 1 insertion(+), 19 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 32b6977f80c0..02eef8f6415f 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -100,7 +100,6 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 		disabled_cpus++;
 		return;
 	}
-
 	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
 		bootup_cpu = " (Bootup-CPU)";
 		boot_cpu_id = m->mpc_apicid;
@@ -118,13 +117,6 @@ static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
 	cpus_complement(tmp_map, cpu_present_map);
 	cpu = first_cpu(tmp_map);
 
-#if MAX_APICS < 255	
-	if ((int)m->mpc_apicid > MAX_APICS) {
-		printk(KERN_ERR "Processor #%d INVALID. (Max ID: %d).\n",
-			m->mpc_apicid, MAX_APICS);
-		return;
-	}
-#endif
 	physid_set(m->mpc_apicid, phys_cpu_present_map);
  	if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
  		/*
@@ -603,8 +595,6 @@ void __init mp_register_lapic_address(u64 address)
 	set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
 	if (boot_cpu_id == -1U)
 		boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
-
-	Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
 }
 
 void __cpuinit mp_register_lapic (u8 id, u8 enabled)
@@ -612,13 +602,7 @@ void __cpuinit mp_register_lapic (u8 id, u8 enabled)
 	struct mpc_config_processor processor;
 	int			boot_cpu = 0;
 	
-	if (id >= MAX_APICS) {
-		printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
-			id, MAX_APICS);
-		return;
-	}
-
-	if (id == boot_cpu_physical_apicid)
+	if (id == boot_cpu_id)
 		boot_cpu = 1;
 
 	processor.mpc_type = MP_PROCESSOR;
diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h
index 2c95a319c056..ed59aa4c6ff9 100644
--- a/include/asm-x86_64/acpi.h
+++ b/include/asm-x86_64/acpi.h
@@ -155,8 +155,6 @@ extern void acpi_reserve_bootmem(void);
 
 #endif /*CONFIG_ACPI_SLEEP*/
 
-#define boot_cpu_physical_apicid boot_cpu_id
-
 extern int acpi_disabled;
 extern int acpi_pci_disabled;
 
-- 
cgit v1.2.1


From aecc63615e15de861db7436c50dade495639132c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Fix coding style and output of the mptable parser

Give the printks a consistent prefix.
Add some missing white space.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/mpparse.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index 02eef8f6415f..20e88f4b564b 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -205,7 +205,7 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 	unsigned char *mpt=((unsigned char *)mpc)+count;
 
 	if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) {
-		printk("SMP mptable: bad signature [%c%c%c%c]!\n",
+		printk("MPTABLE: bad signature [%c%c%c%c]!\n",
 			mpc->mpc_signature[0],
 			mpc->mpc_signature[1],
 			mpc->mpc_signature[2],
@@ -213,31 +213,31 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 		return 0;
 	}
 	if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) {
-		printk("SMP mptable: checksum error!\n");
+		printk("MPTABLE: checksum error!\n");
 		return 0;
 	}
 	if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) {
-		printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n",
+		printk(KERN_ERR "MPTABLE: bad table version (%d)!!\n",
 			mpc->mpc_spec);
 		return 0;
 	}
 	if (!mpc->mpc_lapic) {
-		printk(KERN_ERR "SMP mptable: null local APIC address!\n");
+		printk(KERN_ERR "MPTABLE: null local APIC address!\n");
 		return 0;
 	}
 	memcpy(str,mpc->mpc_oem,8);
-	str[8]=0;
-	printk(KERN_INFO "OEM ID: %s ",str);
+	str[8] = 0;
+	printk(KERN_INFO "MPTABLE: OEM ID: %s ",str);
 
 	memcpy(str,mpc->mpc_productid,12);
-	str[12]=0;
-	printk("Product ID: %s ",str);
+	str[12] = 0;
+	printk("MPTABLE: Product ID: %s ",str);
 
-	printk("APIC at: 0x%X\n",mpc->mpc_lapic);
+	printk("MPTABLE: APIC at: 0x%X\n",mpc->mpc_lapic);
 
 	/* save the local APIC address, it might be non-default */
 	if (!acpi_lapic)
-	mp_lapic_addr = mpc->mpc_lapic;
+		mp_lapic_addr = mpc->mpc_lapic;
 
 	/*
 	 *	Now process the configuration blocks.
@@ -249,7 +249,7 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 				struct mpc_config_processor *m=
 					(struct mpc_config_processor *)mpt;
 				if (!acpi_lapic)
-				MP_processor_info(m);
+					MP_processor_info(m);
 				mpt += sizeof(*m);
 				count += sizeof(*m);
 				break;
@@ -268,8 +268,8 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 				struct mpc_config_ioapic *m=
 					(struct mpc_config_ioapic *)mpt;
 				MP_ioapic_info(m);
-				mpt+=sizeof(*m);
-				count+=sizeof(*m);
+				mpt += sizeof(*m);
+				count += sizeof(*m);
 				break;
 			}
 			case MP_INTSRC:
@@ -278,8 +278,8 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 					(struct mpc_config_intsrc *)mpt;
 
 				MP_intsrc_info(m);
-				mpt+=sizeof(*m);
-				count+=sizeof(*m);
+				mpt += sizeof(*m);
+				count += sizeof(*m);
 				break;
 			}
 			case MP_LINTSRC:
@@ -287,15 +287,15 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 				struct mpc_config_lintsrc *m=
 					(struct mpc_config_lintsrc *)mpt;
 				MP_lintsrc_info(m);
-				mpt+=sizeof(*m);
-				count+=sizeof(*m);
+				mpt += sizeof(*m);
+				count += sizeof(*m);
 				break;
 			}
 		}
 	}
 	clustered_apic_check();
 	if (!num_processors)
-		printk(KERN_ERR "SMP mptable: no processors registered!\n");
+		printk(KERN_ERR "MPTABLE: no processors registered!\n");
 	return num_processors;
 }
 
-- 
cgit v1.2.1


From 7a0a2dff1cac1df82acfa0395bc9bc1bf0bc16ef Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Add a missing check for irq flags tracing in NMI

NMIs are not supposed to track the irq flags, but TRACE_IRQS_IRETQ
did it anyways. Add a check.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 780f9b26169f..4fcc0ad8bbeb 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -773,7 +773,9 @@ paranoid_exit\trace:
 	testl $3,CS(%rsp)
 	jnz   paranoid_userspace\trace
 paranoid_swapgs\trace:
+	.if \trace
 	TRACE_IRQS_IRETQ 0
+	.endif
 	swapgs
 paranoid_restore\trace:
 	RESTORE_ALL 8
-- 
cgit v1.2.1


From b3698c03eb6d4581e879d6bb0f183ed8dda96d37 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Fix boot code head.S warning

When compiling a 64-bit kernel on an Ubuntu 6.06 32bit system (whose GCC is also
a cross-compiler for x86_64) I've seen that head.o is compiled as a 64-bit file
(while it should not) and ld complaining about this during linking:
[AK: it happens on all systems with new binutils]

ld: warning: i386:x86-64 architecture of input file
`arch/x86_64/boot/compressed/head.o' is incompatible with i386 output

I've verified that removing -m64 from compilation flags to turn
"-m64 -traditional -m32" into "-traditional -m32" fixes the issue.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/boot/compressed/Makefile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/boot/compressed/Makefile b/arch/x86_64/boot/compressed/Makefile
index f89d96f11a9f..e70fa6e1da08 100644
--- a/arch/x86_64/boot/compressed/Makefile
+++ b/arch/x86_64/boot/compressed/Makefile
@@ -7,7 +7,8 @@
 #
 
 targets		:= vmlinux vmlinux.bin vmlinux.bin.gz head.o misc.o piggy.o
-EXTRA_AFLAGS	:= -traditional -m32
+EXTRA_AFLAGS	:= -traditional
+AFLAGS		:= $(subst -m64,-m32,$(AFLAGS))
 
 # cannot use EXTRA_CFLAGS because base CFLAGS contains -mkernel which conflicts with
 # -m32
-- 
cgit v1.2.1


From 8380aabb99719af583447133f19a4d8074b5c337 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:37 +0200
Subject: [PATCH] Remove non e820 fallbacks in high level code

Drop support for non e820 BIOS calls to get the memory map.

The boot assembler code still has some support, but not the C code now.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/e820.c | 28 +++++++++-------------------
 include/asm-x86_64/e820.h |  4 ----
 2 files changed, 9 insertions(+), 23 deletions(-)

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index e06c2714ecf3..5d1275a93658 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -536,10 +536,14 @@ static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
 	return 0;
 }
 
-void __init setup_memory_region(void)
+void early_panic(char *msg)
 {
-	char *who = "BIOS-e820";
+	early_printk(msg);
+	panic(msg);
+}
 
+void __init setup_memory_region(void)
+{
 	/*
 	 * Try to copy the BIOS-supplied E820-map.
 	 *
@@ -547,24 +551,10 @@ void __init setup_memory_region(void)
 	 * the next section from 1mb->appropriate_mem_k
 	 */
 	sanitize_e820_map(E820_MAP, &E820_MAP_NR);
-	if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0) {
-		unsigned long mem_size;
-
-		/* compare results from other methods and take the greater */
-		if (ALT_MEM_K < EXT_MEM_K) {
-			mem_size = EXT_MEM_K;
-			who = "BIOS-88";
-		} else {
-			mem_size = ALT_MEM_K;
-			who = "BIOS-e801";
-		}
-
-		e820.nr_map = 0;
-		add_memory_region(0, LOWMEMSIZE(), E820_RAM);
-		add_memory_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
-  	}
+	if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0)
+		early_panic("Cannot find a valid memory map");
 	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
-	e820_print_map(who);
+	e820_print_map("BIOS-e820");
 }
 
 static int __init parse_memopt(char *p)
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index dba012151856..8dac397bf85e 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -22,10 +22,6 @@
 #define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
 #define E820_NVS	4
 
-#define HIGH_MEMORY	(1024*1024)
-
-#define LOWMEMSIZE()	(0x9f000)
-
 #ifndef __ASSEMBLY__
 struct e820entry {
 	u64 addr;	/* start of memory segment */
-- 
cgit v1.2.1


From 0136611c62e8650e354b95c76dff6d2ce6030eff Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] optimize hweight64 for x86_64

Based on patch from David Rientjes <rientjes@google.com>, but
changed by AK.

Optimizes the 64-bit hamming weight for x86_64 processors assuming they
have fast multiplication.  Uses five fewer bitops than the generic
hweight64.  Benchmark on one EMT64 showed ~25% speedup with 2^24
consecutive calls.

Define a new ARCH_HAS_FAST_MULTIPLIER that can be set by other
architectures that can also multiply fast.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/bitops.h |  2 ++
 lib/hweight.c               | 10 ++++++++--
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/include/asm-x86_64/bitops.h b/include/asm-x86_64/bitops.h
index f7ba57b1cc08..5b535eaf5309 100644
--- a/include/asm-x86_64/bitops.h
+++ b/include/asm-x86_64/bitops.h
@@ -399,6 +399,8 @@ static __inline__ int fls(int x)
 	return r+1;
 }
 
+#define ARCH_HAS_FAST_MULTIPLIER 1
+
 #include <asm-generic/bitops/hweight.h>
 
 #endif /* __KERNEL__ */
diff --git a/lib/hweight.c b/lib/hweight.c
index 438257671708..360556a7803d 100644
--- a/lib/hweight.c
+++ b/lib/hweight.c
@@ -1,5 +1,6 @@
 #include <linux/module.h>
 #include <asm/types.h>
+#include <asm/bitops.h>
 
 /**
  * hweightN - returns the hamming weight of a N-bit word
@@ -40,14 +41,19 @@ unsigned long hweight64(__u64 w)
 #if BITS_PER_LONG == 32
 	return hweight32((unsigned int)(w >> 32)) + hweight32((unsigned int)w);
 #elif BITS_PER_LONG == 64
+#ifdef ARCH_HAS_FAST_MULTIPLIER
+	w -= (w >> 1) & 0x5555555555555555ul;
+	w =  (w & 0x3333333333333333ul) + ((w >> 2) & 0x3333333333333333ul);
+	w =  (w + (w >> 4)) & 0x0f0f0f0f0f0f0f0ful;
+	return (w * 0x0101010101010101ul) >> 56;
+#else
 	__u64 res = w - ((w >> 1) & 0x5555555555555555ul);
 	res = (res & 0x3333333333333333ul) + ((res >> 2) & 0x3333333333333333ul);
 	res = (res + (res >> 4)) & 0x0F0F0F0F0F0F0F0Ful;
 	res = res + (res >> 8);
 	res = res + (res >> 16);
 	return (res + (res >> 32)) & 0x00000000000000FFul;
-#else
-#error BITS_PER_LONG not defined
+#endif
 #endif
 }
 EXPORT_SYMBOL(hweight64);
-- 
cgit v1.2.1


From 26374c7b7dca1ff90607c83d9b82e917119f0456 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Reload CS when startup_64 is used.

In long mode the %cs is largely a relic.  However there are a few cases
like iret where it matters that we have a valid value.  Without this
patch it is possible to enter the kernel in startup_64 without setting
%cs to a valid value.  With this patch we don't care what %cs value
we enter the kernel with, so long as the cs shadow register indicates
it is a privileged code segment.

Thanks to Magnus Damm for finding this problem and posting the
first workable patch.  I have moved the jump to set %cs down a
few instructions so we don't need to take an extra jump.  Which
keeps the code simpler.

Signed-of-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/head.S | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index 4a3326ce709c..1e6f80870679 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -185,12 +185,15 @@ startup_64:
 	
 	/* Finally jump to run C code and to be on real kernel address
 	 * Since we are running on identity-mapped space we have to jump
-	 * to the full 64bit address , this is only possible as indirect
-	 * jump
+	 * to the full 64bit address, this is only possible as indirect
+	 * jump.  In addition we need to ensure %cs is set so we make this
+	 * a far return.
 	 */
 	movq	initial_code(%rip),%rax
-	pushq	$0		# fake return address
-	jmp	*%rax
+	pushq	$0		# fake return address to stop unwinder
+	pushq	$__KERNEL_CS	# set correct cs
+	pushq	%rax		# target address in negative space
+	lretq
 
 	/* SMP bootup changes these two */
 	.align	8
-- 
cgit v1.2.1


From f2a9e1dec27189a09ff642f2648e49ad9e76607b Mon Sep 17 00:00:00 2001
From: Ian Campbell <Ian.Campbell@XenSource.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Put .note.* sections into a PT_NOTE segment

This patch updates x86_64 linker script to pack any .note.* sections
into a PT_NOTE segment in the output file.

To do this, we tell ld that we need a PT_NOTE segment.  This requires
us to start explicitly mapping sections to segments, so we also need
to explicitly create PT_LOAD segments for text and data, and map the
sections to them appropriately.  Fortunately, each section will
default to its previous section's segment, so it doesn't take many
changes to vmlinux.lds.S.

The corresponding change is already made for i386 in -mm and I'd like
this patch to join it. The section to segment mappings do change as do
the segment flags so some time in -mm would be good for that reason as
well, just in case.

In particular .data and .bss move from the text segment to the data
segment and .data.cacheline_aligned .data.read_mostly are put in the
data segment instead of a separate one.

I think that it would be possible to exactly match the existing section
to segment mapping and flags but it would be a more intrusive change and
I'm not sure there is a reason for the existing layout other than it is
what you get by default if you don't explicitly specify something else.
If there is a reason for the existing layout then I will of course make
the more intrusive change. If there is no reason we could probably drop
the executable or writable flags from some segments but I don't know how
much attention is paid to them anyway so it might not be worth the
effort.

The vsyscall related sections need to go in a different segment to the
normal data segment and so I invented a "user" segment to contain them.
I believe this should appear to be another data segment as far as the
kernel is concerned so the flags are setup accordingly.

The notes will be used in the Xen paravirt_ops backend to provide
additional information to the domain builder. I am in the process of
converting the xen-unstable kernels and tools over to this scheme at the
moment to support this in the future.

It has been suggested to me that the notes segment should have flags 0
(i.e. not readable) since it is only used by the loader and is not used
at runtime. For now I went with a readable segment since that is what
the i386 patch uses.

AK: dropped NOTES addition right now because the needed infrastructure
for that is not merged yet

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/vmlinux.lds.S | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 8d5a5149bb3a..2802aa963fa3 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -13,6 +13,12 @@ OUTPUT_FORMAT("elf64-x86-64", "elf64-x86-64", "elf64-x86-64")
 OUTPUT_ARCH(i386:x86-64)
 ENTRY(phys_startup_64)
 jiffies_64 = jiffies;
+PHDRS {
+	text PT_LOAD FLAGS(5);	/* R_E */
+	data PT_LOAD FLAGS(7);	/* RWE */
+	user PT_LOAD FLAGS(7);	/* RWE */
+	note PT_NOTE FLAGS(4);	/* R__ */
+}
 SECTIONS
 {
   . = __START_KERNEL;
@@ -31,7 +37,7 @@ SECTIONS
 	KPROBES_TEXT
 	*(.fixup)
 	*(.gnu.warning)
-	} = 0x9090
+	} :text = 0x9090
   				/* out-of-line lock text */
   .text.lock : AT(ADDR(.text.lock) - LOAD_OFFSET) { *(.text.lock) }
 
@@ -57,7 +63,7 @@ SECTIONS
   .data : AT(ADDR(.data) - LOAD_OFFSET) {
 	*(.data)
 	CONSTRUCTORS
-	}
+	} :data
 
   _edata = .;			/* End of data section */
 
@@ -89,7 +95,7 @@ SECTIONS
 #define VVIRT(x) (ADDR(x) - VVIRT_OFFSET)
 
   . = VSYSCALL_ADDR;
-  .vsyscall_0 :	 AT(VSYSCALL_PHYS_ADDR) { *(.vsyscall_0) }
+  .vsyscall_0 :	 AT(VSYSCALL_PHYS_ADDR) { *(.vsyscall_0) } :user
   __vsyscall_0 = VSYSCALL_VIRT_ADDR;
 
   . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
@@ -135,7 +141,7 @@ SECTIONS
   . = ALIGN(8192);		/* init_task */
   .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) {
 	*(.data.init_task)
-  }
+  } :data
 
   . = ALIGN(4096);
   .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) {
-- 
cgit v1.2.1


From 80d2679cbc8e170011c9649fb8fb684ffd7e5c8f Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] x86: Remove incorrect comment about ACPI e820 entries

They cannot be actually freed because the FACS table has a
shared-with-the-BIOS lock.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-i386/e820.h   | 2 +-
 include/asm-x86_64/e820.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-i386/e820.h b/include/asm-i386/e820.h
index ca82acb8cb1f..f7514fb6e8e4 100644
--- a/include/asm-i386/e820.h
+++ b/include/asm-i386/e820.h
@@ -18,7 +18,7 @@
 
 #define E820_RAM	1
 #define E820_RESERVED	2
-#define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
+#define E820_ACPI	3
 #define E820_NVS	4
 
 #define HIGH_MEMORY	(1024*1024)
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 8dac397bf85e..fd4198b49637 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -19,7 +19,7 @@
 
 #define E820_RAM	1
 #define E820_RESERVED	2
-#define E820_ACPI	3 /* usable as RAM once ACPI tables have been read */
+#define E820_ACPI	3
 #define E820_NVS	4
 
 #ifndef __ASSEMBLY__
-- 
cgit v1.2.1


From 53ee11ae0d73f28029a5f0d991bc4dcd7c817e7a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Optimize PDA accesses slightly

Based on a idea by Jeremy Fitzhardinge:

Replace the volatiles and memory clobbers in the PDA access with
telling gcc about access to a proxy PDA structure that doesn't
actually exist. But the dummy accesses give a defined ordering for
read/write accesses.

Also add some memory barriers to the early GS initialization to
make sure no PDA access is moved before it.

Advantage is some .text savings (probably most from better
code for accessing "current"):

   text    data     bss     dec     hex filename
4845647 1223688  615864 6685199  66020f vmlinux
4837780 1223688  615864 6677332  65e354 vmlinux-pda

1.2% smaller code

Cc:  Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c |  3 +++
 include/asm-x86_64/pda.h     | 41 ++++++++++++++++++++++-------------------
 2 files changed, 25 insertions(+), 19 deletions(-)

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index e85cfbb49b63..491361752c70 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -121,7 +121,10 @@ void pda_init(int cpu)
 
 	/* Setup up data that may be needed in __get_free_pages early */
 	asm volatile("movl %0,%%fs ; movl %0,%%gs" :: "r" (0)); 
+	/* Memory clobbers used to order PDA accessed */
+	mb();
 	wrmsrl(MSR_GS_BASE, pda);
+	mb();
 
 	pda->cpunumber = cpu; 
 	pda->irqcount = -1;
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index b47c3df9ed1d..55e21da96e7a 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -36,40 +36,43 @@ extern struct x8664_pda boot_cpu_pda[];
  * There is no fast way to get the base address of the PDA, all the accesses
  * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
  */ 
-#define sizeof_field(type,field)  (sizeof(((type *)0)->field))
-#define typeof_field(type,field)  typeof(((type *)0)->field)
-
 extern void __bad_pda_field(void);
 
+/* proxy_pda doesn't actually exist, but tell gcc it is accessed
+   for all PDA accesses so it gets read/write dependencies right. */
+extern struct x8664_pda _proxy_pda;
+
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
 #define pda_to_op(op,field,val) do { \
-	typedef typeof_field(struct x8664_pda, field) T__; \
-       switch (sizeof_field(struct x8664_pda, field)) { 		\
+	typedef typeof(_proxy_pda.field) T__; \
+       switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm volatile(op "w %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
+asm(op "w %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 4: \
-asm volatile(op "l %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
+asm(op "l %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 8: \
-asm volatile(op "q %0,%%gs:%P1"::"ri" ((T__)val),"i"(pda_offset(field)):"memory"); break; \
-       default: __bad_pda_field(); 					\
+asm(op "q %1,%%gs:%P2": "+m" (_proxy_pda.field) : \
+	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
+default: __bad_pda_field(); 					\
        } \
        } while (0)
 
-/* 
- * AK: PDA read accesses should be neither volatile nor have an memory clobber.
- * Unfortunately removing them causes all hell to break lose currently.
- */
 #define pda_from_op(op,field) ({ \
-       typeof_field(struct x8664_pda, field) ret__; \
-       switch (sizeof_field(struct x8664_pda, field)) { 		\
+       typeof(_proxy_pda.field) ret__; \
+       switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm volatile(op "w %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
+asm(op "w %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 4: \
-asm volatile(op "l %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
+asm(op "l %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 8: \
-asm volatile(op "q %%gs:%P1,%0":"=r" (ret__):"i"(pda_offset(field)):"memory"); break;\
-       default: __bad_pda_field(); 					\
+asm(op "q %%gs:%P1,%0":"=r" (ret__):\
+	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
+default: __bad_pda_field(); 					\
        } \
        ret__; })
 
-- 
cgit v1.2.1


From 575400d1b483fbe9e03c68758059bfaf4e4768d1 Mon Sep 17 00:00:00 2001
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] i386: Fix the EDD code misparsing the command line

The EDD code would scan the command line as a fixed array, without
taking account of either whitespace, null-termination, the old
command-line protocol, late overrides early, or the fact that the
command line may not be reachable from INITSEG.

This should fix those problems, and enable us to use a longer command
line.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/boot/edd.S | 97 ++++++++++++++++++++++++++++++++++++++++------------
 include/linux/edd.h  |  1 +
 2 files changed, 76 insertions(+), 22 deletions(-)

diff --git a/arch/i386/boot/edd.S b/arch/i386/boot/edd.S
index 4b84ea216f2b..34321368011a 100644
--- a/arch/i386/boot/edd.S
+++ b/arch/i386/boot/edd.S
@@ -15,42 +15,95 @@
 #include <asm/setup.h>
 
 #if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
+
+# It is assumed that %ds == INITSEG here
+
 	movb	$0, (EDD_MBR_SIG_NR_BUF)
 	movb	$0, (EDDNR)
 
-# Check the command line for two options:
+# Check the command line for options:
 # edd=of  disables EDD completely  (edd=off)
 # edd=sk  skips the MBR test    (edd=skipmbr)
+# edd=on  re-enables EDD (edd=on)
+
 	pushl	%esi
-    	cmpl	$0, %cs:cmd_line_ptr
-	jz	done_cl
+	movw	$edd_mbr_sig_start, %di	# Default to edd=on
+
 	movl	%cs:(cmd_line_ptr), %esi
-# ds:esi has the pointer to the command line now
-	movl	$(COMMAND_LINE_SIZE-7), %ecx
-# loop through kernel command line one byte at a time
-cl_loop:
-	cmpl	$EDD_CL_EQUALS, (%si)
+	andl	%esi, %esi
+	jz	old_cl			# Old boot protocol?
+
+# Convert to a real-mode pointer in fs:si
+	movl	%esi, %eax
+	shrl	$4, %eax
+	movw	%ax, %fs
+	andw	$0xf, %si
+	jmp	have_cl_pointer
+
+# Old-style boot protocol?
+old_cl:
+	push	%ds			# aka INITSEG
+	pop	%fs
+
+	cmpw	$0xa33f, (0x20)
+	jne	done_cl			# No command line at all?
+	movw	(0x22), %si		# Pointer relative to INITSEG
+
+# fs:si has the pointer to the command line now
+have_cl_pointer:
+
+# Loop through kernel command line one byte at a time.  Just in
+# case the loader is buggy and failed to null-terminate the command line
+# terminate if we get close enough to the end of the segment that we
+# cannot fit "edd=XX"...
+cl_atspace:
+	cmpw	$-5, %si		# Watch for segment wraparound
+	jae	done_cl
+	movl	%fs:(%si), %eax
+	andb	%al, %al		# End of line?
+	jz	done_cl
+	cmpl	$EDD_CL_EQUALS, %eax
 	jz	found_edd_equals
-	incl	%esi
-	loop	cl_loop
-	jmp	done_cl
+	cmpb	$0x20, %al		# <= space consider whitespace
+	ja	cl_skipword
+	incw	%si
+	jmp	cl_atspace
+
+cl_skipword:
+	cmpw	$-5, %si		# Watch for segment wraparound
+	jae	done_cl
+	movb	%fs:(%si), %al		# End of string?
+	andb	%al, %al
+	jz	done_cl
+	cmpb	$0x20, %al
+	jbe	cl_atspace
+	incw	%si
+	jmp	cl_skipword
+
 found_edd_equals:
 # only looking at first two characters after equals
-    	addl	$4, %esi
-	cmpw	$EDD_CL_OFF, (%si)	# edd=of
-	jz	do_edd_off
-	cmpw	$EDD_CL_SKIP, (%si)	# edd=sk
-	jz	do_edd_skipmbr
-	jmp	done_cl
+# late overrides early on the command line, so keep going after finding something
+	movw	%fs:4(%si), %ax
+	cmpw	$EDD_CL_OFF, %ax	# edd=of
+	je	do_edd_off
+	cmpw	$EDD_CL_SKIP, %ax	# edd=sk
+	je	do_edd_skipmbr
+	cmpw	$EDD_CL_ON, %ax		# edd=on
+	je	do_edd_on
+	jmp	cl_skipword
 do_edd_skipmbr:
-    	popl	%esi
-	jmp	edd_start
+	movw	$edd_start, %di
+	jmp	cl_skipword
 do_edd_off:
-	popl	%esi
-	jmp	edd_done
+	movw	$edd_done, %di
+	jmp	cl_skipword
+do_edd_on:
+	movw	$edd_mbr_sig_start, %di
+	jmp	cl_skipword
+
 done_cl:
 	popl	%esi
-
+	jmpw	*%di
 
 # Read the first sector of each BIOS disk device and store the 4-byte signature
 edd_mbr_sig_start:
diff --git a/include/linux/edd.h b/include/linux/edd.h
index 162512b886f7..b2b3e68aa512 100644
--- a/include/linux/edd.h
+++ b/include/linux/edd.h
@@ -52,6 +52,7 @@
 #define EDD_CL_EQUALS   0x3d646465     /* "edd=" */
 #define EDD_CL_OFF      0x666f         /* "of" for off  */
 #define EDD_CL_SKIP     0x6b73         /* "sk" for skipmbr */
+#define EDD_CL_ON       0x6e6f	       /* "on" for on */
 
 #ifndef __ASSEMBLY__
 
-- 
cgit v1.2.1


From f574164491d00d28b727d713685fb5edc9138200 Mon Sep 17 00:00:00 2001
From: Keith Owens <kaos@ocs.com.au>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Remove most of the special cases for the debug IST stack

Remove most of the special cases for the debug IST stack.  This is a
follow on clean up patch, it requires the bug fix patch that adds
orig_ist.

Signed-off-by: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup64.c | 21 +++++----------------
 arch/x86_64/kernel/traps.c   | 21 +--------------------
 include/asm-x86_64/pda.h     |  3 ---
 3 files changed, 6 insertions(+), 39 deletions(-)

diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 491361752c70..9332d2361e08 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -237,28 +237,17 @@ void __cpuinit cpu_init (void)
 	 * set up and load the per-CPU TSS
 	 */
 	for (v = 0; v < N_EXCEPTION_STACKS; v++) {
+		static const unsigned int order[N_EXCEPTION_STACKS] = {
+			[0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
+			[DEBUG_STACK - 1] = DEBUG_STACK_ORDER
+		};
 		if (cpu) {
-			static const unsigned int order[N_EXCEPTION_STACKS] = {
-				[0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
-				[DEBUG_STACK - 1] = DEBUG_STACK_ORDER
-			};
-
 			estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
 			if (!estacks)
 				panic("Cannot allocate exception stack %ld %d\n",
 				      v, cpu); 
 		}
-		switch (v + 1) {
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		case DEBUG_STACK:
-			cpu_pda(cpu)->debugstack = (unsigned long)estacks;
-			estacks += DEBUG_STKSZ;
-			break;
-#endif
-		default:
-			estacks += EXCEPTION_STKSZ;
-			break;
-		}
+		estacks += PAGE_SIZE << order[v];
 		orig_ist->ist[v] = t->ist[v] = (unsigned long)estacks;
 	}
 
diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index 38bc821e457b..fb8486eca1b6 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -162,26 +162,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
 	 * 'stack' is in one of them:
 	 */
 	for (k = 0; k < N_EXCEPTION_STACKS; k++) {
-		unsigned long end;
-
-		/*
-		 * set 'end' to the end of the exception stack.
-		 */
-		switch (k + 1) {
-		/*
-		 * TODO: this block is not needed i think, because
-		 * setup64.c:cpu_init() sets up t->ist[DEBUG_STACK]
-		 * properly too.
-		 */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-		case DEBUG_STACK:
-			end = cpu_pda(cpu)->debugstack + DEBUG_STKSZ;
-			break;
-#endif
-		default:
-			end = per_cpu(orig_ist, cpu).ist[k];
-			break;
-		}
+		unsigned long end = per_cpu(orig_ist, cpu).ist[k];
 		/*
 		 * Is 'stack' above this exception frame's end?
 		 * If yes then skip to the next frame.
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 55e21da96e7a..e2b23e337b94 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -13,9 +13,6 @@ struct x8664_pda {
 	unsigned long data_offset;	/* Per cpu data offset from linker address */
 	unsigned long kernelstack;  /* top of kernel stack for current */ 
 	unsigned long oldrsp; 	    /* user rsp for system call */
-#if DEBUG_STKSZ > EXCEPTION_STKSZ
-	unsigned long debugstack;   /* #DB/#BP stack. */
-#endif
         int irqcount;		    /* Irq nesting counter. Starts with -1 */  	
 	int cpunumber;		    /* Logical CPU number */
 	char *irqstackptr;	/* top of irqstack */
-- 
cgit v1.2.1


From 4bfaaef01a1badb9e8ffb0c0a37cd2379008d21f Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Avoid overwriting the current pgd (V4, x86_64)

kexec: Avoid overwriting the current pgd (V4, x86_64)

This patch upgrades the x86_64-specific kexec code to avoid overwriting the
current pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used
to start a secondary kernel that dumps the memory of the previous kernel.

The code introduces a new set of page tables. These tables are used to provide
an executable identity mapping without overwriting the current pgd.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/machine_kexec.c   |  71 ++++++++-------
 arch/x86_64/kernel/relocate_kernel.S | 171 +++++++++++++++++++++++++++++++----
 include/asm-x86_64/kexec.h           |  29 ++++++
 3 files changed, 218 insertions(+), 53 deletions(-)

diff --git a/arch/x86_64/kernel/machine_kexec.c b/arch/x86_64/kernel/machine_kexec.c
index 2e94c072d84a..0497e3bd5bff 100644
--- a/arch/x86_64/kernel/machine_kexec.c
+++ b/arch/x86_64/kernel/machine_kexec.c
@@ -15,6 +15,15 @@
 #include <asm/mmu_context.h>
 #include <asm/io.h>
 
+#define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
+static u64 kexec_pgd[512] PAGE_ALIGNED;
+static u64 kexec_pud0[512] PAGE_ALIGNED;
+static u64 kexec_pmd0[512] PAGE_ALIGNED;
+static u64 kexec_pte0[512] PAGE_ALIGNED;
+static u64 kexec_pud1[512] PAGE_ALIGNED;
+static u64 kexec_pmd1[512] PAGE_ALIGNED;
+static u64 kexec_pte1[512] PAGE_ALIGNED;
+
 static void init_level2_page(pmd_t *level2p, unsigned long addr)
 {
 	unsigned long end_addr;
@@ -144,32 +153,19 @@ static void load_segments(void)
 		);
 }
 
-typedef NORET_TYPE void (*relocate_new_kernel_t)(unsigned long indirection_page,
-					unsigned long control_code_buffer,
-					unsigned long start_address,
-					unsigned long pgtable) ATTRIB_NORET;
-
-extern const unsigned char relocate_new_kernel[];
-extern const unsigned long relocate_new_kernel_size;
-
 int machine_kexec_prepare(struct kimage *image)
 {
-	unsigned long start_pgtable, control_code_buffer;
+	unsigned long start_pgtable;
 	int result;
 
 	/* Calculate the offsets */
 	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
-	control_code_buffer = start_pgtable + PAGE_SIZE;
 
 	/* Setup the identity mapped 64bit page table */
 	result = init_pgtable(image, start_pgtable);
 	if (result)
 		return result;
 
-	/* Place the code in the reboot code buffer */
-	memcpy(__va(control_code_buffer), relocate_new_kernel,
-						relocate_new_kernel_size);
-
 	return 0;
 }
 
@@ -184,28 +180,34 @@ void machine_kexec_cleanup(struct kimage *image)
  */
 NORET_TYPE void machine_kexec(struct kimage *image)
 {
-	unsigned long page_list;
-	unsigned long control_code_buffer;
-	unsigned long start_pgtable;
-	relocate_new_kernel_t rnk;
+	unsigned long page_list[PAGES_NR];
+	void *control_page;
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
-	/* Calculate the offsets */
-	page_list = image->head;
-	start_pgtable = page_to_pfn(image->control_code_page) << PAGE_SHIFT;
-	control_code_buffer = start_pgtable + PAGE_SIZE;
-
-	/* Set the low half of the page table to my identity mapped
-	 * page table for kexec.  Leave the high half pointing at the
-	 * kernel pages.   Don't bother to flush the global pages
-	 * as that will happen when I fully switch to my identity mapped
-	 * page table anyway.
-	 */
-	memcpy(__va(read_cr3()), __va(start_pgtable), PAGE_SIZE/2);
-	__flush_tlb();
-
+	control_page = page_address(image->control_code_page) + PAGE_SIZE;
+	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+
+	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[PA_PGD] = __pa(kexec_pgd);
+	page_list[VA_PGD] = (unsigned long)kexec_pgd;
+	page_list[PA_PUD_0] = __pa(kexec_pud0);
+	page_list[VA_PUD_0] = (unsigned long)kexec_pud0;
+	page_list[PA_PMD_0] = __pa(kexec_pmd0);
+	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
+	page_list[PA_PTE_0] = __pa(kexec_pte0);
+	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
+	page_list[PA_PUD_1] = __pa(kexec_pud1);
+	page_list[VA_PUD_1] = (unsigned long)kexec_pud1;
+	page_list[PA_PMD_1] = __pa(kexec_pmd1);
+	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
+	page_list[PA_PTE_1] = __pa(kexec_pte1);
+	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
+
+	page_list[PA_TABLE_PAGE] =
+	  (unsigned long)__pa(page_address(image->control_code_page));
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -222,9 +224,10 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	 */
 	set_gdt(phys_to_virt(0),0);
 	set_idt(phys_to_virt(0),0);
+
 	/* now call it */
-	rnk = (relocate_new_kernel_t) control_code_buffer;
-	(*rnk)(page_list, control_code_buffer, image->start, start_pgtable);
+	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
+			image->start);
 }
 
 /* crashkernel=size@addr specifies the location to reserve for
diff --git a/arch/x86_64/kernel/relocate_kernel.S b/arch/x86_64/kernel/relocate_kernel.S
index d24fa9b72a2b..14e95872c6a3 100644
--- a/arch/x86_64/kernel/relocate_kernel.S
+++ b/arch/x86_64/kernel/relocate_kernel.S
@@ -7,31 +7,169 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/page.h>
+#include <asm/kexec.h>
 
-	/*
-	 * Must be relocatable PIC code callable as a C function, that once
-	 * it starts can not use the previous processes stack.
-	 */
-	.globl relocate_new_kernel
+/*
+ * Must be relocatable PIC code callable as a C function
+ */
+
+#define PTR(x) (x << 3)
+#define PAGE_ALIGNED (1 << PAGE_SHIFT)
+#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */
+
+	.text
+	.align PAGE_ALIGNED
 	.code64
+	.globl relocate_kernel
+relocate_kernel:
+	/* %rdi indirection_page
+	 * %rsi page_list
+	 * %rdx start address
+	 */
+
+	/* map the control page at its virtual address */
+
+	movq	$0x0000ff8000000000, %r10        /* mask */
+	mov	$(39 - 3), %cl                   /* bits to shift */
+	movq	PTR(VA_CONTROL_PAGE)(%rsi), %r11 /* address to map */
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PGD)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PUD_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PUD_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PMD_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PMD_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PTE_0)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PTE_0)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	/* identity map the control page at its physical address */
+
+	movq	$0x0000ff8000000000, %r10        /* mask */
+	mov	$(39 - 3), %cl                   /* bits to shift */
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r11 /* address to map */
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PGD)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PUD_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PUD_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PMD_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PMD_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_PTE_1)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
+	shrq	$9, %r10
+	sub	$9, %cl
+
+	movq	%r11, %r9
+	andq	%r10, %r9
+	shrq	%cl, %r9
+
+	movq	PTR(VA_PTE_1)(%rsi), %r8
+	addq	%r8, %r9
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+	orq	$PAGE_ATTR, %r8
+	movq	%r8, (%r9)
+
 relocate_new_kernel:
-	/* %rdi page_list
-	 * %rsi reboot_code_buffer
+	/* %rdi indirection_page
+	 * %rsi page_list
 	 * %rdx start address
-	 * %rcx page_table
-	 * %r8  arg5
-	 * %r9  arg6
 	 */
 
 	/* zero out flags, and disable interrupts */
 	pushq $0
 	popfq
 
-	/* set a new stack at the bottom of our page... */
-	lea   4096(%rsi), %rsp
+	/* get physical address of control page now */
+	/* this is impossible after page table switch */
+	movq	PTR(PA_CONTROL_PAGE)(%rsi), %r8
+
+	/* get physical address of page table now too */
+	movq	PTR(PA_TABLE_PAGE)(%rsi), %rcx
 
-	/* store the parameters back on the stack */
-	pushq	%rdx /* store the start address */
+	/* switch to new set of page tables */
+	movq	PTR(PA_PGD)(%rsi), %r9
+	movq	%r9, %cr3
+
+	/* setup a new stack at the end of the physical control page */
+	lea	4096(%r8), %rsp
+
+	/* jump to identity mapped page */
+	addq	$(identity_mapped - relocate_kernel), %r8
+	pushq	%r8
+	ret
+
+identity_mapped:
+	/* store the start address on the stack */
+	pushq   %rdx
 
 	/* Set cr0 to a known state:
 	 * 31 1 == Paging enabled
@@ -136,8 +274,3 @@ relocate_new_kernel:
 	xorq	%r15, %r15
 
 	ret
-relocate_new_kernel_end:
-
-	.globl relocate_new_kernel_size
-relocate_new_kernel_size:
-	.quad relocate_new_kernel_end - relocate_new_kernel
diff --git a/include/asm-x86_64/kexec.h b/include/asm-x86_64/kexec.h
index c564bae03433..5fab957e1091 100644
--- a/include/asm-x86_64/kexec.h
+++ b/include/asm-x86_64/kexec.h
@@ -1,6 +1,27 @@
 #ifndef _X86_64_KEXEC_H
 #define _X86_64_KEXEC_H
 
+#define PA_CONTROL_PAGE  0
+#define VA_CONTROL_PAGE  1
+#define PA_PGD           2
+#define VA_PGD           3
+#define PA_PUD_0         4
+#define VA_PUD_0         5
+#define PA_PMD_0         6
+#define VA_PMD_0         7
+#define PA_PTE_0         8
+#define VA_PTE_0         9
+#define PA_PUD_1         10
+#define VA_PUD_1         11
+#define PA_PMD_1         12
+#define VA_PMD_1         13
+#define PA_PTE_1         14
+#define VA_PTE_1         15
+#define PA_TABLE_PAGE    16
+#define PAGES_NR         17
+
+#ifndef __ASSEMBLY__
+
 #include <linux/string.h>
 
 #include <asm/page.h>
@@ -64,4 +85,12 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 		newregs->rip = (unsigned long)current_text_addr();
 	}
 }
+
+NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long page_list,
+		unsigned long start_address) ATTRIB_NORET;
+
+#endif /* __ASSEMBLY__ */
+
 #endif /* _X86_64_KEXEC_H */
-- 
cgit v1.2.1


From 3566561bfadffcb5dbc85d576be80c0dbf2cccc9 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus@valinux.co.jp>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] i386: Avoid overwriting the current pgd (V4, i386)

kexec: Avoid overwriting the current pgd (V4, i386)

This patch upgrades the i386-specific kexec code to avoid overwriting the
current pgd. Overwriting the current pgd is bad when CONFIG_CRASH_DUMP is used
to start a secondary kernel that dumps the memory of the previous kernel.

The code introduces a new set of page tables. These tables are used to provide
an executable identity mapping without overwriting the current pgd.

Signed-off-by: Magnus Damm <magnus@valinux.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/machine_kexec.c   | 117 +++++++--------------------
 arch/i386/kernel/relocate_kernel.S | 162 +++++++++++++++++++++++++++++++++----
 include/asm-i386/kexec.h           |  27 +++++++
 3 files changed, 201 insertions(+), 105 deletions(-)

diff --git a/arch/i386/kernel/machine_kexec.c b/arch/i386/kernel/machine_kexec.c
index 66c3dc99a655..91966bafb3dc 100644
--- a/arch/i386/kernel/machine_kexec.c
+++ b/arch/i386/kernel/machine_kexec.c
@@ -21,70 +21,13 @@
 #include <asm/system.h>
 
 #define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
-
-#define L0_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define L1_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define L2_ATTR (_PAGE_PRESENT)
-
-#define LEVEL0_SIZE (1UL << 12UL)
-
-#ifndef CONFIG_X86_PAE
-#define LEVEL1_SIZE (1UL << 22UL)
-static u32 pgtable_level1[1024] PAGE_ALIGNED;
-
-static void identity_map_page(unsigned long address)
-{
-	unsigned long level1_index, level2_index;
-	u32 *pgtable_level2;
-
-	/* Find the current page table */
-	pgtable_level2 = __va(read_cr3());
-
-	/* Find the indexes of the physical address to identity map */
-	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
-	level2_index = address / LEVEL1_SIZE;
-
-	/* Identity map the page table entry */
-	pgtable_level1[level1_index] = address | L0_ATTR;
-	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
-
-	/* Flush the tlb so the new mapping takes effect.
-	 * Global tlb entries are not flushed but that is not an issue.
-	 */
-	load_cr3(pgtable_level2);
-}
-
-#else
-#define LEVEL1_SIZE (1UL << 21UL)
-#define LEVEL2_SIZE (1UL << 30UL)
-static u64 pgtable_level1[512] PAGE_ALIGNED;
-static u64 pgtable_level2[512] PAGE_ALIGNED;
-
-static void identity_map_page(unsigned long address)
-{
-	unsigned long level1_index, level2_index, level3_index;
-	u64 *pgtable_level3;
-
-	/* Find the current page table */
-	pgtable_level3 = __va(read_cr3());
-
-	/* Find the indexes of the physical address to identity map */
-	level1_index = (address % LEVEL1_SIZE)/LEVEL0_SIZE;
-	level2_index = (address % LEVEL2_SIZE)/LEVEL1_SIZE;
-	level3_index = address / LEVEL2_SIZE;
-
-	/* Identity map the page table entry */
-	pgtable_level1[level1_index] = address | L0_ATTR;
-	pgtable_level2[level2_index] = __pa(pgtable_level1) | L1_ATTR;
-	set_64bit(&pgtable_level3[level3_index],
-					       __pa(pgtable_level2) | L2_ATTR);
-
-	/* Flush the tlb so the new mapping takes effect.
-	 * Global tlb entries are not flushed but that is not an issue.
-	 */
-	load_cr3(pgtable_level3);
-}
+static u32 kexec_pgd[1024] PAGE_ALIGNED;
+#ifdef CONFIG_X86_PAE
+static u32 kexec_pmd0[1024] PAGE_ALIGNED;
+static u32 kexec_pmd1[1024] PAGE_ALIGNED;
 #endif
+static u32 kexec_pte0[1024] PAGE_ALIGNED;
+static u32 kexec_pte1[1024] PAGE_ALIGNED;
 
 static void set_idt(void *newidt, __u16 limit)
 {
@@ -128,16 +71,6 @@ static void load_segments(void)
 #undef __STR
 }
 
-typedef asmlinkage NORET_TYPE void (*relocate_new_kernel_t)(
-					unsigned long indirection_page,
-					unsigned long reboot_code_buffer,
-					unsigned long start_address,
-					unsigned int has_pae) ATTRIB_NORET;
-
-extern const unsigned char relocate_new_kernel[];
-extern void relocate_new_kernel_end(void);
-extern const unsigned int relocate_new_kernel_size;
-
 /*
  * A architecture hook called to validate the
  * proposed image and prepare the control pages
@@ -170,25 +103,29 @@ void machine_kexec_cleanup(struct kimage *image)
  */
 NORET_TYPE void machine_kexec(struct kimage *image)
 {
-	unsigned long page_list;
-	unsigned long reboot_code_buffer;
-
-	relocate_new_kernel_t rnk;
+	unsigned long page_list[PAGES_NR];
+	void *control_page;
 
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
-	/* Compute some offsets */
-	reboot_code_buffer = page_to_pfn(image->control_code_page)
-								<< PAGE_SHIFT;
-	page_list = image->head;
-
-	/* Set up an identity mapping for the reboot_code_buffer */
-	identity_map_page(reboot_code_buffer);
-
-	/* copy it out */
-	memcpy((void *)reboot_code_buffer, relocate_new_kernel,
-						relocate_new_kernel_size);
+	control_page = page_address(image->control_code_page);
+	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+
+	page_list[PA_CONTROL_PAGE] = __pa(control_page);
+	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[PA_PGD] = __pa(kexec_pgd);
+	page_list[VA_PGD] = (unsigned long)kexec_pgd;
+#ifdef CONFIG_X86_PAE
+	page_list[PA_PMD_0] = __pa(kexec_pmd0);
+	page_list[VA_PMD_0] = (unsigned long)kexec_pmd0;
+	page_list[PA_PMD_1] = __pa(kexec_pmd1);
+	page_list[VA_PMD_1] = (unsigned long)kexec_pmd1;
+#endif
+	page_list[PA_PTE_0] = __pa(kexec_pte0);
+	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
+	page_list[PA_PTE_1] = __pa(kexec_pte1);
+	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -207,8 +144,8 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	set_idt(phys_to_virt(0),0);
 
 	/* now call it */
-	rnk = (relocate_new_kernel_t) reboot_code_buffer;
-	(*rnk)(page_list, reboot_code_buffer, image->start, cpu_has_pae);
+	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
+			image->start, cpu_has_pae);
 }
 
 /* crashkernel=size@addr specifies the location to reserve for
diff --git a/arch/i386/kernel/relocate_kernel.S b/arch/i386/kernel/relocate_kernel.S
index d312616effa1..f151d6fae462 100644
--- a/arch/i386/kernel/relocate_kernel.S
+++ b/arch/i386/kernel/relocate_kernel.S
@@ -7,16 +7,138 @@
  */
 
 #include <linux/linkage.h>
+#include <asm/page.h>
+#include <asm/kexec.h>
+
+/*
+ * Must be relocatable PIC code callable as a C function
+ */
+
+#define PTR(x) (x << 2)
+#define PAGE_ALIGNED (1 << PAGE_SHIFT)
+#define PAGE_ATTR 0x63 /* _PAGE_PRESENT|_PAGE_RW|_PAGE_ACCESSED|_PAGE_DIRTY */
+#define PAE_PGD_ATTR 0x01 /* _PAGE_PRESENT */
+
+	.text
+	.align PAGE_ALIGNED
+	.globl relocate_kernel
+relocate_kernel:
+	movl	8(%esp), %ebp /* list of pages */
+
+#ifdef CONFIG_X86_PAE
+	/* map the control page at its virtual address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xc0000000, %eax
+	shrl	$27, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PMD_0)(%ebp), %edx
+	orl	$PAE_PGD_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PMD_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x3fe00000, %eax
+	shrl	$18, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_0)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x001ff000, %eax
+	shrl	$9, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	/* identity map the control page at its physical address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xc0000000, %eax
+	shrl	$27, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PMD_1)(%ebp), %edx
+	orl	$PAE_PGD_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PMD_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x3fe00000, %eax
+	shrl	$18, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_1)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x001ff000, %eax
+	shrl	$9, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+#else
+	/* map the control page at its virtual address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xffc00000, %eax
+	shrl	$20, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_0)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_0)(%ebp), %edi
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x003ff000, %eax
+	shrl	$10, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	/* identity map the control page at its physical address */
+
+	movl	PTR(VA_PGD)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0xffc00000, %eax
+	shrl	$20, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_PTE_1)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+
+	movl	PTR(VA_PTE_1)(%ebp), %edi
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %eax
+	andl	$0x003ff000, %eax
+	shrl	$10, %eax
+	addl	%edi, %eax
+
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edx
+	orl	$PAGE_ATTR, %edx
+	movl	%edx, (%eax)
+#endif
 
-	/*
-	 * Must be relocatable PIC code callable as a C function, that once
-	 * it starts can not use the previous processes stack.
-	 */
-	.globl relocate_new_kernel
 relocate_new_kernel:
 	/* read the arguments and say goodbye to the stack */
 	movl  4(%esp), %ebx /* page_list */
-	movl  8(%esp), %ebp /* reboot_code_buffer */
+	movl  8(%esp), %ebp /* list of pages */
 	movl  12(%esp), %edx /* start address */
 	movl  16(%esp), %ecx /* cpu_has_pae */
 
@@ -24,11 +146,26 @@ relocate_new_kernel:
 	pushl $0
 	popfl
 
-	/* set a new stack at the bottom of our page... */
-	lea   4096(%ebp), %esp
+	/* get physical address of control page now */
+	/* this is impossible after page table switch */
+	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edi
 
-	/* store the parameters back on the stack */
-	pushl   %edx /* store the start address */
+	/* switch to new set of page tables */
+	movl	PTR(PA_PGD)(%ebp), %eax
+	movl	%eax, %cr3
+
+	/* setup a new stack at the end of the physical control page */
+	lea	4096(%edi), %esp
+
+	/* jump to identity mapped page */
+	movl    %edi, %eax
+	addl    $(identity_mapped - relocate_kernel), %eax
+	pushl   %eax
+	ret
+
+identity_mapped:
+	/* store the start address on the stack */
+	pushl   %edx
 
 	/* Set cr0 to a known state:
 	 * 31 0 == Paging disabled
@@ -113,8 +250,3 @@ relocate_new_kernel:
 	xorl    %edi, %edi
 	xorl    %ebp, %ebp
 	ret
-relocate_new_kernel_end:
-
-	.globl relocate_new_kernel_size
-relocate_new_kernel_size:
-	.long relocate_new_kernel_end - relocate_new_kernel
diff --git a/include/asm-i386/kexec.h b/include/asm-i386/kexec.h
index 53f0e06672dc..4dfc9f5ed031 100644
--- a/include/asm-i386/kexec.h
+++ b/include/asm-i386/kexec.h
@@ -1,6 +1,26 @@
 #ifndef _I386_KEXEC_H
 #define _I386_KEXEC_H
 
+#define PA_CONTROL_PAGE  0
+#define VA_CONTROL_PAGE  1
+#define PA_PGD           2
+#define VA_PGD           3
+#define PA_PTE_0         4
+#define VA_PTE_0         5
+#define PA_PTE_1         6
+#define VA_PTE_1         7
+#ifdef CONFIG_X86_PAE
+#define PA_PMD_0         8
+#define VA_PMD_0         9
+#define PA_PMD_1         10
+#define VA_PMD_1         11
+#define PAGES_NR         12
+#else
+#define PAGES_NR         8
+#endif
+
+#ifndef __ASSEMBLY__
+
 #include <asm/fixmap.h>
 #include <asm/ptrace.h>
 #include <asm/string.h>
@@ -72,5 +92,12 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
                newregs->eip = (unsigned long)current_text_addr();
        }
 }
+asmlinkage NORET_TYPE void
+relocate_kernel(unsigned long indirection_page,
+		unsigned long control_page,
+		unsigned long start_address,
+		unsigned int has_pae) ATTRIB_NORET;
+
+#endif /* __ASSEMBLY__ */
 
 #endif /* _I386_KEXEC_H */
-- 
cgit v1.2.1


From e8c7391de4cd91b2cbb0c791f69ba1e066595848 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Don't use kernel_text_address in oops context

Because it can take spinlocks.

Suggested by Mathieu Desnoyers

Cc: Mathieu Desnoyers <compudj@krystal.dyndns.org>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/traps.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index fb8486eca1b6..01f2a8d254c2 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -299,7 +299,9 @@ void dump_trace(struct task_struct *tsk, struct pt_regs *regs, unsigned long * s
 #define HANDLE_STACK(cond) \
 	do while (cond) { \
 		unsigned long addr = *stack++; \
-		if (kernel_text_address(addr)) { \
+		if (oops_in_progress ? 		\
+			__kernel_text_address(addr) : \
+			kernel_text_address(addr)) { \
 			/* \
 			 * If the address is either in the text segment of the \
 			 * kernel, or in the region which contains vmalloc'ed \
-- 
cgit v1.2.1


From 027a51cef330433ba5924fc92fb25ef48bcfc1a3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Document my tree in Documentation/HOWTO

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/HOWTO | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/HOWTO b/Documentation/HOWTO
index 915ae8c986c6..1d6560413cc5 100644
--- a/Documentation/HOWTO
+++ b/Documentation/HOWTO
@@ -358,7 +358,8 @@ Here is a list of some of the different kernel trees available:
   quilt trees:
     - USB, PCI, Driver Core, and I2C, Greg Kroah-Hartman <gregkh@suse.de>
 	kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/
-
+    - x86-64, partly i386, Andi Kleen <ak@suse.de>
+        ftp.firstfloor.org:/pub/ak/x86_64/quilt/
 
 Bug Reporting
 -------------
-- 
cgit v1.2.1


From 29a9af60e2120f874d0c600bf9e27617254a0488 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Add comments to the PDA structure to annotate offsets

Change the comments in the pda structure to make the first fields to have
their offset documented and to have the comments aligned.
The stack protector series needs a field at offset 40 (gcc ABI); annotate
upto 40 for that reason.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index e2b23e337b94..6794ffaae433 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -9,13 +9,14 @@
 
 /* Per processor datastructure. %gs points to it while the kernel runs */ 
 struct x8664_pda {
-	struct task_struct *pcurrent;	/* Current process */
-	unsigned long data_offset;	/* Per cpu data offset from linker address */
-	unsigned long kernelstack;  /* top of kernel stack for current */ 
-	unsigned long oldrsp; 	    /* user rsp for system call */
-        int irqcount;		    /* Irq nesting counter. Starts with -1 */  	
-	int cpunumber;		    /* Logical CPU number */
-	char *irqstackptr;	/* top of irqstack */
+	struct task_struct *pcurrent;	/* 0  Current process */
+	unsigned long data_offset;	/* 8 Per cpu data offset from linker
+					   address */
+	unsigned long kernelstack;  /* 16 top of kernel stack for current */
+	unsigned long oldrsp; 	    /* 24 user rsp for system call */
+        int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
+	int cpunumber;		    /* 36 Logical CPU number */
+	char *irqstackptr;	/* 40 top of irqstack */
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
-- 
cgit v1.2.1


From b62a5c740df1e3d49a97349fce0c6a23f633d7fe Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Add the Kconfig option for the stackprotector feature

This patch adds the config options for -fstack-protector.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 arch/x86_64/Kconfig | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 927b392a4aa0..0c61d0019dd7 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -533,6 +533,30 @@ config SECCOMP
 
 	  If unsure, say Y. Only embedded should say N here.
 
+config CC_STACKPROTECTOR
+	bool "Enable -fstack-protector buffer overflow detection (EXPRIMENTAL)"
+	depends on EXPERIMENTAL
+	help
+         This option turns on the -fstack-protector GCC feature. This
+	  feature puts, at the beginning of critical functions, a canary
+	  value on the stack just before the return address, and validates
+	  the value just before actually returning.  Stack based buffer
+	  overflows (that need to overwrite this return address) now also
+	  overwrite the canary, which gets detected and the attack is then
+	  neutralized via a kernel panic.
+
+	  This feature requires gcc version 4.2 or above, or a distribution
+	  gcc with the feature backported. Older versions are automatically
+	  detected and for those versions, this configuration option is ignored.
+
+config CC_STACKPROTECTOR_ALL
+	bool "Use stack-protector for all functions"
+	depends on CC_STACKPROTECTOR
+	help
+	  Normally, GCC only inserts the canary value protection for
+	  functions that use large-ish on-stack buffers. By enabling
+	  this option, GCC will be asked to do this for ALL functions.
+
 source kernel/Kconfig.hz
 
 config REORDER
-- 
cgit v1.2.1


From 0a4254058037eb172758961d0a5b94f4320a1425 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:38 +0200
Subject: [PATCH] Add the canary field to the PDA area and the task struct

This patch adds the per thread cookie field to the task struct and the PDA.
Also it makes sure that the PDA value gets the new cookie value at context
switch, and that a new task gets a new cookie at task creation time.

Signed-off-by: Arjan van Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/process.c | 8 ++++++++
 include/asm-x86_64/pda.h     | 7 ++++++-
 include/linux/sched.h        | 5 +++++
 kernel/fork.c                | 5 +++++
 4 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 9e9a70e50c72..fba8dfeda67c 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -625,6 +625,14 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	unlazy_fpu(prev_p);
 	write_pda(kernelstack,
 		  task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+#ifdef CONFIG_CC_STACKPROTECTOR
+	write_pda(stack_canary, next_p->stack_canary);
+	/*
+	 * Build time only check to make sure the stack_canary is at
+	 * offset 40 in the pda; this is a gcc ABI requirement
+	 */
+	BUILD_BUG_ON(offsetof(struct x8664_pda, stack_canary) != 40);
+#endif
 
 	/*
 	 * Now maybe reload the debug registers and handle I/O bitmaps
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 6794ffaae433..e7773e0af865 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -16,7 +16,12 @@ struct x8664_pda {
 	unsigned long oldrsp; 	    /* 24 user rsp for system call */
         int irqcount;		    /* 32 Irq nesting counter. Starts with -1 */
 	int cpunumber;		    /* 36 Logical CPU number */
-	char *irqstackptr;	/* 40 top of irqstack */
+#ifdef CONFIG_CC_STACKPROTECTOR
+	unsigned long stack_canary;	/* 40 stack canary value */
+					/* gcc-ABI: this canary MUST be at
+					   offset 40!!! */
+#endif
+	char *irqstackptr;
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 807556c5bcd2..9d4aa7f95bc8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -819,6 +819,11 @@ struct task_struct {
 	unsigned did_exec:1;
 	pid_t pid;
 	pid_t tgid;
+
+#ifdef CONFIG_CC_STACKPROTECTOR
+	/* Canary value for the -fstack-protector gcc feature */
+	unsigned long stack_canary;
+#endif
 	/* 
 	 * pointers to (original) parent process, youngest child, younger sibling,
 	 * older sibling, respectively.  (p->father can be replaced with 
diff --git a/kernel/fork.c b/kernel/fork.c
index f9b014e3e700..a0dad84567c9 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -45,6 +45,7 @@
 #include <linux/cn_proc.h>
 #include <linux/delayacct.h>
 #include <linux/taskstats_kern.h>
+#include <linux/random.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -175,6 +176,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
 	tsk->thread_info = ti;
 	setup_thread_stack(tsk, orig);
 
+#ifdef CONFIG_CC_STACKPROTECTOR
+	tsk->stack_canary = get_random_int();
+#endif
+
 	/* One for us, one for whoever does the "release_task()" (usually parent) */
 	atomic_set(&tsk->usage,2);
 	atomic_set(&tsk->fs_excl, 0);
-- 
cgit v1.2.1


From 3162f751d04086a9d006342de63ac8f44fe0f72a Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Add the __stack_chk_fail() function

GCC emits a call to a __stack_chk_fail() function when the stack canary is
not matching the expected value.

Since this is a bad security issue; lets panic the kernel rather than limping
along; the kernel really can't be trusted anymore when this happens.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
---
 kernel/panic.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/kernel/panic.c b/kernel/panic.c
index d2db3e2209e0..6ceb664fb52a 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -271,3 +271,15 @@ void oops_exit(void)
 {
 	do_oops_enter_exit();
 }
+
+#ifdef CONFIG_CC_STACKPROTECTOR
+/*
+ * Called when gcc's -fstack-protector feature is used, and
+ * gcc detects corruption of the on-stack canary value
+ */
+void __stack_chk_fail(void)
+{
+	panic("stack-protector: Kernel stack is corrupted");
+}
+EXPORT_SYMBOL(__stack_chk_fail);
+#endif
-- 
cgit v1.2.1


From 4f7fd4d7a79193ceda4ce77f75e22917d33fa154 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Add the -fstack-protector option to the CFLAGS

Add a feature check that checks that the gcc compiler has stack-protector
support and has the bugfix for PR28281 to make this work in kernel mode.
The easiest solution I could find was to have a shell script in scripts/
to do the detection; if needed we can make this fancier in the future
without making the makefile too complex.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Andi Kleen <ak@suse.de>
CC: Sam Ravnborg <sam@ravnborg.org>
---
 arch/x86_64/Makefile                      | 3 +++
 scripts/gcc-x86_64-has-stack-protector.sh | 6 ++++++
 2 files changed, 9 insertions(+)
 create mode 100644 scripts/gcc-x86_64-has-stack-protector.sh

diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index d6472ddf5f6e..2b8d07c70106 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -58,6 +58,9 @@ cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
 cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
+cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector )
+cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all )
+
 CFLAGS += $(cflags-y)
 CFLAGS_KERNEL += $(cflags-kernel-y)
 AFLAGS += -m64
diff --git a/scripts/gcc-x86_64-has-stack-protector.sh b/scripts/gcc-x86_64-has-stack-protector.sh
new file mode 100644
index 000000000000..325c0a1b03b6
--- /dev/null
+++ b/scripts/gcc-x86_64-has-stack-protector.sh
@@ -0,0 +1,6 @@
+#!/bin/sh
+
+echo "int foo(void) { char X[200]; return 3; }" | $1 -S -xc -c -O0 -mcmodel=kernel -fstack-protector - -o - 2> /dev/null | grep -q "%gs"
+if [ "$?" -eq "0" ] ; then
+	echo $2
+fi
-- 
cgit v1.2.1


From 96e540492ab54423f3693958329e095878f1f12b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Fix a irqcount comment in entry.S

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 4fcc0ad8bbeb..ea32688386fd 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -519,7 +519,12 @@ END(stub_rt_sigreturn)
 	testl $3,CS(%rdi)
 	je 1f
 	swapgs	
-1:	incl	%gs:pda_irqcount	# RED-PEN should check preempt count
+	/* irqcount is used to check if a CPU is already on an interrupt
+	   stack or not. While this is essentially redundant with preempt_count
+	   it is a little cheaper to use a separate counter in the PDA
+	   (short of moving irq_enter into assembly, which would be too
+	    much work) */
+1:	incl	%gs:pda_irqcount
 	cmoveq %gs:pda_irqstackptr,%rsp
 	push    %rbp			# backlink for old unwinder
 	/*
-- 
cgit v1.2.1


From baf5695dd1a49bb48a3daf08726d7f243f42e97e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Use %c instead of %P modifier in pda access

Apparently that is the more official way to get numbers without $ in inline
assembly

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index e7773e0af865..c2aac96ac323 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -51,13 +51,13 @@ extern struct x8664_pda _proxy_pda;
 	typedef typeof(_proxy_pda.field) T__; \
        switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm(op "w %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 4: \
-asm(op "l %1,%%gs:%P2" : "+m" (_proxy_pda.field) : \
+asm(op "l %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
 case 8: \
-asm(op "q %1,%%gs:%P2": "+m" (_proxy_pda.field) : \
+asm(op "q %1,%%gs:%c2": "+m" (_proxy_pda.field) : \
 	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
 default: __bad_pda_field(); 					\
        } \
@@ -67,13 +67,13 @@ default: __bad_pda_field(); 					\
        typeof(_proxy_pda.field) ret__; \
        switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
-asm(op "w %%gs:%P1,%0":"=r" (ret__):\
+asm(op "w %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 4: \
-asm(op "l %%gs:%P1,%0":"=r" (ret__):\
+asm(op "l %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 case 8: \
-asm(op "q %%gs:%P1,%0":"=r" (ret__):\
+asm(op "q %%gs:%c1,%0":"=r" (ret__):\
 	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
 default: __bad_pda_field(); 					\
        } \
-- 
cgit v1.2.1


From 85691f135db78f3548107a0abe383dfab3bc38fa Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Type checking for write_pda()

I just added type checking for assignments the PDA in the i386 PDA code.
Here's the x86-64 equivalent.  (Obviously this doesn't contain the latest
x86-64 PDA change.)

Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index c2aac96ac323..5dadb201f769 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -49,7 +49,8 @@ extern struct x8664_pda _proxy_pda;
 
 #define pda_to_op(op,field,val) do { \
 	typedef typeof(_proxy_pda.field) T__; \
-       switch (sizeof(_proxy_pda.field)) { 		\
+	if (0) { T__ tmp__; tmp__ = (val); }  \
+	switch (sizeof(_proxy_pda.field)) { 		\
 case 2: \
 asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
 	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-- 
cgit v1.2.1


From 7b0bda74f7e77f362eaeee837e7911238acf4c76 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Fix a PDA warning uncovered by the new type checking

Fix
linux/arch/x86_64/kernel/process.c: In function __switch_to:
linux/arch/x86_64/kernel/process.c:626: warning: assignment makes integer from pointer without a cast

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index fba8dfeda67c..885c318f76ab 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -624,7 +624,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
 	   And the AMD workaround requires it to be after DS reload. */
 	unlazy_fpu(prev_p);
 	write_pda(kernelstack,
-		  task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
+	(unsigned long)task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
 #ifdef CONFIG_CC_STACKPROTECTOR
 	write_pda(stack_canary, next_p->stack_canary);
 	/*
-- 
cgit v1.2.1


From 0da5db313317e3195482d3e660a1074857374a89 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: Abstract sensitive instructions

Abstract sensitive instructions in assembler code, replacing them with macros
(which currently are #defined to the native versions).  We use long names:
assembler is case-insensitive, so if something goes wrong and macros do not
expand, it would assemble anyway.

Resulting object files are exactly the same as before.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S    | 38 ++++++++++++++++++++++----------------
 include/asm-i386/spinlock.h |  7 +++++--
 2 files changed, 27 insertions(+), 18 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 4b0845249222..3872fca5c74a 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -76,8 +76,15 @@ DF_MASK		= 0x00000400
 NT_MASK		= 0x00004000
 VM_MASK		= 0x00020000
 
+/* These are replaces for paravirtualization */
+#define DISABLE_INTERRUPTS		cli
+#define ENABLE_INTERRUPTS		sti
+#define ENABLE_INTERRUPTS_SYSEXIT	sti; sysexit
+#define INTERRUPT_RETURN		iret
+#define GET_CR0_INTO_EAX		movl %cr0, %eax
+
 #ifdef CONFIG_PREEMPT
-#define preempt_stop		cli; TRACE_IRQS_OFF
+#define preempt_stop		DISABLE_INTERRUPTS; TRACE_IRQS_OFF
 #else
 #define preempt_stop
 #define resume_kernel		restore_nocheck
@@ -236,7 +243,7 @@ check_userspace:
 	testl $(VM_MASK | 3), %eax
 	jz resume_kernel
 ENTRY(resume_userspace)
- 	cli				# make sure we don't miss an interrupt
+ 	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	movl TI_flags(%ebp), %ecx
@@ -247,7 +254,7 @@ ENTRY(resume_userspace)
 
 #ifdef CONFIG_PREEMPT
 ENTRY(resume_kernel)
-	cli
+	DISABLE_INTERRUPTS
 	cmpl $0,TI_preempt_count(%ebp)	# non-zero preempt_count ?
 	jnz restore_nocheck
 need_resched:
@@ -275,7 +282,7 @@ sysenter_past_esp:
 	 * No need to follow this irqs on/off section: the syscall
 	 * disabled irqs and here we enable it straight after entry:
 	 */
-	sti
+	ENABLE_INTERRUPTS
 	pushl $(__USER_DS)
 	CFI_ADJUST_CFA_OFFSET 4
 	/*CFI_REL_OFFSET ss, 0*/
@@ -320,7 +327,7 @@ sysenter_past_esp:
 	jae syscall_badsys
 	call *sys_call_table(,%eax,4)
 	movl %eax,EAX(%esp)
-	cli
+	DISABLE_INTERRUPTS
 	TRACE_IRQS_OFF
 	movl TI_flags(%ebp), %ecx
 	testw $_TIF_ALLWORK_MASK, %cx
@@ -330,8 +337,7 @@ sysenter_past_esp:
 	movl OLDESP(%esp), %ecx
 	xorl %ebp,%ebp
 	TRACE_IRQS_ON
-	sti
-	sysexit
+	ENABLE_INTERRUPTS_SYSEXIT
 	CFI_ENDPROC
 
 
@@ -356,7 +362,7 @@ syscall_call:
 	call *sys_call_table(,%eax,4)
 	movl %eax,EAX(%esp)		# store the return value
 syscall_exit:
-	cli				# make sure we don't miss an interrupt
+	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	TRACE_IRQS_OFF
@@ -381,11 +387,11 @@ restore_nocheck_notrace:
 	RESTORE_REGS
 	addl $4, %esp
 	CFI_ADJUST_CFA_OFFSET -4
-1:	iret
+1:	INTERRUPT_RETURN
 .section .fixup,"ax"
 iret_exc:
 	TRACE_IRQS_ON
-	sti
+	ENABLE_INTERRUPTS
 	pushl $0			# no error code
 	pushl $do_iret_error
 	jmp error_code
@@ -409,7 +415,7 @@ ldt_ss:
 	 * dosemu and wine happy. */
 	subl $8, %esp		# reserve space for switch16 pointer
 	CFI_ADJUST_CFA_OFFSET 8
-	cli
+	DISABLE_INTERRUPTS
 	TRACE_IRQS_OFF
 	movl %esp, %eax
 	/* Set up the 16bit stack frame with switch32 pointer on top,
@@ -419,7 +425,7 @@ ldt_ss:
 	TRACE_IRQS_IRET
 	RESTORE_REGS
 	lss 20+4(%esp), %esp	# switch to 16bit stack
-1:	iret
+1:	INTERRUPT_RETURN
 .section __ex_table,"a"
 	.align 4
 	.long 1b,iret_exc
@@ -434,7 +440,7 @@ work_pending:
 	jz work_notifysig
 work_resched:
 	call schedule
-	cli				# make sure we don't miss an interrupt
+	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
 					# between sampling and the iret
 	TRACE_IRQS_OFF
@@ -490,7 +496,7 @@ syscall_exit_work:
 	testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl
 	jz work_pending
 	TRACE_IRQS_ON
-	sti				# could let do_syscall_trace() call
+	ENABLE_INTERRUPTS		# could let do_syscall_trace() call
 					# schedule() instead
 	movl %esp, %eax
 	movl $1, %edx
@@ -668,7 +674,7 @@ ENTRY(device_not_available)
 	pushl $-1			# mark this as an int
 	CFI_ADJUST_CFA_OFFSET 4
 	SAVE_ALL
-	movl %cr0, %eax
+	GET_CR0_INTO_EAX
 	testl $0x4, %eax		# EM (math emulation bit)
 	jne device_not_available_emulate
 	preempt_stop
@@ -811,7 +817,7 @@ nmi_16bit_stack:
 	call do_nmi
 	RESTORE_REGS
 	lss 12+4(%esp), %esp		# back to 16bit stack
-1:	iret
+1:	INTERRUPT_RETURN
 	CFI_ENDPROC
 .section __ex_table,"a"
 	.align 4
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index 324329313af8..b0b3043f05e1 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -7,6 +7,9 @@
 #include <asm/processor.h>
 #include <linux/compiler.h>
 
+#define CLI_STRING	"cli"
+#define STI_STRING	"sti"
+
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
  *
@@ -55,12 +58,12 @@ static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long fla
 		"2:\t"
 		"testl $0x200, %1\n\t"
 		"jz 4f\n\t"
-		"sti\n"
+		STI_STRING "\n"
 		"3:\t"
 		"rep;nop\n\t"
 		"cmpb $0, %0\n\t"
 		"jle 3b\n\t"
-		"cli\n\t"
+		CLI_STRING "\n\t"
 		"jmp 1b\n"
 		"4:\t"
 		"rep;nop\n\t"
-- 
cgit v1.2.1


From 78be3706b21a232310590fe00258b224177ac05f Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: Allow a kernel not to be in ring 0

We allow for the fact that the guest kernel may not run in ring 0.  This
requires some abstraction in a few places when setting %cs or checking
privilege level (user vs kernel).

This is Chris' [RFC PATCH 15/33] move segment checks to subarch, except rather
than using #define USER_MODE_MASK which depends on a config option, we use
Zach's more flexible approach of assuming ring 3 == userspace.  I also used
"get_kernel_rpl()" over "get_kernel_cs()" because I think it reads better in
the code...

1) Remove the hardcoded 3 and introduce #define SEGMENT_RPL_MASK 3 2) Add a
get_kernel_rpl() macro, and don't assume it's zero.

And:

Clean up of patch for letting kernel run other than ring 0:

a. Add some comments about the SEGMENT_IS_*_CODE() macros.
b. Add a USER_RPL macro.  (Code was comparing a value to a mask
   in some places and to the magic number 3 in other places.)
c. Add macros for table indicator field and use them.
d. Change the entry.S tests for LDT stack segment to use the macros

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/entry.S   |  9 +++++----
 arch/i386/kernel/process.c |  2 +-
 arch/i386/mm/extable.c     |  2 +-
 arch/i386/mm/fault.c       | 11 ++++-------
 include/asm-i386/ptrace.h  |  5 +++--
 include/asm-i386/segment.h | 17 +++++++++++++++++
 6 files changed, 31 insertions(+), 15 deletions(-)

diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 3872fca5c74a..284f2e908ad0 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -240,8 +240,9 @@ ret_from_intr:
 check_userspace:
 	movl EFLAGS(%esp), %eax		# mix EFLAGS and CS
 	movb CS(%esp), %al
-	testl $(VM_MASK | 3), %eax
-	jz resume_kernel
+	andl $(VM_MASK | SEGMENT_RPL_MASK), %eax
+	cmpl $USER_RPL, %eax
+	jb resume_kernel		# not returning to v8086 or userspace
 ENTRY(resume_userspace)
  	DISABLE_INTERRUPTS		# make sure we don't miss an interrupt
 					# setting need_resched or sigpending
@@ -377,8 +378,8 @@ restore_all:
 	# See comments in process.c:copy_thread() for details.
 	movb OLDSS(%esp), %ah
 	movb CS(%esp), %al
-	andl $(VM_MASK | (4 << 8) | 3), %eax
-	cmpl $((4 << 8) | 3), %eax
+	andl $(VM_MASK | (SEGMENT_TI_MASK << 8) | SEGMENT_RPL_MASK), %eax
+	cmpl $((SEGMENT_LDT << 8) | USER_RPL), %eax
 	CFI_REMEMBER_STATE
 	je ldt_ss			# returning to user-space with LDT SS
 restore_nocheck:
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 220aeca59c3a..8c190ca7ae44 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -338,7 +338,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 	regs.xes = __USER_DS;
 	regs.orig_eax = -1;
 	regs.eip = (unsigned long) kernel_thread_helper;
-	regs.xcs = __KERNEL_CS;
+	regs.xcs = __KERNEL_CS | get_kernel_rpl();
 	regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
 
 	/* Ok, create the new process.. */
diff --git a/arch/i386/mm/extable.c b/arch/i386/mm/extable.c
index de03c5430abc..0ce4f22a2635 100644
--- a/arch/i386/mm/extable.c
+++ b/arch/i386/mm/extable.c
@@ -11,7 +11,7 @@ int fixup_exception(struct pt_regs *regs)
 	const struct exception_table_entry *fixup;
 
 #ifdef CONFIG_PNPBIOS
-	if (unlikely((regs->xcs & ~15) == (GDT_ENTRY_PNPBIOS_BASE << 3)))
+	if (unlikely(SEGMENT_IS_PNP_CODE(regs->xcs)))
 	{
 		extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp;
 		extern u32 pnp_bios_is_utter_crap;
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index 0ce86168a0b1..5e17a3f43b41 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -27,6 +27,7 @@
 #include <asm/uaccess.h>
 #include <asm/desc.h>
 #include <asm/kdebug.h>
+#include <asm/segment.h>
 
 extern void die(const char *,struct pt_regs *,long);
 
@@ -113,10 +114,10 @@ static inline unsigned long get_segment_eip(struct pt_regs *regs,
 	}
 
 	/* The standard kernel/user address space limit. */
-	*eip_limit = (seg & 3) ? USER_DS.seg : KERNEL_DS.seg;
+	*eip_limit = user_mode(regs) ? USER_DS.seg : KERNEL_DS.seg;
 	
 	/* By far the most common cases. */
-	if (likely(seg == __USER_CS || seg == __KERNEL_CS))
+	if (likely(SEGMENT_IS_FLAT_CODE(seg)))
 		return eip;
 
 	/* Check the segment exists, is within the current LDT/GDT size,
@@ -430,11 +431,7 @@ good_area:
 	write = 0;
 	switch (error_code & 3) {
 		default:	/* 3: write, present */
-#ifdef TEST_VERIFY_AREA
-			if (regs->cs == KERNEL_CS)
-				printk("WP fault at %08lx\n", regs->eip);
-#endif
-			/* fall through */
+				/* fall through */
 		case 2:		/* write, not present */
 			if (!(vma->vm_flags & VM_WRITE))
 				goto bad_area;
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index 30a442ec2059..21bb91679c82 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -60,6 +60,7 @@ struct pt_regs {
 #ifdef __KERNEL__
 
 #include <asm/vm86.h>
+#include <asm/segment.h>
 
 struct task_struct;
 extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
@@ -73,11 +74,11 @@ extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int erro
  */
 static inline int user_mode(struct pt_regs *regs)
 {
-	return (regs->xcs & 3) != 0;
+	return (regs->xcs & SEGMENT_RPL_MASK) == USER_RPL;
 }
 static inline int user_mode_vm(struct pt_regs *regs)
 {
-	return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0;
+	return ((regs->xcs & SEGMENT_RPL_MASK) | (regs->eflags & VM_MASK)) >= USER_RPL;
 }
 #define instruction_pointer(regs) ((regs)->eip)
 extern unsigned long profile_pc(struct pt_regs *regs);
diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h
index faf995307b9e..b7ab59685ba7 100644
--- a/include/asm-i386/segment.h
+++ b/include/asm-i386/segment.h
@@ -83,6 +83,11 @@
 
 #define GDT_SIZE (GDT_ENTRIES * 8)
 
+/* Matches __KERNEL_CS and __USER_CS (they must be 2 entries apart) */
+#define SEGMENT_IS_FLAT_CODE(x)  (((x) & 0xec) == GDT_ENTRY_KERNEL_CS * 8)
+/* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */
+#define SEGMENT_IS_PNP_CODE(x)   (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8)
+
 /* Simple and small GDT entries for booting only */
 
 #define GDT_ENTRY_BOOT_CS		2
@@ -112,4 +117,16 @@
  */
 #define IDT_ENTRIES 256
 
+/* Bottom two bits of selector give the ring privilege level */
+#define SEGMENT_RPL_MASK	0x3
+/* Bit 2 is table indicator (LDT/GDT) */
+#define SEGMENT_TI_MASK		0x4
+
+/* User mode is privilege level 3 */
+#define USER_RPL		0x3
+/* LDT segment has TI set, GDT has it cleared */
+#define SEGMENT_LDT		0x4
+#define SEGMENT_GDT		0x0
+
+#define get_kernel_rpl()  0
 #endif
-- 
cgit v1.2.1


From ec5c09269ba9cd3b9cf879390db6d5c7dcdcaf1e Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: Do better early exception handlers

Add early i386 fault handlers with debug information for common faults.
Handles:

        divide error
        invalid opcode
        protection fault
        page fault

Also adds code to detect early recursive/multiple faults and halt the
system when they happen (taken from x86_64.)

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@muc.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
 arch/i386/kernel/head.S | 67 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 67 insertions(+)

diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index a6b8bd89aa27..be9d883c62ce 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -371,8 +371,65 @@ rp_sidt:
 	addl $8,%edi
 	dec %ecx
 	jne rp_sidt
+
+.macro	set_early_handler handler,trapno
+	lea \handler,%edx
+	movl $(__KERNEL_CS << 16),%eax
+	movw %dx,%ax
+	movw $0x8E00,%dx	/* interrupt gate - dpl=0, present */
+	lea idt_table,%edi
+	movl %eax,8*\trapno(%edi)
+	movl %edx,8*\trapno+4(%edi)
+.endm
+
+	set_early_handler handler=early_divide_err,trapno=0
+	set_early_handler handler=early_illegal_opcode,trapno=6
+	set_early_handler handler=early_protection_fault,trapno=13
+	set_early_handler handler=early_page_fault,trapno=14
+
 	ret
 
+early_divide_err:
+	xor %edx,%edx
+	pushl $0	/* fake errcode */
+	jmp early_fault
+
+early_illegal_opcode:
+	movl $6,%edx
+	pushl $0	/* fake errcode */
+	jmp early_fault
+
+early_protection_fault:
+	movl $13,%edx
+	jmp early_fault
+
+early_page_fault:
+	movl $14,%edx
+	jmp early_fault
+
+early_fault:
+	cld
+#ifdef CONFIG_PRINTK
+	movl $(__KERNEL_DS),%eax
+	movl %eax,%ds
+	movl %eax,%es
+	cmpl $2,early_recursion_flag
+	je hlt_loop
+	incl early_recursion_flag
+	movl %cr2,%eax
+	pushl %eax
+	pushl %edx		/* trapno */
+	pushl $fault_msg
+#ifdef CONFIG_EARLY_PRINTK
+	call early_printk
+#else
+	call printk
+#endif
+#endif
+hlt_loop:
+	hlt
+	jmp hlt_loop
+
 /* This is the default interrupt "handler" :-) */
 	ALIGN
 ignore_int:
@@ -386,6 +443,9 @@ ignore_int:
 	movl $(__KERNEL_DS),%eax
 	movl %eax,%ds
 	movl %eax,%es
+	cmpl $2,early_recursion_flag
+	je hlt_loop
+	incl early_recursion_flag
 	pushl 16(%esp)
 	pushl 24(%esp)
 	pushl 32(%esp)
@@ -431,9 +491,16 @@ ENTRY(stack_start)
 
 ready:	.byte 0
 
+early_recursion_flag:
+	.long 0
+
 int_msg:
 	.asciz "Unknown interrupt or fault at EIP %p %p %p\n"
 
+fault_msg:
+	.ascii "Int %d: CR2 %p  err %p  EIP %p  CS %p  flags %p\n"
+	.asciz "Stack: %p %p %p %p %p %p %p %p\n"
+
 /*
  * The IDT and GDT 'descriptors' are a strange 48-bit object
  * only used by the lidt and lgdt instructions. They are not
-- 
cgit v1.2.1


From 54dbc0c9ebefb38840c6b07fa6eabaeb96c921f5 Mon Sep 17 00:00:00 2001
From: "adurbin@google.com" <adurbin@google.com>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] insert IOAPIC(s) and Local APIC into resource map

This patch places the IOAPIC(s) and the Local APIC specified by ACPI
tables into the resource map. The APICs will then be visible within
/proc/iomem

Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/apic.c | 54 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 54 insertions(+)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 6472e321cad7..135ff25e6b44 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -25,6 +25,7 @@
 #include <linux/kernel_stat.h>
 #include <linux/sysdev.h>
 #include <linux/module.h>
+#include <linux/ioport.h>
 
 #include <asm/atomic.h>
 #include <asm/smp.h>
@@ -45,6 +46,11 @@ int apic_calibrate_pmtmr __initdata;
 
 int disable_apic_timer __initdata;
 
+static struct resource lapic_resource = {
+	.name = "Local APIC",
+	.flags = IORESOURCE_MEM | IORESOURCE_BUSY,
+};
+
 /*
  * cpu_mask that denotes the CPUs that needs timer interrupt coming in as
  * IPIs in place of local APIC timers
@@ -585,6 +591,40 @@ static int __init detect_init_APIC (void)
 	return 0;
 }
 
+#ifdef CONFIG_X86_IO_APIC
+static struct resource * __init ioapic_setup_resources(void)
+{
+#define IOAPIC_RESOURCE_NAME_SIZE 11
+	unsigned long n;
+	struct resource *res;
+	char *mem;
+	int i;
+
+	if (nr_ioapics <= 0)
+		return NULL;
+
+	n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource);
+	n *= nr_ioapics;
+
+	res = alloc_bootmem(n);
+
+	if (!res)
+		return NULL;
+
+	memset(res, 0, n);
+	mem = (void *)&res[nr_ioapics];
+
+	for (i = 0; i < nr_ioapics; i++) {
+		res[i].name = mem;
+		res[i].flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+		snprintf(mem, IOAPIC_RESOURCE_NAME_SIZE, "IOAPIC %u", i);
+		mem += IOAPIC_RESOURCE_NAME_SIZE;
+	}
+
+	return res;
+}
+#endif
+
 void __init init_apic_mappings(void)
 {
 	unsigned long apic_phys;
@@ -604,6 +644,11 @@ void __init init_apic_mappings(void)
 	apic_mapped = 1;
 	apic_printk(APIC_VERBOSE,"mapped APIC to %16lx (%16lx)\n", APIC_BASE, apic_phys);
 
+	/* Put local APIC into the resource map. */
+	lapic_resource.start = apic_phys;
+	lapic_resource.end = lapic_resource.start + PAGE_SIZE - 1;
+	insert_resource(&iomem_resource, &lapic_resource);
+
 	/*
 	 * Fetch the APIC ID of the BSP in case we have a
 	 * default configuration (or the MP table is broken).
@@ -613,7 +658,9 @@ void __init init_apic_mappings(void)
 	{
 		unsigned long ioapic_phys, idx = FIX_IO_APIC_BASE_0;
 		int i;
+		struct resource *ioapic_res;
 
+		ioapic_res = ioapic_setup_resources();
 		for (i = 0; i < nr_ioapics; i++) {
 			if (smp_found_config) {
 				ioapic_phys = mp_ioapics[i].mpc_apicaddr;
@@ -625,6 +672,13 @@ void __init init_apic_mappings(void)
 			apic_printk(APIC_VERBOSE,"mapped IOAPIC to %016lx (%016lx)\n",
 					__fix_to_virt(idx), ioapic_phys);
 			idx++;
+
+			if (ioapic_res) {
+				ioapic_res->start = ioapic_phys;
+				ioapic_res->end = ioapic_phys + (4 * 1024) - 1;
+				insert_resource(&iomem_resource, ioapic_res);
+				ioapic_res++;
+			}
 		}
 	}
 }
-- 
cgit v1.2.1


From f0f4c3432e5e1087b3a8c0e6bd4113d3c37497ff Mon Sep 17 00:00:00 2001
From: "adurbin@google.com" <adurbin@google.com>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] i386: add HPET(s) into resource map

Add HPET(s) into resource map. This will allow for the HPET(s) to be
visibile within /proc/iomem.

Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/boot.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 8a6c5b412348..1aaea6ab8c46 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -30,6 +30,8 @@
 #include <linux/module.h>
 #include <linux/dmi.h>
 #include <linux/irq.h>
+#include <linux/bootmem.h>
+#include <linux/ioport.h>
 
 #include <asm/pgtable.h>
 #include <asm/io_apic.h>
@@ -646,6 +648,8 @@ static int __init acpi_parse_sbf(unsigned long phys_addr, unsigned long size)
 static int __init acpi_parse_hpet(unsigned long phys, unsigned long size)
 {
 	struct acpi_table_hpet *hpet_tbl;
+	struct resource *hpet_res;
+	resource_size_t res_start;
 
 	if (!phys || !size)
 		return -EINVAL;
@@ -661,12 +665,26 @@ static int __init acpi_parse_hpet(unsigned long phys, unsigned long size)
 		       "memory.\n");
 		return -1;
 	}
+
+#define HPET_RESOURCE_NAME_SIZE 9
+	hpet_res = alloc_bootmem(sizeof(*hpet_res) + HPET_RESOURCE_NAME_SIZE);
+	if (hpet_res) {
+		memset(hpet_res, 0, sizeof(*hpet_res));
+		hpet_res->name = (void *)&hpet_res[1];
+		hpet_res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+		snprintf((char *)hpet_res->name, HPET_RESOURCE_NAME_SIZE,
+			 "HPET %u", hpet_tbl->number);
+		hpet_res->end = (1 * 1024) - 1;
+	}
+
 #ifdef	CONFIG_X86_64
 	vxtime.hpet_address = hpet_tbl->addr.addrl |
 	    ((long)hpet_tbl->addr.addrh << 32);
 
 	printk(KERN_INFO PREFIX "HPET id: %#x base: %#lx\n",
 	       hpet_tbl->id, vxtime.hpet_address);
+
+	res_start = vxtime.hpet_address;
 #else				/* X86 */
 	{
 		extern unsigned long hpet_address;
@@ -674,9 +692,17 @@ static int __init acpi_parse_hpet(unsigned long phys, unsigned long size)
 		hpet_address = hpet_tbl->addr.addrl;
 		printk(KERN_INFO PREFIX "HPET id: %#x base: %#lx\n",
 		       hpet_tbl->id, hpet_address);
+
+		res_start = hpet_address;
 	}
 #endif				/* X86 */
 
+	if (hpet_res) {
+		hpet_res->start = res_start;
+		hpet_res->end += res_start;
+		insert_resource(&iomem_resource, hpet_res);
+	}
+
 	return 0;
 }
 #else
-- 
cgit v1.2.1


From 3022d734a54cbd2b65eea9a024564821101b4a9a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Fix zeroing on exception in copy_*_user

- Don't zero for __copy_from_user_inatomic following i386.
This will prevent spurious zeros for parallel file system writers when
one does a exception
- The string instruction version didn't zero the output on
exception. Oops.

Also I cleaned up the code a bit while I was at it and added a minor
optimization to the string instruction path.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/x8664_ksyms.c |   1 +
 arch/x86_64/lib/copy_user.S      | 124 +++++++++++++++++++++++++--------------
 include/asm-x86_64/uaccess.h     |   6 +-
 3 files changed, 83 insertions(+), 48 deletions(-)

diff --git a/arch/x86_64/kernel/x8664_ksyms.c b/arch/x86_64/kernel/x8664_ksyms.c
index 370952c4ff22..c3454af5e3a2 100644
--- a/arch/x86_64/kernel/x8664_ksyms.c
+++ b/arch/x86_64/kernel/x8664_ksyms.c
@@ -29,6 +29,7 @@ EXPORT_SYMBOL(__put_user_8);
 EXPORT_SYMBOL(copy_user_generic);
 EXPORT_SYMBOL(copy_from_user);
 EXPORT_SYMBOL(copy_to_user);
+EXPORT_SYMBOL(__copy_from_user_inatomic);
 
 EXPORT_SYMBOL(copy_page);
 EXPORT_SYMBOL(clear_page);
diff --git a/arch/x86_64/lib/copy_user.S b/arch/x86_64/lib/copy_user.S
index 962f3a693c5e..70bebd310408 100644
--- a/arch/x86_64/lib/copy_user.S
+++ b/arch/x86_64/lib/copy_user.S
@@ -9,10 +9,29 @@
 
 #define FIX_ALIGNMENT 1
 
-	#include <asm/current.h>
-	#include <asm/asm-offsets.h>
-	#include <asm/thread_info.h>
-	#include <asm/cpufeature.h>
+#include <asm/current.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+#include <asm/cpufeature.h>
+
+	.macro ALTERNATIVE_JUMP feature,orig,alt
+0:
+	.byte 0xe9	/* 32bit jump */
+	.long \orig-1f	/* by default jump to orig */
+1:
+	.section .altinstr_replacement,"ax"
+2:	.byte 0xe9	             /* near jump with 32bit immediate */
+	.long \alt-1b /* offset */   /* or alternatively to alt */
+	.previous
+	.section .altinstructions,"a"
+	.align 8
+	.quad  0b
+	.quad  2b
+	.byte  \feature		     /* when feature is set */
+	.byte  5
+	.byte  5
+	.previous
+	.endm
 
 /* Standard copy_to_user with segment limit checking */		
 ENTRY(copy_to_user)
@@ -23,25 +42,21 @@ ENTRY(copy_to_user)
 	jc  bad_to_user
 	cmpq threadinfo_addr_limit(%rax),%rcx
 	jae bad_to_user
-2:
-	.byte 0xe9	/* 32bit jump */
-	.long .Lcug-1f
-1:
+	xorl %eax,%eax	/* clear zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
 	CFI_ENDPROC
-ENDPROC(copy_to_user)
 
-	.section .altinstr_replacement,"ax"
-3:	.byte 0xe9			/* replacement jmp with 32 bit immediate */
-	.long copy_user_generic_c-1b	/* offset */
-	.previous
-	.section .altinstructions,"a"
-	.align 8
-	.quad  2b
-	.quad  3b
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  5
-	.byte  5
-	.previous
+ENTRY(copy_user_generic)
+	CFI_STARTPROC
+	movl $1,%ecx	/* set zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
+	CFI_ENDPROC
+
+ENTRY(__copy_from_user_inatomic)
+	CFI_STARTPROC
+	xorl %ecx,%ecx	/* clear zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
+	CFI_ENDPROC
 
 /* Standard copy_from_user with segment limit checking */	
 ENTRY(copy_from_user)
@@ -52,7 +67,8 @@ ENTRY(copy_from_user)
 	jc  bad_from_user
 	cmpq threadinfo_addr_limit(%rax),%rcx
 	jae  bad_from_user
-	/* FALL THROUGH to copy_user_generic */
+	movl $1,%ecx	/* set zero flag */
+	ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string
 	CFI_ENDPROC
 ENDPROC(copy_from_user)
 	
@@ -73,37 +89,26 @@ END(bad_from_user)
 	
 		
 /*
- * copy_user_generic - memory copy with exception handling.
+ * copy_user_generic_unrolled - memory copy with exception handling.
+ * This version is for CPUs like P4 that don't have efficient micro code for rep movsq
  * 	
  * Input:	
  * rdi destination
  * rsi source
  * rdx count
+ * ecx zero flag -- if true zero destination on error
  *
  * Output:		
  * eax uncopied bytes or 0 if successful.
  */
-ENTRY(copy_user_generic)
+ENTRY(copy_user_generic_unrolled)
 	CFI_STARTPROC
-	.byte 0x66,0x66,0x90	/* 5 byte nop for replacement jump */
-	.byte 0x66,0x90
-1:
-	.section .altinstr_replacement,"ax"
-2:	.byte 0xe9	             /* near jump with 32bit immediate */
-	.long copy_user_generic_c-1b /* offset */
-	.previous
-	.section .altinstructions,"a"
-	.align 8
-	.quad  copy_user_generic
-	.quad  2b
-	.byte  X86_FEATURE_REP_GOOD
-	.byte  5
-	.byte  5
-	.previous
-.Lcug:
 	pushq %rbx
 	CFI_ADJUST_CFA_OFFSET 8
 	CFI_REL_OFFSET rbx, 0
+	pushq %rcx
+	CFI_ADJUST_CFA_OFFSET 8
+	CFI_REL_OFFSET rcx, 0
 	xorl %eax,%eax		/*zero for the exception handler */
 
 #ifdef FIX_ALIGNMENT
@@ -179,6 +184,9 @@ ENTRY(copy_user_generic)
 
 	CFI_REMEMBER_STATE
 .Lende:
+	popq %rcx
+	CFI_ADJUST_CFA_OFFSET -8
+	CFI_RESTORE rcx
 	popq %rbx
 	CFI_ADJUST_CFA_OFFSET -8
 	CFI_RESTORE rbx
@@ -265,6 +273,8 @@ ENTRY(copy_user_generic)
 	addl %ecx,%edx
 	/* edx: bytes to zero, rdi: dest, eax:zero */
 .Lzero_rest:
+	cmpl $0,(%rsp)
+	jz   .Le_zero
 	movq %rdx,%rcx
 .Le_byte:
 	xorl %eax,%eax
@@ -286,6 +296,7 @@ ENDPROC(copy_user_generic)
  /* rdi	destination
   * rsi source
   * rdx count
+  * ecx zero flag
   *
   * Output:
   * eax uncopied bytes or 0 if successfull.
@@ -296,25 +307,48 @@ ENDPROC(copy_user_generic)
   * And more would be dangerous because both Intel and AMD have
   * errata with rep movsq > 4GB. If someone feels the need to fix
   * this please consider this.
-   */
-copy_user_generic_c:
+  */
+ENTRY(copy_user_generic_string)
 	CFI_STARTPROC
+	movl %ecx,%r8d		/* save zero flag */
 	movl %edx,%ecx
 	shrl $3,%ecx
 	andl $7,%edx	
+	jz   10f
 1:	rep 
 	movsq 
 	movl %edx,%ecx
 2:	rep
 	movsb
-4:	movl %ecx,%eax
+9:	movl %ecx,%eax
 	ret
-3:	lea (%rdx,%rcx,8),%rax
+
+	/* multiple of 8 byte */
+10:	rep
+	movsq
+	xor %eax,%eax
 	ret
+
+	/* exception handling */
+3:      lea (%rdx,%rcx,8),%rax	/* exception on quad loop */
+	jmp 6f
+5:	movl %ecx,%eax		/* exception on byte loop */
+	/* eax: left over bytes */
+6:	testl %r8d,%r8d		/* zero flag set? */
+	jz 7f
+	movl %eax,%ecx		/* initialize x86 loop counter */
+	push %rax
+	xorl %eax,%eax
+8:	rep
+	stosb 			/* zero the rest */
+11:	pop %rax
+7:	ret
 	CFI_ENDPROC
 END(copy_user_generic_c)
 
 	.section __ex_table,"a"
 	.quad 1b,3b
-	.quad 2b,4b
+	.quad 2b,5b
+	.quad 8b,11b
+	.quad 10b,3b
 	.previous
diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index 1e1fa003daa3..bc6812009bd7 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -238,6 +238,7 @@ do {									\
 
 /* Handles exceptions in both to and from, but doesn't do access_ok */
 extern unsigned long copy_user_generic(void *to, const void *from, unsigned len); 
+extern unsigned long copy_user_generic_dontzero(void *to, const void *from, unsigned len);
 
 extern unsigned long copy_to_user(void __user *to, const void *from, unsigned len); 
 extern unsigned long copy_from_user(void *to, const void __user *from, unsigned len); 
@@ -303,7 +304,6 @@ static __always_inline int __copy_to_user(void __user *dst, const void *src, uns
 	}
 }	
 
-
 static __always_inline int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
@@ -352,7 +352,7 @@ long strlen_user(const char __user *str);
 unsigned long clear_user(void __user *mem, unsigned long len);
 unsigned long __clear_user(void __user *mem, unsigned long len);
 
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
+extern long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
+#define __copy_to_user_inatomic copy_user_generic
 
 #endif /* __X86_64_UACCESS_H */
-- 
cgit v1.2.1


From 95912008ba1fb9d0677c1ce5930aeb0e85ba5710 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Add __must_check to copy_*_user

Following i386.

And also fix the two occurrences that caused warnings in arch/x86_64/*

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ptrace32.c  |  6 ++++--
 include/asm-x86_64/i387.h    |  4 ++--
 include/asm-x86_64/uaccess.h | 46 ++++++++++++++++++++++++++------------------
 3 files changed, 33 insertions(+), 23 deletions(-)

diff --git a/arch/x86_64/ia32/ptrace32.c b/arch/x86_64/ia32/ptrace32.c
index 72bf92a9d375..d18198ed636b 100644
--- a/arch/x86_64/ia32/ptrace32.c
+++ b/arch/x86_64/ia32/ptrace32.c
@@ -375,8 +375,10 @@ asmlinkage long sys32_ptrace(long request, u32 pid, u32 addr, u32 data)
 		ret = -EIO;
 		if (!access_ok(VERIFY_READ, u, sizeof(*u)))
 			break;
-		/* no checking to be bug-to-bug compatible with i386 */
-		__copy_from_user(&child->thread.i387.fxsave, u, sizeof(*u));
+		/* no checking to be bug-to-bug compatible with i386. */
+		/* but silence warning */
+		if (__copy_from_user(&child->thread.i387.fxsave, u, sizeof(*u)))
+			;
 		set_stopped_child_used_math(child);
 		child->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;
 		ret = 0; 
diff --git a/include/asm-x86_64/i387.h b/include/asm-x86_64/i387.h
index 60c0f4853fdb..0217b74cc9fc 100644
--- a/include/asm-x86_64/i387.h
+++ b/include/asm-x86_64/i387.h
@@ -137,8 +137,8 @@ static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)
 #else
 		     : [fx] "cdaSDb" (fx), "0" (0));
 #endif
-	if (unlikely(err))
-		__clear_user(fx, sizeof(struct i387_fxsave_struct));
+	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))
+		err = -EFAULT;
 	/* No need to clear here because the caller clears USED_MATH */
 	return err;
 } 
diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index bc6812009bd7..802a4a068ef6 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -237,14 +237,18 @@ do {									\
  */
 
 /* Handles exceptions in both to and from, but doesn't do access_ok */
-extern unsigned long copy_user_generic(void *to, const void *from, unsigned len); 
-extern unsigned long copy_user_generic_dontzero(void *to, const void *from, unsigned len);
-
-extern unsigned long copy_to_user(void __user *to, const void *from, unsigned len); 
-extern unsigned long copy_from_user(void *to, const void __user *from, unsigned len); 
-extern unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len); 
-
-static __always_inline int __copy_from_user(void *dst, const void __user *src, unsigned size)
+__must_check unsigned long
+copy_user_generic(void *to, const void *from, unsigned len);
+
+__must_check unsigned long
+copy_to_user(void __user *to, const void *from, unsigned len);
+__must_check unsigned long
+copy_from_user(void *to, const void __user *from, unsigned len);
+__must_check unsigned long
+copy_in_user(void __user *to, const void __user *from, unsigned len);
+
+static __always_inline __must_check
+int __copy_from_user(void *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -273,7 +277,8 @@ static __always_inline int __copy_from_user(void *dst, const void __user *src, u
 	}
 }	
 
-static __always_inline int __copy_to_user(void __user *dst, const void *src, unsigned size)
+static __always_inline __must_check
+int __copy_to_user(void __user *dst, const void *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -304,7 +309,8 @@ static __always_inline int __copy_to_user(void __user *dst, const void *src, uns
 	}
 }	
 
-static __always_inline int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
+static __always_inline __must_check
+int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
        int ret = 0;
 	if (!__builtin_constant_p(size))
@@ -344,15 +350,17 @@ static __always_inline int __copy_in_user(void __user *dst, const void __user *s
 	}
 }	
 
-long strncpy_from_user(char *dst, const char __user *src, long count);
-long __strncpy_from_user(char *dst, const char __user *src, long count);
-long strnlen_user(const char __user *str, long n);
-long __strnlen_user(const char __user *str, long n);
-long strlen_user(const char __user *str);
-unsigned long clear_user(void __user *mem, unsigned long len);
-unsigned long __clear_user(void __user *mem, unsigned long len);
-
-extern long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
+__must_check long 
+strncpy_from_user(char *dst, const char __user *src, long count);
+__must_check long 
+__strncpy_from_user(char *dst, const char __user *src, long count);
+__must_check long strnlen_user(const char __user *str, long n);
+__must_check long __strnlen_user(const char __user *str, long n);
+__must_check long strlen_user(const char __user *str);
+__must_check unsigned long clear_user(void __user *mem, unsigned long len);
+__must_check unsigned long __clear_user(void __user *mem, unsigned long len);
+
+__must_check long __copy_from_user_inatomic(void *dst, const void __user *src, unsigned size);
 #define __copy_to_user_inatomic copy_user_generic
 
 #endif /* __X86_64_UACCESS_H */
-- 
cgit v1.2.1


From 758333458aa719bfc26ec16eafd4ad3a9e96014d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:39 +0200
Subject: [PATCH] Check return value of copy_to_user in compat_sys_pselect7

Fix

linux/fs/compat.c: In function compat_sys_pselect7
linux/fs/compat.c:1869: warning: ignoring return value of copy_to_user, declared with attribute warn_unused_result

To make it easier to handle I changed to semantics to not try to
write out a timespec if an error occurred. I hope that's ok.

Cc: dwmw2@infradead.org

Signed-off-by: Andi Kleen <ak@suse.de>
---
 fs/compat.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/compat.c b/fs/compat.c
index e31e9cf96647..ce982f6e8c80 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1855,7 +1855,7 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp,
 
 	} while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec));
 
-	if (tsp && !(current->personality & STICKY_TIMEOUTS)) {
+	if (ret == 0 && tsp && !(current->personality & STICKY_TIMEOUTS)) {
 		struct compat_timespec rts;
 
 		rts.tv_sec = timeout / HZ;
@@ -1866,7 +1866,8 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp,
 		}
 		if (compat_timespec_compare(&rts, &ts) >= 0)
 			rts = ts;
-		copy_to_user(tsp, &rts, sizeof(rts));
+		if (copy_to_user(tsp, &rts, sizeof(rts)))
+			ret = -EFAULT;
 	}
 
 	if (ret == -ERESTARTNOHAND) {
-- 
cgit v1.2.1


From 26c13f2b5bbb03f798f8907db20296347e6c7ca6 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Check return values of __copy_to_user in uname emulation

Quietens some new warnings
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/sys_ia32.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c
index 33cd0a4fe388..b0e82c7947dc 100644
--- a/arch/x86_64/ia32/sys_ia32.c
+++ b/arch/x86_64/ia32/sys_ia32.c
@@ -780,7 +780,7 @@ asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len,
 
 asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
 {
-	int error;
+	int err;
 
 	if (!name)
 		return -EFAULT;
@@ -789,27 +789,31 @@ asmlinkage long sys32_olduname(struct oldold_utsname __user * name)
   
   	down_read(&uts_sem);
 	
-	error = __copy_to_user(&name->sysname,&system_utsname.sysname,__OLD_UTS_LEN);
-	 __put_user(0,name->sysname+__OLD_UTS_LEN);
-	 __copy_to_user(&name->nodename,&system_utsname.nodename,__OLD_UTS_LEN);
-	 __put_user(0,name->nodename+__OLD_UTS_LEN);
-	 __copy_to_user(&name->release,&system_utsname.release,__OLD_UTS_LEN);
-	 __put_user(0,name->release+__OLD_UTS_LEN);
-	 __copy_to_user(&name->version,&system_utsname.version,__OLD_UTS_LEN);
-	 __put_user(0,name->version+__OLD_UTS_LEN);
+	err = __copy_to_user(&name->sysname,&system_utsname.sysname,
+				__OLD_UTS_LEN);
+	err |= __put_user(0,name->sysname+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->nodename,&system_utsname.nodename,
+				__OLD_UTS_LEN);
+	err |= __put_user(0,name->nodename+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->release,&system_utsname.release,
+				__OLD_UTS_LEN);
+	err |= __put_user(0,name->release+__OLD_UTS_LEN);
+	err |= __copy_to_user(&name->version,&system_utsname.version,
+				__OLD_UTS_LEN);
+	err |= __put_user(0,name->version+__OLD_UTS_LEN);
 	 { 
 		 char *arch = "x86_64";
 		 if (personality(current->personality) == PER_LINUX32)
 			 arch = "i686";
 		 
-		 __copy_to_user(&name->machine,arch,strlen(arch)+1);
+		 err |= __copy_to_user(&name->machine,arch,strlen(arch)+1);
 	 }
 	
 	 up_read(&uts_sem);
 	 
-	 error = error ? -EFAULT : 0;
+	 err = err ? -EFAULT : 0;
 	 
-	 return error;
+	 return err;
 }
 
 long sys32_uname(struct old_utsname __user * name)
-- 
cgit v1.2.1


From 383d079bfdfcfccd6a720405a510fe325b3e6576 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Fix some stylistic issues in uaccess.h

- Replace some broken white space.
- Replace __ keywords with standard names

No functional changes.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/uaccess.h | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/include/asm-x86_64/uaccess.h b/include/asm-x86_64/uaccess.h
index 802a4a068ef6..e856570c0689 100644
--- a/include/asm-x86_64/uaccess.h
+++ b/include/asm-x86_64/uaccess.h
@@ -84,7 +84,7 @@ struct exception_table_entry
  */
 
 #define __get_user_x(size,ret,x,ptr) \
-	__asm__ __volatile__("call __get_user_" #size \
+	asm volatile("call __get_user_" #size \
 		:"=a" (ret),"=d" (x) \
 		:"c" (ptr) \
 		:"r8")
@@ -101,7 +101,7 @@ struct exception_table_entry
 	case 8:  __get_user_x(8,__ret_gu,__val_gu,ptr); break;		\
 	default: __get_user_bad(); break;				\
 	}								\
-	(x) = (__typeof__(*(ptr)))__val_gu;				\
+	(x) = (typeof(*(ptr)))__val_gu;				\
 	__ret_gu;							\
 })
 
@@ -112,7 +112,7 @@ extern void __put_user_8(void);
 extern void __put_user_bad(void);
 
 #define __put_user_x(size,ret,x,ptr)					\
-	__asm__ __volatile__("call __put_user_" #size			\
+	asm volatile("call __put_user_" #size			\
 		:"=a" (ret)						\
 		:"c" (ptr),"d" (x)					\
 		:"r8")
@@ -139,7 +139,7 @@ extern void __put_user_bad(void);
 #define __put_user_check(x,ptr,size)			\
 ({							\
 	int __pu_err;					\
-	__typeof__(*(ptr)) __user *__pu_addr = (ptr);	\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);	\
 	switch (size) { 				\
 	case 1: __put_user_x(1,__pu_err,x,__pu_addr); break;	\
 	case 2: __put_user_x(2,__pu_err,x,__pu_addr); break;	\
@@ -173,7 +173,7 @@ struct __large_struct { unsigned long buf[100]; };
  * aliasing issues.
  */
 #define __put_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	__asm__ __volatile__(					\
+	asm volatile(					\
 		"1:	mov"itype" %"rtype"1,%2\n"		\
 		"2:\n"						\
 		".section .fixup,\"ax\"\n"			\
@@ -193,7 +193,7 @@ struct __large_struct { unsigned long buf[100]; };
 	int __gu_err;						\
 	unsigned long __gu_val;					\
 	__get_user_size(__gu_val,(ptr),(size),__gu_err);	\
-	(x) = (__typeof__(*(ptr)))__gu_val;			\
+	(x) = (typeof(*(ptr)))__gu_val;			\
 	__gu_err;						\
 })
 
@@ -217,7 +217,7 @@ do {									\
 } while (0)
 
 #define __get_user_asm(x, addr, err, itype, rtype, ltype, errno)	\
-	__asm__ __volatile__(					\
+	asm volatile(					\
 		"1:	mov"itype" %2,%"rtype"1\n"		\
 		"2:\n"						\
 		".section .fixup,\"ax\"\n"			\
@@ -250,7 +250,7 @@ copy_in_user(void __user *to, const void __user *from, unsigned len);
 static __always_inline __must_check
 int __copy_from_user(void *dst, const void __user *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic(dst,(__force void *)src,size);
 	switch (size) { 
@@ -280,7 +280,7 @@ int __copy_from_user(void *dst, const void __user *src, unsigned size)
 static __always_inline __must_check
 int __copy_to_user(void __user *dst, const void *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic((__force void *)dst,src,size);
 	switch (size) { 
@@ -312,7 +312,7 @@ int __copy_to_user(void __user *dst, const void *src, unsigned size)
 static __always_inline __must_check
 int __copy_in_user(void __user *dst, const void __user *src, unsigned size)
 { 
-       int ret = 0;
+	int ret = 0;
 	if (!__builtin_constant_p(size))
 		return copy_user_generic((__force void *)dst,(__force void *)src,size);
 	switch (size) { 
-- 
cgit v1.2.1


From c1a9d41f4f103bfef2ed0bea1e95b3190e39e448 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Reindent macros in pda.h

Reindent the macros in x86-64 pda.h, making them much more readable.
Follows Jeremy's i386 version of this.

No functional changes

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 85 ++++++++++++++++++++++++++++++------------------
 1 file changed, 53 insertions(+), 32 deletions(-)

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 5dadb201f769..9e3aaf74475d 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -41,46 +41,67 @@ extern struct x8664_pda boot_cpu_pda[];
  */ 
 extern void __bad_pda_field(void);
 
-/* proxy_pda doesn't actually exist, but tell gcc it is accessed
-   for all PDA accesses so it gets read/write dependencies right. */
+/*
+ * proxy_pda doesn't actually exist, but tell gcc it is accessed for
+ * all PDA accesses so it gets read/write dependencies right.
+ */
 extern struct x8664_pda _proxy_pda;
 
 #define pda_offset(field) offsetof(struct x8664_pda, field)
 
-#define pda_to_op(op,field,val) do { \
-	typedef typeof(_proxy_pda.field) T__; \
-	if (0) { T__ tmp__; tmp__ = (val); }  \
-	switch (sizeof(_proxy_pda.field)) { 		\
-case 2: \
-asm(op "w %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
-	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-case 4: \
-asm(op "l %1,%%gs:%c2" : "+m" (_proxy_pda.field) : \
-	"ri" ((T__)val),"i"(pda_offset(field))); break; \
-case 8: \
-asm(op "q %1,%%gs:%c2": "+m" (_proxy_pda.field) : \
-	 "ri" ((T__)val),"i"(pda_offset(field))); break; \
-default: __bad_pda_field(); 					\
-       } \
+#define pda_to_op(op,field,val) do {		\
+	typedef typeof(_proxy_pda.field) T__;	\
+	if (0) { T__ tmp__; tmp__ = (val); }	/* type checking */ \
+	switch (sizeof(_proxy_pda.field)) {	\
+	case 2:					\
+		asm(op "w %1,%%gs:%c2" : 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i"(pda_offset(field))); 	\
+ 		break;				\
+	case 4:					\
+		asm(op "l %1,%%gs:%c2" : 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i" (pda_offset(field))); 	\
+		break;				\
+	case 8:					\
+		asm(op "q %1,%%gs:%c2": 	\
+		    "+m" (_proxy_pda.field) :	\
+		    "ri" ((T__)val),		\
+		    "i"(pda_offset(field))); 	\
+		break;				\
+       default: 				\
+		__bad_pda_field();		\
+       }					\
        } while (0)
 
-#define pda_from_op(op,field) ({ \
-       typeof(_proxy_pda.field) ret__; \
-       switch (sizeof(_proxy_pda.field)) { 		\
-case 2: \
-asm(op "w %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-case 4: \
-asm(op "l %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-case 8: \
-asm(op "q %%gs:%c1,%0":"=r" (ret__):\
-	"i" (pda_offset(field)), "m" (_proxy_pda.field)); break;\
-default: __bad_pda_field(); 					\
-       } \
+#define pda_from_op(op,field) ({		\
+	typeof(_proxy_pda.field) ret__;		\
+	switch (sizeof(_proxy_pda.field)) {	\
+       	case 2:					\
+		asm(op "w %%gs:%c1,%0" : 	\
+		    "=r" (ret__) :		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+	case 4:					\
+		asm(op "l %%gs:%c1,%0":		\
+		    "=r" (ret__):		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+       case 8:					\
+		asm(op "q %%gs:%c1,%0":		\
+		    "=r" (ret__) :		\
+		    "i" (pda_offset(field)), 	\
+		    "m" (_proxy_pda.field)); 	\
+		 break;				\
+       default: 				\
+		__bad_pda_field();		\
+       }					\
        ret__; })
 
-
 #define read_pda(field) pda_from_op("mov",field)
 #define write_pda(field,val) pda_to_op("mov",field,val)
 #define add_pda(field,val) pda_to_op("add",field,val)
-- 
cgit v1.2.1


From fd167e42b237e0688005b3dec380eb5a6e5f3585 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Define __bad_pda_field as noreturn

This quietens so warnings about uninitialized use of the return
value of the pda read operations.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/pda.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 9e3aaf74475d..531f48a6c3af 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -39,7 +39,7 @@ extern struct x8664_pda boot_cpu_pda[];
  * There is no fast way to get the base address of the PDA, all the accesses
  * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
  */ 
-extern void __bad_pda_field(void);
+extern void __bad_pda_field(void) __attribute__((noreturn));
 
 /*
  * proxy_pda doesn't actually exist, but tell gcc it is accessed for
-- 
cgit v1.2.1


From 73bb5117a448bdf0b56232ca28451fe4c534cb3a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Remove unused asm-x86_64/mmx.h

Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/mmx.h | 14 --------------
 1 file changed, 14 deletions(-)
 delete mode 100644 include/asm-x86_64/mmx.h

diff --git a/include/asm-x86_64/mmx.h b/include/asm-x86_64/mmx.h
deleted file mode 100644
index 46b71da99869..000000000000
--- a/include/asm-x86_64/mmx.h
+++ /dev/null
@@ -1,14 +0,0 @@
-#ifndef _ASM_MMX_H
-#define _ASM_MMX_H
-
-/*
- *	MMX 3Dnow! helper operations
- */
-
-#include <linux/types.h>
- 
-extern void *_mmx_memcpy(void *to, const void *from, size_t size);
-extern void mmx_clear_page(void *page);
-extern void mmx_copy_page(void *to, void *from);
-
-#endif
-- 
cgit v1.2.1


From b38337a624c4d3c2c3d9cdf27d952ca94181c6a8 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Mark per cpu data initialization __initdata again

Before 2.6.16 this was changed to work around code that accessed
CPUs not in the possible map. But that code should be all fixed now,
so mark it __initdata again.
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/vmlinux.lds.S | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/arch/x86_64/kernel/vmlinux.lds.S b/arch/x86_64/kernel/vmlinux.lds.S
index 2802aa963fa3..d0564f1bcb0b 100644
--- a/arch/x86_64/kernel/vmlinux.lds.S
+++ b/arch/x86_64/kernel/vmlinux.lds.S
@@ -216,14 +216,12 @@ SECTIONS
   __initramfs_start = .;
   .init.ramfs : AT(ADDR(.init.ramfs) - LOAD_OFFSET) { *(.init.ramfs) }
   __initramfs_end = .;
-  /* temporary here to work around NR_CPUS. If you see this comment in 2.6.17+
-   complain */
-  . = ALIGN(4096);	
-  __init_end = .;	
-  . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
+    . = ALIGN(CONFIG_X86_L1_CACHE_BYTES);
   __per_cpu_start = .;
   .data.percpu  : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) }
   __per_cpu_end = .;
+  . = ALIGN(4096);
+  __init_end = .;
 
   . = ALIGN(4096);
   __nosave_begin = .;
-- 
cgit v1.2.1


From 371c2f2783eb880df333d0e9ac35c78143e79d0a Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386: Remove experimental mark of kexec

kexec has been marked experimental for a year now and all
of the serious problems have been worked through.  So it
is time (if not past time) to remove the experimental mark.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Kconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index e34a4dbd6b92..9b6cddc8c51b 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -739,8 +739,7 @@ config SECCOMP
 source kernel/Kconfig.hz
 
 config KEXEC
-	bool "kexec system call (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	bool "kexec system call"
 	help
 	  kexec is a system call that implements the ability to shutdown your
 	  current kernel, and to start another kernel.  It is like a reboot
-- 
cgit v1.2.1


From 1c9c0a6ca35e9325cea811d734d6ab7352be086b Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Remove experimental mark of kexec

kexec has been marked experimental for a year now and all
of the serious problems have been worked through.  So it
is time (if not past time) to remove the experimental mark.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/Kconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 0c61d0019dd7..1f4212605ef8 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -468,8 +468,7 @@ config X86_MCE_AMD
 	   the DRAM Error Threshold.
 
 config KEXEC
-	bool "kexec system call (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	bool "kexec system call"
 	help
 	  kexec is a system call that implements the ability to shutdown your
 	  current kernel, and to start another kernel.  It is like a reboot
-- 
cgit v1.2.1


From a15da49debaf7f09460a886b0ecd08588410715e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Fix idle notifiers

Previously exit_idle would be called more often than enter_idle

Now instead of using complicated tests just keep track of it
using the per CPU variable as a flip flop.  I moved the idle state into the
PDA to make the access more efficient.

Original bug report and an initial patch from Stephane Eranian,
but redone by AK.

Cc: Stephane Eranian <eranian@hpl.hp.com>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/process.c | 15 +++++++++------
 include/asm-x86_64/pda.h     |  3 ++-
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c
index 885c318f76ab..458006ae19f3 100644
--- a/arch/x86_64/kernel/process.c
+++ b/arch/x86_64/kernel/process.c
@@ -80,25 +80,25 @@ void idle_notifier_unregister(struct notifier_block *n)
 }
 EXPORT_SYMBOL(idle_notifier_unregister);
 
-enum idle_state { CPU_IDLE, CPU_NOT_IDLE };
-static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE;
-
 void enter_idle(void)
 {
-	__get_cpu_var(idle_state) = CPU_IDLE;
+	write_pda(isidle, 1);
 	atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
 }
 
 static void __exit_idle(void)
 {
-	__get_cpu_var(idle_state) = CPU_NOT_IDLE;
+	if (read_pda(isidle) == 0)
+		return;
+	write_pda(isidle, 0);
 	atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
 }
 
 /* Called from interrupts to signify idle end */
 void exit_idle(void)
 {
-	if (current->pid | read_pda(irqcount))
+	/* idle loop has pid 0 */
+	if (current->pid)
 		return;
 	__exit_idle();
 }
@@ -220,6 +220,9 @@ void cpu_idle (void)
 				play_dead();
 			enter_idle();
 			idle();
+			/* In many cases the interrupt that ended idle
+			   has already called exit_idle. But some idle
+			   loops can be woken up without interrupt. */
 			__exit_idle();
 		}
 
diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h
index 531f48a6c3af..14996d962bac 100644
--- a/include/asm-x86_64/pda.h
+++ b/include/asm-x86_64/pda.h
@@ -25,7 +25,8 @@ struct x8664_pda {
 	int nodenumber;		    /* number of current node */
 	unsigned int __softirq_pending;
 	unsigned int __nmi_count;	/* number of NMI on this CPUs */
-	int mmu_state;     
+	short mmu_state;
+	short isidle;
 	struct mm_struct *active_mm;
 	unsigned apic_timer_irqs;
 } ____cacheline_aligned_in_smp;
-- 
cgit v1.2.1


From 5e544d618f0fb21011f36f28d5e3952b9dc109d2 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386/x86-64: PCI: split probing and initialization of type 1
 config space access

First probe if type1/2 accesses work, but then only initialize them at the end.

This is useful for a later patch that needs this information inbetween.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/pci/direct.c     | 25 ++++++++++++++++---------
 arch/i386/pci/init.c       |  9 +++++++--
 arch/i386/pci/mmconfig.c   |  2 +-
 arch/i386/pci/pci.h        |  6 ++++--
 arch/x86_64/pci/mmconfig.c |  2 +-
 5 files changed, 29 insertions(+), 15 deletions(-)

diff --git a/arch/i386/pci/direct.c b/arch/i386/pci/direct.c
index 5d81fb510375..5acf0b4743cf 100644
--- a/arch/i386/pci/direct.c
+++ b/arch/i386/pci/direct.c
@@ -254,7 +254,16 @@ static int __init pci_check_type2(void)
 	return works;
 }
 
-void __init pci_direct_init(void)
+void __init pci_direct_init(int type)
+{
+	printk(KERN_INFO "PCI: Using configuration type %d\n", type);
+	if (type == 1)
+		raw_pci_ops = &pci_direct_conf1;
+	else
+		raw_pci_ops = &pci_direct_conf2;
+}
+
+int __init pci_direct_probe(void)
 {
 	struct resource *region, *region2;
 
@@ -264,19 +273,16 @@ void __init pci_direct_init(void)
 	if (!region)
 		goto type2;
 
-	if (pci_check_type1()) {
-		printk(KERN_INFO "PCI: Using configuration type 1\n");
-		raw_pci_ops = &pci_direct_conf1;
-		return;
-	}
+	if (pci_check_type1())
+		return 1;
 	release_resource(region);
 
  type2:
 	if ((pci_probe & PCI_PROBE_CONF2) == 0)
-		return;
+		return 0;
 	region = request_region(0xCF8, 4, "PCI conf2");
 	if (!region)
-		return;
+		return 0;
 	region2 = request_region(0xC000, 0x1000, "PCI conf2");
 	if (!region2)
 		goto fail2;
@@ -284,10 +290,11 @@ void __init pci_direct_init(void)
 	if (pci_check_type2()) {
 		printk(KERN_INFO "PCI: Using configuration type 2\n");
 		raw_pci_ops = &pci_direct_conf2;
-		return;
+		return 2;
 	}
 
 	release_resource(region2);
  fail2:
 	release_resource(region);
+	return 0;
 }
diff --git a/arch/i386/pci/init.c b/arch/i386/pci/init.c
index 51087a9d9172..d028e1b05c36 100644
--- a/arch/i386/pci/init.c
+++ b/arch/i386/pci/init.c
@@ -6,8 +6,13 @@
    in the right sequence from here. */
 static __init int pci_access_init(void)
 {
+	int type = 0;
+
+#ifdef CONFIG_PCI_DIRECT
+	type = pci_direct_probe();
+#endif
 #ifdef CONFIG_PCI_MMCONFIG
-	pci_mmcfg_init();
+	pci_mmcfg_init(type);
 #endif
 	if (raw_pci_ops)
 		return 0;
@@ -21,7 +26,7 @@ static __init int pci_access_init(void)
 	 * fails.
 	 */
 #ifdef CONFIG_PCI_DIRECT
-	pci_direct_init();
+	pci_direct_init(type);
 #endif
 	return 0;
 }
diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 972180f738d9..44155c5e85d1 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -187,7 +187,7 @@ static __init void unreachable_devices(void)
 	}
 }
 
-void __init pci_mmcfg_init(void)
+void __init pci_mmcfg_init(int type)
 {
 	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
 		return;
diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h
index bf4e79335388..8a7cf1f23684 100644
--- a/arch/i386/pci/pci.h
+++ b/arch/i386/pci/pci.h
@@ -81,7 +81,9 @@ extern int pci_conf1_write(unsigned int seg, unsigned int bus,
 extern int pci_conf1_read(unsigned int seg, unsigned int bus,
 			  unsigned int devfn, int reg, int len, u32 *value);
 
-extern void pci_direct_init(void);
+extern int pci_direct_probe(void);
+extern void pci_direct_init(int type);
 extern void pci_pcbios_init(void);
-extern void pci_mmcfg_init(void);
+extern void pci_mmcfg_init(int type);
 extern void pcibios_sort(void);
+
diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index dd78f4d18df1..d6b90d08476b 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -163,7 +163,7 @@ static __init void unreachable_devices(void)
 	}
 }
 
-void __init pci_mmcfg_init(void)
+void __init pci_mmcfg_init(int type)
 {
 	int i;
 
-- 
cgit v1.2.1


From 9abd79280bbb9f56049f0168f412c3538cadb6eb Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386/x86-64: Only do MCFG e820 check when type 1 works

Needs earlier patch to split type 1 probing from use.

This patch should fix the x86 macs where type 1 PCI config space access
doesn't work, but MCFG does. They also don't have a usable e820 table
so the e820 sanity check failed.

Instead assume now that if type 1 doesn't work then MCFG must work
and don't do the e820 check.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/pci/mmconfig.c   | 4 +++-
 arch/x86_64/pci/mmconfig.c | 4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 44155c5e85d1..2c4f585510c9 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -198,7 +198,9 @@ void __init pci_mmcfg_init(int type)
 	    (pci_mmcfg_config[0].base_address == 0))
 		return;
 
-	if (!e820_all_mapped(pci_mmcfg_config[0].base_address,
+	/* Only do this check when type 1 works. If it doesn't work
+	   assume we run on a Mac and always use MCFG */
+	if (type == 1 && !e820_all_mapped(pci_mmcfg_config[0].base_address,
 			pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN,
 			E820_RESERVED)) {
 		printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n",
diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index d6b90d08476b..8bea51030d66 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -176,7 +176,9 @@ void __init pci_mmcfg_init(int type)
 	    (pci_mmcfg_config[0].base_address == 0))
 		return;
 
-	if (!e820_all_mapped(pci_mmcfg_config[0].base_address,
+	/* Only do this check when type 1 works. If it doesn't work
+           assume we run on a Mac and always use MCFG */
+	if (type == 1 && !e820_all_mapped(pci_mmcfg_config[0].base_address,
 			pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN,
 			E820_RESERVED)) {
 		printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n",
-- 
cgit v1.2.1


From 56dd669a138c40ea6cdae487f233430d12372767 Mon Sep 17 00:00:00 2001
From: Aaron Durbin <adurbin@google.com>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] Insert GART region into resource map

Patch inserts the GART region into the iomem resource map. The GART will then
be visible within /proc/iomem. It will also allow for other users
utilizing the GART to subreserve the region (agp or IOMMU).

Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/aperture.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index 04dbf1662523..f851e1f9c82a 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -17,6 +17,7 @@
 #include <linux/pci_ids.h>
 #include <linux/pci.h>
 #include <linux/bitops.h>
+#include <linux/ioport.h>
 #include <asm/e820.h>
 #include <asm/io.h>
 #include <asm/proto.h>
@@ -33,6 +34,18 @@ int fallback_aper_force __initdata = 0;
 
 int fix_aperture __initdata = 1;
 
+static struct resource gart_resource = {
+	.name	= "GART",
+	.flags	= IORESOURCE_MEM,
+};
+
+static void __init insert_aperture_resource(u32 aper_base, u32 aper_size)
+{
+	gart_resource.start = aper_base;
+	gart_resource.end = aper_base + aper_size - 1;
+	insert_resource(&iomem_resource, &gart_resource);
+}
+
 /* This code runs before the PCI subsystem is initialized, so just
    access the northbridge directly. */
 
@@ -62,6 +75,7 @@ static u32 __init allocate_aperture(void)
 	}
 	printk("Mapping aperture over %d KB of RAM @ %lx\n",
 	       aper_size >> 10, __pa(p)); 
+	insert_aperture_resource((u32)__pa(p), aper_size);
 	return (u32)__pa(p); 
 }
 
@@ -233,8 +247,13 @@ void __init iommu_hole_init(void)
 		last_aper_base = aper_base;
 	} 
 
-	if (!fix && !fallback_aper_force) 
+	if (!fix && !fallback_aper_force) {
+		if (last_aper_base) {
+			unsigned long n = (32 * 1024 * 1024) << last_aper_order;
+			insert_aperture_resource((u32)last_aper_base, n);
+		}
 		return; 
+	}
 
 	if (!fallback_aper_force)
 		aper_alloc = search_agp_bridge(&aper_order, &valid_agp); 
-- 
cgit v1.2.1


From 4c6e052adfe285ede5884e4e8c4d33af33932c13 Mon Sep 17 00:00:00 2001
From: Aaron Durbin <adurbin@google.com>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] MMCONFIG and new Intel motherboards

On Sat, Sep 09, 2006 at 04:14:29PM +0200, Andi Kleen wrote:
> [patch] Looks reasonable, but probably not for 2.6.18 because this stuff
> is already too fragile and it is probably too risky to do any big changes now
> since not enough testing time is left. Can you please resubmit
> it with proper description and signed-off-by line? I can queue it for .19 then
>
> -Andi

Patch inserts PCI memory mapped config region(s) into the resource map.  This
will allow for the MMCCONFIG regions to be marked as busy in the iomem
address space as well as the regions(s) showing up in /proc/iomem.

Signed-off-by: Aaron Durbin <adurbin@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/pci/mmconfig.c | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index 8bea51030d66..8c0722b95cdc 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -163,6 +163,37 @@ static __init void unreachable_devices(void)
 	}
 }
 
+static __init void pci_mmcfg_insert_resources(void)
+{
+#define PCI_MMCFG_RESOURCE_NAME_LEN 19
+	int i;
+	struct resource *res;
+	char *names;
+	unsigned num_buses;
+
+	res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res),
+			pci_mmcfg_config_num, GFP_KERNEL);
+
+	if (!res) {
+		printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n");
+		return;
+	}
+
+	names = (void *)&res[pci_mmcfg_config_num];
+	for (i = 0; i < pci_mmcfg_config_num; i++, res++) {
+		num_buses = pci_mmcfg_config[i].end_bus_number -
+		    pci_mmcfg_config[i].start_bus_number + 1;
+		res->name = names;
+		snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u",
+			pci_mmcfg_config[i].pci_segment_group_number);
+		res->start = pci_mmcfg_config[i].base_address;
+		res->end = res->start + (num_buses << 20) - 1;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+		insert_resource(&iomem_resource, res);
+		names += PCI_MMCFG_RESOURCE_NAME_LEN;
+	}
+}
+
 void __init pci_mmcfg_init(int type)
 {
 	int i;
@@ -206,6 +237,7 @@ void __init pci_mmcfg_init(int type)
 	}
 
 	unreachable_devices();
+	pci_mmcfg_insert_resources();
 
 	raw_pci_ops = &pci_mmcfg;
 	pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
-- 
cgit v1.2.1


From de09bddb9d6f96785be470c832b881e6d72d589f Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386: Add MMCFG resources to i386 too

Following earlier x86-64 patch

Cc: gregkh@suse.de

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/pci/mmconfig.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c
index 2c4f585510c9..05be8db58a8c 100644
--- a/arch/i386/pci/mmconfig.c
+++ b/arch/i386/pci/mmconfig.c
@@ -151,6 +151,38 @@ static struct pci_raw_ops pci_mmcfg = {
 	.write =	pci_mmcfg_write,
 };
 
+
+static __init void pci_mmcfg_insert_resources(void)
+{
+#define PCI_MMCFG_RESOURCE_NAME_LEN 19
+	int i;
+	struct resource *res;
+	char *names;
+	unsigned num_buses;
+
+	res = kcalloc(PCI_MMCFG_RESOURCE_NAME_LEN + sizeof(*res),
+			pci_mmcfg_config_num, GFP_KERNEL);
+
+	if (!res) {
+		printk(KERN_ERR "PCI: Unable to allocate MMCONFIG resources\n");
+		return;
+	}
+
+	names = (void *)&res[pci_mmcfg_config_num];
+	for (i = 0; i < pci_mmcfg_config_num; i++, res++) {
+		num_buses = pci_mmcfg_config[i].end_bus_number -
+		    pci_mmcfg_config[i].start_bus_number + 1;
+		res->name = names;
+		snprintf(names, PCI_MMCFG_RESOURCE_NAME_LEN, "PCI MMCONFIG %u",
+			pci_mmcfg_config[i].pci_segment_group_number);
+		res->start = pci_mmcfg_config[i].base_address;
+		res->end = res->start + (num_buses << 20) - 1;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+		insert_resource(&iomem_resource, res);
+		names += PCI_MMCFG_RESOURCE_NAME_LEN;
+	}
+}
+
 /* K8 systems have some devices (typically in the builtin northbridge)
    that are only accessible using type1
    Normally this can be expressed in the MCFG by not listing them
@@ -187,6 +219,8 @@ static __init void unreachable_devices(void)
 	}
 }
 
+
+
 void __init pci_mmcfg_init(int type)
 {
 	if ((pci_probe & PCI_PROBE_MMCONF) == 0)
@@ -214,4 +248,5 @@ void __init pci_mmcfg_init(int type)
 	pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF;
 
 	unreachable_devices();
+	pci_mmcfg_insert_resources();
 }
-- 
cgit v1.2.1


From 2817716ace8c397685bd702223cc5a8a42c7e997 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Tue, 26 Sep 2006 10:52:40 +0200
Subject: [PATCH] i386: Fix pack_descriptor()

Fix pack_descriptor:
 1. flags are bits 20-23 in the high word
 2. limit's 4 msb are bits 16-19 in the high word

These haven't mattered so far, because all users have had small limits
and a flags setting of 0.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>

===================================================================
---
 include/asm-i386/desc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 5db9e96e8dc1..5874ef119ffd 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -46,7 +46,7 @@ static inline void pack_descriptor(__u32 *a, __u32 *b,
 {
 	*a = ((base & 0xffff) << 16) | (limit & 0xffff);
 	*b = (base & 0xff000000) | ((base & 0xff0000) >> 16) |
-	     ((type & 0xff) << 8) | ((flags & 0xf) << 12);
+		(limit & 0x000f0000) | ((type & 0xff) << 8) | ((flags & 0xf) << 20);
 }
 
 static inline void pack_gate(__u32 *a, __u32 *b,
-- 
cgit v1.2.1


From f354b3a92af9b132b188b9c8ebbfb74de699926d Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] i386: Split multi-line printk in oops output.

Sometimes, bug reports come in where we've had an oops, and the
only record we have is what the reporter saw on screen shortly
before the system locked up completely.  Unfortunatly, syslog
only prints lines beginning with KERN_EMERG to the console, so
some lines get lost.
An example of this can be seen at https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=203723

Some of this information isn't vital to diagnosis, but some parts
are useful, such as the tainted flag.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/traps.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 5c0f4960c67d..c7adb076e811 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -351,8 +351,9 @@ void show_registers(struct pt_regs *regs)
 		ss = regs->xss & 0xffff;
 	}
 	print_modules();
-	printk(KERN_EMERG "CPU:    %d\nEIP:    %04x:[<%08lx>]    %s VLI\n"
-			"EFLAGS: %08lx   (%s %.*s) \n",
+	printk(KERN_EMERG "CPU:    %d\n"
+		KERN_EMERG "EIP:    %04x:[<%08lx>]    %s VLI\n"
+		KERN_EMERG "EFLAGS: %08lx   (%s %.*s)\n",
 		smp_processor_id(), 0xffff & regs->xcs, regs->eip,
 		print_tainted(), regs->eflags, system_utsname.release,
 		(int)strcspn(system_utsname.version, " "),
-- 
cgit v1.2.1


From 35d534a3ff4905c0a7bddbad0fc9df55967742c7 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] x86: - restore i8259A eoi status on resume

Got it. i8259A_resume calls init_8259A(0) unconditionally, even if
auto_eoi has been set. Keep track of the current status and restore that
on resume. This fixes it for AMD64 and i386.

Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/i8259.c   | 6 +++++-
 arch/x86_64/kernel/i8259.c | 6 +++++-
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/arch/i386/kernel/i8259.c b/arch/i386/kernel/i8259.c
index d4756d154f47..ea5f4e7958d8 100644
--- a/arch/i386/kernel/i8259.c
+++ b/arch/i386/kernel/i8259.c
@@ -45,6 +45,8 @@ static void end_8259A_irq (unsigned int irq)
 
 #define shutdown_8259A_irq	disable_8259A_irq
 
+static int i8259A_auto_eoi;
+
 static void mask_and_ack_8259A(unsigned int);
 
 unsigned int startup_8259A_irq(unsigned int irq)
@@ -253,7 +255,7 @@ static void save_ELCR(char *trigger)
 
 static int i8259A_resume(struct sys_device *dev)
 {
-	init_8259A(0);
+	init_8259A(i8259A_auto_eoi);
 	restore_ELCR(irq_trigger);
 	return 0;
 }
@@ -301,6 +303,8 @@ void init_8259A(int auto_eoi)
 {
 	unsigned long flags;
 
+	i8259A_auto_eoi = auto_eoi;
+
 	spin_lock_irqsave(&i8259A_lock, flags);
 
 	outb(0xff, PIC_MASTER_IMR);	/* mask all of 8259A-1 */
diff --git a/arch/x86_64/kernel/i8259.c b/arch/x86_64/kernel/i8259.c
index d2f5acde5749..2dd51f364ea2 100644
--- a/arch/x86_64/kernel/i8259.c
+++ b/arch/x86_64/kernel/i8259.c
@@ -123,6 +123,8 @@ void (*interrupt[NR_IRQS])(void) = {
 
 DEFINE_SPINLOCK(i8259A_lock);
 
+static int i8259A_auto_eoi;
+
 static void end_8259A_irq (unsigned int irq)
 {
 	if (irq > 256) { 
@@ -336,6 +338,8 @@ void init_8259A(int auto_eoi)
 {
 	unsigned long flags;
 
+	i8259A_auto_eoi = auto_eoi;
+
 	spin_lock_irqsave(&i8259A_lock, flags);
 
 	outb(0xff, 0x21);	/* mask all of 8259A-1 */
@@ -394,7 +398,7 @@ static void save_ELCR(char *trigger)
 
 static int i8259A_resume(struct sys_device *dev)
 {
-	init_8259A(0);
+	init_8259A(i8259A_auto_eoi);
 	restore_ELCR(irq_trigger);
 	return 0;
 }
-- 
cgit v1.2.1


From 27fbe5b28a2ffef171c6005f304ea4f80fcdcc01 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Use string instructions for Core2 copy/clear

It is faster than using a unrolled loop for the use cases the kernel
cares about (cached, sizes typically < 4K)

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index e328e3eb8cbe..085f2e09b9dc 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -843,6 +843,8 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c)
 	if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
 	    (c->x86 == 0x6 && c->x86_model >= 0x0e))
 		set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
+	if (c->x86 == 6)
+		set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
 	set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
  	c->x86_max_cores = intel_num_cpu_cores(c);
 
-- 
cgit v1.2.1


From 9ddab42d1e65dfbdb14588346d47fb142090ab40 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Remove outdated comment in x86-64 mmconfig code

Cc: gregkh@suse.de
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/pci/mmconfig.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c
index 8c0722b95cdc..7732f4254d21 100644
--- a/arch/x86_64/pci/mmconfig.c
+++ b/arch/x86_64/pci/mmconfig.c
@@ -218,7 +218,6 @@ void __init pci_mmcfg_init(int type)
 		return;
 	}
 
-	/* RED-PEN i386 doesn't do _nocache right now */
 	pci_mmcfg_virt = kmalloc(sizeof(*pci_mmcfg_virt) * pci_mmcfg_config_num, GFP_KERNEL);
 	if (pci_mmcfg_virt == NULL) {
 		printk("PCI: Can not allocate memory for mmconfig structures\n");
-- 
cgit v1.2.1


From 2049336f60d297c85ac977245b9326ec00396114 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Don't synchronize time reading on single core AMD systems

We do some additional CPU synchronization in gettimeofday et.al. to make
sure the time stamps are always monotonic over multiple CPUs. But on
single core systems that is not needed. So don't do it.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/setup.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 085f2e09b9dc..26524ce3b239 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -728,6 +728,12 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 
 	/* Fix cpuid4 emulation for more */
 	num_cache_leaves = 3;
+
+	/* When there is only one core no need to synchronize RDTSC */
+	if (num_possible_cpus() == 1)
+	        set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
+	else
+	        clear_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
 }
 
 static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
-- 
cgit v1.2.1


From dd54a11004b2c9a1f136225f880e021a43b0eadc Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Remove all traces of signal number conversion

This was old code that was needed for iBCS and x86-64 never supported that.

Pointed out by Albert Cahalan
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_signal.c | 10 +---------
 arch/x86_64/kernel/signal.c    |  5 -----
 2 files changed, 1 insertion(+), 14 deletions(-)

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index 549de439fb2d..2d20f8304b31 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -431,15 +431,7 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		goto give_sigsegv;
 
-	{
-		struct exec_domain *ed = current_thread_info()->exec_domain;
-		err |= __put_user((ed
-		           && ed->signal_invmap
-		           && sig < 32
-		           ? ed->signal_invmap[sig]
-		           : sig),
-		          &frame->sig);
-	}
+	err |= __put_user(sig, &frame->sig);
 	if (err)
 		goto give_sigsegv;
 
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 7f58bc9a056d..49ec324cd141 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -277,11 +277,6 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 #endif
 
 	/* Set up registers for signal handler */
-	{ 
-		struct exec_domain *ed = current_thread_info()->exec_domain;
-		if (unlikely(ed && ed->signal_invmap && sig < 32))
-			sig = ed->signal_invmap[sig];
-	} 
 	regs->rdi = sig;
 	/* In case the signal handler was declared without prototypes */ 
 	regs->rax = 0;	
-- 
cgit v1.2.1


From 536e3ee4fed13d2d4bbf1b775174aba0cadf6aba Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Initialize argument registers for 32bit signal handlers.

In case the user space was compiled with -mregparm=3
Following i386. Pointed out by Albert Cahalan

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_signal.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index 2d20f8304b31..ced15012a3d7 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -478,6 +478,11 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	regs->rsp = (unsigned long) frame;
 	regs->rip = (unsigned long) ka->sa.sa_handler;
 
+	/* Make -mregparm=3 work */
+	regs->rax = sig;
+	regs->rdx = 0;
+	regs->rcx = 0;
+
 	asm volatile("movl %0,%%ds" :: "r" (__USER32_DS)); 
 	asm volatile("movl %0,%%es" :: "r" (__USER32_DS)); 
 
-- 
cgit v1.2.1


From ab2e0b46cb9a197fab7d98e147cac7cd41a14047 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Fix some broken white space in ia32_signal.c

No functional changes
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/ia32/ia32_signal.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/x86_64/ia32/ia32_signal.c b/arch/x86_64/ia32/ia32_signal.c
index ced15012a3d7..a6ba9951e86c 100644
--- a/arch/x86_64/ia32/ia32_signal.c
+++ b/arch/x86_64/ia32/ia32_signal.c
@@ -490,9 +490,9 @@ int ia32_setup_frame(int sig, struct k_sigaction *ka,
 	regs->ss = __USER32_DS; 
 
 	set_fs(USER_DS);
-    regs->eflags &= ~TF_MASK;
-    if (test_thread_flag(TIF_SINGLESTEP))
-        ptrace_notify(SIGTRAP);
+	regs->eflags &= ~TF_MASK;
+	if (test_thread_flag(TIF_SINGLESTEP))
+		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
@@ -587,8 +587,8 @@ int ia32_setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	set_fs(USER_DS);
 	regs->eflags &= ~TF_MASK;
-    if (test_thread_flag(TIF_SINGLESTEP))
-        ptrace_notify(SIGTRAP);
+	if (test_thread_flag(TIF_SINGLESTEP))
+		ptrace_notify(SIGTRAP);
 
 #if DEBUG_SIG
 	printk("SIG deliver (%s:%d): sp=%p pc=%p ra=%p\n",
-- 
cgit v1.2.1


From adf1423698f00d00b267f7dca8231340ce7d65ef Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] i386/x86-64: Work around gcc bug with noreturn functions in
 unwinder

Current gcc generates calls not jumps to noreturn functions. When that happens the
return address can point to the next function, which confuses the unwinder.

This patch works around it by marking asynchronous exception
frames in contrast normal call frames in the unwind information.  Then teach
the unwinder to decode this.

For normal call frames the unwinder now subtracts one from the address which avoids
this problem.  The standard libgcc unwinder uses the same trick.

It doesn't include adjustment of the printed address (i.e. for the original
example, it'd still be kernel_math_error+0 that gets displayed, but the
unwinder wouldn't get confused anymore.

This only works with binutils 2.6.17+ and some versions of H.J.Lu's 2.6.16
unfortunately because earlier binutils don't support .cfi_signal_frame

[AK: added automatic detection of the new binutils and wrote description]

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/Makefile           |  4 ++++
 arch/i386/kernel/entry.S     |  4 ++++
 arch/x86_64/Makefile         |  4 ++++
 arch/x86_64/ia32/ia32entry.S |  4 ++++
 arch/x86_64/kernel/entry.S   |  4 ++++
 include/asm-i386/dwarf2.h    |  7 +++++++
 include/asm-i386/unwind.h    |  5 +++++
 include/asm-x86_64/dwarf2.h  |  6 ++++++
 include/asm-x86_64/unwind.h  |  5 +++++
 kernel/unwind.c              | 35 ++++++++++++++++++++++++++++-------
 scripts/Kbuild.include       |  4 ++--
 11 files changed, 73 insertions(+), 9 deletions(-)

diff --git a/arch/i386/Makefile b/arch/i386/Makefile
index 508cdbeb3a09..7cc0b189b82b 100644
--- a/arch/i386/Makefile
+++ b/arch/i386/Makefile
@@ -50,6 +50,10 @@ CFLAGS				+= $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-op
 cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
+# is .cfi_signal_frame supported too?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
+
 CFLAGS += $(cflags-y)
 
 # Default subarch .c files
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 284f2e908ad0..5a63d6fdb70e 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -183,18 +183,21 @@ VM_MASK		= 0x00020000
 
 #define RING0_INT_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, 3*4;\
 	/*CFI_OFFSET cs, -2*4;*/\
 	CFI_OFFSET eip, -3*4
 
 #define RING0_EC_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, 4*4;\
 	/*CFI_OFFSET cs, -2*4;*/\
 	CFI_OFFSET eip, -3*4
 
 #define RING0_PTREGS_FRAME \
 	CFI_STARTPROC simple;\
+	CFI_SIGNAL_FRAME;\
 	CFI_DEF_CFA esp, OLDESP-EBX;\
 	/*CFI_OFFSET cs, CS-OLDESP;*/\
 	CFI_OFFSET eip, EIP-OLDESP;\
@@ -275,6 +278,7 @@ need_resched:
 	# sysenter call handler stub
 ENTRY(sysenter_entry)
 	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA esp, 0
 	CFI_REGISTER esp, ebp
 	movl TSS_sysenter_esp0(%esp),%esp
diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile
index 2b8d07c70106..1c0f18d4f887 100644
--- a/arch/x86_64/Makefile
+++ b/arch/x86_64/Makefile
@@ -58,6 +58,10 @@ cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,)
 cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,)
 
+# is .cfi_signal_frame supported too?
+cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,)
+AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,)
+
 cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector )
 cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all )
 
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index 32fd32bea07c..b4aa875e175b 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -71,6 +71,7 @@
  */ 	
 ENTRY(ia32_sysenter_target)
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,0
 	CFI_REGISTER	rsp,rbp
 	swapgs
@@ -186,6 +187,7 @@ ENDPROC(ia32_sysenter_target)
  */ 	
 ENTRY(ia32_cstar_target)
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,PDA_STACKOFFSET
 	CFI_REGISTER	rip,rcx
 	/*CFI_REGISTER	rflags,r11*/
@@ -293,6 +295,7 @@ ia32_badarg:
 
 ENTRY(ia32_syscall)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-RIP
 	/*CFI_REL_OFFSET	ss,SS-RIP*/
 	CFI_REL_OFFSET	rsp,RSP-RIP
@@ -370,6 +373,7 @@ ENTRY(ia32_ptregs_common)
 	popq %r11
 	CFI_ENDPROC
 	CFI_STARTPROC32	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-ARGOFFSET
 	CFI_REL_OFFSET	rax,RAX-ARGOFFSET
 	CFI_REL_OFFSET	rcx,RCX-ARGOFFSET
diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index ea32688386fd..4cbc65290ae7 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -123,6 +123,7 @@
 	.macro	CFI_DEFAULT_STACK start=1
 	.if \start
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8
 	.else
 	CFI_DEF_CFA_OFFSET SS+8
@@ -207,6 +208,7 @@ END(ret_from_fork)
 
 ENTRY(system_call)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,PDA_STACKOFFSET
 	CFI_REGISTER	rip,rcx
 	/*CFI_REGISTER	rflags,r11*/
@@ -324,6 +326,7 @@ END(system_call)
  */ 	
 ENTRY(int_ret_from_sys_call)
 	CFI_STARTPROC	simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA	rsp,SS+8-ARGOFFSET
 	/*CFI_REL_OFFSET	ss,SS-ARGOFFSET*/
 	CFI_REL_OFFSET	rsp,RSP-ARGOFFSET
@@ -484,6 +487,7 @@ END(stub_rt_sigreturn)
  */
 	.macro _frame ref
 	CFI_STARTPROC simple
+	CFI_SIGNAL_FRAME
 	CFI_DEF_CFA rsp,SS+8-\ref
 	/*CFI_REL_OFFSET ss,SS-\ref*/
 	CFI_REL_OFFSET rsp,RSP-\ref
diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h
index 5d1a8db5a9b0..6d66398a307d 100644
--- a/include/asm-i386/dwarf2.h
+++ b/include/asm-i386/dwarf2.h
@@ -28,6 +28,12 @@
 #define CFI_RESTORE_STATE .cfi_restore_state
 #define CFI_UNDEFINED .cfi_undefined
 
+#ifdef CONFIG_AS_CFI_SIGNAL_FRAME
+#define CFI_SIGNAL_FRAME .cfi_signal_frame
+#else
+#define CFI_SIGNAL_FRAME
+#endif
+
 #else
 
 /* Due to the structure of pre-exisiting code, don't use assembler line
@@ -48,6 +54,7 @@
 #define CFI_REMEMBER_STATE ignore
 #define CFI_RESTORE_STATE ignore
 #define CFI_UNDEFINED ignore
+#define CFI_SIGNAL_FRAME ignore
 
 #endif
 
diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h
index f0ac399bae3c..5031d693b89d 100644
--- a/include/asm-i386/unwind.h
+++ b/include/asm-i386/unwind.h
@@ -18,6 +18,7 @@ struct unwind_frame_info
 {
 	struct pt_regs regs;
 	struct task_struct *task;
+	unsigned call_frame:1;
 };
 
 #define UNW_PC(frame)        (frame)->regs.eip
@@ -44,6 +45,10 @@ struct unwind_frame_info
 	PTREGS_INFO(edi), \
 	PTREGS_INFO(eip)
 
+#define UNW_DEFAULT_RA(raItem, dataAlign) \
+	((raItem).where == Memory && \
+	 !((raItem).value * (dataAlign) + 4))
+
 static inline void arch_unw_init_frame_info(struct unwind_frame_info *info,
                                             /*const*/ struct pt_regs *regs)
 {
diff --git a/include/asm-x86_64/dwarf2.h b/include/asm-x86_64/dwarf2.h
index 2b9368365fad..eedc08526b0b 100644
--- a/include/asm-x86_64/dwarf2.h
+++ b/include/asm-x86_64/dwarf2.h
@@ -28,6 +28,11 @@
 #define CFI_REMEMBER_STATE .cfi_remember_state
 #define CFI_RESTORE_STATE .cfi_restore_state
 #define CFI_UNDEFINED .cfi_undefined
+#ifdef CONFIG_AS_CFI_SIGNAL_FRAME
+#define CFI_SIGNAL_FRAME .cfi_signal_frame
+#else
+#define CFI_SIGNAL_FRAME
+#endif
 
 #else
 
@@ -45,6 +50,7 @@
 #define CFI_REMEMBER_STATE	#
 #define CFI_RESTORE_STATE	#
 #define CFI_UNDEFINED	#
+#define CFI_SIGNAL_FRAME	#
 
 #endif
 
diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h
index 1f6e9bfb569e..b8fa5cb7ff88 100644
--- a/include/asm-x86_64/unwind.h
+++ b/include/asm-x86_64/unwind.h
@@ -18,6 +18,7 @@ struct unwind_frame_info
 {
 	struct pt_regs regs;
 	struct task_struct *task;
+	unsigned call_frame:1;
 };
 
 #define UNW_PC(frame)        (frame)->regs.rip
@@ -57,6 +58,10 @@ struct unwind_frame_info
 	PTREGS_INFO(r15), \
 	PTREGS_INFO(rip)
 
+#define UNW_DEFAULT_RA(raItem, dataAlign) \
+	((raItem).where == Memory && \
+	 !((raItem).value * (dataAlign) + 8))
+
 static inline void arch_unw_init_frame_info(struct unwind_frame_info *info,
                                             /*const*/ struct pt_regs *regs)
 {
diff --git a/kernel/unwind.c b/kernel/unwind.c
index f69c804c8e62..3430475fcd88 100644
--- a/kernel/unwind.c
+++ b/kernel/unwind.c
@@ -603,6 +603,7 @@ int unwind(struct unwind_frame_info *frame)
 #define FRAME_REG(r, t) (((t *)frame)[reg_info[r].offs])
 	const u32 *fde = NULL, *cie = NULL;
 	const u8 *ptr = NULL, *end = NULL;
+	unsigned long pc = UNW_PC(frame) - frame->call_frame;
 	unsigned long startLoc = 0, endLoc = 0, cfa;
 	unsigned i;
 	signed ptrType = -1;
@@ -612,7 +613,7 @@ int unwind(struct unwind_frame_info *frame)
 
 	if (UNW_PC(frame) == 0)
 		return -EINVAL;
-	if ((table = find_table(UNW_PC(frame))) != NULL
+	if ((table = find_table(pc)) != NULL
 	    && !(table->size & (sizeof(*fde) - 1))) {
 		unsigned long tableSize = table->size;
 
@@ -647,7 +648,7 @@ int unwind(struct unwind_frame_info *frame)
 			                        ptrType & DW_EH_PE_indirect
 			                        ? ptrType
 			                        : ptrType & (DW_EH_PE_FORM|DW_EH_PE_signed));
-			if (UNW_PC(frame) >= startLoc && UNW_PC(frame) < endLoc)
+			if (pc >= startLoc && pc < endLoc)
 				break;
 			cie = NULL;
 		}
@@ -657,16 +658,28 @@ int unwind(struct unwind_frame_info *frame)
 		state.cieEnd = ptr; /* keep here temporarily */
 		ptr = (const u8 *)(cie + 2);
 		end = (const u8 *)(cie + 1) + *cie;
+		frame->call_frame = 1;
 		if ((state.version = *ptr) != 1)
 			cie = NULL; /* unsupported version */
 		else if (*++ptr) {
 			/* check if augmentation size is first (and thus present) */
 			if (*ptr == 'z') {
-				/* check for ignorable (or already handled)
-				 * nul-terminated augmentation string */
-				while (++ptr < end && *ptr)
-					if (strchr("LPR", *ptr) == NULL)
+				while (++ptr < end && *ptr) {
+					switch(*ptr) {
+					/* check for ignorable (or already handled)
+					 * nul-terminated augmentation string */
+					case 'L':
+					case 'P':
+					case 'R':
+						continue;
+					case 'S':
+						frame->call_frame = 0;
+						continue;
+					default:
 						break;
+					}
+					break;
+				}
 			}
 			if (ptr >= end || *ptr)
 				cie = NULL;
@@ -755,7 +768,7 @@ int unwind(struct unwind_frame_info *frame)
 	state.org = startLoc;
 	memcpy(&state.cfa, &badCFA, sizeof(state.cfa));
 	/* process instructions */
-	if (!processCFI(ptr, end, UNW_PC(frame), ptrType, &state)
+	if (!processCFI(ptr, end, pc, ptrType, &state)
 	   || state.loc > endLoc
 	   || state.regs[retAddrReg].where == Nowhere
 	   || state.cfa.reg >= ARRAY_SIZE(reg_info)
@@ -763,6 +776,11 @@ int unwind(struct unwind_frame_info *frame)
 	   || state.cfa.offs % sizeof(unsigned long))
 		return -EIO;
 	/* update frame */
+#ifndef CONFIG_AS_CFI_SIGNAL_FRAME
+	if(frame->call_frame
+	   && !UNW_DEFAULT_RA(state.regs[retAddrReg], state.dataAlign))
+		frame->call_frame = 0;
+#endif
 	cfa = FRAME_REG(state.cfa.reg, unsigned long) + state.cfa.offs;
 	startLoc = min((unsigned long)UNW_SP(frame), cfa);
 	endLoc = max((unsigned long)UNW_SP(frame), cfa);
@@ -866,6 +884,7 @@ int unwind_init_frame_info(struct unwind_frame_info *info,
                            /*const*/ struct pt_regs *regs)
 {
 	info->task = tsk;
+	info->call_frame = 0;
 	arch_unw_init_frame_info(info, regs);
 
 	return 0;
@@ -879,6 +898,7 @@ int unwind_init_blocked(struct unwind_frame_info *info,
                         struct task_struct *tsk)
 {
 	info->task = tsk;
+	info->call_frame = 0;
 	arch_unw_init_blocked(info);
 
 	return 0;
@@ -894,6 +914,7 @@ int unwind_init_running(struct unwind_frame_info *info,
                         void *arg)
 {
 	info->task = current;
+	info->call_frame = 0;
 
 	return arch_unwind_init_running(info, callback, arg);
 }
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index 7adef12a0c26..4f5ff19b992b 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -66,8 +66,8 @@ as-option = $(shell if $(CC) $(CFLAGS) $(1) -Wa,-Z -c -o /dev/null \
 # as-instr
 # Usage: cflags-y += $(call as-instr, instr, option1, option2)
 
-as-instr = $(shell if echo -e "$(1)" | $(AS) -Z -o astest$$$$.out \
-		   2>&1 >/dev/null ; then echo "$(2)"; else echo "$(3)"; fi; \
+as-instr = $(shell if echo -e "$(1)" | $(AS) >/dev/null 2>&1 -W -Z -o astest$$$$.out ; \
+		   then echo "$(2)"; else echo "$(3)"; fi; \
 	           rm -f astest$$$$.out)
 
 # cc-option
-- 
cgit v1.2.1


From 658fdbef66e5e9be79b457edc2cbbb3add840aa9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] Don't leak NT bit into next task

SYSENTER can cause a NT to be set which might cause crashes on the IRET
in the next task.

Following similar i386 patch from Linus.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/kernel/entry.S   | 4 ++++
 arch/x86_64/kernel/setup64.c | 4 ++++
 include/asm-x86_64/system.h  | 5 +++--
 3 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index 4cbc65290ae7..2802524104f3 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -155,6 +155,10 @@
 /* rdi:	prev */	
 ENTRY(ret_from_fork)
 	CFI_DEFAULT_STACK
+	push kernel_eflags(%rip)
+	CFI_ADJUST_CFA_OFFSET 4
+	popf				# reset kernel eflags
+	CFI_ADJUST_CFA_OFFSET -4
 	call schedule_tail
 	GET_THREAD_INFO(%rcx)
 	testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT),threadinfo_flags(%rcx)
diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c
index 9332d2361e08..8c4b80fe71a1 100644
--- a/arch/x86_64/kernel/setup64.c
+++ b/arch/x86_64/kernel/setup64.c
@@ -180,6 +180,8 @@ void __cpuinit check_efer(void)
         }       
 }
 
+unsigned long kernel_eflags;
+
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
@@ -281,4 +283,6 @@ void __cpuinit cpu_init (void)
 	set_debugreg(0UL, 7);
 
 	fpu_init(); 
+
+	raw_local_save_flags(kernel_eflags);
 }
diff --git a/include/asm-x86_64/system.h b/include/asm-x86_64/system.h
index 6bf170bceae1..bd376bc8c4ab 100644
--- a/include/asm-x86_64/system.h
+++ b/include/asm-x86_64/system.h
@@ -14,12 +14,13 @@
 #define __RESTORE(reg,offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
 
 /* frame pointer must be last for get_wchan */
-#define SAVE_CONTEXT    "pushq %%rbp ; movq %%rsi,%%rbp\n\t"
-#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp\n\t"
+#define SAVE_CONTEXT    "pushf ; pushq %%rbp ; movq %%rsi,%%rbp\n\t"
+#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp ; popf\t"
 
 #define __EXTRA_CLOBBER  \
 	,"rcx","rbx","rdx","r8","r9","r10","r11","r12","r13","r14","r15"
 
+/* Save restore flags to clear handle leaking NT */
 #define switch_to(prev,next,last) \
 	asm volatile(SAVE_CONTEXT						    \
 		     "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */	  \
-- 
cgit v1.2.1


From f157cbb1eb9ce3f33a401ec6d20eb3eb852351a3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] i386/x86-64: Make all early PCI scans dependent on CONFIG_PCI

This is useful on systems with broken PCI bus. Affects various
scans in x86-64 and i386's early ACPI quirk scan.

Cc: gregkh@suse.de
Cc: len.brown@intel.com
Cc: Trammell Hudson <hudson@osresearch.net>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/acpi/Makefile | 2 ++
 arch/i386/kernel/setup.c       | 2 ++
 arch/x86_64/Kconfig            | 3 ++-
 arch/x86_64/kernel/Makefile    | 3 ++-
 arch/x86_64/kernel/setup.c     | 2 ++
 5 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/arch/i386/kernel/acpi/Makefile b/arch/i386/kernel/acpi/Makefile
index 7e9ac99354f4..7f7be01f44e6 100644
--- a/arch/i386/kernel/acpi/Makefile
+++ b/arch/i386/kernel/acpi/Makefile
@@ -1,5 +1,7 @@
 obj-$(CONFIG_ACPI)		+= boot.o
+ifneq ($(CONFIG_PCI),)
 obj-$(CONFIG_X86_IO_APIC)	+= earlyquirk.o
+endif
 obj-$(CONFIG_ACPI_SLEEP)	+= sleep.o wakeup.o
 
 ifneq ($(CONFIG_ACPI_PROCESSOR),)
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index ea17567dbe72..7a99b1369fa2 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1437,9 +1437,11 @@ void __init setup_arch(char **cmdline_p)
 	acpi_boot_table_init();
 #endif
 
+#ifdef CONFIG_PCI
 #ifdef CONFIG_X86_IO_APIC
 	check_acpi_pci();	/* Checks more than just ACPI actually */
 #endif
+#endif
 
 #ifdef CONFIG_ACPI
 	acpi_boot_init();
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 1f4212605ef8..c2c68b902347 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -105,6 +105,7 @@ config X86_PC
 
 config X86_VSMP
 	bool "Support for ScaleMP vSMP"
+	depends on PCI
 	 help
 	  Support for ScaleMP vSMP systems.  Say 'Y' here if this kernel is
 	  supposed to run on these EM64T-based machines.  Only choose this option
@@ -291,7 +292,7 @@ config NUMA
 
 config K8_NUMA
        bool "Old style AMD Opteron NUMA detection"
-       depends on NUMA
+       depends on NUMA && PCI
        default y
        help
 	 Enable K8 NUMA node topology detection.  You should say Y here if
diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 000e67e8f028..2466fbd035ee 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -8,7 +8,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_x86_64.o \
 		x8664_ksyms.o i387.o syscall.o vsyscall.o \
 		setup64.o bootflag.o e820.o reboot.o quirks.o i8237.o \
-		pci-dma.o pci-nommu.o alternative.o early-quirks.o
+		pci-dma.o pci-nommu.o alternative.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
 obj-$(CONFIG_X86_MCE)         += mce.o
@@ -39,6 +39,7 @@ obj-$(CONFIG_K8_NB)		+= k8.o
 obj-$(CONFIG_AUDIT)		+= audit.o
 
 obj-$(CONFIG_MODULES)		+= module.o
+obj-$(CONFIG_PCI)		+= early-quirks.o
 
 obj-y				+= topology.o
 obj-y				+= intel_cacheinfo.o
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 26524ce3b239..3d8309b1236e 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -493,7 +493,9 @@ void __init setup_arch(char **cmdline_p)
 
 	paging_init();
 
+#ifdef CONFIG_PCI
 	early_quirks();
+#endif
 
 	/*
 	 * set this early, so we dont allocate cpu0
-- 
cgit v1.2.1


From 8f60774a116ced9b73ae3913d511687889efe725 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] x86: Move direct PCI scanning functions out of line

Saves about 200 bytes of code space.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/pci/Makefile          |  2 +-
 arch/i386/pci/early.c           | 44 +++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/pci/Makefile        |  3 ++-
 include/asm-x86_64/pci-direct.h | 42 ++++-----------------------------------
 4 files changed, 51 insertions(+), 40 deletions(-)
 create mode 100644 arch/i386/pci/early.c

diff --git a/arch/i386/pci/Makefile b/arch/i386/pci/Makefile
index 62ad75c57e6a..1594d2f55c8f 100644
--- a/arch/i386/pci/Makefile
+++ b/arch/i386/pci/Makefile
@@ -11,4 +11,4 @@ pci-y				+= legacy.o irq.o
 pci-$(CONFIG_X86_VISWS)		:= visws.o fixup.o
 pci-$(CONFIG_X86_NUMAQ)		:= numa.o irq.o
 
-obj-y				+= $(pci-y) common.o
+obj-y				+= $(pci-y) common.o early.o
diff --git a/arch/i386/pci/early.c b/arch/i386/pci/early.c
new file mode 100644
index 000000000000..b1f7f40d809b
--- /dev/null
+++ b/arch/i386/pci/early.c
@@ -0,0 +1,44 @@
+#include <linux/kernel.h>
+#include <asm/pci-direct.h>
+#include <asm/io.h>
+
+/* Direct PCI access. This is used for PCI accesses in early boot before
+   the PCI subsystem works. */
+
+#define PDprintk(x...)
+
+u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u32 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inl(0xcfc);
+	if (v != 0xffffffff)
+		PDprintk("%x reading 4 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u8 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inb(0xcfc + (offset&3));
+	PDprintk("%x reading 1 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset)
+{
+	u16 v;
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	v = inw(0xcfc + (offset&2));
+	PDprintk("%x reading 2 from %x: %x\n", slot, offset, v);
+	return v;
+}
+
+void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
+				    u32 val)
+{
+	PDprintk("%x writing to %x: %x\n", slot, offset, val);
+	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
+	outl(val, 0xcfc);
+}
diff --git a/arch/x86_64/pci/Makefile b/arch/x86_64/pci/Makefile
index a3f6ad570179..1eb18f421edf 100644
--- a/arch/x86_64/pci/Makefile
+++ b/arch/x86_64/pci/Makefile
@@ -9,7 +9,7 @@ obj-y		:= i386.o
 obj-$(CONFIG_PCI_DIRECT)+= direct.o
 obj-y		+= fixup.o init.o
 obj-$(CONFIG_ACPI)	+= acpi.o
-obj-y			+= legacy.o irq.o common.o
+obj-y			+= legacy.o irq.o common.o early.o
 # mmconfig has a 64bit special
 obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o direct.o
 
@@ -23,3 +23,4 @@ common-y += ../../i386/pci/common.o
 fixup-y  += ../../i386/pci/fixup.o
 i386-y  += ../../i386/pci/i386.o
 init-y += ../../i386/pci/init.o
+early-y += ../../i386/pci/early.o
diff --git a/include/asm-x86_64/pci-direct.h b/include/asm-x86_64/pci-direct.h
index 036b6ca5b53b..9d916cdaa18e 100644
--- a/include/asm-x86_64/pci-direct.h
+++ b/include/asm-x86_64/pci-direct.h
@@ -2,47 +2,13 @@
 #define ASM_PCI_DIRECT_H 1
 
 #include <linux/types.h>
-#include <asm/io.h>
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
    the PCI subsystem works. */ 
 
-#define PDprintk(x...)
-
-static inline u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u32 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inl(0xcfc); 
-	if (v != 0xffffffff)
-		PDprintk("%x reading 4 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u8 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inb(0xcfc + (offset&3)); 
-	PDprintk("%x reading 1 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset)
-{
-	u16 v; 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	v = inw(0xcfc + (offset&2)); 
-	PDprintk("%x reading 2 from %x: %x\n", slot, offset, v);
-	return v;
-}
-
-static inline void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
-				    u32 val)
-{
-	PDprintk("%x writing to %x: %x\n", slot, offset, val); 
-	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
-	outl(val, 0xcfc); 
-}
+extern u32 read_pci_config(u8 bus, u8 slot, u8 func, u8 offset);
+extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset);
+extern u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset);
+extern void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, u32 val);
 
 #endif
-- 
cgit v1.2.1


From 0637a70a5db98182d9ad3d6ae1ee30acf20afde9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:41 +0200
Subject: [PATCH] x86: Allow disabling early pci scans with pci=noearly or
 disallowing conf1

Some buggy systems can machine check when config space accesses
happen for some non existent devices.  i386/x86-64 do some early
device scans that might trigger this. Allow pci=noearly to disable
this. Also when type 1 is disabling also don't do any early
accesses which are always type1.

This moves the pci= configuration parsing to be a early parameter.
I don't think this can break anything because it only changes
a single global that is only used by PCI.

Cc: gregkh@suse.de
Cc: Trammell Hudson <hudson@osresearch.net>

Signed-off-by: Andi Kleen <ak@suse.de>
---
 Documentation/kernel-parameters.txt | 6 +++++-
 arch/i386/kernel/acpi/earlyquirk.c  | 6 +++++-
 arch/i386/pci/common.c              | 4 ++++
 arch/i386/pci/early.c               | 8 ++++++++
 arch/i386/pci/pci.h                 | 1 +
 arch/x86_64/kernel/aperture.c       | 2 +-
 arch/x86_64/kernel/early-quirks.c   | 4 ++++
 arch/x86_64/kernel/pci-calgary.c    | 3 +++
 arch/x86_64/kernel/vsmp.c           | 3 +++
 arch/x86_64/mm/k8topology.c         | 3 +++
 drivers/pci/pci.c                   | 5 ++---
 include/asm-x86_64/pci-direct.h     | 2 ++
 12 files changed, 41 insertions(+), 6 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 71d05f481727..4ae99f6e9938 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1240,7 +1240,11 @@ running once the system is up.
 				bootloader. This is currently used on
 				IXP2000 systems where the bus has to be
 				configured a certain way for adjunct CPUs.
-
+		noearly		[X86] Don't do any early type 1 scanning.
+				This might help on some broken boards which
+				machine check when some devices' config space
+				is read. But various workarounds are disabled
+				and some IOMMU drivers will not work.
 	pcmv=		[HW,PCMCIA] BadgePAD 4
 
 	pd.		[PARIDE]
diff --git a/arch/i386/kernel/acpi/earlyquirk.c b/arch/i386/kernel/acpi/earlyquirk.c
index 1649a175a206..fe799b11ac0a 100644
--- a/arch/i386/kernel/acpi/earlyquirk.c
+++ b/arch/i386/kernel/acpi/earlyquirk.c
@@ -48,7 +48,11 @@ void __init check_acpi_pci(void)
 	int num, slot, func;
 
 	/* Assume the machine supports type 1. If not it will 
-	   always read ffffffff and should not have any side effect. */
+	   always read ffffffff and should not have any side effect.
+	   Actually a few buggy systems can machine check. Allow the user
+	   to disable it by command line option at least -AK */
+	if (!early_pci_allowed())
+		return;
 
 	/* Poor man's PCI discovery */
 	for (num = 0; num < 32; num++) {
diff --git a/arch/i386/pci/common.c b/arch/i386/pci/common.c
index 0a362e3aeac5..68bce194e688 100644
--- a/arch/i386/pci/common.c
+++ b/arch/i386/pci/common.c
@@ -242,6 +242,10 @@ char * __devinit  pcibios_setup(char *str)
 		acpi_noirq_set();
 		return NULL;
 	}
+	else if (!strcmp(str, "noearly")) {
+		pci_probe |= PCI_PROBE_NOEARLY;
+		return NULL;
+	}
 #ifndef CONFIG_X86_VISWS
 	else if (!strcmp(str, "usepirqmask")) {
 		pci_probe |= PCI_USE_PIRQ_MASK;
diff --git a/arch/i386/pci/early.c b/arch/i386/pci/early.c
index b1f7f40d809b..713d6c866cae 100644
--- a/arch/i386/pci/early.c
+++ b/arch/i386/pci/early.c
@@ -1,6 +1,8 @@
 #include <linux/kernel.h>
+#include <linux/pci.h>
 #include <asm/pci-direct.h>
 #include <asm/io.h>
+#include "pci.h"
 
 /* Direct PCI access. This is used for PCI accesses in early boot before
    the PCI subsystem works. */
@@ -42,3 +44,9 @@ void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset,
 	outl(0x80000000 | (bus<<16) | (slot<<11) | (func<<8) | offset, 0xcf8);
 	outl(val, 0xcfc);
 }
+
+int early_pci_allowed(void)
+{
+	return (pci_probe & (PCI_PROBE_CONF1|PCI_PROBE_NOEARLY)) ==
+			PCI_PROBE_CONF1;
+}
diff --git a/arch/i386/pci/pci.h b/arch/i386/pci/pci.h
index 8a7cf1f23684..1814f74569c6 100644
--- a/arch/i386/pci/pci.h
+++ b/arch/i386/pci/pci.h
@@ -17,6 +17,7 @@
 #define PCI_PROBE_CONF2		0x0004
 #define PCI_PROBE_MMCONF	0x0008
 #define PCI_PROBE_MASK		0x000f
+#define PCI_PROBE_NOEARLY	0x0010
 
 #define PCI_NO_SORT		0x0100
 #define PCI_BIOS_SORT		0x0200
diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index f851e1f9c82a..b487396c4c5b 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -212,7 +212,7 @@ void __init iommu_hole_init(void)
 	u64 aper_base, last_aper_base = 0;
 	int valid_agp = 0;
 
-	if (iommu_aperture_disabled || !fix_aperture)
+	if (iommu_aperture_disabled || !fix_aperture || !early_pci_allowed())
 		return;
 
 	printk("Checking aperture...\n"); 
diff --git a/arch/x86_64/kernel/early-quirks.c b/arch/x86_64/kernel/early-quirks.c
index d637cff1c4b1..208e38a372c1 100644
--- a/arch/x86_64/kernel/early-quirks.c
+++ b/arch/x86_64/kernel/early-quirks.c
@@ -82,6 +82,10 @@ static struct chipset early_qrk[] = {
 void __init early_quirks(void)
 {
 	int num, slot, func;
+
+	if (!early_pci_allowed())
+		return;
+
 	/* Poor man's PCI discovery */
 	for (num = 0; num < 32; num++) {
 		for (slot = 0; slot < 32; slot++) {
diff --git a/arch/x86_64/kernel/pci-calgary.c b/arch/x86_64/kernel/pci-calgary.c
index 466588f95601..cfb09b07ae99 100644
--- a/arch/x86_64/kernel/pci-calgary.c
+++ b/arch/x86_64/kernel/pci-calgary.c
@@ -924,6 +924,9 @@ void __init detect_calgary(void)
 	if (swiotlb || no_iommu || iommu_detected)
 		return;
 
+	if (!early_pci_allowed())
+		return;
+
 	specified_table_size = determine_tce_table_size(end_pfn * PAGE_SIZE);
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
diff --git a/arch/x86_64/kernel/vsmp.c b/arch/x86_64/kernel/vsmp.c
index 92f70c74965f..044e852bd25e 100644
--- a/arch/x86_64/kernel/vsmp.c
+++ b/arch/x86_64/kernel/vsmp.c
@@ -20,6 +20,9 @@ static int __init vsmp_init(void)
 	void *address;
 	unsigned int cap, ctl;
 
+	if (!early_pci_allowed())
+		return 0;
+
 	/* Check if we are running on a ScaleMP vSMP box */
 	if ((read_pci_config_16(0, 0x1f, 0, PCI_VENDOR_ID) != PCI_VENDOR_ID_SCALEMP) ||
 	    (read_pci_config_16(0, 0x1f, 0, PCI_DEVICE_ID) != PCI_DEVICE_ID_SCALEMP_VSMP_CTL))
diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index 7c45c2d2b8b2..5cf594f9230d 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -54,6 +54,9 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 
 	nodes_clear(nodes_parsed);
 
+	if (!early_pci_allowed())
+		return -1;
+
 	nb = find_northbridge(); 
 	if (nb < 0) 
 		return nb;
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 9f79dd6d51ab..684deb6b03aa 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -953,13 +953,12 @@ static int __devinit pci_setup(char *str)
 		}
 		str = k;
 	}
-	return 1;
+	return 0;
 }
+early_param("pci", pci_setup);
 
 device_initcall(pci_init);
 
-__setup("pci=", pci_setup);
-
 #if defined(CONFIG_ISA) || defined(CONFIG_EISA)
 /* FIXME: Some boxes have multiple ISA bridges! */
 struct pci_dev *isa_bridge;
diff --git a/include/asm-x86_64/pci-direct.h b/include/asm-x86_64/pci-direct.h
index 9d916cdaa18e..eba9cb471df3 100644
--- a/include/asm-x86_64/pci-direct.h
+++ b/include/asm-x86_64/pci-direct.h
@@ -11,4 +11,6 @@ extern u8 read_pci_config_byte(u8 bus, u8 slot, u8 func, u8 offset);
 extern u16 read_pci_config_16(u8 bus, u8 slot, u8 func, u8 offset);
 extern void write_pci_config(u8 bus, u8 slot, u8 func, u8 offset, u32 val);
 
+extern int early_pci_allowed(void);
+
 #endif
-- 
cgit v1.2.1


From b89ebd0b0a65d5371aa9ad98e873c4616056ca68 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] Fix unwinder warning in traps.c

Fix

linux/arch/x86_64/kernel/traps.c: In function 'dump_trace':
linux/arch/x86_64/kernel/traps.c:275: warning: cast to pointer from integer of different size

with allnoconfig

Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/asm-x86_64/unwind.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h
index b8fa5cb7ff88..2e7ff10fd775 100644
--- a/include/asm-x86_64/unwind.h
+++ b/include/asm-x86_64/unwind.h
@@ -99,8 +99,8 @@ static inline int arch_unw_user_mode(const struct unwind_frame_info *info)
 
 #else
 
-#define UNW_PC(frame) ((void)(frame), 0)
-#define UNW_SP(frame) ((void)(frame), 0)
+#define UNW_PC(frame) ((void)(frame), 0UL)
+#define UNW_SP(frame) ((void)(frame), 0UL)
 
 static inline int arch_unw_user_mode(const void *info)
 {
-- 
cgit v1.2.1


From 3b171672831b9633c2ed8fa94805255cd4d5af19 Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] Add 64bit jiffies compares (for use with get_jiffies_64)

The current time_before/time_after macros will fail typechecks
when passed u64 values (as returned by get_jiffies_64()). On 64bit
systems, this will just result in a warning about mismatching types
without explicit casts, but since unsigned long and u64
(unsigned long long) are of same size, it will still work.
On 32bit systems, a long is 32bits, so the value from get_jiffies_64()
will be truncated by the cast and thus lose all the precision gained by
64bit jiffies.

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 include/linux/jiffies.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/include/linux/jiffies.h b/include/linux/jiffies.h
index 329ebcffa106..c8d5f207c3d4 100644
--- a/include/linux/jiffies.h
+++ b/include/linux/jiffies.h
@@ -115,6 +115,21 @@ static inline u64 get_jiffies_64(void)
 	 ((long)(a) - (long)(b) >= 0))
 #define time_before_eq(a,b)	time_after_eq(b,a)
 
+/* Same as above, but does so with platform independent 64bit types.
+ * These must be used when utilizing jiffies_64 (i.e. return value of
+ * get_jiffies_64() */
+#define time_after64(a,b)	\
+	(typecheck(__u64, a) &&	\
+	 typecheck(__u64, b) && \
+	 ((__s64)(b) - (__s64)(a) < 0))
+#define time_before64(a,b)	time_after64(b,a)
+
+#define time_after_eq64(a,b)	\
+	(typecheck(__u64, a) && \
+	 typecheck(__u64, b) && \
+	 ((__s64)(a) - (__s64)(b) >= 0))
+#define time_before_eq64(a,b)	time_after_eq64(b,a)
+
 /*
  * Have the 32 bit jiffies value wrap 5 minutes after boot
  * so jiffies wrap bugs show up earlier.
-- 
cgit v1.2.1


From 15d5f8398311f565682959daaca30e3ca7aea600 Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] x86: Refactor thermal throttle processing

Refactor the event processing (syslog messaging and rate limiting)
into separate file therm_throt.c. This allows consistent reporting
of CPU thermal throttle events.

After ACK'ing the interrupt, if the event is current, the user
(p4.c/mce_intel.c) calls therm_throt_process to log (and rate limit)
the event. If that function returns 1, the user has the option to log
things further (such as to mce_log in x86_64).

AK: minor cleanup

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/mcheck/Makefile      |  2 +-
 arch/i386/kernel/cpu/mcheck/p4.c          | 23 ++++--------
 arch/i386/kernel/cpu/mcheck/therm_throt.c | 58 +++++++++++++++++++++++++++++++
 arch/x86_64/kernel/Makefile               |  4 +--
 arch/x86_64/kernel/mce.c                  | 27 ++++++++++++++
 arch/x86_64/kernel/mce_intel.c            | 27 ++++----------
 include/asm-i386/therm_throt.h            |  6 ++++
 include/asm-x86_64/mce.h                  |  2 ++
 include/asm-x86_64/therm_throt.h          |  1 +
 9 files changed, 109 insertions(+), 41 deletions(-)
 create mode 100644 arch/i386/kernel/cpu/mcheck/therm_throt.c
 create mode 100644 include/asm-i386/therm_throt.h
 create mode 100644 include/asm-x86_64/therm_throt.h

diff --git a/arch/i386/kernel/cpu/mcheck/Makefile b/arch/i386/kernel/cpu/mcheck/Makefile
index 30808f3d6715..f1ebe1c1c17a 100644
--- a/arch/i386/kernel/cpu/mcheck/Makefile
+++ b/arch/i386/kernel/cpu/mcheck/Makefile
@@ -1,2 +1,2 @@
-obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o
+obj-y	=	mce.o k7.o p4.o p5.o p6.o winchip.o therm_throt.o
 obj-$(CONFIG_X86_MCE_NONFATAL)	+=	non-fatal.o
diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c
index b95f1b3d53aa..d83a669d376f 100644
--- a/arch/i386/kernel/cpu/mcheck/p4.c
+++ b/arch/i386/kernel/cpu/mcheck/p4.c
@@ -13,6 +13,8 @@
 #include <asm/msr.h>
 #include <asm/apic.h>
 
+#include <asm/therm_throt.h>
+
 #include "mce.h"
 
 /* as supported by the P4/Xeon family */
@@ -44,25 +46,12 @@ static void unexpected_thermal_interrupt(struct pt_regs *regs)
 /* P4/Xeon Thermal transition interrupt handler */
 static void intel_thermal_interrupt(struct pt_regs *regs)
 {
-	u32 l, h;
-	unsigned int cpu = smp_processor_id();
-	static unsigned long next[NR_CPUS];
+	__u64 msr_val;
 
 	ack_APIC_irq();
 
-	if (time_after(next[cpu], jiffies))
-		return;
-
-	next[cpu] = jiffies + HZ*5;
-	rdmsr(MSR_IA32_THERM_STATUS, l, h);
-	if (l & 0x1) {
-		printk(KERN_EMERG "CPU%d: Temperature above threshold\n", cpu);
-		printk(KERN_EMERG "CPU%d: Running in modulated clock mode\n",
-				cpu);
-		add_taint(TAINT_MACHINE_CHECK);
-	} else {
-		printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu);
-	}
+	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
+	therm_throt_process(msr_val & 0x1);
 }
 
 /* Thermal interrupt handler for this CPU setup */
@@ -122,7 +111,7 @@ static void intel_init_thermal(struct cpuinfo_x86 *c)
 	
 	rdmsr (MSR_IA32_MISC_ENABLE, l, h);
 	wrmsr (MSR_IA32_MISC_ENABLE, l | (1<<3), h);
-	
+
 	l = apic_read (APIC_LVTTHMR);
 	apic_write_around (APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk (KERN_INFO "CPU%d: Thermal monitoring enabled\n", cpu);
diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
new file mode 100644
index 000000000000..85eba00d680e
--- /dev/null
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -0,0 +1,58 @@
+/*
+ * linux/arch/i386/kerne/cpu/mcheck/therm_throt.c
+ *
+ * Thermal throttle event support code.
+ *
+ * Author: Dmitriy Zavin (dmitriyz@google.com)
+ *
+ * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.
+ *
+ */
+
+#include <linux/percpu.h>
+#include <linux/cpu.h>
+#include <asm/cpu.h>
+#include <linux/notifier.h>
+#include <asm/therm_throt.h>
+
+/* How long to wait between reporting thermal events */
+#define CHECK_INTERVAL              (300 * HZ)
+
+static DEFINE_PER_CPU(unsigned long, next_check);
+
+/***
+ * therm_throt_process - Process thermal throttling event
+ * @curr: Whether the condition is current or not (boolean), since the
+ *        thermal interrupt normally gets called both when the thermal
+ *        event begins and once the event has ended.
+ *
+ * This function is normally called by the thermal interrupt after the
+ * IRQ has been acknowledged.
+ *
+ * It will take care of rate limiting and printing messages to the syslog.
+ *
+ * Returns: 0 : Event should NOT be further logged, i.e. still in
+ *              "timeout" from previous log message.
+ *          1 : Event should be logged further, and a message has been
+ *              printed to the syslog.
+ */
+int therm_throt_process(int curr)
+{
+	unsigned int cpu = smp_processor_id();
+
+	if (time_before(jiffies, __get_cpu_var(next_check)))
+		return 0;
+
+	__get_cpu_var(next_check) = jiffies + CHECK_INTERVAL;
+
+	/* if we just entered the thermal event */
+	if (curr) {
+		printk(KERN_CRIT "CPU%d: Temperature above threshold, "
+		       "cpu clock throttled\n", cpu);
+		add_taint(TAINT_MACHINE_CHECK);
+	} else {
+		printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu);
+	}
+
+	return 1;
+}
diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 2466fbd035ee..3c7cbff04d3d 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -11,7 +11,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 		pci-dma.o pci-nommu.o alternative.o
 
 obj-$(CONFIG_STACKTRACE)	+= stacktrace.o
-obj-$(CONFIG_X86_MCE)         += mce.o
+obj-$(CONFIG_X86_MCE)		+= mce.o therm_throt.o
 obj-$(CONFIG_X86_MCE_INTEL)	+= mce_intel.o
 obj-$(CONFIG_X86_MCE_AMD)	+= mce_amd.o
 obj-$(CONFIG_MTRR)		+= ../../i386/kernel/cpu/mtrr/
@@ -46,6 +46,7 @@ obj-y				+= intel_cacheinfo.o
 
 CFLAGS_vsyscall.o		:= $(PROFILING) -g0
 
+therm_throt-y                   += ../../i386/kernel/cpu/mcheck/therm_throt.o
 bootflag-y			+= ../../i386/kernel/bootflag.o
 cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o
 topology-y                     += ../../i386/kernel/topology.o
@@ -55,4 +56,3 @@ quirks-y			+= ../../i386/kernel/quirks.o
 i8237-y				+= ../../i386/kernel/i8237.o
 msr-$(subst m,y,$(CONFIG_X86_MSR))  += ../../i386/kernel/msr.o
 alternative-y			+= ../../i386/kernel/alternative.o
-
diff --git a/arch/x86_64/kernel/mce.c b/arch/x86_64/kernel/mce.c
index 1a93c3738404..bbea88801d88 100644
--- a/arch/x86_64/kernel/mce.c
+++ b/arch/x86_64/kernel/mce.c
@@ -274,6 +274,33 @@ void do_machine_check(struct pt_regs * regs, long error_code)
 	atomic_dec(&mce_entry);
 }
 
+#ifdef CONFIG_X86_MCE_INTEL
+/***
+ * mce_log_therm_throt_event - Logs the thermal throttling event to mcelog
+ * @cpu: The CPU on which the event occured.
+ * @status: Event status information
+ *
+ * This function should be called by the thermal interrupt after the
+ * event has been processed and the decision was made to log the event
+ * further.
+ *
+ * The status parameter will be saved to the 'status' field of 'struct mce'
+ * and historically has been the register value of the
+ * MSR_IA32_THERMAL_STATUS (Intel) msr.
+ */
+void mce_log_therm_throt_event(unsigned int cpu, __u64 status)
+{
+	struct mce m;
+
+	memset(&m, 0, sizeof(m));
+	m.cpu = cpu;
+	m.bank = MCE_THERMAL_BANK;
+	m.status = status;
+	rdtscll(m.tsc);
+	mce_log(&m);
+}
+#endif /* CONFIG_X86_MCE_INTEL */
+
 /*
  * Periodic polling timer for "silent" machine check errors.
  */
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index 8f533d2c40cb..dec11219e278 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -11,36 +11,21 @@
 #include <asm/mce.h>
 #include <asm/hw_irq.h>
 #include <asm/idle.h>
-
-static DEFINE_PER_CPU(unsigned long, next_check);
+#include <asm/therm_throt.h>
 
 asmlinkage void smp_thermal_interrupt(void)
 {
-	struct mce m;
+	__u64 msr_val;
 
 	ack_APIC_irq();
 
 	exit_idle();
 	irq_enter();
-	if (time_before(jiffies, __get_cpu_var(next_check)))
-		goto done;
-
-	__get_cpu_var(next_check) = jiffies + HZ*300;
-	memset(&m, 0, sizeof(m));
-	m.cpu = smp_processor_id();
-	m.bank = MCE_THERMAL_BANK;
-	rdtscll(m.tsc);
-	rdmsrl(MSR_IA32_THERM_STATUS, m.status);
-	if (m.status & 0x1) {
-		printk(KERN_EMERG
-			"CPU%d: Temperature above threshold, cpu clock throttled\n", m.cpu);
-		add_taint(TAINT_MACHINE_CHECK);
-	} else {
-		printk(KERN_EMERG "CPU%d: Temperature/speed normal\n", m.cpu);
-	}
 
-	mce_log(&m);
-done:
+	rdmsrl(MSR_IA32_THERM_STATUS, msr_val);
+	if (therm_throt_process(msr_val & 1))
+		mce_log_therm_throt_event(smp_processor_id(), msr_val);
+
 	irq_exit();
 }
 
diff --git a/include/asm-i386/therm_throt.h b/include/asm-i386/therm_throt.h
new file mode 100644
index 000000000000..3c9c22cc10c9
--- /dev/null
+++ b/include/asm-i386/therm_throt.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_I386_THERM_THROT_H__
+#define __ASM_I386_THERM_THROT_H__ 1
+
+int therm_throt_process(int curr);
+
+#endif /* __ASM_I386_THERM_THROT_H__ */
diff --git a/include/asm-x86_64/mce.h b/include/asm-x86_64/mce.h
index d13687dfd691..5a11146d6d9c 100644
--- a/include/asm-x86_64/mce.h
+++ b/include/asm-x86_64/mce.h
@@ -99,6 +99,8 @@ static inline void mce_amd_feature_init(struct cpuinfo_x86 *c)
 }
 #endif
 
+void mce_log_therm_throt_event(unsigned int cpu, __u64 status);
+
 extern atomic_t mce_entry;
 
 #endif
diff --git a/include/asm-x86_64/therm_throt.h b/include/asm-x86_64/therm_throt.h
new file mode 100644
index 000000000000..5aac059007ba
--- /dev/null
+++ b/include/asm-x86_64/therm_throt.h
@@ -0,0 +1 @@
+#include <asm-i386/therm_throt.h>
-- 
cgit v1.2.1


From 66aea9913cf435fe92ebb7bf869b4f15901ab993 Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] i386: Make the jiffies compares use the 64bit safe macros.

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/mcheck/therm_throt.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
index 85eba00d680e..101f7ace00ce 100644
--- a/arch/i386/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -18,7 +18,7 @@
 /* How long to wait between reporting thermal events */
 #define CHECK_INTERVAL              (300 * HZ)
 
-static DEFINE_PER_CPU(unsigned long, next_check);
+static DEFINE_PER_CPU(__u64, next_check);
 
 /***
  * therm_throt_process - Process thermal throttling event
@@ -39,11 +39,12 @@ static DEFINE_PER_CPU(unsigned long, next_check);
 int therm_throt_process(int curr)
 {
 	unsigned int cpu = smp_processor_id();
+	__u64 tmp_jiffs = get_jiffies_64();
 
-	if (time_before(jiffies, __get_cpu_var(next_check)))
+	if (time_before64(tmp_jiffs, __get_cpu_var(next_check)))
 		return 0;
 
-	__get_cpu_var(next_check) = jiffies + CHECK_INTERVAL;
+	__get_cpu_var(next_check) = tmp_jiffs + CHECK_INTERVAL;
 
 	/* if we just entered the thermal event */
 	if (curr) {
-- 
cgit v1.2.1


From 3222b36f46c22f46697a0a53fa8804153a32669f Mon Sep 17 00:00:00 2001
From: Dmitriy Zavin <dmitriyz@google.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] x86: Add a cumulative thermal throttle event counter.

The counter is exported to /sys that keeps track of the
number of thermal events, such that the user knows how bad the
thermal problem might be (since the logging to syslog and mcelog
is rate limited).

AK: Fixed cpu hotplug locking

Signed-off-by: Dmitriy Zavin <dmitriyz@google.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/mcheck/p4.c          |   3 +
 arch/i386/kernel/cpu/mcheck/therm_throt.c | 133 ++++++++++++++++++++++++++++--
 arch/x86_64/kernel/mce_intel.c            |   3 +
 include/asm-i386/therm_throt.h            |   3 +
 4 files changed, 136 insertions(+), 6 deletions(-)

diff --git a/arch/i386/kernel/cpu/mcheck/p4.c b/arch/i386/kernel/cpu/mcheck/p4.c
index d83a669d376f..504434a46011 100644
--- a/arch/i386/kernel/cpu/mcheck/p4.c
+++ b/arch/i386/kernel/cpu/mcheck/p4.c
@@ -115,6 +115,9 @@ static void intel_init_thermal(struct cpuinfo_x86 *c)
 	l = apic_read (APIC_LVTTHMR);
 	apic_write_around (APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk (KERN_INFO "CPU%d: Thermal monitoring enabled\n", cpu);
+
+	/* enable thermal throttle processing */
+	atomic_set(&therm_throt_en, 1);
 	return;
 }
 #endif /* CONFIG_X86_MCE_P4THERMAL */
diff --git a/arch/i386/kernel/cpu/mcheck/therm_throt.c b/arch/i386/kernel/cpu/mcheck/therm_throt.c
index 101f7ace00ce..4f43047de406 100644
--- a/arch/i386/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/i386/kernel/cpu/mcheck/therm_throt.c
@@ -1,15 +1,22 @@
 /*
  * linux/arch/i386/kerne/cpu/mcheck/therm_throt.c
  *
- * Thermal throttle event support code.
+ * Thermal throttle event support code (such as syslog messaging and rate
+ * limiting) that was factored out from x86_64 (mce_intel.c) and i386 (p4.c).
+ * This allows consistent reporting of CPU thermal throttle events.
+ *
+ * Maintains a counter in /sys that keeps track of the number of thermal
+ * events, such that the user knows how bad the thermal problem might be
+ * (since the logging to syslog and mcelog is rate limited).
  *
  * Author: Dmitriy Zavin (dmitriyz@google.com)
  *
  * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c.
- *
+ *          Inspired by Ross Biro's and Al Borchers' counter code.
  */
 
 #include <linux/percpu.h>
+#include <linux/sysdev.h>
 #include <linux/cpu.h>
 #include <asm/cpu.h>
 #include <linux/notifier.h>
@@ -18,15 +25,53 @@
 /* How long to wait between reporting thermal events */
 #define CHECK_INTERVAL              (300 * HZ)
 
-static DEFINE_PER_CPU(__u64, next_check);
+static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES;
+static DEFINE_PER_CPU(unsigned long, thermal_throttle_count);
+atomic_t therm_throt_en = ATOMIC_INIT(0);
+
+#ifdef CONFIG_SYSFS
+#define define_therm_throt_sysdev_one_ro(_name)                              \
+        static SYSDEV_ATTR(_name, 0444, therm_throt_sysdev_show_##_name, NULL)
+
+#define define_therm_throt_sysdev_show_func(name)                            \
+static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev,        \
+                                              char *buf)                     \
+{                                                                            \
+	unsigned int cpu = dev->id;                                          \
+	ssize_t ret;                                                         \
+                                                                             \
+	preempt_disable();              /* CPU hotplug */                    \
+	if (cpu_online(cpu))                                                 \
+		ret = sprintf(buf, "%lu\n",                                  \
+			      per_cpu(thermal_throttle_##name, cpu));        \
+	else                                                                 \
+		ret = 0;                                                     \
+	preempt_enable();                                                    \
+                                                                             \
+	return ret;                                                          \
+}
+
+define_therm_throt_sysdev_show_func(count);
+define_therm_throt_sysdev_one_ro(count);
+
+static struct attribute *thermal_throttle_attrs[] = {
+	&attr_count.attr,
+	NULL
+};
+
+static struct attribute_group thermal_throttle_attr_group = {
+	.attrs = thermal_throttle_attrs,
+	.name = "thermal_throttle"
+};
+#endif /* CONFIG_SYSFS */
 
 /***
- * therm_throt_process - Process thermal throttling event
+ * therm_throt_process - Process thermal throttling event from interrupt
  * @curr: Whether the condition is current or not (boolean), since the
  *        thermal interrupt normally gets called both when the thermal
  *        event begins and once the event has ended.
  *
- * This function is normally called by the thermal interrupt after the
+ * This function is called by the thermal interrupt after the
  * IRQ has been acknowledged.
  *
  * It will take care of rate limiting and printing messages to the syslog.
@@ -41,6 +86,9 @@ int therm_throt_process(int curr)
 	unsigned int cpu = smp_processor_id();
 	__u64 tmp_jiffs = get_jiffies_64();
 
+	if (curr)
+		__get_cpu_var(thermal_throttle_count)++;
+
 	if (time_before64(tmp_jiffs, __get_cpu_var(next_check)))
 		return 0;
 
@@ -49,7 +97,9 @@ int therm_throt_process(int curr)
 	/* if we just entered the thermal event */
 	if (curr) {
 		printk(KERN_CRIT "CPU%d: Temperature above threshold, "
-		       "cpu clock throttled\n", cpu);
+		       "cpu clock throttled (total events = %lu)\n", cpu,
+		       __get_cpu_var(thermal_throttle_count));
+
 		add_taint(TAINT_MACHINE_CHECK);
 	} else {
 		printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu);
@@ -57,3 +107,74 @@ int therm_throt_process(int curr)
 
 	return 1;
 }
+
+#ifdef CONFIG_SYSFS
+/* Add/Remove thermal_throttle interface for CPU device */
+static __cpuinit int thermal_throttle_add_dev(struct sys_device * sys_dev)
+{
+	sysfs_create_group(&sys_dev->kobj, &thermal_throttle_attr_group);
+	return 0;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+static __cpuinit int thermal_throttle_remove_dev(struct sys_device * sys_dev)
+{
+	sysfs_remove_group(&sys_dev->kobj, &thermal_throttle_attr_group);
+	return 0;
+}
+
+/* Mutex protecting device creation against CPU hotplug */
+static DEFINE_MUTEX(therm_cpu_lock);
+
+/* Get notified when a cpu comes on/off. Be hotplug friendly. */
+static __cpuinit int thermal_throttle_cpu_callback(struct notifier_block *nfb,
+						   unsigned long action,
+						   void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+	struct sys_device *sys_dev;
+
+	sys_dev = get_cpu_sysdev(cpu);
+	mutex_lock(&therm_cpu_lock);
+	switch (action) {
+	case CPU_ONLINE:
+		thermal_throttle_add_dev(sys_dev);
+		break;
+	case CPU_DEAD:
+		thermal_throttle_remove_dev(sys_dev);
+		break;
+	}
+	mutex_unlock(&therm_cpu_lock);
+	return NOTIFY_OK;
+}
+
+static struct notifier_block thermal_throttle_cpu_notifier =
+{
+	.notifier_call = thermal_throttle_cpu_callback,
+};
+#endif /* CONFIG_HOTPLUG_CPU */
+
+static __init int thermal_throttle_init_device(void)
+{
+	unsigned int cpu = 0;
+
+	if (!atomic_read(&therm_throt_en))
+		return 0;
+
+	register_hotcpu_notifier(&thermal_throttle_cpu_notifier);
+
+#ifdef CONFIG_HOTPLUG_CPU
+	mutex_lock(&therm_cpu_lock);
+#endif
+	/* connect live CPUs to sysfs */
+	for_each_online_cpu(cpu)
+		thermal_throttle_add_dev(get_cpu_sysdev(cpu));
+#ifdef CONFIG_HOTPLUG_CPU
+	mutex_unlock(&therm_cpu_lock);
+#endif
+
+	return 0;
+}
+
+device_initcall(thermal_throttle_init_device);
+#endif /* CONFIG_SYSFS */
diff --git a/arch/x86_64/kernel/mce_intel.c b/arch/x86_64/kernel/mce_intel.c
index dec11219e278..6551505d8a2c 100644
--- a/arch/x86_64/kernel/mce_intel.c
+++ b/arch/x86_64/kernel/mce_intel.c
@@ -77,6 +77,9 @@ static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c)
 	apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED);
 	printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n",
 		cpu, tm2 ? "TM2" : "TM1");
+
+	/* enable thermal throttle processing */
+	atomic_set(&therm_throt_en, 1);
 	return;
 }
 
diff --git a/include/asm-i386/therm_throt.h b/include/asm-i386/therm_throt.h
index 3c9c22cc10c9..399bf6026b16 100644
--- a/include/asm-i386/therm_throt.h
+++ b/include/asm-i386/therm_throt.h
@@ -1,6 +1,9 @@
 #ifndef __ASM_I386_THERM_THROT_H__
 #define __ASM_I386_THERM_THROT_H__ 1
 
+#include <asm/atomic.h>
+
+extern atomic_t therm_throt_en;
 int therm_throt_process(int curr);
 
 #endif /* __ASM_I386_THERM_THROT_H__ */
-- 
cgit v1.2.1


From dcf10307c3fff2bca4b104ad8fe4e3aa8dcb2ad1 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] i386/x86-64: New Intel feature flags

Add supplemental SSE3 instructions flag, and Direct Cache Access flag.
As described in "Intel Processor idenfication and the CPUID instruction
AP485 Sept 2006"

AK: also added for x86-64

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/proc.c | 4 ++--
 arch/x86_64/kernel/setup.c  | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c
index f54a15268ed7..76aac088a323 100644
--- a/arch/i386/kernel/cpu/proc.c
+++ b/arch/i386/kernel/cpu/proc.c
@@ -46,8 +46,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 
 		/* Intel-defined (#2) */
 		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
-		"tm2", NULL, "cid", NULL, NULL, "cx16", "xtpr", NULL,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
+		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* VIA/Cyrix/Centaur-defined */
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 3d8309b1236e..3c7ad267d8a7 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -1071,8 +1071,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 
 		/* Intel-defined (#2) */
 		"pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
-		"tm2", NULL, "cid", NULL, NULL, "cx16", "xtpr", NULL,
-		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+		"tm2", "ssse3", "cid", NULL, NULL, "cx16", "xtpr", NULL,
+		NULL, NULL, "dca", NULL, NULL, NULL, NULL, NULL,
 		NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
 
 		/* VIA/Cyrix/Centaur-defined */
-- 
cgit v1.2.1


From 3f75f42d7733e73aca5c78326489efd4189e0111 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Tue, 26 Sep 2006 10:52:42 +0200
Subject: [PATCH] Don't set calgary iommu as default y

Most systems don't need it.

Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/x86_64/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index c2c68b902347..294bf77f7222 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -422,7 +422,6 @@ config IOMMU
 
 config CALGARY_IOMMU
 	bool "IBM Calgary IOMMU support"
-	default y
 	select SWIOTLB
 	depends on PCI && EXPERIMENTAL
 	help
-- 
cgit v1.2.1


From 0a2966b48fb784e437520e400ddc94874ddbd4e8 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <christoph@sgi.com>
Date: Mon, 25 Sep 2006 23:30:51 -0700
Subject: [PATCH] Fix longstanding load balancing bug in the scheduler

The scheduler will stop load balancing if the most busy processor contains
processes pinned via processor affinity.

The scheduler currently only does one search for busiest cpu.  If it cannot
pull any tasks away from the busiest cpu because they were pinned then the
scheduler goes into a corner and sulks leaving the idle processors idle.

F.e.  If you have processor 0 busy running four tasks pinned via taskset,
there are none on processor 1 and one just started two processes on
processor 2 then the scheduler will not move one of the two processes away
from processor 2.

This patch fixes that issue by forcing the scheduler to come out of its
corner and retrying the load balancing by considering other processors for
load balancing.

This patch was originally developed by John Hawkes and discussed at

    http://marc.theaimsgroup.com/?l=linux-kernel&m=113901368523205&w=2.

I have removed extraneous material and gone back to equipping struct rq
with the cpu the queue is associated with since this makes the patch much
easier and it is likely that others in the future will have the same
difficulty of figuring out which processor owns which runqueue.

The overhead added through these patches is a single word on the stack if
the kernel is configured to support 32 cpus or less (32 bit).  For 32 bit
environments the maximum number of cpus that can be configued is 255 which
would result in the use of 32 bytes additional on the stack.  On IA64 up to
1k cpus can be configured which will result in the use of 128 additional
bytes on the stack.  The maximum additional cache footprint is one
cacheline.  Typically memory use will be much less than a cacheline and the
additional cpumask will be placed on the stack in a cacheline that already
contains other local variable.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: John Hawkes <hawkes@sgi.com>
Cc: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Peter Williams <pwil3058@bigpond.net.au>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/sched.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 46 insertions(+), 8 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index a234fbee1238..5c848fd4e461 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -238,6 +238,7 @@ struct rq {
 	/* For active balancing */
 	int active_balance;
 	int push_cpu;
+	int cpu;		/* cpu of this runqueue */
 
 	struct task_struct *migration_thread;
 	struct list_head migration_queue;
@@ -267,6 +268,15 @@ struct rq {
 
 static DEFINE_PER_CPU(struct rq, runqueues);
 
+static inline int cpu_of(struct rq *rq)
+{
+#ifdef CONFIG_SMP
+	return rq->cpu;
+#else
+	return 0;
+#endif
+}
+
 /*
  * The domain tree (rq->sd) is protected by RCU's quiescent state transition.
  * See detach_destroy_domains: synchronize_sched for details.
@@ -2211,7 +2221,8 @@ out:
  */
 static struct sched_group *
 find_busiest_group(struct sched_domain *sd, int this_cpu,
-		   unsigned long *imbalance, enum idle_type idle, int *sd_idle)
+		   unsigned long *imbalance, enum idle_type idle, int *sd_idle,
+		   cpumask_t *cpus)
 {
 	struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups;
 	unsigned long max_load, avg_load, total_load, this_load, total_pwr;
@@ -2248,7 +2259,12 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 		sum_weighted_load = sum_nr_running = avg_load = 0;
 
 		for_each_cpu_mask(i, group->cpumask) {
-			struct rq *rq = cpu_rq(i);
+			struct rq *rq;
+
+			if (!cpu_isset(i, *cpus))
+				continue;
+
+			rq = cpu_rq(i);
 
 			if (*sd_idle && !idle_cpu(i))
 				*sd_idle = 0;
@@ -2466,13 +2482,17 @@ ret:
  */
 static struct rq *
 find_busiest_queue(struct sched_group *group, enum idle_type idle,
-		   unsigned long imbalance)
+		   unsigned long imbalance, cpumask_t *cpus)
 {
 	struct rq *busiest = NULL, *rq;
 	unsigned long max_load = 0;
 	int i;
 
 	for_each_cpu_mask(i, group->cpumask) {
+
+		if (!cpu_isset(i, *cpus))
+			continue;
+
 		rq = cpu_rq(i);
 
 		if (rq->nr_running == 1 && rq->raw_weighted_load > imbalance)
@@ -2511,6 +2531,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 	struct sched_group *group;
 	unsigned long imbalance;
 	struct rq *busiest;
+	cpumask_t cpus = CPU_MASK_ALL;
 
 	if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER &&
 	    !sched_smt_power_savings)
@@ -2518,13 +2539,15 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 
 	schedstat_inc(sd, lb_cnt[idle]);
 
-	group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle);
+redo:
+	group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle,
+							&cpus);
 	if (!group) {
 		schedstat_inc(sd, lb_nobusyg[idle]);
 		goto out_balanced;
 	}
 
-	busiest = find_busiest_queue(group, idle, imbalance);
+	busiest = find_busiest_queue(group, idle, imbalance, &cpus);
 	if (!busiest) {
 		schedstat_inc(sd, lb_nobusyq[idle]);
 		goto out_balanced;
@@ -2549,8 +2572,12 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 		double_rq_unlock(this_rq, busiest);
 
 		/* All tasks on this runqueue were pinned by CPU affinity */
-		if (unlikely(all_pinned))
+		if (unlikely(all_pinned)) {
+			cpu_clear(cpu_of(busiest), cpus);
+			if (!cpus_empty(cpus))
+				goto redo;
 			goto out_balanced;
+		}
 	}
 
 	if (!nr_moved) {
@@ -2639,18 +2666,22 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd)
 	unsigned long imbalance;
 	int nr_moved = 0;
 	int sd_idle = 0;
+	cpumask_t cpus = CPU_MASK_ALL;
 
 	if (sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings)
 		sd_idle = 1;
 
 	schedstat_inc(sd, lb_cnt[NEWLY_IDLE]);
-	group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, &sd_idle);
+redo:
+	group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE,
+				&sd_idle, &cpus);
 	if (!group) {
 		schedstat_inc(sd, lb_nobusyg[NEWLY_IDLE]);
 		goto out_balanced;
 	}
 
-	busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance);
+	busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance,
+				&cpus);
 	if (!busiest) {
 		schedstat_inc(sd, lb_nobusyq[NEWLY_IDLE]);
 		goto out_balanced;
@@ -2668,6 +2699,12 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd)
 					minus_1_or_zero(busiest->nr_running),
 					imbalance, sd, NEWLY_IDLE, NULL);
 		spin_unlock(&busiest->lock);
+
+		if (!nr_moved) {
+			cpu_clear(cpu_of(busiest), cpus);
+			if (!cpus_empty(cpus))
+				goto redo;
+		}
 	}
 
 	if (!nr_moved) {
@@ -6747,6 +6784,7 @@ void __init sched_init(void)
 			rq->cpu_load[j] = 0;
 		rq->active_balance = 0;
 		rq->push_cpu = 0;
+		rq->cpu = i;
 		rq->migration_thread = NULL;
 		INIT_LIST_HEAD(&rq->migration_queue);
 #endif
-- 
cgit v1.2.1


From 632bbfeee4f042c05bc65150b4433a297d3fe387 Mon Sep 17 00:00:00 2001
From: Jan Blunck <jblunck@suse.de>
Date: Mon, 25 Sep 2006 23:30:53 -0700
Subject: [PATCH] trigger a syntax error if percpu macros are incorrectly used

get_cpu_var()/per_cpu()/__get_cpu_var() arguments must be simple
identifiers.  Otherwise the arch dependent implementations might break.

This patch enforces the correct usage of the macros by producing a syntax
error if the variable is not a simple identifier.

Signed-off-by: Jan Blunck <jblunck@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/percpu.h |  4 +++-
 include/asm-s390/percpu.h    | 20 +++++++++++---------
 include/asm-x86_64/percpu.h  | 12 +++++++++---
 include/linux/percpu.h       | 10 ++++++++--
 4 files changed, 31 insertions(+), 15 deletions(-)

diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index e160e04290fb..6d45ee5472af 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -14,7 +14,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
     __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
 
 /* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]))
+#define per_cpu(var, cpu) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]); }))
 #define __get_cpu_var(var) per_cpu(var, smp_processor_id())
 #define __raw_get_cpu_var(var) per_cpu(var, raw_smp_processor_id())
 
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 28b3517e787c..495ad99c7635 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -15,18 +15,20 @@
  */
 #if defined(__s390x__) && defined(MODULE)
 
-#define __reloc_hide(var,offset) \
-  (*({ unsigned long *__ptr; \
-       asm ( "larl %0,per_cpu__"#var"@GOTENT" \
-             : "=a" (__ptr) : "X" (per_cpu__##var) ); \
-       (typeof(&per_cpu__##var))((*__ptr) + (offset)); }))
+#define __reloc_hide(var,offset) (*({			\
+	extern int simple_indentifier_##var(void);	\
+	unsigned long *__ptr;				\
+	asm ( "larl %0,per_cpu__"#var"@GOTENT"		\
+	    : "=a" (__ptr) : "X" (per_cpu__##var) );	\
+	(typeof(&per_cpu__##var))((*__ptr) + (offset));	}))
 
 #else
 
-#define __reloc_hide(var, offset) \
-  (*({ unsigned long __ptr; \
-       asm ( "" : "=a" (__ptr) : "0" (&per_cpu__##var) ); \
-       (typeof(&per_cpu__##var)) (__ptr + (offset)); }))
+#define __reloc_hide(var, offset) (*({				\
+	extern int simple_indentifier_##var(void);		\
+	unsigned long __ptr;					\
+	asm ( "" : "=a" (__ptr) : "0" (&per_cpu__##var) );	\
+	(typeof(&per_cpu__##var)) (__ptr + (offset)); }))
 
 #endif
 
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 08dd9f9dda81..bffb2f886a51 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -21,9 +21,15 @@
     __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
 
 /* var is in discarded region: offset to particular copy we want */
-#define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)))
-#define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()))
-#define __raw_get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()))
+#define per_cpu(var, cpu) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu)); }))
+#define __get_cpu_var(var) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
+#define __raw_get_cpu_var(var) (*({			\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __my_cpu_offset()); }))
 
 /* A macro to avoid #include hell... */
 #define percpu_modcopy(pcpudst, src, size)			\
diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index cb9039a21f2a..f926490a7d8b 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -11,8 +11,14 @@
 #define PERCPU_ENOUGH_ROOM 32768
 #endif
 
-/* Must be an lvalue. */
-#define get_cpu_var(var) (*({ preempt_disable(); &__get_cpu_var(var); }))
+/*
+ * Must be an lvalue. Since @var must be a simple identifier,
+ * we force a syntax error here if it isn't.
+ */
+#define get_cpu_var(var) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	preempt_disable();				\
+	&__get_cpu_var(var); }))
 #define put_cpu_var(var) preempt_enable()
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.1


From 3998b9301d3d55be8373add22b6bc5e11c1d9b71 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Mon, 25 Sep 2006 23:30:53 -0700
Subject: [PATCH] jbd: fix commit of ordered data buffers

Original commit code assumes, that when a buffer on BJ_SyncData list is
locked, it is being written to disk.  But this is not true and hence it can
lead to a potential data loss on crash.  Also the code didn't count with
the fact that journal_dirty_data() can steal buffers from committing
transaction and hence could write buffers that no longer belong to the
committing transaction.  Finally it could possibly happen that we tried
writing out one buffer several times.

The patch below tries to solve these problems by a complete rewrite of the
data commit code.  We go through buffers on t_sync_datalist, lock buffers
needing write out and store them in an array.  Buffers are also immediately
refiled to BJ_Locked list or unfiled (if the write out is completed).  When
the array is full or we have to block on buffer lock, we submit all
accumulated buffers for IO.

[suitable for 2.6.18.x around the 2.6.19-rc2 timeframe]

Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/commit.c | 182 +++++++++++++++++++++++++++++++++++---------------------
 1 file changed, 113 insertions(+), 69 deletions(-)

diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 42da60784311..32a8caf0c41e 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -160,6 +160,117 @@ static int journal_write_commit_record(journal_t *journal,
 	return (ret == -EIO);
 }
 
+static void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
+{
+	int i;
+
+	for (i = 0; i < bufs; i++) {
+		wbuf[i]->b_end_io = end_buffer_write_sync;
+		/* We use-up our safety reference in submit_bh() */
+		submit_bh(WRITE, wbuf[i]);
+	}
+}
+
+/*
+ *  Submit all the data buffers to disk
+ */
+static void journal_submit_data_buffers(journal_t *journal,
+				transaction_t *commit_transaction)
+{
+	struct journal_head *jh;
+	struct buffer_head *bh;
+	int locked;
+	int bufs = 0;
+	struct buffer_head **wbuf = journal->j_wbuf;
+
+	/*
+	 * Whenever we unlock the journal and sleep, things can get added
+	 * onto ->t_sync_datalist, so we have to keep looping back to
+	 * write_out_data until we *know* that the list is empty.
+	 *
+	 * Cleanup any flushed data buffers from the data list.  Even in
+	 * abort mode, we want to flush this out as soon as possible.
+	 */
+write_out_data:
+	cond_resched();
+	spin_lock(&journal->j_list_lock);
+
+	while (commit_transaction->t_sync_datalist) {
+		jh = commit_transaction->t_sync_datalist;
+		bh = jh2bh(jh);
+		locked = 0;
+
+		/* Get reference just to make sure buffer does not disappear
+		 * when we are forced to drop various locks */
+		get_bh(bh);
+		/* If the buffer is dirty, we need to submit IO and hence
+		 * we need the buffer lock. We try to lock the buffer without
+		 * blocking. If we fail, we need to drop j_list_lock and do
+		 * blocking lock_buffer().
+		 */
+		if (buffer_dirty(bh)) {
+			if (test_set_buffer_locked(bh)) {
+				BUFFER_TRACE(bh, "needs blocking lock");
+				spin_unlock(&journal->j_list_lock);
+				/* Write out all data to prevent deadlocks */
+				journal_do_submit_data(wbuf, bufs);
+				bufs = 0;
+				lock_buffer(bh);
+				spin_lock(&journal->j_list_lock);
+			}
+			locked = 1;
+		}
+		/* We have to get bh_state lock. Again out of order, sigh. */
+		if (!inverted_lock(journal, bh)) {
+			jbd_lock_bh_state(bh);
+			spin_lock(&journal->j_list_lock);
+		}
+		/* Someone already cleaned up the buffer? */
+		if (!buffer_jbd(bh)
+			|| jh->b_transaction != commit_transaction
+			|| jh->b_jlist != BJ_SyncData) {
+			jbd_unlock_bh_state(bh);
+			if (locked)
+				unlock_buffer(bh);
+			BUFFER_TRACE(bh, "already cleaned up");
+			put_bh(bh);
+			continue;
+		}
+		if (locked && test_clear_buffer_dirty(bh)) {
+			BUFFER_TRACE(bh, "needs writeout, adding to array");
+			wbuf[bufs++] = bh;
+			__journal_file_buffer(jh, commit_transaction,
+						BJ_Locked);
+			jbd_unlock_bh_state(bh);
+			if (bufs == journal->j_wbufsize) {
+				spin_unlock(&journal->j_list_lock);
+				journal_do_submit_data(wbuf, bufs);
+				bufs = 0;
+				goto write_out_data;
+			}
+		}
+		else {
+			BUFFER_TRACE(bh, "writeout complete: unfile");
+			__journal_unfile_buffer(jh);
+			jbd_unlock_bh_state(bh);
+			if (locked)
+				unlock_buffer(bh);
+			journal_remove_journal_head(bh);
+			/* Once for our safety reference, once for
+			 * journal_remove_journal_head() */
+			put_bh(bh);
+			put_bh(bh);
+		}
+
+		if (lock_need_resched(&journal->j_list_lock)) {
+			spin_unlock(&journal->j_list_lock);
+			goto write_out_data;
+		}
+	}
+	spin_unlock(&journal->j_list_lock);
+	journal_do_submit_data(wbuf, bufs);
+}
+
 /*
  * journal_commit_transaction
  *
@@ -313,80 +424,13 @@ void journal_commit_transaction(journal_t *journal)
 	 * Now start flushing things to disk, in the order they appear
 	 * on the transaction lists.  Data blocks go first.
 	 */
-
 	err = 0;
-	/*
-	 * Whenever we unlock the journal and sleep, things can get added
-	 * onto ->t_sync_datalist, so we have to keep looping back to
-	 * write_out_data until we *know* that the list is empty.
-	 */
-	bufs = 0;
-	/*
-	 * Cleanup any flushed data buffers from the data list.  Even in
-	 * abort mode, we want to flush this out as soon as possible.
-	 */
-write_out_data:
-	cond_resched();
-	spin_lock(&journal->j_list_lock);
-
-	while (commit_transaction->t_sync_datalist) {
-		struct buffer_head *bh;
-
-		jh = commit_transaction->t_sync_datalist;
-		commit_transaction->t_sync_datalist = jh->b_tnext;
-		bh = jh2bh(jh);
-		if (buffer_locked(bh)) {
-			BUFFER_TRACE(bh, "locked");
-			if (!inverted_lock(journal, bh))
-				goto write_out_data;
-			__journal_temp_unlink_buffer(jh);
-			__journal_file_buffer(jh, commit_transaction,
-						BJ_Locked);
-			jbd_unlock_bh_state(bh);
-			if (lock_need_resched(&journal->j_list_lock)) {
-				spin_unlock(&journal->j_list_lock);
-				goto write_out_data;
-			}
-		} else {
-			if (buffer_dirty(bh)) {
-				BUFFER_TRACE(bh, "start journal writeout");
-				get_bh(bh);
-				wbuf[bufs++] = bh;
-				if (bufs == journal->j_wbufsize) {
-					jbd_debug(2, "submit %d writes\n",
-							bufs);
-					spin_unlock(&journal->j_list_lock);
-					ll_rw_block(SWRITE, bufs, wbuf);
-					journal_brelse_array(wbuf, bufs);
-					bufs = 0;
-					goto write_out_data;
-				}
-			} else {
-				BUFFER_TRACE(bh, "writeout complete: unfile");
-				if (!inverted_lock(journal, bh))
-					goto write_out_data;
-				__journal_unfile_buffer(jh);
-				jbd_unlock_bh_state(bh);
-				journal_remove_journal_head(bh);
-				put_bh(bh);
-				if (lock_need_resched(&journal->j_list_lock)) {
-					spin_unlock(&journal->j_list_lock);
-					goto write_out_data;
-				}
-			}
-		}
-	}
-
-	if (bufs) {
-		spin_unlock(&journal->j_list_lock);
-		ll_rw_block(SWRITE, bufs, wbuf);
-		journal_brelse_array(wbuf, bufs);
-		spin_lock(&journal->j_list_lock);
-	}
+	journal_submit_data_buffers(journal, commit_transaction);
 
 	/*
 	 * Wait for all previously submitted IO to complete.
 	 */
+	spin_lock(&journal->j_list_lock);
 	while (commit_transaction->t_locked_list) {
 		struct buffer_head *bh;
 
-- 
cgit v1.2.1


From a6ca1b99ed434f3fb41bbed647ed36c0420501e5 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Mon, 25 Sep 2006 23:30:55 -0700
Subject: [PATCH] update to the kernel kmap/kunmap API

Give non-highmem architectures access to the kmap API for the purposes of
overriding (this is what the attached patch does).

The proposal is that we should now require all architectures with coherence
issues to manage data coherence via the kmap/kunmap API.  Thus driver
writers never have to write code like

    kmap(page)
    modify data in page
    flush_kernel_dcache_page(page)
    kunmap(page)

instead, kmap/kunmap will manage the coherence and driver (and filesystem)
writers don't need to worry about how to flush between kmap and kunmap.

For most architectures, the page only needs to be flushed if it was
actually written to *and* there are user mappings of it, so the best
implementation looks to be: clear the page dirty pte bit in the kernel page
tables on kmap and on kunmap, check page->mappings for user maps, and then
the dirty bit, and only flush if it both has user mappings and is dirty.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/highmem.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 85ce7ef9a512..42620e723abb 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -29,6 +29,7 @@ unsigned int nr_free_highpages(void);
 
 static inline unsigned int nr_free_highpages(void) { return 0; }
 
+#ifndef ARCH_HAS_KMAP
 static inline void *kmap(struct page *page)
 {
 	might_sleep();
@@ -41,6 +42,7 @@ static inline void *kmap(struct page *page)
 #define kunmap_atomic(addr, idx)	do { } while (0)
 #define kmap_atomic_pfn(pfn, idx)	page_address(pfn_to_page(pfn))
 #define kmap_atomic_to_page(ptr)	virt_to_page(ptr)
+#endif
 
 #endif /* CONFIG_HIGHMEM */
 
-- 
cgit v1.2.1


From 725d704ecaca4a43f067092c140d4f3271cf2856 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:30:55 -0700
Subject: [PATCH] mm: VM_BUG_ON

Introduce a VM_BUG_ON, which is turned on with CONFIG_DEBUG_VM.  Use this
in the lightweight, inline refcounting functions; PageLRU and PageActive
checks in vmscan, because they're pretty well confined to vmscan.  And in
page allocate/free fastpaths which can be the hottest parts of the kernel
for kbuilds.

Unlike BUG_ON, VM_BUG_ON must not be used to execute statements with
side-effects, and should not be used outside core mm code.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h | 10 +++++++++-
 mm/internal.h      |  4 ++--
 mm/page_alloc.c    | 23 +++++++++++------------
 mm/swap.c          | 12 ++++++------
 mm/vmscan.c        | 16 ++++++++--------
 5 files changed, 36 insertions(+), 29 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 224178a000d2..7d20b25c58fc 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -278,6 +278,12 @@ struct page {
  */
 #include <linux/page-flags.h>
 
+#ifdef CONFIG_DEBUG_VM
+#define VM_BUG_ON(cond) BUG_ON(cond)
+#else
+#define VM_BUG_ON(condition) do { } while(0)
+#endif
+
 /*
  * Methods to modify the page usage count.
  *
@@ -297,7 +303,7 @@ struct page {
  */
 static inline int put_page_testzero(struct page *page)
 {
-	BUG_ON(atomic_read(&page->_count) == 0);
+	VM_BUG_ON(atomic_read(&page->_count) == 0);
 	return atomic_dec_and_test(&page->_count);
 }
 
@@ -307,6 +313,7 @@ static inline int put_page_testzero(struct page *page)
  */
 static inline int get_page_unless_zero(struct page *page)
 {
+	VM_BUG_ON(PageCompound(page));
 	return atomic_inc_not_zero(&page->_count);
 }
 
@@ -323,6 +330,7 @@ static inline void get_page(struct page *page)
 {
 	if (unlikely(PageCompound(page)))
 		page = (struct page *)page_private(page);
+	VM_BUG_ON(atomic_read(&page->_count) == 0);
 	atomic_inc(&page->_count);
 }
 
diff --git a/mm/internal.h b/mm/internal.h
index d20e3cc4aef0..d527b80b292f 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -24,8 +24,8 @@ static inline void set_page_count(struct page *page, int v)
  */
 static inline void set_page_refcounted(struct page *page)
 {
-	BUG_ON(PageCompound(page) && page_private(page) != (unsigned long)page);
-	BUG_ON(atomic_read(&page->_count));
+	VM_BUG_ON(PageCompound(page) && page_private(page) != (unsigned long)page);
+	VM_BUG_ON(atomic_read(&page->_count));
 	set_page_count(page, 1);
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8a52ba9fe693..4b33878e9488 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -127,7 +127,6 @@ static int bad_range(struct zone *zone, struct page *page)
 
 	return 0;
 }
-
 #else
 static inline int bad_range(struct zone *zone, struct page *page)
 {
@@ -218,12 +217,12 @@ static inline void prep_zero_page(struct page *page, int order, gfp_t gfp_flags)
 {
 	int i;
 
-	BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM);
+	VM_BUG_ON((gfp_flags & (__GFP_WAIT | __GFP_HIGHMEM)) == __GFP_HIGHMEM);
 	/*
 	 * clear_highpage() will use KM_USER0, so it's a bug to use __GFP_ZERO
 	 * and __GFP_HIGHMEM from hard or soft interrupt context.
 	 */
-	BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt());
+	VM_BUG_ON((gfp_flags & __GFP_HIGHMEM) && in_interrupt());
 	for (i = 0; i < (1 << order); i++)
 		clear_highpage(page + i);
 }
@@ -347,8 +346,8 @@ static inline void __free_one_page(struct page *page,
 
 	page_idx = page_to_pfn(page) & ((1 << MAX_ORDER) - 1);
 
-	BUG_ON(page_idx & (order_size - 1));
-	BUG_ON(bad_range(zone, page));
+	VM_BUG_ON(page_idx & (order_size - 1));
+	VM_BUG_ON(bad_range(zone, page));
 
 	zone->free_pages += order_size;
 	while (order < MAX_ORDER-1) {
@@ -421,7 +420,7 @@ static void free_pages_bulk(struct zone *zone, int count,
 	while (count--) {
 		struct page *page;
 
-		BUG_ON(list_empty(list));
+		VM_BUG_ON(list_empty(list));
 		page = list_entry(list->prev, struct page, lru);
 		/* have to delete it as __free_one_page list manipulates */
 		list_del(&page->lru);
@@ -512,7 +511,7 @@ static inline void expand(struct zone *zone, struct page *page,
 		area--;
 		high--;
 		size >>= 1;
-		BUG_ON(bad_range(zone, &page[size]));
+		VM_BUG_ON(bad_range(zone, &page[size]));
 		list_add(&page[size].lru, &area->free_list);
 		area->nr_free++;
 		set_page_order(&page[size], high);
@@ -761,8 +760,8 @@ void split_page(struct page *page, unsigned int order)
 {
 	int i;
 
-	BUG_ON(PageCompound(page));
-	BUG_ON(!page_count(page));
+	VM_BUG_ON(PageCompound(page));
+	VM_BUG_ON(!page_count(page));
 	for (i = 1; i < (1 << order); i++)
 		set_page_refcounted(page + i);
 }
@@ -809,7 +808,7 @@ again:
 	local_irq_restore(flags);
 	put_cpu();
 
-	BUG_ON(bad_range(zone, page));
+	VM_BUG_ON(bad_range(zone, page));
 	if (prep_new_page(page, order, gfp_flags))
 		goto again;
 	return page;
@@ -1083,7 +1082,7 @@ fastcall unsigned long get_zeroed_page(gfp_t gfp_mask)
 	 * get_zeroed_page() returns a 32-bit address, which cannot represent
 	 * a highmem page
 	 */
-	BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0);
+	VM_BUG_ON((gfp_mask & __GFP_HIGHMEM) != 0);
 
 	page = alloc_pages(gfp_mask | __GFP_ZERO, 0);
 	if (page)
@@ -1116,7 +1115,7 @@ EXPORT_SYMBOL(__free_pages);
 fastcall void free_pages(unsigned long addr, unsigned int order)
 {
 	if (addr != 0) {
-		BUG_ON(!virt_addr_valid((void *)addr));
+		VM_BUG_ON(!virt_addr_valid((void *)addr));
 		__free_pages(virt_to_page((void *)addr), order);
 	}
 }
diff --git a/mm/swap.c b/mm/swap.c
index 687686a61f7c..600235e43704 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -233,7 +233,7 @@ void fastcall __page_cache_release(struct page *page)
 		struct zone *zone = page_zone(page);
 
 		spin_lock_irqsave(&zone->lru_lock, flags);
-		BUG_ON(!PageLRU(page));
+		VM_BUG_ON(!PageLRU(page));
 		__ClearPageLRU(page);
 		del_page_from_lru(zone, page);
 		spin_unlock_irqrestore(&zone->lru_lock, flags);
@@ -284,7 +284,7 @@ void release_pages(struct page **pages, int nr, int cold)
 				zone = pagezone;
 				spin_lock_irq(&zone->lru_lock);
 			}
-			BUG_ON(!PageLRU(page));
+			VM_BUG_ON(!PageLRU(page));
 			__ClearPageLRU(page);
 			del_page_from_lru(zone, page);
 		}
@@ -337,7 +337,7 @@ void __pagevec_release_nonlru(struct pagevec *pvec)
 	for (i = 0; i < pagevec_count(pvec); i++) {
 		struct page *page = pvec->pages[i];
 
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		if (put_page_testzero(page))
 			pagevec_add(&pages_to_free, page);
 	}
@@ -364,7 +364,7 @@ void __pagevec_lru_add(struct pagevec *pvec)
 			zone = pagezone;
 			spin_lock_irq(&zone->lru_lock);
 		}
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
 		add_page_to_inactive_list(zone, page);
 	}
@@ -391,9 +391,9 @@ void __pagevec_lru_add_active(struct pagevec *pvec)
 			zone = pagezone;
 			spin_lock_irq(&zone->lru_lock);
 		}
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(PageActive(page));
+		VM_BUG_ON(PageActive(page));
 		SetPageActive(page);
 		add_page_to_active_list(zone, page);
 	}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5d4c4d02254d..41a3da3d6ccc 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -440,7 +440,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 		if (TestSetPageLocked(page))
 			goto keep;
 
-		BUG_ON(PageActive(page));
+		VM_BUG_ON(PageActive(page));
 
 		sc->nr_scanned++;
 
@@ -564,7 +564,7 @@ keep_locked:
 		unlock_page(page);
 keep:
 		list_add(&page->lru, &ret_pages);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 	}
 	list_splice(&ret_pages, page_list);
 	if (pagevec_count(&freed_pvec))
@@ -603,7 +603,7 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
 		page = lru_to_page(src);
 		prefetchw_prev_lru_page(page, src, flags);
 
-		BUG_ON(!PageLRU(page));
+		VM_BUG_ON(!PageLRU(page));
 
 		list_del(&page->lru);
 		target = src;
@@ -674,7 +674,7 @@ static unsigned long shrink_inactive_list(unsigned long max_scan,
 		 */
 		while (!list_empty(&page_list)) {
 			page = lru_to_page(&page_list);
-			BUG_ON(PageLRU(page));
+			VM_BUG_ON(PageLRU(page));
 			SetPageLRU(page);
 			list_del(&page->lru);
 			if (PageActive(page))
@@ -797,9 +797,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	while (!list_empty(&l_inactive)) {
 		page = lru_to_page(&l_inactive);
 		prefetchw_prev_lru_page(page, &l_inactive, flags);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(!PageActive(page));
+		VM_BUG_ON(!PageActive(page));
 		ClearPageActive(page);
 
 		list_move(&page->lru, &zone->inactive_list);
@@ -827,9 +827,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	while (!list_empty(&l_active)) {
 		page = lru_to_page(&l_active);
 		prefetchw_prev_lru_page(page, &l_active, flags);
-		BUG_ON(PageLRU(page));
+		VM_BUG_ON(PageLRU(page));
 		SetPageLRU(page);
-		BUG_ON(!PageActive(page));
+		VM_BUG_ON(!PageActive(page));
 		list_move(&page->lru, &zone->active_list);
 		pgmoved++;
 		if (!pagevec_add(&pvec, page)) {
-- 
cgit v1.2.1


From d08b3851da41d0ee60851f2c75b118e1f7a5fc89 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:57 -0700
Subject: [PATCH] mm: tracking shared dirty pages

Tracking of dirty pages in shared writeable mmap()s.

The idea is simple: write protect clean shared writeable pages, catch the
write-fault, make writeable and set dirty.  On page write-back clean all the
PTE dirty bits and write protect them once again.

The implementation is a tad harder, mainly because the default
backing_dev_info capabilities were too loosely maintained.  Hence it is not
enough to test the backing_dev_info for cap_account_dirty.

The current heuristic is as follows, a VMA is eligible when:
 - its shared writeable
    (vm_flags & (VM_WRITE|VM_SHARED)) == (VM_WRITE|VM_SHARED)
 - it is not a 'special' mapping
    (vm_flags & (VM_PFNMAP|VM_INSERTPAGE)) == 0
 - the backing_dev_info is cap_account_dirty
    mapping_cap_account_dirty(vma->vm_file->f_mapping)
 - f_op->mmap() didn't change the default page protection

Page from remap_pfn_range() are explicitly excluded because their COW
semantics are already horrid enough (see vm_normal_page() in do_wp_page()) and
because they don't have a backing store anyway.

mprotect() is taught about the new behaviour as well.  However it overrides
the last condition.

Cleaning the pages on write-back is done with page_mkclean() a new rmap call.
It can be called on any page, but is currently only implemented for mapped
pages, if the page is found the be of a VMA that accounts dirty pages it will
also wrprotect the PTE.

Finally, in fs/buffers.c:try_to_free_buffers(); remove clear_page_dirty() from
under ->private_lock.  This seems to be safe, since ->private_lock is used to
serialize access to the buffers, not the page itself.  This is needed because
clear_page_dirty() will call into page_mkclean() and would thereby violate
locking order.

[dhowells@redhat.com: Provide a page_mkclean() implementation for NOMMU]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/buffer.c          |  2 +-
 include/linux/mm.h   | 34 +++++++++++++++++++++++++++
 include/linux/rmap.h | 14 +++++++++++
 mm/memory.c          | 29 ++++++++++++++++++-----
 mm/mmap.c            | 10 ++++----
 mm/mprotect.c        | 21 +++++++----------
 mm/page-writeback.c  | 17 ++++++++++----
 mm/rmap.c            | 65 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 8 files changed, 162 insertions(+), 30 deletions(-)

diff --git a/fs/buffer.c b/fs/buffer.c
index 71649ef9b658..3b6d701073e7 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -2987,6 +2987,7 @@ int try_to_free_buffers(struct page *page)
 
 	spin_lock(&mapping->private_lock);
 	ret = drop_buffers(page, &buffers_to_free);
+	spin_unlock(&mapping->private_lock);
 	if (ret) {
 		/*
 		 * If the filesystem writes its buffers by hand (eg ext3)
@@ -2998,7 +2999,6 @@ int try_to_free_buffers(struct page *page)
 		 */
 		clear_page_dirty(page);
 	}
-	spin_unlock(&mapping->private_lock);
 out:
 	if (buffers_to_free) {
 		struct buffer_head *bh = buffers_to_free;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7d20b25c58fc..449841413cf1 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -15,6 +15,7 @@
 #include <linux/fs.h>
 #include <linux/mutex.h>
 #include <linux/debug_locks.h>
+#include <linux/backing-dev.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -810,6 +811,39 @@ struct shrinker;
 extern struct shrinker *set_shrinker(int, shrinker_t);
 extern void remove_shrinker(struct shrinker *shrinker);
 
+/*
+ * Some shared mappigns will want the pages marked read-only
+ * to track write events. If so, we'll downgrade vm_page_prot
+ * to the private version (using protection_map[] without the
+ * VM_SHARED bit).
+ */
+static inline int vma_wants_writenotify(struct vm_area_struct *vma)
+{
+	unsigned int vm_flags = vma->vm_flags;
+
+	/* If it was private or non-writable, the write bit is already clear */
+	if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED)))
+		return 0;
+
+	/* The backer wishes to know when pages are first written to? */
+	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
+		return 1;
+
+	/* The open routine did something to the protections already? */
+	if (pgprot_val(vma->vm_page_prot) !=
+	    pgprot_val(protection_map[vm_flags &
+		    (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]))
+		return 0;
+
+	/* Specialty mapping? */
+	if (vm_flags & (VM_PFNMAP|VM_INSERTPAGE))
+		return 0;
+
+	/* Can the mapping track the dirty pages? */
+	return vma->vm_file && vma->vm_file->f_mapping &&
+		mapping_cap_account_dirty(vma->vm_file->f_mapping);
+}
+
 extern pte_t *FASTCALL(get_locked_pte(struct mm_struct *mm, unsigned long addr, spinlock_t **ptl));
 
 int __pud_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address);
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index bf97b0900014..db2c1df4fef9 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -103,6 +103,14 @@ pte_t *page_check_address(struct page *, struct mm_struct *,
  */
 unsigned long page_address_in_vma(struct page *, struct vm_area_struct *);
 
+/*
+ * Cleans the PTEs of shared mappings.
+ * (and since clean PTEs should also be readonly, write protects them too)
+ *
+ * returns the number of cleaned PTEs.
+ */
+int page_mkclean(struct page *);
+
 #else	/* !CONFIG_MMU */
 
 #define anon_vma_init()		do {} while (0)
@@ -112,6 +120,12 @@ unsigned long page_address_in_vma(struct page *, struct vm_area_struct *);
 #define page_referenced(page,l) TestClearPageReferenced(page)
 #define try_to_unmap(page, refs) SWAP_FAIL
 
+static inline int page_mkclean(struct page *page)
+{
+	return 0;
+}
+
+
 #endif	/* CONFIG_MMU */
 
 /*
diff --git a/mm/memory.c b/mm/memory.c
index 109e9866237e..fa941b169071 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1458,14 +1458,19 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 {
 	struct page *old_page, *new_page;
 	pte_t entry;
-	int reuse, ret = VM_FAULT_MINOR;
+	int reuse = 0, ret = VM_FAULT_MINOR;
+	struct page *dirty_page = NULL;
 
 	old_page = vm_normal_page(vma, address, orig_pte);
 	if (!old_page)
 		goto gotten;
 
-	if (unlikely((vma->vm_flags & (VM_SHARED|VM_WRITE)) ==
-				(VM_SHARED|VM_WRITE))) {
+	/*
+	 * Only catch write-faults on shared writable pages, read-only
+	 * shared pages can get COWed by get_user_pages(.write=1, .force=1).
+	 */
+	if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
+					(VM_WRITE|VM_SHARED))) {
 		if (vma->vm_ops && vma->vm_ops->page_mkwrite) {
 			/*
 			 * Notify the address space that the page is about to
@@ -1494,13 +1499,12 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			if (!pte_same(*page_table, orig_pte))
 				goto unlock;
 		}
-
+		dirty_page = old_page;
+		get_page(dirty_page);
 		reuse = 1;
 	} else if (PageAnon(old_page) && !TestSetPageLocked(old_page)) {
 		reuse = can_share_swap_page(old_page);
 		unlock_page(old_page);
-	} else {
-		reuse = 0;
 	}
 
 	if (reuse) {
@@ -1566,6 +1570,10 @@ gotten:
 		page_cache_release(old_page);
 unlock:
 	pte_unmap_unlock(page_table, ptl);
+	if (dirty_page) {
+		set_page_dirty(dirty_page);
+		put_page(dirty_page);
+	}
 	return ret;
 oom:
 	if (old_page)
@@ -2098,6 +2106,7 @@ static int do_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	unsigned int sequence = 0;
 	int ret = VM_FAULT_MINOR;
 	int anon = 0;
+	struct page *dirty_page = NULL;
 
 	pte_unmap(page_table);
 	BUG_ON(vma->vm_flags & VM_PFNMAP);
@@ -2192,6 +2201,10 @@ retry:
 		} else {
 			inc_mm_counter(mm, file_rss);
 			page_add_file_rmap(new_page);
+			if (write_access) {
+				dirty_page = new_page;
+				get_page(dirty_page);
+			}
 		}
 	} else {
 		/* One of our sibling threads was faster, back out. */
@@ -2204,6 +2217,10 @@ retry:
 	lazy_mmu_prot_update(entry);
 unlock:
 	pte_unmap_unlock(page_table, ptl);
+	if (dirty_page) {
+		set_page_dirty(dirty_page);
+		put_page(dirty_page);
+	}
 	return ret;
 oom:
 	page_cache_release(new_page);
diff --git a/mm/mmap.c b/mm/mmap.c
index d799d896d74a..8507ee9cd573 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1105,12 +1105,6 @@ munmap_back:
 			goto free_vma;
 	}
 
-	/* Don't make the VMA automatically writable if it's shared, but the
-	 * backer wishes to know when pages are first written to */
-	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
-		vma->vm_page_prot =
-			protection_map[vm_flags & (VM_READ|VM_WRITE|VM_EXEC)];
-
 	/* We set VM_ACCOUNT in a shared mapping's vm_flags, to inform
 	 * shmem_zero_setup (perhaps called through /dev/zero's ->mmap)
 	 * that memory reservation must be checked; but that reservation
@@ -1128,6 +1122,10 @@ munmap_back:
 	pgoff = vma->vm_pgoff;
 	vm_flags = vma->vm_flags;
 
+	if (vma_wants_writenotify(vma))
+		vma->vm_page_prot =
+			protection_map[vm_flags & (VM_READ|VM_WRITE|VM_EXEC)];
+
 	if (!file || !vma_merge(mm, prev, addr, vma->vm_end,
 			vma->vm_flags, NULL, file, pgoff, vma_policy(vma))) {
 		file = vma->vm_file;
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 638edabaff71..367b7f6c0637 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -123,8 +123,6 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	unsigned long oldflags = vma->vm_flags;
 	long nrpages = (end - start) >> PAGE_SHIFT;
 	unsigned long charged = 0;
-	unsigned int mask;
-	pgprot_t newprot;
 	pgoff_t pgoff;
 	int error;
 
@@ -176,24 +174,21 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	}
 
 success:
-	/* Don't make the VMA automatically writable if it's shared, but the
-	 * backer wishes to know when pages are first written to */
-	mask = VM_READ|VM_WRITE|VM_EXEC|VM_SHARED;
-	if (vma->vm_ops && vma->vm_ops->page_mkwrite)
-		mask &= ~VM_SHARED;
-
-	newprot = protection_map[newflags & mask];
-
 	/*
 	 * vm_flags and vm_page_prot are protected by the mmap_sem
 	 * held in write mode.
 	 */
 	vma->vm_flags = newflags;
-	vma->vm_page_prot = newprot;
+	vma->vm_page_prot = protection_map[newflags &
+		(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
+	if (vma_wants_writenotify(vma))
+		vma->vm_page_prot = protection_map[newflags &
+			(VM_READ|VM_WRITE|VM_EXEC)];
+
 	if (is_vm_hugetlb_page(vma))
-		hugetlb_change_protection(vma, start, end, newprot);
+		hugetlb_change_protection(vma, start, end, vma->vm_page_prot);
 	else
-		change_protection(vma, start, end, newprot);
+		change_protection(vma, start, end, vma->vm_page_prot);
 	vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
 	vm_stat_account(mm, newflags, vma->vm_file, nrpages);
 	return 0;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 77a0bc4e261a..1c87430b7a25 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -23,6 +23,7 @@
 #include <linux/backing-dev.h>
 #include <linux/blkdev.h>
 #include <linux/mpage.h>
+#include <linux/rmap.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
 #include <linux/smp.h>
@@ -550,7 +551,7 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
 		return 0;
 	wbc->for_writepages = 1;
 	if (mapping->a_ops->writepages)
-		ret =  mapping->a_ops->writepages(mapping, wbc);
+		ret = mapping->a_ops->writepages(mapping, wbc);
 	else
 		ret = generic_writepages(mapping, wbc);
 	wbc->for_writepages = 0;
@@ -712,9 +713,15 @@ int test_clear_page_dirty(struct page *page)
 			radix_tree_tag_clear(&mapping->page_tree,
 						page_index(page),
 						PAGECACHE_TAG_DIRTY);
-			if (mapping_cap_account_dirty(mapping))
-				__dec_zone_page_state(page, NR_FILE_DIRTY);
 			write_unlock_irqrestore(&mapping->tree_lock, flags);
+			/*
+			 * We can continue to use `mapping' here because the
+			 * page is locked, which pins the address_space
+			 */
+			if (mapping_cap_account_dirty(mapping)) {
+				page_mkclean(page);
+				dec_zone_page_state(page, NR_FILE_DIRTY);
+			}
 			return 1;
 		}
 		write_unlock_irqrestore(&mapping->tree_lock, flags);
@@ -744,8 +751,10 @@ int clear_page_dirty_for_io(struct page *page)
 
 	if (mapping) {
 		if (TestClearPageDirty(page)) {
-			if (mapping_cap_account_dirty(mapping))
+			if (mapping_cap_account_dirty(mapping)) {
+				page_mkclean(page);
 				dec_zone_page_state(page, NR_FILE_DIRTY);
+			}
 			return 1;
 		}
 		return 0;
diff --git a/mm/rmap.c b/mm/rmap.c
index 40158b59729e..e2155d791d99 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -434,6 +434,71 @@ int page_referenced(struct page *page, int is_locked)
 	return referenced;
 }
 
+static int page_mkclean_one(struct page *page, struct vm_area_struct *vma)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	unsigned long address;
+	pte_t *pte, entry;
+	spinlock_t *ptl;
+	int ret = 0;
+
+	address = vma_address(page, vma);
+	if (address == -EFAULT)
+		goto out;
+
+	pte = page_check_address(page, mm, address, &ptl);
+	if (!pte)
+		goto out;
+
+	if (!pte_dirty(*pte) && !pte_write(*pte))
+		goto unlock;
+
+	entry = ptep_get_and_clear(mm, address, pte);
+	entry = pte_mkclean(entry);
+	entry = pte_wrprotect(entry);
+	ptep_establish(vma, address, pte, entry);
+	lazy_mmu_prot_update(entry);
+	ret = 1;
+
+unlock:
+	pte_unmap_unlock(pte, ptl);
+out:
+	return ret;
+}
+
+static int page_mkclean_file(struct address_space *mapping, struct page *page)
+{
+	pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
+	struct vm_area_struct *vma;
+	struct prio_tree_iter iter;
+	int ret = 0;
+
+	BUG_ON(PageAnon(page));
+
+	spin_lock(&mapping->i_mmap_lock);
+	vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
+		if (vma->vm_flags & VM_SHARED)
+			ret += page_mkclean_one(page, vma);
+	}
+	spin_unlock(&mapping->i_mmap_lock);
+	return ret;
+}
+
+int page_mkclean(struct page *page)
+{
+	int ret = 0;
+
+	BUG_ON(!PageLocked(page));
+
+	if (page_mapped(page)) {
+		struct address_space *mapping = page_mapping(page);
+		if (mapping)
+			ret = page_mkclean_file(mapping, page);
+	}
+
+	return ret;
+}
+
 /**
  * page_set_anon_rmap - setup new anonymous rmap
  * @page:	the page to add the mapping to
-- 
cgit v1.2.1


From edc79b2a46ed854595e40edcf3f8b37f9f14aa3f Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:58 -0700
Subject: [PATCH] mm: balance dirty pages

Now that we can detect writers of shared mappings, throttle them.  Avoids OOM
by surprise.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/writeback.h |  1 +
 mm/memory.c               |  5 +++--
 mm/page-writeback.c       | 10 ++++++++++
 3 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 0422036af4eb..56a23a0e7f2e 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -116,6 +116,7 @@ int sync_page_range(struct inode *inode, struct address_space *mapping,
 			loff_t pos, loff_t count);
 int sync_page_range_nolock(struct inode *inode, struct address_space *mapping,
 			   loff_t pos, loff_t count);
+void set_page_dirty_balance(struct page *page);
 
 /* pdflush.c */
 extern int nr_pdflush_threads;	/* Global so it can be exported to sysctl
diff --git a/mm/memory.c b/mm/memory.c
index fa941b169071..dd7d7fc5ed60 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -49,6 +49,7 @@
 #include <linux/module.h>
 #include <linux/delayacct.h>
 #include <linux/init.h>
+#include <linux/writeback.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -1571,7 +1572,7 @@ gotten:
 unlock:
 	pte_unmap_unlock(page_table, ptl);
 	if (dirty_page) {
-		set_page_dirty(dirty_page);
+		set_page_dirty_balance(dirty_page);
 		put_page(dirty_page);
 	}
 	return ret;
@@ -2218,7 +2219,7 @@ retry:
 unlock:
 	pte_unmap_unlock(page_table, ptl);
 	if (dirty_page) {
-		set_page_dirty(dirty_page);
+		set_page_dirty_balance(dirty_page);
 		put_page(dirty_page);
 	}
 	return ret;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 1c87430b7a25..b9f4c6f1be86 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -244,6 +244,16 @@ static void balance_dirty_pages(struct address_space *mapping)
 		pdflush_operation(background_writeout, 0);
 }
 
+void set_page_dirty_balance(struct page *page)
+{
+	if (set_page_dirty(page)) {
+		struct address_space *mapping = page_mapping(page);
+
+		if (mapping)
+			balance_dirty_pages_ratelimited(mapping);
+	}
+}
+
 /**
  * balance_dirty_pages_ratelimited_nr - balance dirty memory state
  * @mapping: address_space which was dirtied
-- 
cgit v1.2.1


From c1e6098b23bb46e2b488fe9a26f831f867157483 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:59 -0700
Subject: [PATCH] mm: optimize the new mprotect() code a bit

mprotect() resets the page protections, which could result in extra write
faults for those pages whose dirty state we track using write faults and are
dirty already.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mprotect.c | 34 ++++++++++++++++++++++++----------
 1 file changed, 24 insertions(+), 10 deletions(-)

diff --git a/mm/mprotect.c b/mm/mprotect.c
index 367b7f6c0637..955f9d0e38aa 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -27,7 +27,8 @@
 #include <asm/tlbflush.h>
 
 static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
-		unsigned long addr, unsigned long end, pgprot_t newprot)
+		unsigned long addr, unsigned long end, pgprot_t newprot,
+		int dirty_accountable)
 {
 	pte_t *pte, oldpte;
 	spinlock_t *ptl;
@@ -42,7 +43,14 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 			 * bits by wiping the pte and then setting the new pte
 			 * into place.
 			 */
-			ptent = pte_modify(ptep_get_and_clear(mm, addr, pte), newprot);
+			ptent = ptep_get_and_clear(mm, addr, pte);
+			ptent = pte_modify(ptent, newprot);
+			/*
+			 * Avoid taking write faults for pages we know to be
+			 * dirty.
+			 */
+			if (dirty_accountable && pte_dirty(ptent))
+				ptent = pte_mkwrite(ptent);
 			set_pte_at(mm, addr, pte, ptent);
 			lazy_mmu_prot_update(ptent);
 #ifdef CONFIG_MIGRATION
@@ -66,7 +74,8 @@ static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
 }
 
 static inline void change_pmd_range(struct mm_struct *mm, pud_t *pud,
-		unsigned long addr, unsigned long end, pgprot_t newprot)
+		unsigned long addr, unsigned long end, pgprot_t newprot,
+		int dirty_accountable)
 {
 	pmd_t *pmd;
 	unsigned long next;
@@ -76,12 +85,13 @@ static inline void change_pmd_range(struct mm_struct *mm, pud_t *pud,
 		next = pmd_addr_end(addr, end);
 		if (pmd_none_or_clear_bad(pmd))
 			continue;
-		change_pte_range(mm, pmd, addr, next, newprot);
+		change_pte_range(mm, pmd, addr, next, newprot, dirty_accountable);
 	} while (pmd++, addr = next, addr != end);
 }
 
 static inline void change_pud_range(struct mm_struct *mm, pgd_t *pgd,
-		unsigned long addr, unsigned long end, pgprot_t newprot)
+		unsigned long addr, unsigned long end, pgprot_t newprot,
+		int dirty_accountable)
 {
 	pud_t *pud;
 	unsigned long next;
@@ -91,12 +101,13 @@ static inline void change_pud_range(struct mm_struct *mm, pgd_t *pgd,
 		next = pud_addr_end(addr, end);
 		if (pud_none_or_clear_bad(pud))
 			continue;
-		change_pmd_range(mm, pud, addr, next, newprot);
+		change_pmd_range(mm, pud, addr, next, newprot, dirty_accountable);
 	} while (pud++, addr = next, addr != end);
 }
 
 static void change_protection(struct vm_area_struct *vma,
-		unsigned long addr, unsigned long end, pgprot_t newprot)
+		unsigned long addr, unsigned long end, pgprot_t newprot,
+		int dirty_accountable)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	pgd_t *pgd;
@@ -110,7 +121,7 @@ static void change_protection(struct vm_area_struct *vma,
 		next = pgd_addr_end(addr, end);
 		if (pgd_none_or_clear_bad(pgd))
 			continue;
-		change_pud_range(mm, pgd, addr, next, newprot);
+		change_pud_range(mm, pgd, addr, next, newprot, dirty_accountable);
 	} while (pgd++, addr = next, addr != end);
 	flush_tlb_range(vma, start, end);
 }
@@ -125,6 +136,7 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	unsigned long charged = 0;
 	pgoff_t pgoff;
 	int error;
+	int dirty_accountable = 0;
 
 	if (newflags == oldflags) {
 		*pprev = vma;
@@ -181,14 +193,16 @@ success:
 	vma->vm_flags = newflags;
 	vma->vm_page_prot = protection_map[newflags &
 		(VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)];
-	if (vma_wants_writenotify(vma))
+	if (vma_wants_writenotify(vma)) {
 		vma->vm_page_prot = protection_map[newflags &
 			(VM_READ|VM_WRITE|VM_EXEC)];
+		dirty_accountable = 1;
+	}
 
 	if (is_vm_hugetlb_page(vma))
 		hugetlb_change_protection(vma, start, end, vma->vm_page_prot);
 	else
-		change_protection(vma, start, end, vma->vm_page_prot);
+		change_protection(vma, start, end, vma->vm_page_prot, dirty_accountable);
 	vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
 	vm_stat_account(mm, newflags, vma->vm_file, nrpages);
 	return 0;
-- 
cgit v1.2.1


From e88dd6c11c5aef74d8b74a062767add53315533b Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:30:59 -0700
Subject: [PATCH] mm: small cleanup of install_page()

Smallish cleanup to install_page(), could save a memory read (haven't checked
the asm output) and sure looks nicer.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/fremap.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/fremap.c b/mm/fremap.c
index 21b7d0cbc98c..aa30618ec6b2 100644
--- a/mm/fremap.c
+++ b/mm/fremap.c
@@ -79,9 +79,9 @@ int install_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		inc_mm_counter(mm, file_rss);
 
 	flush_icache_page(vma, page);
-	set_pte_at(mm, addr, pte, mk_pte(page, prot));
+	pte_val = mk_pte(page, prot);
+	set_pte_at(mm, addr, pte, pte_val);
 	page_add_file_rmap(page);
-	pte_val = *pte;
 	update_mmu_cache(vma, addr, pte_val);
 	lazy_mmu_prot_update(pte_val);
 	err = 0;
-- 
cgit v1.2.1


From ee6a6457886a80415db209e87033b63f2b06558c Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:31:00 -0700
Subject: [PATCH] mm: fixup do_wp_page()

Wrt. the recent modifications in do_wp_page() Hugh Dickins pointed out:

  "I now realize it's right to the first order (normal case) and to the
   second order (ptrace poke), but not to the third order (ptrace poke
   anon page here to be COWed - perhaps can't occur without intervening
   mprotects)."

This patch restores the old COW behaviour for anonymous pages.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/memory.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index dd7d7fc5ed60..65962534b4ed 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1467,11 +1467,21 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		goto gotten;
 
 	/*
-	 * Only catch write-faults on shared writable pages, read-only
-	 * shared pages can get COWed by get_user_pages(.write=1, .force=1).
+	 * Take out anonymous pages first, anonymous shared vmas are
+	 * not dirty accountable.
 	 */
-	if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
+	if (PageAnon(old_page)) {
+		if (!TestSetPageLocked(old_page)) {
+			reuse = can_share_swap_page(old_page);
+			unlock_page(old_page);
+		}
+	} else if (unlikely((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
 					(VM_WRITE|VM_SHARED))) {
+		/*
+		 * Only catch write-faults on shared writable pages,
+		 * read-only shared pages can get COWed by
+		 * get_user_pages(.write=1, .force=1).
+		 */
 		if (vma->vm_ops && vma->vm_ops->page_mkwrite) {
 			/*
 			 * Notify the address space that the page is about to
@@ -1503,9 +1513,6 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		dirty_page = old_page;
 		get_page(dirty_page);
 		reuse = 1;
-	} else if (PageAnon(old_page) && !TestSetPageLocked(old_page)) {
-		reuse = can_share_swap_page(old_page);
-		unlock_page(old_page);
 	}
 
 	if (reuse) {
-- 
cgit v1.2.1


From 204ec841fbea3e5138168edbc3a76d46747cc987 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:31:01 -0700
Subject: [PATCH] mm: msync() cleanup

With the tracking of dirty pages properly done now, msync doesn't need to scan
the PTEs anymore to determine the dirty status.

From: Hugh Dickins <hugh@veritas.com>

In looking to do that, I made some other tidyups: can remove several
#includes, and sys_msync loop termination not quite right.

Most of those points are criticisms of the existing sys_msync, not of your
patch.  In particular, the loop termination errors were introduced in 2.6.17:
I did notice this shortly before it came out, but decided I was more likely to
get it wrong myself, and make matters worse if I tried to rush a last-minute
fix in.  And it's not terribly likely to go wrong, nor disastrous if it does
go wrong (may miss reporting an unmapped area; may also fsync file of a
following vma).

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/msync.c | 196 +++++++++++--------------------------------------------------
 1 file changed, 33 insertions(+), 163 deletions(-)

diff --git a/mm/msync.c b/mm/msync.c
index d083544df21b..358d73cf7b78 100644
--- a/mm/msync.c
+++ b/mm/msync.c
@@ -7,149 +7,33 @@
 /*
  * The msync() system call.
  */
-#include <linux/slab.h>
-#include <linux/pagemap.h>
 #include <linux/fs.h>
 #include <linux/mm.h>
 #include <linux/mman.h>
-#include <linux/hugetlb.h>
-#include <linux/writeback.h>
 #include <linux/file.h>
 #include <linux/syscalls.h>
 
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-
-static unsigned long msync_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
-				unsigned long addr, unsigned long end)
-{
-	pte_t *pte;
-	spinlock_t *ptl;
-	int progress = 0;
-	unsigned long ret = 0;
-
-again:
-	pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
-	do {
-		struct page *page;
-
-		if (progress >= 64) {
-			progress = 0;
-			if (need_resched() || need_lockbreak(ptl))
-				break;
-		}
-		progress++;
-		if (!pte_present(*pte))
-			continue;
-		if (!pte_maybe_dirty(*pte))
-			continue;
-		page = vm_normal_page(vma, addr, *pte);
-		if (!page)
-			continue;
-		if (ptep_clear_flush_dirty(vma, addr, pte) ||
-				page_test_and_clear_dirty(page))
-			ret += set_page_dirty(page);
-		progress += 3;
-	} while (pte++, addr += PAGE_SIZE, addr != end);
-	pte_unmap_unlock(pte - 1, ptl);
-	cond_resched();
-	if (addr != end)
-		goto again;
-	return ret;
-}
-
-static inline unsigned long msync_pmd_range(struct vm_area_struct *vma,
-			pud_t *pud, unsigned long addr, unsigned long end)
-{
-	pmd_t *pmd;
-	unsigned long next;
-	unsigned long ret = 0;
-
-	pmd = pmd_offset(pud, addr);
-	do {
-		next = pmd_addr_end(addr, end);
-		if (pmd_none_or_clear_bad(pmd))
-			continue;
-		ret += msync_pte_range(vma, pmd, addr, next);
-	} while (pmd++, addr = next, addr != end);
-	return ret;
-}
-
-static inline unsigned long msync_pud_range(struct vm_area_struct *vma,
-			pgd_t *pgd, unsigned long addr, unsigned long end)
-{
-	pud_t *pud;
-	unsigned long next;
-	unsigned long ret = 0;
-
-	pud = pud_offset(pgd, addr);
-	do {
-		next = pud_addr_end(addr, end);
-		if (pud_none_or_clear_bad(pud))
-			continue;
-		ret += msync_pmd_range(vma, pud, addr, next);
-	} while (pud++, addr = next, addr != end);
-	return ret;
-}
-
-static unsigned long msync_page_range(struct vm_area_struct *vma,
-				unsigned long addr, unsigned long end)
-{
-	pgd_t *pgd;
-	unsigned long next;
-	unsigned long ret = 0;
-
-	/* For hugepages we can't go walking the page table normally,
-	 * but that's ok, hugetlbfs is memory based, so we don't need
-	 * to do anything more on an msync().
-	 */
-	if (vma->vm_flags & VM_HUGETLB)
-		return 0;
-
-	BUG_ON(addr >= end);
-	pgd = pgd_offset(vma->vm_mm, addr);
-	flush_cache_range(vma, addr, end);
-	do {
-		next = pgd_addr_end(addr, end);
-		if (pgd_none_or_clear_bad(pgd))
-			continue;
-		ret += msync_pud_range(vma, pgd, addr, next);
-	} while (pgd++, addr = next, addr != end);
-	return ret;
-}
-
 /*
  * MS_SYNC syncs the entire file - including mappings.
  *
- * MS_ASYNC does not start I/O (it used to, up to 2.5.67).  Instead, it just
- * marks the relevant pages dirty.  The application may now run fsync() to
+ * MS_ASYNC does not start I/O (it used to, up to 2.5.67).
+ * Nor does it marks the relevant pages dirty (it used to up to 2.6.17).
+ * Now it doesn't do anything, since dirty pages are properly tracked.
+ *
+ * The application may now run fsync() to
  * write out the dirty pages and wait on the writeout and check the result.
  * Or the application may run fadvise(FADV_DONTNEED) against the fd to start
  * async writeout immediately.
  * So by _not_ starting I/O in MS_ASYNC we provide complete flexibility to
  * applications.
  */
-static int msync_interval(struct vm_area_struct *vma, unsigned long addr,
-			unsigned long end, int flags,
-			unsigned long *nr_pages_dirtied)
-{
-	struct file *file = vma->vm_file;
-
-	if ((flags & MS_INVALIDATE) && (vma->vm_flags & VM_LOCKED))
-		return -EBUSY;
-
-	if (file && (vma->vm_flags & VM_SHARED))
-		*nr_pages_dirtied = msync_page_range(vma, addr, end);
-	return 0;
-}
-
 asmlinkage long sys_msync(unsigned long start, size_t len, int flags)
 {
 	unsigned long end;
+	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	int unmapped_error = 0;
 	int error = -EINVAL;
-	int done = 0;
 
 	if (flags & ~(MS_ASYNC | MS_INVALIDATE | MS_SYNC))
 		goto out;
@@ -169,64 +53,50 @@ asmlinkage long sys_msync(unsigned long start, size_t len, int flags)
 	 * If the interval [start,end) covers some unmapped address ranges,
 	 * just ignore them, but return -ENOMEM at the end.
 	 */
-	down_read(&current->mm->mmap_sem);
-	vma = find_vma(current->mm, start);
-	if (!vma) {
-		error = -ENOMEM;
-		goto out_unlock;
-	}
-	do {
-		unsigned long nr_pages_dirtied = 0;
+	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, start);
+	for (;;) {
 		struct file *file;
 
+		/* Still start < end. */
+		error = -ENOMEM;
+		if (!vma)
+			goto out_unlock;
 		/* Here start < vma->vm_end. */
 		if (start < vma->vm_start) {
-			unmapped_error = -ENOMEM;
 			start = vma->vm_start;
+			if (start >= end)
+				goto out_unlock;
+			unmapped_error = -ENOMEM;
 		}
 		/* Here vma->vm_start <= start < vma->vm_end. */
-		if (end <= vma->vm_end) {
-			if (start < end) {
-				error = msync_interval(vma, start, end, flags,
-							&nr_pages_dirtied);
-				if (error)
-					goto out_unlock;
-			}
-			error = unmapped_error;
-			done = 1;
-		} else {
-			/* Here vma->vm_start <= start < vma->vm_end < end. */
-			error = msync_interval(vma, start, vma->vm_end, flags,
-						&nr_pages_dirtied);
-			if (error)
-				goto out_unlock;
+		if ((flags & MS_INVALIDATE) &&
+				(vma->vm_flags & VM_LOCKED)) {
+			error = -EBUSY;
+			goto out_unlock;
 		}
 		file = vma->vm_file;
 		start = vma->vm_end;
-		if ((flags & MS_ASYNC) && file && nr_pages_dirtied) {
-			get_file(file);
-			up_read(&current->mm->mmap_sem);
-			balance_dirty_pages_ratelimited_nr(file->f_mapping,
-							nr_pages_dirtied);
-			fput(file);
-			down_read(&current->mm->mmap_sem);
-			vma = find_vma(current->mm, start);
-		} else if ((flags & MS_SYNC) && file &&
+		if ((flags & MS_SYNC) && file &&
 				(vma->vm_flags & VM_SHARED)) {
 			get_file(file);
-			up_read(&current->mm->mmap_sem);
+			up_read(&mm->mmap_sem);
 			error = do_fsync(file, 0);
 			fput(file);
-			down_read(&current->mm->mmap_sem);
-			if (error)
-				goto out_unlock;
-			vma = find_vma(current->mm, start);
+			if (error || start >= end)
+				goto out;
+			down_read(&mm->mmap_sem);
+			vma = find_vma(mm, start);
 		} else {
+			if (start >= end) {
+				error = 0;
+				goto out_unlock;
+			}
 			vma = vma->vm_next;
 		}
-	} while (vma && !done);
+	}
 out_unlock:
-	up_read(&current->mm->mmap_sem);
+	up_read(&mm->mmap_sem);
 out:
-	return error;
+	return error ? : unmapped_error;
 }
-- 
cgit v1.2.1


From b221385bc41d6789edde3d2fa0cb20d5045730eb Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 25 Sep 2006 23:31:02 -0700
Subject: [PATCH] mm/: make functions static

This patch makes the following needlessly global functions static:
 - slab.c: kmem_find_general_cachep()
 - swap.c: __page_cache_release()
 - vmalloc.c: __vmalloc_node()

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h      |  2 --
 include/linux/slab.h    |  2 --
 include/linux/vmalloc.h |  2 --
 mm/slab.c               |  3 +--
 mm/swap.c               | 39 +++++++++++++++++++--------------------
 mm/vmalloc.c            |  8 +++++---
 6 files changed, 25 insertions(+), 31 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 449841413cf1..45678b036955 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -318,8 +318,6 @@ static inline int get_page_unless_zero(struct page *page)
 	return atomic_inc_not_zero(&page->_count);
 }
 
-extern void FASTCALL(__page_cache_release(struct page *));
-
 static inline int page_count(struct page *page)
 {
 	if (unlikely(PageCompound(page)))
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 45ad55b70d1c..193c03c547ec 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -67,7 +67,6 @@ extern void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 extern void kmem_cache_free(kmem_cache_t *, void *);
 extern unsigned int kmem_cache_size(kmem_cache_t *);
 extern const char *kmem_cache_name(kmem_cache_t *);
-extern kmem_cache_t *kmem_find_general_cachep(size_t size, gfp_t gfpflags);
 
 /* Size description struct for general caches. */
 struct cache_sizes {
@@ -223,7 +222,6 @@ extern int FASTCALL(kmem_ptr_validate(kmem_cache_t *cachep, void *ptr));
 /* SLOB allocator routines */
 
 void kmem_cache_init(void);
-struct kmem_cache *kmem_find_general_cachep(size_t, gfp_t gfpflags);
 struct kmem_cache *kmem_cache_create(const char *c, size_t, size_t,
 	unsigned long,
 	void (*)(void *, struct kmem_cache *, unsigned long),
diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index 71b6363caaaf..dee88c6b6fa7 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -44,8 +44,6 @@ extern void *vmalloc_32_user(unsigned long size);
 extern void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot);
 extern void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask,
 				pgprot_t prot);
-extern void *__vmalloc_node(unsigned long size, gfp_t gfp_mask,
-				pgprot_t prot, int node);
 extern void vfree(void *addr);
 
 extern void *vmap(struct page **pages, unsigned int count,
diff --git a/mm/slab.c b/mm/slab.c
index 21ba06035700..5870bcbd33cf 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -768,11 +768,10 @@ static inline struct kmem_cache *__find_general_cachep(size_t size,
 	return csizep->cs_cachep;
 }
 
-struct kmem_cache *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
+static struct kmem_cache *kmem_find_general_cachep(size_t size, gfp_t gfpflags)
 {
 	return __find_general_cachep(size, gfpflags);
 }
-EXPORT_SYMBOL(kmem_find_general_cachep);
 
 static size_t slab_mgmt_size(size_t nr_objs, size_t align)
 {
diff --git a/mm/swap.c b/mm/swap.c
index 600235e43704..2e0e871f542f 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -34,6 +34,25 @@
 /* How many pages do we try to swap or page in/out together? */
 int page_cluster;
 
+/*
+ * This path almost never happens for VM activity - pages are normally
+ * freed via pagevecs.  But it gets used by networking.
+ */
+static void fastcall __page_cache_release(struct page *page)
+{
+	if (PageLRU(page)) {
+		unsigned long flags;
+		struct zone *zone = page_zone(page);
+
+		spin_lock_irqsave(&zone->lru_lock, flags);
+		VM_BUG_ON(!PageLRU(page));
+		__ClearPageLRU(page);
+		del_page_from_lru(zone, page);
+		spin_unlock_irqrestore(&zone->lru_lock, flags);
+	}
+	free_hot_page(page);
+}
+
 static void put_compound_page(struct page *page)
 {
 	page = (struct page *)page_private(page);
@@ -222,26 +241,6 @@ int lru_add_drain_all(void)
 }
 #endif
 
-/*
- * This path almost never happens for VM activity - pages are normally
- * freed via pagevecs.  But it gets used by networking.
- */
-void fastcall __page_cache_release(struct page *page)
-{
-	if (PageLRU(page)) {
-		unsigned long flags;
-		struct zone *zone = page_zone(page);
-
-		spin_lock_irqsave(&zone->lru_lock, flags);
-		VM_BUG_ON(!PageLRU(page));
-		__ClearPageLRU(page);
-		del_page_from_lru(zone, page);
-		spin_unlock_irqrestore(&zone->lru_lock, flags);
-	}
-	free_hot_page(page);
-}
-EXPORT_SYMBOL(__page_cache_release);
-
 /*
  * Batched page_cache_release().  Decrement the reference count on all the
  * passed pages.  If it fell to zero then remove the page from the LRU and
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 266162d2ba28..9aad8b0cc6ee 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -24,6 +24,9 @@
 DEFINE_RWLOCK(vmlist_lock);
 struct vm_struct *vmlist;
 
+static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
+			    int node);
+
 static void vunmap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end)
 {
 	pte_t *pte;
@@ -478,8 +481,8 @@ void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
  *	allocator with @gfp_mask flags.  Map them into contiguous
  *	kernel virtual space, using a pagetable protection of @prot.
  */
-void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
-			int node)
+static void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
+			    int node)
 {
 	struct vm_struct *area;
 
@@ -493,7 +496,6 @@ void *__vmalloc_node(unsigned long size, gfp_t gfp_mask, pgprot_t prot,
 
 	return __vmalloc_area_node(area, gfp_mask, prot, node);
 }
-EXPORT_SYMBOL(__vmalloc_node);
 
 void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
 {
-- 
cgit v1.2.1


From 91023300057e96de7f46e95166a3e02394ae72f9 Mon Sep 17 00:00:00 2001
From: keith mannthey <kmannth@us.ibm.com>
Date: Mon, 25 Sep 2006 23:31:03 -0700
Subject: [PATCH] convert i386 NUMA KVA space to bootmem

Address a long standing issue of booting with an initrd on an i386 numa
system.  Currently (and always) the numa kva area is mapped into low memory
by finding the end of low memory and moving that mark down (thus creating
space for the kva).  The issue with this is that Grub loads initrds into
this similar space so when the kernel check the initrd it finds it outside
max_low_pfn and disables it (it thinks the initrd is not mapped into usable
memory) thus initrd enabled kernels can't boot i386 numa :(

My solution to the problem just converts the numa kva area to use the
bootmem allocator to save it's area (instead of moving the end of low
memory).  Using bootmem allows the kva area to be mapped into more diverse
addresses (not just the end of low memory) and enables the kva area to be
mapped below the initrd if present.

I have tested this patch on numaq(no initrd) and summit(initrd) i386 numa
based systems.

[akpm@osdl.org: cleanups]
Signed-off-by: Keith Mannthey <kmannth@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/setup.c  |  3 ++-
 arch/i386/mm/discontig.c  | 29 +++++++++++++++++++++--------
 include/asm-i386/mmzone.h |  6 ++++++
 3 files changed, 29 insertions(+), 9 deletions(-)

diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index f1682206d304..27d4dc0d3ef1 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -53,6 +53,7 @@
 #include <asm/apic.h>
 #include <asm/e820.h>
 #include <asm/mpspec.h>
+#include <asm/mmzone.h>
 #include <asm/setup.h>
 #include <asm/arch_hooks.h>
 #include <asm/sections.h>
@@ -1258,7 +1259,7 @@ void __init setup_bootmem_allocator(void)
 	 */
 	find_smp_config();
 #endif
-
+	numa_kva_reserve();
 #ifdef CONFIG_BLK_DEV_INITRD
 	if (LOADER_TYPE && INITRD_START) {
 		if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) {
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 7c392dc553b8..2e36eff8aff9 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -117,7 +117,8 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags);
 
 void *node_remap_end_vaddr[MAX_NUMNODES];
 void *node_remap_alloc_vaddr[MAX_NUMNODES];
-
+static unsigned long kva_start_pfn;
+static unsigned long kva_pages;
 /*
  * FLAT - support for basic PC memory model with discontig enabled, essentially
  *        a single node with all available processors in it with a flat
@@ -286,7 +287,6 @@ unsigned long __init setup_memory(void)
 {
 	int nid;
 	unsigned long system_start_pfn, system_max_low_pfn;
-	unsigned long reserve_pages;
 
 	/*
 	 * When mapping a NUMA machine we allocate the node_mem_map arrays
@@ -298,14 +298,23 @@ unsigned long __init setup_memory(void)
 	find_max_pfn();
 	get_memcfg_numa();
 
-	reserve_pages = calculate_numa_remap_pages();
+	kva_pages = calculate_numa_remap_pages();
 
 	/* partially used pages are not usable - thus round upwards */
 	system_start_pfn = min_low_pfn = PFN_UP(init_pg_tables_end);
 
-	system_max_low_pfn = max_low_pfn = find_max_low_pfn() - reserve_pages;
-	printk("reserve_pages = %ld find_max_low_pfn() ~ %ld\n",
-			reserve_pages, max_low_pfn + reserve_pages);
+	kva_start_pfn = find_max_low_pfn() - kva_pages;
+
+#ifdef CONFIG_BLK_DEV_INITRD
+	/* Numa kva area is below the initrd */
+	if (LOADER_TYPE && INITRD_START)
+		kva_start_pfn = PFN_DOWN(INITRD_START)  - kva_pages;
+#endif
+	kva_start_pfn -= kva_start_pfn & (PTRS_PER_PTE-1);
+
+	system_max_low_pfn = max_low_pfn = find_max_low_pfn();
+	printk("kva_start_pfn ~ %ld find_max_low_pfn() ~ %ld\n",
+		kva_start_pfn, max_low_pfn);
 	printk("max_pfn = %ld\n", max_pfn);
 #ifdef CONFIG_HIGHMEM
 	highstart_pfn = highend_pfn = max_pfn;
@@ -323,7 +332,7 @@ unsigned long __init setup_memory(void)
 			(ulong) pfn_to_kaddr(max_low_pfn));
 	for_each_online_node(nid) {
 		node_remap_start_vaddr[nid] = pfn_to_kaddr(
-				highstart_pfn + node_remap_offset[nid]);
+				kva_start_pfn + node_remap_offset[nid]);
 		/* Init the node remap allocator */
 		node_remap_end_vaddr[nid] = node_remap_start_vaddr[nid] +
 			(node_remap_size[nid] * PAGE_SIZE);
@@ -338,7 +347,6 @@ unsigned long __init setup_memory(void)
 	}
 	printk("High memory starts at vaddr %08lx\n",
 			(ulong) pfn_to_kaddr(highstart_pfn));
-	vmalloc_earlyreserve = reserve_pages * PAGE_SIZE;
 	for_each_online_node(nid)
 		find_max_pfn_node(nid);
 
@@ -348,6 +356,11 @@ unsigned long __init setup_memory(void)
 	return max_low_pfn;
 }
 
+void __init numa_kva_reserve(void)
+{
+	reserve_bootmem(PFN_PHYS(kva_start_pfn),PFN_PHYS(kva_pages));
+}
+
 void __init zone_sizes_init(void)
 {
 	int nid;
diff --git a/include/asm-i386/mmzone.h b/include/asm-i386/mmzone.h
index 22cb07cc8f32..61b073322006 100644
--- a/include/asm-i386/mmzone.h
+++ b/include/asm-i386/mmzone.h
@@ -38,10 +38,16 @@ static inline void get_memcfg_numa(void)
 }
 
 extern int early_pfn_to_nid(unsigned long pfn);
+extern void numa_kva_reserve(void);
 
 #else /* !CONFIG_NUMA */
+
 #define get_memcfg_numa get_memcfg_numa_flat
 #define get_zholes_size(n) (0)
+
+static inline void numa_kva_reserve(void)
+{
+}
 #endif /* CONFIG_NUMA */
 
 #ifdef CONFIG_DISCONTIGMEM
-- 
cgit v1.2.1


From 2d1a07d487d8b36658404839cdf03a974968cefd Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:03 -0700
Subject: [PATCH] bootmem: remove useless __init in header file

__init in headers is pretty useless because the compiler doesn't check it, and
they get out of sync relatively frequently.  So if you see an __init in a
header file, it's quite unreliable and you need to check the definition
anyway.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 46 +++++++++++++++++++++++-----------------------
 1 file changed, 23 insertions(+), 23 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index e319c649e4fd..c7124d4ce7cf 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -41,23 +41,23 @@ typedef struct bootmem_data {
 	struct list_head list;
 } bootmem_data_t;
 
-extern unsigned long __init bootmem_bootmap_pages (unsigned long);
-extern unsigned long __init init_bootmem (unsigned long addr, unsigned long memend);
-extern void __init free_bootmem (unsigned long addr, unsigned long size);
-extern void * __init __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __init __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __init __alloc_bootmem_low(unsigned long size,
+extern unsigned long bootmem_bootmap_pages (unsigned long);
+extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
+extern void free_bootmem (unsigned long addr, unsigned long size);
+extern void * __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem_low(unsigned long size,
 					 unsigned long align,
 					 unsigned long goal);
-extern void * __init __alloc_bootmem_low_node(pg_data_t *pgdat,
+extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
 					      unsigned long size,
 					      unsigned long align,
 					      unsigned long goal);
-extern void * __init __alloc_bootmem_core(struct bootmem_data *bdata,
+extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
 		unsigned long size, unsigned long align, unsigned long goal,
 		unsigned long limit);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-extern void __init reserve_bootmem (unsigned long addr, unsigned long size);
+extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
 	__alloc_bootmem((x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
@@ -67,12 +67,12 @@ extern void __init reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem_low_pages(x) \
 	__alloc_bootmem_low((x), PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-extern unsigned long __init free_all_bootmem (void);
-extern void * __init __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
-extern unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
-extern void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
-extern void __init free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
-extern unsigned long __init free_all_bootmem_node (pg_data_t *pgdat);
+extern unsigned long free_all_bootmem (void);
+extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
+extern unsigned long init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
+extern void reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
+extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
+extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node((pgdat), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
@@ -94,14 +94,14 @@ static inline void *alloc_remap(int nid, unsigned long size)
 extern unsigned long __meminitdata nr_kernel_pages;
 extern unsigned long nr_all_pages;
 
-extern void *__init alloc_large_system_hash(const char *tablename,
-					    unsigned long bucketsize,
-					    unsigned long numentries,
-					    int scale,
-					    int flags,
-					    unsigned int *_hash_shift,
-					    unsigned int *_hash_mask,
-					    unsigned long limit);
+extern void * alloc_large_system_hash(const char *tablename,
+				      unsigned long bucketsize,
+				      unsigned long numentries,
+				      int scale,
+				      int flags,
+				      unsigned int *_hash_shift,
+				      unsigned int *_hash_mask,
+				      unsigned long limit);
 
 #define HASH_HIGHMEM	0x00000001	/* Consider highmem? */
 #define HASH_EARLY	0x00000002	/* Allocating during early boot? */
-- 
cgit v1.2.1


From 69d49e681d7c7ed864a1ba45efc1e78433df8b9a Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:04 -0700
Subject: [PATCH] bootmem: mark link_bootmem() as part of the __init section

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/bootmem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 50353e0dac12..70f1528a3c8a 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -54,7 +54,7 @@ unsigned long __init bootmem_bootmap_pages (unsigned long pages)
 /*
  * link bdata in order
  */
-static void link_bootmem(bootmem_data_t *bdata)
+static void __init link_bootmem(bootmem_data_t *bdata)
 {
 	bootmem_data_t *ent;
 	if (list_empty(&bdata_list)) {
-- 
cgit v1.2.1


From 71fb2e8f8753b66b1f4295aa264a2eb4e69381e8 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:05 -0700
Subject: [PATCH] bootmem: remove useless parentheses in bootmem header file

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index c7124d4ce7cf..fa2523f47628 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -59,13 +59,13 @@ extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
-	__alloc_bootmem((x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
-	__alloc_bootmem_low((x), SMP_CACHE_BYTES, 0)
+	__alloc_bootmem_low(x, SMP_CACHE_BYTES, 0)
 #define alloc_bootmem_pages(x) \
-	__alloc_bootmem((x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem(x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages(x) \
-	__alloc_bootmem_low((x), PAGE_SIZE, 0)
+	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 extern unsigned long free_all_bootmem (void);
 extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
@@ -75,11 +75,11 @@ extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned lo
 extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
-	__alloc_bootmem_node((pgdat), (x), SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_node(pgdat, x) \
-	__alloc_bootmem_node((pgdat), (x), PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
+	__alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages_node(pgdat, x) \
-	__alloc_bootmem_low_node((pgdat), (x), PAGE_SIZE, 0)
+	__alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
 #ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP
-- 
cgit v1.2.1


From bb0923a66820718f636736b22ce47372f79e3400 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:05 -0700
Subject: [PATCH] bootmem: limit to 80 columns width

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 42 +++++++++++++++++++++++++++++-------------
 mm/bootmem.c            | 28 ++++++++++++++++++----------
 2 files changed, 47 insertions(+), 23 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index fa2523f47628..1e8dddfb3083 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -44,18 +44,24 @@ typedef struct bootmem_data {
 extern unsigned long bootmem_bootmap_pages (unsigned long);
 extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
 extern void free_bootmem (unsigned long addr, unsigned long size);
-extern void * __alloc_bootmem (unsigned long size, unsigned long align, unsigned long goal);
-extern void * __alloc_bootmem_nopanic (unsigned long size, unsigned long align, unsigned long goal);
+extern void * __alloc_bootmem (unsigned long size,
+			       unsigned long align,
+			       unsigned long goal);
+extern void * __alloc_bootmem_nopanic (unsigned long size,
+				       unsigned long align,
+				       unsigned long goal);
 extern void * __alloc_bootmem_low(unsigned long size,
-					 unsigned long align,
-					 unsigned long goal);
+				  unsigned long align,
+				  unsigned long goal);
 extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
-					      unsigned long size,
-					      unsigned long align,
-					      unsigned long goal);
+				       unsigned long size,
+				       unsigned long align,
+				       unsigned long goal);
 extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
-		unsigned long size, unsigned long align, unsigned long goal,
-		unsigned long limit);
+				   unsigned long size,
+				   unsigned long align,
+				   unsigned long goal,
+				   unsigned long limit);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
@@ -68,10 +74,20 @@ extern void reserve_bootmem (unsigned long addr, unsigned long size);
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 extern unsigned long free_all_bootmem (void);
-extern void * __alloc_bootmem_node (pg_data_t *pgdat, unsigned long size, unsigned long align, unsigned long goal);
-extern unsigned long init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn);
-extern void reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size);
-extern void free_bootmem_node (pg_data_t *pgdat, unsigned long addr, unsigned long size);
+extern void * __alloc_bootmem_node (pg_data_t *pgdat,
+				    unsigned long size,
+				    unsigned long align,
+				    unsigned long goal);
+extern unsigned long init_bootmem_node (pg_data_t *pgdat,
+					unsigned long freepfn,
+					unsigned long startpfn,
+					unsigned long endpfn);
+extern void reserve_bootmem_node (pg_data_t *pgdat,
+				  unsigned long physaddr,
+				  unsigned long size);
+extern void free_bootmem_node (pg_data_t *pgdat,
+			       unsigned long addr,
+			       unsigned long size);
 extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 70f1528a3c8a..9083f5029673 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -102,7 +102,8 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
  * might be used for boot-time allocations - or it might get added
  * to the free page pool later on.
  */
-static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long addr, unsigned long size)
+static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
+					unsigned long size)
 {
 	unsigned long i;
 	/*
@@ -127,7 +128,8 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long add
 		}
 }
 
-static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr, unsigned long size)
+static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
+				     unsigned long size)
 {
 	unsigned long i;
 	unsigned long start;
@@ -355,17 +357,20 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	return total;
 }
 
-unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn, unsigned long startpfn, unsigned long endpfn)
+unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn,
+				unsigned long startpfn, unsigned long endpfn)
 {
 	return(init_bootmem_core(pgdat, freepfn, startpfn, endpfn));
 }
 
-void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size)
+void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
+				  unsigned long size)
 {
 	reserve_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr, unsigned long size)
+void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
+			       unsigned long size)
 {
 	free_bootmem_core(pgdat->bdata, physaddr, size);
 }
@@ -399,7 +404,8 @@ unsigned long __init free_all_bootmem (void)
 	return(free_all_bootmem_core(NODE_DATA(0)));
 }
 
-void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
+				      unsigned long goal)
 {
 	bootmem_data_t *bdata;
 	void *ptr;
@@ -410,7 +416,8 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align, u
 	return NULL;
 }
 
-void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem(unsigned long size, unsigned long align,
+			      unsigned long goal)
 {
 	void *mem = __alloc_bootmem_nopanic(size,align,goal);
 	if (mem)
@@ -424,8 +431,8 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align, unsigned
 }
 
 
-void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigned long align,
-				   unsigned long goal)
+void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
+				   unsigned long align, unsigned long goal)
 {
 	void *ptr;
 
@@ -438,7 +445,8 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size, unsigne
 
 #define LOW32LIMIT 0xffffffff
 
-void * __init __alloc_bootmem_low(unsigned long size, unsigned long align, unsigned long goal)
+void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
+				  unsigned long goal)
 {
 	bootmem_data_t *bdata;
 	void *ptr;
-- 
cgit v1.2.1


From e786e86a542ccc1133f333402526ad00b9c088ae Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:06 -0700
Subject: [PATCH] bootmem: remove useless headers inclusions

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h |  5 +----
 mm/bootmem.c            | 10 +++-------
 2 files changed, 4 insertions(+), 11 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 1e8dddfb3083..b2067e4a4486 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -4,11 +4,8 @@
 #ifndef _LINUX_BOOTMEM_H
 #define _LINUX_BOOTMEM_H
 
-#include <asm/pgtable.h>
-#include <asm/dma.h>
-#include <linux/cache.h>
-#include <linux/init.h>
 #include <linux/mmzone.h>
+#include <asm/dma.h>
 
 /*
  *  simple boot-time physical memory area allocator.
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9083f5029673..e136c086fd98 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -8,17 +8,13 @@
  *  free memory collector. It's used to deal with reserved
  *  system memory and memory holes as well.
  */
-
-#include <linux/mm.h>
-#include <linux/kernel_stat.h>
-#include <linux/swap.h>
-#include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/bootmem.h>
-#include <linux/mmzone.h>
 #include <linux/module.h>
-#include <asm/dma.h>
+
+#include <asm/bug.h>
 #include <asm/io.h>
+
 #include "internal.h"
 
 /*
-- 
cgit v1.2.1


From bbc7b92e337ac349ca917f9bf0b6be4743c14f3a Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:07 -0700
Subject: [PATCH] bootmem: use pfn/page conversion macros

It also creates get_mapsize() helper in order to make the code more readable
when it calculates the boot bitmap size.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/bootmem.c | 82 +++++++++++++++++++++++++++++++++---------------------------
 1 file changed, 45 insertions(+), 37 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index e136c086fd98..23c3317c09ed 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -9,6 +9,7 @@
  *  system memory and memory holes as well.
  */
 #include <linux/init.h>
+#include <linux/pfn.h>
 #include <linux/bootmem.h>
 #include <linux/module.h>
 
@@ -68,6 +69,18 @@ static void __init link_bootmem(bootmem_data_t *bdata)
 	return;
 }
 
+/*
+ * Given an initialised bdata, it returns the size of the boot bitmap
+ */
+static unsigned long __init get_mapsize(bootmem_data_t *bdata)
+{
+	unsigned long mapsize;
+	unsigned long start = PFN_DOWN(bdata->node_boot_start);
+	unsigned long end = bdata->node_low_pfn;
+
+	mapsize = ((end - start) + 7) / 8;
+	return ALIGN(mapsize, sizeof(long));
+}
 
 /*
  * Called once to set up the allocator itself.
@@ -76,11 +89,10 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
 	unsigned long mapstart, unsigned long start, unsigned long end)
 {
 	bootmem_data_t *bdata = pgdat->bdata;
-	unsigned long mapsize = ((end - start)+7)/8;
+	unsigned long mapsize;
 
-	mapsize = ALIGN(mapsize, sizeof(long));
-	bdata->node_bootmem_map = phys_to_virt(mapstart << PAGE_SHIFT);
-	bdata->node_boot_start = (start << PAGE_SHIFT);
+	bdata->node_bootmem_map = phys_to_virt(PFN_PHYS(mapstart));
+	bdata->node_boot_start = PFN_PHYS(start);
 	bdata->node_low_pfn = end;
 	link_bootmem(bdata);
 
@@ -88,6 +100,7 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
 	 * Initially all pages are reserved - setup_arch() has to
 	 * register free RAM areas explicitly.
 	 */
+	mapsize = get_mapsize(bdata);
 	memset(bdata->node_bootmem_map, 0xff, mapsize);
 
 	return mapsize;
@@ -101,20 +114,19 @@ static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
 static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 					unsigned long size)
 {
+	unsigned long sidx, eidx;
 	unsigned long i;
+
 	/*
 	 * round up, partially reserved pages are considered
 	 * fully reserved.
 	 */
-	unsigned long sidx = (addr - bdata->node_boot_start)/PAGE_SIZE;
-	unsigned long eidx = (addr + size - bdata->node_boot_start + 
-							PAGE_SIZE-1)/PAGE_SIZE;
-	unsigned long end = (addr + size + PAGE_SIZE-1)/PAGE_SIZE;
-
 	BUG_ON(!size);
-	BUG_ON(sidx >= eidx);
-	BUG_ON((addr >> PAGE_SHIFT) >= bdata->node_low_pfn);
-	BUG_ON(end > bdata->node_low_pfn);
+	BUG_ON(PFN_DOWN(addr) >= bdata->node_low_pfn);
+	BUG_ON(PFN_UP(addr + size) > bdata->node_low_pfn);
+
+	sidx = PFN_DOWN(addr - bdata->node_boot_start);
+	eidx = PFN_UP(addr + size - bdata->node_boot_start);
 
 	for (i = sidx; i < eidx; i++)
 		if (test_and_set_bit(i, bdata->node_bootmem_map)) {
@@ -127,18 +139,15 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata, unsigned long add
 static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 				     unsigned long size)
 {
+	unsigned long sidx, eidx;
 	unsigned long i;
-	unsigned long start;
+
 	/*
 	 * round down end of usable mem, partially free pages are
 	 * considered reserved.
 	 */
-	unsigned long sidx;
-	unsigned long eidx = (addr + size - bdata->node_boot_start)/PAGE_SIZE;
-	unsigned long end = (addr + size)/PAGE_SIZE;
-
 	BUG_ON(!size);
-	BUG_ON(end > bdata->node_low_pfn);
+	BUG_ON(PFN_DOWN(addr + size) > bdata->node_low_pfn);
 
 	if (addr < bdata->last_success)
 		bdata->last_success = addr;
@@ -146,8 +155,8 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 	/*
 	 * Round up the beginning of the address.
 	 */
-	start = (addr + PAGE_SIZE-1) / PAGE_SIZE;
-	sidx = start - (bdata->node_boot_start/PAGE_SIZE);
+	sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start);
+	eidx = PFN_DOWN(addr + size - bdata->node_boot_start);
 
 	for (i = sidx; i < eidx; i++) {
 		if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map)))
@@ -173,7 +182,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	      unsigned long align, unsigned long goal, unsigned long limit)
 {
 	unsigned long offset, remaining_size, areasize, preferred;
-	unsigned long i, start = 0, incr, eidx, end_pfn = bdata->node_low_pfn;
+	unsigned long i, start = 0, incr, eidx, end_pfn;
 	void *ret;
 
 	if(!size) {
@@ -185,23 +194,22 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	if (limit && bdata->node_boot_start >= limit)
 		return NULL;
 
-        limit >>=PAGE_SHIFT;
+	end_pfn = bdata->node_low_pfn;
+	limit = PFN_DOWN(limit);
 	if (limit && end_pfn > limit)
 		end_pfn = limit;
 
-	eidx = end_pfn - (bdata->node_boot_start >> PAGE_SHIFT);
+	eidx = end_pfn - PFN_DOWN(bdata->node_boot_start);
 	offset = 0;
-	if (align &&
-	    (bdata->node_boot_start & (align - 1UL)) != 0)
-		offset = (align - (bdata->node_boot_start & (align - 1UL)));
-	offset >>= PAGE_SHIFT;
+	if (align && (bdata->node_boot_start & (align - 1UL)) != 0)
+		offset = align - (bdata->node_boot_start & (align - 1UL));
+	offset = PFN_DOWN(offset);
 
 	/*
 	 * We try to allocate bootmem pages above 'goal'
 	 * first, then we try to allocate lower pages.
 	 */
-	if (goal && (goal >= bdata->node_boot_start) && 
-	    ((goal >> PAGE_SHIFT) < end_pfn)) {
+	if (goal && goal >= bdata->node_boot_start && PFN_DOWN(goal) < end_pfn) {
 		preferred = goal - bdata->node_boot_start;
 
 		if (bdata->last_success >= preferred)
@@ -210,9 +218,8 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	} else
 		preferred = 0;
 
-	preferred = ALIGN(preferred, align) >> PAGE_SHIFT;
-	preferred += offset;
-	areasize = (size+PAGE_SIZE-1)/PAGE_SIZE;
+	preferred = PFN_DOWN(ALIGN(preferred, align)) + offset;
+	areasize = (size + PAGE_SIZE-1) / PAGE_SIZE;
 	incr = align >> PAGE_SHIFT ? : 1;
 
 restart_scan:
@@ -243,7 +250,7 @@ restart_scan:
 	return NULL;
 
 found:
-	bdata->last_success = start << PAGE_SHIFT;
+	bdata->last_success = PFN_PHYS(start);
 	BUG_ON(start >= eidx);
 
 	/*
@@ -301,8 +308,8 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 
 	count = 0;
 	/* first extant page of the node */
-	pfn = bdata->node_boot_start >> PAGE_SHIFT;
-	idx = bdata->node_low_pfn - (bdata->node_boot_start >> PAGE_SHIFT);
+	pfn = PFN_DOWN(bdata->node_boot_start);
+	idx = bdata->node_low_pfn - pfn;
 	map = bdata->node_bootmem_map;
 	/* Check physaddr is O(LOG2(BITS_PER_LONG)) page aligned */
 	if (bdata->node_boot_start == 0 ||
@@ -343,9 +350,10 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	 */
 	page = virt_to_page(bdata->node_bootmem_map);
 	count = 0;
-	for (i = 0; i < ((bdata->node_low_pfn-(bdata->node_boot_start >> PAGE_SHIFT))/8 + PAGE_SIZE-1)/PAGE_SIZE; i++,page++) {
-		count++;
+	idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT;
+	for (i = 0; i < idx; i++, page++) {
 		__free_pages_bootmem(page, 0);
+		count++;
 	}
 	total += count;
 	bdata->node_bootmem_map = NULL;
-- 
cgit v1.2.1


From f71bf0cac730ccb5ebcdf21747db75ae0445ccde Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Mon, 25 Sep 2006 23:31:08 -0700
Subject: [PATCH] bootmem: miscellaneous coding style fixes

It fixes various coding style issues, specially when spaces are useless.  For
example '*' go next to the function name.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/bootmem.h | 95 +++++++++++++++++++++++++------------------------
 mm/bootmem.c            | 81 ++++++++++++++++++++++-------------------
 2 files changed, 93 insertions(+), 83 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index b2067e4a4486..31e9abb6d977 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -38,29 +38,30 @@ typedef struct bootmem_data {
 	struct list_head list;
 } bootmem_data_t;
 
-extern unsigned long bootmem_bootmap_pages (unsigned long);
-extern unsigned long init_bootmem (unsigned long addr, unsigned long memend);
-extern void free_bootmem (unsigned long addr, unsigned long size);
-extern void * __alloc_bootmem (unsigned long size,
-			       unsigned long align,
-			       unsigned long goal);
-extern void * __alloc_bootmem_nopanic (unsigned long size,
-				       unsigned long align,
-				       unsigned long goal);
-extern void * __alloc_bootmem_low(unsigned long size,
+extern unsigned long bootmem_bootmap_pages(unsigned long);
+extern unsigned long init_bootmem(unsigned long addr, unsigned long memend);
+extern void free_bootmem(unsigned long addr, unsigned long size);
+extern void *__alloc_bootmem(unsigned long size,
+			     unsigned long align,
+			     unsigned long goal);
+extern void *__alloc_bootmem_nopanic(unsigned long size,
+				     unsigned long align,
+				     unsigned long goal);
+extern void *__alloc_bootmem_low(unsigned long size,
+				 unsigned long align,
+				 unsigned long goal);
+extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
+				      unsigned long size,
+				      unsigned long align,
+				      unsigned long goal);
+extern void *__alloc_bootmem_core(struct bootmem_data *bdata,
+				  unsigned long size,
 				  unsigned long align,
-				  unsigned long goal);
-extern void * __alloc_bootmem_low_node(pg_data_t *pgdat,
-				       unsigned long size,
-				       unsigned long align,
-				       unsigned long goal);
-extern void * __alloc_bootmem_core(struct bootmem_data *bdata,
-				   unsigned long size,
-				   unsigned long align,
-				   unsigned long goal,
-				   unsigned long limit);
+				  unsigned long goal,
+				  unsigned long limit);
+
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-extern void reserve_bootmem (unsigned long addr, unsigned long size);
+extern void reserve_bootmem(unsigned long addr, unsigned long size);
 #define alloc_bootmem(x) \
 	__alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
@@ -70,22 +71,24 @@ extern void reserve_bootmem (unsigned long addr, unsigned long size);
 #define alloc_bootmem_low_pages(x) \
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-extern unsigned long free_all_bootmem (void);
-extern void * __alloc_bootmem_node (pg_data_t *pgdat,
-				    unsigned long size,
-				    unsigned long align,
-				    unsigned long goal);
-extern unsigned long init_bootmem_node (pg_data_t *pgdat,
-					unsigned long freepfn,
-					unsigned long startpfn,
-					unsigned long endpfn);
-extern void reserve_bootmem_node (pg_data_t *pgdat,
-				  unsigned long physaddr,
-				  unsigned long size);
-extern void free_bootmem_node (pg_data_t *pgdat,
-			       unsigned long addr,
-			       unsigned long size);
-extern unsigned long free_all_bootmem_node (pg_data_t *pgdat);
+
+extern unsigned long free_all_bootmem(void);
+extern unsigned long free_all_bootmem_node(pg_data_t *pgdat);
+extern void *__alloc_bootmem_node(pg_data_t *pgdat,
+				  unsigned long size,
+				  unsigned long align,
+				  unsigned long goal);
+extern unsigned long init_bootmem_node(pg_data_t *pgdat,
+				       unsigned long freepfn,
+				       unsigned long startpfn,
+				       unsigned long endpfn);
+extern void reserve_bootmem_node(pg_data_t *pgdat,
+				 unsigned long physaddr,
+				 unsigned long size);
+extern void free_bootmem_node(pg_data_t *pgdat,
+			      unsigned long addr,
+			      unsigned long size);
+
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
@@ -102,19 +105,19 @@ static inline void *alloc_remap(int nid, unsigned long size)
 {
 	return NULL;
 }
-#endif
+#endif /* CONFIG_HAVE_ARCH_ALLOC_REMAP */
 
 extern unsigned long __meminitdata nr_kernel_pages;
 extern unsigned long nr_all_pages;
 
-extern void * alloc_large_system_hash(const char *tablename,
-				      unsigned long bucketsize,
-				      unsigned long numentries,
-				      int scale,
-				      int flags,
-				      unsigned int *_hash_shift,
-				      unsigned int *_hash_mask,
-				      unsigned long limit);
+extern void *alloc_large_system_hash(const char *tablename,
+				     unsigned long bucketsize,
+				     unsigned long numentries,
+				     int scale,
+				     int flags,
+				     unsigned int *_hash_shift,
+				     unsigned int *_hash_mask,
+				     unsigned long limit);
 
 #define HASH_HIGHMEM	0x00000001	/* Consider highmem? */
 #define HASH_EARLY	0x00000002	/* Allocating during early boot? */
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 23c3317c09ed..f0f85fa713ca 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -38,7 +38,7 @@ unsigned long saved_max_pfn;
 #endif
 
 /* return the number of _pages_ that will be allocated for the boot bitmap */
-unsigned long __init bootmem_bootmap_pages (unsigned long pages)
+unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
 	unsigned long mapsize;
 
@@ -48,12 +48,14 @@ unsigned long __init bootmem_bootmap_pages (unsigned long pages)
 
 	return mapsize;
 }
+
 /*
  * link bdata in order
  */
 static void __init link_bootmem(bootmem_data_t *bdata)
 {
 	bootmem_data_t *ent;
+
 	if (list_empty(&bdata_list)) {
 		list_add(&bdata->list, &bdata_list);
 		return;
@@ -66,7 +68,6 @@ static void __init link_bootmem(bootmem_data_t *bdata)
 		}
 	}
 	list_add_tail(&bdata->list, &bdata_list);
-	return;
 }
 
 /*
@@ -85,7 +86,7 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata)
 /*
  * Called once to set up the allocator itself.
  */
-static unsigned long __init init_bootmem_core (pg_data_t *pgdat,
+static unsigned long __init init_bootmem_core(pg_data_t *pgdat,
 	unsigned long mapstart, unsigned long start, unsigned long end)
 {
 	bootmem_data_t *bdata = pgdat->bdata;
@@ -185,7 +186,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	unsigned long i, start = 0, incr, eidx, end_pfn;
 	void *ret;
 
-	if(!size) {
+	if (!size) {
 		printk("__alloc_bootmem_core(): zero-sized request\n");
 		BUG();
 	}
@@ -234,7 +235,7 @@ restart_scan:
 		for (j = i + 1; j < i + areasize; ++j) {
 			if (j >= eidx)
 				goto fail_block;
-			if (test_bit (j, bdata->node_bootmem_map))
+			if (test_bit(j, bdata->node_bootmem_map))
 				goto fail_block;
 		}
 		start = i;
@@ -262,19 +263,21 @@ found:
 	    bdata->last_offset && bdata->last_pos+1 == start) {
 		offset = ALIGN(bdata->last_offset, align);
 		BUG_ON(offset > PAGE_SIZE);
-		remaining_size = PAGE_SIZE-offset;
+		remaining_size = PAGE_SIZE - offset;
 		if (size < remaining_size) {
 			areasize = 0;
 			/* last_pos unchanged */
-			bdata->last_offset = offset+size;
-			ret = phys_to_virt(bdata->last_pos*PAGE_SIZE + offset +
-						bdata->node_boot_start);
+			bdata->last_offset = offset + size;
+			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
+					   offset +
+					   bdata->node_boot_start);
 		} else {
 			remaining_size = size - remaining_size;
-			areasize = (remaining_size+PAGE_SIZE-1)/PAGE_SIZE;
-			ret = phys_to_virt(bdata->last_pos*PAGE_SIZE + offset +
-						bdata->node_boot_start);
-			bdata->last_pos = start+areasize-1;
+			areasize = (remaining_size + PAGE_SIZE-1) / PAGE_SIZE;
+			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
+					   offset +
+					   bdata->node_boot_start);
+			bdata->last_pos = start + areasize - 1;
 			bdata->last_offset = remaining_size;
 		}
 		bdata->last_offset &= ~PAGE_MASK;
@@ -287,7 +290,7 @@ found:
 	/*
 	 * Reserve the area now:
 	 */
-	for (i = start; i < start+areasize; i++)
+	for (i = start; i < start + areasize; i++)
 		if (unlikely(test_and_set_bit(i, bdata->node_bootmem_map)))
 			BUG();
 	memset(ret, 0, size);
@@ -338,7 +341,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 				}
 			}
 		} else {
-			i+=BITS_PER_LONG;
+			i += BITS_PER_LONG;
 		}
 		pfn += BITS_PER_LONG;
 	}
@@ -361,51 +364,51 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	return total;
 }
 
-unsigned long __init init_bootmem_node (pg_data_t *pgdat, unsigned long freepfn,
+unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
 				unsigned long startpfn, unsigned long endpfn)
 {
-	return(init_bootmem_core(pgdat, freepfn, startpfn, endpfn));
+	return init_bootmem_core(pgdat, freepfn, startpfn, endpfn);
 }
 
-void __init reserve_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
-				  unsigned long size)
+void __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+				 unsigned long size)
 {
 	reserve_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-void __init free_bootmem_node (pg_data_t *pgdat, unsigned long physaddr,
-			       unsigned long size)
+void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+			      unsigned long size)
 {
 	free_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
-unsigned long __init free_all_bootmem_node (pg_data_t *pgdat)
+unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
 {
-	return(free_all_bootmem_core(pgdat));
+	return free_all_bootmem_core(pgdat);
 }
 
-unsigned long __init init_bootmem (unsigned long start, unsigned long pages)
+unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
 {
 	max_low_pfn = pages;
 	min_low_pfn = start;
-	return(init_bootmem_core(NODE_DATA(0), start, 0, pages));
+	return init_bootmem_core(NODE_DATA(0), start, 0, pages);
 }
 
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-void __init reserve_bootmem (unsigned long addr, unsigned long size)
+void __init reserve_bootmem(unsigned long addr, unsigned long size)
 {
 	reserve_bootmem_core(NODE_DATA(0)->bdata, addr, size);
 }
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
-void __init free_bootmem (unsigned long addr, unsigned long size)
+void __init free_bootmem(unsigned long addr, unsigned long size)
 {
 	free_bootmem_core(NODE_DATA(0)->bdata, addr, size);
 }
 
-unsigned long __init free_all_bootmem (void)
+unsigned long __init free_all_bootmem(void)
 {
-	return(free_all_bootmem_core(NODE_DATA(0)));
+	return free_all_bootmem_core(NODE_DATA(0));
 }
 
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
@@ -414,9 +417,11 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 	bootmem_data_t *bdata;
 	void *ptr;
 
-	list_for_each_entry(bdata, &bdata_list, list)
-		if ((ptr = __alloc_bootmem_core(bdata, size, align, goal, 0)))
-			return(ptr);
+	list_for_each_entry(bdata, &bdata_list, list) {
+		ptr = __alloc_bootmem_core(bdata, size, align, goal, 0);
+		if (ptr)
+			return ptr;
+	}
 	return NULL;
 }
 
@@ -424,6 +429,7 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 			      unsigned long goal)
 {
 	void *mem = __alloc_bootmem_nopanic(size,align,goal);
+
 	if (mem)
 		return mem;
 	/*
@@ -442,7 +448,7 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 
 	ptr = __alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
 	if (ptr)
-		return (ptr);
+		return ptr;
 
 	return __alloc_bootmem(size, align, goal);
 }
@@ -455,10 +461,11 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	bootmem_data_t *bdata;
 	void *ptr;
 
-	list_for_each_entry(bdata, &bdata_list, list)
-		if ((ptr = __alloc_bootmem_core(bdata, size,
-						 align, goal, LOW32LIMIT)))
-			return(ptr);
+	list_for_each_entry(bdata, &bdata_list, list) {
+		ptr = __alloc_bootmem_core(bdata, size, align, goal, LOW32LIMIT);
+		if (ptr)
+			return ptr;
+	}
 
 	/*
 	 * Whoops, we cannot satisfy the allocation request.
-- 
cgit v1.2.1


From 776ed98b842ee8551793f842fe028c8091f3633e Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:09 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: remove two strange uses of MAX_NR_ZONES

I keep seeing zones on various platforms that are never used and wonder why we
compile support for them into the kernel.  Counters show up for HIGHMEM and
DMA32 that are alway zero.

This patch allows the removal of ZONE_DMA32 for non x86_64 architectures and
it will get rid of ZONE_HIGHMEM for arches not using highmem (like 64 bit
architectures).  If an arch does not define CONFIG_HIGHMEM then ZONE_HIGHMEM
will not be defined.  Similarly if an arch does not define CONFIG_ZONE_DMA32
then ZONE_DMA32 will not be defined.

No current architecture uses all the 4 zones (DMA,DMA32,NORMAL,HIGH) that we
have now.  The patchset will reduce the number of zones for all platforms.

On many platforms that do not have DMA32 or HIGHMEM this will reduce the
number of zones by 50%.  F.e.  ia64 only uses DMA and NORMAL.

Large amounts of memory can be saved for larger systemss that may have a few
hundred NUMA nodes.

With ZONE_DMA32 and ZONE_HIGHMEM support optional MAX_NR_ZONES will be 2 for
many non i386 platforms and even for i386 without CONFIG_HIGHMEM set.

Tested on ia64, x86_64 and on i386 with and without highmem.

The patchset consists of 11 patches that are following this message.

One could go even further than this patchset and also make ZONE_DMA optional
because some platforms do not need a separate DMA zone and can do DMA to all
of memory.  This could reduce MAX_NR_ZONES to 1.  Such a patchset will
hopefully follow soon.

This patch:

Fix strange uses of MAX_NR_ZONES

Sometimes we use MAX_NR_ZONES - x to refer to a zone.  Make that explicit.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/init.c   | 2 +-
 arch/x86_64/mm/init.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 89e8486aac34..353453d550a4 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -657,7 +657,7 @@ void __init mem_init(void)
 int arch_add_memory(int nid, u64 start, u64 size)
 {
 	struct pglist_data *pgdata = &contig_page_data;
-	struct zone *zone = pgdata->node_zones + MAX_NR_ZONES-1;
+	struct zone *zone = pgdata->node_zones + ZONE_HIGHMEM;
 	unsigned long start_pfn = start >> PAGE_SHIFT;
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index d14fb2dfbfc4..52fd42c40c86 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -536,7 +536,7 @@ int memory_add_physaddr_to_nid(u64 start)
 int arch_add_memory(int nid, u64 start, u64 size)
 {
 	struct pglist_data *pgdat = NODE_DATA(nid);
-	struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2;
+	struct zone *zone = pgdat->node_zones + ZONE_NORMAL;
 	unsigned long start_pfn = start >> PAGE_SHIFT;
 	unsigned long nr_pages = size >> PAGE_SHIFT;
 	int ret;
-- 
cgit v1.2.1


From f06a96844a577c43249fce25809a4fae07407f46 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:10 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: fix MAX_NR_ZONES array initializations

Fix array initialization in lots of arches

The number of zones may now be reduced from 4 to 2 for many arches.  Fix the
array initialization for the zones array for all architectures so that it is
not initializing a fixed number of elements.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/mm/init.c             | 2 +-
 arch/frv/mm/init.c               | 2 +-
 arch/h8300/mm/init.c             | 2 +-
 arch/i386/kernel/setup.c         | 2 +-
 arch/i386/mm/discontig.c         | 2 +-
 arch/m32r/mm/init.c              | 2 +-
 arch/m68knommu/mm/init.c         | 2 +-
 arch/mips/mm/init.c              | 4 ++--
 arch/mips/sgi-ip27/ip27-memory.c | 2 +-
 arch/parisc/mm/init.c            | 2 +-
 arch/sh64/mm/init.c              | 2 +-
 11 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c
index 917dad1b74c8..550f4907d613 100644
--- a/arch/alpha/mm/init.c
+++ b/arch/alpha/mm/init.c
@@ -270,7 +270,7 @@ callback_init(void * kernel_end)
 void
 paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long zones_size[MAX_NR_ZONES] = {0, };
 	unsigned long dma_pfn, high_pfn;
 
 	dma_pfn = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c
index b5b4286f9dd4..3f3a0ed3539b 100644
--- a/arch/frv/mm/init.c
+++ b/arch/frv/mm/init.c
@@ -98,7 +98,7 @@ void show_mem(void)
  */
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long zones_size[MAX_NR_ZONES] = {0, };
 
 	/* allocate some pages for kernel housekeeping tasks */
 	empty_bad_page_table	= (unsigned long) alloc_bootmem_pages(PAGE_SIZE);
diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index d3d40bdc2d6a..e4f4199f97ab 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -138,7 +138,7 @@ void paging_init(void)
 #endif
 
 	{
-		unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+		unsigned long zones_size[MAX_NR_ZONES] = {0, };
 
 		zones_size[ZONE_DMA]     = 0 >> PAGE_SHIFT;
 		zones_size[ZONE_NORMAL]  = (end_mem - PAGE_OFFSET) >> PAGE_SHIFT;
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 27d4dc0d3ef1..060c68004bec 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1182,7 +1182,7 @@ static unsigned long __init setup_memory(void)
 
 void __init zone_sizes_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long zones_size[MAX_NR_ZONES] = { 0, };
 	unsigned int max_dma, low;
 
 	max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 2e36eff8aff9..07c300f93764 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -367,7 +367,7 @@ void __init zone_sizes_init(void)
 
 
 	for_each_online_node(nid) {
-		unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+		unsigned long zones_size[MAX_NR_ZONES] = {0, };
 		unsigned long *zholes_size;
 		unsigned int max_dma;
 
diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index b71348fec1f4..bbd97c85bc5d 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -100,7 +100,7 @@ void free_initrd_mem(unsigned long, unsigned long);
 #ifndef CONFIG_DISCONTIGMEM
 unsigned long __init zone_sizes_init(void)
 {
-	unsigned long  zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long  zones_size[MAX_NR_ZONES] = {0, };
 	unsigned long  max_dma;
 	unsigned long  low;
 	unsigned long  start_pfn;
diff --git a/arch/m68knommu/mm/init.c b/arch/m68knommu/mm/init.c
index e4c233eef195..06e538d1be3a 100644
--- a/arch/m68knommu/mm/init.c
+++ b/arch/m68knommu/mm/init.c
@@ -136,7 +136,7 @@ void paging_init(void)
 #endif
 
 	{
-		unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+		unsigned long zones_size[MAX_NR_ZONES] = {0, };
 
 		zones_size[ZONE_DMA] = 0 >> PAGE_SHIFT;
 		zones_size[ZONE_NORMAL] = (end_mem - PAGE_OFFSET) >> PAGE_SHIFT;
diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c
index c52497bb102a..5b06349af2d5 100644
--- a/arch/mips/mm/init.c
+++ b/arch/mips/mm/init.c
@@ -163,10 +163,10 @@ static int __init page_is_ram(unsigned long pagenr)
 
 void __init paging_init(void)
 {
-	unsigned long zones_size[] = { [0 ... MAX_NR_ZONES - 1] = 0 };
+	unsigned long zones_size[] = { 0, };
 	unsigned long max_dma, high, low;
 #ifndef CONFIG_FLATMEM
-	unsigned long zholes_size[] = { [0 ... MAX_NR_ZONES - 1] = 0 };
+	unsigned long zholes_size[] = { 0, };
 	unsigned long i, j, pfn;
 #endif
 
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index efe6971fc800..59bfc0fc3f45 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -508,7 +508,7 @@ extern unsigned long setup_zero_pages(void);
 
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long zones_size[MAX_NR_ZONES] = {0, };
 	unsigned node;
 
 	pagetable_init();
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index f2b96f1e0da7..c7329615ef94 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -809,7 +809,7 @@ void __init paging_init(void)
 	flush_tlb_all_local(NULL);
 
 	for (i = 0; i < npmem_ranges; i++) {
-		unsigned long zones_size[MAX_NR_ZONES] = { 0, 0, 0 };
+		unsigned long zones_size[MAX_NR_ZONES] = { 0, };
 
 		/* We have an IOMMU, so all memory can go into a single
 		   ZONE_DMA zone. */
diff --git a/arch/sh64/mm/init.c b/arch/sh64/mm/init.c
index 1169757fb38b..83295bd21aa7 100644
--- a/arch/sh64/mm/init.c
+++ b/arch/sh64/mm/init.c
@@ -110,7 +110,7 @@ void show_mem(void)
  */
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
+	unsigned long zones_size[MAX_NR_ZONES] = {0, };
 
 	pgd_init((unsigned long)swapper_pg_dir);
 	pgd_init((unsigned long)swapper_pg_dir +
-- 
cgit v1.2.1


From 182e8e237349e7b6354f45aee4780b6423fd6a50 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:10 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: make display of highmem counters
 conditional on CONFIG_HIGHMEM

Do not display HIGHMEM memory sizes if CONFIG_HIGHMEM is not set.

Make HIGHMEM dependent texts and make display of highmem counters optional

Some texts are depending on CONFIG_HIGHMEM.

Remove those strings and remove the display of highmem counter values if
CONFIG_HIGHMEM is not set.

[akpm@osdl.org: remove some ifdefs]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/base/node.c | 4 ++++
 fs/proc/proc_misc.c | 4 ++++
 mm/page_alloc.c     | 4 ----
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/base/node.c b/drivers/base/node.c
index e9b0957f15d1..e09f5c2c11ee 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -54,10 +54,12 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       "Node %d MemUsed:      %8lu kB\n"
 		       "Node %d Active:       %8lu kB\n"
 		       "Node %d Inactive:     %8lu kB\n"
+#ifdef CONFIG_HIGHMEM
 		       "Node %d HighTotal:    %8lu kB\n"
 		       "Node %d HighFree:     %8lu kB\n"
 		       "Node %d LowTotal:     %8lu kB\n"
 		       "Node %d LowFree:      %8lu kB\n"
+#endif
 		       "Node %d Dirty:        %8lu kB\n"
 		       "Node %d Writeback:    %8lu kB\n"
 		       "Node %d FilePages:    %8lu kB\n"
@@ -72,10 +74,12 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       nid, K(i.totalram - i.freeram),
 		       nid, K(active),
 		       nid, K(inactive),
+#ifdef CONFIG_HIGHMEM
 		       nid, K(i.totalhigh),
 		       nid, K(i.freehigh),
 		       nid, K(i.totalram - i.totalhigh),
 		       nid, K(i.freeram - i.freehigh),
+#endif
 		       nid, K(node_page_state(nid, NR_FILE_DIRTY)),
 		       nid, K(node_page_state(nid, NR_WRITEBACK)),
 		       nid, K(node_page_state(nid, NR_FILE_PAGES)),
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 942156225447..caa0a51560a0 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -157,10 +157,12 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		"SwapCached:   %8lu kB\n"
 		"Active:       %8lu kB\n"
 		"Inactive:     %8lu kB\n"
+#ifdef CONFIG_HIGHMEM
 		"HighTotal:    %8lu kB\n"
 		"HighFree:     %8lu kB\n"
 		"LowTotal:     %8lu kB\n"
 		"LowFree:      %8lu kB\n"
+#endif
 		"SwapTotal:    %8lu kB\n"
 		"SwapFree:     %8lu kB\n"
 		"Dirty:        %8lu kB\n"
@@ -183,10 +185,12 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		K(total_swapcache_pages),
 		K(active),
 		K(inactive),
+#ifdef CONFIG_HIGHMEM
 		K(i.totalhigh),
 		K(i.freehigh),
 		K(i.totalram-i.totalhigh),
 		K(i.freeram-i.freehigh),
+#endif
 		K(i.totalswap),
 		K(i.freeswap),
 		K(global_page_state(NR_FILE_DIRTY)),
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4b33878e9488..e26491cb5a27 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1281,10 +1281,6 @@ void show_free_areas(void)
 
 	get_zone_counts(&active, &inactive, &free);
 
-	printk("Free pages: %11ukB (%ukB HighMem)\n",
-		K(nr_free_pages()),
-		K(nr_free_highpages()));
-
 	printk("Active:%lu inactive:%lu dirty:%lu writeback:%lu "
 		"unstable:%lu free:%u slab:%lu mapped:%lu pagetables:%lu\n",
 		active,
-- 
cgit v1.2.1


From c1f60a5a419cc60aff27daffb150f5a3a3a79ef4 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:11 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: move HIGHMEM counters into highmem.c/.h

Move totalhigh_pages and nr_free_highpages() into highmem.c/.h

Move the totalhigh_pages definition into highmem.c/.h.  Move the
nr_free_highpages function into highmem.c

[yoichi_yuasa@tripeaks.co.jp: build fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/sgi-ip27/ip27-memory.c |  1 +
 arch/um/kernel/mem.c             |  2 ++
 include/linux/highmem.h          |  3 +++
 include/linux/swap.h             |  1 -
 mm/highmem.c                     | 13 +++++++++++++
 mm/page_alloc.c                  | 15 ---------------
 mm/shmem.c                       |  1 +
 7 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index 59bfc0fc3f45..16e5682b01f1 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -19,6 +19,7 @@
 #include <linux/swap.h>
 #include <linux/bootmem.h>
 #include <linux/pfn.h>
+#include <linux/highmem.h>
 #include <asm/page.h>
 #include <asm/sections.h>
 
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index 61280167c560..b39e624c3291 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -79,8 +79,10 @@ void mem_init(void)
 
 	/* this will put all low memory onto the freelists */
 	totalram_pages = free_all_bootmem();
+#ifdef CONFIG_HIGHMEM
 	totalhigh_pages = highmem >> PAGE_SHIFT;
 	totalram_pages += totalhigh_pages;
+#endif
 	num_physpages = totalram_pages;
 	max_pfn = totalram_pages;
 	printk(KERN_INFO "Memory: %luk available\n", 
diff --git a/include/linux/highmem.h b/include/linux/highmem.h
index 42620e723abb..fd7d12daa94f 100644
--- a/include/linux/highmem.h
+++ b/include/linux/highmem.h
@@ -24,11 +24,14 @@ static inline void flush_kernel_dcache_page(struct page *page)
 
 /* declarations for linux/mm/highmem.c */
 unsigned int nr_free_highpages(void);
+extern unsigned long totalhigh_pages;
 
 #else /* CONFIG_HIGHMEM */
 
 static inline unsigned int nr_free_highpages(void) { return 0; }
 
+#define totalhigh_pages 0
+
 #ifndef ARCH_HAS_KMAP
 static inline void *kmap(struct page *page)
 {
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 5e59184c9096..34a6bc3e6cf3 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -162,7 +162,6 @@ extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *
 
 /* linux/mm/page_alloc.c */
 extern unsigned long totalram_pages;
-extern unsigned long totalhigh_pages;
 extern unsigned long totalreserve_pages;
 extern long nr_swap_pages;
 extern unsigned int nr_free_pages(void);
diff --git a/mm/highmem.c b/mm/highmem.c
index 9b2a5403c447..ee5519b176ee 100644
--- a/mm/highmem.c
+++ b/mm/highmem.c
@@ -46,6 +46,19 @@ static void *mempool_alloc_pages_isa(gfp_t gfp_mask, void *data)
  */
 #ifdef CONFIG_HIGHMEM
 
+unsigned long totalhigh_pages __read_mostly;
+
+unsigned int nr_free_highpages (void)
+{
+	pg_data_t *pgdat;
+	unsigned int pages = 0;
+
+	for_each_online_pgdat(pgdat)
+		pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
+
+	return pages;
+}
+
 static int pkmap_count[LAST_PKMAP];
 static unsigned int last_pkmap_nr;
 static  __cacheline_aligned_in_smp DEFINE_SPINLOCK(kmap_lock);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e26491cb5a27..5cde54695cfb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -51,7 +51,6 @@ EXPORT_SYMBOL(node_online_map);
 nodemask_t node_possible_map __read_mostly = NODE_MASK_ALL;
 EXPORT_SYMBOL(node_possible_map);
 unsigned long totalram_pages __read_mostly;
-unsigned long totalhigh_pages __read_mostly;
 unsigned long totalreserve_pages __read_mostly;
 long nr_swap_pages;
 int percpu_pagelist_fraction;
@@ -1185,20 +1184,6 @@ unsigned int nr_free_pagecache_pages(void)
 {
 	return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER));
 }
-
-#ifdef CONFIG_HIGHMEM
-unsigned int nr_free_highpages (void)
-{
-	pg_data_t *pgdat;
-	unsigned int pages = 0;
-
-	for_each_online_pgdat(pgdat)
-		pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
-
-	return pages;
-}
-#endif
-
 #ifdef CONFIG_NUMA
 static void show_node(struct zone *zone)
 {
diff --git a/mm/shmem.c b/mm/shmem.c
index db21c51531ca..8631be45b40d 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -45,6 +45,7 @@
 #include <linux/namei.h>
 #include <linux/ctype.h>
 #include <linux/migrate.h>
+#include <linux/highmem.h>
 
 #include <asm/uaccess.h>
 #include <asm/div64.h>
-- 
cgit v1.2.1


From 98d2b0ebda72fc39cdefd3720d50b9b3ce409085 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:12 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: page allocator ZONE_HIGHMEM cleanup

page allocator ZONE_HIGHMEM fixups

1. We do not need to do an #ifdef in si_meminfo since both counters
   in use are zero if !CONFIG_HIGHMEM.

2. Add #ifdef in si_meminfo_node instead to avoid referencing zone
   information for ZONE_HIGHMEM if we do not have HIGHMEM
   (may not be there after the following patches).

3. Replace the use of ZONE_HIGHMEM with MAX_NR_ZONES in build_zonelists_node

4. build_zonelists_node: Remove BUG_ON for ZONE_HIGHMEM. Zone will
   be optional soon and thus BUG_ON cannot be triggered anymore.

5. init_free_area_core: Replace a use of ZONE_HIGHMEM with NR_MAX_ZONES.

[akpm@osdl.org: cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5cde54695cfb..13c102b95c57 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1199,13 +1199,8 @@ void si_meminfo(struct sysinfo *val)
 	val->sharedram = 0;
 	val->freeram = nr_free_pages();
 	val->bufferram = nr_blockdev_pages();
-#ifdef CONFIG_HIGHMEM
 	val->totalhigh = totalhigh_pages;
 	val->freehigh = nr_free_highpages();
-#else
-	val->totalhigh = 0;
-	val->freehigh = 0;
-#endif
 	val->mem_unit = PAGE_SIZE;
 }
 
@@ -1218,8 +1213,13 @@ void si_meminfo_node(struct sysinfo *val, int nid)
 
 	val->totalram = pgdat->node_present_pages;
 	val->freeram = nr_free_pages_pgdat(pgdat);
+#ifdef CONFIG_HIGHMEM
 	val->totalhigh = pgdat->node_zones[ZONE_HIGHMEM].present_pages;
 	val->freehigh = pgdat->node_zones[ZONE_HIGHMEM].free_pages;
+#else
+	val->totalhigh = 0;
+	val->freehigh = 0;
+#endif
 	val->mem_unit = PAGE_SIZE;
 }
 #endif
@@ -1344,14 +1344,11 @@ static int __meminit build_zonelists_node(pg_data_t *pgdat,
 {
 	struct zone *zone;
 
-	BUG_ON(zone_type > ZONE_HIGHMEM);
+	BUG_ON(zone_type >= MAX_NR_ZONES);
 
 	do {
 		zone = pgdat->node_zones + zone_type;
 		if (populated_zone(zone)) {
-#ifndef CONFIG_HIGHMEM
-			BUG_ON(zone_type > ZONE_NORMAL);
-#endif
 			zonelist->zones[nr_zones++] = zone;
 			check_highest_zone(zone_type);
 		}
@@ -1981,7 +1978,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		if (zholes_size)
 			realsize -= zholes_size[j];
 
-		if (j < ZONE_HIGHMEM)
+		if (!is_highmem_idx(j))
 			nr_kernel_pages += realsize;
 		nr_all_pages += realsize;
 
-- 
cgit v1.2.1


From 2f1b6248682f8b39ca3c7e549dfc216d26c4109b Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:13 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: use enum to define zones, reformat and
 comment

Use enum for zones and reformat zones dependent information

Add comments explaning the use of zones and add a zones_t type for zone
numbers.

Line up information that will be #ifdefd by the following patches.

[akpm@osdl.org: comment cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     |  7 +++---
 include/linux/mmzone.h | 66 +++++++++++++++++++++++++++++++++++---------------
 mm/page_alloc.c        | 24 +++++++++++++-----
 3 files changed, 69 insertions(+), 28 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 45678b036955..2db4229a0066 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -470,7 +470,7 @@ void split_page(struct page *page, unsigned int order);
 #define SECTIONS_MASK		((1UL << SECTIONS_WIDTH) - 1)
 #define ZONETABLE_MASK		((1UL << ZONETABLE_SHIFT) - 1)
 
-static inline unsigned long page_zonenum(struct page *page)
+static inline enum zone_type page_zonenum(struct page *page)
 {
 	return (page->flags >> ZONES_PGSHIFT) & ZONES_MASK;
 }
@@ -499,11 +499,12 @@ static inline unsigned long page_to_section(struct page *page)
 	return (page->flags >> SECTIONS_PGSHIFT) & SECTIONS_MASK;
 }
 
-static inline void set_page_zone(struct page *page, unsigned long zone)
+static inline void set_page_zone(struct page *page, enum zone_type zone)
 {
 	page->flags &= ~(ZONES_MASK << ZONES_PGSHIFT);
 	page->flags |= (zone & ZONES_MASK) << ZONES_PGSHIFT;
 }
+
 static inline void set_page_node(struct page *page, unsigned long node)
 {
 	page->flags &= ~(NODES_MASK << NODES_PGSHIFT);
@@ -515,7 +516,7 @@ static inline void set_page_section(struct page *page, unsigned long section)
 	page->flags |= (section & SECTIONS_MASK) << SECTIONS_PGSHIFT;
 }
 
-static inline void set_page_links(struct page *page, unsigned long zone,
+static inline void set_page_links(struct page *page, enum zone_type zone,
 	unsigned long node, unsigned long pfn)
 {
 	set_page_zone(page, zone);
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index f45163c528e8..03a5a6eb0ffa 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -88,14 +88,53 @@ struct per_cpu_pageset {
 #define zone_pcp(__z, __cpu) (&(__z)->pageset[(__cpu)])
 #endif
 
-#define ZONE_DMA		0
-#define ZONE_DMA32		1
-#define ZONE_NORMAL		2
-#define ZONE_HIGHMEM		3
+enum zone_type {
+	/*
+	 * ZONE_DMA is used when there are devices that are not able
+	 * to do DMA to all of addressable memory (ZONE_NORMAL). Then we
+	 * carve out the portion of memory that is needed for these devices.
+	 * The range is arch specific.
+	 *
+	 * Some examples
+	 *
+	 * Architecture		Limit
+	 * ---------------------------
+	 * parisc, ia64, sparc	<4G
+	 * s390			<2G
+	 * arm26		<48M
+	 * arm			Various
+	 * alpha		Unlimited or 0-16MB.
+	 *
+	 * i386, x86_64 and multiple other arches
+	 * 			<16M.
+	 */
+	ZONE_DMA,
+	/*
+	 * x86_64 needs two ZONE_DMAs because it supports devices that are
+	 * only able to do DMA to the lower 16M but also 32 bit devices that
+	 * can only do DMA areas below 4G.
+	 */
+	ZONE_DMA32,
+	/*
+	 * Normal addressable memory is in ZONE_NORMAL. DMA operations can be
+	 * performed on pages in ZONE_NORMAL if the DMA devices support
+	 * transfers to all addressable memory.
+	 */
+	ZONE_NORMAL,
+	/*
+	 * A memory area that is only addressable by the kernel through
+	 * mapping portions into its own address space. This is for example
+	 * used by i386 to allow the kernel to address the memory beyond
+	 * 900MB. The kernel will set up special mappings (page
+	 * table entries on i386) for each page that the kernel needs to
+	 * access.
+	 */
+	ZONE_HIGHMEM,
 
-#define MAX_NR_ZONES		4	/* Sync this with ZONES_SHIFT */
-#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
+	MAX_NR_ZONES
+};
 
+#define	ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 
 /*
  * When a memory allocation must conform to specific limitations (such
@@ -126,16 +165,6 @@ struct per_cpu_pageset {
 /* #define GFP_ZONETYPES       (GFP_ZONEMASK + 1) */           /* Non-loner */
 #define GFP_ZONETYPES  ((GFP_ZONEMASK + 1) / 2 + 1)            /* Loner */
 
-/*
- * On machines where it is needed (eg PCs) we divide physical memory
- * into multiple physical zones. On a 32bit PC we have 4 zones:
- *
- * ZONE_DMA	  < 16 MB	ISA DMA capable memory
- * ZONE_DMA32	     0 MB 	Empty
- * ZONE_NORMAL	16-896 MB	direct mapped by the kernel
- * ZONE_HIGHMEM	 > 896 MB	only page cache and user processes
- */
-
 struct zone {
 	/* Fields commonly accessed by the page allocator */
 	unsigned long		free_pages;
@@ -266,7 +295,6 @@ struct zone {
 	char			*name;
 } ____cacheline_internodealigned_in_smp;
 
-
 /*
  * The "priority" of VM scanning is how much of the queues we will scan in one
  * go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the
@@ -373,12 +401,12 @@ static inline int populated_zone(struct zone *zone)
 	return (!!zone->present_pages);
 }
 
-static inline int is_highmem_idx(int idx)
+static inline int is_highmem_idx(enum zone_type idx)
 {
 	return (idx == ZONE_HIGHMEM);
 }
 
-static inline int is_normal_idx(int idx)
+static inline int is_normal_idx(enum zone_type idx)
 {
 	return (idx == ZONE_NORMAL);
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 13c102b95c57..2410a3cb1c53 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -68,7 +68,11 @@ static void __free_pages_ok(struct page *page, unsigned int order);
  * TBD: should special case ZONE_DMA32 machines here - in those we normally
  * don't need any ZONE_NORMAL reservation
  */
-int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = { 256, 256, 32 };
+int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
+	 256,
+	 256,
+	 32
+};
 
 EXPORT_SYMBOL(totalram_pages);
 
@@ -79,7 +83,13 @@ EXPORT_SYMBOL(totalram_pages);
 struct zone *zone_table[1 << ZONETABLE_SHIFT] __read_mostly;
 EXPORT_SYMBOL(zone_table);
 
-static char *zone_names[MAX_NR_ZONES] = { "DMA", "DMA32", "Normal", "HighMem" };
+static char *zone_names[MAX_NR_ZONES] = {
+	 "DMA",
+	 "DMA32",
+	 "Normal",
+	 "HighMem"
+};
+
 int min_free_kbytes = 1024;
 
 unsigned long __meminitdata nr_kernel_pages;
@@ -1487,7 +1497,9 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, k, node, local_node;
+	int i, node, local_node;
+	enum zone_type k;
+	enum zone_type j;
 
 	local_node = pgdat->node_id;
 	for (i = 0; i < GFP_ZONETYPES; i++) {
@@ -1675,8 +1687,8 @@ void zone_init_free_lists(struct pglist_data *pgdat, struct zone *zone,
 }
 
 #define ZONETABLE_INDEX(x, zone_nr)	((x << ZONES_SHIFT) | zone_nr)
-void zonetable_add(struct zone *zone, int nid, int zid, unsigned long pfn,
-		unsigned long size)
+void zonetable_add(struct zone *zone, int nid, enum zone_type zid,
+		unsigned long pfn, unsigned long size)
 {
 	unsigned long snum = pfn_to_section_nr(pfn);
 	unsigned long end = pfn_to_section_nr(pfn + size);
@@ -1960,7 +1972,7 @@ __meminit int init_currently_empty_zone(struct zone *zone,
 static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		unsigned long *zones_size, unsigned long *zholes_size)
 {
-	unsigned long j;
+	enum zone_type j;
 	int nid = pgdat->node_id;
 	unsigned long zone_start_pfn = pgdat->node_start_pfn;
 	int ret;
-- 
cgit v1.2.1


From fb0e7942bdcbbd2f90e61cb4cfa4fa892a873f8a Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:13 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: make ZONE_DMA32 optional

Make ZONE_DMA32 optional

- Add #ifdefs around ZONE_DMA32 specific code and definitions.

- Add CONFIG_ZONE_DMA32 config option and use that for x86_64
  that alone needs this zone.

- Remove the use of CONFIG_DMA_IS_DMA32 and CONFIG_DMA_IS_NORMAL
  for ia64 and fix up the way per node ZVCs are calculated.

- Fall back to prior GFP_ZONEMASK of 0x03 if there is no
  DMA32 zone.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig      |  9 ---------
 arch/x86_64/Kconfig    |  4 ++++
 include/linux/gfp.h    |  2 +-
 include/linux/mmzone.h | 16 +++++++++++++---
 include/linux/vmstat.h |  4 +---
 mm/page_alloc.c        |  6 ++++++
 6 files changed, 25 insertions(+), 16 deletions(-)

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index db274da7dba1..f521f2f60a78 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -66,15 +66,6 @@ config IA64_UNCACHED_ALLOCATOR
 	bool
 	select GENERIC_ALLOCATOR
 
-config DMA_IS_DMA32
-	bool
-	default y
-
-config DMA_IS_NORMAL
-	bool
-	depends on IA64_SGI_SN2
-	default y
-
 config AUDIT_ARCH
 	bool
 	default y
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 6cd4878625f1..581ce9af0ec8 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -24,6 +24,10 @@ config X86
 	bool
 	default y
 
+config ZONE_DMA32
+	bool
+	default y
+
 config LOCKDEP_SUPPORT
 	bool
 	default y
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index cc9e60844484..14610b56c132 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -13,7 +13,7 @@ struct vm_area_struct;
 /* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low three bits) */
 #define __GFP_DMA	((__force gfp_t)0x01u)
 #define __GFP_HIGHMEM	((__force gfp_t)0x02u)
-#ifdef CONFIG_DMA_IS_DMA32
+#ifndef CONFIG_ZONE_DMA32
 #define __GFP_DMA32	((__force gfp_t)0x01)	/* ZONE_DMA is ZONE_DMA32 */
 #elif BITS_PER_LONG < 64
 #define __GFP_DMA32	((__force gfp_t)0x00)	/* ZONE_NORMAL is ZONE_DMA32 */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 03a5a6eb0ffa..adae3c915938 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -109,12 +109,14 @@ enum zone_type {
 	 * 			<16M.
 	 */
 	ZONE_DMA,
+#ifdef CONFIG_ZONE_DMA32
 	/*
 	 * x86_64 needs two ZONE_DMAs because it supports devices that are
 	 * only able to do DMA to the lower 16M but also 32 bit devices that
 	 * can only do DMA areas below 4G.
 	 */
 	ZONE_DMA32,
+#endif
 	/*
 	 * Normal addressable memory is in ZONE_NORMAL. DMA operations can be
 	 * performed on pages in ZONE_NORMAL if the DMA devices support
@@ -161,9 +163,13 @@ enum zone_type {
  *
  * NOTE! Make sure this matches the zones in <linux/gfp.h>
  */
-#define GFP_ZONEMASK	0x07
-/* #define GFP_ZONETYPES       (GFP_ZONEMASK + 1) */           /* Non-loner */
-#define GFP_ZONETYPES  ((GFP_ZONEMASK + 1) / 2 + 1)            /* Loner */
+#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
+
+#ifdef CONFIG_ZONE_DMA32
+#define GFP_ZONEMASK		0x07
+#else
+#define GFP_ZONEMASK		0x03
+#endif
 
 struct zone {
 	/* Fields commonly accessed by the page allocator */
@@ -429,7 +435,11 @@ static inline int is_normal(struct zone *zone)
 
 static inline int is_dma32(struct zone *zone)
 {
+#ifdef CONFIG_ZONE_DMA32
 	return zone == zone->zone_pgdat->node_zones + ZONE_DMA32;
+#else
+	return 0;
+#endif
 }
 
 static inline int is_dma(struct zone *zone)
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 2d9b1b60798a..9c6e62c56ec2 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -124,12 +124,10 @@ static inline unsigned long node_page_state(int node,
 	struct zone *zones = NODE_DATA(node)->node_zones;
 
 	return
-#ifndef CONFIG_DMA_IS_NORMAL
-#if !defined(CONFIG_DMA_IS_DMA32) && BITS_PER_LONG >= 64
+#ifdef CONFIG_ZONE_DMA32
 		zone_page_state(&zones[ZONE_DMA32], item) +
 #endif
 		zone_page_state(&zones[ZONE_NORMAL], item) +
-#endif
 #ifdef CONFIG_HIGHMEM
 		zone_page_state(&zones[ZONE_HIGHMEM], item) +
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 2410a3cb1c53..5b5cbb5e1816 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -70,7 +70,9 @@ static void __free_pages_ok(struct page *page, unsigned int order);
  */
 int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
 	 256,
+#ifdef CONFIG_ZONE_DMA32
 	 256,
+#endif
 	 32
 };
 
@@ -85,7 +87,9 @@ EXPORT_SYMBOL(zone_table);
 
 static char *zone_names[MAX_NR_ZONES] = {
 	 "DMA",
+#ifdef CONFIG_ZONE_DMA32
 	 "DMA32",
+#endif
 	 "Normal",
 	 "HighMem"
 };
@@ -1373,8 +1377,10 @@ static inline int highest_zone(int zone_bits)
 	int res = ZONE_NORMAL;
 	if (zone_bits & (__force int)__GFP_HIGHMEM)
 		res = ZONE_HIGHMEM;
+#ifdef CONFIG_ZONE_DMA32
 	if (zone_bits & (__force int)__GFP_DMA32)
 		res = ZONE_DMA32;
+#endif
 	if (zone_bits & (__force int)__GFP_DMA)
 		res = ZONE_DMA;
 	return res;
-- 
cgit v1.2.1


From e53ef38d05dd59ed281a35590e4a5b64d8ff4c52 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:14 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: make ZONE_HIGHMEM optional

Make ZONE_HIGHMEM optional

- ifdef out code and definitions related to CONFIG_HIGHMEM

- __GFP_HIGHMEM falls back to normal allocations if there is no
  ZONE_HIGHMEM

- GFP_ZONEMASK becomes 0x01 if there is no DMA32 and no HIGHMEM
  zone.

[jdike@addtoit.com: build fix]
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/mem.c   |  2 ++
 include/linux/gfp.h    | 18 ++++++++++--------
 include/linux/mmzone.h | 36 +++++++++++++++++++++++++++++++++---
 mm/page_alloc.c        |  6 ++++++
 4 files changed, 51 insertions(+), 11 deletions(-)

diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index b39e624c3291..b1cd5c6e468b 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -226,7 +226,9 @@ void paging_init(void)
 	for(i=0;i<sizeof(zones_size)/sizeof(zones_size[0]);i++) 
 		zones_size[i] = 0;
 	zones_size[ZONE_DMA] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT);
+#ifdef CONFIG_HIGHMEM
 	zones_size[ZONE_HIGHMEM] = highmem >> PAGE_SHIFT;
+#endif
 	free_area_init(zones_size);
 
 	/*
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 14610b56c132..2a2153ebfe0b 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -9,17 +9,19 @@ struct vm_area_struct;
 
 /*
  * GFP bitmasks..
+ *
+ * Zone modifiers (see linux/mmzone.h - low three bits)
+ *
+ * These may be masked by GFP_ZONEMASK to make allocations with this bit
+ * set fall back to ZONE_NORMAL.
+ *
+ * Do not put any conditional on these. If necessary modify the definitions
+ * without the underscores and use the consistently. The definitions here may
+ * be used in bit comparisons.
  */
-/* Zone modifiers in GFP_ZONEMASK (see linux/mmzone.h - low three bits) */
 #define __GFP_DMA	((__force gfp_t)0x01u)
 #define __GFP_HIGHMEM	((__force gfp_t)0x02u)
-#ifndef CONFIG_ZONE_DMA32
-#define __GFP_DMA32	((__force gfp_t)0x01)	/* ZONE_DMA is ZONE_DMA32 */
-#elif BITS_PER_LONG < 64
-#define __GFP_DMA32	((__force gfp_t)0x00)	/* ZONE_NORMAL is ZONE_DMA32 */
-#else
-#define __GFP_DMA32	((__force gfp_t)0x04)	/* Has own ZONE_DMA32 */
-#endif
+#define __GFP_DMA32	((__force gfp_t)0x04u)
 
 /*
  * Action modifiers - doesn't change the zoning
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index adae3c915938..76d33e688593 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -123,6 +123,7 @@ enum zone_type {
 	 * transfers to all addressable memory.
 	 */
 	ZONE_NORMAL,
+#ifdef CONFIG_HIGHMEM
 	/*
 	 * A memory area that is only addressable by the kernel through
 	 * mapping portions into its own address space. This is for example
@@ -132,11 +133,10 @@ enum zone_type {
 	 * access.
 	 */
 	ZONE_HIGHMEM,
-
+#endif
 	MAX_NR_ZONES
 };
 
-#define	ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 
 /*
  * When a memory allocation must conform to specific limitations (such
@@ -163,12 +163,34 @@ enum zone_type {
  *
  * NOTE! Make sure this matches the zones in <linux/gfp.h>
  */
-#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
 
 #ifdef CONFIG_ZONE_DMA32
+
+#ifdef CONFIG_HIGHMEM
+#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
 #define GFP_ZONEMASK		0x07
+#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
 #else
+#define GFP_ZONETYPES		((0x07 + 1) / 2 + 1)    /* Loner */
+/* Mask __GFP_HIGHMEM */
+#define GFP_ZONEMASK		0x05
+#define ZONES_SHIFT		2
+#endif
+
+#else
+#ifdef CONFIG_HIGHMEM
+
 #define GFP_ZONEMASK		0x03
+#define ZONES_SHIFT		2
+#define GFP_ZONETYPES		3
+
+#else
+
+#define GFP_ZONEMASK		0x01
+#define ZONES_SHIFT		1
+#define GFP_ZONETYPES		2
+
+#endif
 #endif
 
 struct zone {
@@ -409,7 +431,11 @@ static inline int populated_zone(struct zone *zone)
 
 static inline int is_highmem_idx(enum zone_type idx)
 {
+#ifdef CONFIG_HIGHMEM
 	return (idx == ZONE_HIGHMEM);
+#else
+	return 0;
+#endif
 }
 
 static inline int is_normal_idx(enum zone_type idx)
@@ -425,7 +451,11 @@ static inline int is_normal_idx(enum zone_type idx)
  */
 static inline int is_highmem(struct zone *zone)
 {
+#ifdef CONFIG_HIGHMEM
 	return zone == zone->zone_pgdat->node_zones + ZONE_HIGHMEM;
+#else
+	return 0;
+#endif
 }
 
 static inline int is_normal(struct zone *zone)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5b5cbb5e1816..6c7c2dd1b3ed 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -73,7 +73,9 @@ int sysctl_lowmem_reserve_ratio[MAX_NR_ZONES-1] = {
 #ifdef CONFIG_ZONE_DMA32
 	 256,
 #endif
+#ifdef CONFIG_HIGHMEM
 	 32
+#endif
 };
 
 EXPORT_SYMBOL(totalram_pages);
@@ -91,7 +93,9 @@ static char *zone_names[MAX_NR_ZONES] = {
 	 "DMA32",
 #endif
 	 "Normal",
+#ifdef CONFIG_HIGHMEM
 	 "HighMem"
+#endif
 };
 
 int min_free_kbytes = 1024;
@@ -1375,8 +1379,10 @@ static int __meminit build_zonelists_node(pg_data_t *pgdat,
 static inline int highest_zone(int zone_bits)
 {
 	int res = ZONE_NORMAL;
+#ifdef CONFIG_HIGHMEM
 	if (zone_bits & (__force int)__GFP_HIGHMEM)
 		res = ZONE_HIGHMEM;
+#endif
 #ifdef CONFIG_ZONE_DMA32
 	if (zone_bits & (__force int)__GFP_DMA32)
 		res = ZONE_DMA32;
-- 
cgit v1.2.1


From 27bf71c2a7e596ed34e9bf2d4a5030321a09a1ad Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:15 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: remove display of counters for
 unconfigured zones

eventcounters: Do not display counters for zones that are not available on an
arch

Do not define or display counters for the DMA32 and the HIGHMEM zone if such
zones were not configured.

[akpm@osdl.org: s390 fix]
[heiko.carstens@de.ibm.com: s390 fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/appldata/appldata_mem.c |  3 +--
 include/linux/vmstat.h            | 14 ++++++++++++-
 mm/vmstat.c                       | 43 ++++++++++++++++++---------------------
 3 files changed, 34 insertions(+), 26 deletions(-)

diff --git a/arch/s390/appldata/appldata_mem.c b/arch/s390/appldata/appldata_mem.c
index ab3b0765a64e..8aea3698a77b 100644
--- a/arch/s390/appldata/appldata_mem.c
+++ b/arch/s390/appldata/appldata_mem.c
@@ -117,8 +117,7 @@ static void appldata_get_mem_data(void *data)
 	mem_data->pgpgout    = ev[PGPGOUT] >> 1;
 	mem_data->pswpin     = ev[PSWPIN];
 	mem_data->pswpout    = ev[PSWPOUT];
-	mem_data->pgalloc    = ev[PGALLOC_HIGH] + ev[PGALLOC_NORMAL] +
-			       ev[PGALLOC_DMA];
+	mem_data->pgalloc    = ev[PGALLOC_NORMAL] + ev[PGALLOC_DMA];
 	mem_data->pgfault    = ev[PGFAULT];
 	mem_data->pgmajfault = ev[PGMAJFAULT];
 
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 9c6e62c56ec2..176c7f797339 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -18,7 +18,19 @@
  * generated will simply be the increment of a global address.
  */
 
-#define FOR_ALL_ZONES(x) x##_DMA, x##_DMA32, x##_NORMAL, x##_HIGH
+#ifdef CONFIG_ZONE_DMA32
+#define DMA32_ZONE(xx) xx##_DMA32,
+#else
+#define DMA32_ZONE(xx)
+#endif
+
+#ifdef CONFIG_HIGHMEM
+#define HIGHMEM_ZONE(xx) , xx##_HIGH
+#else
+#define HIGHMEM_ZONE(xx)
+#endif
+
+#define FOR_ALL_ZONES(xx) xx##_DMA, DMA32_ZONE(xx) xx##_NORMAL HIGHMEM_ZONE(xx)
 
 enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
 		FOR_ALL_ZONES(PGALLOC),
diff --git a/mm/vmstat.c b/mm/vmstat.c
index c1b5f4106b38..04a9093f649e 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -435,6 +435,21 @@ struct seq_operations fragmentation_op = {
 	.show	= frag_show,
 };
 
+#ifdef CONFIG_ZONE_DMA32
+#define TEXT_FOR_DMA32(xx) xx "_dma32",
+#else
+#define TEXT_FOR_DMA32(xx)
+#endif
+
+#ifdef CONFIG_HIGHMEM
+#define TEXT_FOR_HIGHMEM(xx) xx "_high",
+#else
+#define TEXT_FOR_HIGHMEM(xx)
+#endif
+
+#define TEXTS_FOR_ZONES(xx) xx "_dma", TEXT_FOR_DMA32(xx) xx "_normal", \
+					TEXT_FOR_HIGHMEM(xx)
+
 static char *vmstat_text[] = {
 	/* Zoned VM counters */
 	"nr_anon_pages",
@@ -462,10 +477,7 @@ static char *vmstat_text[] = {
 	"pswpin",
 	"pswpout",
 
-	"pgalloc_dma",
-	"pgalloc_dma32",
-	"pgalloc_normal",
-	"pgalloc_high",
+	TEXTS_FOR_ZONES("pgalloc")
 
 	"pgfree",
 	"pgactivate",
@@ -474,25 +486,10 @@ static char *vmstat_text[] = {
 	"pgfault",
 	"pgmajfault",
 
-	"pgrefill_dma",
-	"pgrefill_dma32",
-	"pgrefill_normal",
-	"pgrefill_high",
-
-	"pgsteal_dma",
-	"pgsteal_dma32",
-	"pgsteal_normal",
-	"pgsteal_high",
-
-	"pgscan_kswapd_dma",
-	"pgscan_kswapd_dma32",
-	"pgscan_kswapd_normal",
-	"pgscan_kswapd_high",
-
-	"pgscan_direct_dma",
-	"pgscan_direct_dma32",
-	"pgscan_direct_normal",
-	"pgscan_direct_high",
+	TEXTS_FOR_ZONES("pgrefill")
+	TEXTS_FOR_ZONES("pgsteal")
+	TEXTS_FOR_ZONES("pgscan_kswapd")
+	TEXTS_FOR_ZONES("pgscan_direct")
 
 	"pginodesteal",
 	"slabs_scanned",
-- 
cgit v1.2.1


From b9b15780f808efa2c897f337644ba7a2bec03ecc Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:16 -0700
Subject: [PATCH] reduce MAX_NR_ZONES: fix i386 SRAT check for MAX_NR_ZONES

We cannot check MAX_NR_ZONES since it not defined in the preprocessor
anymore.

So remove the check.

The maximum number of zones per node for i386 is 3 since i386 does not
support ZONE_DMA32.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/srat.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/arch/i386/kernel/srat.c b/arch/i386/kernel/srat.c
index b1809c9a0899..83db411b3aa7 100644
--- a/arch/i386/kernel/srat.c
+++ b/arch/i386/kernel/srat.c
@@ -42,7 +42,7 @@
 #define PXM_BITMAP_LEN (MAX_PXM_DOMAINS / 8) 
 static u8 pxm_bitmap[PXM_BITMAP_LEN];	/* bitmap of proximity domains */
 
-#define MAX_CHUNKS_PER_NODE	4
+#define MAX_CHUNKS_PER_NODE	3
 #define MAXCHUNKS		(MAX_CHUNKS_PER_NODE * MAX_NUMNODES)
 struct node_memory_chunk_s {
 	unsigned long	start_pfn;
@@ -135,9 +135,6 @@ static void __init parse_memory_affinity_structure (char *sratp)
 		 "enabled and removable" : "enabled" ) );
 }
 
-#if MAX_NR_ZONES != 4
-#error "MAX_NR_ZONES != 4, chunk_to_zone requires review"
-#endif
 /* Take a chunk of pages from page frame cstart to cend and count the number
  * of pages in each zone, returned via zones[].
  */
-- 
cgit v1.2.1


From 4e4785bcf0c8503224fa6c17d8e0228de781bff6 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:17 -0700
Subject: [PATCH] mempolicies: fix policy_zone check

There is a check in zonelist_policy that compares pieces of the bitmap
obtained from a gfp mask via GFP_ZONETYPES with a zone number in function
zonelist_policy().

The bitmap is an ORed mask of __GFP_DMA, __GFP_DMA32 and __GFP_HIGHMEM.
The policy_zone is a zone number with the possible values of ZONE_DMA,
ZONE_DMA32, ZONE_HIGHMEM and ZONE_NORMAL. These are two different domains
of values.

For some reason seemed to work before the zone reduction patchset (It
definitely works on SGI boxes since we just have one zone and the check
cannot fail).

With the zone reduction patchset this check definitely fails on systems
with two zones if the system actually has memory in both zones.

This is because ZONE_NORMAL is selected using no __GFP flag at
all and thus gfp_zone(gfpmask) == 0. ZONE_DMA is selected when __GFP_DMA
is set. __GFP_DMA is 0x01.  So gfp_zone(gfpmask) == 1.

policy_zone is set to ZONE_NORMAL (==1) if ZONE_NORMAL and ZONE_DMA are
populated.

For ZONE_NORMAL gfp_zone(<no _GFP_DMA>) yields 0 which is <
policy_zone(ZONE_NORMAL) and so policy is not applied to regular memory
allocations!

Instead gfp_zone(__GFP_DMA) == 1 which results in policy being applied
to DMA allocations!

What we realy want in that place is to establish the highest allowable
zone for a given gfp_mask. If the highest zone is higher or equal to the
policy_zone then memory policies need to be applied. We have such
a highest_zone() function in page_alloc.c.

So move the highest_zone() function from mm/page_alloc.c into
include/linux/gfp.h.  On the way we simplify the function and use the new
zone_type that was also introduced with the zone reduction patchset plus we
also specify the right type for the gfp flags parameter.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 15 +++++++++++++++
 mm/mempolicy.c      |  2 +-
 mm/page_alloc.c     | 16 ----------------
 3 files changed, 16 insertions(+), 17 deletions(-)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 2a2153ebfe0b..a0992d392d79 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -85,6 +85,21 @@ static inline int gfp_zone(gfp_t gfp)
 	return zone;
 }
 
+static inline enum zone_type highest_zone(gfp_t flags)
+{
+	if (flags & __GFP_DMA)
+		return ZONE_DMA;
+#ifdef CONFIG_ZONE_DMA32
+	if (flags & __GFP_DMA32)
+		return ZONE_DMA32;
+#endif
+#ifdef CONFIG_HIGHMEM
+	if (flags & __GFP_HIGHMEM)
+		return ZONE_HIGHMEM;
+#endif
+	return ZONE_NORMAL;
+}
+
 /*
  * There is only one page-allocator function, and two main namespaces to
  * it. The alloc_page*() variants return 'struct page *' and as such
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index a9963ceddd65..9870624d72a6 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1096,7 +1096,7 @@ static struct zonelist *zonelist_policy(gfp_t gfp, struct mempolicy *policy)
 	case MPOL_BIND:
 		/* Lower zones don't get a policy applied */
 		/* Careful: current->mems_allowed might have moved */
-		if (gfp_zone(gfp) >= policy_zone)
+		if (highest_zone(gfp) >= policy_zone)
 			if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist))
 				return policy->v.zonelist;
 		/*FALL THROUGH*/
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 6c7c2dd1b3ed..25f39865e324 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1376,22 +1376,6 @@ static int __meminit build_zonelists_node(pg_data_t *pgdat,
 	return nr_zones;
 }
 
-static inline int highest_zone(int zone_bits)
-{
-	int res = ZONE_NORMAL;
-#ifdef CONFIG_HIGHMEM
-	if (zone_bits & (__force int)__GFP_HIGHMEM)
-		res = ZONE_HIGHMEM;
-#endif
-#ifdef CONFIG_ZONE_DMA32
-	if (zone_bits & (__force int)__GFP_DMA32)
-		res = ZONE_DMA32;
-#endif
-	if (zone_bits & (__force int)__GFP_DMA)
-		res = ZONE_DMA;
-	return res;
-}
-
 #ifdef CONFIG_NUMA
 #define MAX_NODE_LOAD (num_online_nodes())
 static int __meminitdata node_load[MAX_NUMNODES];
-- 
cgit v1.2.1


From 2f6726e54a9410e2e4cee864947c05e954051916 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:18 -0700
Subject: [PATCH] Apply type enum zone_type

After we have done this we can now do some typing cleanup.

The memory policy layer keeps a policy_zone that specifies
the zone that gets memory policies applied. This variable
can now be of type enum zone_type.

The check_highest_zone function and the build_zonelists funnctionm must
then also take a enum zone_type parameter.

Plus there are a number of loops over zones that also should use
zone_type.

We run into some troubles at some points with functions that need a
zone_type variable to become -1. Fix that up.

[pj@sgi.com: fix set_mempolicy() crash]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mempolicy.h |  4 ++--
 mm/mempolicy.c            | 11 ++++++++---
 mm/page_alloc.c           | 27 +++++++++++++++++----------
 3 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 72440f0a443d..09f0f575ddff 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -162,9 +162,9 @@ extern struct zonelist *huge_zonelist(struct vm_area_struct *vma,
 		unsigned long addr);
 extern unsigned slab_node(struct mempolicy *policy);
 
-extern int policy_zone;
+extern enum zone_type policy_zone;
 
-static inline void check_highest_zone(int k)
+static inline void check_highest_zone(enum zone_type k)
 {
 	if (k > policy_zone)
 		policy_zone = k;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 9870624d72a6..7da4142ce960 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -105,7 +105,7 @@ static struct kmem_cache *sn_cache;
 
 /* Highest zone. An specific allocation for a zone below that is not
    policied. */
-int policy_zone = ZONE_DMA;
+enum zone_type policy_zone = ZONE_DMA;
 
 struct mempolicy default_policy = {
 	.refcnt = ATOMIC_INIT(1), /* never free it */
@@ -137,7 +137,8 @@ static int mpol_check_policy(int mode, nodemask_t *nodes)
 static struct zonelist *bind_zonelist(nodemask_t *nodes)
 {
 	struct zonelist *zl;
-	int num, max, nd, k;
+	int num, max, nd;
+	enum zone_type k;
 
 	max = 1 + MAX_NR_ZONES * nodes_weight(*nodes);
 	zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL);
@@ -148,12 +149,16 @@ static struct zonelist *bind_zonelist(nodemask_t *nodes)
 	   lower zones etc. Avoid empty zones because the memory allocator
 	   doesn't like them. If you implement node hot removal you
 	   have to fix that. */
-	for (k = policy_zone; k >= 0; k--) { 
+	k = policy_zone;
+	while (1) {
 		for_each_node_mask(nd, *nodes) { 
 			struct zone *z = &NODE_DATA(nd)->node_zones[k];
 			if (z->present_pages > 0) 
 				zl->zones[num++] = z;
 		}
+		if (k == 0)
+			break;
+		k--;
 	}
 	zl->zones[num] = NULL;
 	return zl;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 25f39865e324..9c44b9a39d30 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -637,7 +637,8 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
  */
 void drain_node_pages(int nodeid)
 {
-	int i, z;
+	int i;
+	enum zone_type z;
 	unsigned long flags;
 
 	for (z = 0; z < MAX_NR_ZONES; z++) {
@@ -1158,7 +1159,8 @@ EXPORT_SYMBOL(nr_free_pages);
 #ifdef CONFIG_NUMA
 unsigned int nr_free_pages_pgdat(pg_data_t *pgdat)
 {
-	unsigned int i, sum = 0;
+	unsigned int sum = 0;
+	enum zone_type i;
 
 	for (i = 0; i < MAX_NR_ZONES; i++)
 		sum += pgdat->node_zones[i].free_pages;
@@ -1358,21 +1360,22 @@ void show_free_areas(void)
  * Add all populated zones of a node to the zonelist.
  */
 static int __meminit build_zonelists_node(pg_data_t *pgdat,
-			struct zonelist *zonelist, int nr_zones, int zone_type)
+			struct zonelist *zonelist, int nr_zones, enum zone_type zone_type)
 {
 	struct zone *zone;
 
 	BUG_ON(zone_type >= MAX_NR_ZONES);
+	zone_type++;
 
 	do {
+		zone_type--;
 		zone = pgdat->node_zones + zone_type;
 		if (populated_zone(zone)) {
 			zonelist->zones[nr_zones++] = zone;
 			check_highest_zone(zone_type);
 		}
-		zone_type--;
 
-	} while (zone_type >= 0);
+	} while (zone_type);
 	return nr_zones;
 }
 
@@ -1441,10 +1444,11 @@ static int __meminit find_next_best_node(int node, nodemask_t *used_node_mask)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, k, node, local_node;
+	int i, j, node, local_node;
 	int prev_node, load;
 	struct zonelist *zonelist;
 	nodemask_t used_mask;
+	enum zone_type k;
 
 	/* initialize zonelists */
 	for (i = 0; i < GFP_ZONETYPES; i++) {
@@ -1628,7 +1632,7 @@ static void __init calculate_zone_totalpages(struct pglist_data *pgdat,
 		unsigned long *zones_size, unsigned long *zholes_size)
 {
 	unsigned long realtotalpages, totalpages = 0;
-	int i;
+	enum zone_type i;
 
 	for (i = 0; i < MAX_NR_ZONES; i++)
 		totalpages += zones_size[i];
@@ -2116,7 +2120,7 @@ static void calculate_totalreserve_pages(void)
 {
 	struct pglist_data *pgdat;
 	unsigned long reserve_pages = 0;
-	int i, j;
+	enum zone_type i, j;
 
 	for_each_online_pgdat(pgdat) {
 		for (i = 0; i < MAX_NR_ZONES; i++) {
@@ -2149,7 +2153,7 @@ static void calculate_totalreserve_pages(void)
 static void setup_per_zone_lowmem_reserve(void)
 {
 	struct pglist_data *pgdat;
-	int j, idx;
+	enum zone_type j, idx;
 
 	for_each_online_pgdat(pgdat) {
 		for (j = 0; j < MAX_NR_ZONES; j++) {
@@ -2158,9 +2162,12 @@ static void setup_per_zone_lowmem_reserve(void)
 
 			zone->lowmem_reserve[j] = 0;
 
-			for (idx = j-1; idx >= 0; idx--) {
+			idx = j;
+			while (idx) {
 				struct zone *lower_zone;
 
+				idx--;
+
 				if (sysctl_lowmem_reserve_ratio[idx] < 1)
 					sysctl_lowmem_reserve_ratio[idx] = 1;
 
-- 
cgit v1.2.1


From 19655d3487001d7df0e10e9cbfc27c758b77c2b5 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:19 -0700
Subject: [PATCH] linearly index zone->node_zonelists[]

I wonder why we need this bitmask indexing into zone->node_zonelists[]?

We always start with the highest zone and then include all lower zones
if we build zonelists.

Are there really cases where we need allocation from ZONE_DMA or
ZONE_HIGHMEM but not ZONE_NORMAL? It seems that the current implementation
of highest_zone() makes that already impossible.

If we go linear on the index then gfp_zone() == highest_zone() and a lot
of definitions fall by the wayside.

We can now revert back to the use of gfp_zone() in mempolicy.c ;-)

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h    | 12 +-----------
 include/linux/mmzone.h | 52 +++++---------------------------------------------
 mm/mempolicy.c         |  2 +-
 mm/page_alloc.c        | 27 +++++++++++---------------
 4 files changed, 18 insertions(+), 75 deletions(-)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index a0992d392d79..63ab88a36f39 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -12,9 +12,6 @@ struct vm_area_struct;
  *
  * Zone modifiers (see linux/mmzone.h - low three bits)
  *
- * These may be masked by GFP_ZONEMASK to make allocations with this bit
- * set fall back to ZONE_NORMAL.
- *
  * Do not put any conditional on these. If necessary modify the definitions
  * without the underscores and use the consistently. The definitions here may
  * be used in bit comparisons.
@@ -78,14 +75,7 @@ struct vm_area_struct;
 #define GFP_DMA32	__GFP_DMA32
 
 
-static inline int gfp_zone(gfp_t gfp)
-{
-	int zone = GFP_ZONEMASK & (__force int) gfp;
-	BUG_ON(zone >= GFP_ZONETYPES);
-	return zone;
-}
-
-static inline enum zone_type highest_zone(gfp_t flags)
+static inline enum zone_type gfp_zone(gfp_t flags)
 {
 	if (flags & __GFP_DMA)
 		return ZONE_DMA;
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 76d33e688593..7fe317164b73 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -137,60 +137,18 @@ enum zone_type {
 	MAX_NR_ZONES
 };
 
-
 /*
  * When a memory allocation must conform to specific limitations (such
  * as being suitable for DMA) the caller will pass in hints to the
  * allocator in the gfp_mask, in the zone modifier bits.  These bits
  * are used to select a priority ordered list of memory zones which
- * match the requested limits.  GFP_ZONEMASK defines which bits within
- * the gfp_mask should be considered as zone modifiers.  Each valid
- * combination of the zone modifier bits has a corresponding list
- * of zones (in node_zonelists).  Thus for two zone modifiers there
- * will be a maximum of 4 (2 ** 2) zonelists, for 3 modifiers there will
- * be 8 (2 ** 3) zonelists.  GFP_ZONETYPES defines the number of possible
- * combinations of zone modifiers in "zone modifier space".
- *
- * As an optimisation any zone modifier bits which are only valid when
- * no other zone modifier bits are set (loners) should be placed in
- * the highest order bits of this field.  This allows us to reduce the
- * extent of the zonelists thus saving space.  For example in the case
- * of three zone modifier bits, we could require up to eight zonelists.
- * If the left most zone modifier is a "loner" then the highest valid
- * zonelist would be four allowing us to allocate only five zonelists.
- * Use the first form for GFP_ZONETYPES when the left most bit is not
- * a "loner", otherwise use the second.
- *
- * NOTE! Make sure this matches the zones in <linux/gfp.h>
+ * match the requested limits. See gfp_zone() in include/linux/gfp.h
  */
 
-#ifdef CONFIG_ZONE_DMA32
-
-#ifdef CONFIG_HIGHMEM
-#define GFP_ZONETYPES		((GFP_ZONEMASK + 1) / 2 + 1)    /* Loner */
-#define GFP_ZONEMASK		0x07
-#define ZONES_SHIFT		2	/* ceil(log2(MAX_NR_ZONES)) */
-#else
-#define GFP_ZONETYPES		((0x07 + 1) / 2 + 1)    /* Loner */
-/* Mask __GFP_HIGHMEM */
-#define GFP_ZONEMASK		0x05
-#define ZONES_SHIFT		2
-#endif
-
-#else
-#ifdef CONFIG_HIGHMEM
-
-#define GFP_ZONEMASK		0x03
-#define ZONES_SHIFT		2
-#define GFP_ZONETYPES		3
-
+#if !defined(CONFIG_ZONE_DMA32) && !defined(CONFIG_HIGHMEM)
+#define ZONES_SHIFT 1
 #else
-
-#define GFP_ZONEMASK		0x01
-#define ZONES_SHIFT		1
-#define GFP_ZONETYPES		2
-
-#endif
+#define ZONES_SHIFT 2
 #endif
 
 struct zone {
@@ -360,7 +318,7 @@ struct zonelist {
 struct bootmem_data;
 typedef struct pglist_data {
 	struct zone node_zones[MAX_NR_ZONES];
-	struct zonelist node_zonelists[GFP_ZONETYPES];
+	struct zonelist node_zonelists[MAX_NR_ZONES];
 	int nr_zones;
 #ifdef CONFIG_FLAT_NODE_MEM_MAP
 	struct page *node_mem_map;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 7da4142ce960..c3429a710ab1 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1101,7 +1101,7 @@ static struct zonelist *zonelist_policy(gfp_t gfp, struct mempolicy *policy)
 	case MPOL_BIND:
 		/* Lower zones don't get a policy applied */
 		/* Careful: current->mems_allowed might have moved */
-		if (highest_zone(gfp) >= policy_zone)
+		if (gfp_zone(gfp) >= policy_zone)
 			if (cpuset_zonelist_valid_mems_allowed(policy->v.zonelist))
 				return policy->v.zonelist;
 		/*FALL THROUGH*/
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9c44b9a39d30..208a6b03aa78 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1444,14 +1444,14 @@ static int __meminit find_next_best_node(int node, nodemask_t *used_node_mask)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, j, node, local_node;
+	int j, node, local_node;
+	enum zone_type i;
 	int prev_node, load;
 	struct zonelist *zonelist;
 	nodemask_t used_mask;
-	enum zone_type k;
 
 	/* initialize zonelists */
-	for (i = 0; i < GFP_ZONETYPES; i++) {
+	for (i = 0; i < MAX_NR_ZONES; i++) {
 		zonelist = pgdat->node_zonelists + i;
 		zonelist->zones[0] = NULL;
 	}
@@ -1481,13 +1481,11 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 			node_load[node] += load;
 		prev_node = node;
 		load--;
-		for (i = 0; i < GFP_ZONETYPES; i++) {
+		for (i = 0; i < MAX_NR_ZONES; i++) {
 			zonelist = pgdat->node_zonelists + i;
 			for (j = 0; zonelist->zones[j] != NULL; j++);
 
-			k = highest_zone(i);
-
-	 		j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+	 		j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 			zonelist->zones[j] = NULL;
 		}
 	}
@@ -1497,19 +1495,16 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 
 static void __meminit build_zonelists(pg_data_t *pgdat)
 {
-	int i, node, local_node;
-	enum zone_type k;
-	enum zone_type j;
+	int node, local_node;
+	enum zone_type i,j;
 
 	local_node = pgdat->node_id;
-	for (i = 0; i < GFP_ZONETYPES; i++) {
+	for (i = 0; i < MAX_NR_ZONES; i++) {
 		struct zonelist *zonelist;
 
 		zonelist = pgdat->node_zonelists + i;
 
-		j = 0;
-		k = highest_zone(i);
- 		j = build_zonelists_node(pgdat, zonelist, j, k);
+ 		j = build_zonelists_node(pgdat, zonelist, 0, i);
  		/*
  		 * Now we build the zonelist so that it contains the zones
  		 * of all the other nodes.
@@ -1521,12 +1516,12 @@ static void __meminit build_zonelists(pg_data_t *pgdat)
 		for (node = local_node + 1; node < MAX_NUMNODES; node++) {
 			if (!node_online(node))
 				continue;
-			j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+			j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 		}
 		for (node = 0; node < local_node; node++) {
 			if (!node_online(node))
 				continue;
-			j = build_zonelists_node(NODE_DATA(node), zonelist, j, k);
+			j = build_zonelists_node(NODE_DATA(node), zonelist, j, i);
 		}
 
 		zonelist->zones[j] = NULL;
-- 
cgit v1.2.1


From 8bc719d3cab8414938f9ea6e33b58d8810d18068 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:20 -0700
Subject: [PATCH] out of memory notifier

Add a notifer chain to the out of memory killer.  If one of the registered
callbacks could release some memory, do not kill the process but return and
retry the allocation that forced the oom killer to run.

The purpose of the notifier is to add a safety net in the presence of
memory ballooners.  If the resource manager inflated the balloon to a size
where memory allocations can not be satisfied anymore, it is better to
deflate the balloon a bit instead of killing processes.

The implementation for the s390 ballooner is included.

[akpm@osdl.org: cleanups]
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/mm/cmm.c   | 155 +++++++++++++++++++++++++++++++--------------------
 include/linux/swap.h |   4 ++
 mm/oom_kill.c        |  22 ++++++++
 3 files changed, 121 insertions(+), 60 deletions(-)

diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c
index 786a44dba5bf..f2e00df99bae 100644
--- a/arch/s390/mm/cmm.c
+++ b/arch/s390/mm/cmm.c
@@ -15,6 +15,7 @@
 #include <linux/sched.h>
 #include <linux/sysctl.h>
 #include <linux/ctype.h>
+#include <linux/swap.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -34,17 +35,18 @@ struct cmm_page_array {
 	unsigned long pages[CMM_NR_PAGES];
 };
 
-static long cmm_pages = 0;
-static long cmm_timed_pages = 0;
-static volatile long cmm_pages_target = 0;
-static volatile long cmm_timed_pages_target = 0;
-static long cmm_timeout_pages = 0;
-static long cmm_timeout_seconds = 0;
+static long cmm_pages;
+static long cmm_timed_pages;
+static volatile long cmm_pages_target;
+static volatile long cmm_timed_pages_target;
+static long cmm_timeout_pages;
+static long cmm_timeout_seconds;
 
-static struct cmm_page_array *cmm_page_list = NULL;
-static struct cmm_page_array *cmm_timed_page_list = NULL;
+static struct cmm_page_array *cmm_page_list;
+static struct cmm_page_array *cmm_timed_page_list;
+static DEFINE_SPINLOCK(cmm_lock);
 
-static unsigned long cmm_thread_active = 0;
+static unsigned long cmm_thread_active;
 static struct work_struct cmm_thread_starter;
 static wait_queue_head_t cmm_thread_wait;
 static struct timer_list cmm_timer;
@@ -53,58 +55,89 @@ static void cmm_timer_fn(unsigned long);
 static void cmm_set_timer(void);
 
 static long
-cmm_alloc_pages(long pages, long *counter, struct cmm_page_array **list)
+cmm_alloc_pages(long nr, long *counter, struct cmm_page_array **list)
 {
-	struct cmm_page_array *pa;
-	unsigned long page;
+	struct cmm_page_array *pa, *npa;
+	unsigned long addr;
 
-	pa = *list;
-	while (pages) {
-		page = __get_free_page(GFP_NOIO);
-		if (!page)
+	while (nr) {
+		addr = __get_free_page(GFP_NOIO);
+		if (!addr)
 			break;
+		spin_lock(&cmm_lock);
+		pa = *list;
 		if (!pa || pa->index >= CMM_NR_PAGES) {
 			/* Need a new page for the page list. */
-			pa = (struct cmm_page_array *)
+			spin_unlock(&cmm_lock);
+			npa = (struct cmm_page_array *)
 				__get_free_page(GFP_NOIO);
-			if (!pa) {
-				free_page(page);
+			if (!npa) {
+				free_page(addr);
 				break;
 			}
-			pa->next = *list;
-			pa->index = 0;
-			*list = pa;
+			spin_lock(&cmm_lock);
+			pa = *list;
+			if (!pa || pa->index >= CMM_NR_PAGES) {
+				npa->next = pa;
+				npa->index = 0;
+				pa = npa;
+				*list = pa;
+			} else
+				free_page((unsigned long) npa);
 		}
-		diag10(page);
-		pa->pages[pa->index++] = page;
+		diag10(addr);
+		pa->pages[pa->index++] = addr;
 		(*counter)++;
-		pages--;
+		spin_unlock(&cmm_lock);
+		nr--;
 	}
-	return pages;
+	return nr;
 }
 
-static void
-cmm_free_pages(long pages, long *counter, struct cmm_page_array **list)
+static long
+cmm_free_pages(long nr, long *counter, struct cmm_page_array **list)
 {
 	struct cmm_page_array *pa;
-	unsigned long page;
+	unsigned long addr;
 
+	spin_lock(&cmm_lock);
 	pa = *list;
-	while (pages) {
+	while (nr) {
 		if (!pa || pa->index <= 0)
 			break;
-		page = pa->pages[--pa->index];
+		addr = pa->pages[--pa->index];
 		if (pa->index == 0) {
 			pa = pa->next;
 			free_page((unsigned long) *list);
 			*list = pa;
 		}
-		free_page(page);
+		free_page(addr);
 		(*counter)--;
-		pages--;
+		nr--;
 	}
+	spin_unlock(&cmm_lock);
+	return nr;
 }
 
+static int cmm_oom_notify(struct notifier_block *self,
+			  unsigned long dummy, void *parm)
+{
+	unsigned long *freed = parm;
+	long nr = 256;
+
+	nr = cmm_free_pages(nr, &cmm_timed_pages, &cmm_timed_page_list);
+	if (nr > 0)
+		nr = cmm_free_pages(nr, &cmm_pages, &cmm_page_list);
+	cmm_pages_target = cmm_pages;
+	cmm_timed_pages_target = cmm_timed_pages;
+	*freed += 256 - nr;
+	return NOTIFY_OK;
+}
+
+static struct notifier_block cmm_oom_nb = {
+	.notifier_call = cmm_oom_notify
+};
+
 static int
 cmm_thread(void *dummy)
 {
@@ -177,21 +210,21 @@ cmm_set_timer(void)
 static void
 cmm_timer_fn(unsigned long ignored)
 {
-	long pages;
+	long nr;
 
-	pages = cmm_timed_pages_target - cmm_timeout_pages;
-	if (pages < 0)
+	nr = cmm_timed_pages_target - cmm_timeout_pages;
+	if (nr < 0)
 		cmm_timed_pages_target = 0;
 	else
-		cmm_timed_pages_target = pages;
+		cmm_timed_pages_target = nr;
 	cmm_kick_thread();
 	cmm_set_timer();
 }
 
 void
-cmm_set_pages(long pages)
+cmm_set_pages(long nr)
 {
-	cmm_pages_target = pages;
+	cmm_pages_target = nr;
 	cmm_kick_thread();
 }
 
@@ -202,9 +235,9 @@ cmm_get_pages(void)
 }
 
 void
-cmm_add_timed_pages(long pages)
+cmm_add_timed_pages(long nr)
 {
-	cmm_timed_pages_target += pages;
+	cmm_timed_pages_target += nr;
 	cmm_kick_thread();
 }
 
@@ -215,9 +248,9 @@ cmm_get_timed_pages(void)
 }
 
 void
-cmm_set_timeout(long pages, long seconds)
+cmm_set_timeout(long nr, long seconds)
 {
-	cmm_timeout_pages = pages;
+	cmm_timeout_pages = nr;
 	cmm_timeout_seconds = seconds;
 	cmm_set_timer();
 }
@@ -245,7 +278,7 @@ cmm_pages_handler(ctl_table *ctl, int write, struct file *filp,
 		  void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	char buf[16], *p;
-	long pages;
+	long nr;
 	int len;
 
 	if (!*lenp || (*ppos && !write)) {
@@ -260,17 +293,17 @@ cmm_pages_handler(ctl_table *ctl, int write, struct file *filp,
 			return -EFAULT;
 		buf[sizeof(buf) - 1] = '\0';
 		cmm_skip_blanks(buf, &p);
-		pages = simple_strtoul(p, &p, 0);
+		nr = simple_strtoul(p, &p, 0);
 		if (ctl == &cmm_table[0])
-			cmm_set_pages(pages);
+			cmm_set_pages(nr);
 		else
-			cmm_add_timed_pages(pages);
+			cmm_add_timed_pages(nr);
 	} else {
 		if (ctl == &cmm_table[0])
-			pages = cmm_get_pages();
+			nr = cmm_get_pages();
 		else
-			pages = cmm_get_timed_pages();
-		len = sprintf(buf, "%ld\n", pages);
+			nr = cmm_get_timed_pages();
+		len = sprintf(buf, "%ld\n", nr);
 		if (len > *lenp)
 			len = *lenp;
 		if (copy_to_user(buffer, buf, len))
@@ -286,7 +319,7 @@ cmm_timeout_handler(ctl_table *ctl, int write, struct file *filp,
 		    void __user *buffer, size_t *lenp, loff_t *ppos)
 {
 	char buf[64], *p;
-	long pages, seconds;
+	long nr, seconds;
 	int len;
 
 	if (!*lenp || (*ppos && !write)) {
@@ -301,10 +334,10 @@ cmm_timeout_handler(ctl_table *ctl, int write, struct file *filp,
 			return -EFAULT;
 		buf[sizeof(buf) - 1] = '\0';
 		cmm_skip_blanks(buf, &p);
-		pages = simple_strtoul(p, &p, 0);
+		nr = simple_strtoul(p, &p, 0);
 		cmm_skip_blanks(p, &p);
 		seconds = simple_strtoul(p, &p, 0);
-		cmm_set_timeout(pages, seconds);
+		cmm_set_timeout(nr, seconds);
 	} else {
 		len = sprintf(buf, "%ld %ld\n",
 			      cmm_timeout_pages, cmm_timeout_seconds);
@@ -357,7 +390,7 @@ static struct ctl_table cmm_dir_table[] = {
 static void
 cmm_smsg_target(char *from, char *msg)
 {
-	long pages, seconds;
+	long nr, seconds;
 
 	if (strlen(sender) > 0 && strcmp(from, sender) != 0)
 		return;
@@ -366,27 +399,27 @@ cmm_smsg_target(char *from, char *msg)
 	if (strncmp(msg, "SHRINK", 6) == 0) {
 		if (!cmm_skip_blanks(msg + 6, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_set_pages(pages);
+			cmm_set_pages(nr);
 	} else if (strncmp(msg, "RELEASE", 7) == 0) {
 		if (!cmm_skip_blanks(msg + 7, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_add_timed_pages(pages);
+			cmm_add_timed_pages(nr);
 	} else if (strncmp(msg, "REUSE", 5) == 0) {
 		if (!cmm_skip_blanks(msg + 5, &msg))
 			return;
-		pages = simple_strtoul(msg, &msg, 0);
+		nr = simple_strtoul(msg, &msg, 0);
 		if (!cmm_skip_blanks(msg, &msg))
 			return;
 		seconds = simple_strtoul(msg, &msg, 0);
 		cmm_skip_blanks(msg, &msg);
 		if (*msg == '\0')
-			cmm_set_timeout(pages, seconds);
+			cmm_set_timeout(nr, seconds);
 	}
 }
 #endif
@@ -402,6 +435,7 @@ cmm_init (void)
 #ifdef CONFIG_CMM_IUCV
 	smsg_register_callback(SMSG_PREFIX, cmm_smsg_target);
 #endif
+	register_oom_notifier(&cmm_oom_nb);
 	INIT_WORK(&cmm_thread_starter, (void *) cmm_start_thread, NULL);
 	init_waitqueue_head(&cmm_thread_wait);
 	init_timer(&cmm_timer);
@@ -411,6 +445,7 @@ cmm_init (void)
 static void
 cmm_exit(void)
 {
+	unregister_oom_notifier(&cmm_oom_nb);
 	cmm_free_pages(cmm_pages, &cmm_pages, &cmm_page_list);
 	cmm_free_pages(cmm_timed_pages, &cmm_timed_pages, &cmm_timed_page_list);
 #ifdef CONFIG_CMM_PROC
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 34a6bc3e6cf3..32db06c8ffe0 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -10,6 +10,8 @@
 #include <asm/atomic.h>
 #include <asm/page.h>
 
+struct notifier_block;
+
 #define SWAP_FLAG_PREFER	0x8000	/* set if swap priority specified */
 #define SWAP_FLAG_PRIO_MASK	0x7fff
 #define SWAP_FLAG_PRIO_SHIFT	0
@@ -156,6 +158,8 @@ struct swap_list_t {
 
 /* linux/mm/oom_kill.c */
 extern void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order);
+extern int register_oom_notifier(struct notifier_block *nb);
+extern int unregister_oom_notifier(struct notifier_block *nb);
 
 /* linux/mm/memory.c */
 extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index b9af136e5cfa..7d056843fa2d 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -21,6 +21,8 @@
 #include <linux/timex.h>
 #include <linux/jiffies.h>
 #include <linux/cpuset.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
 
 int sysctl_panic_on_oom;
 /* #define DEBUG */
@@ -306,6 +308,20 @@ static int oom_kill_process(struct task_struct *p, unsigned long points,
 	return oom_kill_task(p, message);
 }
 
+static BLOCKING_NOTIFIER_HEAD(oom_notify_list);
+
+int register_oom_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_register(&oom_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(register_oom_notifier);
+
+int unregister_oom_notifier(struct notifier_block *nb)
+{
+	return blocking_notifier_chain_unregister(&oom_notify_list, nb);
+}
+EXPORT_SYMBOL_GPL(unregister_oom_notifier);
+
 /**
  * out_of_memory - kill the "best" process when we run out of memory
  *
@@ -318,6 +334,12 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 {
 	struct task_struct *p;
 	unsigned long points = 0;
+	unsigned long freed = 0;
+
+	blocking_notifier_call_chain(&oom_notify_list, 0, &freed);
+	if (freed > 0)
+		/* Got some memory back in the last second. */
+		return;
 
 	if (printk_ratelimit()) {
 		printk("oom-killer: gfp_mask=0x%x, order=%d\n",
-- 
cgit v1.2.1


From 7ff6f08295d90ab20d25200ef485ebb45b1b8d71 Mon Sep 17 00:00:00 2001
From: Martin Peschke <mp3@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:21 -0700
Subject: [PATCH] CPU hotplug compatible alloc_percpu()

This patch splits alloc_percpu() up into two phases.  Likewise for
free_percpu().  This allows clients to limit initial allocations to online
cpu's, and to populate or depopulate per-cpu data at run time as needed:

  struct my_struct *obj;

  /* initial allocation for online cpu's */
  obj = percpu_alloc(sizeof(struct my_struct), GFP_KERNEL);

  ...

  /* populate per-cpu data for cpu coming online */
  ptr = percpu_populate(obj, sizeof(struct my_struct), GFP_KERNEL, cpu);

  ...

  /* access per-cpu object */
  ptr = percpu_ptr(obj, smp_processor_id());

  ...

  /* depopulate per-cpu data for cpu going offline */
  percpu_depopulate(obj, cpu);

  ...

  /* final removal */
  percpu_free(obj);

Signed-off-by: Martin Peschke <mp3@de.ibm.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/percpu.h |  79 +++++++++++++++++------
 mm/slab.c              | 166 ++++++++++++++++++++++++++++++++-----------------
 2 files changed, 169 insertions(+), 76 deletions(-)

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index f926490a7d8b..3835a9642f13 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -1,9 +1,12 @@
 #ifndef __LINUX_PERCPU_H
 #define __LINUX_PERCPU_H
+
 #include <linux/spinlock.h> /* For preempt_disable() */
 #include <linux/slab.h> /* For kmalloc() */
 #include <linux/smp.h>
 #include <linux/string.h> /* For memset() */
+#include <linux/cpumask.h>
+
 #include <asm/percpu.h>
 
 /* Enough to cover all DEFINE_PER_CPUs in kernel, including modules. */
@@ -27,39 +30,77 @@ struct percpu_data {
 	void *ptrs[NR_CPUS];
 };
 
+#define __percpu_disguise(pdata) (struct percpu_data *)~(unsigned long)(pdata)
 /* 
- * Use this to get to a cpu's version of the per-cpu object allocated using
- * alloc_percpu.  Non-atomic access to the current CPU's version should
+ * Use this to get to a cpu's version of the per-cpu object dynamically
+ * allocated. Non-atomic access to the current CPU's version should
  * probably be combined with get_cpu()/put_cpu().
  */ 
-#define per_cpu_ptr(ptr, cpu)                   \
-({                                              \
-        struct percpu_data *__p = (struct percpu_data *)~(unsigned long)(ptr); \
-        (__typeof__(ptr))__p->ptrs[(cpu)];	\
+#define percpu_ptr(ptr, cpu)                              \
+({                                                        \
+        struct percpu_data *__p = __percpu_disguise(ptr); \
+        (__typeof__(ptr))__p->ptrs[(cpu)];	          \
 })
 
-extern void *__alloc_percpu(size_t size);
-extern void free_percpu(const void *);
+extern void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu);
+extern void percpu_depopulate(void *__pdata, int cpu);
+extern int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+				  cpumask_t *mask);
+extern void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask);
+extern void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask);
+extern void percpu_free(void *__pdata);
 
 #else /* CONFIG_SMP */
 
-#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
+#define percpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
+
+static inline void percpu_depopulate(void *__pdata, int cpu)
+{
+}
+
+static inline void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+{
+}
 
-static inline void *__alloc_percpu(size_t size)
+static inline void *percpu_populate(void *__pdata, size_t size, gfp_t gfp,
+				    int cpu)
 {
-	void *ret = kmalloc(size, GFP_KERNEL);
-	if (ret)
-		memset(ret, 0, size);
-	return ret;
+	return percpu_ptr(__pdata, cpu);
 }
-static inline void free_percpu(const void *ptr)
-{	
-	kfree(ptr);
+
+static inline int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+					 cpumask_t *mask)
+{
+	return 0;
+}
+
+static inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+{
+	return kzalloc(size, gfp);
+}
+
+static inline void percpu_free(void *__pdata)
+{
+	kfree(__pdata);
 }
 
 #endif /* CONFIG_SMP */
 
-/* Simple wrapper for the common case: zeros memory. */
-#define alloc_percpu(type)	((type *)(__alloc_percpu(sizeof(type))))
+#define percpu_populate_mask(__pdata, size, gfp, mask) \
+	__percpu_populate_mask((__pdata), (size), (gfp), &(mask))
+#define percpu_depopulate_mask(__pdata, mask) \
+	__percpu_depopulate_mask((__pdata), &(mask))
+#define percpu_alloc_mask(size, gfp, mask) \
+	__percpu_alloc_mask((size), (gfp), &(mask))
+
+#define percpu_alloc(size, gfp) percpu_alloc_mask((size), (gfp), cpu_online_map)
+
+/* (legacy) interface for use without CPU hotplug handling */
+
+#define __alloc_percpu(size)	percpu_alloc_mask((size), GFP_KERNEL, \
+						  cpu_possible_map)
+#define alloc_percpu(type)	(type *)__alloc_percpu(sizeof(type))
+#define free_percpu(ptr)	percpu_free((ptr))
+#define per_cpu_ptr(ptr, cpu)	percpu_ptr((ptr), (cpu))
 
 #endif /* __LINUX_PERCPU_H */
diff --git a/mm/slab.c b/mm/slab.c
index 5870bcbd33cf..00584dbbec03 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3371,52 +3371,127 @@ EXPORT_SYMBOL(__kmalloc_track_caller);
 
 #ifdef CONFIG_SMP
 /**
- * __alloc_percpu - allocate one copy of the object for every present
- * cpu in the system, zeroing them.
- * Objects should be dereferenced using the per_cpu_ptr macro only.
+ * percpu_depopulate - depopulate per-cpu data for given cpu
+ * @__pdata: per-cpu data to depopulate
+ * @cpu: depopulate per-cpu data for this cpu
  *
- * @size: how many bytes of memory are required.
+ * Depopulating per-cpu data for a cpu going offline would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
  */
-void *__alloc_percpu(size_t size)
+void percpu_depopulate(void *__pdata, int cpu)
 {
-	int i;
-	struct percpu_data *pdata = kmalloc(sizeof(*pdata), GFP_KERNEL);
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	if (pdata->ptrs[cpu]) {
+		kfree(pdata->ptrs[cpu]);
+		pdata->ptrs[cpu] = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(percpu_depopulate);
 
-	if (!pdata)
-		return NULL;
+/**
+ * percpu_depopulate_mask - depopulate per-cpu data for some cpu's
+ * @__pdata: per-cpu data to depopulate
+ * @mask: depopulate per-cpu data for cpu's selected through mask bits
+ */
+void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+{
+	int cpu;
+	for_each_cpu_mask(cpu, *mask)
+		percpu_depopulate(__pdata, cpu);
+}
+EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
 
-	/*
-	 * Cannot use for_each_online_cpu since a cpu may come online
-	 * and we have no way of figuring out how to fix the array
-	 * that we have allocated then....
-	 */
-	for_each_possible_cpu(i) {
-		int node = cpu_to_node(i);
+/**
+ * percpu_populate - populate per-cpu data for given cpu
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @cpu: populate per-data for this cpu
+ *
+ * Populating per-cpu data for a cpu coming online would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
+ * Per-cpu object is populated with zeroed buffer.
+ */
+void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
+{
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	int node = cpu_to_node(cpu);
 
-		if (node_online(node))
-			pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL, node);
-		else
-			pdata->ptrs[i] = kmalloc(size, GFP_KERNEL);
+	BUG_ON(pdata->ptrs[cpu]);
+	if (node_online(node)) {
+		/* FIXME: kzalloc_node(size, gfp, node) */
+		pdata->ptrs[cpu] = kmalloc_node(size, gfp, node);
+		if (pdata->ptrs[cpu])
+			memset(pdata->ptrs[cpu], 0, size);
+	} else
+		pdata->ptrs[cpu] = kzalloc(size, gfp);
+	return pdata->ptrs[cpu];
+}
+EXPORT_SYMBOL_GPL(percpu_populate);
 
-		if (!pdata->ptrs[i])
-			goto unwind_oom;
-		memset(pdata->ptrs[i], 0, size);
-	}
+/**
+ * percpu_populate_mask - populate per-cpu data for more cpu's
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-cpu data for cpu's selected through mask bits
+ *
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+			   cpumask_t *mask)
+{
+	cpumask_t populated = CPU_MASK_NONE;
+	int cpu;
+
+	for_each_cpu_mask(cpu, *mask)
+		if (unlikely(!percpu_populate(__pdata, size, gfp, cpu))) {
+			__percpu_depopulate_mask(__pdata, &populated);
+			return -ENOMEM;
+		} else
+			cpu_set(cpu, populated);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__percpu_populate_mask);
 
-	/* Catch derefs w/o wrappers */
-	return (void *)(~(unsigned long)pdata);
+/**
+ * percpu_alloc_mask - initial setup of per-cpu data
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-data for cpu's selected through mask bits
+ *
+ * Populating per-cpu data for all online cpu's would be a typical use case,
+ * which is simplified by the percpu_alloc() wrapper.
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+{
+	void *pdata = kzalloc(sizeof(struct percpu_data), gfp);
+	void *__pdata = __percpu_disguise(pdata);
 
-unwind_oom:
-	while (--i >= 0) {
-		if (!cpu_possible(i))
-			continue;
-		kfree(pdata->ptrs[i]);
-	}
+	if (unlikely(!pdata))
+		return NULL;
+	if (likely(!__percpu_populate_mask(__pdata, size, gfp, mask)))
+		return __pdata;
 	kfree(pdata);
 	return NULL;
 }
-EXPORT_SYMBOL(__alloc_percpu);
-#endif
+EXPORT_SYMBOL_GPL(__percpu_alloc_mask);
+
+/**
+ * percpu_free - final cleanup of per-cpu data
+ * @__pdata: object to clean up
+ *
+ * We simply clean up any per-cpu object left. No need for the client to
+ * track and specify through a bis mask which per-cpu objects are to free.
+ */
+void percpu_free(void *__pdata)
+{
+	__percpu_depopulate_mask(__pdata, &cpu_possible_map);
+	kfree(__percpu_disguise(__pdata));
+}
+EXPORT_SYMBOL_GPL(percpu_free);
+#endif	/* CONFIG_SMP */
 
 /**
  * kmem_cache_free - Deallocate an object
@@ -3463,29 +3538,6 @@ void kfree(const void *objp)
 }
 EXPORT_SYMBOL(kfree);
 
-#ifdef CONFIG_SMP
-/**
- * free_percpu - free previously allocated percpu memory
- * @objp: pointer returned by alloc_percpu.
- *
- * Don't free memory not originally allocated by alloc_percpu()
- * The complemented objp is to check for that.
- */
-void free_percpu(const void *objp)
-{
-	int i;
-	struct percpu_data *p = (struct percpu_data *)(~(unsigned long)objp);
-
-	/*
-	 * We allocate for all cpus so we cannot use for online cpu here.
-	 */
-	for_each_possible_cpu(i)
-	    kfree(p->ptrs[i]);
-	kfree(p);
-}
-EXPORT_SYMBOL(free_percpu);
-#endif
-
 unsigned int kmem_cache_size(struct kmem_cache *cachep)
 {
 	return obj_size(cachep);
-- 
cgit v1.2.1


From bfa5bf6d6446f0028187a727f792fbc7934228ad Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Mon, 25 Sep 2006 23:31:22 -0700
Subject: [PATCH] Add kerneldocs for some functions in mm/memory.c

These functions are already documented quite well with long comments.  Now
add kerneldoc style header to make this turn up in everyones favorite doc
format.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Cc: "Randy.Dunlap" <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/memory.c | 34 +++++++++++++++++++++++++++-------
 1 file changed, 27 insertions(+), 7 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index 65962534b4ed..92a3ebd8d795 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1227,7 +1227,12 @@ out:
 	return retval;
 }
 
-/*
+/**
+ * vm_insert_page - insert single page into user vma
+ * @vma: user vma to map to
+ * @addr: target user address of this page
+ * @page: source kernel page
+ *
  * This allows drivers to insert individual pages they've allocated
  * into a user vma.
  *
@@ -1319,7 +1324,16 @@ static inline int remap_pud_range(struct mm_struct *mm, pgd_t *pgd,
 	return 0;
 }
 
-/*  Note: this is only safe if the mm semaphore is held when called. */
+/**
+ * remap_pfn_range - remap kernel memory to userspace
+ * @vma: user vma to map to
+ * @addr: target user address to start at
+ * @pfn: physical address of kernel memory
+ * @size: size of map area
+ * @prot: page protection flags for this mapping
+ *
+ *  Note: this is only safe if the mm semaphore is held when called.
+ */
 int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr,
 		    unsigned long pfn, unsigned long size, pgprot_t prot)
 {
@@ -1801,9 +1815,10 @@ void unmap_mapping_range(struct address_space *mapping,
 }
 EXPORT_SYMBOL(unmap_mapping_range);
 
-/*
- * Handle all mappings that got truncated by a "truncate()"
- * system call.
+/**
+ * vmtruncate - unmap mappings "freed" by truncate() syscall
+ * @inode: inode of the file used
+ * @offset: file offset to start truncating
  *
  * NOTE! We have to be ready to update the memory sharing
  * between the file and the memory map for a potential last
@@ -1872,11 +1887,16 @@ int vmtruncate_range(struct inode *inode, loff_t offset, loff_t end)
 }
 EXPORT_UNUSED_SYMBOL(vmtruncate_range);  /*  June 2006  */
 
-/* 
+/**
+ * swapin_readahead - swap in pages in hope we need them soon
+ * @entry: swap entry of this memory
+ * @addr: address to start
+ * @vma: user vma this addresses belong to
+ *
  * Primitive swap readahead code. We simply read an aligned block of
  * (1 << page_cluster) entries in the swap area. This method is chosen
  * because it doesn't cost us any seek time.  We also make sure to queue
- * the 'original' request together with the readahead ones...  
+ * the 'original' request together with the readahead ones...
  *
  * This has been extended to use the NUMA policies from the mm triggering
  * the readahead.
-- 
cgit v1.2.1


From 28e4d965e6131ace1e813e93aebca89ac6b82dc1 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:23 -0700
Subject: [PATCH] mm: remove_mapping() safeness

Some users of remove_mapping had been unsafe.

Modify the remove_mapping precondition to ensure the caller has locked the
page and obtained the correct mapping.  Modify callers to ensure the
mapping is the correct one.

[hugh@veritas.com: swapper_space fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 41a3da3d6ccc..16180587fd7d 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -377,8 +377,8 @@ static pageout_t pageout(struct page *page, struct address_space *mapping)
 
 int remove_mapping(struct address_space *mapping, struct page *page)
 {
-	if (!mapping)
-		return 0;		/* truncate got there first */
+	BUG_ON(!PageLocked(page));
+	BUG_ON(mapping != page_mapping(page));
 
 	write_lock_irq(&mapping->tree_lock);
 
@@ -547,7 +547,7 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 				goto free_it;
 		}
 
-		if (!remove_mapping(mapping, page))
+		if (!mapping || !remove_mapping(mapping, page))
 			goto keep_locked;
 
 free_it:
-- 
cgit v1.2.1


From db37648cd6ce9b828abd6d49aa3d269926ee7b7d Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:24 -0700
Subject: [PATCH] mm: non syncing lock_page()

lock_page needs the caller to have a reference on the page->mapping inode
due to sync_page, ergo set_page_dirty_lock is obviously buggy according to
its comments.

Solve it by introducing a new lock_page_nosync which does not do a sync_page.

akpm: unpleasant solution to an unpleasant problem.  If it goes wrong it could
cause great slowdowns while the lock_page() caller waits for kblockd to
perform the unplug.  And if a filesystem has special sync_page() requirements
(none presently do), permanent hangs are possible.

otoh, set_page_dirty_lock() is usually (always?) called against userspace
pages.  They are always up-to-date, so there shouldn't be any pending read I/O
against these pages.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/pagemap.h | 15 +++++++++++++++
 mm/filemap.c            | 17 +++++++++++++++++
 mm/page-writeback.c     |  2 +-
 3 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 0a2f5d27f60e..64f950925151 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -130,14 +130,29 @@ static inline pgoff_t linear_page_index(struct vm_area_struct *vma,
 }
 
 extern void FASTCALL(__lock_page(struct page *page));
+extern void FASTCALL(__lock_page_nosync(struct page *page));
 extern void FASTCALL(unlock_page(struct page *page));
 
+/*
+ * lock_page may only be called if we have the page's inode pinned.
+ */
 static inline void lock_page(struct page *page)
 {
 	might_sleep();
 	if (TestSetPageLocked(page))
 		__lock_page(page);
 }
+
+/*
+ * lock_page_nosync should only be used if we can't pin the page's inode.
+ * Doesn't play quite so well with block device plugging.
+ */
+static inline void lock_page_nosync(struct page *page)
+{
+	might_sleep();
+	if (TestSetPageLocked(page))
+		__lock_page_nosync(page);
+}
 	
 /*
  * This is exported only for wait_on_page_locked/wait_on_page_writeback.
diff --git a/mm/filemap.c b/mm/filemap.c
index b9a60c43b61a..d5af1cab4268 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -488,6 +488,12 @@ struct page *page_cache_alloc_cold(struct address_space *x)
 EXPORT_SYMBOL(page_cache_alloc_cold);
 #endif
 
+static int __sleep_on_page_lock(void *word)
+{
+	io_schedule();
+	return 0;
+}
+
 /*
  * In order to wait for pages to become available there must be
  * waitqueues associated with pages. By using a hash table of
@@ -577,6 +583,17 @@ void fastcall __lock_page(struct page *page)
 }
 EXPORT_SYMBOL(__lock_page);
 
+/*
+ * Variant of lock_page that does not require the caller to hold a reference
+ * on the page's mapping.
+ */
+void fastcall __lock_page_nosync(struct page *page)
+{
+	DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);
+	__wait_on_bit_lock(page_waitqueue(page), &wait, __sleep_on_page_lock,
+							TASK_UNINTERRUPTIBLE);
+}
+
 /**
  * find_get_page - find and get a page reference
  * @mapping: the address_space to search
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index b9f4c6f1be86..555752907dc3 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -701,7 +701,7 @@ int set_page_dirty_lock(struct page *page)
 {
 	int ret;
 
-	lock_page(page);
+	lock_page_nosync(page);
 	ret = set_page_dirty(page);
 	unlock_page(page);
 	return ret;
-- 
cgit v1.2.1


From ca5f9703dffa012cc46166e6206c5a992910e041 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Mon, 25 Sep 2006 23:31:25 -0700
Subject: [PATCH] slab: respect architecture and caller mandated alignment

As explained by Heiko, on s390 (32-bit) ARCH_KMALLOC_MINALIGN is set to
eight because their common I/O layer allocates data structures that need to
have an eight byte alignment.  This does not work when CONFIG_SLAB_DEBUG is
enabled because kmem_cache_create will override alignment to BYTES_PER_WORD
which is four.

So change kmem_cache_create to ensure cache alignment is always at minimum
what the architecture or caller mandates even if slab debugging is enabled.

Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index 00584dbbec03..d47d0e186973 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2096,6 +2096,15 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 	} else {
 		ralign = BYTES_PER_WORD;
 	}
+
+	/*
+	 * Redzoning and user store require word alignment. Note this will be
+	 * overridden by architecture or caller mandated alignment if either
+	 * is greater than BYTES_PER_WORD.
+	 */
+	if (flags & SLAB_RED_ZONE || flags & SLAB_STORE_USER)
+		ralign = BYTES_PER_WORD;
+
 	/* 2) arch mandated alignment: disables debug if necessary */
 	if (ralign < ARCH_SLAB_MINALIGN) {
 		ralign = ARCH_SLAB_MINALIGN;
@@ -2109,8 +2118,7 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 			flags &= ~(SLAB_RED_ZONE | SLAB_STORE_USER);
 	}
 	/*
-	 * 4) Store it. Note that the debug code below can reduce
-	 *    the alignment to BYTES_PER_WORD.
+	 * 4) Store it.
 	 */
 	align = ralign;
 
@@ -2122,20 +2130,19 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 #if DEBUG
 	cachep->obj_size = size;
 
+	/*
+	 * Both debugging options require word-alignment which is calculated
+	 * into align above.
+	 */
 	if (flags & SLAB_RED_ZONE) {
-		/* redzoning only works with word aligned caches */
-		align = BYTES_PER_WORD;
-
 		/* add space for red zone words */
 		cachep->obj_offset += BYTES_PER_WORD;
 		size += 2 * BYTES_PER_WORD;
 	}
 	if (flags & SLAB_STORE_USER) {
-		/* user store requires word alignment and
-		 * one word storage behind the end of the real
-		 * object.
+		/* user store requires one word storage behind the end of
+		 * the real object.
 		 */
-		align = BYTES_PER_WORD;
 		size += BYTES_PER_WORD;
 	}
 #if FORCED_DEBUG && defined(CONFIG_DEBUG_PAGEALLOC)
-- 
cgit v1.2.1


From 6ddab3b9ebebc88bfdd8107c64f12d7e4480c559 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 25 Sep 2006 23:31:26 -0700
Subject: [PATCH] mm: swap write failure fixup

Currently we can silently drop data if the write to swap failed.  It
usually doesn't result in data-corruption because on page-in the process
will receive SIGBUS (assuming write-failure implies read-failure).

This assumption might or might not be valid.

This patch will avoid the page being discarded after a failed write.  But
will print a warning the sysadmin _should_ take to heart, if a lot of swap
space becomes un-writeable, OOM is not far off.

Tested by making the write fail 'randomly' once every 50 writes or so.

[akpm@osdl.org: printk warning fix]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_io.c | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/mm/page_io.c b/mm/page_io.c
index 88029948d00a..d2f0a5783370 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -52,8 +52,23 @@ static int end_swap_bio_write(struct bio *bio, unsigned int bytes_done, int err)
 	if (bio->bi_size)
 		return 1;
 
-	if (!uptodate)
+	if (!uptodate) {
 		SetPageError(page);
+		/*
+		 * We failed to write the page out to swap-space.
+		 * Re-dirty the page in order to avoid it being reclaimed.
+		 * Also print a dire warning that things will go BAD (tm)
+		 * very quickly.
+		 *
+		 * Also clear PG_reclaim to avoid rotate_reclaimable_page()
+		 */
+		set_page_dirty(page);
+		printk(KERN_ALERT "Write-error on swap-device (%u:%u:%Lu)\n",
+				imajor(bio->bi_bdev->bd_inode),
+				iminor(bio->bi_bdev->bd_inode),
+				(unsigned long long)bio->bi_sector);
+		ClearPageReclaim(page);
+	}
 	end_page_writeback(page);
 	bio_put(bio);
 	return 0;
@@ -70,6 +85,10 @@ static int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err)
 	if (!uptodate) {
 		SetPageError(page);
 		ClearPageUptodate(page);
+		printk(KERN_ALERT "Read-error on swap-device (%u:%u:%Lu)\n",
+				imajor(bio->bi_bdev->bd_inode),
+				iminor(bio->bi_bdev->bd_inode),
+				(unsigned long long)bio->bi_sector);
 	} else {
 		SetPageUptodate(page);
 	}
-- 
cgit v1.2.1


From 408d85441cd5a9bd6bc851d677a10c605ed8db5f Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:27 -0700
Subject: [PATCH] oom: use unreclaimable info

__alloc_pages currently starts shooting if page reclaim has failed to free up
swap_cluster_max pages in one run through the priorities.  This is not always
a good indicator on its own, so make use of the all_unreclaimable logic as
well: don't consider going OOM until all zones we're interested in are
unreclaimable.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 16180587fd7d..ba18d0c36b83 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -62,6 +62,8 @@ struct scan_control {
 	int swap_cluster_max;
 
 	int swappiness;
+
+	int all_unreclaimable;
 };
 
 /*
@@ -925,6 +927,7 @@ static unsigned long shrink_zones(int priority, struct zone **zones,
 	unsigned long nr_reclaimed = 0;
 	int i;
 
+	sc->all_unreclaimable = 1;
 	for (i = 0; zones[i] != NULL; i++) {
 		struct zone *zone = zones[i];
 
@@ -941,6 +944,8 @@ static unsigned long shrink_zones(int priority, struct zone **zones,
 		if (zone->all_unreclaimable && priority != DEF_PRIORITY)
 			continue;	/* Let kswapd poll it */
 
+		sc->all_unreclaimable = 0;
+
 		nr_reclaimed += shrink_zone(priority, zone, sc);
 	}
 	return nr_reclaimed;
@@ -1021,6 +1026,9 @@ unsigned long try_to_free_pages(struct zone **zones, gfp_t gfp_mask)
 		if (sc.nr_scanned && priority < DEF_PRIORITY - 2)
 			blk_congestion_wait(WRITE, HZ/10);
 	}
+	/* top priority shrink_caches still had more to do? don't OOM, then */
+	if (!sc.all_unreclaimable)
+		ret = 1;
 out:
 	for (i = 0; zones[i] != 0; i++) {
 		struct zone *zone = zones[i];
-- 
cgit v1.2.1


From 4ff1ffb4870b007b86f21e5f27eeb11498c4c077 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:28 -0700
Subject: [PATCH] oom: reclaim_mapped on oom

Potentially it takes several scans of the lru lists before we can even start
reclaiming pages.

mapped pages, with young ptes can take 2 passes on the active list + one on
the inactive list.  But reclaim_mapped may not always kick in instantly, so it
could take even more than that.

Raise the threshold for marking a zone as all_unreclaimable from a factor of 4
time the pages in the zone to 6.  Introduce a mechanism to force
reclaim_mapped if we've reached a factor 3 and still haven't made progress.

Previously, a customer doing stress testing was able to easily OOM the box
after using only a small fraction of its swap (~100MB).  After the patches, it
would only OOM after having used up all swap (~800MB).

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index ba18d0c36b83..8f35d7d585cb 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -697,6 +697,11 @@ done:
 	return nr_reclaimed;
 }
 
+static inline int zone_is_near_oom(struct zone *zone)
+{
+	return zone->pages_scanned >= (zone->nr_active + zone->nr_inactive)*3;
+}
+
 /*
  * This moves pages from the active list to the inactive list.
  *
@@ -732,6 +737,9 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 		long distress;
 		long swap_tendency;
 
+		if (zone_is_near_oom(zone))
+			goto force_reclaim_mapped;
+
 		/*
 		 * `distress' is a measure of how much trouble we're having
 		 * reclaiming pages.  0 -> no problems.  100 -> great trouble.
@@ -767,6 +775,7 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 		 * memory onto the inactive list.
 		 */
 		if (swap_tendency >= 100)
+force_reclaim_mapped:
 			reclaim_mapped = 1;
 	}
 
@@ -1161,7 +1170,7 @@ scan:
 			if (zone->all_unreclaimable)
 				continue;
 			if (nr_slab == 0 && zone->pages_scanned >=
-				    (zone->nr_active + zone->nr_inactive) * 4)
+				    (zone->nr_active + zone->nr_inactive) * 6)
 				zone->all_unreclaimable = 1;
 			/*
 			 * If we've done a decent amount of scanning and
-- 
cgit v1.2.1


From 7887a3da753e1ba8244556cc9a2b38c815bfe256 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:29 -0700
Subject: [PATCH] oom: cpuset hint

cpuset_excl_nodes_overlap does not always indicate that killing a task will
not free any memory we for us.  For example, we may be asking for an
allocation from _anywhere_ in the machine, or the task in question may be
pinning memory that is outside its cpuset.  Fix this by just causing
cpuset_excl_nodes_overlap to reduce the badness rather than disallow it.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 7d056843fa2d..4f815b06ac1b 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -128,6 +128,14 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 	if (cap_t(p->cap_effective) & CAP_TO_MASK(CAP_SYS_RAWIO))
 		points /= 4;
 
+	/*
+	 * If p's nodes don't overlap ours, it may still help to kill p
+	 * because p may have allocated or otherwise mapped memory on
+	 * this node before. However it will be less likely.
+	 */
+	if (!cpuset_excl_nodes_overlap(p))
+		points /= 8;
+
 	/*
 	 * Adjust the score by oomkilladj.
 	 */
@@ -198,9 +206,6 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 			continue;
 		if (p->oomkilladj == OOM_DISABLE)
 			continue;
-		/* If p's nodes don't overlap ours, it won't help to kill p. */
-		if (!cpuset_excl_nodes_overlap(p))
-			continue;
 
 		/*
 		 * This is in the process of releasing memory so wait for it
-- 
cgit v1.2.1


From 50ec3bbffbe8a96347c54832d48110a5bc9e9ff8 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:29 -0700
Subject: [PATCH] oom: handle current exiting

If current *is* exiting, it should actually be allowed to access reserved
memory rather than OOM kill something else.  Can't do this via a straight
check in page_alloc.c because that would allow multiple tasks to use up
reserves.  Instead cause current to OOM-kill itself which will mark it as
TIF_MEMDIE.

The current procedure of simply aborting the OOM-kill if a task is exiting can
lead to OOM deadlocks.

In the case of killing a PF_EXITING task, don't make a lot of noise about it.
This becomes more important in future patches, where we can "kill" OOM_DISABLE
tasks.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 35 +++++++++++++++++++++++++++++++----
 1 file changed, 31 insertions(+), 4 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 4f815b06ac1b..0131bae2a16d 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -210,11 +210,26 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 		/*
 		 * This is in the process of releasing memory so wait for it
 		 * to finish before killing some other task by mistake.
+		 *
+		 * However, if p is the current task, we allow the 'kill' to
+		 * go ahead if it is exiting: this will simply set TIF_MEMDIE,
+		 * which will allow it to gain access to memory reserves in
+		 * the process of exiting and releasing its resources.
+		 * Otherwise we could get an OOM deadlock.
 		 */
 		releasing = test_tsk_thread_flag(p, TIF_MEMDIE) ||
 						p->flags & PF_EXITING;
-		if (releasing && !(p->flags & PF_DEAD))
+		if (releasing) {
+			/* PF_DEAD tasks have already released their mm */
+			if (p->flags & PF_DEAD)
+				continue;
+			if (p->flags & PF_EXITING && p == current) {
+				chosen = p;
+				*ppoints = ULONG_MAX;
+				break;
+			}
 			return ERR_PTR(-1UL);
+		}
 		if (p->flags & PF_SWAPOFF)
 			return p;
 
@@ -248,8 +263,11 @@ static void __oom_kill_task(struct task_struct *p, const char *message)
 		return;
 	}
 	task_unlock(p);
-	printk(KERN_ERR "%s: Killed process %d (%s).\n",
+
+	if (message) {
+		printk(KERN_ERR "%s: Killed process %d (%s).\n",
 				message, p->pid, p->comm);
+	}
 
 	/*
 	 * We give our sacrificial lamb high priority and access to
@@ -300,8 +318,17 @@ static int oom_kill_process(struct task_struct *p, unsigned long points,
 	struct task_struct *c;
 	struct list_head *tsk;
 
-	printk(KERN_ERR "Out of Memory: Kill process %d (%s) score %li and "
-		"children.\n", p->pid, p->comm, points);
+	/*
+	 * If the task is already exiting, don't alarm the sysadmin or kill
+	 * its children or threads, just set TIF_MEMDIE so it can die quickly
+	 */
+	if (p->flags & PF_EXITING) {
+		__oom_kill_task(p, NULL);
+		return 0;
+	}
+
+	printk(KERN_ERR "Out of Memory: Kill process %d (%s) score %li"
+			" and children.\n", p->pid, p->comm, points);
 	/* Try to kill a child first */
 	list_for_each(tsk, &p->children) {
 		c = list_entry(tsk, struct task_struct, sibling);
-- 
cgit v1.2.1


From 4a3ede107e422a0c53d28024b0aa902ca22a8768 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:30 -0700
Subject: [PATCH] oom: handle oom_disable exiting

Having the oomkilladj == OOM_DISABLE check before the releasing check means
that oomkilladj == OOM_DISABLE tasks exiting will not stop the OOM killer.

Moving the test down will give the desired behaviour.  Also: it will allow
them to "OOM-kill" themselves if they are exiting.  As per the previous patch,
this is required to prevent OOM killer deadlocks (and they don't actually get
killed, because they're already exiting -- they're simply allowed access to
memory reserves).

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 0131bae2a16d..55a05f1ef76d 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -204,8 +204,6 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 		/* skip the init task with pid == 1 */
 		if (p->pid == 1)
 			continue;
-		if (p->oomkilladj == OOM_DISABLE)
-			continue;
 
 		/*
 		 * This is in the process of releasing memory so wait for it
@@ -230,6 +228,8 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 			}
 			return ERR_PTR(-1UL);
 		}
+		if (p->oomkilladj == OOM_DISABLE)
+			continue;
 		if (p->flags & PF_SWAPOFF)
 			return p;
 
-- 
cgit v1.2.1


From af5b912435de32fbede08cee949429823ed49781 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:31 -0700
Subject: [PATCH] oom: swapoff tasks tweak

PF_SWAPOFF processes currently cause select_bad_process to return straight
away.  Instead, give them high priority, so we will kill them first, however
we also first ensure no parallel OOM kills are happening at the same time.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 55a05f1ef76d..f1aba7e7b760 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -59,6 +59,12 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 		return 0;
 	}
 
+	/*
+	 * swapoff can easily use up all memory, so kill those first.
+	 */
+	if (p->flags & PF_SWAPOFF)
+		return ULONG_MAX;
+
 	/*
 	 * The memory size of the process is the basis for the badness.
 	 */
@@ -230,8 +236,6 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 		}
 		if (p->oomkilladj == OOM_DISABLE)
 			continue;
-		if (p->flags & PF_SWAPOFF)
-			return p;
 
 		points = badness(p, uptime.tv_sec);
 		if (points > *ppoints || !chosen) {
-- 
cgit v1.2.1


From 5081dde33f7a61d28d9b185cc386f12cb837c7a4 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:32 -0700
Subject: [PATCH] oom: kthread infinite loop fix

Skip kernel threads, rather than having them return 0 from badness.
Theoretically, badness might truncate all results to 0, thus a kernel thread
might be picked first, causing an infinite loop.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index f1aba7e7b760..12cd4735dc29 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -207,6 +207,9 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 		unsigned long points;
 		int releasing;
 
+		/* skip kernel threads */
+		if (!p->mm)
+			continue;
 		/* skip the init task with pid == 1 */
 		if (p->pid == 1)
 			continue;
-- 
cgit v1.2.1


From b72f160443cb78b2f8addae6e331d2adaa70f869 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:32 -0700
Subject: [PATCH] oom: more printk

Print the name of the task invoking the OOM killer.  Could make debugging
easier.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 12cd4735dc29..c5e384000585 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -381,8 +381,9 @@ void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 		return;
 
 	if (printk_ratelimit()) {
-		printk("oom-killer: gfp_mask=0x%x, order=%d\n",
-			gfp_mask, order);
+		printk(KERN_WARNING "%s invoked oom-killer: "
+			"gfp_mask=0x%x, order=%d, oomkilladj=%d\n",
+			current->comm, gfp_mask, order, current->oomkilladj);
 		dump_stack();
 		show_mem();
 	}
-- 
cgit v1.2.1


From dfd54cbcc0b834652389ce99b5e656ea5f44a3c1 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 25 Sep 2006 23:31:33 -0700
Subject: [PATCH] bootmem: use MAX_DMA_ADDRESS instead of LOW32LIMIT

Introduce ARCH_LOW_ADDRESS_LIMIT which can be set per architecture to
override the 4GB default limit used by the bootmem allocater within
__alloc_bootmem_low() and __alloc_bootmem_low_node().  E.g.  s390 needs a
2GB limit instead of 4GB.

Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/processor.h |  2 ++
 mm/bootmem.c                 | 11 ++++++++---
 2 files changed, 10 insertions(+), 3 deletions(-)

diff --git a/include/asm-s390/processor.h b/include/asm-s390/processor.h
index a3a4e5fd30d7..578c2209fa76 100644
--- a/include/asm-s390/processor.h
+++ b/include/asm-s390/processor.h
@@ -337,6 +337,8 @@ struct notifier_block;
 int register_idle_notifier(struct notifier_block *nb);
 int unregister_idle_notifier(struct notifier_block *nb);
 
+#define ARCH_LOW_ADDRESS_LIMIT	0x7fffffffUL
+
 #endif
 
 /*
diff --git a/mm/bootmem.c b/mm/bootmem.c
index f0f85fa713ca..d53112fcb404 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -15,6 +15,7 @@
 
 #include <asm/bug.h>
 #include <asm/io.h>
+#include <asm/processor.h>
 
 #include "internal.h"
 
@@ -453,7 +454,9 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 	return __alloc_bootmem(size, align, goal);
 }
 
-#define LOW32LIMIT 0xffffffff
+#ifndef ARCH_LOW_ADDRESS_LIMIT
+#define ARCH_LOW_ADDRESS_LIMIT	0xffffffffUL
+#endif
 
 void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 				  unsigned long goal)
@@ -462,7 +465,8 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	void *ptr;
 
 	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = __alloc_bootmem_core(bdata, size, align, goal, LOW32LIMIT);
+		ptr = __alloc_bootmem_core(bdata, size, align, goal,
+						ARCH_LOW_ADDRESS_LIMIT);
 		if (ptr)
 			return ptr;
 	}
@@ -478,5 +482,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size,
 				       unsigned long align, unsigned long goal)
 {
-	return __alloc_bootmem_core(pgdat->bdata, size, align, goal, LOW32LIMIT);
+	return __alloc_bootmem_core(pgdat->bdata, size, align, goal,
+				    ARCH_LOW_ADDRESS_LIMIT);
 }
-- 
cgit v1.2.1


From e5ac9c5aec7c4bc57fa93f2d37d760a22cb7bd33 Mon Sep 17 00:00:00 2001
From: Ravikiran G Thirumalai <kiran@scalex86.org>
Date: Mon, 25 Sep 2006 23:31:34 -0700
Subject: [PATCH] Add some comments to slab.c

Also, checks if we get a valid slabp_cache for off slab slab-descriptors.
We should always get this.  If we don't, then in that case we, will have to
disable off-slab descriptors for this cache and do the calculations again.
This is a rare case, so add a BUG_ON, for now, just in case.

Signed-off-by: Alok N Kataria <alok.kataria@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 29 +++++++++++++++++++++++++++--
 1 file changed, 27 insertions(+), 2 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index d47d0e186973..3ad2f64998fd 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2206,8 +2206,17 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 		cachep->gfpflags |= GFP_DMA;
 	cachep->buffer_size = size;
 
-	if (flags & CFLGS_OFF_SLAB)
+	if (flags & CFLGS_OFF_SLAB) {
 		cachep->slabp_cache = kmem_find_general_cachep(slab_size, 0u);
+		/*
+		 * This is a possibility for one of the malloc_sizes caches.
+		 * But since we go off slab only for object size greater than
+		 * PAGE_SIZE/8, and malloc_sizes gets created in ascending order,
+		 * this should not happen at all.
+		 * But leave a BUG_ON for some lucky dude.
+		 */
+		BUG_ON(!cachep->slabp_cache);
+	}
 	cachep->ctor = ctor;
 	cachep->dtor = dtor;
 	cachep->name = name;
@@ -2441,7 +2450,17 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
-/* Get the memory for a slab management obj. */
+/*
+ * Get the memory for a slab management obj.
+ * For a slab cache when the slab descriptor is off-slab, slab descriptors
+ * always come from malloc_sizes caches.  The slab descriptor cannot
+ * come from the same cache which is getting created because,
+ * when we are searching for an appropriate cache for these
+ * descriptors in kmem_cache_create, we search through the malloc_sizes array.
+ * If we are creating a malloc_sizes cache here it would not be visible to
+ * kmem_find_general_cachep till the initialization is complete.
+ * Hence we cannot have slabp_cache same as the original cache.
+ */
 static struct slab *alloc_slabmgmt(struct kmem_cache *cachep, void *objp,
 				   int colour_off, gfp_t local_flags,
 				   int nodeid)
@@ -3125,6 +3144,12 @@ static void free_block(struct kmem_cache *cachep, void **objpp, int nr_objects,
 		if (slabp->inuse == 0) {
 			if (l3->free_objects > l3->free_limit) {
 				l3->free_objects -= cachep->num;
+				/* No need to drop any previously held
+				 * lock here, even if we have a off-slab slab
+				 * descriptor it is guaranteed to come from
+				 * a different cache, refer to comments before
+				 * alloc_slabmgmt.
+				 */
 				slab_destroy(cachep, slabp);
 			} else {
 				list_add(&slabp->list, &l3->slabs_free);
-- 
cgit v1.2.1


From da6052f7b33abe55fbfd7d2213815f58c00a88d4 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 25 Sep 2006 23:31:35 -0700
Subject: [PATCH] update some mm/ comments

Let's try to keep mm/ comments more useful and up to date. This is a start.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h         | 68 +++++++++++++++++++++++++++-------------------
 include/linux/page-flags.h | 35 ++++++++++++++----------
 mm/filemap.c               |  8 +++---
 3 files changed, 64 insertions(+), 47 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 2db4229a0066..f2018775b995 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -219,7 +219,8 @@ struct inode;
  * Each physical page in the system has a struct page associated with
  * it to keep track of whatever it is we are using the page for at the
  * moment. Note that we have no way to track which tasks are using
- * a page.
+ * a page, though if it is a pagecache page, rmap structures can tell us
+ * who is mapping it.
  */
 struct page {
 	unsigned long flags;		/* Atomic flags, some possibly
@@ -299,8 +300,7 @@ struct page {
  */
 
 /*
- * Drop a ref, return true if the logical refcount fell to zero (the page has
- * no users)
+ * Drop a ref, return true if the refcount fell to zero (the page has no users)
  */
 static inline int put_page_testzero(struct page *page)
 {
@@ -356,43 +356,55 @@ void split_page(struct page *page, unsigned int order);
  * For the non-reserved pages, page_count(page) denotes a reference count.
  *   page_count() == 0 means the page is free. page->lru is then used for
  *   freelist management in the buddy allocator.
- *   page_count() == 1 means the page is used for exactly one purpose
- *   (e.g. a private data page of one process).
+ *   page_count() > 0  means the page has been allocated.
  *
- * A page may be used for kmalloc() or anyone else who does a
- * __get_free_page(). In this case the page_count() is at least 1, and
- * all other fields are unused but should be 0 or NULL. The
- * management of this page is the responsibility of the one who uses
- * it.
+ * Pages are allocated by the slab allocator in order to provide memory
+ * to kmalloc and kmem_cache_alloc. In this case, the management of the
+ * page, and the fields in 'struct page' are the responsibility of mm/slab.c
+ * unless a particular usage is carefully commented. (the responsibility of
+ * freeing the kmalloc memory is the caller's, of course).
  *
- * The other pages (we may call them "process pages") are completely
+ * A page may be used by anyone else who does a __get_free_page().
+ * In this case, page_count still tracks the references, and should only
+ * be used through the normal accessor functions. The top bits of page->flags
+ * and page->virtual store page management information, but all other fields
+ * are unused and could be used privately, carefully. The management of this
+ * page is the responsibility of the one who allocated it, and those who have
+ * subsequently been given references to it.
+ *
+ * The other pages (we may call them "pagecache pages") are completely
  * managed by the Linux memory manager: I/O, buffers, swapping etc.
  * The following discussion applies only to them.
  *
- * A page may belong to an inode's memory mapping. In this case,
- * page->mapping is the pointer to the inode, and page->index is the
- * file offset of the page, in units of PAGE_CACHE_SIZE.
+ * A pagecache page contains an opaque `private' member, which belongs to the
+ * page's address_space. Usually, this is the address of a circular list of
+ * the page's disk buffers. PG_private must be set to tell the VM to call
+ * into the filesystem to release these pages.
  *
- * A page contains an opaque `private' member, which belongs to the
- * page's address_space.  Usually, this is the address of a circular
- * list of the page's disk buffers.
+ * A page may belong to an inode's memory mapping. In this case, page->mapping
+ * is the pointer to the inode, and page->index is the file offset of the page,
+ * in units of PAGE_CACHE_SIZE.
  *
- * For pages belonging to inodes, the page_count() is the number of
- * attaches, plus 1 if `private' contains something, plus one for
- * the page cache itself.
+ * If pagecache pages are not associated with an inode, they are said to be
+ * anonymous pages. These may become associated with the swapcache, and in that
+ * case PG_swapcache is set, and page->private is an offset into the swapcache.
  *
- * Instead of keeping dirty/clean pages in per address-space lists, we instead
- * now tag pages as dirty/under writeback in the radix tree.
+ * In either case (swapcache or inode backed), the pagecache itself holds one
+ * reference to the page. Setting PG_private should also increment the
+ * refcount. The each user mapping also has a reference to the page.
  *
- * There is also a per-mapping radix tree mapping index to the page
- * in memory if present. The tree is rooted at mapping->root.  
+ * The pagecache pages are stored in a per-mapping radix tree, which is
+ * rooted at mapping->page_tree, and indexed by offset.
+ * Where 2.4 and early 2.6 kernels kept dirty/clean pages in per-address_space
+ * lists, we instead now tag pages as dirty/writeback in the radix tree.
  *
- * All process pages can do I/O:
+ * All pagecache pages may be subject to I/O:
  * - inode pages may need to be read from disk,
  * - inode pages which have been modified and are MAP_SHARED may need
- *   to be written to disk,
- * - private pages which have been modified may need to be swapped out
- *   to swap space and (later) to be read back into memory.
+ *   to be written back to the inode on disk,
+ * - anonymous pages (including MAP_PRIVATE file mappings) which have been
+ *   modified may need to be swapped out to swap space and (later) to be read
+ *   back into memory.
  */
 
 /*
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 5748642e9f36..9d7921dd50f0 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -13,24 +13,25 @@
  * PG_reserved is set for special pages, which can never be swapped out. Some
  * of them might not even exist (eg empty_bad_page)...
  *
- * The PG_private bitflag is set if page->private contains a valid value.
+ * The PG_private bitflag is set on pagecache pages if they contain filesystem
+ * specific data (which is normally at page->private). It can be used by
+ * private allocations for its own usage.
  *
- * During disk I/O, PG_locked is used. This bit is set before I/O and
- * reset when I/O completes. page_waitqueue(page) is a wait queue of all tasks
- * waiting for the I/O on this page to complete.
+ * During initiation of disk I/O, PG_locked is set. This bit is set before I/O
+ * and cleared when writeback _starts_ or when read _completes_. PG_writeback
+ * is set before writeback starts and cleared when it finishes.
+ *
+ * PG_locked also pins a page in pagecache, and blocks truncation of the file
+ * while it is held.
+ *
+ * page_waitqueue(page) is a wait queue of all tasks waiting for the page
+ * to become unlocked.
  *
  * PG_uptodate tells whether the page's contents is valid.  When a read
  * completes, the page becomes uptodate, unless a disk I/O error happened.
  *
- * For choosing which pages to swap out, inode pages carry a PG_referenced bit,
- * which is set any time the system accesses that page through the (mapping,
- * index) hash table.  This referenced bit, together with the referenced bit
- * in the page tables, is used to manipulate page->age and move the page across
- * the active, inactive_dirty and inactive_clean lists.
- *
- * Note that the referenced bit, the page->lru list_head and the active,
- * inactive_dirty and inactive_clean lists are protected by the
- * zone->lru_lock, and *NOT* by the usual PG_locked bit!
+ * PG_referenced, PG_reclaim are used for page reclaim for anonymous and
+ * file-backed pagecache (see mm/vmscan.c).
  *
  * PG_error is set to indicate that an I/O error occurred on this page.
  *
@@ -42,6 +43,10 @@
  * space, they need to be kmapped separately for doing IO on the pages.  The
  * struct page (these bits with information) are always mapped into kernel
  * address space...
+ *
+ * PG_buddy is set to indicate that the page is free and in the buddy system
+ * (see mm/page_alloc.c).
+ *
  */
 
 /*
@@ -74,7 +79,7 @@
 #define PG_checked		 8	/* kill me in 2.5.<early>. */
 #define PG_arch_1		 9
 #define PG_reserved		10
-#define PG_private		11	/* Has something at ->private */
+#define PG_private		11	/* If pagecache, has fs-private data */
 
 #define PG_writeback		12	/* Page is under writeback */
 #define PG_nosave		13	/* Used for system suspend/resume */
@@ -83,7 +88,7 @@
 
 #define PG_mappedtodisk		16	/* Has blocks allocated on-disk */
 #define PG_reclaim		17	/* To be reclaimed asap */
-#define PG_nosave_free		18	/* Free, should not be written */
+#define PG_nosave_free		18	/* Used for system suspend/resume */
 #define PG_buddy		19	/* Page is free, on buddy lists */
 
 
diff --git a/mm/filemap.c b/mm/filemap.c
index d5af1cab4268..afcdc72b5e90 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -599,8 +599,8 @@ void fastcall __lock_page_nosync(struct page *page)
  * @mapping: the address_space to search
  * @offset: the page index
  *
- * A rather lightweight function, finding and getting a reference to a
- * hashed page atomically.
+ * Is there a pagecache struct page at the given (mapping, offset) tuple?
+ * If yes, increment its refcount and return it; if no, return NULL.
  */
 struct page * find_get_page(struct address_space *mapping, unsigned long offset)
 {
@@ -987,7 +987,7 @@ page_not_up_to_date:
 		/* Get exclusive access to the page ... */
 		lock_page(page);
 
-		/* Did it get unhashed before we got the lock? */
+		/* Did it get truncated before we got the lock? */
 		if (!page->mapping) {
 			unlock_page(page);
 			page_cache_release(page);
@@ -1627,7 +1627,7 @@ no_cached_page:
 page_not_uptodate:
 	lock_page(page);
 
-	/* Did it get unhashed while we waited for it? */
+	/* Did it get truncated while we waited for it? */
 	if (!page->mapping) {
 		unlock_page(page);
 		goto err;
-- 
cgit v1.2.1


From dbe5e69d2d6e591996ea2b817b887d03b60bb143 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 25 Sep 2006 23:31:36 -0700
Subject: [PATCH] slab: optimize kmalloc_node the same way as kmalloc

[akpm@osdl.org: export fix]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/slab.h | 25 ++++++++++++++++++++++++-
 mm/slab.c            |  4 ++--
 2 files changed, 26 insertions(+), 3 deletions(-)

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 193c03c547ec..2f6bef6a98c9 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -202,7 +202,30 @@ extern int slab_is_available(void);
 
 #ifdef CONFIG_NUMA
 extern void *kmem_cache_alloc_node(kmem_cache_t *, gfp_t flags, int node);
-extern void *kmalloc_node(size_t size, gfp_t flags, int node);
+extern void *__kmalloc_node(size_t size, gfp_t flags, int node);
+
+static inline void *kmalloc_node(size_t size, gfp_t flags, int node)
+{
+	if (__builtin_constant_p(size)) {
+		int i = 0;
+#define CACHE(x) \
+		if (size <= x) \
+			goto found; \
+		else \
+			i++;
+#include "kmalloc_sizes.h"
+#undef CACHE
+		{
+			extern void __you_cannot_kmalloc_that_much(void);
+			__you_cannot_kmalloc_that_much();
+		}
+found:
+		return kmem_cache_alloc_node((flags & GFP_DMA) ?
+			malloc_sizes[i].cs_dmacachep :
+			malloc_sizes[i].cs_cachep, flags, node);
+	}
+	return __kmalloc_node(size, flags, node);
+}
 #else
 static inline void *kmem_cache_alloc_node(kmem_cache_t *cachep, gfp_t flags, int node)
 {
diff --git a/mm/slab.c b/mm/slab.c
index 3ad2f64998fd..5e59ce7a46c8 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3348,7 +3348,7 @@ void *kmem_cache_alloc_node(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 }
 EXPORT_SYMBOL(kmem_cache_alloc_node);
 
-void *kmalloc_node(size_t size, gfp_t flags, int node)
+void *__kmalloc_node(size_t size, gfp_t flags, int node)
 {
 	struct kmem_cache *cachep;
 
@@ -3357,7 +3357,7 @@ void *kmalloc_node(size_t size, gfp_t flags, int node)
 		return NULL;
 	return kmem_cache_alloc_node(cachep, flags, node);
 }
-EXPORT_SYMBOL(kmalloc_node);
+EXPORT_SYMBOL(__kmalloc_node);
 #endif
 
 /**
-- 
cgit v1.2.1


From 117f6eb1d8b8deb6f19fc88fc15bdb413c2a0c79 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:37 -0700
Subject: [PATCH] slab: extract __kmem_cache_destroy from kmem_cache_destroy

The ability to free memory allocated to a slab cache is also useful if an
error occurs during setup of a slab.  So extract the function.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 38 ++++++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 16 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index 5e59ce7a46c8..c714741b253b 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -1833,6 +1833,27 @@ static void set_up_list3s(struct kmem_cache *cachep, int index)
 	}
 }
 
+static void __kmem_cache_destroy(struct kmem_cache *cachep)
+{
+	int i;
+	struct kmem_list3 *l3;
+
+	for_each_online_cpu(i)
+	    kfree(cachep->array[i]);
+
+	/* NUMA: free the list3 structures */
+	for_each_online_node(i) {
+		l3 = cachep->nodelists[i];
+		if (l3) {
+			kfree(l3->shared);
+			free_alien_cache(l3->alien);
+			kfree(l3);
+		}
+	}
+	kmem_cache_free(&cache_cache, cachep);
+}
+
+
 /**
  * calculate_slab_order - calculate size (page order) of slabs
  * @cachep: pointer to the cache that is being created
@@ -2404,9 +2425,6 @@ EXPORT_SYMBOL(kmem_cache_shrink);
  */
 int kmem_cache_destroy(struct kmem_cache *cachep)
 {
-	int i;
-	struct kmem_list3 *l3;
-
 	BUG_ON(!cachep || in_interrupt());
 
 	/* Don't let CPUs to come and go */
@@ -2432,19 +2450,7 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU))
 		synchronize_rcu();
 
-	for_each_online_cpu(i)
-	    kfree(cachep->array[i]);
-
-	/* NUMA: free the list3 structures */
-	for_each_online_node(i) {
-		l3 = cachep->nodelists[i];
-		if (l3) {
-			kfree(l3->shared);
-			free_alien_cache(l3->alien);
-			kfree(l3);
-		}
-	}
-	kmem_cache_free(&cache_cache, cachep);
+	__kmem_cache_destroy(cachep);
 	unlock_cpu_hotplug();
 	return 0;
 }
-- 
cgit v1.2.1


From 2ed3a4ef95ef1a13a424378c34ebd9b7e593f212 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:38 -0700
Subject: [PATCH] slab: do not panic when alloc_kmemlist fails and slab is up

It is fairly easy to get a system to oops by simply sizing a cache via
/proc in such a way that one of the chaches (shared is easiest) becomes
bigger than the maximum allowed slab allocation size.  This occurs because
enable_cpucache() fails if it cannot reallocate some caches.

However, enable_cpucache() is used for multiple purposes: resizing caches,
cache creation and bootstrap.

If the slab is already up then we already have working caches.  The resize
can fail without a problem.  We just need to return the proper error code.
F.e.  after this patch:

# echo "size-64 10000 50 1000" >/proc/slabinfo
-bash: echo: write error: Cannot allocate memory

notice no OOPS.

If we are doing a kmem_cache_create() then we also should not panic but
return -ENOMEM.

If on the other hand we do not have a fully bootstrapped slab allocator yet
then we should indeed panic since we are unable to bring up the slab to its
full functionality.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 35 +++++++++++++++++------------------
 1 file changed, 17 insertions(+), 18 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index c714741b253b..3233c4c7cbce 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -313,7 +313,7 @@ static int drain_freelist(struct kmem_cache *cache,
 			struct kmem_list3 *l3, int tofree);
 static void free_block(struct kmem_cache *cachep, void **objpp, int len,
 			int node);
-static void enable_cpucache(struct kmem_cache *cachep);
+static int enable_cpucache(struct kmem_cache *cachep);
 static void cache_reap(void *unused);
 
 /*
@@ -1490,7 +1490,8 @@ void __init kmem_cache_init(void)
 		struct kmem_cache *cachep;
 		mutex_lock(&cache_chain_mutex);
 		list_for_each_entry(cachep, &cache_chain, next)
-			enable_cpucache(cachep);
+			if (enable_cpucache(cachep))
+				BUG();
 		mutex_unlock(&cache_chain_mutex);
 	}
 
@@ -1924,12 +1925,11 @@ static size_t calculate_slab_order(struct kmem_cache *cachep,
 	return left_over;
 }
 
-static void setup_cpu_cache(struct kmem_cache *cachep)
+static int setup_cpu_cache(struct kmem_cache *cachep)
 {
-	if (g_cpucache_up == FULL) {
-		enable_cpucache(cachep);
-		return;
-	}
+	if (g_cpucache_up == FULL)
+		return enable_cpucache(cachep);
+
 	if (g_cpucache_up == NONE) {
 		/*
 		 * Note: the first kmem_cache_create must create the cache
@@ -1976,6 +1976,7 @@ static void setup_cpu_cache(struct kmem_cache *cachep)
 	cpu_cache_get(cachep)->touched = 0;
 	cachep->batchcount = 1;
 	cachep->limit = BOOT_CPUCACHE_ENTRIES;
+	return 0;
 }
 
 /**
@@ -2242,8 +2243,11 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 	cachep->dtor = dtor;
 	cachep->name = name;
 
-
-	setup_cpu_cache(cachep);
+	if (setup_cpu_cache(cachep)) {
+		__kmem_cache_destroy(cachep);
+		cachep = NULL;
+		goto oops;
+	}
 
 	/* cache setup completed, link it into the list */
 	list_add(&cachep->next, &cache_chain);
@@ -3693,7 +3697,7 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 				int batchcount, int shared)
 {
 	struct ccupdate_struct new;
-	int i, err;
+	int i;
 
 	memset(&new.new, 0, sizeof(new.new));
 	for_each_online_cpu(i) {
@@ -3724,17 +3728,11 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 		kfree(ccold);
 	}
 
-	err = alloc_kmemlist(cachep);
-	if (err) {
-		printk(KERN_ERR "alloc_kmemlist failed for %s, error %d.\n",
-		       cachep->name, -err);
-		BUG();
-	}
-	return 0;
+	return alloc_kmemlist(cachep);
 }
 
 /* Called with cache_chain_mutex held always */
-static void enable_cpucache(struct kmem_cache *cachep)
+static int enable_cpucache(struct kmem_cache *cachep)
 {
 	int err;
 	int limit, shared;
@@ -3786,6 +3784,7 @@ static void enable_cpucache(struct kmem_cache *cachep)
 	if (err)
 		printk(KERN_ERR "enable_cpucache failed for %s, error %d.\n",
 		       cachep->name, -err);
+	return err;
 }
 
 /*
-- 
cgit v1.2.1


From 056c62418cc639bf2fe962c6a6ee56054b838bc7 Mon Sep 17 00:00:00 2001
From: Ravikiran G Thirumalai <kiran@scalex86.org>
Date: Mon, 25 Sep 2006 23:31:38 -0700
Subject: [PATCH] slab: fix lockdep warnings

Place the alien array cache locks of on slab malloc slab caches on a
seperate lockdep class.  This avoids false positives from lockdep

[akpm@osdl.org: build fix]
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 57 +++++++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 43 insertions(+), 14 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index 3233c4c7cbce..2b37a62f6314 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -674,6 +674,8 @@ static struct kmem_cache cache_cache = {
 #endif
 };
 
+#define BAD_ALIEN_MAGIC 0x01020304ul
+
 #ifdef CONFIG_LOCKDEP
 
 /*
@@ -682,29 +684,53 @@ static struct kmem_cache cache_cache = {
  * The locking for this is tricky in that it nests within the locks
  * of all other slabs in a few places; to deal with this special
  * locking we put on-slab caches into a separate lock-class.
+ *
+ * We set lock class for alien array caches which are up during init.
+ * The lock annotation will be lost if all cpus of a node goes down and
+ * then comes back up during hotplug
  */
-static struct lock_class_key on_slab_key;
+static struct lock_class_key on_slab_l3_key;
+static struct lock_class_key on_slab_alc_key;
+
+static inline void init_lock_keys(void)
 
-static inline void init_lock_keys(struct cache_sizes *s)
 {
 	int q;
-
-	for (q = 0; q < MAX_NUMNODES; q++) {
-		if (!s->cs_cachep->nodelists[q] || OFF_SLAB(s->cs_cachep))
-			continue;
-		lockdep_set_class(&s->cs_cachep->nodelists[q]->list_lock,
-				  &on_slab_key);
+	struct cache_sizes *s = malloc_sizes;
+
+	while (s->cs_size != ULONG_MAX) {
+		for_each_node(q) {
+			struct array_cache **alc;
+			int r;
+			struct kmem_list3 *l3 = s->cs_cachep->nodelists[q];
+			if (!l3 || OFF_SLAB(s->cs_cachep))
+				continue;
+			lockdep_set_class(&l3->list_lock, &on_slab_l3_key);
+			alc = l3->alien;
+			/*
+			 * FIXME: This check for BAD_ALIEN_MAGIC
+			 * should go away when common slab code is taught to
+			 * work even without alien caches.
+			 * Currently, non NUMA code returns BAD_ALIEN_MAGIC
+			 * for alloc_alien_cache,
+			 */
+			if (!alc || (unsigned long)alc == BAD_ALIEN_MAGIC)
+				continue;
+			for_each_node(r) {
+				if (alc[r])
+					lockdep_set_class(&alc[r]->lock,
+					     &on_slab_alc_key);
+			}
+		}
+		s++;
 	}
 }
-
 #else
-static inline void init_lock_keys(struct cache_sizes *s)
+static inline void init_lock_keys(void)
 {
 }
 #endif
 
-
-
 /* Guard access to the cache-chain. */
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
@@ -1091,7 +1117,7 @@ static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
 
 static inline struct array_cache **alloc_alien_cache(int node, int limit)
 {
-	return (struct array_cache **) 0x01020304ul;
+	return (struct array_cache **)BAD_ALIEN_MAGIC;
 }
 
 static inline void free_alien_cache(struct array_cache **ac_ptr)
@@ -1421,7 +1447,6 @@ void __init kmem_cache_init(void)
 					ARCH_KMALLOC_FLAGS|SLAB_PANIC,
 					NULL, NULL);
 		}
-		init_lock_keys(sizes);
 
 		sizes->cs_dmacachep = kmem_cache_create(names->name_dma,
 					sizes->cs_size,
@@ -1495,6 +1520,10 @@ void __init kmem_cache_init(void)
 		mutex_unlock(&cache_chain_mutex);
 	}
 
+	/* Annotate slab for lockdep -- annotate the malloc caches */
+	init_lock_keys();
+
+
 	/* Done! */
 	g_cpucache_up = FULL;
 
-- 
cgit v1.2.1


From 9b819d204cf602eab1a53a9ec4b8d2ca51e02a1d Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:40 -0700
Subject: [PATCH] Add __GFP_THISNODE to avoid fallback to other nodes and
 ignore cpuset/memory policy restrictions

Add a new gfp flag __GFP_THISNODE to avoid fallback to other nodes.  This
flag is essential if a kernel component requires memory to be located on a
certain node.  It will be needed for alloc_pages_node() to force allocation
on the indicated node and for alloc_pages() to force allocation on the
current node.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 3 ++-
 kernel/cpuset.c     | 2 +-
 mm/mempolicy.c      | 2 +-
 mm/page_alloc.c     | 3 +++
 4 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 63ab88a36f39..0eda5b6dedbd 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -45,6 +45,7 @@ struct vm_area_struct;
 #define __GFP_ZERO	((__force gfp_t)0x8000u)/* Return zeroed page on success */
 #define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */
 #define __GFP_HARDWALL   ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */
+#define __GFP_THISNODE	((__force gfp_t)0x40000u)/* No fallback, no policies */
 
 #define __GFP_BITS_SHIFT 20	/* Room for 20 __GFP_FOO bits */
 #define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
@@ -53,7 +54,7 @@ struct vm_area_struct;
 #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \
 			__GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \
 			__GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \
-			__GFP_NOMEMALLOC|__GFP_HARDWALL)
+			__GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE)
 
 /* This equals 0, but use constants in case they ever change */
 #define GFP_NOWAIT	(GFP_ATOMIC & ~__GFP_HIGH)
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 4ea6f0dc2fc5..76940361273e 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2316,7 +2316,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 	const struct cpuset *cs;	/* current cpuset ancestors */
 	int allowed;			/* is allocation in zone z allowed? */
 
-	if (in_interrupt())
+	if (in_interrupt() || (gfp_mask & __GFP_THISNODE))
 		return 1;
 	node = z->zone_pgdat->node_id;
 	might_sleep_if(!(gfp_mask & __GFP_HARDWALL));
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index c3429a710ab1..8002e1faccda 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1290,7 +1290,7 @@ struct page *alloc_pages_current(gfp_t gfp, unsigned order)
 
 	if ((gfp & __GFP_WAIT) && !in_interrupt())
 		cpuset_update_task_memory_state();
-	if (!pol || in_interrupt())
+	if (!pol || in_interrupt() || (gfp & __GFP_THISNODE))
 		pol = &default_policy;
 	if (pol->policy == MPOL_INTERLEAVE)
 		return alloc_page_interleave(gfp, order, interleave_nodes(pol));
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 208a6b03aa78..ea498788af53 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -893,6 +893,9 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 	 * See also cpuset_zone_allowed() comment in kernel/cpuset.c.
 	 */
 	do {
+		if (unlikely((gfp_mask & __GFP_THISNODE) &&
+			(*z)->zone_pgdat != zonelist->zones[0]->zone_pgdat))
+				break;
 		if ((alloc_flags & ALLOC_CPUSET) &&
 				!cpuset_zone_allowed(*z, gfp_mask))
 			continue;
-- 
cgit v1.2.1


From 3d99cfb5f46191fc68f1343feeb2cf835001f7d7 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:43 -0700
Subject: [PATCH] sys_move_pages: Do not fall back to other nodes

If the user specified a node where we should move the page to then we
really do not want any other node.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/migrate.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/migrate.c b/mm/migrate.c
index 3f1e0c2c942c..6196f45c5263 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -741,7 +741,9 @@ static struct page *new_page_node(struct page *p, unsigned long private,
 
 	*result = &pm->status;
 
-	return alloc_pages_node(pm->node, GFP_HIGHUSER, 0);
+	return alloc_pages_node(pm->node,
+		GFP_HIGHUSER | __GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY,
+		0);
 }
 
 /*
-- 
cgit v1.2.1


From bd1b1677b5d4f39deda068bf5cc43ce3aaec839f Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:44 -0700
Subject: [PATCH] Guarantee that the uncached allocator gets pages on the
 correct node

The uncached allocator manages per node pools.  Specify __GFP_THISNODE in
order to force allocation on the indicated node or fail.  The uncached
allocator has already logic to deal with failing allocations.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/uncached.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index 4c73a6763669..f813caa80570 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -98,7 +98,8 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
 
 	/* attempt to allocate a granule's worth of cached memory pages */
 
-	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO,
+	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO |
+				 __GFP_THISNODE | __GFP_NORETRY | __GFP_NOWARN,
 				IA64_GRANULE_SHIFT-PAGE_SHIFT);
 	if (!page) {
 		mutex_unlock(&uc_pool->add_chunk_mutex);
-- 
cgit v1.2.1


From 1192d526412b1b8ccb1493064cea06efc12c772b Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:45 -0700
Subject: [PATCH] Cleanup: Add zone pointer to get_page_from_freelist

There are frequent references to *z in get_page_from_freelist.

Add an explicit zone variable that can be used in all these places.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index ea498788af53..e8a71657ac4a 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -887,35 +887,37 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 	struct zone **z = zonelist->zones;
 	struct page *page = NULL;
 	int classzone_idx = zone_idx(*z);
+	struct zone *zone;
 
 	/*
 	 * Go through the zonelist once, looking for a zone with enough free.
 	 * See also cpuset_zone_allowed() comment in kernel/cpuset.c.
 	 */
 	do {
+		zone = *z;
 		if (unlikely((gfp_mask & __GFP_THISNODE) &&
-			(*z)->zone_pgdat != zonelist->zones[0]->zone_pgdat))
+			zone->zone_pgdat != zonelist->zones[0]->zone_pgdat))
 				break;
 		if ((alloc_flags & ALLOC_CPUSET) &&
-				!cpuset_zone_allowed(*z, gfp_mask))
+				!cpuset_zone_allowed(zone, gfp_mask))
 			continue;
 
 		if (!(alloc_flags & ALLOC_NO_WATERMARKS)) {
 			unsigned long mark;
 			if (alloc_flags & ALLOC_WMARK_MIN)
-				mark = (*z)->pages_min;
+				mark = zone->pages_min;
 			else if (alloc_flags & ALLOC_WMARK_LOW)
-				mark = (*z)->pages_low;
+				mark = zone->pages_low;
 			else
-				mark = (*z)->pages_high;
-			if (!zone_watermark_ok(*z, order, mark,
+				mark = zone->pages_high;
+			if (!zone_watermark_ok(zone , order, mark,
 				    classzone_idx, alloc_flags))
 				if (!zone_reclaim_mode ||
-				    !zone_reclaim(*z, gfp_mask, order))
+				    !zone_reclaim(zone, gfp_mask, order))
 					continue;
 		}
 
-		page = buffered_rmqueue(zonelist, *z, order, gfp_mask);
+		page = buffered_rmqueue(zonelist, zone, order, gfp_mask);
 		if (page) {
 			break;
 		}
-- 
cgit v1.2.1


From fbd98167e653535c5816be154f2149c0efa7757d Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:45 -0700
Subject: [PATCH] Profiling: require buffer allocation on the correct node

Profiling really suffers with off node buffers.  Fail if no memory is
available on the nodes.  The profiling code can deal with these failures
should they occur.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/profile.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/kernel/profile.c b/kernel/profile.c
index d5bd75e7501c..fb660c7d35ba 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -309,13 +309,17 @@ static int __devinit profile_cpu_callback(struct notifier_block *info,
 		node = cpu_to_node(cpu);
 		per_cpu(cpu_profile_flip, cpu) = 0;
 		if (!per_cpu(cpu_profile_hits, cpu)[1]) {
-			page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0);
+			page = alloc_pages_node(node,
+					GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+					0);
 			if (!page)
 				return NOTIFY_BAD;
 			per_cpu(cpu_profile_hits, cpu)[1] = page_address(page);
 		}
 		if (!per_cpu(cpu_profile_hits, cpu)[0]) {
-			page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0);
+			page = alloc_pages_node(node,
+					GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+					0);
 			if (!page)
 				goto out_free;
 			per_cpu(cpu_profile_hits, cpu)[0] = page_address(page);
@@ -491,12 +495,16 @@ static int __init create_hash_tables(void)
 		int node = cpu_to_node(cpu);
 		struct page *page;
 
-		page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0);
+		page = alloc_pages_node(node,
+				GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+				0);
 		if (!page)
 			goto out_cleanup;
 		per_cpu(cpu_profile_hits, cpu)[1]
 				= (struct profile_hit *)page_address(page);
-		page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0);
+		page = alloc_pages_node(node,
+				GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+				0);
 		if (!page)
 			goto out_cleanup;
 		per_cpu(cpu_profile_hits, cpu)[0]
-- 
cgit v1.2.1


From 980128f223fa3c75e3ebdde650c9f1bcabd4c0a2 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:46 -0700
Subject: [PATCH] Define easier to handle GFP_THISNODE

In many places we will need to use the same combination of flags.  Specify
a single GFP_THISNODE definition for ease of use in gfp.h.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/uncached.c | 3 +--
 include/linux/gfp.h         | 2 ++
 mm/migrate.c                | 4 +---
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index f813caa80570..c58e933694d5 100644
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -98,8 +98,7 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
 
 	/* attempt to allocate a granule's worth of cached memory pages */
 
-	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO |
-				 __GFP_THISNODE | __GFP_NORETRY | __GFP_NOWARN,
+	page = alloc_pages_node(nid, GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
 				IA64_GRANULE_SHIFT-PAGE_SHIFT);
 	if (!page) {
 		mutex_unlock(&uc_pool->add_chunk_mutex);
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 0eda5b6dedbd..8b34aabfe4c6 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -67,6 +67,8 @@ struct vm_area_struct;
 #define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
 			 __GFP_HIGHMEM)
 
+#define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
+
 /* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
    platforms, used as appropriate on others */
 
diff --git a/mm/migrate.c b/mm/migrate.c
index 6196f45c5263..20a8c2687b1e 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -741,9 +741,7 @@ static struct page *new_page_node(struct page *p, unsigned long private,
 
 	*result = &pm->status;
 
-	return alloc_pages_node(pm->node,
-		GFP_HIGHUSER | __GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY,
-		0);
+	return alloc_pages_node(pm->node, GFP_HIGHUSER | GFP_THISNODE, 0);
 }
 
 /*
-- 
cgit v1.2.1


From d2e7b7d0aa021847c59f882b066e7d3812902870 Mon Sep 17 00:00:00 2001
From: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Date: Mon, 25 Sep 2006 23:31:47 -0700
Subject: [PATCH] fix potential stack overflow in mm/slab.c

On High end systems (1024 or so cpus) this can potentially cause stack
overflow. Fix the stack usage.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index 2b37a62f6314..619337a5cb2b 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3725,22 +3725,26 @@ static void do_ccupdate_local(void *info)
 static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 				int batchcount, int shared)
 {
-	struct ccupdate_struct new;
+	struct ccupdate_struct *new;
 	int i;
 
-	memset(&new.new, 0, sizeof(new.new));
+	new = kzalloc(sizeof(*new), GFP_KERNEL);
+	if (!new)
+		return -ENOMEM;
+
 	for_each_online_cpu(i) {
-		new.new[i] = alloc_arraycache(cpu_to_node(i), limit,
+		new->new[i] = alloc_arraycache(cpu_to_node(i), limit,
 						batchcount);
-		if (!new.new[i]) {
+		if (!new->new[i]) {
 			for (i--; i >= 0; i--)
-				kfree(new.new[i]);
+				kfree(new->new[i]);
+			kfree(new);
 			return -ENOMEM;
 		}
 	}
-	new.cachep = cachep;
+	new->cachep = cachep;
 
-	on_each_cpu(do_ccupdate_local, (void *)&new, 1, 1);
+	on_each_cpu(do_ccupdate_local, (void *)new, 1, 1);
 
 	check_irq_on();
 	cachep->batchcount = batchcount;
@@ -3748,7 +3752,7 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 	cachep->shared = shared;
 
 	for_each_online_cpu(i) {
-		struct array_cache *ccold = new.new[i];
+		struct array_cache *ccold = new->new[i];
 		if (!ccold)
 			continue;
 		spin_lock_irq(&cachep->nodelists[cpu_to_node(i)]->list_lock);
@@ -3756,7 +3760,7 @@ static int do_tune_cpucache(struct kmem_cache *cachep, int limit,
 		spin_unlock_irq(&cachep->nodelists[cpu_to_node(i)]->list_lock);
 		kfree(ccold);
 	}
-
+	kfree(new);
 	return alloc_kmemlist(cachep);
 }
 
@@ -4274,6 +4278,7 @@ static int leaks_show(struct seq_file *m, void *p)
 		show_symbol(m, n[2*i+2]);
 		seq_putc(m, '\n');
 	}
+
 	return 0;
 }
 
-- 
cgit v1.2.1


From 46a82b2d5591335277ed2930611f6acb4ce654ed Mon Sep 17 00:00:00 2001
From: Dave McCracken <dmccr@us.ibm.com>
Date: Mon, 25 Sep 2006 23:31:48 -0700
Subject: [PATCH] Standardize pxx_page macros

One of the changes necessary for shared page tables is to standardize the
pxx_page macros.  pte_page and pmd_page have always returned the struct
page associated with their entry, while pte_page_kernel and pmd_page_kernel
have returned the kernel virtual address.  pud_page and pgd_page, on the
other hand, return the kernel virtual address.

Shared page tables needs pud_page and pgd_page to return the actual page
structures.  There are very few actual users of these functions, so it is
simple to standardize their usage.

Since this is basic cleanup, I am submitting these changes as a standalone
patch.  Per Hugh Dickins' comments about it, I am also changing the
pxx_page_kernel macros to pxx_page_vaddr to clarify their meaning.

Signed-off-by: Dave McCracken <dmccr@us.ibm.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/arm/mm/ioremap.c               |  2 +-
 arch/sh/mm/cache-sh7705.c           |  2 +-
 arch/sparc/mm/srmmu.c               |  2 +-
 arch/sparc/mm/sun4c.c               |  2 +-
 arch/um/kernel/skas/mmu.c           |  2 +-
 arch/x86_64/mm/fault.c              |  6 +++---
 include/asm-alpha/mmzone.h          |  1 +
 include/asm-alpha/pgtable.h         |  9 +++++----
 include/asm-arm/pgtable.h           |  8 ++++----
 include/asm-arm26/pgtable.h         |  8 ++++----
 include/asm-cris/pgtable.h          |  4 ++--
 include/asm-frv/pgtable.h           |  8 ++++----
 include/asm-generic/4level-fixup.h  |  4 ++++
 include/asm-generic/pgtable-nopmd.h |  2 +-
 include/asm-generic/pgtable-nopud.h |  2 +-
 include/asm-i386/pgtable-3level.h   |  2 +-
 include/asm-i386/pgtable.h          |  4 ++--
 include/asm-ia64/pgtable.h          | 14 ++++++++------
 include/asm-m32r/pgtable-2level.h   |  6 +++++-
 include/asm-m32r/pgtable.h          |  4 ++--
 include/asm-m68k/motorola_pgtable.h |  1 +
 include/asm-mips/pgtable-32.h       |  4 ++--
 include/asm-mips/pgtable-64.h       | 10 ++++++----
 include/asm-mips/pgtable.h          |  2 +-
 include/asm-parisc/pgtable.h        |  9 +++++----
 include/asm-powerpc/pgtable-4k.h    |  5 +++--
 include/asm-powerpc/pgtable.h       | 11 ++++++-----
 include/asm-ppc/pgtable.h           |  8 ++++----
 include/asm-s390/pgtable.h          | 10 ++++++----
 include/asm-sh/pgtable-2level.h     |  5 ++++-
 include/asm-sh/pgtable.h            |  4 ++--
 include/asm-sh64/pgtable.h          |  6 ++++--
 include/asm-sparc/pgtable.h         |  4 ++--
 include/asm-sparc64/pgtable.h       |  5 +++--
 include/asm-um/pgtable-2level.h     |  2 +-
 include/asm-um/pgtable-3level.h     |  5 +++--
 include/asm-um/pgtable.h            |  4 ++--
 include/asm-x86_64/pgtable.h        | 16 ++++++++--------
 include/asm-xtensa/pgtable.h        |  4 ++--
 39 files changed, 117 insertions(+), 90 deletions(-)

diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index 88a999df0ab3..591fc3187c7f 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -177,7 +177,7 @@ static void unmap_area_sections(unsigned long virt, unsigned long size)
 			 * Free the page table, if there was one.
 			 */
 			if ((pmd_val(pmd) & PMD_TYPE_MASK) == PMD_TYPE_TABLE)
-				pte_free_kernel(pmd_page_kernel(pmd));
+				pte_free_kernel(pmd_page_vaddr(pmd));
 		}
 
 		addr += PGDIR_SIZE;
diff --git a/arch/sh/mm/cache-sh7705.c b/arch/sh/mm/cache-sh7705.c
index ad8ed7d41e16..bf94eedb0a8e 100644
--- a/arch/sh/mm/cache-sh7705.c
+++ b/arch/sh/mm/cache-sh7705.c
@@ -30,7 +30,7 @@
 
 #define __pte_offset(address) \
 		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
+#define pte_offset(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
 		__pte_offset(address))
 
 static inline void cache_wback_all(void)
diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index 16e13f663ab0..b27a506309ee 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -2175,7 +2175,7 @@ void __init ld_mmu_srmmu(void)
 
 	BTFIXUPSET_CALL(pte_pfn, srmmu_pte_pfn, BTFIXUPCALL_NORM);
 	BTFIXUPSET_CALL(pmd_page, srmmu_pmd_page, BTFIXUPCALL_NORM);
-	BTFIXUPSET_CALL(pgd_page, srmmu_pgd_page, BTFIXUPCALL_NORM);
+	BTFIXUPSET_CALL(pgd_page_vaddr, srmmu_pgd_page, BTFIXUPCALL_NORM);
 
 	BTFIXUPSET_SETHI(none_mask, 0xF0000000);
 
diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c
index 7fdddf3c7e16..436021ceb2e7 100644
--- a/arch/sparc/mm/sun4c.c
+++ b/arch/sparc/mm/sun4c.c
@@ -2280,5 +2280,5 @@ void __init ld_mmu_sun4c(void)
 
 	/* These should _never_ get called with two level tables. */
 	BTFIXUPSET_CALL(pgd_set, sun4c_pgd_set, BTFIXUPCALL_NOP);
-	BTFIXUPSET_CALL(pgd_page, sun4c_pgd_page, BTFIXUPCALL_RETO0);
+	BTFIXUPSET_CALL(pgd_page_vaddr, sun4c_pgd_page, BTFIXUPCALL_RETO0);
 }
diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c
index 624ca238d1fd..79c22707a637 100644
--- a/arch/um/kernel/skas/mmu.c
+++ b/arch/um/kernel/skas/mmu.c
@@ -55,7 +55,7 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc,
 	 * destroy_context_skas.
 	 */
 
-        mm->context.skas.last_page_table = pmd_page_kernel(*pmd);
+        mm->context.skas.last_page_table = pmd_page_vaddr(*pmd);
 #ifdef CONFIG_3_LEVEL_PGTABLES
         mm->context.skas.last_pmd = (unsigned long) __va(pud_val(*pud));
 #endif
diff --git a/arch/x86_64/mm/fault.c b/arch/x86_64/mm/fault.c
index ac8ea66ccb94..4198798e1469 100644
--- a/arch/x86_64/mm/fault.c
+++ b/arch/x86_64/mm/fault.c
@@ -299,7 +299,7 @@ static int vmalloc_fault(unsigned long address)
 	if (pgd_none(*pgd))
 		set_pgd(pgd, *pgd_ref);
 	else
-		BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+		BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
 
 	/* Below here mismatches are bugs because these lower tables
 	   are shared */
@@ -308,7 +308,7 @@ static int vmalloc_fault(unsigned long address)
 	pud_ref = pud_offset(pgd_ref, address);
 	if (pud_none(*pud_ref))
 		return -1;
-	if (pud_none(*pud) || pud_page(*pud) != pud_page(*pud_ref))
+	if (pud_none(*pud) || pud_page_vaddr(*pud) != pud_page_vaddr(*pud_ref))
 		BUG();
 	pmd = pmd_offset(pud, address);
 	pmd_ref = pmd_offset(pud_ref, address);
@@ -641,7 +641,7 @@ void vmalloc_sync_all(void)
 				if (pgd_none(*pgd))
 					set_pgd(pgd, *pgd_ref);
 				else
-					BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
+					BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
 			}
 			spin_unlock(&pgd_lock);
 			set_bit(pgd_index(address), insync);
diff --git a/include/asm-alpha/mmzone.h b/include/asm-alpha/mmzone.h
index 64d0ab98fcd8..8af56ce346ad 100644
--- a/include/asm-alpha/mmzone.h
+++ b/include/asm-alpha/mmzone.h
@@ -75,6 +75,7 @@ PLAT_NODE_DATA_LOCALNR(unsigned long p, int n)
 #define VALID_PAGE(page)	(((page) - mem_map) < max_mapnr)
 
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> 32))
+#define pgd_page(pgd)		(pfn_to_page(pgd_val(pgd) >> 32))
 #define pte_pfn(pte)		(pte_val(pte) >> 32)
 
 #define mk_pte(page, pgprot)						     \
diff --git a/include/asm-alpha/pgtable.h b/include/asm-alpha/pgtable.h
index 93eaa58b7961..49ac9bee7ced 100644
--- a/include/asm-alpha/pgtable.h
+++ b/include/asm-alpha/pgtable.h
@@ -230,16 +230,17 @@ extern inline void pgd_set(pgd_t * pgdp, pmd_t * pmdp)
 
 
 extern inline unsigned long
-pmd_page_kernel(pmd_t pmd)
+pmd_page_vaddr(pmd_t pmd)
 {
 	return ((pmd_val(pmd) & _PFN_MASK) >> (32-PAGE_SHIFT)) + PAGE_OFFSET;
 }
 
 #ifndef CONFIG_DISCONTIGMEM
 #define pmd_page(pmd)	(mem_map + ((pmd_val(pmd) & _PFN_MASK) >> 32))
+#define pgd_page(pgd)	(mem_map + ((pgd_val(pgd) & _PFN_MASK) >> 32))
 #endif
 
-extern inline unsigned long pgd_page(pgd_t pgd)
+extern inline unsigned long pgd_page_vaddr(pgd_t pgd)
 { return PAGE_OFFSET + ((pgd_val(pgd) & _PFN_MASK) >> (32-PAGE_SHIFT)); }
 
 extern inline int pte_none(pte_t pte)		{ return !pte_val(pte); }
@@ -293,13 +294,13 @@ extern inline pte_t pte_mkyoung(pte_t pte)	{ pte_val(pte) |= __ACCESS_BITS; retu
 /* Find an entry in the second-level page table.. */
 extern inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 {
-	return (pmd_t *) pgd_page(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
+	return (pmd_t *) pgd_page_vaddr(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
 }
 
 /* Find an entry in the third-level page table.. */
 extern inline pte_t * pte_offset_kernel(pmd_t * dir, unsigned long address)
 {
-	return (pte_t *) pmd_page_kernel(*dir)
+	return (pte_t *) pmd_page_vaddr(*dir)
 		+ ((address >> PAGE_SHIFT) & (PTRS_PER_PAGE - 1));
 }
 
diff --git a/include/asm-arm/pgtable.h b/include/asm-arm/pgtable.h
index 8d3919c6458c..4d10d319fa34 100644
--- a/include/asm-arm/pgtable.h
+++ b/include/asm-arm/pgtable.h
@@ -224,9 +224,9 @@ extern struct page *empty_zero_page;
 #define pte_none(pte)		(!pte_val(pte))
 #define pte_clear(mm,addr,ptep)	set_pte_at((mm),(addr),(ptep), __pte(0))
 #define pte_page(pte)		(pfn_to_page(pte_pfn(pte)))
-#define pte_offset_kernel(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map_nested(dir,addr)	(pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_kernel(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map_nested(dir,addr)	(pmd_page_vaddr(*(dir)) + __pte_index(addr))
 #define pte_unmap(pte)		do { } while (0)
 #define pte_unmap_nested(pte)	do { } while (0)
 
@@ -291,7 +291,7 @@ PTE_BIT_FUNC(mkyoung,   |= L_PTE_YOUNG);
 		clean_pmd_entry(pmdp);	\
 	} while (0)
 
-static inline pte_t *pmd_page_kernel(pmd_t pmd)
+static inline pte_t *pmd_page_vaddr(pmd_t pmd)
 {
 	unsigned long ptr;
 
diff --git a/include/asm-arm26/pgtable.h b/include/asm-arm26/pgtable.h
index 19ac9101a6bb..63a8881fae13 100644
--- a/include/asm-arm26/pgtable.h
+++ b/include/asm-arm26/pgtable.h
@@ -186,12 +186,12 @@ extern struct page *empty_zero_page;
  * return a pointer to memory (no special alignment)
  */
 #define pmd_page(pmd)  ((struct page *)(pmd_val((pmd)) & ~_PMD_PRESENT))
-#define pmd_page_kernel(pmd) ((pte_t *)(pmd_val((pmd)) & ~_PMD_PRESENT))
+#define pmd_page_vaddr(pmd) ((pte_t *)(pmd_val((pmd)) & ~_PMD_PRESENT))
 
-#define pte_offset_kernel(dir,addr)     (pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_kernel(dir,addr)     (pmd_page_vaddr(*(dir)) + __pte_index(addr))
 
-#define pte_offset_map(dir,addr)        (pmd_page_kernel(*(dir)) + __pte_index(addr))
-#define pte_offset_map_nested(dir,addr) (pmd_page_kernel(*(dir)) + __pte_index(addr))
+#define pte_offset_map(dir,addr)        (pmd_page_vaddr(*(dir)) + __pte_index(addr))
+#define pte_offset_map_nested(dir,addr) (pmd_page_vaddr(*(dir)) + __pte_index(addr))
 #define pte_unmap(pte)                  do { } while (0)
 #define pte_unmap_nested(pte)           do { } while (0)
 
diff --git a/include/asm-cris/pgtable.h b/include/asm-cris/pgtable.h
index 5d76c1c0d6c9..c94a7107019c 100644
--- a/include/asm-cris/pgtable.h
+++ b/include/asm-cris/pgtable.h
@@ -253,7 +253,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 { pmd_val(*pmdp) = _PAGE_TABLE | (unsigned long) ptep; }
 
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
-#define pmd_page_kernel(pmd)	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd)	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /* to find an entry in a page-table-directory. */
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
@@ -271,7 +271,7 @@ static inline pgd_t * pgd_offset(struct mm_struct * mm, unsigned long address)
 #define __pte_offset(address) \
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 #define pte_offset_map(dir, address) \
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
 #define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
diff --git a/include/asm-frv/pgtable.h b/include/asm-frv/pgtable.h
index 7af7485e889e..2fb3c6f05e03 100644
--- a/include/asm-frv/pgtable.h
+++ b/include/asm-frv/pgtable.h
@@ -217,7 +217,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long address)
 }
 
 #define pgd_page(pgd)				(pud_page((pud_t){ pgd }))
-#define pgd_page_kernel(pgd)			(pud_page_kernel((pud_t){ pgd }))
+#define pgd_page_vaddr(pgd)			(pud_page_vaddr((pud_t){ pgd }))
 
 /*
  * allocating and freeing a pud is trivial: the 1-entry pud is
@@ -246,7 +246,7 @@ static inline void pud_clear(pud_t *pud)	{ }
 #define set_pud(pudptr, pudval)			set_pmd((pmd_t *)(pudptr), (pmd_t) { pudval })
 
 #define pud_page(pud)				(pmd_page((pmd_t){ pud }))
-#define pud_page_kernel(pud)			(pmd_page_kernel((pmd_t){ pud }))
+#define pud_page_vaddr(pud)			(pmd_page_vaddr((pmd_t){ pud }))
 
 /*
  * (pmds are folded into pgds so this doesn't get actually called,
@@ -362,7 +362,7 @@ static inline pmd_t *pmd_offset(pud_t *dir, unsigned long address)
 #define	pmd_bad(x)	(pmd_val(x) & xAMPRx_SS)
 #define pmd_clear(xp)	do { __set_pmd(xp, 0); } while(0)
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -458,7 +458,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 
 #if defined(CONFIG_HIGHPTE)
 #define pte_offset_map(dir, address) \
diff --git a/include/asm-generic/4level-fixup.h b/include/asm-generic/4level-fixup.h
index 68c6fea994d9..7b88d3931e34 100644
--- a/include/asm-generic/4level-fixup.h
+++ b/include/asm-generic/4level-fixup.h
@@ -21,6 +21,10 @@
 #define pud_present(pud)		1
 #define pud_ERROR(pud)			do { } while (0)
 #define pud_clear(pud)			pgd_clear(pud)
+#define pud_val(pud)			pgd_val(pud)
+#define pud_populate(mm, pud, pmd)	pgd_populate(mm, pud, pmd)
+#define pud_page(pud)			pgd_page(pud)
+#define pud_page_vaddr(pud)		pgd_page_vaddr(pud)
 
 #undef pud_free_tlb
 #define pud_free_tlb(tlb, x)            do { } while (0)
diff --git a/include/asm-generic/pgtable-nopmd.h b/include/asm-generic/pgtable-nopmd.h
index c8d53ba20e19..29ff5d84d8c3 100644
--- a/include/asm-generic/pgtable-nopmd.h
+++ b/include/asm-generic/pgtable-nopmd.h
@@ -47,7 +47,7 @@ static inline pmd_t * pmd_offset(pud_t * pud, unsigned long address)
 #define __pmd(x)				((pmd_t) { __pud(x) } )
 
 #define pud_page(pud)				(pmd_page((pmd_t){ pud }))
-#define pud_page_kernel(pud)			(pmd_page_kernel((pmd_t){ pud }))
+#define pud_page_vaddr(pud)			(pmd_page_vaddr((pmd_t){ pud }))
 
 /*
  * allocating and freeing a pmd is trivial: the 1-entry pmd is
diff --git a/include/asm-generic/pgtable-nopud.h b/include/asm-generic/pgtable-nopud.h
index 82e29f0ce467..566464500558 100644
--- a/include/asm-generic/pgtable-nopud.h
+++ b/include/asm-generic/pgtable-nopud.h
@@ -44,7 +44,7 @@ static inline pud_t * pud_offset(pgd_t * pgd, unsigned long address)
 #define __pud(x)				((pud_t) { __pgd(x) } )
 
 #define pgd_page(pgd)				(pud_page((pud_t){ pgd }))
-#define pgd_page_kernel(pgd)			(pud_page_kernel((pud_t){ pgd }))
+#define pgd_page_vaddr(pgd)			(pud_page_vaddr((pud_t){ pgd }))
 
 /*
  * allocating and freeing a pud is trivial: the 1-entry pud is
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index dccb1b3337ad..807ed9e366de 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -77,7 +77,7 @@ static inline void pud_clear (pud_t * pud) { }
 #define pud_page(pud) \
 ((struct page *) __va(pud_val(pud) & PAGE_MASK))
 
-#define pud_page_kernel(pud) \
+#define pud_page_vaddr(pud) \
 ((unsigned long) __va(pud_val(pud) & PAGE_MASK))
 
 
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 09697fec3d2b..c04b3d0f484e 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -364,11 +364,11 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 
 #define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 		((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
diff --git a/include/asm-ia64/pgtable.h b/include/asm-ia64/pgtable.h
index 228981cadf8f..553182747722 100644
--- a/include/asm-ia64/pgtable.h
+++ b/include/asm-ia64/pgtable.h
@@ -275,21 +275,23 @@ ia64_phys_addr_valid (unsigned long addr)
 #define pmd_bad(pmd)			(!ia64_phys_addr_valid(pmd_val(pmd)))
 #define pmd_present(pmd)		(pmd_val(pmd) != 0UL)
 #define pmd_clear(pmdp)			(pmd_val(*(pmdp)) = 0UL)
-#define pmd_page_kernel(pmd)		((unsigned long) __va(pmd_val(pmd) & _PFN_MASK))
+#define pmd_page_vaddr(pmd)		((unsigned long) __va(pmd_val(pmd) & _PFN_MASK))
 #define pmd_page(pmd)			virt_to_page((pmd_val(pmd) + PAGE_OFFSET))
 
 #define pud_none(pud)			(!pud_val(pud))
 #define pud_bad(pud)			(!ia64_phys_addr_valid(pud_val(pud)))
 #define pud_present(pud)		(pud_val(pud) != 0UL)
 #define pud_clear(pudp)			(pud_val(*(pudp)) = 0UL)
-#define pud_page(pud)			((unsigned long) __va(pud_val(pud) & _PFN_MASK))
+#define pud_page_vaddr(pud)		((unsigned long) __va(pud_val(pud) & _PFN_MASK))
+#define pud_page(pud)			virt_to_page((pud_val(pud) + PAGE_OFFSET))
 
 #ifdef CONFIG_PGTABLE_4
 #define pgd_none(pgd)			(!pgd_val(pgd))
 #define pgd_bad(pgd)			(!ia64_phys_addr_valid(pgd_val(pgd)))
 #define pgd_present(pgd)		(pgd_val(pgd) != 0UL)
 #define pgd_clear(pgdp)			(pgd_val(*(pgdp)) = 0UL)
-#define pgd_page(pgd)			((unsigned long) __va(pgd_val(pgd) & _PFN_MASK))
+#define pgd_page_vaddr(pgd)		((unsigned long) __va(pgd_val(pgd) & _PFN_MASK))
+#define pgd_page(pgd)			virt_to_page((pgd_val(pgd) + PAGE_OFFSET))
 #endif
 
 /*
@@ -360,19 +362,19 @@ pgd_offset (struct mm_struct *mm, unsigned long address)
 #ifdef CONFIG_PGTABLE_4
 /* Find an entry in the second-level page table.. */
 #define pud_offset(dir,addr) \
-	((pud_t *) pgd_page(*(dir)) + (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
+	((pud_t *) pgd_page_vaddr(*(dir)) + (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
 #endif
 
 /* Find an entry in the third-level page table.. */
 #define pmd_offset(dir,addr) \
-	((pmd_t *) pud_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+	((pmd_t *) pud_page_vaddr(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
 
 /*
  * Find an entry in the third-level page table.  This looks more complicated than it
  * should be because some platforms place page tables in high memory.
  */
 #define pte_index(addr)	 	(((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir,addr)	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(addr))
+#define pte_offset_kernel(dir,addr)	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir,addr)	pte_offset_kernel(dir, addr)
 #define pte_offset_map_nested(dir,addr)	pte_offset_map(dir, addr)
 #define pte_unmap(pte)			do { } while (0)
diff --git a/include/asm-m32r/pgtable-2level.h b/include/asm-m32r/pgtable-2level.h
index be0f167e344a..6a674e3d37a2 100644
--- a/include/asm-m32r/pgtable-2level.h
+++ b/include/asm-m32r/pgtable-2level.h
@@ -52,9 +52,13 @@ static inline int pgd_present(pgd_t pgd)	{ return 1; }
 #define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
 #define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
+#ifndef CONFIG_DISCONTIGMEM
+#define pgd_page(pgd)	(mem_map + ((pgd_val(pgd) >> PAGE_SHIFT) - PFN_BASE))
+#endif /* !CONFIG_DISCONTIGMEM */
+
 static inline pmd_t *pmd_offset(pgd_t * dir, unsigned long address)
 {
 	return (pmd_t *) dir;
diff --git a/include/asm-m32r/pgtable.h b/include/asm-m32r/pgtable.h
index 1983b7f4527a..1c15ba7ce319 100644
--- a/include/asm-m32r/pgtable.h
+++ b/include/asm-m32r/pgtable.h
@@ -336,7 +336,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 	pmd_val(*pmdp) = (((unsigned long) ptep) & PAGE_MASK);
 }
 
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -358,7 +358,7 @@ static inline void pmd_set(pmd_t * pmdp, pte_t * ptep)
 #define pte_index(address)	\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address)	\
-	((pte_t *)pmd_page_kernel(*(dir)) + pte_index(address))
+	((pte_t *)pmd_page_vaddr(*(dir)) + pte_index(address))
 #define pte_offset_map(dir, address)	\
 	((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
 #define pte_offset_map_nested(dir, address)	pte_offset_map(dir, address)
diff --git a/include/asm-m68k/motorola_pgtable.h b/include/asm-m68k/motorola_pgtable.h
index 1ccc7338a54b..61e4406ed96a 100644
--- a/include/asm-m68k/motorola_pgtable.h
+++ b/include/asm-m68k/motorola_pgtable.h
@@ -150,6 +150,7 @@ static inline void pgd_set(pgd_t *pgdp, pmd_t *pmdp)
 #define pgd_bad(pgd)		((pgd_val(pgd) & _DESCTYPE_MASK) != _PAGE_TABLE)
 #define pgd_present(pgd)	(pgd_val(pgd) & _PAGE_TABLE)
 #define pgd_clear(pgdp)		({ pgd_val(*pgdp) = 0; })
+#define pgd_page(pgd)		(mem_map + ((unsigned long)(__va(pgd_val(pgd)) - PAGE_OFFSET) >> PAGE_SHIFT))
 
 #define pte_ERROR(e) \
 	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
diff --git a/include/asm-mips/pgtable-32.h b/include/asm-mips/pgtable-32.h
index 4b26d8528133..d20f2e9b28be 100644
--- a/include/asm-mips/pgtable-32.h
+++ b/include/asm-mips/pgtable-32.h
@@ -156,9 +156,9 @@ pfn_pte(unsigned long pfn, pgprot_t prot)
 #define __pte_offset(address)						\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset(dir, address)					\
-	((pte_t *) (pmd_page_kernel(*dir)) + __pte_offset(address))
+	((pte_t *) (pmd_page_vaddr(*dir)) + __pte_offset(address))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 
 #define pte_offset_map(dir, address)                                    \
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h
index e3db93212eab..c59a1e21f5b0 100644
--- a/include/asm-mips/pgtable-64.h
+++ b/include/asm-mips/pgtable-64.h
@@ -178,24 +178,26 @@ static inline void pud_clear(pud_t *pudp)
 /* to find an entry in a page-table-directory */
 #define pgd_offset(mm,addr)	((mm)->pgd + pgd_index(addr))
 
-static inline unsigned long pud_page(pud_t pud)
+static inline unsigned long pud_page_vaddr(pud_t pud)
 {
 	return pud_val(pud);
 }
+#define pud_phys(pud)		(pud_val(pud) - PAGE_OFFSET)
+#define pud_page(pud)		(pfn_to_page(pud_phys(pud) >> PAGE_SHIFT))
 
 /* Find an entry in the second-level page table.. */
 static inline pmd_t *pmd_offset(pud_t * pud, unsigned long address)
 {
-	return (pmd_t *) pud_page(*pud) + pmd_index(address);
+	return (pmd_t *) pud_page_vaddr(*pud) + pmd_index(address);
 }
 
 /* Find an entry in the third-level page table.. */
 #define __pte_offset(address)						\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset(dir, address)					\
-	((pte_t *) (pmd_page_kernel(*dir)) + __pte_offset(address))
+	((pte_t *) (pmd_page_vaddr(*dir)) + __pte_offset(address))
 #define pte_offset_kernel(dir, address)					\
-	((pte_t *) pmd_page_kernel(*(dir)) +  __pte_offset(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  __pte_offset(address))
 #define pte_offset_map(dir, address)					\
 	((pte_t *)page_address(pmd_page(*(dir))) + __pte_offset(address))
 #define pte_offset_map_nested(dir, address)				\
diff --git a/include/asm-mips/pgtable.h b/include/asm-mips/pgtable.h
index a36ca1be17f2..1ca4d1e185c7 100644
--- a/include/asm-mips/pgtable.h
+++ b/include/asm-mips/pgtable.h
@@ -87,7 +87,7 @@ extern void paging_init(void);
  */
 #define pmd_phys(pmd)		(pmd_val(pmd) - PAGE_OFFSET)
 #define pmd_page(pmd)		(pfn_to_page(pmd_phys(pmd) >> PAGE_SHIFT))
-#define pmd_page_kernel(pmd)	pmd_val(pmd)
+#define pmd_page_vaddr(pmd)	pmd_val(pmd)
 
 #if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32_R1)
 
diff --git a/include/asm-parisc/pgtable.h b/include/asm-parisc/pgtable.h
index 5066c54dae0a..c0b61e0d1497 100644
--- a/include/asm-parisc/pgtable.h
+++ b/include/asm-parisc/pgtable.h
@@ -303,7 +303,8 @@ static inline void pmd_clear(pmd_t *pmd) {
 
 
 #if PT_NLEVELS == 3
-#define pgd_page(pgd) ((unsigned long) __va(pgd_address(pgd)))
+#define pgd_page_vaddr(pgd) ((unsigned long) __va(pgd_address(pgd)))
+#define pgd_page(pgd)	virt_to_page((void *)pgd_page_vaddr(pgd))
 
 /* For 64 bit we have three level tables */
 
@@ -382,7 +383,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #define pte_page(pte)		(pfn_to_page(pte_pfn(pte)))
 
-#define pmd_page_kernel(pmd)	((unsigned long) __va(pmd_address(pmd)))
+#define pmd_page_vaddr(pmd)	((unsigned long) __va(pmd_address(pmd)))
 
 #define __pmd_page(pmd) ((unsigned long) __va(pmd_address(pmd)))
 #define pmd_page(pmd)	virt_to_page((void *)__pmd_page(pmd))
@@ -400,7 +401,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #if PT_NLEVELS == 3
 #define pmd_offset(dir,address) \
-((pmd_t *) pgd_page(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
+((pmd_t *) pgd_page_vaddr(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
 #else
 #define pmd_offset(dir,addr) ((pmd_t *) dir)
 #endif
@@ -408,7 +409,7 @@ extern inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 /* Find an entry in the third-level page table.. */ 
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
 #define pte_offset_kernel(pmd, address) \
-	((pte_t *) pmd_page_kernel(*(pmd)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(pmd)) + pte_index(address))
 #define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_offset_map_nested(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_unmap(pte) do { } while (0)
diff --git a/include/asm-powerpc/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h
index e7036155672e..345d9b07b3e2 100644
--- a/include/asm-powerpc/pgtable-4k.h
+++ b/include/asm-powerpc/pgtable-4k.h
@@ -88,10 +88,11 @@
 #define pgd_bad(pgd)		(pgd_val(pgd) == 0)
 #define pgd_present(pgd)	(pgd_val(pgd) != 0)
 #define pgd_clear(pgdp)		(pgd_val(*(pgdp)) = 0)
-#define pgd_page(pgd)		(pgd_val(pgd) & ~PGD_MASKED_BITS)
+#define pgd_page_vaddr(pgd)	(pgd_val(pgd) & ~PGD_MASKED_BITS)
+#define pgd_page(pgd)		virt_to_page(pgd_page_vaddr(pgd))
 
 #define pud_offset(pgdp, addr)	\
-  (((pud_t *) pgd_page(*(pgdp))) + \
+  (((pud_t *) pgd_page_vaddr(*(pgdp))) + \
     (((addr) >> PUD_SHIFT) & (PTRS_PER_PUD - 1)))
 
 #define pud_ERROR(e) \
diff --git a/include/asm-powerpc/pgtable.h b/include/asm-powerpc/pgtable.h
index 8dbf5ad8150f..10f52743f4ff 100644
--- a/include/asm-powerpc/pgtable.h
+++ b/include/asm-powerpc/pgtable.h
@@ -196,8 +196,8 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 				 || (pmd_val(pmd) & PMD_BAD_BITS))
 #define	pmd_present(pmd)	(pmd_val(pmd) != 0)
 #define	pmd_clear(pmdp)		(pmd_val(*(pmdp)) = 0)
-#define pmd_page_kernel(pmd)	(pmd_val(pmd) & ~PMD_MASKED_BITS)
-#define pmd_page(pmd)		virt_to_page(pmd_page_kernel(pmd))
+#define pmd_page_vaddr(pmd)	(pmd_val(pmd) & ~PMD_MASKED_BITS)
+#define pmd_page(pmd)		virt_to_page(pmd_page_vaddr(pmd))
 
 #define pud_set(pudp, pudval)	(pud_val(*(pudp)) = (pudval))
 #define pud_none(pud)		(!pud_val(pud))
@@ -205,7 +205,8 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 				 || (pud_val(pud) & PUD_BAD_BITS))
 #define pud_present(pud)	(pud_val(pud) != 0)
 #define pud_clear(pudp)		(pud_val(*(pudp)) = 0)
-#define pud_page(pud)		(pud_val(pud) & ~PUD_MASKED_BITS)
+#define pud_page_vaddr(pud)	(pud_val(pud) & ~PUD_MASKED_BITS)
+#define pud_page(pud)		virt_to_page(pud_page_vaddr(pud))
 
 #define pgd_set(pgdp, pudp)	({pgd_val(*(pgdp)) = (unsigned long)(pudp);})
 
@@ -219,10 +220,10 @@ static inline pte_t pfn_pte(unsigned long pfn, pgprot_t pgprot)
 #define pgd_offset(mm, address)	 ((mm)->pgd + pgd_index(address))
 
 #define pmd_offset(pudp,addr) \
-  (((pmd_t *) pud_page(*(pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+  (((pmd_t *) pud_page_vaddr(*(pudp))) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
 
 #define pte_offset_kernel(dir,addr) \
-  (((pte_t *) pmd_page_kernel(*(dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))
+  (((pte_t *) pmd_page_vaddr(*(dir))) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))
 
 #define pte_offset_map(dir,addr)	pte_offset_kernel((dir), (addr))
 #define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir), (addr))
diff --git a/include/asm-ppc/pgtable.h b/include/asm-ppc/pgtable.h
index 51fa7c662917..b1fdbf40dba2 100644
--- a/include/asm-ppc/pgtable.h
+++ b/include/asm-ppc/pgtable.h
@@ -526,7 +526,7 @@ static inline int pgd_bad(pgd_t pgd)		{ return 0; }
 static inline int pgd_present(pgd_t pgd)	{ return 1; }
 #define pgd_clear(xp)				do { } while (0)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 	((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
 /*
@@ -720,12 +720,12 @@ extern pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
  * of the pte page.  -- paulus
  */
 #ifndef CONFIG_BOOKE
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd)		\
 	(mem_map + (pmd_val(pmd) >> PAGE_SHIFT))
 #else
-#define pmd_page_kernel(pmd)	\
+#define pmd_page_vaddr(pmd)	\
 	((unsigned long) (pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd)		\
 	(mem_map + (__pa(pmd_val(pmd)) >> PAGE_SHIFT))
@@ -748,7 +748,7 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 #define pte_index(address)		\
 	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, addr)	\
-	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(addr))
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir, addr)		\
 	((pte_t *) kmap_atomic(pmd_page(*(dir)), KM_PTE0) + pte_index(addr))
 #define pte_offset_map_nested(dir, addr)	\
diff --git a/include/asm-s390/pgtable.h b/include/asm-s390/pgtable.h
index 1a07028d575e..e965309fedac 100644
--- a/include/asm-s390/pgtable.h
+++ b/include/asm-s390/pgtable.h
@@ -664,11 +664,13 @@ static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
 #define pte_pfn(x) (pte_val(x) >> PAGE_SHIFT)
 #define pte_page(x) pfn_to_page(pte_pfn(x))
 
-#define pmd_page_kernel(pmd) (pmd_val(pmd) & PAGE_MASK)
+#define pmd_page_vaddr(pmd) (pmd_val(pmd) & PAGE_MASK)
 
 #define pmd_page(pmd) (mem_map+(pmd_val(pmd) >> PAGE_SHIFT))
 
-#define pgd_page_kernel(pgd) (pgd_val(pgd) & PAGE_MASK)
+#define pgd_page_vaddr(pgd) (pgd_val(pgd) & PAGE_MASK)
+
+#define pgd_page(pgd) (mem_map+(pgd_val(pgd) >> PAGE_SHIFT))
 
 /* to find an entry in a page-table-directory */
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
@@ -690,14 +692,14 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 /* Find an entry in the second-level page table.. */
 #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
 #define pmd_offset(dir,addr) \
-	((pmd_t *) pgd_page_kernel(*(dir)) + pmd_index(addr))
+	((pmd_t *) pgd_page_vaddr(*(dir)) + pmd_index(addr))
 
 #endif /* __s390x__ */
 
 /* Find an entry in the third-level page table.. */
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
 #define pte_offset_kernel(pmd, address) \
-	((pte_t *) pmd_page_kernel(*(pmd)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(pmd)) + pte_index(address))
 #define pte_offset_map(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_offset_map_nested(pmd, address) pte_offset_kernel(pmd, address)
 #define pte_unmap(pte) do { } while (0)
diff --git a/include/asm-sh/pgtable-2level.h b/include/asm-sh/pgtable-2level.h
index b0528aa3cb1f..b525db6f61c6 100644
--- a/include/asm-sh/pgtable-2level.h
+++ b/include/asm-sh/pgtable-2level.h
@@ -50,9 +50,12 @@ static inline void pgd_clear (pgd_t * pgdp) 	{ }
 #define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
 #define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)
 
-#define pgd_page(pgd) \
+#define pgd_page_vaddr(pgd) \
 ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
 
+#define pgd_page(pgd) \
+	(phys_to_page(pgd_val(pgd)))
+
 static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 {
 	return (pmd_t *) dir;
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index dcd23a03683d..40d41a78041e 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -225,7 +225,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t _prot)
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 { set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK) | pgprot_val(newprot))); return pte; }
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 #define pmd_page(pmd) \
@@ -242,7 +242,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_index(address) \
 		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) + pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
 #define pte_offset_map(dir, address) pte_offset_kernel(dir, address)
 #define pte_offset_map_nested(dir, address) pte_offset_kernel(dir, address)
 #define pte_unmap(pte)		do { } while (0)
diff --git a/include/asm-sh64/pgtable.h b/include/asm-sh64/pgtable.h
index 54c7821893f5..6b97c4cb1d64 100644
--- a/include/asm-sh64/pgtable.h
+++ b/include/asm-sh64/pgtable.h
@@ -190,7 +190,9 @@ static inline int pgd_bad(pgd_t pgd)		{ return 0; }
 #endif
 
 
-#define pgd_page(pgd_entry)	((unsigned long) (pgd_val(pgd_entry) & PAGE_MASK))
+#define pgd_page_vaddr(pgd_entry)	((unsigned long) (pgd_val(pgd_entry) & PAGE_MASK))
+#define pgd_page(pgd)	(virt_to_page(pgd_val(pgd)))
+
 
 /*
  * PMD defines. Middle level.
@@ -219,7 +221,7 @@ static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
 #define pmd_none(pmd_entry)	(pmd_val((pmd_entry)) == _PMD_EMPTY)
 #define pmd_bad(pmd_entry)	((pmd_val(pmd_entry) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
 
-#define pmd_page_kernel(pmd_entry) \
+#define pmd_page_vaddr(pmd_entry) \
 	((unsigned long) __va(pmd_val(pmd_entry) & PAGE_MASK))
 
 #define pmd_page(pmd) \
diff --git a/include/asm-sparc/pgtable.h b/include/asm-sparc/pgtable.h
index 226c6475c9a2..4f0a5ba0d6a0 100644
--- a/include/asm-sparc/pgtable.h
+++ b/include/asm-sparc/pgtable.h
@@ -143,10 +143,10 @@ extern unsigned long empty_zero_page;
 /*
  */
 BTFIXUPDEF_CALL_CONST(struct page *, pmd_page, pmd_t)
-BTFIXUPDEF_CALL_CONST(unsigned long, pgd_page, pgd_t)
+BTFIXUPDEF_CALL_CONST(unsigned long, pgd_page_vaddr, pgd_t)
 
 #define pmd_page(pmd) BTFIXUP_CALL(pmd_page)(pmd)
-#define pgd_page(pgd) BTFIXUP_CALL(pgd_page)(pgd)
+#define pgd_page_vaddr(pgd) BTFIXUP_CALL(pgd_page_vaddr)(pgd)
 
 BTFIXUPDEF_SETHI(none_mask)
 BTFIXUPDEF_CALL_CONST(int, pte_present, pte_t)
diff --git a/include/asm-sparc64/pgtable.h b/include/asm-sparc64/pgtable.h
index ebfe395cfb87..b12be7a869f6 100644
--- a/include/asm-sparc64/pgtable.h
+++ b/include/asm-sparc64/pgtable.h
@@ -630,8 +630,9 @@ static inline unsigned long pte_present(pte_t pte)
 #define __pmd_page(pmd)		\
 	((unsigned long) __va((((unsigned long)pmd_val(pmd))<<11UL)))
 #define pmd_page(pmd) 			virt_to_page((void *)__pmd_page(pmd))
-#define pud_page(pud)		\
+#define pud_page_vaddr(pud)		\
 	((unsigned long) __va((((unsigned long)pud_val(pud))<<11UL)))
+#define pud_page(pud) 			virt_to_page((void *)pud_page_vaddr(pud))
 #define pmd_none(pmd)			(!pmd_val(pmd))
 #define pmd_bad(pmd)			(0)
 #define pmd_present(pmd)		(pmd_val(pmd) != 0U)
@@ -653,7 +654,7 @@ static inline unsigned long pte_present(pte_t pte)
 
 /* Find an entry in the second-level page table.. */
 #define pmd_offset(pudp, address)	\
-	((pmd_t *) pud_page(*(pudp)) + \
+	((pmd_t *) pud_page_vaddr(*(pudp)) + \
 	 (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1)))
 
 /* Find an entry in the third-level page table.. */
diff --git a/include/asm-um/pgtable-2level.h b/include/asm-um/pgtable-2level.h
index ffe017f6b64b..6050e0eb257e 100644
--- a/include/asm-um/pgtable-2level.h
+++ b/include/asm-um/pgtable-2level.h
@@ -41,7 +41,7 @@ static inline void pgd_mkuptodate(pgd_t pgd)	{ }
 #define pfn_pte(pfn, prot) __pte(pfn_to_phys(pfn) | pgprot_val(prot))
 #define pfn_pmd(pfn, prot) __pmd(pfn_to_phys(pfn) | pgprot_val(prot))
 
-#define pmd_page_kernel(pmd) \
+#define pmd_page_vaddr(pmd) \
 	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
diff --git a/include/asm-um/pgtable-3level.h b/include/asm-um/pgtable-3level.h
index 786c25727289..ca0c2a92a112 100644
--- a/include/asm-um/pgtable-3level.h
+++ b/include/asm-um/pgtable-3level.h
@@ -74,11 +74,12 @@ extern inline void pud_clear (pud_t *pud)
         set_pud(pud, __pud(0));
 }
 
-#define pud_page(pud) \
+#define pud_page(pud) phys_to_page(pud_val(pud) & PAGE_MASK)
+#define pud_page_vaddr(pud) \
 	((struct page *) __va(pud_val(pud) & PAGE_MASK))
 
 /* Find an entry in the second-level page table.. */
-#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \
+#define pmd_offset(pud, address) ((pmd_t *) pud_page_vaddr(*(pud)) + \
 			pmd_index(address))
 
 static inline unsigned long pte_pfn(pte_t pte)
diff --git a/include/asm-um/pgtable.h b/include/asm-um/pgtable.h
index ac64eb955868..4862daf8b906 100644
--- a/include/asm-um/pgtable.h
+++ b/include/asm-um/pgtable.h
@@ -349,7 +349,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 	return pte; 
 }
 
-#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
 
 /*
  * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
@@ -389,7 +389,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
  */
 #define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir, address) \
-	((pte_t *) pmd_page_kernel(*(dir)) +  pte_index(address))
+	((pte_t *) pmd_page_vaddr(*(dir)) +  pte_index(address))
 #define pte_offset_map(dir, address) \
 	((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
 #define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index a31ab4e68a9b..51eba2395171 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -101,9 +101,6 @@ static inline void pgd_clear (pgd_t * pgd)
 	set_pgd(pgd, __pgd(0));
 }
 
-#define pud_page(pud) \
-((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
-
 #define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte, 0))
 
 struct mm_struct;
@@ -326,7 +323,8 @@ static inline int pmd_large(pmd_t pte) {
 /*
  * Level 4 access.
  */
-#define pgd_page(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
+#define pgd_page_vaddr(pgd) ((unsigned long) __va((unsigned long)pgd_val(pgd) & PTE_MASK))
+#define pgd_page(pgd)		(pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT))
 #define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
 #define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
 #define pgd_offset_k(address) (init_level4_pgt + pgd_index(address))
@@ -335,16 +333,18 @@ static inline int pmd_large(pmd_t pte) {
 
 /* PUD - Level3 access */
 /* to find an entry in a page-table-directory. */
+#define pud_page_vaddr(pud) ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
+#define pud_page(pud)		(pfn_to_page(pud_val(pud) >> PAGE_SHIFT))
 #define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
-#define pud_offset(pgd, address) ((pud_t *) pgd_page(*(pgd)) + pud_index(address))
+#define pud_offset(pgd, address) ((pud_t *) pgd_page_vaddr(*(pgd)) + pud_index(address))
 #define pud_present(pud) (pud_val(pud) & _PAGE_PRESENT)
 
 /* PMD  - Level 2 access */
-#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
 #define pmd_page(pmd)		(pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
 
 #define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
-#define pmd_offset(dir, address) ((pmd_t *) pud_page(*(dir)) + \
+#define pmd_offset(dir, address) ((pmd_t *) pud_page_vaddr(*(dir)) + \
 			pmd_index(address))
 #define pmd_none(x)	(!pmd_val(x))
 #define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
@@ -382,7 +382,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 
 #define pte_index(address) \
 		(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
+#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
 			pte_index(address))
 
 /* x86-64 always has all page tables mapped. */
diff --git a/include/asm-xtensa/pgtable.h b/include/asm-xtensa/pgtable.h
index 7b15afb70c56..a47cc734c20c 100644
--- a/include/asm-xtensa/pgtable.h
+++ b/include/asm-xtensa/pgtable.h
@@ -218,7 +218,7 @@ extern pgd_t swapper_pg_dir[PAGE_SIZE/sizeof(pgd_t)];
 /*
  * The pmd contains the kernel virtual address of the pte page.
  */
-#define pmd_page_kernel(pmd) ((unsigned long)(pmd_val(pmd) & PAGE_MASK))
+#define pmd_page_vaddr(pmd) ((unsigned long)(pmd_val(pmd) & PAGE_MASK))
 #define pmd_page(pmd) virt_to_page(pmd_val(pmd))
 
 /*
@@ -349,7 +349,7 @@ ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 /* Find an entry in the third-level page table.. */
 #define pte_index(address)	(((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
 #define pte_offset_kernel(dir,addr) 					\
-	((pte_t*) pmd_page_kernel(*(dir)) + pte_index(addr))
+	((pte_t*) pmd_page_vaddr(*(dir)) + pte_index(addr))
 #define pte_offset_map(dir,addr)	pte_offset_kernel((dir),(addr))
 #define pte_offset_map_nested(dir,addr)	pte_offset_kernel((dir),(addr))
 
-- 
cgit v1.2.1


From 006d22d9bbb7e66279ba5cc4556b54eeaf8fd556 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:48 -0700
Subject: [PATCH] Optimize free_one_page

Free one_page currently adds the page to a fake list and calls
free_page_bulk.  Fee_page_bulk takes it off again and then calles
__free_one_page.

Make free_one_page go directly to __free_one_page.  Saves list on / off and
a temporary list in free_one_page for higher ordered pages.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e8a71657ac4a..cc6483047567 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -448,9 +448,11 @@ static void free_pages_bulk(struct zone *zone, int count,
 
 static void free_one_page(struct zone *zone, struct page *page, int order)
 {
-	LIST_HEAD(list);
-	list_add(&page->lru, &list);
-	free_pages_bulk(zone, 1, &list, order);
+	spin_lock(&zone->lock);
+	zone->all_unreclaimable = 0;
+	zone->pages_scanned = 0;
+	__free_one_page(page, zone ,order);
+	spin_unlock(&zone->lock);
 }
 
 static void __free_pages_ok(struct page *page, unsigned int order)
-- 
cgit v1.2.1


From 39bbcb8f88154c4ac9853baf3f1134af4c987517 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:49 -0700
Subject: [PATCH] mm: do not check unpopulated zones for draining and counter
 updates

If a zone is unpopulated then we do not need to check for pages that are to
be drained and also not for vm counters that may need to be updated.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 5 ++++-
 mm/vmstat.c     | 3 +++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cc6483047567..f7ea020c23ea 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -633,7 +633,7 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
 #ifdef CONFIG_NUMA
 /*
  * Called from the slab reaper to drain pagesets on a particular node that
- * belong to the currently executing processor.
+ * belongs to the currently executing processor.
  * Note that this function must be called with the thread pinned to
  * a single processor.
  */
@@ -647,6 +647,9 @@ void drain_node_pages(int nodeid)
 		struct zone *zone = NODE_DATA(nodeid)->node_zones + z;
 		struct per_cpu_pageset *pset;
 
+		if (!populated_zone(zone))
+			continue;
+
 		pset = zone_pcp(zone, smp_processor_id());
 		for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) {
 			struct per_cpu_pages *pcp;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 04a9093f649e..968c0072e19a 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -321,6 +321,9 @@ void refresh_cpu_vm_stats(int cpu)
 	for_each_zone(zone) {
 		struct per_cpu_pageset *pcp;
 
+		if (!populated_zone(zone))
+			continue;
+
 		pcp = zone_pcp(zone, cpu);
 
 		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
-- 
cgit v1.2.1


From d00bcc98d7ec2c87391c9d9e1cca519ef64d33ef Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:50 -0700
Subject: [PATCH] Extract the allocpercpu functions from the slab allocator

The allocpercpu functions __alloc_percpu and __free_percpu() are heavily
using the slab allocator.  However, they are conceptually slab.  This also
simplifies SLOB (at this point slob may be broken in mm.  This should fix
it).

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/Makefile      |   2 +-
 mm/allocpercpu.c | 129 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 mm/slab.c        | 124 ----------------------------------------------------
 mm/slob.c        |  45 -------------------
 4 files changed, 130 insertions(+), 170 deletions(-)
 create mode 100644 mm/allocpercpu.c

diff --git a/mm/Makefile b/mm/Makefile
index 9dd824c11eeb..60c56c0b5e10 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -23,4 +23,4 @@ obj-$(CONFIG_SLAB) += slab.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
 obj-$(CONFIG_MIGRATION) += migrate.o
-
+obj-$(CONFIG_SMP) += allocpercpu.o
diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c
new file mode 100644
index 000000000000..eaa9abeea536
--- /dev/null
+++ b/mm/allocpercpu.c
@@ -0,0 +1,129 @@
+/*
+ * linux/mm/allocpercpu.c
+ *
+ * Separated from slab.c August 11, 2006 Christoph Lameter <clameter@sgi.com>
+ */
+#include <linux/mm.h>
+#include <linux/module.h>
+
+/**
+ * percpu_depopulate - depopulate per-cpu data for given cpu
+ * @__pdata: per-cpu data to depopulate
+ * @cpu: depopulate per-cpu data for this cpu
+ *
+ * Depopulating per-cpu data for a cpu going offline would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
+ */
+void percpu_depopulate(void *__pdata, int cpu)
+{
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	if (pdata->ptrs[cpu]) {
+		kfree(pdata->ptrs[cpu]);
+		pdata->ptrs[cpu] = NULL;
+	}
+}
+EXPORT_SYMBOL_GPL(percpu_depopulate);
+
+/**
+ * percpu_depopulate_mask - depopulate per-cpu data for some cpu's
+ * @__pdata: per-cpu data to depopulate
+ * @mask: depopulate per-cpu data for cpu's selected through mask bits
+ */
+void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+{
+	int cpu;
+	for_each_cpu_mask(cpu, *mask)
+		percpu_depopulate(__pdata, cpu);
+}
+EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
+
+/**
+ * percpu_populate - populate per-cpu data for given cpu
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @cpu: populate per-data for this cpu
+ *
+ * Populating per-cpu data for a cpu coming online would be a typical
+ * use case. You need to register a cpu hotplug handler for that purpose.
+ * Per-cpu object is populated with zeroed buffer.
+ */
+void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
+{
+	struct percpu_data *pdata = __percpu_disguise(__pdata);
+	int node = cpu_to_node(cpu);
+
+	BUG_ON(pdata->ptrs[cpu]);
+	if (node_online(node)) {
+		/* FIXME: kzalloc_node(size, gfp, node) */
+		pdata->ptrs[cpu] = kmalloc_node(size, gfp, node);
+		if (pdata->ptrs[cpu])
+			memset(pdata->ptrs[cpu], 0, size);
+	} else
+		pdata->ptrs[cpu] = kzalloc(size, gfp);
+	return pdata->ptrs[cpu];
+}
+EXPORT_SYMBOL_GPL(percpu_populate);
+
+/**
+ * percpu_populate_mask - populate per-cpu data for more cpu's
+ * @__pdata: per-cpu data to populate further
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-cpu data for cpu's selected through mask bits
+ *
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+			   cpumask_t *mask)
+{
+	cpumask_t populated = CPU_MASK_NONE;
+	int cpu;
+
+	for_each_cpu_mask(cpu, *mask)
+		if (unlikely(!percpu_populate(__pdata, size, gfp, cpu))) {
+			__percpu_depopulate_mask(__pdata, &populated);
+			return -ENOMEM;
+		} else
+			cpu_set(cpu, populated);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(__percpu_populate_mask);
+
+/**
+ * percpu_alloc_mask - initial setup of per-cpu data
+ * @size: size of per-cpu object
+ * @gfp: may sleep or not etc.
+ * @mask: populate per-data for cpu's selected through mask bits
+ *
+ * Populating per-cpu data for all online cpu's would be a typical use case,
+ * which is simplified by the percpu_alloc() wrapper.
+ * Per-cpu objects are populated with zeroed buffers.
+ */
+void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
+{
+	void *pdata = kzalloc(sizeof(struct percpu_data), gfp);
+	void *__pdata = __percpu_disguise(pdata);
+
+	if (unlikely(!pdata))
+		return NULL;
+	if (likely(!__percpu_populate_mask(__pdata, size, gfp, mask)))
+		return __pdata;
+	kfree(pdata);
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(__percpu_alloc_mask);
+
+/**
+ * percpu_free - final cleanup of per-cpu data
+ * @__pdata: object to clean up
+ *
+ * We simply clean up any per-cpu object left. No need for the client to
+ * track and specify through a bis mask which per-cpu objects are to free.
+ */
+void percpu_free(void *__pdata)
+{
+	__percpu_depopulate_mask(__pdata, &cpu_possible_map);
+	kfree(__percpu_disguise(__pdata));
+}
+EXPORT_SYMBOL_GPL(percpu_free);
diff --git a/mm/slab.c b/mm/slab.c
index 619337a5cb2b..13b5050f84cc 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3440,130 +3440,6 @@ void *__kmalloc_track_caller(size_t size, gfp_t flags, void *caller)
 EXPORT_SYMBOL(__kmalloc_track_caller);
 #endif
 
-#ifdef CONFIG_SMP
-/**
- * percpu_depopulate - depopulate per-cpu data for given cpu
- * @__pdata: per-cpu data to depopulate
- * @cpu: depopulate per-cpu data for this cpu
- *
- * Depopulating per-cpu data for a cpu going offline would be a typical
- * use case. You need to register a cpu hotplug handler for that purpose.
- */
-void percpu_depopulate(void *__pdata, int cpu)
-{
-	struct percpu_data *pdata = __percpu_disguise(__pdata);
-	if (pdata->ptrs[cpu]) {
-		kfree(pdata->ptrs[cpu]);
-		pdata->ptrs[cpu] = NULL;
-	}
-}
-EXPORT_SYMBOL_GPL(percpu_depopulate);
-
-/**
- * percpu_depopulate_mask - depopulate per-cpu data for some cpu's
- * @__pdata: per-cpu data to depopulate
- * @mask: depopulate per-cpu data for cpu's selected through mask bits
- */
-void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
-{
-	int cpu;
-	for_each_cpu_mask(cpu, *mask)
-		percpu_depopulate(__pdata, cpu);
-}
-EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
-
-/**
- * percpu_populate - populate per-cpu data for given cpu
- * @__pdata: per-cpu data to populate further
- * @size: size of per-cpu object
- * @gfp: may sleep or not etc.
- * @cpu: populate per-data for this cpu
- *
- * Populating per-cpu data for a cpu coming online would be a typical
- * use case. You need to register a cpu hotplug handler for that purpose.
- * Per-cpu object is populated with zeroed buffer.
- */
-void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
-{
-	struct percpu_data *pdata = __percpu_disguise(__pdata);
-	int node = cpu_to_node(cpu);
-
-	BUG_ON(pdata->ptrs[cpu]);
-	if (node_online(node)) {
-		/* FIXME: kzalloc_node(size, gfp, node) */
-		pdata->ptrs[cpu] = kmalloc_node(size, gfp, node);
-		if (pdata->ptrs[cpu])
-			memset(pdata->ptrs[cpu], 0, size);
-	} else
-		pdata->ptrs[cpu] = kzalloc(size, gfp);
-	return pdata->ptrs[cpu];
-}
-EXPORT_SYMBOL_GPL(percpu_populate);
-
-/**
- * percpu_populate_mask - populate per-cpu data for more cpu's
- * @__pdata: per-cpu data to populate further
- * @size: size of per-cpu object
- * @gfp: may sleep or not etc.
- * @mask: populate per-cpu data for cpu's selected through mask bits
- *
- * Per-cpu objects are populated with zeroed buffers.
- */
-int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
-			   cpumask_t *mask)
-{
-	cpumask_t populated = CPU_MASK_NONE;
-	int cpu;
-
-	for_each_cpu_mask(cpu, *mask)
-		if (unlikely(!percpu_populate(__pdata, size, gfp, cpu))) {
-			__percpu_depopulate_mask(__pdata, &populated);
-			return -ENOMEM;
-		} else
-			cpu_set(cpu, populated);
-	return 0;
-}
-EXPORT_SYMBOL_GPL(__percpu_populate_mask);
-
-/**
- * percpu_alloc_mask - initial setup of per-cpu data
- * @size: size of per-cpu object
- * @gfp: may sleep or not etc.
- * @mask: populate per-data for cpu's selected through mask bits
- *
- * Populating per-cpu data for all online cpu's would be a typical use case,
- * which is simplified by the percpu_alloc() wrapper.
- * Per-cpu objects are populated with zeroed buffers.
- */
-void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
-{
-	void *pdata = kzalloc(sizeof(struct percpu_data), gfp);
-	void *__pdata = __percpu_disguise(pdata);
-
-	if (unlikely(!pdata))
-		return NULL;
-	if (likely(!__percpu_populate_mask(__pdata, size, gfp, mask)))
-		return __pdata;
-	kfree(pdata);
-	return NULL;
-}
-EXPORT_SYMBOL_GPL(__percpu_alloc_mask);
-
-/**
- * percpu_free - final cleanup of per-cpu data
- * @__pdata: object to clean up
- *
- * We simply clean up any per-cpu object left. No need for the client to
- * track and specify through a bis mask which per-cpu objects are to free.
- */
-void percpu_free(void *__pdata)
-{
-	__percpu_depopulate_mask(__pdata, &cpu_possible_map);
-	kfree(__percpu_disguise(__pdata));
-}
-EXPORT_SYMBOL_GPL(percpu_free);
-#endif	/* CONFIG_SMP */
-
 /**
  * kmem_cache_free - Deallocate an object
  * @cachep: The cache the allocation was from.
diff --git a/mm/slob.c b/mm/slob.c
index 7b52b20b9607..4c28a421b270 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -343,48 +343,3 @@ void kmem_cache_init(void)
 atomic_t slab_reclaim_pages = ATOMIC_INIT(0);
 EXPORT_SYMBOL(slab_reclaim_pages);
 
-#ifdef CONFIG_SMP
-
-void *__alloc_percpu(size_t size)
-{
-	int i;
-	struct percpu_data *pdata = kmalloc(sizeof (*pdata), GFP_KERNEL);
-
-	if (!pdata)
-		return NULL;
-
-	for_each_possible_cpu(i) {
-		pdata->ptrs[i] = kmalloc(size, GFP_KERNEL);
-		if (!pdata->ptrs[i])
-			goto unwind_oom;
-		memset(pdata->ptrs[i], 0, size);
-	}
-
-	/* Catch derefs w/o wrappers */
-	return (void *) (~(unsigned long) pdata);
-
-unwind_oom:
-	while (--i >= 0) {
-		if (!cpu_possible(i))
-			continue;
-		kfree(pdata->ptrs[i]);
-	}
-	kfree(pdata);
-	return NULL;
-}
-EXPORT_SYMBOL(__alloc_percpu);
-
-void
-free_percpu(const void *objp)
-{
-	int i;
-	struct percpu_data *p = (struct percpu_data *) (~(unsigned long) objp);
-
-	for_each_possible_cpu(i)
-		kfree(p->ptrs[i]);
-
-	kfree(p);
-}
-EXPORT_SYMBOL(free_percpu);
-
-#endif
-- 
cgit v1.2.1


From 8417bba4b151346ed475fcc923693c9e3be89063 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:51 -0700
Subject: [PATCH] Replace min_unmapped_ratio by min_unmapped_pages in struct
 zone

*_pages is a better description of the role of the variable.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 2 +-
 mm/page_alloc.c        | 4 ++--
 mm/vmscan.c            | 2 +-
 3 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 7fe317164b73..a703527e2b45 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -169,7 +169,7 @@ struct zone {
 	/*
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
-	unsigned long		min_unmapped_ratio;
+	unsigned long		min_unmapped_pages;
 	struct per_cpu_pageset	*pageset[NR_CPUS];
 #else
 	struct per_cpu_pageset	pageset[NR_CPUS];
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f7ea020c23ea..5da6bc4e0a6b 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2002,7 +2002,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		zone->spanned_pages = size;
 		zone->present_pages = realsize;
 #ifdef CONFIG_NUMA
-		zone->min_unmapped_ratio = (realsize*sysctl_min_unmapped_ratio)
+		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
 #endif
 		zone->name = zone_names[j];
@@ -2313,7 +2313,7 @@ int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write,
 		return rc;
 
 	for_each_zone(zone)
-		zone->min_unmapped_ratio = (zone->present_pages *
+		zone->min_unmapped_pages = (zone->present_pages *
 				sysctl_min_unmapped_ratio) / 100;
 	return 0;
 }
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 8f35d7d585cb..5154c25e8440 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1618,7 +1618,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * unmapped file backed pages.
 	 */
 	if (zone_page_state(zone, NR_FILE_PAGES) -
-	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_ratio)
+	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages)
 		return 0;
 
 	/*
-- 
cgit v1.2.1


From 972d1a7b140569084439a81265a0f15b74e924e0 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:51 -0700
Subject: [PATCH] ZVC: Support NR_SLAB_RECLAIMABLE / NR_SLAB_UNRECLAIMABLE

Remove the atomic counter for slab_reclaim_pages and replace the counter
and NR_SLAB with two ZVC counter that account for unreclaimable and
reclaimable slab pages: NR_SLAB_RECLAIMABLE and NR_SLAB_UNRECLAIMABLE.

Change the check in vmscan.c to refer to to NR_SLAB_RECLAIMABLE.  The
intend seems to be to check for slab pages that could be freed.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/pgtable.c |  4 +++-
 drivers/base/node.c    |  9 +++++++--
 fs/proc/proc_misc.c    |  7 ++++++-
 include/linux/mmzone.h |  3 ++-
 include/linux/slab.h   |  2 --
 mm/mmap.c              |  2 +-
 mm/nommu.c             |  2 +-
 mm/page_alloc.c        |  3 ++-
 mm/slab.c              | 24 +++++++++++-------------
 mm/slob.c              |  4 ----
 mm/vmscan.c            |  2 +-
 mm/vmstat.c            |  3 ++-
 12 files changed, 36 insertions(+), 29 deletions(-)

diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index bd98768d8764..a9f4910a22f8 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -60,7 +60,9 @@ void show_mem(void)
 	printk(KERN_INFO "%lu pages writeback\n",
 					global_page_state(NR_WRITEBACK));
 	printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
+	printk(KERN_INFO "%lu pages slab\n",
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+		global_page_state(NR_SLAB_UNRECLAIMABLE));
 	printk(KERN_INFO "%lu pages pagetables\n",
 					global_page_state(NR_PAGETABLE));
 }
diff --git a/drivers/base/node.c b/drivers/base/node.c
index e09f5c2c11ee..001e6f6b9c1b 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -68,7 +68,9 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       "Node %d PageTables:   %8lu kB\n"
 		       "Node %d NFS_Unstable: %8lu kB\n"
 		       "Node %d Bounce:       %8lu kB\n"
-		       "Node %d Slab:         %8lu kB\n",
+		       "Node %d Slab:         %8lu kB\n"
+		       "Node %d SReclaimable: %8lu kB\n"
+		       "Node %d SUnreclaim:   %8lu kB\n",
 		       nid, K(i.totalram),
 		       nid, K(i.freeram),
 		       nid, K(i.totalram - i.freeram),
@@ -88,7 +90,10 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       nid, K(node_page_state(nid, NR_PAGETABLE)),
 		       nid, K(node_page_state(nid, NR_UNSTABLE_NFS)),
 		       nid, K(node_page_state(nid, NR_BOUNCE)),
-		       nid, K(node_page_state(nid, NR_SLAB)));
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE) +
+				node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE)));
 	n += hugetlb_report_node_meminfo(nid, buf + n);
 	return n;
 }
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index caa0a51560a0..5bbd60896050 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -170,6 +170,8 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		"AnonPages:    %8lu kB\n"
 		"Mapped:       %8lu kB\n"
 		"Slab:         %8lu kB\n"
+		"SReclaimable: %8lu kB\n"
+		"SUnreclaim:   %8lu kB\n"
 		"PageTables:   %8lu kB\n"
 		"NFS_Unstable: %8lu kB\n"
 		"Bounce:       %8lu kB\n"
@@ -197,7 +199,10 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		K(global_page_state(NR_WRITEBACK)),
 		K(global_page_state(NR_ANON_PAGES)),
 		K(global_page_state(NR_FILE_MAPPED)),
-		K(global_page_state(NR_SLAB)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE) +
+				global_page_state(NR_SLAB_UNRECLAIMABLE)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE)),
+		K(global_page_state(NR_SLAB_UNRECLAIMABLE)),
 		K(global_page_state(NR_PAGETABLE)),
 		K(global_page_state(NR_UNSTABLE_NFS)),
 		K(global_page_state(NR_BOUNCE)),
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index a703527e2b45..08c41b9f92e0 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -51,7 +51,8 @@ enum zone_stat_item {
 	NR_FILE_MAPPED,	/* pagecache pages mapped into pagetables.
 			   only modified from process context */
 	NR_FILE_PAGES,
-	NR_SLAB,	/* Pages used by slab allocator */
+	NR_SLAB_RECLAIMABLE,
+	NR_SLAB_UNRECLAIMABLE,
 	NR_PAGETABLE,	/* used for pagetables */
 	NR_FILE_DIRTY,
 	NR_WRITEBACK,
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 2f6bef6a98c9..66d6eb78d1c6 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -284,8 +284,6 @@ extern kmem_cache_t	*fs_cachep;
 extern kmem_cache_t	*sighand_cachep;
 extern kmem_cache_t	*bio_cachep;
 
-extern atomic_t slab_reclaim_pages;
-
 #endif	/* __KERNEL__ */
 
 #endif	/* _LINUX_SLAB_H */
diff --git a/mm/mmap.c b/mm/mmap.c
index 8507ee9cd573..eea8eefd51a8 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -116,7 +116,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root
diff --git a/mm/nommu.c b/mm/nommu.c
index c576df71e3bb..d99dea31e443 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1133,7 +1133,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5da6bc4e0a6b..47e98423b30d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1304,7 +1304,8 @@ void show_free_areas(void)
 		global_page_state(NR_WRITEBACK),
 		global_page_state(NR_UNSTABLE_NFS),
 		nr_free_pages(),
-		global_page_state(NR_SLAB),
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+			global_page_state(NR_SLAB_UNRECLAIMABLE),
 		global_page_state(NR_FILE_MAPPED),
 		global_page_state(NR_PAGETABLE));
 
diff --git a/mm/slab.c b/mm/slab.c
index 13b5050f84cc..7a48eb1a60c8 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -735,14 +735,6 @@ static inline void init_lock_keys(void)
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
 
-/*
- * vm_enough_memory() looks at this to determine how many slab-allocated pages
- * are possibly freeable under pressure
- *
- * SLAB_RECLAIM_ACCOUNT turns this on per-slab
- */
-atomic_t slab_reclaim_pages;
-
 /*
  * chicken and egg problem: delay the per-cpu array allocation
  * until the general caches are up.
@@ -1580,8 +1572,11 @@ static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 
 	nr_pages = (1 << cachep->gfporder);
 	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_add(nr_pages, &slab_reclaim_pages);
-	add_zone_page_state(page_zone(page), NR_SLAB, nr_pages);
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_RECLAIMABLE, nr_pages);
+	else
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_UNRECLAIMABLE, nr_pages);
 	for (i = 0; i < nr_pages; i++)
 		__SetPageSlab(page + i);
 	return page_address(page);
@@ -1596,7 +1591,12 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	struct page *page = virt_to_page(addr);
 	const unsigned long nr_freed = i;
 
-	sub_zone_page_state(page_zone(page), NR_SLAB, nr_freed);
+	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_RECLAIMABLE, nr_freed);
+	else
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_UNRECLAIMABLE, nr_freed);
 	while (i--) {
 		BUG_ON(!PageSlab(page));
 		__ClearPageSlab(page);
@@ -1605,8 +1605,6 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	if (current->reclaim_state)
 		current->reclaim_state->reclaimed_slab += nr_freed;
 	free_pages((unsigned long)addr, cachep->gfporder);
-	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_sub(1 << cachep->gfporder, &slab_reclaim_pages);
 }
 
 static void kmem_rcu_free(struct rcu_head *head)
diff --git a/mm/slob.c b/mm/slob.c
index 4c28a421b270..20188627347c 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -339,7 +339,3 @@ void kmem_cache_init(void)
 
 	mod_timer(&slob_timer, jiffies + HZ);
 }
-
-atomic_t slab_reclaim_pages = ATOMIC_INIT(0);
-EXPORT_SYMBOL(slab_reclaim_pages);
-
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5154c25e8440..349797ba4bac 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1378,7 +1378,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages)
 	for_each_zone(zone)
 		lru_pages += zone->nr_active + zone->nr_inactive;
 
-	nr_slab = global_page_state(NR_SLAB);
+	nr_slab = global_page_state(NR_SLAB_RECLAIMABLE);
 	/* If slab caches are huge, it's better to hit them first */
 	while (nr_slab >= lru_pages) {
 		reclaim_state.reclaimed_slab = 0;
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 968c0072e19a..490d8c1a0ded 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -458,7 +458,8 @@ static char *vmstat_text[] = {
 	"nr_anon_pages",
 	"nr_mapped",
 	"nr_file_pages",
-	"nr_slab",
+	"nr_slab_reclaimable",
+	"nr_slab_unreclaimable",
 	"nr_page_table_pages",
 	"nr_dirty",
 	"nr_writeback",
-- 
cgit v1.2.1


From 0ff38490c836dc379ff7ec45b10a15a662f4e5f6 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:52 -0700
Subject: [PATCH] zone_reclaim: dynamic slab reclaim

Currently one can enable slab reclaim by setting an explicit option in
/proc/sys/vm/zone_reclaim_mode.  Slab reclaim is then used as a final
option if the freeing of unmapped file backed pages is not enough to free
enough pages to allow a local allocation.

However, that means that the slab can grow excessively and that most memory
of a node may be used by slabs.  We have had a case where a machine with
46GB of memory was using 40-42GB for slab.  Zone reclaim was effective in
dealing with pagecache pages.  However, slab reclaim was only done during
global reclaim (which is a bit rare on NUMA systems).

This patch implements slab reclaim during zone reclaim.  Zone reclaim
occurs if there is a danger of an off node allocation.  At that point we

1. Shrink the per node page cache if the number of pagecache
   pages is more than min_unmapped_ratio percent of pages in a zone.

2. Shrink the slab cache if the number of the nodes reclaimable slab pages
   (patch depends on earlier one that implements that counter)
   are more than min_slab_ratio (a new /proc/sys/vm tunable).

The shrinking of the slab cache is a bit problematic since it is not node
specific.  So we simply calculate what point in the slab we want to reach
(current per node slab use minus the number of pages that neeed to be
allocated) and then repeately run the global reclaim until that is
unsuccessful or we have reached the limit.  I hope we will have zone based
slab reclaim at some point which will make that easier.

The default for the min_slab_ratio is 5%

Also remove the slab option from /proc/sys/vm/zone_reclaim_mode.

[akpm@osdl.org: cleanups]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/sysctl/vm.txt | 27 +++++++++++++++------
 include/linux/mmzone.h      |  3 +++
 include/linux/swap.h        |  1 +
 include/linux/sysctl.h      |  1 +
 kernel/sysctl.c             | 11 +++++++++
 mm/page_alloc.c             | 17 +++++++++++++
 mm/vmscan.c                 | 58 +++++++++++++++++++++++++++++----------------
 7 files changed, 90 insertions(+), 28 deletions(-)

diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 7cee90223d3a..20d0d797f539 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -29,6 +29,7 @@ Currently, these files are in /proc/sys/vm:
 - drop-caches
 - zone_reclaim_mode
 - min_unmapped_ratio
+- min_slab_ratio
 - panic_on_oom
 
 ==============================================================
@@ -138,7 +139,6 @@ This is value ORed together of
 1	= Zone reclaim on
 2	= Zone reclaim writes dirty pages out
 4	= Zone reclaim swaps pages
-8	= Also do a global slab reclaim pass
 
 zone_reclaim_mode is set during bootup to 1 if it is determined that pages
 from remote zones will cause a measurable performance reduction. The
@@ -162,18 +162,13 @@ Allowing regular swap effectively restricts allocations to the local
 node unless explicitly overridden by memory policies or cpuset
 configurations.
 
-It may be advisable to allow slab reclaim if the system makes heavy
-use of files and builds up large slab caches. However, the slab
-shrink operation is global, may take a long time and free slabs
-in all nodes of the system.
-
 =============================================================
 
 min_unmapped_ratio:
 
 This is available only on NUMA kernels.
 
-A percentage of the file backed pages in each zone.  Zone reclaim will only
+A percentage of the total pages in each zone.  Zone reclaim will only
 occur if more than this percentage of pages are file backed and unmapped.
 This is to insure that a minimal amount of local pages is still available for
 file I/O even if the node is overallocated.
@@ -182,6 +177,24 @@ The default is 1 percent.
 
 =============================================================
 
+min_slab_ratio:
+
+This is available only on NUMA kernels.
+
+A percentage of the total pages in each zone.  On Zone reclaim
+(fallback from the local zone occurs) slabs will be reclaimed if more
+than this percentage of pages in a zone are reclaimable slab pages.
+This insures that the slab growth stays under control even in NUMA
+systems that rarely perform global reclaim.
+
+The default is 5 percent.
+
+Note that slab reclaim is triggered in a per zone / node fashion.
+The process of reclaiming slab memory is currently not node specific
+and may not be fast.
+
+=============================================================
+
 panic_on_oom
 
 This enables or disables panic on out-of-memory feature.  If this is set to 1,
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 08c41b9f92e0..3693f1a52788 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -171,6 +171,7 @@ struct zone {
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
 	unsigned long		min_unmapped_pages;
+	unsigned long		min_slab_pages;
 	struct per_cpu_pageset	*pageset[NR_CPUS];
 #else
 	struct per_cpu_pageset	pageset[NR_CPUS];
@@ -448,6 +449,8 @@ int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *, int, struct file
 					void __user *, size_t *, loff_t *);
 int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *, int,
 			struct file *, void __user *, size_t *, loff_t *);
+int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *, int,
+			struct file *, void __user *, size_t *, loff_t *);
 
 #include <linux/topology.h>
 /* Returns the number of the current Node. */
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 32db06c8ffe0..a2f5ad7c2d2e 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -193,6 +193,7 @@ extern long vm_total_pages;
 #ifdef CONFIG_NUMA
 extern int zone_reclaim_mode;
 extern int sysctl_min_unmapped_ratio;
+extern int sysctl_min_slab_ratio;
 extern int zone_reclaim(struct zone *, gfp_t, unsigned int);
 #else
 #define zone_reclaim_mode 0
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 736ed917a4f8..eca555781d05 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -191,6 +191,7 @@ enum
 	VM_MIN_UNMAPPED=32,	/* Set min percent of unmapped pages */
 	VM_PANIC_ON_OOM=33,	/* panic at out-of-memory */
 	VM_VDSO_ENABLED=34,	/* map VDSO into new processes? */
+	VM_MIN_SLAB=35,		 /* Percent pages ignored by zone reclaim */
 };
 
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 362a0cc37138..fd43c3e6786b 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -943,6 +943,17 @@ static ctl_table vm_table[] = {
 		.extra1		= &zero,
 		.extra2		= &one_hundred,
 	},
+	{
+		.ctl_name	= VM_MIN_SLAB,
+		.procname	= "min_slab_ratio",
+		.data		= &sysctl_min_slab_ratio,
+		.maxlen		= sizeof(sysctl_min_slab_ratio),
+		.mode		= 0644,
+		.proc_handler	= &sysctl_min_slab_ratio_sysctl_handler,
+		.strategy	= &sysctl_intvec,
+		.extra1		= &zero,
+		.extra2		= &one_hundred,
+	},
 #endif
 #ifdef CONFIG_X86_32
 	{
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 47e98423b30d..cf913bdd433e 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2005,6 +2005,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 #ifdef CONFIG_NUMA
 		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
+		zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100;
 #endif
 		zone->name = zone_names[j];
 		spin_lock_init(&zone->lock);
@@ -2318,6 +2319,22 @@ int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write,
 				sysctl_min_unmapped_ratio) / 100;
 	return 0;
 }
+
+int sysctl_min_slab_ratio_sysctl_handler(ctl_table *table, int write,
+	struct file *file, void __user *buffer, size_t *length, loff_t *ppos)
+{
+	struct zone *zone;
+	int rc;
+
+	rc = proc_dointvec_minmax(table, write, file, buffer, length, ppos);
+	if (rc)
+		return rc;
+
+	for_each_zone(zone)
+		zone->min_slab_pages = (zone->present_pages *
+				sysctl_min_slab_ratio) / 100;
+	return 0;
+}
 #endif
 
 /*
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 349797ba4bac..089e943c4d38 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1527,7 +1527,6 @@ int zone_reclaim_mode __read_mostly;
 #define RECLAIM_ZONE (1<<0)	/* Run shrink_cache on the zone */
 #define RECLAIM_WRITE (1<<1)	/* Writeout pages during reclaim */
 #define RECLAIM_SWAP (1<<2)	/* Swap pages out during reclaim */
-#define RECLAIM_SLAB (1<<3)	/* Do a global slab shrink if the zone is out of memory */
 
 /*
  * Priority for ZONE_RECLAIM. This determines the fraction of pages
@@ -1542,6 +1541,12 @@ int zone_reclaim_mode __read_mostly;
  */
 int sysctl_min_unmapped_ratio = 1;
 
+/*
+ * If the number of slab pages in a zone grows beyond this percentage then
+ * slab reclaim needs to occur.
+ */
+int sysctl_min_slab_ratio = 5;
+
 /*
  * Try to free up some pages from this zone through reclaim.
  */
@@ -1573,29 +1578,37 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	reclaim_state.reclaimed_slab = 0;
 	p->reclaim_state = &reclaim_state;
 
-	/*
-	 * Free memory by calling shrink zone with increasing priorities
-	 * until we have enough memory freed.
-	 */
-	priority = ZONE_RECLAIM_PRIORITY;
-	do {
-		nr_reclaimed += shrink_zone(priority, zone, &sc);
-		priority--;
-	} while (priority >= 0 && nr_reclaimed < nr_pages);
+	if (zone_page_state(zone, NR_FILE_PAGES) -
+		zone_page_state(zone, NR_FILE_MAPPED) >
+		zone->min_unmapped_pages) {
+		/*
+		 * Free memory by calling shrink zone with increasing
+		 * priorities until we have enough memory freed.
+		 */
+		priority = ZONE_RECLAIM_PRIORITY;
+		do {
+			nr_reclaimed += shrink_zone(priority, zone, &sc);
+			priority--;
+		} while (priority >= 0 && nr_reclaimed < nr_pages);
+	}
 
-	if (nr_reclaimed < nr_pages && (zone_reclaim_mode & RECLAIM_SLAB)) {
+	if (zone_page_state(zone, NR_SLAB_RECLAIMABLE) > zone->min_slab_pages) {
 		/*
 		 * shrink_slab() does not currently allow us to determine how
-		 * many pages were freed in this zone. So we just shake the slab
-		 * a bit and then go off node for this particular allocation
-		 * despite possibly having freed enough memory to allocate in
-		 * this zone.  If we freed local memory then the next
-		 * allocations will be local again.
+		 * many pages were freed in this zone. So we take the current
+		 * number of slab pages and shake the slab until it is reduced
+		 * by the same nr_pages that we used for reclaiming unmapped
+		 * pages.
 		 *
-		 * shrink_slab will free memory on all zones and may take
-		 * a long time.
+		 * Note that shrink_slab will free memory on all zones and may
+		 * take a long time.
 		 */
-		shrink_slab(sc.nr_scanned, gfp_mask, order);
+		unsigned long limit = zone_page_state(zone,
+				NR_SLAB_RECLAIMABLE) - nr_pages;
+
+		while (shrink_slab(sc.nr_scanned, gfp_mask, order) &&
+			zone_page_state(zone, NR_SLAB_RECLAIMABLE) > limit)
+			;
 	}
 
 	p->reclaim_state = NULL;
@@ -1609,7 +1622,8 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	int node_id;
 
 	/*
-	 * Zone reclaim reclaims unmapped file backed pages.
+	 * Zone reclaim reclaims unmapped file backed pages and
+	 * slab pages if we are over the defined limits.
 	 *
 	 * A small portion of unmapped file backed pages is needed for
 	 * file I/O otherwise pages read by file I/O will be immediately
@@ -1618,7 +1632,9 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * unmapped file backed pages.
 	 */
 	if (zone_page_state(zone, NR_FILE_PAGES) -
-	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages)
+	    zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages
+	    && zone_page_state(zone, NR_SLAB_RECLAIMABLE)
+			<= zone->min_slab_pages)
 		return 0;
 
 	/*
-- 
cgit v1.2.1


From 83e33a4711760469f5c3861b8ffea4947656d4eb Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:53 -0700
Subject: [PATCH] zone reclaim with slab: avoid unecessary off node allocations

Minor performance fix.

If we reclaimed enough slab pages from a zone then we can avoid going off
node with the current allocation.  Take care of updating nr_reclaimed when
reclaiming from the slab.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 089e943c4d38..b950f193816e 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1566,6 +1566,7 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 		.gfp_mask = gfp_mask,
 		.swappiness = vm_swappiness,
 	};
+	unsigned long slab_reclaimable;
 
 	disable_swap_token();
 	cond_resched();
@@ -1592,7 +1593,8 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 		} while (priority >= 0 && nr_reclaimed < nr_pages);
 	}
 
-	if (zone_page_state(zone, NR_SLAB_RECLAIMABLE) > zone->min_slab_pages) {
+	slab_reclaimable = zone_page_state(zone, NR_SLAB_RECLAIMABLE);
+	if (slab_reclaimable > zone->min_slab_pages) {
 		/*
 		 * shrink_slab() does not currently allow us to determine how
 		 * many pages were freed in this zone. So we take the current
@@ -1603,12 +1605,17 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 		 * Note that shrink_slab will free memory on all zones and may
 		 * take a long time.
 		 */
-		unsigned long limit = zone_page_state(zone,
-				NR_SLAB_RECLAIMABLE) - nr_pages;
-
 		while (shrink_slab(sc.nr_scanned, gfp_mask, order) &&
-			zone_page_state(zone, NR_SLAB_RECLAIMABLE) > limit)
+			zone_page_state(zone, NR_SLAB_RECLAIMABLE) >
+				slab_reclaimable - nr_pages)
 			;
+
+		/*
+		 * Update nr_reclaimed by the number of slab pages we
+		 * reclaimed from this zone.
+		 */
+		nr_reclaimed += slab_reclaimable -
+			zone_page_state(zone, NR_SLAB_RECLAIMABLE);
 	}
 
 	p->reclaim_state = NULL;
-- 
cgit v1.2.1


From 5a291b98b2116d669449885abef3000f747504b3 Mon Sep 17 00:00:00 2001
From: Ram Gupta <ram.gupta5@gmail.com>
Date: Mon, 25 Sep 2006 23:31:54 -0700
Subject: [PATCH] oom-kill: update comments to reflect current code

Update the comments for __oom_kill_task() to reflect the code changes.

Signed-off-by: Ram Gupta <r.gupta@astronautics.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index c5e384000585..f1c0ef1fd21f 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -250,9 +250,9 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 }
 
 /**
- * We must be careful though to never send SIGKILL a process with
- * CAP_SYS_RAW_IO set, send SIGTERM instead (but it's unlikely that
- * we select a process with CAP_SYS_RAW_IO set).
+ * Send SIGKILL to the selected  process irrespective of  CAP_SYS_RAW_IO
+ * flag though it's unlikely that  we select a process with CAP_SYS_RAW_IO
+ * set.
  */
 static void __oom_kill_task(struct task_struct *p, const char *message)
 {
-- 
cgit v1.2.1


From 4415cc8df630b05d3a54267d5f3e5c0b63a4ec05 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:55 -0700
Subject: [PATCH] Hugepages: Use page_to_nid rather than traversing zone
 pointers

I found two location in hugetlb.c where we chase pointer instead of using
page_to_nid().  Page_to_nid is more effective and can get the node directly
from page flags.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/hugetlb.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index df499973255f..3aceadce1a76 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -177,7 +177,7 @@ static void update_and_free_page(struct page *page)
 {
 	int i;
 	nr_huge_pages--;
-	nr_huge_pages_node[page_zone(page)->zone_pgdat->node_id]--;
+	nr_huge_pages_node[page_to_nid(page)]--;
 	for (i = 0; i < (HPAGE_SIZE / PAGE_SIZE); i++) {
 		page[i].flags &= ~(1 << PG_locked | 1 << PG_error | 1 << PG_referenced |
 				1 << PG_dirty | 1 << PG_active | 1 << PG_reserved |
@@ -191,7 +191,8 @@ static void update_and_free_page(struct page *page)
 #ifdef CONFIG_HIGHMEM
 static void try_to_free_low(unsigned long count)
 {
-	int i, nid;
+	int i;
+
 	for (i = 0; i < MAX_NUMNODES; ++i) {
 		struct page *page, *next;
 		list_for_each_entry_safe(page, next, &hugepage_freelists[i], lru) {
@@ -199,9 +200,8 @@ static void try_to_free_low(unsigned long count)
 				continue;
 			list_del(&page->lru);
 			update_and_free_page(page);
-			nid = page_zone(page)->zone_pgdat->node_id;
 			free_huge_pages--;
-			free_huge_pages_node[nid]--;
+			free_huge_pages_node[page_to_nid(page)]--;
 			if (count >= nr_huge_pages)
 				return;
 		}
-- 
cgit v1.2.1


From 89fa30242facca249aead2aac03c4c69764f911c Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Mon, 25 Sep 2006 23:31:55 -0700
Subject: [PATCH] NUMA: Add zone_to_nid function

There are many places where we need to determine the node of a zone.
Currently we use a difficult to read sequence of pointer dereferencing.
Put that into an inline function and use throughout VM.  Maybe we can find
a way to optimize the lookup in the future.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mm/discontig.c | 2 +-
 arch/parisc/mm/init.c    | 2 +-
 include/linux/mm.h       | 7 ++++++-
 kernel/cpuset.c          | 4 ++--
 mm/hugetlb.c             | 2 +-
 mm/mempolicy.c           | 6 +++---
 mm/oom_kill.c            | 3 +--
 mm/page_alloc.c          | 2 +-
 mm/vmscan.c              | 2 +-
 9 files changed, 17 insertions(+), 13 deletions(-)

diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 07c300f93764..fb5d8b747de4 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -422,7 +422,7 @@ void __init set_highmem_pages_init(int bad_ppro)
 		zone_end_pfn = zone_start_pfn + zone->spanned_pages;
 
 		printk("Initializing %s for node %d (%08lx:%08lx)\n",
-				zone->name, zone->zone_pgdat->node_id,
+				zone->name, zone_to_nid(zone),
 				zone_start_pfn, zone_end_pfn);
 
 		for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) {
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index c7329615ef94..25ad28d63e88 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -551,7 +551,7 @@ void show_mem(void)
 
 				printk("Zone list for zone %d on node %d: ", j, i);
 				for (k = 0; zl->zones[k] != NULL; k++) 
-					printk("[%d/%s] ", zl->zones[k]->zone_pgdat->node_id, zl->zones[k]->name);
+					printk("[%d/%s] ", zone_to_nid(zl->zones[k]), zl->zones[k]->name);
 				printk("\n");
 			}
 		}
diff --git a/include/linux/mm.h b/include/linux/mm.h
index f2018775b995..856f0ee7e84a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -499,12 +499,17 @@ static inline struct zone *page_zone(struct page *page)
 	return zone_table[page_zone_id(page)];
 }
 
+static inline unsigned long zone_to_nid(struct zone *zone)
+{
+	return zone->zone_pgdat->node_id;
+}
+
 static inline unsigned long page_to_nid(struct page *page)
 {
 	if (FLAGS_HAS_NODE)
 		return (page->flags >> NODES_PGSHIFT) & NODES_MASK;
 	else
-		return page_zone(page)->zone_pgdat->node_id;
+		return zone_to_nid(page_zone(page));
 }
 static inline unsigned long page_to_section(struct page *page)
 {
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 76940361273e..cff41511269f 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -2245,7 +2245,7 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl)
 	int i;
 
 	for (i = 0; zl->zones[i]; i++) {
-		int nid = zl->zones[i]->zone_pgdat->node_id;
+		int nid = zone_to_nid(zl->zones[i]);
 
 		if (node_isset(nid, current->mems_allowed))
 			return 1;
@@ -2318,7 +2318,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 
 	if (in_interrupt() || (gfp_mask & __GFP_THISNODE))
 		return 1;
-	node = z->zone_pgdat->node_id;
+	node = zone_to_nid(z);
 	might_sleep_if(!(gfp_mask & __GFP_HARDWALL));
 	if (node_isset(node, current->mems_allowed))
 		return 1;
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 3aceadce1a76..7c7d03dbf73d 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -72,7 +72,7 @@ static struct page *dequeue_huge_page(struct vm_area_struct *vma,
 	struct zone **z;
 
 	for (z = zonelist->zones; *z; z++) {
-		nid = (*z)->zone_pgdat->node_id;
+		nid = zone_to_nid(*z);
 		if (cpuset_zone_allowed(*z, GFP_HIGHUSER) &&
 		    !list_empty(&hugepage_freelists[nid]))
 			break;
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 8002e1faccda..38f89650bc84 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -487,7 +487,7 @@ static void get_zonemask(struct mempolicy *p, nodemask_t *nodes)
 	switch (p->policy) {
 	case MPOL_BIND:
 		for (i = 0; p->v.zonelist->zones[i]; i++)
-			node_set(p->v.zonelist->zones[i]->zone_pgdat->node_id,
+			node_set(zone_to_nid(p->v.zonelist->zones[i]),
 				*nodes);
 		break;
 	case MPOL_DEFAULT:
@@ -1145,7 +1145,7 @@ unsigned slab_node(struct mempolicy *policy)
 		 * Follow bind policy behavior and start allocation at the
 		 * first node.
 		 */
-		return policy->v.zonelist->zones[0]->zone_pgdat->node_id;
+		return zone_to_nid(policy->v.zonelist->zones[0]);
 
 	case MPOL_PREFERRED:
 		if (policy->v.preferred_node >= 0)
@@ -1649,7 +1649,7 @@ void mpol_rebind_policy(struct mempolicy *pol, const nodemask_t *newmask)
 
 		nodes_clear(nodes);
 		for (z = pol->v.zonelist->zones; *z; z++)
-			node_set((*z)->zone_pgdat->node_id, nodes);
+			node_set(zone_to_nid(*z), nodes);
 		nodes_remap(tmp, nodes, *mpolmask, *newmask);
 		nodes = tmp;
 
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index f1c0ef1fd21f..bada3d03119f 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -177,8 +177,7 @@ static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask)
 
 	for (z = zonelist->zones; *z; z++)
 		if (cpuset_zone_allowed(*z, gfp_mask))
-			node_clear((*z)->zone_pgdat->node_id,
-					nodes);
+			node_clear(zone_to_nid(*z), nodes);
 		else
 			return CONSTRAINT_CPUSET;
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cf913bdd433e..51070b6d593f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1217,7 +1217,7 @@ unsigned int nr_free_pagecache_pages(void)
 #ifdef CONFIG_NUMA
 static void show_node(struct zone *zone)
 {
-	printk("Node %d ", zone->zone_pgdat->node_id);
+	printk("Node %ld ", zone_to_nid(zone));
 }
 #else
 #define show_node(zone)	do { } while (0)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index b950f193816e..87779dda4ec6 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1661,7 +1661,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 	 * over remote processors and spread off node memory allocations
 	 * as wide as possible.
 	 */
-	node_id = zone->zone_pgdat->node_id;
+	node_id = zone_to_nid(zone);
 	mask = node_to_cpumask(node_id);
 	if (!cpus_empty(mask) && node_id != numa_node_id())
 		return 0;
-- 
cgit v1.2.1


From 62bac0185ad3dfef11d9602980445c54d45199c6 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:56 -0700
Subject: [PATCH] selinux: eliminate selinux_task_ctxid

Eliminate selinux_task_ctxid since it duplicates selinux_task_get_sid.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h    | 15 ---------------
 kernel/auditsc.c           |  2 +-
 security/selinux/exports.c |  9 ---------
 3 files changed, 1 insertion(+), 25 deletions(-)

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index aad4e390d6a5..79e4707ca772 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -69,16 +69,6 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
  */
 void selinux_audit_set_callback(int (*callback)(void));
 
-/**
- *	selinux_task_ctxid - determine a context ID for a process.
- *	@tsk: the task object
- *	@ctxid: ID value returned via this
- *
- *	On return, ctxid will contain an ID for the context.  This value
- *	should only be used opaquely.
- */
-void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid);
-
 /**
  *     selinux_ctxid_to_string - map a security context ID to a string
  *     @ctxid: security context ID to be converted.
@@ -166,11 +156,6 @@ static inline void selinux_audit_set_callback(int (*callback)(void))
 	return;
 }
 
-static inline void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid)
-{
-	*ctxid = 0;
-}
-
 static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
 {
        *ctx = NULL;
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 1bd8827a0102..331e17010393 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -385,7 +385,7 @@ static int audit_filter_rules(struct task_struct *tsk,
 			   logged upon error */
 			if (f->se_rule) {
 				if (need_sid) {
-					selinux_task_ctxid(tsk, &sid);
+					selinux_get_task_sid(tsk, &sid);
 					need_sid = 0;
 				}
 				result = selinux_audit_rule_match(sid, f->type,
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index 9d7737db5e51..ee0fb47f81ae 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -21,15 +21,6 @@
 #include "security.h"
 #include "objsec.h"
 
-void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid)
-{
-	struct task_security_struct *tsec = tsk->security;
-	if (selinux_enabled)
-		*ctxid = tsec->sid;
-	else
-		*ctxid = 0;
-}
-
 int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
 {
 	if (selinux_enabled)
-- 
cgit v1.2.1


From 1a70cd40cb291c25b67ec0da715a49d76719329d Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:57 -0700
Subject: [PATCH] selinux: rename selinux_ctxid_to_string

Rename selinux_ctxid_to_string to selinux_sid_to_string to be
consistent with other interfaces.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h    |  8 ++++----
 kernel/audit.c             | 14 +++++++-------
 kernel/auditfilter.c       |  2 +-
 kernel/auditsc.c           |  4 ++--
 security/selinux/exports.c |  4 ++--
 5 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index 79e4707ca772..df9098de4c99 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -70,8 +70,8 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
 void selinux_audit_set_callback(int (*callback)(void));
 
 /**
- *     selinux_ctxid_to_string - map a security context ID to a string
- *     @ctxid: security context ID to be converted.
+ *     selinux_sid_to_string - map a security context ID to a string
+ *     @sid: security context ID to be converted.
  *     @ctx: address of context string to be returned
  *     @ctxlen: length of returned context string.
  *
@@ -79,7 +79,7 @@ void selinux_audit_set_callback(int (*callback)(void));
  *     string will be allocated internally, and the caller must call
  *     kfree() on it after use.
  */
-int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen);
+int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen);
 
 /**
  *     selinux_get_inode_sid - get the inode's security context ID
@@ -156,7 +156,7 @@ static inline void selinux_audit_set_callback(int (*callback)(void))
 	return;
 }
 
-static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
+static inline int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
 {
        *ctx = NULL;
        *ctxlen = 0;
diff --git a/kernel/audit.c b/kernel/audit.c
index 963fd15c9621..f9889ee77825 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -244,7 +244,7 @@ static int audit_set_rate_limit(int limit, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -267,7 +267,7 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -293,7 +293,7 @@ static int audit_set_enabled(int state, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -321,7 +321,7 @@ static int audit_set_failure(int state, uid_t loginuid, u32 sid)
 		char *ctx = NULL;
 		u32 len;
 		int rc;
-		if ((rc = selinux_ctxid_to_string(sid, &ctx, &len)))
+		if ((rc = selinux_sid_to_string(sid, &ctx, &len)))
 			return rc;
 		else
 			audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE,
@@ -538,7 +538,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		if (status_get->mask & AUDIT_STATUS_PID) {
 			int old   = audit_pid;
 			if (sid) {
-				if ((err = selinux_ctxid_to_string(
+				if ((err = selinux_sid_to_string(
 						sid, &ctx, &len)))
 					return err;
 				else
@@ -576,7 +576,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 						 "user pid=%d uid=%u auid=%u",
 						 pid, uid, loginuid);
 				if (sid) {
-					if (selinux_ctxid_to_string(
+					if (selinux_sid_to_string(
 							sid, &ctx, &len)) {
 						audit_log_format(ab, 
 							" ssid=%u", sid);
@@ -614,7 +614,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 					   loginuid, sid);
 		break;
 	case AUDIT_SIGNAL_INFO:
-		err = selinux_ctxid_to_string(audit_sig_sid, &ctx, &len);
+		err = selinux_sid_to_string(audit_sig_sid, &ctx, &len);
 		if (err)
 			return err;
 		sig_data = kmalloc(sizeof(*sig_data) + len, GFP_KERNEL);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index a44879b0c72f..1a58a81fb09d 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1398,7 +1398,7 @@ static void audit_log_rule_change(uid_t loginuid, u32 sid, char *action,
 	if (sid) {
 		char *ctx = NULL;
 		u32 len;
-		if (selinux_ctxid_to_string(sid, &ctx, &len))
+		if (selinux_sid_to_string(sid, &ctx, &len))
 			audit_log_format(ab, " ssid=%u", sid);
 		else
 			audit_log_format(ab, " subj=%s", ctx);
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 331e17010393..fb83c5cb8c32 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -898,7 +898,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 			if (axi->osid != 0) {
 				char *ctx = NULL;
 				u32 len;
-				if (selinux_ctxid_to_string(
+				if (selinux_sid_to_string(
 						axi->osid, &ctx, &len)) {
 					audit_log_format(ab, " osid=%u",
 							axi->osid);
@@ -1005,7 +1005,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
 		if (n->osid != 0) {
 			char *ctx = NULL;
 			u32 len;
-			if (selinux_ctxid_to_string(
+			if (selinux_sid_to_string(
 				n->osid, &ctx, &len)) {
 				audit_log_format(ab, " osid=%u", n->osid);
 				call_panic = 2;
diff --git a/security/selinux/exports.c b/security/selinux/exports.c
index ee0fb47f81ae..b6f96943be1f 100644
--- a/security/selinux/exports.c
+++ b/security/selinux/exports.c
@@ -21,10 +21,10 @@
 #include "security.h"
 #include "objsec.h"
 
-int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen)
+int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen)
 {
 	if (selinux_enabled)
-		return security_sid_to_context(ctxid, ctx, ctxlen);
+		return security_sid_to_context(sid, ctx, ctxlen);
 	else {
 		*ctx = NULL;
 		*ctxlen = 0;
-- 
cgit v1.2.1


From 9a2f44f01a67a6ecca71515af999895b45a2aeb0 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:58 -0700
Subject: [PATCH] selinux: replace ctxid with sid in selinux_audit_rule_match
 interface

Replace ctxid with sid in selinux_audit_rule_match interface for
consistency with other interfaces.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/selinux.h        | 6 +++---
 security/selinux/ss/services.c | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/linux/selinux.h b/include/linux/selinux.h
index df9098de4c99..d1b7ca6c1c57 100644
--- a/include/linux/selinux.h
+++ b/include/linux/selinux.h
@@ -46,7 +46,7 @@ void selinux_audit_rule_free(struct selinux_audit_rule *rule);
 
 /**
  *	selinux_audit_rule_match - determine if a context ID matches a rule.
- *	@ctxid: the context ID to check
+ *	@sid: the context ID to check
  *	@field: the field this rule refers to
  *	@op: the operater the rule uses
  *	@rule: pointer to the audit rule to check against
@@ -55,7 +55,7 @@ void selinux_audit_rule_free(struct selinux_audit_rule *rule);
  *	Returns 1 if the context id matches the rule, 0 if it does not, and
  *	-errno on failure.
  */
-int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                              struct selinux_audit_rule *rule,
                              struct audit_context *actx);
 
@@ -144,7 +144,7 @@ static inline void selinux_audit_rule_free(struct selinux_audit_rule *rule)
 	return;
 }
 
-static inline int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+static inline int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                                            struct selinux_audit_rule *rule,
                                            struct audit_context *actx)
 {
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 22ed17c17718..988079f45294 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2003,7 +2003,7 @@ int selinux_audit_rule_init(u32 field, u32 op, char *rulestr,
 	return rc;
 }
 
-int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
+int selinux_audit_rule_match(u32 sid, u32 field, u32 op,
                              struct selinux_audit_rule *rule,
                              struct audit_context *actx)
 {
@@ -2026,11 +2026,11 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op,
 		goto out;
 	}
 
-	ctxt = sidtab_search(&sidtab, ctxid);
+	ctxt = sidtab_search(&sidtab, sid);
 	if (!ctxt) {
 		audit_log(actx, GFP_ATOMIC, AUDIT_SELINUX_ERR,
 		          "selinux_audit_rule_match: unrecognized SID %d\n",
-		          ctxid);
+		          sid);
 		match = -ENOENT;
 		goto out;
 	}
-- 
cgit v1.2.1


From 016b9bdb81d9c9c7800e4e224ade38d8b37669d3 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:31:58 -0700
Subject: [PATCH] selinux: enable configuration of max policy version

Enable configuration of SELinux maximum supported policy version to support
legacy userland (init) that does not gracefully handle kernels that support
newer policy versions two or more beyond the installed policy, as in FC3
and FC4.

[bunk@stusta.de: improve Kconfig help text]
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Acked-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/Kconfig            | 37 +++++++++++++++++++++++++++++++++++++
 security/selinux/include/security.h |  6 +++++-
 2 files changed, 42 insertions(+), 1 deletion(-)

diff --git a/security/selinux/Kconfig b/security/selinux/Kconfig
index 814ddc42f1f4..5c64c746b062 100644
--- a/security/selinux/Kconfig
+++ b/security/selinux/Kconfig
@@ -124,3 +124,40 @@ config SECURITY_SELINUX_ENABLE_SECMARK_DEFAULT
 
 	  If you are unsure what do do here, select N.
 
+config SECURITY_SELINUX_POLICYDB_VERSION_MAX
+	bool "NSA SELinux maximum supported policy format version"
+	depends on SECURITY_SELINUX
+	default n
+	help
+	  This option enables the maximum policy format version supported
+	  by SELinux to be set to a particular value.  This value is reported
+	  to userspace via /selinux/policyvers and used at policy load time.
+	  It can be adjusted downward to support legacy userland (init) that
+	  does not correctly handle kernels that support newer policy versions.
+
+	  Examples:
+	  For the Fedora Core 3 or 4 Linux distributions, enable this option
+	  and set the value via the next option. For Fedore Core 5 and later,
+	  do not enable this option.
+
+	  If you are unsure how to answer this question, answer N.
+
+config SECURITY_SELINUX_POLICYDB_VERSION_MAX_VALUE
+	int "NSA SELinux maximum supported policy format version value"
+	depends on SECURITY_SELINUX_POLICYDB_VERSION_MAX
+	range 15 20
+	default 19
+	help
+	  This option sets the value for the maximum policy format version
+	  supported by SELinux.
+
+	  Examples:
+	  For Fedora Core 3, use 18.
+	  For Fedora Core 4, use 19.
+
+	  If you are unsure how to answer this question, look for the
+	  policy format version supported by your policy toolchain, by
+	  running 'checkpolicy -V'. Or look at what policy you have
+	  installed under /etc/selinux/$SELINUXTYPE/policy, where
+	  SELINUXTYPE is defined in your /etc/selinux/config.
+
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index 911954a692fa..aa21ca1721af 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -27,7 +27,11 @@
 
 /* Range of policy versions we understand*/
 #define POLICYDB_VERSION_MIN   POLICYDB_VERSION_BASE
-#define POLICYDB_VERSION_MAX   POLICYDB_VERSION_AVTAB
+#ifdef CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX
+#define POLICYDB_VERSION_MAX	CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX_VALUE
+#else
+#define POLICYDB_VERSION_MAX	POLICYDB_VERSION_AVTAB
+#endif
 
 extern int selinux_enabled;
 extern int selinux_mls_enabled;
-- 
cgit v1.2.1


From f3f8771420737004da55159c2f2dc0b6f483a4ef Mon Sep 17 00:00:00 2001
From: Darrel Goeddel <dgoeddel@TrustedCS.com>
Date: Mon, 25 Sep 2006 23:31:59 -0700
Subject: [PATCH] selinux: add support for range transitions on object classes

Introduces support for policy version 21.  This version of the binary
kernel policy allows for defining range transitions on security classes
other than the process security class.  As always, backwards compatibility
for older formats is retained.  The security class is read in as specified
when using the new format, while the "process" security class is assumed
when using an older policy format.

Signed-off-by: Darrel Goeddel <dgoeddel@trustedcs.com>
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Acked-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/Kconfig            |  2 +-
 security/selinux/include/security.h |  3 ++-
 security/selinux/ss/mls.c           | 21 ++++++++++-----------
 security/selinux/ss/policydb.c      | 27 ++++++++++++++++++++-------
 security/selinux/ss/policydb.h      |  7 ++++---
 5 files changed, 37 insertions(+), 23 deletions(-)

diff --git a/security/selinux/Kconfig b/security/selinux/Kconfig
index 5c64c746b062..293dbd6246c1 100644
--- a/security/selinux/Kconfig
+++ b/security/selinux/Kconfig
@@ -145,7 +145,7 @@ config SECURITY_SELINUX_POLICYDB_VERSION_MAX
 config SECURITY_SELINUX_POLICYDB_VERSION_MAX_VALUE
 	int "NSA SELinux maximum supported policy format version value"
 	depends on SECURITY_SELINUX_POLICYDB_VERSION_MAX
-	range 15 20
+	range 15 21
 	default 19
 	help
 	  This option sets the value for the maximum policy format version
diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h
index aa21ca1721af..1ef79172cc8c 100644
--- a/security/selinux/include/security.h
+++ b/security/selinux/include/security.h
@@ -24,13 +24,14 @@
 #define POLICYDB_VERSION_VALIDATETRANS	19
 #define POLICYDB_VERSION_MLS		19
 #define POLICYDB_VERSION_AVTAB		20
+#define POLICYDB_VERSION_RANGETRANS	21
 
 /* Range of policy versions we understand*/
 #define POLICYDB_VERSION_MIN   POLICYDB_VERSION_BASE
 #ifdef CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX
 #define POLICYDB_VERSION_MAX	CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX_VALUE
 #else
-#define POLICYDB_VERSION_MAX	POLICYDB_VERSION_AVTAB
+#define POLICYDB_VERSION_MAX	POLICYDB_VERSION_RANGETRANS
 #endif
 
 extern int selinux_enabled;
diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c
index 119bd6078ba1..c713af23250a 100644
--- a/security/selinux/ss/mls.c
+++ b/security/selinux/ss/mls.c
@@ -530,22 +530,21 @@ int mls_compute_sid(struct context *scontext,
 		    u32 specified,
 		    struct context *newcontext)
 {
+	struct range_trans *rtr;
+
 	if (!selinux_mls_enabled)
 		return 0;
 
 	switch (specified) {
 	case AVTAB_TRANSITION:
-		if (tclass == SECCLASS_PROCESS) {
-			struct range_trans *rangetr;
-			/* Look for a range transition rule. */
-			for (rangetr = policydb.range_tr; rangetr;
-			     rangetr = rangetr->next) {
-				if (rangetr->dom == scontext->type &&
-				    rangetr->type == tcontext->type) {
-					/* Set the range from the rule */
-					return mls_range_set(newcontext,
-					                     &rangetr->range);
-				}
+		/* Look for a range transition rule. */
+		for (rtr = policydb.range_tr; rtr; rtr = rtr->next) {
+			if (rtr->source_type == scontext->type &&
+			    rtr->target_type == tcontext->type &&
+			    rtr->target_class == tclass) {
+				/* Set the range from the rule */
+				return mls_range_set(newcontext,
+				                     &rtr->target_range);
 			}
 		}
 		/* Fallthrough */
diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c
index f03960e697ce..b18895302555 100644
--- a/security/selinux/ss/policydb.c
+++ b/security/selinux/ss/policydb.c
@@ -96,6 +96,11 @@ static struct policydb_compat_info policydb_compat[] = {
 		.sym_num        = SYM_NUM,
 		.ocon_num       = OCON_NUM,
 	},
+	{
+		.version        = POLICYDB_VERSION_RANGETRANS,
+		.sym_num        = SYM_NUM,
+		.ocon_num       = OCON_NUM,
+	},
 };
 
 static struct policydb_compat_info *policydb_lookup_compat(int version)
@@ -645,15 +650,15 @@ void policydb_destroy(struct policydb *p)
 
 	for (rt = p->range_tr; rt; rt = rt -> next) {
 		if (lrt) {
-			ebitmap_destroy(&lrt->range.level[0].cat);
-			ebitmap_destroy(&lrt->range.level[1].cat);
+			ebitmap_destroy(&lrt->target_range.level[0].cat);
+			ebitmap_destroy(&lrt->target_range.level[1].cat);
 			kfree(lrt);
 		}
 		lrt = rt;
 	}
 	if (lrt) {
-		ebitmap_destroy(&lrt->range.level[0].cat);
-		ebitmap_destroy(&lrt->range.level[1].cat);
+		ebitmap_destroy(&lrt->target_range.level[0].cat);
+		ebitmap_destroy(&lrt->target_range.level[1].cat);
 		kfree(lrt);
 	}
 
@@ -1829,6 +1834,7 @@ int policydb_read(struct policydb *p, void *fp)
 	}
 
 	if (p->policyvers >= POLICYDB_VERSION_MLS) {
+		int new_rangetr = p->policyvers >= POLICYDB_VERSION_RANGETRANS;
 		rc = next_entry(buf, fp, sizeof(u32));
 		if (rc < 0)
 			goto bad;
@@ -1847,9 +1853,16 @@ int policydb_read(struct policydb *p, void *fp)
 			rc = next_entry(buf, fp, (sizeof(u32) * 2));
 			if (rc < 0)
 				goto bad;
-			rt->dom = le32_to_cpu(buf[0]);
-			rt->type = le32_to_cpu(buf[1]);
-			rc = mls_read_range_helper(&rt->range, fp);
+			rt->source_type = le32_to_cpu(buf[0]);
+			rt->target_type = le32_to_cpu(buf[1]);
+			if (new_rangetr) {
+				rc = next_entry(buf, fp, sizeof(u32));
+				if (rc < 0)
+					goto bad;
+				rt->target_class = le32_to_cpu(buf[0]);
+			} else
+				rt->target_class = SECCLASS_PROCESS;
+			rc = mls_read_range_helper(&rt->target_range, fp);
 			if (rc)
 				goto bad;
 			lrt = rt;
diff --git a/security/selinux/ss/policydb.h b/security/selinux/ss/policydb.h
index b1340711f721..8319d5ff5944 100644
--- a/security/selinux/ss/policydb.h
+++ b/security/selinux/ss/policydb.h
@@ -106,9 +106,10 @@ struct cat_datum {
 };
 
 struct range_trans {
-	u32 dom;			/* current process domain */
-	u32 type;			/* program executable type */
-	struct mls_range range;		/* new range */
+	u32 source_type;
+	u32 target_type;
+	u32 target_class;
+	struct mls_range target_range;
 	struct range_trans *next;
 };
 
-- 
cgit v1.2.1


From 296fddf7513c155adbd3a443d12add1f62b5cddb Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Mon, 25 Sep 2006 23:32:00 -0700
Subject: [PATCH] SELinux: eliminate inode_security_set_security

inode_security_set_sid is only called by security_inode_init_security, which
is called when a new file is being created and needs to have its incore
security state initialized and its security xattr set.  This helper used to be
called in other places in the past, but now only has the one.  So this patch
rolls inode_security_set_sid directly back into security_inode_init_security.
There also is no need to hold the isec->sem while doing this, as the inode is
not available to other threads at this point in time.

Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/hooks.c | 27 +++++++--------------------
 1 file changed, 7 insertions(+), 20 deletions(-)

diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 5a66c4c09f7a..14a78199ee1f 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -1364,25 +1364,6 @@ static inline u32 file_to_av(struct file *file)
 	return av;
 }
 
-/* Set an inode's SID to a specified value. */
-static int inode_security_set_sid(struct inode *inode, u32 sid)
-{
-	struct inode_security_struct *isec = inode->i_security;
-	struct superblock_security_struct *sbsec = inode->i_sb->s_security;
-
-	if (!sbsec->initialized) {
-		/* Defer initialization to selinux_complete_init. */
-		return 0;
-	}
-
-	down(&isec->sem);
-	isec->sclass = inode_mode_to_security_class(inode->i_mode);
-	isec->sid = sid;
-	isec->initialized = 1;
-	up(&isec->sem);
-	return 0;
-}
-
 /* Hook functions begin here. */
 
 static int selinux_ptrace(struct task_struct *parent, struct task_struct *child)
@@ -2091,7 +2072,13 @@ static int selinux_inode_init_security(struct inode *inode, struct inode *dir,
 		}
 	}
 
-	inode_security_set_sid(inode, newsid);
+	/* Possibly defer initialization to selinux_complete_init. */
+	if (sbsec->initialized) {
+		struct inode_security_struct *isec = inode->i_security;
+		isec->sclass = inode_mode_to_security_class(inode->i_mode);
+		isec->sid = newsid;
+		isec->initialized = 1;
+	}
 
 	if (!ss_initialized || sbsec->behavior == SECURITY_FS_USE_MNTPOINT)
 		return -EOPNOTSUPP;
-- 
cgit v1.2.1


From 23970741720360de9dd0a4e87fbeb1d5927aa474 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Mon, 25 Sep 2006 23:32:01 -0700
Subject: [PATCH] SELinux: change isec semaphore to a mutex

This patch converts the remaining isec->sem into a mutex.  Very similar
locking is provided as before only in the faster smaller mutex rather than a
semaphore.  An out_unlock path is introduced rather than the conditional
unlocking found in the original code.

Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/hooks.c          | 30 ++++++++++++++----------------
 security/selinux/include/objsec.h |  2 +-
 security/selinux/ss/services.c    |  4 ++--
 3 files changed, 17 insertions(+), 19 deletions(-)

diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 14a78199ee1f..63ad57ab44fc 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -71,6 +71,7 @@
 #include <linux/audit.h>
 #include <linux/string.h>
 #include <linux/selinux.h>
+#include <linux/mutex.h>
 
 #include "avc.h"
 #include "objsec.h"
@@ -185,7 +186,7 @@ static int inode_alloc_security(struct inode *inode)
 		return -ENOMEM;
 
 	memset(isec, 0, sizeof(*isec));
-	init_MUTEX(&isec->sem);
+	mutex_init(&isec->lock);
 	INIT_LIST_HEAD(&isec->list);
 	isec->inode = inode;
 	isec->sid = SECINITSID_UNLABELED;
@@ -843,15 +844,13 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 	char *context = NULL;
 	unsigned len = 0;
 	int rc = 0;
-	int hold_sem = 0;
 
 	if (isec->initialized)
 		goto out;
 
-	down(&isec->sem);
-	hold_sem = 1;
+	mutex_lock(&isec->lock);
 	if (isec->initialized)
-		goto out;
+		goto out_unlock;
 
 	sbsec = inode->i_sb->s_security;
 	if (!sbsec->initialized) {
@@ -862,7 +861,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 		if (list_empty(&isec->list))
 			list_add(&isec->list, &sbsec->isec_head);
 		spin_unlock(&sbsec->isec_lock);
-		goto out;
+		goto out_unlock;
 	}
 
 	switch (sbsec->behavior) {
@@ -885,7 +884,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 			printk(KERN_WARNING "%s:  no dentry for dev=%s "
 			       "ino=%ld\n", __FUNCTION__, inode->i_sb->s_id,
 			       inode->i_ino);
-			goto out;
+			goto out_unlock;
 		}
 
 		len = INITCONTEXTLEN;
@@ -893,7 +892,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 		if (!context) {
 			rc = -ENOMEM;
 			dput(dentry);
-			goto out;
+			goto out_unlock;
 		}
 		rc = inode->i_op->getxattr(dentry, XATTR_NAME_SELINUX,
 					   context, len);
@@ -903,7 +902,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 						   NULL, 0);
 			if (rc < 0) {
 				dput(dentry);
-				goto out;
+				goto out_unlock;
 			}
 			kfree(context);
 			len = rc;
@@ -911,7 +910,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 			if (!context) {
 				rc = -ENOMEM;
 				dput(dentry);
-				goto out;
+				goto out_unlock;
 			}
 			rc = inode->i_op->getxattr(dentry,
 						   XATTR_NAME_SELINUX,
@@ -924,7 +923,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 				       "%d for dev=%s ino=%ld\n", __FUNCTION__,
 				       -rc, inode->i_sb->s_id, inode->i_ino);
 				kfree(context);
-				goto out;
+				goto out_unlock;
 			}
 			/* Map ENODATA to the default file SID */
 			sid = sbsec->def_sid;
@@ -960,7 +959,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 					     isec->sclass,
 					     &sid);
 		if (rc)
-			goto out;
+			goto out_unlock;
 		isec->sid = sid;
 		break;
 	case SECURITY_FS_USE_MNTPOINT:
@@ -978,7 +977,7 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 							  isec->sclass,
 							  &sid);
 				if (rc)
-					goto out;
+					goto out_unlock;
 				isec->sid = sid;
 			}
 		}
@@ -987,12 +986,11 @@ static int inode_doinit_with_dentry(struct inode *inode, struct dentry *opt_dent
 
 	isec->initialized = 1;
 
+out_unlock:
+	mutex_unlock(&isec->lock);
 out:
 	if (isec->sclass == SECCLASS_FILE)
 		isec->sclass = inode_mode_to_security_class(inode->i_mode);
-
-	if (hold_sem)
-		up(&isec->sem);
 	return rc;
 }
 
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 0a39bfd1319f..7d5a0289878b 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -44,7 +44,7 @@ struct inode_security_struct {
 	u32 sid;             /* SID of this object */
 	u16 sclass;       /* security class of this object */
 	unsigned char initialized;     /* initialization flag */
-	struct semaphore sem;
+	struct mutex lock;
 	unsigned char inherit;         /* inherit SID from parent entry */
 };
 
diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c
index 988079f45294..0c219a1b3243 100644
--- a/security/selinux/ss/services.c
+++ b/security/selinux/ss/services.c
@@ -2578,7 +2578,7 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
 	sock = SOCKET_I(inode);
 	isec = inode->i_security;
 	sksec = sock->sk->sk_security;
-	down(&isec->sem);
+	mutex_lock(&isec->lock);
 	if (unlikely(sksec->nlbl_state == NLBL_REQUIRE &&
 		     (mask & (MAY_WRITE | MAY_APPEND)))) {
 		lock_sock(sock->sk);
@@ -2586,7 +2586,7 @@ int selinux_netlbl_inode_permission(struct inode *inode, int mask)
 		release_sock(sock->sk);
 	} else
 		rc = 0;
-	up(&isec->sem);
+	mutex_unlock(&isec->lock);
 
 	return rc;
 }
-- 
cgit v1.2.1


From bc7e982b84aceef0a040c88ff659eb5c83818f72 Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Mon, 25 Sep 2006 23:32:02 -0700
Subject: [PATCH] SELinux: convert sbsec semaphore to a mutex

This patch converts the semaphore in the superblock security struct to a
mutex.  No locking changes or other code changes are done.

Signed-off-by: Eric Paris <eparis@redhat.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/hooks.c          | 7 +++----
 security/selinux/include/objsec.h | 2 +-
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 63ad57ab44fc..55cec4d6f117 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -51,7 +51,6 @@
 #include <net/ip.h>		/* for sysctl_local_port_range[] */
 #include <net/tcp.h>		/* struct or_callable used in sock_rcv_skb */
 #include <asm/uaccess.h>
-#include <asm/semaphore.h>
 #include <asm/ioctls.h>
 #include <linux/bitops.h>
 #include <linux/interrupt.h>
@@ -243,7 +242,7 @@ static int superblock_alloc_security(struct super_block *sb)
 	if (!sbsec)
 		return -ENOMEM;
 
-	init_MUTEX(&sbsec->sem);
+	mutex_init(&sbsec->lock);
 	INIT_LIST_HEAD(&sbsec->list);
 	INIT_LIST_HEAD(&sbsec->isec_head);
 	spin_lock_init(&sbsec->isec_lock);
@@ -595,7 +594,7 @@ static int superblock_doinit(struct super_block *sb, void *data)
 	struct inode *inode = root->d_inode;
 	int rc = 0;
 
-	down(&sbsec->sem);
+	mutex_lock(&sbsec->lock);
 	if (sbsec->initialized)
 		goto out;
 
@@ -690,7 +689,7 @@ next_inode:
 	}
 	spin_unlock(&sbsec->isec_lock);
 out:
-	up(&sbsec->sem);
+	mutex_unlock(&sbsec->lock);
 	return rc;
 }
 
diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h
index 7d5a0289878b..ef2267fea8bd 100644
--- a/security/selinux/include/objsec.h
+++ b/security/selinux/include/objsec.h
@@ -63,7 +63,7 @@ struct superblock_security_struct {
 	unsigned int behavior;          /* labeling behavior */
 	unsigned char initialized;      /* initialization flag */
 	unsigned char proc;             /* proc fs */
-	struct semaphore sem;
+	struct mutex lock;
 	struct list_head isec_head;
 	spinlock_t isec_lock;
 };
-- 
cgit v1.2.1


From b20c8122a3204496fca8b5343c93b60fe11dad04 Mon Sep 17 00:00:00 2001
From: Stephen Smalley <sds@tycho.nsa.gov>
Date: Mon, 25 Sep 2006 23:32:03 -0700
Subject: [PATCH] selinux: fix tty locking

Take tty_mutex when accessing ->signal->tty in selinux code.  Noted by Alan
Cox.  Longer term, we are looking at refactoring the code to provide better
encapsulation of the tty layer, but this is a simple fix that addresses the
immediate bug.

Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: Alan Cox <alan@redhat.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 security/selinux/hooks.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 55cec4d6f117..e4d81a42fca4 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -1689,10 +1689,12 @@ static inline void flush_unauthorized_files(struct files_struct * files)
 {
 	struct avc_audit_data ad;
 	struct file *file, *devnull = NULL;
-	struct tty_struct *tty = current->signal->tty;
+	struct tty_struct *tty;
 	struct fdtable *fdt;
 	long j = -1;
 
+	mutex_lock(&tty_mutex);
+	tty = current->signal->tty;
 	if (tty) {
 		file_list_lock();
 		file = list_entry(tty->tty_files.next, typeof(*file), f_u.fu_list);
@@ -1712,6 +1714,7 @@ static inline void flush_unauthorized_files(struct files_struct * files)
 		}
 		file_list_unlock();
 	}
+	mutex_unlock(&tty_mutex);
 
 	/* Revalidate access to inherited open files. */
 
-- 
cgit v1.2.1


From 8d6b5eeea5eb644232cbbbe1c927fdf051e60fa5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:04 -0700
Subject: [PATCH] binfmt_elf: consistently use loff_t

As David Howells <dhowells@redhat.com> points out, binfmt_elf sometimes uses
off_t, sometimes uses loff_t.  Use loff_t throughout.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/binfmt_elf.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 672a3b90bc55..64802aabd1ac 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1262,7 +1262,7 @@ static void fill_elf_header(struct elfhdr *elf, int segs)
 	return;
 }
 
-static void fill_elf_note_phdr(struct elf_phdr *phdr, int sz, off_t offset)
+static void fill_elf_note_phdr(struct elf_phdr *phdr, int sz, loff_t offset)
 {
 	phdr->p_type = PT_NOTE;
 	phdr->p_offset = offset;
@@ -1428,7 +1428,7 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file)
 	int i;
 	struct vm_area_struct *vma;
 	struct elfhdr *elf = NULL;
-	off_t offset = 0, dataoff;
+	loff_t offset = 0, dataoff;
 	unsigned long limit = current->signal->rlim[RLIMIT_CORE].rlim_cur;
 	int numnote;
 	struct memelfnote *notes = NULL;
@@ -1661,11 +1661,11 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file)
 	ELF_CORE_WRITE_EXTRA_DATA;
 #endif
 
-	if ((off_t)file->f_pos != offset) {
+	if (file->f_pos != offset) {
 		/* Sanity check */
 		printk(KERN_WARNING
-		       "elf_core_dump: file->f_pos (%ld) != offset (%ld)\n",
-		       (off_t)file->f_pos, offset);
+		       "elf_core_dump: file->f_pos (%Ld) != offset (%Ld)\n",
+		       file->f_pos, offset);
 	}
 
 end_coredump:
-- 
cgit v1.2.1


From 1bcbba306048ed86b935d57a95d887c23d52c94b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:04 -0700
Subject: [PATCH] FRV: Use the generic IRQ stuff

Make the FRV arch use the generic IRQ code rather than having its own
routines for doing so.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/Kconfig                    |   8 +-
 arch/frv/kernel/Makefile            |   5 +-
 arch/frv/kernel/irq-mb93091.c       | 157 +++++---
 arch/frv/kernel/irq-mb93093.c       | 115 ++++--
 arch/frv/kernel/irq-mb93493.c       | 167 +++++---
 arch/frv/kernel/irq-routing.c       | 291 --------------
 arch/frv/kernel/irq.c               | 750 ++++++------------------------------
 arch/frv/kernel/setup.c             |   1 -
 arch/frv/kernel/time.c              |   1 -
 arch/frv/mb93090-mb00/pci-irq.c     |   1 -
 include/asm-frv/cpu-irqs.h          |  54 ++-
 include/asm-frv/hardirq.h           |   5 +
 include/asm-frv/irq-routing.h       |  70 ----
 include/asm-frv/irq.h               |  26 +-
 include/asm-frv/mb93091-fpga-irqs.h |   6 +-
 include/asm-frv/mb93093-fpga-irqs.h |   6 +-
 include/asm-frv/mb93493-irqs.h      |   6 +-
 include/asm-frv/mb93493-regs.h      |   2 +
 18 files changed, 450 insertions(+), 1221 deletions(-)
 delete mode 100644 arch/frv/kernel/irq-routing.c
 delete mode 100644 include/asm-frv/irq-routing.h

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index a601a17cf568..5833808fb823 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -27,7 +27,7 @@ config GENERIC_CALIBRATE_DELAY
 
 config GENERIC_HARDIRQS
 	bool
-	default n
+	default y
 
 config GENERIC_TIME
 	bool
@@ -251,6 +251,12 @@ config MB93091_NO_MB
 endchoice
 endif
 
+config FUJITSU_MB93493
+	bool "MB93493 Multimedia chip"
+	help
+	  Select this option if the MB93493 multimedia chip is going to be
+	  used.
+
 choice
 	prompt "GP-Relative data support"
 	default GPREL_DATA_8
diff --git a/arch/frv/kernel/Makefile b/arch/frv/kernel/Makefile
index 5a827b349b5e..32db3499c461 100644
--- a/arch/frv/kernel/Makefile
+++ b/arch/frv/kernel/Makefile
@@ -10,15 +10,14 @@ extra-y:= head.o init_task.o vmlinux.lds
 obj-y := $(heads-y) entry.o entry-table.o break.o switch_to.o kernel_thread.o \
 	 process.o traps.o ptrace.o signal.o dma.o \
 	 sys_frv.o time.o semaphore.o setup.o frv_ksyms.o \
-	 debug-stub.o irq.o irq-routing.o sleep.o uaccess.o
+	 debug-stub.o irq.o sleep.o uaccess.o
 
 obj-$(CONFIG_GDBSTUB)		+= gdb-stub.o gdb-io.o
 
 obj-$(CONFIG_MB93091_VDK)	+= irq-mb93091.o
-obj-$(CONFIG_MB93093_PDK)	+= irq-mb93093.o
-obj-$(CONFIG_FUJITSU_MB93493)	+= irq-mb93493.o
 obj-$(CONFIG_PM)		+= pm.o cmode.o
 obj-$(CONFIG_MB93093_PDK)	+= pm-mb93093.o
+obj-$(CONFIG_FUJITSU_MB93493)	+= irq-mb93493.o
 obj-$(CONFIG_SYSCTL)		+= sysctl.o
 obj-$(CONFIG_FUTEX)		+= futex.o
 obj-$(CONFIG_MODULES)		+= module.o
diff --git a/arch/frv/kernel/irq-mb93091.c b/arch/frv/kernel/irq-mb93091.c
index 1381abcd5cc9..635d23437666 100644
--- a/arch/frv/kernel/irq-mb93091.c
+++ b/arch/frv/kernel/irq-mb93091.c
@@ -24,7 +24,6 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 
 #define __reg16(ADDR) (*(volatile unsigned short *)(ADDR))
 
@@ -33,83 +32,129 @@
 #define __get_IFR()	({ __reg16(0xffc0000c); })
 #define __clr_IFR(M)	do { __reg16(0xffc0000c) = ~(M); wmb(); } while(0)
 
-static void frv_fpga_doirq(struct irq_source *source);
-static void frv_fpga_control(struct irq_group *group, int irq, int on);
 
-/*****************************************************************************/
 /*
- * FPGA IRQ multiplexor
+ * on-motherboard FPGA PIC operations
  */
-static struct irq_source frv_fpga[4] = {
-#define __FPGA(X, M)					\
-	[X] = {						\
-		.muxname	= "fpga."#X,		\
-		.irqmask	= M,			\
-		.doirq		= frv_fpga_doirq,	\
-	}
+static void frv_fpga_enable(unsigned int irq)
+{
+	uint16_t imr = __get_IMR();
 
-	__FPGA(0, 0x0028),
-	__FPGA(1, 0x0050),
-	__FPGA(2, 0x1c00),
-	__FPGA(3, 0x6386),
-};
-
-static struct irq_group frv_fpga_irqs = {
-	.first_irq	= IRQ_BASE_FPGA,
-	.control	= frv_fpga_control,
-	.sources = {
-		[ 1] = &frv_fpga[3],
-		[ 2] = &frv_fpga[3],
-		[ 3] = &frv_fpga[0],
-		[ 4] = &frv_fpga[1],
-		[ 5] = &frv_fpga[0],
-		[ 6] = &frv_fpga[1],
-		[ 7] = &frv_fpga[3],
-		[ 8] = &frv_fpga[3],
-		[ 9] = &frv_fpga[3],
-		[10] = &frv_fpga[2],
-		[11] = &frv_fpga[2],
-		[12] = &frv_fpga[2],
-		[13] = &frv_fpga[3],
-		[14] = &frv_fpga[3],
-	},
-};
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
 
+	__set_IMR(imr);
+}
 
-static void frv_fpga_control(struct irq_group *group, int index, int on)
+static void frv_fpga_disable(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	if (on)
-		imr &= ~(1 << index);
-	else
-		imr |= 1 << index;
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 
 	__set_IMR(imr);
 }
 
-static void frv_fpga_doirq(struct irq_source *source)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_end(unsigned int irq)
+{
+}
+
+static struct irq_chip frv_fpga_pic = {
+	.name		= "mb93091",
+	.enable		= frv_fpga_enable,
+	.disable	= frv_fpga_disable,
+	.ack		= frv_fpga_ack,
+	.mask		= frv_fpga_disable,
+	.unmask		= frv_fpga_enable,
+	.end		= frv_fpga_end,
+};
+
+/*
+ * FPGA PIC interrupt handler
+ */
+static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
-	uint16_t mask, imr;
+	uint16_t imr, mask = (unsigned long) _mask;
+	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
-	mask = source->irqmask & ~imr & __get_IFR();
-	if (mask) {
-		__set_IMR(imr | mask);
-		__clr_IFR(mask);
-		distribute_irqs(&frv_fpga_irqs, mask);
-		__set_IMR(imr);
+	mask = mask & ~imr & __get_IFR();
+
+	/* poll all the triggered IRQs */
+	while (mask) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(mask));
+		irq = 31 - irq;
+		mask &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
+			iret |= IRQ_HANDLED;
 	}
+
+	return iret;
 }
 
+/*
+ * define an interrupt action for each FPGA PIC output
+ * - use dev_id to indicate the FPGA PIC input to output mappings
+ */
+static struct irqaction fpga_irq[4]  = {
+	[0] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.0",
+		.dev_id		= (void *) 0x0028UL,
+	},
+	[1] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.1",
+		.dev_id		= (void *) 0x0050UL,
+	},
+	[2] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.2",
+		.dev_id		= (void *) 0x1c00UL,
+	},
+	[3] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.3",
+		.dev_id		= (void *) 0x6386UL,
+	}
+};
+
+/*
+ * initialise the motherboard FPGA's PIC
+ */
 void __init fpga_init(void)
 {
+	int irq;
+
+	/* all PIC inputs are all set to be low-level driven, apart from the
+	 * NMI button (15) which is fixed at falling-edge
+	 */
 	__set_IMR(0x7ffe);
 	__clr_IFR(0x0000);
 
-	frv_irq_route_external(&frv_fpga[0], IRQ_CPU_EXTERNAL0);
-	frv_irq_route_external(&frv_fpga[1], IRQ_CPU_EXTERNAL1);
-	frv_irq_route_external(&frv_fpga[2], IRQ_CPU_EXTERNAL2);
-	frv_irq_route_external(&frv_fpga[3], IRQ_CPU_EXTERNAL3);
-	frv_irq_set_group(&frv_fpga_irqs);
+	for (irq = IRQ_BASE_FPGA + 1; irq <= IRQ_BASE_FPGA + 14; irq++)
+		set_irq_chip_and_handler(irq, &frv_fpga_pic, handle_level_irq);
+
+	set_irq_chip_and_handler(IRQ_FPGA_NMI, &frv_fpga_pic, handle_edge_irq);
+
+	/* the FPGA drives the first four external IRQ inputs on the CPU PIC */
+	setup_irq(IRQ_CPU_EXTERNAL0, &fpga_irq[0]);
+	setup_irq(IRQ_CPU_EXTERNAL1, &fpga_irq[1]);
+	setup_irq(IRQ_CPU_EXTERNAL2, &fpga_irq[2]);
+	setup_irq(IRQ_CPU_EXTERNAL3, &fpga_irq[3]);
 }
diff --git a/arch/frv/kernel/irq-mb93093.c b/arch/frv/kernel/irq-mb93093.c
index 48b2a6420888..f60db7988e78 100644
--- a/arch/frv/kernel/irq-mb93093.c
+++ b/arch/frv/kernel/irq-mb93093.c
@@ -1,6 +1,6 @@
 /* irq-mb93093.c: MB93093 FPGA interrupt handling
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -24,7 +24,6 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 
 #define __reg16(ADDR) (*(volatile unsigned short *)(__region_CS2 + (ADDR)))
 
@@ -33,66 +32,100 @@
 #define __get_IFR()	({ __reg16(0x02); })
 #define __clr_IFR(M)	do { __reg16(0x02) = ~(M); wmb(); } while(0)
 
-static void frv_fpga_doirq(struct irq_source *source);
-static void frv_fpga_control(struct irq_group *group, int irq, int on);
-
-/*****************************************************************************/
 /*
- * FPGA IRQ multiplexor
+ * off-CPU FPGA PIC operations
  */
-static struct irq_source frv_fpga[4] = {
-#define __FPGA(X, M)					\
-	[X] = {						\
-		.muxname	= "fpga."#X,		\
-		.irqmask	= M,			\
-		.doirq		= frv_fpga_doirq,	\
-	}
-
-	__FPGA(0, 0x0700),
-};
+static void frv_fpga_enable(unsigned int irq)
+{
+	uint16_t imr = __get_IMR();
 
-static struct irq_group frv_fpga_irqs = {
-	.first_irq	= IRQ_BASE_FPGA,
-	.control	= frv_fpga_control,
-	.sources = {
-		[ 8] = &frv_fpga[0],
-		[ 9] = &frv_fpga[0],
-		[10] = &frv_fpga[0],
-	},
-};
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
 
+	__set_IMR(imr);
+}
 
-static void frv_fpga_control(struct irq_group *group, int index, int on)
+static void frv_fpga_disable(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	if (on)
-		imr &= ~(1 << index);
-	else
-		imr |= 1 << index;
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 
 	__set_IMR(imr);
 }
 
-static void frv_fpga_doirq(struct irq_source *source)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_end(unsigned int irq)
+{
+}
+
+static struct irq_chip frv_fpga_pic = {
+	.name		= "mb93093",
+	.enable		= frv_fpga_enable,
+	.disable	= frv_fpga_disable,
+	.ack		= frv_fpga_ack,
+	.mask		= frv_fpga_disable,
+	.unmask		= frv_fpga_enable,
+	.end		= frv_fpga_end,
+};
+
+/*
+ * FPGA PIC interrupt handler
+ */
+static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
-	uint16_t mask, imr;
+	uint16_t imr, mask = (unsigned long) _mask;
+	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
-	mask = source->irqmask & ~imr & __get_IFR();
-	if (mask) {
-		__set_IMR(imr | mask);
-		__clr_IFR(mask);
-		distribute_irqs(&frv_fpga_irqs, mask);
-		__set_IMR(imr);
+	mask = mask & ~imr & __get_IFR();
+
+	/* poll all the triggered IRQs */
+	while (mask) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(mask));
+		irq = 31 - irq;
+		mask &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
+			iret |= IRQ_HANDLED;
 	}
+
+	return iret;
 }
 
+/*
+ * define an interrupt action for each FPGA PIC output
+ * - use dev_id to indicate the FPGA PIC input to output mappings
+ */
+static struct irqaction fpga_irq[1]  = {
+	[0] = {
+		.handler	= fpga_interrupt,
+		.flags		= IRQF_DISABLED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "fpga.0",
+		.dev_id		= (void *) 0x0700UL,
+	}
+};
+
+/*
+ * initialise the motherboard FPGA's PIC
+ */
 void __init fpga_init(void)
 {
+	int irq;
+
+	/* all PIC inputs are all set to be edge triggered */
 	__set_IMR(0x0700);
 	__clr_IFR(0x0000);
 
-	frv_irq_route_external(&frv_fpga[0], IRQ_CPU_EXTERNAL2);
-	frv_irq_set_group(&frv_fpga_irqs);
+	for (irq = IRQ_BASE_FPGA + 8; irq <= IRQ_BASE_FPGA + 10; irq++)
+		set_irq_chip_and_handler(irq, &frv_fpga_pic, handle_edge_irq);
+
+	/* the FPGA drives external IRQ input #2 on the CPU PIC */
+	setup_irq(IRQ_CPU_EXTERNAL2, &fpga_irq[0]);
 }
diff --git a/arch/frv/kernel/irq-mb93493.c b/arch/frv/kernel/irq-mb93493.c
index 988d035640e1..8ad9abfc7c13 100644
--- a/arch/frv/kernel/irq-mb93493.c
+++ b/arch/frv/kernel/irq-mb93493.c
@@ -1,6 +1,6 @@
 /* irq-mb93493.c: MB93493 companion chip interrupt handler
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -24,84 +24,133 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 #include <asm/mb93493-irqs.h>
+#include <asm/mb93493-regs.h>
 
-static void frv_mb93493_doirq(struct irq_source *source);
+#define IRQ_ROUTE_ONE(X) (X##_ROUTE << (X - IRQ_BASE_MB93493))
+
+#define IRQ_ROUTING					\
+	(IRQ_ROUTE_ONE(IRQ_MB93493_VDC)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_VCC)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_AUDIO_OUT)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_I2C_0)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_I2C_1)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_USB)		|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_LOCAL_BUS)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_PCMCIA)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_GPIO)	|	\
+	 IRQ_ROUTE_ONE(IRQ_MB93493_AUDIO_IN))
 
-/*****************************************************************************/
 /*
- * MB93493 companion chip IRQ multiplexor
+ * daughter board PIC operations
  */
-static struct irq_source frv_mb93493[2] = {
-	[0] = {
-		.muxname		= "mb93493.0",
-		.muxdata		= __region_CS3 + 0x3d0,
-		.doirq			= frv_mb93493_doirq,
-		.irqmask		= 0x0000,
-	},
-	[1] = {
-		.muxname		= "mb93493.1",
-		.muxdata		= __region_CS3 + 0x3d4,
-		.doirq			= frv_mb93493_doirq,
-		.irqmask		= 0x0000,
-	},
-};
-
-static void frv_mb93493_control(struct irq_group *group, int index, int on)
+static void frv_mb93493_enable(unsigned int irq)
 {
-	struct irq_source *source;
 	uint32_t iqsr;
+	volatile void *piqsr;
 
-	if ((frv_mb93493[0].irqmask & (1 << index)))
-		source = &frv_mb93493[0];
+	if (IRQ_ROUTING & (1 << (irq - IRQ_BASE_MB93493)))
+		piqsr = __addr_MB93493_IQSR(1);
 	else
-		source = &frv_mb93493[1];
+		piqsr = __addr_MB93493_IQSR(0);
 
-	iqsr = readl(source->muxdata);
-	if (on)
-		iqsr |= 1 << (index + 16);
+	iqsr = readl(piqsr);
+	iqsr |= 1 << (irq - IRQ_BASE_MB93493 + 16);
+	writel(iqsr, piqsr);
+}
+
+static void frv_mb93493_disable(unsigned int irq)
+{
+	uint32_t iqsr;
+	volatile void *piqsr;
+
+	if (IRQ_ROUTING & (1 << (irq - IRQ_BASE_MB93493)))
+		piqsr = __addr_MB93493_IQSR(1);
 	else
-		iqsr &= ~(1 << (index + 16));
+		piqsr = __addr_MB93493_IQSR(0);
 
-	writel(iqsr, source->muxdata);
+	iqsr = readl(piqsr);
+	iqsr &= ~(1 << (irq - IRQ_BASE_MB93493 + 16));
+	writel(iqsr, piqsr);
 }
 
-static struct irq_group frv_mb93493_irqs = {
-	.first_irq	= IRQ_BASE_MB93493,
-	.control	= frv_mb93493_control,
-};
-
-static void frv_mb93493_doirq(struct irq_source *source)
+static void frv_mb93493_ack(unsigned int irq)
 {
-	uint32_t mask = readl(source->muxdata);
-	mask = mask & (mask >> 16) & 0xffff;
+}
 
-	if (mask)
-		distribute_irqs(&frv_mb93493_irqs, mask);
+static void frv_mb93493_end(unsigned int irq)
+{
 }
 
-static void __init mb93493_irq_route(int irq, int source)
+static struct irq_chip frv_mb93493_pic = {
+	.name		= "mb93093",
+	.enable		= frv_mb93493_enable,
+	.disable	= frv_mb93493_disable,
+	.ack		= frv_mb93493_ack,
+	.mask		= frv_mb93493_disable,
+	.unmask		= frv_mb93493_enable,
+	.end		= frv_mb93493_end,
+};
+
+/*
+ * MB93493 PIC interrupt handler
+ */
+static irqreturn_t mb93493_interrupt(int irq, void *_piqsr, struct pt_regs *regs)
 {
-	frv_mb93493[source].irqmask |= 1 << (irq - IRQ_BASE_MB93493);
-	frv_mb93493_irqs.sources[irq - IRQ_BASE_MB93493] = &frv_mb93493[source];
+	volatile void *piqsr = _piqsr;
+	irqreturn_t iret = 0;
+	uint32_t iqsr;
+
+	iqsr = readl(piqsr);
+	iqsr = iqsr & (iqsr >> 16) & 0xffff;
+
+	/* poll all the triggered IRQs */
+	while (iqsr) {
+		int irq;
+
+		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(iqsr));
+		irq = 31 - irq;
+		iqsr &= ~(1 << irq);
+
+		if (__do_IRQ(IRQ_BASE_MB93493 + irq, regs))
+			iret |= IRQ_HANDLED;
+	}
+
+	return iret;
 }
 
-void __init route_mb93493_irqs(void)
+/*
+ * define an interrupt action for each MB93493 PIC output
+ * - use dev_id to indicate the MB93493 PIC input to output mappings
+ */
+static struct irqaction mb93493_irq[2]  = {
+	[0] = {
+		.handler	= mb93493_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "mb93493.0",
+		.dev_id		= (void *) __addr_MB93493_IQSR(0),
+	},
+	[1] = {
+		.handler	= mb93493_interrupt,
+		.flags		= IRQF_DISABLED | IRQF_SHARED,
+		.mask		= CPU_MASK_NONE,
+		.name		= "mb93493.1",
+		.dev_id		= (void *) __addr_MB93493_IQSR(1),
+	}
+};
+
+/*
+ * initialise the motherboard MB93493's PIC
+ */
+void __init mb93493_init(void)
 {
-	frv_irq_route_external(&frv_mb93493[0], IRQ_CPU_MB93493_0);
-	frv_irq_route_external(&frv_mb93493[1], IRQ_CPU_MB93493_1);
-
-	frv_irq_set_group(&frv_mb93493_irqs);
-
-	mb93493_irq_route(IRQ_MB93493_VDC,		IRQ_MB93493_VDC_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_VCC,		IRQ_MB93493_VCC_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_AUDIO_IN,		IRQ_MB93493_AUDIO_IN_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_I2C_0,		IRQ_MB93493_I2C_0_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_I2C_1,		IRQ_MB93493_I2C_1_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_USB,		IRQ_MB93493_USB_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_LOCAL_BUS,	IRQ_MB93493_LOCAL_BUS_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_PCMCIA,		IRQ_MB93493_PCMCIA_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_GPIO,		IRQ_MB93493_GPIO_ROUTE);
-	mb93493_irq_route(IRQ_MB93493_AUDIO_OUT,	IRQ_MB93493_AUDIO_OUT_ROUTE);
+	int irq;
+
+	for (irq = IRQ_BASE_MB93493 + 0; irq <= IRQ_BASE_MB93493 + 10; irq++)
+		set_irq_chip_and_handler(irq, &frv_mb93493_pic, handle_edge_irq);
+
+	/* the MB93493 drives external IRQ inputs on the CPU PIC */
+	setup_irq(IRQ_CPU_MB93493_0, &mb93493_irq[0]);
+	setup_irq(IRQ_CPU_MB93493_1, &mb93493_irq[1]);
 }
diff --git a/arch/frv/kernel/irq-routing.c b/arch/frv/kernel/irq-routing.c
deleted file mode 100644
index 53886adf47de..000000000000
--- a/arch/frv/kernel/irq-routing.c
+++ /dev/null
@@ -1,291 +0,0 @@
-/* irq-routing.c: IRQ routing
- *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#include <linux/sched.h>
-#include <linux/random.h>
-#include <linux/init.h>
-#include <linux/serial_reg.h>
-#include <asm/io.h>
-#include <asm/irq-routing.h>
-#include <asm/irc-regs.h>
-#include <asm/serial-regs.h>
-#include <asm/dma.h>
-
-struct irq_level frv_irq_levels[16] = {
-	[0 ... 15] = {
-		.lock	= SPIN_LOCK_UNLOCKED,
-	}
-};
-
-struct irq_group *irq_groups[NR_IRQ_GROUPS];
-
-extern struct irq_group frv_cpu_irqs;
-
-void __init frv_irq_route(struct irq_source *source, int irqlevel)
-{
-	source->level = &frv_irq_levels[irqlevel];
-	source->next = frv_irq_levels[irqlevel].sources;
-	frv_irq_levels[irqlevel].sources = source;
-}
-
-void __init frv_irq_route_external(struct irq_source *source, int irq)
-{
-	int irqlevel = 0;
-
-	switch (irq) {
-	case IRQ_CPU_EXTERNAL0:	irqlevel = IRQ_XIRQ0_LEVEL; break;
-	case IRQ_CPU_EXTERNAL1:	irqlevel = IRQ_XIRQ1_LEVEL; break;
-	case IRQ_CPU_EXTERNAL2:	irqlevel = IRQ_XIRQ2_LEVEL; break;
-	case IRQ_CPU_EXTERNAL3:	irqlevel = IRQ_XIRQ3_LEVEL; break;
-	case IRQ_CPU_EXTERNAL4:	irqlevel = IRQ_XIRQ4_LEVEL; break;
-	case IRQ_CPU_EXTERNAL5:	irqlevel = IRQ_XIRQ5_LEVEL; break;
-	case IRQ_CPU_EXTERNAL6:	irqlevel = IRQ_XIRQ6_LEVEL; break;
-	case IRQ_CPU_EXTERNAL7:	irqlevel = IRQ_XIRQ7_LEVEL; break;
-	default: BUG();
-	}
-
-	source->level = &frv_irq_levels[irqlevel];
-	source->next = frv_irq_levels[irqlevel].sources;
-	frv_irq_levels[irqlevel].sources = source;
-}
-
-void __init frv_irq_set_group(struct irq_group *group)
-{
-	irq_groups[group->first_irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP] = group;
-}
-
-void distribute_irqs(struct irq_group *group, unsigned long irqmask)
-{
-	struct irqaction *action;
-	int irq;
-
-	while (irqmask) {
-		asm("scan %1,gr0,%0" : "=r"(irq) : "r"(irqmask));
-		if (irq < 0 || irq > 31)
-			asm volatile("break");
-		irq = 31 - irq;
-
-		irqmask &= ~(1 << irq);
-		action = group->actions[irq];
-
-		irq += group->first_irq;
-
-		if (action) {
-			int status = 0;
-
-//			if (!(action->flags & IRQF_DISABLED))
-//				local_irq_enable();
-
-			do {
-				status |= action->flags;
-				action->handler(irq, action->dev_id, __frame);
-				action = action->next;
-			} while (action);
-
-			if (status & IRQF_SAMPLE_RANDOM)
-				add_interrupt_randomness(irq);
-			local_irq_disable();
-		}
-	}
-}
-
-/*****************************************************************************/
-/*
- * CPU UART interrupts
- */
-static void frv_cpuuart_doirq(struct irq_source *source)
-{
-//	uint8_t iir = readb(source->muxdata + UART_IIR * 8);
-//	if ((iir & 0x0f) != UART_IIR_NO_INT)
-		distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpuuart[2] = {
-#define __CPUUART(X, A)						\
-	[X] = {							\
-		.muxname	= "uart",			\
-		.muxdata	= (volatile void __iomem *)(unsigned long)A,\
-		.irqmask	= 1 << IRQ_CPU_UART##X,		\
-		.doirq		= frv_cpuuart_doirq,		\
-	}
-
-	__CPUUART(0, UART0_BASE),
-	__CPUUART(1, UART1_BASE),
-};
-
-/*****************************************************************************/
-/*
- * CPU DMA interrupts
- */
-static void frv_cpudma_doirq(struct irq_source *source)
-{
-	uint32_t cstr = readl(source->muxdata + DMAC_CSTRx);
-	if (cstr & DMAC_CSTRx_INT)
-		distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpudma[8] = {
-#define __CPUDMA(X, A)						\
-	[X] = {							\
-		.muxname	= "dma",			\
-		.muxdata	= (volatile void __iomem *)(unsigned long)A,\
-		.irqmask	= 1 << IRQ_CPU_DMA##X,		\
-		.doirq		= frv_cpudma_doirq,		\
-	}
-
-	__CPUDMA(0, 0xfe000900),
-	__CPUDMA(1, 0xfe000980),
-	__CPUDMA(2, 0xfe000a00),
-	__CPUDMA(3, 0xfe000a80),
-	__CPUDMA(4, 0xfe001000),
-	__CPUDMA(5, 0xfe001080),
-	__CPUDMA(6, 0xfe001100),
-	__CPUDMA(7, 0xfe001180),
-};
-
-/*****************************************************************************/
-/*
- * CPU timer interrupts - can't tell whether they've generated an interrupt or not
- */
-static void frv_cputimer_doirq(struct irq_source *source)
-{
-	distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cputimer[3] = {
-#define __CPUTIMER(X)						\
-	[X] = {							\
-		.muxname	= "timer",			\
-		.muxdata	= NULL,				\
-		.irqmask	= 1 << IRQ_CPU_TIMER##X,	\
-		.doirq		= frv_cputimer_doirq,		\
-	}
-
-	__CPUTIMER(0),
-	__CPUTIMER(1),
-	__CPUTIMER(2),
-};
-
-/*****************************************************************************/
-/*
- * external CPU interrupts - can't tell directly whether they've generated an interrupt or not
- */
-static void frv_cpuexternal_doirq(struct irq_source *source)
-{
-	distribute_irqs(&frv_cpu_irqs, source->irqmask);
-}
-
-struct irq_source frv_cpuexternal[8] = {
-#define __CPUEXTERNAL(X)					\
-	[X] = {							\
-		.muxname	= "ext",			\
-		.muxdata	= NULL,				\
-		.irqmask	= 1 << IRQ_CPU_EXTERNAL##X,	\
-		.doirq		= frv_cpuexternal_doirq,	\
-	}
-
-	__CPUEXTERNAL(0),
-	__CPUEXTERNAL(1),
-	__CPUEXTERNAL(2),
-	__CPUEXTERNAL(3),
-	__CPUEXTERNAL(4),
-	__CPUEXTERNAL(5),
-	__CPUEXTERNAL(6),
-	__CPUEXTERNAL(7),
-};
-
-#define set_IRR(N,A,B,C,D) __set_IRR(N, (A << 28) | (B << 24) | (C << 20) | (D << 16))
-
-struct irq_group frv_cpu_irqs = {
-	.sources = {
-		[IRQ_CPU_UART0]		= &frv_cpuuart[0],
-		[IRQ_CPU_UART1]		= &frv_cpuuart[1],
-		[IRQ_CPU_TIMER0]	= &frv_cputimer[0],
-		[IRQ_CPU_TIMER1]	= &frv_cputimer[1],
-		[IRQ_CPU_TIMER2]	= &frv_cputimer[2],
-		[IRQ_CPU_DMA0]		= &frv_cpudma[0],
-		[IRQ_CPU_DMA1]		= &frv_cpudma[1],
-		[IRQ_CPU_DMA2]		= &frv_cpudma[2],
-		[IRQ_CPU_DMA3]		= &frv_cpudma[3],
-		[IRQ_CPU_DMA4]		= &frv_cpudma[4],
-		[IRQ_CPU_DMA5]		= &frv_cpudma[5],
-		[IRQ_CPU_DMA6]		= &frv_cpudma[6],
-		[IRQ_CPU_DMA7]		= &frv_cpudma[7],
-		[IRQ_CPU_EXTERNAL0]	= &frv_cpuexternal[0],
-		[IRQ_CPU_EXTERNAL1]	= &frv_cpuexternal[1],
-		[IRQ_CPU_EXTERNAL2]	= &frv_cpuexternal[2],
-		[IRQ_CPU_EXTERNAL3]	= &frv_cpuexternal[3],
-		[IRQ_CPU_EXTERNAL4]	= &frv_cpuexternal[4],
-		[IRQ_CPU_EXTERNAL5]	= &frv_cpuexternal[5],
-		[IRQ_CPU_EXTERNAL6]	= &frv_cpuexternal[6],
-		[IRQ_CPU_EXTERNAL7]	= &frv_cpuexternal[7],
-	},
-};
-
-/*****************************************************************************/
-/*
- * route the CPU's interrupt sources
- */
-void __init route_cpu_irqs(void)
-{
-	frv_irq_set_group(&frv_cpu_irqs);
-
-	__set_IITMR(0, 0x003f0000);	/* DMA0-3, TIMER0-2 IRQ detect levels */
-	__set_IITMR(1, 0x20000000);	/* ERR0-1, UART0-1, DMA4-7 IRQ detect levels */
-
-	/* route UART and error interrupts */
-	frv_irq_route(&frv_cpuuart[0],	IRQ_UART0_LEVEL);
-	frv_irq_route(&frv_cpuuart[1],	IRQ_UART1_LEVEL);
-
-	set_IRR(6, IRQ_GDBSTUB_LEVEL, IRQ_GDBSTUB_LEVEL, IRQ_UART1_LEVEL, IRQ_UART0_LEVEL);
-
-	/* route DMA channel interrupts */
-	frv_irq_route(&frv_cpudma[0],	IRQ_DMA0_LEVEL);
-	frv_irq_route(&frv_cpudma[1],	IRQ_DMA1_LEVEL);
-	frv_irq_route(&frv_cpudma[2],	IRQ_DMA2_LEVEL);
-	frv_irq_route(&frv_cpudma[3],	IRQ_DMA3_LEVEL);
-	frv_irq_route(&frv_cpudma[4],	IRQ_DMA4_LEVEL);
-	frv_irq_route(&frv_cpudma[5],	IRQ_DMA5_LEVEL);
-	frv_irq_route(&frv_cpudma[6],	IRQ_DMA6_LEVEL);
-	frv_irq_route(&frv_cpudma[7],	IRQ_DMA7_LEVEL);
-
-	set_IRR(4, IRQ_DMA3_LEVEL, IRQ_DMA2_LEVEL, IRQ_DMA1_LEVEL, IRQ_DMA0_LEVEL);
-	set_IRR(7, IRQ_DMA7_LEVEL, IRQ_DMA6_LEVEL, IRQ_DMA5_LEVEL, IRQ_DMA4_LEVEL);
-
-	/* route timer interrupts */
-	frv_irq_route(&frv_cputimer[0],	IRQ_TIMER0_LEVEL);
-	frv_irq_route(&frv_cputimer[1],	IRQ_TIMER1_LEVEL);
-	frv_irq_route(&frv_cputimer[2],	IRQ_TIMER2_LEVEL);
-
-	set_IRR(5, 0, IRQ_TIMER2_LEVEL, IRQ_TIMER1_LEVEL, IRQ_TIMER0_LEVEL);
-
-	/* route external interrupts */
-	frv_irq_route(&frv_cpuexternal[0], IRQ_XIRQ0_LEVEL);
-	frv_irq_route(&frv_cpuexternal[1], IRQ_XIRQ1_LEVEL);
-	frv_irq_route(&frv_cpuexternal[2], IRQ_XIRQ2_LEVEL);
-	frv_irq_route(&frv_cpuexternal[3], IRQ_XIRQ3_LEVEL);
-	frv_irq_route(&frv_cpuexternal[4], IRQ_XIRQ4_LEVEL);
-	frv_irq_route(&frv_cpuexternal[5], IRQ_XIRQ5_LEVEL);
-	frv_irq_route(&frv_cpuexternal[6], IRQ_XIRQ6_LEVEL);
-	frv_irq_route(&frv_cpuexternal[7], IRQ_XIRQ7_LEVEL);
-
-	set_IRR(2, IRQ_XIRQ7_LEVEL, IRQ_XIRQ6_LEVEL, IRQ_XIRQ5_LEVEL, IRQ_XIRQ4_LEVEL);
-	set_IRR(3, IRQ_XIRQ3_LEVEL, IRQ_XIRQ2_LEVEL, IRQ_XIRQ1_LEVEL, IRQ_XIRQ0_LEVEL);
-
-#if defined(CONFIG_MB93091_VDK)
-	__set_TM1(0x55550000);		/* XIRQ7-0 all active low */
-#elif defined(CONFIG_MB93093_PDK)
-	__set_TM1(0x15550000);		/* XIRQ7 active high, 6-0 all active low */
-#else
-#error dont know external IRQ trigger levels for this setup
-#endif
-
-} /* end route_cpu_irqs() */
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index 08967010be04..e1ab9f2e43fb 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -1,6 +1,6 @@
 /* irq.c: FRV IRQ handling
  *
- * Copyright (C) 2003, 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2003, 2004, 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -9,13 +9,6 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-/*
- * (mostly architecture independent, will move to kernel/irq.c in 2.5.)
- *
- * IRQs are in fact implemented a bit like signal handlers for the kernel.
- * Naturally it's not a 1:1 relation, but there are similarities.
- */
-
 #include <linux/ptrace.h>
 #include <linux/errno.h>
 #include <linux/signal.h>
@@ -43,19 +36,16 @@
 #include <asm/delay.h>
 #include <asm/irq.h>
 #include <asm/irc-regs.h>
-#include <asm/irq-routing.h>
 #include <asm/gdb-stub.h>
 
-extern void __init fpga_init(void);
-extern void __init route_mb93493_irqs(void);
-
-static void register_irq_proc (unsigned int irq);
+#define set_IRR(N,A,B,C,D) __set_IRR(N, (A << 28) | (B << 24) | (C << 20) | (D << 16))
 
-/*
- * Special irq handlers.
- */
+extern void __init fpga_init(void);
+#ifdef CONFIG_FUJITSU_MB93493
+extern void __init mb93493_init(void);
+#endif
 
-irqreturn_t no_action(int cpl, void *dev_id, struct pt_regs *regs) { return IRQ_HANDLED; }
+#define __reg16(ADDR) (*(volatile unsigned short *)(ADDR))
 
 atomic_t irq_err_count;
 
@@ -64,215 +54,99 @@ atomic_t irq_err_count;
  */
 int show_interrupts(struct seq_file *p, void *v)
 {
-	struct irqaction *action;
-	struct irq_group *group;
+	int i = *(loff_t *) v, cpu;
+	struct irqaction * action;
 	unsigned long flags;
-	int level, grp, ix, i, j;
-
-	i = *(loff_t *) v;
-
-	switch (i) {
-	case 0:
-		seq_printf(p, "           ");
-		for_each_online_cpu(j)
-			seq_printf(p, "CPU%d       ",j);
-
-		seq_putc(p, '\n');
-		break;
 
-	case 1 ... NR_IRQ_GROUPS * NR_IRQ_ACTIONS_PER_GROUP:
-		local_irq_save(flags);
-
-		grp = (i - 1) / NR_IRQ_ACTIONS_PER_GROUP;
-		group = irq_groups[grp];
-		if (!group)
-			goto skip;
-
-		ix = (i - 1) % NR_IRQ_ACTIONS_PER_GROUP;
-		action = group->actions[ix];
-		if (!action)
-			goto skip;
-
-		seq_printf(p, "%3d: ", i - 1);
-
-#ifndef CONFIG_SMP
-		seq_printf(p, "%10u ", kstat_irqs(i));
-#else
-		for_each_online_cpu(j)
-			seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
-#endif
-
-		level = group->sources[ix]->level - frv_irq_levels;
-
-		seq_printf(p, " %12s@%x", group->sources[ix]->muxname, level);
-		seq_printf(p, "  %s", action->name);
-
-		for (action = action->next; action; action = action->next)
-			seq_printf(p, ", %s", action->name);
+	if (i == 0) {
+		char cpuname[12];
 
+		seq_printf(p, "    ");
+		for_each_present_cpu(cpu) {
+			sprintf(cpuname, "CPU%d", cpu);
+			seq_printf(p, " %10s", cpuname);
+		}
 		seq_putc(p, '\n');
-skip:
-		local_irq_restore(flags);
-		break;
+	}
 
-	case NR_IRQ_GROUPS * NR_IRQ_ACTIONS_PER_GROUP + 1:
-		seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
-		break;
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (action) {
+			seq_printf(p, "%3d: ", i);
+			for_each_present_cpu(cpu)
+				seq_printf(p, "%10u ", kstat_cpu(cpu).irqs[i]);
+			seq_printf(p, " %10s", irq_desc[i].chip->name ? : "-");
+			seq_printf(p, "  %s", action->name);
+			for (action = action->next;
+			     action;
+			     action = action->next)
+				seq_printf(p, ", %s", action->name);
+
+			seq_putc(p, '\n');
+		}
 
-	default:
-		break;
+		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
+	} else if (i == NR_IRQS) {
+		seq_printf(p, "Err: %10u\n", atomic_read(&irq_err_count));
 	}
 
 	return 0;
 }
 
-
 /*
- * Generic enable/disable code: this just calls
- * down into the PIC-specific version for the actual
- * hardware disable after having gotten the irq
- * controller lock.
+ * on-CPU PIC operations
  */
-
-/**
- *	disable_irq_nosync - disable an irq without waiting
- *	@irq: Interrupt to disable
- *
- *	Disable the selected interrupt line.  Disables and Enables are
- *	nested.
- *	Unlike disable_irq(), this function does not ensure existing
- *	instances of the IRQ handler have completed before returning.
- *
- *	This function may be called from IRQ context.
- */
-
-void disable_irq_nosync(unsigned int irq)
+static void frv_cpupic_enable(unsigned int irqlevel)
 {
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	unsigned long flags;
-	int idx = irq & (NR_IRQ_ACTIONS_PER_GROUP - 1);
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[idx];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	if (group->control) {
-		if (!group->disable_cnt[idx]++)
-			group->control(group, idx, 0);
-	} else if (!level->disable_count++) {
-		__set_MASK(level - frv_irq_levels);
-	}
-
-	spin_unlock_irqrestore(&level->lock, flags);
+	__clr_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(disable_irq_nosync);
-
-/**
- *	disable_irq - disable an irq and wait for completion
- *	@irq: Interrupt to disable
- *
- *	Disable the selected interrupt line.  Enables and Disables are
- *	nested.
- *	This function waits for any pending IRQ handlers for this interrupt
- *	to complete before returning. If you use this function while
- *	holding a resource the IRQ handler may need you will deadlock.
- *
- *	This function may be called - with care - from IRQ context.
- */
-
-void disable_irq(unsigned int irq)
+static void frv_cpupic_disable(unsigned int irqlevel)
 {
-	disable_irq_nosync(irq);
-
-#ifdef CONFIG_SMP
-	if (!local_irq_count(smp_processor_id())) {
-		do {
-			barrier();
-		} while (irq_desc[irq].status & IRQ_INPROGRESS);
-	}
-#endif
+	__set_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(disable_irq);
-
-/**
- *	enable_irq - enable handling of an irq
- *	@irq: Interrupt to enable
- *
- *	Undoes the effect of one call to disable_irq().  If this
- *	matches the last disable, processing of interrupts on this
- *	IRQ line is re-enabled.
- *
- *	This function may be called from IRQ context.
- */
-
-void enable_irq(unsigned int irq)
+static void frv_cpupic_ack(unsigned int irqlevel)
 {
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	unsigned long flags;
-	int idx = irq & (NR_IRQ_ACTIONS_PER_GROUP - 1);
-	int count;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[idx];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	if (group->control)
-		count = group->disable_cnt[idx];
-	else
-		count = level->disable_count;
-
-	switch (count) {
-	case 1:
-		if (group->control) {
-			if (group->actions[idx])
-				group->control(group, idx, 1);
-		} else {
-			if (level->usage)
-				__clr_MASK(level - frv_irq_levels);
-		}
-		/* fall-through */
+	__set_MASK(irqlevel);
+	__clr_RC(irqlevel);
+	__clr_IRL();
+}
 
-	default:
-		count--;
-		break;
+static void frv_cpupic_mask(unsigned int irqlevel)
+{
+	__set_MASK(irqlevel);
+}
 
-	case 0:
-		printk("enable_irq(%u) unbalanced from %p\n", irq, __builtin_return_address(0));
-	}
+static void frv_cpupic_mask_ack(unsigned int irqlevel)
+{
+	__set_MASK(irqlevel);
+	__clr_RC(irqlevel);
+	__clr_IRL();
+}
 
-	if (group->control)
-		group->disable_cnt[idx] = count;
-	else
-		level->disable_count = count;
+static void frv_cpupic_unmask(unsigned int irqlevel)
+{
+	__clr_MASK(irqlevel);
+}
 
-	spin_unlock_irqrestore(&level->lock, flags);
+static void frv_cpupic_end(unsigned int irqlevel)
+{
+	__clr_MASK(irqlevel);
 }
 
-EXPORT_SYMBOL(enable_irq);
+static struct irq_chip frv_cpu_pic = {
+	.name		= "cpu",
+	.enable		= frv_cpupic_enable,
+	.disable	= frv_cpupic_disable,
+	.ack		= frv_cpupic_ack,
+	.mask		= frv_cpupic_mask,
+	.mask_ack	= frv_cpupic_mask_ack,
+	.unmask		= frv_cpupic_unmask,
+	.end		= frv_cpupic_end,
+};
 
-/*****************************************************************************/
 /*
  * handles all normal device IRQ's
  * - registers are referred to by the __frame variable (GR28)
@@ -281,463 +155,65 @@ EXPORT_SYMBOL(enable_irq);
  */
 asmlinkage void do_IRQ(void)
 {
-	struct irq_source *source;
-	int level, cpu;
-
 	irq_enter();
-
-	level = (__frame->tbr >> 4) & 0xf;
-	cpu = smp_processor_id();
-
-	if ((unsigned long) __frame - (unsigned long) (current + 1) < 512)
-		BUG();
-
-	__set_MASK(level);
-	__clr_RC(level);
-	__clr_IRL();
-
-	kstat_this_cpu.irqs[level]++;
-
-	for (source = frv_irq_levels[level].sources; source; source = source->next)
-		source->doirq(source);
-
-	__clr_MASK(level);
-
+	__do_IRQ(__get_IRL(), __frame);
 	irq_exit();
+}
 
-} /* end do_IRQ() */
-
-/*****************************************************************************/
 /*
  * handles all NMIs when not co-opted by the debugger
  * - registers are referred to by the __frame variable (GR28)
  */
 asmlinkage void do_NMI(void)
 {
-} /* end do_NMI() */
-
-/*****************************************************************************/
-/**
- *	request_irq - allocate an interrupt line
- *	@irq: Interrupt line to allocate
- *	@handler: Function to be called when the IRQ occurs
- *	@irqflags: Interrupt type flags
- *	@devname: An ascii name for the claiming device
- *	@dev_id: A cookie passed back to the handler function
- *
- *	This call allocates interrupt resources and enables the
- *	interrupt line and IRQ handling. From the point this
- *	call is made your handler function may be invoked. Since
- *	your handler function must clear any interrupt the board
- *	raises, you must take care both to initialise your hardware
- *	and to set up the interrupt handler in the right order.
- *
- *	Dev_id must be globally unique. Normally the address of the
- *	device data structure is used as the cookie. Since the handler
- *	receives this value it makes sense to use it.
- *
- *	If your interrupt is shared you must pass a non NULL dev_id
- *	as this is required when freeing the interrupt.
- *
- *	Flags:
- *
- *	IRQF_SHARED		Interrupt is shared
- *
- *	IRQF_DISABLED	Disable local interrupts while processing
- *
- *	IRQF_SAMPLE_RANDOM	The interrupt can be used for entropy
- *
- */
-
-int request_irq(unsigned int irq,
-		irqreturn_t (*handler)(int, void *, struct pt_regs *),
-		unsigned long irqflags,
-		const char * devname,
-		void *dev_id)
-{
-	int retval;
-	struct irqaction *action;
-
-#if 1
-	/*
-	 * Sanity-check: shared interrupts should REALLY pass in
-	 * a real dev-ID, otherwise we'll have trouble later trying
-	 * to figure out which interrupt is which (messes up the
-	 * interrupt freeing logic etc).
-	 */
-	if (irqflags & IRQF_SHARED) {
-		if (!dev_id)
-			printk("Bad boy: %s (at 0x%x) called us without a dev_id!\n",
-			       devname, (&irq)[-1]);
-	}
-#endif
-
-	if ((irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP) >= NR_IRQ_GROUPS)
-		return -EINVAL;
-	if (!handler)
-		return -EINVAL;
-
-	action = (struct irqaction *) kmalloc(sizeof(struct irqaction), GFP_KERNEL);
-	if (!action)
-		return -ENOMEM;
-
-	action->handler = handler;
-	action->flags = irqflags;
-	action->mask = CPU_MASK_NONE;
-	action->name = devname;
-	action->next = NULL;
-	action->dev_id = dev_id;
-
-	retval = setup_irq(irq, action);
-	if (retval)
-		kfree(action);
-	return retval;
-}
-
-EXPORT_SYMBOL(request_irq);
-
-/**
- *	free_irq - free an interrupt
- *	@irq: Interrupt line to free
- *	@dev_id: Device identity to free
- *
- *	Remove an interrupt handler. The handler is removed and if the
- *	interrupt line is no longer in use by any driver it is disabled.
- *	On a shared IRQ the caller must ensure the interrupt is disabled
- *	on the card it drives before calling this function. The function
- *	does not return until any executing interrupts for this IRQ
- *	have completed.
- *
- *	This function may be called from interrupt context.
- *
- *	Bugs: Attempting to free an irq in a handler for the same irq hangs
- *	      the machine.
- */
-
-void free_irq(unsigned int irq, void *dev_id)
-{
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	struct irqaction **p, **pp;
-	unsigned long flags;
-
-	if ((irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP) >= NR_IRQ_GROUPS)
-		return;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-	if (!source)
-		BUG();
-
-	level = source->level;
-	p = &group->actions[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-
-	spin_lock_irqsave(&level->lock, flags);
-
-	for (pp = p; *pp; pp = &(*pp)->next) {
-		struct irqaction *action = *pp;
-
-		if (action->dev_id != dev_id)
-			continue;
-
-		/* found it - remove from the list of entries */
-		*pp = action->next;
-
-		level->usage--;
-
-		if (p == pp && group->control)
-			group->control(group, irq & (NR_IRQ_ACTIONS_PER_GROUP - 1), 0);
-
-		if (level->usage == 0)
-			__set_MASK(level - frv_irq_levels);
-
-		spin_unlock_irqrestore(&level->lock,flags);
-
-#ifdef CONFIG_SMP
-		/* Wait to make sure it's not being used on another CPU */
-		while (desc->status & IRQ_INPROGRESS)
-			barrier();
-#endif
-		kfree(action);
-		return;
-	}
-}
-
-EXPORT_SYMBOL(free_irq);
-
-/*
- * IRQ autodetection code..
- *
- * This depends on the fact that any interrupt that comes in on to an
- * unassigned IRQ will cause GxICR_DETECT to be set
- */
-
-static DECLARE_MUTEX(probe_sem);
-
-/**
- *	probe_irq_on	- begin an interrupt autodetect
- *
- *	Commence probing for an interrupt. The interrupts are scanned
- *	and a mask of potential interrupt lines is returned.
- *
- */
-
-unsigned long probe_irq_on(void)
-{
-	down(&probe_sem);
-	return 0;
 }
 
-EXPORT_SYMBOL(probe_irq_on);
-
-/*
- * Return a mask of triggered interrupts (this
- * can handle only legacy ISA interrupts).
- */
-
-/**
- *	probe_irq_mask - scan a bitmap of interrupt lines
- *	@val:	mask of interrupts to consider
- *
- *	Scan the ISA bus interrupt lines and return a bitmap of
- *	active interrupts. The interrupt probe logic state is then
- *	returned to its previous value.
- *
- *	Note: we need to scan all the irq's even though we will
- *	only return ISA irq numbers - just so that we reset them
- *	all to a known state.
- */
-unsigned int probe_irq_mask(unsigned long xmask)
-{
-	up(&probe_sem);
-	return 0;
-}
-
-EXPORT_SYMBOL(probe_irq_mask);
-
 /*
- * Return the one interrupt that triggered (this can
- * handle any interrupt source).
- */
-
-/**
- *	probe_irq_off	- end an interrupt autodetect
- *	@xmask: mask of potential interrupts (unused)
- *
- *	Scans the unused interrupt lines and returns the line which
- *	appears to have triggered the interrupt. If no interrupt was
- *	found then zero is returned. If more than one interrupt is
- *	found then minus the first candidate is returned to indicate
- *	their is doubt.
- *
- *	The interrupt probe logic state is returned to its previous
- *	value.
- *
- *	BUGS: When used in a module (which arguably shouldnt happen)
- *	nothing prevents two IRQ probe callers from overlapping. The
- *	results of this are non-optimal.
+ * initialise the interrupt system
  */
-
-int probe_irq_off(unsigned long xmask)
-{
-	up(&probe_sem);
-	return -1;
-}
-
-EXPORT_SYMBOL(probe_irq_off);
-
-/* this was setup_x86_irq but it seems pretty generic */
-int setup_irq(unsigned int irq, struct irqaction *new)
-{
-	struct irq_source *source;
-	struct irq_group *group;
-	struct irq_level *level;
-	struct irqaction **p, **pp;
-	unsigned long flags;
-
-	group = irq_groups[irq >> NR_IRQ_LOG2_ACTIONS_PER_GROUP];
-	if (!group)
-		BUG();
-
-	source = group->sources[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-	if (!source)
-		BUG();
-
-	level = source->level;
-
-	p = &group->actions[irq & (NR_IRQ_ACTIONS_PER_GROUP - 1)];
-
-	/*
-	 * Some drivers like serial.c use request_irq() heavily,
-	 * so we have to be careful not to interfere with a
-	 * running system.
-	 */
-	if (new->flags & IRQF_SAMPLE_RANDOM) {
-		/*
-		 * This function might sleep, we want to call it first,
-		 * outside of the atomic block.
-		 * Yes, this might clear the entropy pool if the wrong
-		 * driver is attempted to be loaded, without actually
-		 * installing a new handler, but is this really a problem,
-		 * only the sysadmin is able to do this.
-		 */
-		rand_initialize_irq(irq);
-	}
-
-	/* must juggle the interrupt processing stuff with interrupts disabled */
-	spin_lock_irqsave(&level->lock, flags);
-
-	/* can't share interrupts unless all parties agree to */
-	if (level->usage != 0 && !(level->flags & new->flags & IRQF_SHARED)) {
-		spin_unlock_irqrestore(&level->lock,flags);
-		return -EBUSY;
-	}
-
-	/* add new interrupt at end of irq queue */
-	pp = p;
-	while (*pp)
-		pp = &(*pp)->next;
-
-	*pp = new;
-
-	level->usage++;
-	level->flags = new->flags;
-
-	/* turn the interrupts on */
-	if (level->usage == 1)
-		__clr_MASK(level - frv_irq_levels);
-
-	if (p == pp && group->control)
-		group->control(group, irq & (NR_IRQ_ACTIONS_PER_GROUP - 1), 1);
-
-	spin_unlock_irqrestore(&level->lock, flags);
-	register_irq_proc(irq);
-	return 0;
-}
-
-static struct proc_dir_entry * root_irq_dir;
-static struct proc_dir_entry * irq_dir [NR_IRQS];
-
-#define HEX_DIGITS 8
-
-static unsigned int parse_hex_value (const char __user *buffer,
-				     unsigned long count, unsigned long *ret)
-{
-	unsigned char hexnum [HEX_DIGITS];
-	unsigned long value;
-	int i;
-
-	if (!count)
-		return -EINVAL;
-	if (count > HEX_DIGITS)
-		count = HEX_DIGITS;
-	if (copy_from_user(hexnum, buffer, count))
-		return -EFAULT;
-
-	/*
-	 * Parse the first 8 characters as a hex string, any non-hex char
-	 * is end-of-string. '00e1', 'e1', '00E1', 'E1' are all the same.
-	 */
-	value = 0;
-
-	for (i = 0; i < count; i++) {
-		unsigned int c = hexnum[i];
-
-		switch (c) {
-			case '0' ... '9': c -= '0'; break;
-			case 'a' ... 'f': c -= 'a'-10; break;
-			case 'A' ... 'F': c -= 'A'-10; break;
-		default:
-			goto out;
-		}
-		value = (value << 4) | c;
-	}
-out:
-	*ret = value;
-	return 0;
-}
-
-
-static int prof_cpu_mask_read_proc (char *page, char **start, off_t off,
-			int count, int *eof, void *data)
-{
-	unsigned long *mask = (unsigned long *) data;
-	if (count < HEX_DIGITS+1)
-		return -EINVAL;
-	return sprintf (page, "%08lx\n", *mask);
-}
-
-static int prof_cpu_mask_write_proc (struct file *file, const char __user *buffer,
-					unsigned long count, void *data)
-{
-	unsigned long *mask = (unsigned long *) data, full_count = count, err;
-	unsigned long new_value;
-
-	show_state();
-	err = parse_hex_value(buffer, count, &new_value);
-	if (err)
-		return err;
-
-	*mask = new_value;
-	return full_count;
-}
-
-#define MAX_NAMELEN 10
-
-static void register_irq_proc (unsigned int irq)
-{
-	char name [MAX_NAMELEN];
-
-	if (!root_irq_dir || irq_dir[irq])
-		return;
-
-	memset(name, 0, MAX_NAMELEN);
-	sprintf(name, "%d", irq);
-
-	/* create /proc/irq/1234 */
-	irq_dir[irq] = proc_mkdir(name, root_irq_dir);
-}
-
-unsigned long prof_cpu_mask = -1;
-
-void init_irq_proc (void)
+void __init init_IRQ(void)
 {
-	struct proc_dir_entry *entry;
-	int i;
+	int level;
 
-	/* create /proc/irq */
-	root_irq_dir = proc_mkdir("irq", NULL);
+	for (level = 1; level <= 14; level++)
+		set_irq_chip_and_handler(level, &frv_cpu_pic,
+					 handle_level_irq);
 
-	/* create /proc/irq/prof_cpu_mask */
-	entry = create_proc_entry("prof_cpu_mask", 0600, root_irq_dir);
-	if (!entry)
-	    return;
+	set_irq_handler(IRQ_CPU_TIMER0, handle_edge_irq);
 
-	entry->nlink = 1;
-	entry->data = (void *)&prof_cpu_mask;
-	entry->read_proc = prof_cpu_mask_read_proc;
-	entry->write_proc = prof_cpu_mask_write_proc;
-
-	/*
-	 * Create entries for all existing IRQs.
+	/* set the trigger levels for internal interrupt sources
+	 * - timers all falling-edge
+	 * - ERR0 is rising-edge
+	 * - all others are high-level
 	 */
-	for (i = 0; i < NR_IRQS; i++)
-		register_irq_proc(i);
-}
+	__set_IITMR(0, 0x003f0000);	/* DMA0-3, TIMER0-2 */
+	__set_IITMR(1, 0x20000000);	/* ERR0-1, UART0-1, DMA4-7 */
+
+	/* route internal interrupts */
+	set_IRR(4, IRQ_DMA3_LEVEL, IRQ_DMA2_LEVEL, IRQ_DMA1_LEVEL,
+		IRQ_DMA0_LEVEL);
+	set_IRR(5, 0, IRQ_TIMER2_LEVEL, IRQ_TIMER1_LEVEL, IRQ_TIMER0_LEVEL);
+	set_IRR(6, IRQ_GDBSTUB_LEVEL, IRQ_GDBSTUB_LEVEL,
+		IRQ_UART1_LEVEL, IRQ_UART0_LEVEL);
+	set_IRR(7, IRQ_DMA7_LEVEL, IRQ_DMA6_LEVEL, IRQ_DMA5_LEVEL,
+		IRQ_DMA4_LEVEL);
+
+	/* route external interrupts */
+	set_IRR(2, IRQ_XIRQ7_LEVEL, IRQ_XIRQ6_LEVEL, IRQ_XIRQ5_LEVEL,
+		IRQ_XIRQ4_LEVEL);
+	set_IRR(3, IRQ_XIRQ3_LEVEL, IRQ_XIRQ2_LEVEL, IRQ_XIRQ1_LEVEL,
+		IRQ_XIRQ0_LEVEL);
+
+#if defined(CONFIG_MB93091_VDK)
+	__set_TM1(0x55550000);		/* XIRQ7-0 all active low */
+#elif defined(CONFIG_MB93093_PDK)
+	__set_TM1(0x15550000);		/* XIRQ7 active high, 6-0 all active low */
+#else
+#error dont know external IRQ trigger levels for this setup
+#endif
 
-/*****************************************************************************/
-/*
- * initialise the interrupt system
- */
-void __init init_IRQ(void)
-{
-	route_cpu_irqs();
 	fpga_init();
 #ifdef CONFIG_FUJITSU_MB93493
-	route_mb93493_irqs();
+	mb93493_init();
 #endif
-} /* end init_IRQ() */
+}
diff --git a/arch/frv/kernel/setup.c b/arch/frv/kernel/setup.c
index af08ccd4ed6e..d96a57e5f030 100644
--- a/arch/frv/kernel/setup.c
+++ b/arch/frv/kernel/setup.c
@@ -43,7 +43,6 @@
 #include <asm/mb-regs.h>
 #include <asm/mb93493-regs.h>
 #include <asm/gdb-stub.h>
-#include <asm/irq-routing.h>
 #include <asm/io.h>
 
 #ifdef CONFIG_BLK_DEV_INITRD
diff --git a/arch/frv/kernel/time.c b/arch/frv/kernel/time.c
index 68a77fe3bb40..3d0284bccb94 100644
--- a/arch/frv/kernel/time.c
+++ b/arch/frv/kernel/time.c
@@ -26,7 +26,6 @@
 #include <asm/timer-regs.h>
 #include <asm/mb-regs.h>
 #include <asm/mb86943a.h>
-#include <asm/irq-routing.h>
 
 #include <linux/timex.h>
 
diff --git a/arch/frv/mb93090-mb00/pci-irq.c b/arch/frv/mb93090-mb00/pci-irq.c
index 2278c80bd88c..ba587523c015 100644
--- a/arch/frv/mb93090-mb00/pci-irq.c
+++ b/arch/frv/mb93090-mb00/pci-irq.c
@@ -15,7 +15,6 @@
 
 #include <asm/io.h>
 #include <asm/smp.h>
-#include <asm/irq-routing.h>
 
 #include "pci-frv.h"
 
diff --git a/include/asm-frv/cpu-irqs.h b/include/asm-frv/cpu-irqs.h
index 5cd691e1f8c4..478f3498fcfe 100644
--- a/include/asm-frv/cpu-irqs.h
+++ b/include/asm-frv/cpu-irqs.h
@@ -14,36 +14,6 @@
 
 #ifndef __ASSEMBLY__
 
-#include <asm/irq-routing.h>
-
-#define IRQ_BASE_CPU		(NR_IRQ_ACTIONS_PER_GROUP * 0)
-
-/* IRQ IDs presented to drivers */
-enum {
-	IRQ_CPU__UNUSED = IRQ_BASE_CPU,
-	IRQ_CPU_UART0,
-	IRQ_CPU_UART1,
-	IRQ_CPU_TIMER0,
-	IRQ_CPU_TIMER1,
-	IRQ_CPU_TIMER2,
-	IRQ_CPU_DMA0,
-	IRQ_CPU_DMA1,
-	IRQ_CPU_DMA2,
-	IRQ_CPU_DMA3,
-	IRQ_CPU_DMA4,
-	IRQ_CPU_DMA5,
-	IRQ_CPU_DMA6,
-	IRQ_CPU_DMA7,
-	IRQ_CPU_EXTERNAL0,
-	IRQ_CPU_EXTERNAL1,
-	IRQ_CPU_EXTERNAL2,
-	IRQ_CPU_EXTERNAL3,
-	IRQ_CPU_EXTERNAL4,
-	IRQ_CPU_EXTERNAL5,
-	IRQ_CPU_EXTERNAL6,
-	IRQ_CPU_EXTERNAL7,
-};
-
 /* IRQ to level mappings */
 #define IRQ_GDBSTUB_LEVEL	15
 #define IRQ_UART_LEVEL		13
@@ -82,6 +52,30 @@ enum {
 #define IRQ_XIRQ6_LEVEL		7
 #define IRQ_XIRQ7_LEVEL		8
 
+/* IRQ IDs presented to drivers */
+#define IRQ_CPU__UNUSED		IRQ_BASE_CPU
+#define IRQ_CPU_UART0		(IRQ_BASE_CPU + IRQ_UART0_LEVEL)
+#define IRQ_CPU_UART1		(IRQ_BASE_CPU + IRQ_UART1_LEVEL)
+#define IRQ_CPU_TIMER0		(IRQ_BASE_CPU + IRQ_TIMER0_LEVEL)
+#define IRQ_CPU_TIMER1		(IRQ_BASE_CPU + IRQ_TIMER1_LEVEL)
+#define IRQ_CPU_TIMER2		(IRQ_BASE_CPU + IRQ_TIMER2_LEVEL)
+#define IRQ_CPU_DMA0		(IRQ_BASE_CPU + IRQ_DMA0_LEVEL)
+#define IRQ_CPU_DMA1		(IRQ_BASE_CPU + IRQ_DMA1_LEVEL)
+#define IRQ_CPU_DMA2		(IRQ_BASE_CPU + IRQ_DMA2_LEVEL)
+#define IRQ_CPU_DMA3		(IRQ_BASE_CPU + IRQ_DMA3_LEVEL)
+#define IRQ_CPU_DMA4		(IRQ_BASE_CPU + IRQ_DMA4_LEVEL)
+#define IRQ_CPU_DMA5		(IRQ_BASE_CPU + IRQ_DMA5_LEVEL)
+#define IRQ_CPU_DMA6		(IRQ_BASE_CPU + IRQ_DMA6_LEVEL)
+#define IRQ_CPU_DMA7		(IRQ_BASE_CPU + IRQ_DMA7_LEVEL)
+#define IRQ_CPU_EXTERNAL0	(IRQ_BASE_CPU + IRQ_XIRQ0_LEVEL)
+#define IRQ_CPU_EXTERNAL1	(IRQ_BASE_CPU + IRQ_XIRQ1_LEVEL)
+#define IRQ_CPU_EXTERNAL2	(IRQ_BASE_CPU + IRQ_XIRQ2_LEVEL)
+#define IRQ_CPU_EXTERNAL3	(IRQ_BASE_CPU + IRQ_XIRQ3_LEVEL)
+#define IRQ_CPU_EXTERNAL4	(IRQ_BASE_CPU + IRQ_XIRQ4_LEVEL)
+#define IRQ_CPU_EXTERNAL5	(IRQ_BASE_CPU + IRQ_XIRQ5_LEVEL)
+#define IRQ_CPU_EXTERNAL6	(IRQ_BASE_CPU + IRQ_XIRQ6_LEVEL)
+#define IRQ_CPU_EXTERNAL7	(IRQ_BASE_CPU + IRQ_XIRQ7_LEVEL)
+
 #endif /* !__ASSEMBLY__ */
 
 #endif /* _ASM_CPU_IRQS_H */
diff --git a/include/asm-frv/hardirq.h b/include/asm-frv/hardirq.h
index 7581b5a7559a..fc47515822a2 100644
--- a/include/asm-frv/hardirq.h
+++ b/include/asm-frv/hardirq.h
@@ -26,5 +26,10 @@ typedef struct {
 #error SMP not available on FR-V
 #endif /* CONFIG_SMP */
 
+extern atomic_t irq_err_count;
+static inline void ack_bad_irq(int irq)
+{
+	atomic_inc(&irq_err_count);
+}
 
 #endif
diff --git a/include/asm-frv/irq-routing.h b/include/asm-frv/irq-routing.h
deleted file mode 100644
index ac3ab900a1dc..000000000000
--- a/include/asm-frv/irq-routing.h
+++ /dev/null
@@ -1,70 +0,0 @@
-/* irq-routing.h: multiplexed IRQ routing
- *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version
- * 2 of the License, or (at your option) any later version.
- */
-
-#ifndef _ASM_IRQ_ROUTING_H
-#define _ASM_IRQ_ROUTING_H
-
-#ifndef __ASSEMBLY__
-
-#include <linux/spinlock.h>
-#include <asm/irq.h>
-
-struct irq_source;
-struct irq_level;
-
-/*
- * IRQ action distribution sets
- */
-struct irq_group {
-	int			first_irq;	/* first IRQ distributed here */
-	void (*control)(struct irq_group *group, int index, int on);
-
-	struct irqaction	*actions[NR_IRQ_ACTIONS_PER_GROUP];	/* IRQ action chains */
-	struct irq_source	*sources[NR_IRQ_ACTIONS_PER_GROUP];	/* IRQ sources */
-	int			disable_cnt[NR_IRQ_ACTIONS_PER_GROUP];	/* disable counts */
-};
-
-/*
- * IRQ source manager
- */
-struct irq_source {
-	struct irq_source	*next;
-	struct irq_level	*level;
-	const char		*muxname;
-	volatile void __iomem	*muxdata;
-	unsigned long		irqmask;
-
-	void (*doirq)(struct irq_source *source);
-};
-
-/*
- * IRQ level management (per CPU IRQ priority / entry vector)
- */
-struct irq_level {
-	int			usage;
-	int			disable_count;
-	unsigned long		flags;		/* current IRQF_DISABLED and IRQF_SHARED settings */
-	spinlock_t		lock;
-	struct irq_source	*sources;
-};
-
-extern struct irq_level frv_irq_levels[16];
-extern struct irq_group *irq_groups[NR_IRQ_GROUPS];
-
-extern void frv_irq_route(struct irq_source *source, int irqlevel);
-extern void frv_irq_route_external(struct irq_source *source, int irq);
-extern void frv_irq_set_group(struct irq_group *group);
-extern void distribute_irqs(struct irq_group *group, unsigned long irqmask);
-extern void route_cpu_irqs(void);
-
-#endif /* !__ASSEMBLY__ */
-
-#endif /* _ASM_IRQ_ROUTING_H */
diff --git a/include/asm-frv/irq.h b/include/asm-frv/irq.h
index 58b619215a50..8fefd6b827aa 100644
--- a/include/asm-frv/irq.h
+++ b/include/asm-frv/irq.h
@@ -1,6 +1,6 @@
 /* irq.h: FRV IRQ definitions
  *
- * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Copyright (C) 2006 Red Hat, Inc. All Rights Reserved.
  * Written by David Howells (dhowells@redhat.com)
  *
  * This program is free software; you can redistribute it and/or
@@ -12,32 +12,22 @@
 #ifndef _ASM_IRQ_H_
 #define _ASM_IRQ_H_
 
-
-/*
- * the system has an on-CPU PIC and another PIC on the FPGA and other PICs on other peripherals,
- * so we do some routing in irq-routing.[ch] to reduce the number of false-positives seen by
- * drivers
- */
-
 /* this number is used when no interrupt has been assigned */
 #define NO_IRQ				(-1)
 
-#define NR_IRQ_LOG2_ACTIONS_PER_GROUP	5
-#define NR_IRQ_ACTIONS_PER_GROUP	(1 << NR_IRQ_LOG2_ACTIONS_PER_GROUP)
-#define NR_IRQ_GROUPS			4
-#define NR_IRQS				(NR_IRQ_ACTIONS_PER_GROUP * NR_IRQ_GROUPS)
+#define NR_IRQS				48
+#define IRQ_BASE_CPU			(0 * 16)
+#define IRQ_BASE_FPGA			(1 * 16)
+#define IRQ_BASE_MB93493		(2 * 16)
 
 /* probe returns a 32-bit IRQ mask:-/ */
-#define MIN_PROBE_IRQ	(NR_IRQS - 32)
+#define MIN_PROBE_IRQ			(NR_IRQS - 32)
 
+#ifndef __ASSEMBLY__
 static inline int irq_canonicalize(int irq)
 {
 	return irq;
 }
-
-extern void disable_irq_nosync(unsigned int irq);
-extern void disable_irq(unsigned int irq);
-extern void enable_irq(unsigned int irq);
-
+#endif
 
 #endif /* _ASM_IRQ_H_ */
diff --git a/include/asm-frv/mb93091-fpga-irqs.h b/include/asm-frv/mb93091-fpga-irqs.h
index 341bfc52a0eb..19778c5ba9d6 100644
--- a/include/asm-frv/mb93091-fpga-irqs.h
+++ b/include/asm-frv/mb93091-fpga-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93091_FPGA_IRQS_H
 #define _ASM_MB93091_FPGA_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_FPGA		(NR_IRQ_ACTIONS_PER_GROUP * 1)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93093-fpga-irqs.h b/include/asm-frv/mb93093-fpga-irqs.h
index 1e0f11c2fcdb..590266b1a6d3 100644
--- a/include/asm-frv/mb93093-fpga-irqs.h
+++ b/include/asm-frv/mb93093-fpga-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93093_FPGA_IRQS_H
 #define _ASM_MB93093_FPGA_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_FPGA		(NR_IRQ_ACTIONS_PER_GROUP * 1)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93493-irqs.h b/include/asm-frv/mb93493-irqs.h
index 15096e731325..82c7aeddd333 100644
--- a/include/asm-frv/mb93493-irqs.h
+++ b/include/asm-frv/mb93493-irqs.h
@@ -12,11 +12,9 @@
 #ifndef _ASM_MB93493_IRQS_H
 #define _ASM_MB93493_IRQS_H
 
-#ifndef __ASSEMBLY__
-
-#include <asm/irq-routing.h>
+#include <asm/irq.h>
 
-#define IRQ_BASE_MB93493	(NR_IRQ_ACTIONS_PER_GROUP * 2)
+#ifndef __ASSEMBLY__
 
 /* IRQ IDs presented to drivers */
 enum {
diff --git a/include/asm-frv/mb93493-regs.h b/include/asm-frv/mb93493-regs.h
index c54aa9d14468..8a1f6aac8cf1 100644
--- a/include/asm-frv/mb93493-regs.h
+++ b/include/asm-frv/mb93493-regs.h
@@ -15,6 +15,7 @@
 #include <asm/mb-regs.h>
 #include <asm/mb93493-irqs.h>
 
+#define __addr_MB93493(X)	((volatile unsigned long *)(__region_CS3 + (X)))
 #define __get_MB93493(X)	({ *(volatile unsigned long *)(__region_CS3 + (X)); })
 
 #define __set_MB93493(X,V)						\
@@ -26,6 +27,7 @@ do {									\
 #define __set_MB93493_STSR(X,V)	__set_MB93493(0x3c0 + (X) * 4, (V))
 #define MB93493_STSR_EN
 
+#define __addr_MB93493_IQSR(X)	__addr_MB93493(0x3d0 + (X) * 4)
 #define __get_MB93493_IQSR(X)	__get_MB93493(0x3d0 + (X) * 4)
 #define __set_MB93493_IQSR(X,V)	__set_MB93493(0x3d0 + (X) * 4, (V))
 
-- 
cgit v1.2.1


From 88d6e19900366781739df033e9c0e2532e715fa5 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:06 -0700
Subject: [PATCH] FRV: improve FRV's use of generic IRQ handling

Improve FRV's use of generic IRQ handling:

 (*) Use generic_handle_irq() rather than __do_IRQ() as the latter is obsolete.

 (*) Don't implement enable() and disable() ops as these will fall back to
     using unmask() and mask().

 (*) Provide mask_ack() functions to avoid a call each to mask() and ack().

 (*) Make the cascade handlers always return IRQ_HANDLED.

 (*) Implement the mask() and unmask() functions in the same order as they're
     listed in the ops table.

Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/kernel/irq-mb93091.c | 36 +++++++++++++++++++-----------------
 arch/frv/kernel/irq-mb93093.c | 36 +++++++++++++++++++-----------------
 arch/frv/kernel/irq-mb93493.c | 35 ++++++++++++++---------------------
 arch/frv/kernel/irq.c         | 15 +--------------
 4 files changed, 53 insertions(+), 69 deletions(-)

diff --git a/arch/frv/kernel/irq-mb93091.c b/arch/frv/kernel/irq-mb93091.c
index 635d23437666..369bc0a7443d 100644
--- a/arch/frv/kernel/irq-mb93091.c
+++ b/arch/frv/kernel/irq-mb93091.c
@@ -36,41 +36,45 @@
 /*
  * on-motherboard FPGA PIC operations
  */
-static void frv_fpga_enable(unsigned int irq)
+static void frv_fpga_mask(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 
 	__set_IMR(imr);
 }
 
-static void frv_fpga_disable(unsigned int irq)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_mask_ack(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
 	imr |= 1 << (irq - IRQ_BASE_FPGA);
-
 	__set_IMR(imr);
-}
 
-static void frv_fpga_ack(unsigned int irq)
-{
 	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
 }
 
-static void frv_fpga_end(unsigned int irq)
+static void frv_fpga_unmask(unsigned int irq)
 {
+	uint16_t imr = __get_IMR();
+
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
+
+	__set_IMR(imr);
 }
 
 static struct irq_chip frv_fpga_pic = {
 	.name		= "mb93091",
-	.enable		= frv_fpga_enable,
-	.disable	= frv_fpga_disable,
 	.ack		= frv_fpga_ack,
-	.mask		= frv_fpga_disable,
-	.unmask		= frv_fpga_enable,
-	.end		= frv_fpga_end,
+	.mask		= frv_fpga_mask,
+	.mask_ack	= frv_fpga_mask_ack,
+	.unmask		= frv_fpga_unmask,
 };
 
 /*
@@ -79,7 +83,6 @@ static struct irq_chip frv_fpga_pic = {
 static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
 	uint16_t imr, mask = (unsigned long) _mask;
-	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
 	mask = mask & ~imr & __get_IFR();
@@ -92,11 +95,10 @@ static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 		irq = 31 - irq;
 		mask &= ~(1 << irq);
 
-		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
-			iret |= IRQ_HANDLED;
+		generic_handle_irq(IRQ_BASE_FPGA + irq, regs);
 	}
 
-	return iret;
+	return IRQ_HANDLED;
 }
 
 /*
diff --git a/arch/frv/kernel/irq-mb93093.c b/arch/frv/kernel/irq-mb93093.c
index f60db7988e78..a43a22158956 100644
--- a/arch/frv/kernel/irq-mb93093.c
+++ b/arch/frv/kernel/irq-mb93093.c
@@ -35,40 +35,44 @@
 /*
  * off-CPU FPGA PIC operations
  */
-static void frv_fpga_enable(unsigned int irq)
+static void frv_fpga_mask(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
-	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
-
+	imr |= 1 << (irq - IRQ_BASE_FPGA);
 	__set_IMR(imr);
 }
 
-static void frv_fpga_disable(unsigned int irq)
+static void frv_fpga_ack(unsigned int irq)
+{
+	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
+}
+
+static void frv_fpga_mask_ack(unsigned int irq)
 {
 	uint16_t imr = __get_IMR();
 
 	imr |= 1 << (irq - IRQ_BASE_FPGA);
-
 	__set_IMR(imr);
-}
 
-static void frv_fpga_ack(unsigned int irq)
-{
 	__clr_IFR(1 << (irq - IRQ_BASE_FPGA));
 }
 
-static void frv_fpga_end(unsigned int irq)
+static void frv_fpga_unmask(unsigned int irq)
 {
+	uint16_t imr = __get_IMR();
+
+	imr &= ~(1 << (irq - IRQ_BASE_FPGA));
+
+	__set_IMR(imr);
 }
 
 static struct irq_chip frv_fpga_pic = {
 	.name		= "mb93093",
-	.enable		= frv_fpga_enable,
-	.disable	= frv_fpga_disable,
 	.ack		= frv_fpga_ack,
-	.mask		= frv_fpga_disable,
-	.unmask		= frv_fpga_enable,
+	.mask		= frv_fpga_mask,
+	.mask_ack	= frv_fpga_mask_ack,
+	.unmask		= frv_fpga_unmask,
 	.end		= frv_fpga_end,
 };
 
@@ -78,7 +82,6 @@ static struct irq_chip frv_fpga_pic = {
 static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 {
 	uint16_t imr, mask = (unsigned long) _mask;
-	irqreturn_t iret = 0;
 
 	imr = __get_IMR();
 	mask = mask & ~imr & __get_IFR();
@@ -91,11 +94,10 @@ static irqreturn_t fpga_interrupt(int irq, void *_mask, struct pt_regs *regs)
 		irq = 31 - irq;
 		mask &= ~(1 << irq);
 
-		if (__do_IRQ(IRQ_BASE_FPGA + irq, regs))
-			iret |= IRQ_HANDLED;
+		generic_irq_handle(IRQ_BASE_FPGA + irq, regs);
 	}
 
-	return iret;
+	return IRQ_HANDLED;
 }
 
 /*
diff --git a/arch/frv/kernel/irq-mb93493.c b/arch/frv/kernel/irq-mb93493.c
index 8ad9abfc7c13..39c0188a3498 100644
--- a/arch/frv/kernel/irq-mb93493.c
+++ b/arch/frv/kernel/irq-mb93493.c
@@ -43,8 +43,9 @@
 
 /*
  * daughter board PIC operations
+ * - there is no way to ACK interrupts in the MB93493 chip
  */
-static void frv_mb93493_enable(unsigned int irq)
+static void frv_mb93493_mask(unsigned int irq)
 {
 	uint32_t iqsr;
 	volatile void *piqsr;
@@ -55,11 +56,15 @@ static void frv_mb93493_enable(unsigned int irq)
 		piqsr = __addr_MB93493_IQSR(0);
 
 	iqsr = readl(piqsr);
-	iqsr |= 1 << (irq - IRQ_BASE_MB93493 + 16);
+	iqsr &= ~(1 << (irq - IRQ_BASE_MB93493 + 16));
 	writel(iqsr, piqsr);
 }
 
-static void frv_mb93493_disable(unsigned int irq)
+static void frv_mb93493_ack(unsigned int irq)
+{
+}
+
+static void frv_mb93493_unmask(unsigned int irq)
 {
 	uint32_t iqsr;
 	volatile void *piqsr;
@@ -70,26 +75,16 @@ static void frv_mb93493_disable(unsigned int irq)
 		piqsr = __addr_MB93493_IQSR(0);
 
 	iqsr = readl(piqsr);
-	iqsr &= ~(1 << (irq - IRQ_BASE_MB93493 + 16));
+	iqsr |= 1 << (irq - IRQ_BASE_MB93493 + 16);
 	writel(iqsr, piqsr);
 }
 
-static void frv_mb93493_ack(unsigned int irq)
-{
-}
-
-static void frv_mb93493_end(unsigned int irq)
-{
-}
-
 static struct irq_chip frv_mb93493_pic = {
 	.name		= "mb93093",
-	.enable		= frv_mb93493_enable,
-	.disable	= frv_mb93493_disable,
 	.ack		= frv_mb93493_ack,
-	.mask		= frv_mb93493_disable,
-	.unmask		= frv_mb93493_enable,
-	.end		= frv_mb93493_end,
+	.mask		= frv_mb93493_mask,
+	.mask_ack	= frv_mb93493_mask,
+	.unmask		= frv_mb93493_unmask,
 };
 
 /*
@@ -98,7 +93,6 @@ static struct irq_chip frv_mb93493_pic = {
 static irqreturn_t mb93493_interrupt(int irq, void *_piqsr, struct pt_regs *regs)
 {
 	volatile void *piqsr = _piqsr;
-	irqreturn_t iret = 0;
 	uint32_t iqsr;
 
 	iqsr = readl(piqsr);
@@ -112,11 +106,10 @@ static irqreturn_t mb93493_interrupt(int irq, void *_piqsr, struct pt_regs *regs
 		irq = 31 - irq;
 		iqsr &= ~(1 << irq);
 
-		if (__do_IRQ(IRQ_BASE_MB93493 + irq, regs))
-			iret |= IRQ_HANDLED;
+		generic_handle_irq(IRQ_BASE_MB93493 + irq, regs);
 	}
 
-	return iret;
+	return IRQ_HANDLED;
 }
 
 /*
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index e1ab9f2e43fb..5ac041c7c0a4 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -97,19 +97,8 @@ int show_interrupts(struct seq_file *p, void *v)
 /*
  * on-CPU PIC operations
  */
-static void frv_cpupic_enable(unsigned int irqlevel)
-{
-	__clr_MASK(irqlevel);
-}
-
-static void frv_cpupic_disable(unsigned int irqlevel)
-{
-	__set_MASK(irqlevel);
-}
-
 static void frv_cpupic_ack(unsigned int irqlevel)
 {
-	__set_MASK(irqlevel);
 	__clr_RC(irqlevel);
 	__clr_IRL();
 }
@@ -138,8 +127,6 @@ static void frv_cpupic_end(unsigned int irqlevel)
 
 static struct irq_chip frv_cpu_pic = {
 	.name		= "cpu",
-	.enable		= frv_cpupic_enable,
-	.disable	= frv_cpupic_disable,
 	.ack		= frv_cpupic_ack,
 	.mask		= frv_cpupic_mask,
 	.mask_ack	= frv_cpupic_mask_ack,
@@ -156,7 +143,7 @@ static struct irq_chip frv_cpu_pic = {
 asmlinkage void do_IRQ(void)
 {
 	irq_enter();
-	__do_IRQ(__get_IRL(), __frame);
+	generic_handle_irq(__get_IRL(), __frame);
 	irq_exit();
 }
 
-- 
cgit v1.2.1


From af8c65b57aaa4ae321af34dbfc5ca7f5625263fe Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:07 -0700
Subject: [PATCH] FRV: permit __do_IRQ() to be dispensed with

Permit __do_IRQ() to be dispensed with based on a configuration option.

Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/Kconfig    | 4 ++++
 include/linux/irq.h | 6 ++++++
 kernel/irq/handle.c | 2 ++
 3 files changed, 12 insertions(+)

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 5833808fb823..f7b171b92ea2 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -29,6 +29,10 @@ config GENERIC_HARDIRQS
 	bool
 	default y
 
+config GENERIC_HARDIRQS_NO__DO_IRQ
+	bool
+	default y
+
 config GENERIC_TIME
 	bool
 	default y
diff --git a/include/linux/irq.h b/include/linux/irq.h
index fbf6d901e9c2..48d3cb3b6a47 100644
--- a/include/linux/irq.h
+++ b/include/linux/irq.h
@@ -320,7 +320,9 @@ handle_irq_name(void fastcall (*handle)(unsigned int, struct irq_desc *,
  * Monolithic do_IRQ implementation.
  * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly)
  */
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
 extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs);
+#endif
 
 /*
  * Architectures call this to let the generic IRQ layer
@@ -332,10 +334,14 @@ static inline void generic_handle_irq(unsigned int irq, struct pt_regs *regs)
 {
 	struct irq_desc *desc = irq_desc + irq;
 
+#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
+	desc->handle_irq(irq, desc, regs);
+#else
 	if (likely(desc->handle_irq))
 		desc->handle_irq(irq, desc, regs);
 	else
 		__do_IRQ(irq, regs);
+#endif
 }
 
 /* Handling of unhandled and spurious interrupts: */
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 48a53f68af96..4c6cdbaed661 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -154,6 +154,7 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct pt_regs *regs,
 	return retval;
 }
 
+#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ
 /**
  * __do_IRQ - original all in one highlevel IRQ handler
  * @irq:	the interrupt number
@@ -253,6 +254,7 @@ out:
 
 	return 1;
 }
+#endif
 
 #ifdef CONFIG_TRACE_IRQFLAGS
 
-- 
cgit v1.2.1


From 92fc707208bb2e601c24b5ab65db37bcb361b658 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:07 -0700
Subject: [PATCH] FRV: Fix fls() to handle bit 31 being set correctly

Fix FRV fls() to handle bit 31 being set correctly (it should return 32 not 0).

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 980ae1b0cd28..97fb746f76c7 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -161,16 +161,29 @@ static inline int __test_bit(int nr, const volatile void * addr)
 #include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/find.h>
 
-/*
- * fls: find last bit set.
+/**
+ * fls - find last bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as ffs:
+ * - return 32..1 to indicate bit 31..0 most significant bit set
+ * - return 0 to indicate no bits set
  */
 #define fls(x)						\
 ({							\
 	int bit;					\
 							\
-	asm("scan %1,gr0,%0" : "=r"(bit) : "r"(x));	\
+	asm("	subcc	%1,gr0,gr0,icc0		\n"	\
+	    "	ckne	icc0,cc4		\n"	\
+	    "	cscan.p	%1,gr0,%0	,cc4,#1	\n"	\
+	    "	csub	%0,%0,%0	,cc4,#0	\n"	\
+	    "   csub    %2,%0,%0	,cc4,#1	\n"	\
+	    : "=&r"(bit)				\
+	    : "r"(x), "r"(32)				\
+	    : "icc0", "cc4"				\
+	    );						\
 							\
-	bit ? 33 - bit : bit;				\
+	bit;						\
 })
 
 #include <asm-generic/bitops/fls64.h>
-- 
cgit v1.2.1


From a8ad27d03f17e6154c61e81d4a7028c56ca6390d Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:08 -0700
Subject: [PATCH] FRV: Implement fls64()

Implement fls64() for FRV without recource to conditional jumps.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 42 +++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 41 insertions(+), 1 deletion(-)

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 97fb746f76c7..591eecc1f8cd 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -186,7 +186,47 @@ static inline int __test_bit(int nr, const volatile void * addr)
 	bit;						\
 })
 
-#include <asm-generic/bitops/fls64.h>
+/**
+ * fls64 - find last bit set in a 64-bit value
+ * @n: the value to search
+ *
+ * This is defined the same way as ffs:
+ * - return 64..1 to indicate bit 63..0 most significant bit set
+ * - return 0 to indicate no bits set
+ */
+static inline __attribute__((const))
+int fls64(u64 n)
+{
+	union {
+		u64 ll;
+		struct { u32 h, l; };
+	} _;
+	int bit, x, y;
+
+	_.ll = n;
+
+	asm("	subcc.p		%3,gr0,gr0,icc0		\n"
+	    "	subcc		%4,gr0,gr0,icc1		\n"
+	    "	ckne		icc0,cc4		\n"
+	    "	ckne		icc1,cc5		\n"
+	    "	norcr		cc4,cc5,cc6		\n"
+	    "	csub.p		%0,%0,%0	,cc6,1	\n"
+	    "	orcr		cc5,cc4,cc4		\n"
+	    "	andcr		cc4,cc5,cc4		\n"
+	    "	cscan.p		%3,gr0,%0	,cc4,0	\n"
+	    "   setlos		#64,%1			\n"
+	    "	cscan.p		%4,gr0,%0	,cc4,1	\n"
+	    "   setlos		#32,%2			\n"
+	    "	csub.p		%1,%0,%0	,cc4,0	\n"
+	    "	csub		%2,%0,%0	,cc4,1	\n"
+	    : "=&r"(bit), "=r"(x), "=r"(y)
+	    : "0r"(_.h), "r"(_.l)
+	    : "icc0", "icc1", "cc4", "cc5", "cc6"
+	    );
+	return bit;
+
+}
+
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/hweight.h>
 
-- 
cgit v1.2.1


From cf134483b2cd657039b305777215c531a1009947 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Mon, 25 Sep 2006 23:32:09 -0700
Subject: [PATCH] FRV: Optimise ffs()

Optimise ffs(x) by using fls(x & x - 1) which we optimise to use the SCAN
instruction.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-frv/bitops.h | 33 +++++++++++++++++++++++++++++++--
 1 file changed, 31 insertions(+), 2 deletions(-)

diff --git a/include/asm-frv/bitops.h b/include/asm-frv/bitops.h
index 591eecc1f8cd..1f70d47148bd 100644
--- a/include/asm-frv/bitops.h
+++ b/include/asm-frv/bitops.h
@@ -157,8 +157,6 @@ static inline int __test_bit(int nr, const volatile void * addr)
  __constant_test_bit((nr),(addr)) : \
  __test_bit((nr),(addr)))
 
-#include <asm-generic/bitops/ffs.h>
-#include <asm-generic/bitops/__ffs.h>
 #include <asm-generic/bitops/find.h>
 
 /**
@@ -227,6 +225,37 @@ int fls64(u64 n)
 
 }
 
+/**
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * - return 32..1 to indicate bit 31..0 most least significant bit set
+ * - return 0 to indicate no bits set
+ */
+static inline __attribute__((const))
+int ffs(int x)
+{
+	/* Note: (x & -x) gives us a mask that is the least significant
+	 * (rightmost) 1-bit of the value in x.
+	 */
+	return fls(x & -x);
+}
+
+/**
+ * __ffs - find first bit set
+ * @x: the word to search
+ *
+ * - return 31..0 to indicate bit 31..0 most least significant bit set
+ * - if no bits are set in x, the result is undefined
+ */
+static inline __attribute__((const))
+int __ffs(unsigned long x)
+{
+	int bit;
+	asm("scan %1,gr0,%0" : "=r"(bit) : "r"(x & -x));
+	return 31 - bit;
+}
+
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/hweight.h>
 
-- 
cgit v1.2.1


From 53e62d3aaa60590d4a69b4e07c29f448b5151047 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 25 Sep 2006 23:32:10 -0700
Subject: [PATCH] Alchemy: Delete unused pt_regs * argument from
 au1xxx_dbdma_chan_alloc

The third argument of au1xxx_dbdma_chan_alloc's callback function is not
used anywhere.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Cc: David Howells <dhowells@redhat.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/au1000/common/dbdma.c             | 10 ++++++----
 drivers/ide/mips/au1xxx-ide.c               |  4 ++--
 drivers/mmc/au1xmmc.c                       |  2 +-
 include/asm-mips/mach-au1x00/au1xxx_dbdma.h |  6 +++---
 sound/oss/au1550_ac97.c                     |  6 ++----
 5 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/arch/mips/au1000/common/dbdma.c b/arch/mips/au1000/common/dbdma.c
index 98244d51c154..c4fae8ff4671 100644
--- a/arch/mips/au1000/common/dbdma.c
+++ b/arch/mips/au1000/common/dbdma.c
@@ -230,7 +230,7 @@ EXPORT_SYMBOL(au1xxx_ddma_add_device);
 */
 u32
 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
-       void (*callback)(int, void *, struct pt_regs *), void *callparam)
+       void (*callback)(int, void *), void *callparam)
 {
 	unsigned long   flags;
 	u32		used, chan, rv;
@@ -248,8 +248,10 @@ au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
 		au1xxx_dbdma_init();
 	dbdma_initialized = 1;
 
-	if ((stp = find_dbdev_id(srcid)) == NULL) return 0;
-	if ((dtp = find_dbdev_id(destid)) == NULL) return 0;
+	if ((stp = find_dbdev_id(srcid)) == NULL)
+		return 0;
+	if ((dtp = find_dbdev_id(destid)) == NULL)
+		return 0;
 
 	used = 0;
 	rv = 0;
@@ -869,7 +871,7 @@ dbdma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	au_sync();
 
 	if (ctp->chan_callback)
-		(ctp->chan_callback)(irq, ctp->chan_callparam, regs);
+		(ctp->chan_callback)(irq, ctp->chan_callparam);
 
 	ctp->cur_ptr = phys_to_virt(DSCR_GET_NXTPTR(dp->dscr_nxtptr));
 	return IRQ_RETVAL(1);
diff --git a/drivers/ide/mips/au1xxx-ide.c b/drivers/ide/mips/au1xxx-ide.c
index 71f27e955d87..c7854ea57b52 100644
--- a/drivers/ide/mips/au1xxx-ide.c
+++ b/drivers/ide/mips/au1xxx-ide.c
@@ -476,13 +476,13 @@ static int auide_dma_lostirq(ide_drive_t *drive)
 	return 0;
 }
 
-static void auide_ddma_tx_callback(int irq, void *param, struct pt_regs *regs)
+static void auide_ddma_tx_callback(int irq, void *param)
 {
 	_auide_hwif *ahwif = (_auide_hwif*)param;
 	ahwif->drive->waiting_for_dma = 0;
 }
 
-static void auide_ddma_rx_callback(int irq, void *param, struct pt_regs *regs)
+static void auide_ddma_rx_callback(int irq, void *param)
 {
 	_auide_hwif *ahwif = (_auide_hwif*)param;
 	ahwif->drive->waiting_for_dma = 0;
diff --git a/drivers/mmc/au1xmmc.c b/drivers/mmc/au1xmmc.c
index fb606165af3b..61268da13957 100644
--- a/drivers/mmc/au1xmmc.c
+++ b/drivers/mmc/au1xmmc.c
@@ -731,7 +731,7 @@ static void au1xmmc_set_ios(struct mmc_host* mmc, struct mmc_ios* ios)
 	}
 }
 
-static void au1xmmc_dma_callback(int irq, void *dev_id, struct pt_regs *regs)
+static void au1xmmc_dma_callback(int irq, void *dev_id)
 {
 	struct au1xmmc_host *host = (struct au1xmmc_host *) dev_id;
 
diff --git a/include/asm-mips/mach-au1x00/au1xxx_dbdma.h b/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
index d5b38a247e5a..eeb0c3115b6a 100644
--- a/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
+++ b/include/asm-mips/mach-au1x00/au1xxx_dbdma.h
@@ -316,7 +316,7 @@ typedef struct dbdma_chan_config {
 	au1x_ddma_desc_t	*chan_desc_base;
 	au1x_ddma_desc_t	*get_ptr, *put_ptr, *cur_ptr;
 	void			*chan_callparam;
-	void (*chan_callback)(int, void *, struct pt_regs *);
+	void (*chan_callback)(int, void *);
 } chan_tab_t;
 
 #define DEV_FLAGS_INUSE		(1 << 0)
@@ -334,8 +334,8 @@ typedef struct dbdma_chan_config {
  * meaningful name.  The 'callback' is called during dma completion
  * interrupt.
  */
-u32 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
-       void (*callback)(int, void *, struct pt_regs *), void *callparam);
+extern u32 au1xxx_dbdma_chan_alloc(u32 srcid, u32 destid,
+	void (*callback)(int, void *), void *callparam);
 
 #define DBDMA_MEM_CHAN	DSCR_CMD0_ALWAYS
 
diff --git a/sound/oss/au1550_ac97.c b/sound/oss/au1550_ac97.c
index 4cdb86252d67..219795171c71 100644
--- a/sound/oss/au1550_ac97.c
+++ b/sound/oss/au1550_ac97.c
@@ -719,8 +719,7 @@ prog_dmabuf_dac(struct au1550_state *s)
 }
 
 
-static void
-dac_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+static void dac_dma_interrupt(int irq, void *dev_id)
 {
 	struct au1550_state *s = (struct au1550_state *) dev_id;
 	struct dmabuf  *db = &s->dma_dac;
@@ -754,8 +753,7 @@ dac_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 }
 
 
-static void
-adc_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+static void adc_dma_interrupt(int irq, void *dev_id)
 {
 	struct	au1550_state *s = (struct au1550_state *)dev_id;
 	struct	dmabuf  *dp = &s->dma_adc;
-- 
cgit v1.2.1


From 5f97f7f9400de47ae837170bb274e90ad3934386 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 25 Sep 2006 23:32:13 -0700
Subject: [PATCH] avr32 architecture

This adds support for the Atmel AVR32 architecture as well as the AT32AP7000
CPU and the AT32STK1000 development board.

AVR32 is a new high-performance 32-bit RISC microprocessor core, designed for
cost-sensitive embedded applications, with particular emphasis on low power
consumption and high code density.  The AVR32 architecture is not binary
compatible with earlier 8-bit AVR architectures.

The AVR32 architecture, including the instruction set, is described by the
AVR32 Architecture Manual, available from

http://www.atmel.com/dyn/resources/prod_documents/doc32000.pdf

The Atmel AT32AP7000 is the first CPU implementing the AVR32 architecture.  It
features a 7-stage pipeline, 16KB instruction and data caches and a full
Memory Management Unit.  It also comes with a large set of integrated
peripherals, many of which are shared with the AT91 ARM-based controllers from
Atmel.

Full data sheet is available from

http://www.atmel.com/dyn/resources/prod_documents/doc32003.pdf

while the CPU core implementation including caches and MMU is documented by
the AVR32 AP Technical Reference, available from

http://www.atmel.com/dyn/resources/prod_documents/doc32001.pdf

Information about the AT32STK1000 development board can be found at

http://www.atmel.com/dyn/products/tools_card.asp?tool_id=3918

including a BSP CD image with an earlier version of this patch, development
tools (binaries and source/patches) and a root filesystem image suitable for
booting from SD card.

Alternatively, there's a preliminary "getting started" guide available at
http://avr32linux.org/twiki/bin/view/Main/GettingStarted which provides links
to the sources and patches you will need in order to set up a cross-compiling
environment for avr32-linux.

This patch, as well as the other patches included with the BSP and the
toolchain patches, is actively supported by Atmel Corporation.

[dmccr@us.ibm.com: Fix more pxx_page macro locations]
[bunk@stusta.de: fix `make defconfig']
Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Dave McCracken <dmccr@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 MAINTAINERS                                      |  17 +
 arch/avr32/Kconfig                               | 196 +++++
 arch/avr32/Kconfig.debug                         |  19 +
 arch/avr32/Makefile                              |  84 +++
 arch/avr32/boards/atstk1000/Makefile             |   2 +
 arch/avr32/boards/atstk1000/atstk1002.c          |  37 +
 arch/avr32/boards/atstk1000/setup.c              |  59 ++
 arch/avr32/boards/atstk1000/spi.c                |  27 +
 arch/avr32/boot/images/Makefile                  |  62 ++
 arch/avr32/boot/u-boot/Makefile                  |   3 +
 arch/avr32/boot/u-boot/empty.S                   |   1 +
 arch/avr32/boot/u-boot/head.S                    |  60 ++
 arch/avr32/configs/atstk1002_defconfig           | 754 ++++++++++++++++++++
 arch/avr32/kernel/Makefile                       |  18 +
 arch/avr32/kernel/asm-offsets.c                  |  25 +
 arch/avr32/kernel/avr32_ksyms.c                  |  55 ++
 arch/avr32/kernel/cpu.c                          | 327 +++++++++
 arch/avr32/kernel/entry-avr32b.S                 | 678 ++++++++++++++++++
 arch/avr32/kernel/head.S                         |  45 ++
 arch/avr32/kernel/init_task.c                    |  38 +
 arch/avr32/kernel/irq.c                          |  71 ++
 arch/avr32/kernel/kprobes.c                      | 270 +++++++
 arch/avr32/kernel/module.c                       | 324 +++++++++
 arch/avr32/kernel/process.c                      | 276 ++++++++
 arch/avr32/kernel/ptrace.c                       | 371 ++++++++++
 arch/avr32/kernel/semaphore.c                    | 148 ++++
 arch/avr32/kernel/setup.c                        | 335 +++++++++
 arch/avr32/kernel/signal.c                       | 328 +++++++++
 arch/avr32/kernel/switch_to.S                    |  35 +
 arch/avr32/kernel/sys_avr32.c                    |  51 ++
 arch/avr32/kernel/syscall-stubs.S                | 102 +++
 arch/avr32/kernel/syscall_table.S                | 289 ++++++++
 arch/avr32/kernel/time.c                         | 238 +++++++
 arch/avr32/kernel/traps.c                        | 425 +++++++++++
 arch/avr32/kernel/vmlinux.lds.c                  | 139 ++++
 arch/avr32/lib/Makefile                          |  10 +
 arch/avr32/lib/__avr32_asr64.S                   |  31 +
 arch/avr32/lib/__avr32_lsl64.S                   |  31 +
 arch/avr32/lib/__avr32_lsr64.S                   |  31 +
 arch/avr32/lib/clear_user.S                      |  76 ++
 arch/avr32/lib/copy_user.S                       | 119 ++++
 arch/avr32/lib/csum_partial.S                    |  47 ++
 arch/avr32/lib/csum_partial_copy_generic.S       |  99 +++
 arch/avr32/lib/delay.c                           |  55 ++
 arch/avr32/lib/findbit.S                         | 154 ++++
 arch/avr32/lib/io-readsl.S                       |  24 +
 arch/avr32/lib/io-readsw.S                       |  43 ++
 arch/avr32/lib/io-writesl.S                      |  20 +
 arch/avr32/lib/io-writesw.S                      |  38 +
 arch/avr32/lib/libgcc.h                          |  33 +
 arch/avr32/lib/longlong.h                        |  98 +++
 arch/avr32/lib/memcpy.S                          |  62 ++
 arch/avr32/lib/memset.S                          |  72 ++
 arch/avr32/lib/strncpy_from_user.S               |  60 ++
 arch/avr32/lib/strnlen_user.S                    |  67 ++
 arch/avr32/mach-at32ap/Makefile                  |   2 +
 arch/avr32/mach-at32ap/at32ap.c                  |  90 +++
 arch/avr32/mach-at32ap/at32ap7000.c              | 866 +++++++++++++++++++++++
 arch/avr32/mach-at32ap/clock.c                   | 148 ++++
 arch/avr32/mach-at32ap/clock.h                   |  30 +
 arch/avr32/mach-at32ap/extint.c                  | 171 +++++
 arch/avr32/mach-at32ap/intc.c                    | 133 ++++
 arch/avr32/mach-at32ap/intc.h                    | 327 +++++++++
 arch/avr32/mach-at32ap/pio.c                     | 118 +++
 arch/avr32/mach-at32ap/pio.h                     | 178 +++++
 arch/avr32/mach-at32ap/sm.c                      | 289 ++++++++
 arch/avr32/mach-at32ap/sm.h                      | 240 +++++++
 arch/avr32/mm/Makefile                           |   6 +
 arch/avr32/mm/cache.c                            | 150 ++++
 arch/avr32/mm/clear_page.S                       |  25 +
 arch/avr32/mm/copy_page.S                        |  28 +
 arch/avr32/mm/dma-coherent.c                     | 139 ++++
 arch/avr32/mm/fault.c                            | 315 +++++++++
 arch/avr32/mm/init.c                             | 480 +++++++++++++
 arch/avr32/mm/ioremap.c                          | 197 ++++++
 arch/avr32/mm/tlb.c                              | 378 ++++++++++
 include/asm-avr32/Kbuild                         |   3 +
 include/asm-avr32/a.out.h                        |  26 +
 include/asm-avr32/addrspace.h                    |  43 ++
 include/asm-avr32/arch-at32ap/at91rm9200_pdc.h   |  36 +
 include/asm-avr32/arch-at32ap/at91rm9200_usart.h | 123 ++++
 include/asm-avr32/arch-at32ap/board.h            |  35 +
 include/asm-avr32/arch-at32ap/init.h             |  21 +
 include/asm-avr32/arch-at32ap/portmux.h          |  16 +
 include/asm-avr32/arch-at32ap/sm.h               |  27 +
 include/asm-avr32/asm.h                          | 102 +++
 include/asm-avr32/atomic.h                       | 201 ++++++
 include/asm-avr32/auxvec.h                       |   4 +
 include/asm-avr32/bitops.h                       | 296 ++++++++
 include/asm-avr32/bug.h                          |  47 ++
 include/asm-avr32/bugs.h                         |  15 +
 include/asm-avr32/byteorder.h                    |  25 +
 include/asm-avr32/cache.h                        |  29 +
 include/asm-avr32/cachectl.h                     |  11 +
 include/asm-avr32/cacheflush.h                   | 129 ++++
 include/asm-avr32/checksum.h                     | 156 ++++
 include/asm-avr32/cputime.h                      |   6 +
 include/asm-avr32/current.h                      |  15 +
 include/asm-avr32/delay.h                        |  26 +
 include/asm-avr32/div64.h                        |   6 +
 include/asm-avr32/dma-mapping.h                  | 320 +++++++++
 include/asm-avr32/dma.h                          |   8 +
 include/asm-avr32/elf.h                          | 110 +++
 include/asm-avr32/emergency-restart.h            |   6 +
 include/asm-avr32/errno.h                        |   6 +
 include/asm-avr32/fcntl.h                        |   6 +
 include/asm-avr32/futex.h                        |   6 +
 include/asm-avr32/hardirq.h                      |  34 +
 include/asm-avr32/hw_irq.h                       |   9 +
 include/asm-avr32/intc.h                         | 128 ++++
 include/asm-avr32/io.h                           | 253 +++++++
 include/asm-avr32/ioctl.h                        |   6 +
 include/asm-avr32/ioctls.h                       |  83 +++
 include/asm-avr32/ipcbuf.h                       |  29 +
 include/asm-avr32/irq.h                          |  10 +
 include/asm-avr32/irqflags.h                     |  68 ++
 include/asm-avr32/kdebug.h                       |  38 +
 include/asm-avr32/kmap_types.h                   |  30 +
 include/asm-avr32/kprobes.h                      |  34 +
 include/asm-avr32/linkage.h                      |   7 +
 include/asm-avr32/local.h                        |   6 +
 include/asm-avr32/mach/serial_at91.h             |  33 +
 include/asm-avr32/mman.h                         |  17 +
 include/asm-avr32/mmu.h                          |  10 +
 include/asm-avr32/mmu_context.h                  | 148 ++++
 include/asm-avr32/module.h                       |  28 +
 include/asm-avr32/msgbuf.h                       |  31 +
 include/asm-avr32/mutex.h                        |   9 +
 include/asm-avr32/namei.h                        |   7 +
 include/asm-avr32/numnodes.h                     |   7 +
 include/asm-avr32/ocd.h                          |  78 ++
 include/asm-avr32/page.h                         | 112 +++
 include/asm-avr32/param.h                        |  23 +
 include/asm-avr32/pci.h                          |   8 +
 include/asm-avr32/percpu.h                       |   6 +
 include/asm-avr32/pgalloc.h                      |  96 +++
 include/asm-avr32/pgtable-2level.h               |  47 ++
 include/asm-avr32/pgtable.h                      | 408 +++++++++++
 include/asm-avr32/poll.h                         |  27 +
 include/asm-avr32/posix_types.h                  | 129 ++++
 include/asm-avr32/processor.h                    | 147 ++++
 include/asm-avr32/ptrace.h                       | 154 ++++
 include/asm-avr32/resource.h                     |   6 +
 include/asm-avr32/scatterlist.h                  |  21 +
 include/asm-avr32/sections.h                     |   6 +
 include/asm-avr32/semaphore.h                    | 109 +++
 include/asm-avr32/sembuf.h                       |  25 +
 include/asm-avr32/setup.h                        | 141 ++++
 include/asm-avr32/shmbuf.h                       |  42 ++
 include/asm-avr32/shmparam.h                     |   6 +
 include/asm-avr32/sigcontext.h                   |  34 +
 include/asm-avr32/siginfo.h                      |   6 +
 include/asm-avr32/signal.h                       | 168 +++++
 include/asm-avr32/socket.h                       |  53 ++
 include/asm-avr32/sockios.h                      |  12 +
 include/asm-avr32/stat.h                         |  79 +++
 include/asm-avr32/statfs.h                       |   6 +
 include/asm-avr32/string.h                       |  17 +
 include/asm-avr32/sysreg.h                       | 332 +++++++++
 include/asm-avr32/system.h                       | 155 ++++
 include/asm-avr32/termbits.h                     | 173 +++++
 include/asm-avr32/termios.h                      |  80 +++
 include/asm-avr32/thread_info.h                  | 106 +++
 include/asm-avr32/timex.h                        |  40 ++
 include/asm-avr32/tlb.h                          |  32 +
 include/asm-avr32/tlbflush.h                     |  40 ++
 include/asm-avr32/topology.h                     |   6 +
 include/asm-avr32/traps.h                        |  23 +
 include/asm-avr32/types.h                        |  70 ++
 include/asm-avr32/uaccess.h                      | 335 +++++++++
 include/asm-avr32/ucontext.h                     |  12 +
 include/asm-avr32/unaligned.h                    |  25 +
 include/asm-avr32/unistd.h                       | 387 ++++++++++
 include/asm-avr32/user.h                         |  65 ++
 include/linux/elf-em.h                           |   1 +
 lib/Kconfig.debug                                |   4 +-
 176 files changed, 18124 insertions(+), 2 deletions(-)
 create mode 100644 arch/avr32/Kconfig
 create mode 100644 arch/avr32/Kconfig.debug
 create mode 100644 arch/avr32/Makefile
 create mode 100644 arch/avr32/boards/atstk1000/Makefile
 create mode 100644 arch/avr32/boards/atstk1000/atstk1002.c
 create mode 100644 arch/avr32/boards/atstk1000/setup.c
 create mode 100644 arch/avr32/boards/atstk1000/spi.c
 create mode 100644 arch/avr32/boot/images/Makefile
 create mode 100644 arch/avr32/boot/u-boot/Makefile
 create mode 100644 arch/avr32/boot/u-boot/empty.S
 create mode 100644 arch/avr32/boot/u-boot/head.S
 create mode 100644 arch/avr32/configs/atstk1002_defconfig
 create mode 100644 arch/avr32/kernel/Makefile
 create mode 100644 arch/avr32/kernel/asm-offsets.c
 create mode 100644 arch/avr32/kernel/avr32_ksyms.c
 create mode 100644 arch/avr32/kernel/cpu.c
 create mode 100644 arch/avr32/kernel/entry-avr32b.S
 create mode 100644 arch/avr32/kernel/head.S
 create mode 100644 arch/avr32/kernel/init_task.c
 create mode 100644 arch/avr32/kernel/irq.c
 create mode 100644 arch/avr32/kernel/kprobes.c
 create mode 100644 arch/avr32/kernel/module.c
 create mode 100644 arch/avr32/kernel/process.c
 create mode 100644 arch/avr32/kernel/ptrace.c
 create mode 100644 arch/avr32/kernel/semaphore.c
 create mode 100644 arch/avr32/kernel/setup.c
 create mode 100644 arch/avr32/kernel/signal.c
 create mode 100644 arch/avr32/kernel/switch_to.S
 create mode 100644 arch/avr32/kernel/sys_avr32.c
 create mode 100644 arch/avr32/kernel/syscall-stubs.S
 create mode 100644 arch/avr32/kernel/syscall_table.S
 create mode 100644 arch/avr32/kernel/time.c
 create mode 100644 arch/avr32/kernel/traps.c
 create mode 100644 arch/avr32/kernel/vmlinux.lds.c
 create mode 100644 arch/avr32/lib/Makefile
 create mode 100644 arch/avr32/lib/__avr32_asr64.S
 create mode 100644 arch/avr32/lib/__avr32_lsl64.S
 create mode 100644 arch/avr32/lib/__avr32_lsr64.S
 create mode 100644 arch/avr32/lib/clear_user.S
 create mode 100644 arch/avr32/lib/copy_user.S
 create mode 100644 arch/avr32/lib/csum_partial.S
 create mode 100644 arch/avr32/lib/csum_partial_copy_generic.S
 create mode 100644 arch/avr32/lib/delay.c
 create mode 100644 arch/avr32/lib/findbit.S
 create mode 100644 arch/avr32/lib/io-readsl.S
 create mode 100644 arch/avr32/lib/io-readsw.S
 create mode 100644 arch/avr32/lib/io-writesl.S
 create mode 100644 arch/avr32/lib/io-writesw.S
 create mode 100644 arch/avr32/lib/libgcc.h
 create mode 100644 arch/avr32/lib/longlong.h
 create mode 100644 arch/avr32/lib/memcpy.S
 create mode 100644 arch/avr32/lib/memset.S
 create mode 100644 arch/avr32/lib/strncpy_from_user.S
 create mode 100644 arch/avr32/lib/strnlen_user.S
 create mode 100644 arch/avr32/mach-at32ap/Makefile
 create mode 100644 arch/avr32/mach-at32ap/at32ap.c
 create mode 100644 arch/avr32/mach-at32ap/at32ap7000.c
 create mode 100644 arch/avr32/mach-at32ap/clock.c
 create mode 100644 arch/avr32/mach-at32ap/clock.h
 create mode 100644 arch/avr32/mach-at32ap/extint.c
 create mode 100644 arch/avr32/mach-at32ap/intc.c
 create mode 100644 arch/avr32/mach-at32ap/intc.h
 create mode 100644 arch/avr32/mach-at32ap/pio.c
 create mode 100644 arch/avr32/mach-at32ap/pio.h
 create mode 100644 arch/avr32/mach-at32ap/sm.c
 create mode 100644 arch/avr32/mach-at32ap/sm.h
 create mode 100644 arch/avr32/mm/Makefile
 create mode 100644 arch/avr32/mm/cache.c
 create mode 100644 arch/avr32/mm/clear_page.S
 create mode 100644 arch/avr32/mm/copy_page.S
 create mode 100644 arch/avr32/mm/dma-coherent.c
 create mode 100644 arch/avr32/mm/fault.c
 create mode 100644 arch/avr32/mm/init.c
 create mode 100644 arch/avr32/mm/ioremap.c
 create mode 100644 arch/avr32/mm/tlb.c
 create mode 100644 include/asm-avr32/Kbuild
 create mode 100644 include/asm-avr32/a.out.h
 create mode 100644 include/asm-avr32/addrspace.h
 create mode 100644 include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
 create mode 100644 include/asm-avr32/arch-at32ap/at91rm9200_usart.h
 create mode 100644 include/asm-avr32/arch-at32ap/board.h
 create mode 100644 include/asm-avr32/arch-at32ap/init.h
 create mode 100644 include/asm-avr32/arch-at32ap/portmux.h
 create mode 100644 include/asm-avr32/arch-at32ap/sm.h
 create mode 100644 include/asm-avr32/asm.h
 create mode 100644 include/asm-avr32/atomic.h
 create mode 100644 include/asm-avr32/auxvec.h
 create mode 100644 include/asm-avr32/bitops.h
 create mode 100644 include/asm-avr32/bug.h
 create mode 100644 include/asm-avr32/bugs.h
 create mode 100644 include/asm-avr32/byteorder.h
 create mode 100644 include/asm-avr32/cache.h
 create mode 100644 include/asm-avr32/cachectl.h
 create mode 100644 include/asm-avr32/cacheflush.h
 create mode 100644 include/asm-avr32/checksum.h
 create mode 100644 include/asm-avr32/cputime.h
 create mode 100644 include/asm-avr32/current.h
 create mode 100644 include/asm-avr32/delay.h
 create mode 100644 include/asm-avr32/div64.h
 create mode 100644 include/asm-avr32/dma-mapping.h
 create mode 100644 include/asm-avr32/dma.h
 create mode 100644 include/asm-avr32/elf.h
 create mode 100644 include/asm-avr32/emergency-restart.h
 create mode 100644 include/asm-avr32/errno.h
 create mode 100644 include/asm-avr32/fcntl.h
 create mode 100644 include/asm-avr32/futex.h
 create mode 100644 include/asm-avr32/hardirq.h
 create mode 100644 include/asm-avr32/hw_irq.h
 create mode 100644 include/asm-avr32/intc.h
 create mode 100644 include/asm-avr32/io.h
 create mode 100644 include/asm-avr32/ioctl.h
 create mode 100644 include/asm-avr32/ioctls.h
 create mode 100644 include/asm-avr32/ipcbuf.h
 create mode 100644 include/asm-avr32/irq.h
 create mode 100644 include/asm-avr32/irqflags.h
 create mode 100644 include/asm-avr32/kdebug.h
 create mode 100644 include/asm-avr32/kmap_types.h
 create mode 100644 include/asm-avr32/kprobes.h
 create mode 100644 include/asm-avr32/linkage.h
 create mode 100644 include/asm-avr32/local.h
 create mode 100644 include/asm-avr32/mach/serial_at91.h
 create mode 100644 include/asm-avr32/mman.h
 create mode 100644 include/asm-avr32/mmu.h
 create mode 100644 include/asm-avr32/mmu_context.h
 create mode 100644 include/asm-avr32/module.h
 create mode 100644 include/asm-avr32/msgbuf.h
 create mode 100644 include/asm-avr32/mutex.h
 create mode 100644 include/asm-avr32/namei.h
 create mode 100644 include/asm-avr32/numnodes.h
 create mode 100644 include/asm-avr32/ocd.h
 create mode 100644 include/asm-avr32/page.h
 create mode 100644 include/asm-avr32/param.h
 create mode 100644 include/asm-avr32/pci.h
 create mode 100644 include/asm-avr32/percpu.h
 create mode 100644 include/asm-avr32/pgalloc.h
 create mode 100644 include/asm-avr32/pgtable-2level.h
 create mode 100644 include/asm-avr32/pgtable.h
 create mode 100644 include/asm-avr32/poll.h
 create mode 100644 include/asm-avr32/posix_types.h
 create mode 100644 include/asm-avr32/processor.h
 create mode 100644 include/asm-avr32/ptrace.h
 create mode 100644 include/asm-avr32/resource.h
 create mode 100644 include/asm-avr32/scatterlist.h
 create mode 100644 include/asm-avr32/sections.h
 create mode 100644 include/asm-avr32/semaphore.h
 create mode 100644 include/asm-avr32/sembuf.h
 create mode 100644 include/asm-avr32/setup.h
 create mode 100644 include/asm-avr32/shmbuf.h
 create mode 100644 include/asm-avr32/shmparam.h
 create mode 100644 include/asm-avr32/sigcontext.h
 create mode 100644 include/asm-avr32/siginfo.h
 create mode 100644 include/asm-avr32/signal.h
 create mode 100644 include/asm-avr32/socket.h
 create mode 100644 include/asm-avr32/sockios.h
 create mode 100644 include/asm-avr32/stat.h
 create mode 100644 include/asm-avr32/statfs.h
 create mode 100644 include/asm-avr32/string.h
 create mode 100644 include/asm-avr32/sysreg.h
 create mode 100644 include/asm-avr32/system.h
 create mode 100644 include/asm-avr32/termbits.h
 create mode 100644 include/asm-avr32/termios.h
 create mode 100644 include/asm-avr32/thread_info.h
 create mode 100644 include/asm-avr32/timex.h
 create mode 100644 include/asm-avr32/tlb.h
 create mode 100644 include/asm-avr32/tlbflush.h
 create mode 100644 include/asm-avr32/topology.h
 create mode 100644 include/asm-avr32/traps.h
 create mode 100644 include/asm-avr32/types.h
 create mode 100644 include/asm-avr32/uaccess.h
 create mode 100644 include/asm-avr32/ucontext.h
 create mode 100644 include/asm-avr32/unaligned.h
 create mode 100644 include/asm-avr32/unistd.h
 create mode 100644 include/asm-avr32/user.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 23348c0d37bc..767a43434ae8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -443,6 +443,23 @@ W:	http://people.redhat.com/sgrubb/audit/
 T:	git kernel.org:/pub/scm/linux/kernel/git/dwmw2/audit-2.6.git
 S:	Maintained
 
+AVR32 ARCHITECTURE
+P:	Atmel AVR32 Support Team
+M:	avr32@atmel.com
+P:	Haavard Skinnemoen
+M:	hskinnemoen@atmel.com
+W:	http://www.atmel.com/products/AVR32/
+W:	http://avr32linux.org/
+W:	http://avrfreaks.net/
+S:	Supported
+
+AVR32/AT32AP MACHINE SUPPORT
+P:	Atmel AVR32 Support Team
+M:	avr32@atmel.com
+P:	Haavard Skinnemoen
+M:	hskinnemoen@atmel.com
+S:	Supported
+
 AX.25 NETWORK LAYER
 P:	Ralf Baechle
 M:	ralf@linux-mips.org
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
new file mode 100644
index 000000000000..5f1694eea842
--- /dev/null
+++ b/arch/avr32/Kconfig
@@ -0,0 +1,196 @@
+#
+# For a description of the syntax of this configuration file,
+# see Documentation/kbuild/kconfig-language.txt.
+#
+
+mainmenu "Linux Kernel Configuration"
+
+config AVR32
+	bool
+	default y
+	# With EMBEDDED=n, we get lots of stuff automatically selected
+	# that we usually don't need on AVR32.
+	select EMBEDDED
+	help
+	  AVR32 is a high-performance 32-bit RISC microprocessor core,
+	  designed for cost-sensitive embedded applications, with particular
+	  emphasis on low power consumption and high code density.
+
+	  There is an AVR32 Linux project with a web page at
+	  http://avr32linux.org/.
+
+config UID16
+	bool
+
+config GENERIC_HARDIRQS
+	bool
+	default y
+
+config HARDIRQS_SW_RESEND
+	bool
+	default y
+
+config GENERIC_IRQ_PROBE
+	bool
+	default y
+
+config RWSEM_GENERIC_SPINLOCK
+	bool
+	default y
+
+config GENERIC_TIME
+	bool
+	default y
+
+config RWSEM_XCHGADD_ALGORITHM
+	bool
+
+config GENERIC_BUST_SPINLOCK
+	bool
+
+config GENERIC_HWEIGHT
+	bool
+	default y
+
+config GENERIC_CALIBRATE_DELAY
+	bool
+	default y
+
+source "init/Kconfig"
+
+menu "System Type and features"
+
+config SUBARCH_AVR32B
+	bool
+config MMU
+	bool
+config PERFORMANCE_COUNTERS
+	bool
+
+config PLATFORM_AT32AP
+	bool
+	select SUBARCH_AVR32B
+	select MMU
+	select PERFORMANCE_COUNTERS
+
+choice
+	prompt "AVR32 CPU type"
+	default CPU_AT32AP7000
+
+config CPU_AT32AP7000
+	bool "AT32AP7000"
+	select PLATFORM_AT32AP
+endchoice
+
+#
+# CPU Daughterboards for ATSTK1000
+config BOARD_ATSTK1002
+	bool
+
+choice
+	prompt "AVR32 board type"
+	default BOARD_ATSTK1000
+
+config BOARD_ATSTK1000
+	bool "ATSTK1000 evaluation board"
+	select BOARD_ATSTK1002 if CPU_AT32AP7000
+endchoice
+
+choice
+	prompt "Boot loader type"
+	default LOADER_U_BOOT
+
+config	LOADER_U_BOOT
+	bool "U-Boot (or similar) bootloader"
+endchoice
+
+config LOAD_ADDRESS
+	hex
+	default 0x10000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+
+config ENTRY_ADDRESS
+	hex
+	default 0x90000000 if LOADER_U_BOOT=y && CPU_AT32AP7000=y
+
+config PHYS_OFFSET
+	hex
+	default 0x10000000 if CPU_AT32AP7000=y
+
+source "kernel/Kconfig.preempt"
+
+config HAVE_ARCH_BOOTMEM_NODE
+	bool
+	default n
+
+config ARCH_HAVE_MEMORY_PRESENT
+	bool
+	default n
+
+config NEED_NODE_MEMMAP_SIZE
+	bool
+	default n
+
+config ARCH_FLATMEM_ENABLE
+	bool
+	default y
+
+config ARCH_DISCONTIGMEM_ENABLE
+	bool
+	default n
+
+config ARCH_SPARSEMEM_ENABLE
+	bool
+	default n
+
+source "mm/Kconfig"
+
+config OWNERSHIP_TRACE
+	bool "Ownership trace support"
+	default y
+	help
+	  Say Y to generate an Ownership Trace message on every context switch,
+	  enabling Nexus-compliant debuggers to keep track of the PID of the
+	  currently executing task.
+
+# FPU emulation goes here
+
+source "kernel/Kconfig.hz"
+
+config CMDLINE
+	string "Default kernel command line"
+	default ""
+	help
+	  If you don't have a boot loader capable of passing a command line string
+	  to the kernel, you may specify one here. As a minimum, you should specify
+	  the memory size and the root device (e.g., mem=8M, root=/dev/nfs).
+
+endmenu
+
+menu "Bus options"
+
+config PCI
+	bool
+
+source "drivers/pci/Kconfig"
+
+source "drivers/pcmcia/Kconfig"
+
+endmenu
+
+menu "Executable file formats"
+source "fs/Kconfig.binfmt"
+endmenu
+
+source "net/Kconfig"
+
+source "drivers/Kconfig"
+
+source "fs/Kconfig"
+
+source "arch/avr32/Kconfig.debug"
+
+source "security/Kconfig"
+
+source "crypto/Kconfig"
+
+source "lib/Kconfig"
diff --git a/arch/avr32/Kconfig.debug b/arch/avr32/Kconfig.debug
new file mode 100644
index 000000000000..64ace00fe6cb
--- /dev/null
+++ b/arch/avr32/Kconfig.debug
@@ -0,0 +1,19 @@
+menu "Kernel hacking"
+
+config TRACE_IRQFLAGS_SUPPORT
+	bool
+	default y
+
+source "lib/Kconfig.debug"
+
+config KPROBES
+	bool "Kprobes"
+	depends on DEBUG_KERNEL
+	help
+	  Kprobes allows you to trap at almost any kernel address and
+          execute a callback function.  register_kprobe() establishes
+          a probepoint and specifies the callback.  Kprobes is useful
+          for kernel debugging, non-intrusive instrumentation and testing.
+          If in doubt, say "N".
+
+endmenu
diff --git a/arch/avr32/Makefile b/arch/avr32/Makefile
new file mode 100644
index 000000000000..cefc95a73980
--- /dev/null
+++ b/arch/avr32/Makefile
@@ -0,0 +1,84 @@
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (C) 2004-2006 Atmel Corporation.
+
+# Default target when executing plain make
+.PHONY: all
+all: uImage vmlinux.elf linux.lst
+
+KBUILD_DEFCONFIG	:= atstk1002_defconfig
+
+CFLAGS		+= -pipe -fno-builtin -mno-pic
+AFLAGS		+= -mrelax -mno-pic
+CFLAGS_MODULE	+= -mno-relax
+LDFLAGS_vmlinux	+= --relax
+
+cpuflags-$(CONFIG_CPU_AP7000)	+= -mcpu=ap7000
+
+CFLAGS		+= $(cpuflags-y)
+AFLAGS		+= $(cpuflags-y)
+
+CHECKFLAGS	+= -D__avr32__
+
+LIBGCC		:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
+
+head-$(CONFIG_LOADER_U_BOOT)		+= arch/avr32/boot/u-boot/head.o
+head-y					+= arch/avr32/kernel/head.o
+core-$(CONFIG_PLATFORM_AT32AP)		+= arch/avr32/mach-at32ap/
+core-$(CONFIG_BOARD_ATSTK1000)		+= arch/avr32/boards/atstk1000/
+core-$(CONFIG_LOADER_U_BOOT)		+= arch/avr32/boot/u-boot/
+core-y					+= arch/avr32/kernel/
+core-y					+= arch/avr32/mm/
+libs-y					+= arch/avr32/lib/ #$(LIBGCC)
+
+archincdir-$(CONFIG_PLATFORM_AT32AP)	:= arch-at32ap
+
+include/asm-avr32/.arch: $(wildcard include/config/platform/*.h) include/config/auto.conf
+	@echo '  SYMLINK include/asm-avr32/arch -> include/asm-avr32/$(archincdir-y)'
+ifneq ($(KBUILD_SRC),)
+	$(Q)mkdir -p include/asm-avr32
+	$(Q)ln -fsn $(srctree)/include/asm-avr32/$(archincdir-y) include/asm-avr32/arch
+else
+	$(Q)ln -fsn $(archincdir-y) include/asm-avr32/arch
+endif
+	@touch $@
+
+archprepare: include/asm-avr32/.arch
+
+BOOT_TARGETS := vmlinux.elf vmlinux.bin uImage uImage.srec
+
+.PHONY: $(BOOT_TARGETS) install
+
+boot := arch/$(ARCH)/boot/images
+
+             KBUILD_IMAGE := $(boot)/uImage
+vmlinux.elf: KBUILD_IMAGE := $(boot)/vmlinux.elf
+vmlinux.cso: KBUILD_IMAGE := $(boot)/vmlinux.cso
+uImage.srec: KBUILD_IMAGE := $(boot)/uImage.srec
+uImage:      KBUILD_IMAGE := $(boot)/uImage
+
+quiet_cmd_listing = LST     $@
+      cmd_listing = avr32-linux-objdump $(OBJDUMPFLAGS) -lS $< > $@
+quiet_cmd_disasm  = DIS     $@
+      cmd_disasm  = avr32-linux-objdump $(OBJDUMPFLAGS) -d $< > $@
+
+vmlinux.elf vmlinux.bin uImage.srec uImage vmlinux.cso: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+
+install: vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) $@
+
+linux.s: vmlinux
+	$(call if_changed,disasm)
+
+linux.lst: vmlinux
+	$(call if_changed,listing)
+
+define archhelp
+  @echo '* vmlinux.elf		- ELF image with load address 0'
+  @echo '  vmlinux.cso		- PathFinder CSO image'
+  @echo '  uImage		- Create a bootable image for U-Boot'
+endef
diff --git a/arch/avr32/boards/atstk1000/Makefile b/arch/avr32/boards/atstk1000/Makefile
new file mode 100644
index 000000000000..c3e36ece8651
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/Makefile
@@ -0,0 +1,2 @@
+obj-y				+= setup.o spi.o
+obj-$(CONFIG_BOARD_ATSTK1002)	+= atstk1002.o
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
new file mode 100644
index 000000000000..49164e9aadd6
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -0,0 +1,37 @@
+/*
+ * ATSTK1002 daughterboard-specific init code
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+
+#include <asm/arch/board.h>
+
+struct eth_platform_data __initdata eth0_data = {
+	.valid		= 1,
+	.mii_phy_addr	= 0x10,
+	.is_rmii	= 0,
+	.hw_addr	= { 0x6a, 0x87, 0x71, 0x14, 0xcd, 0xcb },
+};
+
+extern struct lcdc_platform_data atstk1000_fb0_data;
+
+static int __init atstk1002_init(void)
+{
+	at32_add_system_devices();
+
+	at32_add_device_usart(1);	/* /dev/ttyS0 */
+	at32_add_device_usart(2);	/* /dev/ttyS1 */
+	at32_add_device_usart(3);	/* /dev/ttyS2 */
+
+	at32_add_device_eth(0, &eth0_data);
+	at32_add_device_spi(0);
+	at32_add_device_lcdc(0, &atstk1000_fb0_data);
+
+	return 0;
+}
+postcore_initcall(atstk1002_init);
diff --git a/arch/avr32/boards/atstk1000/setup.c b/arch/avr32/boards/atstk1000/setup.c
new file mode 100644
index 000000000000..191ab85de9a3
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/setup.c
@@ -0,0 +1,59 @@
+/*
+ * ATSTK1000 board-specific setup code.
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/bootmem.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/linkage.h>
+
+#include <asm/setup.h>
+
+#include <asm/arch/board.h>
+
+/* Initialized by bootloader-specific startup code. */
+struct tag *bootloader_tags __initdata;
+
+struct lcdc_platform_data __initdata atstk1000_fb0_data;
+
+asmlinkage void __init board_early_init(void)
+{
+	extern void sdram_init(void);
+
+#ifdef CONFIG_LOADER_STANDALONE
+	sdram_init();
+#endif
+}
+
+void __init board_setup_fbmem(unsigned long fbmem_start,
+			      unsigned long fbmem_size)
+{
+	if (!fbmem_size)
+		return;
+
+	if (!fbmem_start) {
+		void *fbmem;
+
+		fbmem = alloc_bootmem_low_pages(fbmem_size);
+		fbmem_start = __pa(fbmem);
+	} else {
+		pg_data_t *pgdat;
+
+		for_each_online_pgdat(pgdat) {
+			if (fbmem_start >= pgdat->bdata->node_boot_start
+			    && fbmem_start <= pgdat->bdata->node_low_pfn)
+				reserve_bootmem_node(pgdat, fbmem_start,
+						     fbmem_size);
+		}
+	}
+
+	printk("%luKiB framebuffer memory at address 0x%08lx\n",
+	       fbmem_size >> 10, fbmem_start);
+	atstk1000_fb0_data.fbmem_start = fbmem_start;
+	atstk1000_fb0_data.fbmem_size = fbmem_size;
+}
diff --git a/arch/avr32/boards/atstk1000/spi.c b/arch/avr32/boards/atstk1000/spi.c
new file mode 100644
index 000000000000..567726c82c6e
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/spi.c
@@ -0,0 +1,27 @@
+/*
+ * ATSTK1000 SPI devices
+ *
+ * Copyright (C) 2005 Atmel Norway
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/device.h>
+#include <linux/spi/spi.h>
+
+static struct spi_board_info spi_board_info[] __initdata = {
+	{
+		.modalias	= "ltv350qv",
+		.max_speed_hz	= 16000000,
+		.bus_num	= 0,
+		.chip_select	= 1,
+	},
+};
+
+static int board_init_spi(void)
+{
+	spi_register_board_info(spi_board_info, ARRAY_SIZE(spi_board_info));
+	return 0;
+}
+arch_initcall(board_init_spi);
diff --git a/arch/avr32/boot/images/Makefile b/arch/avr32/boot/images/Makefile
new file mode 100644
index 000000000000..ccd74eeecec3
--- /dev/null
+++ b/arch/avr32/boot/images/Makefile
@@ -0,0 +1,62 @@
+#
+# Copyright (C) 2004-2006 Atmel Corporation
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+
+MKIMAGE		:= $(srctree)/scripts/mkuboot.sh
+
+extra-y		:= vmlinux.bin vmlinux.gz
+
+OBJCOPYFLAGS_vmlinux.bin := -O binary
+$(obj)/vmlinux.bin: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+$(obj)/vmlinux.gz: $(obj)/vmlinux.bin FORCE
+	$(call if_changed,gzip)
+
+quiet_cmd_uimage = UIMAGE $@
+      cmd_uimage = $(CONFIG_SHELL) $(MKIMAGE) -A avr32 -O linux -T kernel	\
+		-C gzip -a $(CONFIG_LOAD_ADDRESS) -e $(CONFIG_ENTRY_ADDRESS)	\
+		-n 'Linux-$(KERNELRELEASE)' -d $< $@
+
+targets += uImage uImage.srec
+$(obj)/uImage: $(obj)/vmlinux.gz
+	$(call if_changed,uimage)
+	@echo '  Image $@ is ready'
+
+OBJCOPYFLAGS_uImage.srec := -I binary -O srec
+$(obj)/uImage.srec: $(obj)/uImage
+	$(call if_changed,objcopy)
+
+OBJCOPYFLAGS_vmlinux.elf := --change-section-lma .text-0x80000000 \
+			    --change-section-lma __ex_table-0x80000000 \
+			    --change-section-lma .rodata-0x80000000 \
+			    --change-section-lma .data-0x80000000 \
+			    --change-section-lma .init-0x80000000 \
+			    --change-section-lma .bss-0x80000000 \
+			    --change-section-lma .initrd-0x80000000 \
+			    --change-section-lma __param-0x80000000 \
+			    --change-section-lma __ksymtab-0x80000000 \
+			    --change-section-lma __ksymtab_gpl-0x80000000 \
+			    --change-section-lma __kcrctab-0x80000000 \
+			    --change-section-lma __kcrctab_gpl-0x80000000 \
+			    --change-section-lma __ksymtab_strings-0x80000000 \
+			    --change-section-lma .got-0x80000000 \
+			    --set-start 0xa0000000
+$(obj)/vmlinux.elf: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+quiet_cmd_sfdwarf = SFDWARF $@
+      cmd_sfdwarf = sfdwarf $< TO $@ GNUAVR IW $(SFDWARF_FLAGS) > $(obj)/sfdwarf.log
+
+$(obj)/vmlinux.cso: $(obj)/vmlinux.elf FORCE
+	$(call if_changed,sfdwarf)
+
+install: $(BOOTIMAGE)
+	sh $(srctree)/install-kernel.sh $<
+
+# Generated files to be removed upon make clean
+clean-files	:= vmlinux* uImage uImage.srec
diff --git a/arch/avr32/boot/u-boot/Makefile b/arch/avr32/boot/u-boot/Makefile
new file mode 100644
index 000000000000..125ddc96c275
--- /dev/null
+++ b/arch/avr32/boot/u-boot/Makefile
@@ -0,0 +1,3 @@
+extra-y		:= head.o
+
+obj-y		:= empty.o
diff --git a/arch/avr32/boot/u-boot/empty.S b/arch/avr32/boot/u-boot/empty.S
new file mode 100644
index 000000000000..8ac91a5f12f0
--- /dev/null
+++ b/arch/avr32/boot/u-boot/empty.S
@@ -0,0 +1 @@
+/* Empty file */
diff --git a/arch/avr32/boot/u-boot/head.S b/arch/avr32/boot/u-boot/head.S
new file mode 100644
index 000000000000..4488fa27fe94
--- /dev/null
+++ b/arch/avr32/boot/u-boot/head.S
@@ -0,0 +1,60 @@
+/*
+ * Startup code for use with the u-boot bootloader.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/setup.h>
+
+	/*
+	 * The kernel is loaded where we want it to be and all caches
+	 * have just been flushed. We get two parameters from u-boot:
+	 *
+	 * r12 contains a magic number (ATAG_MAGIC)
+	 * r11 points to a tag table providing information about
+	 *     the system.
+	 */
+	.section .init.text,"ax"
+	.global _start
+_start:
+	/* Check if the boot loader actually provided a tag table */
+	lddpc	r0, magic_number
+	cp.w	r12, r0
+	brne	no_tag_table
+
+	/* Initialize .bss */
+	lddpc	r2, bss_start_addr
+	lddpc   r3, end_addr
+	mov	r0, 0
+	mov	r1, 0
+1:      st.d    r2++, r0
+	cp      r2, r3
+	brlo    1b
+
+	/*
+	 * Save the tag table address for later use. This must be done
+	 * _after_ .bss has been initialized...
+	 */
+	lddpc	r0, tag_table_addr
+	st.w	r0[0], r11
+
+	/* Jump to loader-independent setup code */
+	rjmp	kernel_entry
+
+	.align	2
+magic_number:
+	.long	ATAG_MAGIC
+tag_table_addr:
+	.long	bootloader_tags
+bss_start_addr:
+	.long   __bss_start
+end_addr:
+	.long   _end
+
+no_tag_table:
+	sub	r12, pc, (. - 2f)
+	bral	panic
+2:	.asciz	"Boot loader didn't provide correct magic number\n"
diff --git a/arch/avr32/configs/atstk1002_defconfig b/arch/avr32/configs/atstk1002_defconfig
new file mode 100644
index 000000000000..1d22255009fd
--- /dev/null
+++ b/arch/avr32/configs/atstk1002_defconfig
@@ -0,0 +1,754 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.18-rc1
+# Tue Jul 11 12:41:36 2006
+#
+CONFIG_AVR32=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_HARDIRQS_SW_RESEND=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_SWAP=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_BASE_FULL is not set
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+CONFIG_SHMEM=y
+# CONFIG_SLAB is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=1
+CONFIG_SLOB=y
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+
+#
+# Block layer
+#
+# CONFIG_BLK_DEV_IO_TRACE is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System Type and features
+#
+CONFIG_SUBARCH_AVR32B=y
+CONFIG_MMU=y
+CONFIG_PERFORMANCE_COUNTERS=y
+CONFIG_PLATFORM_AT32AP=y
+CONFIG_CPU_AT32AP7000=y
+CONFIG_BOARD_ATSTK1002=y
+CONFIG_BOARD_ATSTK1000=y
+CONFIG_LOADER_U_BOOT=y
+CONFIG_LOAD_ADDRESS=0x10000000
+CONFIG_ENTRY_ADDRESS=0x90000000
+CONFIG_PHYS_OFFSET=0x10000000
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_HAVE_ARCH_BOOTMEM_NODE is not set
+# CONFIG_ARCH_HAVE_MEMORY_PRESENT is not set
+# CONFIG_NEED_NODE_MEMMAP_SIZE is not set
+CONFIG_ARCH_FLATMEM_ENABLE=y
+# CONFIG_ARCH_DISCONTIGMEM_ENABLE is not set
+# CONFIG_ARCH_SPARSEMEM_ENABLE is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+# CONFIG_OWNERSHIP_TRACE is not set
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+CONFIG_CMDLINE=""
+
+#
+# Bus options
+#
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_NET_TCPPROBE is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+# CONFIG_PREVENT_FIRMWARE_BUILD is not set
+# CONFIG_FW_LOADER is not set
+# CONFIG_DEBUG_DRIVER is not set
+# CONFIG_SYS_HYPERVISOR is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=m
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+CONFIG_BLK_DEV_NBD=m
+CONFIG_BLK_DEV_RAM=m
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+CONFIG_DUMMY=y
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+CONFIG_MACB=y
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+CONFIG_PPP=m
+# CONFIG_PPP_MULTILINK is not set
+# CONFIG_PPP_FILTER is not set
+CONFIG_PPP_ASYNC=m
+# CONFIG_PPP_SYNC_TTY is not set
+CONFIG_PPP_DEFLATE=m
+# CONFIG_PPP_BSDCOMP is not set
+# CONFIG_PPP_MPPE is not set
+# CONFIG_PPPOE is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_AT91=y
+CONFIG_SERIAL_AT91_CONSOLE=y
+# CONFIG_SERIAL_AT91_TTYAT is not set
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+# CONFIG_LEGACY_PTYS is not set
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+CONFIG_SPI=y
+# CONFIG_SPI_DEBUG is not set
+CONFIG_SPI_MASTER=y
+
+#
+# SPI Master Controller Drivers
+#
+CONFIG_SPI_ATMEL=m
+# CONFIG_SPI_BITBANG is not set
+
+#
+# SPI Protocol Masters
+#
+
+#
+# Dallas's 1-wire bus
+#
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FIRMWARE_EDID is not set
+CONFIG_FB=m
+CONFIG_FB_CFB_FILLRECT=m
+CONFIG_FB_CFB_COPYAREA=m
+CONFIG_FB_CFB_IMAGEBLIT=m
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+CONFIG_FB_SIDSA=m
+CONFIG_FB_SIDSA_DEFAULT_BPP=24
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_VIRTUAL is not set
+
+#
+# Logo configuration
+#
+# CONFIG_LOGO is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+# CONFIG_BACKLIGHT_CLASS_DEVICE is not set
+CONFIG_LCD_CLASS_DEVICE=m
+CONFIG_LCD_DEVICE=y
+CONFIG_LCD_LTV350QV=m
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# DMA Engine support
+#
+# CONFIG_DMA_ENGINE is not set
+
+#
+# DMA Clients
+#
+
+#
+# DMA Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+CONFIG_MINIX_FS=m
+CONFIG_ROMFS_FS=m
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+CONFIG_CONFIGFS_FS=m
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+CONFIG_CIFS=m
+# CONFIG_CIFS_STATS is not set
+# CONFIG_CIFS_WEAK_PW_HASH is not set
+# CONFIG_CIFS_XATTR is not set
+# CONFIG_CIFS_DEBUG2 is not set
+# CONFIG_CIFS_EXPERIMENTAL is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=m
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=m
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+CONFIG_NLS_CODEPAGE_850=m
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+CONFIG_NLS_ISO8859_1=m
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+
+#
+# Kernel hacking
+#
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
+CONFIG_PRINTK_TIME=y
+CONFIG_MAGIC_SYSRQ=y
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_KERNEL=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_MUTEXES is not set
+# CONFIG_DEBUG_RWSEMS is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_LOCKING_API_SELFTESTS is not set
+# CONFIG_DEBUG_KOBJECT is not set
+CONFIG_DEBUG_BUGVERBOSE=y
+# CONFIG_DEBUG_INFO is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_DEBUG_VM is not set
+CONFIG_FRAME_POINTER=y
+# CONFIG_UNWIND_INFO is not set
+CONFIG_FORCED_INLINING=y
+# CONFIG_RCU_TORTURE_TEST is not set
+CONFIG_KPROBES=y
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+CONFIG_CRC_CCITT=m
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=m
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=m
+CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/avr32/kernel/Makefile b/arch/avr32/kernel/Makefile
new file mode 100644
index 000000000000..90e5afff54a2
--- /dev/null
+++ b/arch/avr32/kernel/Makefile
@@ -0,0 +1,18 @@
+#
+# Makefile for the Linux/AVR32 kernel.
+#
+
+extra-y				:= head.o vmlinux.lds
+
+obj-$(CONFIG_SUBARCH_AVR32B)	+= entry-avr32b.o
+obj-y				+= syscall_table.o syscall-stubs.o irq.o
+obj-y				+= setup.o traps.o semaphore.o ptrace.o
+obj-y				+= signal.o sys_avr32.o process.o time.o
+obj-y				+= init_task.o switch_to.o cpu.o
+obj-$(CONFIG_MODULES)		+= module.o avr32_ksyms.o
+obj-$(CONFIG_KPROBES)		+= kprobes.o
+
+USE_STANDARD_AS_RULE		:= true
+
+%.lds: %.lds.c FORCE
+	$(call if_changed_dep,cpp_lds_S)
diff --git a/arch/avr32/kernel/asm-offsets.c b/arch/avr32/kernel/asm-offsets.c
new file mode 100644
index 000000000000..97d865865667
--- /dev/null
+++ b/arch/avr32/kernel/asm-offsets.c
@@ -0,0 +1,25 @@
+/*
+ * Generate definitions needed by assembly language modules.
+ * This code generates raw asm output which is post-processed
+ * to extract and format the required data.
+ */
+
+#include <linux/thread_info.h>
+
+#define DEFINE(sym, val) \
+        asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+
+#define BLANK() asm volatile("\n->" : : )
+
+#define OFFSET(sym, str, mem) \
+        DEFINE(sym, offsetof(struct str, mem));
+
+void foo(void)
+{
+	OFFSET(TI_task, thread_info, task);
+	OFFSET(TI_exec_domain, thread_info, exec_domain);
+	OFFSET(TI_flags, thread_info, flags);
+	OFFSET(TI_cpu, thread_info, cpu);
+	OFFSET(TI_preempt_count, thread_info, preempt_count);
+	OFFSET(TI_restart_block, thread_info, restart_block);
+}
diff --git a/arch/avr32/kernel/avr32_ksyms.c b/arch/avr32/kernel/avr32_ksyms.c
new file mode 100644
index 000000000000..04f767a272b7
--- /dev/null
+++ b/arch/avr32/kernel/avr32_ksyms.c
@@ -0,0 +1,55 @@
+/*
+ * Export AVR32-specific functions for loadable modules.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+
+#include <asm/checksum.h>
+#include <asm/uaccess.h>
+#include <asm/delay.h>
+
+/*
+ * GCC functions
+ */
+extern unsigned long long __avr32_lsl64(unsigned long long u, unsigned long b);
+extern unsigned long long __avr32_lsr64(unsigned long long u, unsigned long b);
+extern unsigned long long __avr32_asr64(unsigned long long u, unsigned long b);
+EXPORT_SYMBOL(__avr32_lsl64);
+EXPORT_SYMBOL(__avr32_lsr64);
+EXPORT_SYMBOL(__avr32_asr64);
+
+/*
+ * String functions
+ */
+EXPORT_SYMBOL(memset);
+EXPORT_SYMBOL(memcpy);
+
+/*
+ * Userspace access stuff.
+ */
+EXPORT_SYMBOL(copy_from_user);
+EXPORT_SYMBOL(copy_to_user);
+EXPORT_SYMBOL(__copy_user);
+EXPORT_SYMBOL(strncpy_from_user);
+EXPORT_SYMBOL(__strncpy_from_user);
+EXPORT_SYMBOL(clear_user);
+EXPORT_SYMBOL(__clear_user);
+EXPORT_SYMBOL(csum_partial);
+EXPORT_SYMBOL(csum_partial_copy_generic);
+
+/* Delay loops (lib/delay.S) */
+EXPORT_SYMBOL(__ndelay);
+EXPORT_SYMBOL(__udelay);
+EXPORT_SYMBOL(__const_udelay);
+
+/* Bit operations (lib/findbit.S) */
+EXPORT_SYMBOL(find_first_zero_bit);
+EXPORT_SYMBOL(find_next_zero_bit);
+EXPORT_SYMBOL(find_first_bit);
+EXPORT_SYMBOL(find_next_bit);
+EXPORT_SYMBOL(generic_find_next_zero_le_bit);
diff --git a/arch/avr32/kernel/cpu.c b/arch/avr32/kernel/cpu.c
new file mode 100644
index 000000000000..342452ba2049
--- /dev/null
+++ b/arch/avr32/kernel/cpu.c
@@ -0,0 +1,327 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/sysdev.h>
+#include <linux/seq_file.h>
+#include <linux/cpu.h>
+#include <linux/percpu.h>
+#include <linux/param.h>
+#include <linux/errno.h>
+
+#include <asm/setup.h>
+#include <asm/sysreg.h>
+
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+#ifdef CONFIG_PERFORMANCE_COUNTERS
+
+/*
+ * XXX: If/when a SMP-capable implementation of AVR32 will ever be
+ * made, we must make sure that the code executes on the correct CPU.
+ */
+static ssize_t show_pc0event(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "0x%lx\n", (pccr >> 12) & 0x3f);
+}
+static ssize_t store_pc0event(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf || val > 0x3f)
+		return -EINVAL;
+	val = (val << 12) | (sysreg_read(PCCR) & 0xfffc0fff);
+	sysreg_write(PCCR, val);
+	return count;
+}
+static ssize_t show_pc0count(struct sys_device *dev, char *buf)
+{
+	unsigned long pcnt0;
+
+	pcnt0 = sysreg_read(PCNT0);
+	return sprintf(buf, "%lu\n", pcnt0);
+}
+static ssize_t store_pc0count(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCNT0, val);
+
+	return count;
+}
+
+static ssize_t show_pc1event(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "0x%lx\n", (pccr >> 18) & 0x3f);
+}
+static ssize_t store_pc1event(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf || val > 0x3f)
+		return -EINVAL;
+	val = (val << 18) | (sysreg_read(PCCR) & 0xff03ffff);
+	sysreg_write(PCCR, val);
+	return count;
+}
+static ssize_t show_pc1count(struct sys_device *dev, char *buf)
+{
+	unsigned long pcnt1;
+
+	pcnt1 = sysreg_read(PCNT1);
+	return sprintf(buf, "%lu\n", pcnt1);
+}
+static ssize_t store_pc1count(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCNT1, val);
+
+	return count;
+}
+
+static ssize_t show_pccycles(struct sys_device *dev, char *buf)
+{
+	unsigned long pccnt;
+
+	pccnt = sysreg_read(PCCNT);
+	return sprintf(buf, "%lu\n", pccnt);
+}
+static ssize_t store_pccycles(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	sysreg_write(PCCNT, val);
+
+	return count;
+}
+
+static ssize_t show_pcenable(struct sys_device *dev, char *buf)
+{
+	unsigned long pccr;
+
+	pccr = sysreg_read(PCCR);
+	return sprintf(buf, "%c\n", (pccr & 1)?'1':'0');
+}
+static ssize_t store_pcenable(struct sys_device *dev, const char *buf,
+			      size_t count)
+{
+	unsigned long pccr, val;
+	char *endp;
+
+	val = simple_strtoul(buf, &endp, 0);
+	if (endp == buf)
+		return -EINVAL;
+	if (val)
+		val = 1;
+
+	pccr = sysreg_read(PCCR);
+	pccr = (pccr & ~1UL) | val;
+	sysreg_write(PCCR, pccr);
+
+	return count;
+}
+
+static SYSDEV_ATTR(pc0event, 0600, show_pc0event, store_pc0event);
+static SYSDEV_ATTR(pc0count, 0600, show_pc0count, store_pc0count);
+static SYSDEV_ATTR(pc1event, 0600, show_pc1event, store_pc1event);
+static SYSDEV_ATTR(pc1count, 0600, show_pc1count, store_pc1count);
+static SYSDEV_ATTR(pccycles, 0600, show_pccycles, store_pccycles);
+static SYSDEV_ATTR(pcenable, 0600, show_pcenable, store_pcenable);
+
+#endif /* CONFIG_PERFORMANCE_COUNTERS */
+
+static int __init topology_init(void)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu) {
+		struct cpu *c = &per_cpu(cpu_devices, cpu);
+
+		register_cpu(c, cpu);
+
+#ifdef CONFIG_PERFORMANCE_COUNTERS
+		sysdev_create_file(&c->sysdev, &attr_pc0event);
+		sysdev_create_file(&c->sysdev, &attr_pc0count);
+		sysdev_create_file(&c->sysdev, &attr_pc1event);
+		sysdev_create_file(&c->sysdev, &attr_pc1count);
+		sysdev_create_file(&c->sysdev, &attr_pccycles);
+		sysdev_create_file(&c->sysdev, &attr_pcenable);
+#endif
+	}
+
+	return 0;
+}
+
+subsys_initcall(topology_init);
+
+static const char *cpu_names[] = {
+	"Morgan",
+	"AP7000",
+};
+#define NR_CPU_NAMES ARRAY_SIZE(cpu_names)
+
+static const char *arch_names[] = {
+	"AVR32A",
+	"AVR32B",
+};
+#define NR_ARCH_NAMES ARRAY_SIZE(arch_names)
+
+static const char *mmu_types[] = {
+	"No MMU",
+	"ITLB and DTLB",
+	"Shared TLB",
+	"MPU"
+};
+
+void __init setup_processor(void)
+{
+	unsigned long config0, config1;
+	unsigned cpu_id, cpu_rev, arch_id, arch_rev, mmu_type;
+	unsigned tmp;
+
+	config0 = sysreg_read(CONFIG0); /* 0x0000013e; */
+	config1 = sysreg_read(CONFIG1); /* 0x01f689a2; */
+	cpu_id = config0 >> 24;
+	cpu_rev = (config0 >> 16) & 0xff;
+	arch_id = (config0 >> 13) & 0x07;
+	arch_rev = (config0 >> 10) & 0x07;
+	mmu_type = (config0 >> 7) & 0x03;
+
+	boot_cpu_data.arch_type = arch_id;
+	boot_cpu_data.cpu_type = cpu_id;
+	boot_cpu_data.arch_revision = arch_rev;
+	boot_cpu_data.cpu_revision = cpu_rev;
+	boot_cpu_data.tlb_config = mmu_type;
+
+	tmp = (config1 >> 13) & 0x07;
+	if (tmp) {
+		boot_cpu_data.icache.ways = 1 << ((config1 >> 10) & 0x07);
+		boot_cpu_data.icache.sets = 1 << ((config1 >> 16) & 0x0f);
+		boot_cpu_data.icache.linesz = 1 << (tmp + 1);
+	}
+	tmp = (config1 >> 3) & 0x07;
+	if (tmp) {
+		boot_cpu_data.dcache.ways = 1 << (config1 & 0x07);
+		boot_cpu_data.dcache.sets = 1 << ((config1 >> 6) & 0x0f);
+		boot_cpu_data.dcache.linesz = 1 << (tmp + 1);
+	}
+
+	if ((cpu_id >= NR_CPU_NAMES) || (arch_id >= NR_ARCH_NAMES)) {
+		printk ("Unknown CPU configuration (ID %02x, arch %02x), "
+			"continuing anyway...\n",
+			cpu_id, arch_id);
+		return;
+	}
+
+	printk ("CPU: %s [%02x] revision %d (%s revision %d)\n",
+		cpu_names[cpu_id], cpu_id, cpu_rev,
+		arch_names[arch_id], arch_rev);
+	printk ("CPU: MMU configuration: %s\n", mmu_types[mmu_type]);
+	printk ("CPU: features:");
+	if (config0 & (1 << 6))
+		printk(" fpu");
+	if (config0 & (1 << 5))
+		printk(" java");
+	if (config0 & (1 << 4))
+		printk(" perfctr");
+	if (config0 & (1 << 3))
+		printk(" ocd");
+	printk("\n");
+}
+
+#ifdef CONFIG_PROC_FS
+static int c_show(struct seq_file *m, void *v)
+{
+	unsigned int icache_size, dcache_size;
+	unsigned int cpu = smp_processor_id();
+
+	icache_size = boot_cpu_data.icache.ways *
+		boot_cpu_data.icache.sets *
+		boot_cpu_data.icache.linesz;
+	dcache_size = boot_cpu_data.dcache.ways *
+		boot_cpu_data.dcache.sets *
+		boot_cpu_data.dcache.linesz;
+
+	seq_printf(m, "processor\t: %d\n", cpu);
+
+	if (boot_cpu_data.arch_type < NR_ARCH_NAMES)
+		seq_printf(m, "cpu family\t: %s revision %d\n",
+			   arch_names[boot_cpu_data.arch_type],
+			   boot_cpu_data.arch_revision);
+	if (boot_cpu_data.cpu_type < NR_CPU_NAMES)
+		seq_printf(m, "cpu type\t: %s revision %d\n",
+			   cpu_names[boot_cpu_data.cpu_type],
+			   boot_cpu_data.cpu_revision);
+
+	seq_printf(m, "i-cache\t\t: %dK (%u ways x %u sets x %u)\n",
+		   icache_size >> 10,
+		   boot_cpu_data.icache.ways,
+		   boot_cpu_data.icache.sets,
+		   boot_cpu_data.icache.linesz);
+	seq_printf(m, "d-cache\t\t: %dK (%u ways x %u sets x %u)\n",
+		   dcache_size >> 10,
+		   boot_cpu_data.dcache.ways,
+		   boot_cpu_data.dcache.sets,
+		   boot_cpu_data.dcache.linesz);
+	seq_printf(m, "bogomips\t: %lu.%02lu\n",
+		   boot_cpu_data.loops_per_jiffy / (500000/HZ),
+		   (boot_cpu_data.loops_per_jiffy / (5000/HZ)) % 100);
+
+	return 0;
+}
+
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+	return *pos < 1 ? (void *)1 : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	++*pos;
+	return NULL;
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+
+}
+
+struct seq_operations cpuinfo_op = {
+	.start	= c_start,
+	.next	= c_next,
+	.stop	= c_stop,
+	.show	= c_show
+};
+#endif /* CONFIG_PROC_FS */
diff --git a/arch/avr32/kernel/entry-avr32b.S b/arch/avr32/kernel/entry-avr32b.S
new file mode 100644
index 000000000000..eeb66792bc37
--- /dev/null
+++ b/arch/avr32/kernel/entry-avr32b.S
@@ -0,0 +1,678 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * This file contains the low-level entry-points into the kernel, that is,
+ * exception handlers, debug trap handlers, interrupt handlers and the
+ * system call handler.
+ */
+#include <linux/errno.h>
+
+#include <asm/asm.h>
+#include <asm/hardirq.h>
+#include <asm/irq.h>
+#include <asm/ocd.h>
+#include <asm/page.h>
+#include <asm/pgtable.h>
+#include <asm/ptrace.h>
+#include <asm/sysreg.h>
+#include <asm/thread_info.h>
+#include <asm/unistd.h>
+
+#ifdef CONFIG_PREEMPT
+# define preempt_stop		mask_interrupts
+#else
+# define preempt_stop
+# define fault_resume_kernel	fault_restore_all
+#endif
+
+#define __MASK(x)	((1 << (x)) - 1)
+#define IRQ_MASK	((__MASK(SOFTIRQ_BITS) << SOFTIRQ_SHIFT) | \
+			 (__MASK(HARDIRQ_BITS) << HARDIRQ_SHIFT))
+
+	.section .ex.text,"ax",@progbits
+	.align	2
+exception_vectors:
+	bral	handle_critical
+	.align	2
+	bral	handle_critical
+	.align	2
+	bral	do_bus_error_write
+	.align	2
+	bral	do_bus_error_read
+	.align	2
+	bral	do_nmi_ll
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	handle_debug
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	do_fpe_ll
+	.align	2
+	bral	do_illegal_opcode_ll
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_address_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	handle_protection_fault
+	.align	2
+	bral	do_dtlb_modified
+
+	/*
+	 * r0 :	PGD/PT/PTE
+	 * r1 : Offending address
+	 * r2 : Scratch register
+	 * r3 : Cause (5, 12 or 13)
+	 */
+#define	tlbmiss_save	pushm	r0-r3
+#define tlbmiss_restore	popm	r0-r3
+
+	.section .tlbx.ex.text,"ax",@progbits
+	.global	itlb_miss
+itlb_miss:
+	tlbmiss_save
+	rjmp	tlb_miss_common
+
+	.section .tlbr.ex.text,"ax",@progbits
+dtlb_miss_read:
+	tlbmiss_save
+	rjmp	tlb_miss_common
+
+	.section .tlbw.ex.text,"ax",@progbits
+dtlb_miss_write:
+	tlbmiss_save
+
+	.global	tlb_miss_common
+tlb_miss_common:
+	mfsr	r0, SYSREG_PTBR
+	mfsr	r1, SYSREG_TLBEAR
+
+	/* Is it the vmalloc space? */
+	bld	r1, 31
+	brcs	handle_vmalloc_miss
+
+	/* First level lookup */
+pgtbl_lookup:
+	lsr	r2, r1, PGDIR_SHIFT
+	ld.w	r0, r0[r2 << 2]
+	bld	r0, _PAGE_BIT_PRESENT
+	brcc	page_table_not_present
+
+	/* TODO: Check access rights on page table if necessary */
+
+	/* Translate to virtual address in P1. */
+	andl	r0, 0xf000
+	sbr	r0, 31
+
+	/* Second level lookup */
+	lsl	r1, (32 - PGDIR_SHIFT)
+	lsr	r1, (32 - PGDIR_SHIFT) + PAGE_SHIFT
+	add	r2, r0, r1 << 2
+	ld.w	r1, r2[0]
+	bld	r1, _PAGE_BIT_PRESENT
+	brcc	page_not_present
+
+	/* Mark the page as accessed */
+	sbr	r1, _PAGE_BIT_ACCESSED
+	st.w	r2[0], r1
+
+	/* Drop software flags */
+	andl	r1, _PAGE_FLAGS_HARDWARE_MASK & 0xffff
+	mtsr	SYSREG_TLBELO, r1
+
+	/* Figure out which entry we want to replace */
+	mfsr	r0, SYSREG_TLBARLO
+	clz	r2, r0
+	brcc	1f
+	mov	r1, -1			/* All entries have been accessed, */
+	mtsr	SYSREG_TLBARLO, r1	/* so reset TLBAR */
+	mov	r2, 0			/* and start at 0 */
+1:	mfsr	r1, SYSREG_MMUCR
+	lsl	r2, 14
+	andl	r1, 0x3fff, COH
+	or	r1, r2
+	mtsr	SYSREG_MMUCR, r1
+
+	tlbw
+
+	tlbmiss_restore
+	rete
+
+handle_vmalloc_miss:
+	/* Simply do the lookup in init's page table */
+	mov	r0, lo(swapper_pg_dir)
+	orh	r0, hi(swapper_pg_dir)
+	rjmp	pgtbl_lookup
+
+
+	/* ---                    System Call                    --- */
+
+	.section .scall.text,"ax",@progbits
+system_call:
+	pushm	r12		/* r12_orig */
+	stmts	--sp, r0-lr
+	zero_fp
+	mfsr	r0, SYSREG_RAR_SUP
+	mfsr	r1, SYSREG_RSR_SUP
+	stm	--sp, r0-r1
+
+	/* check for syscall tracing */
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_SYSCALL_TRACE
+	brcs	syscall_trace_enter
+
+syscall_trace_cont:
+	cp.w	r8, NR_syscalls
+	brhs	syscall_badsys
+
+	lddpc   lr, syscall_table_addr
+	ld.w    lr, lr[r8 << 2]
+	mov	r8, r5		/* 5th argument (6th is pushed by stub) */
+	icall   lr
+
+	.global	syscall_return
+syscall_return:
+	get_thread_info r0
+	mask_interrupts		/* make sure we don't miss an interrupt
+				   setting need_resched or sigpending
+				   between sampling and the rets */
+
+	/* Store the return value so that the correct value is loaded below */
+	stdsp   sp[REG_R12], r12
+
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_ALLWORK_MASK, COH
+	brne	syscall_exit_work
+
+syscall_exit_cont:
+	popm	r8-r9
+	mtsr	SYSREG_RAR_SUP, r8
+	mtsr	SYSREG_RSR_SUP, r9
+	ldmts	sp++, r0-lr
+	sub	sp, -4		/* r12_orig */
+	rets
+
+	.align	2
+syscall_table_addr:
+	.long   sys_call_table
+
+syscall_badsys:
+	mov	r12, -ENOSYS
+	rjmp	syscall_return
+
+	.global ret_from_fork
+ret_from_fork:
+	rcall   schedule_tail
+
+	/* check for syscall tracing */
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_ALLWORK_MASK, COH
+	brne	syscall_exit_work
+	rjmp    syscall_exit_cont
+
+syscall_trace_enter:
+	pushm	r8-r12
+	rcall	syscall_trace
+	popm	r8-r12
+	rjmp	syscall_trace_cont
+
+syscall_exit_work:
+	bld	r1, TIF_SYSCALL_TRACE
+	brcc	1f
+	unmask_interrupts
+	rcall	syscall_trace
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+
+1:	bld	r1, TIF_NEED_RESCHED
+	brcc	2f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+2:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	3f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+3:	bld	r1, TIF_BREAKPOINT
+	brcc	syscall_exit_cont
+	mfsr	r3, SYSREG_TLBEHI
+	lddsp	r2, sp[REG_PC]
+	andl	r3, 0xff, COH
+	lsl	r3, 1
+	sbr	r3, 30
+	sbr	r3, 0
+	mtdr	DBGREG_BWA2A, r2
+	mtdr	DBGREG_BWC2A, r3
+	rjmp	syscall_exit_cont
+
+
+	/* The slow path of the TLB miss handler */
+page_table_not_present:
+page_not_present:
+	tlbmiss_restore
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_page_fault
+	rjmp	ret_from_exception
+
+	/* This function expects to find offending PC in SYSREG_RAR_EX */
+save_full_context_ex:
+	mfsr	r8, SYSREG_RSR_EX
+	mov	r12, r8
+	andh	r8, (MODE_MASK >> 16), COH
+	mfsr	r11, SYSREG_RAR_EX
+	brne	2f
+
+1:	pushm	r11, r12	/* PC and SR */
+	unmask_exceptions
+	ret	r12
+
+2:	sub	r10, sp, -(FRAME_SIZE_FULL - REG_LR)
+	stdsp	sp[4], r10	/* replace saved SP */
+	rjmp	1b
+
+	/* Low-level exception handlers */
+handle_critical:
+	pushm	r12
+	pushm	r0-r12
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_critical_exception
+
+	/* We should never get here... */
+bad_return:
+	sub	r12, pc, (. - 1f)
+	bral	panic
+	.align	2
+1:	.asciz	"Return from critical exception!"
+
+	.align	1
+do_bus_error_write:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mov	r11, 1
+	rjmp	1f
+
+do_bus_error_read:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mov	r11, 0
+1:	mfsr	r12, SYSREG_BEAR
+	mov	r10, sp
+	rcall	do_bus_error
+	rjmp	ret_from_exception
+
+	.align	1
+do_nmi_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	/* FIXME: Make sure RAR_NMI and RSR_NMI are pushed instead of *_EX */
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_nmi
+	rjmp	bad_return
+
+handle_address_fault:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_address_exception
+	rjmp	ret_from_exception
+
+handle_protection_fault:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_page_fault
+	rjmp	ret_from_exception
+
+	.align	1
+do_illegal_opcode_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	mfsr	r12, SYSREG_ECR
+	mov	r11, sp
+	rcall	do_illegal_opcode
+	rjmp	ret_from_exception
+
+do_dtlb_modified:
+	pushm	r0-r3
+	mfsr	r1, SYSREG_TLBEAR
+	mfsr	r0, SYSREG_PTBR
+	lsr	r2, r1, PGDIR_SHIFT
+	ld.w	r0, r0[r2 << 2]
+	lsl	r1, (32 - PGDIR_SHIFT)
+	lsr	r1, (32 - PGDIR_SHIFT) + PAGE_SHIFT
+
+	/* Translate to virtual address in P1 */
+	andl	r0, 0xf000
+	sbr	r0, 31
+	add	r2, r0, r1 << 2
+	ld.w	r3, r2[0]
+	sbr	r3, _PAGE_BIT_DIRTY
+	mov	r0, r3
+	st.w	r2[0], r3
+
+	/* The page table is up-to-date. Update the TLB entry as well */
+	andl	r0, lo(_PAGE_FLAGS_HARDWARE_MASK)
+	mtsr	SYSREG_TLBELO, r0
+
+	/* MMUCR[DRP] is updated automatically, so let's go... */
+	tlbw
+
+	popm	r0-r3
+	rete
+
+do_fpe_ll:
+	sub	sp, 4
+	stmts	--sp, r0-lr
+	rcall	save_full_context_ex
+	unmask_interrupts
+	mov	r12, 26
+	mov	r11, sp
+	rcall	do_fpe
+	rjmp	ret_from_exception
+
+ret_from_exception:
+	mask_interrupts
+	lddsp	r4, sp[REG_SR]
+	andh	r4, (MODE_MASK >> 16), COH
+	brne	fault_resume_kernel
+
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_WORK_MASK, COH
+	brne	fault_exit_work
+
+fault_resume_user:
+	popm	r8-r9
+	mask_exceptions
+	mtsr	SYSREG_RAR_EX, r8
+	mtsr	SYSREG_RSR_EX, r9
+	ldmts	sp++, r0-lr
+	sub	sp, -4
+	rete
+
+fault_resume_kernel:
+#ifdef CONFIG_PREEMPT
+	get_thread_info	r0
+	ld.w	r2, r0[TI_preempt_count]
+	cp.w	r2, 0
+	brne	1f
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1f
+	lddsp	r4, sp[REG_SR]
+	bld	r4, SYSREG_GM_OFFSET
+	brcs	1f
+	rcall	preempt_schedule_irq
+1:
+#endif
+
+	popm	r8-r9
+	mask_exceptions
+	mfsr	r1, SYSREG_SR
+	mtsr	SYSREG_RAR_EX, r8
+	mtsr	SYSREG_RSR_EX, r9
+	popm	lr
+	sub	sp, -4		/* ignore SP */
+	popm	r0-r12
+	sub	sp, -4		/* ignore r12_orig */
+	rete
+
+irq_exit_work:
+	/* Switch to exception mode so that we can share the same code. */
+	mfsr	r8, SYSREG_SR
+	cbr	r8, SYSREG_M0_OFFSET
+	orh	r8, hi(SYSREG_BIT(M1) | SYSREG_BIT(M2))
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	get_thread_info r0
+	ld.w	r1, r0[TI_flags]
+
+fault_exit_work:
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	fault_exit_work
+
+1:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	2f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	fault_exit_work
+
+2:	bld	r1, TIF_BREAKPOINT
+	brcc	fault_resume_user
+	mfsr	r3, SYSREG_TLBEHI
+	lddsp	r2, sp[REG_PC]
+	andl	r3, 0xff, COH
+	lsl	r3, 1
+	sbr	r3, 30
+	sbr	r3, 0
+	mtdr	DBGREG_BWA2A, r2
+	mtdr	DBGREG_BWC2A, r3
+	rjmp	fault_resume_user
+
+	/* If we get a debug trap from privileged context we end up here */
+handle_debug_priv:
+	/* Fix up LR and SP in regs. r11 contains the mode we came from */
+	mfsr	r8, SYSREG_SR
+	mov	r9, r8
+	andh	r8, hi(~MODE_MASK)
+	or	r8, r11
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	stdsp	sp[REG_LR], lr
+	mtsr	SYSREG_SR, r9
+	sub	pc, -2
+	sub	r10, sp, -FRAME_SIZE_FULL
+	stdsp	sp[REG_SP], r10
+	mov	r12, sp
+	rcall	do_debug_priv
+
+	/* Now, put everything back */
+	ssrf	SR_EM_BIT
+	popm	r10, r11
+	mtsr	SYSREG_RAR_DBG, r10
+	mtsr	SYSREG_RSR_DBG, r11
+	mfsr	r8, SYSREG_SR
+	mov	r9, r8
+	andh	r8, hi(~MODE_MASK)
+	andh	r11, hi(MODE_MASK)
+	or	r8, r11
+	mtsr	SYSREG_SR, r8
+	sub	pc, -2
+	popm	lr
+	mtsr	SYSREG_SR, r9
+	sub	pc, -2
+	sub	sp, -4		/* skip SP */
+	popm	r0-r12
+	sub	sp, -4
+	retd
+
+	/*
+	 * At this point, everything is masked, that is, interrupts,
+	 * exceptions and debugging traps. We might get called from
+	 * interrupt or exception context in some rare cases, but this
+	 * will be taken care of by do_debug(), so we're not going to
+	 * do a 100% correct context save here.
+	 */
+handle_debug:
+	sub	sp, 4		/* r12_orig */
+	stmts	--sp, r0-lr
+	mfsr	r10, SYSREG_RAR_DBG
+	mfsr	r11, SYSREG_RSR_DBG
+	unmask_exceptions
+	pushm	r10,r11
+	andh	r11, (MODE_MASK >> 16), COH
+	brne	handle_debug_priv
+
+	mov	r12, sp
+	rcall	do_debug
+
+	lddsp	r10, sp[REG_SR]
+	andh	r10, (MODE_MASK >> 16), COH
+	breq	debug_resume_user
+
+debug_restore_all:
+	popm	r10,r11
+	mask_exceptions
+	mtsr	SYSREG_RSR_DBG, r11
+	mtsr	SYSREG_RAR_DBG, r10
+	ldmts	sp++, r0-lr
+	sub	sp, -4
+	retd
+
+debug_resume_user:
+	get_thread_info r0
+	mask_interrupts
+
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_DBGWORK_MASK, COH
+	breq	debug_restore_all
+
+1:	bld	r1, TIF_NEED_RESCHED
+	brcc	2f
+	unmask_interrupts
+	rcall	schedule
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+2:	mov	r2, _TIF_SIGPENDING | _TIF_RESTORE_SIGMASK
+	tst	r1, r2
+	breq	3f
+	unmask_interrupts
+	mov	r12, sp
+	mov	r11, r0
+	rcall	do_notify_resume
+	mask_interrupts
+	ld.w	r1, r0[TI_flags]
+	rjmp	1b
+
+3:	bld	r1, TIF_SINGLE_STEP
+	brcc	debug_restore_all
+	mfdr	r2, DBGREG_DC
+	sbr	r2, DC_SS_BIT
+	mtdr	DBGREG_DC, r2
+	rjmp	debug_restore_all
+
+	.set	rsr_int0,	SYSREG_RSR_INT0
+	.set	rsr_int1,	SYSREG_RSR_INT1
+	.set	rsr_int2,	SYSREG_RSR_INT2
+	.set	rsr_int3,	SYSREG_RSR_INT3
+	.set	rar_int0,	SYSREG_RAR_INT0
+	.set	rar_int1,	SYSREG_RAR_INT1
+	.set	rar_int2,	SYSREG_RAR_INT2
+	.set	rar_int3,	SYSREG_RAR_INT3
+
+	.macro	IRQ_LEVEL level
+	.type	irq_level\level, @function
+irq_level\level:
+	sub	sp, 4		/* r12_orig */
+	stmts	--sp,r0-lr
+	mfsr	r8, rar_int\level
+	mfsr	r9, rsr_int\level
+	pushm	r8-r9
+
+	mov	r11, sp
+	mov	r12, \level
+
+	rcall	do_IRQ
+
+	lddsp	r4, sp[REG_SR]
+	andh	r4, (MODE_MASK >> 16), COH
+#ifdef CONFIG_PREEMPT
+	brne	2f
+#else
+	brne	1f
+#endif
+
+	get_thread_info	r0
+	ld.w	r1, r0[TI_flags]
+	andl	r1, _TIF_WORK_MASK, COH
+	brne	irq_exit_work
+
+1:	popm	r8-r9
+	mtsr	rar_int\level, r8
+	mtsr	rsr_int\level, r9
+	ldmts	sp++,r0-lr
+	sub	sp, -4		/* ignore r12_orig */
+	rete
+
+#ifdef CONFIG_PREEMPT
+2:
+	get_thread_info	r0
+	ld.w	r2, r0[TI_preempt_count]
+	cp.w	r2, 0
+	brne	1b
+	ld.w	r1, r0[TI_flags]
+	bld	r1, TIF_NEED_RESCHED
+	brcc	1b
+	lddsp	r4, sp[REG_SR]
+	bld	r4, SYSREG_GM_OFFSET
+	brcs	1b
+	rcall	preempt_schedule_irq
+	rjmp	1b
+#endif
+	.endm
+
+	.section .irq.text,"ax",@progbits
+
+	.global	irq_level0
+	.global	irq_level1
+	.global	irq_level2
+	.global	irq_level3
+	IRQ_LEVEL 0
+	IRQ_LEVEL 1
+	IRQ_LEVEL 2
+	IRQ_LEVEL 3
diff --git a/arch/avr32/kernel/head.S b/arch/avr32/kernel/head.S
new file mode 100644
index 000000000000..773b7ad87be9
--- /dev/null
+++ b/arch/avr32/kernel/head.S
@@ -0,0 +1,45 @@
+/*
+ * Non-board-specific low-level startup code
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/sysreg.h>
+
+	.section .init.text,"ax"
+	.global kernel_entry
+kernel_entry:
+	/* Initialize status register */
+	lddpc   r0, init_sr
+	mtsr	SYSREG_SR, r0
+
+	/* Set initial stack pointer */
+	lddpc   sp, stack_addr
+	sub	sp, -THREAD_SIZE
+
+#ifdef CONFIG_FRAME_POINTER
+	/* Mark last stack frame */
+	mov	lr, 0
+	mov	r7, 0
+#endif
+
+	/* Set up the PIO, SDRAM controller, early printk, etc. */
+	rcall	board_early_init
+
+	/* Start the show */
+	lddpc   pc, kernel_start_addr
+
+	.align  2
+init_sr:
+	.long   0x007f0000	/* Supervisor mode, everything masked */
+stack_addr:
+	.long   init_thread_union
+kernel_start_addr:
+	.long   start_kernel
diff --git a/arch/avr32/kernel/init_task.c b/arch/avr32/kernel/init_task.c
new file mode 100644
index 000000000000..effcacf9d1a2
--- /dev/null
+++ b/arch/avr32/kernel/init_task.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/sched.h>
+#include <linux/init_task.h>
+#include <linux/mqueue.h>
+
+#include <asm/pgtable.h>
+
+static struct fs_struct init_fs = INIT_FS;
+static struct files_struct init_files = INIT_FILES;
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+struct mm_struct init_mm = INIT_MM(init_mm);
+
+EXPORT_SYMBOL(init_mm);
+
+/*
+ * Initial thread structure. Must be aligned on an 8192-byte boundary.
+ */
+union thread_union init_thread_union
+	__attribute__((__section__(".data.init_task"))) =
+		{ INIT_THREAD_INFO(init_task) };
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+struct task_struct init_task = INIT_TASK(init_task);
+
+EXPORT_SYMBOL(init_task);
diff --git a/arch/avr32/kernel/irq.c b/arch/avr32/kernel/irq.c
new file mode 100644
index 000000000000..856f3548e664
--- /dev/null
+++ b/arch/avr32/kernel/irq.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on arch/i386/kernel/irq.c
+ *   Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This file contains the code used by various IRQ handling routines:
+ * asking for different IRQ's should be done through these routines
+ * instead of just grabbing them. Thus setups with different IRQ numbers
+ * shouldn't result in any weird surprises, and installing new handlers
+ * should be easier.
+ *
+ * IRQ's are in fact implemented a bit like signal handlers for the kernel.
+ * Naturally it's not a 1:1 relation, but there are similarities.
+ */
+
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/sysdev.h>
+
+/*
+ * 'what should we do if we get a hw irq event on an illegal vector'.
+ * each architecture has to answer this themselves.
+ */
+void ack_bad_irq(unsigned int irq)
+{
+	printk("unexpected IRQ %u\n", irq);
+}
+
+#ifdef CONFIG_PROC_FS
+int show_interrupts(struct seq_file *p, void *v)
+{
+	int i = *(loff_t *)v, cpu;
+	struct irqaction *action;
+	unsigned long flags;
+
+	if (i == 0) {
+		seq_puts(p, "           ");
+		for_each_online_cpu(cpu)
+			seq_printf(p, "CPU%d       ", cpu);
+		seq_putc(p, '\n');
+	}
+
+	if (i < NR_IRQS) {
+		spin_lock_irqsave(&irq_desc[i].lock, flags);
+		action = irq_desc[i].action;
+		if (!action)
+			goto unlock;
+
+		seq_printf(p, "%3d: ", i);
+		for_each_online_cpu(cpu)
+			seq_printf(p, "%10u ", kstat_cpu(cpu).irqs[i]);
+		seq_printf(p, "  %s", action->name);
+		for (action = action->next; action; action = action->next)
+			seq_printf(p, ", %s", action->name);
+
+		seq_putc(p, '\n');
+	unlock:
+		spin_unlock_irqrestore(&irq_desc[i].lock, flags);
+	}
+
+	return 0;
+}
+#endif
diff --git a/arch/avr32/kernel/kprobes.c b/arch/avr32/kernel/kprobes.c
new file mode 100644
index 000000000000..6caf9e8d8080
--- /dev/null
+++ b/arch/avr32/kernel/kprobes.c
@@ -0,0 +1,270 @@
+/*
+ *  Kernel Probes (KProbes)
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * Based on arch/ppc64/kernel/kprobes.c
+ *  Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kprobes.h>
+#include <linux/ptrace.h>
+
+#include <asm/cacheflush.h>
+#include <asm/kdebug.h>
+#include <asm/ocd.h>
+
+DEFINE_PER_CPU(struct kprobe *, current_kprobe);
+static unsigned long kprobe_status;
+static struct pt_regs jprobe_saved_regs;
+
+int __kprobes arch_prepare_kprobe(struct kprobe *p)
+{
+	int ret = 0;
+
+	if ((unsigned long)p->addr & 0x01) {
+		printk("Attempt to register kprobe at an unaligned address\n");
+		ret = -EINVAL;
+	}
+
+	/* XXX: Might be a good idea to check if p->addr is a valid
+	 * kernel address as well... */
+
+	if (!ret) {
+		pr_debug("copy kprobe at %p\n", p->addr);
+		memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
+		p->opcode = *p->addr;
+	}
+
+	return ret;
+}
+
+void __kprobes arch_arm_kprobe(struct kprobe *p)
+{
+	pr_debug("arming kprobe at %p\n", p->addr);
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+void __kprobes arch_disarm_kprobe(struct kprobe *p)
+{
+	pr_debug("disarming kprobe at %p\n", p->addr);
+	*p->addr = p->opcode;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long dc;
+
+	pr_debug("preparing to singlestep over %p (PC=%08lx)\n",
+		 p->addr, regs->pc);
+
+	BUG_ON(!(sysreg_read(SR) & SYSREG_BIT(SR_D)));
+
+	dc = __mfdr(DBGREG_DC);
+	dc |= DC_SS;
+	__mtdr(DBGREG_DC, dc);
+
+	/*
+	 * We must run the instruction from its original location
+	 * since it may actually reference PC.
+	 *
+	 * TODO: Do the instruction replacement directly in icache.
+	 */
+	*p->addr = p->opcode;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
+{
+	unsigned long dc;
+
+	pr_debug("resuming execution at PC=%08lx\n", regs->pc);
+
+	dc = __mfdr(DBGREG_DC);
+	dc &= ~DC_SS;
+	__mtdr(DBGREG_DC, dc);
+
+	*p->addr = BREAKPOINT_INSTRUCTION;
+	flush_icache_range((unsigned long)p->addr,
+			   (unsigned long)p->addr + sizeof(kprobe_opcode_t));
+}
+
+static void __kprobes set_current_kprobe(struct kprobe *p)
+{
+	__get_cpu_var(current_kprobe) = p;
+}
+
+static int __kprobes kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *p;
+	void *addr = (void *)regs->pc;
+	int ret = 0;
+
+	pr_debug("kprobe_handler: kprobe_running=%d\n",
+		 kprobe_running());
+
+	/*
+	 * We don't want to be preempted for the entire
+	 * duration of kprobe processing
+	 */
+	preempt_disable();
+
+	/* Check that we're not recursing */
+	if (kprobe_running()) {
+		p = get_kprobe(addr);
+		if (p) {
+			if (kprobe_status == KPROBE_HIT_SS) {
+				printk("FIXME: kprobe hit while single-stepping!\n");
+				goto no_kprobe;
+			}
+
+			printk("FIXME: kprobe hit while handling another kprobe\n");
+			goto no_kprobe;
+		} else {
+			p = kprobe_running();
+			if (p->break_handler && p->break_handler(p, regs))
+				goto ss_probe;
+		}
+		/* If it's not ours, can't be delete race, (we hold lock). */
+		goto no_kprobe;
+	}
+
+	p = get_kprobe(addr);
+	if (!p)
+		goto no_kprobe;
+
+	kprobe_status = KPROBE_HIT_ACTIVE;
+	set_current_kprobe(p);
+	if (p->pre_handler && p->pre_handler(p, regs))
+		/* handler has already set things up, so skip ss setup */
+		return 1;
+
+ss_probe:
+	prepare_singlestep(p, regs);
+	kprobe_status = KPROBE_HIT_SS;
+	return 1;
+
+no_kprobe:
+	return ret;
+}
+
+static int __kprobes post_kprobe_handler(struct pt_regs *regs)
+{
+	struct kprobe *cur = kprobe_running();
+
+	pr_debug("post_kprobe_handler, cur=%p\n", cur);
+
+	if (!cur)
+		return 0;
+
+	if (cur->post_handler) {
+		kprobe_status = KPROBE_HIT_SSDONE;
+		cur->post_handler(cur, regs, 0);
+	}
+
+	resume_execution(cur, regs);
+	reset_current_kprobe();
+	preempt_enable_no_resched();
+
+	return 1;
+}
+
+static int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr)
+{
+	struct kprobe *cur = kprobe_running();
+
+	pr_debug("kprobe_fault_handler: trapnr=%d\n", trapnr);
+
+	if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
+		return 1;
+
+	if (kprobe_status & KPROBE_HIT_SS) {
+		resume_execution(cur, regs);
+		preempt_enable_no_resched();
+	}
+	return 0;
+}
+
+/*
+ * Wrapper routine to for handling exceptions.
+ */
+int __kprobes kprobe_exceptions_notify(struct notifier_block *self,
+				       unsigned long val, void *data)
+{
+	struct die_args *args = (struct die_args *)data;
+	int ret = NOTIFY_DONE;
+
+	pr_debug("kprobe_exceptions_notify: val=%lu, data=%p\n",
+		 val, data);
+
+	switch (val) {
+	case DIE_BREAKPOINT:
+		if (kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_SSTEP:
+		if (post_kprobe_handler(args->regs))
+			ret = NOTIFY_STOP;
+		break;
+	case DIE_FAULT:
+		if (kprobe_running()
+		    && kprobe_fault_handler(args->regs, args->trapnr))
+			ret = NOTIFY_STOP;
+		break;
+	default:
+		break;
+	}
+
+	return ret;
+}
+
+int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	struct jprobe *jp = container_of(p, struct jprobe, kp);
+
+	memcpy(&jprobe_saved_regs, regs, sizeof(struct pt_regs));
+
+	/*
+	 * TODO: We should probably save some of the stack here as
+	 * well, since gcc may pass arguments on the stack for certain
+	 * functions (lots of arguments, large aggregates, varargs)
+	 */
+
+	/* setup return addr to the jprobe handler routine */
+	regs->pc = (unsigned long)jp->entry;
+	return 1;
+}
+
+void __kprobes jprobe_return(void)
+{
+	asm volatile("breakpoint" ::: "memory");
+}
+
+int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	/*
+	 * FIXME - we should ideally be validating that we got here 'cos
+	 * of the "trap" in jprobe_return() above, before restoring the
+	 * saved regs...
+	 */
+	memcpy(regs, &jprobe_saved_regs, sizeof(struct pt_regs));
+	return 1;
+}
+
+int __init arch_init_kprobes(void)
+{
+	printk("KPROBES: Enabling monitor mode (MM|DBE)...\n");
+	__mtdr(DBGREG_DC, DC_MM | DC_DBE);
+
+	/* TODO: Register kretprobe trampoline */
+	return 0;
+}
diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c
new file mode 100644
index 000000000000..dfc32f2817b6
--- /dev/null
+++ b/arch/avr32/kernel/module.c
@@ -0,0 +1,324 @@
+/*
+ * AVR32-specific kernel module loader
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * GOT initialization parts are based on the s390 version
+ *   Copyright (C) 2002, 2003 IBM Deutschland Entwicklung GmbH,
+ *                            IBM Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/moduleloader.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/elf.h>
+#include <linux/vmalloc.h>
+
+void *module_alloc(unsigned long size)
+{
+	if (size == 0)
+		return NULL;
+	return vmalloc(size);
+}
+
+void module_free(struct module *mod, void *module_region)
+{
+	vfree(mod->arch.syminfo);
+	mod->arch.syminfo = NULL;
+
+	vfree(module_region);
+	/* FIXME: if module_region == mod->init_region, trim exception
+	 * table entries. */
+}
+
+static inline int check_rela(Elf32_Rela *rela, struct module *module,
+			     char *strings, Elf32_Sym *symbols)
+{
+	struct mod_arch_syminfo *info;
+
+	info = module->arch.syminfo + ELF32_R_SYM(rela->r_info);
+	switch (ELF32_R_TYPE(rela->r_info)) {
+	case R_AVR32_GOT32:
+	case R_AVR32_GOT16:
+	case R_AVR32_GOT8:
+	case R_AVR32_GOT21S:
+	case R_AVR32_GOT18SW:	/* mcall */
+	case R_AVR32_GOT16S:	/* ld.w */
+		if (rela->r_addend != 0) {
+			printk(KERN_ERR
+			       "GOT relocation against %s at offset %u with addend\n",
+			       strings + symbols[ELF32_R_SYM(rela->r_info)].st_name,
+			       rela->r_offset);
+			return -ENOEXEC;
+		}
+		if (info->got_offset == -1UL) {
+			info->got_offset = module->arch.got_size;
+			module->arch.got_size += sizeof(void *);
+		}
+		pr_debug("GOT[%3lu] %s\n", info->got_offset,
+			 strings + symbols[ELF32_R_SYM(rela->r_info)].st_name);
+		break;
+	}
+
+	return 0;
+}
+
+int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
+			      char *secstrings, struct module *module)
+{
+	Elf32_Shdr *symtab;
+	Elf32_Sym *symbols;
+	Elf32_Rela *rela;
+	char *strings;
+	int nrela, i, j;
+	int ret;
+
+	/* Find the symbol table */
+	symtab = NULL;
+	for (i = 0; i < hdr->e_shnum; i++)
+		switch (sechdrs[i].sh_type) {
+		case SHT_SYMTAB:
+			symtab = &sechdrs[i];
+			break;
+		}
+	if (!symtab) {
+		printk(KERN_ERR "module %s: no symbol table\n", module->name);
+		return -ENOEXEC;
+	}
+
+	/* Allocate room for one syminfo structure per symbol. */
+	module->arch.nsyms = symtab->sh_size / sizeof(Elf_Sym);
+	module->arch.syminfo = vmalloc(module->arch.nsyms
+				   * sizeof(struct mod_arch_syminfo));
+	if (!module->arch.syminfo)
+		return -ENOMEM;
+
+	symbols = (void *)hdr + symtab->sh_offset;
+	strings = (void *)hdr + sechdrs[symtab->sh_link].sh_offset;
+	for (i = 0; i < module->arch.nsyms; i++) {
+		if (symbols[i].st_shndx == SHN_UNDEF &&
+		    strcmp(strings + symbols[i].st_name,
+			   "_GLOBAL_OFFSET_TABLE_") == 0)
+			/* "Define" it as absolute. */
+			symbols[i].st_shndx = SHN_ABS;
+		module->arch.syminfo[i].got_offset = -1UL;
+		module->arch.syminfo[i].got_initialized = 0;
+	}
+
+	/* Allocate GOT entries for symbols that need it. */
+	module->arch.got_size = 0;
+	for (i = 0; i < hdr->e_shnum; i++) {
+		if (sechdrs[i].sh_type != SHT_RELA)
+			continue;
+		nrela = sechdrs[i].sh_size / sizeof(Elf32_Rela);
+		rela = (void *)hdr + sechdrs[i].sh_offset;
+		for (j = 0; j < nrela; j++) {
+			ret = check_rela(rela + j, module,
+					 strings, symbols);
+			if (ret)
+				goto out_free_syminfo;
+		}
+	}
+
+	/*
+	 * Increase core size to make room for GOT and set start
+	 * offset for GOT.
+	 */
+	module->core_size = ALIGN(module->core_size, 4);
+	module->arch.got_offset = module->core_size;
+	module->core_size += module->arch.got_size;
+
+	return 0;
+
+out_free_syminfo:
+	vfree(module->arch.syminfo);
+	module->arch.syminfo = NULL;
+
+	return ret;
+}
+
+static inline int reloc_overflow(struct module *module, const char *reloc_name,
+				 Elf32_Addr relocation)
+{
+	printk(KERN_ERR "module %s: Value %lx does not fit relocation %s\n",
+	       module->name, (unsigned long)relocation, reloc_name);
+	return -ENOEXEC;
+}
+
+#define get_u16(loc)		(*((uint16_t *)loc))
+#define put_u16(loc, val)	(*((uint16_t *)loc) = (val))
+
+int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
+		       unsigned int symindex, unsigned int relindex,
+		       struct module *module)
+{
+	Elf32_Shdr *symsec = sechdrs + symindex;
+	Elf32_Shdr *relsec = sechdrs + relindex;
+	Elf32_Shdr *dstsec = sechdrs + relsec->sh_info;
+	Elf32_Rela *rel = (void *)relsec->sh_addr;
+	unsigned int i;
+	int ret = 0;
+
+	for (i = 0; i < relsec->sh_size / sizeof(Elf32_Rela); i++, rel++) {
+		struct mod_arch_syminfo *info;
+		Elf32_Sym *sym;
+		Elf32_Addr relocation;
+		uint32_t *location;
+		uint32_t value;
+
+		location = (void *)dstsec->sh_addr + rel->r_offset;
+		sym = (Elf32_Sym *)symsec->sh_addr + ELF32_R_SYM(rel->r_info);
+		relocation = sym->st_value + rel->r_addend;
+
+		info = module->arch.syminfo + ELF32_R_SYM(rel->r_info);
+
+		/* Initialize GOT entry if necessary */
+		switch (ELF32_R_TYPE(rel->r_info)) {
+		case R_AVR32_GOT32:
+		case R_AVR32_GOT16:
+		case R_AVR32_GOT8:
+		case R_AVR32_GOT21S:
+		case R_AVR32_GOT18SW:
+		case R_AVR32_GOT16S:
+			if (!info->got_initialized) {
+				Elf32_Addr *gotent;
+
+				gotent = (module->module_core
+					  + module->arch.got_offset
+					  + info->got_offset);
+				*gotent = relocation;
+				info->got_initialized = 1;
+			}
+
+			relocation = info->got_offset;
+			break;
+		}
+
+		switch (ELF32_R_TYPE(rel->r_info)) {
+		case R_AVR32_32:
+		case R_AVR32_32_CPENT:
+			*location = relocation;
+			break;
+		case R_AVR32_22H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xffe00001) != 0
+			    && (relocation & 0xffc00001) != 0xffc00000)
+				return reloc_overflow(module,
+						      "R_AVR32_22H_PCREL",
+						      relocation);
+			relocation >>= 1;
+
+			value = *location;
+			value = ((value & 0xe1ef0000)
+				 | (relocation & 0xffff)
+				 | ((relocation & 0x10000) << 4)
+				 | ((relocation & 0x1e0000) << 8));
+			*location = value;
+			break;
+		case R_AVR32_11H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xfffffc01) != 0
+			    && (relocation & 0xfffff801) != 0xfffff800)
+				return reloc_overflow(module,
+						      "R_AVR32_11H_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf00c)
+				 | ((relocation & 0x1fe) << 3)
+				 | ((relocation & 0x600) >> 9));
+			put_u16(location, value);
+			break;
+		case R_AVR32_9H_PCREL:
+			relocation -= (Elf32_Addr)location;
+			if ((relocation & 0xffffff01) != 0
+			    && (relocation & 0xfffffe01) != 0xfffffe00)
+				return reloc_overflow(module,
+						      "R_AVR32_9H_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf00f)
+				 | ((relocation & 0x1fe) << 3));
+			put_u16(location, value);
+			break;
+		case R_AVR32_9UW_PCREL:
+			relocation -= ((Elf32_Addr)location) & 0xfffffffc;
+			if ((relocation & 0xfffffc03) != 0)
+				return reloc_overflow(module,
+						      "R_AVR32_9UW_PCREL",
+						      relocation);
+			value = get_u16(location);
+			value = ((value & 0xf80f)
+				 | ((relocation & 0x1fc) << 2));
+			put_u16(location, value);
+			break;
+		case R_AVR32_GOTPC:
+			/*
+			 * R6 = PC - (PC - GOT)
+			 *
+			 * At this point, relocation contains the
+			 * value of PC.  Just subtract the value of
+			 * GOT, and we're done.
+			 */
+			pr_debug("GOTPC: PC=0x%lx, got_offset=0x%lx, core=0x%p\n",
+				 relocation, module->arch.got_offset,
+				 module->module_core);
+			relocation -= ((unsigned long)module->module_core
+				       + module->arch.got_offset);
+			*location = relocation;
+			break;
+		case R_AVR32_GOT18SW:
+			if ((relocation & 0xfffe0003) != 0
+			    && (relocation & 0xfffc0003) != 0xffff0000)
+				return reloc_overflow(module, "R_AVR32_GOT18SW",
+						     relocation);
+			relocation >>= 2;
+			/* fall through */
+		case R_AVR32_GOT16S:
+			if ((relocation & 0xffff8000) != 0
+			    && (relocation & 0xffff0000) != 0xffff0000)
+				return reloc_overflow(module, "R_AVR32_GOT16S",
+						      relocation);
+			pr_debug("GOT reloc @ 0x%lx -> %lu\n",
+				 rel->r_offset, relocation);
+			value = *location;
+			value = ((value & 0xffff0000)
+				 | (relocation & 0xffff));
+			*location = value;
+			break;
+
+		default:
+			printk(KERN_ERR "module %s: Unknown relocation: %u\n",
+			       module->name, ELF32_R_TYPE(rel->r_info));
+			return -ENOEXEC;
+		}
+	}
+
+	return ret;
+}
+
+int apply_relocate(Elf32_Shdr *sechdrs, const char *strtab,
+		   unsigned int symindex, unsigned int relindex,
+		   struct module *module)
+{
+	printk(KERN_ERR "module %s: REL relocations are not supported\n",
+		module->name);
+	return -ENOEXEC;
+}
+
+int module_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
+		    struct module *module)
+{
+	vfree(module->arch.syminfo);
+	module->arch.syminfo = NULL;
+
+	return 0;
+}
+
+void module_arch_cleanup(struct module *module)
+{
+
+}
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
new file mode 100644
index 000000000000..317dc50945f2
--- /dev/null
+++ b/arch/avr32/kernel/process.c
@@ -0,0 +1,276 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/fs.h>
+#include <linux/ptrace.h>
+#include <linux/reboot.h>
+#include <linux/unistd.h>
+
+#include <asm/sysreg.h>
+#include <asm/ocd.h>
+
+void (*pm_power_off)(void) = NULL;
+EXPORT_SYMBOL(pm_power_off);
+
+/*
+ * This file handles the architecture-dependent parts of process handling..
+ */
+
+void cpu_idle(void)
+{
+	/* endless idle loop with no priority at all */
+	while (1) {
+		/* TODO: Enter sleep mode */
+		while (!need_resched())
+			cpu_relax();
+		preempt_enable_no_resched();
+		schedule();
+		preempt_disable();
+	}
+}
+
+void machine_halt(void)
+{
+}
+
+void machine_power_off(void)
+{
+}
+
+void machine_restart(char *cmd)
+{
+	__mtdr(DBGREG_DC, DC_DBE);
+	__mtdr(DBGREG_DC, DC_RES);
+	while (1) ;
+}
+
+/*
+ * PC is actually discarded when returning from a system call -- the
+ * return address must be stored in LR. This function will make sure
+ * LR points to do_exit before starting the thread.
+ *
+ * Also, when returning from fork(), r12 is 0, so we must copy the
+ * argument as well.
+ *
+ *  r0 : The argument to the main thread function
+ *  r1 : The address of do_exit
+ *  r2 : The address of the main thread function
+ */
+asmlinkage extern void kernel_thread_helper(void);
+__asm__("	.type	kernel_thread_helper, @function\n"
+	"kernel_thread_helper:\n"
+	"	mov	r12, r0\n"
+	"	mov	lr, r2\n"
+	"	mov	pc, r1\n"
+	"	.size	kernel_thread_helper, . - kernel_thread_helper");
+
+int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
+{
+	struct pt_regs regs;
+
+	memset(&regs, 0, sizeof(regs));
+
+	regs.r0 = (unsigned long)arg;
+	regs.r1 = (unsigned long)fn;
+	regs.r2 = (unsigned long)do_exit;
+	regs.lr = (unsigned long)kernel_thread_helper;
+	regs.pc = (unsigned long)kernel_thread_helper;
+	regs.sr = MODE_SUPERVISOR;
+
+	return do_fork(flags | CLONE_VM | CLONE_UNTRACED,
+		       0, &regs, 0, NULL, NULL);
+}
+EXPORT_SYMBOL(kernel_thread);
+
+/*
+ * Free current thread data structures etc
+ */
+void exit_thread(void)
+{
+	/* nothing to do */
+}
+
+void flush_thread(void)
+{
+	/* nothing to do */
+}
+
+void release_thread(struct task_struct *dead_task)
+{
+	/* do nothing */
+}
+
+static const char *cpu_modes[] = {
+	"Application", "Supervisor", "Interrupt level 0", "Interrupt level 1",
+	"Interrupt level 2", "Interrupt level 3", "Exception", "NMI"
+};
+
+void show_regs(struct pt_regs *regs)
+{
+	unsigned long sp = regs->sp;
+	unsigned long lr = regs->lr;
+	unsigned long mode = (regs->sr & MODE_MASK) >> MODE_SHIFT;
+
+	if (!user_mode(regs))
+		sp = (unsigned long)regs + FRAME_SIZE_FULL;
+
+	print_symbol("PC is at %s\n", instruction_pointer(regs));
+	print_symbol("LR is at %s\n", lr);
+	printk("pc : [<%08lx>]    lr : [<%08lx>]    %s\n"
+	       "sp : %08lx  r12: %08lx  r11: %08lx\n",
+	       instruction_pointer(regs),
+	       lr, print_tainted(), sp, regs->r12, regs->r11);
+	printk("r10: %08lx  r9 : %08lx  r8 : %08lx\n",
+	       regs->r10, regs->r9, regs->r8);
+	printk("r7 : %08lx  r6 : %08lx  r5 : %08lx  r4 : %08lx\n",
+	       regs->r7, regs->r6, regs->r5, regs->r4);
+	printk("r3 : %08lx  r2 : %08lx  r1 : %08lx  r0 : %08lx\n",
+	       regs->r3, regs->r2, regs->r1, regs->r0);
+	printk("Flags: %c%c%c%c%c\n",
+	       regs->sr & SR_Q ? 'Q' : 'q',
+	       regs->sr & SR_V ? 'V' : 'v',
+	       regs->sr & SR_N ? 'N' : 'n',
+	       regs->sr & SR_Z ? 'Z' : 'z',
+	       regs->sr & SR_C ? 'C' : 'c');
+	printk("Mode bits: %c%c%c%c%c%c%c%c%c\n",
+	       regs->sr & SR_H ? 'H' : 'h',
+	       regs->sr & SR_R ? 'R' : 'r',
+	       regs->sr & SR_J ? 'J' : 'j',
+	       regs->sr & SR_EM ? 'E' : 'e',
+	       regs->sr & SR_I3M ? '3' : '.',
+	       regs->sr & SR_I2M ? '2' : '.',
+	       regs->sr & SR_I1M ? '1' : '.',
+	       regs->sr & SR_I0M ? '0' : '.',
+	       regs->sr & SR_GM ? 'G' : 'g');
+	printk("CPU Mode: %s\n", cpu_modes[mode]);
+
+	show_trace(NULL, (unsigned long *)sp, regs);
+}
+EXPORT_SYMBOL(show_regs);
+
+/* Fill in the fpu structure for a core dump. This is easy -- we don't have any */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu)
+{
+	/* Not valid */
+	return 0;
+}
+
+asmlinkage void ret_from_fork(void);
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
+		unsigned long unused,
+		struct task_struct *p, struct pt_regs *regs)
+{
+	struct pt_regs *childregs;
+
+	childregs = ((struct pt_regs *)(THREAD_SIZE + (unsigned long)p->thread_info)) - 1;
+	*childregs = *regs;
+
+	if (user_mode(regs))
+		childregs->sp = usp;
+	else
+		childregs->sp = (unsigned long)p->thread_info + THREAD_SIZE;
+
+	childregs->r12 = 0; /* Set return value for child */
+
+	p->thread.cpu_context.sr = MODE_SUPERVISOR | SR_GM;
+	p->thread.cpu_context.ksp = (unsigned long)childregs;
+	p->thread.cpu_context.pc = (unsigned long)ret_from_fork;
+
+	return 0;
+}
+
+/* r12-r8 are dummy parameters to force the compiler to use the stack */
+asmlinkage int sys_fork(struct pt_regs *regs)
+{
+	return do_fork(SIGCHLD, regs->sp, regs, 0, NULL, NULL);
+}
+
+asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
+			 unsigned long parent_tidptr,
+			 unsigned long child_tidptr, struct pt_regs *regs)
+{
+	if (!newsp)
+		newsp = regs->sp;
+	return do_fork(clone_flags, newsp, regs, 0,
+		       (int __user *)parent_tidptr,
+		       (int __user *)child_tidptr);
+}
+
+asmlinkage int sys_vfork(struct pt_regs *regs)
+{
+	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->sp, regs,
+		       0, NULL, NULL);
+}
+
+asmlinkage int sys_execve(char __user *ufilename, char __user *__user *uargv,
+			  char __user *__user *uenvp, struct pt_regs *regs)
+{
+	int error;
+	char *filename;
+
+	filename = getname(ufilename);
+	error = PTR_ERR(filename);
+	if (IS_ERR(filename))
+		goto out;
+
+	error = do_execve(filename, uargv, uenvp, regs);
+	if (error == 0)
+		current->ptrace &= ~PT_DTRACE;
+	putname(filename);
+
+out:
+	return error;
+}
+
+
+/*
+ * This function is supposed to answer the question "who called
+ * schedule()?"
+ */
+unsigned long get_wchan(struct task_struct *p)
+{
+	unsigned long pc;
+	unsigned long stack_page;
+
+	if (!p || p == current || p->state == TASK_RUNNING)
+		return 0;
+
+	stack_page = (unsigned long)p->thread_info;
+	BUG_ON(!stack_page);
+
+	/*
+	 * The stored value of PC is either the address right after
+	 * the call to __switch_to() or ret_from_fork.
+	 */
+	pc = thread_saved_pc(p);
+	if (in_sched_functions(pc)) {
+#ifdef CONFIG_FRAME_POINTER
+		unsigned long fp = p->thread.cpu_context.r7;
+		BUG_ON(fp < stack_page || fp > (THREAD_SIZE + stack_page));
+		pc = *(unsigned long *)fp;
+#else
+		/*
+		 * We depend on the frame size of schedule here, which
+		 * is actually quite ugly. It might be possible to
+		 * determine the frame size automatically at build
+		 * time by doing this:
+		 *   - compile sched.c
+		 *   - disassemble the resulting sched.o
+		 *   - look for 'sub sp,??' shortly after '<schedule>:'
+		 */
+		unsigned long sp = p->thread.cpu_context.ksp + 16;
+		BUG_ON(sp < stack_page || sp > (THREAD_SIZE + stack_page));
+		pc = *(unsigned long *)sp;
+#endif
+	}
+
+	return pc;
+}
diff --git a/arch/avr32/kernel/ptrace.c b/arch/avr32/kernel/ptrace.c
new file mode 100644
index 000000000000..3c89e59029ab
--- /dev/null
+++ b/arch/avr32/kernel/ptrace.c
@@ -0,0 +1,371 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#undef DEBUG
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/smp_lock.h>
+#include <linux/ptrace.h>
+#include <linux/errno.h>
+#include <linux/user.h>
+#include <linux/security.h>
+#include <linux/unistd.h>
+#include <linux/notifier.h>
+
+#include <asm/traps.h>
+#include <asm/uaccess.h>
+#include <asm/ocd.h>
+#include <asm/mmu_context.h>
+#include <asm/kdebug.h>
+
+static struct pt_regs *get_user_regs(struct task_struct *tsk)
+{
+	return (struct pt_regs *)((unsigned long) tsk->thread_info +
+				  THREAD_SIZE - sizeof(struct pt_regs));
+}
+
+static void ptrace_single_step(struct task_struct *tsk)
+{
+	pr_debug("ptrace_single_step: pid=%u, SR=0x%08lx\n",
+		 tsk->pid, tsk->thread.cpu_context.sr);
+	if (!(tsk->thread.cpu_context.sr & SR_D)) {
+		/*
+		 * Set a breakpoint at the current pc to force the
+		 * process into debug mode.  The syscall/exception
+		 * exit code will set a breakpoint at the return
+		 * address when this flag is set.
+		 */
+		pr_debug("ptrace_single_step: Setting TIF_BREAKPOINT\n");
+		set_tsk_thread_flag(tsk, TIF_BREAKPOINT);
+	}
+
+	/* The monitor code will do the actual step for us */
+	set_tsk_thread_flag(tsk, TIF_SINGLE_STEP);
+}
+
+/*
+ * Called by kernel/ptrace.c when detaching
+ *
+ * Make sure any single step bits, etc. are not set
+ */
+void ptrace_disable(struct task_struct *child)
+{
+	clear_tsk_thread_flag(child, TIF_SINGLE_STEP);
+}
+
+/*
+ * Handle hitting a breakpoint
+ */
+static void ptrace_break(struct task_struct *tsk, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code  = TRAP_BRKPT;
+	info.si_addr  = (void __user *)instruction_pointer(regs);
+
+	pr_debug("ptrace_break: Sending SIGTRAP to PID %u (pc = 0x%p)\n",
+		 tsk->pid, info.si_addr);
+	force_sig_info(SIGTRAP, &info, tsk);
+}
+
+/*
+ * Read the word at offset "offset" into the task's "struct user". We
+ * actually access the pt_regs struct stored on the kernel stack.
+ */
+static int ptrace_read_user(struct task_struct *tsk, unsigned long offset,
+			    unsigned long __user *data)
+{
+	unsigned long *regs;
+	unsigned long value;
+
+	pr_debug("ptrace_read_user(%p, %#lx, %p)\n",
+		 tsk, offset, data);
+
+	if (offset & 3 || offset >= sizeof(struct user)) {
+		printk("ptrace_read_user: invalid offset 0x%08lx\n", offset);
+		return -EIO;
+	}
+
+	regs = (unsigned long *)get_user_regs(tsk);
+
+	value = 0;
+	if (offset < sizeof(struct pt_regs))
+		value = regs[offset / sizeof(regs[0])];
+
+	return put_user(value, data);
+}
+
+/*
+ * Write the word "value" to offset "offset" into the task's "struct
+ * user". We actually access the pt_regs struct stored on the kernel
+ * stack.
+ */
+static int ptrace_write_user(struct task_struct *tsk, unsigned long offset,
+			     unsigned long value)
+{
+	unsigned long *regs;
+
+	if (offset & 3 || offset >= sizeof(struct user)) {
+		printk("ptrace_write_user: invalid offset 0x%08lx\n", offset);
+		return -EIO;
+	}
+
+	if (offset >= sizeof(struct pt_regs))
+		return 0;
+
+	regs = (unsigned long *)get_user_regs(tsk);
+	regs[offset / sizeof(regs[0])] = value;
+
+	return 0;
+}
+
+static int ptrace_getregs(struct task_struct *tsk, void __user *uregs)
+{
+	struct pt_regs *regs = get_user_regs(tsk);
+
+	return copy_to_user(uregs, regs, sizeof(*regs)) ? -EFAULT : 0;
+}
+
+static int ptrace_setregs(struct task_struct *tsk, const void __user *uregs)
+{
+	struct pt_regs newregs;
+	int ret;
+
+	ret = -EFAULT;
+	if (copy_from_user(&newregs, uregs, sizeof(newregs)) == 0) {
+		struct pt_regs *regs = get_user_regs(tsk);
+
+		ret = -EINVAL;
+		if (valid_user_regs(&newregs)) {
+			*regs = newregs;
+			ret = 0;
+		}
+	}
+
+	return ret;
+}
+
+long arch_ptrace(struct task_struct *child, long request, long addr, long data)
+{
+	unsigned long tmp;
+	int ret;
+
+	pr_debug("arch_ptrace(%ld, %ld, %#lx, %#lx)\n",
+		 request, child->pid, addr, data);
+
+	pr_debug("ptrace: Enabling monitor mode...\n");
+	__mtdr(DBGREG_DC, __mfdr(DBGREG_DC) | DC_MM | DC_DBE);
+
+	switch (request) {
+	/* Read the word at location addr in the child process */
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKDATA:
+		ret = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
+		if (ret == sizeof(tmp))
+			ret = put_user(tmp, (unsigned long __user *)data);
+		else
+			ret = -EIO;
+		break;
+
+	case PTRACE_PEEKUSR:
+		ret = ptrace_read_user(child, addr,
+				       (unsigned long __user *)data);
+		break;
+
+	/* Write the word in data at location addr */
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEDATA:
+		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
+		if (ret == sizeof(data))
+			ret = 0;
+		else
+			ret = -EIO;
+		break;
+
+	case PTRACE_POKEUSR:
+		ret = ptrace_write_user(child, addr, data);
+		break;
+
+	/* continue and stop at next (return from) syscall */
+	case PTRACE_SYSCALL:
+	/* restart after signal */
+	case PTRACE_CONT:
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+		if (request == PTRACE_SYSCALL)
+			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		else
+			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		child->exit_code = data;
+		/* XXX: Are we sure no breakpoints are active here? */
+		wake_up_process(child);
+		ret = 0;
+		break;
+
+	/*
+	 * Make the child exit. Best I can do is send it a
+	 * SIGKILL. Perhaps it should be put in the status that it
+	 * wants to exit.
+	 */
+	case PTRACE_KILL:
+		ret = 0;
+		if (child->exit_state == EXIT_ZOMBIE)
+			break;
+		child->exit_code = SIGKILL;
+		wake_up_process(child);
+		break;
+
+	/*
+	 * execute single instruction.
+	 */
+	case PTRACE_SINGLESTEP:
+		ret = -EIO;
+		if (!valid_signal(data))
+			break;
+		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+		ptrace_single_step(child);
+		child->exit_code = data;
+		wake_up_process(child);
+		ret = 0;
+		break;
+
+	/* Detach a process that was attached */
+	case PTRACE_DETACH:
+		ret = ptrace_detach(child, data);
+		break;
+
+	case PTRACE_GETREGS:
+		ret = ptrace_getregs(child, (void __user *)data);
+		break;
+
+	case PTRACE_SETREGS:
+		ret = ptrace_setregs(child, (const void __user *)data);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+		break;
+	}
+
+	pr_debug("sys_ptrace returning %d (DC = 0x%08lx)\n", ret, __mfdr(DBGREG_DC));
+	return ret;
+}
+
+asmlinkage void syscall_trace(void)
+{
+	pr_debug("syscall_trace called\n");
+	if (!test_thread_flag(TIF_SYSCALL_TRACE))
+		return;
+	if (!(current->ptrace & PT_PTRACED))
+		return;
+
+	pr_debug("syscall_trace: notifying parent\n");
+	/* The 0x80 provides a way for the tracing parent to
+	 * distinguish between a syscall stop and SIGTRAP delivery */
+	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
+				 ? 0x80 : 0));
+
+	/*
+	 * this isn't the same as continuing with a signal, but it
+	 * will do for normal use.  strace only continues with a
+	 * signal if the stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		pr_debug("syscall_trace: sending signal %d to PID %u\n",
+			 current->exit_code, current->pid);
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}
+
+asmlinkage void do_debug_priv(struct pt_regs *regs)
+{
+	unsigned long dc, ds;
+	unsigned long die_val;
+
+	ds = __mfdr(DBGREG_DS);
+
+	pr_debug("do_debug_priv: pc = %08lx, ds = %08lx\n", regs->pc, ds);
+
+	if (ds & DS_SSS)
+		die_val = DIE_SSTEP;
+	else
+		die_val = DIE_BREAKPOINT;
+
+	if (notify_die(die_val, regs, 0, SIGTRAP) == NOTIFY_STOP)
+		return;
+
+	if (likely(ds & DS_SSS)) {
+		extern void itlb_miss(void);
+		extern void tlb_miss_common(void);
+		struct thread_info *ti;
+
+		dc = __mfdr(DBGREG_DC);
+		dc &= ~DC_SS;
+		__mtdr(DBGREG_DC, dc);
+
+		ti = current_thread_info();
+		ti->flags |= _TIF_BREAKPOINT;
+
+		/* The TLB miss handlers don't check thread flags */
+		if ((regs->pc >= (unsigned long)&itlb_miss)
+		    && (regs->pc <= (unsigned long)&tlb_miss_common)) {
+			__mtdr(DBGREG_BWA2A, sysreg_read(RAR_EX));
+			__mtdr(DBGREG_BWC2A, 0x40000001 | (get_asid() << 1));
+		}
+
+		/*
+		 * If we're running in supervisor mode, the breakpoint
+		 * will take us where we want directly, no need to
+		 * single step.
+		 */
+		if ((regs->sr & MODE_MASK) != MODE_SUPERVISOR)
+			ti->flags |= TIF_SINGLE_STEP;
+	} else {
+		panic("Unable to handle debug trap at pc = %08lx\n",
+		      regs->pc);
+	}
+}
+
+/*
+ * Handle breakpoints, single steps and other debuggy things. To keep
+ * things simple initially, we run with interrupts and exceptions
+ * disabled all the time.
+ */
+asmlinkage void do_debug(struct pt_regs *regs)
+{
+	unsigned long dc, ds;
+
+	ds = __mfdr(DBGREG_DS);
+	pr_debug("do_debug: pc = %08lx, ds = %08lx\n", regs->pc, ds);
+
+	if (test_thread_flag(TIF_BREAKPOINT)) {
+		pr_debug("TIF_BREAKPOINT set\n");
+		/* We're taking care of it */
+		clear_thread_flag(TIF_BREAKPOINT);
+		__mtdr(DBGREG_BWC2A, 0);
+	}
+
+	if (test_thread_flag(TIF_SINGLE_STEP)) {
+		pr_debug("TIF_SINGLE_STEP set, ds = 0x%08lx\n", ds);
+		if (ds & DS_SSS) {
+			dc = __mfdr(DBGREG_DC);
+			dc &= ~DC_SS;
+			__mtdr(DBGREG_DC, dc);
+
+			clear_thread_flag(TIF_SINGLE_STEP);
+			ptrace_break(current, regs);
+		}
+	} else {
+		/* regular breakpoint */
+		ptrace_break(current, regs);
+	}
+}
diff --git a/arch/avr32/kernel/semaphore.c b/arch/avr32/kernel/semaphore.c
new file mode 100644
index 000000000000..1e2705a05016
--- /dev/null
+++ b/arch/avr32/kernel/semaphore.c
@@ -0,0 +1,148 @@
+/*
+ * AVR32 sempahore implementation.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/i386/kernel/semaphore.c
+ *  Copyright (C) 1999 Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sched.h>
+#include <linux/errno.h>
+#include <linux/module.h>
+
+#include <asm/semaphore.h>
+#include <asm/atomic.h>
+
+/*
+ * Semaphores are implemented using a two-way counter:
+ * The "count" variable is decremented for each process
+ * that tries to acquire the semaphore, while the "sleeping"
+ * variable is a count of such acquires.
+ *
+ * Notably, the inline "up()" and "down()" functions can
+ * efficiently test if they need to do any extra work (up
+ * needs to do something only if count was negative before
+ * the increment operation.
+ *
+ * "sleeping" and the contention routine ordering is protected
+ * by the spinlock in the semaphore's waitqueue head.
+ *
+ * Note that these functions are only called when there is
+ * contention on the lock, and as such all this is the
+ * "non-critical" part of the whole semaphore business. The
+ * critical part is the inline stuff in <asm/semaphore.h>
+ * where we want to avoid any extra jumps and calls.
+ */
+
+/*
+ * Logic:
+ *  - only on a boundary condition do we need to care. When we go
+ *    from a negative count to a non-negative, we wake people up.
+ *  - when we go from a non-negative count to a negative do we
+ *    (a) synchronize with the "sleeper" count and (b) make sure
+ *    that we're on the wakeup list before we synchronize so that
+ *    we cannot lose wakeup events.
+ */
+
+void __up(struct semaphore *sem)
+{
+	wake_up(&sem->wait);
+}
+EXPORT_SYMBOL(__up);
+
+void __sched __down(struct semaphore *sem)
+{
+	struct task_struct *tsk = current;
+        DECLARE_WAITQUEUE(wait, tsk);
+        unsigned long flags;
+
+        tsk->state = TASK_UNINTERRUPTIBLE;
+        spin_lock_irqsave(&sem->wait.lock, flags);
+        add_wait_queue_exclusive_locked(&sem->wait, &wait);
+
+        sem->sleepers++;
+        for (;;) {
+                int sleepers = sem->sleepers;
+
+                /*
+                 * Add "everybody else" into it. They aren't
+                 * playing, because we own the spinlock in
+                 * the wait_queue_head.
+                 */
+                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
+                        sem->sleepers = 0;
+                        break;
+                }
+                sem->sleepers = 1;      /* us - see -1 above */
+                spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+                schedule();
+
+                spin_lock_irqsave(&sem->wait.lock, flags);
+                tsk->state = TASK_UNINTERRUPTIBLE;
+        }
+        remove_wait_queue_locked(&sem->wait, &wait);
+        wake_up_locked(&sem->wait);
+        spin_unlock_irqrestore(&sem->wait.lock, flags);
+        tsk->state = TASK_RUNNING;
+}
+EXPORT_SYMBOL(__down);
+
+int __sched __down_interruptible(struct semaphore *sem)
+{
+	int retval = 0;
+	struct task_struct *tsk = current;
+        DECLARE_WAITQUEUE(wait, tsk);
+        unsigned long flags;
+
+        tsk->state = TASK_INTERRUPTIBLE;
+        spin_lock_irqsave(&sem->wait.lock, flags);
+        add_wait_queue_exclusive_locked(&sem->wait, &wait);
+
+        sem->sleepers++;
+        for (;;) {
+                int sleepers = sem->sleepers;
+
+		/*
+		 * With signals pending, this turns into the trylock
+		 * failure case - we won't be sleeping, and we can't
+		 * get the lock as it has contention. Just correct the
+		 * count and exit.
+		 */
+		if (signal_pending(current)) {
+			retval = -EINTR;
+			sem->sleepers = 0;
+			atomic_add(sleepers, &sem->count);
+			break;
+		}
+
+                /*
+                 * Add "everybody else" into it. They aren't
+                 * playing, because we own the spinlock in
+                 * the wait_queue_head.
+                 */
+                if (atomic_add_return(sleepers - 1, &sem->count) >= 0) {
+                        sem->sleepers = 0;
+                        break;
+                }
+                sem->sleepers = 1;      /* us - see -1 above */
+                spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+                schedule();
+
+                spin_lock_irqsave(&sem->wait.lock, flags);
+                tsk->state = TASK_INTERRUPTIBLE;
+        }
+        remove_wait_queue_locked(&sem->wait, &wait);
+        wake_up_locked(&sem->wait);
+        spin_unlock_irqrestore(&sem->wait.lock, flags);
+
+        tsk->state = TASK_RUNNING;
+	return retval;
+}
+EXPORT_SYMBOL(__down_interruptible);
diff --git a/arch/avr32/kernel/setup.c b/arch/avr32/kernel/setup.c
new file mode 100644
index 000000000000..5d68f3c6990b
--- /dev/null
+++ b/arch/avr32/kernel/setup.c
@@ -0,0 +1,335 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/console.h>
+#include <linux/ioport.h>
+#include <linux/bootmem.h>
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <linux/root_dev.h>
+#include <linux/cpu.h>
+
+#include <asm/sections.h>
+#include <asm/processor.h>
+#include <asm/pgtable.h>
+#include <asm/setup.h>
+#include <asm/sysreg.h>
+
+#include <asm/arch/board.h>
+#include <asm/arch/init.h>
+
+extern int root_mountflags;
+
+/*
+ * Bootloader-provided information about physical memory
+ */
+struct tag_mem_range *mem_phys;
+struct tag_mem_range *mem_reserved;
+struct tag_mem_range *mem_ramdisk;
+
+/*
+ * Initialize loops_per_jiffy as 5000000 (500MIPS).
+ * Better make it too large than too small...
+ */
+struct avr32_cpuinfo boot_cpu_data = {
+	.loops_per_jiffy = 5000000
+};
+EXPORT_SYMBOL(boot_cpu_data);
+
+static char command_line[COMMAND_LINE_SIZE];
+
+/*
+ * Should be more than enough, but if you have a _really_ complex
+ * setup, you might need to increase the size of this...
+ */
+static struct tag_mem_range __initdata mem_range_cache[32];
+static unsigned mem_range_next_free;
+
+/*
+ * Standard memory resources
+ */
+static struct resource mem_res[] = {
+	{
+		.name	= "Kernel code",
+		.start	= 0,
+		.end	= 0,
+		.flags	= IORESOURCE_MEM
+	},
+	{
+		.name	= "Kernel data",
+		.start	= 0,
+		.end	= 0,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+#define kernel_code	mem_res[0]
+#define kernel_data	mem_res[1]
+
+/*
+ * Early framebuffer allocation. Works as follows:
+ *   - If fbmem_size is zero, nothing will be allocated or reserved.
+ *   - If fbmem_start is zero when setup_bootmem() is called,
+ *     fbmem_size bytes will be allocated from the bootmem allocator.
+ *   - If fbmem_start is nonzero, an area of size fbmem_size will be
+ *     reserved at the physical address fbmem_start if necessary. If
+ *     the area isn't in a memory region known to the kernel, it will
+ *     be left alone.
+ *
+ * Board-specific code may use these variables to set up platform data
+ * for the framebuffer driver if fbmem_size is nonzero.
+ */
+static unsigned long __initdata fbmem_start;
+static unsigned long __initdata fbmem_size;
+
+/*
+ * "fbmem=xxx[kKmM]" allocates the specified amount of boot memory for
+ * use as framebuffer.
+ *
+ * "fbmem=xxx[kKmM]@yyy[kKmM]" defines a memory region of size xxx and
+ * starting at yyy to be reserved for use as framebuffer.
+ *
+ * The kernel won't verify that the memory region starting at yyy
+ * actually contains usable RAM.
+ */
+static int __init early_parse_fbmem(char *p)
+{
+	fbmem_size = memparse(p, &p);
+	if (*p == '@')
+		fbmem_start = memparse(p, &p);
+	return 0;
+}
+early_param("fbmem", early_parse_fbmem);
+
+static inline void __init resource_init(void)
+{
+	struct tag_mem_range *region;
+
+	kernel_code.start = __pa(init_mm.start_code);
+	kernel_code.end = __pa(init_mm.end_code - 1);
+	kernel_data.start = __pa(init_mm.end_code);
+	kernel_data.end = __pa(init_mm.brk - 1);
+
+	for (region = mem_phys; region; region = region->next) {
+		struct resource *res;
+		unsigned long phys_start, phys_end;
+
+		if (region->size == 0)
+			continue;
+
+		phys_start = region->addr;
+		phys_end = phys_start + region->size - 1;
+
+		res = alloc_bootmem_low(sizeof(*res));
+		res->name = "System RAM";
+		res->start = phys_start;
+		res->end = phys_end;
+		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
+
+		request_resource (&iomem_resource, res);
+
+		if (kernel_code.start >= res->start &&
+		    kernel_code.end <= res->end)
+			request_resource (res, &kernel_code);
+		if (kernel_data.start >= res->start &&
+		    kernel_data.end <= res->end)
+			request_resource (res, &kernel_data);
+	}
+}
+
+static int __init parse_tag_core(struct tag *tag)
+{
+	if (tag->hdr.size > 2) {
+		if ((tag->u.core.flags & 1) == 0)
+			root_mountflags &= ~MS_RDONLY;
+		ROOT_DEV = new_decode_dev(tag->u.core.rootdev);
+	}
+	return 0;
+}
+__tagtable(ATAG_CORE, parse_tag_core);
+
+static int __init parse_tag_mem_range(struct tag *tag,
+				      struct tag_mem_range **root)
+{
+	struct tag_mem_range *cur, **pprev;
+	struct tag_mem_range *new;
+
+	/*
+	 * Ignore zero-sized entries. If we're running standalone, the
+	 * SDRAM code may emit such entries if something goes
+	 * wrong...
+	 */
+	if (tag->u.mem_range.size == 0)
+		return 0;
+
+	/*
+	 * Copy the data so the bootmem init code doesn't need to care
+	 * about it.
+	 */
+	if (mem_range_next_free >=
+	    (sizeof(mem_range_cache) / sizeof(mem_range_cache[0])))
+		panic("Physical memory map too complex!\n");
+
+	new = &mem_range_cache[mem_range_next_free++];
+	*new = tag->u.mem_range;
+
+	pprev = root;
+	cur = *root;
+	while (cur) {
+		pprev = &cur->next;
+		cur = cur->next;
+	}
+
+	*pprev = new;
+	new->next = NULL;
+
+	return 0;
+}
+
+static int __init parse_tag_mem(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_phys);
+}
+__tagtable(ATAG_MEM, parse_tag_mem);
+
+static int __init parse_tag_cmdline(struct tag *tag)
+{
+	strlcpy(saved_command_line, tag->u.cmdline.cmdline, COMMAND_LINE_SIZE);
+	return 0;
+}
+__tagtable(ATAG_CMDLINE, parse_tag_cmdline);
+
+static int __init parse_tag_rdimg(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_ramdisk);
+}
+__tagtable(ATAG_RDIMG, parse_tag_rdimg);
+
+static int __init parse_tag_clock(struct tag *tag)
+{
+	/*
+	 * We'll figure out the clocks by peeking at the system
+	 * manager regs directly.
+	 */
+	return 0;
+}
+__tagtable(ATAG_CLOCK, parse_tag_clock);
+
+static int __init parse_tag_rsvd_mem(struct tag *tag)
+{
+	return parse_tag_mem_range(tag, &mem_reserved);
+}
+__tagtable(ATAG_RSVD_MEM, parse_tag_rsvd_mem);
+
+static int __init parse_tag_ethernet(struct tag *tag)
+{
+#if 0
+	const struct platform_device *pdev;
+
+	/*
+	 * We really need a bus type that supports "classes"...this
+	 * will do for now (until we must handle other kinds of
+	 * ethernet controllers)
+	 */
+	pdev = platform_get_device("macb", tag->u.ethernet.mac_index);
+	if (pdev && pdev->dev.platform_data) {
+		struct eth_platform_data *data = pdev->dev.platform_data;
+
+		data->valid = 1;
+		data->mii_phy_addr = tag->u.ethernet.mii_phy_addr;
+		memcpy(data->hw_addr, tag->u.ethernet.hw_address,
+		       sizeof(data->hw_addr));
+	}
+#endif
+	return 0;
+}
+__tagtable(ATAG_ETHERNET, parse_tag_ethernet);
+
+/*
+ * Scan the tag table for this tag, and call its parse function. The
+ * tag table is built by the linker from all the __tagtable
+ * declarations.
+ */
+static int __init parse_tag(struct tag *tag)
+{
+	extern struct tagtable __tagtable_begin, __tagtable_end;
+	struct tagtable *t;
+
+	for (t = &__tagtable_begin; t < &__tagtable_end; t++)
+		if (tag->hdr.tag == t->tag) {
+			t->parse(tag);
+			break;
+		}
+
+	return t < &__tagtable_end;
+}
+
+/*
+ * Parse all tags in the list we got from the boot loader
+ */
+static void __init parse_tags(struct tag *t)
+{
+	for (; t->hdr.tag != ATAG_NONE; t = tag_next(t))
+		if (!parse_tag(t))
+			printk(KERN_WARNING
+			       "Ignoring unrecognised tag 0x%08x\n",
+			       t->hdr.tag);
+}
+
+void __init setup_arch (char **cmdline_p)
+{
+	struct clk *cpu_clk;
+
+	parse_tags(bootloader_tags);
+
+	setup_processor();
+	setup_platform();
+
+	cpu_clk = clk_get(NULL, "cpu");
+	if (IS_ERR(cpu_clk)) {
+		printk(KERN_WARNING "Warning: Unable to get CPU clock\n");
+	} else {
+		unsigned long cpu_hz = clk_get_rate(cpu_clk);
+
+		/*
+		 * Well, duh, but it's probably a good idea to
+		 * increment the use count.
+		 */
+		clk_enable(cpu_clk);
+
+		boot_cpu_data.clk = cpu_clk;
+		boot_cpu_data.loops_per_jiffy = cpu_hz * 4;
+		printk("CPU: Running at %lu.%03lu MHz\n",
+		       ((cpu_hz + 500) / 1000) / 1000,
+		       ((cpu_hz + 500) / 1000) % 1000);
+	}
+
+	init_mm.start_code = (unsigned long) &_text;
+	init_mm.end_code = (unsigned long) &_etext;
+	init_mm.end_data = (unsigned long) &_edata;
+	init_mm.brk = (unsigned long) &_end;
+
+	strlcpy(command_line, saved_command_line, COMMAND_LINE_SIZE);
+	*cmdline_p = command_line;
+	parse_early_param();
+
+	setup_bootmem();
+
+	board_setup_fbmem(fbmem_start, fbmem_size);
+
+#ifdef CONFIG_VT
+	conswitchp = &dummy_con;
+#endif
+
+	paging_init();
+
+	resource_init();
+}
diff --git a/arch/avr32/kernel/signal.c b/arch/avr32/kernel/signal.c
new file mode 100644
index 000000000000..33096651c24f
--- /dev/null
+++ b/arch/avr32/kernel/signal.c
@@ -0,0 +1,328 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/sh/kernel/signal.c
+ *  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
+ *  Copyright (C) 1991, 1992  Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/errno.h>
+#include <linux/ptrace.h>
+#include <linux/unistd.h>
+#include <linux/suspend.h>
+
+#include <asm/uaccess.h>
+#include <asm/ucontext.h>
+
+#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
+
+asmlinkage int sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+			       struct pt_regs *regs)
+{
+	return do_sigaltstack(uss, uoss, regs->sp);
+}
+
+struct rt_sigframe
+{
+	struct siginfo info;
+	struct ucontext uc;
+	unsigned long retcode;
+};
+
+static int
+restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc)
+{
+	int err = 0;
+
+#define COPY(x)		err |= __get_user(regs->x, &sc->x)
+	COPY(sr);
+	COPY(pc);
+	COPY(lr);
+	COPY(sp);
+	COPY(r12);
+	COPY(r11);
+	COPY(r10);
+	COPY(r9);
+	COPY(r8);
+	COPY(r7);
+	COPY(r6);
+	COPY(r5);
+	COPY(r4);
+	COPY(r3);
+	COPY(r2);
+	COPY(r1);
+	COPY(r0);
+#undef	COPY
+
+	/*
+	 * Don't allow anyone to pretend they're running in supervisor
+	 * mode or something...
+	 */
+	err |= !valid_user_regs(regs);
+
+	return err;
+}
+
+
+asmlinkage int sys_rt_sigreturn(struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	sigset_t set;
+
+	frame = (struct rt_sigframe __user *)regs->sp;
+	pr_debug("SIG return: frame = %p\n", frame);
+
+	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
+		goto badframe;
+
+	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
+		goto badframe;
+
+	sigdelsetmask(&set, ~_BLOCKABLE);
+	spin_lock_irq(&current->sighand->siglock);
+	current->blocked = set;
+	recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext))
+		goto badframe;
+
+	pr_debug("Context restored: pc = %08lx, lr = %08lx, sp = %08lx\n",
+		 regs->pc, regs->lr, regs->sp);
+
+	return regs->r12;
+
+badframe:
+	force_sig(SIGSEGV, current);
+	return 0;
+}
+
+static int
+setup_sigcontext(struct sigcontext __user *sc, struct pt_regs *regs)
+{
+	int err = 0;
+
+#define COPY(x)		err |= __put_user(regs->x, &sc->x)
+	COPY(sr);
+	COPY(pc);
+	COPY(lr);
+	COPY(sp);
+	COPY(r12);
+	COPY(r11);
+	COPY(r10);
+	COPY(r9);
+	COPY(r8);
+	COPY(r7);
+	COPY(r6);
+	COPY(r5);
+	COPY(r4);
+	COPY(r3);
+	COPY(r2);
+	COPY(r1);
+	COPY(r0);
+#undef	COPY
+
+	return err;
+}
+
+static inline void __user *
+get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, int framesize)
+{
+	unsigned long sp = regs->sp;
+
+	if ((ka->sa.sa_flags & SA_ONSTACK) && !sas_ss_flags(sp))
+		sp = current->sas_ss_sp + current->sas_ss_size;
+
+	return (void __user *)((sp - framesize) & ~3);
+}
+
+static int
+setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+	       sigset_t *set, struct pt_regs *regs)
+{
+	struct rt_sigframe __user *frame;
+	int err = 0;
+
+	frame = get_sigframe(ka, regs, sizeof(*frame));
+	err = -EFAULT;
+	if (!access_ok(VERIFY_WRITE, frame, sizeof (*frame)))
+		goto out;
+
+	/*
+	 * Set up the return code:
+	 *
+	 *	mov	r8, __NR_rt_sigreturn
+	 *	scall
+	 *
+	 * Note: This will blow up since we're using a non-executable
+	 * stack. Better use SA_RESTORER.
+	 */
+#if __NR_rt_sigreturn > 127
+# error __NR_rt_sigreturn must be < 127 to fit in a short mov
+#endif
+	err = __put_user(0x3008d733 | (__NR_rt_sigreturn << 20),
+			 &frame->retcode);
+
+	err |= copy_siginfo_to_user(&frame->info, info);
+
+	/* Set up the ucontext */
+	err |= __put_user(0, &frame->uc.uc_flags);
+	err |= __put_user(NULL, &frame->uc.uc_link);
+	err |= __put_user((void __user *)current->sas_ss_sp,
+			  &frame->uc.uc_stack.ss_sp);
+	err |= __put_user(sas_ss_flags(regs->sp),
+			  &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(current->sas_ss_size,
+			  &frame->uc.uc_stack.ss_size);
+	err |= setup_sigcontext(&frame->uc.uc_mcontext, regs);
+	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
+
+	if (err)
+		goto out;
+
+	regs->r12 = sig;
+	regs->r11 = (unsigned long) &frame->info;
+	regs->r10 = (unsigned long) &frame->uc;
+	regs->sp = (unsigned long) frame;
+	if (ka->sa.sa_flags & SA_RESTORER)
+		regs->lr = (unsigned long)ka->sa.sa_restorer;
+	else {
+		printk(KERN_NOTICE "[%s:%d] did not set SA_RESTORER\n",
+		       current->comm, current->pid);
+		regs->lr = (unsigned long) &frame->retcode;
+	}
+
+	pr_debug("SIG deliver [%s:%d]: sig=%d sp=0x%lx pc=0x%lx->0x%p lr=0x%lx\n",
+		 current->comm, current->pid, sig, regs->sp,
+		 regs->pc, ka->sa.sa_handler, regs->lr);
+
+	regs->pc = (unsigned long) ka->sa.sa_handler;
+
+out:
+	return err;
+}
+
+static inline void restart_syscall(struct pt_regs *regs)
+{
+	if (regs->r12 == -ERESTART_RESTARTBLOCK)
+		regs->r8 = __NR_restart_syscall;
+	else
+		regs->r12 = regs->r12_orig;
+	regs->pc -= 2;
+}
+
+static inline void
+handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
+	      sigset_t *oldset, struct pt_regs *regs, int syscall)
+{
+	int ret;
+
+	/*
+	 * Set up the stack frame
+	 */
+	ret = setup_rt_frame(sig, ka, info, oldset, regs);
+
+	/*
+	 * Check that the resulting registers are sane
+	 */
+	ret |= !valid_user_regs(regs);
+
+	/*
+	 * Block the signal if we were unsuccessful.
+	 */
+	if (ret != 0 || !(ka->sa.sa_flags & SA_NODEFER)) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked, &current->blocked,
+			  &ka->sa.sa_mask);
+		sigaddset(&current->blocked, sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	if (ret == 0)
+		return;
+
+	force_sigsegv(sig, current);
+}
+
+/*
+ * Note that 'init' is a special process: it doesn't get signals it
+ * doesn't want to handle. Thus you cannot kill init even with a
+ * SIGKILL even by mistake.
+ */
+int do_signal(struct pt_regs *regs, sigset_t *oldset, int syscall)
+{
+	siginfo_t info;
+	int signr;
+	struct k_sigaction ka;
+
+	/*
+	 * We want the common case to go fast, which is why we may in
+	 * certain cases get here from kernel mode. Just return
+	 * without doing anything if so.
+	 */
+	if (!user_mode(regs))
+		return 0;
+
+	if (try_to_freeze()) {
+		signr = 0;
+		if (!signal_pending(current))
+			goto no_signal;
+	}
+
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else if (!oldset)
+		oldset = &current->blocked;
+
+	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
+no_signal:
+	if (syscall) {
+		switch (regs->r12) {
+		case -ERESTART_RESTARTBLOCK:
+		case -ERESTARTNOHAND:
+			if (signr > 0) {
+				regs->r12 = -EINTR;
+				break;
+			}
+			/* fall through */
+		case -ERESTARTSYS:
+			if (signr > 0 && !(ka.sa.sa_flags & SA_RESTART)) {
+				regs->r12 = -EINTR;
+				break;
+			}
+			/* fall through */
+		case -ERESTARTNOINTR:
+			restart_syscall(regs);
+		}
+	}
+
+	if (signr == 0) {
+		/* No signal to deliver -- put the saved sigmask back */
+		if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+			clear_thread_flag(TIF_RESTORE_SIGMASK);
+			sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+		}
+		return 0;
+	}
+
+	handle_signal(signr, &ka, &info, oldset, regs, syscall);
+	return 1;
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, struct thread_info *ti)
+{
+	int syscall = 0;
+
+	if ((sysreg_read(SR) & MODE_MASK) == MODE_SUPERVISOR)
+		syscall = 1;
+
+	if (ti->flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, &current->blocked, syscall);
+}
diff --git a/arch/avr32/kernel/switch_to.S b/arch/avr32/kernel/switch_to.S
new file mode 100644
index 000000000000..a48d046723c5
--- /dev/null
+++ b/arch/avr32/kernel/switch_to.S
@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <asm/sysreg.h>
+
+	.text
+	.global	__switch_to
+	.type	__switch_to, @function
+
+	/* Switch thread context from "prev" to "next", returning "last"
+	 *   r12 :	prev
+	 *   r11 :	&prev->thread + 1
+	 *   r10 :	&next->thread
+	 */
+__switch_to:
+	stm	--r11, r0,r1,r2,r3,r4,r5,r6,r7,sp,lr
+	mfsr	r9, SYSREG_SR
+	st.w	--r11, r9
+	ld.w	r8, r10++
+	/*
+	 * schedule() may have been called from a mode with a different
+	 * set of registers. Make sure we don't lose anything here.
+	 */
+	pushm	r10,r12
+	mtsr	SYSREG_SR, r8
+	frs			/* flush the return stack */
+	sub	pc, -2		/* flush the pipeline */
+	popm	r10,r12
+	ldm	r10++, r0,r1,r2,r3,r4,r5,r6,r7,sp,pc
+	.size	__switch_to, . - __switch_to
diff --git a/arch/avr32/kernel/sys_avr32.c b/arch/avr32/kernel/sys_avr32.c
new file mode 100644
index 000000000000..6ec5693da448
--- /dev/null
+++ b/arch/avr32/kernel/sys_avr32.c
@@ -0,0 +1,51 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm.h>
+#include <linux/unistd.h>
+
+#include <asm/mman.h>
+#include <asm/uaccess.h>
+
+asmlinkage int sys_pipe(unsigned long __user *filedes)
+{
+	int fd[2];
+	int error;
+
+	error = do_pipe(fd);
+	if (!error) {
+		if (copy_to_user(filedes, fd, sizeof(fd)))
+			error = -EFAULT;
+	}
+	return error;
+}
+
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+			  unsigned long prot, unsigned long flags,
+			  unsigned long fd, off_t offset)
+{
+	int error = -EBADF;
+	struct file *file = NULL;
+
+	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
+	if (!(flags & MAP_ANONYMOUS)) {
+		file = fget(fd);
+		if (!file)
+			return error;
+	}
+
+	down_write(&current->mm->mmap_sem);
+	error = do_mmap_pgoff(file, addr, len, prot, flags, offset);
+	up_write(&current->mm->mmap_sem);
+
+	if (file)
+		fput(file);
+	return error;
+}
diff --git a/arch/avr32/kernel/syscall-stubs.S b/arch/avr32/kernel/syscall-stubs.S
new file mode 100644
index 000000000000..7589a9b426cb
--- /dev/null
+++ b/arch/avr32/kernel/syscall-stubs.S
@@ -0,0 +1,102 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * Stubs for syscalls that require access to pt_regs or that take more
+ * than five parameters.
+ */
+
+#define ARG6	r3
+
+	.text
+	.global __sys_rt_sigsuspend
+	.type	__sys_rt_sigsuspend,@function
+__sys_rt_sigsuspend:
+	mov	r10, sp
+	rjmp	sys_rt_sigsuspend
+
+	.global	__sys_sigaltstack
+	.type	__sys_sigaltstack,@function
+__sys_sigaltstack:
+	mov	r10, sp
+	rjmp	sys_sigaltstack
+
+	.global	__sys_rt_sigreturn
+	.type	__sys_rt_sigreturn,@function
+__sys_rt_sigreturn:
+	mov	r12, sp
+	rjmp	sys_rt_sigreturn
+
+	.global	__sys_fork
+	.type	__sys_fork,@function
+__sys_fork:
+	mov	r12, sp
+	rjmp	sys_fork
+
+	.global	__sys_clone
+	.type	__sys_clone,@function
+__sys_clone:
+	mov	r8, sp
+	rjmp	sys_clone
+
+	.global	__sys_vfork
+	.type	__sys_vfork,@function
+__sys_vfork:
+	mov	r12, sp
+	rjmp	sys_vfork
+
+	.global	__sys_execve
+	.type	__sys_execve,@function
+__sys_execve:
+	mov	r9, sp
+	rjmp	sys_execve
+
+	.global	__sys_mmap2
+	.type	__sys_mmap2,@function
+__sys_mmap2:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_mmap2
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_sendto
+	.type	__sys_sendto,@function
+__sys_sendto:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_sendto
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_recvfrom
+	.type	__sys_recvfrom,@function
+__sys_recvfrom:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_recvfrom
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_pselect6
+	.type	__sys_pselect6,@function
+__sys_pselect6:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_pselect6
+	sub	sp, -4
+	popm	pc
+
+	.global	__sys_splice
+	.type	__sys_splice,@function
+__sys_splice:
+	pushm	lr
+	st.w	--sp, ARG6
+	rcall	sys_splice
+	sub	sp, -4
+	popm	pc
diff --git a/arch/avr32/kernel/syscall_table.S b/arch/avr32/kernel/syscall_table.S
new file mode 100644
index 000000000000..63b206965d05
--- /dev/null
+++ b/arch/avr32/kernel/syscall_table.S
@@ -0,0 +1,289 @@
+/*
+ * AVR32 system call table
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
+#define sys_nfsservctl sys_ni_syscall
+#endif
+
+#if !defined(CONFIG_SYSV_IPC)
+# define sys_ipc	sys_ni_syscall
+#endif
+
+	.section .rodata,"a",@progbits
+	.type	sys_call_table,@object
+	.global	sys_call_table
+	.align	2
+sys_call_table:
+	.long	sys_restart_syscall
+	.long	sys_exit
+	.long	__sys_fork
+	.long	sys_read
+	.long	sys_write
+	.long	sys_open		/* 5 */
+	.long	sys_close
+	.long	sys_umask
+	.long	sys_creat
+	.long	sys_link
+	.long	sys_unlink		/* 10 */
+	.long	__sys_execve
+	.long	sys_chdir
+	.long	sys_time
+	.long	sys_mknod
+	.long	sys_chmod		/* 15 */
+	.long	sys_chown
+	.long	sys_lchown
+	.long	sys_lseek
+	.long	sys_llseek
+	.long	sys_getpid		/* 20 */
+	.long	sys_mount
+	.long	sys_umount
+	.long	sys_setuid
+	.long	sys_getuid
+	.long	sys_stime		/* 25 */
+	.long	sys_ptrace
+	.long	sys_alarm
+	.long	sys_pause
+	.long	sys_utime
+	.long	sys_newstat		/* 30 */
+	.long	sys_newfstat
+	.long	sys_newlstat
+	.long	sys_access
+	.long	sys_chroot
+	.long	sys_sync		/* 35 */
+	.long	sys_fsync
+	.long	sys_kill
+	.long	sys_rename
+	.long	sys_mkdir
+	.long	sys_rmdir		/* 40 */
+	.long	sys_dup
+	.long	sys_pipe
+	.long	sys_times
+	.long	__sys_clone
+	.long	sys_brk			/* 45 */
+	.long	sys_setgid
+	.long	sys_getgid
+	.long	sys_getcwd
+	.long	sys_geteuid
+	.long	sys_getegid		/* 50 */
+	.long	sys_acct
+	.long	sys_setfsuid
+	.long	sys_setfsgid
+	.long	sys_ioctl
+	.long	sys_fcntl		/* 55 */
+	.long	sys_setpgid
+	.long	sys_mremap
+	.long	sys_setresuid
+	.long	sys_getresuid
+	.long	sys_setreuid		/* 60 */
+	.long	sys_setregid
+	.long	sys_ustat
+	.long	sys_dup2
+	.long	sys_getppid
+	.long	sys_getpgrp		/* 65 */
+	.long	sys_setsid
+	.long	sys_rt_sigaction
+	.long	__sys_rt_sigreturn
+	.long	sys_rt_sigprocmask
+	.long	sys_rt_sigpending	/* 70 */
+	.long	sys_rt_sigtimedwait
+	.long	sys_rt_sigqueueinfo
+	.long	__sys_rt_sigsuspend
+	.long	sys_sethostname
+	.long	sys_setrlimit		/* 75 */
+	.long	sys_getrlimit
+	.long	sys_getrusage
+	.long	sys_gettimeofday
+	.long	sys_settimeofday
+	.long	sys_getgroups		/* 80 */
+	.long	sys_setgroups
+	.long	sys_select
+	.long	sys_symlink
+	.long	sys_fchdir
+	.long	sys_readlink		/* 85 */
+	.long	sys_pread64
+	.long	sys_pwrite64
+	.long	sys_swapon
+	.long	sys_reboot
+	.long	__sys_mmap2		/* 90 */
+	.long	sys_munmap
+	.long	sys_truncate
+	.long	sys_ftruncate
+	.long	sys_fchmod
+	.long	sys_fchown		/* 95 */
+	.long	sys_getpriority
+	.long	sys_setpriority
+	.long	sys_wait4
+	.long	sys_statfs
+	.long	sys_fstatfs		/* 100 */
+	.long	sys_vhangup
+	.long	__sys_sigaltstack
+	.long	sys_syslog
+	.long	sys_setitimer
+	.long	sys_getitimer		/* 105 */
+	.long	sys_swapoff
+	.long	sys_sysinfo
+	.long	sys_ipc
+	.long	sys_sendfile
+	.long	sys_setdomainname	/* 110 */
+	.long	sys_newuname
+	.long	sys_adjtimex
+	.long	sys_mprotect
+	.long	__sys_vfork
+	.long	sys_init_module		/* 115 */
+	.long	sys_delete_module
+	.long	sys_quotactl
+	.long	sys_getpgid
+	.long	sys_bdflush
+	.long	sys_sysfs		/* 120 */
+	.long	sys_personality
+	.long	sys_ni_syscall		/* reserved for afs_syscall */
+	.long	sys_getdents
+	.long	sys_flock
+	.long	sys_msync		/* 125 */
+	.long	sys_readv
+	.long	sys_writev
+	.long	sys_getsid
+	.long	sys_fdatasync
+	.long	sys_sysctl		/* 130 */
+	.long	sys_mlock
+	.long	sys_munlock
+	.long	sys_mlockall
+	.long	sys_munlockall
+	.long	sys_sched_setparam		/* 135 */
+	.long	sys_sched_getparam
+	.long	sys_sched_setscheduler
+	.long	sys_sched_getscheduler
+	.long	sys_sched_yield
+	.long	sys_sched_get_priority_max	/* 140 */
+	.long	sys_sched_get_priority_min
+	.long	sys_sched_rr_get_interval
+	.long	sys_nanosleep
+	.long	sys_poll
+	.long	sys_nfsservctl		/* 145 */
+	.long	sys_setresgid
+	.long	sys_getresgid
+	.long	sys_prctl
+	.long	sys_socket
+	.long	sys_bind		/* 150 */
+	.long	sys_connect
+	.long	sys_listen
+	.long	sys_accept
+	.long	sys_getsockname
+	.long	sys_getpeername		/* 155 */
+	.long	sys_socketpair
+	.long	sys_send
+	.long	sys_recv
+	.long	__sys_sendto
+	.long	__sys_recvfrom		/* 160 */
+	.long	sys_shutdown
+	.long	sys_setsockopt
+	.long	sys_getsockopt
+	.long	sys_sendmsg
+	.long	sys_recvmsg		/* 165 */
+	.long	sys_truncate64
+	.long	sys_ftruncate64
+	.long	sys_stat64
+	.long	sys_lstat64
+	.long	sys_fstat64		/* 170 */
+	.long	sys_pivot_root
+	.long	sys_mincore
+	.long	sys_madvise
+	.long	sys_getdents64
+	.long	sys_fcntl64		/* 175 */
+	.long	sys_gettid
+	.long	sys_readahead
+	.long	sys_setxattr
+	.long	sys_lsetxattr
+	.long	sys_fsetxattr		/* 180 */
+	.long	sys_getxattr
+	.long	sys_lgetxattr
+	.long	sys_fgetxattr
+	.long	sys_listxattr
+	.long	sys_llistxattr		/* 185 */
+	.long	sys_flistxattr
+	.long	sys_removexattr
+	.long	sys_lremovexattr
+	.long	sys_fremovexattr
+	.long	sys_tkill		/* 190 */
+	.long	sys_sendfile64
+	.long	sys_futex
+	.long	sys_sched_setaffinity
+	.long	sys_sched_getaffinity
+	.long	sys_capget		/* 195 */
+	.long	sys_capset
+	.long	sys_io_setup
+	.long	sys_io_destroy
+	.long	sys_io_getevents
+	.long	sys_io_submit		/* 200 */
+	.long	sys_io_cancel
+	.long	sys_fadvise64
+	.long	sys_exit_group
+	.long	sys_lookup_dcookie
+	.long	sys_epoll_create	/* 205 */
+	.long	sys_epoll_ctl
+	.long	sys_epoll_wait
+	.long	sys_remap_file_pages
+	.long	sys_set_tid_address
+	.long	sys_timer_create	/* 210 */
+	.long	sys_timer_settime
+	.long	sys_timer_gettime
+	.long	sys_timer_getoverrun
+	.long	sys_timer_delete
+	.long	sys_clock_settime	/* 215 */
+	.long	sys_clock_gettime
+	.long	sys_clock_getres
+	.long	sys_clock_nanosleep
+	.long	sys_statfs64
+	.long	sys_fstatfs64		/* 220 */
+	.long	sys_tgkill
+	.long	sys_ni_syscall		/* reserved for TUX */
+	.long	sys_utimes
+	.long	sys_fadvise64_64
+	.long	sys_cacheflush		/* 225 */
+	.long	sys_ni_syscall		/* sys_vserver */
+	.long	sys_mq_open
+	.long	sys_mq_unlink
+	.long	sys_mq_timedsend
+	.long	sys_mq_timedreceive	/* 230 */
+	.long	sys_mq_notify
+	.long	sys_mq_getsetattr
+	.long	sys_kexec_load
+	.long	sys_waitid
+	.long	sys_add_key		/* 235 */
+	.long	sys_request_key
+	.long	sys_keyctl
+	.long	sys_ioprio_set
+	.long	sys_ioprio_get
+	.long	sys_inotify_init	/* 240 */
+	.long	sys_inotify_add_watch
+	.long	sys_inotify_rm_watch
+	.long	sys_openat
+	.long	sys_mkdirat
+	.long	sys_mknodat		/* 245 */
+	.long	sys_fchownat
+	.long	sys_futimesat
+	.long	sys_fstatat64
+	.long	sys_unlinkat
+	.long	sys_renameat		/* 250 */
+	.long	sys_linkat
+	.long	sys_symlinkat
+	.long	sys_readlinkat
+	.long	sys_fchmodat
+	.long	sys_faccessat		/* 255 */
+	.long	__sys_pselect6
+	.long	sys_ppoll
+	.long	sys_unshare
+	.long	sys_set_robust_list
+	.long	sys_get_robust_list	/* 260 */
+	.long	__sys_splice
+	.long	sys_sync_file_range
+	.long	sys_tee
+	.long	sys_vmsplice
+	.long	sys_ni_syscall		/* r8 is saturated at nr_syscalls */
diff --git a/arch/avr32/kernel/time.c b/arch/avr32/kernel/time.c
new file mode 100644
index 000000000000..b0e6b5855a38
--- /dev/null
+++ b/arch/avr32/kernel/time.c
@@ -0,0 +1,238 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on MIPS implementation arch/mips/kernel/time.c
+ *   Copyright 2001 MontaVista Software Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/clocksource.h>
+#include <linux/time.h>
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/kernel_stat.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/profile.h>
+#include <linux/sysdev.h>
+
+#include <asm/div64.h>
+#include <asm/sysreg.h>
+#include <asm/io.h>
+#include <asm/sections.h>
+
+static cycle_t read_cycle_count(void)
+{
+	return (cycle_t)sysreg_read(COUNT);
+}
+
+static struct clocksource clocksource_avr32 = {
+	.name		= "avr32",
+	.rating		= 350,
+	.read		= read_cycle_count,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.shift		= 16,
+	.is_continuous	= 1,
+};
+
+/*
+ * By default we provide the null RTC ops
+ */
+static unsigned long null_rtc_get_time(void)
+{
+	return mktime(2004, 1, 1, 0, 0, 0);
+}
+
+static int null_rtc_set_time(unsigned long sec)
+{
+	return 0;
+}
+
+static unsigned long (*rtc_get_time)(void) = null_rtc_get_time;
+static int (*rtc_set_time)(unsigned long) = null_rtc_set_time;
+
+/* how many counter cycles in a jiffy? */
+static unsigned long cycles_per_jiffy;
+
+/* cycle counter value at the previous timer interrupt */
+static unsigned int timerhi, timerlo;
+
+/* the count value for the next timer interrupt */
+static unsigned int expirelo;
+
+static void avr32_timer_ack(void)
+{
+	unsigned int count;
+
+	/* Ack this timer interrupt and set the next one */
+	expirelo += cycles_per_jiffy;
+	if (expirelo == 0) {
+		printk(KERN_DEBUG "expirelo == 0\n");
+		sysreg_write(COMPARE, expirelo + 1);
+	} else {
+		sysreg_write(COMPARE, expirelo);
+	}
+
+	/* Check to see if we have missed any timer interrupts */
+	count = sysreg_read(COUNT);
+	if ((count - expirelo) < 0x7fffffff) {
+		expirelo = count + cycles_per_jiffy;
+		sysreg_write(COMPARE, expirelo);
+	}
+}
+
+static unsigned int avr32_hpt_read(void)
+{
+	return sysreg_read(COUNT);
+}
+
+/*
+ * Taken from MIPS c0_hpt_timer_init().
+ *
+ * Why is it so complicated, and what is "count"?  My assumption is
+ * that `count' specifies the "reference cycle", i.e. the cycle since
+ * reset that should mean "zero". The reason COUNT is written twice is
+ * probably to make sure we don't get any timer interrupts while we
+ * are messing with the counter.
+ */
+static void avr32_hpt_init(unsigned int count)
+{
+	count = sysreg_read(COUNT) - count;
+	expirelo = (count / cycles_per_jiffy + 1) * cycles_per_jiffy;
+	sysreg_write(COUNT, expirelo - cycles_per_jiffy);
+	sysreg_write(COMPARE, expirelo);
+	sysreg_write(COUNT, count);
+}
+
+/*
+ * Scheduler clock - returns current time in nanosec units.
+ */
+unsigned long long sched_clock(void)
+{
+	/* There must be better ways...? */
+	return (unsigned long long)jiffies * (1000000000 / HZ);
+}
+
+/*
+ * local_timer_interrupt() does profiling and process accounting on a
+ * per-CPU basis.
+ *
+ * In UP mode, it is invoked from the (global) timer_interrupt.
+ */
+static void local_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	if (current->pid)
+		profile_tick(CPU_PROFILING, regs);
+	update_process_times(user_mode(regs));
+}
+
+static irqreturn_t
+timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	unsigned int count;
+
+	/* ack timer interrupt and try to set next interrupt */
+	count = avr32_hpt_read();
+	avr32_timer_ack();
+
+	/* Update timerhi/timerlo for intra-jiffy calibration */
+	timerhi += count < timerlo;	/* Wrap around */
+	timerlo = count;
+
+	/*
+	 * Call the generic timer interrupt handler
+	 */
+	write_seqlock(&xtime_lock);
+	do_timer(regs);
+	write_sequnlock(&xtime_lock);
+
+	/*
+	 * In UP mode, we call local_timer_interrupt() to do profiling
+	 * and process accounting.
+	 *
+	 * SMP is not supported yet.
+	 */
+	local_timer_interrupt(irq, dev_id, regs);
+
+	return IRQ_HANDLED;
+}
+
+static struct irqaction timer_irqaction = {
+	.handler	= timer_interrupt,
+	.flags		= IRQF_DISABLED,
+	.name		= "timer",
+};
+
+void __init time_init(void)
+{
+	unsigned long mult, shift, count_hz;
+	int ret;
+
+	xtime.tv_sec = rtc_get_time();
+	xtime.tv_nsec = 0;
+
+	set_normalized_timespec(&wall_to_monotonic,
+				-xtime.tv_sec, -xtime.tv_nsec);
+
+	printk("Before time_init: count=%08lx, compare=%08lx\n",
+	       (unsigned long)sysreg_read(COUNT),
+	       (unsigned long)sysreg_read(COMPARE));
+
+	count_hz = clk_get_rate(boot_cpu_data.clk);
+	shift = clocksource_avr32.shift;
+	mult = clocksource_hz2mult(count_hz, shift);
+	clocksource_avr32.mult = mult;
+
+	printk("Cycle counter: mult=%lu, shift=%lu\n", mult, shift);
+
+	{
+		u64 tmp;
+
+		tmp = TICK_NSEC;
+		tmp <<= shift;
+		tmp += mult / 2;
+		do_div(tmp, mult);
+
+		cycles_per_jiffy = tmp;
+	}
+
+	/* This sets up the high precision timer for the first interrupt. */
+	avr32_hpt_init(avr32_hpt_read());
+
+	printk("After time_init: count=%08lx, compare=%08lx\n",
+	       (unsigned long)sysreg_read(COUNT),
+	       (unsigned long)sysreg_read(COMPARE));
+
+	ret = clocksource_register(&clocksource_avr32);
+	if (ret)
+		printk(KERN_ERR
+		       "timer: could not register clocksource: %d\n", ret);
+
+	ret = setup_irq(0, &timer_irqaction);
+	if (ret)
+		printk("timer: could not request IRQ 0: %d\n", ret);
+}
+
+static struct sysdev_class timer_class = {
+	set_kset_name("timer"),
+};
+
+static struct sys_device timer_device = {
+	.id	= 0,
+	.cls	= &timer_class,
+};
+
+static int __init init_timer_sysfs(void)
+{
+	int err = sysdev_class_register(&timer_class);
+	if (!err)
+		err = sysdev_register(&timer_device);
+	return err;
+}
+
+device_initcall(init_timer_sysfs);
diff --git a/arch/avr32/kernel/traps.c b/arch/avr32/kernel/traps.c
new file mode 100644
index 000000000000..7e803f4d7a12
--- /dev/null
+++ b/arch/avr32/kernel/traps.c
@@ -0,0 +1,425 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#undef DEBUG
+#include <linux/sched.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/notifier.h>
+
+#include <asm/traps.h>
+#include <asm/sysreg.h>
+#include <asm/addrspace.h>
+#include <asm/ocd.h>
+#include <asm/mmu_context.h>
+#include <asm/uaccess.h>
+
+static void dump_mem(const char *str, unsigned long bottom, unsigned long top)
+{
+	unsigned long p;
+	int i;
+
+	printk("%s(0x%08lx to 0x%08lx)\n", str, bottom, top);
+
+	for (p = bottom & ~31; p < top; ) {
+		printk("%04lx: ", p & 0xffff);
+
+		for (i = 0; i < 8; i++, p += 4) {
+			unsigned int val;
+
+			if (p < bottom || p >= top)
+				printk("         ");
+			else {
+				if (__get_user(val, (unsigned int __user *)p)) {
+					printk("\n");
+					goto out;
+				}
+				printk("%08x ", val);
+			}
+		}
+		printk("\n");
+	}
+
+out:
+	return;
+}
+
+#ifdef CONFIG_FRAME_POINTER
+static inline void __show_trace(struct task_struct *tsk, unsigned long *sp,
+				struct pt_regs *regs)
+{
+	unsigned long __user *fp;
+	unsigned long __user *last_fp = NULL;
+
+	if (regs) {
+		fp = (unsigned long __user *)regs->r7;
+	} else if (tsk == current) {
+		register unsigned long __user *real_fp __asm__("r7");
+		fp = real_fp;
+	} else {
+		fp = (unsigned long __user *)tsk->thread.cpu_context.r7;
+	}
+
+	/*
+	 * Walk the stack until (a) we get an exception, (b) the frame
+	 * pointer becomes zero, or (c) the frame pointer gets stuck
+	 * at the same value.
+	 */
+	while (fp && fp != last_fp) {
+		unsigned long lr, new_fp = 0;
+
+		last_fp = fp;
+		if (__get_user(lr, fp))
+			break;
+		if (fp && __get_user(new_fp, fp + 1))
+			break;
+		fp = (unsigned long __user *)new_fp;
+
+		printk(" [<%08lx>] ", lr);
+		print_symbol("%s\n", lr);
+	}
+	printk("\n");
+}
+#else
+static inline void __show_trace(struct task_struct *tsk, unsigned long *sp,
+				struct pt_regs *regs)
+{
+	unsigned long addr;
+
+	while (!kstack_end(sp)) {
+		addr = *sp++;
+		if (kernel_text_address(addr)) {
+			printk(" [<%08lx>] ", addr);
+			print_symbol("%s\n", addr);
+		}
+	}
+}
+#endif
+
+void show_trace(struct task_struct *tsk, unsigned long *sp,
+		       struct pt_regs *regs)
+{
+	if (regs &&
+	    (((regs->sr & MODE_MASK) == MODE_EXCEPTION) ||
+	     ((regs->sr & MODE_MASK) == MODE_USER)))
+		return;
+
+	printk ("Call trace:");
+#ifdef CONFIG_KALLSYMS
+	printk("\n");
+#endif
+
+	__show_trace(tsk, sp, regs);
+	printk("\n");
+}
+
+void show_stack(struct task_struct *tsk, unsigned long *sp)
+{
+	unsigned long stack;
+
+	if (!tsk)
+		tsk = current;
+	if (sp == 0) {
+		if (tsk == current) {
+			register unsigned long *real_sp __asm__("sp");
+			sp = real_sp;
+		} else {
+			sp = (unsigned long *)tsk->thread.cpu_context.ksp;
+		}
+	}
+
+	stack = (unsigned long)sp;
+	dump_mem("Stack: ", stack,
+		 THREAD_SIZE + (unsigned long)tsk->thread_info);
+	show_trace(tsk, sp, NULL);
+}
+
+void dump_stack(void)
+{
+	show_stack(NULL, NULL);
+}
+EXPORT_SYMBOL(dump_stack);
+
+ATOMIC_NOTIFIER_HEAD(avr32_die_chain);
+
+int register_die_notifier(struct notifier_block *nb)
+{
+	pr_debug("register_die_notifier: %p\n", nb);
+
+	return atomic_notifier_chain_register(&avr32_die_chain, nb);
+}
+EXPORT_SYMBOL(register_die_notifier);
+
+int unregister_die_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&avr32_die_chain, nb);
+}
+EXPORT_SYMBOL(unregister_die_notifier);
+
+static DEFINE_SPINLOCK(die_lock);
+
+void __die(const char *str, struct pt_regs *regs, unsigned long err,
+	   const char *file, const char *func, unsigned long line)
+{
+	struct task_struct *tsk = current;
+	static int die_counter;
+
+	console_verbose();
+	spin_lock_irq(&die_lock);
+	bust_spinlocks(1);
+
+	printk(KERN_ALERT "%s", str);
+	if (file && func)
+		printk(" in %s:%s, line %ld", file, func, line);
+	printk("[#%d]:\n", ++die_counter);
+	print_modules();
+	show_regs(regs);
+	printk("Process %s (pid: %d, stack limit = 0x%p)\n",
+	       tsk->comm, tsk->pid, tsk->thread_info + 1);
+
+	if (!user_mode(regs) || in_interrupt()) {
+		dump_mem("Stack: ", regs->sp,
+			 THREAD_SIZE + (unsigned long)tsk->thread_info);
+	}
+
+	bust_spinlocks(0);
+	spin_unlock_irq(&die_lock);
+	do_exit(SIGSEGV);
+}
+
+void __die_if_kernel(const char *str, struct pt_regs *regs, unsigned long err,
+		     const char *file, const char *func, unsigned long line)
+{
+	if (!user_mode(regs))
+		__die(str, regs, err, file, func, line);
+}
+
+asmlinkage void do_nmi(unsigned long ecr, struct pt_regs *regs)
+{
+#ifdef CONFIG_SUBARCH_AVR32B
+	/*
+	 * The exception entry always saves RSR_EX. For NMI, this is
+	 * wrong; it should be RSR_NMI
+	 */
+	regs->sr = sysreg_read(RSR_NMI);
+#endif
+
+	printk("NMI taken!!!!\n");
+	die("NMI", regs, ecr);
+	BUG();
+}
+
+asmlinkage void do_critical_exception(unsigned long ecr, struct pt_regs *regs)
+{
+	printk("Unable to handle critical exception %lu at pc = %08lx!\n",
+	       ecr, regs->pc);
+	die("Oops", regs, ecr);
+	BUG();
+}
+
+asmlinkage void do_address_exception(unsigned long ecr, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	die_if_kernel("Oops: Address exception in kernel mode", regs, ecr);
+
+#ifdef DEBUG
+	if (ecr == ECR_ADDR_ALIGN_X)
+		pr_debug("Instruction Address Exception at pc = %08lx\n",
+			 regs->pc);
+	else if (ecr == ECR_ADDR_ALIGN_R)
+		pr_debug("Data Address Exception (Read) at pc = %08lx\n",
+			 regs->pc);
+	else if (ecr == ECR_ADDR_ALIGN_W)
+		pr_debug("Data Address Exception (Write) at pc = %08lx\n",
+			 regs->pc);
+	else
+		BUG();
+
+	show_regs(regs);
+#endif
+
+	info.si_signo = SIGBUS;
+	info.si_errno = 0;
+	info.si_code = BUS_ADRALN;
+	info.si_addr = (void __user *)regs->pc;
+
+	force_sig_info(SIGBUS, &info, current);
+}
+
+/* This way of handling undefined instructions is stolen from ARM */
+static LIST_HEAD(undef_hook);
+static spinlock_t undef_lock = SPIN_LOCK_UNLOCKED;
+
+void register_undef_hook(struct undef_hook *hook)
+{
+	spin_lock_irq(&undef_lock);
+	list_add(&hook->node, &undef_hook);
+	spin_unlock_irq(&undef_lock);
+}
+
+void unregister_undef_hook(struct undef_hook *hook)
+{
+	spin_lock_irq(&undef_lock);
+	list_del(&hook->node);
+	spin_unlock_irq(&undef_lock);
+}
+
+static int do_cop_absent(u32 insn)
+{
+	int cop_nr;
+	u32 cpucr;
+	if ( (insn & 0xfdf00000) == 0xf1900000 )
+		/* LDC0 */
+		cop_nr = 0;
+	else
+		cop_nr = (insn >> 13) & 0x7;
+
+	/* Try enabling the coprocessor */
+	cpucr = sysreg_read(CPUCR);
+	cpucr |= (1 << (24 + cop_nr));
+	sysreg_write(CPUCR, cpucr);
+
+	cpucr = sysreg_read(CPUCR);
+	if ( !(cpucr & (1 << (24 + cop_nr))) ){
+		printk("Coprocessor #%i not found!\n", cop_nr);
+		return -1;
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_BUG
+#ifdef CONFIG_DEBUG_BUGVERBOSE
+static inline void do_bug_verbose(struct pt_regs *regs, u32 insn)
+{
+	char *file;
+	u16 line;
+	char c;
+
+	if (__get_user(line, (u16 __user *)(regs->pc + 2)))
+		return;
+	if (__get_user(file, (char * __user *)(regs->pc + 4))
+	    || (unsigned long)file < PAGE_OFFSET
+	    || __get_user(c, file))
+		file = "<bad filename>";
+
+	printk(KERN_ALERT "kernel BUG at %s:%d!\n", file, line);
+}
+#else
+static inline void do_bug_verbose(struct pt_regs *regs, u32 insn)
+{
+
+}
+#endif
+#endif
+
+asmlinkage void do_illegal_opcode(unsigned long ecr, struct pt_regs *regs)
+{
+	u32 insn;
+	struct undef_hook *hook;
+	siginfo_t info;
+	void __user *pc;
+
+	if (!user_mode(regs))
+		goto kernel_trap;
+
+	local_irq_enable();
+
+	pc = (void __user *)instruction_pointer(regs);
+	if (__get_user(insn, (u32 __user *)pc))
+		goto invalid_area;
+
+        if (ecr == ECR_COPROC_ABSENT) {
+		if (do_cop_absent(insn) == 0)
+			return;
+        }
+
+	spin_lock_irq(&undef_lock);
+	list_for_each_entry(hook, &undef_hook, node) {
+		if ((insn & hook->insn_mask) == hook->insn_val) {
+			if (hook->fn(regs, insn) == 0) {
+				spin_unlock_irq(&undef_lock);
+				return;
+			}
+		}
+	}
+	spin_unlock_irq(&undef_lock);
+
+invalid_area:
+
+#ifdef DEBUG
+	printk("Illegal instruction at pc = %08lx\n", regs->pc);
+	if (regs->pc < TASK_SIZE) {
+		unsigned long ptbr, pgd, pte, *p;
+
+		ptbr = sysreg_read(PTBR);
+		p = (unsigned long *)ptbr;
+		pgd = p[regs->pc >> 22];
+		p = (unsigned long *)((pgd & 0x1ffff000) | 0x80000000);
+		pte = p[(regs->pc >> 12) & 0x3ff];
+		printk("page table: 0x%08lx -> 0x%08lx -> 0x%08lx\n", ptbr, pgd, pte);
+	}
+#endif
+
+	info.si_signo = SIGILL;
+	info.si_errno = 0;
+	info.si_addr = (void __user *)regs->pc;
+	switch (ecr) {
+	case ECR_ILLEGAL_OPCODE:
+	case ECR_UNIMPL_INSTRUCTION:
+		info.si_code = ILL_ILLOPC;
+		break;
+	case ECR_PRIVILEGE_VIOLATION:
+		info.si_code = ILL_PRVOPC;
+		break;
+	case ECR_COPROC_ABSENT:
+		info.si_code = ILL_COPROC;
+		break;
+	default:
+		BUG();
+	}
+
+	force_sig_info(SIGILL, &info, current);
+	return;
+
+kernel_trap:
+#ifdef CONFIG_BUG
+	if (__kernel_text_address(instruction_pointer(regs))) {
+		insn = *(u16 *)instruction_pointer(regs);
+		if (insn == AVR32_BUG_OPCODE) {
+			do_bug_verbose(regs, insn);
+			die("Kernel BUG", regs, 0);
+			return;
+		}
+	}
+#endif
+
+	die("Oops: Illegal instruction in kernel code", regs, ecr);
+}
+
+asmlinkage void do_fpe(unsigned long ecr, struct pt_regs *regs)
+{
+	siginfo_t info;
+
+	printk("Floating-point exception at pc = %08lx\n", regs->pc);
+
+	/* We have no FPU... */
+	info.si_signo = SIGILL;
+	info.si_errno = 0;
+	info.si_addr = (void __user *)regs->pc;
+	info.si_code = ILL_COPROC;
+
+	force_sig_info(SIGILL, &info, current);
+}
+
+
+void __init trap_init(void)
+{
+
+}
diff --git a/arch/avr32/kernel/vmlinux.lds.c b/arch/avr32/kernel/vmlinux.lds.c
new file mode 100644
index 000000000000..cdd627c6b7dc
--- /dev/null
+++ b/arch/avr32/kernel/vmlinux.lds.c
@@ -0,0 +1,139 @@
+/*
+ * AVR32 linker script for the Linux kernel
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#define LOAD_OFFSET 0x00000000
+#include <asm-generic/vmlinux.lds.h>
+
+OUTPUT_FORMAT("elf32-avr32", "elf32-avr32", "elf32-avr32")
+OUTPUT_ARCH(avr32)
+ENTRY(_start)
+
+/* Big endian */
+jiffies = jiffies_64 + 4;
+
+SECTIONS
+{
+	. = CONFIG_ENTRY_ADDRESS;
+	.init		: AT(ADDR(.init) - LOAD_OFFSET) {
+		_stext = .;
+		__init_begin = .;
+			_sinittext = .;
+			*(.text.reset)
+			*(.init.text)
+			_einittext = .;
+		. = ALIGN(4);
+		__tagtable_begin = .;
+			*(.taglist)
+		__tagtable_end = .;
+			*(.init.data)
+		. = ALIGN(16);
+		__setup_start = .;
+			*(.init.setup)
+		__setup_end = .;
+		. = ALIGN(4);
+		__initcall_start = .;
+			*(.initcall1.init)
+			*(.initcall2.init)
+			*(.initcall3.init)
+			*(.initcall4.init)
+			*(.initcall5.init)
+			*(.initcall6.init)
+			*(.initcall7.init)
+		__initcall_end = .;
+		__con_initcall_start = .;
+			*(.con_initcall.init)
+		__con_initcall_end = .;
+		__security_initcall_start = .;
+			*(.security_initcall.init)
+		__security_initcall_end = .;
+		. = ALIGN(32);
+		__initramfs_start = .;
+			*(.init.ramfs)
+		__initramfs_end = .;
+		. = ALIGN(4096);
+		__init_end = .;
+	}
+
+	. = ALIGN(8192);
+	.text		: AT(ADDR(.text) - LOAD_OFFSET) {
+		_evba = .;
+		_text = .;
+		*(.ex.text)
+		. = 0x50;
+		*(.tlbx.ex.text)
+		. = 0x60;
+		*(.tlbr.ex.text)
+		. = 0x70;
+		*(.tlbw.ex.text)
+		. = 0x100;
+		*(.scall.text)
+		*(.irq.text)
+		*(.text)
+		SCHED_TEXT
+		LOCK_TEXT
+		KPROBES_TEXT
+		*(.fixup)
+		*(.gnu.warning)
+		_etext = .;
+	} = 0xd703d703
+
+	. = ALIGN(4);
+	__ex_table	: AT(ADDR(__ex_table) - LOAD_OFFSET) {
+		__start___ex_table = .;
+		*(__ex_table)
+		__stop___ex_table = .;
+	}
+
+	RODATA
+
+	. = ALIGN(8192);
+
+	.data		: AT(ADDR(.data) - LOAD_OFFSET) {
+		_data = .;
+		_sdata = .;
+		/*
+		 * First, the init task union, aligned to an 8K boundary.
+		 */
+		*(.data.init_task)
+
+		/* Then, the cacheline aligned data */
+		. = ALIGN(32);
+		*(.data.cacheline_aligned)
+
+		/* And the rest... */
+		*(.data.rel*)
+		*(.data)
+		CONSTRUCTORS
+
+		_edata = .;
+	}
+
+
+	. = ALIGN(8);
+	.bss    	: AT(ADDR(.bss) - LOAD_OFFSET) {
+		__bss_start = .;
+		*(.bss)
+		*(COMMON)
+		. = ALIGN(8);
+		__bss_stop = .;
+		_end = .;
+	}
+
+	/* When something in the kernel is NOT compiled as a module, the module
+	 * cleanup code and data are put into these segments. Both can then be
+	 * thrown away, as cleanup code is never called unless it's a module.
+	 */
+	/DISCARD/       	: {
+		*(.exit.text)
+		*(.exit.data)
+		*(.exitcall.exit)
+	}
+
+	DWARF_DEBUG
+}
diff --git a/arch/avr32/lib/Makefile b/arch/avr32/lib/Makefile
new file mode 100644
index 000000000000..09ac43e40522
--- /dev/null
+++ b/arch/avr32/lib/Makefile
@@ -0,0 +1,10 @@
+#
+# Makefile for AVR32-specific library files
+#
+
+lib-y	:= copy_user.o clear_user.o
+lib-y	+= strncpy_from_user.o strnlen_user.o
+lib-y	+= delay.o memset.o memcpy.o findbit.o
+lib-y	+= csum_partial.o csum_partial_copy_generic.o
+lib-y	+= io-readsw.o io-readsl.o io-writesw.o io-writesl.o
+lib-y	+= __avr32_lsl64.o __avr32_lsr64.o __avr32_asr64.o
diff --git a/arch/avr32/lib/__avr32_asr64.S b/arch/avr32/lib/__avr32_asr64.S
new file mode 100644
index 000000000000..368b6bca4c76
--- /dev/null
+++ b/arch/avr32/lib/__avr32_asr64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_asr64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_asr64
+	.type	__avr32_asr64,@function
+__avr32_asr64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsl	r8, r11, r9
+	lsr	r10, r10, r12
+	asr	r11, r11, r12
+	or	r10, r8
+	retal	r12
+
+1:	neg	r9
+	asr	r10, r11, r9
+	asr	r11, 31
+	retal	r12
diff --git a/arch/avr32/lib/__avr32_lsl64.S b/arch/avr32/lib/__avr32_lsl64.S
new file mode 100644
index 000000000000..f1dbc2b36257
--- /dev/null
+++ b/arch/avr32/lib/__avr32_lsl64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_lsl64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_lsl64
+	.type	__avr32_lsl64,@function
+__avr32_lsl64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsr	r8, r10, r9
+	lsl	r10, r10, r12
+	lsl	r11, r11, r12
+	or	r11, r8
+	retal	r12
+
+1:	neg	r9
+	lsl	r11, r10, r9
+	mov	r10, 0
+	retal	r12
diff --git a/arch/avr32/lib/__avr32_lsr64.S b/arch/avr32/lib/__avr32_lsr64.S
new file mode 100644
index 000000000000..e65bb7f0d24c
--- /dev/null
+++ b/arch/avr32/lib/__avr32_lsr64.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * DWtype __avr32_lsr64(DWtype u, word_type b)
+	 */
+	.text
+	.global	__avr32_lsr64
+	.type	__avr32_lsr64,@function
+__avr32_lsr64:
+	cp.w	r12, 0
+	reteq	r12
+
+	rsub	r9, r12, 32
+	brle	1f
+
+	lsl	r8, r11, r9
+	lsr	r11, r11, r12
+	lsr	r10, r10, r12
+	or	r10, r8
+	retal	r12
+
+1:	neg	r9
+	lsr	r10, r11, r9
+	mov	r11, 0
+	retal	r12
diff --git a/arch/avr32/lib/clear_user.S b/arch/avr32/lib/clear_user.S
new file mode 100644
index 000000000000..d8991b6f8eb7
--- /dev/null
+++ b/arch/avr32/lib/clear_user.S
@@ -0,0 +1,76 @@
+/*
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	.text
+	.align	1
+	.global	clear_user
+	.type	clear_user, "function"
+clear_user:
+	branch_if_kernel r8, __clear_user
+	ret_if_privileged r8, r12, r11, r11
+
+	.global	__clear_user
+	.type	__clear_user, "function"
+__clear_user:
+	mov	r9, r12
+	mov	r8, 0
+	andl	r9, 3, COH
+	brne	5f
+
+1:	sub	r11, 4
+	brlt	2f
+
+10:	st.w	r12++, r8
+	sub	r11, 4
+	brge	10b
+
+2:	sub	r11, -4
+	reteq	0
+
+	/* Unaligned count or address */
+	bld	r11, 1
+	brcc	12f
+11:	st.h	r12++, r8
+	sub	r11, 2
+	reteq	0
+12:	st.b	r12++, r8
+	retal	0
+
+	/* Unaligned address */
+5:	cp.w	r11, 4
+	brlt	2b
+
+	lsl	r9, 2
+	add	pc, pc, r9
+13:	st.b	r12++, r8
+	sub	r11, 1
+14:	st.b	r12++, r8
+	sub	r11, 1
+15:	st.b	r12++, r8
+	sub	r11, 1
+	rjmp	1b
+
+	.size	clear_user, . - clear_user
+	.size	__clear_user, . - __clear_user
+
+	.section .fixup, "ax"
+	.align	1
+18:	sub	r11, -4
+19:	retal	r11
+
+	.section __ex_table, "a"
+	.align	2
+	.long	10b, 18b
+	.long	11b, 19b
+	.long	12b, 19b
+	.long	13b, 19b
+	.long	14b, 19b
+	.long	15b, 19b
diff --git a/arch/avr32/lib/copy_user.S b/arch/avr32/lib/copy_user.S
new file mode 100644
index 000000000000..ea59c04b07de
--- /dev/null
+++ b/arch/avr32/lib/copy_user.S
@@ -0,0 +1,119 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	/*
+	 * __kernel_size_t
+	 * __copy_user(void *to, const void *from, __kernel_size_t n)
+	 *
+	 * Returns the number of bytes not copied. Might be off by
+	 * max 3 bytes if we get a fault in the main loop.
+	 *
+	 * The address-space checking functions simply fall through to
+	 * the non-checking version.
+	 */
+	.text
+	.align	1
+	.global	copy_from_user
+	.type	copy_from_user, @function
+copy_from_user:
+	branch_if_kernel r8, __copy_user
+	ret_if_privileged r8, r11, r10, r10
+	rjmp	__copy_user
+	.size	copy_from_user, . - copy_from_user
+
+	.global	copy_to_user
+	.type	copy_to_user, @function
+copy_to_user:
+	branch_if_kernel r8, __copy_user
+	ret_if_privileged r8, r12, r10, r10
+	.size	copy_to_user, . - copy_to_user
+
+	.global	__copy_user
+	.type	__copy_user, @function
+__copy_user:
+	mov	r9, r11
+	andl	r9, 3, COH
+	brne	6f
+
+	/* At this point, from is word-aligned */
+1:	sub	r10, 4
+	brlt	3f
+
+2:
+10:	ld.w	r8, r11++
+11:	st.w	r12++, r8
+	sub	r10, 4
+	brge	2b
+
+3:	sub	r10, -4
+	reteq	0
+
+	/*
+	 * Handle unaligned count. Need to be careful with r10 here so
+	 * that we return the correct value even if we get a fault
+	 */
+4:
+20:	ld.ub	r8, r11++
+21:	st.b	r12++, r8
+	sub	r10, 1
+	reteq	0
+22:	ld.ub	r8, r11++
+23:	st.b	r12++, r8
+	sub	r10, 1
+	reteq	0
+24:	ld.ub	r8, r11++
+25:	st.b	r12++, r8
+	retal	0
+
+	/* Handle unaligned from-pointer */
+6:	cp.w	r10, 4
+	brlt	4b
+	rsub	r9, r9, 4
+
+30:	ld.ub	r8, r11++
+31:	st.b	r12++, r8
+	sub	r10, 1
+	sub	r9, 1
+	breq	1b
+32:	ld.ub	r8, r11++
+33:	st.b	r12++, r8
+	sub	r10, 1
+	sub	r9, 1
+	breq	1b
+34:	ld.ub	r8, r11++
+35:	st.b	r12++, r8
+	sub	r10, 1
+	rjmp	1b
+	.size	__copy_user, . - __copy_user
+
+	.section .fixup,"ax"
+	.align	1
+19:	sub	r10, -4
+29:	retal	r10
+
+	.section __ex_table,"a"
+	.align	2
+	.long	10b, 19b
+	.long	11b, 19b
+	.long	20b, 29b
+	.long	21b, 29b
+	.long	22b, 29b
+	.long	23b, 29b
+	.long	24b, 29b
+	.long	25b, 29b
+	.long	30b, 29b
+	.long	31b, 29b
+	.long	32b, 29b
+	.long	33b, 29b
+	.long	34b, 29b
+	.long	35b, 29b
diff --git a/arch/avr32/lib/csum_partial.S b/arch/avr32/lib/csum_partial.S
new file mode 100644
index 000000000000..6a262b528eb7
--- /dev/null
+++ b/arch/avr32/lib/csum_partial.S
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * unsigned int csum_partial(const unsigned char *buff,
+	 *			     int len, unsigned int sum)
+	 */
+	.text
+	.global	csum_partial
+	.type	csum_partial,"function"
+	.align	1
+csum_partial:
+	/* checksum complete words, aligned or not */
+3:	sub	r11, 4
+	brlt	5f
+4:	ld.w	r9, r12++
+	add	r10, r9
+	acr	r10
+	sub	r11, 4
+	brge	4b
+
+	/* return if we had a whole number of words */
+5:	sub	r11, -4
+	reteq	r10
+
+	/* checksum any remaining bytes at the end */
+	mov	r9, 0
+	mov	r8, 0
+	cp	r11, 2
+	brlt	6f
+	ld.uh	r9, r12++
+	sub	r11, 2
+	breq	7f
+	lsl	r9, 16
+6:	ld.ub	r8, r12++
+	lsl	r8, 8
+7:	or	r9, r8
+	add	r10, r9
+	acr	r10
+
+	retal	r10
+	.size	csum_partial, . - csum_partial
diff --git a/arch/avr32/lib/csum_partial_copy_generic.S b/arch/avr32/lib/csum_partial_copy_generic.S
new file mode 100644
index 000000000000..a3a0f9b8929c
--- /dev/null
+++ b/arch/avr32/lib/csum_partial_copy_generic.S
@@ -0,0 +1,99 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/errno.h>
+#include <asm/asm.h>
+
+	/*
+	 * unsigned int csum_partial_copy_generic(const char *src, char *dst, int len
+	 *					  int sum, int *src_err_ptr,
+	 *					  int *dst_err_ptr)
+	 *
+	 * Copy src to dst while checksumming, otherwise like csum_partial.
+	 */
+
+	.macro ld_src size, reg, ptr
+9999:	ld.\size \reg, \ptr
+	.section __ex_table, "a"
+	.long	9999b, fixup_ld_src
+	.previous
+	.endm
+
+	.macro st_dst size, ptr, reg
+9999:	st.\size \ptr, \reg
+	.section __ex_table, "a"
+	.long	9999b, fixup_st_dst
+	.previous
+	.endm
+
+	.text
+	.global	csum_partial_copy_generic
+	.type	csum_partial_copy_generic,"function"
+	.align	1
+csum_partial_copy_generic:
+	pushm	r4-r7,lr
+
+	/* The inner loop */
+1:	sub	r10, 4
+	brlt	5f
+2:	ld_src	w, r5, r12++
+	st_dst	w, r11++, r5
+	add	r9, r5
+	acr	r9
+	sub	r10, 4
+	brge	2b
+
+	/* return if we had a whole number of words */
+5:	sub	r10, -4
+	brne	7f
+
+6:	mov	r12, r9
+	popm	r4-r7,pc
+
+	/* handle additional bytes at the tail */
+7:	mov	r5, 0
+	mov	r4, 32
+8:	ld_src	ub, r6, r12++
+	st_dst	b, r11++, r6
+	lsl	r5, 8
+	sub	r4, 8
+	bfins	r5, r6, 0, 8
+	sub	r10, 1
+	brne	8b
+
+	lsl	r5, r5, r4
+	add	r9, r5
+	acr	r9
+	rjmp	6b
+
+	/* Exception handler */
+	.section .fixup,"ax"
+	.align	1
+fixup_ld_src:
+	mov	r9, -EFAULT
+	cp.w	r8, 0
+	breq	1f
+	st.w	r8[0], r9
+
+1:	/*
+	 * TODO: zero the complete destination - computing the rest
+	 * is too much work
+	 */
+
+	mov	r9, 0
+	rjmp	6b
+
+fixup_st_dst:
+	mov	r9, -EFAULT
+	lddsp	r8, sp[20]
+	cp.w	r8, 0
+	breq	1f
+	st.w	r8[0], r9
+1:	mov	r9, 0
+	rjmp	6b
+
+	.previous
diff --git a/arch/avr32/lib/delay.c b/arch/avr32/lib/delay.c
new file mode 100644
index 000000000000..462c8307b680
--- /dev/null
+++ b/arch/avr32/lib/delay.c
@@ -0,0 +1,55 @@
+/*
+ *      Precise Delay Loops for avr32
+ *
+ *      Copyright (C) 1993 Linus Torvalds
+ *      Copyright (C) 1997 Martin Mares <mj@atrey.karlin.mff.cuni.cz>
+ *	Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/delay.h>
+#include <linux/module.h>
+#include <linux/types.h>
+
+#include <asm/delay.h>
+#include <asm/processor.h>
+#include <asm/sysreg.h>
+
+int read_current_timer(unsigned long *timer_value)
+{
+	*timer_value = sysreg_read(COUNT);
+	return 0;
+}
+
+void __delay(unsigned long loops)
+{
+	unsigned bclock, now;
+
+	bclock = sysreg_read(COUNT);
+	do {
+		now = sysreg_read(COUNT);
+	} while ((now - bclock) < loops);
+}
+
+inline void __const_udelay(unsigned long xloops)
+{
+	unsigned long long loops;
+
+	asm("mulu.d %0, %1, %2"
+	    : "=r"(loops)
+	    : "r"(current_cpu_data.loops_per_jiffy * HZ), "r"(xloops));
+	__delay(loops >> 32);
+}
+
+void __udelay(unsigned long usecs)
+{
+	__const_udelay(usecs * 0x000010c7); /* 2**32 / 1000000 (rounded up) */
+}
+
+void __ndelay(unsigned long nsecs)
+{
+	__const_udelay(nsecs * 0x00005); /* 2**32 / 1000000000 (rounded up) */
+}
diff --git a/arch/avr32/lib/findbit.S b/arch/avr32/lib/findbit.S
new file mode 100644
index 000000000000..2b4856f4bf7c
--- /dev/null
+++ b/arch/avr32/lib/findbit.S
@@ -0,0 +1,154 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+
+	.text
+	/*
+	 * unsigned long find_first_zero_bit(const unsigned long *addr,
+	 *				     unsigned long size)
+	 */
+ENTRY(find_first_zero_bit)
+	cp.w	r11, 0
+	reteq	r11
+	mov	r9, r11
+1:	ld.w	r8, r12[0]
+	com	r8
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/*
+	 * unsigned long find_next_zero_bit(const unsigned long *addr,
+	 *				    unsigned long size,
+	 *				    unsigned long offset)
+	 */
+ENTRY(find_next_zero_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ld.w	r8, r12[0]
+	com	r8
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ld.w	r8, r12[0]
+	com	r8
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/* Common return path for when a bit is actually found. */
+.L_found:
+	brev	r8
+	clz	r10, r8
+	rsub	r9, r11
+	add	r10, r9
+
+	/* XXX: If we don't have to return exactly "size" when the bit
+	   is not found, we may drop this "min" thing */
+	min	r12, r11, r10
+	retal	r12
+
+	/*
+	 * unsigned long find_first_bit(const unsigned long *addr,
+	 *				unsigned long size)
+	 */
+ENTRY(find_first_bit)
+	cp.w	r11, 0
+	reteq	r11
+	mov	r9, r11
+1:	ld.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+	/*
+	 * unsigned long find_next_bit(const unsigned long *addr,
+	 *			       unsigned long size,
+	 *			       unsigned long offset)
+	 */
+ENTRY(find_next_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ld.w	r8, r12[0]
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ld.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
+
+ENTRY(generic_find_next_zero_le_bit)
+	lsr	r8, r10, 5
+	sub	r9, r11, r10
+	retle	r11
+
+	lsl	r8, 2
+	add	r12, r8
+	andl	r10, 31, COH
+	breq	1f
+
+	/* offset is not word-aligned. Handle the first (32 - r10) bits */
+	ldswp.w	r8, r12[0]
+	sub	r12, -4
+	lsr	r8, r8, r10
+	brne	.L_found
+
+	/* r9 = r9 - (32 - r10) = r9 + r10 - 32 */
+	add	r9, r10
+	sub	r9, 32
+	retle	r11
+
+	/* Main loop. offset must be word-aligned */
+1:	ldswp.w	r8, r12[0]
+	cp.w	r8, 0
+	brne	.L_found
+	sub	r12, -4
+	sub	r9, 32
+	brgt	1b
+	retal	r11
diff --git a/arch/avr32/lib/io-readsl.S b/arch/avr32/lib/io-readsl.S
new file mode 100644
index 000000000000..b103511ed6c4
--- /dev/null
+++ b/arch/avr32/lib/io-readsl.S
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	.global	__raw_readsl
+	.type	__raw_readsl,@function
+__raw_readsl:
+	cp.w	r10, 0
+	reteq	r12
+
+	/*
+	 * If r11 isn't properly aligned, we might get an exception on
+	 * some implementations. But there's not much we can do about it.
+	 */
+1:	ld.w	r8, r12[0]
+	sub	r10, 1
+	st.w	r11++, r8
+	brne	1b
+
+	retal	r12
diff --git a/arch/avr32/lib/io-readsw.S b/arch/avr32/lib/io-readsw.S
new file mode 100644
index 000000000000..456be9909027
--- /dev/null
+++ b/arch/avr32/lib/io-readsw.S
@@ -0,0 +1,43 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+.Lnot_word_aligned:
+	/*
+	 * Bad alignment will cause a hardware exception, which is as
+	 * good as anything. No need for us to check for proper alignment.
+	 */
+	ld.uh	r8, r12[0]
+	sub	r10, 1
+	st.h	r11++, r8
+
+	/* fall through */
+
+	.global	__raw_readsw
+	.type	__raw_readsw,@function
+__raw_readsw:
+	cp.w	r10, 0
+	reteq	r12
+	mov	r9, 3
+	tst	r11, r9
+	brne	.Lnot_word_aligned
+
+	sub	r10, 2
+	brlt	2f
+
+1:	ldins.h	r8:t, r12[0]
+	ldins.h	r8:b, r12[0]
+	st.w	r11++, r8
+	sub	r10, 2
+	brge	1b
+
+2:	sub	r10, -2
+	reteq	r12
+
+	ld.uh	r8, r12[0]
+	st.h	r11++, r8
+	retal	r12
diff --git a/arch/avr32/lib/io-writesl.S b/arch/avr32/lib/io-writesl.S
new file mode 100644
index 000000000000..22138b3a16e5
--- /dev/null
+++ b/arch/avr32/lib/io-writesl.S
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	.global	__raw_writesl
+	.type	__raw_writesl,@function
+__raw_writesl:
+	cp.w	r10, 0
+	reteq	r12
+
+1:	ld.w	r8, r11++
+	sub	r10, 1
+	st.w	r12[0], r8
+	brne	1b
+
+	retal	r12
diff --git a/arch/avr32/lib/io-writesw.S b/arch/avr32/lib/io-writesw.S
new file mode 100644
index 000000000000..8c4a53f1c52a
--- /dev/null
+++ b/arch/avr32/lib/io-writesw.S
@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+.Lnot_word_aligned:
+	ld.uh	r8, r11++
+	sub	r10, 1
+	st.h	r12[0], r8
+
+	.global	__raw_writesw
+	.type	__raw_writesw,@function
+__raw_writesw:
+	cp.w	r10, 0
+	mov	r9, 3
+	reteq	r12
+	tst	r11, r9
+	brne	.Lnot_word_aligned
+
+	sub	r10, 2
+	brlt	2f
+
+1:	ld.w	r8, r11++
+	bfextu	r9, r8, 16, 16
+	st.h	r12[0], r9
+	st.h	r12[0], r8
+	sub	r10, 2
+	brge	1b
+
+2:	sub	r10, -2
+	reteq	r12
+
+	ld.uh	r8, r11++
+	st.h	r12[0], r8
+	retal	r12
diff --git a/arch/avr32/lib/libgcc.h b/arch/avr32/lib/libgcc.h
new file mode 100644
index 000000000000..5a091b5e3618
--- /dev/null
+++ b/arch/avr32/lib/libgcc.h
@@ -0,0 +1,33 @@
+/* Definitions for various functions 'borrowed' from gcc-3.4.3 */
+
+#define BITS_PER_UNIT	8
+
+typedef		 int QItype	__attribute__ ((mode (QI)));
+typedef unsigned int UQItype	__attribute__ ((mode (QI)));
+typedef		 int HItype	__attribute__ ((mode (HI)));
+typedef unsigned int UHItype	__attribute__ ((mode (HI)));
+typedef 	 int SItype	__attribute__ ((mode (SI)));
+typedef unsigned int USItype	__attribute__ ((mode (SI)));
+typedef		 int DItype	__attribute__ ((mode (DI)));
+typedef unsigned int UDItype	__attribute__ ((mode (DI)));
+typedef 	float SFtype	__attribute__ ((mode (SF)));
+typedef		float DFtype	__attribute__ ((mode (DF)));
+typedef int word_type __attribute__ ((mode (__word__)));
+
+#define W_TYPE_SIZE (4 * BITS_PER_UNIT)
+#define Wtype	SItype
+#define UWtype	USItype
+#define HWtype	SItype
+#define UHWtype	USItype
+#define DWtype	DItype
+#define UDWtype	UDItype
+#define __NW(a,b)	__ ## a ## si ## b
+#define __NDW(a,b)	__ ## a ## di ## b
+
+struct DWstruct {Wtype high, low;};
+
+typedef union
+{
+  struct DWstruct s;
+  DWtype ll;
+} DWunion;
diff --git a/arch/avr32/lib/longlong.h b/arch/avr32/lib/longlong.h
new file mode 100644
index 000000000000..cd5e369ac437
--- /dev/null
+++ b/arch/avr32/lib/longlong.h
@@ -0,0 +1,98 @@
+/* longlong.h -- definitions for mixed size 32/64 bit arithmetic.
+   Copyright (C) 1991, 1992, 1994, 1995, 1996, 1997, 1998, 1999, 2000
+   Free Software Foundation, Inc.
+
+   This definition file is free software; you can redistribute it
+   and/or modify it under the terms of the GNU General Public
+   License as published by the Free Software Foundation; either
+   version 2, or (at your option) any later version.
+
+   This definition file is distributed in the hope that it will be
+   useful, but WITHOUT ANY WARRANTY; without even the implied
+   warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+   See the GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 59 Temple Place - Suite 330,
+   Boston, MA 02111-1307, USA.  */
+
+/* Borrowed from gcc-3.4.3 */
+
+#define __BITS4 (W_TYPE_SIZE / 4)
+#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
+#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
+#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
+
+#define count_leading_zeros(count, x) ((count) = __builtin_clz(x))
+
+#define __udiv_qrnnd_c(q, r, n1, n0, d) \
+  do {									\
+    UWtype __d1, __d0, __q1, __q0;					\
+    UWtype __r1, __r0, __m;						\
+    __d1 = __ll_highpart (d);						\
+    __d0 = __ll_lowpart (d);						\
+									\
+    __r1 = (n1) % __d1;							\
+    __q1 = (n1) / __d1;							\
+    __m = (UWtype) __q1 * __d0;						\
+    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
+    if (__r1 < __m)							\
+      {									\
+	__q1--, __r1 += (d);						\
+	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
+	  if (__r1 < __m)						\
+	    __q1--, __r1 += (d);					\
+      }									\
+    __r1 -= __m;							\
+									\
+    __r0 = __r1 % __d1;							\
+    __q0 = __r1 / __d1;							\
+    __m = (UWtype) __q0 * __d0;						\
+    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
+    if (__r0 < __m)							\
+      {									\
+	__q0--, __r0 += (d);						\
+	if (__r0 >= (d))						\
+	  if (__r0 < __m)						\
+	    __q0--, __r0 += (d);					\
+      }									\
+    __r0 -= __m;							\
+									\
+    (q) = (UWtype) __q1 * __ll_B | __q0;				\
+    (r) = __r0;								\
+  } while (0)
+
+#define udiv_qrnnd __udiv_qrnnd_c
+
+#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
+  do {									\
+    UWtype __x;								\
+    __x = (al) - (bl);							\
+    (sh) = (ah) - (bh) - (__x > (al));					\
+    (sl) = __x;								\
+  } while (0)
+
+#define umul_ppmm(w1, w0, u, v)						\
+  do {									\
+    UWtype __x0, __x1, __x2, __x3;					\
+    UHWtype __ul, __vl, __uh, __vh;					\
+									\
+    __ul = __ll_lowpart (u);						\
+    __uh = __ll_highpart (u);						\
+    __vl = __ll_lowpart (v);						\
+    __vh = __ll_highpart (v);						\
+									\
+    __x0 = (UWtype) __ul * __vl;					\
+    __x1 = (UWtype) __ul * __vh;					\
+    __x2 = (UWtype) __uh * __vl;					\
+    __x3 = (UWtype) __uh * __vh;					\
+									\
+    __x1 += __ll_highpart (__x0);/* this can't give carry */		\
+    __x1 += __x2;		/* but this indeed can */		\
+    if (__x1 < __x2)		/* did we get it? */			\
+      __x3 += __ll_B;		/* yes, add it in the proper pos.  */	\
+									\
+    (w1) = __x3 + __ll_highpart (__x1);					\
+    (w0) = __ll_lowpart (__x1) * __ll_B + __ll_lowpart (__x0);		\
+  } while (0)
diff --git a/arch/avr32/lib/memcpy.S b/arch/avr32/lib/memcpy.S
new file mode 100644
index 000000000000..0abb26142b64
--- /dev/null
+++ b/arch/avr32/lib/memcpy.S
@@ -0,0 +1,62 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+	/*
+	 * void *memcpy(void *to, const void *from, unsigned long n)
+	 *
+	 * This implementation does word-aligned loads in the main loop,
+	 * possibly sacrificing alignment of stores.
+	 *
+	 * Hopefully, in most cases, both "to" and "from" will be
+	 * word-aligned to begin with.
+	 */
+	.text
+	.global	memcpy
+	.type	memcpy, @function
+memcpy:
+	mov	r9, r11
+	andl	r9, 3, COH
+	brne	1f
+
+	/* At this point, "from" is word-aligned */
+2:	sub	r10, 4
+	mov	r9, r12
+	brlt	4f
+
+3:	ld.w	r8, r11++
+	sub	r10, 4
+	st.w	r12++, r8
+	brge	3b
+
+4:	neg	r10
+	reteq	r9
+
+	/* Handle unaligned count */
+	lsl	r10, 2
+	add	pc, pc, r10
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	retal	r9
+
+	/* Handle unaligned "from" pointer */
+1:	sub	r10, 4
+	brlt	4b
+	add	r10, r9
+	lsl	r9, 2
+	add	pc, pc, r9
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	ld.ub	r8, r11++
+	st.b	r12++, r8
+	rjmp	2b
diff --git a/arch/avr32/lib/memset.S b/arch/avr32/lib/memset.S
new file mode 100644
index 000000000000..40da32c0480c
--- /dev/null
+++ b/arch/avr32/lib/memset.S
@@ -0,0 +1,72 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/arm/lib/memset.S
+ *   Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * ASM optimised string functions
+ */
+#include <asm/asm.h>
+
+	/*
+	 * r12:	void *b
+	 * r11:	int c
+	 * r10:	size_t len
+	 *
+	 * Returns b in r12
+	 */
+	.text
+	.global	memset
+	.type	memset, @function
+	.align	5
+memset:
+	mov	r9, r12
+	mov	r8, r12
+	or	r11, r11, r11 << 8
+	andl	r9, 3, COH
+	brne	1f
+
+2:	or	r11, r11, r11 << 16
+	sub	r10, 4
+	brlt	5f
+
+	/* Let's do some real work */
+4:	st.w	r8++, r11
+	sub	r10, 4
+	brge	4b
+
+	/*
+	 * When we get here, we've got less than 4 bytes to set. r10
+	 * might be negative.
+	 */
+5:	sub	r10, -4
+	reteq	r12
+
+	/* Fastpath ends here, exactly 32 bytes from memset */
+
+	/* Handle unaligned count or pointer */
+	bld	r10, 1
+	brcc	6f
+	st.b	r8++, r11
+	st.b	r8++, r11
+	bld	r10, 0
+	retcc	r12
+6:	st.b	r8++, r11
+	retal	r12
+
+	/* Handle unaligned pointer */
+1:	sub	r10, 4
+	brlt	5b
+	add	r10, r9
+	lsl	r9, 1
+	add	pc, r9
+	st.b	r8++, r11
+	st.b	r8++, r11
+	st.b	r8++, r11
+	rjmp	2b
+
+	.size	memset, . - memset
diff --git a/arch/avr32/lib/strncpy_from_user.S b/arch/avr32/lib/strncpy_from_user.S
new file mode 100644
index 000000000000..72bd50599ec6
--- /dev/null
+++ b/arch/avr32/lib/strncpy_from_user.S
@@ -0,0 +1,60 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/errno.h>
+
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/asm.h>
+
+	/*
+	 * long strncpy_from_user(char *dst, const char *src, long count)
+	 *
+	 * On success, returns the length of the string, not including
+	 * the terminating NUL.
+	 *
+	 * If the string is longer than count, returns count
+	 *
+	 * If userspace access fails, returns -EFAULT
+	 */
+	.text
+	.align	1
+	.global	strncpy_from_user
+	.type	strncpy_from_user, "function"
+strncpy_from_user:
+	mov	r9, -EFAULT
+	branch_if_kernel r8, __strncpy_from_user
+	ret_if_privileged r8, r11, r10, r9
+
+	.global	__strncpy_from_user
+	.type	__strncpy_from_user, "function"
+__strncpy_from_user:
+	cp.w	r10, 0
+	reteq	0
+
+	mov	r9, r10
+
+1:	ld.ub	r8, r11++
+	st.b	r12++, r8
+	cp.w	r8, 0
+	breq	2f
+	sub	r9, 1
+	brne	1b
+
+2:	sub	r10, r9
+	retal	r10
+
+	.section .fixup, "ax"
+	.align	1
+3:	mov	r12, -EFAULT
+	retal	r12
+
+	.section __ex_table, "a"
+	.align	2
+	.long	1b, 3b
diff --git a/arch/avr32/lib/strnlen_user.S b/arch/avr32/lib/strnlen_user.S
new file mode 100644
index 000000000000..65ce11afa66a
--- /dev/null
+++ b/arch/avr32/lib/strnlen_user.S
@@ -0,0 +1,67 @@
+/*
+ * Copy to/from userspace with optional address space checking.
+ *
+ * Copyright 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <asm/page.h>
+#include <asm/thread_info.h>
+#include <asm/processor.h>
+#include <asm/asm.h>
+
+	.text
+	.align	1
+	.global	strnlen_user
+	.type	strnlen_user, "function"
+strnlen_user:
+	branch_if_kernel r8, __strnlen_user
+	sub	r8, r11, 1
+	add	r8, r12
+	retcs	0
+	brmi	adjust_length	/* do a closer inspection */
+
+	.global	__strnlen_user
+	.type	__strnlen_user, "function"
+__strnlen_user:
+	mov	r10, r12
+
+10:	ld.ub	r8, r12++
+	cp.w	r8, 0
+	breq	2f
+	sub	r11, 1
+	brne	10b
+
+	sub	r12, -1
+2:	sub	r12, r10
+	retal	r12
+
+
+	.type	adjust_length, "function"
+adjust_length:
+	cp.w	r12, 0		/* addr must always be < TASK_SIZE */
+	retmi	0
+
+	pushm	lr
+	lddpc	lr, _task_size
+	sub	r11, lr, r12
+	mov	r9, r11
+	rcall	__strnlen_user
+	cp.w	r12, r9
+	brgt	1f
+	popm	pc
+1:	popm	pc, r12=0
+
+	.align	2
+_task_size:
+	.long	TASK_SIZE
+
+	.section .fixup, "ax"
+	.align	1
+19:	retal	0
+
+	.section __ex_table, "a"
+	.align	2
+	.long	10b, 19b
diff --git a/arch/avr32/mach-at32ap/Makefile b/arch/avr32/mach-at32ap/Makefile
new file mode 100644
index 000000000000..4b10853eb614
--- /dev/null
+++ b/arch/avr32/mach-at32ap/Makefile
@@ -0,0 +1,2 @@
+obj-y				+= at32ap.o clock.o pio.o intc.o extint.o
+obj-$(CONFIG_CPU_AT32AP7000)	+= at32ap7000.o
diff --git a/arch/avr32/mach-at32ap/at32ap.c b/arch/avr32/mach-at32ap/at32ap.c
new file mode 100644
index 000000000000..f7cedf5aabea
--- /dev/null
+++ b/arch/avr32/mach-at32ap/at32ap.c
@@ -0,0 +1,90 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/init.h>
+#include <asm/arch/sm.h>
+
+struct at32_sm system_manager;
+
+static int __init at32_sm_init(void)
+{
+	struct resource *regs;
+	struct at32_sm *sm = &system_manager;
+	int ret = -ENXIO;
+
+	regs = platform_get_resource(&at32_sm_device, IORESOURCE_MEM, 0);
+	if (!regs)
+		goto fail;
+
+	spin_lock_init(&sm->lock);
+	sm->pdev = &at32_sm_device;
+
+	ret = -ENOMEM;
+	sm->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!sm->regs)
+		goto fail;
+
+	return 0;
+
+fail:
+	printk(KERN_ERR "Failed to initialize System Manager: %d\n", ret);
+	return ret;
+}
+
+void __init setup_platform(void)
+{
+	at32_sm_init();
+	at32_clock_init();
+	at32_portmux_init();
+
+	/* FIXME: This doesn't belong here */
+	at32_setup_serial_console(1);
+}
+
+static int __init pdc_probe(struct platform_device *pdev)
+{
+	struct clk *pclk, *hclk;
+
+	pclk = clk_get(&pdev->dev, "pclk");
+	if (IS_ERR(pclk)) {
+		dev_err(&pdev->dev, "no pclk defined\n");
+		return PTR_ERR(pclk);
+	}
+	hclk = clk_get(&pdev->dev, "hclk");
+	if (IS_ERR(hclk)) {
+		dev_err(&pdev->dev, "no hclk defined\n");
+		clk_put(pclk);
+		return PTR_ERR(hclk);
+	}
+
+	clk_enable(pclk);
+	clk_enable(hclk);
+
+	dev_info(&pdev->dev, "Atmel Peripheral DMA Controller enabled\n");
+	return 0;
+}
+
+static struct platform_driver pdc_driver = {
+	.probe		= pdc_probe,
+	.driver		= {
+		.name	= "pdc",
+	},
+};
+
+static int __init pdc_init(void)
+{
+	return platform_driver_register(&pdc_driver);
+}
+arch_initcall(pdc_init);
diff --git a/arch/avr32/mach-at32ap/at32ap7000.c b/arch/avr32/mach-at32ap/at32ap7000.c
new file mode 100644
index 000000000000..e8c6893a1c23
--- /dev/null
+++ b/arch/avr32/mach-at32ap/at32ap7000.c
@@ -0,0 +1,866 @@
+/*
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/board.h>
+#include <asm/arch/portmux.h>
+#include <asm/arch/sm.h>
+
+#include "clock.h"
+#include "pio.h"
+#include "sm.h"
+
+#define PBMEM(base)					\
+	{						\
+		.start		= base,			\
+		.end		= base + 0x3ff,		\
+		.flags		= IORESOURCE_MEM,	\
+	}
+#define IRQ(num)					\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+#define NAMED_IRQ(num, _name)				\
+	{						\
+		.start		= num,			\
+		.end		= num,			\
+		.name		= _name,		\
+		.flags		= IORESOURCE_IRQ,	\
+	}
+
+#define DEFINE_DEV(_name, _id)					\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+#define DEFINE_DEV_DATA(_name, _id)				\
+static struct platform_device _name##_id##_device = {		\
+	.name		= #_name,				\
+	.id		= _id,					\
+	.dev		= {					\
+		.platform_data	= &_name##_id##_data,		\
+	},							\
+	.resource	= _name##_id##_resource,		\
+	.num_resources	= ARRAY_SIZE(_name##_id##_resource),	\
+}
+
+#define DEV_CLK(_name, devname, bus, _index)			\
+static struct clk devname##_##_name = {				\
+	.name		= #_name,				\
+	.dev		= &devname##_device.dev,		\
+	.parent		= &bus##_clk,				\
+	.mode		= bus##_clk_mode,			\
+	.get_rate	= bus##_clk_get_rate,			\
+	.index		= _index,				\
+}
+
+enum {
+	PIOA,
+	PIOB,
+	PIOC,
+	PIOD,
+};
+
+enum {
+	FUNC_A,
+	FUNC_B,
+};
+
+unsigned long at32ap7000_osc_rates[3] = {
+	[0] = 32768,
+	/* FIXME: these are ATSTK1002-specific */
+	[1] = 20000000,
+	[2] = 12000000,
+};
+
+static unsigned long osc_get_rate(struct clk *clk)
+{
+	return at32ap7000_osc_rates[clk->index];
+}
+
+static unsigned long pll_get_rate(struct clk *clk, unsigned long control)
+{
+	unsigned long div, mul, rate;
+
+	if (!(control & SM_BIT(PLLEN)))
+		return 0;
+
+	div = SM_BFEXT(PLLDIV, control) + 1;
+	mul = SM_BFEXT(PLLMUL, control) + 1;
+
+	rate = clk->parent->get_rate(clk->parent);
+	rate = (rate + div / 2) / div;
+	rate *= mul;
+
+	return rate;
+}
+
+static unsigned long pll0_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = sm_readl(&system_manager, PM_PLL0);
+
+	return pll_get_rate(clk, control);
+}
+
+static unsigned long pll1_get_rate(struct clk *clk)
+{
+	u32 control;
+
+	control = sm_readl(&system_manager, PM_PLL1);
+
+	return pll_get_rate(clk, control);
+}
+
+/*
+ * The AT32AP7000 has five primary clock sources: One 32kHz
+ * oscillator, two crystal oscillators and two PLLs.
+ */
+static struct clk osc32k = {
+	.name		= "osc32k",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 0,
+};
+static struct clk osc0 = {
+	.name		= "osc0",
+	.get_rate	= osc_get_rate,
+	.users		= 1,
+	.index		= 1,
+};
+static struct clk osc1 = {
+	.name		= "osc1",
+	.get_rate	= osc_get_rate,
+	.index		= 2,
+};
+static struct clk pll0 = {
+	.name		= "pll0",
+	.get_rate	= pll0_get_rate,
+	.parent		= &osc0,
+};
+static struct clk pll1 = {
+	.name		= "pll1",
+	.get_rate	= pll1_get_rate,
+	.parent		= &osc0,
+};
+
+/*
+ * The main clock can be either osc0 or pll0.  The boot loader may
+ * have chosen one for us, so we don't really know which one until we
+ * have a look at the SM.
+ */
+static struct clk *main_clock;
+
+/*
+ * Synchronous clocks are generated from the main clock. The clocks
+ * must satisfy the constraint
+ *   fCPU >= fHSB >= fPB
+ * i.e. each clock must not be faster than its parent.
+ */
+static unsigned long bus_clk_get_rate(struct clk *clk, unsigned int shift)
+{
+	return main_clock->get_rate(main_clock) >> shift;
+};
+
+static void cpu_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_CPU_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_CPU_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long cpu_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(CPUDIV))
+		shift = SM_BFEXT(CPUSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void hsb_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_HSB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_HSB_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long hsb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(HSBDIV))
+		shift = SM_BFEXT(HSBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pba_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_PBA_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_PBA_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long pba_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(PBADIV))
+		shift = SM_BFEXT(PBASEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static void pbb_clk_mode(struct clk *clk, int enabled)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned long flags;
+	u32 mask;
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mask = sm_readl(sm, PM_PBB_MASK);
+	if (enabled)
+		mask |= 1 << clk->index;
+	else
+		mask &= ~(1 << clk->index);
+	sm_writel(sm, PM_PBB_MASK, mask);
+	spin_unlock_irqrestore(&sm->lock, flags);
+}
+
+static unsigned long pbb_clk_get_rate(struct clk *clk)
+{
+	unsigned long cksel, shift = 0;
+
+	cksel = sm_readl(&system_manager, PM_CKSEL);
+	if (cksel & SM_BIT(PBBDIV))
+		shift = SM_BFEXT(PBBSEL, cksel) + 1;
+
+	return bus_clk_get_rate(clk, shift);
+}
+
+static struct clk cpu_clk = {
+	.name		= "cpu",
+	.get_rate	= cpu_clk_get_rate,
+	.users		= 1,
+};
+static struct clk hsb_clk = {
+	.name		= "hsb",
+	.parent		= &cpu_clk,
+	.get_rate	= hsb_clk_get_rate,
+};
+static struct clk pba_clk = {
+	.name		= "pba",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pba_clk_get_rate,
+	.index		= 1,
+};
+static struct clk pbb_clk = {
+	.name		= "pbb",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= pbb_clk_get_rate,
+	.users		= 1,
+	.index		= 2,
+};
+
+/* --------------------------------------------------------------------
+ *  Generic Clock operations
+ * -------------------------------------------------------------------- */
+
+static void genclk_mode(struct clk *clk, int enabled)
+{
+	u32 control;
+
+	BUG_ON(clk->index > 7);
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+	if (enabled)
+		control |= SM_BIT(CEN);
+	else
+		control &= ~SM_BIT(CEN);
+	sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index, control);
+}
+
+static unsigned long genclk_get_rate(struct clk *clk)
+{
+	u32 control;
+	unsigned long div = 1;
+
+	BUG_ON(clk->index > 7);
+
+	if (!clk->parent)
+		return 0;
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+	if (control & SM_BIT(DIVEN))
+		div = 2 * (SM_BFEXT(DIV, control) + 1);
+
+	return clk->parent->get_rate(clk->parent) / div;
+}
+
+static long genclk_set_rate(struct clk *clk, unsigned long rate, int apply)
+{
+	u32 control;
+	unsigned long parent_rate, actual_rate, div;
+
+	BUG_ON(clk->index > 7);
+
+	if (!clk->parent)
+		return 0;
+
+	parent_rate = clk->parent->get_rate(clk->parent);
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+
+	if (rate > 3 * parent_rate / 4) {
+		actual_rate = parent_rate;
+		control &= ~SM_BIT(DIVEN);
+	} else {
+		div = (parent_rate + rate) / (2 * rate) - 1;
+		control = SM_BFINS(DIV, div, control) | SM_BIT(DIVEN);
+		actual_rate = parent_rate / (2 * (div + 1));
+	}
+
+	printk("clk %s: new rate %lu (actual rate %lu)\n",
+	       clk->name, rate, actual_rate);
+
+	if (apply)
+		sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index,
+			  control);
+
+	return actual_rate;
+}
+
+int genclk_set_parent(struct clk *clk, struct clk *parent)
+{
+	u32 control;
+
+	BUG_ON(clk->index > 7);
+
+	printk("clk %s: new parent %s (was %s)\n",
+	       clk->name, parent->name,
+	       clk->parent ? clk->parent->name : "(null)");
+
+	control = sm_readl(&system_manager, PM_GCCTRL + 4 * clk->index);
+
+	if (parent == &osc1 || parent == &pll1)
+		control |= SM_BIT(OSCSEL);
+	else if (parent == &osc0 || parent == &pll0)
+		control &= ~SM_BIT(OSCSEL);
+	else
+		return -EINVAL;
+
+	if (parent == &pll0 || parent == &pll1)
+		control |= SM_BIT(PLLSEL);
+	else
+		control &= ~SM_BIT(PLLSEL);
+
+	sm_writel(&system_manager, PM_GCCTRL + 4 * clk->index, control);
+	clk->parent = parent;
+
+	return 0;
+}
+
+/* --------------------------------------------------------------------
+ *  System peripherals
+ * -------------------------------------------------------------------- */
+static struct resource sm_resource[] = {
+	PBMEM(0xfff00000),
+	NAMED_IRQ(19, "eim"),
+	NAMED_IRQ(20, "pm"),
+	NAMED_IRQ(21, "rtc"),
+};
+struct platform_device at32_sm_device = {
+	.name		= "sm",
+	.id		= 0,
+	.resource	= sm_resource,
+	.num_resources	= ARRAY_SIZE(sm_resource),
+};
+DEV_CLK(pclk, at32_sm, pbb, 0);
+
+static struct resource intc0_resource[] = {
+	PBMEM(0xfff00400),
+};
+struct platform_device at32_intc0_device = {
+	.name		= "intc",
+	.id		= 0,
+	.resource	= intc0_resource,
+	.num_resources	= ARRAY_SIZE(intc0_resource),
+};
+DEV_CLK(pclk, at32_intc0, pbb, 1);
+
+static struct clk ebi_clk = {
+	.name		= "ebi",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+};
+static struct clk hramc_clk = {
+	.name		= "hramc",
+	.parent		= &hsb_clk,
+	.mode		= hsb_clk_mode,
+	.get_rate	= hsb_clk_get_rate,
+	.users		= 1,
+};
+
+static struct platform_device pdc_device = {
+	.name		= "pdc",
+	.id		= 0,
+};
+DEV_CLK(hclk, pdc, hsb, 4);
+DEV_CLK(pclk, pdc, pba, 16);
+
+static struct clk pico_clk = {
+	.name		= "pico",
+	.parent		= &cpu_clk,
+	.mode		= cpu_clk_mode,
+	.get_rate	= cpu_clk_get_rate,
+	.users		= 1,
+};
+
+/* --------------------------------------------------------------------
+ *  PIO
+ * -------------------------------------------------------------------- */
+
+static struct resource pio0_resource[] = {
+	PBMEM(0xffe02800),
+	IRQ(13),
+};
+DEFINE_DEV(pio, 0);
+DEV_CLK(mck, pio0, pba, 10);
+
+static struct resource pio1_resource[] = {
+	PBMEM(0xffe02c00),
+	IRQ(14),
+};
+DEFINE_DEV(pio, 1);
+DEV_CLK(mck, pio1, pba, 11);
+
+static struct resource pio2_resource[] = {
+	PBMEM(0xffe03000),
+	IRQ(15),
+};
+DEFINE_DEV(pio, 2);
+DEV_CLK(mck, pio2, pba, 12);
+
+static struct resource pio3_resource[] = {
+	PBMEM(0xffe03400),
+	IRQ(16),
+};
+DEFINE_DEV(pio, 3);
+DEV_CLK(mck, pio3, pba, 13);
+
+void __init at32_add_system_devices(void)
+{
+	system_manager.eim_first_irq = NR_INTERNAL_IRQS;
+
+	platform_device_register(&at32_sm_device);
+	platform_device_register(&at32_intc0_device);
+	platform_device_register(&pdc_device);
+
+	platform_device_register(&pio0_device);
+	platform_device_register(&pio1_device);
+	platform_device_register(&pio2_device);
+	platform_device_register(&pio3_device);
+}
+
+/* --------------------------------------------------------------------
+ *  USART
+ * -------------------------------------------------------------------- */
+
+static struct resource usart0_resource[] = {
+	PBMEM(0xffe00c00),
+	IRQ(7),
+};
+DEFINE_DEV(usart, 0);
+DEV_CLK(usart, usart0, pba, 4);
+
+static struct resource usart1_resource[] = {
+	PBMEM(0xffe01000),
+	IRQ(7),
+};
+DEFINE_DEV(usart, 1);
+DEV_CLK(usart, usart1, pba, 4);
+
+static struct resource usart2_resource[] = {
+	PBMEM(0xffe01400),
+	IRQ(8),
+};
+DEFINE_DEV(usart, 2);
+DEV_CLK(usart, usart2, pba, 5);
+
+static struct resource usart3_resource[] = {
+	PBMEM(0xffe01800),
+	IRQ(9),
+};
+DEFINE_DEV(usart, 3);
+DEV_CLK(usart, usart3, pba, 6);
+
+static inline void configure_usart0_pins(void)
+{
+	portmux_set_func(PIOA,  8, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOA,  9, FUNC_B);	/* TXD	*/
+}
+
+static inline void configure_usart1_pins(void)
+{
+	portmux_set_func(PIOA, 17, FUNC_A);	/* RXD	*/
+	portmux_set_func(PIOA, 18, FUNC_A);	/* TXD	*/
+}
+
+static inline void configure_usart2_pins(void)
+{
+	portmux_set_func(PIOB, 26, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOB, 27, FUNC_B);	/* TXD	*/
+}
+
+static inline void configure_usart3_pins(void)
+{
+	portmux_set_func(PIOB, 18, FUNC_B);	/* RXD	*/
+	portmux_set_func(PIOB, 17, FUNC_B);	/* TXD	*/
+}
+
+static struct platform_device *setup_usart(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &usart0_device;
+		configure_usart0_pins();
+		break;
+	case 1:
+		pdev = &usart1_device;
+		configure_usart1_pins();
+		break;
+	case 2:
+		pdev = &usart2_device;
+		configure_usart2_pins();
+		break;
+	case 3:
+		pdev = &usart3_device;
+		configure_usart3_pins();
+		break;
+	default:
+		pdev = NULL;
+		break;
+	}
+
+	return pdev;
+}
+
+struct platform_device *__init at32_add_device_usart(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	pdev = setup_usart(id);
+	if (pdev)
+		platform_device_register(pdev);
+
+	return pdev;
+}
+
+struct platform_device *at91_default_console_device;
+
+void __init at32_setup_serial_console(unsigned int usart_id)
+{
+	at91_default_console_device = setup_usart(usart_id);
+}
+
+/* --------------------------------------------------------------------
+ *  Ethernet
+ * -------------------------------------------------------------------- */
+
+static struct eth_platform_data macb0_data;
+static struct resource macb0_resource[] = {
+	PBMEM(0xfff01800),
+	IRQ(25),
+};
+DEFINE_DEV_DATA(macb, 0);
+DEV_CLK(hclk, macb0, hsb, 8);
+DEV_CLK(pclk, macb0, pbb, 6);
+
+struct platform_device *__init
+at32_add_device_eth(unsigned int id, struct eth_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &macb0_device;
+
+		portmux_set_func(PIOC,  3, FUNC_A);	/* TXD0	*/
+		portmux_set_func(PIOC,  4, FUNC_A);	/* TXD1	*/
+		portmux_set_func(PIOC,  7, FUNC_A);	/* TXEN	*/
+		portmux_set_func(PIOC,  8, FUNC_A);	/* TXCK */
+		portmux_set_func(PIOC,  9, FUNC_A);	/* RXD0	*/
+		portmux_set_func(PIOC, 10, FUNC_A);	/* RXD1	*/
+		portmux_set_func(PIOC, 13, FUNC_A);	/* RXER	*/
+		portmux_set_func(PIOC, 15, FUNC_A);	/* RXDV	*/
+		portmux_set_func(PIOC, 16, FUNC_A);	/* MDC	*/
+		portmux_set_func(PIOC, 17, FUNC_A);	/* MDIO	*/
+
+		if (!data->is_rmii) {
+			portmux_set_func(PIOC,  0, FUNC_A);	/* COL	*/
+			portmux_set_func(PIOC,  1, FUNC_A);	/* CRS	*/
+			portmux_set_func(PIOC,  2, FUNC_A);	/* TXER	*/
+			portmux_set_func(PIOC,  5, FUNC_A);	/* TXD2	*/
+			portmux_set_func(PIOC,  6, FUNC_A);	/* TXD3 */
+			portmux_set_func(PIOC, 11, FUNC_A);	/* RXD2	*/
+			portmux_set_func(PIOC, 12, FUNC_A);	/* RXD3	*/
+			portmux_set_func(PIOC, 14, FUNC_A);	/* RXCK	*/
+			portmux_set_func(PIOC, 18, FUNC_A);	/* SPD	*/
+		}
+		break;
+
+	default:
+		return NULL;
+	}
+
+	memcpy(pdev->dev.platform_data, data, sizeof(struct eth_platform_data));
+	platform_device_register(pdev);
+
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  SPI
+ * -------------------------------------------------------------------- */
+static struct resource spi0_resource[] = {
+	PBMEM(0xffe00000),
+	IRQ(3),
+};
+DEFINE_DEV(spi, 0);
+DEV_CLK(mck, spi0, pba, 0);
+
+struct platform_device *__init at32_add_device_spi(unsigned int id)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &spi0_device;
+		portmux_set_func(PIOA,  0, FUNC_A);	/* MISO	 */
+		portmux_set_func(PIOA,  1, FUNC_A);	/* MOSI	 */
+		portmux_set_func(PIOA,  2, FUNC_A);	/* SCK	 */
+		portmux_set_func(PIOA,  3, FUNC_A);	/* NPCS0 */
+		portmux_set_func(PIOA,  4, FUNC_A);	/* NPCS1 */
+		portmux_set_func(PIOA,  5, FUNC_A);	/* NPCS2 */
+		break;
+
+	default:
+		return NULL;
+	}
+
+	platform_device_register(pdev);
+	return pdev;
+}
+
+/* --------------------------------------------------------------------
+ *  LCDC
+ * -------------------------------------------------------------------- */
+static struct lcdc_platform_data lcdc0_data;
+static struct resource lcdc0_resource[] = {
+	{
+		.start		= 0xff000000,
+		.end		= 0xff000fff,
+		.flags		= IORESOURCE_MEM,
+	},
+	IRQ(1),
+};
+DEFINE_DEV_DATA(lcdc, 0);
+DEV_CLK(hclk, lcdc0, hsb, 7);
+static struct clk lcdc0_pixclk = {
+	.name		= "pixclk",
+	.dev		= &lcdc0_device.dev,
+	.mode		= genclk_mode,
+	.get_rate	= genclk_get_rate,
+	.set_rate	= genclk_set_rate,
+	.set_parent	= genclk_set_parent,
+	.index		= 7,
+};
+
+struct platform_device *__init
+at32_add_device_lcdc(unsigned int id, struct lcdc_platform_data *data)
+{
+	struct platform_device *pdev;
+
+	switch (id) {
+	case 0:
+		pdev = &lcdc0_device;
+		portmux_set_func(PIOC, 19, FUNC_A);	/* CC	  */
+		portmux_set_func(PIOC, 20, FUNC_A);	/* HSYNC  */
+		portmux_set_func(PIOC, 21, FUNC_A);	/* PCLK	  */
+		portmux_set_func(PIOC, 22, FUNC_A);	/* VSYNC  */
+		portmux_set_func(PIOC, 23, FUNC_A);	/* DVAL	  */
+		portmux_set_func(PIOC, 24, FUNC_A);	/* MODE	  */
+		portmux_set_func(PIOC, 25, FUNC_A);	/* PWR	  */
+		portmux_set_func(PIOC, 26, FUNC_A);	/* DATA0  */
+		portmux_set_func(PIOC, 27, FUNC_A);	/* DATA1  */
+		portmux_set_func(PIOC, 28, FUNC_A);	/* DATA2  */
+		portmux_set_func(PIOC, 29, FUNC_A);	/* DATA3  */
+		portmux_set_func(PIOC, 30, FUNC_A);	/* DATA4  */
+		portmux_set_func(PIOC, 31, FUNC_A);	/* DATA5  */
+		portmux_set_func(PIOD,  0, FUNC_A);	/* DATA6  */
+		portmux_set_func(PIOD,  1, FUNC_A);	/* DATA7  */
+		portmux_set_func(PIOD,  2, FUNC_A);	/* DATA8  */
+		portmux_set_func(PIOD,  3, FUNC_A);	/* DATA9  */
+		portmux_set_func(PIOD,  4, FUNC_A);	/* DATA10 */
+		portmux_set_func(PIOD,  5, FUNC_A);	/* DATA11 */
+		portmux_set_func(PIOD,  6, FUNC_A);	/* DATA12 */
+		portmux_set_func(PIOD,  7, FUNC_A);	/* DATA13 */
+		portmux_set_func(PIOD,  8, FUNC_A);	/* DATA14 */
+		portmux_set_func(PIOD,  9, FUNC_A);	/* DATA15 */
+		portmux_set_func(PIOD, 10, FUNC_A);	/* DATA16 */
+		portmux_set_func(PIOD, 11, FUNC_A);	/* DATA17 */
+		portmux_set_func(PIOD, 12, FUNC_A);	/* DATA18 */
+		portmux_set_func(PIOD, 13, FUNC_A);	/* DATA19 */
+		portmux_set_func(PIOD, 14, FUNC_A);	/* DATA20 */
+		portmux_set_func(PIOD, 15, FUNC_A);	/* DATA21 */
+		portmux_set_func(PIOD, 16, FUNC_A);	/* DATA22 */
+		portmux_set_func(PIOD, 17, FUNC_A);	/* DATA23 */
+
+		clk_set_parent(&lcdc0_pixclk, &pll0);
+		clk_set_rate(&lcdc0_pixclk, clk_get_rate(&pll0));
+		break;
+
+	default:
+		return NULL;
+	}
+
+	memcpy(pdev->dev.platform_data, data,
+	       sizeof(struct lcdc_platform_data));
+
+	platform_device_register(pdev);
+	return pdev;
+}
+
+struct clk *at32_clock_list[] = {
+	&osc32k,
+	&osc0,
+	&osc1,
+	&pll0,
+	&pll1,
+	&cpu_clk,
+	&hsb_clk,
+	&pba_clk,
+	&pbb_clk,
+	&at32_sm_pclk,
+	&at32_intc0_pclk,
+	&ebi_clk,
+	&hramc_clk,
+	&pdc_hclk,
+	&pdc_pclk,
+	&pico_clk,
+	&pio0_mck,
+	&pio1_mck,
+	&pio2_mck,
+	&pio3_mck,
+	&usart0_usart,
+	&usart1_usart,
+	&usart2_usart,
+	&usart3_usart,
+	&macb0_hclk,
+	&macb0_pclk,
+	&spi0_mck,
+	&lcdc0_hclk,
+	&lcdc0_pixclk,
+};
+unsigned int at32_nr_clocks = ARRAY_SIZE(at32_clock_list);
+
+void __init at32_portmux_init(void)
+{
+	at32_init_pio(&pio0_device);
+	at32_init_pio(&pio1_device);
+	at32_init_pio(&pio2_device);
+	at32_init_pio(&pio3_device);
+}
+
+void __init at32_clock_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	u32 cpu_mask = 0, hsb_mask = 0, pba_mask = 0, pbb_mask = 0;
+	int i;
+
+	if (sm_readl(sm, PM_MCCTRL) & SM_BIT(PLLSEL))
+		main_clock = &pll0;
+	else
+		main_clock = &osc0;
+
+	if (sm_readl(sm, PM_PLL0) & SM_BIT(PLLOSC))
+		pll0.parent = &osc1;
+	if (sm_readl(sm, PM_PLL1) & SM_BIT(PLLOSC))
+		pll1.parent = &osc1;
+
+	/*
+	 * Turn on all clocks that have at least one user already, and
+	 * turn off everything else. We only do this for module
+	 * clocks, and even though it isn't particularly pretty to
+	 * check the address of the mode function, it should do the
+	 * trick...
+	 */
+	for (i = 0; i < ARRAY_SIZE(at32_clock_list); i++) {
+		struct clk *clk = at32_clock_list[i];
+
+		if (clk->mode == &cpu_clk_mode)
+			cpu_mask |= 1 << clk->index;
+		else if (clk->mode == &hsb_clk_mode)
+			hsb_mask |= 1 << clk->index;
+		else if (clk->mode == &pba_clk_mode)
+			pba_mask |= 1 << clk->index;
+		else if (clk->mode == &pbb_clk_mode)
+			pbb_mask |= 1 << clk->index;
+	}
+
+	sm_writel(sm, PM_CPU_MASK, cpu_mask);
+	sm_writel(sm, PM_HSB_MASK, hsb_mask);
+	sm_writel(sm, PM_PBA_MASK, pba_mask);
+	sm_writel(sm, PM_PBB_MASK, pbb_mask);
+}
diff --git a/arch/avr32/mach-at32ap/clock.c b/arch/avr32/mach-at32ap/clock.c
new file mode 100644
index 000000000000..3d0d1097389f
--- /dev/null
+++ b/arch/avr32/mach-at32ap/clock.c
@@ -0,0 +1,148 @@
+/*
+ * Clock management for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on arch/arm/mach-at91rm9200/clock.c
+ *   Copyright (C) 2005 David Brownell
+ *   Copyright (C) 2005 Ivan Kokshaysky
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/device.h>
+#include <linux/string.h>
+
+#include "clock.h"
+
+static spinlock_t clk_lock = SPIN_LOCK_UNLOCKED;
+
+struct clk *clk_get(struct device *dev, const char *id)
+{
+	int i;
+
+	for (i = 0; i < at32_nr_clocks; i++) {
+		struct clk *clk = at32_clock_list[i];
+
+		if (clk->dev == dev && strcmp(id, clk->name) == 0)
+			return clk;
+	}
+
+	return ERR_PTR(-ENOENT);
+}
+EXPORT_SYMBOL(clk_get);
+
+void clk_put(struct clk *clk)
+{
+	/* clocks are static for now, we can't free them */
+}
+EXPORT_SYMBOL(clk_put);
+
+static void __clk_enable(struct clk *clk)
+{
+	if (clk->parent)
+		__clk_enable(clk->parent);
+	if (clk->users++ == 0 && clk->mode)
+		clk->mode(clk, 1);
+}
+
+int clk_enable(struct clk *clk)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	__clk_enable(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return 0;
+}
+EXPORT_SYMBOL(clk_enable);
+
+static void __clk_disable(struct clk *clk)
+{
+	BUG_ON(clk->users == 0);
+
+	if (--clk->users == 0 && clk->mode)
+		clk->mode(clk, 0);
+	if (clk->parent)
+		__clk_disable(clk->parent);
+}
+
+void clk_disable(struct clk *clk)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	__clk_disable(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+}
+EXPORT_SYMBOL(clk_disable);
+
+unsigned long clk_get_rate(struct clk *clk)
+{
+	unsigned long flags;
+	unsigned long rate;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	rate = clk->get_rate(clk);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return rate;
+}
+EXPORT_SYMBOL(clk_get_rate);
+
+long clk_round_rate(struct clk *clk, unsigned long rate)
+{
+	unsigned long flags, actual_rate;
+
+	if (!clk->set_rate)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	actual_rate = clk->set_rate(clk, rate, 0);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return actual_rate;
+}
+EXPORT_SYMBOL(clk_round_rate);
+
+int clk_set_rate(struct clk *clk, unsigned long rate)
+{
+	unsigned long flags;
+	long ret;
+
+	if (!clk->set_rate)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	ret = clk->set_rate(clk, rate, 1);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return (ret < 0) ? ret : 0;
+}
+EXPORT_SYMBOL(clk_set_rate);
+
+int clk_set_parent(struct clk *clk, struct clk *parent)
+{
+	unsigned long flags;
+	int ret;
+
+	if (!clk->set_parent)
+		return -ENOSYS;
+
+	spin_lock_irqsave(&clk_lock, flags);
+	ret = clk->set_parent(clk, parent);
+	spin_unlock_irqrestore(&clk_lock, flags);
+
+	return ret;
+}
+EXPORT_SYMBOL(clk_set_parent);
+
+struct clk *clk_get_parent(struct clk *clk)
+{
+	return clk->parent;
+}
+EXPORT_SYMBOL(clk_get_parent);
diff --git a/arch/avr32/mach-at32ap/clock.h b/arch/avr32/mach-at32ap/clock.h
new file mode 100644
index 000000000000..f953f044ba4d
--- /dev/null
+++ b/arch/avr32/mach-at32ap/clock.h
@@ -0,0 +1,30 @@
+/*
+ * Clock management for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on arch/arm/mach-at91rm9200/clock.c
+ *   Copyright (C) 2005 David Brownell
+ *   Copyright (C) 2005 Ivan Kokshaysky
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/clk.h>
+
+struct clk {
+	const char	*name;		/* Clock name/function */
+	struct device	*dev;		/* Device the clock is used by */
+	struct clk	*parent;	/* Parent clock, if any */
+	void		(*mode)(struct clk *clk, int enabled);
+	unsigned long	(*get_rate)(struct clk *clk);
+	long		(*set_rate)(struct clk *clk, unsigned long rate,
+				    int apply);
+	int		(*set_parent)(struct clk *clk, struct clk *parent);
+	u16		users;		/* Enabled if non-zero */
+	u16		index;		/* Sibling index */
+};
+
+extern struct clk *at32_clock_list[];
+extern unsigned int at32_nr_clocks;
diff --git a/arch/avr32/mach-at32ap/extint.c b/arch/avr32/mach-at32ap/extint.c
new file mode 100644
index 000000000000..7da9c5f7a0eb
--- /dev/null
+++ b/arch/avr32/mach-at32ap/extint.c
@@ -0,0 +1,171 @@
+/*
+ * External interrupt handling for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+#include <linux/random.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/sm.h>
+
+#include "sm.h"
+
+static void eim_ack_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_ICR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_mask_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_IDR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_mask_ack_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_ICR, 1 << (irq - sm->eim_first_irq));
+	sm_writel(sm, EIM_IDR, 1 << (irq - sm->eim_first_irq));
+}
+
+static void eim_unmask_irq(unsigned int irq)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	sm_writel(sm, EIM_IER, 1 << (irq - sm->eim_first_irq));
+}
+
+static int eim_set_irq_type(unsigned int irq, unsigned int flow_type)
+{
+	struct at32_sm *sm = get_irq_chip_data(irq);
+	unsigned int i = irq - sm->eim_first_irq;
+	u32 mode, edge, level;
+	unsigned long flags;
+	int ret = 0;
+
+	flow_type &= IRQ_TYPE_SENSE_MASK;
+
+	spin_lock_irqsave(&sm->lock, flags);
+
+	mode = sm_readl(sm, EIM_MODE);
+	edge = sm_readl(sm, EIM_EDGE);
+	level = sm_readl(sm, EIM_LEVEL);
+
+	switch (flow_type) {
+	case IRQ_TYPE_LEVEL_LOW:
+		mode |= 1 << i;
+		level &= ~(1 << i);
+		break;
+	case IRQ_TYPE_LEVEL_HIGH:
+		mode |= 1 << i;
+		level |= 1 << i;
+		break;
+	case IRQ_TYPE_EDGE_RISING:
+		mode &= ~(1 << i);
+		edge |= 1 << i;
+		break;
+	case IRQ_TYPE_EDGE_FALLING:
+		mode &= ~(1 << i);
+		edge &= ~(1 << i);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+	sm_writel(sm, EIM_MODE, mode);
+	sm_writel(sm, EIM_EDGE, edge);
+	sm_writel(sm, EIM_LEVEL, level);
+
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	return ret;
+}
+
+struct irq_chip eim_chip = {
+	.name		= "eim",
+	.ack		= eim_ack_irq,
+	.mask		= eim_mask_irq,
+	.mask_ack	= eim_mask_ack_irq,
+	.unmask		= eim_unmask_irq,
+	.set_type	= eim_set_irq_type,
+};
+
+static void demux_eim_irq(unsigned int irq, struct irq_desc *desc,
+			  struct pt_regs *regs)
+{
+	struct at32_sm *sm = desc->handler_data;
+	struct irq_desc *ext_desc;
+	unsigned long status, pending;
+	unsigned int i, ext_irq;
+
+	spin_lock(&sm->lock);
+
+	status = sm_readl(sm, EIM_ISR);
+	pending = status & sm_readl(sm, EIM_IMR);
+
+	while (pending) {
+		i = fls(pending) - 1;
+		pending &= ~(1 << i);
+
+		ext_irq = i + sm->eim_first_irq;
+		ext_desc = irq_desc + ext_irq;
+		ext_desc->handle_irq(ext_irq, ext_desc, regs);
+	}
+
+	spin_unlock(&sm->lock);
+}
+
+static int __init eim_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned int i;
+	unsigned int nr_irqs;
+	unsigned int int_irq;
+	u32 pattern;
+
+	/*
+	 * The EIM is really the same module as SM, so register
+	 * mapping, etc. has been taken care of already.
+	 */
+
+	/*
+	 * Find out how many interrupt lines that are actually
+	 * implemented in hardware.
+	 */
+	sm_writel(sm, EIM_IDR, ~0UL);
+	sm_writel(sm, EIM_MODE, ~0UL);
+	pattern = sm_readl(sm, EIM_MODE);
+	nr_irqs = fls(pattern);
+
+	sm->eim_chip = &eim_chip;
+
+	for (i = 0; i < nr_irqs; i++) {
+		set_irq_chip(sm->eim_first_irq + i, &eim_chip);
+		set_irq_chip_data(sm->eim_first_irq + i, sm);
+	}
+
+	int_irq = platform_get_irq_byname(sm->pdev, "eim");
+
+	set_irq_chained_handler(int_irq, demux_eim_irq);
+	set_irq_data(int_irq, sm);
+
+	printk("EIM: External Interrupt Module at 0x%p, IRQ %u\n",
+	       sm->regs, int_irq);
+	printk("EIM: Handling %u external IRQs, starting with IRQ %u\n",
+	       nr_irqs, sm->eim_first_irq);
+
+	return 0;
+}
+arch_initcall(eim_init);
diff --git a/arch/avr32/mach-at32ap/intc.c b/arch/avr32/mach-at32ap/intc.c
new file mode 100644
index 000000000000..74f8c9f2f03d
--- /dev/null
+++ b/arch/avr32/mach-at32ap/intc.c
@@ -0,0 +1,133 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include "intc.h"
+
+struct intc {
+	void __iomem	*regs;
+	struct irq_chip	chip;
+};
+
+extern struct platform_device at32_intc0_device;
+
+/*
+ * TODO: We may be able to implement mask/unmask by setting IxM flags
+ * in the status register.
+ */
+static void intc_mask_irq(unsigned int irq)
+{
+
+}
+
+static void intc_unmask_irq(unsigned int irq)
+{
+
+}
+
+static struct intc intc0 = {
+	.chip = {
+		.name		= "intc",
+		.mask		= intc_mask_irq,
+		.unmask		= intc_unmask_irq,
+	},
+};
+
+/*
+ * All interrupts go via intc at some point.
+ */
+asmlinkage void do_IRQ(int level, struct pt_regs *regs)
+{
+	struct irq_desc *desc;
+	unsigned int irq;
+	unsigned long status_reg;
+
+	local_irq_disable();
+
+	irq_enter();
+
+	irq = intc_readl(&intc0, INTCAUSE0 - 4 * level);
+	desc = irq_desc + irq;
+	desc->handle_irq(irq, desc, regs);
+
+	/*
+	 * Clear all interrupt level masks so that we may handle
+	 * interrupts during softirq processing.  If this is a nested
+	 * interrupt, interrupts must stay globally disabled until we
+	 * return.
+	 */
+	status_reg = sysreg_read(SR);
+	status_reg &= ~(SYSREG_BIT(I0M) | SYSREG_BIT(I1M)
+			| SYSREG_BIT(I2M) | SYSREG_BIT(I3M));
+	sysreg_write(SR, status_reg);
+
+	irq_exit();
+}
+
+void __init init_IRQ(void)
+{
+	extern void _evba(void);
+	extern void irq_level0(void);
+	struct resource *regs;
+	struct clk *pclk;
+	unsigned int i;
+	u32 offset, readback;
+
+	regs = platform_get_resource(&at32_intc0_device, IORESOURCE_MEM, 0);
+	if (!regs) {
+		printk(KERN_EMERG "intc: no mmio resource defined\n");
+		goto fail;
+	}
+	pclk = clk_get(&at32_intc0_device.dev, "pclk");
+	if (IS_ERR(pclk)) {
+		printk(KERN_EMERG "intc: no clock defined\n");
+		goto fail;
+	}
+
+	clk_enable(pclk);
+
+	intc0.regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!intc0.regs) {
+		printk(KERN_EMERG "intc: failed to map registers (0x%08lx)\n",
+		       (unsigned long)regs->start);
+		goto fail;
+	}
+
+	/*
+	 * Initialize all interrupts to level 0 (lowest priority). The
+	 * priority level may be changed by calling
+	 * irq_set_priority().
+	 *
+	 */
+	offset = (unsigned long)&irq_level0 - (unsigned long)&_evba;
+	for (i = 0; i < NR_INTERNAL_IRQS; i++) {
+		intc_writel(&intc0, INTPR0 + 4 * i, offset);
+		readback = intc_readl(&intc0, INTPR0 + 4 * i);
+		if (readback == offset)
+			set_irq_chip_and_handler(i, &intc0.chip,
+						 handle_simple_irq);
+	}
+
+	/* Unmask all interrupt levels */
+	sysreg_write(SR, (sysreg_read(SR)
+			  & ~(SR_I3M | SR_I2M | SR_I1M | SR_I0M)));
+
+	return;
+
+fail:
+	panic("Interrupt controller initialization failed!\n");
+}
+
diff --git a/arch/avr32/mach-at32ap/intc.h b/arch/avr32/mach-at32ap/intc.h
new file mode 100644
index 000000000000..d289ca2fff13
--- /dev/null
+++ b/arch/avr32/mach-at32ap/intc.h
@@ -0,0 +1,327 @@
+/*
+ * Automatically generated by gen-header.xsl
+ */
+#ifndef __ASM_AVR32_PERIHP_INTC_H__
+#define __ASM_AVR32_PERIHP_INTC_H__
+
+#define INTC_NUM_INT_GRPS            33
+
+#define INTC_INTPR0                  0x0
+# define INTC_INTPR0_INTLEV_OFFSET   30
+# define INTC_INTPR0_INTLEV_SIZE     2
+# define INTC_INTPR0_OFFSET_OFFSET   0
+# define INTC_INTPR0_OFFSET_SIZE     24
+#define INTC_INTREQ0                 0x100
+# define INTC_INTREQ0_IREQUEST0_OFFSET 0
+# define INTC_INTREQ0_IREQUEST0_SIZE 1
+# define INTC_INTREQ0_IREQUEST1_OFFSET 1
+# define INTC_INTREQ0_IREQUEST1_SIZE 1
+#define INTC_INTPR1                  0x4
+# define INTC_INTPR1_INTLEV_OFFSET   30
+# define INTC_INTPR1_INTLEV_SIZE     2
+# define INTC_INTPR1_OFFSET_OFFSET   0
+# define INTC_INTPR1_OFFSET_SIZE     24
+#define INTC_INTREQ1                 0x104
+# define INTC_INTREQ1_IREQUEST32_OFFSET 0
+# define INTC_INTREQ1_IREQUEST32_SIZE 1
+# define INTC_INTREQ1_IREQUEST33_OFFSET 1
+# define INTC_INTREQ1_IREQUEST33_SIZE 1
+# define INTC_INTREQ1_IREQUEST34_OFFSET 2
+# define INTC_INTREQ1_IREQUEST34_SIZE 1
+# define INTC_INTREQ1_IREQUEST35_OFFSET 3
+# define INTC_INTREQ1_IREQUEST35_SIZE 1
+# define INTC_INTREQ1_IREQUEST36_OFFSET 4
+# define INTC_INTREQ1_IREQUEST36_SIZE 1
+# define INTC_INTREQ1_IREQUEST37_OFFSET 5
+# define INTC_INTREQ1_IREQUEST37_SIZE 1
+#define INTC_INTPR2                  0x8
+# define INTC_INTPR2_INTLEV_OFFSET   30
+# define INTC_INTPR2_INTLEV_SIZE     2
+# define INTC_INTPR2_OFFSET_OFFSET   0
+# define INTC_INTPR2_OFFSET_SIZE     24
+#define INTC_INTREQ2                 0x108
+# define INTC_INTREQ2_IREQUEST64_OFFSET 0
+# define INTC_INTREQ2_IREQUEST64_SIZE 1
+# define INTC_INTREQ2_IREQUEST65_OFFSET 1
+# define INTC_INTREQ2_IREQUEST65_SIZE 1
+# define INTC_INTREQ2_IREQUEST66_OFFSET 2
+# define INTC_INTREQ2_IREQUEST66_SIZE 1
+# define INTC_INTREQ2_IREQUEST67_OFFSET 3
+# define INTC_INTREQ2_IREQUEST67_SIZE 1
+# define INTC_INTREQ2_IREQUEST68_OFFSET 4
+# define INTC_INTREQ2_IREQUEST68_SIZE 1
+#define INTC_INTPR3                  0xc
+# define INTC_INTPR3_INTLEV_OFFSET   30
+# define INTC_INTPR3_INTLEV_SIZE     2
+# define INTC_INTPR3_OFFSET_OFFSET   0
+# define INTC_INTPR3_OFFSET_SIZE     24
+#define INTC_INTREQ3                 0x10c
+# define INTC_INTREQ3_IREQUEST96_OFFSET 0
+# define INTC_INTREQ3_IREQUEST96_SIZE 1
+#define INTC_INTPR4                  0x10
+# define INTC_INTPR4_INTLEV_OFFSET   30
+# define INTC_INTPR4_INTLEV_SIZE     2
+# define INTC_INTPR4_OFFSET_OFFSET   0
+# define INTC_INTPR4_OFFSET_SIZE     24
+#define INTC_INTREQ4                 0x110
+# define INTC_INTREQ4_IREQUEST128_OFFSET 0
+# define INTC_INTREQ4_IREQUEST128_SIZE 1
+#define INTC_INTPR5                  0x14
+# define INTC_INTPR5_INTLEV_OFFSET   30
+# define INTC_INTPR5_INTLEV_SIZE     2
+# define INTC_INTPR5_OFFSET_OFFSET   0
+# define INTC_INTPR5_OFFSET_SIZE     24
+#define INTC_INTREQ5                 0x114
+# define INTC_INTREQ5_IREQUEST160_OFFSET 0
+# define INTC_INTREQ5_IREQUEST160_SIZE 1
+#define INTC_INTPR6                  0x18
+# define INTC_INTPR6_INTLEV_OFFSET   30
+# define INTC_INTPR6_INTLEV_SIZE     2
+# define INTC_INTPR6_OFFSET_OFFSET   0
+# define INTC_INTPR6_OFFSET_SIZE     24
+#define INTC_INTREQ6                 0x118
+# define INTC_INTREQ6_IREQUEST192_OFFSET 0
+# define INTC_INTREQ6_IREQUEST192_SIZE 1
+#define INTC_INTPR7                  0x1c
+# define INTC_INTPR7_INTLEV_OFFSET   30
+# define INTC_INTPR7_INTLEV_SIZE     2
+# define INTC_INTPR7_OFFSET_OFFSET   0
+# define INTC_INTPR7_OFFSET_SIZE     24
+#define INTC_INTREQ7                 0x11c
+# define INTC_INTREQ7_IREQUEST224_OFFSET 0
+# define INTC_INTREQ7_IREQUEST224_SIZE 1
+#define INTC_INTPR8                  0x20
+# define INTC_INTPR8_INTLEV_OFFSET   30
+# define INTC_INTPR8_INTLEV_SIZE     2
+# define INTC_INTPR8_OFFSET_OFFSET   0
+# define INTC_INTPR8_OFFSET_SIZE     24
+#define INTC_INTREQ8                 0x120
+# define INTC_INTREQ8_IREQUEST256_OFFSET 0
+# define INTC_INTREQ8_IREQUEST256_SIZE 1
+#define INTC_INTPR9                  0x24
+# define INTC_INTPR9_INTLEV_OFFSET   30
+# define INTC_INTPR9_INTLEV_SIZE     2
+# define INTC_INTPR9_OFFSET_OFFSET   0
+# define INTC_INTPR9_OFFSET_SIZE     24
+#define INTC_INTREQ9                 0x124
+# define INTC_INTREQ9_IREQUEST288_OFFSET 0
+# define INTC_INTREQ9_IREQUEST288_SIZE 1
+#define INTC_INTPR10                 0x28
+# define INTC_INTPR10_INTLEV_OFFSET  30
+# define INTC_INTPR10_INTLEV_SIZE    2
+# define INTC_INTPR10_OFFSET_OFFSET  0
+# define INTC_INTPR10_OFFSET_SIZE    24
+#define INTC_INTREQ10                0x128
+# define INTC_INTREQ10_IREQUEST320_OFFSET 0
+# define INTC_INTREQ10_IREQUEST320_SIZE 1
+#define INTC_INTPR11                 0x2c
+# define INTC_INTPR11_INTLEV_OFFSET  30
+# define INTC_INTPR11_INTLEV_SIZE    2
+# define INTC_INTPR11_OFFSET_OFFSET  0
+# define INTC_INTPR11_OFFSET_SIZE    24
+#define INTC_INTREQ11                0x12c
+# define INTC_INTREQ11_IREQUEST352_OFFSET 0
+# define INTC_INTREQ11_IREQUEST352_SIZE 1
+#define INTC_INTPR12                 0x30
+# define INTC_INTPR12_INTLEV_OFFSET  30
+# define INTC_INTPR12_INTLEV_SIZE    2
+# define INTC_INTPR12_OFFSET_OFFSET  0
+# define INTC_INTPR12_OFFSET_SIZE    24
+#define INTC_INTREQ12                0x130
+# define INTC_INTREQ12_IREQUEST384_OFFSET 0
+# define INTC_INTREQ12_IREQUEST384_SIZE 1
+#define INTC_INTPR13                 0x34
+# define INTC_INTPR13_INTLEV_OFFSET  30
+# define INTC_INTPR13_INTLEV_SIZE    2
+# define INTC_INTPR13_OFFSET_OFFSET  0
+# define INTC_INTPR13_OFFSET_SIZE    24
+#define INTC_INTREQ13                0x134
+# define INTC_INTREQ13_IREQUEST416_OFFSET 0
+# define INTC_INTREQ13_IREQUEST416_SIZE 1
+#define INTC_INTPR14                 0x38
+# define INTC_INTPR14_INTLEV_OFFSET  30
+# define INTC_INTPR14_INTLEV_SIZE    2
+# define INTC_INTPR14_OFFSET_OFFSET  0
+# define INTC_INTPR14_OFFSET_SIZE    24
+#define INTC_INTREQ14                0x138
+# define INTC_INTREQ14_IREQUEST448_OFFSET 0
+# define INTC_INTREQ14_IREQUEST448_SIZE 1
+#define INTC_INTPR15                 0x3c
+# define INTC_INTPR15_INTLEV_OFFSET  30
+# define INTC_INTPR15_INTLEV_SIZE    2
+# define INTC_INTPR15_OFFSET_OFFSET  0
+# define INTC_INTPR15_OFFSET_SIZE    24
+#define INTC_INTREQ15                0x13c
+# define INTC_INTREQ15_IREQUEST480_OFFSET 0
+# define INTC_INTREQ15_IREQUEST480_SIZE 1
+#define INTC_INTPR16                 0x40
+# define INTC_INTPR16_INTLEV_OFFSET  30
+# define INTC_INTPR16_INTLEV_SIZE    2
+# define INTC_INTPR16_OFFSET_OFFSET  0
+# define INTC_INTPR16_OFFSET_SIZE    24
+#define INTC_INTREQ16                0x140
+# define INTC_INTREQ16_IREQUEST512_OFFSET 0
+# define INTC_INTREQ16_IREQUEST512_SIZE 1
+#define INTC_INTPR17                 0x44
+# define INTC_INTPR17_INTLEV_OFFSET  30
+# define INTC_INTPR17_INTLEV_SIZE    2
+# define INTC_INTPR17_OFFSET_OFFSET  0
+# define INTC_INTPR17_OFFSET_SIZE    24
+#define INTC_INTREQ17                0x144
+# define INTC_INTREQ17_IREQUEST544_OFFSET 0
+# define INTC_INTREQ17_IREQUEST544_SIZE 1
+#define INTC_INTPR18                 0x48
+# define INTC_INTPR18_INTLEV_OFFSET  30
+# define INTC_INTPR18_INTLEV_SIZE    2
+# define INTC_INTPR18_OFFSET_OFFSET  0
+# define INTC_INTPR18_OFFSET_SIZE    24
+#define INTC_INTREQ18                0x148
+# define INTC_INTREQ18_IREQUEST576_OFFSET 0
+# define INTC_INTREQ18_IREQUEST576_SIZE 1
+#define INTC_INTPR19                 0x4c
+# define INTC_INTPR19_INTLEV_OFFSET  30
+# define INTC_INTPR19_INTLEV_SIZE    2
+# define INTC_INTPR19_OFFSET_OFFSET  0
+# define INTC_INTPR19_OFFSET_SIZE    24
+#define INTC_INTREQ19                0x14c
+# define INTC_INTREQ19_IREQUEST608_OFFSET 0
+# define INTC_INTREQ19_IREQUEST608_SIZE 1
+# define INTC_INTREQ19_IREQUEST609_OFFSET 1
+# define INTC_INTREQ19_IREQUEST609_SIZE 1
+# define INTC_INTREQ19_IREQUEST610_OFFSET 2
+# define INTC_INTREQ19_IREQUEST610_SIZE 1
+# define INTC_INTREQ19_IREQUEST611_OFFSET 3
+# define INTC_INTREQ19_IREQUEST611_SIZE 1
+#define INTC_INTPR20                 0x50
+# define INTC_INTPR20_INTLEV_OFFSET  30
+# define INTC_INTPR20_INTLEV_SIZE    2
+# define INTC_INTPR20_OFFSET_OFFSET  0
+# define INTC_INTPR20_OFFSET_SIZE    24
+#define INTC_INTREQ20                0x150
+# define INTC_INTREQ20_IREQUEST640_OFFSET 0
+# define INTC_INTREQ20_IREQUEST640_SIZE 1
+#define INTC_INTPR21                 0x54
+# define INTC_INTPR21_INTLEV_OFFSET  30
+# define INTC_INTPR21_INTLEV_SIZE    2
+# define INTC_INTPR21_OFFSET_OFFSET  0
+# define INTC_INTPR21_OFFSET_SIZE    24
+#define INTC_INTREQ21                0x154
+# define INTC_INTREQ21_IREQUEST672_OFFSET 0
+# define INTC_INTREQ21_IREQUEST672_SIZE 1
+#define INTC_INTPR22                 0x58
+# define INTC_INTPR22_INTLEV_OFFSET  30
+# define INTC_INTPR22_INTLEV_SIZE    2
+# define INTC_INTPR22_OFFSET_OFFSET  0
+# define INTC_INTPR22_OFFSET_SIZE    24
+#define INTC_INTREQ22                0x158
+# define INTC_INTREQ22_IREQUEST704_OFFSET 0
+# define INTC_INTREQ22_IREQUEST704_SIZE 1
+# define INTC_INTREQ22_IREQUEST705_OFFSET 1
+# define INTC_INTREQ22_IREQUEST705_SIZE 1
+# define INTC_INTREQ22_IREQUEST706_OFFSET 2
+# define INTC_INTREQ22_IREQUEST706_SIZE 1
+#define INTC_INTPR23                 0x5c
+# define INTC_INTPR23_INTLEV_OFFSET  30
+# define INTC_INTPR23_INTLEV_SIZE    2
+# define INTC_INTPR23_OFFSET_OFFSET  0
+# define INTC_INTPR23_OFFSET_SIZE    24
+#define INTC_INTREQ23                0x15c
+# define INTC_INTREQ23_IREQUEST736_OFFSET 0
+# define INTC_INTREQ23_IREQUEST736_SIZE 1
+# define INTC_INTREQ23_IREQUEST737_OFFSET 1
+# define INTC_INTREQ23_IREQUEST737_SIZE 1
+# define INTC_INTREQ23_IREQUEST738_OFFSET 2
+# define INTC_INTREQ23_IREQUEST738_SIZE 1
+#define INTC_INTPR24                 0x60
+# define INTC_INTPR24_INTLEV_OFFSET  30
+# define INTC_INTPR24_INTLEV_SIZE    2
+# define INTC_INTPR24_OFFSET_OFFSET  0
+# define INTC_INTPR24_OFFSET_SIZE    24
+#define INTC_INTREQ24                0x160
+# define INTC_INTREQ24_IREQUEST768_OFFSET 0
+# define INTC_INTREQ24_IREQUEST768_SIZE 1
+#define INTC_INTPR25                 0x64
+# define INTC_INTPR25_INTLEV_OFFSET  30
+# define INTC_INTPR25_INTLEV_SIZE    2
+# define INTC_INTPR25_OFFSET_OFFSET  0
+# define INTC_INTPR25_OFFSET_SIZE    24
+#define INTC_INTREQ25                0x164
+# define INTC_INTREQ25_IREQUEST800_OFFSET 0
+# define INTC_INTREQ25_IREQUEST800_SIZE 1
+#define INTC_INTPR26                 0x68
+# define INTC_INTPR26_INTLEV_OFFSET  30
+# define INTC_INTPR26_INTLEV_SIZE    2
+# define INTC_INTPR26_OFFSET_OFFSET  0
+# define INTC_INTPR26_OFFSET_SIZE    24
+#define INTC_INTREQ26                0x168
+# define INTC_INTREQ26_IREQUEST832_OFFSET 0
+# define INTC_INTREQ26_IREQUEST832_SIZE 1
+#define INTC_INTPR27                 0x6c
+# define INTC_INTPR27_INTLEV_OFFSET  30
+# define INTC_INTPR27_INTLEV_SIZE    2
+# define INTC_INTPR27_OFFSET_OFFSET  0
+# define INTC_INTPR27_OFFSET_SIZE    24
+#define INTC_INTREQ27                0x16c
+# define INTC_INTREQ27_IREQUEST864_OFFSET 0
+# define INTC_INTREQ27_IREQUEST864_SIZE 1
+#define INTC_INTPR28                 0x70
+# define INTC_INTPR28_INTLEV_OFFSET  30
+# define INTC_INTPR28_INTLEV_SIZE    2
+# define INTC_INTPR28_OFFSET_OFFSET  0
+# define INTC_INTPR28_OFFSET_SIZE    24
+#define INTC_INTREQ28                0x170
+# define INTC_INTREQ28_IREQUEST896_OFFSET 0
+# define INTC_INTREQ28_IREQUEST896_SIZE 1
+#define INTC_INTPR29                 0x74
+# define INTC_INTPR29_INTLEV_OFFSET  30
+# define INTC_INTPR29_INTLEV_SIZE    2
+# define INTC_INTPR29_OFFSET_OFFSET  0
+# define INTC_INTPR29_OFFSET_SIZE    24
+#define INTC_INTREQ29                0x174
+# define INTC_INTREQ29_IREQUEST928_OFFSET 0
+# define INTC_INTREQ29_IREQUEST928_SIZE 1
+#define INTC_INTPR30                 0x78
+# define INTC_INTPR30_INTLEV_OFFSET  30
+# define INTC_INTPR30_INTLEV_SIZE    2
+# define INTC_INTPR30_OFFSET_OFFSET  0
+# define INTC_INTPR30_OFFSET_SIZE    24
+#define INTC_INTREQ30                0x178
+# define INTC_INTREQ30_IREQUEST960_OFFSET 0
+# define INTC_INTREQ30_IREQUEST960_SIZE 1
+#define INTC_INTPR31                 0x7c
+# define INTC_INTPR31_INTLEV_OFFSET  30
+# define INTC_INTPR31_INTLEV_SIZE    2
+# define INTC_INTPR31_OFFSET_OFFSET  0
+# define INTC_INTPR31_OFFSET_SIZE    24
+#define INTC_INTREQ31                0x17c
+# define INTC_INTREQ31_IREQUEST992_OFFSET 0
+# define INTC_INTREQ31_IREQUEST992_SIZE 1
+#define INTC_INTPR32                 0x80
+# define INTC_INTPR32_INTLEV_OFFSET  30
+# define INTC_INTPR32_INTLEV_SIZE    2
+# define INTC_INTPR32_OFFSET_OFFSET  0
+# define INTC_INTPR32_OFFSET_SIZE    24
+#define INTC_INTREQ32                0x180
+# define INTC_INTREQ32_IREQUEST1024_OFFSET 0
+# define INTC_INTREQ32_IREQUEST1024_SIZE 1
+#define INTC_INTCAUSE0               0x20c
+# define INTC_INTCAUSE0_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE0_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE1               0x208
+# define INTC_INTCAUSE1_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE1_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE2               0x204
+# define INTC_INTCAUSE2_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE2_CAUSEGRP_SIZE 6
+#define INTC_INTCAUSE3               0x200
+# define INTC_INTCAUSE3_CAUSEGRP_OFFSET 0
+# define INTC_INTCAUSE3_CAUSEGRP_SIZE 6
+
+#define INTC_BIT(name)               (1 << INTC_##name##_OFFSET)
+#define INTC_MKBF(name, value)       (((value) & ((1 << INTC_##name##_SIZE) - 1)) << INTC_##name##_OFFSET)
+#define INTC_GETBF(name, value)      (((value) >> INTC_##name##_OFFSET) & ((1 << INTC_##name##_SIZE) - 1))
+
+#define intc_readl(port,reg)         readl((port)->regs + INTC_##reg)
+#define intc_writel(port,reg,value)  writel((value), (port)->regs + INTC_##reg)
+
+#endif /* __ASM_AVR32_PERIHP_INTC_H__ */
diff --git a/arch/avr32/mach-at32ap/pio.c b/arch/avr32/mach-at32ap/pio.c
new file mode 100644
index 000000000000..d3aabfca8598
--- /dev/null
+++ b/arch/avr32/mach-at32ap/pio.c
@@ -0,0 +1,118 @@
+/*
+ * Atmel PIO2 Port Multiplexer support
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/clk.h>
+#include <linux/debugfs.h>
+#include <linux/fs.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+
+#include <asm/arch/portmux.h>
+
+#include "pio.h"
+
+#define MAX_NR_PIO_DEVICES		8
+
+struct pio_device {
+	void __iomem *regs;
+	const struct platform_device *pdev;
+	struct clk *clk;
+	u32 alloc_mask;
+	char name[32];
+};
+
+static struct pio_device pio_dev[MAX_NR_PIO_DEVICES];
+
+void portmux_set_func(unsigned int portmux_id, unsigned int pin_id,
+		      unsigned int function_id)
+{
+	struct pio_device *pio;
+	u32 mask = 1 << pin_id;
+
+	BUG_ON(portmux_id >= MAX_NR_PIO_DEVICES);
+
+	pio = &pio_dev[portmux_id];
+
+	if (function_id)
+		pio_writel(pio, BSR, mask);
+	else
+		pio_writel(pio, ASR, mask);
+	pio_writel(pio, PDR, mask);
+}
+
+static int __init pio_probe(struct platform_device *pdev)
+{
+	struct pio_device *pio = NULL;
+
+	BUG_ON(pdev->id >= MAX_NR_PIO_DEVICES);
+	pio = &pio_dev[pdev->id];
+	BUG_ON(!pio->regs);
+
+	/* TODO: Interrupts */
+
+	platform_set_drvdata(pdev, pio);
+
+	printk(KERN_INFO "%s: Atmel Port Multiplexer at 0x%p (irq %d)\n",
+	       pio->name, pio->regs, platform_get_irq(pdev, 0));
+
+	return 0;
+}
+
+static struct platform_driver pio_driver = {
+	.probe		= pio_probe,
+	.driver		= {
+		.name		= "pio",
+	},
+};
+
+static int __init pio_init(void)
+{
+	return platform_driver_register(&pio_driver);
+}
+subsys_initcall(pio_init);
+
+void __init at32_init_pio(struct platform_device *pdev)
+{
+	struct resource *regs;
+	struct pio_device *pio;
+
+	if (pdev->id > MAX_NR_PIO_DEVICES) {
+		dev_err(&pdev->dev, "only %d PIO devices supported\n",
+			MAX_NR_PIO_DEVICES);
+		return;
+	}
+
+	pio = &pio_dev[pdev->id];
+	snprintf(pio->name, sizeof(pio->name), "pio%d", pdev->id);
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!regs) {
+		dev_err(&pdev->dev, "no mmio resource defined\n");
+		return;
+	}
+
+	pio->clk = clk_get(&pdev->dev, "mck");
+	if (IS_ERR(pio->clk))
+		/*
+		 * This is a fatal error, but if we continue we might
+		 * be so lucky that we manage to initialize the
+		 * console and display this message...
+		 */
+		dev_err(&pdev->dev, "no mck clock defined\n");
+	else
+		clk_enable(pio->clk);
+
+	pio->pdev = pdev;
+	pio->regs = ioremap(regs->start, regs->end - regs->start + 1);
+
+	pio_writel(pio, ODR, ~0UL);
+	pio_writel(pio, PER, ~0UL);
+}
diff --git a/arch/avr32/mach-at32ap/pio.h b/arch/avr32/mach-at32ap/pio.h
new file mode 100644
index 000000000000..cfea12351599
--- /dev/null
+++ b/arch/avr32/mach-at32ap/pio.h
@@ -0,0 +1,178 @@
+/*
+ * Atmel PIO2 Port Multiplexer support
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ARCH_AVR32_AT32AP_PIO_H__
+#define __ARCH_AVR32_AT32AP_PIO_H__
+
+/* PIO register offsets */
+#define PIO_PER                                0x0000
+#define PIO_PDR                                0x0004
+#define PIO_PSR                                0x0008
+#define PIO_OER                                0x0010
+#define PIO_ODR                                0x0014
+#define PIO_OSR                                0x0018
+#define PIO_IFER                               0x0020
+#define PIO_IFDR                               0x0024
+#define PIO_ISFR                               0x0028
+#define PIO_SODR                               0x0030
+#define PIO_CODR                               0x0034
+#define PIO_ODSR                               0x0038
+#define PIO_PDSR                               0x003c
+#define PIO_IER                                0x0040
+#define PIO_IDR                                0x0044
+#define PIO_IMR                                0x0048
+#define PIO_ISR                                0x004c
+#define PIO_MDER                               0x0050
+#define PIO_MDDR                               0x0054
+#define PIO_MDSR                               0x0058
+#define PIO_PUDR                               0x0060
+#define PIO_PUER                               0x0064
+#define PIO_PUSR                               0x0068
+#define PIO_ASR                                0x0070
+#define PIO_BSR                                0x0074
+#define PIO_ABSR                               0x0078
+#define PIO_OWER                               0x00a0
+#define PIO_OWDR                               0x00a4
+#define PIO_OWSR                               0x00a8
+
+/* Bitfields in PER */
+
+/* Bitfields in PDR */
+
+/* Bitfields in PSR */
+
+/* Bitfields in OER */
+
+/* Bitfields in ODR */
+
+/* Bitfields in OSR */
+
+/* Bitfields in IFER */
+
+/* Bitfields in IFDR */
+
+/* Bitfields in ISFR */
+
+/* Bitfields in SODR */
+
+/* Bitfields in CODR */
+
+/* Bitfields in ODSR */
+
+/* Bitfields in PDSR */
+
+/* Bitfields in IER */
+
+/* Bitfields in IDR */
+
+/* Bitfields in IMR */
+
+/* Bitfields in ISR */
+
+/* Bitfields in MDER */
+
+/* Bitfields in MDDR */
+
+/* Bitfields in MDSR */
+
+/* Bitfields in PUDR */
+
+/* Bitfields in PUER */
+
+/* Bitfields in PUSR */
+
+/* Bitfields in ASR */
+
+/* Bitfields in BSR */
+
+/* Bitfields in ABSR */
+#define PIO_P0_OFFSET                          0
+#define PIO_P0_SIZE                            1
+#define PIO_P1_OFFSET                          1
+#define PIO_P1_SIZE                            1
+#define PIO_P2_OFFSET                          2
+#define PIO_P2_SIZE                            1
+#define PIO_P3_OFFSET                          3
+#define PIO_P3_SIZE                            1
+#define PIO_P4_OFFSET                          4
+#define PIO_P4_SIZE                            1
+#define PIO_P5_OFFSET                          5
+#define PIO_P5_SIZE                            1
+#define PIO_P6_OFFSET                          6
+#define PIO_P6_SIZE                            1
+#define PIO_P7_OFFSET                          7
+#define PIO_P7_SIZE                            1
+#define PIO_P8_OFFSET                          8
+#define PIO_P8_SIZE                            1
+#define PIO_P9_OFFSET                          9
+#define PIO_P9_SIZE                            1
+#define PIO_P10_OFFSET                         10
+#define PIO_P10_SIZE                           1
+#define PIO_P11_OFFSET                         11
+#define PIO_P11_SIZE                           1
+#define PIO_P12_OFFSET                         12
+#define PIO_P12_SIZE                           1
+#define PIO_P13_OFFSET                         13
+#define PIO_P13_SIZE                           1
+#define PIO_P14_OFFSET                         14
+#define PIO_P14_SIZE                           1
+#define PIO_P15_OFFSET                         15
+#define PIO_P15_SIZE                           1
+#define PIO_P16_OFFSET                         16
+#define PIO_P16_SIZE                           1
+#define PIO_P17_OFFSET                         17
+#define PIO_P17_SIZE                           1
+#define PIO_P18_OFFSET                         18
+#define PIO_P18_SIZE                           1
+#define PIO_P19_OFFSET                         19
+#define PIO_P19_SIZE                           1
+#define PIO_P20_OFFSET                         20
+#define PIO_P20_SIZE                           1
+#define PIO_P21_OFFSET                         21
+#define PIO_P21_SIZE                           1
+#define PIO_P22_OFFSET                         22
+#define PIO_P22_SIZE                           1
+#define PIO_P23_OFFSET                         23
+#define PIO_P23_SIZE                           1
+#define PIO_P24_OFFSET                         24
+#define PIO_P24_SIZE                           1
+#define PIO_P25_OFFSET                         25
+#define PIO_P25_SIZE                           1
+#define PIO_P26_OFFSET                         26
+#define PIO_P26_SIZE                           1
+#define PIO_P27_OFFSET                         27
+#define PIO_P27_SIZE                           1
+#define PIO_P28_OFFSET                         28
+#define PIO_P28_SIZE                           1
+#define PIO_P29_OFFSET                         29
+#define PIO_P29_SIZE                           1
+#define PIO_P30_OFFSET                         30
+#define PIO_P30_SIZE                           1
+#define PIO_P31_OFFSET                         31
+#define PIO_P31_SIZE                           1
+
+/* Bitfields in OWER */
+
+/* Bitfields in OWDR */
+
+/* Bitfields in OWSR */
+
+/* Bit manipulation macros */
+#define PIO_BIT(name)                          (1 << PIO_##name##_OFFSET)
+#define PIO_BF(name,value)                     (((value) & ((1 << PIO_##name##_SIZE) - 1)) << PIO_##name##_OFFSET)
+#define PIO_BFEXT(name,value)                  (((value) >> PIO_##name##_OFFSET) & ((1 << PIO_##name##_SIZE) - 1))
+#define PIO_BFINS(name,value,old)              (((old) & ~(((1 << PIO_##name##_SIZE) - 1) << PIO_##name##_OFFSET)) | PIO_BF(name,value))
+
+/* Register access macros */
+#define pio_readl(port,reg)                    readl((port)->regs + PIO_##reg)
+#define pio_writel(port,reg,value)             writel((value), (port)->regs + PIO_##reg)
+
+void at32_init_pio(struct platform_device *pdev);
+
+#endif /* __ARCH_AVR32_AT32AP_PIO_H__ */
diff --git a/arch/avr32/mach-at32ap/sm.c b/arch/avr32/mach-at32ap/sm.c
new file mode 100644
index 000000000000..03306eb0345e
--- /dev/null
+++ b/arch/avr32/mach-at32ap/sm.c
@@ -0,0 +1,289 @@
+/*
+ * System Manager driver for AT32AP CPUs
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/random.h>
+#include <linux/spinlock.h>
+
+#include <asm/intc.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+
+#include <asm/arch/sm.h>
+
+#include "sm.h"
+
+#define SM_EIM_IRQ_RESOURCE	1
+#define SM_PM_IRQ_RESOURCE	2
+#define SM_RTC_IRQ_RESOURCE	3
+
+#define to_eim(irqc) container_of(irqc, struct at32_sm, irqc)
+
+struct at32_sm system_manager;
+
+int __init at32_sm_init(void)
+{
+	struct resource *regs;
+	struct at32_sm *sm = &system_manager;
+	int ret = -ENXIO;
+
+	regs = platform_get_resource(&at32_sm_device, IORESOURCE_MEM, 0);
+	if (!regs)
+		goto fail;
+
+	spin_lock_init(&sm->lock);
+	sm->pdev = &at32_sm_device;
+
+	ret = -ENOMEM;
+	sm->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!sm->regs)
+		goto fail;
+
+	return 0;
+
+fail:
+	printk(KERN_ERR "Failed to initialize System Manager: %d\n", ret);
+	return ret;
+}
+
+/*
+ * External Interrupt Module (EIM).
+ *
+ * EIM gets level- or edge-triggered interrupts of either polarity
+ * from the outside and converts it to active-high level-triggered
+ * interrupts that the internal interrupt controller can handle. EIM
+ * also provides masking/unmasking of interrupts, as well as
+ * acknowledging of edge-triggered interrupts.
+ */
+
+static irqreturn_t spurious_eim_interrupt(int irq, void *dev_id,
+					  struct pt_regs *regs)
+{
+	printk(KERN_WARNING "Spurious EIM interrupt %d\n", irq);
+	disable_irq(irq);
+	return IRQ_NONE;
+}
+
+static struct irqaction eim_spurious_action = {
+	.handler = spurious_eim_interrupt,
+};
+
+static irqreturn_t eim_handle_irq(int irq, void *dev_id, struct pt_regs *regs)
+{
+	struct irq_controller * irqc = dev_id;
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long pending;
+
+	/*
+	 * No need to disable interrupts globally.  The interrupt
+	 * level relevant to this group must be masked all the time,
+	 * so we know that this particular EIM instance will not be
+	 * re-entered.
+	 */
+	spin_lock(&sm->lock);
+
+	pending = intc_get_pending(sm->irqc.irq_group);
+	if (unlikely(!pending)) {
+		printk(KERN_ERR "EIM (group %u): No interrupts pending!\n",
+		       sm->irqc.irq_group);
+		goto unlock;
+	}
+
+	do {
+		struct irqaction *action;
+		unsigned int i;
+
+		i = fls(pending) - 1;
+		pending &= ~(1 << i);
+		action = sm->action[i];
+
+		/* Acknowledge the interrupt */
+		sm_writel(sm, EIM_ICR, 1 << i);
+
+		spin_unlock(&sm->lock);
+
+		if (action->flags & SA_INTERRUPT)
+			local_irq_disable();
+		action->handler(sm->irqc.first_irq + i, action->dev_id, regs);
+		local_irq_enable();
+		spin_lock(&sm->lock);
+		if (action->flags & SA_SAMPLE_RANDOM)
+			add_interrupt_randomness(sm->irqc.first_irq + i);
+	} while (pending);
+
+unlock:
+	spin_unlock(&sm->lock);
+	return IRQ_HANDLED;
+}
+
+static void eim_mask(struct irq_controller *irqc, unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned int i;
+
+	i = irq - sm->irqc.first_irq;
+	sm_writel(sm, EIM_IDR, 1 << i);
+}
+
+static void eim_unmask(struct irq_controller *irqc, unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned int i;
+
+	i = irq - sm->irqc.first_irq;
+	sm_writel(sm, EIM_IER, 1 << i);
+}
+
+static int eim_setup(struct irq_controller *irqc, unsigned int irq,
+		struct irqaction *action)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	sm->action[irq - sm->irqc.first_irq] = action;
+	/* Acknowledge earlier interrupts */
+	sm_writel(sm, EIM_ICR, (1<<(irq - sm->irqc.first_irq)));
+	eim_unmask(irqc, irq);
+	return 0;
+}
+
+static void eim_free(struct irq_controller *irqc, unsigned int irq,
+		void *dev)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	eim_mask(irqc, irq);
+	sm->action[irq - sm->irqc.first_irq] = &eim_spurious_action;
+}
+
+static int eim_set_type(struct irq_controller *irqc, unsigned int irq,
+			unsigned int type)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long flags;
+	u32 value, pattern;
+
+	spin_lock_irqsave(&sm->lock, flags);
+
+	pattern = 1 << (irq - sm->irqc.first_irq);
+
+	value = sm_readl(sm, EIM_MODE);
+	if (type & IRQ_TYPE_LEVEL)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_MODE, value);
+	value = sm_readl(sm, EIM_EDGE);
+	if (type & IRQ_EDGE_RISING)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_EDGE, value);
+	value = sm_readl(sm, EIM_LEVEL);
+	if (type & IRQ_LEVEL_HIGH)
+		value |= pattern;
+	else
+		value &= ~pattern;
+	sm_writel(sm, EIM_LEVEL, value);
+
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	return 0;
+}
+
+static unsigned int eim_get_type(struct irq_controller *irqc,
+				 unsigned int irq)
+{
+	struct at32_sm *sm = to_eim(irqc);
+	unsigned long flags;
+	unsigned int type = 0;
+	u32 mode, edge, level, pattern;
+
+	pattern = 1 << (irq - sm->irqc.first_irq);
+
+	spin_lock_irqsave(&sm->lock, flags);
+	mode = sm_readl(sm, EIM_MODE);
+	edge = sm_readl(sm, EIM_EDGE);
+	level = sm_readl(sm, EIM_LEVEL);
+	spin_unlock_irqrestore(&sm->lock, flags);
+
+	if (mode & pattern)
+		type |= IRQ_TYPE_LEVEL;
+	if (edge & pattern)
+		type |= IRQ_EDGE_RISING;
+	if (level & pattern)
+		type |= IRQ_LEVEL_HIGH;
+
+	return type;
+}
+
+static struct irq_controller_class eim_irq_class = {
+	.typename	= "EIM",
+	.handle		= eim_handle_irq,
+	.setup		= eim_setup,
+	.free		= eim_free,
+	.mask		= eim_mask,
+	.unmask		= eim_unmask,
+	.set_type	= eim_set_type,
+	.get_type	= eim_get_type,
+};
+
+static int __init eim_init(void)
+{
+	struct at32_sm *sm = &system_manager;
+	unsigned int i;
+	u32 pattern;
+	int ret;
+
+	/*
+	 * The EIM is really the same module as SM, so register
+	 * mapping, etc. has been taken care of already.
+	 */
+
+	/*
+	 * Find out how many interrupt lines that are actually
+	 * implemented in hardware.
+	 */
+	sm_writel(sm, EIM_IDR, ~0UL);
+	sm_writel(sm, EIM_MODE, ~0UL);
+	pattern = sm_readl(sm, EIM_MODE);
+	sm->irqc.nr_irqs = fls(pattern);
+
+	ret = -ENOMEM;
+	sm->action = kmalloc(sizeof(*sm->action) * sm->irqc.nr_irqs,
+			     GFP_KERNEL);
+	if (!sm->action)
+		goto out;
+
+	for (i = 0; i < sm->irqc.nr_irqs; i++)
+		sm->action[i] = &eim_spurious_action;
+
+	spin_lock_init(&sm->lock);
+	sm->irqc.irq_group = sm->pdev->resource[SM_EIM_IRQ_RESOURCE].start;
+	sm->irqc.class = &eim_irq_class;
+
+	ret = intc_register_controller(&sm->irqc);
+	if (ret < 0)
+		goto out_free_actions;
+
+	printk("EIM: External Interrupt Module at 0x%p, IRQ group %u\n",
+	       sm->regs, sm->irqc.irq_group);
+	printk("EIM: Handling %u external IRQs, starting with IRQ%u\n",
+	       sm->irqc.nr_irqs, sm->irqc.first_irq);
+
+	return 0;
+
+out_free_actions:
+	kfree(sm->action);
+out:
+	return ret;
+}
+arch_initcall(eim_init);
diff --git a/arch/avr32/mach-at32ap/sm.h b/arch/avr32/mach-at32ap/sm.h
new file mode 100644
index 000000000000..27565822ae2a
--- /dev/null
+++ b/arch/avr32/mach-at32ap/sm.h
@@ -0,0 +1,240 @@
+/*
+ * Register definitions for SM
+ *
+ * System Manager
+ */
+#ifndef __ASM_AVR32_SM_H__
+#define __ASM_AVR32_SM_H__
+
+/* SM register offsets */
+#define SM_PM_MCCTRL                            0x0000
+#define SM_PM_CKSEL                             0x0004
+#define SM_PM_CPU_MASK                          0x0008
+#define SM_PM_HSB_MASK                          0x000c
+#define SM_PM_PBA_MASK                         0x0010
+#define SM_PM_PBB_MASK                         0x0014
+#define SM_PM_PLL0                              0x0020
+#define SM_PM_PLL1                              0x0024
+#define SM_PM_VCTRL                             0x0030
+#define SM_PM_VMREF                             0x0034
+#define SM_PM_VMV                               0x0038
+#define SM_PM_IER                               0x0040
+#define SM_PM_IDR                               0x0044
+#define SM_PM_IMR                               0x0048
+#define SM_PM_ISR                               0x004c
+#define SM_PM_ICR                               0x0050
+#define SM_PM_GCCTRL                            0x0060
+#define SM_RTC_CTRL                             0x0080
+#define SM_RTC_VAL                              0x0084
+#define SM_RTC_TOP                              0x0088
+#define SM_RTC_IER                              0x0090
+#define SM_RTC_IDR                              0x0094
+#define SM_RTC_IMR                              0x0098
+#define SM_RTC_ISR                              0x009c
+#define SM_RTC_ICR                              0x00a0
+#define SM_WDT_CTRL                             0x00b0
+#define SM_WDT_CLR                              0x00b4
+#define SM_WDT_EXT                              0x00b8
+#define SM_RC_RCAUSE                            0x00c0
+#define SM_EIM_IER                              0x0100
+#define SM_EIM_IDR                              0x0104
+#define SM_EIM_IMR                              0x0108
+#define SM_EIM_ISR                              0x010c
+#define SM_EIM_ICR                              0x0110
+#define SM_EIM_MODE                             0x0114
+#define SM_EIM_EDGE                             0x0118
+#define SM_EIM_LEVEL                            0x011c
+#define SM_EIM_TEST                             0x0120
+#define SM_EIM_NMIC                             0x0124
+
+/* Bitfields in PM_MCCTRL */
+
+/* Bitfields in PM_CKSEL */
+#define SM_CPUSEL_OFFSET                        0
+#define SM_CPUSEL_SIZE                          3
+#define SM_CPUDIV_OFFSET                        7
+#define SM_CPUDIV_SIZE                          1
+#define SM_HSBSEL_OFFSET                        8
+#define SM_HSBSEL_SIZE                          3
+#define SM_HSBDIV_OFFSET                        15
+#define SM_HSBDIV_SIZE                          1
+#define SM_PBASEL_OFFSET                       16
+#define SM_PBASEL_SIZE                         3
+#define SM_PBADIV_OFFSET                       23
+#define SM_PBADIV_SIZE                         1
+#define SM_PBBSEL_OFFSET                       24
+#define SM_PBBSEL_SIZE                         3
+#define SM_PBBDIV_OFFSET                       31
+#define SM_PBBDIV_SIZE                         1
+
+/* Bitfields in PM_CPU_MASK */
+
+/* Bitfields in PM_HSB_MASK */
+
+/* Bitfields in PM_PBA_MASK */
+
+/* Bitfields in PM_PBB_MASK */
+
+/* Bitfields in PM_PLL0 */
+#define SM_PLLEN_OFFSET                         0
+#define SM_PLLEN_SIZE                           1
+#define SM_PLLOSC_OFFSET                        1
+#define SM_PLLOSC_SIZE                          1
+#define SM_PLLOPT_OFFSET                        2
+#define SM_PLLOPT_SIZE                          3
+#define SM_PLLDIV_OFFSET                        8
+#define SM_PLLDIV_SIZE                          8
+#define SM_PLLMUL_OFFSET                        16
+#define SM_PLLMUL_SIZE                          8
+#define SM_PLLCOUNT_OFFSET                      24
+#define SM_PLLCOUNT_SIZE                        6
+#define SM_PLLTEST_OFFSET                       31
+#define SM_PLLTEST_SIZE                         1
+
+/* Bitfields in PM_PLL1 */
+
+/* Bitfields in PM_VCTRL */
+#define SM_VAUTO_OFFSET                         0
+#define SM_VAUTO_SIZE                           1
+#define SM_PM_VCTRL_VAL_OFFSET                  8
+#define SM_PM_VCTRL_VAL_SIZE                    7
+
+/* Bitfields in PM_VMREF */
+#define SM_REFSEL_OFFSET                        0
+#define SM_REFSEL_SIZE                          4
+
+/* Bitfields in PM_VMV */
+#define SM_PM_VMV_VAL_OFFSET                    0
+#define SM_PM_VMV_VAL_SIZE                      8
+
+/* Bitfields in PM_IER */
+
+/* Bitfields in PM_IDR */
+
+/* Bitfields in PM_IMR */
+
+/* Bitfields in PM_ISR */
+
+/* Bitfields in PM_ICR */
+#define SM_LOCK0_OFFSET                         0
+#define SM_LOCK0_SIZE                           1
+#define SM_LOCK1_OFFSET                         1
+#define SM_LOCK1_SIZE                           1
+#define SM_WAKE_OFFSET                          2
+#define SM_WAKE_SIZE                            1
+#define SM_VOK_OFFSET                           3
+#define SM_VOK_SIZE                             1
+#define SM_VMRDY_OFFSET                         4
+#define SM_VMRDY_SIZE                           1
+#define SM_CKRDY_OFFSET                         5
+#define SM_CKRDY_SIZE                           1
+
+/* Bitfields in PM_GCCTRL */
+#define SM_OSCSEL_OFFSET                        0
+#define SM_OSCSEL_SIZE                          1
+#define SM_PLLSEL_OFFSET                        1
+#define SM_PLLSEL_SIZE                          1
+#define SM_CEN_OFFSET                           2
+#define SM_CEN_SIZE                             1
+#define SM_CPC_OFFSET                           3
+#define SM_CPC_SIZE                             1
+#define SM_DIVEN_OFFSET                         4
+#define SM_DIVEN_SIZE                           1
+#define SM_DIV_OFFSET                           8
+#define SM_DIV_SIZE                             8
+
+/* Bitfields in RTC_CTRL */
+#define SM_PCLR_OFFSET                          1
+#define SM_PCLR_SIZE                            1
+#define SM_TOPEN_OFFSET                         2
+#define SM_TOPEN_SIZE                           1
+#define SM_CLKEN_OFFSET                         3
+#define SM_CLKEN_SIZE                           1
+#define SM_PSEL_OFFSET                          8
+#define SM_PSEL_SIZE                            16
+
+/* Bitfields in RTC_VAL */
+#define SM_RTC_VAL_VAL_OFFSET                   0
+#define SM_RTC_VAL_VAL_SIZE                     31
+
+/* Bitfields in RTC_TOP */
+#define SM_RTC_TOP_VAL_OFFSET                   0
+#define SM_RTC_TOP_VAL_SIZE                     32
+
+/* Bitfields in RTC_IER */
+
+/* Bitfields in RTC_IDR */
+
+/* Bitfields in RTC_IMR */
+
+/* Bitfields in RTC_ISR */
+
+/* Bitfields in RTC_ICR */
+#define SM_TOPI_OFFSET                          0
+#define SM_TOPI_SIZE                            1
+
+/* Bitfields in WDT_CTRL */
+#define SM_KEY_OFFSET                           24
+#define SM_KEY_SIZE                             8
+
+/* Bitfields in WDT_CLR */
+
+/* Bitfields in WDT_EXT */
+
+/* Bitfields in RC_RCAUSE */
+#define SM_POR_OFFSET                           0
+#define SM_POR_SIZE                             1
+#define SM_BOD_OFFSET                           1
+#define SM_BOD_SIZE                             1
+#define SM_EXT_OFFSET                           2
+#define SM_EXT_SIZE                             1
+#define SM_WDT_OFFSET                           3
+#define SM_WDT_SIZE                             1
+#define SM_NTAE_OFFSET                          4
+#define SM_NTAE_SIZE                            1
+#define SM_SERP_OFFSET                          5
+#define SM_SERP_SIZE                            1
+
+/* Bitfields in EIM_IER */
+
+/* Bitfields in EIM_IDR */
+
+/* Bitfields in EIM_IMR */
+
+/* Bitfields in EIM_ISR */
+
+/* Bitfields in EIM_ICR */
+
+/* Bitfields in EIM_MODE */
+
+/* Bitfields in EIM_EDGE */
+#define SM_INT0_OFFSET                          0
+#define SM_INT0_SIZE                            1
+#define SM_INT1_OFFSET                          1
+#define SM_INT1_SIZE                            1
+#define SM_INT2_OFFSET                          2
+#define SM_INT2_SIZE                            1
+#define SM_INT3_OFFSET                          3
+#define SM_INT3_SIZE                            1
+
+/* Bitfields in EIM_LEVEL */
+
+/* Bitfields in EIM_TEST */
+#define SM_TESTEN_OFFSET                        31
+#define SM_TESTEN_SIZE                          1
+
+/* Bitfields in EIM_NMIC */
+#define SM_EN_OFFSET                            0
+#define SM_EN_SIZE                              1
+
+/* Bit manipulation macros */
+#define SM_BIT(name)                            (1 << SM_##name##_OFFSET)
+#define SM_BF(name,value)                       (((value) & ((1 << SM_##name##_SIZE) - 1)) << SM_##name##_OFFSET)
+#define SM_BFEXT(name,value)                    (((value) >> SM_##name##_OFFSET) & ((1 << SM_##name##_SIZE) - 1))
+#define SM_BFINS(name,value,old)                (((old) & ~(((1 << SM_##name##_SIZE) - 1) << SM_##name##_OFFSET)) | SM_BF(name,value))
+
+/* Register access macros */
+#define sm_readl(port,reg)                      readl((port)->regs + SM_##reg)
+#define sm_writel(port,reg,value)               writel((value), (port)->regs + SM_##reg)
+
+#endif /* __ASM_AVR32_SM_H__ */
diff --git a/arch/avr32/mm/Makefile b/arch/avr32/mm/Makefile
new file mode 100644
index 000000000000..0066491f90d4
--- /dev/null
+++ b/arch/avr32/mm/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the Linux/AVR32 kernel.
+#
+
+obj-y				+= init.o clear_page.o copy_page.o dma-coherent.o
+obj-y				+= ioremap.o cache.o fault.o tlb.o
diff --git a/arch/avr32/mm/cache.c b/arch/avr32/mm/cache.c
new file mode 100644
index 000000000000..450515b245a0
--- /dev/null
+++ b/arch/avr32/mm/cache.c
@@ -0,0 +1,150 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/highmem.h>
+#include <linux/unistd.h>
+
+#include <asm/cacheflush.h>
+#include <asm/cachectl.h>
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+
+/*
+ * If you attempt to flush anything more than this, you need superuser
+ * privileges.  The value is completely arbitrary.
+ */
+#define CACHEFLUSH_MAX_LEN	1024
+
+void invalidate_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+
+	//printk("invalidate dcache: %p + %u\n", start, size);
+
+	/* You asked for it, you got it */
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		invalidate_dcache_line((void *)v);
+}
+
+void clean_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		clean_dcache_line((void *)v);
+	flush_write_buffer();
+}
+
+void flush_dcache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		flush_dcache_line((void *)v);
+	flush_write_buffer();
+}
+
+void invalidate_icache_region(void *start, size_t size)
+{
+	unsigned long v, begin, end, linesz;
+
+	linesz = boot_cpu_data.icache.linesz;
+	begin = (unsigned long)start & ~(linesz - 1);
+	end = ((unsigned long)start + size + linesz - 1) & ~(linesz - 1);
+
+	for (v = begin; v < end; v += linesz)
+		invalidate_icache_line((void *)v);
+}
+
+static inline void __flush_icache_range(unsigned long start, unsigned long end)
+{
+	unsigned long v, linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	for (v = start; v < end; v += linesz) {
+		clean_dcache_line((void *)v);
+		invalidate_icache_line((void *)v);
+	}
+
+	flush_write_buffer();
+}
+
+/*
+ * This one is called after a module has been loaded.
+ */
+void flush_icache_range(unsigned long start, unsigned long end)
+{
+	unsigned long linesz;
+
+	linesz = boot_cpu_data.dcache.linesz;
+	__flush_icache_range(start & ~(linesz - 1),
+			     (end + linesz - 1) & ~(linesz - 1));
+}
+
+/*
+ * This one is called from do_no_page(), do_swap_page() and install_page().
+ */
+void flush_icache_page(struct vm_area_struct *vma, struct page *page)
+{
+	if (vma->vm_flags & VM_EXEC) {
+		void *v = kmap(page);
+		__flush_icache_range((unsigned long)v, (unsigned long)v + PAGE_SIZE);
+		kunmap(v);
+	}
+}
+
+/*
+ * This one is used by copy_to_user_page()
+ */
+void flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
+			     unsigned long addr, int len)
+{
+	if (vma->vm_flags & VM_EXEC)
+		flush_icache_range(addr, addr + len);
+}
+
+asmlinkage int sys_cacheflush(int operation, void __user *addr, size_t len)
+{
+	int ret;
+
+	if (len > CACHEFLUSH_MAX_LEN) {
+		ret = -EPERM;
+		if (!capable(CAP_SYS_ADMIN))
+			goto out;
+	}
+
+	ret = -EFAULT;
+	if (!access_ok(VERIFY_WRITE, addr, len))
+		goto out;
+
+	switch (operation) {
+	case CACHE_IFLUSH:
+		flush_icache_range((unsigned long)addr,
+				   (unsigned long)addr + len);
+		ret = 0;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+out:
+	return ret;
+}
diff --git a/arch/avr32/mm/clear_page.S b/arch/avr32/mm/clear_page.S
new file mode 100644
index 000000000000..5d70dca00699
--- /dev/null
+++ b/arch/avr32/mm/clear_page.S
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/page.h>
+
+/*
+ * clear_page
+ * r12: P1 address (to)
+ */
+	.text
+	.global clear_page
+clear_page:
+	sub	r9, r12, -PAGE_SIZE
+	mov     r10, 0
+	mov	r11, 0
+0:      st.d    r12++, r10
+	cp      r12, r9
+	brne	0b
+	mov     pc, lr
diff --git a/arch/avr32/mm/copy_page.S b/arch/avr32/mm/copy_page.S
new file mode 100644
index 000000000000..c2b3752946b8
--- /dev/null
+++ b/arch/avr32/mm/copy_page.S
@@ -0,0 +1,28 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/linkage.h>
+#include <asm/page.h>
+
+/*
+ * copy_page
+ *
+ * r12		to (P1 address)
+ * r11		from (P1 address)
+ * r8-r10	scratch
+ */
+	.text
+	.global copy_page
+copy_page:
+	sub	r10, r11, -(1 << PAGE_SHIFT)
+	/* pref	r11[0] */
+1:	/* pref	r11[8] */
+	ld.d	r8, r11++
+	st.d	r12++, r8
+	cp	r11, r10
+	brlo	1b
+	mov	pc, lr
diff --git a/arch/avr32/mm/dma-coherent.c b/arch/avr32/mm/dma-coherent.c
new file mode 100644
index 000000000000..44ab8a7bdae2
--- /dev/null
+++ b/arch/avr32/mm/dma-coherent.c
@@ -0,0 +1,139 @@
+/*
+ *  Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/dma-mapping.h>
+
+#include <asm/addrspace.h>
+#include <asm/cacheflush.h>
+
+void dma_cache_sync(void *vaddr, size_t size, int direction)
+{
+	/*
+	 * No need to sync an uncached area
+	 */
+	if (PXSEG(vaddr) == P2SEG)
+		return;
+
+	switch (direction) {
+	case DMA_FROM_DEVICE:		/* invalidate only */
+		dma_cache_inv(vaddr, size);
+		break;
+	case DMA_TO_DEVICE:		/* writeback only */
+		dma_cache_wback(vaddr, size);
+		break;
+	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
+		dma_cache_wback_inv(vaddr, size);
+		break;
+	default:
+		BUG();
+	}
+}
+EXPORT_SYMBOL(dma_cache_sync);
+
+static struct page *__dma_alloc(struct device *dev, size_t size,
+				dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page, *free, *end;
+	int order;
+
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	page = alloc_pages(gfp, order);
+	if (!page)
+		return NULL;
+	split_page(page, order);
+
+	/*
+	 * When accessing physical memory with valid cache data, we
+	 * get a cache hit even if the virtual memory region is marked
+	 * as uncached.
+	 *
+	 * Since the memory is newly allocated, there is no point in
+	 * doing a writeback. If the previous owner cares, he should
+	 * have flushed the cache before releasing the memory.
+	 */
+	invalidate_dcache_region(phys_to_virt(page_to_phys(page)), size);
+
+	*handle = page_to_bus(page);
+	free = page + (size >> PAGE_SHIFT);
+	end = page + (1 << order);
+
+	/*
+	 * Free any unused pages
+	 */
+	while (free < end) {
+		__free_page(free);
+		free++;
+	}
+
+	return page;
+}
+
+static void __dma_free(struct device *dev, size_t size,
+		       struct page *page, dma_addr_t handle)
+{
+	struct page *end = page + (PAGE_ALIGN(size) >> PAGE_SHIFT);
+
+	while (page < end)
+		__free_page(page++);
+}
+
+void *dma_alloc_coherent(struct device *dev, size_t size,
+			 dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page;
+	void *ret = NULL;
+
+	page = __dma_alloc(dev, size, handle, gfp);
+	if (page)
+		ret = phys_to_uncached(page_to_phys(page));
+
+	return ret;
+}
+EXPORT_SYMBOL(dma_alloc_coherent);
+
+void dma_free_coherent(struct device *dev, size_t size,
+		       void *cpu_addr, dma_addr_t handle)
+{
+	void *addr = phys_to_cached(uncached_to_phys(cpu_addr));
+	struct page *page;
+
+	pr_debug("dma_free_coherent addr %p (phys %08lx) size %u\n",
+		 cpu_addr, (unsigned long)handle, (unsigned)size);
+	BUG_ON(!virt_addr_valid(addr));
+	page = virt_to_page(addr);
+	__dma_free(dev, size, page, handle);
+}
+EXPORT_SYMBOL(dma_free_coherent);
+
+#if 0
+void *dma_alloc_writecombine(struct device *dev, size_t size,
+			     dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page;
+
+	page = __dma_alloc(dev, size, handle, gfp);
+
+	/* Now, map the page into P3 with write-combining turned on */
+	return __ioremap(page_to_phys(page), size, _PAGE_BUFFER);
+}
+EXPORT_SYMBOL(dma_alloc_writecombine);
+
+void dma_free_writecombine(struct device *dev, size_t size,
+			   void *cpu_addr, dma_addr_t handle)
+{
+	struct page *page;
+
+	iounmap(cpu_addr);
+
+	page = bus_to_page(handle);
+	__dma_free(dev, size, page, handle);
+}
+EXPORT_SYMBOL(dma_free_writecombine);
+#endif
diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c
new file mode 100644
index 000000000000..678557260a35
--- /dev/null
+++ b/arch/avr32/mm/fault.c
@@ -0,0 +1,315 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/arch/sh/mm/fault.c:
+ *   Copyright (C) 1999  Niibe Yutaka
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/pagemap.h>
+
+#include <asm/kdebug.h>
+#include <asm/mmu_context.h>
+#include <asm/sysreg.h>
+#include <asm/uaccess.h>
+#include <asm/tlb.h>
+
+#ifdef DEBUG
+static void dump_code(unsigned long pc)
+{
+	char *p = (char *)pc;
+	char val;
+	int i;
+
+
+	printk(KERN_DEBUG "Code:");
+	for (i = 0; i < 16; i++) {
+		if (__get_user(val, p + i))
+			break;
+		printk(" %02x", val);
+	}
+	printk("\n");
+}
+#endif
+
+#ifdef CONFIG_KPROBES
+ATOMIC_NOTIFIER_HEAD(notify_page_fault_chain);
+
+/* Hook to register for page fault notifications */
+int register_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_register(&notify_page_fault_chain, nb);
+}
+
+int unregister_page_fault_notifier(struct notifier_block *nb)
+{
+	return atomic_notifier_chain_unregister(&notify_page_fault_chain, nb);
+}
+
+static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
+				    int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.trapnr = trap,
+	};
+	return atomic_notifier_call_chain(&notify_page_fault_chain, val, &args);
+}
+#else
+static inline int notify_page_fault(enum die_val val, struct pt_regs *regs,
+				    int trap, int sig)
+{
+	return NOTIFY_DONE;
+}
+#endif
+
+/*
+ * This routine handles page faults. It determines the address and the
+ * problem, and then passes it off to one of the appropriate routines.
+ *
+ * ecr is the Exception Cause Register. Possible values are:
+ *   5:  Page not found (instruction access)
+ *   6:  Protection fault (instruction access)
+ *   12: Page not found (read access)
+ *   13: Page not found (write access)
+ *   14: Protection fault (read access)
+ *   15: Protection fault (write access)
+ */
+asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs)
+{
+	struct task_struct *tsk;
+	struct mm_struct *mm;
+	struct vm_area_struct *vma;
+	const struct exception_table_entry *fixup;
+	unsigned long address;
+	unsigned long page;
+	int writeaccess = 0;
+
+	if (notify_page_fault(DIE_PAGE_FAULT, regs,
+			      ecr, SIGSEGV) == NOTIFY_STOP)
+		return;
+
+	address = sysreg_read(TLBEAR);
+
+	tsk = current;
+	mm = tsk->mm;
+
+	/*
+	 * If we're in an interrupt or have no user context, we must
+	 * not take the fault...
+	 */
+	if (in_atomic() || !mm || regs->sr & SYSREG_BIT(GM))
+		goto no_context;
+
+	local_irq_enable();
+
+	down_read(&mm->mmap_sem);
+
+	vma = find_vma(mm, address);
+	if (!vma)
+		goto bad_area;
+	if (vma->vm_start <= address)
+		goto good_area;
+	if (!(vma->vm_flags & VM_GROWSDOWN))
+		goto bad_area;
+	if (expand_stack(vma, address))
+		goto bad_area;
+
+	/*
+	 * Ok, we have a good vm_area for this memory access, so we
+	 * can handle it...
+	 */
+good_area:
+	//pr_debug("good area: vm_flags = 0x%lx\n", vma->vm_flags);
+	switch (ecr) {
+	case ECR_PROTECTION_X:
+	case ECR_TLB_MISS_X:
+		if (!(vma->vm_flags & VM_EXEC))
+			goto bad_area;
+		break;
+	case ECR_PROTECTION_R:
+	case ECR_TLB_MISS_R:
+		if (!(vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)))
+			goto bad_area;
+		break;
+	case ECR_PROTECTION_W:
+	case ECR_TLB_MISS_W:
+		if (!(vma->vm_flags & VM_WRITE))
+			goto bad_area;
+		writeaccess = 1;
+		break;
+	default:
+		panic("Unhandled case %lu in do_page_fault!", ecr);
+	}
+
+	/*
+	 * If for any reason at all we couldn't handle the fault, make
+	 * sure we exit gracefully rather than endlessly redo the
+	 * fault.
+	 */
+survive:
+	switch (handle_mm_fault(mm, vma, address, writeaccess)) {
+	case VM_FAULT_MINOR:
+		tsk->min_flt++;
+		break;
+	case VM_FAULT_MAJOR:
+		tsk->maj_flt++;
+		break;
+	case VM_FAULT_SIGBUS:
+		goto do_sigbus;
+	case VM_FAULT_OOM:
+		goto out_of_memory;
+	default:
+		BUG();
+	}
+
+	up_read(&mm->mmap_sem);
+	return;
+
+	/*
+	 * Something tried to access memory that isn't in our memory
+	 * map. Fix it, but check if it's kernel or user first...
+	 */
+bad_area:
+	pr_debug("Bad area [%s:%u]: addr %08lx, ecr %lu\n",
+		 tsk->comm, tsk->pid, address, ecr);
+
+	up_read(&mm->mmap_sem);
+
+	if (user_mode(regs)) {
+		/* Hmm...we have to pass address and ecr somehow... */
+		/* tsk->thread.address = address;
+		   tsk->thread.error_code = ecr; */
+#ifdef DEBUG
+		show_regs(regs);
+		dump_code(regs->pc);
+
+		page = sysreg_read(PTBR);
+		printk("ptbr = %08lx", page);
+		if (page) {
+			page = ((unsigned long *)page)[address >> 22];
+			printk(" pgd = %08lx", page);
+			if (page & _PAGE_PRESENT) {
+				page &= PAGE_MASK;
+				address &= 0x003ff000;
+				page = ((unsigned long *)__va(page))[address >> PAGE_SHIFT];
+				printk(" pte = %08lx\n", page);
+			}
+		}
+#endif
+		pr_debug("Sending SIGSEGV to PID %d...\n",
+			tsk->pid);
+		force_sig(SIGSEGV, tsk);
+		return;
+	}
+
+no_context:
+	pr_debug("No context\n");
+
+	/* Are we prepared to handle this kernel fault? */
+	fixup = search_exception_tables(regs->pc);
+	if (fixup) {
+		regs->pc = fixup->fixup;
+		pr_debug("Found fixup at %08lx\n", fixup->fixup);
+		return;
+	}
+
+	/*
+	 * Oops. The kernel tried to access some bad page. We'll have
+	 * to terminate things with extreme prejudice.
+	 */
+	if (address < PAGE_SIZE)
+		printk(KERN_ALERT
+		       "Unable to handle kernel NULL pointer dereference");
+	else
+		printk(KERN_ALERT
+		       "Unable to handle kernel paging request");
+	printk(" at virtual address %08lx\n", address);
+	printk(KERN_ALERT "pc = %08lx\n", regs->pc);
+
+	page = sysreg_read(PTBR);
+	printk(KERN_ALERT "ptbr = %08lx", page);
+	if (page) {
+		page = ((unsigned long *)page)[address >> 22];
+		printk(" pgd = %08lx", page);
+		if (page & _PAGE_PRESENT) {
+			page &= PAGE_MASK;
+			address &= 0x003ff000;
+			page = ((unsigned long *)__va(page))[address >> PAGE_SHIFT];
+			printk(" pte = %08lx\n", page);
+		}
+	}
+	die("\nOops", regs, ecr);
+	do_exit(SIGKILL);
+
+	/*
+	 * We ran out of memory, or some other thing happened to us
+	 * that made us unable to handle the page fault gracefully.
+	 */
+out_of_memory:
+	printk("Out of memory\n");
+	up_read(&mm->mmap_sem);
+	if (current->pid == 1) {
+		yield();
+		down_read(&mm->mmap_sem);
+		goto survive;
+	}
+	printk("VM: Killing process %s\n", tsk->comm);
+	if (user_mode(regs))
+		do_exit(SIGKILL);
+	goto no_context;
+
+do_sigbus:
+	up_read(&mm->mmap_sem);
+
+	/*
+	 * Send a sigbus, regardless of whether we were in kernel or
+	 * user mode.
+	 */
+	/* address, error_code, trap_no, ... */
+#ifdef DEBUG
+	show_regs(regs);
+	dump_code(regs->pc);
+#endif
+	pr_debug("Sending SIGBUS to PID %d...\n", tsk->pid);
+	force_sig(SIGBUS, tsk);
+
+	/* Kernel mode? Handle exceptions or die */
+	if (!user_mode(regs))
+		goto no_context;
+}
+
+asmlinkage void do_bus_error(unsigned long addr, int write_access,
+			     struct pt_regs *regs)
+{
+	printk(KERN_ALERT
+	       "Bus error at physical address 0x%08lx (%s access)\n",
+	       addr, write_access ? "write" : "read");
+	printk(KERN_INFO "DTLB dump:\n");
+	dump_dtlb();
+	die("Bus Error", regs, write_access);
+	do_exit(SIGKILL);
+}
+
+/*
+ * This functionality is currently not possible to implement because
+ * we're using segmentation to ensure a fixed mapping of the kernel
+ * virtual address space.
+ *
+ * It would be possible to implement this, but it would require us to
+ * disable segmentation at startup and load the kernel mappings into
+ * the TLB like any other pages. There will be lots of trickery to
+ * avoid recursive invocation of the TLB miss handler, though...
+ */
+#ifdef CONFIG_DEBUG_PAGEALLOC
+void kernel_map_pages(struct page *page, int numpages, int enable)
+{
+
+}
+EXPORT_SYMBOL(kernel_map_pages);
+#endif
diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
new file mode 100644
index 000000000000..3e6c41039808
--- /dev/null
+++ b/arch/avr32/mm/init.c
@@ -0,0 +1,480 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/init.h>
+#include <linux/initrd.h>
+#include <linux/mmzone.h>
+#include <linux/bootmem.h>
+#include <linux/pagemap.h>
+#include <linux/pfn.h>
+#include <linux/nodemask.h>
+
+#include <asm/page.h>
+#include <asm/mmu_context.h>
+#include <asm/tlb.h>
+#include <asm/io.h>
+#include <asm/dma.h>
+#include <asm/setup.h>
+#include <asm/sections.h>
+
+DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
+
+pgd_t swapper_pg_dir[PTRS_PER_PGD];
+
+struct page *empty_zero_page;
+
+/*
+ * Cache of MMU context last used.
+ */
+unsigned long mmu_context_cache = NO_CONTEXT;
+
+#define START_PFN	(NODE_DATA(0)->bdata->node_boot_start >> PAGE_SHIFT)
+#define MAX_LOW_PFN	(NODE_DATA(0)->bdata->node_low_pfn)
+
+void show_mem(void)
+{
+	int total = 0, reserved = 0, cached = 0;
+	int slab = 0, free = 0, shared = 0;
+	pg_data_t *pgdat;
+
+	printk("Mem-info:\n");
+	show_free_areas();
+
+	for_each_online_pgdat(pgdat) {
+		struct page *page, *end;
+
+		page = pgdat->node_mem_map;
+		end = page + pgdat->node_spanned_pages;
+
+		do {
+			total++;
+			if (PageReserved(page))
+				reserved++;
+			else if (PageSwapCache(page))
+				cached++;
+			else if (PageSlab(page))
+				slab++;
+			else if (!page_count(page))
+				free++;
+			else
+				shared += page_count(page) - 1;
+			page++;
+		} while (page < end);
+	}
+
+	printk ("%d pages of RAM\n", total);
+	printk ("%d free pages\n", free);
+	printk ("%d reserved pages\n", reserved);
+	printk ("%d slab pages\n", slab);
+	printk ("%d pages shared\n", shared);
+	printk ("%d pages swap cached\n", cached);
+}
+
+static void __init print_memory_map(const char *what,
+				    struct tag_mem_range *mem)
+{
+	printk ("%s:\n", what);
+	for (; mem; mem = mem->next) {
+		printk ("  %08lx - %08lx\n",
+			(unsigned long)mem->addr,
+			(unsigned long)(mem->addr + mem->size));
+	}
+}
+
+#define MAX_LOWMEM	HIGHMEM_START
+#define MAX_LOWMEM_PFN	PFN_DOWN(MAX_LOWMEM)
+
+/*
+ * Sort a list of memory regions in-place by ascending address.
+ *
+ * We're using bubble sort because we only have singly linked lists
+ * with few elements.
+ */
+static void __init sort_mem_list(struct tag_mem_range **pmem)
+{
+	int done;
+	struct tag_mem_range **a, **b;
+
+	if (!*pmem)
+		return;
+
+	do {
+		done = 1;
+		a = pmem, b = &(*pmem)->next;
+		while (*b) {
+			if ((*a)->addr > (*b)->addr) {
+				struct tag_mem_range *tmp;
+				tmp = (*b)->next;
+				(*b)->next = *a;
+				*a = *b;
+				*b = tmp;
+				done = 0;
+			}
+			a = &(*a)->next;
+			b = &(*a)->next;
+		}
+	} while (!done);
+}
+
+/*
+ * Find a free memory region large enough for storing the
+ * bootmem bitmap.
+ */
+static unsigned long __init
+find_bootmap_pfn(const struct tag_mem_range *mem)
+{
+	unsigned long bootmap_pages, bootmap_len;
+	unsigned long node_pages = PFN_UP(mem->size);
+	unsigned long bootmap_addr = mem->addr;
+	struct tag_mem_range *reserved = mem_reserved;
+	struct tag_mem_range *ramdisk = mem_ramdisk;
+	unsigned long kern_start = virt_to_phys(_stext);
+	unsigned long kern_end = virt_to_phys(_end);
+
+	bootmap_pages = bootmem_bootmap_pages(node_pages);
+	bootmap_len = bootmap_pages << PAGE_SHIFT;
+
+	/*
+	 * Find a large enough region without reserved pages for
+	 * storing the bootmem bitmap. We can take advantage of the
+	 * fact that all lists have been sorted.
+	 *
+	 * We have to check explicitly reserved regions as well as the
+	 * kernel image and any RAMDISK images...
+	 *
+	 * Oh, and we have to make sure we don't overwrite the taglist
+	 * since we're going to use it until the bootmem allocator is
+	 * fully up and running.
+	 */
+	while (1) {
+		if ((bootmap_addr < kern_end) &&
+		    ((bootmap_addr + bootmap_len) > kern_start))
+			bootmap_addr = kern_end;
+
+		while (reserved &&
+		       (bootmap_addr >= (reserved->addr + reserved->size)))
+			reserved = reserved->next;
+
+		if (reserved &&
+		    ((bootmap_addr + bootmap_len) >= reserved->addr)) {
+			bootmap_addr = reserved->addr + reserved->size;
+			continue;
+		}
+
+		while (ramdisk &&
+		       (bootmap_addr >= (ramdisk->addr + ramdisk->size)))
+			ramdisk = ramdisk->next;
+
+		if (!ramdisk ||
+		    ((bootmap_addr + bootmap_len) < ramdisk->addr))
+			break;
+
+		bootmap_addr = ramdisk->addr + ramdisk->size;
+	}
+
+	if ((PFN_UP(bootmap_addr) + bootmap_len) >= (mem->addr + mem->size))
+		return ~0UL;
+
+	return PFN_UP(bootmap_addr);
+}
+
+void __init setup_bootmem(void)
+{
+	unsigned bootmap_size;
+	unsigned long first_pfn, bootmap_pfn, pages;
+	unsigned long max_pfn, max_low_pfn;
+	unsigned long kern_start = virt_to_phys(_stext);
+	unsigned long kern_end = virt_to_phys(_end);
+	unsigned node = 0;
+	struct tag_mem_range *bank, *res;
+
+	sort_mem_list(&mem_phys);
+	sort_mem_list(&mem_reserved);
+
+	print_memory_map("Physical memory", mem_phys);
+	print_memory_map("Reserved memory", mem_reserved);
+
+	nodes_clear(node_online_map);
+
+	if (mem_ramdisk) {
+#ifdef CONFIG_BLK_DEV_INITRD
+		initrd_start = __va(mem_ramdisk->addr);
+		initrd_end = initrd_start + mem_ramdisk->size;
+
+		print_memory_map("RAMDISK images", mem_ramdisk);
+		if (mem_ramdisk->next)
+			printk(KERN_WARNING
+			       "Warning: Only the first RAMDISK image "
+			       "will be used\n");
+		sort_mem_list(&mem_ramdisk);
+#else
+		printk(KERN_WARNING "RAM disk image present, but "
+		       "no initrd support in kernel!\n");
+#endif
+	}
+
+	if (mem_phys->next)
+		printk(KERN_WARNING "Only using first memory bank\n");
+
+	for (bank = mem_phys; bank; bank = NULL) {
+		first_pfn = PFN_UP(bank->addr);
+		max_low_pfn = max_pfn = PFN_DOWN(bank->addr + bank->size);
+		bootmap_pfn = find_bootmap_pfn(bank);
+		if (bootmap_pfn > max_pfn)
+			panic("No space for bootmem bitmap!\n");
+
+		if (max_low_pfn > MAX_LOWMEM_PFN) {
+			max_low_pfn = MAX_LOWMEM_PFN;
+#ifndef CONFIG_HIGHMEM
+			/*
+			 * Lowmem is memory that can be addressed
+			 * directly through P1/P2
+			 */
+			printk(KERN_WARNING
+			       "Node %u: Only %ld MiB of memory will be used.\n",
+			       node, MAX_LOWMEM >> 20);
+			printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
+#else
+#error HIGHMEM is not supported by AVR32 yet
+#endif
+		}
+
+		/* Initialize the boot-time allocator with low memory only. */
+		bootmap_size = init_bootmem_node(NODE_DATA(node), bootmap_pfn,
+						 first_pfn, max_low_pfn);
+
+		printk("Node %u: bdata = %p, bdata->node_bootmem_map = %p\n",
+		       node, NODE_DATA(node)->bdata,
+		       NODE_DATA(node)->bdata->node_bootmem_map);
+
+		/*
+		 * Register fully available RAM pages with the bootmem
+		 * allocator.
+		 */
+		pages = max_low_pfn - first_pfn;
+		free_bootmem_node (NODE_DATA(node), PFN_PHYS(first_pfn),
+				   PFN_PHYS(pages));
+
+		/*
+		 * Reserve space for the kernel image (if present in
+		 * this node)...
+		 */
+		if ((kern_start >= PFN_PHYS(first_pfn)) &&
+		    (kern_start < PFN_PHYS(max_pfn))) {
+			printk("Node %u: Kernel image %08lx - %08lx\n",
+			       node, kern_start, kern_end);
+			reserve_bootmem_node(NODE_DATA(node), kern_start,
+					     kern_end - kern_start);
+		}
+
+		/* ...the bootmem bitmap... */
+		reserve_bootmem_node(NODE_DATA(node),
+				     PFN_PHYS(bootmap_pfn),
+				     bootmap_size);
+
+		/* ...any RAMDISK images... */
+		for (res = mem_ramdisk; res; res = res->next) {
+			if (res->addr > PFN_PHYS(max_pfn))
+				break;
+
+			if (res->addr >= PFN_PHYS(first_pfn)) {
+				printk("Node %u: RAMDISK %08lx - %08lx\n",
+				       node,
+				       (unsigned long)res->addr,
+				       (unsigned long)(res->addr + res->size));
+				reserve_bootmem_node(NODE_DATA(node),
+						     res->addr, res->size);
+			}
+		}
+
+		/* ...and any other reserved regions. */
+		for (res = mem_reserved; res; res = res->next) {
+			if (res->addr > PFN_PHYS(max_pfn))
+				break;
+
+			if (res->addr >= PFN_PHYS(first_pfn)) {
+				printk("Node %u: Reserved %08lx - %08lx\n",
+				       node,
+				       (unsigned long)res->addr,
+				       (unsigned long)(res->addr + res->size));
+				reserve_bootmem_node(NODE_DATA(node),
+						     res->addr, res->size);
+			}
+		}
+
+		node_set_online(node);
+	}
+}
+
+/*
+ * paging_init() sets up the page tables
+ *
+ * This routine also unmaps the page at virtual kernel address 0, so
+ * that we can trap those pesky NULL-reference errors in the kernel.
+ */
+void __init paging_init(void)
+{
+	extern unsigned long _evba;
+	void *zero_page;
+	int nid;
+
+	/*
+	 * Make sure we can handle exceptions before enabling
+	 * paging. Not that we should ever _get_ any exceptions this
+	 * early, but you never know...
+	 */
+	printk("Exception vectors start at %p\n", &_evba);
+	sysreg_write(EVBA, (unsigned long)&_evba);
+
+	/*
+	 * Since we are ready to handle exceptions now, we should let
+	 * the CPU generate them...
+	 */
+	__asm__ __volatile__ ("csrf %0" : : "i"(SR_EM_BIT));
+
+	/*
+	 * Allocate the zero page. The allocator will panic if it
+	 * can't satisfy the request, so no need to check.
+	 */
+	zero_page = alloc_bootmem_low_pages_node(NODE_DATA(0),
+						 PAGE_SIZE);
+
+	{
+		pgd_t *pg_dir;
+		int i;
+
+		pg_dir = swapper_pg_dir;
+		sysreg_write(PTBR, (unsigned long)pg_dir);
+
+		for (i = 0; i < PTRS_PER_PGD; i++)
+			pgd_val(pg_dir[i]) = 0;
+
+		enable_mmu();
+		printk ("CPU: Paging enabled\n");
+	}
+
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		unsigned long zones_size[MAX_NR_ZONES];
+		unsigned long low, start_pfn;
+
+		start_pfn = pgdat->bdata->node_boot_start;
+		start_pfn >>= PAGE_SHIFT;
+		low = pgdat->bdata->node_low_pfn;
+
+		memset(zones_size, 0, sizeof(zones_size));
+		zones_size[ZONE_NORMAL] = low - start_pfn;
+
+		printk("Node %u: start_pfn = 0x%lx, low = 0x%lx\n",
+		       nid, start_pfn, low);
+
+		free_area_init_node(nid, pgdat, zones_size, start_pfn, NULL);
+
+		printk("Node %u: mem_map starts at %p\n",
+		       pgdat->node_id, pgdat->node_mem_map);
+	}
+
+	mem_map = NODE_DATA(0)->node_mem_map;
+
+	memset(zero_page, 0, PAGE_SIZE);
+	empty_zero_page = virt_to_page(zero_page);
+	flush_dcache_page(empty_zero_page);
+}
+
+void __init mem_init(void)
+{
+	int codesize, reservedpages, datasize, initsize;
+	int nid, i;
+
+	reservedpages = 0;
+	high_memory = NULL;
+
+	/* this will put all low memory onto the freelists */
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		unsigned long node_pages = 0;
+		void *node_high_memory;
+
+		num_physpages += pgdat->node_present_pages;
+
+		if (pgdat->node_spanned_pages != 0)
+			node_pages = free_all_bootmem_node(pgdat);
+
+		totalram_pages += node_pages;
+
+		for (i = 0; i < node_pages; i++)
+			if (PageReserved(pgdat->node_mem_map + i))
+				reservedpages++;
+
+		node_high_memory = (void *)((pgdat->node_start_pfn
+					     + pgdat->node_spanned_pages)
+					    << PAGE_SHIFT);
+		if (node_high_memory > high_memory)
+			high_memory = node_high_memory;
+	}
+
+	max_mapnr = MAP_NR(high_memory);
+
+	codesize = (unsigned long)_etext - (unsigned long)_text;
+	datasize = (unsigned long)_edata - (unsigned long)_data;
+	initsize = (unsigned long)__init_end - (unsigned long)__init_begin;
+
+	printk ("Memory: %luk/%luk available (%dk kernel code, "
+		"%dk reserved, %dk data, %dk init)\n",
+		(unsigned long)nr_free_pages() << (PAGE_SHIFT - 10),
+		totalram_pages << (PAGE_SHIFT - 10),
+		codesize >> 10,
+		reservedpages << (PAGE_SHIFT - 10),
+		datasize >> 10,
+		initsize >> 10);
+}
+
+static inline void free_area(unsigned long addr, unsigned long end, char *s)
+{
+	unsigned int size = (end - addr) >> 10;
+
+	for (; addr < end; addr += PAGE_SIZE) {
+		struct page *page = virt_to_page(addr);
+		ClearPageReserved(page);
+		init_page_count(page);
+		free_page(addr);
+		totalram_pages++;
+	}
+
+	if (size && s)
+		printk(KERN_INFO "Freeing %s memory: %dK (%lx - %lx)\n",
+		       s, size, end - (size << 10), end);
+}
+
+void free_initmem(void)
+{
+	free_area((unsigned long)__init_begin, (unsigned long)__init_end,
+		  "init");
+}
+
+#ifdef CONFIG_BLK_DEV_INITRD
+
+static int keep_initrd;
+
+void free_initrd_mem(unsigned long start, unsigned long end)
+{
+	if (!keep_initrd)
+		free_area(start, end, "initrd");
+}
+
+static int __init keepinitrd_setup(char *__unused)
+{
+	keep_initrd = 1;
+	return 1;
+}
+
+__setup("keepinitrd", keepinitrd_setup);
+#endif
diff --git a/arch/avr32/mm/ioremap.c b/arch/avr32/mm/ioremap.c
new file mode 100644
index 000000000000..536021877df6
--- /dev/null
+++ b/arch/avr32/mm/ioremap.c
@@ -0,0 +1,197 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+
+#include <asm/io.h>
+#include <asm/pgtable.h>
+#include <asm/cacheflush.h>
+#include <asm/tlbflush.h>
+#include <asm/addrspace.h>
+
+static inline int remap_area_pte(pte_t *pte, unsigned long address,
+				  unsigned long end, unsigned long phys_addr,
+				  pgprot_t prot)
+{
+	unsigned long pfn;
+
+	pfn = phys_addr >> PAGE_SHIFT;
+	do {
+		WARN_ON(!pte_none(*pte));
+
+		set_pte(pte, pfn_pte(pfn, prot));
+		address += PAGE_SIZE;
+		pfn++;
+		pte++;
+	} while (address && (address < end));
+
+	return 0;
+}
+
+static inline int remap_area_pmd(pmd_t *pmd, unsigned long address,
+				 unsigned long end, unsigned long phys_addr,
+				 pgprot_t prot)
+{
+	unsigned long next;
+
+	phys_addr -= address;
+
+	do {
+		pte_t *pte = pte_alloc_kernel(pmd, address);
+		if (!pte)
+			return -ENOMEM;
+
+		next = (address + PMD_SIZE) & PMD_MASK;
+		if (remap_area_pte(pte, address, next,
+				   address + phys_addr, prot))
+			return -ENOMEM;
+
+		address = next;
+		pmd++;
+	} while (address && (address < end));
+	return 0;
+}
+
+static int remap_area_pud(pud_t *pud, unsigned long address,
+			  unsigned long end, unsigned long phys_addr,
+			  pgprot_t prot)
+{
+	unsigned long next;
+
+	phys_addr -= address;
+
+	do {
+		pmd_t *pmd = pmd_alloc(&init_mm, pud, address);
+		if (!pmd)
+			return -ENOMEM;
+		next = (address + PUD_SIZE) & PUD_MASK;
+		if (remap_area_pmd(pmd, address, next,
+				   phys_addr + address, prot))
+			return -ENOMEM;
+
+		address = next;
+		pud++;
+	} while (address && address < end);
+
+	return 0;
+}
+
+static int remap_area_pages(unsigned long address, unsigned long phys_addr,
+			    size_t size, pgprot_t prot)
+{
+	unsigned long end = address + size;
+	unsigned long next;
+	pgd_t *pgd;
+	int err = 0;
+
+	phys_addr -= address;
+
+	pgd = pgd_offset_k(address);
+	flush_cache_all();
+	BUG_ON(address >= end);
+
+	spin_lock(&init_mm.page_table_lock);
+	do {
+		pud_t *pud = pud_alloc(&init_mm, pgd, address);
+
+		err = -ENOMEM;
+		if (!pud)
+			break;
+
+		next = (address + PGDIR_SIZE) & PGDIR_MASK;
+		if (next < address || next > end)
+			next = end;
+		err = remap_area_pud(pud, address, next,
+				     phys_addr + address, prot);
+		if (err)
+			break;
+
+		address = next;
+		pgd++;
+	} while (address && (address < end));
+
+	spin_unlock(&init_mm.page_table_lock);
+	flush_tlb_all();
+	return err;
+}
+
+/*
+ * Re-map an arbitrary physical address space into the kernel virtual
+ * address space. Needed when the kernel wants to access physical
+ * memory directly.
+ */
+void __iomem *__ioremap(unsigned long phys_addr, size_t size,
+			unsigned long flags)
+{
+	void *addr;
+	struct vm_struct *area;
+	unsigned long offset, last_addr;
+	pgprot_t prot;
+
+	/*
+	 * Check if we can simply use the P4 segment. This area is
+	 * uncacheable, so if caching/buffering is requested, we can't
+	 * use it.
+	 */
+	if ((phys_addr >= P4SEG) && (flags == 0))
+		return (void __iomem *)phys_addr;
+
+	/* Don't allow wraparound or zero size */
+	last_addr = phys_addr + size - 1;
+	if (!size || last_addr < phys_addr)
+		return NULL;
+
+	/*
+	 * XXX: When mapping regular RAM, we'd better make damn sure
+	 * it's never used for anything else.  But this is really the
+	 * caller's responsibility...
+	 */
+	if (PHYSADDR(P2SEGADDR(phys_addr)) == phys_addr)
+		return (void __iomem *)P2SEGADDR(phys_addr);
+
+	/* Mappings have to be page-aligned */
+	offset = phys_addr & ~PAGE_MASK;
+	phys_addr &= PAGE_MASK;
+	size = PAGE_ALIGN(last_addr + 1) - phys_addr;
+
+	prot = __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY
+			| _PAGE_ACCESSED | _PAGE_TYPE_SMALL | flags);
+
+	/*
+	 * Ok, go for it..
+	 */
+	area = get_vm_area(size, VM_IOREMAP);
+	if (!area)
+		return NULL;
+	area->phys_addr = phys_addr;
+	addr = area->addr;
+	if (remap_area_pages((unsigned long)addr, phys_addr, size, prot)) {
+		vunmap(addr);
+		return NULL;
+	}
+
+	return (void __iomem *)(offset + (char *)addr);
+}
+EXPORT_SYMBOL(__ioremap);
+
+void __iounmap(void __iomem *addr)
+{
+	struct vm_struct *p;
+
+	if ((unsigned long)addr >= P4SEG)
+		return;
+
+	p = remove_vm_area((void *)(PAGE_MASK & (unsigned long __force)addr));
+	if (unlikely(!p)) {
+		printk (KERN_ERR "iounmap: bad address %p\n", addr);
+		return;
+	}
+
+	kfree (p);
+}
+EXPORT_SYMBOL(__iounmap);
diff --git a/arch/avr32/mm/tlb.c b/arch/avr32/mm/tlb.c
new file mode 100644
index 000000000000..5d0523bbe298
--- /dev/null
+++ b/arch/avr32/mm/tlb.c
@@ -0,0 +1,378 @@
+/*
+ * AVR32 TLB operations
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/mm.h>
+
+#include <asm/mmu_context.h>
+
+#define _TLBEHI_I	0x100
+
+void show_dtlb_entry(unsigned int index)
+{
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+
+	local_irq_save(flags);
+	mmucr_save = sysreg_read(MMUCR);
+	tlbehi_save = sysreg_read(TLBEHI);
+	mmucr = mmucr_save & 0x13;
+	mmucr |= index << 14;
+	sysreg_write(MMUCR, mmucr);
+
+	asm volatile("tlbr" : : : "memory");
+	cpu_sync_pipeline();
+
+	tlbehi = sysreg_read(TLBEHI);
+	tlbelo = sysreg_read(TLBELO);
+
+	printk("%2u: %c %c %02x   %05x %05x %o  %o  %c %c %c %c\n",
+	       index,
+	       (tlbehi & 0x200)?'1':'0',
+	       (tlbelo & 0x100)?'1':'0',
+	       (tlbehi & 0xff),
+	       (tlbehi >> 12), (tlbelo >> 12),
+	       (tlbelo >> 4) & 7, (tlbelo >> 2) & 3,
+	       (tlbelo & 0x200)?'1':'0',
+	       (tlbelo & 0x080)?'1':'0',
+	       (tlbelo & 0x001)?'1':'0',
+	       (tlbelo & 0x002)?'1':'0');
+
+	sysreg_write(MMUCR, mmucr_save);
+	sysreg_write(TLBEHI, tlbehi_save);
+	cpu_sync_pipeline();
+	local_irq_restore(flags);
+}
+
+void dump_dtlb(void)
+{
+	unsigned int i;
+
+	printk("ID  V G ASID VPN   PFN   AP SZ C B W D\n");
+	for (i = 0; i < 32; i++)
+		show_dtlb_entry(i);
+}
+
+static unsigned long last_mmucr;
+
+static inline void set_replacement_pointer(unsigned shift)
+{
+	unsigned long mmucr, mmucr_save;
+
+	mmucr = mmucr_save = sysreg_read(MMUCR);
+
+	/* Does this mapping already exist? */
+	__asm__ __volatile__(
+		"	tlbs\n"
+		"	mfsr %0, %1"
+		: "=r"(mmucr)
+		: "i"(SYSREG_MMUCR));
+
+	if (mmucr & SYSREG_BIT(MMUCR_N)) {
+		/* Not found -- pick a not-recently-accessed entry */
+		unsigned long rp;
+		unsigned long tlbar = sysreg_read(TLBARLO);
+
+		rp = 32 - fls(tlbar);
+		if (rp == 32) {
+			rp = 0;
+			sysreg_write(TLBARLO, -1L);
+		}
+
+		mmucr &= 0x13;
+		mmucr |= (rp << shift);
+
+		sysreg_write(MMUCR, mmucr);
+	}
+
+	last_mmucr = mmucr;
+}
+
+static void update_dtlb(unsigned long address, pte_t pte, unsigned long asid)
+{
+	unsigned long vpn;
+
+	vpn = (address & MMU_VPN_MASK) | _TLBEHI_VALID | asid;
+	sysreg_write(TLBEHI, vpn);
+	cpu_sync_pipeline();
+
+	set_replacement_pointer(14);
+
+	sysreg_write(TLBELO, pte_val(pte) & _PAGE_FLAGS_HARDWARE_MASK);
+
+	/* Let's go */
+	asm volatile("nop\n\ttlbw" : : : "memory");
+	cpu_sync_pipeline();
+}
+
+void update_mmu_cache(struct vm_area_struct *vma,
+		      unsigned long address, pte_t pte)
+{
+	unsigned long flags;
+
+	/* ptrace may call this routine */
+	if (vma && current->active_mm != vma->vm_mm)
+		return;
+
+	local_irq_save(flags);
+	update_dtlb(address, pte, get_asid());
+	local_irq_restore(flags);
+}
+
+void __flush_tlb_page(unsigned long asid, unsigned long page)
+{
+	unsigned long mmucr, tlbehi;
+
+	page |= asid;
+	sysreg_write(TLBEHI, page);
+	cpu_sync_pipeline();
+	asm volatile("tlbs");
+	mmucr = sysreg_read(MMUCR);
+
+	if (!(mmucr & SYSREG_BIT(MMUCR_N))) {
+		unsigned long tlbarlo;
+		unsigned long entry;
+
+		/* Clear the "valid" bit */
+		tlbehi = sysreg_read(TLBEHI);
+		tlbehi &= ~_TLBEHI_VALID;
+		sysreg_write(TLBEHI, tlbehi);
+		cpu_sync_pipeline();
+
+		/* mark the entry as "not accessed" */
+		entry = (mmucr >> 14) & 0x3f;
+		tlbarlo = sysreg_read(TLBARLO);
+		tlbarlo |= (0x80000000 >> entry);
+		sysreg_write(TLBARLO, tlbarlo);
+
+		/* update the entry with valid bit clear */
+		asm volatile("tlbw");
+		cpu_sync_pipeline();
+	}
+}
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
+{
+	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+		unsigned long flags, asid;
+		unsigned long saved_asid = MMU_NO_ASID;
+
+		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		page &= PAGE_MASK;
+
+		local_irq_save(flags);
+		if (vma->vm_mm != current->mm) {
+			saved_asid = get_asid();
+			set_asid(asid);
+		}
+
+		__flush_tlb_page(asid, page);
+
+		if (saved_asid != MMU_NO_ASID)
+			set_asid(saved_asid);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		int size;
+
+		local_irq_save(flags);
+		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+		if (size > (MMU_DTLB_ENTRIES / 4)) { /* Too many entries to flush */
+			mm->context = NO_CONTEXT;
+			if (mm == current->mm)
+				activate_context(mm);
+		} else {
+			unsigned long asid = mm->context & MMU_CONTEXT_ASID_MASK;
+			unsigned long saved_asid = MMU_NO_ASID;
+
+			start &= PAGE_MASK;
+			end += (PAGE_SIZE - 1);
+			end &= PAGE_MASK;
+			if (mm != current->mm) {
+				saved_asid = get_asid();
+				set_asid(asid);
+			}
+
+			while (start < end) {
+				__flush_tlb_page(asid, start);
+				start += PAGE_SIZE;
+			}
+			if (saved_asid != MMU_NO_ASID)
+				set_asid(saved_asid);
+		}
+		local_irq_restore(flags);
+	}
+}
+
+/*
+ * TODO: If this is only called for addresses > TASK_SIZE, we can probably
+ * skip the ASID stuff and just use the Global bit...
+ */
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	unsigned long flags;
+	int size;
+
+	local_irq_save(flags);
+	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	if (size > (MMU_DTLB_ENTRIES / 4)) { /* Too many entries to flush */
+		flush_tlb_all();
+	} else {
+		unsigned long asid = init_mm.context & MMU_CONTEXT_ASID_MASK;
+		unsigned long saved_asid = get_asid();
+
+		start &= PAGE_MASK;
+		end += (PAGE_SIZE - 1);
+		end &= PAGE_MASK;
+		set_asid(asid);
+		while (start < end) {
+			__flush_tlb_page(asid, start);
+			start += PAGE_SIZE;
+		}
+		set_asid(saved_asid);
+	}
+	local_irq_restore(flags);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	/* Invalidate all TLB entries of this process by getting a new ASID */
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+
+		local_irq_save(flags);
+		mm->context = NO_CONTEXT;
+		if (mm == current->mm)
+			activate_context(mm);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_all(void)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	sysreg_write(MMUCR, sysreg_read(MMUCR) | SYSREG_BIT(MMUCR_I));
+	local_irq_restore(flags);
+}
+
+#ifdef CONFIG_PROC_FS
+
+#include <linux/seq_file.h>
+#include <linux/proc_fs.h>
+#include <linux/init.h>
+
+static void *tlb_start(struct seq_file *tlb, loff_t *pos)
+{
+	static unsigned long tlb_index;
+
+	if (*pos >= 32)
+		return NULL;
+
+	tlb_index = 0;
+	return &tlb_index;
+}
+
+static void *tlb_next(struct seq_file *tlb, void *v, loff_t *pos)
+{
+	unsigned long *index = v;
+
+	if (*index >= 31)
+		return NULL;
+
+	++*pos;
+	++*index;
+	return index;
+}
+
+static void tlb_stop(struct seq_file *tlb, void *v)
+{
+
+}
+
+static int tlb_show(struct seq_file *tlb, void *v)
+{
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+	unsigned long *index = v;
+
+	if (*index == 0)
+		seq_puts(tlb, "ID  V G ASID VPN   PFN   AP SZ C B W D\n");
+
+	BUG_ON(*index >= 32);
+
+	local_irq_save(flags);
+	mmucr_save = sysreg_read(MMUCR);
+	tlbehi_save = sysreg_read(TLBEHI);
+	mmucr = mmucr_save & 0x13;
+	mmucr |= *index << 14;
+	sysreg_write(MMUCR, mmucr);
+
+	asm volatile("tlbr" : : : "memory");
+	cpu_sync_pipeline();
+
+	tlbehi = sysreg_read(TLBEHI);
+	tlbelo = sysreg_read(TLBELO);
+
+	sysreg_write(MMUCR, mmucr_save);
+	sysreg_write(TLBEHI, tlbehi_save);
+	cpu_sync_pipeline();
+	local_irq_restore(flags);
+
+	seq_printf(tlb, "%2lu: %c %c %02x   %05x %05x %o  %o  %c %c %c %c\n",
+	       *index,
+	       (tlbehi & 0x200)?'1':'0',
+	       (tlbelo & 0x100)?'1':'0',
+	       (tlbehi & 0xff),
+	       (tlbehi >> 12), (tlbelo >> 12),
+	       (tlbelo >> 4) & 7, (tlbelo >> 2) & 3,
+	       (tlbelo & 0x200)?'1':'0',
+	       (tlbelo & 0x080)?'1':'0',
+	       (tlbelo & 0x001)?'1':'0',
+	       (tlbelo & 0x002)?'1':'0');
+
+	return 0;
+}
+
+static struct seq_operations tlb_ops = {
+	.start		= tlb_start,
+	.next		= tlb_next,
+	.stop		= tlb_stop,
+	.show		= tlb_show,
+};
+
+static int tlb_open(struct inode *inode, struct file *file)
+{
+	return seq_open(file, &tlb_ops);
+}
+
+static struct file_operations proc_tlb_operations = {
+	.open		= tlb_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int __init proctlb_init(void)
+{
+	struct proc_dir_entry *entry;
+
+	entry = create_proc_entry("tlb", 0, NULL);
+	if (entry)
+		entry->proc_fops = &proc_tlb_operations;
+	return 0;
+}
+late_initcall(proctlb_init);
+#endif /* CONFIG_PROC_FS */
diff --git a/include/asm-avr32/Kbuild b/include/asm-avr32/Kbuild
new file mode 100644
index 000000000000..8770e73ce938
--- /dev/null
+++ b/include/asm-avr32/Kbuild
@@ -0,0 +1,3 @@
+include include/asm-generic/Kbuild.asm
+
+headers-y	+= cachectl.h
diff --git a/include/asm-avr32/a.out.h b/include/asm-avr32/a.out.h
new file mode 100644
index 000000000000..50bf6e31a143
--- /dev/null
+++ b/include/asm-avr32/a.out.h
@@ -0,0 +1,26 @@
+#ifndef __ASM_AVR32_A_OUT_H
+#define __ASM_AVR32_A_OUT_H
+
+struct exec
+{
+  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
+  unsigned a_text;		/* length of text, in bytes */
+  unsigned a_data;		/* length of data, in bytes */
+  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
+  unsigned a_syms;		/* length of symbol table data in file, in bytes */
+  unsigned a_entry;		/* start address */
+  unsigned a_trsize;		/* length of relocation info for text, in bytes */
+  unsigned a_drsize;		/* length of relocation info for data, in bytes */
+};
+
+#define N_TRSIZE(a)	((a).a_trsize)
+#define N_DRSIZE(a)	((a).a_drsize)
+#define N_SYMSIZE(a)	((a).a_syms)
+
+#ifdef __KERNEL__
+
+#define STACK_TOP	TASK_SIZE
+
+#endif
+
+#endif /* __ASM_AVR32_A_OUT_H */
diff --git a/include/asm-avr32/addrspace.h b/include/asm-avr32/addrspace.h
new file mode 100644
index 000000000000..366794858ec7
--- /dev/null
+++ b/include/asm-avr32/addrspace.h
@@ -0,0 +1,43 @@
+/*
+ * Defitions for the address spaces of the AVR32 CPUs. Heavily based on
+ * include/asm-sh/addrspace.h
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ADDRSPACE_H
+#define __ASM_AVR32_ADDRSPACE_H
+
+#ifdef CONFIG_MMU
+
+/* Memory segments when segmentation is enabled */
+#define P0SEG		0x00000000
+#define P1SEG		0x80000000
+#define P2SEG		0xa0000000
+#define P3SEG		0xc0000000
+#define P4SEG		0xe0000000
+
+/* Returns the privileged segment base of a given address */
+#define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
+
+/* Returns the physical address of a PnSEG (n=1,2) address */
+#define PHYSADDR(a)	(((unsigned long)(a)) & 0x1fffffff)
+
+/*
+ * Map an address to a certain privileged segment
+ */
+#define P1SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P1SEG))
+#define P2SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P2SEG))
+#define P3SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P3SEG))
+#define P4SEGADDR(a) ((__typeof__(a))(((unsigned long)(a) & 0x1fffffff) \
+				      | P4SEG))
+
+#endif /* CONFIG_MMU */
+
+#endif /* __ASM_AVR32_ADDRSPACE_H */
diff --git a/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h b/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
new file mode 100644
index 000000000000..ce1150d4438d
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/at91rm9200_pdc.h
@@ -0,0 +1,36 @@
+/*
+ * include/asm-arm/arch-at91rm9200/at91rm9200_pdc.h
+ *
+ * Copyright (C) 2005 Ivan Kokshaysky
+ * Copyright (C) SAN People
+ *
+ * Peripheral Data Controller (PDC) registers.
+ * Based on AT91RM9200 datasheet revision E.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91RM9200_PDC_H
+#define AT91RM9200_PDC_H
+
+#define AT91_PDC_RPR		0x100	/* Receive Pointer Register */
+#define AT91_PDC_RCR		0x104	/* Receive Counter Register */
+#define AT91_PDC_TPR		0x108	/* Transmit Pointer Register */
+#define AT91_PDC_TCR		0x10c	/* Transmit Counter Register */
+#define AT91_PDC_RNPR		0x110	/* Receive Next Pointer Register */
+#define AT91_PDC_RNCR		0x114	/* Receive Next Counter Register */
+#define AT91_PDC_TNPR		0x118	/* Transmit Next Pointer Register */
+#define AT91_PDC_TNCR		0x11c	/* Transmit Next Counter Register */
+
+#define AT91_PDC_PTCR		0x120	/* Transfer Control Register */
+#define		AT91_PDC_RXTEN		(1 << 0)	/* Receiver Transfer Enable */
+#define		AT91_PDC_RXTDIS		(1 << 1)	/* Receiver Transfer Disable */
+#define		AT91_PDC_TXTEN		(1 << 8)	/* Transmitter Transfer Enable */
+#define		AT91_PDC_TXTDIS		(1 << 9)	/* Transmitter Transfer Disable */
+
+#define AT91_PDC_PTSR		0x124	/* Transfer Status Register */
+
+#endif
diff --git a/include/asm-avr32/arch-at32ap/at91rm9200_usart.h b/include/asm-avr32/arch-at32ap/at91rm9200_usart.h
new file mode 100644
index 000000000000..79f851e31b9c
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/at91rm9200_usart.h
@@ -0,0 +1,123 @@
+/*
+ * include/asm-arm/arch-at91rm9200/at91rm9200_usart.h
+ *
+ * Copyright (C) 2005 Ivan Kokshaysky
+ * Copyright (C) SAN People
+ *
+ * USART registers.
+ * Based on AT91RM9200 datasheet revision E.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#ifndef AT91RM9200_USART_H
+#define AT91RM9200_USART_H
+
+#define AT91_US_CR		0x00			/* Control Register */
+#define		AT91_US_RSTRX		(1 <<  2)		/* Reset Receiver */
+#define		AT91_US_RSTTX		(1 <<  3)		/* Reset Transmitter */
+#define		AT91_US_RXEN		(1 <<  4)		/* Receiver Enable */
+#define		AT91_US_RXDIS		(1 <<  5)		/* Receiver Disable */
+#define		AT91_US_TXEN		(1 <<  6)		/* Transmitter Enable */
+#define		AT91_US_TXDIS		(1 <<  7)		/* Transmitter Disable */
+#define		AT91_US_RSTSTA		(1 <<  8)		/* Reset Status Bits */
+#define		AT91_US_STTBRK		(1 <<  9)		/* Start Break */
+#define		AT91_US_STPBRK		(1 << 10)		/* Stop Break */
+#define		AT91_US_STTTO		(1 << 11)		/* Start Time-out */
+#define		AT91_US_SENDA		(1 << 12)		/* Send Address */
+#define		AT91_US_RSTIT		(1 << 13)		/* Reset Iterations */
+#define		AT91_US_RSTNACK		(1 << 14)		/* Reset Non Acknowledge */
+#define		AT91_US_RETTO		(1 << 15)		/* Rearm Time-out */
+#define		AT91_US_DTREN		(1 << 16)		/* Data Terminal Ready Enable */
+#define		AT91_US_DTRDIS		(1 << 17)		/* Data Terminal Ready Disable */
+#define		AT91_US_RTSEN		(1 << 18)		/* Request To Send Enable */
+#define		AT91_US_RTSDIS		(1 << 19)		/* Request To Send Disable */
+
+#define AT91_US_MR		0x04			/* Mode Register */
+#define		AT91_US_USMODE		(0xf <<  0)		/* Mode of the USART */
+#define			AT91_US_USMODE_NORMAL		0
+#define			AT91_US_USMODE_RS485		1
+#define			AT91_US_USMODE_HWHS		2
+#define			AT91_US_USMODE_MODEM		3
+#define			AT91_US_USMODE_ISO7816_T0	4
+#define			AT91_US_USMODE_ISO7816_T1	6
+#define			AT91_US_USMODE_IRDA		8
+#define		AT91_US_USCLKS		(3   <<  4)		/* Clock Selection */
+#define		AT91_US_CHRL		(3   <<  6)		/* Character Length */
+#define			AT91_US_CHRL_5			(0 <<  6)
+#define			AT91_US_CHRL_6			(1 <<  6)
+#define			AT91_US_CHRL_7			(2 <<  6)
+#define			AT91_US_CHRL_8			(3 <<  6)
+#define		AT91_US_SYNC		(1 <<  8)		/* Synchronous Mode Select */
+#define		AT91_US_PAR		(7 <<  9)		/* Parity Type */
+#define			AT91_US_PAR_EVEN		(0 <<  9)
+#define			AT91_US_PAR_ODD			(1 <<  9)
+#define			AT91_US_PAR_SPACE		(2 <<  9)
+#define			AT91_US_PAR_MARK		(3 <<  9)
+#define			AT91_US_PAR_NONE		(4 <<  9)
+#define			AT91_US_PAR_MULTI_DROP		(6 <<  9)
+#define		AT91_US_NBSTOP		(3 << 12)		/* Number of Stop Bits */
+#define			AT91_US_NBSTOP_1		(0 << 12)
+#define			AT91_US_NBSTOP_1_5		(1 << 12)
+#define			AT91_US_NBSTOP_2		(2 << 12)
+#define		AT91_US_CHMODE		(3 << 14)		/* Channel Mode */
+#define			AT91_US_CHMODE_NORMAL		(0 << 14)
+#define			AT91_US_CHMODE_ECHO		(1 << 14)
+#define			AT91_US_CHMODE_LOC_LOOP		(2 << 14)
+#define			AT91_US_CHMODE_REM_LOOP		(3 << 14)
+#define		AT91_US_MSBF		(1 << 16)		/* Bit Order */
+#define		AT91_US_MODE9		(1 << 17)		/* 9-bit Character Length */
+#define		AT91_US_CLKO		(1 << 18)		/* Clock Output Select */
+#define		AT91_US_OVER		(1 << 19)		/* Oversampling Mode */
+#define		AT91_US_INACK		(1 << 20)		/* Inhibit Non Acknowledge */
+#define		AT91_US_DSNACK		(1 << 21)		/* Disable Successive NACK */
+#define		AT91_US_MAX_ITER	(7 << 24)		/* Max Iterations */
+#define		AT91_US_FILTER		(1 << 28)		/* Infrared Receive Line Filter */
+
+#define AT91_US_IER		0x08			/* Interrupt Enable Register */
+#define		AT91_US_RXRDY		(1 <<  0)		/* Receiver Ready */
+#define		AT91_US_TXRDY		(1 <<  1)		/* Transmitter Ready */
+#define		AT91_US_RXBRK		(1 <<  2)		/* Break Received / End of Break */
+#define		AT91_US_ENDRX		(1 <<  3)		/* End of Receiver Transfer */
+#define		AT91_US_ENDTX		(1 <<  4)		/* End of Transmitter Transfer */
+#define		AT91_US_OVRE		(1 <<  5)		/* Overrun Error */
+#define		AT91_US_FRAME		(1 <<  6)		/* Framing Error */
+#define		AT91_US_PARE		(1 <<  7)		/* Parity Error */
+#define		AT91_US_TIMEOUT		(1 <<  8)		/* Receiver Time-out */
+#define		AT91_US_TXEMPTY		(1 <<  9)		/* Transmitter Empty */
+#define		AT91_US_ITERATION	(1 << 10)		/* Max number of Repetitions Reached */
+#define		AT91_US_TXBUFE		(1 << 11)		/* Transmission Buffer Empty */
+#define		AT91_US_RXBUFF		(1 << 12)		/* Reception Buffer Full */
+#define		AT91_US_NACK		(1 << 13)		/* Non Acknowledge */
+#define		AT91_US_RIIC		(1 << 16)		/* Ring Indicator Input Change */
+#define		AT91_US_DSRIC		(1 << 17)		/* Data Set Ready Input Change */
+#define		AT91_US_DCDIC		(1 << 18)		/* Data Carrier Detect Input Change */
+#define		AT91_US_CTSIC		(1 << 19)		/* Clear to Send Input Change */
+#define		AT91_US_RI		(1 << 20)		/* RI */
+#define		AT91_US_DSR		(1 << 21)		/* DSR */
+#define		AT91_US_DCD		(1 << 22)		/* DCD */
+#define		AT91_US_CTS		(1 << 23)		/* CTS */
+
+#define AT91_US_IDR		0x0c			/* Interrupt Disable Register */
+#define AT91_US_IMR		0x10			/* Interrupt Mask Register */
+#define AT91_US_CSR		0x14			/* Channel Status Register */
+#define AT91_US_RHR		0x18			/* Receiver Holding Register */
+#define AT91_US_THR		0x1c			/* Transmitter Holding Register */
+
+#define AT91_US_BRGR		0x20			/* Baud Rate Generator Register */
+#define		AT91_US_CD		(0xffff << 0)		/* Clock Divider */
+
+#define AT91_US_RTOR		0x24			/* Receiver Time-out Register */
+#define		AT91_US_TO		(0xffff << 0)		/* Time-out Value */
+
+#define AT91_US_TTGR		0x28			/* Transmitter Timeguard Register */
+#define		AT91_US_TG		(0xff << 0)		/* Timeguard Value */
+
+#define AT91_US_FIDI		0x40			/* FI DI Ratio Register */
+#define AT91_US_NER		0x44			/* Number of Errors Register */
+#define AT91_US_IF		0x4c			/* IrDA Filter Register */
+
+#endif
diff --git a/include/asm-avr32/arch-at32ap/board.h b/include/asm-avr32/arch-at32ap/board.h
new file mode 100644
index 000000000000..39368e18ab20
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/board.h
@@ -0,0 +1,35 @@
+/*
+ * Platform data definitions.
+ */
+#ifndef __ASM_ARCH_BOARD_H
+#define __ASM_ARCH_BOARD_H
+
+#include <linux/types.h>
+
+/* Add basic devices: system manager, interrupt controller, portmuxes, etc. */
+void at32_add_system_devices(void);
+
+#define AT91_NR_UART	4
+extern struct platform_device *at91_default_console_device;
+
+struct platform_device *at32_add_device_usart(unsigned int id);
+
+struct eth_platform_data {
+	u8	valid;
+	u8	mii_phy_addr;
+	u8	is_rmii;
+	u8	hw_addr[6];
+};
+struct platform_device *
+at32_add_device_eth(unsigned int id, struct eth_platform_data *data);
+
+struct platform_device *at32_add_device_spi(unsigned int id);
+
+struct lcdc_platform_data {
+	unsigned long fbmem_start;
+	unsigned long fbmem_size;
+};
+struct platform_device *
+at32_add_device_lcdc(unsigned int id, struct lcdc_platform_data *data);
+
+#endif /* __ASM_ARCH_BOARD_H */
diff --git a/include/asm-avr32/arch-at32ap/init.h b/include/asm-avr32/arch-at32ap/init.h
new file mode 100644
index 000000000000..43722634e069
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/init.h
@@ -0,0 +1,21 @@
+/*
+ * AT32AP platform initialization calls.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32AP_INIT_H__
+#define __ASM_AVR32_AT32AP_INIT_H__
+
+void setup_platform(void);
+
+/* Called by setup_platform */
+void at32_clock_init(void);
+void at32_portmux_init(void);
+
+void at32_setup_serial_console(unsigned int usart_id);
+
+#endif /* __ASM_AVR32_AT32AP_INIT_H__ */
diff --git a/include/asm-avr32/arch-at32ap/portmux.h b/include/asm-avr32/arch-at32ap/portmux.h
new file mode 100644
index 000000000000..4d50421262a1
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/portmux.h
@@ -0,0 +1,16 @@
+/*
+ * AT32 portmux interface.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32_PORTMUX_H__
+#define __ASM_AVR32_AT32_PORTMUX_H__
+
+void portmux_set_func(unsigned int portmux_id, unsigned int pin_id,
+		      unsigned int function_id);
+
+#endif /* __ASM_AVR32_AT32_PORTMUX_H__ */
diff --git a/include/asm-avr32/arch-at32ap/sm.h b/include/asm-avr32/arch-at32ap/sm.h
new file mode 100644
index 000000000000..265a9ead20bf
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/sm.h
@@ -0,0 +1,27 @@
+/*
+ * AT32 System Manager interface.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_AT32_SM_H__
+#define __ASM_AVR32_AT32_SM_H__
+
+struct irq_chip;
+struct platform_device;
+
+struct at32_sm {
+	spinlock_t lock;
+	void __iomem *regs;
+	struct irq_chip *eim_chip;
+	unsigned int eim_first_irq;
+	struct platform_device *pdev;
+};
+
+extern struct platform_device at32_sm_device;
+extern struct at32_sm system_manager;
+
+#endif /* __ASM_AVR32_AT32_SM_H__ */
diff --git a/include/asm-avr32/asm.h b/include/asm-avr32/asm.h
new file mode 100644
index 000000000000..515c7618952b
--- /dev/null
+++ b/include/asm-avr32/asm.h
@@ -0,0 +1,102 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ASM_H__
+#define __ASM_AVR32_ASM_H__
+
+#include <asm/sysreg.h>
+#include <asm/asm-offsets.h>
+#include <asm/thread_info.h>
+
+#define mask_interrupts		ssrf	SR_GM_BIT
+#define mask_exceptions		ssrf	SR_EM_BIT
+#define unmask_interrupts	csrf	SR_GM_BIT
+#define unmask_exceptions	csrf	SR_EM_BIT
+
+#ifdef CONFIG_FRAME_POINTER
+	.macro	save_fp
+	st.w	--sp, r7
+	.endm
+	.macro	restore_fp
+	ld.w	r7, sp++
+	.endm
+	.macro	zero_fp
+	mov	r7, 0
+	.endm
+#else
+	.macro	save_fp
+	.endm
+	.macro	restore_fp
+	.endm
+	.macro	zero_fp
+	.endm
+#endif
+	.macro	get_thread_info reg
+	mov	\reg, sp
+	andl	\reg, ~(THREAD_SIZE - 1) & 0xffff
+	.endm
+
+	/* Save and restore registers */
+	.macro	save_min sr, tmp=lr
+	pushm	lr
+	mfsr	\tmp, \sr
+	zero_fp
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_min sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	popm	lr
+	.endm
+
+	.macro	save_half sr, tmp=lr
+	save_fp
+	pushm	r8-r9,r10,r11,r12,lr
+	zero_fp
+	mfsr	\tmp, \sr
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_half sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	popm	r8-r9,r10,r11,r12,lr
+	restore_fp
+	.endm
+
+	.macro	save_full_user sr, tmp=lr
+	stmts	--sp, r0,r1,r2,r3,r4,r5,r6,r7,r8,r9,r10,r11,r12,sp,lr
+	st.w	--sp, lr
+	zero_fp
+	mfsr	\tmp, \sr
+	st.w	--sp, \tmp
+	.endm
+
+	.macro	restore_full_user sr, tmp=lr
+	ld.w	\tmp, sp++
+	mtsr	\sr, \tmp
+	ld.w	lr, sp++
+	ldmts	sp++, r0,r1,r2,r3,r4,r5,r6,r7,r8,r9,r10,r11,r12,sp,lr
+	.endm
+
+	/* uaccess macros */
+	.macro branch_if_kernel scratch, label
+	get_thread_info \scratch
+	ld.w	\scratch, \scratch[TI_flags]
+	bld	\scratch, TIF_USERSPACE
+	brcc	\label
+	.endm
+
+	.macro ret_if_privileged scratch, addr, size, ret
+	sub	\scratch, \size, 1
+	add	\scratch, \addr
+	retcs	\ret
+	retmi	\ret
+	.endm
+
+#endif /* __ASM_AVR32_ASM_H__ */
diff --git a/include/asm-avr32/atomic.h b/include/asm-avr32/atomic.h
new file mode 100644
index 000000000000..e0b9c44c126c
--- /dev/null
+++ b/include/asm-avr32/atomic.h
@@ -0,0 +1,201 @@
+/*
+ * Atomic operations that C can't guarantee us.  Useful for
+ * resource counting etc.
+ *
+ * But use these as seldom as possible since they are slower than
+ * regular operations.
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_ATOMIC_H
+#define __ASM_AVR32_ATOMIC_H
+
+#include <asm/system.h>
+
+typedef struct { volatile int counter; } atomic_t;
+#define ATOMIC_INIT(i)  { (i) }
+
+#define atomic_read(v)		((v)->counter)
+#define atomic_set(v, i)	(((v)->counter) = i)
+
+/*
+ * atomic_sub_return - subtract the atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically subtracts @i from @v. Returns the resulting value.
+ */
+static inline int atomic_sub_return(int i, atomic_t *v)
+{
+	int result;
+
+	asm volatile(
+		"/* atomic_sub_return */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	sub	%0, %3\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(i)
+		: "cc");
+
+	return result;
+}
+
+/*
+ * atomic_add_return - add integer to atomic variable
+ * @i: integer value to add
+ * @v: pointer of type atomic_t
+ *
+ * Atomically adds @i to @v. Returns the resulting value.
+ */
+static inline int atomic_add_return(int i, atomic_t *v)
+{
+	int result;
+
+	if (__builtin_constant_p(i))
+		result = atomic_sub_return(-i, v);
+	else
+		asm volatile(
+			"/* atomic_add_return */\n"
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %1\n"
+			"	add	%0, %3\n"
+			"	stcond	%2, %0\n"
+			"	brne	1b"
+			: "=&r"(result), "=o"(v->counter)
+			: "m"(v->counter), "r"(i)
+			: "cc", "memory");
+
+	return result;
+}
+
+/*
+ * atomic_sub_unless - sub unless the number is a given value
+ * @v: pointer of type atomic_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * If the atomic value v is not equal to u, this function subtracts a
+ * from v, and returns non zero. If v is equal to u then it returns
+ * zero. This is done as an atomic operation.
+*/
+static inline int atomic_sub_unless(atomic_t *v, int a, int u)
+{
+	int tmp, result = 0;
+
+	asm volatile(
+		"/* atomic_sub_unless */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %3\n"
+		"	cp.w	%0, %5\n"
+		"	breq	1f\n"
+		"	sub	%0, %4\n"
+		"	stcond	%2, %0\n"
+		"	brne	1b\n"
+		"	mov	%1, 1\n"
+		"1:"
+		: "=&r"(tmp), "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(a), "ir"(u)
+		: "cc", "memory");
+
+	return result;
+}
+
+/*
+ * atomic_add_unless - add unless the number is a given value
+ * @v: pointer of type atomic_t
+ * @a: the amount to add to v...
+ * @u: ...unless v is equal to u.
+ *
+ * If the atomic value v is not equal to u, this function adds a to v,
+ * and returns non zero. If v is equal to u then it returns zero. This
+ * is done as an atomic operation.
+*/
+static inline int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int tmp, result;
+
+	if (__builtin_constant_p(a))
+		result = atomic_sub_unless(v, -a, u);
+	else {
+		result = 0;
+		asm volatile(
+			"/* atomic_add_unless */\n"
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	cp.w	%0, %5\n"
+			"	breq	1f\n"
+			"	add	%0, %4\n"
+			"	stcond	%2, %0\n"
+			"	brne	1b\n"
+			"	mov	%1, 1\n"
+			"1:"
+			: "=&r"(tmp), "=&r"(result), "=o"(v->counter)
+			: "m"(v->counter), "r"(a), "ir"(u)
+			: "cc", "memory");
+	}
+
+	return result;
+}
+
+/*
+ * atomic_sub_if_positive - conditionally subtract integer from atomic variable
+ * @i: integer value to subtract
+ * @v: pointer of type atomic_t
+ *
+ * Atomically test @v and subtract @i if @v is greater or equal than @i.
+ * The function returns the old value of @v minus @i.
+ */
+static inline int atomic_sub_if_positive(int i, atomic_t *v)
+{
+	int result;
+
+	asm volatile(
+		"/* atomic_sub_if_positive */\n"
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	sub	%0, %3\n"
+		"	brlt	1f\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b\n"
+		"1:"
+		: "=&r"(result), "=o"(v->counter)
+		: "m"(v->counter), "ir"(i)
+		: "cc", "memory");
+
+	return result;
+}
+
+#define atomic_xchg(v, new)	(xchg(&((v)->counter), new))
+#define atomic_cmpxchg(v, o, n)	((int)cmpxchg(&((v)->counter), (o), (n)))
+
+#define atomic_sub(i, v)	(void)atomic_sub_return(i, v)
+#define atomic_add(i, v)	(void)atomic_add_return(i, v)
+#define atomic_dec(v)		atomic_sub(1, (v))
+#define atomic_inc(v)		atomic_add(1, (v))
+
+#define atomic_dec_return(v)	atomic_sub_return(1, v)
+#define atomic_inc_return(v)	atomic_add_return(1, v)
+
+#define atomic_sub_and_test(i, v) (atomic_sub_return(i, v) == 0)
+#define atomic_inc_and_test(v) (atomic_add_return(1, v) == 0)
+#define atomic_dec_and_test(v) (atomic_sub_return(1, v) == 0)
+#define atomic_add_negative(i, v) (atomic_add_return(i, v) < 0)
+
+#define atomic_inc_not_zero(v)	atomic_add_unless(v, 1, 0)
+#define atomic_dec_if_positive(v) atomic_sub_if_positive(1, v)
+
+#define smp_mb__before_atomic_dec()	barrier()
+#define smp_mb__after_atomic_dec()	barrier()
+#define smp_mb__before_atomic_inc()	barrier()
+#define smp_mb__after_atomic_inc()	barrier()
+
+#include <asm-generic/atomic.h>
+
+#endif /*  __ASM_AVR32_ATOMIC_H */
diff --git a/include/asm-avr32/auxvec.h b/include/asm-avr32/auxvec.h
new file mode 100644
index 000000000000..d5dd435bf8f4
--- /dev/null
+++ b/include/asm-avr32/auxvec.h
@@ -0,0 +1,4 @@
+#ifndef __ASM_AVR32_AUXVEC_H
+#define __ASM_AVR32_AUXVEC_H
+
+#endif /* __ASM_AVR32_AUXVEC_H */
diff --git a/include/asm-avr32/bitops.h b/include/asm-avr32/bitops.h
new file mode 100644
index 000000000000..5299f8c8e11d
--- /dev/null
+++ b/include/asm-avr32/bitops.h
@@ -0,0 +1,296 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_BITOPS_H
+#define __ASM_AVR32_BITOPS_H
+
+#include <asm/byteorder.h>
+#include <asm/system.h>
+
+/*
+ * clear_bit() doesn't provide any barrier for the compiler
+ */
+#define smp_mb__before_clear_bit()	barrier()
+#define smp_mb__after_clear_bit()	barrier()
+
+/*
+ * set_bit - Atomically set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * This function is atomic and may not be reordered.  See __set_bit()
+ * if you do not require the atomic guarantees.
+ *
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void set_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long tmp;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	sbr	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "i"(nr)
+			: "cc");
+	} else {
+		unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	or	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "r"(mask)
+			: "cc");
+	}
+}
+
+/*
+ * clear_bit - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * clear_bit() is atomic and may not be reordered.  However, it does
+ * not contain a memory barrier, so if it is used for locking purposes,
+ * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
+ * in order to ensure changes are visible on other processors.
+ */
+static inline void clear_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long tmp;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	cbr	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "i"(nr)
+			: "cc");
+	} else {
+		unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %2\n"
+			"	andn	%0, %3\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p)
+			: "m"(*p), "r"(mask)
+			: "cc");
+	}
+}
+
+/*
+ * change_bit - Toggle a bit in memory
+ * @nr: Bit to change
+ * @addr: Address to start counting from
+ *
+ * change_bit() is atomic and may not be reordered.
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void change_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%0, %2\n"
+		"	eor	%0, %3\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(tmp), "=o"(*p)
+		: "m"(*p), "r"(mask)
+		: "cc");
+}
+
+/*
+ * test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_set_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	sbr	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "i"(nr)
+			: "memory", "cc");
+	} else {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%2, %3\n"
+			"	or	%0, %2, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "r"(mask)
+			: "memory", "cc");
+	}
+
+	return (old & mask) != 0;
+}
+
+/*
+ * test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_clear_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	if (__builtin_constant_p(nr)) {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	cbr	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "i"(nr)
+			: "memory", "cc");
+	} else {
+		asm volatile(
+			"1:	ssrf	5\n"
+			"	ld.w	%0, %3\n"
+			"	mov	%2, %0\n"
+			"	andn	%0, %4\n"
+			"	stcond	%1, %0\n"
+			"	brne	1b"
+			: "=&r"(tmp), "=o"(*p), "=&r"(old)
+			: "m"(*p), "r"(mask)
+			: "memory", "cc");
+	}
+
+	return (old & mask) != 0;
+}
+
+/*
+ * test_and_change_bit - Change a bit and return its old value
+ * @nr: Bit to change
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int test_and_change_bit(int nr, volatile void * addr)
+{
+	unsigned long *p = ((unsigned long *)addr) + nr / BITS_PER_LONG;
+	unsigned long mask = 1UL << (nr % BITS_PER_LONG);
+	unsigned long tmp, old;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%2, %3\n"
+		"	eor	%0, %2, %4\n"
+		"	stcond	%1, %0\n"
+		"	brne	1b"
+		: "=&r"(tmp), "=o"(*p), "=&r"(old)
+		: "m"(*p), "r"(mask)
+		: "memory", "cc");
+
+	return (old & mask) != 0;
+}
+
+#include <asm-generic/bitops/non-atomic.h>
+
+/* Find First bit Set */
+static inline unsigned long __ffs(unsigned long word)
+{
+	unsigned long result;
+
+	asm("brev %1\n\t"
+	    "clz %0,%1"
+	    : "=r"(result), "=&r"(word)
+	    : "1"(word));
+	return result;
+}
+
+/* Find First Zero */
+static inline unsigned long ffz(unsigned long word)
+{
+	return __ffs(~word);
+}
+
+/* Find Last bit Set */
+static inline int fls(unsigned long word)
+{
+	unsigned long result;
+
+	asm("clz %0,%1" : "=r"(result) : "r"(word));
+	return 32 - result;
+}
+
+unsigned long find_first_zero_bit(const unsigned long *addr,
+				  unsigned long size);
+unsigned long find_next_zero_bit(const unsigned long *addr,
+				 unsigned long size,
+				 unsigned long offset);
+unsigned long find_first_bit(const unsigned long *addr,
+			     unsigned long size);
+unsigned long find_next_bit(const unsigned long *addr,
+				 unsigned long size,
+				 unsigned long offset);
+
+/*
+ * ffs: find first bit set. This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ *
+ * The difference is that bit numbering starts at 1, and if no bit is set,
+ * the function returns 0.
+ */
+static inline int ffs(unsigned long word)
+{
+	if(word == 0)
+		return 0;
+	return __ffs(word) + 1;
+}
+
+#include <asm-generic/bitops/fls64.h>
+#include <asm-generic/bitops/sched.h>
+#include <asm-generic/bitops/hweight.h>
+
+#include <asm-generic/bitops/ext2-non-atomic.h>
+#include <asm-generic/bitops/ext2-atomic.h>
+#include <asm-generic/bitops/minix-le.h>
+
+#endif /* __ASM_AVR32_BITOPS_H */
diff --git a/include/asm-avr32/bug.h b/include/asm-avr32/bug.h
new file mode 100644
index 000000000000..521766bc9366
--- /dev/null
+++ b/include/asm-avr32/bug.h
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_BUG_H
+#define __ASM_AVR32_BUG_H
+
+#ifdef CONFIG_BUG
+
+/*
+ * According to our Chief Architect, this compact opcode is very
+ * unlikely to ever be implemented.
+ */
+#define AVR32_BUG_OPCODE	0x5df0
+
+#ifdef CONFIG_DEBUG_BUGVERBOSE
+
+#define BUG()								\
+	do {								\
+		asm volatile(".hword	%0\n\t"				\
+			     ".hword	%1\n\t"				\
+			     ".long	%2"				\
+			     :						\
+			     : "n"(AVR32_BUG_OPCODE),			\
+			       "i"(__LINE__), "X"(__FILE__));		\
+	} while (0)
+
+#else
+
+#define BUG()								\
+	do {								\
+		asm volatile(".hword	%0\n\t"				\
+			     : : "n"(AVR32_BUG_OPCODE));		\
+	} while (0)
+
+#endif /* CONFIG_DEBUG_BUGVERBOSE */
+
+#define HAVE_ARCH_BUG
+
+#endif /* CONFIG_BUG */
+
+#include <asm-generic/bug.h>
+
+#endif /* __ASM_AVR32_BUG_H */
diff --git a/include/asm-avr32/bugs.h b/include/asm-avr32/bugs.h
new file mode 100644
index 000000000000..7635e770622e
--- /dev/null
+++ b/include/asm-avr32/bugs.h
@@ -0,0 +1,15 @@
+/*
+ * This is included by init/main.c to check for architecture-dependent bugs.
+ *
+ * Needs:
+ *      void check_bugs(void);
+ */
+#ifndef __ASM_AVR32_BUGS_H
+#define __ASM_AVR32_BUGS_H
+
+static void __init check_bugs(void)
+{
+	cpu_data->loops_per_jiffy = loops_per_jiffy;
+}
+
+#endif /* __ASM_AVR32_BUGS_H */
diff --git a/include/asm-avr32/byteorder.h b/include/asm-avr32/byteorder.h
new file mode 100644
index 000000000000..402ff4125cdc
--- /dev/null
+++ b/include/asm-avr32/byteorder.h
@@ -0,0 +1,25 @@
+/*
+ * AVR32 endian-conversion functions.
+ */
+#ifndef __ASM_AVR32_BYTEORDER_H
+#define __ASM_AVR32_BYTEORDER_H
+
+#include <asm/types.h>
+#include <linux/compiler.h>
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_bswap_32(unsigned long x);
+extern unsigned short __builtin_bswap_16(unsigned short x);
+#endif
+
+#define __arch__swab32(x) __builtin_bswap_32(x)
+#define __arch__swab16(x) __builtin_bswap_16(x)
+
+#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
+# define __BYTEORDER_HAS_U64__
+# define __SWAB_64_THRU_32__
+#endif
+
+#include <linux/byteorder/big_endian.h>
+
+#endif /* __ASM_AVR32_BYTEORDER_H */
diff --git a/include/asm-avr32/cache.h b/include/asm-avr32/cache.h
new file mode 100644
index 000000000000..dabb955f3c00
--- /dev/null
+++ b/include/asm-avr32/cache.h
@@ -0,0 +1,29 @@
+#ifndef __ASM_AVR32_CACHE_H
+#define __ASM_AVR32_CACHE_H
+
+#define L1_CACHE_SHIFT 5
+#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
+
+#ifndef __ASSEMBLER__
+struct cache_info {
+	unsigned int ways;
+	unsigned int sets;
+	unsigned int linesz;
+};
+#endif /* __ASSEMBLER */
+
+/* Cache operation constants */
+#define ICACHE_FLUSH		0x00
+#define ICACHE_INVALIDATE	0x01
+#define ICACHE_LOCK		0x02
+#define ICACHE_UNLOCK		0x03
+#define ICACHE_PREFETCH		0x04
+
+#define DCACHE_FLUSH		0x08
+#define DCACHE_LOCK		0x09
+#define DCACHE_UNLOCK		0x0a
+#define DCACHE_INVALIDATE	0x0b
+#define DCACHE_CLEAN		0x0c
+#define DCACHE_CLEAN_INVAL	0x0d
+
+#endif /* __ASM_AVR32_CACHE_H */
diff --git a/include/asm-avr32/cachectl.h b/include/asm-avr32/cachectl.h
new file mode 100644
index 000000000000..4faf1ce60061
--- /dev/null
+++ b/include/asm-avr32/cachectl.h
@@ -0,0 +1,11 @@
+#ifndef __ASM_AVR32_CACHECTL_H
+#define __ASM_AVR32_CACHECTL_H
+
+/*
+ * Operations that can be performed through the cacheflush system call
+ */
+
+/* Clean the data cache, then invalidate the icache */
+#define CACHE_IFLUSH	0
+
+#endif /* __ASM_AVR32_CACHECTL_H */
diff --git a/include/asm-avr32/cacheflush.h b/include/asm-avr32/cacheflush.h
new file mode 100644
index 000000000000..f1bf1708980e
--- /dev/null
+++ b/include/asm-avr32/cacheflush.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_CACHEFLUSH_H
+#define __ASM_AVR32_CACHEFLUSH_H
+
+/* Keep includes the same across arches.  */
+#include <linux/mm.h>
+
+#define CACHE_OP_ICACHE_INVALIDATE	0x01
+#define CACHE_OP_DCACHE_INVALIDATE	0x0b
+#define CACHE_OP_DCACHE_CLEAN		0x0c
+#define CACHE_OP_DCACHE_CLEAN_INVAL	0x0d
+
+/*
+ * Invalidate any cacheline containing virtual address vaddr without
+ * writing anything back to memory.
+ *
+ * Note that this function may corrupt unrelated data structures when
+ * applied on buffers that are not cacheline aligned in both ends.
+ */
+static inline void invalidate_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_INVALIDATE)
+		     : "memory");
+}
+
+/*
+ * Make sure any cacheline containing virtual address vaddr is written
+ * to memory.
+ */
+static inline void clean_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_CLEAN)
+		     : "memory");
+}
+
+/*
+ * Make sure any cacheline containing virtual address vaddr is written
+ * to memory and then invalidate it.
+ */
+static inline void flush_dcache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_DCACHE_CLEAN_INVAL)
+		     : "memory");
+}
+
+/*
+ * Invalidate any instruction cacheline containing virtual address
+ * vaddr.
+ */
+static inline void invalidate_icache_line(void *vaddr)
+{
+	asm volatile("cache %0[0], %1"
+		     :
+		     : "r"(vaddr), "n"(CACHE_OP_ICACHE_INVALIDATE)
+		     : "memory");
+}
+
+/*
+ * Applies the above functions on all lines that are touched by the
+ * specified virtual address range.
+ */
+void invalidate_dcache_region(void *start, size_t len);
+void clean_dcache_region(void *start, size_t len);
+void flush_dcache_region(void *start, size_t len);
+void invalidate_icache_region(void *start, size_t len);
+
+/*
+ * Make sure any pending writes are completed before continuing.
+ */
+#define flush_write_buffer() asm volatile("sync 0" : : : "memory")
+
+/*
+ * The following functions are called when a virtual mapping changes.
+ * We do not need to flush anything in this case.
+ */
+#define flush_cache_all()			do { } while (0)
+#define flush_cache_mm(mm)			do { } while (0)
+#define flush_cache_range(vma, start, end)	do { } while (0)
+#define flush_cache_page(vma, vmaddr, pfn)	do { } while (0)
+#define flush_cache_vmap(start, end)		do { } while (0)
+#define flush_cache_vunmap(start, end)		do { } while (0)
+
+/*
+ * I think we need to implement this one to be able to reliably
+ * execute pages from RAMDISK. However, if we implement the
+ * flush_dcache_*() functions, it might not be needed anymore.
+ *
+ * #define flush_icache_page(vma, page)		do { } while (0)
+ */
+extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
+
+/*
+ * These are (I think) related to D-cache aliasing.  We might need to
+ * do something here, but only for certain configurations.  No such
+ * configurations exist at this time.
+ */
+#define flush_dcache_page(page)			do { } while (0)
+#define flush_dcache_mmap_lock(page)		do { } while (0)
+#define flush_dcache_mmap_unlock(page)		do { } while (0)
+
+/*
+ * These are for I/D cache coherency. In this case, we do need to
+ * flush with all configurations.
+ */
+extern void flush_icache_range(unsigned long start, unsigned long end);
+extern void flush_icache_user_range(struct vm_area_struct *vma,
+				    struct page *page,
+				    unsigned long addr, int len);
+
+#define copy_to_user_page(vma, page, vaddr, dst, src, len) do {	\
+	memcpy(dst, src, len);					\
+	flush_icache_user_range(vma, page, vaddr, len);		\
+} while(0)
+#define copy_from_user_page(vma, page, vaddr, dst, src, len)	\
+	memcpy(dst, src, len)
+
+#endif /* __ASM_AVR32_CACHEFLUSH_H */
diff --git a/include/asm-avr32/checksum.h b/include/asm-avr32/checksum.h
new file mode 100644
index 000000000000..41b7af09edc4
--- /dev/null
+++ b/include/asm-avr32/checksum.h
@@ -0,0 +1,156 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_CHECKSUM_H
+#define __ASM_AVR32_CHECKSUM_H
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+unsigned int csum_partial(const unsigned char * buff, int len,
+			  unsigned int sum);
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums, and handles user-space pointer exceptions correctly, when needed.
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+unsigned int csum_partial_copy_generic(const char *src, char *dst, int len,
+				       int sum, int *src_err_ptr,
+				       int *dst_err_ptr);
+
+/*
+ *	Note: when you get a NULL pointer exception here this means someone
+ *	passed in an incorrect kernel address to one of these functions.
+ *
+ *	If you use these functions directly please don't forget the
+ *	verify_area().
+ */
+static inline
+unsigned int csum_partial_copy_nocheck(const char *src, char *dst,
+				       int len, int sum)
+{
+	return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
+}
+
+static inline
+unsigned int csum_partial_copy_from_user (const char __user *src, char *dst,
+					  int len, int sum, int *err_ptr)
+{
+	return csum_partial_copy_generic((const char __force *)src, dst, len,
+					 sum, err_ptr, NULL);
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+static inline unsigned short ip_fast_csum(unsigned char *iph,
+					  unsigned int ihl)
+{
+	unsigned int sum, tmp;
+
+	__asm__ __volatile__(
+		"	ld.w	%0, %1++\n"
+		"	ld.w	%3, %1++\n"
+		"	sub	%2, 4\n"
+		"	add	%0, %3\n"
+		"	ld.w	%3, %1++\n"
+		"	adc	%0, %0, %3\n"
+		"	ld.w	%3, %1++\n"
+		"	adc	%0, %0, %3\n"
+		"	acr	%0\n"
+		"1:	ld.w	%3, %1++\n"
+		"	add	%0, %3\n"
+		"	acr	%0\n"
+		"	sub	%2, 1\n"
+		"	brne	1b\n"
+		"	lsl	%3, %0, 16\n"
+		"	andl	%0, 0\n"
+		"	mov	%2, 0xffff\n"
+		"	add	%0, %3\n"
+		"	adc	%0, %0, %2\n"
+		"	com	%0\n"
+		"	lsr	%0, 16\n"
+		: "=r"(sum), "=r"(iph), "=r"(ihl), "=r"(tmp)
+		: "1"(iph), "2"(ihl)
+		: "memory", "cc");
+	return sum;
+}
+
+/*
+ *	Fold a partial checksum
+ */
+
+static inline unsigned int csum_fold(unsigned int sum)
+{
+	unsigned int tmp;
+
+	asm("	bfextu	%1, %0, 0, 16\n"
+	    "	lsr	%0, 16\n"
+	    "	add	%0, %1\n"
+	    "	bfextu	%1, %0, 16, 16\n"
+	    "	add	%0, %1"
+	    : "=&r"(sum), "=&r"(tmp)
+	    : "0"(sum));
+
+	return ~sum;
+}
+
+static inline unsigned long csum_tcpudp_nofold(unsigned long saddr,
+					       unsigned long daddr,
+					       unsigned short len,
+					       unsigned short proto,
+					       unsigned int sum)
+{
+	asm("	add	%0, %1\n"
+	    "	adc	%0, %0, %2\n"
+	    "	adc	%0, %0, %3\n"
+	    "	acr	%0"
+	    : "=r"(sum)
+	    : "r"(daddr), "r"(saddr), "r"(ntohs(len) | (proto << 16)),
+	      "0"(sum)
+	    : "cc");
+
+	return sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+static inline unsigned short int csum_tcpudp_magic(unsigned long saddr,
+						   unsigned long daddr,
+						   unsigned short len,
+						   unsigned short proto,
+						   unsigned int sum)
+{
+	return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum));
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+
+static inline unsigned short ip_compute_csum(unsigned char * buff, int len)
+{
+    return csum_fold(csum_partial(buff, len, 0));
+}
+
+#endif /* __ASM_AVR32_CHECKSUM_H */
diff --git a/include/asm-avr32/cputime.h b/include/asm-avr32/cputime.h
new file mode 100644
index 000000000000..e87e0f81cbeb
--- /dev/null
+++ b/include/asm-avr32/cputime.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_CPUTIME_H
+#define __ASM_AVR32_CPUTIME_H
+
+#include <asm-generic/cputime.h>
+
+#endif /* __ASM_AVR32_CPUTIME_H */
diff --git a/include/asm-avr32/current.h b/include/asm-avr32/current.h
new file mode 100644
index 000000000000..c7b0549eab8a
--- /dev/null
+++ b/include/asm-avr32/current.h
@@ -0,0 +1,15 @@
+#ifndef __ASM_AVR32_CURRENT_H
+#define __ASM_AVR32_CURRENT_H
+
+#include <linux/thread_info.h>
+
+struct task_struct;
+
+inline static struct task_struct * get_current(void)
+{
+	return current_thread_info()->task;
+}
+
+#define current get_current()
+
+#endif /* __ASM_AVR32_CURRENT_H */
diff --git a/include/asm-avr32/delay.h b/include/asm-avr32/delay.h
new file mode 100644
index 000000000000..cc3b2e3343b3
--- /dev/null
+++ b/include/asm-avr32/delay.h
@@ -0,0 +1,26 @@
+#ifndef __ASM_AVR32_DELAY_H
+#define __ASM_AVR32_DELAY_H
+
+/*
+ * Copyright (C) 1993 Linus Torvalds
+ *
+ * Delay routines calling functions in arch/avr32/lib/delay.c
+ */
+
+extern void __bad_udelay(void);
+extern void __bad_ndelay(void);
+
+extern void __udelay(unsigned long usecs);
+extern void __ndelay(unsigned long nsecs);
+extern void __const_udelay(unsigned long usecs);
+extern void __delay(unsigned long loops);
+
+#define udelay(n) (__builtin_constant_p(n) ? \
+	((n) > 20000 ? __bad_udelay() : __const_udelay((n) * 0x10c6ul)) : \
+	__udelay(n))
+
+#define ndelay(n) (__builtin_constant_p(n) ? \
+	((n) > 20000 ? __bad_ndelay() : __const_udelay((n) * 5ul)) : \
+	__ndelay(n))
+
+#endif /* __ASM_AVR32_DELAY_H */
diff --git a/include/asm-avr32/div64.h b/include/asm-avr32/div64.h
new file mode 100644
index 000000000000..d7ddd4fdeca6
--- /dev/null
+++ b/include/asm-avr32/div64.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_DIV64_H
+#define __ASM_AVR32_DIV64_H
+
+#include <asm-generic/div64.h>
+
+#endif /* __ASM_AVR32_DIV64_H */
diff --git a/include/asm-avr32/dma-mapping.h b/include/asm-avr32/dma-mapping.h
new file mode 100644
index 000000000000..4c40cb41cdf8
--- /dev/null
+++ b/include/asm-avr32/dma-mapping.h
@@ -0,0 +1,320 @@
+#ifndef __ASM_AVR32_DMA_MAPPING_H
+#define __ASM_AVR32_DMA_MAPPING_H
+
+#include <linux/mm.h>
+#include <linux/device.h>
+#include <asm/scatterlist.h>
+#include <asm/processor.h>
+#include <asm/cacheflush.h>
+#include <asm/io.h>
+
+extern void dma_cache_sync(void *vaddr, size_t size, int direction);
+
+/*
+ * Return whether the given device DMA address mask can be supported
+ * properly.  For example, if your device can only drive the low 24-bits
+ * during bus mastering, then you would pass 0x00ffffff as the mask
+ * to this function.
+ */
+static inline int dma_supported(struct device *dev, u64 mask)
+{
+	/* Fix when needed. I really don't know of any limitations */
+	return 1;
+}
+
+static inline int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
+		return -EIO;
+
+	*dev->dma_mask = dma_mask;
+	return 0;
+}
+
+/**
+ * dma_alloc_coherent - allocate consistent memory for DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: required memory size
+ * @handle: bus-specific DMA address
+ *
+ * Allocate some uncached, unbuffered memory for a device for
+ * performing DMA.  This function allocates pages, and will
+ * return the CPU-viewed address, and sets @handle to be the
+ * device-viewed address.
+ */
+extern void *dma_alloc_coherent(struct device *dev, size_t size,
+				dma_addr_t *handle, gfp_t gfp);
+
+/**
+ * dma_free_coherent - free memory allocated by dma_alloc_coherent
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: size of memory originally requested in dma_alloc_coherent
+ * @cpu_addr: CPU-view address returned from dma_alloc_coherent
+ * @handle: device-view address returned from dma_alloc_coherent
+ *
+ * Free (and unmap) a DMA buffer previously allocated by
+ * dma_alloc_coherent().
+ *
+ * References to memory and mappings associated with cpu_addr/handle
+ * during and after this call executing are illegal.
+ */
+extern void dma_free_coherent(struct device *dev, size_t size,
+			      void *cpu_addr, dma_addr_t handle);
+
+/**
+ * dma_alloc_writecombine - allocate write-combining memory for DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: required memory size
+ * @handle: bus-specific DMA address
+ *
+ * Allocate some uncached, buffered memory for a device for
+ * performing DMA.  This function allocates pages, and will
+ * return the CPU-viewed address, and sets @handle to be the
+ * device-viewed address.
+ */
+extern void *dma_alloc_writecombine(struct device *dev, size_t size,
+				    dma_addr_t *handle, gfp_t gfp);
+
+/**
+ * dma_free_coherent - free memory allocated by dma_alloc_writecombine
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @size: size of memory originally requested in dma_alloc_writecombine
+ * @cpu_addr: CPU-view address returned from dma_alloc_writecombine
+ * @handle: device-view address returned from dma_alloc_writecombine
+ *
+ * Free (and unmap) a DMA buffer previously allocated by
+ * dma_alloc_writecombine().
+ *
+ * References to memory and mappings associated with cpu_addr/handle
+ * during and after this call executing are illegal.
+ */
+extern void dma_free_writecombine(struct device *dev, size_t size,
+				  void *cpu_addr, dma_addr_t handle);
+
+/**
+ * dma_map_single - map a single buffer for streaming DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @cpu_addr: CPU direct mapped address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Ensure that any data held in the cache is appropriately discarded
+ * or written back.
+ *
+ * The device owns this memory once this call has completed.  The CPU
+ * can regain ownership by calling dma_unmap_single() or dma_sync_single().
+ */
+static inline dma_addr_t
+dma_map_single(struct device *dev, void *cpu_addr, size_t size,
+	       enum dma_data_direction direction)
+{
+	dma_cache_sync(cpu_addr, size, direction);
+	return virt_to_bus(cpu_addr);
+}
+
+/**
+ * dma_unmap_single - unmap a single buffer previously mapped
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a single streaming mode DMA translation.  The handle and size
+ * must match what was provided in the previous dma_map_single() call.
+ * All other usages are undefined.
+ *
+ * After this call, reads by the CPU to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+static inline void
+dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
+		 enum dma_data_direction direction)
+{
+
+}
+
+/**
+ * dma_map_page - map a portion of a page for streaming DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @page: page that buffer resides in
+ * @offset: offset into page for start of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Ensure that any data held in the cache is appropriately discarded
+ * or written back.
+ *
+ * The device owns this memory once this call has completed.  The CPU
+ * can regain ownership by calling dma_unmap_page() or dma_sync_single().
+ */
+static inline dma_addr_t
+dma_map_page(struct device *dev, struct page *page,
+	     unsigned long offset, size_t size,
+	     enum dma_data_direction direction)
+{
+	return dma_map_single(dev, page_address(page) + offset,
+			      size, direction);
+}
+
+/**
+ * dma_unmap_page - unmap a buffer previously mapped through dma_map_page()
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a single streaming mode DMA translation.  The handle and size
+ * must match what was provided in the previous dma_map_single() call.
+ * All other usages are undefined.
+ *
+ * After this call, reads by the CPU to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+static inline void
+dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
+	       enum dma_data_direction direction)
+{
+	dma_unmap_single(dev, dma_address, size, direction);
+}
+
+/**
+ * dma_map_sg - map a set of SG buffers for streaming mode DMA
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Map a set of buffers described by scatterlist in streaming
+ * mode for DMA.  This is the scatter-gather version of the
+ * above pci_map_single interface.  Here the scatter gather list
+ * elements are each tagged with the appropriate dma address
+ * and length.  They are obtained via sg_dma_{address,length}(SG).
+ *
+ * NOTE: An implementation may be able to use a smaller number of
+ *       DMA address/length pairs than there are SG table elements.
+ *       (for example via virtual mapping capabilities)
+ *       The routine returns the number of addr/length pairs actually
+ *       used, at most nents.
+ *
+ * Device ownership issues as mentioned above for pci_map_single are
+ * the same here.
+ */
+static inline int
+dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
+	   enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		char *virt;
+
+		sg[i].dma_address = page_to_bus(sg[i].page) + sg[i].offset;
+		virt = page_address(sg[i].page) + sg[i].offset;
+		dma_cache_sync(virt, sg[i].length, direction);
+	}
+
+	return nents;
+}
+
+/**
+ * dma_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Unmap a set of streaming mode DMA translations.
+ * Again, CPU read rules concerning calls here are the same as for
+ * pci_unmap_single() above.
+ */
+static inline void
+dma_unmap_sg(struct device *dev, struct scatterlist *sg, int nhwentries,
+	     enum dma_data_direction direction)
+{
+
+}
+
+/**
+ * dma_sync_single_for_cpu
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @handle: DMA address of buffer
+ * @size: size of buffer to map
+ * @dir: DMA transfer direction
+ *
+ * Make physical memory consistent for a single streaming mode DMA
+ * translation after a transfer.
+ *
+ * If you perform a dma_map_single() but wish to interrogate the
+ * buffer using the cpu, yet do not wish to teardown the DMA mapping,
+ * you must call this function before doing so.  At the next point you
+ * give the DMA address back to the card, you must first perform a
+ * dma_sync_single_for_device, and then the device again owns the
+ * buffer.
+ */
+static inline void
+dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
+			size_t size, enum dma_data_direction direction)
+{
+	dma_cache_sync(bus_to_virt(dma_handle), size, direction);
+}
+
+static inline void
+dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
+			   size_t size, enum dma_data_direction direction)
+{
+	dma_cache_sync(bus_to_virt(dma_handle), size, direction);
+}
+
+/**
+ * dma_sync_sg_for_cpu
+ * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices
+ * @sg: list of buffers
+ * @nents: number of buffers to map
+ * @dir: DMA transfer direction
+ *
+ * Make physical memory consistent for a set of streaming
+ * mode DMA translations after a transfer.
+ *
+ * The same as dma_sync_single_for_* but for a scatter-gather list,
+ * same rules and usage.
+ */
+static inline void
+dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
+		    int nents, enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		dma_cache_sync(page_address(sg[i].page) + sg[i].offset,
+			       sg[i].length, direction);
+	}
+}
+
+static inline void
+dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
+		       int nents, enum dma_data_direction direction)
+{
+	int i;
+
+	for (i = 0; i < nents; i++) {
+		dma_cache_sync(page_address(sg[i].page) + sg[i].offset,
+			       sg[i].length, direction);
+	}
+}
+
+/* Now for the API extensions over the pci_ one */
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+
+static inline int dma_is_consistent(dma_addr_t dma_addr)
+{
+	return 1;
+}
+
+static inline int dma_get_cache_alignment(void)
+{
+	return boot_cpu_data.dcache.linesz;
+}
+
+#endif /* __ASM_AVR32_DMA_MAPPING_H */
diff --git a/include/asm-avr32/dma.h b/include/asm-avr32/dma.h
new file mode 100644
index 000000000000..9e91205590ac
--- /dev/null
+++ b/include/asm-avr32/dma.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_AVR32_DMA_H
+#define __ASM_AVR32_DMA_H
+
+/* The maximum address that we can perform a DMA transfer to on this platform.
+ * Not really applicable to AVR32, but some functions need it. */
+#define MAX_DMA_ADDRESS		0xffffffff
+
+#endif /* __ASM_AVR32_DMA_H */
diff --git a/include/asm-avr32/elf.h b/include/asm-avr32/elf.h
new file mode 100644
index 000000000000..d334b4994d2d
--- /dev/null
+++ b/include/asm-avr32/elf.h
@@ -0,0 +1,110 @@
+#ifndef __ASM_AVR32_ELF_H
+#define __ASM_AVR32_ELF_H
+
+/* AVR32 relocation numbers */
+#define R_AVR32_NONE		0
+#define R_AVR32_32		1
+#define R_AVR32_16		2
+#define R_AVR32_8		3
+#define R_AVR32_32_PCREL	4
+#define R_AVR32_16_PCREL	5
+#define R_AVR32_8_PCREL		6
+#define R_AVR32_DIFF32		7
+#define R_AVR32_DIFF16		8
+#define R_AVR32_DIFF8		9
+#define R_AVR32_GOT32		10
+#define R_AVR32_GOT16		11
+#define R_AVR32_GOT8		12
+#define R_AVR32_21S		13
+#define R_AVR32_16U		14
+#define R_AVR32_16S		15
+#define R_AVR32_8S		16
+#define R_AVR32_8S_EXT		17
+#define R_AVR32_22H_PCREL	18
+#define R_AVR32_18W_PCREL	19
+#define R_AVR32_16B_PCREL	20
+#define R_AVR32_16N_PCREL	21
+#define R_AVR32_14UW_PCREL	22
+#define R_AVR32_11H_PCREL	23
+#define R_AVR32_10UW_PCREL	24
+#define R_AVR32_9H_PCREL	25
+#define R_AVR32_9UW_PCREL	26
+#define R_AVR32_HI16		27
+#define R_AVR32_LO16		28
+#define R_AVR32_GOTPC		29
+#define R_AVR32_GOTCALL		30
+#define R_AVR32_LDA_GOT		31
+#define R_AVR32_GOT21S		32
+#define R_AVR32_GOT18SW		33
+#define R_AVR32_GOT16S		34
+#define R_AVR32_GOT7UW		35
+#define R_AVR32_32_CPENT	36
+#define R_AVR32_CPCALL		37
+#define R_AVR32_16_CP		38
+#define R_AVR32_9W_CP		39
+#define R_AVR32_RELATIVE	40
+#define R_AVR32_GLOB_DAT	41
+#define R_AVR32_JMP_SLOT	42
+#define R_AVR32_ALIGN		43
+
+/*
+ * ELF register definitions..
+ */
+
+#include <asm/ptrace.h>
+#include <asm/user.h>
+
+typedef unsigned long elf_greg_t;
+
+#define ELF_NGREG (sizeof (struct pt_regs) / sizeof (elf_greg_t))
+typedef elf_greg_t elf_gregset_t[ELF_NGREG];
+
+typedef struct user_fpu_struct elf_fpregset_t;
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ */
+#define elf_check_arch(x) ( (x)->e_machine == EM_AVR32 )
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+#define ELF_CLASS	ELFCLASS32
+#ifdef __LITTLE_ENDIAN__
+#define ELF_DATA	ELFDATA2LSB
+#else
+#define ELF_DATA	ELFDATA2MSB
+#endif
+#define ELF_ARCH	EM_AVR32
+
+#define USE_ELF_CORE_DUMP
+#define ELF_EXEC_PAGESIZE	4096
+
+/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
+   use of this is to invoke "./ld.so someprog" to test out a new version of
+   the loader.  We need to make sure that it is out of the way of the program
+   that it will "exec", and that there is sufficient room for the brk.  */
+
+#define ELF_ET_DYN_BASE         (2 * TASK_SIZE / 3)
+
+
+/* This yields a mask that user programs can use to figure out what
+   instruction set this CPU supports.  This could be done in user space,
+   but it's not easy, and we've already done it here.  */
+
+#define ELF_HWCAP	(0)
+
+/* This yields a string that ld.so will use to load implementation
+   specific libraries for optimization.  This is more specific in
+   intent than poking at uname or /proc/cpuinfo.
+
+   For the moment, we have only optimizations for the Intel generations,
+   but that could change... */
+
+#define ELF_PLATFORM  (NULL)
+
+#ifdef __KERNEL__
+#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
+#endif
+
+#endif /* __ASM_AVR32_ELF_H */
diff --git a/include/asm-avr32/emergency-restart.h b/include/asm-avr32/emergency-restart.h
new file mode 100644
index 000000000000..3e7e014776ba
--- /dev/null
+++ b/include/asm-avr32/emergency-restart.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_EMERGENCY_RESTART_H
+#define __ASM_AVR32_EMERGENCY_RESTART_H
+
+#include <asm-generic/emergency-restart.h>
+
+#endif /* __ASM_AVR32_EMERGENCY_RESTART_H */
diff --git a/include/asm-avr32/errno.h b/include/asm-avr32/errno.h
new file mode 100644
index 000000000000..558a7249f06d
--- /dev/null
+++ b/include/asm-avr32/errno.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_ERRNO_H
+#define __ASM_AVR32_ERRNO_H
+
+#include <asm-generic/errno.h>
+
+#endif /* __ASM_AVR32_ERRNO_H */
diff --git a/include/asm-avr32/fcntl.h b/include/asm-avr32/fcntl.h
new file mode 100644
index 000000000000..14c0c4402b11
--- /dev/null
+++ b/include/asm-avr32/fcntl.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_FCNTL_H
+#define __ASM_AVR32_FCNTL_H
+
+#include <asm-generic/fcntl.h>
+
+#endif /* __ASM_AVR32_FCNTL_H */
diff --git a/include/asm-avr32/futex.h b/include/asm-avr32/futex.h
new file mode 100644
index 000000000000..10419f14a68a
--- /dev/null
+++ b/include/asm-avr32/futex.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_FUTEX_H
+#define __ASM_AVR32_FUTEX_H
+
+#include <asm-generic/futex.h>
+
+#endif /* __ASM_AVR32_FUTEX_H */
diff --git a/include/asm-avr32/hardirq.h b/include/asm-avr32/hardirq.h
new file mode 100644
index 000000000000..267354356f60
--- /dev/null
+++ b/include/asm-avr32/hardirq.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_AVR32_HARDIRQ_H
+#define __ASM_AVR32_HARDIRQ_H
+
+#include <linux/threads.h>
+#include <asm/irq.h>
+
+#ifndef __ASSEMBLY__
+
+#include <linux/cache.h>
+
+/* entry.S is sensitive to the offsets of these fields */
+typedef struct {
+	unsigned int __softirq_pending;
+} ____cacheline_aligned irq_cpustat_t;
+
+void ack_bad_irq(unsigned int irq);
+
+/* Standard mappings for irq_cpustat_t above */
+#include <linux/irq_cpustat.h>
+
+#endif /* __ASSEMBLY__ */
+
+#define HARDIRQ_BITS	12
+
+/*
+ * The hardirq mask has to be large enough to have
+ * space for potentially all IRQ sources in the system
+ * nesting on a single CPU:
+ */
+#if (1 << HARDIRQ_BITS) < NR_IRQS
+# error HARDIRQ_BITS is too low!
+#endif
+
+#endif /* __ASM_AVR32_HARDIRQ_H */
diff --git a/include/asm-avr32/hw_irq.h b/include/asm-avr32/hw_irq.h
new file mode 100644
index 000000000000..218b0a6bfd1b
--- /dev/null
+++ b/include/asm-avr32/hw_irq.h
@@ -0,0 +1,9 @@
+#ifndef __ASM_AVR32_HW_IRQ_H
+#define __ASM_AVR32_HW_IRQ_H
+
+static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i)
+{
+	/* Nothing to do */
+}
+
+#endif /* __ASM_AVR32_HW_IRQ_H */
diff --git a/include/asm-avr32/intc.h b/include/asm-avr32/intc.h
new file mode 100644
index 000000000000..1ac9ca75e8fd
--- /dev/null
+++ b/include/asm-avr32/intc.h
@@ -0,0 +1,128 @@
+#ifndef __ASM_AVR32_INTC_H
+#define __ASM_AVR32_INTC_H
+
+#include <linux/sysdev.h>
+#include <linux/interrupt.h>
+
+struct irq_controller;
+struct irqaction;
+struct pt_regs;
+
+struct platform_device;
+
+/* Information about the internal interrupt controller */
+struct intc_device {
+	/* ioremapped address of configuration block */
+	void __iomem *regs;
+
+	/* the physical device */
+	struct platform_device *pdev;
+
+	/* Number of interrupt lines per group. */
+	unsigned int irqs_per_group;
+
+	/* The highest group ID + 1 */
+	unsigned int nr_groups;
+
+	/*
+	 * Bitfield indicating which groups are actually in use.  The
+	 * size of the array is
+	 * ceil(group_max / (8 * sizeof(unsigned int))).
+	 */
+	unsigned int group_mask[];
+};
+
+struct irq_controller_class {
+	/*
+	 * A short name identifying this kind of controller.
+	 */
+	const char *typename;
+	/*
+	 * Handle the IRQ.  Must do any necessary acking and masking.
+	 */
+	irqreturn_t (*handle)(int irq, void *dev_id, struct pt_regs *regs);
+	/*
+	 * Register a new IRQ handler.
+	 */
+	int (*setup)(struct irq_controller *ctrl, unsigned int irq,
+		     struct irqaction *action);
+	/*
+	 * Unregister a IRQ handler.
+	 */
+	void (*free)(struct irq_controller *ctrl, unsigned int irq,
+		     void *dev_id);
+	/*
+	 * Mask the IRQ in the interrupt controller.
+	 */
+	void (*mask)(struct irq_controller *ctrl, unsigned int irq);
+	/*
+	 * Unmask the IRQ in the interrupt controller.
+	 */
+	void (*unmask)(struct irq_controller *ctrl, unsigned int irq);
+	/*
+	 * Set the type of the IRQ. See below for possible types.
+	 * Return -EINVAL if a given type is not supported
+	 */
+	int (*set_type)(struct irq_controller *ctrl, unsigned int irq,
+			unsigned int type);
+	/*
+	 * Return the IRQ type currently set
+	 */
+	unsigned int (*get_type)(struct irq_controller *ctrl, unsigned int irq);
+};
+
+struct irq_controller {
+	struct irq_controller_class *class;
+	unsigned int irq_group;
+	unsigned int first_irq;
+	unsigned int nr_irqs;
+	struct list_head list;
+};
+
+struct intc_group_desc {
+	struct irq_controller *ctrl;
+	irqreturn_t (*handle)(int, void *, struct pt_regs *);
+	unsigned long flags;
+	void *dev_id;
+	const char *devname;
+};
+
+/*
+ * The internal interrupt controller.  Defined in board/part-specific
+ * devices.c.
+ * TODO: Should probably be defined per-cpu.
+ */
+extern struct intc_device intc;
+
+extern int request_internal_irq(unsigned int irq,
+				irqreturn_t (*handler)(int, void *, struct pt_regs *),
+				unsigned long irqflags,
+				const char *devname, void *dev_id);
+extern void free_internal_irq(unsigned int irq);
+
+/* Only used by time_init() */
+extern int setup_internal_irq(unsigned int irq, struct intc_group_desc *desc);
+
+/*
+ * Set interrupt priority for a given group. `group' can be found by
+ * using irq_to_group(irq). Priority can be from 0 (lowest) to 3
+ * (highest). Higher-priority interrupts will preempt lower-priority
+ * interrupts (unless interrupts are masked globally).
+ *
+ * This function does not check for conflicts within a group.
+ */
+extern int intc_set_priority(unsigned int group,
+			     unsigned int priority);
+
+/*
+ * Returns a bitmask of pending interrupts in a group.
+ */
+extern unsigned long intc_get_pending(unsigned int group);
+
+/*
+ * Register a new external interrupt controller.  Returns the first
+ * external IRQ number that is assigned to the new controller.
+ */
+extern int intc_register_controller(struct irq_controller *ctrl);
+
+#endif /* __ASM_AVR32_INTC_H */
diff --git a/include/asm-avr32/io.h b/include/asm-avr32/io.h
new file mode 100644
index 000000000000..2fc8f111dce9
--- /dev/null
+++ b/include/asm-avr32/io.h
@@ -0,0 +1,253 @@
+#ifndef __ASM_AVR32_IO_H
+#define __ASM_AVR32_IO_H
+
+#include <linux/string.h>
+
+#ifdef __KERNEL__
+
+#include <asm/addrspace.h>
+#include <asm/byteorder.h>
+
+/* virt_to_phys will only work when address is in P1 or P2 */
+static __inline__ unsigned long virt_to_phys(volatile void *address)
+{
+	return PHYSADDR(address);
+}
+
+static __inline__ void * phys_to_virt(unsigned long address)
+{
+	return (void *)P1SEGADDR(address);
+}
+
+#define cached_to_phys(addr)	((unsigned long)PHYSADDR(addr))
+#define uncached_to_phys(addr)	((unsigned long)PHYSADDR(addr))
+#define phys_to_cached(addr)	((void *)P1SEGADDR(addr))
+#define phys_to_uncached(addr)	((void *)P2SEGADDR(addr))
+
+/*
+ * Generic IO read/write.  These perform native-endian accesses.  Note
+ * that some architectures will want to re-define __raw_{read,write}w.
+ */
+extern void __raw_writesb(unsigned int addr, const void *data, int bytelen);
+extern void __raw_writesw(unsigned int addr, const void *data, int wordlen);
+extern void __raw_writesl(unsigned int addr, const void *data, int longlen);
+
+extern void __raw_readsb(unsigned int addr, void *data, int bytelen);
+extern void __raw_readsw(unsigned int addr, void *data, int wordlen);
+extern void __raw_readsl(unsigned int addr, void *data, int longlen);
+
+static inline void writeb(unsigned char b, volatile void __iomem *addr)
+{
+	*(volatile unsigned char __force *)addr = b;
+}
+static inline void writew(unsigned short b, volatile void __iomem *addr)
+{
+	*(volatile unsigned short __force *)addr = b;
+}
+static inline void writel(unsigned int b, volatile void __iomem *addr)
+{
+	*(volatile unsigned int __force *)addr = b;
+}
+#define __raw_writeb writeb
+#define __raw_writew writew
+#define __raw_writel writel
+
+static inline unsigned char readb(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned char __force *)addr;
+}
+static inline unsigned short readw(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned short __force *)addr;
+}
+static inline unsigned int readl(const volatile void __iomem *addr)
+{
+	return *(const volatile unsigned int __force *)addr;
+}
+#define __raw_readb readb
+#define __raw_readw readw
+#define __raw_readl readl
+
+#define writesb(p, d, l)	__raw_writesb((unsigned int)p, d, l)
+#define writesw(p, d, l)	__raw_writesw((unsigned int)p, d, l)
+#define writesl(p, d, l)	__raw_writesl((unsigned int)p, d, l)
+
+#define readsb(p, d, l)		__raw_readsb((unsigned int)p, d, l)
+#define readsw(p, d, l)		__raw_readsw((unsigned int)p, d, l)
+#define readsl(p, d, l)		__raw_readsl((unsigned int)p, d, l)
+
+/*
+ * These two are only here because ALSA _thinks_ it needs them...
+ */
+static inline void memcpy_fromio(void * to, const volatile void __iomem *from,
+				 unsigned long count)
+{
+	char *p = to;
+	while (count) {
+		count--;
+		*p = readb(from);
+		p++;
+		from++;
+	}
+}
+
+static inline void  memcpy_toio(volatile void __iomem *to, const void * from,
+				unsigned long count)
+{
+	const char *p = from;
+	while (count) {
+		count--;
+		writeb(*p, to);
+		p++;
+		to++;
+	}
+}
+
+static inline void memset_io(volatile void __iomem *addr, unsigned char val,
+			     unsigned long count)
+{
+	memset((void __force *)addr, val, count);
+}
+
+/*
+ * Bad read/write accesses...
+ */
+extern void __readwrite_bug(const char *fn);
+
+#define IO_SPACE_LIMIT	0xffffffff
+
+/* Convert I/O port address to virtual address */
+#define __io(p)		((void __iomem *)phys_to_uncached(p))
+
+/*
+ *  IO port access primitives
+ *  -------------------------
+ *
+ * The AVR32 doesn't have special IO access instructions; all IO is memory
+ * mapped. Note that these are defined to perform little endian accesses
+ * only. Their primary purpose is to access PCI and ISA peripherals.
+ *
+ * Note that for a big endian machine, this implies that the following
+ * big endian mode connectivity is in place.
+ *
+ * The machine specific io.h include defines __io to translate an "IO"
+ * address to a memory address.
+ *
+ * Note that we prevent GCC re-ordering or caching values in expressions
+ * by introducing sequence points into the in*() definitions.  Note that
+ * __raw_* do not guarantee this behaviour.
+ *
+ * The {in,out}[bwl] macros are for emulating x86-style PCI/ISA IO space.
+ */
+#define outb(v, p)		__raw_writeb(v, __io(p))
+#define outw(v, p)		__raw_writew(cpu_to_le16(v), __io(p))
+#define outl(v, p)		__raw_writel(cpu_to_le32(v), __io(p))
+
+#define inb(p)			__raw_readb(__io(p))
+#define inw(p)			le16_to_cpu(__raw_readw(__io(p)))
+#define inl(p)			le32_to_cpu(__raw_readl(__io(p)))
+
+static inline void __outsb(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outb(*(u8 *)addr, port);
+		addr++;
+	}
+}
+
+static inline void __insb(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u8 *)addr = inb(port);
+		addr++;
+	}
+}
+
+static inline void __outsw(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outw(*(u16 *)addr, port);
+		addr += 2;
+	}
+}
+
+static inline void __insw(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u16 *)addr = inw(port);
+		addr += 2;
+	}
+}
+
+static inline void __outsl(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		outl(*(u32 *)addr, port);
+		addr += 4;
+	}
+}
+
+static inline void __insl(unsigned long port, void *addr, unsigned int count)
+{
+	while (count--) {
+		*(u32 *)addr = inl(port);
+		addr += 4;
+	}
+}
+
+#define outsb(port, addr, count)	__outsb(port, addr, count)
+#define insb(port, addr, count)		__insb(port, addr, count)
+#define outsw(port, addr, count)	__outsw(port, addr, count)
+#define insw(port, addr, count)		__insw(port, addr, count)
+#define outsl(port, addr, count)	__outsl(port, addr, count)
+#define insl(port, addr, count)		__insl(port, addr, count)
+
+extern void __iomem *__ioremap(unsigned long offset, size_t size,
+			       unsigned long flags);
+extern void __iounmap(void __iomem *addr);
+
+/*
+ * ioremap	-   map bus memory into CPU space
+ * @offset	bus address of the memory
+ * @size	size of the resource to map
+ *
+ * ioremap performs a platform specific sequence of operations to make
+ * bus memory CPU accessible via the readb/.../writel functions and
+ * the other mmio helpers. The returned address is not guaranteed to
+ * be usable directly as a virtual address.
+ */
+#define ioremap(offset, size)			\
+	__ioremap((offset), (size), 0)
+
+#define iounmap(addr)				\
+	__iounmap(addr)
+
+#define cached(addr) P1SEGADDR(addr)
+#define uncached(addr) P2SEGADDR(addr)
+
+#define virt_to_bus virt_to_phys
+#define bus_to_virt phys_to_virt
+#define page_to_bus page_to_phys
+#define bus_to_page phys_to_page
+
+#define dma_cache_wback_inv(_start, _size)	\
+	flush_dcache_region(_start, _size)
+#define dma_cache_inv(_start, _size)		\
+	invalidate_dcache_region(_start, _size)
+#define dma_cache_wback(_start, _size)		\
+	clean_dcache_region(_start, _size)
+
+/*
+ * Convert a physical pointer to a virtual kernel pointer for /dev/mem
+ * access
+ */
+#define xlate_dev_mem_ptr(p)    __va(p)
+
+/*
+ * Convert a virtual cached pointer to an uncached pointer
+ */
+#define xlate_dev_kmem_ptr(p)   p
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_AVR32_IO_H */
diff --git a/include/asm-avr32/ioctl.h b/include/asm-avr32/ioctl.h
new file mode 100644
index 000000000000..c8472c1398ef
--- /dev/null
+++ b/include/asm-avr32/ioctl.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_IOCTL_H
+#define __ASM_AVR32_IOCTL_H
+
+#include <asm-generic/ioctl.h>
+
+#endif /* __ASM_AVR32_IOCTL_H */
diff --git a/include/asm-avr32/ioctls.h b/include/asm-avr32/ioctls.h
new file mode 100644
index 000000000000..0500426b7186
--- /dev/null
+++ b/include/asm-avr32/ioctls.h
@@ -0,0 +1,83 @@
+#ifndef __ASM_AVR32_IOCTLS_H
+#define __ASM_AVR32_IOCTLS_H
+
+#include <asm/ioctl.h>
+
+/* 0x54 is just a magic number to make these relatively unique ('T') */
+
+#define TCGETS		0x5401
+#define TCSETS		0x5402 /* Clashes with SNDCTL_TMR_START sound ioctl */
+#define TCSETSW		0x5403
+#define TCSETSF		0x5404
+#define TCGETA		0x5405
+#define TCSETA		0x5406
+#define TCSETAW		0x5407
+#define TCSETAF		0x5408
+#define TCSBRK		0x5409
+#define TCXONC		0x540A
+#define TCFLSH		0x540B
+#define TIOCEXCL	0x540C
+#define TIOCNXCL	0x540D
+#define TIOCSCTTY	0x540E
+#define TIOCGPGRP	0x540F
+#define TIOCSPGRP	0x5410
+#define TIOCOUTQ	0x5411
+#define TIOCSTI		0x5412
+#define TIOCGWINSZ	0x5413
+#define TIOCSWINSZ	0x5414
+#define TIOCMGET	0x5415
+#define TIOCMBIS	0x5416
+#define TIOCMBIC	0x5417
+#define TIOCMSET	0x5418
+#define TIOCGSOFTCAR	0x5419
+#define TIOCSSOFTCAR	0x541A
+#define FIONREAD	0x541B
+#define TIOCINQ		FIONREAD
+#define TIOCLINUX	0x541C
+#define TIOCCONS	0x541D
+#define TIOCGSERIAL	0x541E
+#define TIOCSSERIAL	0x541F
+#define TIOCPKT		0x5420
+#define FIONBIO		0x5421
+#define TIOCNOTTY	0x5422
+#define TIOCSETD	0x5423
+#define TIOCGETD	0x5424
+#define TCSBRKP		0x5425	/* Needed for POSIX tcsendbreak() */
+/* #define TIOCTTYGSTRUCT 0x5426 - Former debugging-only ioctl */
+#define TIOCSBRK	0x5427  /* BSD compatibility */
+#define TIOCCBRK	0x5428  /* BSD compatibility */
+#define TIOCGSID	0x5429  /* Return the session ID of FD */
+#define TIOCGPTN	_IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
+#define TIOCSPTLCK	_IOW('T',0x31, int)  /* Lock/unlock Pty */
+
+#define FIONCLEX	0x5450
+#define FIOCLEX		0x5451
+#define FIOASYNC	0x5452
+#define TIOCSERCONFIG	0x5453
+#define TIOCSERGWILD	0x5454
+#define TIOCSERSWILD	0x5455
+#define TIOCGLCKTRMIOS	0x5456
+#define TIOCSLCKTRMIOS	0x5457
+#define TIOCSERGSTRUCT	0x5458 /* For debugging only */
+#define TIOCSERGETLSR   0x5459 /* Get line status register */
+#define TIOCSERGETMULTI 0x545A /* Get multiport config  */
+#define TIOCSERSETMULTI 0x545B /* Set multiport config */
+
+#define TIOCMIWAIT	0x545C	/* wait for a change on serial input line(s) */
+#define TIOCGICOUNT	0x545D	/* read serial port inline interrupt counts */
+#define TIOCGHAYESESP   0x545E  /* Get Hayes ESP configuration */
+#define TIOCSHAYESESP   0x545F  /* Set Hayes ESP configuration */
+#define FIOQSIZE	0x5460
+
+/* Used for packet mode */
+#define TIOCPKT_DATA		 0
+#define TIOCPKT_FLUSHREAD	 1
+#define TIOCPKT_FLUSHWRITE	 2
+#define TIOCPKT_STOP		 4
+#define TIOCPKT_START		 8
+#define TIOCPKT_NOSTOP		16
+#define TIOCPKT_DOSTOP		32
+
+#define TIOCSER_TEMT    0x01	/* Transmitter physically empty */
+
+#endif /* __ASM_AVR32_IOCTLS_H */
diff --git a/include/asm-avr32/ipcbuf.h b/include/asm-avr32/ipcbuf.h
new file mode 100644
index 000000000000..1552c9698f5e
--- /dev/null
+++ b/include/asm-avr32/ipcbuf.h
@@ -0,0 +1,29 @@
+#ifndef __ASM_AVR32_IPCBUF_H
+#define __ASM_AVR32_IPCBUF_H
+
+/*
+* The user_ipc_perm structure for AVR32 architecture.
+* Note extra padding because this structure is passed back and forth
+* between kernel and user space.
+*
+* Pad space is left for:
+* - 32-bit mode_t and seq
+* - 2 miscellaneous 32-bit values
+*/
+
+struct ipc64_perm
+{
+        __kernel_key_t          key;
+        __kernel_uid32_t        uid;
+        __kernel_gid32_t        gid;
+        __kernel_uid32_t        cuid;
+        __kernel_gid32_t        cgid;
+        __kernel_mode_t         mode;
+        unsigned short          __pad1;
+        unsigned short          seq;
+        unsigned short          __pad2;
+        unsigned long           __unused1;
+        unsigned long           __unused2;
+};
+
+#endif /* __ASM_AVR32_IPCBUF_H */
diff --git a/include/asm-avr32/irq.h b/include/asm-avr32/irq.h
new file mode 100644
index 000000000000..f7e725707dd7
--- /dev/null
+++ b/include/asm-avr32/irq.h
@@ -0,0 +1,10 @@
+#ifndef __ASM_AVR32_IRQ_H
+#define __ASM_AVR32_IRQ_H
+
+#define NR_INTERNAL_IRQS	64
+#define NR_EXTERNAL_IRQS	64
+#define NR_IRQS			(NR_INTERNAL_IRQS + NR_EXTERNAL_IRQS)
+
+#define irq_canonicalize(i)	(i)
+
+#endif /* __ASM_AVR32_IOCTLS_H */
diff --git a/include/asm-avr32/irqflags.h b/include/asm-avr32/irqflags.h
new file mode 100644
index 000000000000..93570daac38a
--- /dev/null
+++ b/include/asm-avr32/irqflags.h
@@ -0,0 +1,68 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_IRQFLAGS_H
+#define __ASM_AVR32_IRQFLAGS_H
+
+#include <asm/sysreg.h>
+
+static inline unsigned long __raw_local_save_flags(void)
+{
+	return sysreg_read(SR);
+}
+
+#define raw_local_save_flags(x)					\
+	do { (x) = __raw_local_save_flags(); } while (0)
+
+/*
+ * This will restore ALL status register flags, not only the interrupt
+ * mask flag.
+ *
+ * The empty asm statement informs the compiler of this fact while
+ * also serving as a barrier.
+ */
+static inline void raw_local_irq_restore(unsigned long flags)
+{
+	sysreg_write(SR, flags);
+	asm volatile("" : : : "memory", "cc");
+}
+
+static inline void raw_local_irq_disable(void)
+{
+	asm volatile("ssrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
+}
+
+static inline void raw_local_irq_enable(void)
+{
+	asm volatile("csrf %0" : : "n"(SYSREG_GM_OFFSET) : "memory");
+}
+
+static inline int raw_irqs_disabled_flags(unsigned long flags)
+{
+	return (flags & SYSREG_BIT(GM)) != 0;
+}
+
+static inline int raw_irqs_disabled(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	return raw_irqs_disabled_flags(flags);
+}
+
+static inline unsigned long __raw_local_irq_save(void)
+{
+	unsigned long flags = __raw_local_save_flags();
+
+	raw_local_irq_disable();
+
+	return flags;
+}
+
+#define raw_local_irq_save(flags)				\
+	do { (flags) = __raw_local_irq_save(); } while (0)
+
+#endif /* __ASM_AVR32_IRQFLAGS_H */
diff --git a/include/asm-avr32/kdebug.h b/include/asm-avr32/kdebug.h
new file mode 100644
index 000000000000..f583b643ffb2
--- /dev/null
+++ b/include/asm-avr32/kdebug.h
@@ -0,0 +1,38 @@
+#ifndef __ASM_AVR32_KDEBUG_H
+#define __ASM_AVR32_KDEBUG_H
+
+#include <linux/notifier.h>
+
+struct pt_regs;
+
+struct die_args {
+	struct pt_regs *regs;
+	int trapnr;
+};
+
+int register_die_notifier(struct notifier_block *nb);
+int unregister_die_notifier(struct notifier_block *nb);
+int register_page_fault_notifier(struct notifier_block *nb);
+int unregister_page_fault_notifier(struct notifier_block *nb);
+extern struct atomic_notifier_head avr32_die_chain;
+
+/* Grossly misnamed. */
+enum die_val {
+	DIE_FAULT,
+	DIE_BREAKPOINT,
+	DIE_SSTEP,
+	DIE_PAGE_FAULT,
+};
+
+static inline int notify_die(enum die_val val, struct pt_regs *regs,
+			     int trap, int sig)
+{
+	struct die_args args = {
+		.regs = regs,
+		.trapnr = trap,
+	};
+
+	return atomic_notifier_call_chain(&avr32_die_chain, val, &args);
+}
+
+#endif /* __ASM_AVR32_KDEBUG_H */
diff --git a/include/asm-avr32/kmap_types.h b/include/asm-avr32/kmap_types.h
new file mode 100644
index 000000000000..b7f5c6870107
--- /dev/null
+++ b/include/asm-avr32/kmap_types.h
@@ -0,0 +1,30 @@
+#ifndef __ASM_AVR32_KMAP_TYPES_H
+#define __ASM_AVR32_KMAP_TYPES_H
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+# define D(n) __KM_FENCE_##n ,
+#else
+# define D(n)
+#endif
+
+enum km_type {
+D(0)	KM_BOUNCE_READ,
+D(1)	KM_SKB_SUNRPC_DATA,
+D(2)	KM_SKB_DATA_SOFTIRQ,
+D(3)	KM_USER0,
+D(4)	KM_USER1,
+D(5)	KM_BIO_SRC_IRQ,
+D(6)	KM_BIO_DST_IRQ,
+D(7)	KM_PTE0,
+D(8)	KM_PTE1,
+D(9)	KM_PTE2,
+D(10)	KM_IRQ0,
+D(11)	KM_IRQ1,
+D(12)	KM_SOFTIRQ0,
+D(13)	KM_SOFTIRQ1,
+D(14)	KM_TYPE_NR
+};
+
+#undef D
+
+#endif /* __ASM_AVR32_KMAP_TYPES_H */
diff --git a/include/asm-avr32/kprobes.h b/include/asm-avr32/kprobes.h
new file mode 100644
index 000000000000..09a5cbe2f896
--- /dev/null
+++ b/include/asm-avr32/kprobes.h
@@ -0,0 +1,34 @@
+/*
+ * Kernel Probes (KProbes)
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ * Copyright (C) IBM Corporation, 2002, 2004
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_KPROBES_H
+#define __ASM_AVR32_KPROBES_H
+
+#include <linux/types.h>
+
+typedef u16	kprobe_opcode_t;
+#define BREAKPOINT_INSTRUCTION	0xd673	/* breakpoint */
+#define MAX_INSN_SIZE		2
+
+#define ARCH_INACTIVE_KPROBE_COUNT 1
+
+#define arch_remove_kprobe(p)	do { } while (0)
+
+/* Architecture specific copy of original instruction */
+struct arch_specific_insn {
+	kprobe_opcode_t	insn[MAX_INSN_SIZE];
+};
+
+extern int kprobe_exceptions_notify(struct notifier_block *self,
+				    unsigned long val, void *data);
+
+#define flush_insn_slot(p)	do { } while (0)
+
+#endif /* __ASM_AVR32_KPROBES_H */
diff --git a/include/asm-avr32/linkage.h b/include/asm-avr32/linkage.h
new file mode 100644
index 000000000000..f7b285e910d4
--- /dev/null
+++ b/include/asm-avr32/linkage.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_LINKAGE_H
+#define __ASM_LINKAGE_H
+
+#define __ALIGN .balign 2
+#define __ALIGN_STR ".balign 2"
+
+#endif /* __ASM_LINKAGE_H */
diff --git a/include/asm-avr32/local.h b/include/asm-avr32/local.h
new file mode 100644
index 000000000000..1c1619694da3
--- /dev/null
+++ b/include/asm-avr32/local.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_LOCAL_H
+#define __ASM_AVR32_LOCAL_H
+
+#include <asm-generic/local.h>
+
+#endif /* __ASM_AVR32_LOCAL_H */
diff --git a/include/asm-avr32/mach/serial_at91.h b/include/asm-avr32/mach/serial_at91.h
new file mode 100644
index 000000000000..1290bb32802d
--- /dev/null
+++ b/include/asm-avr32/mach/serial_at91.h
@@ -0,0 +1,33 @@
+/*
+ *  linux/include/asm-arm/mach/serial_at91.h
+ *
+ *  Based on serial_sa1100.h  by Nicolas Pitre
+ *
+ *  Copyright (C) 2002 ATMEL Rousset
+ *
+ *  Low level machine dependent UART functions.
+ */
+
+struct uart_port;
+
+/*
+ * This is a temporary structure for registering these
+ * functions; it is intended to be discarded after boot.
+ */
+struct at91_port_fns {
+	void	(*set_mctrl)(struct uart_port *, u_int);
+	u_int	(*get_mctrl)(struct uart_port *);
+	void	(*enable_ms)(struct uart_port *);
+	void	(*pm)(struct uart_port *, u_int, u_int);
+	int	(*set_wake)(struct uart_port *, u_int);
+	int	(*open)(struct uart_port *);
+	void	(*close)(struct uart_port *);
+};
+
+#if defined(CONFIG_SERIAL_AT91)
+void at91_register_uart_fns(struct at91_port_fns *fns);
+#else
+#define at91_register_uart_fns(fns) do { } while (0)
+#endif
+
+
diff --git a/include/asm-avr32/mman.h b/include/asm-avr32/mman.h
new file mode 100644
index 000000000000..648f91e7187a
--- /dev/null
+++ b/include/asm-avr32/mman.h
@@ -0,0 +1,17 @@
+#ifndef __ASM_AVR32_MMAN_H__
+#define __ASM_AVR32_MMAN_H__
+
+#include <asm-generic/mman.h>
+
+#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
+#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
+#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
+#define MAP_LOCKED	0x2000		/* pages are locked */
+#define MAP_NORESERVE	0x4000		/* don't check for reservations */
+#define MAP_POPULATE	0x8000		/* populate (prefault) page tables */
+#define MAP_NONBLOCK	0x10000		/* do not block on IO */
+
+#define MCL_CURRENT	1		/* lock all current mappings */
+#define MCL_FUTURE	2		/* lock all future mappings */
+
+#endif /* __ASM_AVR32_MMAN_H__ */
diff --git a/include/asm-avr32/mmu.h b/include/asm-avr32/mmu.h
new file mode 100644
index 000000000000..60c2d2650d32
--- /dev/null
+++ b/include/asm-avr32/mmu.h
@@ -0,0 +1,10 @@
+#ifndef __ASM_AVR32_MMU_H
+#define __ASM_AVR32_MMU_H
+
+/* Default "unsigned long" context */
+typedef unsigned long mm_context_t;
+
+#define MMU_ITLB_ENTRIES	64
+#define MMU_DTLB_ENTRIES	64
+
+#endif /* __ASM_AVR32_MMU_H */
diff --git a/include/asm-avr32/mmu_context.h b/include/asm-avr32/mmu_context.h
new file mode 100644
index 000000000000..31add1ae8089
--- /dev/null
+++ b/include/asm-avr32/mmu_context.h
@@ -0,0 +1,148 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * ASID handling taken from SH implementation.
+ *   Copyright (C) 1999 Niibe Yutaka
+ *   Copyright (C) 2003 Paul Mundt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_MMU_CONTEXT_H
+#define __ASM_AVR32_MMU_CONTEXT_H
+
+#include <asm/tlbflush.h>
+#include <asm/pgalloc.h>
+#include <asm/sysreg.h>
+
+/*
+ * The MMU "context" consists of two things:
+ *    (a) TLB cache version
+ *    (b) ASID (Address Space IDentifier)
+ */
+#define MMU_CONTEXT_ASID_MASK		0x000000ff
+#define MMU_CONTEXT_VERSION_MASK	0xffffff00
+#define MMU_CONTEXT_FIRST_VERSION       0x00000100
+#define NO_CONTEXT			0
+
+#define MMU_NO_ASID			0x100
+
+/* Virtual Page Number mask */
+#define MMU_VPN_MASK	0xfffff000
+
+/* Cache of MMU context last used */
+extern unsigned long mmu_context_cache;
+
+/*
+ * Get MMU context if needed
+ */
+static inline void
+get_mmu_context(struct mm_struct *mm)
+{
+	unsigned long mc = mmu_context_cache;
+
+	if (((mm->context ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
+		/* It's up to date, do nothing */
+		return;
+
+	/* It's old, we need to get new context with new version */
+	mc = ++mmu_context_cache;
+	if (!(mc & MMU_CONTEXT_ASID_MASK)) {
+		/*
+		 * We have exhausted all ASIDs of this version.
+		 * Flush the TLB and start new cycle.
+		 */
+		flush_tlb_all();
+		/*
+		 * Fix version. Note that we avoid version #0
+		 * to distinguish NO_CONTEXT.
+		 */
+		if (!mc)
+			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
+	}
+	mm->context = mc;
+}
+
+/*
+ * Initialize the context related info for a new mm_struct
+ * instance.
+ */
+static inline int init_new_context(struct task_struct *tsk,
+				       struct mm_struct *mm)
+{
+	mm->context = NO_CONTEXT;
+	return 0;
+}
+
+/*
+ * Destroy context related info for an mm_struct that is about
+ * to be put to rest.
+ */
+static inline void destroy_context(struct mm_struct *mm)
+{
+	/* Do nothing */
+}
+
+static inline void set_asid(unsigned long asid)
+{
+	/* XXX: We're destroying TLBEHI[8:31] */
+	sysreg_write(TLBEHI, asid & MMU_CONTEXT_ASID_MASK);
+	cpu_sync_pipeline();
+}
+
+static inline unsigned long get_asid(void)
+{
+	unsigned long asid;
+
+	asid = sysreg_read(TLBEHI);
+	return asid & MMU_CONTEXT_ASID_MASK;
+}
+
+static inline void activate_context(struct mm_struct *mm)
+{
+	get_mmu_context(mm);
+	set_asid(mm->context & MMU_CONTEXT_ASID_MASK);
+}
+
+static inline void switch_mm(struct mm_struct *prev,
+				 struct mm_struct *next,
+				 struct task_struct *tsk)
+{
+	if (likely(prev != next)) {
+		unsigned long __pgdir = (unsigned long)next->pgd;
+
+		sysreg_write(PTBR, __pgdir);
+		activate_context(next);
+	}
+}
+
+#define deactivate_mm(tsk,mm) do { } while(0)
+
+#define activate_mm(prev, next) switch_mm((prev), (next), NULL)
+
+static inline void
+enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
+{
+}
+
+
+static inline void enable_mmu(void)
+{
+	sysreg_write(MMUCR, (SYSREG_BIT(MMUCR_S)
+			     | SYSREG_BIT(E)
+			     | SYSREG_BIT(MMUCR_I)));
+	nop(); nop(); nop(); nop(); nop(); nop(); nop(); nop();
+
+	if (mmu_context_cache == NO_CONTEXT)
+		mmu_context_cache = MMU_CONTEXT_FIRST_VERSION;
+
+	set_asid(mmu_context_cache & MMU_CONTEXT_ASID_MASK);
+}
+
+static inline void disable_mmu(void)
+{
+	sysreg_write(MMUCR, SYSREG_BIT(MMUCR_S));
+}
+
+#endif /* __ASM_AVR32_MMU_CONTEXT_H */
diff --git a/include/asm-avr32/module.h b/include/asm-avr32/module.h
new file mode 100644
index 000000000000..451444538a1b
--- /dev/null
+++ b/include/asm-avr32/module.h
@@ -0,0 +1,28 @@
+#ifndef __ASM_AVR32_MODULE_H
+#define __ASM_AVR32_MODULE_H
+
+struct mod_arch_syminfo {
+	unsigned long got_offset;
+	int got_initialized;
+};
+
+struct mod_arch_specific {
+	/* Starting offset of got in the module core memory. */
+	unsigned long got_offset;
+	/* Size of the got. */
+	unsigned long got_size;
+	/* Number of symbols in syminfo. */
+	int nsyms;
+	/* Additional symbol information (got offsets). */
+	struct mod_arch_syminfo *syminfo;
+};
+
+#define Elf_Shdr		Elf32_Shdr
+#define Elf_Sym			Elf32_Sym
+#define Elf_Ehdr		Elf32_Ehdr
+
+#define MODULE_PROC_FAMILY "AVR32v1"
+
+#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY
+
+#endif /* __ASM_AVR32_MODULE_H */
diff --git a/include/asm-avr32/msgbuf.h b/include/asm-avr32/msgbuf.h
new file mode 100644
index 000000000000..ac18bc4da7f7
--- /dev/null
+++ b/include/asm-avr32/msgbuf.h
@@ -0,0 +1,31 @@
+#ifndef __ASM_AVR32_MSGBUF_H
+#define __ASM_AVR32_MSGBUF_H
+
+/*
+ * The msqid64_ds structure for i386 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct msqid64_ds {
+	struct ipc64_perm msg_perm;
+	__kernel_time_t msg_stime;	/* last msgsnd time */
+	unsigned long	__unused1;
+	__kernel_time_t msg_rtime;	/* last msgrcv time */
+	unsigned long	__unused2;
+	__kernel_time_t msg_ctime;	/* last change time */
+	unsigned long	__unused3;
+	unsigned long  msg_cbytes;	/* current number of bytes on queue */
+	unsigned long  msg_qnum;	/* number of messages in queue */
+	unsigned long  msg_qbytes;	/* max number of bytes on queue */
+	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
+	__kernel_pid_t msg_lrpid;	/* last receive pid */
+	unsigned long  __unused4;
+	unsigned long  __unused5;
+};
+
+#endif /* __ASM_AVR32_MSGBUF_H */
diff --git a/include/asm-avr32/mutex.h b/include/asm-avr32/mutex.h
new file mode 100644
index 000000000000..458c1f7fbc18
--- /dev/null
+++ b/include/asm-avr32/mutex.h
@@ -0,0 +1,9 @@
+/*
+ * Pull in the generic implementation for the mutex fastpath.
+ *
+ * TODO: implement optimized primitives instead, or leave the generic
+ * implementation in place, or pick the atomic_xchg() based generic
+ * implementation. (see asm-generic/mutex-xchg.h for details)
+ */
+
+#include <asm-generic/mutex-dec.h>
diff --git a/include/asm-avr32/namei.h b/include/asm-avr32/namei.h
new file mode 100644
index 000000000000..f0a26de06cab
--- /dev/null
+++ b/include/asm-avr32/namei.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_AVR32_NAMEI_H
+#define __ASM_AVR32_NAMEI_H
+
+/* This dummy routine may be changed to something useful */
+#define __emul_prefix() NULL
+
+#endif /* __ASM_AVR32_NAMEI_H */
diff --git a/include/asm-avr32/numnodes.h b/include/asm-avr32/numnodes.h
new file mode 100644
index 000000000000..0b864d7ce330
--- /dev/null
+++ b/include/asm-avr32/numnodes.h
@@ -0,0 +1,7 @@
+#ifndef __ASM_AVR32_NUMNODES_H
+#define __ASM_AVR32_NUMNODES_H
+
+/* Max 4 nodes */
+#define NODES_SHIFT	2
+
+#endif /* __ASM_AVR32_NUMNODES_H */
diff --git a/include/asm-avr32/ocd.h b/include/asm-avr32/ocd.h
new file mode 100644
index 000000000000..46f73180a127
--- /dev/null
+++ b/include/asm-avr32/ocd.h
@@ -0,0 +1,78 @@
+/*
+ * AVR32 OCD Registers
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_OCD_H
+#define __ASM_AVR32_OCD_H
+
+/* Debug Registers */
+#define DBGREG_DID		  0
+#define DBGREG_DC		  8
+#define DBGREG_DS		 16
+#define DBGREG_RWCS		 28
+#define DBGREG_RWA		 36
+#define DBGREG_RWD		 40
+#define DBGREG_WT		 44
+#define DBGREG_DTC		 52
+#define DBGREG_DTSA0		 56
+#define DBGREG_DTSA1		 60
+#define DBGREG_DTEA0		 72
+#define DBGREG_DTEA1		 76
+#define DBGREG_BWC0A		 88
+#define DBGREG_BWC0B		 92
+#define DBGREG_BWC1A		 96
+#define DBGREG_BWC1B		100
+#define DBGREG_BWC2A		104
+#define DBGREG_BWC2B		108
+#define DBGREG_BWC3A		112
+#define DBGREG_BWC3B		116
+#define DBGREG_BWA0A		120
+#define DBGREG_BWA0B		124
+#define DBGREG_BWA1A		128
+#define DBGREG_BWA1B		132
+#define DBGREG_BWA2A		136
+#define DBGREG_BWA2B		140
+#define DBGREG_BWA3A		144
+#define DBGREG_BWA3B		148
+#define DBGREG_BWD3A		153
+#define DBGREG_BWD3B		156
+
+#define DBGREG_PID		284
+
+#define SABAH_OCD		0x01
+#define SABAH_ICACHE		0x02
+#define SABAH_MEM_CACHED	0x04
+#define SABAH_MEM_UNCACHED	0x05
+
+/* Fields in the Development Control register */
+#define DC_SS_BIT		8
+
+#define DC_SS			(1 <<  DC_SS_BIT)
+#define DC_DBE			(1 << 13)
+#define DC_RID			(1 << 27)
+#define DC_ORP			(1 << 28)
+#define DC_MM			(1 << 29)
+#define DC_RES			(1 << 30)
+
+/* Fields in the Development Status register */
+#define DS_SSS			(1 <<  0)
+#define DS_SWB			(1 <<  1)
+#define DS_HWB			(1 <<  2)
+#define DS_BP_SHIFT		8
+#define DS_BP_MASK		(0xff << DS_BP_SHIFT)
+
+#define __mfdr(addr)							\
+({									\
+	register unsigned long value;					\
+	asm volatile("mfdr	%0, %1" : "=r"(value) : "i"(addr));	\
+	value;								\
+})
+#define __mtdr(addr, value)						\
+	asm volatile("mtdr	%0, %1" : : "i"(addr), "r"(value))
+
+#endif /* __ASM_AVR32_OCD_H */
diff --git a/include/asm-avr32/page.h b/include/asm-avr32/page.h
new file mode 100644
index 000000000000..0f630b3e9932
--- /dev/null
+++ b/include/asm-avr32/page.h
@@ -0,0 +1,112 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PAGE_H
+#define __ASM_AVR32_PAGE_H
+
+#ifdef __KERNEL__
+
+/* PAGE_SHIFT determines the page size */
+#define PAGE_SHIFT	12
+#ifdef __ASSEMBLY__
+#define PAGE_SIZE	(1 << PAGE_SHIFT)
+#else
+#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#endif
+#define PAGE_MASK	(~(PAGE_SIZE-1))
+#define PTE_MASK	PAGE_MASK
+
+#ifndef __ASSEMBLY__
+
+#include <asm/addrspace.h>
+
+extern void clear_page(void *to);
+extern void copy_page(void *to, void *from);
+
+#define clear_user_page(page, vaddr, pg)	clear_page(page)
+#define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
+
+/*
+ * These are used to make use of C type-checking..
+ */
+typedef struct { unsigned long pte; } pte_t;
+typedef struct { unsigned long pgd; } pgd_t;
+typedef struct { unsigned long pgprot; } pgprot_t;
+
+#define pte_val(x)		((x).pte)
+#define pgd_val(x)		((x).pgd)
+#define pgprot_val(x)		((x).pgprot)
+
+#define __pte(x)		((pte_t) { (x) })
+#define __pgd(x)		((pgd_t) { (x) })
+#define __pgprot(x)		((pgprot_t) { (x) })
+
+/* FIXME: These should be removed soon */
+extern unsigned long memory_start, memory_end;
+
+/* Pure 2^n version of get_order */
+static inline int get_order(unsigned long size)
+{
+	unsigned lz;
+
+	size = (size - 1) >> PAGE_SHIFT;
+	asm("clz %0, %1" : "=r"(lz) : "r"(size));
+	return 32 - lz;
+}
+
+#endif /* !__ASSEMBLY__ */
+
+/* Align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
+
+/*
+ * The hardware maps the virtual addresses 0x80000000 -> 0x9fffffff
+ * permanently to the physical addresses 0x00000000 -> 0x1fffffff when
+ * segmentation is enabled. We want to make use of this in order to
+ * minimize TLB pressure.
+ */
+#define PAGE_OFFSET		(0x80000000UL)
+
+/*
+ * ALSA uses virt_to_page() on DMA pages, which I'm not entirely sure
+ * is a good idea. Anyway, we can't simply subtract PAGE_OFFSET here
+ * in that case, so we'll have to mask out the three most significant
+ * bits of the address instead...
+ *
+ * What's the difference between __pa() and virt_to_phys() anyway?
+ */
+#define __pa(x)		PHYSADDR(x)
+#define __va(x)		((void *)(P1SEGADDR(x)))
+
+#define MAP_NR(addr)	(((unsigned long)(addr) - PAGE_OFFSET) >> PAGE_SHIFT)
+
+#define phys_to_page(phys)	(pfn_to_page(phys >> PAGE_SHIFT))
+#define page_to_phys(page)	(page_to_pfn(page) << PAGE_SHIFT)
+
+#ifndef CONFIG_NEED_MULTIPLE_NODES
+
+#define PHYS_PFN_OFFSET		(CONFIG_PHYS_OFFSET >> PAGE_SHIFT)
+
+#define pfn_to_page(pfn)	(mem_map + ((pfn) - PHYS_PFN_OFFSET))
+#define page_to_pfn(page)	((unsigned long)((page) - mem_map) + PHYS_PFN_OFFSET)
+#define pfn_valid(pfn)		((pfn) >= PHYS_PFN_OFFSET && (pfn) < (PHYS_PFN_OFFSET + max_mapnr))
+#endif /* CONFIG_NEED_MULTIPLE_NODES */
+
+#define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+#define virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+
+#define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE |	\
+				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
+/*
+ * Memory above this physical address will be considered highmem.
+ */
+#define HIGHMEM_START		0x20000000UL
+
+#endif /* __KERNEL__ */
+
+#endif /* __ASM_AVR32_PAGE_H */
diff --git a/include/asm-avr32/param.h b/include/asm-avr32/param.h
new file mode 100644
index 000000000000..34bc8d4c3b29
--- /dev/null
+++ b/include/asm-avr32/param.h
@@ -0,0 +1,23 @@
+#ifndef __ASM_AVR32_PARAM_H
+#define __ASM_AVR32_PARAM_H
+
+#ifdef __KERNEL__
+# define HZ		CONFIG_HZ
+# define USER_HZ	100		/* User interfaces are in "ticks" */
+# define CLOCKS_PER_SEC	(USER_HZ)	/* frequency at which times() counts */
+#endif
+
+#ifndef HZ
+# define HZ		100
+#endif
+
+/* TODO: Should be configurable */
+#define EXEC_PAGESIZE	4096
+
+#ifndef NOGROUP
+# define NOGROUP	(-1)
+#endif
+
+#define MAXHOSTNAMELEN	64
+
+#endif /* __ASM_AVR32_PARAM_H */
diff --git a/include/asm-avr32/pci.h b/include/asm-avr32/pci.h
new file mode 100644
index 000000000000..0f5f134b896a
--- /dev/null
+++ b/include/asm-avr32/pci.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_AVR32_PCI_H__
+#define __ASM_AVR32_PCI_H__
+
+/* We don't support PCI yet, but some drivers require this file anyway */
+
+#define PCI_DMA_BUS_IS_PHYS	(1)
+
+#endif /* __ASM_AVR32_PCI_H__ */
diff --git a/include/asm-avr32/percpu.h b/include/asm-avr32/percpu.h
new file mode 100644
index 000000000000..69227b4cd0d4
--- /dev/null
+++ b/include/asm-avr32/percpu.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_PERCPU_H
+#define __ASM_AVR32_PERCPU_H
+
+#include <asm-generic/percpu.h>
+
+#endif /* __ASM_AVR32_PERCPU_H */
diff --git a/include/asm-avr32/pgalloc.h b/include/asm-avr32/pgalloc.h
new file mode 100644
index 000000000000..7492cfb92ced
--- /dev/null
+++ b/include/asm-avr32/pgalloc.h
@@ -0,0 +1,96 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGALLOC_H
+#define __ASM_AVR32_PGALLOC_H
+
+#include <asm/processor.h>
+#include <linux/threads.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+
+#define pmd_populate_kernel(mm, pmd, pte) \
+	set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
+
+static __inline__ void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
+				    struct page *pte)
+{
+	set_pmd(pmd, __pmd(_PAGE_TABLE + page_to_phys(pte)));
+}
+
+/*
+ * Allocate and free page tables
+ */
+static __inline__ pgd_t *pgd_alloc(struct mm_struct *mm)
+{
+	unsigned int pgd_size = (USER_PTRS_PER_PGD * sizeof(pgd_t));
+	pgd_t *pgd = (pgd_t *)kmalloc(pgd_size, GFP_KERNEL);
+
+	if (pgd)
+		memset(pgd, 0, pgd_size);
+
+	return pgd;
+}
+
+static inline void pgd_free(pgd_t *pgd)
+{
+	kfree(pgd);
+}
+
+static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
+					  unsigned long address)
+{
+	int count = 0;
+	pte_t *pte;
+
+	do {
+		pte = (pte_t *) __get_free_page(GFP_KERNEL | __GFP_REPEAT);
+		if (pte)
+			clear_page(pte);
+		else {
+			current->state = TASK_UNINTERRUPTIBLE;
+			schedule_timeout(HZ);
+		}
+	} while (!pte && (count++ < 10));
+
+	return pte;
+}
+
+static inline struct page *pte_alloc_one(struct mm_struct *mm,
+					 unsigned long address)
+{
+	int count = 0;
+	struct page *pte;
+
+	do {
+		pte = alloc_pages(GFP_KERNEL, 0);
+		if (pte)
+			clear_page(page_address(pte));
+		else {
+			current->state = TASK_UNINTERRUPTIBLE;
+			schedule_timeout(HZ);
+		}
+	} while (!pte && (count++ < 10));
+
+	return pte;
+}
+
+static inline void pte_free_kernel(pte_t *pte)
+{
+	free_page((unsigned long)pte);
+}
+
+static inline void pte_free(struct page *pte)
+{
+	__free_page(pte);
+}
+
+#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
+
+#define check_pgt_cache() do { } while(0)
+
+#endif /* __ASM_AVR32_PGALLOC_H */
diff --git a/include/asm-avr32/pgtable-2level.h b/include/asm-avr32/pgtable-2level.h
new file mode 100644
index 000000000000..425dd567b5b9
--- /dev/null
+++ b/include/asm-avr32/pgtable-2level.h
@@ -0,0 +1,47 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGTABLE_2LEVEL_H
+#define __ASM_AVR32_PGTABLE_2LEVEL_H
+
+#include <asm-generic/pgtable-nopmd.h>
+
+/*
+ * Traditional 2-level paging structure
+ */
+#define PGDIR_SHIFT	22
+#define PTRS_PER_PGD	1024
+
+#define PTRS_PER_PTE	1024
+
+#ifndef __ASSEMBLY__
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep, pteval)
+
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define pte_pfn(x)		((unsigned long)(((x).pte >> PAGE_SHIFT)))
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PGTABLE_2LEVEL_H */
diff --git a/include/asm-avr32/pgtable.h b/include/asm-avr32/pgtable.h
new file mode 100644
index 000000000000..6b8ca9db2bd5
--- /dev/null
+++ b/include/asm-avr32/pgtable.h
@@ -0,0 +1,408 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PGTABLE_H
+#define __ASM_AVR32_PGTABLE_H
+
+#include <asm/addrspace.h>
+
+#ifndef __ASSEMBLY__
+#include <linux/sched.h>
+
+#endif /* !__ASSEMBLY__ */
+
+/*
+ * Use two-level page tables just as the i386 (without PAE)
+ */
+#include <asm/pgtable-2level.h>
+
+/*
+ * The following code might need some cleanup when the values are
+ * final...
+ */
+#define PMD_SIZE	(1UL << PMD_SHIFT)
+#define PMD_MASK	(~(PMD_SIZE-1))
+#define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
+#define PGDIR_MASK	(~(PGDIR_SIZE-1))
+
+#define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
+#define FIRST_USER_ADDRESS	0
+
+#define PTE_PHYS_MASK	0x1ffff000
+
+#ifndef __ASSEMBLY__
+extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
+extern void paging_init(void);
+
+/*
+ * ZERO_PAGE is a global shared page that is always zero: used for
+ * zero-mapped memory areas etc.
+ */
+extern struct page *empty_zero_page;
+#define ZERO_PAGE(vaddr) (empty_zero_page)
+
+/*
+ * Just any arbitrary offset to the start of the vmalloc VM area: the
+ * current 8 MiB value just means that there will be a 8 MiB "hole"
+ * after the uncached physical memory (P2 segment) until the vmalloc
+ * area starts. That means that any out-of-bounds memory accesses will
+ * hopefully be caught; we don't know if the end of the P1/P2 segments
+ * are actually used for anything, but it is anyway safer to let the
+ * MMU catch these kinds of errors than to rely on the memory bus.
+ *
+ * A "hole" of the same size is added to the end of the P3 segment as
+ * well. It might seem wasteful to use 16 MiB of virtual address space
+ * on this, but we do have 512 MiB of it...
+ *
+ * The vmalloc() routines leave a hole of 4 KiB between each vmalloced
+ * area for the same reason.
+ */
+#define VMALLOC_OFFSET	(8 * 1024 * 1024)
+#define VMALLOC_START	(P3SEG + VMALLOC_OFFSET)
+#define VMALLOC_END	(P4SEG - VMALLOC_OFFSET)
+#endif /* !__ASSEMBLY__ */
+
+/*
+ * Page flags. Some of these flags are not directly supported by
+ * hardware, so we have to emulate them.
+ */
+#define _TLBEHI_BIT_VALID	9
+#define _TLBEHI_VALID		(1 << _TLBEHI_BIT_VALID)
+
+#define _PAGE_BIT_WT		0  /* W-bit   : write-through */
+#define _PAGE_BIT_DIRTY		1  /* D-bit   : page changed */
+#define _PAGE_BIT_SZ0		2  /* SZ0-bit : Size of page */
+#define _PAGE_BIT_SZ1		3  /* SZ1-bit : Size of page */
+#define _PAGE_BIT_EXECUTE	4  /* X-bit   : execute access allowed */
+#define _PAGE_BIT_RW		5  /* AP0-bit : write access allowed */
+#define _PAGE_BIT_USER		6  /* AP1-bit : user space access allowed */
+#define _PAGE_BIT_BUFFER	7  /* B-bit   : bufferable */
+#define _PAGE_BIT_GLOBAL	8  /* G-bit   : global (ignore ASID) */
+#define _PAGE_BIT_CACHABLE	9  /* C-bit   : cachable */
+
+/* If we drop support for 1K pages, we get two extra bits */
+#define _PAGE_BIT_PRESENT	10
+#define _PAGE_BIT_ACCESSED	11 /* software: page was accessed */
+
+/* The following flags are only valid when !PRESENT */
+#define _PAGE_BIT_FILE		0 /* software: pagecache or swap? */
+
+#define _PAGE_WT		(1 << _PAGE_BIT_WT)
+#define _PAGE_DIRTY		(1 << _PAGE_BIT_DIRTY)
+#define _PAGE_EXECUTE		(1 << _PAGE_BIT_EXECUTE)
+#define _PAGE_RW		(1 << _PAGE_BIT_RW)
+#define _PAGE_USER		(1 << _PAGE_BIT_USER)
+#define _PAGE_BUFFER		(1 << _PAGE_BIT_BUFFER)
+#define _PAGE_GLOBAL		(1 << _PAGE_BIT_GLOBAL)
+#define _PAGE_CACHABLE		(1 << _PAGE_BIT_CACHABLE)
+
+/* Software flags */
+#define _PAGE_ACCESSED		(1 << _PAGE_BIT_ACCESSED)
+#define _PAGE_PRESENT		(1 << _PAGE_BIT_PRESENT)
+#define _PAGE_FILE		(1 << _PAGE_BIT_FILE)
+
+/*
+ * Page types, i.e. sizes. _PAGE_TYPE_NONE corresponds to what is
+ * usually called _PAGE_PROTNONE on other architectures.
+ *
+ * XXX: Find out if _PAGE_PROTNONE is equivalent with !_PAGE_USER. If
+ * so, we can encode all possible page sizes (although we can't really
+ * support 1K pages anyway due to the _PAGE_PRESENT and _PAGE_ACCESSED
+ * bits)
+ *
+ */
+#define _PAGE_TYPE_MASK		((1 << _PAGE_BIT_SZ0) | (1 << _PAGE_BIT_SZ1))
+#define _PAGE_TYPE_NONE		(0 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_SMALL	(1 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_MEDIUM	(2 << _PAGE_BIT_SZ0)
+#define _PAGE_TYPE_LARGE	(3 << _PAGE_BIT_SZ0)
+
+/*
+ * Mask which drop software flags. We currently can't handle more than
+ * 512 MiB of physical memory, so we can use bits 29-31 for other
+ * stuff.  With a fixed 4K page size, we can use bits 10-11 as well as
+ * bits 2-3 (SZ)
+ */
+#define _PAGE_FLAGS_HARDWARE_MASK	0xfffff3ff
+
+#define _PAGE_FLAGS_CACHE_MASK	(_PAGE_CACHABLE | _PAGE_BUFFER | _PAGE_WT)
+
+/* TODO: Check for saneness */
+/* User-mode page table flags (to be set in a pgd or pmd entry) */
+#define _PAGE_TABLE		(_PAGE_PRESENT | _PAGE_TYPE_SMALL | _PAGE_RW \
+				 | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
+/* Kernel-mode page table flags */
+#define _KERNPG_TABLE		(_PAGE_PRESENT | _PAGE_TYPE_SMALL | _PAGE_RW \
+				 | _PAGE_ACCESSED | _PAGE_DIRTY)
+/* Flags that may be modified by software */
+#define _PAGE_CHG_MASK		(PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY \
+				 | _PAGE_FLAGS_CACHE_MASK)
+
+#define _PAGE_FLAGS_READ	(_PAGE_CACHABLE	| _PAGE_BUFFER)
+#define _PAGE_FLAGS_WRITE	(_PAGE_FLAGS_READ | _PAGE_RW | _PAGE_DIRTY)
+
+#define _PAGE_NORMAL(x)	__pgprot((x) | _PAGE_PRESENT | _PAGE_TYPE_SMALL	\
+				 | _PAGE_ACCESSED)
+
+#define PAGE_NONE	(_PAGE_ACCESSED | _PAGE_TYPE_NONE)
+#define PAGE_READ	(_PAGE_FLAGS_READ | _PAGE_USER)
+#define PAGE_EXEC	(_PAGE_FLAGS_READ | _PAGE_EXECUTE | _PAGE_USER)
+#define PAGE_WRITE	(_PAGE_FLAGS_WRITE | _PAGE_USER)
+#define PAGE_KERNEL	_PAGE_NORMAL(_PAGE_FLAGS_WRITE | _PAGE_EXECUTE | _PAGE_GLOBAL)
+#define PAGE_KERNEL_RO	_PAGE_NORMAL(_PAGE_FLAGS_READ | _PAGE_EXECUTE | _PAGE_GLOBAL)
+
+#define _PAGE_P(x)	_PAGE_NORMAL((x) & ~(_PAGE_RW | _PAGE_DIRTY))
+#define _PAGE_S(x)	_PAGE_NORMAL(x)
+
+#define PAGE_COPY	_PAGE_P(PAGE_WRITE | PAGE_READ)
+
+#ifndef __ASSEMBLY__
+/*
+ * The hardware supports flags for write- and execute access. Read is
+ * always allowed if the page is loaded into the TLB, so the "-w-",
+ * "--x" and "-wx" mappings are implemented as "rw-", "r-x" and "rwx",
+ * respectively.
+ *
+ * The "---" case is handled by software; the page will simply not be
+ * loaded into the TLB if the page type is _PAGE_TYPE_NONE.
+ */
+
+#define __P000	__pgprot(PAGE_NONE)
+#define __P001	_PAGE_P(PAGE_READ)
+#define __P010	_PAGE_P(PAGE_WRITE)
+#define __P011	_PAGE_P(PAGE_WRITE | PAGE_READ)
+#define __P100	_PAGE_P(PAGE_EXEC)
+#define __P101	_PAGE_P(PAGE_EXEC | PAGE_READ)
+#define __P110	_PAGE_P(PAGE_EXEC | PAGE_WRITE)
+#define __P111	_PAGE_P(PAGE_EXEC | PAGE_WRITE | PAGE_READ)
+
+#define __S000	__pgprot(PAGE_NONE)
+#define __S001	_PAGE_S(PAGE_READ)
+#define __S010	_PAGE_S(PAGE_WRITE)
+#define __S011	_PAGE_S(PAGE_WRITE | PAGE_READ)
+#define __S100	_PAGE_S(PAGE_EXEC)
+#define __S101	_PAGE_S(PAGE_EXEC | PAGE_READ)
+#define __S110	_PAGE_S(PAGE_EXEC | PAGE_WRITE)
+#define __S111	_PAGE_S(PAGE_EXEC | PAGE_WRITE | PAGE_READ)
+
+#define pte_none(x)	(!pte_val(x))
+#define pte_present(x)	(pte_val(x) & _PAGE_PRESENT)
+
+#define pte_clear(mm,addr,xp)					\
+	do {							\
+		set_pte_at(mm, addr, xp, __pte(0));		\
+	} while (0)
+
+/*
+ * The following only work if pte_present() is true.
+ * Undefined behaviour if not..
+ */
+static inline int pte_read(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_USER;
+}
+static inline int pte_write(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_RW;
+}
+static inline int pte_exec(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_EXECUTE;
+}
+static inline int pte_dirty(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_DIRTY;
+}
+static inline int pte_young(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_ACCESSED;
+}
+
+/*
+ * The following only work if pte_present() is not true.
+ */
+static inline int pte_file(pte_t pte)
+{
+	return pte_val(pte) & _PAGE_FILE;
+}
+
+/* Mutator functions for PTE bits */
+static inline pte_t pte_rdprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_USER));
+	return pte;
+}
+static inline pte_t pte_wrprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_RW));
+	return pte;
+}
+static inline pte_t pte_exprotect(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_EXECUTE));
+	return pte;
+}
+static inline pte_t pte_mkclean(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_DIRTY));
+	return pte;
+}
+static inline pte_t pte_mkold(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) & ~_PAGE_ACCESSED));
+	return pte;
+}
+static inline pte_t pte_mkread(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_USER));
+	return pte;
+}
+static inline pte_t pte_mkwrite(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_RW));
+	return pte;
+}
+static inline pte_t pte_mkexec(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_EXECUTE));
+	return pte;
+}
+static inline pte_t pte_mkdirty(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_DIRTY));
+	return pte;
+}
+static inline pte_t pte_mkyoung(pte_t pte)
+{
+	set_pte(&pte, __pte(pte_val(pte) | _PAGE_ACCESSED));
+	return pte;
+}
+
+#define pmd_none(x)	(!pmd_val(x))
+#define pmd_present(x)	(pmd_val(x) & _PAGE_PRESENT)
+#define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
+#define	pmd_bad(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER))	\
+			 != _KERNPG_TABLE)
+
+/*
+ * Permanent address of a page. We don't support highmem, so this is
+ * trivial.
+ */
+#define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
+#define pte_page(x) 	phys_to_page(pte_val(x) & PTE_PHYS_MASK)
+
+/*
+ * Mark the prot value as uncacheable and unbufferable
+ */
+#define pgprot_noncached(prot)						\
+	__pgprot(pgprot_val(prot) & ~(_PAGE_BUFFER | _PAGE_CACHABLE))
+
+/*
+ * Mark the prot value as uncacheable but bufferable
+ */
+#define pgprot_writecombine(prot)					\
+	__pgprot((pgprot_val(prot) & ~_PAGE_CACHABLE) | _PAGE_BUFFER)
+
+/*
+ * Conversion functions: convert a page and protection to a page entry,
+ * and a page entry and page directory to the page they refer to.
+ *
+ * extern pte_t mk_pte(struct page *page, pgprot_t pgprot)
+ */
+#define mk_pte(page, pgprot)	pfn_pte(page_to_pfn(page), (pgprot))
+
+static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
+{
+	set_pte(&pte, __pte((pte_val(pte) & _PAGE_CHG_MASK)
+			    | pgprot_val(newprot)));
+	return pte;
+}
+
+#define page_pte(page)	page_pte_prot(page, __pgprot(0))
+
+#define pmd_page_vaddr(pmd)					\
+	((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
+
+#define pmd_page(pmd)	(phys_to_page(pmd_val(pmd)))
+
+/* to find an entry in a page-table-directory. */
+#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
+#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+#define pgd_offset_current(address)				\
+	((pgd_t *)__mfsr(SYSREG_PTBR) + pgd_index(address))
+
+/* to find an entry in a kernel page-table-directory */
+#define pgd_offset_k(address) pgd_offset(&init_mm, address)
+
+/* Find an entry in the third-level page table.. */
+#define pte_index(address)				\
+	((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
+#define pte_offset(dir, address)					\
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
+#define pte_offset_kernel(dir, address)					\
+	((pte_t *) pmd_page_vaddr(*(dir)) + pte_index(address))
+#define pte_offset_map(dir, address) pte_offset_kernel(dir, address)
+#define pte_offset_map_nested(dir, address) pte_offset_kernel(dir, address)
+#define pte_unmap(pte)		do { } while (0)
+#define pte_unmap_nested(pte)	do { } while (0)
+
+struct vm_area_struct;
+extern void update_mmu_cache(struct vm_area_struct * vma,
+			     unsigned long address, pte_t pte);
+
+/*
+ * Encode and decode a swap entry
+ *
+ * Constraints:
+ *   _PAGE_FILE at bit 0
+ *   _PAGE_TYPE_* at bits 2-3 (for emulating _PAGE_PROTNONE)
+ *   _PAGE_PRESENT at bit 10
+ *
+ * We encode the type into bits 4-9 and offset into bits 11-31. This
+ * gives us a 21 bits offset, or 2**21 * 4K = 8G usable swap space per
+ * device, and 64 possible types.
+ *
+ * NOTE: We should set ZEROs at the position of _PAGE_PRESENT
+ *       and _PAGE_PROTNONE bits
+ */
+#define __swp_type(x)		(((x).val >> 4) & 0x3f)
+#define __swp_offset(x)		((x).val >> 11)
+#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 4) | ((offset) << 11) })
+#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
+#define __swp_entry_to_pte(x)	((pte_t) { (x).val })
+
+/*
+ * Encode and decode a nonlinear file mapping entry. We have to
+ * preserve _PAGE_FILE and _PAGE_PRESENT here. _PAGE_TYPE_* isn't
+ * necessary, since _PAGE_FILE implies !_PAGE_PROTNONE (?)
+ */
+#define PTE_FILE_MAX_BITS	30
+#define pte_to_pgoff(pte)	(((pte_val(pte) >> 1) & 0x1ff)		\
+				 | ((pte_val(pte) >> 11) << 9))
+#define pgoff_to_pte(off)	((pte_t) { ((((off) & 0x1ff) << 1)	\
+					    | (((off) >> 9) << 11)	\
+					    | _PAGE_FILE) })
+
+typedef pte_t *pte_addr_t;
+
+#define kern_addr_valid(addr)	(1)
+
+#define io_remap_pfn_range(vma, vaddr, pfn, size, prot)	\
+	remap_pfn_range(vma, vaddr, pfn, size, prot)
+
+#define MK_IOSPACE_PFN(space, pfn)	(pfn)
+#define GET_IOSPACE(pfn)		0
+#define GET_PFN(pfn)			(pfn)
+
+/* No page table caches to initialize (?) */
+#define pgtable_cache_init()	do { } while(0)
+
+#include <asm-generic/pgtable.h>
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PGTABLE_H */
diff --git a/include/asm-avr32/poll.h b/include/asm-avr32/poll.h
new file mode 100644
index 000000000000..736e29755dfc
--- /dev/null
+++ b/include/asm-avr32/poll.h
@@ -0,0 +1,27 @@
+#ifndef __ASM_AVR32_POLL_H
+#define __ASM_AVR32_POLL_H
+
+/* These are specified by iBCS2 */
+#define POLLIN		0x0001
+#define POLLPRI		0x0002
+#define POLLOUT		0x0004
+#define POLLERR		0x0008
+#define POLLHUP		0x0010
+#define POLLNVAL	0x0020
+
+/* The rest seem to be more-or-less nonstandard. Check them! */
+#define POLLRDNORM	0x0040
+#define POLLRDBAND	0x0080
+#define POLLWRNORM	0x0100
+#define POLLWRBAND	0x0200
+#define POLLMSG		0x0400
+#define POLLREMOVE	0x1000
+#define POLLRDHUP	0x2000
+
+struct pollfd {
+	int fd;
+	short events;
+	short revents;
+};
+
+#endif /* __ASM_AVR32_POLL_H */
diff --git a/include/asm-avr32/posix_types.h b/include/asm-avr32/posix_types.h
new file mode 100644
index 000000000000..2831b039b349
--- /dev/null
+++ b/include/asm-avr32/posix_types.h
@@ -0,0 +1,129 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_POSIX_TYPES_H
+#define __ASM_AVR32_POSIX_TYPES_H
+
+/*
+ * This file is generally used by user-level software, so you need to
+ * be a little careful about namespace pollution etc.  Also, we cannot
+ * assume GCC is being used.
+ */
+
+typedef unsigned long   __kernel_ino_t;
+typedef unsigned short  __kernel_mode_t;
+typedef unsigned short  __kernel_nlink_t;
+typedef long            __kernel_off_t;
+typedef int             __kernel_pid_t;
+typedef unsigned short  __kernel_ipc_pid_t;
+typedef unsigned int	__kernel_uid_t;
+typedef unsigned int	__kernel_gid_t;
+typedef unsigned long	__kernel_size_t;
+typedef int             __kernel_ssize_t;
+typedef int             __kernel_ptrdiff_t;
+typedef long            __kernel_time_t;
+typedef long            __kernel_suseconds_t;
+typedef long            __kernel_clock_t;
+typedef int             __kernel_timer_t;
+typedef int             __kernel_clockid_t;
+typedef int             __kernel_daddr_t;
+typedef char *          __kernel_caddr_t;
+typedef unsigned short  __kernel_uid16_t;
+typedef unsigned short  __kernel_gid16_t;
+typedef unsigned int    __kernel_uid32_t;
+typedef unsigned int    __kernel_gid32_t;
+
+typedef unsigned short  __kernel_old_uid_t;
+typedef unsigned short  __kernel_old_gid_t;
+typedef unsigned short  __kernel_old_dev_t;
+
+#ifdef __GNUC__
+typedef long long       __kernel_loff_t;
+#endif
+
+typedef struct {
+#if defined(__KERNEL__) || defined(__USE_ALL)
+    int     val[2];
+#else /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+    int     __val[2];
+#endif /* !defined(__KERNEL__) && !defined(__USE_ALL) */
+} __kernel_fsid_t;
+
+#if defined(__KERNEL__)
+
+#undef  __FD_SET
+static __inline__ void __FD_SET(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    __fdsetp->fds_bits[__tmp] |= (1UL<<__rem);
+}
+
+#undef  __FD_CLR
+static __inline__ void __FD_CLR(unsigned long __fd, __kernel_fd_set *__fdsetp)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    __fdsetp->fds_bits[__tmp] &= ~(1UL<<__rem);
+}
+
+
+#undef  __FD_ISSET
+static __inline__ int __FD_ISSET(unsigned long __fd, const __kernel_fd_set *__p)
+{
+    unsigned long __tmp = __fd / __NFDBITS;
+    unsigned long __rem = __fd % __NFDBITS;
+    return (__p->fds_bits[__tmp] & (1UL<<__rem)) != 0;
+}
+
+/*
+ * This will unroll the loop for the normal constant case (8 ints,
+ * for a 256-bit fd_set)
+ */
+#undef  __FD_ZERO
+static __inline__ void __FD_ZERO(__kernel_fd_set *__p)
+{
+    unsigned long *__tmp = __p->fds_bits;
+    int __i;
+
+    if (__builtin_constant_p(__FDSET_LONGS)) {
+        switch (__FDSET_LONGS) {
+            case 16:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                __tmp[ 4] = 0; __tmp[ 5] = 0;
+                __tmp[ 6] = 0; __tmp[ 7] = 0;
+                __tmp[ 8] = 0; __tmp[ 9] = 0;
+                __tmp[10] = 0; __tmp[11] = 0;
+                __tmp[12] = 0; __tmp[13] = 0;
+                __tmp[14] = 0; __tmp[15] = 0;
+                return;
+
+            case 8:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                __tmp[ 4] = 0; __tmp[ 5] = 0;
+                __tmp[ 6] = 0; __tmp[ 7] = 0;
+                return;
+
+            case 4:
+                __tmp[ 0] = 0; __tmp[ 1] = 0;
+                __tmp[ 2] = 0; __tmp[ 3] = 0;
+                return;
+        }
+    }
+    __i = __FDSET_LONGS;
+    while (__i) {
+        __i--;
+        *__tmp = 0;
+        __tmp++;
+    }
+}
+
+#endif /* defined(__KERNEL__) */
+
+#endif /* __ASM_AVR32_POSIX_TYPES_H */
diff --git a/include/asm-avr32/processor.h b/include/asm-avr32/processor.h
new file mode 100644
index 000000000000..f6913778a45f
--- /dev/null
+++ b/include/asm-avr32/processor.h
@@ -0,0 +1,147 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PROCESSOR_H
+#define __ASM_AVR32_PROCESSOR_H
+
+#include <asm/page.h>
+#include <asm/cache.h>
+
+#define TASK_SIZE	0x80000000
+
+#ifndef __ASSEMBLY__
+
+static inline void *current_text_addr(void)
+{
+	register void *pc asm("pc");
+	return pc;
+}
+
+enum arch_type {
+	ARCH_AVR32A,
+	ARCH_AVR32B,
+	ARCH_MAX
+};
+
+enum cpu_type {
+	CPU_MORGAN,
+	CPU_AT32AP,
+	CPU_MAX
+};
+
+enum tlb_config {
+	TLB_NONE,
+	TLB_SPLIT,
+	TLB_UNIFIED,
+	TLB_INVALID
+};
+
+struct avr32_cpuinfo {
+	struct clk *clk;
+	unsigned long loops_per_jiffy;
+	enum arch_type arch_type;
+	enum cpu_type cpu_type;
+	unsigned short arch_revision;
+	unsigned short cpu_revision;
+	enum tlb_config tlb_config;
+
+	struct cache_info icache;
+	struct cache_info dcache;
+};
+
+extern struct avr32_cpuinfo boot_cpu_data;
+
+#ifdef CONFIG_SMP
+extern struct avr32_cpuinfo cpu_data[];
+#define current_cpu_data cpu_data[smp_processor_id()]
+#else
+#define cpu_data (&boot_cpu_data)
+#define current_cpu_data boot_cpu_data
+#endif
+
+/* This decides where the kernel will search for a free chunk of vm
+ * space during mmap's
+ */
+#define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
+
+#define cpu_relax()		barrier()
+#define cpu_sync_pipeline()	asm volatile("sub pc, -2" : : : "memory")
+
+struct cpu_context {
+	unsigned long sr;
+	unsigned long pc;
+	unsigned long ksp;	/* Kernel stack pointer */
+	unsigned long r7;
+	unsigned long r6;
+	unsigned long r5;
+	unsigned long r4;
+	unsigned long r3;
+	unsigned long r2;
+	unsigned long r1;
+	unsigned long r0;
+};
+
+/* This struct contains the CPU context as stored by switch_to() */
+struct thread_struct {
+	struct cpu_context cpu_context;
+	unsigned long single_step_addr;
+	u16 single_step_insn;
+};
+
+#define INIT_THREAD {						\
+	.cpu_context = {					\
+		.ksp = sizeof(init_stack) + (long)&init_stack,	\
+	},							\
+}
+
+/*
+ * Do necessary setup to start up a newly executed thread.
+ */
+#define start_thread(regs, new_pc, new_sp)	 \
+	do {					 \
+		set_fs(USER_DS);		 \
+		memset(regs, 0, sizeof(*regs));	 \
+		regs->sr = MODE_USER;		 \
+		regs->pc = new_pc & ~1;		 \
+		regs->sp = new_sp;		 \
+	} while(0)
+
+struct task_struct;
+
+/* Free all resources held by a thread */
+extern void release_thread(struct task_struct *);
+
+/* Create a kernel thread without removing it from tasklists */
+extern int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags);
+
+/* Prepare to copy thread state - unlazy all lazy status */
+#define prepare_to_copy(tsk) do { } while(0)
+
+/* Return saved PC of a blocked thread */
+#define thread_saved_pc(tsk)    ((tsk)->thread.cpu_context.pc)
+
+struct pt_regs;
+void show_trace(struct task_struct *task, unsigned long *stack,
+		struct pt_regs *regs);
+
+extern unsigned long get_wchan(struct task_struct *p);
+
+#define KSTK_EIP(tsk)	((tsk)->thread.cpu_context.pc)
+#define KSTK_ESP(tsk)	((tsk)->thread.cpu_context.ksp)
+
+#define ARCH_HAS_PREFETCH
+
+static inline void prefetch(const void *x)
+{
+	const char *c = x;
+	asm volatile("pref %0" : : "r"(c));
+}
+#define PREFETCH_STRIDE	L1_CACHE_BYTES
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PROCESSOR_H */
diff --git a/include/asm-avr32/ptrace.h b/include/asm-avr32/ptrace.h
new file mode 100644
index 000000000000..60f0f19a81f1
--- /dev/null
+++ b/include/asm-avr32/ptrace.h
@@ -0,0 +1,154 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_PTRACE_H
+#define __ASM_AVR32_PTRACE_H
+
+#define PTRACE_GETREGS		12
+#define PTRACE_SETREGS		13
+
+/*
+ * Status Register bits
+ */
+#define SR_H		0x40000000
+#define SR_R		0x20000000
+#define SR_J		0x10000000
+#define SR_DM		0x08000000
+#define SR_D		0x04000000
+#define MODE_NMI	0x01c00000
+#define MODE_EXCEPTION	0x01800000
+#define MODE_INT3	0x01400000
+#define MODE_INT2	0x01000000
+#define MODE_INT1	0x00c00000
+#define MODE_INT0	0x00800000
+#define MODE_SUPERVISOR	0x00400000
+#define MODE_USER	0x00000000
+#define MODE_MASK	0x01c00000
+#define SR_EM		0x00200000
+#define SR_I3M		0x00100000
+#define SR_I2M		0x00080000
+#define SR_I1M		0x00040000
+#define SR_I0M		0x00020000
+#define SR_GM		0x00010000
+
+#define SR_H_BIT	30
+#define SR_R_BIT	29
+#define SR_J_BIT	28
+#define SR_DM_BIT	27
+#define SR_D_BIT	26
+#define MODE_SHIFT	22
+#define SR_EM_BIT	21
+#define SR_I3M_BIT	20
+#define SR_I2M_BIT	19
+#define SR_I1M_BIT	18
+#define SR_I0M_BIT	17
+#define SR_GM_BIT	16
+
+/* The user-visible part */
+#define SR_L		0x00000020
+#define SR_Q		0x00000010
+#define SR_V		0x00000008
+#define SR_N		0x00000004
+#define SR_Z		0x00000002
+#define SR_C		0x00000001
+
+#define SR_L_BIT	5
+#define SR_Q_BIT	4
+#define SR_V_BIT	3
+#define SR_N_BIT	2
+#define SR_Z_BIT	1
+#define SR_C_BIT	0
+
+/*
+ * The order is defined by the stmts instruction. r0 is stored first,
+ * so it gets the highest address.
+ *
+ * Registers 0-12 are general-purpose registers (r12 is normally used for
+ * the function return value).
+ * Register 13 is the stack pointer
+ * Register 14 is the link register
+ * Register 15 is the program counter (retrieved from the RAR sysreg)
+ */
+#define FRAME_SIZE_FULL 72
+#define REG_R12_ORIG	68
+#define REG_R0		64
+#define REG_R1		60
+#define REG_R2		56
+#define REG_R3		52
+#define REG_R4		48
+#define REG_R5		44
+#define REG_R6		40
+#define REG_R7		36
+#define REG_R8		32
+#define REG_R9		28
+#define REG_R10		24
+#define REG_R11		20
+#define REG_R12		16
+#define REG_SP		12
+#define REG_LR		 8
+
+#define FRAME_SIZE_MIN	 8
+#define REG_PC		 4
+#define REG_SR		 0
+
+#ifndef __ASSEMBLY__
+struct pt_regs {
+	/* These are always saved */
+	unsigned long sr;
+	unsigned long pc;
+
+	/* These are sometimes saved */
+	unsigned long lr;
+	unsigned long sp;
+	unsigned long r12;
+	unsigned long r11;
+	unsigned long r10;
+	unsigned long r9;
+	unsigned long r8;
+	unsigned long r7;
+	unsigned long r6;
+	unsigned long r5;
+	unsigned long r4;
+	unsigned long r3;
+	unsigned long r2;
+	unsigned long r1;
+	unsigned long r0;
+
+	/* Only saved on system call */
+	unsigned long r12_orig;
+};
+
+#ifdef __KERNEL__
+# define user_mode(regs) (((regs)->sr & MODE_MASK) == MODE_USER)
+extern void show_regs (struct pt_regs *);
+
+static __inline__ int valid_user_regs(struct pt_regs *regs)
+{
+	/*
+	 * Some of the Java bits might be acceptable if/when we
+	 * implement some support for that stuff...
+	 */
+	if ((regs->sr & 0xffff0000) == 0)
+		return 1;
+
+	/*
+	 * Force status register flags to be sane and report this
+	 * illegal behaviour...
+	 */
+	regs->sr &= 0x0000ffff;
+	return 0;
+}
+
+#define instruction_pointer(regs) ((regs)->pc)
+
+#define profile_pc(regs) instruction_pointer(regs)
+
+#endif /* __KERNEL__ */
+
+#endif /* ! __ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_PTRACE_H */
diff --git a/include/asm-avr32/resource.h b/include/asm-avr32/resource.h
new file mode 100644
index 000000000000..c6dd101472b1
--- /dev/null
+++ b/include/asm-avr32/resource.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_RESOURCE_H
+#define __ASM_AVR32_RESOURCE_H
+
+#include <asm-generic/resource.h>
+
+#endif /* __ASM_AVR32_RESOURCE_H */
diff --git a/include/asm-avr32/scatterlist.h b/include/asm-avr32/scatterlist.h
new file mode 100644
index 000000000000..bfe7d753423c
--- /dev/null
+++ b/include/asm-avr32/scatterlist.h
@@ -0,0 +1,21 @@
+#ifndef __ASM_AVR32_SCATTERLIST_H
+#define __ASM_AVR32_SCATTERLIST_H
+
+struct scatterlist {
+    struct page		*page;
+    unsigned int	offset;
+    dma_addr_t		dma_address;
+    unsigned int	length;
+};
+
+/* These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#define sg_dma_len(sg)		((sg)->length)
+
+#define ISA_DMA_THRESHOLD (0xffffffff)
+
+#endif /* __ASM_AVR32_SCATTERLIST_H */
diff --git a/include/asm-avr32/sections.h b/include/asm-avr32/sections.h
new file mode 100644
index 000000000000..aa14252e4181
--- /dev/null
+++ b/include/asm-avr32/sections.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_SECTIONS_H
+#define __ASM_AVR32_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+#endif /* __ASM_AVR32_SECTIONS_H */
diff --git a/include/asm-avr32/semaphore.h b/include/asm-avr32/semaphore.h
new file mode 100644
index 000000000000..ef99ddccc10c
--- /dev/null
+++ b/include/asm-avr32/semaphore.h
@@ -0,0 +1,109 @@
+/*
+ * SMP- and interrupt-safe semaphores.
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Based on include/asm-i386/semaphore.h
+ *   Copyright (C) 1996 Linus Torvalds
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SEMAPHORE_H
+#define __ASM_AVR32_SEMAPHORE_H
+
+#include <linux/linkage.h>
+
+#include <asm/system.h>
+#include <asm/atomic.h>
+#include <linux/wait.h>
+#include <linux/rwsem.h>
+
+struct semaphore {
+	atomic_t count;
+	int sleepers;
+	wait_queue_head_t wait;
+};
+
+#define __SEMAPHORE_INITIALIZER(name, n)				\
+{									\
+	.count		= ATOMIC_INIT(n),				\
+	.wait		= __WAIT_QUEUE_HEAD_INITIALIZER((name).wait)	\
+}
+
+#define __DECLARE_SEMAPHORE_GENERIC(name,count) \
+	struct semaphore name = __SEMAPHORE_INITIALIZER(name,count)
+
+#define DECLARE_MUTEX(name) __DECLARE_SEMAPHORE_GENERIC(name,1)
+#define DECLARE_MUTEX_LOCKED(name) __DECLARE_SEMAPHORE_GENERIC(name,0)
+
+static inline void sema_init (struct semaphore *sem, int val)
+{
+	atomic_set(&sem->count, val);
+	sem->sleepers = 0;
+	init_waitqueue_head(&sem->wait);
+}
+
+static inline void init_MUTEX (struct semaphore *sem)
+{
+	sema_init(sem, 1);
+}
+
+static inline void init_MUTEX_LOCKED (struct semaphore *sem)
+{
+	sema_init(sem, 0);
+}
+
+void __down(struct semaphore * sem);
+int  __down_interruptible(struct semaphore * sem);
+void __up(struct semaphore * sem);
+
+/*
+ * This is ugly, but we want the default case to fall through.
+ * "__down_failed" is a special asm handler that calls the C
+ * routine that actually waits. See arch/i386/kernel/semaphore.c
+ */
+static inline void down(struct semaphore * sem)
+{
+	might_sleep();
+	if (unlikely(atomic_dec_return (&sem->count) < 0))
+		__down (sem);
+}
+
+/*
+ * Interruptible try to acquire a semaphore.  If we obtained
+ * it, return zero.  If we were interrupted, returns -EINTR
+ */
+static inline int down_interruptible(struct semaphore * sem)
+{
+	int ret = 0;
+
+	might_sleep();
+	if (unlikely(atomic_dec_return (&sem->count) < 0))
+		ret = __down_interruptible (sem);
+	return ret;
+}
+
+/*
+ * Non-blockingly attempt to down() a semaphore.
+ * Returns zero if we acquired it
+ */
+static inline int down_trylock(struct semaphore * sem)
+{
+	return atomic_dec_if_positive(&sem->count) < 0;
+}
+
+/*
+ * Note! This is subtle. We jump to wake people up only if
+ * the semaphore was negative (== somebody was waiting on it).
+ * The default case (no contention) will result in NO
+ * jumps for both down() and up().
+ */
+static inline void up(struct semaphore * sem)
+{
+	if (unlikely(atomic_inc_return (&sem->count) <= 0))
+		__up (sem);
+}
+
+#endif /*__ASM_AVR32_SEMAPHORE_H */
diff --git a/include/asm-avr32/sembuf.h b/include/asm-avr32/sembuf.h
new file mode 100644
index 000000000000..e472216e0c97
--- /dev/null
+++ b/include/asm-avr32/sembuf.h
@@ -0,0 +1,25 @@
+#ifndef __ASM_AVR32_SEMBUF_H
+#define __ASM_AVR32_SEMBUF_H
+
+/*
+* The semid64_ds structure for AVR32 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct semid64_ds {
+        struct ipc64_perm sem_perm;             /* permissions .. see ipc.h */
+        __kernel_time_t sem_otime;              /* last semop time */
+        unsigned long   __unused1;
+        __kernel_time_t sem_ctime;              /* last change time */
+        unsigned long   __unused2;
+        unsigned long   sem_nsems;              /* no. of semaphores in array */
+        unsigned long   __unused3;
+        unsigned long   __unused4;
+};
+
+#endif /* __ASM_AVR32_SEMBUF_H */
diff --git a/include/asm-avr32/setup.h b/include/asm-avr32/setup.h
new file mode 100644
index 000000000000..10193da4113b
--- /dev/null
+++ b/include/asm-avr32/setup.h
@@ -0,0 +1,141 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * Based on linux/include/asm-arm/setup.h
+ *   Copyright (C) 1997-1999 Russel King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SETUP_H__
+#define __ASM_AVR32_SETUP_H__
+
+#define COMMAND_LINE_SIZE 256
+
+/* Magic number indicating that a tag table is present */
+#define ATAG_MAGIC	0xa2a25441
+
+#ifndef __ASSEMBLY__
+
+/*
+ * Generic memory range, used by several tags.
+ *
+ *   addr is always physical.
+ *   size is measured in bytes.
+ *   next is for use by the OS, e.g. for grouping regions into
+ *        linked lists.
+ */
+struct tag_mem_range {
+	u32			addr;
+	u32			size;
+	struct tag_mem_range *	next;
+};
+
+/* The list ends with an ATAG_NONE node. */
+#define ATAG_NONE	0x00000000
+
+struct tag_header {
+	u32 size;
+	u32 tag;
+};
+
+/* The list must start with an ATAG_CORE node */
+#define ATAG_CORE	0x54410001
+
+struct tag_core {
+	u32 flags;
+	u32 pagesize;
+	u32 rootdev;
+};
+
+/* it is allowed to have multiple ATAG_MEM nodes */
+#define ATAG_MEM	0x54410002
+/* ATAG_MEM uses tag_mem_range */
+
+/* command line: \0 terminated string */
+#define ATAG_CMDLINE	0x54410003
+
+struct tag_cmdline {
+	char	cmdline[1];	/* this is the minimum size */
+};
+
+/* Ramdisk image (may be compressed) */
+#define ATAG_RDIMG	0x54410004
+/* ATAG_RDIMG uses tag_mem_range */
+
+/* Information about various clocks present in the system */
+#define ATAG_CLOCK	0x54410005
+
+struct tag_clock {
+	u32	clock_id;	/* Which clock are we talking about? */
+	u32	clock_flags;	/* Special features */
+	u64	clock_hz;	/* Clock speed in Hz */
+};
+
+/* The clock types we know about */
+#define CLOCK_BOOTCPU	0
+
+/* Memory reserved for the system (e.g. the bootloader) */
+#define ATAG_RSVD_MEM	0x54410006
+/* ATAG_RSVD_MEM uses tag_mem_range */
+
+/* Ethernet information */
+
+#define ATAG_ETHERNET	0x54410007
+
+struct tag_ethernet {
+	u8	mac_index;
+	u8	mii_phy_addr;
+	u8	hw_address[6];
+};
+
+#define ETH_INVALID_PHY	0xff
+
+struct tag {
+	struct tag_header hdr;
+	union {
+		struct tag_core core;
+		struct tag_mem_range mem_range;
+		struct tag_cmdline cmdline;
+		struct tag_clock clock;
+		struct tag_ethernet ethernet;
+	} u;
+};
+
+struct tagtable {
+	u32	tag;
+	int	(*parse)(struct tag *);
+};
+
+#define __tag __attribute_used__ __attribute__((__section__(".taglist")))
+#define __tagtable(tag, fn)						\
+	static struct tagtable __tagtable_##fn __tag = { tag, fn }
+
+#define tag_member_present(tag,member)					\
+	((unsigned long)(&((struct tag *)0L)->member + 1)		\
+	 <= (tag)->hdr.size * 4)
+
+#define tag_next(t)	((struct tag *)((u32 *)(t) + (t)->hdr.size))
+#define tag_size(type)	((sizeof(struct tag_header) + sizeof(struct type)) >> 2)
+
+#define for_each_tag(t,base)						\
+	for (t = base; t->hdr.size; t = tag_next(t))
+
+extern struct tag_mem_range *mem_phys;
+extern struct tag_mem_range *mem_reserved;
+extern struct tag_mem_range *mem_ramdisk;
+
+extern struct tag *bootloader_tags;
+
+extern void setup_bootmem(void);
+extern void setup_processor(void);
+extern void board_setup_fbmem(unsigned long fbmem_start,
+			      unsigned long fbmem_size);
+
+/* Chip-specific hook to enable the use of SDRAM */
+void chip_enable_sdram(void);
+
+#endif /* !__ASSEMBLY__ */
+
+#endif /* __ASM_AVR32_SETUP_H__ */
diff --git a/include/asm-avr32/shmbuf.h b/include/asm-avr32/shmbuf.h
new file mode 100644
index 000000000000..c62fba41739a
--- /dev/null
+++ b/include/asm-avr32/shmbuf.h
@@ -0,0 +1,42 @@
+#ifndef __ASM_AVR32_SHMBUF_H
+#define __ASM_AVR32_SHMBUF_H
+
+/*
+ * The shmid64_ds structure for i386 architecture.
+ * Note extra padding because this structure is passed back and forth
+ * between kernel and user space.
+ *
+ * Pad space is left for:
+ * - 64-bit time_t to solve y2038 problem
+ * - 2 miscellaneous 32-bit values
+ */
+
+struct shmid64_ds {
+	struct ipc64_perm	shm_perm;	/* operation perms */
+	size_t			shm_segsz;	/* size of segment (bytes) */
+	__kernel_time_t		shm_atime;	/* last attach time */
+	unsigned long		__unused1;
+	__kernel_time_t		shm_dtime;	/* last detach time */
+	unsigned long		__unused2;
+	__kernel_time_t		shm_ctime;	/* last change time */
+	unsigned long		__unused3;
+	__kernel_pid_t		shm_cpid;	/* pid of creator */
+	__kernel_pid_t		shm_lpid;	/* pid of last operator */
+	unsigned long		shm_nattch;	/* no. of current attaches */
+	unsigned long		__unused4;
+	unsigned long		__unused5;
+};
+
+struct shminfo64 {
+	unsigned long	shmmax;
+	unsigned long	shmmin;
+	unsigned long	shmmni;
+	unsigned long	shmseg;
+	unsigned long	shmall;
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+	unsigned long	__unused3;
+	unsigned long	__unused4;
+};
+
+#endif /* __ASM_AVR32_SHMBUF_H */
diff --git a/include/asm-avr32/shmparam.h b/include/asm-avr32/shmparam.h
new file mode 100644
index 000000000000..3681266c77f7
--- /dev/null
+++ b/include/asm-avr32/shmparam.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_SHMPARAM_H
+#define __ASM_AVR32_SHMPARAM_H
+
+#define SHMLBA PAGE_SIZE	/* attach addr a multiple of this */
+
+#endif /* __ASM_AVR32_SHMPARAM_H */
diff --git a/include/asm-avr32/sigcontext.h b/include/asm-avr32/sigcontext.h
new file mode 100644
index 000000000000..e04062b5f39f
--- /dev/null
+++ b/include/asm-avr32/sigcontext.h
@@ -0,0 +1,34 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SIGCONTEXT_H
+#define __ASM_AVR32_SIGCONTEXT_H
+
+struct sigcontext {
+	unsigned long	oldmask;
+
+	/* CPU registers */
+	unsigned long	sr;
+	unsigned long	pc;
+	unsigned long	lr;
+	unsigned long	sp;
+	unsigned long	r12;
+	unsigned long	r11;
+	unsigned long	r10;
+	unsigned long	r9;
+	unsigned long	r8;
+	unsigned long	r7;
+	unsigned long	r6;
+	unsigned long	r5;
+	unsigned long	r4;
+	unsigned long	r3;
+	unsigned long	r2;
+	unsigned long	r1;
+	unsigned long	r0;
+};
+
+#endif /* __ASM_AVR32_SIGCONTEXT_H */
diff --git a/include/asm-avr32/siginfo.h b/include/asm-avr32/siginfo.h
new file mode 100644
index 000000000000..5ee93f40a8a8
--- /dev/null
+++ b/include/asm-avr32/siginfo.h
@@ -0,0 +1,6 @@
+#ifndef _AVR32_SIGINFO_H
+#define _AVR32_SIGINFO_H
+
+#include <asm-generic/siginfo.h>
+
+#endif
diff --git a/include/asm-avr32/signal.h b/include/asm-avr32/signal.h
new file mode 100644
index 000000000000..caffefeeba1f
--- /dev/null
+++ b/include/asm-avr32/signal.h
@@ -0,0 +1,168 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SIGNAL_H
+#define __ASM_AVR32_SIGNAL_H
+
+#include <linux/types.h>
+
+/* Avoid too many header ordering problems.  */
+struct siginfo;
+
+#ifdef __KERNEL__
+/* Most things should be clean enough to redefine this at will, if care
+   is taken to make libc match.  */
+
+#define _NSIG		64
+#define _NSIG_BPW	32
+#define _NSIG_WORDS	(_NSIG / _NSIG_BPW)
+
+typedef unsigned long old_sigset_t;		/* at least 32 bits */
+
+typedef struct {
+	unsigned long sig[_NSIG_WORDS];
+} sigset_t;
+
+#else
+/* Here we must cater to libcs that poke about in kernel headers.  */
+
+#define NSIG		32
+typedef unsigned long sigset_t;
+
+#endif /* __KERNEL__ */
+
+#define SIGHUP		 1
+#define SIGINT		 2
+#define SIGQUIT		 3
+#define SIGILL		 4
+#define SIGTRAP		 5
+#define SIGABRT		 6
+#define SIGIOT		 6
+#define SIGBUS		 7
+#define SIGFPE		 8
+#define SIGKILL		 9
+#define SIGUSR1		10
+#define SIGSEGV		11
+#define SIGUSR2		12
+#define SIGPIPE		13
+#define SIGALRM		14
+#define SIGTERM		15
+#define SIGSTKFLT	16
+#define SIGCHLD		17
+#define SIGCONT		18
+#define SIGSTOP		19
+#define SIGTSTP		20
+#define SIGTTIN		21
+#define SIGTTOU		22
+#define SIGURG		23
+#define SIGXCPU		24
+#define SIGXFSZ		25
+#define SIGVTALRM	26
+#define SIGPROF		27
+#define SIGWINCH	28
+#define SIGIO		29
+#define SIGPOLL		SIGIO
+/*
+#define SIGLOST		29
+*/
+#define SIGPWR		30
+#define SIGSYS		31
+#define	SIGUNUSED	31
+
+/* These should not be considered constants from userland.  */
+#define SIGRTMIN	32
+#define SIGRTMAX	(_NSIG-1)
+
+/*
+ * SA_FLAGS values:
+ *
+ * SA_NOCLDSTOP		flag to turn off SIGCHLD when children stop.
+ * SA_NOCLDWAIT		flag on SIGCHLD to inhibit zombies.
+ * SA_SIGINFO		deliver the signal with SIGINFO structs
+ * SA_ONSTACK		indicates that a registered stack_t will be used.
+ * SA_RESTART		flag to get restarting signals (which were the default long ago)
+ * SA_NODEFER		prevents the current signal from being masked in the handler.
+ * SA_RESETHAND		clears the handler when the signal is delivered.
+ *
+ * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single
+ * Unix names RESETHAND and NODEFER respectively.
+ */
+#define SA_NOCLDSTOP	0x00000001
+#define SA_NOCLDWAIT	0x00000002
+#define SA_SIGINFO	0x00000004
+#define SA_RESTORER	0x04000000
+#define SA_ONSTACK	0x08000000
+#define SA_RESTART	0x10000000
+#define SA_NODEFER	0x40000000
+#define SA_RESETHAND	0x80000000
+
+#define SA_NOMASK	SA_NODEFER
+#define SA_ONESHOT	SA_RESETHAND
+
+/*
+ * sigaltstack controls
+ */
+#define SS_ONSTACK	1
+#define SS_DISABLE	2
+
+#define MINSIGSTKSZ	2048
+#define SIGSTKSZ	8192
+
+#include <asm-generic/signal.h>
+
+#ifdef __KERNEL__
+struct old_sigaction {
+	__sighandler_t sa_handler;
+	old_sigset_t sa_mask;
+	unsigned long sa_flags;
+	__sigrestore_t sa_restorer;
+};
+
+struct sigaction {
+	__sighandler_t sa_handler;
+	unsigned long sa_flags;
+	__sigrestore_t sa_restorer;
+	sigset_t sa_mask;		/* mask last for extensibility */
+};
+
+struct k_sigaction {
+	struct sigaction sa;
+};
+#else
+/* Here we must cater to libcs that poke about in kernel headers.  */
+
+struct sigaction {
+	union {
+		__sighandler_t _sa_handler;
+		void (*_sa_sigaction)(int, struct siginfo *, void *);
+	} _u;
+	sigset_t sa_mask;
+	unsigned long sa_flags;
+	void (*sa_restorer)(void);
+};
+
+#define sa_handler	_u._sa_handler
+#define sa_sigaction	_u._sa_sigaction
+
+#endif /* __KERNEL__ */
+
+typedef struct sigaltstack {
+	void __user *ss_sp;
+	int ss_flags;
+	size_t ss_size;
+} stack_t;
+
+#ifdef __KERNEL__
+
+#include <asm/sigcontext.h>
+#undef __HAVE_ARCH_SIG_BITOPS
+
+#define ptrace_signal_deliver(regs, cookie) do { } while (0)
+
+#endif /* __KERNEL__ */
+
+#endif
diff --git a/include/asm-avr32/socket.h b/include/asm-avr32/socket.h
new file mode 100644
index 000000000000..543229de8173
--- /dev/null
+++ b/include/asm-avr32/socket.h
@@ -0,0 +1,53 @@
+#ifndef __ASM_AVR32_SOCKET_H
+#define __ASM_AVR32_SOCKET_H
+
+#include <asm/sockios.h>
+
+/* For setsockopt(2) */
+#define SOL_SOCKET	1
+
+#define SO_DEBUG	1
+#define SO_REUSEADDR	2
+#define SO_TYPE		3
+#define SO_ERROR	4
+#define SO_DONTROUTE	5
+#define SO_BROADCAST	6
+#define SO_SNDBUF	7
+#define SO_RCVBUF	8
+#define SO_SNDBUFFORCE	32
+#define SO_RCVBUFFORCE	33
+#define SO_KEEPALIVE	9
+#define SO_OOBINLINE	10
+#define SO_NO_CHECK	11
+#define SO_PRIORITY	12
+#define SO_LINGER	13
+#define SO_BSDCOMPAT	14
+/* To add :#define SO_REUSEPORT 15 */
+#define SO_PASSCRED	16
+#define SO_PEERCRED	17
+#define SO_RCVLOWAT	18
+#define SO_SNDLOWAT	19
+#define SO_RCVTIMEO	20
+#define SO_SNDTIMEO	21
+
+/* Security levels - as per NRL IPv6 - don't actually do anything */
+#define SO_SECURITY_AUTHENTICATION		22
+#define SO_SECURITY_ENCRYPTION_TRANSPORT	23
+#define SO_SECURITY_ENCRYPTION_NETWORK		24
+
+#define SO_BINDTODEVICE	25
+
+/* Socket filtering */
+#define SO_ATTACH_FILTER        26
+#define SO_DETACH_FILTER        27
+
+#define SO_PEERNAME		28
+#define SO_TIMESTAMP		29
+#define SCM_TIMESTAMP		SO_TIMESTAMP
+
+#define SO_ACCEPTCONN		30
+
+#define SO_PEERSEC		31
+#define SO_PASSSEC		34
+
+#endif /* __ASM_AVR32_SOCKET_H */
diff --git a/include/asm-avr32/sockios.h b/include/asm-avr32/sockios.h
new file mode 100644
index 000000000000..84f3d65b3b3b
--- /dev/null
+++ b/include/asm-avr32/sockios.h
@@ -0,0 +1,12 @@
+#ifndef __ASM_AVR32_SOCKIOS_H
+#define __ASM_AVR32_SOCKIOS_H
+
+/* Socket-level I/O control calls. */
+#define FIOSETOWN 	0x8901
+#define SIOCSPGRP	0x8902
+#define FIOGETOWN	0x8903
+#define SIOCGPGRP	0x8904
+#define SIOCATMARK	0x8905
+#define SIOCGSTAMP	0x8906		/* Get stamp */
+
+#endif /* __ASM_AVR32_SOCKIOS_H */
diff --git a/include/asm-avr32/stat.h b/include/asm-avr32/stat.h
new file mode 100644
index 000000000000..e72881e10230
--- /dev/null
+++ b/include/asm-avr32/stat.h
@@ -0,0 +1,79 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_STAT_H
+#define __ASM_AVR32_STAT_H
+
+struct __old_kernel_stat {
+        unsigned short st_dev;
+        unsigned short st_ino;
+        unsigned short st_mode;
+        unsigned short st_nlink;
+        unsigned short st_uid;
+        unsigned short st_gid;
+        unsigned short st_rdev;
+        unsigned long  st_size;
+        unsigned long  st_atime;
+        unsigned long  st_mtime;
+        unsigned long  st_ctime;
+};
+
+struct stat {
+        unsigned long st_dev;
+        unsigned long st_ino;
+        unsigned short st_mode;
+        unsigned short st_nlink;
+        unsigned short st_uid;
+        unsigned short st_gid;
+        unsigned long  st_rdev;
+        unsigned long  st_size;
+        unsigned long  st_blksize;
+        unsigned long  st_blocks;
+        unsigned long  st_atime;
+        unsigned long  st_atime_nsec;
+        unsigned long  st_mtime;
+        unsigned long  st_mtime_nsec;
+        unsigned long  st_ctime;
+        unsigned long  st_ctime_nsec;
+        unsigned long  __unused4;
+        unsigned long  __unused5;
+};
+
+#define STAT_HAVE_NSEC 1
+
+struct stat64 {
+	unsigned long long st_dev;
+
+	unsigned long long st_ino;
+	unsigned int	st_mode;
+	unsigned int	st_nlink;
+
+	unsigned long	st_uid;
+	unsigned long	st_gid;
+
+	unsigned long long st_rdev;
+
+	long long	st_size;
+	unsigned long	__pad1;		/* align 64-bit st_blocks */
+	unsigned long	st_blksize;
+
+	unsigned long long st_blocks;	/* Number 512-byte blocks allocated. */
+
+	unsigned long	st_atime;
+	unsigned long	st_atime_nsec;
+
+	unsigned long	st_mtime;
+	unsigned long	st_mtime_nsec;
+
+	unsigned long	st_ctime;
+	unsigned long	st_ctime_nsec;
+
+	unsigned long	__unused1;
+	unsigned long	__unused2;
+};
+
+#endif /* __ASM_AVR32_STAT_H */
diff --git a/include/asm-avr32/statfs.h b/include/asm-avr32/statfs.h
new file mode 100644
index 000000000000..2961bd18c50e
--- /dev/null
+++ b/include/asm-avr32/statfs.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_STATFS_H
+#define __ASM_AVR32_STATFS_H
+
+#include <asm-generic/statfs.h>
+
+#endif /* __ASM_AVR32_STATFS_H */
diff --git a/include/asm-avr32/string.h b/include/asm-avr32/string.h
new file mode 100644
index 000000000000..c91a623cd585
--- /dev/null
+++ b/include/asm-avr32/string.h
@@ -0,0 +1,17 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_STRING_H
+#define __ASM_AVR32_STRING_H
+
+#define __HAVE_ARCH_MEMSET
+extern void *memset(void *b, int c, size_t len);
+
+#define __HAVE_ARCH_MEMCPY
+extern void *memcpy(void *to, const void *from, size_t len);
+
+#endif /* __ASM_AVR32_STRING_H */
diff --git a/include/asm-avr32/sysreg.h b/include/asm-avr32/sysreg.h
new file mode 100644
index 000000000000..f91975f330f6
--- /dev/null
+++ b/include/asm-avr32/sysreg.h
@@ -0,0 +1,332 @@
+/*
+ * AVR32 System Registers
+ *
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SYSREG_H__
+#define __ASM_AVR32_SYSREG_H__
+
+/* sysreg register offsets */
+#define SYSREG_SR                               0x0000
+#define SYSREG_EVBA                             0x0004
+#define SYSREG_ACBA                             0x0008
+#define SYSREG_CPUCR                            0x000c
+#define SYSREG_ECR                              0x0010
+#define SYSREG_RSR_SUP                          0x0014
+#define SYSREG_RSR_INT0                         0x0018
+#define SYSREG_RSR_INT1                         0x001c
+#define SYSREG_RSR_INT2                         0x0020
+#define SYSREG_RSR_INT3                         0x0024
+#define SYSREG_RSR_EX                           0x0028
+#define SYSREG_RSR_NMI                          0x002c
+#define SYSREG_RSR_DBG                          0x0030
+#define SYSREG_RAR_SUP                          0x0034
+#define SYSREG_RAR_INT0                         0x0038
+#define SYSREG_RAR_INT1                         0x003c
+#define SYSREG_RAR_INT2                         0x0040
+#define SYSREG_RAR_INT3                         0x0044
+#define SYSREG_RAR_EX                           0x0048
+#define SYSREG_RAR_NMI                          0x004c
+#define SYSREG_RAR_DBG                          0x0050
+#define SYSREG_JECR                             0x0054
+#define SYSREG_JOSP                             0x0058
+#define SYSREG_JAVA_LV0                         0x005c
+#define SYSREG_JAVA_LV1                         0x0060
+#define SYSREG_JAVA_LV2                         0x0064
+#define SYSREG_JAVA_LV3                         0x0068
+#define SYSREG_JAVA_LV4                         0x006c
+#define SYSREG_JAVA_LV5                         0x0070
+#define SYSREG_JAVA_LV6                         0x0074
+#define SYSREG_JAVA_LV7                         0x0078
+#define SYSREG_JTBA                             0x007c
+#define SYSREG_JBCR                             0x0080
+#define SYSREG_CONFIG0                          0x0100
+#define SYSREG_CONFIG1                          0x0104
+#define SYSREG_COUNT                            0x0108
+#define SYSREG_COMPARE                          0x010c
+#define SYSREG_TLBEHI                           0x0110
+#define SYSREG_TLBELO                           0x0114
+#define SYSREG_PTBR                             0x0118
+#define SYSREG_TLBEAR                           0x011c
+#define SYSREG_MMUCR                            0x0120
+#define SYSREG_TLBARLO                          0x0124
+#define SYSREG_TLBARHI                          0x0128
+#define SYSREG_PCCNT                            0x012c
+#define SYSREG_PCNT0                            0x0130
+#define SYSREG_PCNT1                            0x0134
+#define SYSREG_PCCR                             0x0138
+#define SYSREG_BEAR                             0x013c
+
+/* Bitfields in SR */
+#define SYSREG_SR_C_OFFSET                      0
+#define SYSREG_SR_C_SIZE                        1
+#define SYSREG_Z_OFFSET                         1
+#define SYSREG_Z_SIZE                           1
+#define SYSREG_SR_N_OFFSET                      2
+#define SYSREG_SR_N_SIZE                        1
+#define SYSREG_SR_V_OFFSET                      3
+#define SYSREG_SR_V_SIZE                        1
+#define SYSREG_Q_OFFSET                         4
+#define SYSREG_Q_SIZE                           1
+#define SYSREG_GM_OFFSET                        16
+#define SYSREG_GM_SIZE                          1
+#define SYSREG_I0M_OFFSET                       17
+#define SYSREG_I0M_SIZE                         1
+#define SYSREG_I1M_OFFSET                       18
+#define SYSREG_I1M_SIZE                         1
+#define SYSREG_I2M_OFFSET                       19
+#define SYSREG_I2M_SIZE                         1
+#define SYSREG_I3M_OFFSET                       20
+#define SYSREG_I3M_SIZE                         1
+#define SYSREG_EM_OFFSET                        21
+#define SYSREG_EM_SIZE                          1
+#define SYSREG_M0_OFFSET                        22
+#define SYSREG_M0_SIZE                          1
+#define SYSREG_M1_OFFSET                        23
+#define SYSREG_M1_SIZE                          1
+#define SYSREG_M2_OFFSET                        24
+#define SYSREG_M2_SIZE                          1
+#define SYSREG_SR_D_OFFSET                      26
+#define SYSREG_SR_D_SIZE                        1
+#define SYSREG_DM_OFFSET                        27
+#define SYSREG_DM_SIZE                          1
+#define SYSREG_SR_J_OFFSET                      28
+#define SYSREG_SR_J_SIZE                        1
+#define SYSREG_R_OFFSET                         29
+#define SYSREG_R_SIZE                           1
+#define SYSREG_H_OFFSET                         30
+#define SYSREG_H_SIZE                           1
+
+/* Bitfields in EVBA */
+
+/* Bitfields in ACBA */
+
+/* Bitfields in CPUCR */
+#define SYSREG_BI_OFFSET                        0
+#define SYSREG_BI_SIZE                          1
+#define SYSREG_BE_OFFSET                        1
+#define SYSREG_BE_SIZE                          1
+#define SYSREG_FE_OFFSET                        2
+#define SYSREG_FE_SIZE                          1
+#define SYSREG_RE_OFFSET                        3
+#define SYSREG_RE_SIZE                          1
+#define SYSREG_IBE_OFFSET                       4
+#define SYSREG_IBE_SIZE                         1
+#define SYSREG_IEE_OFFSET                       5
+#define SYSREG_IEE_SIZE                         1
+
+/* Bitfields in ECR */
+#define SYSREG_ECR_OFFSET                       0
+#define SYSREG_ECR_SIZE                         32
+
+/* Bitfields in RSR_SUP */
+
+/* Bitfields in RSR_INT0 */
+
+/* Bitfields in RSR_INT1 */
+
+/* Bitfields in RSR_INT2 */
+
+/* Bitfields in RSR_INT3 */
+
+/* Bitfields in RSR_EX */
+
+/* Bitfields in RSR_NMI */
+
+/* Bitfields in RSR_DBG */
+
+/* Bitfields in RAR_SUP */
+
+/* Bitfields in RAR_INT0 */
+
+/* Bitfields in RAR_INT1 */
+
+/* Bitfields in RAR_INT2 */
+
+/* Bitfields in RAR_INT3 */
+
+/* Bitfields in RAR_EX */
+
+/* Bitfields in RAR_NMI */
+
+/* Bitfields in RAR_DBG */
+
+/* Bitfields in JECR */
+
+/* Bitfields in JOSP */
+
+/* Bitfields in JAVA_LV0 */
+
+/* Bitfields in JAVA_LV1 */
+
+/* Bitfields in JAVA_LV2 */
+
+/* Bitfields in JAVA_LV3 */
+
+/* Bitfields in JAVA_LV4 */
+
+/* Bitfields in JAVA_LV5 */
+
+/* Bitfields in JAVA_LV6 */
+
+/* Bitfields in JAVA_LV7 */
+
+/* Bitfields in JTBA */
+
+/* Bitfields in JBCR */
+
+/* Bitfields in CONFIG0 */
+#define SYSREG_CONFIG0_D_OFFSET                 1
+#define SYSREG_CONFIG0_D_SIZE                   1
+#define SYSREG_CONFIG0_S_OFFSET                 2
+#define SYSREG_CONFIG0_S_SIZE                   1
+#define SYSREG_O_OFFSET                         3
+#define SYSREG_O_SIZE                           1
+#define SYSREG_P_OFFSET                         4
+#define SYSREG_P_SIZE                           1
+#define SYSREG_CONFIG0_J_OFFSET                 5
+#define SYSREG_CONFIG0_J_SIZE                   1
+#define SYSREG_F_OFFSET                         6
+#define SYSREG_F_SIZE                           1
+#define SYSREG_MMUT_OFFSET                      7
+#define SYSREG_MMUT_SIZE                        3
+#define SYSREG_AR_OFFSET                        10
+#define SYSREG_AR_SIZE                          3
+#define SYSREG_AT_OFFSET                        13
+#define SYSREG_AT_SIZE                          3
+#define SYSREG_PROCESSORREVISION_OFFSET         16
+#define SYSREG_PROCESSORREVISION_SIZE           8
+#define SYSREG_PROCESSORID_OFFSET               24
+#define SYSREG_PROCESSORID_SIZE                 8
+
+/* Bitfields in CONFIG1 */
+#define SYSREG_DASS_OFFSET                      0
+#define SYSREG_DASS_SIZE                        3
+#define SYSREG_DLSZ_OFFSET                      3
+#define SYSREG_DLSZ_SIZE                        3
+#define SYSREG_DSET_OFFSET                      6
+#define SYSREG_DSET_SIZE                        4
+#define SYSREG_IASS_OFFSET                      10
+#define SYSREG_IASS_SIZE                        2
+#define SYSREG_ILSZ_OFFSET                      13
+#define SYSREG_ILSZ_SIZE                        3
+#define SYSREG_ISET_OFFSET                      16
+#define SYSREG_ISET_SIZE                        4
+#define SYSREG_DMMUSZ_OFFSET                    20
+#define SYSREG_DMMUSZ_SIZE                      6
+#define SYSREG_IMMUSZ_OFFSET                    26
+#define SYSREG_IMMUSZ_SIZE                      6
+
+/* Bitfields in COUNT */
+
+/* Bitfields in COMPARE */
+
+/* Bitfields in TLBEHI */
+#define SYSREG_ASID_OFFSET                      0
+#define SYSREG_ASID_SIZE                        8
+#define SYSREG_TLBEHI_I_OFFSET                  8
+#define SYSREG_TLBEHI_I_SIZE                    1
+#define SYSREG_TLBEHI_V_OFFSET                  9
+#define SYSREG_TLBEHI_V_SIZE                    1
+#define SYSREG_VPN_OFFSET                       10
+#define SYSREG_VPN_SIZE                         22
+
+/* Bitfields in TLBELO */
+#define SYSREG_W_OFFSET                         0
+#define SYSREG_W_SIZE                           1
+#define SYSREG_TLBELO_D_OFFSET                  1
+#define SYSREG_TLBELO_D_SIZE                    1
+#define SYSREG_SZ_OFFSET                        2
+#define SYSREG_SZ_SIZE                          2
+#define SYSREG_AP_OFFSET                        4
+#define SYSREG_AP_SIZE                          3
+#define SYSREG_B_OFFSET                         7
+#define SYSREG_B_SIZE                           1
+#define SYSREG_G_OFFSET                         8
+#define SYSREG_G_SIZE                           1
+#define SYSREG_TLBELO_C_OFFSET                  9
+#define SYSREG_TLBELO_C_SIZE                    1
+#define SYSREG_PFN_OFFSET                       10
+#define SYSREG_PFN_SIZE                         22
+
+/* Bitfields in PTBR */
+
+/* Bitfields in TLBEAR */
+
+/* Bitfields in MMUCR */
+#define SYSREG_E_OFFSET                         0
+#define SYSREG_E_SIZE                           1
+#define SYSREG_M_OFFSET                         1
+#define SYSREG_M_SIZE                           1
+#define SYSREG_MMUCR_I_OFFSET                   2
+#define SYSREG_MMUCR_I_SIZE                     1
+#define SYSREG_MMUCR_N_OFFSET                   3
+#define SYSREG_MMUCR_N_SIZE                     1
+#define SYSREG_MMUCR_S_OFFSET                   4
+#define SYSREG_MMUCR_S_SIZE                     1
+#define SYSREG_DLA_OFFSET                       8
+#define SYSREG_DLA_SIZE                         6
+#define SYSREG_DRP_OFFSET                       14
+#define SYSREG_DRP_SIZE                         6
+#define SYSREG_ILA_OFFSET                       20
+#define SYSREG_ILA_SIZE                         6
+#define SYSREG_IRP_OFFSET                       26
+#define SYSREG_IRP_SIZE                         6
+
+/* Bitfields in TLBARLO */
+
+/* Bitfields in TLBARHI */
+
+/* Bitfields in PCCNT */
+
+/* Bitfields in PCNT0 */
+
+/* Bitfields in PCNT1 */
+
+/* Bitfields in PCCR */
+
+/* Bitfields in BEAR */
+
+/* Constants for ECR */
+#define ECR_UNRECOVERABLE                       0
+#define ECR_TLB_MULTIPLE                        1
+#define ECR_BUS_ERROR_WRITE                     2
+#define ECR_BUS_ERROR_READ                      3
+#define ECR_NMI                                 4
+#define ECR_ADDR_ALIGN_X                        5
+#define ECR_PROTECTION_X                        6
+#define ECR_DEBUG                               7
+#define ECR_ILLEGAL_OPCODE                      8
+#define ECR_UNIMPL_INSTRUCTION                  9
+#define ECR_PRIVILEGE_VIOLATION                 10
+#define ECR_FPE                                 11
+#define ECR_COPROC_ABSENT                       12
+#define ECR_ADDR_ALIGN_R                        13
+#define ECR_ADDR_ALIGN_W                        14
+#define ECR_PROTECTION_R                        15
+#define ECR_PROTECTION_W                        16
+#define ECR_DTLB_MODIFIED                       17
+#define ECR_TLB_MISS_X                          20
+#define ECR_TLB_MISS_R                          24
+#define ECR_TLB_MISS_W                          28
+
+/* Bit manipulation macros */
+#define SYSREG_BIT(name)                        (1 << SYSREG_##name##_OFFSET)
+#define SYSREG_BF(name,value)                   (((value) & ((1 << SYSREG_##name##_SIZE) - 1)) << SYSREG_##name##_OFFSET)
+#define SYSREG_BFEXT(name,value)                (((value) >> SYSREG_##name##_OFFSET) & ((1 << SYSREG_##name##_SIZE) - 1))
+#define SYSREG_BFINS(name,value,old)            (((old) & ~(((1 << SYSREG_##name##_SIZE) - 1) << SYSREG_##name##_OFFSET)) | SYSREG_BF(name,value))
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_mfsr(unsigned long reg);
+extern void __builtin_mtsr(unsigned long reg, unsigned long value);
+#endif
+
+/* Register access macros */
+#define sysreg_read(reg)                        __builtin_mfsr(SYSREG_##reg)
+#define sysreg_write(reg, value)                __builtin_mtsr(SYSREG_##reg, value)
+
+#endif /* __ASM_AVR32_SYSREG_H__ */
diff --git a/include/asm-avr32/system.h b/include/asm-avr32/system.h
new file mode 100644
index 000000000000..ac596058697d
--- /dev/null
+++ b/include/asm-avr32/system.h
@@ -0,0 +1,155 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_SYSTEM_H
+#define __ASM_AVR32_SYSTEM_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+
+#include <asm/ptrace.h>
+#include <asm/sysreg.h>
+
+#define xchg(ptr,x) \
+	((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
+
+#define nop() asm volatile("nop")
+
+#define mb()			asm volatile("" : : : "memory")
+#define rmb()			mb()
+#define wmb()			asm volatile("sync 0" : : : "memory")
+#define read_barrier_depends()  do { } while(0)
+#define set_mb(var, value)      do { var = value; mb(); } while(0)
+
+/*
+ * Help PathFinder and other Nexus-compliant debuggers keep track of
+ * the current PID by emitting an Ownership Trace Message each time we
+ * switch task.
+ */
+#ifdef CONFIG_OWNERSHIP_TRACE
+#include <asm/ocd.h>
+#define finish_arch_switch(prev)			\
+	do {						\
+		__mtdr(DBGREG_PID, prev->pid);		\
+		__mtdr(DBGREG_PID, current->pid);	\
+	} while(0)
+#endif
+
+/*
+ * switch_to(prev, next, last) should switch from task `prev' to task
+ * `next'. `prev' will never be the same as `next'.
+ *
+ * We just delegate everything to the __switch_to assembly function,
+ * which is implemented in arch/avr32/kernel/switch_to.S
+ *
+ * mb() tells GCC not to cache `current' across this call.
+ */
+struct cpu_context;
+struct task_struct;
+extern struct task_struct *__switch_to(struct task_struct *,
+				       struct cpu_context *,
+				       struct cpu_context *);
+#define switch_to(prev, next, last)					\
+	do {								\
+		last = __switch_to(prev, &prev->thread.cpu_context + 1,	\
+				   &next->thread.cpu_context);		\
+	} while (0)
+
+#ifdef CONFIG_SMP
+# error "The AVR32 port does not support SMP"
+#else
+# define smp_mb()		barrier()
+# define smp_rmb()		barrier()
+# define smp_wmb()		barrier()
+# define smp_read_barrier_depends() do { } while(0)
+#endif
+
+#include <linux/irqflags.h>
+
+extern void __xchg_called_with_bad_pointer(void);
+
+#ifdef __CHECKER__
+extern unsigned long __builtin_xchg(void *ptr, unsigned long x);
+#endif
+
+#define xchg_u32(val, m) __builtin_xchg((void *)m, val)
+
+static inline unsigned long __xchg(unsigned long x,
+				       volatile void *ptr,
+				       int size)
+{
+	switch(size) {
+	case 4:
+		return xchg_u32(x, ptr);
+	default:
+		__xchg_called_with_bad_pointer();
+		return x;
+	}
+}
+
+static inline unsigned long __cmpxchg_u32(volatile int *m, unsigned long old,
+					  unsigned long new)
+{
+	__u32 ret;
+
+	asm volatile(
+		"1:	ssrf	5\n"
+		"	ld.w	%[ret], %[m]\n"
+		"	cp.w	%[ret], %[old]\n"
+		"	brne	2f\n"
+		"	stcond	%[m], %[new]\n"
+		"	brne	1b\n"
+		"2:\n"
+		: [ret] "=&r"(ret), [m] "=m"(*m)
+		: "m"(m), [old] "ir"(old), [new] "r"(new)
+		: "memory", "cc");
+	return ret;
+}
+
+extern unsigned long __cmpxchg_u64_unsupported_on_32bit_kernels(
+        volatile int * m, unsigned long old, unsigned long new);
+#define __cmpxchg_u64 __cmpxchg_u64_unsupported_on_32bit_kernels
+
+/* This function doesn't exist, so you'll get a linker error
+   if something tries to do an invalid cmpxchg().  */
+extern void __cmpxchg_called_with_bad_pointer(void);
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
+				      unsigned long new, int size)
+{
+	switch (size) {
+	case 4:
+		return __cmpxchg_u32(ptr, old, new);
+	case 8:
+		return __cmpxchg_u64(ptr, old, new);
+	}
+
+	__cmpxchg_called_with_bad_pointer();
+	return old;
+}
+
+#define cmpxchg(ptr, old, new)					\
+	((typeof(*(ptr)))__cmpxchg((ptr), (unsigned long)(old),	\
+				   (unsigned long)(new),	\
+				   sizeof(*(ptr))))
+
+struct pt_regs;
+extern void __die(const char *, struct pt_regs *, unsigned long,
+		  const char *, const char *, unsigned long);
+extern void __die_if_kernel(const char *, struct pt_regs *, unsigned long,
+			    const char *, const char *, unsigned long);
+
+#define die(msg, regs, err)					\
+	__die(msg, regs, err, __FILE__ ":", __FUNCTION__, __LINE__)
+#define die_if_kernel(msg, regs, err)					\
+	__die_if_kernel(msg, regs, err, __FILE__ ":", __FUNCTION__, __LINE__)
+
+#define arch_align_stack(x)	(x)
+
+#endif /* __ASM_AVR32_SYSTEM_H */
diff --git a/include/asm-avr32/termbits.h b/include/asm-avr32/termbits.h
new file mode 100644
index 000000000000..9dc6eacafa33
--- /dev/null
+++ b/include/asm-avr32/termbits.h
@@ -0,0 +1,173 @@
+#ifndef __ASM_AVR32_TERMBITS_H
+#define __ASM_AVR32_TERMBITS_H
+
+#include <linux/posix_types.h>
+
+typedef unsigned char	cc_t;
+typedef unsigned int	speed_t;
+typedef unsigned int	tcflag_t;
+
+#define NCCS 19
+struct termios {
+	tcflag_t c_iflag;		/* input mode flags */
+	tcflag_t c_oflag;		/* output mode flags */
+	tcflag_t c_cflag;		/* control mode flags */
+	tcflag_t c_lflag;		/* local mode flags */
+	cc_t c_line;			/* line discipline */
+	cc_t c_cc[NCCS];		/* control characters */
+};
+
+/* c_cc characters */
+#define VINTR 0
+#define VQUIT 1
+#define VERASE 2
+#define VKILL 3
+#define VEOF 4
+#define VTIME 5
+#define VMIN 6
+#define VSWTC 7
+#define VSTART 8
+#define VSTOP 9
+#define VSUSP 10
+#define VEOL 11
+#define VREPRINT 12
+#define VDISCARD 13
+#define VWERASE 14
+#define VLNEXT 15
+#define VEOL2 16
+
+/* c_iflag bits */
+#define IGNBRK	0000001
+#define BRKINT	0000002
+#define IGNPAR	0000004
+#define PARMRK	0000010
+#define INPCK	0000020
+#define ISTRIP	0000040
+#define INLCR	0000100
+#define IGNCR	0000200
+#define ICRNL	0000400
+#define IUCLC	0001000
+#define IXON	0002000
+#define IXANY	0004000
+#define IXOFF	0010000
+#define IMAXBEL	0020000
+#define IUTF8	0040000
+
+/* c_oflag bits */
+#define OPOST	0000001
+#define OLCUC	0000002
+#define ONLCR	0000004
+#define OCRNL	0000010
+#define ONOCR	0000020
+#define ONLRET	0000040
+#define OFILL	0000100
+#define OFDEL	0000200
+#define NLDLY	0000400
+#define   NL0	0000000
+#define   NL1	0000400
+#define CRDLY	0003000
+#define   CR0	0000000
+#define   CR1	0001000
+#define   CR2	0002000
+#define   CR3	0003000
+#define TABDLY	0014000
+#define   TAB0	0000000
+#define   TAB1	0004000
+#define   TAB2	0010000
+#define   TAB3	0014000
+#define   XTABS	0014000
+#define BSDLY	0020000
+#define   BS0	0000000
+#define   BS1	0020000
+#define VTDLY	0040000
+#define   VT0	0000000
+#define   VT1	0040000
+#define FFDLY	0100000
+#define   FF0	0000000
+#define   FF1	0100000
+
+/* c_cflag bit meaning */
+#define CBAUD	0010017
+#define  B0	0000000		/* hang up */
+#define  B50	0000001
+#define  B75	0000002
+#define  B110	0000003
+#define  B134	0000004
+#define  B150	0000005
+#define  B200	0000006
+#define  B300	0000007
+#define  B600	0000010
+#define  B1200	0000011
+#define  B1800	0000012
+#define  B2400	0000013
+#define  B4800	0000014
+#define  B9600	0000015
+#define  B19200	0000016
+#define  B38400	0000017
+#define EXTA B19200
+#define EXTB B38400
+#define CSIZE	0000060
+#define   CS5	0000000
+#define   CS6	0000020
+#define   CS7	0000040
+#define   CS8	0000060
+#define CSTOPB	0000100
+#define CREAD	0000200
+#define PARENB	0000400
+#define PARODD	0001000
+#define HUPCL	0002000
+#define CLOCAL	0004000
+#define CBAUDEX 0010000
+#define    B57600 0010001
+#define   B115200 0010002
+#define   B230400 0010003
+#define   B460800 0010004
+#define   B500000 0010005
+#define   B576000 0010006
+#define   B921600 0010007
+#define  B1000000 0010010
+#define  B1152000 0010011
+#define  B1500000 0010012
+#define  B2000000 0010013
+#define  B2500000 0010014
+#define  B3000000 0010015
+#define  B3500000 0010016
+#define  B4000000 0010017
+#define CIBAUD	  002003600000	/* input baud rate (not used) */
+#define CMSPAR	  010000000000		/* mark or space (stick) parity */
+#define CRTSCTS	  020000000000		/* flow control */
+
+/* c_lflag bits */
+#define ISIG	0000001
+#define ICANON	0000002
+#define XCASE	0000004
+#define ECHO	0000010
+#define ECHOE	0000020
+#define ECHOK	0000040
+#define ECHONL	0000100
+#define NOFLSH	0000200
+#define TOSTOP	0000400
+#define ECHOCTL	0001000
+#define ECHOPRT	0002000
+#define ECHOKE	0004000
+#define FLUSHO	0010000
+#define PENDIN	0040000
+#define IEXTEN	0100000
+
+/* tcflow() and TCXONC use these */
+#define	TCOOFF		0
+#define	TCOON		1
+#define	TCIOFF		2
+#define	TCION		3
+
+/* tcflush() and TCFLSH use these */
+#define	TCIFLUSH	0
+#define	TCOFLUSH	1
+#define	TCIOFLUSH	2
+
+/* tcsetattr uses these */
+#define	TCSANOW		0
+#define	TCSADRAIN	1
+#define	TCSAFLUSH	2
+
+#endif /* __ASM_AVR32_TERMBITS_H */
diff --git a/include/asm-avr32/termios.h b/include/asm-avr32/termios.h
new file mode 100644
index 000000000000..615bc0639e5c
--- /dev/null
+++ b/include/asm-avr32/termios.h
@@ -0,0 +1,80 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TERMIOS_H
+#define __ASM_AVR32_TERMIOS_H
+
+#include <asm/termbits.h>
+#include <asm/ioctls.h>
+
+struct winsize {
+	unsigned short ws_row;
+	unsigned short ws_col;
+	unsigned short ws_xpixel;
+	unsigned short ws_ypixel;
+};
+
+#define NCC 8
+struct termio {
+	unsigned short c_iflag;		/* input mode flags */
+	unsigned short c_oflag;		/* output mode flags */
+	unsigned short c_cflag;		/* control mode flags */
+	unsigned short c_lflag;		/* local mode flags */
+	unsigned char c_line;		/* line discipline */
+	unsigned char c_cc[NCC];	/* control characters */
+};
+
+/* modem lines */
+#define TIOCM_LE	0x001
+#define TIOCM_DTR	0x002
+#define TIOCM_RTS	0x004
+#define TIOCM_ST	0x008
+#define TIOCM_SR	0x010
+#define TIOCM_CTS	0x020
+#define TIOCM_CAR	0x040
+#define TIOCM_RNG	0x080
+#define TIOCM_DSR	0x100
+#define TIOCM_CD	TIOCM_CAR
+#define TIOCM_RI	TIOCM_RNG
+#define TIOCM_OUT1	0x2000
+#define TIOCM_OUT2	0x4000
+#define TIOCM_LOOP	0x8000
+
+/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
+
+/* line disciplines */
+#define N_TTY		0
+#define N_SLIP		1
+#define N_MOUSE		2
+#define N_PPP		3
+#define N_STRIP		4
+#define N_AX25		5
+#define N_X25		6	/* X.25 async */
+#define N_6PACK		7
+#define N_MASC		8	/* Reserved for Mobitex module <kaz@cafe.net> */
+#define N_R3964		9	/* Reserved for Simatic R3964 module */
+#define N_PROFIBUS_FDL	10	/* Reserved for Profibus <Dave@mvhi.com> */
+#define N_IRDA		11	/* Linux IR - http://irda.sourceforge.net/ */
+#define N_SMSBLOCK	12	/* SMS block mode - for talking to GSM data cards about SMS messages */
+#define N_HDLC		13	/* synchronous HDLC */
+#define N_SYNC_PPP	14	/* synchronous PPP */
+#define N_HCI		15  /* Bluetooth HCI UART */
+
+#ifdef __KERNEL__
+/*	intr=^C		quit=^\		erase=del	kill=^U
+	eof=^D		vtime=\0	vmin=\1		sxtc=\0
+	start=^Q	stop=^S		susp=^Z		eol=\0
+	reprint=^R	discard=^U	werase=^W	lnext=^V
+	eol2=\0
+*/
+#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
+
+#include <asm-generic/termios.h>
+
+#endif	/* __KERNEL__ */
+
+#endif	/* __ASM_AVR32_TERMIOS_H */
diff --git a/include/asm-avr32/thread_info.h b/include/asm-avr32/thread_info.h
new file mode 100644
index 000000000000..d1f5b35ebd54
--- /dev/null
+++ b/include/asm-avr32/thread_info.h
@@ -0,0 +1,106 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_THREAD_INFO_H
+#define __ASM_AVR32_THREAD_INFO_H
+
+#include <asm/page.h>
+
+#define THREAD_SIZE_ORDER	1
+#define THREAD_SIZE		(PAGE_SIZE << THREAD_SIZE_ORDER)
+
+#ifndef __ASSEMBLY__
+#include <asm/types.h>
+
+struct task_struct;
+struct exec_domain;
+
+struct thread_info {
+	struct task_struct	*task;		/* main task structure */
+	struct exec_domain	*exec_domain;	/* execution domain */
+	unsigned long		flags;		/* low level flags */
+	__u32			cpu;
+	__s32			preempt_count;	/* 0 => preemptable, <0 => BUG */
+	struct restart_block	restart_block;
+	__u8			supervisor_stack[0];
+};
+
+#define INIT_THREAD_INFO(tsk)						\
+{									\
+	.task		= &tsk,						\
+	.exec_domain	= &default_exec_domain,				\
+	.flags		= 0,						\
+	.cpu		= 0,						\
+	.preempt_count	= 1,						\
+	.restart_block	= {						\
+		.fn	= do_no_restart_syscall				\
+	}								\
+}
+
+#define init_thread_info	(init_thread_union.thread_info)
+#define init_stack		(init_thread_union.stack)
+
+/*
+ * Get the thread information struct from C.
+ * We do the usual trick and use the lower end of the stack for this
+ */
+static inline struct thread_info *current_thread_info(void)
+{
+	unsigned long addr = ~(THREAD_SIZE - 1);
+
+	asm("and %0, sp" : "=r"(addr) : "0"(addr));
+	return (struct thread_info *)addr;
+}
+
+/* thread information allocation */
+#define alloc_thread_info(ti) \
+	((struct thread_info *) __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER))
+#define free_thread_info(ti) free_pages((unsigned long)(ti), 1)
+#define get_thread_info(ti) get_task_struct((ti)->task)
+#define put_thread_info(ti) put_task_struct((ti)->task)
+
+#endif /* !__ASSEMBLY__ */
+
+#define PREEMPT_ACTIVE		0x40000000
+
+/*
+ * Thread information flags
+ * - these are process state flags that various assembly files may need to access
+ * - pending work-to-be-done flags are in LSW
+ * - other flags in MSW
+ */
+#define TIF_SYSCALL_TRACE       0       /* syscall trace active */
+#define TIF_NOTIFY_RESUME       1       /* resumption notification requested */
+#define TIF_SIGPENDING          2       /* signal pending */
+#define TIF_NEED_RESCHED        3       /* rescheduling necessary */
+#define TIF_POLLING_NRFLAG      4       /* true if poll_idle() is polling
+					   TIF_NEED_RESCHED */
+#define TIF_BREAKPOINT		5	/* true if we should break after return */
+#define TIF_SINGLE_STEP		6	/* single step after next break */
+#define TIF_MEMDIE		7
+#define TIF_RESTORE_SIGMASK	8	/* restore signal mask in do_signal */
+#define TIF_USERSPACE		31      /* true if FS sets userspace */
+
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
+#define _TIF_BREAKPOINT		(1 << TIF_BREAKPOINT)
+#define _TIF_SINGLE_STEP	(1 << TIF_SINGLE_STEP)
+#define _TIF_MEMDIE		(1 << TIF_MEMDIE)
+#define _TIF_RESTORE_SIGMASK	(1 << TIF_RESTORE_SIGMASK)
+
+/* XXX: These two masks must never span more than 16 bits! */
+/* work to do on interrupt/exception return */
+#define _TIF_WORK_MASK		0x0000013e
+/* work to do on any return to userspace */
+#define _TIF_ALLWORK_MASK	0x0000013f
+/* work to do on return from debug mode */
+#define _TIF_DBGWORK_MASK	0x0000017e
+
+#endif /* __ASM_AVR32_THREAD_INFO_H */
diff --git a/include/asm-avr32/timex.h b/include/asm-avr32/timex.h
new file mode 100644
index 000000000000..5e44ecb3ce0c
--- /dev/null
+++ b/include/asm-avr32/timex.h
@@ -0,0 +1,40 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TIMEX_H
+#define __ASM_AVR32_TIMEX_H
+
+/*
+ * This is the frequency of the timer used for Linux's timer interrupt.
+ * The value should be defined as accurate as possible or under certain
+ * circumstances Linux timekeeping might become inaccurate or fail.
+ *
+ * For many system the exact clockrate of the timer isn't known but due to
+ * the way this value is used we can get away with a wrong value as long
+ * as this value is:
+ *
+ *  - a multiple of HZ
+ *  - a divisor of the actual rate
+ *
+ * 500000 is a good such cheat value.
+ *
+ * The obscure number 1193182 is the same as used by the original i8254
+ * time in legacy PC hardware; the chip is never found in AVR32 systems.
+ */
+#define CLOCK_TICK_RATE		500000	/* Underlying HZ */
+
+typedef unsigned long cycles_t;
+
+static inline cycles_t get_cycles (void)
+{
+	return 0;
+}
+
+extern int read_current_timer(unsigned long *timer_value);
+#define ARCH_HAS_READ_CURRENT_TIMER	1
+
+#endif /* __ASM_AVR32_TIMEX_H */
diff --git a/include/asm-avr32/tlb.h b/include/asm-avr32/tlb.h
new file mode 100644
index 000000000000..5c55f9ce7c7d
--- /dev/null
+++ b/include/asm-avr32/tlb.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TLB_H
+#define __ASM_AVR32_TLB_H
+
+#define tlb_start_vma(tlb, vma) \
+	flush_cache_range(vma, vma->vm_start, vma->vm_end)
+
+#define tlb_end_vma(tlb, vma) \
+	flush_tlb_range(vma, vma->vm_start, vma->vm_end)
+
+#define __tlb_remove_tlb_entry(tlb, pte, address) do { } while(0)
+
+/*
+ * Flush whole TLB for MM
+ */
+#define tlb_flush(tlb) flush_tlb_mm((tlb)->mm)
+
+#include <asm-generic/tlb.h>
+
+/*
+ * For debugging purposes
+ */
+extern void show_dtlb_entry(unsigned int index);
+extern void dump_dtlb(void);
+
+#endif /* __ASM_AVR32_TLB_H */
diff --git a/include/asm-avr32/tlbflush.h b/include/asm-avr32/tlbflush.h
new file mode 100644
index 000000000000..730e268f81f3
--- /dev/null
+++ b/include/asm-avr32/tlbflush.h
@@ -0,0 +1,40 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TLBFLUSH_H
+#define __ASM_AVR32_TLBFLUSH_H
+
+#include <asm/mmu.h>
+
+/*
+ * TLB flushing:
+ *
+ *  - flush_tlb() flushes the current mm struct TLBs
+ *  - flush_tlb_all() flushes all processes' TLB entries
+ *  - flush_tlb_mm(mm) flushes the specified mm context TLBs
+ *  - flush_tlb_page(vma, vmaddr) flushes one page
+ *  - flush_tlb_range(vma, start, end) flushes a range of pages
+ *  - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
+ *  - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
+ */
+extern void flush_tlb(void);
+extern void flush_tlb_all(void);
+extern void flush_tlb_mm(struct mm_struct *mm);
+extern void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+			    unsigned long end);
+extern void flush_tlb_page(struct vm_area_struct *vma, unsigned long page);
+extern void __flush_tlb_page(unsigned long asid, unsigned long page);
+
+static inline void flush_tlb_pgtables(struct mm_struct *mm,
+				      unsigned long start, unsigned long end)
+{
+	/* Nothing to do */
+}
+
+extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
+
+#endif /* __ASM_AVR32_TLBFLUSH_H */
diff --git a/include/asm-avr32/topology.h b/include/asm-avr32/topology.h
new file mode 100644
index 000000000000..5b766cbb4806
--- /dev/null
+++ b/include/asm-avr32/topology.h
@@ -0,0 +1,6 @@
+#ifndef __ASM_AVR32_TOPOLOGY_H
+#define __ASM_AVR32_TOPOLOGY_H
+
+#include <asm-generic/topology.h>
+
+#endif /* __ASM_AVR32_TOPOLOGY_H */
diff --git a/include/asm-avr32/traps.h b/include/asm-avr32/traps.h
new file mode 100644
index 000000000000..6a8fb944f414
--- /dev/null
+++ b/include/asm-avr32/traps.h
@@ -0,0 +1,23 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TRAPS_H
+#define __ASM_AVR32_TRAPS_H
+
+#include <linux/list.h>
+
+struct undef_hook {
+	struct list_head node;
+	u32 insn_mask;
+	u32 insn_val;
+	int (*fn)(struct pt_regs *regs, u32 insn);
+};
+
+void register_undef_hook(struct undef_hook *hook);
+void unregister_undef_hook(struct undef_hook *hook);
+
+#endif /* __ASM_AVR32_TRAPS_H */
diff --git a/include/asm-avr32/types.h b/include/asm-avr32/types.h
new file mode 100644
index 000000000000..3f47db9675af
--- /dev/null
+++ b/include/asm-avr32/types.h
@@ -0,0 +1,70 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_TYPES_H
+#define __ASM_AVR32_TYPES_H
+
+#ifndef __ASSEMBLY__
+
+typedef unsigned short umode_t;
+
+/*
+ * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the
+ * header files exported to user space
+ */
+typedef __signed__ char __s8;
+typedef unsigned char __u8;
+
+typedef __signed__ short __s16;
+typedef unsigned short __u16;
+
+typedef __signed__ int __s32;
+typedef unsigned int __u32;
+
+#if defined(__GNUC__) && !defined(__STRICT_ANSI__)
+typedef __signed__ long long __s64;
+typedef unsigned long long __u64;
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+/*
+ * These aren't exported outside the kernel to avoid name space clashes
+ */
+#ifdef __KERNEL__
+
+#define BITS_PER_LONG 32
+
+#ifndef __ASSEMBLY__
+
+typedef signed char s8;
+typedef unsigned char u8;
+
+typedef signed short s16;
+typedef unsigned short u16;
+
+typedef signed int s32;
+typedef unsigned int u32;
+
+typedef signed long long s64;
+typedef unsigned long long u64;
+
+/* Dma addresses are 32-bits wide.  */
+
+typedef u32 dma_addr_t;
+
+#ifdef CONFIG_LBD
+typedef u64 sector_t;
+#define HAVE_SECTOR_T
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __KERNEL__ */
+
+
+#endif /* __ASM_AVR32_TYPES_H */
diff --git a/include/asm-avr32/uaccess.h b/include/asm-avr32/uaccess.h
new file mode 100644
index 000000000000..821deb5a9d28
--- /dev/null
+++ b/include/asm-avr32/uaccess.h
@@ -0,0 +1,335 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_UACCESS_H
+#define __ASM_AVR32_UACCESS_H
+
+#include <linux/errno.h>
+#include <linux/sched.h>
+
+#define VERIFY_READ	0
+#define VERIFY_WRITE	1
+
+typedef struct {
+	unsigned int is_user_space;
+} mm_segment_t;
+
+/*
+ * The fs value determines whether argument validity checking should be
+ * performed or not.  If get_fs() == USER_DS, checking is performed, with
+ * get_fs() == KERNEL_DS, checking is bypassed.
+ *
+ * For historical reasons (Data Segment Register?), these macros are misnamed.
+ */
+#define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
+#define segment_eq(a,b)	((a).is_user_space == (b).is_user_space)
+
+#define USER_ADDR_LIMIT 0x80000000
+
+#define KERNEL_DS	MAKE_MM_SEG(0)
+#define USER_DS		MAKE_MM_SEG(1)
+
+#define get_ds()	(KERNEL_DS)
+
+static inline mm_segment_t get_fs(void)
+{
+	return MAKE_MM_SEG(test_thread_flag(TIF_USERSPACE));
+}
+
+static inline void set_fs(mm_segment_t s)
+{
+	if (s.is_user_space)
+		set_thread_flag(TIF_USERSPACE);
+	else
+		clear_thread_flag(TIF_USERSPACE);
+}
+
+/*
+ * Test whether a block of memory is a valid user space address.
+ * Returns 0 if the range is valid, nonzero otherwise.
+ *
+ * We do the following checks:
+ *   1. Is the access from kernel space?
+ *   2. Does (addr + size) set the carry bit?
+ *   3. Is (addr + size) a negative number (i.e. >= 0x80000000)?
+ *
+ * If yes on the first check, access is granted.
+ * If no on any of the others, access is denied.
+ */
+#define __range_ok(addr, size)						\
+	(test_thread_flag(TIF_USERSPACE)				\
+	 && (((unsigned long)(addr) >= 0x80000000)			\
+	     || ((unsigned long)(size) > 0x80000000)			\
+	     || (((unsigned long)(addr) + (unsigned long)(size)) > 0x80000000)))
+
+#define access_ok(type, addr, size) (likely(__range_ok(addr, size) == 0))
+
+static inline int
+verify_area(int type, const void __user *addr, unsigned long size)
+{
+	return access_ok(type, addr, size) ? 0 : -EFAULT;
+}
+
+/* Generic arbitrary sized copy. Return the number of bytes NOT copied */
+extern __kernel_size_t __copy_user(void *to, const void *from,
+				   __kernel_size_t n);
+
+extern __kernel_size_t copy_to_user(void __user *to, const void *from,
+				    __kernel_size_t n);
+extern __kernel_size_t copy_from_user(void *to, const void __user *from,
+				      __kernel_size_t n);
+
+static inline __kernel_size_t __copy_to_user(void __user *to, const void *from,
+					     __kernel_size_t n)
+{
+	return __copy_user((void __force *)to, from, n);
+}
+static inline __kernel_size_t __copy_from_user(void *to,
+					       const void __user *from,
+					       __kernel_size_t n)
+{
+	return __copy_user(to, (const void __force *)from, n);
+}
+
+#define __copy_to_user_inatomic __copy_to_user
+#define __copy_from_user_inatomic __copy_from_user
+
+/*
+ * put_user: - Write a simple value into user space.
+ * @x:   Value to copy to user space.
+ * @ptr: Destination address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple value from kernel space to user
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and @x must be assignable
+ * to the result of dereferencing @ptr.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ */
+#define put_user(x,ptr)	\
+	__put_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * get_user: - Get a simple variable from user space.
+ * @x:   Variable to store result.
+ * @ptr: Source address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple variable from user space to kernel
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and the result of
+ * dereferencing @ptr must be assignable to @x without a cast.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ * On error, the variable @x is set to zero.
+ */
+#define get_user(x,ptr) \
+	__get_user_check((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * __put_user: - Write a simple value into user space, with less checking.
+ * @x:   Value to copy to user space.
+ * @ptr: Destination address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple value from kernel space to user
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and @x must be assignable
+ * to the result of dereferencing @ptr.
+ *
+ * Caller must check the pointer with access_ok() before calling this
+ * function.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ */
+#define __put_user(x,ptr) \
+	__put_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+/*
+ * __get_user: - Get a simple variable from user space, with less checking.
+ * @x:   Variable to store result.
+ * @ptr: Source address, in user space.
+ *
+ * Context: User context only.  This function may sleep.
+ *
+ * This macro copies a single simple variable from user space to kernel
+ * space.  It supports simple types like char and int, but not larger
+ * data types like structures or arrays.
+ *
+ * @ptr must have pointer-to-simple-variable type, and the result of
+ * dereferencing @ptr must be assignable to @x without a cast.
+ *
+ * Caller must check the pointer with access_ok() before calling this
+ * function.
+ *
+ * Returns zero on success, or -EFAULT on error.
+ * On error, the variable @x is set to zero.
+ */
+#define __get_user(x,ptr) \
+	__get_user_nocheck((x),(ptr),sizeof(*(ptr)))
+
+extern int __get_user_bad(void);
+extern int __put_user_bad(void);
+
+#define __get_user_nocheck(x, ptr, size)				\
+({									\
+	typeof(*(ptr)) __gu_val = (typeof(*(ptr)) __force)0;		\
+	int __gu_err = 0;						\
+									\
+	switch (size) {							\
+	case 1: __get_user_asm("ub", __gu_val, ptr, __gu_err); break;	\
+	case 2: __get_user_asm("uh", __gu_val, ptr, __gu_err); break;	\
+	case 4: __get_user_asm("w", __gu_val, ptr, __gu_err); break;	\
+	case 8: __get_user_asm("d", __gu_val, ptr, __gu_err); break;	\
+	default: __gu_err = __get_user_bad(); break;			\
+	}								\
+									\
+	x = __gu_val;							\
+	__gu_err;							\
+})
+
+#define __get_user_check(x, ptr, size)					\
+({									\
+	typeof(*(ptr)) __gu_val = (typeof(*(ptr)) __force)0;		\
+	const typeof(*(ptr)) __user * __gu_addr = (ptr);		\
+	int __gu_err = 0;						\
+									\
+	if (access_ok(VERIFY_READ, __gu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__get_user_asm("ub", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 2:							\
+			__get_user_asm("uh", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 4:							\
+			__get_user_asm("w", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		case 8:							\
+			__get_user_asm("d", __gu_val, __gu_addr,	\
+				       __gu_err);			\
+			break;						\
+		default:						\
+			__gu_err = __get_user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__gu_err = -EFAULT;					\
+	}								\
+	x = __gu_val;							\
+	__gu_err;							\
+})
+
+#define __get_user_asm(suffix, __gu_val, ptr, __gu_err)			\
+	asm volatile(							\
+		"1:	ld." suffix "	%1, %3			\n"	\
+		"2:						\n"	\
+		"	.section .fixup, \"ax\"			\n"	\
+		"3:	mov	%0, %4				\n"	\
+		"	rjmp	2b				\n"	\
+		"	.previous				\n"	\
+		"	.section __ex_table, \"a\"		\n"	\
+		"	.long	1b, 3b				\n"	\
+		"	.previous				\n"	\
+		: "=r"(__gu_err), "=r"(__gu_val)			\
+		: "0"(__gu_err), "m"(*(ptr)), "i"(-EFAULT))
+
+#define __put_user_nocheck(x, ptr, size)				\
+({									\
+	typeof(*(ptr)) __pu_val;					\
+	int __pu_err = 0;						\
+									\
+	__pu_val = (x);							\
+	switch (size) {							\
+	case 1: __put_user_asm("b", ptr, __pu_val, __pu_err); break;	\
+	case 2: __put_user_asm("h", ptr, __pu_val, __pu_err); break;	\
+	case 4: __put_user_asm("w", ptr, __pu_val, __pu_err); break;	\
+	case 8: __put_user_asm("d", ptr, __pu_val, __pu_err); break;	\
+	default: __pu_err = __put_user_bad(); break;			\
+	}								\
+	__pu_err;							\
+})
+
+#define __put_user_check(x, ptr, size)					\
+({									\
+	typeof(*(ptr)) __pu_val;					\
+	typeof(*(ptr)) __user *__pu_addr = (ptr);			\
+	int __pu_err = 0;						\
+									\
+	__pu_val = (x);							\
+	if (access_ok(VERIFY_WRITE, __pu_addr, size)) {			\
+		switch (size) {						\
+		case 1:							\
+			__put_user_asm("b", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 2:							\
+			__put_user_asm("h", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 4:							\
+			__put_user_asm("w", __pu_addr, __pu_val,	\
+				       __pu_err);			\
+			break;						\
+		case 8:							\
+			__put_user_asm("d", __pu_addr, __pu_val,		\
+				       __pu_err);			\
+			break;						\
+		default:						\
+			__pu_err = __put_user_bad();			\
+			break;						\
+		}							\
+	} else {							\
+		__pu_err = -EFAULT;					\
+	}								\
+	__pu_err;							\
+})
+
+#define __put_user_asm(suffix, ptr, __pu_val, __gu_err)			\
+	asm volatile(							\
+		"1:	st." suffix "	%1, %3			\n"	\
+		"2:						\n"	\
+		"	.section .fixup, \"ax\"			\n"	\
+		"3:	mov	%0, %4				\n"	\
+		"	rjmp	2b				\n"	\
+		"	.previous				\n"	\
+		"	.section __ex_table, \"a\"		\n"	\
+		"	.long	1b, 3b				\n"	\
+		"	.previous				\n"	\
+		: "=r"(__gu_err), "=m"(*(ptr))				\
+		: "0"(__gu_err), "r"(__pu_val), "i"(-EFAULT))
+
+extern __kernel_size_t clear_user(void __user *addr, __kernel_size_t size);
+extern __kernel_size_t __clear_user(void __user *addr, __kernel_size_t size);
+
+extern long strncpy_from_user(char *dst, const char __user *src, long count);
+extern long __strncpy_from_user(char *dst, const char __user *src, long count);
+
+extern long strnlen_user(const char __user *__s, long __n);
+extern long __strnlen_user(const char __user *__s, long __n);
+
+#define strlen_user(s) strnlen_user(s, ~0UL >> 1)
+
+struct exception_table_entry
+{
+	unsigned long insn, fixup;
+};
+
+#endif /* __ASM_AVR32_UACCESS_H */
diff --git a/include/asm-avr32/ucontext.h b/include/asm-avr32/ucontext.h
new file mode 100644
index 000000000000..ac7259c2a799
--- /dev/null
+++ b/include/asm-avr32/ucontext.h
@@ -0,0 +1,12 @@
+#ifndef __ASM_AVR32_UCONTEXT_H
+#define __ASM_AVR32_UCONTEXT_H
+
+struct ucontext {
+	unsigned long		uc_flags;
+	struct ucontext	*	uc_link;
+	stack_t			uc_stack;
+	struct sigcontext	uc_mcontext;
+	sigset_t		uc_sigmask;
+};
+
+#endif /* __ASM_AVR32_UCONTEXT_H */
diff --git a/include/asm-avr32/unaligned.h b/include/asm-avr32/unaligned.h
new file mode 100644
index 000000000000..3042723fcbfd
--- /dev/null
+++ b/include/asm-avr32/unaligned.h
@@ -0,0 +1,25 @@
+#ifndef __ASM_AVR32_UNALIGNED_H
+#define __ASM_AVR32_UNALIGNED_H
+
+/*
+ * AVR32 can handle some unaligned accesses, depending on the
+ * implementation.  The AVR32 AP implementation can handle unaligned
+ * words, but halfwords must be halfword-aligned, and doublewords must
+ * be word-aligned.
+ *
+ * TODO: Make all this CPU-specific and optimize.
+ */
+
+#include <linux/string.h>
+
+/* Use memmove here, so gcc does not insert a __builtin_memcpy. */
+
+#define get_unaligned(ptr) \
+  ({ __typeof__(*(ptr)) __tmp; memmove(&__tmp, (ptr), sizeof(*(ptr))); __tmp; })
+
+#define put_unaligned(val, ptr)				\
+  ({ __typeof__(*(ptr)) __tmp = (val);			\
+     memmove((ptr), &__tmp, sizeof(*(ptr)));		\
+     (void)0; })
+
+#endif /* __ASM_AVR32_UNALIGNED_H */
diff --git a/include/asm-avr32/unistd.h b/include/asm-avr32/unistd.h
new file mode 100644
index 000000000000..1f528f92690d
--- /dev/null
+++ b/include/asm-avr32/unistd.h
@@ -0,0 +1,387 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_UNISTD_H
+#define __ASM_AVR32_UNISTD_H
+
+/*
+ * This file contains the system call numbers.
+ */
+
+#define __NR_restart_syscall      0
+#define __NR_exit		  1
+#define __NR_fork		  2
+#define __NR_read		  3
+#define __NR_write		  4
+#define __NR_open		  5
+#define __NR_close		  6
+#define __NR_umask		  7
+#define __NR_creat		  8
+#define __NR_link		  9
+#define __NR_unlink		 10
+#define __NR_execve		 11
+#define __NR_chdir		 12
+#define __NR_time		 13
+#define __NR_mknod		 14
+#define __NR_chmod		 15
+#define __NR_chown		 16
+#define __NR_lchown		 17
+#define __NR_lseek		 18
+#define __NR__llseek		 19
+#define __NR_getpid		 20
+#define __NR_mount		 21
+#define __NR_umount2		 22
+#define __NR_setuid		 23
+#define __NR_getuid		 24
+#define __NR_stime		 25
+#define __NR_ptrace		 26
+#define __NR_alarm		 27
+#define __NR_pause		 28
+#define __NR_utime		 29
+#define __NR_stat		 30
+#define __NR_fstat		 31
+#define __NR_lstat		 32
+#define __NR_access		 33
+#define __NR_chroot		 34
+#define __NR_sync		 35
+#define __NR_fsync		 36
+#define __NR_kill		 37
+#define __NR_rename		 38
+#define __NR_mkdir		 39
+#define __NR_rmdir		 40
+#define __NR_dup		 41
+#define __NR_pipe		 42
+#define __NR_times		 43
+#define __NR_clone		 44
+#define __NR_brk		 45
+#define __NR_setgid		 46
+#define __NR_getgid		 47
+#define __NR_getcwd		 48
+#define __NR_geteuid		 49
+#define __NR_getegid		 50
+#define __NR_acct		 51
+#define __NR_setfsuid		 52
+#define __NR_setfsgid		 53
+#define __NR_ioctl		 54
+#define __NR_fcntl		 55
+#define __NR_setpgid		 56
+#define __NR_mremap		 57
+#define __NR_setresuid		 58
+#define __NR_getresuid		 59
+#define __NR_setreuid		 60
+#define __NR_setregid		 61
+#define __NR_ustat		 62
+#define __NR_dup2		 63
+#define __NR_getppid		 64
+#define __NR_getpgrp		 65
+#define __NR_setsid		 66
+#define __NR_rt_sigaction	 67
+#define __NR_rt_sigreturn	 68
+#define __NR_rt_sigprocmask	 69
+#define __NR_rt_sigpending	 70
+#define __NR_rt_sigtimedwait	 71
+#define __NR_rt_sigqueueinfo	 72
+#define __NR_rt_sigsuspend	 73
+#define __NR_sethostname	 74
+#define __NR_setrlimit		 75
+#define __NR_getrlimit		 76	/* SuS compliant getrlimit */
+#define __NR_getrusage		 77
+#define __NR_gettimeofday	 78
+#define __NR_settimeofday	 79
+#define __NR_getgroups		 80
+#define __NR_setgroups		 81
+#define __NR_select		 82
+#define __NR_symlink		 83
+#define __NR_fchdir		 84
+#define __NR_readlink		 85
+#define __NR_pread		 86
+#define __NR_pwrite		 87
+#define __NR_swapon		 88
+#define __NR_reboot		 89
+#define __NR_mmap2		 90
+#define __NR_munmap		 91
+#define __NR_truncate		 92
+#define __NR_ftruncate		 93
+#define __NR_fchmod		 94
+#define __NR_fchown		 95
+#define __NR_getpriority	 96
+#define __NR_setpriority	 97
+#define __NR_wait4		 98
+#define __NR_statfs		 99
+#define __NR_fstatfs		100
+#define __NR_vhangup		101
+#define __NR_sigaltstack	102
+#define __NR_syslog		103
+#define __NR_setitimer		104
+#define __NR_getitimer		105
+#define __NR_swapoff		106
+#define __NR_sysinfo		107
+#define __NR_ipc		108
+#define __NR_sendfile		109
+#define __NR_setdomainname	110
+#define __NR_uname		111
+#define __NR_adjtimex		112
+#define __NR_mprotect		113
+#define __NR_vfork		114
+#define __NR_init_module	115
+#define __NR_delete_module	116
+#define __NR_quotactl		117
+#define __NR_getpgid		118
+#define __NR_bdflush		119
+#define __NR_sysfs		120
+#define __NR_personality	121
+#define __NR_afs_syscall	122 /* Syscall for Andrew File System */
+#define __NR_getdents		123
+#define __NR_flock		124
+#define __NR_msync		125
+#define __NR_readv		126
+#define __NR_writev		127
+#define __NR_getsid		128
+#define __NR_fdatasync		129
+#define __NR__sysctl		130
+#define __NR_mlock		131
+#define __NR_munlock		132
+#define __NR_mlockall		133
+#define __NR_munlockall		134
+#define __NR_sched_setparam		135
+#define __NR_sched_getparam		136
+#define __NR_sched_setscheduler		137
+#define __NR_sched_getscheduler		138
+#define __NR_sched_yield		139
+#define __NR_sched_get_priority_max	140
+#define __NR_sched_get_priority_min	141
+#define __NR_sched_rr_get_interval	142
+#define __NR_nanosleep		143
+#define __NR_poll		144
+#define __NR_nfsservctl		145
+#define __NR_setresgid		146
+#define __NR_getresgid		147
+#define __NR_prctl              148
+#define __NR_socket		149
+#define __NR_bind		150
+#define __NR_connect		151
+#define __NR_listen		152
+#define __NR_accept		153
+#define __NR_getsockname	154
+#define __NR_getpeername	155
+#define __NR_socketpair		156
+#define __NR_send		157
+#define __NR_recv		158
+#define __NR_sendto		159
+#define __NR_recvfrom		160
+#define __NR_shutdown		161
+#define __NR_setsockopt		162
+#define __NR_getsockopt		163
+#define __NR_sendmsg		164
+#define __NR_recvmsg		165
+#define __NR_truncate64		166
+#define __NR_ftruncate64	167
+#define __NR_stat64		168
+#define __NR_lstat64		169
+#define __NR_fstat64		170
+#define __NR_pivot_root		171
+#define __NR_mincore		172
+#define __NR_madvise		173
+#define __NR_getdents64		174
+#define __NR_fcntl64		175
+#define __NR_gettid		176
+#define __NR_readahead		177
+#define __NR_setxattr		178
+#define __NR_lsetxattr		179
+#define __NR_fsetxattr		180
+#define __NR_getxattr		181
+#define __NR_lgetxattr		182
+#define __NR_fgetxattr		183
+#define __NR_listxattr		184
+#define __NR_llistxattr		185
+#define __NR_flistxattr		186
+#define __NR_removexattr	187
+#define __NR_lremovexattr	188
+#define __NR_fremovexattr	189
+#define __NR_tkill		190
+#define __NR_sendfile64		191
+#define __NR_futex		192
+#define __NR_sched_setaffinity	193
+#define __NR_sched_getaffinity	194
+#define __NR_capget		195
+#define __NR_capset		196
+#define __NR_io_setup		197
+#define __NR_io_destroy		198
+#define __NR_io_getevents	199
+#define __NR_io_submit		200
+#define __NR_io_cancel		201
+#define __NR_fadvise64		202
+#define __NR_exit_group		203
+#define __NR_lookup_dcookie	204
+#define __NR_epoll_create	205
+#define __NR_epoll_ctl		206
+#define __NR_epoll_wait		207
+#define __NR_remap_file_pages	208
+#define __NR_set_tid_address	209
+
+#define __NR_timer_create	210
+#define __NR_timer_settime	211
+#define __NR_timer_gettime	212
+#define __NR_timer_getoverrun	213
+#define __NR_timer_delete	214
+#define __NR_clock_settime	215
+#define __NR_clock_gettime	216
+#define __NR_clock_getres	217
+#define __NR_clock_nanosleep	218
+#define __NR_statfs64		219
+#define __NR_fstatfs64		220
+#define __NR_tgkill		221
+				/* 222 reserved for tux */
+#define __NR_utimes		223
+#define __NR_fadvise64_64	224
+
+#define __NR_cacheflush		225
+
+#define __NR_vserver		226
+#define __NR_mq_open		227
+#define __NR_mq_unlink		228
+#define __NR_mq_timedsend	229
+#define __NR_mq_timedreceive	230
+#define __NR_mq_notify		231
+#define __NR_mq_getsetattr	232
+#define __NR_kexec_load		233
+#define __NR_waitid		234
+#define __NR_add_key		235
+#define __NR_request_key	236
+#define __NR_keyctl		237
+#define __NR_ioprio_set		238
+#define __NR_ioprio_get		239
+#define __NR_inotify_init	240
+#define __NR_inotify_add_watch	241
+#define __NR_inotify_rm_watch	242
+#define __NR_openat		243
+#define __NR_mkdirat		244
+#define __NR_mknodat		245
+#define __NR_fchownat		246
+#define __NR_futimesat		247
+#define __NR_fstatat64		248
+#define __NR_unlinkat		249
+#define __NR_renameat		250
+#define __NR_linkat		251
+#define __NR_symlinkat		252
+#define __NR_readlinkat		253
+#define __NR_fchmodat		254
+#define __NR_faccessat		255
+#define __NR_pselect6		256
+#define __NR_ppoll		257
+#define __NR_unshare		258
+#define __NR_set_robust_list	259
+#define __NR_get_robust_list	260
+#define __NR_splice		261
+#define __NR_sync_file_range	262
+#define __NR_tee		263
+#define __NR_vmsplice		264
+
+#define NR_syscalls		265
+
+
+/*
+ * AVR32 calling convention for system calls:
+ *   - System call number in r8
+ *   - Parameters in r12 and downwards to r9 as well as r6 and r5.
+ *   - Return value in r12
+ */
+
+/*
+ * user-visible error numbers are in the range -1 - -124: see
+ * <asm-generic/errno.h>
+ */
+
+#define __syscall_return(type, res) do {				\
+		if ((unsigned long)(res) >= (unsigned long)(-125)) {	\
+			errno = -(res);					\
+			res = -1;					\
+		}							\
+		return (type) (res);					\
+	} while (0)
+
+#ifdef __KERNEL__
+#define __ARCH_WANT_IPC_PARSE_VERSION
+#define __ARCH_WANT_STAT64
+#define __ARCH_WANT_SYS_ALARM
+#define __ARCH_WANT_SYS_GETHOSTNAME
+#define __ARCH_WANT_SYS_PAUSE
+#define __ARCH_WANT_SYS_TIME
+#define __ARCH_WANT_SYS_UTIME
+#define __ARCH_WANT_SYS_WAITPID
+#define __ARCH_WANT_SYS_FADVISE64
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_LLSEEK
+#define __ARCH_WANT_SYS_GETPGRP
+#define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
+#endif
+
+#if defined(__KERNEL_SYSCALLS__) || defined(__CHECKER__)
+
+#include <linux/types.h>
+#include <linux/linkage.h>
+#include <asm/signal.h>
+
+struct pt_regs;
+
+/*
+ * we need this inline - forking from kernel space will result
+ * in NO COPY ON WRITE (!!!), until an execve is executed. This
+ * is no problem, but for the stack. This is handled by not letting
+ * main() use the stack at all after fork(). Thus, no function
+ * calls - which means inline code for fork too, as otherwise we
+ * would use the stack upon exit from 'fork()'.
+ *
+ * Actually only pause and fork are needed inline, so that there
+ * won't be any messing with the stack from main(), but we define
+ * some others too.
+ */
+static inline int execve(const char *file, char **argv, char **envp)
+{
+	register long scno asm("r8") = __NR_execve;
+	register long sc1 asm("r12") = (long)file;
+	register long sc2 asm("r11") = (long)argv;
+	register long sc3 asm("r10") = (long)envp;
+	int res;
+
+	asm volatile("scall"
+		     : "=r"(sc1)
+		     : "r"(scno), "0"(sc1), "r"(sc2), "r"(sc3)
+		     : "lr", "memory");
+	res = sc1;
+	__syscall_return(int, res);
+}
+
+asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize);
+asmlinkage int sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
+			       struct pt_regs *regs);
+asmlinkage int sys_rt_sigreturn(struct pt_regs *regs);
+asmlinkage int sys_pipe(unsigned long __user *filedes);
+asmlinkage long sys_mmap2(unsigned long addr, unsigned long len,
+			  unsigned long prot, unsigned long flags,
+			  unsigned long fd, off_t offset);
+asmlinkage int sys_cacheflush(int operation, void __user *addr, size_t len);
+asmlinkage int sys_fork(struct pt_regs *regs);
+asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
+			 unsigned long parent_tidptr,
+			 unsigned long child_tidptr, struct pt_regs *regs);
+asmlinkage int sys_vfork(struct pt_regs *regs);
+asmlinkage int sys_execve(char __user *ufilename, char __user *__user *uargv,
+			  char __user *__user *uenvp, struct pt_regs *regs);
+
+#endif
+
+/*
+ * "Conditional" syscalls
+ *
+ * What we want is __attribute__((weak,alias("sys_ni_syscall"))),
+ * but it doesn't work on all toolchains, so we just do it by hand
+ */
+#define cond_syscall(x) asm(".weak\t" #x "\n\t.set\t" #x ",sys_ni_syscall");
+
+#endif /* __ASM_AVR32_UNISTD_H */
diff --git a/include/asm-avr32/user.h b/include/asm-avr32/user.h
new file mode 100644
index 000000000000..060fb3acee49
--- /dev/null
+++ b/include/asm-avr32/user.h
@@ -0,0 +1,65 @@
+/*
+ * Copyright (C) 2004-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Note: We may not need these definitions for AVR32, as we don't
+ * support a.out.
+ */
+#ifndef __ASM_AVR32_USER_H
+#define __ASM_AVR32_USER_H
+
+#include <linux/types.h>
+#include <asm/ptrace.h>
+#include <asm/page.h>
+
+/*
+ * Core file format: The core file is written in such a way that gdb
+ * can understand it and provide useful information to the user (under
+ * linux we use the `trad-core' bfd).  The file contents are as follows:
+ *
+ *  upage: 1 page consisting of a user struct that tells gdb
+ *	what is present in the file.  Directly after this is a
+ *	copy of the task_struct, which is currently not used by gdb,
+ *	but it may come in handy at some point.  All of the registers
+ *	are stored as part of the upage.  The upage should always be
+ *	only one page long.
+ *  data: The data segment follows next.  We use current->end_text to
+ *	current->brk to pick up all of the user variables, plus any memory
+ *	that may have been sbrk'ed.  No attempt is made to determine if a
+ *	page is demand-zero or if a page is totally unused, we just cover
+ *	the entire range.  All of the addresses are rounded in such a way
+ *	that an integral number of pages is written.
+ *  stack: We need the stack information in order to get a meaningful
+ *	backtrace.  We need to write the data from usp to
+ *	current->start_stack, so we round each of these in order to be able
+ *	to write an integer number of pages.
+ */
+
+struct user_fpu_struct {
+	/* We have no FPU (yet) */
+};
+
+struct user {
+	struct pt_regs	regs;			/* entire machine state */
+	size_t		u_tsize;		/* text size (pages) */
+	size_t		u_dsize;		/* data size (pages) */
+	size_t		u_ssize;		/* stack size (pages) */
+	unsigned long	start_code;		/* text starting address */
+	unsigned long	start_data;		/* data starting address */
+	unsigned long	start_stack;		/* stack starting address */
+	long int	signal;			/* signal causing core dump */
+	struct regs *	u_ar0;			/* help gdb find registers */
+	unsigned long	magic;			/* identifies a core file */
+	char		u_comm[32];		/* user command name */
+};
+
+#define NBPG			PAGE_SIZE
+#define UPAGES			1
+#define HOST_TEXT_START_ADDR	(u.start_code)
+#define HOST_DATA_START_ADDR	(u.start_data)
+#define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
+
+#endif /* __ASM_AVR32_USER_H */
diff --git a/include/linux/elf-em.h b/include/linux/elf-em.h
index 6a5796c81c90..666e0a5f00fc 100644
--- a/include/linux/elf-em.h
+++ b/include/linux/elf-em.h
@@ -31,6 +31,7 @@
 #define EM_M32R		88	/* Renesas M32R */
 #define EM_H8_300	46	/* Renesas H8/300,300H,H8S */
 #define EM_FRV		0x5441	/* Fujitsu FR-V */
+#define EM_AVR32	0x18ad	/* Atmel AVR32 */
 
 /*
  * This is an interim value that we will use until the committee comes
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 554ee688a9f8..3f21cc79a134 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -277,7 +277,7 @@ config DEBUG_HIGHMEM
 config DEBUG_BUGVERBOSE
 	bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED
 	depends on BUG
-	depends on ARM || ARM26 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV
+	depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV
 	default !EMBEDDED
 	help
 	  Say Y here to make BUG() panics output the file name and line number
@@ -315,7 +315,7 @@ config DEBUG_VM
 
 config FRAME_POINTER
 	bool "Compile the kernel with frame pointers"
-	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390)
+	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || AVR32)
 	default y if DEBUG_INFO && UML
 	help
 	  If you say Y here the resulting kernel image will be slightly larger
-- 
cgit v1.2.1


From bc157b75960f1f33566074e820342690216629b9 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 25 Sep 2006 23:32:16 -0700
Subject: [PATCH] AVR32 MTD: Static Memory Controller driver

This patchset adds the necessary drivers and infrastructure to access the
external flash on the ATSTK1000 board through the MTD subsystem.  With this
stuff in place, it will be possible to use a jffs2 filesystem stored in the
external flash as a root filesystem.  It might also be possible to update the
boot loader if you drop the write protection of partition 0.

As suggested by David Woodhouse, I reworked the patches to use the physmap
driver instead of introducing a separate mapping driver for the ATSTK1000.
I've also cleaned up the hsmc header by removing useless comments and
converting spaces to tabs (my headerfile generator needs some work.)

Unfortunately, I couldn't unlock the flash in fixup_use_atmel_lock because the
erase regions hadn't been set up yet, so I had to do it from cfi_amdstd_setup
instead.

This patch:

This adds a simple API for configuring the static memory controller along with
an implementation for the Atmel HSMC.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/avr32/mach-at32ap/Makefile     |   2 +-
 arch/avr32/mach-at32ap/at32ap7000.c |  10 +++
 arch/avr32/mach-at32ap/hsmc.c       | 164 ++++++++++++++++++++++++++++++++++++
 arch/avr32/mach-at32ap/hsmc.h       | 127 ++++++++++++++++++++++++++++
 include/asm-avr32/arch-at32ap/smc.h |  60 +++++++++++++
 5 files changed, 362 insertions(+), 1 deletion(-)
 create mode 100644 arch/avr32/mach-at32ap/hsmc.c
 create mode 100644 arch/avr32/mach-at32ap/hsmc.h
 create mode 100644 include/asm-avr32/arch-at32ap/smc.h

diff --git a/arch/avr32/mach-at32ap/Makefile b/arch/avr32/mach-at32ap/Makefile
index 4b10853eb614..f62eb6915510 100644
--- a/arch/avr32/mach-at32ap/Makefile
+++ b/arch/avr32/mach-at32ap/Makefile
@@ -1,2 +1,2 @@
-obj-y				+= at32ap.o clock.o pio.o intc.o extint.o
+obj-y				+= at32ap.o clock.o pio.o intc.o extint.o hsmc.o
 obj-$(CONFIG_CPU_AT32AP7000)	+= at32ap7000.o
diff --git a/arch/avr32/mach-at32ap/at32ap7000.c b/arch/avr32/mach-at32ap/at32ap7000.c
index e8c6893a1c23..37982b60398e 100644
--- a/arch/avr32/mach-at32ap/at32ap7000.c
+++ b/arch/avr32/mach-at32ap/at32ap7000.c
@@ -450,6 +450,13 @@ static struct clk hramc_clk = {
 	.users		= 1,
 };
 
+static struct resource smc0_resource[] = {
+	PBMEM(0xfff03400),
+};
+DEFINE_DEV(smc, 0);
+DEV_CLK(pclk, smc0, pbb, 13);
+DEV_CLK(mck, smc0, hsb, 0);
+
 static struct platform_device pdc_device = {
 	.name		= "pdc",
 	.id		= 0,
@@ -503,6 +510,7 @@ void __init at32_add_system_devices(void)
 
 	platform_device_register(&at32_sm_device);
 	platform_device_register(&at32_intc0_device);
+	platform_device_register(&smc0_device);
 	platform_device_register(&pdc_device);
 
 	platform_device_register(&pio0_device);
@@ -796,6 +804,8 @@ struct clk *at32_clock_list[] = {
 	&at32_intc0_pclk,
 	&ebi_clk,
 	&hramc_clk,
+	&smc0_pclk,
+	&smc0_mck,
 	&pdc_hclk,
 	&pdc_pclk,
 	&pico_clk,
diff --git a/arch/avr32/mach-at32ap/hsmc.c b/arch/avr32/mach-at32ap/hsmc.c
new file mode 100644
index 000000000000..7691721928a7
--- /dev/null
+++ b/arch/avr32/mach-at32ap/hsmc.c
@@ -0,0 +1,164 @@
+/*
+ * Static Memory Controller for AT32 chips
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#define DEBUG
+#include <linux/clk.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#include <asm/io.h>
+#include <asm/arch/smc.h>
+
+#include "hsmc.h"
+
+#define NR_CHIP_SELECTS 6
+
+struct hsmc {
+	void __iomem *regs;
+	struct clk *pclk;
+	struct clk *mck;
+};
+
+static struct hsmc *hsmc;
+
+int smc_set_configuration(int cs, const struct smc_config *config)
+{
+	unsigned long mul;
+	unsigned long offset;
+	u32 setup, pulse, cycle, mode;
+
+	if (!hsmc)
+		return -ENODEV;
+	if (cs >= NR_CHIP_SELECTS)
+		return -EINVAL;
+
+	/*
+	 * cycles = x / T = x * f
+	 *   = ((x * 1000000000) * ((f * 65536) / 1000000000)) / 65536
+	 *   = ((x * 1000000000) * (((f / 10000) * 65536) / 100000)) / 65536
+	 */
+	mul = (clk_get_rate(hsmc->mck) / 10000) << 16;
+	mul /= 100000;
+
+#define ns2cyc(x) ((((x) * mul) + 65535) >> 16)
+
+	setup = (HSMC_BF(NWE_SETUP, ns2cyc(config->nwe_setup))
+		 | HSMC_BF(NCS_WR_SETUP, ns2cyc(config->ncs_write_setup))
+		 | HSMC_BF(NRD_SETUP, ns2cyc(config->nrd_setup))
+		 | HSMC_BF(NCS_RD_SETUP, ns2cyc(config->ncs_read_setup)));
+	pulse = (HSMC_BF(NWE_PULSE, ns2cyc(config->nwe_pulse))
+		 | HSMC_BF(NCS_WR_PULSE, ns2cyc(config->ncs_write_pulse))
+		 | HSMC_BF(NRD_PULSE, ns2cyc(config->nrd_pulse))
+		 | HSMC_BF(NCS_RD_PULSE, ns2cyc(config->ncs_read_pulse)));
+	cycle = (HSMC_BF(NWE_CYCLE, ns2cyc(config->write_cycle))
+		 | HSMC_BF(NRD_CYCLE, ns2cyc(config->read_cycle)));
+
+	switch (config->bus_width) {
+	case 1:
+		mode = HSMC_BF(DBW, HSMC_DBW_8_BITS);
+		break;
+	case 2:
+		mode = HSMC_BF(DBW, HSMC_DBW_16_BITS);
+		break;
+	case 4:
+		mode = HSMC_BF(DBW, HSMC_DBW_32_BITS);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (config->nrd_controlled)
+		mode |= HSMC_BIT(READ_MODE);
+	if (config->nwe_controlled)
+		mode |= HSMC_BIT(WRITE_MODE);
+	if (config->byte_write)
+		mode |= HSMC_BIT(BAT);
+
+	pr_debug("smc cs%d: setup/%08x pulse/%08x cycle/%08x mode/%08x\n",
+		 cs, setup, pulse, cycle, mode);
+
+	offset = cs * 0x10;
+	hsmc_writel(hsmc, SETUP0 + offset, setup);
+	hsmc_writel(hsmc, PULSE0 + offset, pulse);
+	hsmc_writel(hsmc, CYCLE0 + offset, cycle);
+	hsmc_writel(hsmc, MODE0 + offset, mode);
+	hsmc_readl(hsmc, MODE0); /* I/O barrier */
+
+	return 0;
+}
+EXPORT_SYMBOL(smc_set_configuration);
+
+static int hsmc_probe(struct platform_device *pdev)
+{
+	struct resource *regs;
+	struct clk *pclk, *mck;
+	int ret;
+
+	if (hsmc)
+		return -EBUSY;
+
+	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!regs)
+		return -ENXIO;
+	pclk = clk_get(&pdev->dev, "pclk");
+	if (IS_ERR(pclk))
+		return PTR_ERR(pclk);
+	mck = clk_get(&pdev->dev, "mck");
+	if (IS_ERR(mck)) {
+		ret = PTR_ERR(mck);
+		goto out_put_pclk;
+	}
+
+	ret = -ENOMEM;
+	hsmc = kzalloc(sizeof(struct hsmc), GFP_KERNEL);
+	if (!hsmc)
+		goto out_put_clocks;
+
+	clk_enable(pclk);
+	clk_enable(mck);
+
+	hsmc->pclk = pclk;
+	hsmc->mck = mck;
+	hsmc->regs = ioremap(regs->start, regs->end - regs->start + 1);
+	if (!hsmc->regs)
+		goto out_disable_clocks;
+
+	dev_info(&pdev->dev, "Atmel Static Memory Controller at 0x%08lx\n",
+		 (unsigned long)regs->start);
+
+	platform_set_drvdata(pdev, hsmc);
+
+	return 0;
+
+out_disable_clocks:
+	clk_disable(mck);
+	clk_disable(pclk);
+	kfree(hsmc);
+out_put_clocks:
+	clk_put(mck);
+out_put_pclk:
+	clk_put(pclk);
+	hsmc = NULL;
+	return ret;
+}
+
+static struct platform_driver hsmc_driver = {
+	.probe		= hsmc_probe,
+	.driver		= {
+		.name	= "smc",
+	},
+};
+
+static int __init hsmc_init(void)
+{
+	return platform_driver_register(&hsmc_driver);
+}
+arch_initcall(hsmc_init);
diff --git a/arch/avr32/mach-at32ap/hsmc.h b/arch/avr32/mach-at32ap/hsmc.h
new file mode 100644
index 000000000000..5681276fafdb
--- /dev/null
+++ b/arch/avr32/mach-at32ap/hsmc.h
@@ -0,0 +1,127 @@
+/*
+ * Register definitions for Atmel Static Memory Controller (SMC)
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ASM_AVR32_HSMC_H__
+#define __ASM_AVR32_HSMC_H__
+
+/* HSMC register offsets */
+#define HSMC_SETUP0				0x0000
+#define HSMC_PULSE0				0x0004
+#define HSMC_CYCLE0				0x0008
+#define HSMC_MODE0				0x000c
+#define HSMC_SETUP1				0x0010
+#define HSMC_PULSE1				0x0014
+#define HSMC_CYCLE1				0x0018
+#define HSMC_MODE1				0x001c
+#define HSMC_SETUP2				0x0020
+#define HSMC_PULSE2				0x0024
+#define HSMC_CYCLE2				0x0028
+#define HSMC_MODE2				0x002c
+#define HSMC_SETUP3				0x0030
+#define HSMC_PULSE3				0x0034
+#define HSMC_CYCLE3				0x0038
+#define HSMC_MODE3				0x003c
+#define HSMC_SETUP4				0x0040
+#define HSMC_PULSE4				0x0044
+#define HSMC_CYCLE4				0x0048
+#define HSMC_MODE4				0x004c
+#define HSMC_SETUP5				0x0050
+#define HSMC_PULSE5				0x0054
+#define HSMC_CYCLE5				0x0058
+#define HSMC_MODE5				0x005c
+
+/* Bitfields in SETUP0 */
+#define HSMC_NWE_SETUP_OFFSET			0
+#define HSMC_NWE_SETUP_SIZE			6
+#define HSMC_NCS_WR_SETUP_OFFSET		8
+#define HSMC_NCS_WR_SETUP_SIZE			6
+#define HSMC_NRD_SETUP_OFFSET			16
+#define HSMC_NRD_SETUP_SIZE			6
+#define HSMC_NCS_RD_SETUP_OFFSET		24
+#define HSMC_NCS_RD_SETUP_SIZE			6
+
+/* Bitfields in PULSE0 */
+#define HSMC_NWE_PULSE_OFFSET			0
+#define HSMC_NWE_PULSE_SIZE			7
+#define HSMC_NCS_WR_PULSE_OFFSET		8
+#define HSMC_NCS_WR_PULSE_SIZE			7
+#define HSMC_NRD_PULSE_OFFSET			16
+#define HSMC_NRD_PULSE_SIZE			7
+#define HSMC_NCS_RD_PULSE_OFFSET		24
+#define HSMC_NCS_RD_PULSE_SIZE			7
+
+/* Bitfields in CYCLE0 */
+#define HSMC_NWE_CYCLE_OFFSET			0
+#define HSMC_NWE_CYCLE_SIZE			9
+#define HSMC_NRD_CYCLE_OFFSET			16
+#define HSMC_NRD_CYCLE_SIZE			9
+
+/* Bitfields in MODE0 */
+#define HSMC_READ_MODE_OFFSET			0
+#define HSMC_READ_MODE_SIZE			1
+#define HSMC_WRITE_MODE_OFFSET			1
+#define HSMC_WRITE_MODE_SIZE			1
+#define HSMC_EXNW_MODE_OFFSET			4
+#define HSMC_EXNW_MODE_SIZE			2
+#define HSMC_BAT_OFFSET				8
+#define HSMC_BAT_SIZE				1
+#define HSMC_DBW_OFFSET				12
+#define HSMC_DBW_SIZE				2
+#define HSMC_TDF_CYCLES_OFFSET			16
+#define HSMC_TDF_CYCLES_SIZE			4
+#define HSMC_TDF_MODE_OFFSET			20
+#define HSMC_TDF_MODE_SIZE			1
+#define HSMC_PMEN_OFFSET			24
+#define HSMC_PMEN_SIZE				1
+#define HSMC_PS_OFFSET				28
+#define HSMC_PS_SIZE				2
+
+/* Constants for READ_MODE */
+#define HSMC_READ_MODE_NCS_CONTROLLED		0
+#define HSMC_READ_MODE_NRD_CONTROLLED		1
+
+/* Constants for WRITE_MODE */
+#define HSMC_WRITE_MODE_NCS_CONTROLLED		0
+#define HSMC_WRITE_MODE_NWE_CONTROLLED		1
+
+/* Constants for EXNW_MODE */
+#define HSMC_EXNW_MODE_DISABLED			0
+#define HSMC_EXNW_MODE_RESERVED			1
+#define HSMC_EXNW_MODE_FROZEN			2
+#define HSMC_EXNW_MODE_READY			3
+
+/* Constants for BAT */
+#define HSMC_BAT_BYTE_SELECT			0
+#define HSMC_BAT_BYTE_WRITE			1
+
+/* Constants for DBW */
+#define HSMC_DBW_8_BITS				0
+#define HSMC_DBW_16_BITS			1
+#define HSMC_DBW_32_BITS			2
+
+/* Bit manipulation macros */
+#define HSMC_BIT(name)							\
+	(1 << HSMC_##name##_OFFSET)
+#define HSMC_BF(name,value)						\
+	(((value) & ((1 << HSMC_##name##_SIZE) - 1))			\
+	 << HSMC_##name##_OFFSET)
+#define HSMC_BFEXT(name,value)						\
+	(((value) >> HSMC_##name##_OFFSET)				\
+	 & ((1 << HSMC_##name##_SIZE) - 1))
+#define HSMC_BFINS(name,value,old)					\
+	(((old) & ~(((1 << HSMC_##name##_SIZE) - 1)			\
+		    << HSMC_##name##_OFFSET)) | HSMC_BF(name,value))
+
+/* Register access macros */
+#define hsmc_readl(port,reg)						\
+	readl((port)->regs + HSMC_##reg)
+#define hsmc_writel(port,reg,value)					\
+	writel((value), (port)->regs + HSMC_##reg)
+
+#endif /* __ASM_AVR32_HSMC_H__ */
diff --git a/include/asm-avr32/arch-at32ap/smc.h b/include/asm-avr32/arch-at32ap/smc.h
new file mode 100644
index 000000000000..3732b328303d
--- /dev/null
+++ b/include/asm-avr32/arch-at32ap/smc.h
@@ -0,0 +1,60 @@
+/*
+ * Static Memory Controller for AT32 chips
+ *
+ * Copyright (C) 2006 Atmel Corporation
+ *
+ * Inspired by the OMAP2 General-Purpose Memory Controller interface
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __ARCH_AT32AP_SMC_H
+#define __ARCH_AT32AP_SMC_H
+
+/*
+ * All timing parameters are in nanoseconds.
+ */
+struct smc_config {
+	/* Delay from address valid to assertion of given strobe */
+	u16		ncs_read_setup;
+	u16		nrd_setup;
+	u16		ncs_write_setup;
+	u16		nwe_setup;
+
+	/* Pulse length of given strobe */
+	u16		ncs_read_pulse;
+	u16		nrd_pulse;
+	u16		ncs_write_pulse;
+	u16		nwe_pulse;
+
+	/* Total cycle length of given operation */
+	u16		read_cycle;
+	u16		write_cycle;
+
+	/* Bus width in bytes */
+	u8		bus_width;
+
+	/*
+	 * 0: Data is sampled on rising edge of NCS
+	 * 1: Data is sampled on rising edge of NRD
+	 */
+	unsigned int	nrd_controlled:1;
+
+	/*
+	 * 0: Data is driven on falling edge of NCS
+	 * 1: Data is driven on falling edge of NWR
+	 */
+	unsigned int	nwe_controlled:1;
+
+	/*
+	 * 0: Byte select access type
+	 * 1: Byte write access type
+	 */
+	unsigned int	byte_write:1;
+};
+
+extern int smc_set_configuration(int cs, const struct smc_config *config);
+extern struct smc_config *smc_get_configuration(int cs);
+
+#endif /* __ARCH_AT32AP_SMC_H */
-- 
cgit v1.2.1


From 2514183dff2d5282cb745af34f56d1b98e5b2df8 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Mon, 25 Sep 2006 23:32:17 -0700
Subject: [PATCH] AVR32 MTD: AT49BV6416 platform device for ATSTK1000

FRegister a platform device for the AT49BV6416 NOR flash chip on the ATSTK1000
development board for use by the physmap MTD driver.

The SMC timings are set up before the platform device is registered so that no
board-specific mapping driver is necessary.

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/avr32/boards/atstk1000/Makefile |  2 +-
 arch/avr32/boards/atstk1000/flash.c  | 95 ++++++++++++++++++++++++++++++++++++
 2 files changed, 96 insertions(+), 1 deletion(-)
 create mode 100644 arch/avr32/boards/atstk1000/flash.c

diff --git a/arch/avr32/boards/atstk1000/Makefile b/arch/avr32/boards/atstk1000/Makefile
index c3e36ece8651..df9499480530 100644
--- a/arch/avr32/boards/atstk1000/Makefile
+++ b/arch/avr32/boards/atstk1000/Makefile
@@ -1,2 +1,2 @@
-obj-y				+= setup.o spi.o
+obj-y				+= setup.o spi.o flash.o
 obj-$(CONFIG_BOARD_ATSTK1002)	+= atstk1002.o
diff --git a/arch/avr32/boards/atstk1000/flash.c b/arch/avr32/boards/atstk1000/flash.c
new file mode 100644
index 000000000000..aac4300cca12
--- /dev/null
+++ b/arch/avr32/boards/atstk1000/flash.c
@@ -0,0 +1,95 @@
+/*
+ * ATSTK1000 board-specific flash initialization
+ *
+ * Copyright (C) 2005-2006 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/physmap.h>
+
+#include <asm/arch/smc.h>
+
+static struct smc_config flash_config __initdata = {
+	.ncs_read_setup		= 0,
+	.nrd_setup		= 40,
+	.ncs_write_setup	= 0,
+	.nwe_setup		= 10,
+
+	.ncs_read_pulse		= 80,
+	.nrd_pulse		= 40,
+	.ncs_write_pulse	= 65,
+	.nwe_pulse		= 55,
+
+	.read_cycle		= 120,
+	.write_cycle		= 120,
+
+	.bus_width		= 2,
+	.nrd_controlled		= 1,
+	.nwe_controlled		= 1,
+	.byte_write		= 1,
+};
+
+static struct mtd_partition flash_parts[] = {
+	{
+		.name           = "u-boot",
+		.offset         = 0x00000000,
+		.size           = 0x00020000,           /* 128 KiB */
+		.mask_flags     = MTD_WRITEABLE,
+	},
+	{
+		.name           = "root",
+		.offset         = 0x00020000,
+		.size           = 0x007d0000,
+	},
+	{
+		.name           = "env",
+		.offset         = 0x007f0000,
+		.size           = 0x00010000,
+		.mask_flags     = MTD_WRITEABLE,
+	},
+};
+
+static struct physmap_flash_data flash_data = {
+	.width		= 2,
+	.nr_parts	= ARRAY_SIZE(flash_parts),
+	.parts		= flash_parts,
+};
+
+static struct resource flash_resource = {
+	.start		= 0x00000000,
+	.end		= 0x007fffff,
+	.flags		= IORESOURCE_MEM,
+};
+
+static struct platform_device flash_device = {
+	.name		= "physmap-flash",
+	.id		= 0,
+	.resource	= &flash_resource,
+	.num_resources	= 1,
+	.dev		= {
+		.platform_data = &flash_data,
+	},
+};
+
+/* This needs to be called after the SMC has been initialized */
+static int __init atstk1000_flash_init(void)
+{
+	int ret;
+
+	ret = smc_set_configuration(0, &flash_config);
+	if (ret < 0) {
+		printk(KERN_ERR "atstk1000: failed to set NOR flash timing\n");
+		return ret;
+	}
+
+	platform_device_register(&flash_device);
+
+	return 0;
+}
+device_initcall(atstk1000_flash_init);
-- 
cgit v1.2.1


From 1447c27d38faf8fb03d4599e8082e507453ea3cf Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Mon, 25 Sep 2006 23:32:17 -0700
Subject: [PATCH] hpet rtc emulation: add watchdog timer

To prevent the emulated RTC timer from stopping when interrupts are delayed
for too long, disable interrupts around all of the register initialization,
and check that the interrupt handler did not schedule the next interrupt in
the past.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Vojtech Pavlik <vojtech@suse.cz>
Cc: Robert Picco <Robert.Picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/time_hpet.c | 37 +++++++++++++++++++++++++++++++------
 arch/x86_64/kernel/time.c    | 37 +++++++++++++++++++++++++++++++------
 2 files changed, 62 insertions(+), 12 deletions(-)

diff --git a/arch/i386/kernel/time_hpet.c b/arch/i386/kernel/time_hpet.c
index 14a1376fedd1..6bf14a4e995e 100644
--- a/arch/i386/kernel/time_hpet.c
+++ b/arch/i386/kernel/time_hpet.c
@@ -301,23 +301,25 @@ int hpet_rtc_timer_init(void)
 		hpet_rtc_int_freq = DEFAULT_RTC_INT_FREQ;
 
 	local_irq_save(flags);
+
 	cnt = hpet_readl(HPET_COUNTER);
 	cnt += ((hpet_tick*HZ)/hpet_rtc_int_freq);
 	hpet_writel(cnt, HPET_T1_CMP);
 	hpet_t1_cmp = cnt;
-	local_irq_restore(flags);
 
 	cfg = hpet_readl(HPET_T1_CFG);
 	cfg &= ~HPET_TN_PERIODIC;
 	cfg |= HPET_TN_ENABLE | HPET_TN_32BIT;
 	hpet_writel(cfg, HPET_T1_CFG);
 
+	local_irq_restore(flags);
+
 	return 1;
 }
 
 static void hpet_rtc_timer_reinit(void)
 {
-	unsigned int cfg, cnt;
+	unsigned int cfg, cnt, ticks_per_int, lost_ints;
 
 	if (unlikely(!(PIE_on | AIE_on | UIE_on))) {
 		cfg = hpet_readl(HPET_T1_CFG);
@@ -332,10 +334,33 @@ static void hpet_rtc_timer_reinit(void)
 		hpet_rtc_int_freq = DEFAULT_RTC_INT_FREQ;
 
 	/* It is more accurate to use the comparator value than current count.*/
-	cnt = hpet_t1_cmp;
-	cnt += hpet_tick*HZ/hpet_rtc_int_freq;
-	hpet_writel(cnt, HPET_T1_CMP);
-	hpet_t1_cmp = cnt;
+	ticks_per_int = hpet_tick * HZ / hpet_rtc_int_freq;
+	hpet_t1_cmp += ticks_per_int;
+	hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
+
+	/*
+	 * If the interrupt handler was delayed too long, the write above tries
+	 * to schedule the next interrupt in the past and the hardware would
+	 * not interrupt until the counter had wrapped around.
+	 * So we have to check that the comparator wasn't set to a past time.
+	 */
+	cnt = hpet_readl(HPET_COUNTER);
+	if (unlikely((int)(cnt - hpet_t1_cmp) > 0)) {
+		lost_ints = (cnt - hpet_t1_cmp) / ticks_per_int + 1;
+		/* Make sure that, even with the time needed to execute
+		 * this code, the next scheduled interrupt has been moved
+		 * back to the future: */
+		lost_ints++;
+
+		hpet_t1_cmp += lost_ints * ticks_per_int;
+		hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
+
+		if (PIE_on)
+			PIE_count += lost_ints;
+
+		printk(KERN_WARNING "rtc: lost some interrupts at %ldHz.\n",
+		       hpet_rtc_int_freq);
+	}
 }
 
 /*
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 7a9b18224182..7700e6cd2bd9 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -1148,23 +1148,25 @@ int hpet_rtc_timer_init(void)
 		hpet_rtc_int_freq = DEFAULT_RTC_INT_FREQ;
 
 	local_irq_save(flags);
+
 	cnt = hpet_readl(HPET_COUNTER);
 	cnt += ((hpet_tick*HZ)/hpet_rtc_int_freq);
 	hpet_writel(cnt, HPET_T1_CMP);
 	hpet_t1_cmp = cnt;
-	local_irq_restore(flags);
 
 	cfg = hpet_readl(HPET_T1_CFG);
 	cfg &= ~HPET_TN_PERIODIC;
 	cfg |= HPET_TN_ENABLE | HPET_TN_32BIT;
 	hpet_writel(cfg, HPET_T1_CFG);
 
+	local_irq_restore(flags);
+
 	return 1;
 }
 
 static void hpet_rtc_timer_reinit(void)
 {
-	unsigned int cfg, cnt;
+	unsigned int cfg, cnt, ticks_per_int, lost_ints;
 
 	if (unlikely(!(PIE_on | AIE_on | UIE_on))) {
 		cfg = hpet_readl(HPET_T1_CFG);
@@ -1179,10 +1181,33 @@ static void hpet_rtc_timer_reinit(void)
 		hpet_rtc_int_freq = DEFAULT_RTC_INT_FREQ;
 
 	/* It is more accurate to use the comparator value than current count.*/
-	cnt = hpet_t1_cmp;
-	cnt += hpet_tick*HZ/hpet_rtc_int_freq;
-	hpet_writel(cnt, HPET_T1_CMP);
-	hpet_t1_cmp = cnt;
+	ticks_per_int = hpet_tick * HZ / hpet_rtc_int_freq;
+	hpet_t1_cmp += ticks_per_int;
+	hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
+
+	/*
+	 * If the interrupt handler was delayed too long, the write above tries
+	 * to schedule the next interrupt in the past and the hardware would
+	 * not interrupt until the counter had wrapped around.
+	 * So we have to check that the comparator wasn't set to a past time.
+	 */
+	cnt = hpet_readl(HPET_COUNTER);
+	if (unlikely((int)(cnt - hpet_t1_cmp) > 0)) {
+		lost_ints = (cnt - hpet_t1_cmp) / ticks_per_int + 1;
+		/* Make sure that, even with the time needed to execute
+		 * this code, the next scheduled interrupt has been moved
+		 * back to the future: */
+		lost_ints++;
+
+		hpet_t1_cmp += lost_ints * ticks_per_int;
+		hpet_writel(hpet_t1_cmp, HPET_T1_CMP);
+
+		if (PIE_on)
+			PIE_count += lost_ints;
+
+		printk(KERN_WARNING "rtc: lost some interrupts at %ldHz.\n",
+		       hpet_rtc_int_freq);
+	}
 }
 
 /*
-- 
cgit v1.2.1


From 99325326a57b6a56595bb097655bee9fd27d77b0 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Mon, 25 Sep 2006 23:32:19 -0700
Subject: [PATCH] i386: show_registers(): try harder to print failing code

show_registers() tries to dump failing code starting 43 bytes before the
offending instruction, but this address can be bad, for example in a device
driver where the failing instruction is less than 43 bytes from the start
of the driver's code.  When that happens, try to dump code starting at the
failing instruction instead of printing no code at all.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/traps.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 7e9edafffd8a..4fcc6690be99 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -313,6 +313,8 @@ void show_registers(struct pt_regs *regs)
 	 */
 	if (in_kernel) {
 		u8 __user *eip;
+		int code_bytes = 64;
+		unsigned char c;
 
 		printk("\n" KERN_EMERG "Stack: ");
 		show_stack_log_lvl(NULL, regs, (unsigned long *)esp, KERN_EMERG);
@@ -320,9 +322,12 @@ void show_registers(struct pt_regs *regs)
 		printk(KERN_EMERG "Code: ");
 
 		eip = (u8 __user *)regs->eip - 43;
-		for (i = 0; i < 64; i++, eip++) {
-			unsigned char c;
-
+		if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
+			/* try starting at EIP */
+			eip = (u8 __user *)regs->eip;
+			code_bytes = 32;
+		}
+		for (i = 0; i < code_bytes; i++, eip++) {
 			if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
 				printk(" Bad EIP value.");
 				break;
-- 
cgit v1.2.1


From 3a750363e6075a28e5542ce93a69c620c0cfd605 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Mon, 25 Sep 2006 23:32:20 -0700
Subject: [PATCH] Use BUG_ON(foo) instead of "if (foo) BUG()" in
 include/asm-i386/dma-mapping.h

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/dma-mapping.h | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/include/asm-i386/dma-mapping.h b/include/asm-i386/dma-mapping.h
index 9cf20cacf76e..576ae01d71c8 100644
--- a/include/asm-i386/dma-mapping.h
+++ b/include/asm-i386/dma-mapping.h
@@ -21,8 +21,7 @@ static inline dma_addr_t
 dma_map_single(struct device *dev, void *ptr, size_t size,
 	       enum dma_data_direction direction)
 {
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 	WARN_ON(size == 0);
 	flush_write_buffers();
 	return virt_to_phys(ptr);
@@ -32,8 +31,7 @@ static inline void
 dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 		 enum dma_data_direction direction)
 {
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 }
 
 static inline int
@@ -42,8 +40,7 @@ dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 {
 	int i;
 
-	if (direction == DMA_NONE)
-		BUG();
+	BUG_ON(direction == DMA_NONE);
 	WARN_ON(nents == 0 || sg[0].length == 0);
 
 	for (i = 0; i < nents; i++ ) {
-- 
cgit v1.2.1


From c94a62aae6ebc99b416e55c023b6f5a1d19a400b Mon Sep 17 00:00:00 2001
From: Neil Horman <nhorman@tuxdriver.com>
Date: Mon, 25 Sep 2006 23:32:21 -0700
Subject: [PATCH] apm: clean up module initalization

Clean up module initalization for apm.c.  I had started by auditing for
proper return code checks in misc_register, but I found that in the event
of an initalization failure, a proc file and a kernel thread were left
hanging out.  this patch properly cleans up those loose ends on any
initalization failure.

Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apm.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c
index 8591f2fa920c..24fd577861f1 100644
--- a/arch/i386/kernel/apm.c
+++ b/arch/i386/kernel/apm.c
@@ -2339,6 +2339,7 @@ static int __init apm_init(void)
 	ret = kernel_thread(apm, NULL, CLONE_KERNEL | SIGCHLD);
 	if (ret < 0) {
 		printk(KERN_ERR "apm: disabled - Unable to start kernel thread.\n");
+		remove_proc_entry("apm", NULL);
 		return -ENOMEM;
 	}
 
@@ -2348,7 +2349,13 @@ static int __init apm_init(void)
 		return 0;
 	}
 
-	misc_register(&apm_device);
+	/*
+	 * Note we don't actually care if the misc_device cannot be registered.
+	 * this driver can do its job without it, even if userspace can't
+	 * control it.  just log the error
+	 */
+	if (misc_register(&apm_device))
+		printk(KERN_WARNING "apm: Could not register misc device.\n");
 
 	if (HZ != 100)
 		idle_period = (idle_period * HZ) / 100;
-- 
cgit v1.2.1


From 05f4a3ec94281347e05c81eafefcfe5ea545c94c Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:22 -0700
Subject: [PATCH] x86: remove locally-defined ldt structure in favour of
 standard type

arch/i386/kernel/reboot.c defines its own struct to describe an ldt entry: it
should use struct Xgt_desc_struct (currently load_ldt is a macro, so doesn't
complain: paravirt patches make it warn).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/reboot.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c
index 54cfeabbc5e4..84278e0093a2 100644
--- a/arch/i386/kernel/reboot.c
+++ b/arch/i386/kernel/reboot.c
@@ -145,14 +145,10 @@ real_mode_gdt_entries [3] =
 	0x000092000100ffffULL	/* 16-bit real-mode 64k data at 0x00000100 */
 };
 
-static struct
-{
-	unsigned short       size __attribute__ ((packed));
-	unsigned long long * base __attribute__ ((packed));
-}
-real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, real_mode_gdt_entries },
-real_mode_idt = { 0x3ff, NULL },
-no_idt = { 0, NULL };
+static struct Xgt_desc_struct
+real_mode_gdt = { sizeof (real_mode_gdt_entries) - 1, (long)real_mode_gdt_entries },
+real_mode_idt = { 0x3ff, 0 },
+no_idt = { 0, 0 };
 
 
 /* This is 16-bit protected mode code to disable paging and the cache,
-- 
cgit v1.2.1


From 027a8c7e6067a1bcdef6775d1b1c08729dfbae51 Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@sous-sol.org>
Date: Mon, 25 Sep 2006 23:32:23 -0700
Subject: [PATCH] x86: implement always-locked bit ops, for memory shared with
 an SMP hypervisor

Add "always lock'd" implementations of set_bit, clear_bit and change_bit and
the corresponding test_and_ functions.  Also add "always lock'd"
implementation of cmpxchg.  These give guaranteed strong synchronisation and
are required for non-SMP kernels running on an SMP hypervisor.

Signed-off-by: Ian Pratt <ian.pratt@xensource.com>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/sync_bitops.h | 156 +++++++++++++++++++++++++++++++++++++++++
 include/asm-i386/system.h      |  36 ++++++++++
 2 files changed, 192 insertions(+)
 create mode 100644 include/asm-i386/sync_bitops.h

diff --git a/include/asm-i386/sync_bitops.h b/include/asm-i386/sync_bitops.h
new file mode 100644
index 000000000000..c94d51c993ee
--- /dev/null
+++ b/include/asm-i386/sync_bitops.h
@@ -0,0 +1,156 @@
+#ifndef _I386_SYNC_BITOPS_H
+#define _I386_SYNC_BITOPS_H
+
+/*
+ * Copyright 1992, Linus Torvalds.
+ */
+
+/*
+ * These have to be done with inline assembly: that way the bit-setting
+ * is guaranteed to be atomic. All bit operations return 0 if the bit
+ * was cleared before the operation and != 0 if it was not.
+ *
+ * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
+ */
+
+#define ADDR (*(volatile long *) addr)
+
+/**
+ * sync_set_bit - Atomically set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * This function is atomic and may not be reordered.  See __set_bit()
+ * if you do not require the atomic guarantees.
+ *
+ * Note: there are no guarantees that this function will not be reordered
+ * on non x86 architectures, so if you are writting portable code,
+ * make sure not to rely on its reordering guarantees.
+ *
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void sync_set_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btsl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_clear_bit - Clears a bit in memory
+ * @nr: Bit to clear
+ * @addr: Address to start counting from
+ *
+ * sync_clear_bit() is atomic and may not be reordered.  However, it does
+ * not contain a memory barrier, so if it is used for locking purposes,
+ * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit()
+ * in order to ensure changes are visible on other processors.
+ */
+static inline void sync_clear_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btrl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_change_bit - Toggle a bit in memory
+ * @nr: Bit to change
+ * @addr: Address to start counting from
+ *
+ * change_bit() is atomic and may not be reordered. It may be
+ * reordered on other architectures than x86.
+ * Note that @nr may be almost arbitrarily large; this function is not
+ * restricted to acting on a single-word quantity.
+ */
+static inline void sync_change_bit(int nr, volatile unsigned long * addr)
+{
+	__asm__ __volatile__("lock; btcl %1,%0"
+			     :"+m" (ADDR)
+			     :"Ir" (nr)
+			     : "memory");
+}
+
+/**
+ * sync_test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It may be reordered on other architectures than x86.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_set_bit(int nr, volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btsl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+/**
+ * sync_test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It can be reorderdered on other architectures other than x86.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_clear_bit(int nr, volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btrl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+/**
+ * sync_test_and_change_bit - Change a bit and return its old value
+ * @nr: Bit to change
+ * @addr: Address to count from
+ *
+ * This operation is atomic and cannot be reordered.
+ * It also implies a memory barrier.
+ */
+static inline int sync_test_and_change_bit(int nr, volatile unsigned long* addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("lock; btcl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit),"+m" (ADDR)
+			     :"Ir" (nr) : "memory");
+	return oldbit;
+}
+
+static __always_inline int sync_const_test_bit(int nr, const volatile unsigned long *addr)
+{
+	return ((1UL << (nr & 31)) &
+		(((const volatile unsigned int *)addr)[nr >> 5])) != 0;
+}
+
+static inline int sync_var_test_bit(int nr, const volatile unsigned long * addr)
+{
+	int oldbit;
+
+	__asm__ __volatile__("btl %2,%1\n\tsbbl %0,%0"
+			     :"=r" (oldbit)
+			     :"m" (ADDR),"Ir" (nr));
+	return oldbit;
+}
+
+#define sync_test_bit(nr,addr)			\
+	(__builtin_constant_p(nr) ?		\
+	 sync_constant_test_bit((nr),(addr)) :	\
+	 sync_var_test_bit((nr),(addr)))
+
+#undef ADDR
+
+#endif /* _I386_SYNC_BITOPS_H */
diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h
index 098bcee94e38..a6dabbcd6e6a 100644
--- a/include/asm-i386/system.h
+++ b/include/asm-i386/system.h
@@ -267,6 +267,9 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 #define cmpxchg(ptr,o,n)\
 	((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
 					(unsigned long)(n),sizeof(*(ptr))))
+#define sync_cmpxchg(ptr,o,n)\
+	((__typeof__(*(ptr)))__sync_cmpxchg((ptr),(unsigned long)(o),\
+					(unsigned long)(n),sizeof(*(ptr))))
 #endif
 
 static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
@@ -296,6 +299,39 @@ static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
 	return old;
 }
 
+/*
+ * Always use locked operations when touching memory shared with a
+ * hypervisor, since the system may be SMP even if the guest kernel
+ * isn't.
+ */
+static inline unsigned long __sync_cmpxchg(volatile void *ptr,
+					    unsigned long old,
+					    unsigned long new, int size)
+{
+	unsigned long prev;
+	switch (size) {
+	case 1:
+		__asm__ __volatile__("lock; cmpxchgb %b1,%2"
+				     : "=a"(prev)
+				     : "q"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	case 2:
+		__asm__ __volatile__("lock; cmpxchgw %w1,%2"
+				     : "=a"(prev)
+				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	case 4:
+		__asm__ __volatile__("lock; cmpxchgl %1,%2"
+				     : "=a"(prev)
+				     : "r"(new), "m"(*__xg(ptr)), "0"(old)
+				     : "memory");
+		return prev;
+	}
+	return old;
+}
+
 #ifndef CONFIG_X86_CMPXCHG
 /*
  * Building a kernel capable running on 80386. It may be necessary to
-- 
cgit v1.2.1


From 9f093394d75cd9c5df82c7a99c5eb5d7ce7ba199 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:24 -0700
Subject: [PATCH] x86: roll all the cpuid asm into one __cpuid call

It's a little neater, and also means only one place to patch for
paravirtualization.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/processor.h | 60 ++++++++++++++++++++------------------------
 1 file changed, 27 insertions(+), 33 deletions(-)

diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index b32346d62e10..2277127696d2 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -143,6 +143,18 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {}
 #define X86_EFLAGS_VIP	0x00100000 /* Virtual Interrupt Pending */
 #define X86_EFLAGS_ID	0x00200000 /* CPUID detection flag */
 
+static inline void __cpuid(unsigned int *eax, unsigned int *ebx,
+			   unsigned int *ecx, unsigned int *edx)
+{
+	/* ecx is often an input as well as an output. */
+	__asm__("cpuid"
+		: "=a" (*eax),
+		  "=b" (*ebx),
+		  "=c" (*ecx),
+		  "=d" (*edx)
+		: "0" (*eax), "2" (*ecx));
+}
+
 /*
  * Generic CPUID function
  * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
@@ -150,24 +162,18 @@ static inline void detect_ht(struct cpuinfo_x86 *c) {}
  */
 static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx)
 {
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op), "c"(0));
+	*eax = op;
+	*ecx = 0;
+	__cpuid(eax, ebx, ecx, edx);
 }
 
 /* Some CPUID calls want 'count' to be placed in ecx */
 static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
-	       	int *edx)
+			       int *edx)
 {
-	__asm__("cpuid"
-		: "=a" (*eax),
-		  "=b" (*ebx),
-		  "=c" (*ecx),
-		  "=d" (*edx)
-		: "0" (op), "c" (count));
+	*eax = op;
+	*ecx = count;
+	__cpuid(eax, ebx, ecx, edx);
 }
 
 /*
@@ -175,42 +181,30 @@ static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
  */
 static inline unsigned int cpuid_eax(unsigned int op)
 {
-	unsigned int eax;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax)
-		: "0" (op)
-		: "bx", "cx", "dx");
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return eax;
 }
 static inline unsigned int cpuid_ebx(unsigned int op)
 {
-	unsigned int eax, ebx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=b" (ebx)
-		: "0" (op)
-		: "cx", "dx" );
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return ebx;
 }
 static inline unsigned int cpuid_ecx(unsigned int op)
 {
-	unsigned int eax, ecx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=c" (ecx)
-		: "0" (op)
-		: "bx", "dx" );
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return ecx;
 }
 static inline unsigned int cpuid_edx(unsigned int op)
 {
-	unsigned int eax, edx;
+	unsigned int eax, ebx, ecx, edx;
 
-	__asm__("cpuid"
-		: "=a" (eax), "=d" (edx)
-		: "0" (op)
-		: "bx", "cx");
+	cpuid(op, &eax, &ebx, &ecx, &edx);
 	return edx;
 }
 
-- 
cgit v1.2.1


From 052e79941a042e5be4feffa03b1fd60d93fb9e9a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@xensource.com>
Date: Mon, 25 Sep 2006 23:32:25 -0700
Subject: [PATCH] x86: make __FIXADDR_TOP variable to allow it to make space
 for a hypervisor

Make __FIXADDR_TOP a variable, so that it can be set to not get in the way of
address space a hypervisor may want to reserve.

Original patch by Gerd Hoffmann <kraxel@suse.de>

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig         |  1 +
 arch/i386/mm/init.c       | 42 ++++++++++++++++++++++++++++++++++++++++++
 arch/i386/mm/pgtable.c    | 26 ++++++++++++++++++++++++++
 include/asm-i386/fixmap.h |  7 ++++++-
 4 files changed, 75 insertions(+), 1 deletion(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index b2751eadbc56..1f83efb7a60e 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -794,6 +794,7 @@ config HOTPLUG_CPU
 config COMPAT_VDSO
 	bool "Compat VDSO support"
 	default y
+	depends on !PARAVIRT
 	help
 	  Map the VDSO to the predictable old-style address too.
 	---help---
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 353453d550a4..efd0bcdac65d 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -629,6 +629,48 @@ void __init mem_init(void)
 		(unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
 	       );
 
+#if 1 /* double-sanity-check paranoia */
+	printk("virtual kernel memory layout:\n"
+	       "    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+#ifdef CONFIG_HIGHMEM
+	       "    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+#endif
+	       "    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+	       "    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+	       "      .init : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+	       "      .data : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+	       "      .text : 0x%08lx - 0x%08lx   (%4ld kB)\n",
+	       FIXADDR_START, FIXADDR_TOP,
+	       (FIXADDR_TOP - FIXADDR_START) >> 10,
+
+#ifdef CONFIG_HIGHMEM
+	       PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
+	       (LAST_PKMAP*PAGE_SIZE) >> 10,
+#endif
+
+	       VMALLOC_START, VMALLOC_END,
+	       (VMALLOC_END - VMALLOC_START) >> 20,
+
+	       (unsigned long)__va(0), (unsigned long)high_memory,
+	       ((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,
+
+	       (unsigned long)&__init_begin, (unsigned long)&__init_end,
+	       ((unsigned long)&__init_end - (unsigned long)&__init_begin) >> 10,
+
+	       (unsigned long)&_etext, (unsigned long)&_edata,
+	       ((unsigned long)&_edata - (unsigned long)&_etext) >> 10,
+
+	       (unsigned long)&_text, (unsigned long)&_etext,
+	       ((unsigned long)&_etext - (unsigned long)&_text) >> 10);
+
+#ifdef CONFIG_HIGHMEM
+	BUG_ON(PKMAP_BASE+LAST_PKMAP*PAGE_SIZE > FIXADDR_START);
+	BUG_ON(VMALLOC_END                     > PKMAP_BASE);
+#endif
+	BUG_ON(VMALLOC_START                   > VMALLOC_END);
+	BUG_ON((unsigned long)high_memory      > VMALLOC_START);
+#endif /* double-sanity-check paranoia */
+
 #ifdef CONFIG_X86_PAE
 	if (!cpu_has_pae)
 		panic("cannot execute a PAE-enabled kernel on a PAE-less CPU!");
diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c
index a9f4910a22f8..10126e3f8174 100644
--- a/arch/i386/mm/pgtable.c
+++ b/arch/i386/mm/pgtable.c
@@ -12,6 +12,7 @@
 #include <linux/slab.h>
 #include <linux/pagemap.h>
 #include <linux/spinlock.h>
+#include <linux/module.h>
 
 #include <asm/system.h>
 #include <asm/pgtable.h>
@@ -139,6 +140,12 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
 	__flush_tlb_one(vaddr);
 }
 
+static int fixmaps;
+#ifndef CONFIG_COMPAT_VDSO
+unsigned long __FIXADDR_TOP = 0xfffff000;
+EXPORT_SYMBOL(__FIXADDR_TOP);
+#endif
+
 void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
 {
 	unsigned long address = __fix_to_virt(idx);
@@ -148,6 +155,25 @@ void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
 		return;
 	}
 	set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
+	fixmaps++;
+}
+
+/**
+ * reserve_top_address - reserves a hole in the top of kernel address space
+ * @reserve - size of hole to reserve
+ *
+ * Can be used to relocate the fixmap area and poke a hole in the top
+ * of kernel address space to make room for a hypervisor.
+ */
+void reserve_top_address(unsigned long reserve)
+{
+	BUG_ON(fixmaps > 0);
+#ifdef CONFIG_COMPAT_VDSO
+	BUG_ON(reserve != 0);
+#else
+	__FIXADDR_TOP = -reserve - PAGE_SIZE;
+	__VMALLOC_RESERVE += reserve;
+#endif
 }
 
 pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
diff --git a/include/asm-i386/fixmap.h b/include/asm-i386/fixmap.h
index a48cc3f7ccc6..02428cb36621 100644
--- a/include/asm-i386/fixmap.h
+++ b/include/asm-i386/fixmap.h
@@ -19,7 +19,11 @@
  * Leave one empty page between vmalloc'ed areas and
  * the start of the fixmap.
  */
-#define __FIXADDR_TOP	0xfffff000
+#ifndef CONFIG_COMPAT_VDSO
+extern unsigned long __FIXADDR_TOP;
+#else
+#define __FIXADDR_TOP  0xfffff000
+#endif
 
 #ifndef __ASSEMBLY__
 #include <linux/kernel.h>
@@ -93,6 +97,7 @@ enum fixed_addresses {
 
 extern void __set_fixmap (enum fixed_addresses idx,
 					unsigned long phys, pgprot_t flags);
+extern void reserve_top_address(unsigned long reserve);
 
 #define set_fixmap(idx, phys) \
 		__set_fixmap(idx, phys, PAGE_KERNEL)
-- 
cgit v1.2.1


From 461a9afff5e731d6337c0f5b08a1e727ccd57e0a Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Mon, 25 Sep 2006 23:32:25 -0700
Subject: [PATCH] x86: add a bootparameter to reserve high linear address space

Add a boot parameter to reserve high linear address space for hypervisors.
This is necessary to allow dynamically loaded hypervisor modules, which might
not happen until userspace is already running, and also provides a useful tool
to benchmark the performance impact of reduced lowmem address space.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |  5 +++++
 arch/i386/kernel/setup.c            | 18 ++++++++++++++++++
 2 files changed, 23 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 71d05f481727..766abdab94e7 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1363,6 +1363,11 @@ running once the system is up.
 
 	reserve=	[KNL,BUGS] Force the kernel to ignore some iomem area
 
+	reservetop=	[IA-32]
+			Format: nn[KMG]
+			Reserves a hole at the top of the kernel virtual
+			address space.
+
 	resume=		[SWSUSP]
 			Specify the partition device for software suspend
 
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 060c68004bec..16d99444cf66 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -934,6 +934,24 @@ static void __init parse_cmdline_early (char ** cmdline_p)
 	}
 }
 
+/*
+ * reservetop=size reserves a hole at the top of the kernel address space which
+ * a hypervisor can load into later.  Needed for dynamically loaded hypervisors,
+ * so relocating the fixmap can be done before paging initialization.
+ */
+static int __init parse_reservetop(char *arg)
+{
+	unsigned long address;
+
+	if (!arg)
+		return -EINVAL;
+
+	address = memparse(arg, &arg);
+	reserve_top_address(address);
+	return 0;
+}
+early_param("reservetop", parse_reservetop);
+
 /*
  * Callback for efi_memory_walk.
  */
-- 
cgit v1.2.1


From 9c9b8b388296ad5a306ab238dc677cfe6ff4cb12 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@xensource.com>
Date: Mon, 25 Sep 2006 23:32:26 -0700
Subject: [PATCH] x86: put .note.* sections into a PT_NOTE segment in vmlinux

This patch will pack any .note.* section into a PT_NOTE segment in the output
file.

To do this, we tell ld that we need a PT_NOTE segment.  This requires us to
start explicitly mapping sections to segments, so we also need to explicitly
create PT_LOAD segments for text and data, and map the sections to them
appropriately.  Fortunately, each section will default to its previous
section's segment, so it doesn't take many changes to vmlinux.lds.S.

This only changes i386 for now, but I presume the corresponding changes for
other architectures will be as simple.

This change also adds <linux/elfnote.h>, which defines C and Assembler macros
for actually creating ELF notes.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Hollis Blanchard <hollisb@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/vmlinux.lds.S    | 12 +++++-
 include/asm-generic/vmlinux.lds.h |  3 ++
 include/linux/elfnote.h           | 88 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 101 insertions(+), 2 deletions(-)
 create mode 100644 include/linux/elfnote.h

diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 2d4f1386e2b1..1e7ac1c44ddc 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -13,6 +13,12 @@ OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
 OUTPUT_ARCH(i386)
 ENTRY(phys_startup_32)
 jiffies = jiffies_64;
+
+PHDRS {
+	text PT_LOAD FLAGS(5);	/* R_E */
+	data PT_LOAD FLAGS(7);	/* RWE */
+	note PT_NOTE FLAGS(4);	/* R__ */
+}
 SECTIONS
 {
   . = __KERNEL_START;
@@ -26,7 +32,7 @@ SECTIONS
 	KPROBES_TEXT
 	*(.fixup)
 	*(.gnu.warning)
-	} = 0x9090
+	} :text = 0x9090
 
   _etext = .;			/* End of text section */
 
@@ -48,7 +54,7 @@ SECTIONS
   .data : AT(ADDR(.data) - LOAD_OFFSET) {	/* Data */
 	*(.data)
 	CONSTRUCTORS
-	}
+	} :data
 
   . = ALIGN(4096);
   __nosave_begin = .;
@@ -184,4 +190,6 @@ SECTIONS
   STABS_DEBUG
 
   DWARF_DEBUG
+
+  NOTES
 }
diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index db5a3732f106..253ae1328271 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -194,3 +194,6 @@
 		.stab.index 0 : { *(.stab.index) }			\
 		.stab.indexstr 0 : { *(.stab.indexstr) }		\
 		.comment 0 : { *(.comment) }
+
+#define NOTES								\
+		.notes : { *(.note.*) } :note
diff --git a/include/linux/elfnote.h b/include/linux/elfnote.h
new file mode 100644
index 000000000000..16f9f8ebffd9
--- /dev/null
+++ b/include/linux/elfnote.h
@@ -0,0 +1,88 @@
+#ifndef _LINUX_ELFNOTE_H
+#define _LINUX_ELFNOTE_H
+/*
+ * Helper macros to generate ELF Note structures, which are put into a
+ * PT_NOTE segment of the final vmlinux image.  These are useful for
+ * including name-value pairs of metadata into the kernel binary (or
+ * modules?) for use by external programs.
+ *
+ * Each note has three parts: a name, a type and a desc.  The name is
+ * intended to distinguish the note's originator, so it would be a
+ * company, project, subsystem, etc; it must be in a suitable form for
+ * use in a section name.  The type is an integer which is used to tag
+ * the data, and is considered to be within the "name" namespace (so
+ * "FooCo"'s type 42 is distinct from "BarProj"'s type 42).  The
+ * "desc" field is the actual data.  There are no constraints on the
+ * desc field's contents, though typically they're fairly small.
+ *
+ * All notes from a given NAME are put into a section named
+ * .note.NAME.  When the kernel image is finally linked, all the notes
+ * are packed into a single .notes section, which is mapped into the
+ * PT_NOTE segment.  Because notes for a given name are grouped into
+ * the same section, they'll all be adjacent the output file.
+ *
+ * This file defines macros for both C and assembler use.  Their
+ * syntax is slightly different, but they're semantically similar.
+ *
+ * See the ELF specification for more detail about ELF notes.
+ */
+
+#ifdef __ASSEMBLER__
+/*
+ * Generate a structure with the same shape as Elf{32,64}_Nhdr (which
+ * turn out to be the same size and shape), followed by the name and
+ * desc data with appropriate padding.  The 'desc' argument includes
+ * the assembler pseudo op defining the type of the data: .asciz
+ * "hello, world"
+ */
+.macro ELFNOTE name type desc:vararg
+.pushsection ".note.\name"
+  .align 4
+  .long 2f - 1f			/* namesz */
+  .long 4f - 3f			/* descsz */
+  .long \type
+1:.asciz "\name"
+2:.align 4
+3:\desc
+4:.align 4
+.popsection
+.endm
+#else	/* !__ASSEMBLER__ */
+#include <linux/elf.h>
+/*
+ * Use an anonymous structure which matches the shape of
+ * Elf{32,64}_Nhdr, but includes the name and desc data.  The size and
+ * type of name and desc depend on the macro arguments.  "name" must
+ * be a literal string, and "desc" must be passed by value.  You may
+ * only define one note per line, since __LINE__ is used to generate
+ * unique symbols.
+ */
+#define _ELFNOTE_PASTE(a,b)	a##b
+#define _ELFNOTE(size, name, unique, type, desc)			\
+	static const struct {						\
+		struct elf##size##_note _nhdr;				\
+		unsigned char _name[sizeof(name)]			\
+		__attribute__((aligned(sizeof(Elf##size##_Word))));	\
+		typeof(desc) _desc					\
+			     __attribute__((aligned(sizeof(Elf##size##_Word)))); \
+	} _ELFNOTE_PASTE(_note_, unique)				\
+		__attribute_used__					\
+		__attribute__((section(".note." name),			\
+			       aligned(sizeof(Elf##size##_Word)),	\
+			       unused)) = {				\
+		{							\
+			sizeof(name),					\
+			sizeof(desc),					\
+			type,						\
+		},							\
+		name,							\
+		desc							\
+	}
+#define ELFNOTE(size, name, type, desc)		\
+	_ELFNOTE(size, name, __LINE__, type, desc)
+
+#define ELFNOTE32(name, type, desc) ELFNOTE(32, name, type, desc)
+#define ELFNOTE64(name, type, desc) ELFNOTE(64, name, type, desc)
+#endif	/* __ASSEMBLER__ */
+
+#endif /* _LINUX_ELFNOTE_H */
-- 
cgit v1.2.1


From 5091e746848f74c9a2c0579b4ef8d8cd1a6b135d Mon Sep 17 00:00:00 2001
From: Ian Campbell <Ian.Campbell@xensource.com>
Date: Mon, 25 Sep 2006 23:32:28 -0700
Subject: [PATCH] Translate asm version of ELFNOTE macro into preprocessor
 macro

I've come across some problems with the assembly version of the ELFNOTE
macro currently in -mm. (in
x86-put-note-sections-into-a-pt_note-segment-in-vmlinux.patch)

The first is that older gas does not support :varargs in .macro
definitions (in my testing 2.17 does while 2.15 does not, I don't know
when it became supported). The Changes file says binutils >= 2.12 so I
think we need to avoid using it. There are no other uses in mainline or
-mm. Old gas appears to just ignore it so you get "too many arguments"
type errors.

Secondly it seems that passing strings as arguments to assembler macros
is broken without varargs. It looks like they get unquoted or each
character is treated as a separate argument or something and this causes
all manner of grief. I think this is because of the use of -traditional
when compiling assembly files.

Therefore I have translated the assembler macro into a pre-processor
macro.

I added the desctype as a separate argument instead of including it with
the descdata as the previous version did since -traditional means the
ELFNOTE definition after the #else needs to have the same number of
arguments (I think so anyway, the -traditional CPP semantics are pretty
fscking strange!).

With this patch I am able to define elfnotes in assembly like this with
both old and new assemblers.

	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz, "linux")
	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_VERSION,  .asciz, "2.6")
	ELFNOTE(Xen, XEN_ELFNOTE_XEN_VERSION,    .asciz, "xen-3.0")
	ELFNOTE(Xen, XEN_ELFNOTE_VIRT_BASE,      .long,  __PAGE_OFFSET)

Which seems reasonable enough.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/elfnote.h | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

diff --git a/include/linux/elfnote.h b/include/linux/elfnote.h
index 16f9f8ebffd9..67396db141e8 100644
--- a/include/linux/elfnote.h
+++ b/include/linux/elfnote.h
@@ -31,22 +31,24 @@
 /*
  * Generate a structure with the same shape as Elf{32,64}_Nhdr (which
  * turn out to be the same size and shape), followed by the name and
- * desc data with appropriate padding.  The 'desc' argument includes
- * the assembler pseudo op defining the type of the data: .asciz
- * "hello, world"
+ * desc data with appropriate padding.  The 'desctype' argument is the
+ * assembler pseudo op defining the type of the data e.g. .asciz while
+ * 'descdata' is the data itself e.g.  "hello, world".
+ *
+ * e.g. ELFNOTE(XYZCo, 42, .asciz, "forty-two")
+ *      ELFNOTE(XYZCo, 12, .long, 0xdeadbeef)
  */
-.macro ELFNOTE name type desc:vararg
-.pushsection ".note.\name"
-  .align 4
-  .long 2f - 1f			/* namesz */
-  .long 4f - 3f			/* descsz */
-  .long \type
-1:.asciz "\name"
-2:.align 4
-3:\desc
-4:.align 4
-.popsection
-.endm
+#define ELFNOTE(name, type, desctype, descdata)	\
+.pushsection .note.name			;	\
+  .align 4				;	\
+  .long 2f - 1f		/* namesz */	;	\
+  .long 4f - 3f		/* descsz */	;	\
+  .long type				;	\
+1:.asciz "name"				;	\
+2:.align 4				;	\
+3:desctype descdata			;	\
+4:.align 4				;	\
+.popsection				;
 #else	/* !__ASSEMBLER__ */
 #include <linux/elf.h>
 /*
-- 
cgit v1.2.1


From 753b9f86e7aef76c2beda32668ce528f90cb1733 Mon Sep 17 00:00:00 2001
From: Dave Hansen <haveblue@us.ibm.com>
Date: Mon, 25 Sep 2006 23:32:29 -0700
Subject: [PATCH] x86: enable VMSPLIT for highmem kernels

The current VMSPLIT Kconfig option is disabled whenever highmem is on.
This is a bit screwy because the people who need to change VMSPLIT the most
tend to be the ones with highmem and constrained lowmem.

So, remove the highmem dependency.  But, re-include the dependency for the
"full 1GB of lowmem" option.  You can't have the full 1GB of lowmem and
highmem because of the need for the vmalloc(), kmap(), etc...  areas.

I thought there would be at least a bit of tweaking to do to
get it to work, but everything seems OK.

Boot tested on a 4GB x86 machine, and a 12GB 3-node NUMA-Q:

elm3b82:~# cat /proc/meminfo
MemTotal:      3695412 kB
MemFree:       3659540 kB
...
LowTotal:      2909008 kB
LowFree:       2892324 kB
...
elm3b82:~# zgrep PAE /proc/config.gz
CONFIG_X86_PAE=y

larry:~# cat /proc/meminfo
MemTotal:     11845900 kB
MemFree:      11786748 kB
...
LowTotal:      2855180 kB
LowFree:       2830092 kB

Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 1f83efb7a60e..6189b0c28d6f 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -494,7 +494,7 @@ config HIGHMEM64G
 endchoice
 
 choice
-	depends on EXPERIMENTAL && !X86_PAE
+	depends on EXPERIMENTAL
 	prompt "Memory split" if EMBEDDED
 	default VMSPLIT_3G
 	help
@@ -516,6 +516,7 @@ choice
 	config VMSPLIT_3G
 		bool "3G/1G user/kernel split"
 	config VMSPLIT_3G_OPT
+		depends on !HIGHMEM
 		bool "3G/1G user/kernel split (for full 1G low memory)"
 	config VMSPLIT_2G
 		bool "2G/2G user/kernel split"
-- 
cgit v1.2.1


From 673eae8230a192f07b8715b872d6925521e9738d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:29 -0700
Subject: [PATCH] x86: trivial pgtable.h __ASSEMBLY__ move

Parsing generic pgtable.h in assembler is simply crazy.  None of this file is
needed in assembler code, and C inline functions and structures routine break
one or more different compiles.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/pgtable.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index c2059a3a0621..349260cd86ed 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_GENERIC_PGTABLE_H
 #define _ASM_GENERIC_PGTABLE_H
 
+#ifndef __ASSEMBLY__
+
 #ifndef __HAVE_ARCH_PTEP_ESTABLISH
 /*
  * Establish a new mapping:
@@ -188,7 +190,6 @@ static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addres
 })
 #endif
 
-#ifndef __ASSEMBLY__
 /*
  * When walking page tables, we usually want to skip any p?d_none entries;
  * and any p?d_bad entries - reporting the error before resetting to none.
-- 
cgit v1.2.1


From 6049742dbcecf170e903638a029f4dc280b9d53d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:30 -0700
Subject: [PATCH] x86: trivial move of __HAVE macros in i386 pagetable headers

Move the __HAVE_ARCH_PTEP defines to accompany the function definitions.
Anything else is just a complete nightmare to track through the 2/3-level
paging code, and this caused duplicate definitions to be needed (pte_same),
which could have easily been taken care of with the asm-generic pgtable
functions.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable-2level.h |  3 ++-
 include/asm-i386/pgtable-3level.h |  2 ++
 include/asm-i386/pgtable.h        | 10 ++++------
 3 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h
index 2756d4b04c27..201c86a6711e 100644
--- a/include/asm-i386/pgtable-2level.h
+++ b/include/asm-i386/pgtable-2level.h
@@ -21,8 +21,9 @@
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 #define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 #define ptep_get_and_clear(mm,addr,xp)	__pte(xchg(&(xp)->pte_low, 0))
-#define pte_same(a, b)		((a).pte_low == (b).pte_low)
+
 #define pte_page(x)		pfn_to_page(pte_pfn(x))
 #define pte_none(x)		(!(x).pte_low)
 #define pte_pfn(x)		((unsigned long)(((x).pte_low >> PAGE_SHIFT)))
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index 807ed9e366de..0d899173232e 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -105,6 +105,7 @@ static inline void pmd_clear(pmd_t *pmd)
 	*(tmp + 1) = 0;
 }
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
 static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	pte_t res;
@@ -117,6 +118,7 @@ static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 	return res;
 }
 
+#define __HAVE_ARCH_PTE_SAME
 static inline int pte_same(pte_t a, pte_t b)
 {
 	return a.pte_low == b.pte_low && a.pte_high == b.pte_high;
diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index c04b3d0f484e..38bf97514436 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -246,6 +246,7 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
 static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
 	if (!pte_dirty(*ptep))
@@ -253,6 +254,7 @@ static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned
 	return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte_low);
 }
 
+#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
 static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
 	if (!pte_young(*ptep))
@@ -260,6 +262,7 @@ static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned
 	return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte_low);
 }
 
+#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
 static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
 {
 	pte_t pte;
@@ -272,6 +275,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 	return pte;
 }
 
+#define __HAVE_ARCH_PTEP_SET_WRPROTECT
 static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
 	clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
@@ -441,12 +445,6 @@ extern void noexec_setup(const char *str);
 #define GET_IOSPACE(pfn)		0
 #define GET_PFN(pfn)			(pfn)
 
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
-#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
-#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
-#define __HAVE_ARCH_PTEP_SET_WRPROTECT
-#define __HAVE_ARCH_PTE_SAME
 #include <asm-generic/pgtable.h>
 
 #endif /* _I386_PGTABLE_H */
-- 
cgit v1.2.1


From 2965a0e6da0ccd8971ccf2c00a02bfa6e212acdb Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:31 -0700
Subject: [PATCH] x86: trivial move of ptep_set_access_flags

Move ptep_set_access_flags to be closer to the other ptep accessors, and make
the indentation standard.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/pgtable.h | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/include/asm-i386/pgtable.h b/include/asm-i386/pgtable.h
index 38bf97514436..0dc051a8078b 100644
--- a/include/asm-i386/pgtable.h
+++ b/include/asm-i386/pgtable.h
@@ -246,6 +246,22 @@ static inline pte_t pte_mkhuge(pte_t pte)	{ (pte).pte_low |= _PAGE_PSE; return p
 # include <asm/pgtable-2level.h>
 #endif
 
+/*
+ * We only update the dirty/accessed state if we set
+ * the dirty bit by hand in the kernel, since the hardware
+ * will do the accessed bit for us, and we don't want to
+ * race with other CPU's that might be updating the dirty
+ * bit at the same time.
+ */
+#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
+#define ptep_set_access_flags(vma, address, ptep, entry, dirty)		\
+do {									\
+	if (dirty) {							\
+		(ptep)->pte_low = (entry).pte_low;			\
+		flush_tlb_page(vma, address);				\
+	}								\
+} while (0)
+
 #define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
 static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
 {
@@ -415,23 +431,8 @@ extern void noexec_setup(const char *str);
 /*
  * The i386 doesn't have any external MMU info: the kernel page
  * tables contain all the necessary information.
- *
- * Also, we only update the dirty/accessed state if we set
- * the dirty bit by hand in the kernel, since the hardware
- * will do the accessed bit for us, and we don't want to
- * race with other CPU's that might be updating the dirty
- * bit at the same time.
  */
 #define update_mmu_cache(vma,address,pte) do { } while (0)
-#define  __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
-#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
-	do {								  \
-		if (__dirty) {						  \
-			(__ptep)->pte_low = (__entry).pte_low;	  	  \
-			flush_tlb_page(__vma, __address);		  \
-		}							  \
-	} while (0)
-
 #endif /* !__ASSEMBLY__ */
 
 #ifdef CONFIG_FLATMEM
-- 
cgit v1.2.1


From 27d26666fc495166036f4ee2208df25ae7f89ab8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Mon, 25 Sep 2006 23:32:32 -0700
Subject: [PATCH] x86: remove unused include from efi_stub.S

Remove unnecessary include from efi_stub.S

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/efi_stub.S | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/i386/kernel/efi_stub.S b/arch/i386/kernel/efi_stub.S
index d3ee73a3eee3..ef00bb77d7e4 100644
--- a/arch/i386/kernel/efi_stub.S
+++ b/arch/i386/kernel/efi_stub.S
@@ -7,7 +7,6 @@
 
 #include <linux/linkage.h>
 #include <asm/page.h>
-#include <asm/pgtable.h>
 
 /*
  * efi_call_phys(void *, ...) is a function with variable parameters.
-- 
cgit v1.2.1


From eaa70773e750cc09d60938bceacd028bc76b8e3a Mon Sep 17 00:00:00 2001
From: Stephane Eranian <eranian@hpl.hp.com>
Date: Mon, 25 Sep 2006 23:32:32 -0700
Subject: [PATCH] i386: add smp_call_function_single

Continiung the series of small patches necessary for the perfmon subsystem,
here is a patch that adds support for the smp_call_function_single()
function for i386.  It exists for almost all other architectures but i386.
The perfmon subsystem needs it in one case to free some state on a
designated remote CPU.

Signed-off-by: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smp.c | 66 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 66 insertions(+)

diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index c10789d7a9d3..304243ed7a30 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -634,3 +634,69 @@ fastcall void smp_call_function_interrupt(struct pt_regs *regs)
 	}
 }
 
+/*
+ * this function sends a 'generic call function' IPI to one other CPU
+ * in the system.
+ *
+ * cpu is a standard Linux logical CPU number.
+ */
+static void
+__smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+				int nonatomic, int wait)
+{
+	struct call_data_struct data;
+	int cpus = 1;
+
+	data.func = func;
+	data.info = info;
+	atomic_set(&data.started, 0);
+	data.wait = wait;
+	if (wait)
+		atomic_set(&data.finished, 0);
+
+	call_data = &data;
+	wmb();
+	/* Send a message to all other CPUs and wait for them to respond */
+	send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_VECTOR);
+
+	/* Wait for response */
+	while (atomic_read(&data.started) != cpus)
+		cpu_relax();
+
+	if (!wait)
+		return;
+
+	while (atomic_read(&data.finished) != cpus)
+		cpu_relax();
+}
+
+/*
+ * smp_call_function_single - Run a function on another CPU
+ * @func: The function to run. This must be fast and non-blocking.
+ * @info: An arbitrary pointer to pass to the function.
+ * @nonatomic: Currently unused.
+ * @wait: If true, wait until function has completed on other CPUs.
+ *
+ * Retrurns 0 on success, else a negative status code.
+ *
+ * Does not return until the remote CPU is nearly ready to execute <func>
+ * or is or has executed.
+ */
+
+int smp_call_function_single (int cpu, void (*func) (void *info), void *info,
+	int nonatomic, int wait)
+{
+	/* prevent preemption and reschedule on another processor */
+	int me = get_cpu();
+	if (cpu == me) {
+		WARN_ON(1);
+		put_cpu();
+		return -EBUSY;
+	}
+	spin_lock_bh(&call_lock);
+	__smp_call_function_single(cpu, func, info, nonatomic, wait);
+	spin_unlock_bh(&call_lock);
+	put_cpu();
+	return 0;
+}
+EXPORT_SYMBOL(smp_call_function_single);
-- 
cgit v1.2.1


From a3bc0dbc81d36fd38991b4373f6de8e1a507605a Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:33 -0700
Subject: [PATCH] smp_call_function_single() cleanup

If we're going to implement smp_call_function_single() on three architecture
with the same prototype then it should have a declaration in a
non-arch-specific header file.

Move it into <linux/smp.h>.

Cc: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smp.c              | 4 ++--
 arch/ia64/kernel/perfmon.c          | 1 +
 arch/ia64/sn/kernel/sn2/sn_hwperf.c | 3 ++-
 arch/x86_64/kernel/smpboot.c        | 3 ++-
 include/asm-ia64/smp.h              | 2 --
 include/asm-x86_64/smp.h            | 2 --
 include/linux/smp.h                 | 3 +++
 7 files changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/i386/kernel/smp.c b/arch/i386/kernel/smp.c
index 304243ed7a30..465188e2d701 100644
--- a/arch/i386/kernel/smp.c
+++ b/arch/i386/kernel/smp.c
@@ -683,8 +683,8 @@ __smp_call_function_single(int cpu, void (*func) (void *info), void *info,
  * or is or has executed.
  */
 
-int smp_call_function_single (int cpu, void (*func) (void *info), void *info,
-	int nonatomic, int wait)
+int smp_call_function_single(int cpu, void (*func) (void *info), void *info,
+			int nonatomic, int wait)
 {
 	/* prevent preemption and reschedule on another processor */
 	int me = get_cpu();
diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 84a7e52f56f6..7bb7696e4ce2 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -34,6 +34,7 @@
 #include <linux/file.h>
 #include <linux/poll.h>
 #include <linux/vfs.h>
+#include <linux/smp.h>
 #include <linux/pagemap.h>
 #include <linux/mount.h>
 #include <linux/bitops.h>
diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
index 9a8a29339d2d..b632b9c1e3b3 100644
--- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c
+++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c
@@ -32,9 +32,10 @@
 #include <linux/cpumask.h>
 #include <linux/smp_lock.h>
 #include <linux/nodemask.h>
+#include <linux/smp.h>
+
 #include <asm/processor.h>
 #include <asm/topology.h>
-#include <asm/smp.h>
 #include <asm/semaphore.h>
 #include <asm/uaccess.h>
 #include <asm/sal.h>
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 975380207b46..3ae9ffddddc0 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -46,9 +46,10 @@
 #include <linux/bootmem.h>
 #include <linux/thread_info.h>
 #include <linux/module.h>
-
 #include <linux/delay.h>
 #include <linux/mc146818rtc.h>
+#include <linux/smp.h>
+
 #include <asm/mtrr.h>
 #include <asm/pgalloc.h>
 #include <asm/desc.h>
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 719ff309ce09..74bde1c2bb1a 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -122,8 +122,6 @@ extern void __init smp_build_cpu_map(void);
 extern void __init init_smp_config (void);
 extern void smp_do_timer (struct pt_regs *regs);
 
-extern int smp_call_function_single (int cpuid, void (*func) (void *info), void *info,
-				     int retry, int wait);
 extern void smp_send_reschedule (int cpu);
 extern void lock_ipi_calllock(void);
 extern void unlock_ipi_calllock(void);
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6805e1feb300..ce97f65e1d10 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -48,8 +48,6 @@ extern void unlock_ipi_call_lock(void);
 extern int smp_num_siblings;
 extern void smp_send_reschedule(int cpu);
 void smp_stop_cpu(void);
-extern int smp_call_function_single(int cpuid, void (*func) (void *info),
-				void *info, int retry, int wait);
 
 extern cpumask_t cpu_sibling_map[NR_CPUS];
 extern cpumask_t cpu_core_map[NR_CPUS];
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 837e8bce1349..51649987f691 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -53,6 +53,9 @@ extern void smp_cpus_done(unsigned int max_cpus);
  */
 int smp_call_function(void(*func)(void *info), void *info, int retry, int wait);
 
+int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
+				int retry, int wait);
+
 /*
  * Call a function on all processors
  */
-- 
cgit v1.2.1


From 0f0f1b400ce3c4780b9d974bc69e8b558d99aba4 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Mon, 25 Sep 2006 23:32:34 -0700
Subject: [PATCH] Voyager: tty locking

Voyager fiddles with current->signal.tty without locking.  It turns out
that the code in question has already cleared current->signal.tty correctly
because daemonize() does the right thing already.

The signal handling also appears to be incorrect as it does an unprotected
sigfillset that also appears unneccessary.  As I don't have a bowtie and am
therefore not a qualified voyager maintainer I leave that to James.

Signed-off-by: Alan Cox <alan@redhat.com>
Acked-by: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mach-voyager/voyager_thread.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/i386/mach-voyager/voyager_thread.c b/arch/i386/mach-voyager/voyager_thread.c
index 50f6de6ff64d..f39887359e8e 100644
--- a/arch/i386/mach-voyager/voyager_thread.c
+++ b/arch/i386/mach-voyager/voyager_thread.c
@@ -130,7 +130,6 @@ thread(void *unused)
 	init_timer(&wakeup_timer);
 
 	sigfillset(&current->blocked);
-	current->signal->tty = NULL;
 
 	printk(KERN_NOTICE "Voyager starting monitor thread\n");
 
-- 
cgit v1.2.1


From c7f40ff15aba95bc09a759024d62b2c344ef0856 Mon Sep 17 00:00:00 2001
From: john stultz <johnstul@us.ibm.com>
Date: Mon, 25 Sep 2006 23:32:35 -0700
Subject: [PATCH] i386: Kill references to xtime

Remove all references to xtime in i386 and replace them w/
get/set_timeofday().  Requires some ugly and uncertain changes to APM, but
has been lightly tested to work.

Signed-off-by: John Stultz <johnstul@us.ibm.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Mikael Pettersson <mikpe@it.uu.se>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/apm.c  | 17 +++++------------
 arch/i386/kernel/time.c | 26 +++++++++++++++-----------
 2 files changed, 20 insertions(+), 23 deletions(-)

diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c
index 24fd577861f1..ff9ce4b5eaa8 100644
--- a/arch/i386/kernel/apm.c
+++ b/arch/i386/kernel/apm.c
@@ -1154,9 +1154,11 @@ out:
 
 static void set_time(void)
 {
+	struct timespec ts;
 	if (got_clock_diff) {	/* Must know time zone in order to set clock */
-		xtime.tv_sec = get_cmos_time() + clock_cmos_diff;
-		xtime.tv_nsec = 0; 
+		ts.tv_sec = get_cmos_time() + clock_cmos_diff;
+		ts.tv_nsec = 0;
+		do_settimeofday(&ts);
 	} 
 }
 
@@ -1232,13 +1234,8 @@ static int suspend(int vetoable)
 	restore_processor_state();
 
 	local_irq_disable();
-	write_seqlock(&xtime_lock);
-	spin_lock(&i8253_lock);
-	reinit_timer();
 	set_time();
-
-	spin_unlock(&i8253_lock);
-	write_sequnlock(&xtime_lock);
+	reinit_timer();
 
 	if (err == APM_NO_ERROR)
 		err = APM_SUCCESS;
@@ -1365,9 +1362,7 @@ static void check_events(void)
 			ignore_bounce = 1;
 			if ((event != APM_NORMAL_RESUME)
 			    || (ignore_normal_resume == 0)) {
-				write_seqlock_irq(&xtime_lock);
 				set_time();
-				write_sequnlock_irq(&xtime_lock);
 				device_resume();
 				pm_send_all(PM_RESUME, (void *)0);
 				queue_event(event, NULL);
@@ -1383,9 +1378,7 @@ static void check_events(void)
 			break;
 
 		case APM_UPDATE_TIME:
-			write_seqlock_irq(&xtime_lock);
 			set_time();
-			write_sequnlock_irq(&xtime_lock);
 			break;
 
 		case APM_CRITICAL_SUSPEND:
diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index edd00f6cee37..6f333e7fb23c 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -288,7 +288,7 @@ static int timer_resume(struct sys_device *dev)
 	unsigned long flags;
 	unsigned long sec;
 	unsigned long sleep_length;
-
+	struct timespec ts;
 #ifdef CONFIG_HPET_TIMER
 	if (is_hpet_enabled())
 		hpet_reenable();
@@ -296,9 +296,11 @@ static int timer_resume(struct sys_device *dev)
 	setup_pit_timer();
 	sec = get_cmos_time() + clock_cmos_diff;
 	sleep_length = (get_cmos_time() - sleep_start) * HZ;
+
+	ts.tv_sec = sec;
+	ts.tv_nsec = 0;
+	do_settimeofday(&ts);
 	write_seqlock_irqsave(&xtime_lock, flags);
-	xtime.tv_sec = sec;
-	xtime.tv_nsec = 0;
 	jiffies_64 += sleep_length;
 	wall_jiffies += sleep_length;
 	write_sequnlock_irqrestore(&xtime_lock, flags);
@@ -334,10 +336,11 @@ extern void (*late_time_init)(void);
 /* Duplicate of time_init() below, with hpet_enable part added */
 static void __init hpet_time_init(void)
 {
-	xtime.tv_sec = get_cmos_time();
-	xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-	set_normalized_timespec(&wall_to_monotonic,
-		-xtime.tv_sec, -xtime.tv_nsec);
+	struct timespec ts;
+	ts.tv_sec = get_cmos_time();
+	ts.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
+
+	do_settimeofday(&ts);
 
 	if ((hpet_enable() >= 0) && hpet_use_timer) {
 		printk("Using HPET for base-timer\n");
@@ -349,6 +352,7 @@ static void __init hpet_time_init(void)
 
 void __init time_init(void)
 {
+	struct timespec ts;
 #ifdef CONFIG_HPET_TIMER
 	if (is_hpet_capable()) {
 		/*
@@ -359,10 +363,10 @@ void __init time_init(void)
 		return;
 	}
 #endif
-	xtime.tv_sec = get_cmos_time();
-	xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
-	set_normalized_timespec(&wall_to_monotonic,
-		-xtime.tv_sec, -xtime.tv_nsec);
+	ts.tv_sec = get_cmos_time();
+	ts.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
+
+	do_settimeofday(&ts);
 
 	time_init_hook();
 }
-- 
cgit v1.2.1


From 182daa55679d9b3557d27609e7ab4bfab749ac5b Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Mon, 25 Sep 2006 23:32:36 -0700
Subject: [PATCH] mtrr: Add lock annotations for prepare_set and post_set

The functions prepare_set and post_set in kernel/cpu/mtrr/generic.c wrap
the spinlock set_atomicity_lock: prepare_set returns with the lock held,
and post_set releases the lock without acquiring it.  Add lock annotations
to these two functions so that sparse can check callers for lock pairing,
and so that sparse will not complain about these functions since they
intentionally use locks in this manner.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/mtrr/generic.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/i386/kernel/cpu/mtrr/generic.c b/arch/i386/kernel/cpu/mtrr/generic.c
index 169ac8e0db68..0b61eed8bbd8 100644
--- a/arch/i386/kernel/cpu/mtrr/generic.c
+++ b/arch/i386/kernel/cpu/mtrr/generic.c
@@ -243,7 +243,7 @@ static DEFINE_SPINLOCK(set_atomicity_lock);
  * has been called.
  */
 
-static void prepare_set(void)
+static void prepare_set(void) __acquires(set_atomicity_lock)
 {
 	unsigned long cr0;
 
@@ -274,7 +274,7 @@ static void prepare_set(void)
 	mtrr_wrmsr(MTRRdefType_MSR, deftype_lo & 0xf300UL, deftype_hi);
 }
 
-static void post_set(void)
+static void post_set(void) __releases(set_atomicity_lock)
 {
 	/*  Flush TLBs (no need to flush caches - they are disabled)  */
 	__flush_tlb();
-- 
cgit v1.2.1


From 060ec3d52db417a4fa554b6e14594ca62418c326 Mon Sep 17 00:00:00 2001
From: "Fernando J. Pereda" <ferdy@gentoo.org>
Date: Mon, 25 Sep 2006 23:32:37 -0700
Subject: [PATCH] alpha: Fix ALPHA_EV56 dependencies typo

There appears to be a typo in the EV56 config option. NORITAKE and PRIMO are
be able to set a variation of either.

Signed-off-by: Daniel Drake <dsd@gentoo.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/alpha/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index 213c7850d5fb..2b36afd8e969 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -381,7 +381,7 @@ config ALPHA_EV56
 
 config ALPHA_EV56
 	prompt "EV56 CPU (speed >= 333MHz)?"
-	depends on ALPHA_NORITAKE && ALPHA_PRIMO
+	depends on ALPHA_NORITAKE || ALPHA_PRIMO
 
 config ALPHA_EV56
 	prompt "EV56 CPU (speed >= 400MHz)?"
-- 
cgit v1.2.1


From 930631edd4b1fe2781d9fe90edbe35d89dfc94cc Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Mon, 25 Sep 2006 23:32:40 -0700
Subject: [PATCH] add DIV_ROUND_UP()

Add the DIV_ROUND_UP() helper macro: divide `n' by `d', rounding up.

Stolen from the gfs2 tree(!) because the swsusp patches need it.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 2b2ae4fdce8b..e44a37e2c71c 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -33,6 +33,7 @@ extern const char linux_banner[];
 #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
 #define ALIGN(x,a) (((x)+(a)-1UL)&~((a)-1UL))
 #define FIELD_SIZEOF(t, f) (sizeof(((t*)0)->f))
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
 #define roundup(x, y) ((((x) + ((y) - 1)) / (y)) * (y))
 
 #define	KERN_EMERG	"<0>"	/* system is unusable			*/
-- 
cgit v1.2.1


From 3a4f7577c9ef393ca80c783f02ffbc125de771c7 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:41 -0700
Subject: [PATCH] swsusp: add write-speed instrumentation

Add some instrumentation to the swsusp writeout code to show what bandwidth
we're achieving.

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/swap.c | 32 +++++++++++++++++++++++++++++---
 1 file changed, 29 insertions(+), 3 deletions(-)

diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index f1dd146bd64d..79b66e734bdb 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -146,6 +146,26 @@ static void release_swap_writer(struct swap_map_handle *handle)
 	handle->bitmap = NULL;
 }
 
+static void show_speed(struct timeval *start, struct timeval *stop,
+			unsigned nr_pages, char *msg)
+{
+	s64 elapsed_centisecs64;
+	int centisecs;
+	int k;
+	int kps;
+
+	elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start);
+	do_div(elapsed_centisecs64, NSEC_PER_SEC / 100);
+	centisecs = elapsed_centisecs64;
+	if (centisecs == 0)
+		centisecs = 1;	/* avoid div-by-zero */
+	k = nr_pages * (PAGE_SIZE / 1024);
+	kps = (k * 100) / centisecs;
+	printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k,
+			centisecs / 100, centisecs % 100,
+			kps / 1000, (kps % 1000) / 10);
+}
+
 static int get_swap_writer(struct swap_map_handle *handle)
 {
 	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL);
@@ -206,17 +226,21 @@ static int flush_swap_writer(struct swap_map_handle *handle)
 
 static int save_image(struct swap_map_handle *handle,
                       struct snapshot_handle *snapshot,
-                      unsigned int nr_pages)
+                      unsigned int nr_to_write)
 {
 	unsigned int m;
 	int ret;
 	int error = 0;
+	int nr_pages;
+	struct timeval start;
+	struct timeval stop;
 
-	printk("Saving image data pages (%u pages) ...     ", nr_pages);
-	m = nr_pages / 100;
+	printk("Saving image data pages (%u pages) ...     ", nr_to_write);
+	m = nr_to_write / 100;
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	do_gettimeofday(&start);
 	do {
 		ret = snapshot_read_next(snapshot, PAGE_SIZE);
 		if (ret > 0) {
@@ -228,8 +252,10 @@ static int save_image(struct swap_map_handle *handle,
 			nr_pages++;
 		}
 	} while (ret > 0);
+	do_gettimeofday(&stop);
 	if (!error)
 		printk("\b\b\b\bdone\n");
+	show_speed(&start, &stop, nr_to_write, "Wrote");
 	return error;
 }
 
-- 
cgit v1.2.1


From ab954160350c91c77ae03740ef90458c3ad5412c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:42 -0700
Subject: [PATCH] swsusp: write speedup

Switch the swsusp writeout code from 4k-at-a-time to 4MB-at-a-time.

Crufty old PIII testbox:
	12.9 MB/s -> 20.9 MB/s

Sony Vaio:
	14.7 MB/s -> 26.5 MB/s

The implementation is crude.  A better one would use larger BIOs, but wouldn't
gain any performance.

The memcpys will be mostly pipelined with the IO and basically come for free.

The ENOMEM path has not been tested.  It should be.

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/swap.h |  5 ++-
 kernel/power/swap.c  | 93 ++++++++++++++++++++++++++++++++++++++++++----------
 mm/page_io.c         | 25 ++++++++++----
 3 files changed, 98 insertions(+), 25 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index a2f5ad7c2d2e..3d434cbffe26 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -12,6 +12,8 @@
 
 struct notifier_block;
 
+struct bio;
+
 #define SWAP_FLAG_PREFER	0x8000	/* set if swap priority specified */
 #define SWAP_FLAG_PRIO_MASK	0x7fff
 #define SWAP_FLAG_PRIO_SHIFT	0
@@ -216,7 +218,8 @@ extern void swap_unplug_io_fn(struct backing_dev_info *, struct page *);
 /* linux/mm/page_io.c */
 extern int swap_readpage(struct file *, struct page *);
 extern int swap_writepage(struct page *page, struct writeback_control *wbc);
-extern int rw_swap_page_sync(int, swp_entry_t, struct page *);
+extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
+				struct bio **bio_chain);
 
 /* linux/mm/swap_state.c */
 extern struct address_space swapper_space;
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 79b66e734bdb..2dc883d361d5 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -49,18 +49,16 @@ static int mark_swapfiles(swp_entry_t start)
 {
 	int error;
 
-	rw_swap_page_sync(READ,
-			  swp_entry(root_swap, 0),
-			  virt_to_page((unsigned long)&swsusp_header));
+	rw_swap_page_sync(READ, swp_entry(root_swap, 0),
+			  virt_to_page((unsigned long)&swsusp_header), NULL);
 	if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
 	    !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
 		memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
 		memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
 		swsusp_header.image = start;
-		error = rw_swap_page_sync(WRITE,
-					  swp_entry(root_swap, 0),
-					  virt_to_page((unsigned long)
-						       &swsusp_header));
+		error = rw_swap_page_sync(WRITE, swp_entry(root_swap, 0),
+				virt_to_page((unsigned long)&swsusp_header),
+				NULL);
 	} else {
 		pr_debug("swsusp: Partition is not swap space.\n");
 		error = -ENODEV;
@@ -88,16 +86,37 @@ static int swsusp_swap_check(void) /* This is called before saving image */
  *	write_page - Write one page to given swap location.
  *	@buf:		Address we're writing.
  *	@offset:	Offset of the swap page we're writing to.
+ *	@bio_chain:	Link the next write BIO here
  */
 
-static int write_page(void *buf, unsigned long offset)
+static int write_page(void *buf, unsigned long offset, struct bio **bio_chain)
 {
 	swp_entry_t entry;
 	int error = -ENOSPC;
 
 	if (offset) {
+		struct page *page = virt_to_page(buf);
+
+		if (bio_chain) {
+			/*
+			 * Whether or not we successfully allocated a copy page,
+			 * we take a ref on the page here.  It gets undone in
+			 * wait_on_bio_chain().
+			 */
+			struct page *page_copy;
+			page_copy = alloc_page(GFP_ATOMIC);
+			if (page_copy == NULL) {
+				WARN_ON_ONCE(1);
+				bio_chain = NULL;	/* Go synchronous */
+				get_page(page);
+			} else {
+				memcpy(page_address(page_copy),
+					page_address(page), PAGE_SIZE);
+				page = page_copy;
+			}
+		}
 		entry = swp_entry(root_swap, offset);
-		error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf));
+		error = rw_swap_page_sync(WRITE, entry, page, bio_chain);
 	}
 	return error;
 }
@@ -185,37 +204,68 @@ static int get_swap_writer(struct swap_map_handle *handle)
 	return 0;
 }
 
-static int swap_write_page(struct swap_map_handle *handle, void *buf)
+static int wait_on_bio_chain(struct bio **bio_chain)
 {
-	int error;
+	struct bio *bio;
+	struct bio *next_bio;
+	int ret = 0;
+
+	if (bio_chain == NULL)
+		return 0;
+
+	bio = *bio_chain;
+	while (bio) {
+		struct page *page;
+
+		next_bio = bio->bi_private;
+		page = bio->bi_io_vec[0].bv_page;
+		wait_on_page_locked(page);
+		if (!PageUptodate(page) || PageError(page))
+			ret = -EIO;
+		put_page(page);
+		bio_put(bio);
+		bio = next_bio;
+	}
+	*bio_chain = NULL;
+	return ret;
+}
+
+static int swap_write_page(struct swap_map_handle *handle, void *buf,
+				struct bio **bio_chain)
+{
+	int error = 0;
 	unsigned long offset;
 
 	if (!handle->cur)
 		return -EINVAL;
 	offset = alloc_swap_page(root_swap, handle->bitmap);
-	error = write_page(buf, offset);
+	error = write_page(buf, offset, bio_chain);
 	if (error)
 		return error;
 	handle->cur->entries[handle->k++] = offset;
 	if (handle->k >= MAP_PAGE_ENTRIES) {
+		error = wait_on_bio_chain(bio_chain);
+		if (error)
+			goto out;
 		offset = alloc_swap_page(root_swap, handle->bitmap);
 		if (!offset)
 			return -ENOSPC;
 		handle->cur->next_swap = offset;
-		error = write_page(handle->cur, handle->cur_swap);
+		error = write_page(handle->cur, handle->cur_swap, NULL);
 		if (error)
-			return error;
+			goto out;
 		memset(handle->cur, 0, PAGE_SIZE);
 		handle->cur_swap = offset;
 		handle->k = 0;
 	}
-	return 0;
+out:
+	return error;
 }
 
 static int flush_swap_writer(struct swap_map_handle *handle)
 {
 	if (handle->cur && handle->cur_swap)
-		return write_page(handle->cur, handle->cur_swap);
+		return write_page(handle->cur, handle->cur_swap, NULL);
 	else
 		return -EINVAL;
 }
@@ -232,6 +282,8 @@ static int save_image(struct swap_map_handle *handle,
 	int ret;
 	int error = 0;
 	int nr_pages;
+	int err2;
+	struct bio *bio;
 	struct timeval start;
 	struct timeval stop;
 
@@ -240,11 +292,13 @@ static int save_image(struct swap_map_handle *handle,
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	bio = NULL;
 	do_gettimeofday(&start);
 	do {
 		ret = snapshot_read_next(snapshot, PAGE_SIZE);
 		if (ret > 0) {
-			error = swap_write_page(handle, data_of(*snapshot));
+			error = swap_write_page(handle, data_of(*snapshot),
+						&bio);
 			if (error)
 				break;
 			if (!(nr_pages % m))
@@ -252,7 +306,10 @@ static int save_image(struct swap_map_handle *handle,
 			nr_pages++;
 		}
 	} while (ret > 0);
+	err2 = wait_on_bio_chain(&bio);
 	do_gettimeofday(&stop);
+	if (!error)
+		error = err2;
 	if (!error)
 		printk("\b\b\b\bdone\n");
 	show_speed(&start, &stop, nr_to_write, "Wrote");
@@ -307,7 +364,7 @@ int swsusp_write(void)
 	error = get_swap_writer(&handle);
 	if (!error) {
 		unsigned long start = handle.cur_swap;
-		error = swap_write_page(&handle, header);
+		error = swap_write_page(&handle, header, NULL);
 		if (!error)
 			error = save_image(&handle, &snapshot,
 					header->pages - 1);
diff --git a/mm/page_io.c b/mm/page_io.c
index d2f0a5783370..f46a9862b7ef 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -156,10 +156,12 @@ out:
  * We use end_swap_bio_read() even for writes, because it happens to do what
  * we want.
  */
-int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page)
+int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
+			struct bio **bio_chain)
 {
 	struct bio *bio;
 	int ret = 0;
+	int bio_rw;
 
 	lock_page(page);
 
@@ -170,11 +172,22 @@ int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page)
 		goto out;
 	}
 
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	wait_on_page_locked(page);
-
-	if (!PageUptodate(page) || PageError(page))
-		ret = -EIO;
+	bio_rw = rw;
+	if (!bio_chain)
+		bio_rw |= (1 << BIO_RW_SYNC);
+	if (bio_chain)
+		bio_get(bio);
+	submit_bio(bio_rw, bio);
+	if (bio_chain == NULL) {
+		wait_on_page_locked(page);
+
+		if (!PageUptodate(page) || PageError(page))
+			ret = -EIO;
+	}
+	if (bio_chain) {
+		bio->bi_private = *bio_chain;
+		*bio_chain = bio;
+	}
 out:
 	return ret;
 }
-- 
cgit v1.2.1


From 8c002494b55119a3fd1dddee83b4fb75cfda47e5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:43 -0700
Subject: [PATCH] swsusp: add read-speed instrumentation

Add some instrumentation to the swsusp readin code to show what bandwidth
we're achieving.

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/swap.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 2dc883d361d5..9ab989572164 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -522,12 +522,15 @@ static int load_image(struct swap_map_handle *handle,
 	unsigned int m;
 	int ret;
 	int error = 0;
+	struct timeval start;
+	struct timeval stop;
 
 	printk("Loading image data pages (%u pages) ...     ", nr_pages);
 	m = nr_pages / 100;
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	do_gettimeofday(&start);
 	do {
 		ret = snapshot_write_next(snapshot, PAGE_SIZE);
 		if (ret > 0) {
@@ -539,11 +542,13 @@ static int load_image(struct swap_map_handle *handle,
 			nr_pages++;
 		}
 	} while (ret > 0);
+	do_gettimeofday(&stop);
 	if (!error) {
 		printk("\b\b\b\bdone\n");
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
+	show_speed(&start, &stop, nr_pages, "Read");
 	return error;
 }
 
-- 
cgit v1.2.1


From 546e0d271941dd1ff6961e2a1f7eac75f1fc277e Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 23:32:44 -0700
Subject: [PATCH] swsusp: read speedup

Implement async reads for swsusp resuming.

Crufty old PIII testbox:
	15.7 MB/s -> 20.3 MB/s

Sony Vaio:
	14.6 MB/s -> 33.3 MB/s

I didn't implement the post-resume bio_set_pages_dirty().  I don't really
understand why resume needs to run set_page_dirty() against these pages.

It might be a worry that this code modifies PG_Uptodate, PG_Error and
PG_Locked against the image pages.  Can this possibly affect the resumed-into
kernel?  Hopefully not, if we're atomically restoring its mem_map?

Cc: Pavel Machek <pavel@ucw.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Jens Axboe <axboe@suse.de>
Cc: Laurent Riffard <laurent.riffard@free.fr>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/swap.h    |   1 +
 kernel/power/power.h    |   1 +
 kernel/power/snapshot.c |  11 ++--
 kernel/power/swap.c     | 138 ++++++++++++++++++++++++------------------------
 mm/page_io.c            |   2 +-
 5 files changed, 81 insertions(+), 72 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 3d434cbffe26..e7c36ba2a2db 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -220,6 +220,7 @@ extern int swap_readpage(struct file *, struct page *);
 extern int swap_writepage(struct page *page, struct writeback_control *wbc);
 extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page,
 				struct bio **bio_chain);
+extern int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err);
 
 /* linux/mm/swap_state.c */
 extern struct address_space swapper_space;
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 57a792982fb9..59ce712f082d 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -58,6 +58,7 @@ struct snapshot_handle {
 	struct pbe	*pbe, *last_pbe;
 	void		*buffer;
 	unsigned int	buf_offset;
+	int		sync_read;
 };
 
 #define data_of(handle)	((handle).buffer + (handle).buf_offset)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 75d4886e648e..591301ae8b7d 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -314,7 +314,7 @@ static unsigned int unsafe_pages;
  *	and we count them using unsafe_pages
  */
 
-static inline void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
+static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 {
 	void *res;
 
@@ -828,13 +828,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 	}
 	if (!handle->offset)
 		handle->buffer = buffer;
+	handle->sync_read = 1;
 	if (handle->prev < handle->page) {
 		if (!handle->prev) {
-			error = load_header(handle, (struct swsusp_info *)buffer);
+			error = load_header(handle,
+					(struct swsusp_info *)buffer);
 			if (error)
 				return error;
 		} else if (handle->prev <= nr_meta_pages) {
-			handle->pbe = unpack_orig_addresses(buffer, handle->pbe);
+			handle->pbe = unpack_orig_addresses(buffer,
+							handle->pbe);
 			if (!handle->pbe) {
 				error = prepare_image(handle);
 				if (error)
@@ -842,10 +845,12 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 				handle->pbe = pagedir_nosave;
 				handle->last_pbe = NULL;
 				handle->buffer = get_buffer(handle);
+				handle->sync_read = 0;
 			}
 		} else {
 			handle->pbe = handle->pbe->next;
 			handle->buffer = get_buffer(handle);
+			handle->sync_read = 0;
 		}
 		handle->prev = handle->page;
 	}
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 9ab989572164..8309d20b2563 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -22,6 +22,7 @@
 #include <linux/device.h>
 #include <linux/buffer_head.h>
 #include <linux/bio.h>
+#include <linux/blkdev.h>
 #include <linux/swap.h>
 #include <linux/swapops.h>
 #include <linux/pm.h>
@@ -214,6 +215,8 @@ static int wait_on_bio_chain(struct bio **bio_chain)
 		return 0;
 
 	bio = *bio_chain;
+	if (bio == NULL)
+		return 0;
 	while (bio) {
 		struct page *page;
 
@@ -349,7 +352,8 @@ int swsusp_write(void)
 	int error;
 
 	if ((error = swsusp_swap_check())) {
-		printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
+		printk(KERN_ERR "swsusp: Cannot find swap device, try "
+				"swapon -a.\n");
 		return error;
 	}
 	memset(&snapshot, 0, sizeof(struct snapshot_handle));
@@ -381,27 +385,6 @@ int swsusp_write(void)
 	return error;
 }
 
-/*
- *	Using bio to read from swap.
- *	This code requires a bit more work than just using buffer heads
- *	but, it is the recommended way for 2.5/2.6.
- *	The following are to signal the beginning and end of I/O. Bios
- *	finish asynchronously, while we want them to happen synchronously.
- *	A simple atomic_t, and a wait loop take care of this problem.
- */
-
-static atomic_t io_done = ATOMIC_INIT(0);
-
-static int end_io(struct bio *bio, unsigned int num, int err)
-{
-	if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) {
-		printk(KERN_ERR "I/O error reading swsusp image.\n");
-		return -EIO;
-	}
-	atomic_set(&io_done, 0);
-	return 0;
-}
-
 static struct block_device *resume_bdev;
 
 /**
@@ -409,15 +392,15 @@ static struct block_device *resume_bdev;
  *	@rw:	READ or WRITE.
  *	@off	physical offset of page.
  *	@page:	page we're reading or writing.
+ *	@bio_chain: list of pending biod (for async reading)
  *
  *	Straight from the textbook - allocate and initialize the bio.
- *	If we're writing, make sure the page is marked as dirty.
- *	Then submit it and wait.
+ *	If we're reading, make sure the page is marked as dirty.
+ *	Then submit it and, if @bio_chain == NULL, wait.
  */
-
-static int submit(int rw, pgoff_t page_off, void *page)
+static int submit(int rw, pgoff_t page_off, struct page *page,
+			struct bio **bio_chain)
 {
-	int error = 0;
 	struct bio *bio;
 
 	bio = bio_alloc(GFP_ATOMIC, 1);
@@ -425,33 +408,40 @@ static int submit(int rw, pgoff_t page_off, void *page)
 		return -ENOMEM;
 	bio->bi_sector = page_off * (PAGE_SIZE >> 9);
 	bio->bi_bdev = resume_bdev;
-	bio->bi_end_io = end_io;
+	bio->bi_end_io = end_swap_bio_read;
 
-	if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
-		printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
-		error = -EFAULT;
-		goto Done;
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
+		printk("swsusp: ERROR: adding page to bio at %ld\n", page_off);
+		bio_put(bio);
+		return -EFAULT;
 	}
 
-	atomic_set(&io_done, 1);
-	submit_bio(rw | (1 << BIO_RW_SYNC), bio);
-	while (atomic_read(&io_done))
-		yield();
-	if (rw == READ)
-		bio_set_pages_dirty(bio);
- Done:
-	bio_put(bio);
-	return error;
+	lock_page(page);
+	bio_get(bio);
+
+	if (bio_chain == NULL) {
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+		wait_on_page_locked(page);
+		if (rw == READ)
+			bio_set_pages_dirty(bio);
+		bio_put(bio);
+	} else {
+		get_page(page);
+		bio->bi_private = *bio_chain;
+		*bio_chain = bio;
+		submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+	}
+	return 0;
 }
 
-static int bio_read_page(pgoff_t page_off, void *page)
+static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain)
 {
-	return submit(READ, page_off, page);
+	return submit(READ, page_off, virt_to_page(addr), bio_chain);
 }
 
-static int bio_write_page(pgoff_t page_off, void *page)
+static int bio_write_page(pgoff_t page_off, void *addr)
 {
-	return submit(WRITE, page_off, page);
+	return submit(WRITE, page_off, virt_to_page(addr), NULL);
 }
 
 /**
@@ -476,7 +466,7 @@ static int get_swap_reader(struct swap_map_handle *handle,
 	handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
 	if (!handle->cur)
 		return -ENOMEM;
-	error = bio_read_page(swp_offset(start), handle->cur);
+	error = bio_read_page(swp_offset(start), handle->cur, NULL);
 	if (error) {
 		release_swap_reader(handle);
 		return error;
@@ -485,7 +475,8 @@ static int get_swap_reader(struct swap_map_handle *handle,
 	return 0;
 }
 
-static int swap_read_page(struct swap_map_handle *handle, void *buf)
+static int swap_read_page(struct swap_map_handle *handle, void *buf,
+				struct bio **bio_chain)
 {
 	unsigned long offset;
 	int error;
@@ -495,16 +486,17 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf)
 	offset = handle->cur->entries[handle->k];
 	if (!offset)
 		return -EFAULT;
-	error = bio_read_page(offset, buf);
+	error = bio_read_page(offset, buf, bio_chain);
 	if (error)
 		return error;
 	if (++handle->k >= MAP_PAGE_ENTRIES) {
+		error = wait_on_bio_chain(bio_chain);
 		handle->k = 0;
 		offset = handle->cur->next_swap;
 		if (!offset)
 			release_swap_reader(handle);
-		else
-			error = bio_read_page(offset, handle->cur);
+		else if (!error)
+			error = bio_read_page(offset, handle->cur, NULL);
 	}
 	return error;
 }
@@ -517,38 +509,48 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf)
 
 static int load_image(struct swap_map_handle *handle,
                       struct snapshot_handle *snapshot,
-                      unsigned int nr_pages)
+                      unsigned int nr_to_read)
 {
 	unsigned int m;
-	int ret;
 	int error = 0;
 	struct timeval start;
 	struct timeval stop;
+	struct bio *bio;
+	int err2;
+	unsigned nr_pages;
 
-	printk("Loading image data pages (%u pages) ...     ", nr_pages);
-	m = nr_pages / 100;
+	printk("Loading image data pages (%u pages) ...     ", nr_to_read);
+	m = nr_to_read / 100;
 	if (!m)
 		m = 1;
 	nr_pages = 0;
+	bio = NULL;
 	do_gettimeofday(&start);
-	do {
-		ret = snapshot_write_next(snapshot, PAGE_SIZE);
-		if (ret > 0) {
-			error = swap_read_page(handle, data_of(*snapshot));
-			if (error)
-				break;
-			if (!(nr_pages % m))
-				printk("\b\b\b\b%3d%%", nr_pages / m);
-			nr_pages++;
-		}
-	} while (ret > 0);
+	for ( ; ; ) {
+		error = snapshot_write_next(snapshot, PAGE_SIZE);
+		if (error <= 0)
+			break;
+		error = swap_read_page(handle, data_of(*snapshot), &bio);
+		if (error)
+			break;
+		if (snapshot->sync_read)
+			error = wait_on_bio_chain(&bio);
+		if (error)
+			break;
+		if (!(nr_pages % m))
+			printk("\b\b\b\b%3d%%", nr_pages / m);
+		nr_pages++;
+	}
+	err2 = wait_on_bio_chain(&bio);
 	do_gettimeofday(&stop);
+	if (!error)
+		error = err2;
 	if (!error) {
 		printk("\b\b\b\bdone\n");
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
-	show_speed(&start, &stop, nr_pages, "Read");
+	show_speed(&start, &stop, nr_to_read, "Read");
 	return error;
 }
 
@@ -571,7 +573,7 @@ int swsusp_read(void)
 	header = (struct swsusp_info *)data_of(snapshot);
 	error = get_swap_reader(&handle, swsusp_header.image);
 	if (!error)
-		error = swap_read_page(&handle, header);
+		error = swap_read_page(&handle, header, NULL);
 	if (!error)
 		error = load_image(&handle, &snapshot, header->pages - 1);
 	release_swap_reader(&handle);
@@ -597,7 +599,7 @@ int swsusp_check(void)
 	if (!IS_ERR(resume_bdev)) {
 		set_blocksize(resume_bdev, PAGE_SIZE);
 		memset(&swsusp_header, 0, sizeof(swsusp_header));
-		if ((error = bio_read_page(0, &swsusp_header)))
+		if ((error = bio_read_page(0, &swsusp_header, NULL)))
 			return error;
 		if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
 			memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
diff --git a/mm/page_io.c b/mm/page_io.c
index f46a9862b7ef..d4840ecbf8f9 100644
--- a/mm/page_io.c
+++ b/mm/page_io.c
@@ -74,7 +74,7 @@ static int end_swap_bio_write(struct bio *bio, unsigned int bytes_done, int err)
 	return 0;
 }
 
-static int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err)
+int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err)
 {
 	const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags);
 	struct page *page = bio->bi_io_vec[0].bv_page;
-- 
cgit v1.2.1


From ae83c5eef59ffe6eb61110b8c8fd1ea3e0881712 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:45 -0700
Subject: [PATCH] swsusp: clean up browsing of pfns

Clean up some loops over pfns for each zone in snapshot.c: reduce the
number of additions to perform, rework detection of saveable pages and make
the code a bit less difficult to understand, hopefully.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 62 +++++++++++++++++++++++++------------------------
 1 file changed, 32 insertions(+), 30 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 591301ae8b7d..979096c27773 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -156,7 +156,7 @@ static inline int save_highmem(void) {return 0;}
 static inline int restore_highmem(void) {return 0;}
 #endif
 
-static int pfn_is_nosave(unsigned long pfn)
+static inline int pfn_is_nosave(unsigned long pfn)
 {
 	unsigned long nosave_begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT;
 	unsigned long nosave_end_pfn = PAGE_ALIGN(__pa(&__nosave_end)) >> PAGE_SHIFT;
@@ -167,43 +167,43 @@ static int pfn_is_nosave(unsigned long pfn)
  *	saveable - Determine whether a page should be cloned or not.
  *	@pfn:	The page
  *
- *	We save a page if it's Reserved, and not in the range of pages
- *	statically defined as 'unsaveable', or if it isn't reserved, and
- *	isn't part of a free chunk of pages.
+ *	We save a page if it isn't Nosave, and is not in the range of pages
+ *	statically defined as 'unsaveable', and it
+ *	isn't a part of a free chunk of pages.
  */
 
-static int saveable(struct zone *zone, unsigned long *zone_pfn)
+static struct page *saveable_page(unsigned long pfn)
 {
-	unsigned long pfn = *zone_pfn + zone->zone_start_pfn;
 	struct page *page;
 
 	if (!pfn_valid(pfn))
-		return 0;
+		return NULL;
 
 	page = pfn_to_page(pfn);
-	BUG_ON(PageReserved(page) && PageNosave(page));
+
 	if (PageNosave(page))
-		return 0;
+		return NULL;
 	if (PageReserved(page) && pfn_is_nosave(pfn))
-		return 0;
+		return NULL;
 	if (PageNosaveFree(page))
-		return 0;
+		return NULL;
 
-	return 1;
+	return page;
 }
 
 unsigned int count_data_pages(void)
 {
 	struct zone *zone;
-	unsigned long zone_pfn;
+	unsigned long pfn, max_zone_pfn;
 	unsigned int n = 0;
 
 	for_each_zone (zone) {
 		if (is_highmem(zone))
 			continue;
 		mark_free_pages(zone);
-		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-			n += saveable(zone, &zone_pfn);
+		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+			n += !!saveable_page(pfn);
 	}
 	return n;
 }
@@ -211,7 +211,7 @@ unsigned int count_data_pages(void)
 static void copy_data_pages(struct pbe *pblist)
 {
 	struct zone *zone;
-	unsigned long zone_pfn;
+	unsigned long pfn, max_zone_pfn;
 	struct pbe *pbe, *p;
 
 	pbe = pblist;
@@ -224,13 +224,14 @@ static void copy_data_pages(struct pbe *pblist)
 			SetPageNosaveFree(virt_to_page(p));
 		for_each_pbe (p, pblist)
 			SetPageNosaveFree(virt_to_page(p->address));
-		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) {
-			if (saveable(zone, &zone_pfn)) {
-				struct page *page;
+		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) {
+			struct page *page = saveable_page(pfn);
+
+			if (page) {
 				long *src, *dst;
 				int n;
 
-				page = pfn_to_page(zone_pfn + zone->zone_start_pfn);
 				BUG_ON(!pbe);
 				pbe->orig_address = (unsigned long)page_address(page);
 				/* copy_page and memcpy are not usable for copying task structs. */
@@ -383,13 +384,14 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
 void swsusp_free(void)
 {
 	struct zone *zone;
-	unsigned long zone_pfn;
+	unsigned long pfn, max_zone_pfn;
 
 	for_each_zone(zone) {
-		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-			if (pfn_valid(zone_pfn + zone->zone_start_pfn)) {
-				struct page *page;
-				page = pfn_to_page(zone_pfn + zone->zone_start_pfn);
+		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+			if (pfn_valid(pfn)) {
+				struct page *page = pfn_to_page(pfn);
+
 				if (PageNosave(page) && PageNosaveFree(page)) {
 					ClearPageNosave(page);
 					ClearPageNosaveFree(page);
@@ -598,7 +600,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 static int mark_unsafe_pages(struct pbe *pblist)
 {
 	struct zone *zone;
-	unsigned long zone_pfn;
+	unsigned long pfn, max_zone_pfn;
 	struct pbe *p;
 
 	if (!pblist) /* a sanity check */
@@ -606,10 +608,10 @@ static int mark_unsafe_pages(struct pbe *pblist)
 
 	/* Clear page flags */
 	for_each_zone (zone) {
-		for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-			if (pfn_valid(zone_pfn + zone->zone_start_pfn))
-				ClearPageNosaveFree(pfn_to_page(zone_pfn +
-					zone->zone_start_pfn));
+		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+			if (pfn_valid(pfn))
+				ClearPageNosaveFree(pfn_to_page(pfn));
 	}
 
 	/* Mark orig addresses */
-- 
cgit v1.2.1


From fb13a28b0f5ada60861868c4fa48a12bd0cb8dea Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:46 -0700
Subject: [PATCH] swsusp: struct snapshot_handle cleanup

Add comments describing struct snapshot_handle and its members, change the
confusing name of its member 'page' to 'cur'.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/power.h    | 64 ++++++++++++++++++++++++++++++++++++++++++-------
 kernel/power/snapshot.c | 40 +++++++++++++++----------------
 2 files changed, 76 insertions(+), 28 deletions(-)

diff --git a/kernel/power/power.h b/kernel/power/power.h
index 59ce712f082d..1cefcf87a694 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -50,17 +50,65 @@ extern asmlinkage int swsusp_arch_resume(void);
 
 extern unsigned int count_data_pages(void);
 
+/**
+ *	Auxiliary structure used for reading the snapshot image data and
+ *	metadata from and writing them to the list of page backup entries
+ *	(PBEs) which is the main data structure of swsusp.
+ *
+ *	Using struct snapshot_handle we can transfer the image, including its
+ *	metadata, as a continuous sequence of bytes with the help of
+ *	snapshot_read_next() and snapshot_write_next().
+ *
+ *	The code that writes the image to a storage or transfers it to
+ *	the user land is required to use snapshot_read_next() for this
+ *	purpose and it should not make any assumptions regarding the internal
+ *	structure of the image.  Similarly, the code that reads the image from
+ *	a storage or transfers it from the user land is required to use
+ *	snapshot_write_next().
+ *
+ *	This may allow us to change the internal structure of the image
+ *	in the future with considerably less effort.
+ */
+
 struct snapshot_handle {
-	loff_t		offset;
-	unsigned int	page;
-	unsigned int	page_offset;
-	unsigned int	prev;
-	struct pbe	*pbe, *last_pbe;
-	void		*buffer;
-	unsigned int	buf_offset;
-	int		sync_read;
+	loff_t		offset;	/* number of the last byte ready for reading
+				 * or writing in the sequence
+				 */
+	unsigned int	cur;	/* number of the block of PAGE_SIZE bytes the
+				 * next operation will refer to (ie. current)
+				 */
+	unsigned int	cur_offset;	/* offset with respect to the current
+					 * block (for the next operation)
+					 */
+	unsigned int	prev;	/* number of the block of PAGE_SIZE bytes that
+				 * was the current one previously
+				 */
+	struct pbe	*pbe;	/* PBE that corresponds to 'buffer' */
+	struct pbe	*last_pbe;	/* When the image is restored (eg. read
+					 * from disk) we can store some image
+					 * data directly in the page frames
+					 * in which they were before suspend.
+					 * In such a case the PBEs that
+					 * correspond to them will be unused.
+					 * This is the last PBE, so far, that
+					 * does not correspond to such data.
+					 */
+	void		*buffer;	/* address of the block to read from
+					 * or write to
+					 */
+	unsigned int	buf_offset;	/* location to read from or write to,
+					 * given as a displacement from 'buffer'
+					 */
+	int		sync_read;	/* Set to one to notify the caller of
+					 * snapshot_write_next() that it may
+					 * need to call wait_on_bio_chain()
+					 */
 };
 
+/* This macro returns the address from/to which the caller of
+ * snapshot_read_next()/snapshot_write_next() is allowed to
+ * read/write data after the function returns
+ */
 #define data_of(handle)	((handle).buffer + (handle).buf_offset)
 
 extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 979096c27773..81fe8de9e604 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -555,7 +555,7 @@ static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pb
 
 int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 {
-	if (handle->page > nr_meta_pages + nr_copy_pages)
+	if (handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
@@ -568,8 +568,8 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 		handle->buffer = buffer;
 		handle->pbe = pagedir_nosave;
 	}
-	if (handle->prev < handle->page) {
-		if (handle->page <= nr_meta_pages) {
+	if (handle->prev < handle->cur) {
+		if (handle->cur <= nr_meta_pages) {
 			handle->pbe = pack_orig_addresses(buffer, handle->pbe);
 			if (!handle->pbe)
 				handle->pbe = pagedir_nosave;
@@ -577,15 +577,15 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 			handle->buffer = (void *)handle->pbe->address;
 			handle->pbe = handle->pbe->next;
 		}
-		handle->prev = handle->page;
+		handle->prev = handle->cur;
 	}
-	handle->buf_offset = handle->page_offset;
-	if (handle->page_offset + count >= PAGE_SIZE) {
-		count = PAGE_SIZE - handle->page_offset;
-		handle->page_offset = 0;
-		handle->page++;
+	handle->buf_offset = handle->cur_offset;
+	if (handle->cur_offset + count >= PAGE_SIZE) {
+		count = PAGE_SIZE - handle->cur_offset;
+		handle->cur_offset = 0;
+		handle->cur++;
 	} else {
-		handle->page_offset += count;
+		handle->cur_offset += count;
 	}
 	handle->offset += count;
 	return count;
@@ -820,7 +820,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 {
 	int error = 0;
 
-	if (handle->prev && handle->page > nr_meta_pages + nr_copy_pages)
+	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
@@ -831,7 +831,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 	if (!handle->offset)
 		handle->buffer = buffer;
 	handle->sync_read = 1;
-	if (handle->prev < handle->page) {
+	if (handle->prev < handle->cur) {
 		if (!handle->prev) {
 			error = load_header(handle,
 					(struct swsusp_info *)buffer);
@@ -854,15 +854,15 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 			handle->buffer = get_buffer(handle);
 			handle->sync_read = 0;
 		}
-		handle->prev = handle->page;
+		handle->prev = handle->cur;
 	}
-	handle->buf_offset = handle->page_offset;
-	if (handle->page_offset + count >= PAGE_SIZE) {
-		count = PAGE_SIZE - handle->page_offset;
-		handle->page_offset = 0;
-		handle->page++;
+	handle->buf_offset = handle->cur_offset;
+	if (handle->cur_offset + count >= PAGE_SIZE) {
+		count = PAGE_SIZE - handle->cur_offset;
+		handle->cur_offset = 0;
+		handle->cur++;
 	} else {
-		handle->page_offset += count;
+		handle->cur_offset += count;
 	}
 	handle->offset += count;
 	return count;
@@ -871,5 +871,5 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 int snapshot_image_loaded(struct snapshot_handle *handle)
 {
 	return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
-		handle->page <= nr_meta_pages + nr_copy_pages);
+		handle->cur <= nr_meta_pages + nr_copy_pages);
 }
-- 
cgit v1.2.1


From e8eff5ac294e12531c4195e0c15a222d3c9015e5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:46 -0700
Subject: [PATCH] Make swsusp avoid memory holes and reserved memory regions on
 x86_64

On x86_64 machines with more than 2 GB of RAM there are large memory gaps
(with no corresponding kernel virtual addresses) and reserved memory
regions between areas of usable physical RAM.  Moreover, if CONFIG_FLATMEM
is set, they appear within the normal zone.  swsusp should not try to save
them, so the corresponding page structs have to be marked as 'nosave'.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Mel Gorman <mel@csn.ul.ie>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/e820.c  | 48 ++++++++++++++++++++++++++++++++++++++++++++++
 arch/x86_64/kernel/setup.c |  1 +
 include/asm-x86_64/e820.h  |  1 +
 3 files changed, 50 insertions(+)

diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index d6d7f731f6f0..708a3cd9a27e 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -16,6 +16,7 @@
 #include <linux/string.h>
 #include <linux/kexec.h>
 #include <linux/module.h>
+#include <linux/mm.h>
 
 #include <asm/pgtable.h>
 #include <asm/page.h>
@@ -297,6 +298,53 @@ void __init e820_reserve_resources(void)
 	}
 }
 
+/* Mark pages corresponding to given address range as nosave */
+static void __init
+e820_mark_nosave_range(unsigned long start, unsigned long end)
+{
+	unsigned long pfn, max_pfn;
+
+	if (start >= end)
+		return;
+
+	printk("Nosave address range: %016lx - %016lx\n", start, end);
+	max_pfn = end >> PAGE_SHIFT;
+	for (pfn = start >> PAGE_SHIFT; pfn < max_pfn; pfn++)
+		if (pfn_valid(pfn))
+			SetPageNosave(pfn_to_page(pfn));
+}
+
+/*
+ * Find the ranges of physical addresses that do not correspond to
+ * e820 RAM areas and mark the corresponding pages as nosave for software
+ * suspend and suspend to RAM.
+ *
+ * This function requires the e820 map to be sorted and without any
+ * overlapping entries and assumes the first e820 area to be RAM.
+ */
+void __init e820_mark_nosave_regions(void)
+{
+	int i;
+	unsigned long paddr;
+
+	paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE);
+	for (i = 1; i < e820.nr_map; i++) {
+		struct e820entry *ei = &e820.map[i];
+
+		if (paddr < ei->addr)
+			e820_mark_nosave_range(paddr,
+					round_up(ei->addr, PAGE_SIZE));
+
+		paddr = round_down(ei->addr + ei->size, PAGE_SIZE);
+		if (ei->type != E820_RAM)
+			e820_mark_nosave_range(round_up(ei->addr, PAGE_SIZE),
+					paddr);
+
+		if (paddr >= (end_pfn << PAGE_SHIFT))
+			break;
+	}
+}
+
 /* 
  * Add a memory region to the kernel e820 map.
  */ 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 34afad704824..4b39f0da17f3 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -689,6 +689,7 @@ void __init setup_arch(char **cmdline_p)
 	 */
 	probe_roms();
 	e820_reserve_resources(); 
+	e820_mark_nosave_regions();
 
 	request_resource(&iomem_resource, &video_ram_resource);
 
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index 670a3388e70a..f65674832318 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -46,6 +46,7 @@ extern void setup_memory_region(void);
 extern void contig_e820_setup(void); 
 extern unsigned long e820_end_of_ram(void);
 extern void e820_reserve_resources(void);
+extern void e820_mark_nosave_regions(void);
 extern void e820_print_map(char *who);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
-- 
cgit v1.2.1


From e3920fb42c8ddfe63befb54d95c0e13eabacea9b Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:48 -0700
Subject: [PATCH] Disable CPU hotplug during suspend

The current suspend code has to be run on one CPU, so we use the CPU
hotplug to take the non-boot CPUs offline on SMP machines.  However, we
should also make sure that these CPUs will not be enabled by someone else
after we have disabled them.

The functions disable_nonboot_cpus() and enable_nonboot_cpus() are moved to
kernel/cpu.c, because they now refer to some stuff in there that should
better be static.  Also it's better if disable_nonboot_cpus() returns an
error instead of panicking if something goes wrong, and
enable_nonboot_cpus() has no reason to panic(), because the CPUs may have
been enabled by the userland before it tries to take them online.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/cpu.h     |   8 +++
 include/linux/suspend.h |   8 ---
 kernel/cpu.c            | 138 +++++++++++++++++++++++++++++++++++++++++-------
 kernel/power/Makefile   |   2 -
 kernel/power/disk.c     |   7 ++-
 kernel/power/main.c     |  10 ++--
 kernel/power/smp.c      |  62 ----------------------
 kernel/power/user.c     |  14 +++--
 8 files changed, 145 insertions(+), 104 deletions(-)
 delete mode 100644 kernel/power/smp.c

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 8fb344a9abd8..3fef7d67aedc 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -89,4 +89,12 @@ int cpu_down(unsigned int cpu);
 static inline int cpu_is_offline(int cpu) { return 0; }
 #endif
 
+#ifdef CONFIG_SUSPEND_SMP
+extern int disable_nonboot_cpus(void);
+extern void enable_nonboot_cpus(void);
+#else
+static inline int disable_nonboot_cpus(void) { return 0; }
+static inline void enable_nonboot_cpus(void) {}
+#endif
+
 #endif /* _LINUX_CPU_H_ */
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 96e31aa64cc7..6e8a06c950f4 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -57,14 +57,6 @@ static inline int software_suspend(void)
 }
 #endif /* CONFIG_PM */
 
-#ifdef CONFIG_SUSPEND_SMP
-extern void disable_nonboot_cpus(void);
-extern void enable_nonboot_cpus(void);
-#else
-static inline void disable_nonboot_cpus(void) {}
-static inline void enable_nonboot_cpus(void) {}
-#endif
-
 void save_processor_state(void);
 void restore_processor_state(void);
 struct saved_context;
diff --git a/kernel/cpu.c b/kernel/cpu.c
index f230f9ae01c2..32c96628463e 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -21,6 +21,11 @@ static DEFINE_MUTEX(cpu_bitmask_lock);
 
 static __cpuinitdata BLOCKING_NOTIFIER_HEAD(cpu_chain);
 
+/* If set, cpu_up and cpu_down will return -EBUSY and do nothing.
+ * Should always be manipulated under cpu_add_remove_lock
+ */
+static int cpu_hotplug_disabled;
+
 #ifdef CONFIG_HOTPLUG_CPU
 
 /* Crappy recursive lock-takers in cpufreq! Complain loudly about idiots */
@@ -108,30 +113,25 @@ static int take_cpu_down(void *unused)
 	return 0;
 }
 
-int cpu_down(unsigned int cpu)
+/* Requires cpu_add_remove_lock to be held */
+static int _cpu_down(unsigned int cpu)
 {
 	int err;
 	struct task_struct *p;
 	cpumask_t old_allowed, tmp;
 
-	mutex_lock(&cpu_add_remove_lock);
-	if (num_online_cpus() == 1) {
-		err = -EBUSY;
-		goto out;
-	}
+	if (num_online_cpus() == 1)
+		return -EBUSY;
 
-	if (!cpu_online(cpu)) {
-		err = -EINVAL;
-		goto out;
-	}
+	if (!cpu_online(cpu))
+		return -EINVAL;
 
 	err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE,
 						(void *)(long)cpu);
 	if (err == NOTIFY_BAD) {
 		printk("%s: attempt to take down CPU %u failed\n",
 				__FUNCTION__, cpu);
-		err = -EINVAL;
-		goto out;
+		return -EINVAL;
 	}
 
 	/* Ensure that we are not runnable on dying cpu */
@@ -179,22 +179,32 @@ out_thread:
 	err = kthread_stop(p);
 out_allowed:
 	set_cpus_allowed(current, old_allowed);
-out:
+	return err;
+}
+
+int cpu_down(unsigned int cpu)
+{
+	int err = 0;
+
+	mutex_lock(&cpu_add_remove_lock);
+	if (cpu_hotplug_disabled)
+		err = -EBUSY;
+	else
+		err = _cpu_down(cpu);
+
 	mutex_unlock(&cpu_add_remove_lock);
 	return err;
 }
 #endif /*CONFIG_HOTPLUG_CPU*/
 
-int __devinit cpu_up(unsigned int cpu)
+/* Requires cpu_add_remove_lock to be held */
+static int __devinit _cpu_up(unsigned int cpu)
 {
 	int ret;
 	void *hcpu = (void *)(long)cpu;
 
-	mutex_lock(&cpu_add_remove_lock);
-	if (cpu_online(cpu) || !cpu_present(cpu)) {
-		ret = -EINVAL;
-		goto out;
-	}
+	if (cpu_online(cpu) || !cpu_present(cpu))
+		return -EINVAL;
 
 	ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu);
 	if (ret == NOTIFY_BAD) {
@@ -219,7 +229,95 @@ out_notify:
 	if (ret != 0)
 		blocking_notifier_call_chain(&cpu_chain,
 				CPU_UP_CANCELED, hcpu);
+
+	return ret;
+}
+
+int __devinit cpu_up(unsigned int cpu)
+{
+	int err = 0;
+
+	mutex_lock(&cpu_add_remove_lock);
+	if (cpu_hotplug_disabled)
+		err = -EBUSY;
+	else
+		err = _cpu_up(cpu);
+
+	mutex_unlock(&cpu_add_remove_lock);
+	return err;
+}
+
+#ifdef CONFIG_SUSPEND_SMP
+static cpumask_t frozen_cpus;
+
+int disable_nonboot_cpus(void)
+{
+	int cpu, first_cpu, error;
+
+	mutex_lock(&cpu_add_remove_lock);
+	first_cpu = first_cpu(cpu_present_map);
+	if (!cpu_online(first_cpu)) {
+		error = _cpu_up(first_cpu);
+		if (error) {
+			printk(KERN_ERR "Could not bring CPU%d up.\n",
+				first_cpu);
+			goto out;
+		}
+	}
+	error = set_cpus_allowed(current, cpumask_of_cpu(first_cpu));
+	if (error) {
+		printk(KERN_ERR "Could not run on CPU%d\n", first_cpu);
+		goto out;
+	}
+	/* We take down all of the non-boot CPUs in one shot to avoid races
+	 * with the userspace trying to use the CPU hotplug at the same time
+	 */
+	cpus_clear(frozen_cpus);
+	printk("Disabling non-boot CPUs ...\n");
+	for_each_online_cpu(cpu) {
+		if (cpu == first_cpu)
+			continue;
+		error = _cpu_down(cpu);
+		if (!error) {
+			cpu_set(cpu, frozen_cpus);
+			printk("CPU%d is down\n", cpu);
+		} else {
+			printk(KERN_ERR "Error taking CPU%d down: %d\n",
+				cpu, error);
+			break;
+		}
+	}
+	if (!error) {
+		BUG_ON(num_online_cpus() > 1);
+		/* Make sure the CPUs won't be enabled by someone else */
+		cpu_hotplug_disabled = 1;
+	} else {
+		printk(KERN_ERR "Non-boot CPUs are not disabled");
+	}
 out:
 	mutex_unlock(&cpu_add_remove_lock);
-	return ret;
+	return error;
+}
+
+void enable_nonboot_cpus(void)
+{
+	int cpu, error;
+
+	/* Allow everyone to use the CPU hotplug again */
+	mutex_lock(&cpu_add_remove_lock);
+	cpu_hotplug_disabled = 0;
+	mutex_unlock(&cpu_add_remove_lock);
+
+	printk("Enabling non-boot CPUs ...\n");
+	for_each_cpu_mask(cpu, frozen_cpus) {
+		error = cpu_up(cpu);
+		if (!error) {
+			printk("CPU%d is up\n", cpu);
+			continue;
+		}
+		printk(KERN_WARNING "Error taking CPU%d up: %d\n",
+			cpu, error);
+	}
+	cpus_clear(frozen_cpus);
 }
+#endif
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index 8d0af3d37a4b..38725f526afc 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -7,6 +7,4 @@ obj-y				:= main.o process.o console.o
 obj-$(CONFIG_PM_LEGACY)		+= pm.o
 obj-$(CONFIG_SOFTWARE_SUSPEND)	+= swsusp.o disk.o snapshot.o swap.o user.o
 
-obj-$(CONFIG_SUSPEND_SMP)	+= smp.o
-
 obj-$(CONFIG_MAGIC_SYSRQ)	+= poweroff.o
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index e13e74067845..7c7b9b65e365 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -18,6 +18,7 @@
 #include <linux/fs.h>
 #include <linux/mount.h>
 #include <linux/pm.h>
+#include <linux/cpu.h>
 
 #include "power.h"
 
@@ -72,7 +73,10 @@ static int prepare_processes(void)
 	int error;
 
 	pm_prepare_console();
-	disable_nonboot_cpus();
+
+	error = disable_nonboot_cpus();
+	if (error)
+		goto enable_cpus;
 
 	if (freeze_processes()) {
 		error = -EBUSY;
@@ -84,6 +88,7 @@ static int prepare_processes(void)
 		return 0;
 thaw:
 	thaw_processes();
+enable_cpus:
 	enable_nonboot_cpus();
 	pm_restore_console();
 	return error;
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6d295c776794..4d403323a7bb 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/pm.h>
 #include <linux/console.h>
+#include <linux/cpu.h>
 
 #include "power.h"
 
@@ -51,7 +52,7 @@ void pm_set_ops(struct pm_ops * ops)
 
 static int suspend_prepare(suspend_state_t state)
 {
-	int error = 0;
+	int error;
 	unsigned int free_pages;
 
 	if (!pm_ops || !pm_ops->enter)
@@ -59,12 +60,9 @@ static int suspend_prepare(suspend_state_t state)
 
 	pm_prepare_console();
 
-	disable_nonboot_cpus();
-
-	if (num_online_cpus() != 1) {
-		error = -EPERM;
+	error = disable_nonboot_cpus();
+	if (error)
 		goto Enable_cpu;
-	}
 
 	if (freeze_processes()) {
 		error = -EAGAIN;
diff --git a/kernel/power/smp.c b/kernel/power/smp.c
deleted file mode 100644
index 5957312b2d68..000000000000
--- a/kernel/power/smp.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * drivers/power/smp.c - Functions for stopping other CPUs.
- *
- * Copyright 2004 Pavel Machek <pavel@suse.cz>
- * Copyright (C) 2002-2003 Nigel Cunningham <ncunningham@clear.net.nz>
- *
- * This file is released under the GPLv2.
- */
-
-#undef DEBUG
-
-#include <linux/smp_lock.h>
-#include <linux/interrupt.h>
-#include <linux/suspend.h>
-#include <linux/module.h>
-#include <linux/cpu.h>
-#include <asm/atomic.h>
-#include <asm/tlbflush.h>
-
-/* This is protected by pm_sem semaphore */
-static cpumask_t frozen_cpus;
-
-void disable_nonboot_cpus(void)
-{
-	int cpu, error;
-
-	error = 0;
-	cpus_clear(frozen_cpus);
-	printk("Freezing cpus ...\n");
-	for_each_online_cpu(cpu) {
-		if (cpu == 0)
-			continue;
-		error = cpu_down(cpu);
-		if (!error) {
-			cpu_set(cpu, frozen_cpus);
-			printk("CPU%d is down\n", cpu);
-			continue;
-		}
-		printk("Error taking cpu %d down: %d\n", cpu, error);
-	}
-	BUG_ON(raw_smp_processor_id() != 0);
-	if (error)
-		panic("cpus not sleeping");
-}
-
-void enable_nonboot_cpus(void)
-{
-	int cpu, error;
-
-	printk("Thawing cpus ...\n");
-	for_each_cpu_mask(cpu, frozen_cpus) {
-		error = cpu_up(cpu);
-		if (!error) {
-			printk("CPU%d is up\n", cpu);
-			continue;
-		}
-		printk("Error taking cpu %d up: %d\n", cpu, error);
-		panic("Not enough cpus");
-	}
-	cpus_clear(frozen_cpus);
-}
-
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 3f1539fbe48a..0ef5e4ba39e5 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -19,6 +19,7 @@
 #include <linux/swapops.h>
 #include <linux/pm.h>
 #include <linux/fs.h>
+#include <linux/cpu.h>
 
 #include <asm/uaccess.h>
 
@@ -139,12 +140,15 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		if (data->frozen)
 			break;
 		down(&pm_sem);
-		disable_nonboot_cpus();
-		if (freeze_processes()) {
-			thaw_processes();
-			enable_nonboot_cpus();
-			error = -EBUSY;
+		error = disable_nonboot_cpus();
+		if (!error) {
+			error = freeze_processes();
+			if (error) {
+				thaw_processes();
+				error = -EBUSY;
+			}
 		}
+		enable_nonboot_cpus();
 		up(&pm_sem);
 		if (!error)
 			data->frozen = 1;
-- 
cgit v1.2.1


From f623f0db8e6aa86a37be86167e4ff478821a9f4f Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:49 -0700
Subject: [PATCH] swsusp: Fix mark_free_pages

Clean up mm/page_alloc.c#mark_free_pages() and make it avoid clearing
PageNosaveFree for PageNosave pages.  This allows us to get rid of an ugly
hack in kernel/power/snapshot.c#copy_data_pages().

Additionally, the page-copying loop in copy_data_pages() is moved to an
inline function.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 27 +++++++++++++--------------
 mm/page_alloc.c         | 24 ++++++++++++++++--------
 2 files changed, 29 insertions(+), 22 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 81fe8de9e604..4afb7900002b 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -208,37 +208,36 @@ unsigned int count_data_pages(void)
 	return n;
 }
 
+static inline void copy_data_page(long *dst, long *src)
+{
+	int n;
+
+	/* copy_page and memcpy are not usable for copying task structs. */
+	for (n = PAGE_SIZE / sizeof(long); n; n--)
+		*dst++ = *src++;
+}
+
 static void copy_data_pages(struct pbe *pblist)
 {
 	struct zone *zone;
 	unsigned long pfn, max_zone_pfn;
-	struct pbe *pbe, *p;
+	struct pbe *pbe;
 
 	pbe = pblist;
 	for_each_zone (zone) {
 		if (is_highmem(zone))
 			continue;
 		mark_free_pages(zone);
-		/* This is necessary for swsusp_free() */
-		for_each_pb_page (p, pblist)
-			SetPageNosaveFree(virt_to_page(p));
-		for_each_pbe (p, pblist)
-			SetPageNosaveFree(virt_to_page(p->address));
 		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
 		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) {
 			struct page *page = saveable_page(pfn);
 
 			if (page) {
-				long *src, *dst;
-				int n;
+				void *ptr = page_address(page);
 
 				BUG_ON(!pbe);
-				pbe->orig_address = (unsigned long)page_address(page);
-				/* copy_page and memcpy are not usable for copying task structs. */
-				dst = (long *)pbe->address;
-				src = (long *)pbe->orig_address;
-				for (n = PAGE_SIZE / sizeof(long); n; n--)
-					*dst++ = *src++;
+				copy_data_page((void *)pbe->address, ptr);
+				pbe->orig_address = (unsigned long)ptr;
 				pbe = pbe->next;
 			}
 		}
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 51070b6d593f..9810f0a60db7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -694,7 +694,8 @@ static void __drain_pages(unsigned int cpu)
 
 void mark_free_pages(struct zone *zone)
 {
-	unsigned long zone_pfn, flags;
+	unsigned long pfn, max_zone_pfn;
+	unsigned long flags;
 	int order;
 	struct list_head *curr;
 
@@ -702,18 +703,25 @@ void mark_free_pages(struct zone *zone)
 		return;
 
 	spin_lock_irqsave(&zone->lock, flags);
-	for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
-		ClearPageNosaveFree(pfn_to_page(zone_pfn + zone->zone_start_pfn));
+
+	max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
+	for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+		if (pfn_valid(pfn)) {
+			struct page *page = pfn_to_page(pfn);
+
+			if (!PageNosave(page))
+				ClearPageNosaveFree(page);
+		}
 
 	for (order = MAX_ORDER - 1; order >= 0; --order)
 		list_for_each(curr, &zone->free_area[order].free_list) {
-			unsigned long start_pfn, i;
+			unsigned long i;
 
-			start_pfn = page_to_pfn(list_entry(curr, struct page, lru));
+			pfn = page_to_pfn(list_entry(curr, struct page, lru));
+			for (i = 0; i < (1UL << order); i++)
+				SetPageNosaveFree(pfn_to_page(pfn + i));
+		}
 
-			for (i=0; i < (1<<order); i++)
-				SetPageNosaveFree(pfn_to_page(start_pfn+i));
-	}
 	spin_unlock_irqrestore(&zone->lock, flags);
 }
 
-- 
cgit v1.2.1


From f6143aa60ed71e58578bc92cc64d98158a694d99 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:50 -0700
Subject: [PATCH] swsusp: Reorder memory-allocating functions

Move some functions in kernel/power/snapshot.c to a better place (in the
same file) and introduce free_image_page() (will be necessary in the
future).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 93 ++++++++++++++++++++++++++++---------------------
 1 file changed, 53 insertions(+), 40 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 4afb7900002b..7ad0c0465524 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -156,6 +156,58 @@ static inline int save_highmem(void) {return 0;}
 static inline int restore_highmem(void) {return 0;}
 #endif
 
+/**
+ *	@safe_needed - on resume, for storing the PBE list and the image,
+ *	we can only use memory pages that do not conflict with the pages
+ *	used before suspend.
+ *
+ *	The unsafe pages are marked with the PG_nosave_free flag
+ *	and we count them using unsafe_pages
+ */
+
+static unsigned int unsafe_pages;
+
+static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
+{
+	void *res;
+
+	res = (void *)get_zeroed_page(gfp_mask);
+	if (safe_needed)
+		while (res && PageNosaveFree(virt_to_page(res))) {
+			/* The page is unsafe, mark it for swsusp_free() */
+			SetPageNosave(virt_to_page(res));
+			unsafe_pages++;
+			res = (void *)get_zeroed_page(gfp_mask);
+		}
+	if (res) {
+		SetPageNosave(virt_to_page(res));
+		SetPageNosaveFree(virt_to_page(res));
+	}
+	return res;
+}
+
+unsigned long get_safe_page(gfp_t gfp_mask)
+{
+	return (unsigned long)alloc_image_page(gfp_mask, 1);
+}
+
+/**
+ *	free_image_page - free page represented by @addr, allocated with
+ *	alloc_image_page (page flags set by it must be cleared)
+ */
+
+static inline void free_image_page(void *addr, int clear_nosave_free)
+{
+	ClearPageNosave(virt_to_page(addr));
+	if (clear_nosave_free)
+		ClearPageNosaveFree(virt_to_page(addr));
+	free_page((unsigned long)addr);
+}
+
+/**
+ *	pfn_is_nosave - check if given pfn is in the 'nosave' section
+ */
+
 static inline int pfn_is_nosave(unsigned long pfn)
 {
 	unsigned long nosave_begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT;
@@ -245,7 +297,6 @@ static void copy_data_pages(struct pbe *pblist)
 	BUG_ON(pbe);
 }
 
-
 /**
  *	free_pagedir - free pages allocated with alloc_pagedir()
  */
@@ -256,10 +307,7 @@ static void free_pagedir(struct pbe *pblist, int clear_nosave_free)
 
 	while (pblist) {
 		pbe = (pblist + PB_PAGE_SKIP)->next;
-		ClearPageNosave(virt_to_page(pblist));
-		if (clear_nosave_free)
-			ClearPageNosaveFree(virt_to_page(pblist));
-		free_page((unsigned long)pblist);
+		free_image_page(pblist, clear_nosave_free);
 		pblist = pbe;
 	}
 }
@@ -303,41 +351,6 @@ static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages)
 	}
 }
 
-static unsigned int unsafe_pages;
-
-/**
- *	@safe_needed - on resume, for storing the PBE list and the image,
- *	we can only use memory pages that do not conflict with the pages
- *	used before suspend.
- *
- *	The unsafe pages are marked with the PG_nosave_free flag
- *	and we count them using unsafe_pages
- */
-
-static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
-{
-	void *res;
-
-	res = (void *)get_zeroed_page(gfp_mask);
-	if (safe_needed)
-		while (res && PageNosaveFree(virt_to_page(res))) {
-			/* The page is unsafe, mark it for swsusp_free() */
-			SetPageNosave(virt_to_page(res));
-			unsafe_pages++;
-			res = (void *)get_zeroed_page(gfp_mask);
-		}
-	if (res) {
-		SetPageNosave(virt_to_page(res));
-		SetPageNosaveFree(virt_to_page(res));
-	}
-	return res;
-}
-
-unsigned long get_safe_page(gfp_t gfp_mask)
-{
-	return (unsigned long)alloc_image_page(gfp_mask, 1);
-}
-
 /**
  *	alloc_pagedir - Allocate the page directory.
  *
-- 
cgit v1.2.1


From cd560bb2f9e2cd451bb3942af43da19632ba4a8e Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:50 -0700
Subject: [PATCH] swsusp: Fix alloc_pagedir

Get rid of the FIXME in kernel/power/snapshot.c#alloc_pagedir() and
simplify the functions called by it.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 32 +++++++++++++++++---------------
 1 file changed, 17 insertions(+), 15 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 7ad0c0465524..4ca372f2bc14 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -316,12 +316,12 @@ static void free_pagedir(struct pbe *pblist, int clear_nosave_free)
  *	fill_pb_page - Create a list of PBEs on a given memory page
  */
 
-static inline void fill_pb_page(struct pbe *pbpage)
+static inline void fill_pb_page(struct pbe *pbpage, unsigned int n)
 {
 	struct pbe *p;
 
 	p = pbpage;
-	pbpage += PB_PAGE_SKIP;
+	pbpage += n - 1;
 	do
 		p->next = p + 1;
 	while (++p < pbpage);
@@ -330,24 +330,26 @@ static inline void fill_pb_page(struct pbe *pbpage)
 /**
  *	create_pbe_list - Create a list of PBEs on top of a given chain
  *	of memory pages allocated with alloc_pagedir()
+ *
+ *	This function assumes that pages allocated by alloc_image_page() will
+ *	always be zeroed.
  */
 
 static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages)
 {
-	struct pbe *pbpage, *p;
+	struct pbe *pbpage;
 	unsigned int num = PBES_PER_PAGE;
 
 	for_each_pb_page (pbpage, pblist) {
 		if (num >= nr_pages)
 			break;
 
-		fill_pb_page(pbpage);
+		fill_pb_page(pbpage, PBES_PER_PAGE);
 		num += PBES_PER_PAGE;
 	}
 	if (pbpage) {
-		for (num -= PBES_PER_PAGE - 1, p = pbpage; num < nr_pages; p++, num++)
-			p->next = p + 1;
-		p->next = NULL;
+		num -= PBES_PER_PAGE;
+		fill_pb_page(pbpage, nr_pages - num);
 	}
 }
 
@@ -374,17 +376,17 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
 		return NULL;
 
 	pblist = alloc_image_page(gfp_mask, safe_needed);
-	/* FIXME: rewrite this ugly loop */
-	for (pbe = pblist, num = PBES_PER_PAGE; pbe && num < nr_pages;
-        		pbe = pbe->next, num += PBES_PER_PAGE) {
+	pbe = pblist;
+	for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) {
+		if (!pbe) {
+			free_pagedir(pblist, 1);
+			return NULL;
+		}
 		pbe += PB_PAGE_SKIP;
 		pbe->next = alloc_image_page(gfp_mask, safe_needed);
+		pbe = pbe->next;
 	}
-	if (!pbe) { /* get_zeroed_page() failed */
-		free_pagedir(pblist, 1);
-		pblist = NULL;
-        } else
-		create_pbe_list(pblist, nr_pages);
+	create_pbe_list(pblist, nr_pages);
 	return pblist;
 }
 
-- 
cgit v1.2.1


From dcbb5a54f6e3984efa24772394f2225b11495c55 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:51 -0700
Subject: [PATCH] swsusp: clean up suspend header

Remove some things that are no longer used or defined elsewhere from suspend.h
and make the inline version of software_suspend() return the right error code.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/suspend.h | 15 +++------------
 1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 6e8a06c950f4..c11cacf1a13b 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -10,11 +10,11 @@
 #include <linux/pm.h>
 
 /* page backup entry */
-typedef struct pbe {
+struct pbe {
 	unsigned long address;		/* address of the copy */
 	unsigned long orig_address;	/* original address of page */
 	struct pbe *next;
-} suspend_pagedir_t;
+};
 
 #define for_each_pbe(pbe, pblist) \
 	for (pbe = pblist ; pbe ; pbe = pbe->next)
@@ -25,15 +25,6 @@ typedef struct pbe {
 #define for_each_pb_page(pbe, pblist) \
 	for (pbe = pblist ; pbe ; pbe = (pbe+PB_PAGE_SKIP)->next)
 
-
-#define SWAP_FILENAME_MAXLENGTH	32
-
-
-extern dev_t swsusp_resume_device;
-   
-/* mm/vmscan.c */
-extern int shrink_mem(void);
-
 /* mm/page_alloc.c */
 extern void drain_local_pages(void);
 extern void mark_free_pages(struct zone *zone);
@@ -53,7 +44,7 @@ static inline void pm_restore_console(void) {}
 static inline int software_suspend(void)
 {
 	printk("Warning: fake suspend called\n");
-	return -EPERM;
+	return -ENOSYS;
 }
 #endif /* CONFIG_PM */
 
-- 
cgit v1.2.1


From 75534b50cc658e951bcb213c2763c81e9f7b0b48 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:52 -0700
Subject: [PATCH] Change the name of pagedir_nosave

The name of the pagedir_nosave variable does not make sense any more, so it
seems reasonable to change it to something more meaningful.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/power/swsusp.S         |  2 +-
 arch/powerpc/kernel/swsusp_32.S  |  4 ++--
 arch/x86_64/kernel/suspend_asm.S |  2 +-
 kernel/power/power.h             |  2 --
 kernel/power/snapshot.c          | 26 ++++++++++++++------------
 5 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/arch/i386/power/swsusp.S b/arch/i386/power/swsusp.S
index c893b897217f..8a2b50a0aaad 100644
--- a/arch/i386/power/swsusp.S
+++ b/arch/i386/power/swsusp.S
@@ -32,7 +32,7 @@ ENTRY(swsusp_arch_resume)
 	movl	$swsusp_pg_dir-__PAGE_OFFSET, %ecx
 	movl	%ecx, %cr3
 
-	movl	pagedir_nosave, %edx
+	movl	restore_pblist, %edx
 	.p2align 4,,7
 
 copy_loop:
diff --git a/arch/powerpc/kernel/swsusp_32.S b/arch/powerpc/kernel/swsusp_32.S
index 7369f9a6ad25..69e8f86aa4f8 100644
--- a/arch/powerpc/kernel/swsusp_32.S
+++ b/arch/powerpc/kernel/swsusp_32.S
@@ -159,8 +159,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC)
 	isync
 
 	/* Load ptr the list of pages to copy in r3 */
-	lis	r11,(pagedir_nosave - KERNELBASE)@h
-	ori	r11,r11,pagedir_nosave@l
+	lis	r11,(restore_pblist - KERNELBASE)@h
+	ori	r11,r11,restore_pblist@l
 	lwz	r10,0(r11)
 
 	/* Copy the pages. This is a very basic implementation, to
diff --git a/arch/x86_64/kernel/suspend_asm.S b/arch/x86_64/kernel/suspend_asm.S
index 320b6fb00cca..bfbe00763c68 100644
--- a/arch/x86_64/kernel/suspend_asm.S
+++ b/arch/x86_64/kernel/suspend_asm.S
@@ -54,7 +54,7 @@ ENTRY(restore_image)
 	movq	%rcx, %cr3;
 	movq	%rax, %cr4;  # turn PGE back on
 
-	movq	pagedir_nosave(%rip), %rdx
+	movq	restore_pblist(%rip), %rdx
 loop:
 	testq	%rdx, %rdx
 	jz	done
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 1cefcf87a694..e18ba207e784 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -38,8 +38,6 @@ extern struct subsystem power_subsys;
 /* References to section boundaries */
 extern const void __nosave_begin, __nosave_end;
 
-extern struct pbe *pagedir_nosave;
-
 /* Preferred image size in bytes (default 500 MB) */
 extern unsigned long image_size;
 extern int in_suspend;
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 4ca372f2bc14..1d276b3ae152 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -34,7 +34,9 @@
 
 #include "power.h"
 
-struct pbe *pagedir_nosave;
+/* List of PBEs used for creating and restoring the suspend image */
+struct pbe *restore_pblist;
+
 static unsigned int nr_copy_pages;
 static unsigned int nr_meta_pages;
 static unsigned long *buffer;
@@ -415,7 +417,7 @@ void swsusp_free(void)
 	}
 	nr_copy_pages = 0;
 	nr_meta_pages = 0;
-	pagedir_nosave = NULL;
+	restore_pblist = NULL;
 	buffer = NULL;
 }
 
@@ -490,15 +492,15 @@ asmlinkage int swsusp_save(void)
 		return -ENOMEM;
 	}
 
-	pagedir_nosave = swsusp_alloc(nr_pages);
-	if (!pagedir_nosave)
+	restore_pblist = swsusp_alloc(nr_pages);
+	if (!restore_pblist)
 		return -ENOMEM;
 
 	/* During allocating of suspend pagedir, new cold pages may appear.
 	 * Kill them.
 	 */
 	drain_local_pages();
-	copy_data_pages(pagedir_nosave);
+	copy_data_pages(restore_pblist);
 
 	/*
 	 * End of critical section. From now on, we can write to memory,
@@ -580,13 +582,13 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 	if (!handle->offset) {
 		init_header((struct swsusp_info *)buffer);
 		handle->buffer = buffer;
-		handle->pbe = pagedir_nosave;
+		handle->pbe = restore_pblist;
 	}
 	if (handle->prev < handle->cur) {
 		if (handle->cur <= nr_meta_pages) {
 			handle->pbe = pack_orig_addresses(buffer, handle->pbe);
 			if (!handle->pbe)
-				handle->pbe = pagedir_nosave;
+				handle->pbe = restore_pblist;
 		} else {
 			handle->buffer = (void *)handle->pbe->address;
 			handle->pbe = handle->pbe->next;
@@ -689,7 +691,7 @@ static int load_header(struct snapshot_handle *handle,
 		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0);
 		if (!pblist)
 			return -ENOMEM;
-		pagedir_nosave = pblist;
+		restore_pblist = pblist;
 		handle->pbe = pblist;
 		nr_copy_pages = info->image_pages;
 		nr_meta_pages = info->pages - info->image_pages - 1;
@@ -716,7 +718,7 @@ static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
 
 /**
  *	prepare_image - use metadata contained in the PBE list
- *	pointed to by pagedir_nosave to mark the pages that will
+ *	pointed to by restore_pblist to mark the pages that will
  *	be overwritten in the process of restoring the system
  *	memory state from the image ("unsafe" pages) and allocate
  *	memory for the image
@@ -741,7 +743,7 @@ static int prepare_image(struct snapshot_handle *handle)
 	unsigned int nr_pages = nr_copy_pages;
 	struct pbe *p, *pblist = NULL;
 
-	p = pagedir_nosave;
+	p = restore_pblist;
 	error = mark_unsafe_pages(p);
 	if (!error) {
 		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1);
@@ -773,7 +775,7 @@ static int prepare_image(struct snapshot_handle *handle)
 		}
 	}
 	if (!error) {
-		pagedir_nosave = pblist;
+		restore_pblist = pblist;
 	} else {
 		handle->pbe = NULL;
 		swsusp_free();
@@ -858,7 +860,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 				error = prepare_image(handle);
 				if (error)
 					return error;
-				handle->pbe = pagedir_nosave;
+				handle->pbe = restore_pblist;
 				handle->last_pbe = NULL;
 				handle->buffer = get_buffer(handle);
 				handle->sync_read = 0;
-- 
cgit v1.2.1


From 0bcd888d64684f896ffa70c1d16a42b00753c184 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:52 -0700
Subject: [PATCH] swsusp: Introduce some helpful constants

Introduce some constants that hopefully will help improve the readability of
code in kernel/power/snapshot.c.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 1d276b3ae152..d0d691f976d8 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -167,6 +167,11 @@ static inline int restore_highmem(void) {return 0;}
  *	and we count them using unsafe_pages
  */
 
+#define PG_ANY		0
+#define PG_SAFE		1
+#define PG_UNSAFE_CLEAR	1
+#define PG_UNSAFE_KEEP	0
+
 static unsigned int unsafe_pages;
 
 static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
@@ -190,7 +195,7 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 
 unsigned long get_safe_page(gfp_t gfp_mask)
 {
-	return (unsigned long)alloc_image_page(gfp_mask, 1);
+	return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE);
 }
 
 /**
@@ -381,7 +386,7 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
 	pbe = pblist;
 	for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) {
 		if (!pbe) {
-			free_pagedir(pblist, 1);
+			free_pagedir(pblist, PG_UNSAFE_CLEAR);
 			return NULL;
 		}
 		pbe += PB_PAGE_SKIP;
@@ -458,12 +463,13 @@ static struct pbe *swsusp_alloc(unsigned int nr_pages)
 {
 	struct pbe *pblist;
 
-	if (!(pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, 0))) {
+	pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, PG_ANY);
+	if (!pblist) {
 		printk(KERN_ERR "suspend: Allocating pagedir failed.\n");
 		return NULL;
 	}
 
-	if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, 0)) {
+	if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, PG_ANY)) {
 		printk(KERN_ERR "suspend: Allocating image pages failed.\n");
 		swsusp_free();
 		return NULL;
@@ -575,7 +581,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 		return 0;
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
-		buffer = alloc_image_page(GFP_ATOMIC, 0);
+		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
 		if (!buffer)
 			return -ENOMEM;
 	}
@@ -688,7 +694,7 @@ static int load_header(struct snapshot_handle *handle,
 
 	error = check_header(info);
 	if (!error) {
-		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0);
+		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY);
 		if (!pblist)
 			return -ENOMEM;
 		restore_pblist = pblist;
@@ -746,10 +752,10 @@ static int prepare_image(struct snapshot_handle *handle)
 	p = restore_pblist;
 	error = mark_unsafe_pages(p);
 	if (!error) {
-		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1);
+		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE);
 		if (pblist)
 			copy_page_backup_list(pblist, p);
-		free_pagedir(p, 0);
+		free_pagedir(p, PG_UNSAFE_KEEP);
 		if (!pblist)
 			error = -ENOMEM;
 	}
@@ -840,7 +846,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 		return 0;
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
-		buffer = alloc_image_page(GFP_ATOMIC, 0);
+		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
 		if (!buffer)
 			return -ENOMEM;
 	}
-- 
cgit v1.2.1


From b788db79896ef2a5817b9395ad63573b254a6d93 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:54 -0700
Subject: [PATCH] swsusp: Introduce memory bitmaps

Introduce the memory bitmap data structure and make swsusp use in the suspend
phase.

The current swsusp's internal data structure is not very efficient from the
memory usage point of view, so it seems reasonable to replace it with a data
structure that will require less memory, such as a pair of bitmaps.

The idea is to use bitmaps that may be allocated as sets of individual pages,
so that we can avoid making allocations of order greater than 0.  For this
reason the memory bitmap structure consists of several linked lists of objects
that contain pointers to memory pages with the actual bitmap data.  Still, for
a typical system all of these lists fit in a single page, so it's reasonable
to introduce an additional mechanism allowing us to allocate all of them
efficiently without sacrificing the generality of the design.  This is done
with the help of the chain_allocator structure and associated functions.

We need to use two memory bitmaps during the suspend phase of the
suspend-resume cycle.  One of them is necessary for marking the saveable
pages, and the second is used to mark the pages in which to store the copies
of them (aka image pages).

First, the bitmaps are created and we allocate as many image pages as needed
(the corresponding bits in the second bitmap are set as soon as the pages are
allocated).  Second, the bits corresponding to the saveable pages are set in
the first bitmap and the saveable pages are copied to the image pages.
Finally, the first bitmap is used to save the kernel virtual addresses of the
saveable pages and the second one is used to save the contents of the image
pages.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/power.h    |   3 +-
 kernel/power/snapshot.c | 602 ++++++++++++++++++++++++++++++++++++++++++------
 kernel/power/swsusp.c   |   5 +-
 3 files changed, 540 insertions(+), 70 deletions(-)

diff --git a/kernel/power/power.h b/kernel/power/power.h
index e18ba207e784..6e9e2acc34f8 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -109,9 +109,10 @@ struct snapshot_handle {
  */
 #define data_of(handle)	((handle).buffer + (handle).buf_offset)
 
+extern unsigned int snapshot_additional_pages(struct zone *zone);
 extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
-int snapshot_image_loaded(struct snapshot_handle *handle);
+extern int snapshot_image_loaded(struct snapshot_handle *handle);
 
 #define SNAPSHOT_IOC_MAGIC	'3'
 #define SNAPSHOT_FREEZE			_IO(SNAPSHOT_IOC_MAGIC, 1)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index d0d691f976d8..852e0df41719 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -211,6 +211,467 @@ static inline void free_image_page(void *addr, int clear_nosave_free)
 	free_page((unsigned long)addr);
 }
 
+/* struct linked_page is used to build chains of pages */
+
+#define LINKED_PAGE_DATA_SIZE	(PAGE_SIZE - sizeof(void *))
+
+struct linked_page {
+	struct linked_page *next;
+	char data[LINKED_PAGE_DATA_SIZE];
+} __attribute__((packed));
+
+static inline void
+free_list_of_pages(struct linked_page *list, int clear_page_nosave)
+{
+	while (list) {
+		struct linked_page *lp = list->next;
+
+		free_image_page(list, clear_page_nosave);
+		list = lp;
+	}
+}
+
+/**
+  *	struct chain_allocator is used for allocating small objects out of
+  *	a linked list of pages called 'the chain'.
+  *
+  *	The chain grows each time when there is no room for a new object in
+  *	the current page.  The allocated objects cannot be freed individually.
+  *	It is only possible to free them all at once, by freeing the entire
+  *	chain.
+  *
+  *	NOTE: The chain allocator may be inefficient if the allocated objects
+  *	are not much smaller than PAGE_SIZE.
+  */
+
+struct chain_allocator {
+	struct linked_page *chain;	/* the chain */
+	unsigned int used_space;	/* total size of objects allocated out
+					 * of the current page
+					 */
+	gfp_t gfp_mask;		/* mask for allocating pages */
+	int safe_needed;	/* if set, only "safe" pages are allocated */
+};
+
+static void
+chain_init(struct chain_allocator *ca, gfp_t gfp_mask, int safe_needed)
+{
+	ca->chain = NULL;
+	ca->used_space = LINKED_PAGE_DATA_SIZE;
+	ca->gfp_mask = gfp_mask;
+	ca->safe_needed = safe_needed;
+}
+
+static void *chain_alloc(struct chain_allocator *ca, unsigned int size)
+{
+	void *ret;
+
+	if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) {
+		struct linked_page *lp;
+
+		lp = alloc_image_page(ca->gfp_mask, ca->safe_needed);
+		if (!lp)
+			return NULL;
+
+		lp->next = ca->chain;
+		ca->chain = lp;
+		ca->used_space = 0;
+	}
+	ret = ca->chain->data + ca->used_space;
+	ca->used_space += size;
+	return ret;
+}
+
+static void chain_free(struct chain_allocator *ca, int clear_page_nosave)
+{
+	free_list_of_pages(ca->chain, clear_page_nosave);
+	memset(ca, 0, sizeof(struct chain_allocator));
+}
+
+/**
+ *	Data types related to memory bitmaps.
+ *
+ *	Memory bitmap is a structure consiting of many linked lists of
+ *	objects.  The main list's elements are of type struct zone_bitmap
+ *	and each of them corresonds to one zone.  For each zone bitmap
+ *	object there is a list of objects of type struct bm_block that
+ *	represent each blocks of bit chunks in which information is
+ *	stored.
+ *
+ *	struct memory_bitmap contains a pointer to the main list of zone
+ *	bitmap objects, a struct bm_position used for browsing the bitmap,
+ *	and a pointer to the list of pages used for allocating all of the
+ *	zone bitmap objects and bitmap block objects.
+ *
+ *	NOTE: It has to be possible to lay out the bitmap in memory
+ *	using only allocations of order 0.  Additionally, the bitmap is
+ *	designed to work with arbitrary number of zones (this is over the
+ *	top for now, but let's avoid making unnecessary assumptions ;-).
+ *
+ *	struct zone_bitmap contains a pointer to a list of bitmap block
+ *	objects and a pointer to the bitmap block object that has been
+ *	most recently used for setting bits.  Additionally, it contains the
+ *	pfns that correspond to the start and end of the represented zone.
+ *
+ *	struct bm_block contains a pointer to the memory page in which
+ *	information is stored (in the form of a block of bit chunks
+ *	of type unsigned long each).  It also contains the pfns that
+ *	correspond to the start and end of the represented memory area and
+ *	the number of bit chunks in the block.
+ *
+ *	NOTE: Memory bitmaps are used for two types of operations only:
+ *	"set a bit" and "find the next bit set".  Moreover, the searching
+ *	is always carried out after all of the "set a bit" operations
+ *	on given bitmap.
+ */
+
+#define BM_END_OF_MAP	(~0UL)
+
+#define BM_CHUNKS_PER_BLOCK	(PAGE_SIZE / sizeof(long))
+#define BM_BITS_PER_CHUNK	(sizeof(long) << 3)
+#define BM_BITS_PER_BLOCK	(PAGE_SIZE << 3)
+
+struct bm_block {
+	struct bm_block *next;		/* next element of the list */
+	unsigned long start_pfn;	/* pfn represented by the first bit */
+	unsigned long end_pfn;	/* pfn represented by the last bit plus 1 */
+	unsigned int size;	/* number of bit chunks */
+	unsigned long *data;	/* chunks of bits representing pages */
+};
+
+struct zone_bitmap {
+	struct zone_bitmap *next;	/* next element of the list */
+	unsigned long start_pfn;	/* minimal pfn in this zone */
+	unsigned long end_pfn;		/* maximal pfn in this zone plus 1 */
+	struct bm_block *bm_blocks;	/* list of bitmap blocks */
+	struct bm_block *cur_block;	/* recently used bitmap block */
+};
+
+/* strcut bm_position is used for browsing memory bitmaps */
+
+struct bm_position {
+	struct zone_bitmap *zone_bm;
+	struct bm_block *block;
+	int chunk;
+	int bit;
+};
+
+struct memory_bitmap {
+	struct zone_bitmap *zone_bm_list;	/* list of zone bitmaps */
+	struct linked_page *p_list;	/* list of pages used to store zone
+					 * bitmap objects and bitmap block
+					 * objects
+					 */
+	struct bm_position cur;	/* most recently used bit position */
+};
+
+/* Functions that operate on memory bitmaps */
+
+static inline void memory_bm_reset_chunk(struct memory_bitmap *bm)
+{
+	bm->cur.chunk = 0;
+	bm->cur.bit = -1;
+}
+
+static void memory_bm_position_reset(struct memory_bitmap *bm)
+{
+	struct zone_bitmap *zone_bm;
+
+	zone_bm = bm->zone_bm_list;
+	bm->cur.zone_bm = zone_bm;
+	bm->cur.block = zone_bm->bm_blocks;
+	memory_bm_reset_chunk(bm);
+}
+
+static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free);
+
+/**
+ *	create_bm_block_list - create a list of block bitmap objects
+ */
+
+static inline struct bm_block *
+create_bm_block_list(unsigned int nr_blocks, struct chain_allocator *ca)
+{
+	struct bm_block *bblist = NULL;
+
+	while (nr_blocks-- > 0) {
+		struct bm_block *bb;
+
+		bb = chain_alloc(ca, sizeof(struct bm_block));
+		if (!bb)
+			return NULL;
+
+		bb->next = bblist;
+		bblist = bb;
+	}
+	return bblist;
+}
+
+/**
+ *	create_zone_bm_list - create a list of zone bitmap objects
+ */
+
+static inline struct zone_bitmap *
+create_zone_bm_list(unsigned int nr_zones, struct chain_allocator *ca)
+{
+	struct zone_bitmap *zbmlist = NULL;
+
+	while (nr_zones-- > 0) {
+		struct zone_bitmap *zbm;
+
+		zbm = chain_alloc(ca, sizeof(struct zone_bitmap));
+		if (!zbm)
+			return NULL;
+
+		zbm->next = zbmlist;
+		zbmlist = zbm;
+	}
+	return zbmlist;
+}
+
+/**
+  *	memory_bm_create - allocate memory for a memory bitmap
+  */
+
+static int
+memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
+{
+	struct chain_allocator ca;
+	struct zone *zone;
+	struct zone_bitmap *zone_bm;
+	struct bm_block *bb;
+	unsigned int nr;
+
+	chain_init(&ca, gfp_mask, safe_needed);
+
+	/* Compute the number of zones */
+	nr = 0;
+	for_each_zone (zone)
+		if (populated_zone(zone) && !is_highmem(zone))
+			nr++;
+
+	/* Allocate the list of zones bitmap objects */
+	zone_bm = create_zone_bm_list(nr, &ca);
+	bm->zone_bm_list = zone_bm;
+	if (!zone_bm) {
+		chain_free(&ca, PG_UNSAFE_CLEAR);
+		return -ENOMEM;
+	}
+
+	/* Initialize the zone bitmap objects */
+	for_each_zone (zone) {
+		unsigned long pfn;
+
+		if (!populated_zone(zone) || is_highmem(zone))
+			continue;
+
+		zone_bm->start_pfn = zone->zone_start_pfn;
+		zone_bm->end_pfn = zone->zone_start_pfn + zone->spanned_pages;
+		/* Allocate the list of bitmap block objects */
+		nr = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
+		bb = create_bm_block_list(nr, &ca);
+		zone_bm->bm_blocks = bb;
+		zone_bm->cur_block = bb;
+		if (!bb)
+			goto Free;
+
+		nr = zone->spanned_pages;
+		pfn = zone->zone_start_pfn;
+		/* Initialize the bitmap block objects */
+		while (bb) {
+			unsigned long *ptr;
+
+			ptr = alloc_image_page(gfp_mask, safe_needed);
+			bb->data = ptr;
+			if (!ptr)
+				goto Free;
+
+			bb->start_pfn = pfn;
+			if (nr >= BM_BITS_PER_BLOCK) {
+				pfn += BM_BITS_PER_BLOCK;
+				bb->size = BM_CHUNKS_PER_BLOCK;
+				nr -= BM_BITS_PER_BLOCK;
+			} else {
+				/* This is executed only once in the loop */
+				pfn += nr;
+				bb->size = DIV_ROUND_UP(nr, BM_BITS_PER_CHUNK);
+			}
+			bb->end_pfn = pfn;
+			bb = bb->next;
+		}
+		zone_bm = zone_bm->next;
+	}
+	bm->p_list = ca.chain;
+	memory_bm_position_reset(bm);
+	return 0;
+
+Free:
+	bm->p_list = ca.chain;
+	memory_bm_free(bm, PG_UNSAFE_CLEAR);
+	return -ENOMEM;
+}
+
+/**
+  *	memory_bm_free - free memory occupied by the memory bitmap @bm
+  */
+
+static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free)
+{
+	struct zone_bitmap *zone_bm;
+
+	/* Free the list of bit blocks for each zone_bitmap object */
+	zone_bm = bm->zone_bm_list;
+	while (zone_bm) {
+		struct bm_block *bb;
+
+		bb = zone_bm->bm_blocks;
+		while (bb) {
+			if (bb->data)
+				free_image_page(bb->data, clear_nosave_free);
+			bb = bb->next;
+		}
+		zone_bm = zone_bm->next;
+	}
+	free_list_of_pages(bm->p_list, clear_nosave_free);
+	bm->zone_bm_list = NULL;
+}
+
+/**
+ *	memory_bm_set_bit - set the bit in the bitmap @bm that corresponds
+ *	to given pfn.  The cur_zone_bm member of @bm and the cur_block member
+ *	of @bm->cur_zone_bm are updated.
+ *
+ *	If the bit cannot be set, the function returns -EINVAL .
+ */
+
+static int
+memory_bm_set_bit(struct memory_bitmap *bm, unsigned long pfn)
+{
+	struct zone_bitmap *zone_bm;
+	struct bm_block *bb;
+
+	/* Check if the pfn is from the current zone */
+	zone_bm = bm->cur.zone_bm;
+	if (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) {
+		zone_bm = bm->zone_bm_list;
+		/* We don't assume that the zones are sorted by pfns */
+		while (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) {
+			zone_bm = zone_bm->next;
+			if (unlikely(!zone_bm))
+				return -EINVAL;
+		}
+		bm->cur.zone_bm = zone_bm;
+	}
+	/* Check if the pfn corresponds to the current bitmap block */
+	bb = zone_bm->cur_block;
+	if (pfn < bb->start_pfn)
+		bb = zone_bm->bm_blocks;
+
+	while (pfn >= bb->end_pfn) {
+		bb = bb->next;
+		if (unlikely(!bb))
+			return -EINVAL;
+	}
+	zone_bm->cur_block = bb;
+	pfn -= bb->start_pfn;
+	set_bit(pfn % BM_BITS_PER_CHUNK, bb->data + pfn / BM_BITS_PER_CHUNK);
+	return 0;
+}
+
+/* Two auxiliary functions for memory_bm_next_pfn */
+
+/* Find the first set bit in the given chunk, if there is one */
+
+static inline int next_bit_in_chunk(int bit, unsigned long *chunk_p)
+{
+	bit++;
+	while (bit < BM_BITS_PER_CHUNK) {
+		if (test_bit(bit, chunk_p))
+			return bit;
+
+		bit++;
+	}
+	return -1;
+}
+
+/* Find a chunk containing some bits set in given block of bits */
+
+static inline int next_chunk_in_block(int n, struct bm_block *bb)
+{
+	n++;
+	while (n < bb->size) {
+		if (bb->data[n])
+			return n;
+
+		n++;
+	}
+	return -1;
+}
+
+/**
+ *	memory_bm_next_pfn - find the pfn that corresponds to the next set bit
+ *	in the bitmap @bm.  If the pfn cannot be found, BM_END_OF_MAP is
+ *	returned.
+ *
+ *	It is required to run memory_bm_position_reset() before the first call to
+ *	this function.
+ */
+
+static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm)
+{
+	struct zone_bitmap *zone_bm;
+	struct bm_block *bb;
+	int chunk;
+	int bit;
+
+	do {
+		bb = bm->cur.block;
+		do {
+			chunk = bm->cur.chunk;
+			bit = bm->cur.bit;
+			do {
+				bit = next_bit_in_chunk(bit, bb->data + chunk);
+				if (bit >= 0)
+					goto Return_pfn;
+
+				chunk = next_chunk_in_block(chunk, bb);
+				bit = -1;
+			} while (chunk >= 0);
+			bb = bb->next;
+			bm->cur.block = bb;
+			memory_bm_reset_chunk(bm);
+		} while (bb);
+		zone_bm = bm->cur.zone_bm->next;
+		if (zone_bm) {
+			bm->cur.zone_bm = zone_bm;
+			bm->cur.block = zone_bm->bm_blocks;
+			memory_bm_reset_chunk(bm);
+		}
+	} while (zone_bm);
+	memory_bm_position_reset(bm);
+	return BM_END_OF_MAP;
+
+Return_pfn:
+	bm->cur.chunk = chunk;
+	bm->cur.bit = bit;
+	return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit;
+}
+
+/**
+ *	snapshot_additional_pages - estimate the number of additional pages
+ *	be needed for setting up the suspend image data structures for given
+ *	zone (usually the returned value is greater than the exact number)
+ */
+
+unsigned int snapshot_additional_pages(struct zone *zone)
+{
+	unsigned int res;
+
+	res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK);
+	res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE);
+	return res;
+}
+
 /**
  *	pfn_is_nosave - check if given pfn is in the 'nosave' section
  */
@@ -276,32 +737,38 @@ static inline void copy_data_page(long *dst, long *src)
 		*dst++ = *src++;
 }
 
-static void copy_data_pages(struct pbe *pblist)
+static void
+copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
 {
 	struct zone *zone;
-	unsigned long pfn, max_zone_pfn;
-	struct pbe *pbe;
+	unsigned long pfn;
 
-	pbe = pblist;
 	for_each_zone (zone) {
+		unsigned long max_zone_pfn;
+
 		if (is_highmem(zone))
 			continue;
+
 		mark_free_pages(zone);
 		max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages;
-		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) {
-			struct page *page = saveable_page(pfn);
-
-			if (page) {
-				void *ptr = page_address(page);
-
-				BUG_ON(!pbe);
-				copy_data_page((void *)pbe->address, ptr);
-				pbe->orig_address = (unsigned long)ptr;
-				pbe = pbe->next;
-			}
-		}
+		for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++)
+			if (saveable_page(pfn))
+				memory_bm_set_bit(orig_bm, pfn);
 	}
-	BUG_ON(pbe);
+	memory_bm_position_reset(orig_bm);
+	memory_bm_position_reset(copy_bm);
+	do {
+		pfn = memory_bm_next_pfn(orig_bm);
+		if (likely(pfn != BM_END_OF_MAP)) {
+			struct page *page;
+			void *src;
+
+			page = pfn_to_page(pfn);
+			src = page_address(page);
+			page = pfn_to_page(memory_bm_next_pfn(copy_bm));
+			copy_data_page(page_address(page), src);
+		}
+	} while (pfn != BM_END_OF_MAP);
 }
 
 /**
@@ -447,37 +914,43 @@ static int enough_free_mem(unsigned int nr_pages)
 		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
 }
 
-static int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed)
+static int
+swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm,
+		unsigned int nr_pages)
 {
-	struct pbe *p;
+	int error;
 
-	for_each_pbe (p, pblist) {
-		p->address = (unsigned long)alloc_image_page(gfp_mask, safe_needed);
-		if (!p->address)
-			return -ENOMEM;
-	}
-	return 0;
-}
+	error = memory_bm_create(orig_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY);
+	if (error)
+		goto Free;
 
-static struct pbe *swsusp_alloc(unsigned int nr_pages)
-{
-	struct pbe *pblist;
+	error = memory_bm_create(copy_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY);
+	if (error)
+		goto Free;
 
-	pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, PG_ANY);
-	if (!pblist) {
-		printk(KERN_ERR "suspend: Allocating pagedir failed.\n");
-		return NULL;
-	}
+	while (nr_pages-- > 0) {
+		struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD);
+		if (!page)
+			goto Free;
 
-	if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, PG_ANY)) {
-		printk(KERN_ERR "suspend: Allocating image pages failed.\n");
-		swsusp_free();
-		return NULL;
+		SetPageNosave(page);
+		SetPageNosaveFree(page);
+		memory_bm_set_bit(copy_bm, page_to_pfn(page));
 	}
+	return 0;
 
-	return pblist;
+Free:
+	swsusp_free();
+	return -ENOMEM;
 }
 
+/* Memory bitmap used for marking saveable pages */
+static struct memory_bitmap orig_bm;
+/* Memory bitmap used for marking allocated pages that will contain the copies
+ * of saveable pages
+ */
+static struct memory_bitmap copy_bm;
+
 asmlinkage int swsusp_save(void)
 {
 	unsigned int nr_pages;
@@ -498,15 +971,14 @@ asmlinkage int swsusp_save(void)
 		return -ENOMEM;
 	}
 
-	restore_pblist = swsusp_alloc(nr_pages);
-	if (!restore_pblist)
+	if (swsusp_alloc(&orig_bm, &copy_bm, nr_pages))
 		return -ENOMEM;
 
 	/* During allocating of suspend pagedir, new cold pages may appear.
 	 * Kill them.
 	 */
 	drain_local_pages();
-	copy_data_pages(restore_pblist);
+	copy_data_pages(&copy_bm, &orig_bm);
 
 	/*
 	 * End of critical section. From now on, we can write to memory,
@@ -535,22 +1007,23 @@ static void init_header(struct swsusp_info *info)
 }
 
 /**
- *	pack_orig_addresses - the .orig_address fields of the PBEs from the
- *	list starting at @pbe are stored in the array @buf[] (1 page)
+ *	pack_addresses - the addresses corresponding to pfns found in the
+ *	bitmap @bm are stored in the array @buf[] (1 page)
  */
 
-static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pbe)
+static inline void
+pack_addresses(unsigned long *buf, struct memory_bitmap *bm)
 {
 	int j;
 
-	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-		buf[j] = pbe->orig_address;
-		pbe = pbe->next;
+	for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
+		unsigned long pfn = memory_bm_next_pfn(bm);
+
+		if (unlikely(pfn == BM_END_OF_MAP))
+			break;
+
+		buf[j] = (unsigned long)page_address(pfn_to_page(pfn));
 	}
-	if (!pbe)
-		for (; j < PAGE_SIZE / sizeof(long); j++)
-			buf[j] = 0;
-	return pbe;
 }
 
 /**
@@ -579,6 +1052,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 {
 	if (handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
+
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
 		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
@@ -588,16 +1062,17 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 	if (!handle->offset) {
 		init_header((struct swsusp_info *)buffer);
 		handle->buffer = buffer;
-		handle->pbe = restore_pblist;
+		memory_bm_position_reset(&orig_bm);
+		memory_bm_position_reset(&copy_bm);
 	}
 	if (handle->prev < handle->cur) {
 		if (handle->cur <= nr_meta_pages) {
-			handle->pbe = pack_orig_addresses(buffer, handle->pbe);
-			if (!handle->pbe)
-				handle->pbe = restore_pblist;
+			memset(buffer, 0, PAGE_SIZE);
+			pack_addresses(buffer, &orig_bm);
 		} else {
-			handle->buffer = (void *)handle->pbe->address;
-			handle->pbe = handle->pbe->next;
+			unsigned long pfn = memory_bm_next_pfn(&copy_bm);
+
+			handle->buffer = page_address(pfn_to_page(pfn));
 		}
 		handle->prev = handle->cur;
 	}
@@ -736,12 +1211,7 @@ static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
  *	of "safe" which will be used later
  */
 
-struct safe_page {
-	struct safe_page *next;
-	char padding[PAGE_SIZE - sizeof(void *)];
-};
-
-static struct safe_page *safe_pages;
+static struct linked_page *safe_pages;
 
 static int prepare_image(struct snapshot_handle *handle)
 {
@@ -763,9 +1233,9 @@ static int prepare_image(struct snapshot_handle *handle)
 	if (!error && nr_pages > unsafe_pages) {
 		nr_pages -= unsafe_pages;
 		while (nr_pages--) {
-			struct safe_page *ptr;
+			struct linked_page *ptr;
 
-			ptr = (struct safe_page *)get_zeroed_page(GFP_ATOMIC);
+			ptr = (void *)get_zeroed_page(GFP_ATOMIC);
 			if (!ptr) {
 				error = -ENOMEM;
 				break;
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 17f669c83012..8ef677ea0cea 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -193,14 +193,13 @@ int swsusp_shrink_memory(void)
 	printk("Shrinking memory...  ");
 	do {
 		size = 2 * count_highmem_pages();
-		size += size / 50 + count_data_pages();
-		size += (size + PBES_PER_PAGE - 1) / PBES_PER_PAGE +
-			PAGES_FOR_IO;
+		size += size / 50 + count_data_pages() + PAGES_FOR_IO;
 		tmp = size;
 		for_each_zone (zone)
 			if (!is_highmem(zone) && populated_zone(zone)) {
 				tmp -= zone->free_pages;
 				tmp += zone->lowmem_reserve[ZONE_NORMAL];
+				tmp += snapshot_additional_pages(zone);
 			}
 		if (tmp > 0) {
 			tmp = __shrink_memory(tmp);
-- 
cgit v1.2.1


From 940864ddabdb180e02041c4dcd46ba6f9eee732f Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:55 -0700
Subject: [PATCH] swsusp: Use memory bitmaps during resume

Make swsusp use memory bitmaps to store its internal information during the
resume phase of the suspend-resume cycle.

If the pfns of saveable pages are saved during the suspend phase instead of
the kernel virtual addresses of these pages, we can use them during the resume
phase directly to set the corresponding bits in a memory bitmap.  Then, this
bitmap is used to mark the page frames corresponding to the pages that were
saveable before the suspend (aka "unsafe" page frames).

Next, we allocate as many page frames as needed to store the entire suspend
image and make sure that there will be some extra free "safe" page frames for
the list of PBEs constructed later.  Subsequently, the image is loaded and, if
possible, the data loaded from it are written into their "original" page
frames (ie.  the ones they had occupied before the suspend).

The image data that cannot be written into their "original" page frames are
loaded into "safe" page frames and their "original" kernel virtual addresses,
as well as the addresses of the "safe" pages containing their copies, are
stored in a list of PBEs.  Finally, the list of PBEs is used to copy the
remaining image data into their "original" page frames (this is done
atomically, by the architecture-dependent parts of swsusp).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/suspend.h |   9 --
 kernel/power/power.h    |  11 +-
 kernel/power/snapshot.c | 416 +++++++++++++++++++++---------------------------
 kernel/power/swap.c     |   4 +-
 kernel/power/user.c     |   1 +
 5 files changed, 186 insertions(+), 255 deletions(-)

diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index c11cacf1a13b..b1237f16ecde 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -16,15 +16,6 @@ struct pbe {
 	struct pbe *next;
 };
 
-#define for_each_pbe(pbe, pblist) \
-	for (pbe = pblist ; pbe ; pbe = pbe->next)
-
-#define PBES_PER_PAGE      (PAGE_SIZE/sizeof(struct pbe))
-#define PB_PAGE_SKIP       (PBES_PER_PAGE-1)
-
-#define for_each_pb_page(pbe, pblist) \
-	for (pbe = pblist ; pbe ; pbe = (pbe+PB_PAGE_SKIP)->next)
-
 /* mm/page_alloc.c */
 extern void drain_local_pages(void);
 extern void mark_free_pages(struct zone *zone);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 6e9e2acc34f8..bfe999f7b272 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -81,16 +81,6 @@ struct snapshot_handle {
 	unsigned int	prev;	/* number of the block of PAGE_SIZE bytes that
 				 * was the current one previously
 				 */
-	struct pbe	*pbe;	/* PBE that corresponds to 'buffer' */
-	struct pbe	*last_pbe;	/* When the image is restored (eg. read
-					 * from disk) we can store some image
-					 * data directly in the page frames
-					 * in which they were before suspend.
-					 * In such a case the PBEs that
-					 * correspond to them will be unused.
-					 * This is the last PBE, so far, that
-					 * does not correspond to such data.
-					 */
 	void		*buffer;	/* address of the block to read from
 					 * or write to
 					 */
@@ -113,6 +103,7 @@ extern unsigned int snapshot_additional_pages(struct zone *zone);
 extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
 extern int snapshot_image_loaded(struct snapshot_handle *handle);
+extern void snapshot_free_unused_memory(struct snapshot_handle *handle);
 
 #define SNAPSHOT_IOC_MAGIC	'3'
 #define SNAPSHOT_FREEZE			_IO(SNAPSHOT_IOC_MAGIC, 1)
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 852e0df41719..1b84313cbab5 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -39,7 +39,7 @@ struct pbe *restore_pblist;
 
 static unsigned int nr_copy_pages;
 static unsigned int nr_meta_pages;
-static unsigned long *buffer;
+static void *buffer;
 
 #ifdef CONFIG_HIGHMEM
 unsigned int count_highmem_pages(void)
@@ -172,7 +172,7 @@ static inline int restore_highmem(void) {return 0;}
 #define PG_UNSAFE_CLEAR	1
 #define PG_UNSAFE_KEEP	0
 
-static unsigned int unsafe_pages;
+static unsigned int allocated_unsafe_pages;
 
 static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 {
@@ -183,7 +183,7 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed)
 		while (res && PageNosaveFree(virt_to_page(res))) {
 			/* The page is unsafe, mark it for swsusp_free() */
 			SetPageNosave(virt_to_page(res));
-			unsafe_pages++;
+			allocated_unsafe_pages++;
 			res = (void *)get_zeroed_page(gfp_mask);
 		}
 	if (res) {
@@ -772,101 +772,10 @@ copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm)
 }
 
 /**
- *	free_pagedir - free pages allocated with alloc_pagedir()
- */
-
-static void free_pagedir(struct pbe *pblist, int clear_nosave_free)
-{
-	struct pbe *pbe;
-
-	while (pblist) {
-		pbe = (pblist + PB_PAGE_SKIP)->next;
-		free_image_page(pblist, clear_nosave_free);
-		pblist = pbe;
-	}
-}
-
-/**
- *	fill_pb_page - Create a list of PBEs on a given memory page
- */
-
-static inline void fill_pb_page(struct pbe *pbpage, unsigned int n)
-{
-	struct pbe *p;
-
-	p = pbpage;
-	pbpage += n - 1;
-	do
-		p->next = p + 1;
-	while (++p < pbpage);
-}
-
-/**
- *	create_pbe_list - Create a list of PBEs on top of a given chain
- *	of memory pages allocated with alloc_pagedir()
+ *	swsusp_free - free pages allocated for the suspend.
  *
- *	This function assumes that pages allocated by alloc_image_page() will
- *	always be zeroed.
- */
-
-static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages)
-{
-	struct pbe *pbpage;
-	unsigned int num = PBES_PER_PAGE;
-
-	for_each_pb_page (pbpage, pblist) {
-		if (num >= nr_pages)
-			break;
-
-		fill_pb_page(pbpage, PBES_PER_PAGE);
-		num += PBES_PER_PAGE;
-	}
-	if (pbpage) {
-		num -= PBES_PER_PAGE;
-		fill_pb_page(pbpage, nr_pages - num);
-	}
-}
-
-/**
- *	alloc_pagedir - Allocate the page directory.
- *
- *	First, determine exactly how many pages we need and
- *	allocate them.
- *
- *	We arrange the pages in a chain: each page is an array of PBES_PER_PAGE
- *	struct pbe elements (pbes) and the last element in the page points
- *	to the next page.
- *
- *	On each page we set up a list of struct_pbe elements.
- */
-
-static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask,
-				 int safe_needed)
-{
-	unsigned int num;
-	struct pbe *pblist, *pbe;
-
-	if (!nr_pages)
-		return NULL;
-
-	pblist = alloc_image_page(gfp_mask, safe_needed);
-	pbe = pblist;
-	for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) {
-		if (!pbe) {
-			free_pagedir(pblist, PG_UNSAFE_CLEAR);
-			return NULL;
-		}
-		pbe += PB_PAGE_SKIP;
-		pbe->next = alloc_image_page(gfp_mask, safe_needed);
-		pbe = pbe->next;
-	}
-	create_pbe_list(pblist, nr_pages);
-	return pblist;
-}
-
-/**
- * Free pages we allocated for suspend. Suspend pages are alocated
- * before atomic copy, so we need to free them after resume.
+ *	Suspend pages are alocated before the atomic copy is made, so we
+ *	need to release them after the resume.
  */
 
 void swsusp_free(void)
@@ -904,14 +813,18 @@ void swsusp_free(void)
 static int enough_free_mem(unsigned int nr_pages)
 {
 	struct zone *zone;
-	unsigned int n = 0;
+	unsigned int free = 0, meta = 0;
 
 	for_each_zone (zone)
-		if (!is_highmem(zone))
-			n += zone->free_pages;
-	pr_debug("swsusp: available memory: %u pages\n", n);
-	return n > (nr_pages + PAGES_FOR_IO +
-		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+		if (!is_highmem(zone)) {
+			free += zone->free_pages;
+			meta += snapshot_additional_pages(zone);
+		}
+
+	pr_debug("swsusp: pages needed: %u + %u + %u, available pages: %u\n",
+		nr_pages, PAGES_FOR_IO, meta, free);
+
+	return free > nr_pages + PAGES_FOR_IO + meta;
 }
 
 static int
@@ -961,11 +874,6 @@ asmlinkage int swsusp_save(void)
 	nr_pages = count_data_pages();
 	printk("swsusp: Need to copy %u pages\n", nr_pages);
 
-	pr_debug("swsusp: pages needed: %u + %lu + %u, free: %u\n",
-		 nr_pages,
-		 (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE,
-		 PAGES_FOR_IO, nr_free_pages());
-
 	if (!enough_free_mem(nr_pages)) {
 		printk(KERN_ERR "swsusp: Not enough free memory\n");
 		return -ENOMEM;
@@ -1007,22 +915,19 @@ static void init_header(struct swsusp_info *info)
 }
 
 /**
- *	pack_addresses - the addresses corresponding to pfns found in the
- *	bitmap @bm are stored in the array @buf[] (1 page)
+ *	pack_pfns - pfns corresponding to the set bits found in the bitmap @bm
+ *	are stored in the array @buf[] (1 page at a time)
  */
 
 static inline void
-pack_addresses(unsigned long *buf, struct memory_bitmap *bm)
+pack_pfns(unsigned long *buf, struct memory_bitmap *bm)
 {
 	int j;
 
 	for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
-		unsigned long pfn = memory_bm_next_pfn(bm);
-
-		if (unlikely(pfn == BM_END_OF_MAP))
+		buf[j] = memory_bm_next_pfn(bm);
+		if (unlikely(buf[j] == BM_END_OF_MAP))
 			break;
-
-		buf[j] = (unsigned long)page_address(pfn_to_page(pfn));
 	}
 }
 
@@ -1068,7 +973,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
 	if (handle->prev < handle->cur) {
 		if (handle->cur <= nr_meta_pages) {
 			memset(buffer, 0, PAGE_SIZE);
-			pack_addresses(buffer, &orig_bm);
+			pack_pfns(buffer, &orig_bm);
 		} else {
 			unsigned long pfn = memory_bm_next_pfn(&copy_bm);
 
@@ -1094,14 +999,10 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count)
  *	had been used before suspend
  */
 
-static int mark_unsafe_pages(struct pbe *pblist)
+static int mark_unsafe_pages(struct memory_bitmap *bm)
 {
 	struct zone *zone;
 	unsigned long pfn, max_zone_pfn;
-	struct pbe *p;
-
-	if (!pblist) /* a sanity check */
-		return -EINVAL;
 
 	/* Clear page flags */
 	for_each_zone (zone) {
@@ -1111,30 +1012,37 @@ static int mark_unsafe_pages(struct pbe *pblist)
 				ClearPageNosaveFree(pfn_to_page(pfn));
 	}
 
-	/* Mark orig addresses */
-	for_each_pbe (p, pblist) {
-		if (virt_addr_valid(p->orig_address))
-			SetPageNosaveFree(virt_to_page(p->orig_address));
-		else
-			return -EFAULT;
-	}
+	/* Mark pages that correspond to the "original" pfns as "unsafe" */
+	memory_bm_position_reset(bm);
+	do {
+		pfn = memory_bm_next_pfn(bm);
+		if (likely(pfn != BM_END_OF_MAP)) {
+			if (likely(pfn_valid(pfn)))
+				SetPageNosaveFree(pfn_to_page(pfn));
+			else
+				return -EFAULT;
+		}
+	} while (pfn != BM_END_OF_MAP);
 
-	unsafe_pages = 0;
+	allocated_unsafe_pages = 0;
 
 	return 0;
 }
 
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
+static void
+duplicate_memory_bitmap(struct memory_bitmap *dst, struct memory_bitmap *src)
 {
-	/* We assume both lists contain the same number of elements */
-	while (src) {
-		dst->orig_address = src->orig_address;
-		dst = dst->next;
-		src = src->next;
+	unsigned long pfn;
+
+	memory_bm_position_reset(src);
+	pfn = memory_bm_next_pfn(src);
+	while (pfn != BM_END_OF_MAP) {
+		memory_bm_set_bit(dst, pfn);
+		pfn = memory_bm_next_pfn(src);
 	}
 }
 
-static int check_header(struct swsusp_info *info)
+static inline int check_header(struct swsusp_info *info)
 {
 	char *reason = NULL;
 
@@ -1161,19 +1069,14 @@ static int check_header(struct swsusp_info *info)
  *	load header - check the image header and copy data from it
  */
 
-static int load_header(struct snapshot_handle *handle,
-                              struct swsusp_info *info)
+static int
+load_header(struct swsusp_info *info)
 {
 	int error;
-	struct pbe *pblist;
 
+	restore_pblist = NULL;
 	error = check_header(info);
 	if (!error) {
-		pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY);
-		if (!pblist)
-			return -ENOMEM;
-		restore_pblist = pblist;
-		handle->pbe = pblist;
 		nr_copy_pages = info->image_pages;
 		nr_meta_pages = info->pages - info->image_pages - 1;
 	}
@@ -1181,108 +1084,137 @@ static int load_header(struct snapshot_handle *handle,
 }
 
 /**
- *	unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- *	the PBEs in the list starting at @pbe
+ *	unpack_orig_pfns - for each element of @buf[] (1 page at a time) set
+ *	the corresponding bit in the memory bitmap @bm
  */
 
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
-                                                struct pbe *pbe)
+static inline void
+unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm)
 {
 	int j;
 
-	for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
-		pbe->orig_address = buf[j];
-		pbe = pbe->next;
+	for (j = 0; j < PAGE_SIZE / sizeof(long); j++) {
+		if (unlikely(buf[j] == BM_END_OF_MAP))
+			break;
+
+		memory_bm_set_bit(bm, buf[j]);
 	}
-	return pbe;
 }
 
 /**
- *	prepare_image - use metadata contained in the PBE list
- *	pointed to by restore_pblist to mark the pages that will
- *	be overwritten in the process of restoring the system
- *	memory state from the image ("unsafe" pages) and allocate
- *	memory for the image
+ *	prepare_image - use the memory bitmap @bm to mark the pages that will
+ *	be overwritten in the process of restoring the system memory state
+ *	from the suspend image ("unsafe" pages) and allocate memory for the
+ *	image.
  *
- *	The idea is to allocate the PBE list first and then
- *	allocate as many pages as it's needed for the image data,
- *	but not to assign these pages to the PBEs initially.
- *	Instead, we just mark them as allocated and create a list
- *	of "safe" which will be used later
+ *	The idea is to allocate a new memory bitmap first and then allocate
+ *	as many pages as needed for the image data, but not to assign these
+ *	pages to specific tasks initially.  Instead, we just mark them as
+ *	allocated and create a list of "safe" pages that will be used later.
  */
 
-static struct linked_page *safe_pages;
+#define PBES_PER_LINKED_PAGE	(LINKED_PAGE_DATA_SIZE / sizeof(struct pbe))
+
+static struct linked_page *safe_pages_list;
 
-static int prepare_image(struct snapshot_handle *handle)
+static int
+prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm)
 {
-	int error = 0;
-	unsigned int nr_pages = nr_copy_pages;
-	struct pbe *p, *pblist = NULL;
+	unsigned int nr_pages;
+	struct linked_page *sp_list, *lp;
+	int error;
 
-	p = restore_pblist;
-	error = mark_unsafe_pages(p);
-	if (!error) {
-		pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE);
-		if (pblist)
-			copy_page_backup_list(pblist, p);
-		free_pagedir(p, PG_UNSAFE_KEEP);
-		if (!pblist)
+	error = mark_unsafe_pages(bm);
+	if (error)
+		goto Free;
+
+	error = memory_bm_create(new_bm, GFP_ATOMIC, PG_SAFE);
+	if (error)
+		goto Free;
+
+	duplicate_memory_bitmap(new_bm, bm);
+	memory_bm_free(bm, PG_UNSAFE_KEEP);
+	/* Reserve some safe pages for potential later use.
+	 *
+	 * NOTE: This way we make sure there will be enough safe pages for the
+	 * chain_alloc() in get_buffer().  It is a bit wasteful, but
+	 * nr_copy_pages cannot be greater than 50% of the memory anyway.
+	 */
+	sp_list = NULL;
+	/* nr_copy_pages cannot be lesser than allocated_unsafe_pages */
+	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE);
+	while (nr_pages > 0) {
+		lp = alloc_image_page(GFP_ATOMIC, PG_SAFE);
+		if (!lp) {
 			error = -ENOMEM;
+			goto Free;
+		}
+		lp->next = sp_list;
+		sp_list = lp;
+		nr_pages--;
 	}
-	safe_pages = NULL;
-	if (!error && nr_pages > unsafe_pages) {
-		nr_pages -= unsafe_pages;
-		while (nr_pages--) {
-			struct linked_page *ptr;
-
-			ptr = (void *)get_zeroed_page(GFP_ATOMIC);
-			if (!ptr) {
-				error = -ENOMEM;
-				break;
-			}
-			if (!PageNosaveFree(virt_to_page(ptr))) {
-				/* The page is "safe", add it to the list */
-				ptr->next = safe_pages;
-				safe_pages = ptr;
-			}
-			/* Mark the page as allocated */
-			SetPageNosave(virt_to_page(ptr));
-			SetPageNosaveFree(virt_to_page(ptr));
+	/* Preallocate memory for the image */
+	safe_pages_list = NULL;
+	nr_pages = nr_copy_pages - allocated_unsafe_pages;
+	while (nr_pages > 0) {
+		lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC);
+		if (!lp) {
+			error = -ENOMEM;
+			goto Free;
+		}
+		if (!PageNosaveFree(virt_to_page(lp))) {
+			/* The page is "safe", add it to the list */
+			lp->next = safe_pages_list;
+			safe_pages_list = lp;
 		}
+		/* Mark the page as allocated */
+		SetPageNosave(virt_to_page(lp));
+		SetPageNosaveFree(virt_to_page(lp));
+		nr_pages--;
 	}
-	if (!error) {
-		restore_pblist = pblist;
-	} else {
-		handle->pbe = NULL;
-		swsusp_free();
+	/* Free the reserved safe pages so that chain_alloc() can use them */
+	while (sp_list) {
+		lp = sp_list->next;
+		free_image_page(sp_list, PG_UNSAFE_CLEAR);
+		sp_list = lp;
 	}
+	return 0;
+
+Free:
+	swsusp_free();
 	return error;
 }
 
-static void *get_buffer(struct snapshot_handle *handle)
+/**
+ *	get_buffer - compute the address that snapshot_write_next() should
+ *	set for its caller to write to.
+ */
+
+static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca)
 {
-	struct pbe *pbe = handle->pbe, *last = handle->last_pbe;
-	struct page *page = virt_to_page(pbe->orig_address);
+	struct pbe *pbe;
+	struct page *page = pfn_to_page(memory_bm_next_pfn(bm));
 
-	if (PageNosave(page) && PageNosaveFree(page)) {
-		/*
-		 * We have allocated the "original" page frame and we can
-		 * use it directly to store the read page
+	if (PageNosave(page) && PageNosaveFree(page))
+		/* We have allocated the "original" page frame and we can
+		 * use it directly to store the loaded page.
 		 */
-		pbe->address = 0;
-		if (last && last->next)
-			last->next = NULL;
-		return (void *)pbe->orig_address;
-	}
-	/*
-	 * The "original" page frame has not been allocated and we have to
-	 * use a "safe" page frame to store the read page
+		return page_address(page);
+
+	/* The "original" page frame has not been allocated and we have to
+	 * use a "safe" page frame to store the loaded page.
 	 */
-	pbe->address = (unsigned long)safe_pages;
-	safe_pages = safe_pages->next;
-	if (last)
-		last->next = pbe;
-	handle->last_pbe = pbe;
+	pbe = chain_alloc(ca, sizeof(struct pbe));
+	if (!pbe) {
+		swsusp_free();
+		return NULL;
+	}
+	pbe->orig_address = (unsigned long)page_address(page);
+	pbe->address = (unsigned long)safe_pages_list;
+	safe_pages_list = safe_pages_list->next;
+	pbe->next = restore_pblist;
+	restore_pblist = pbe;
 	return (void *)pbe->address;
 }
 
@@ -1310,10 +1242,13 @@ static void *get_buffer(struct snapshot_handle *handle)
 
 int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 {
+	static struct chain_allocator ca;
 	int error = 0;
 
+	/* Check if we have already loaded the entire image */
 	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
 		return 0;
+
 	if (!buffer) {
 		/* This makes the buffer be freed by swsusp_free() */
 		buffer = alloc_image_page(GFP_ATOMIC, PG_ANY);
@@ -1324,26 +1259,32 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 		handle->buffer = buffer;
 	handle->sync_read = 1;
 	if (handle->prev < handle->cur) {
-		if (!handle->prev) {
-			error = load_header(handle,
-					(struct swsusp_info *)buffer);
+		if (handle->prev == 0) {
+			error = load_header(buffer);
+			if (error)
+				return error;
+
+			error = memory_bm_create(&copy_bm, GFP_ATOMIC, PG_ANY);
 			if (error)
 				return error;
+
 		} else if (handle->prev <= nr_meta_pages) {
-			handle->pbe = unpack_orig_addresses(buffer,
-							handle->pbe);
-			if (!handle->pbe) {
-				error = prepare_image(handle);
+			unpack_orig_pfns(buffer, &copy_bm);
+			if (handle->prev == nr_meta_pages) {
+				error = prepare_image(&orig_bm, &copy_bm);
 				if (error)
 					return error;
-				handle->pbe = restore_pblist;
-				handle->last_pbe = NULL;
-				handle->buffer = get_buffer(handle);
+
+				chain_init(&ca, GFP_ATOMIC, PG_SAFE);
+				memory_bm_position_reset(&orig_bm);
+				restore_pblist = NULL;
+				handle->buffer = get_buffer(&orig_bm, &ca);
 				handle->sync_read = 0;
+				if (!handle->buffer)
+					return -ENOMEM;
 			}
 		} else {
-			handle->pbe = handle->pbe->next;
-			handle->buffer = get_buffer(handle);
+			handle->buffer = get_buffer(&orig_bm, &ca);
 			handle->sync_read = 0;
 		}
 		handle->prev = handle->cur;
@@ -1362,6 +1303,13 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count)
 
 int snapshot_image_loaded(struct snapshot_handle *handle)
 {
-	return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
-		handle->cur <= nr_meta_pages + nr_copy_pages);
+	return !(!nr_copy_pages ||
+			handle->cur <= nr_meta_pages + nr_copy_pages);
+}
+
+void snapshot_free_unused_memory(struct snapshot_handle *handle)
+{
+	/* Free only if we have loaded the image entirely */
+	if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages)
+		memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR);
 }
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 8309d20b2563..9b2ee5344dee 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -331,8 +331,7 @@ static int enough_swap(unsigned int nr_pages)
 	unsigned int free_swap = count_swap_pages(root_swap, 1);
 
 	pr_debug("swsusp: free swap pages: %u\n", free_swap);
-	return free_swap > (nr_pages + PAGES_FOR_IO +
-		(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+	return free_swap > nr_pages + PAGES_FOR_IO;
 }
 
 /**
@@ -547,6 +546,7 @@ static int load_image(struct swap_map_handle *handle,
 		error = err2;
 	if (!error) {
 		printk("\b\b\b\bdone\n");
+		snapshot_free_unused_memory(snapshot);
 		if (!snapshot_image_loaded(snapshot))
 			error = -ENODATA;
 	}
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 0ef5e4ba39e5..2e4499f3e4d9 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -193,6 +193,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			error = -EPERM;
 			break;
 		}
+		snapshot_free_unused_memory(&data->handle);
 		down(&pm_sem);
 		pm_prepare_console();
 		error = device_suspend(PMSG_FREEZE);
-- 
cgit v1.2.1


From c8eb8b4025175f967af0ba8e933f23aa9954dc35 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:56 -0700
Subject: [PATCH] PM: make it possible to disable console suspending

Change suspend_console() so that it waits for all consoles to flush the
remaining messages and make it possible to switch the console suspending off
with the help of a Kconfig option.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: Stefan Seyfried <seife@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/console.h |  5 +++++
 kernel/power/Kconfig    | 11 +++++++++++
 kernel/printk.c         |  3 +++
 3 files changed, 19 insertions(+)

diff --git a/include/linux/console.h b/include/linux/console.h
index 3bdf2155e565..76a1807726eb 100644
--- a/include/linux/console.h
+++ b/include/linux/console.h
@@ -120,9 +120,14 @@ extern void console_stop(struct console *);
 extern void console_start(struct console *);
 extern int is_console_locked(void);
 
+#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND
 /* Suspend and resume console messages over PM events */
 extern void suspend_console(void);
 extern void resume_console(void);
+#else
+static inline void suspend_console(void) {}
+static inline void resume_console(void) {}
+#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */
 
 /* Some debug stub to catch some of the obvious races in the VT code */
 #if 1
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 619ecabf7c58..4b6e2f18e056 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -36,6 +36,17 @@ config PM_DEBUG
 	code. This is helpful when debugging and reporting various PM bugs, 
 	like suspend support.
 
+config DISABLE_CONSOLE_SUSPEND
+	bool "Keep console(s) enabled during suspend/resume (DANGEROUS)"
+	depends on PM && PM_DEBUG
+	default n
+	---help---
+	This option turns off the console suspend mechanism that prevents
+	debug messages from reaching the console during the suspend/resume
+	operations.  This may be helpful when debugging device drivers'
+	suspend/resume routines, but may itself lead to problems, for example
+	if netconsole is used.
+
 config PM_TRACE
 	bool "Suspend/resume event tracing"
 	depends on PM && PM_DEBUG && X86_32 && EXPERIMENTAL
diff --git a/kernel/printk.c b/kernel/printk.c
index 1149365e989e..771f5e861bcd 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -721,6 +721,7 @@ int __init add_preferred_console(char *name, int idx, char *options)
 	return 0;
 }
 
+#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND
 /**
  * suspend_console - suspend the console subsystem
  *
@@ -728,6 +729,7 @@ int __init add_preferred_console(char *name, int idx, char *options)
  */
 void suspend_console(void)
 {
+	printk("Suspending console(s)\n");
 	acquire_console_sem();
 	console_suspended = 1;
 }
@@ -737,6 +739,7 @@ void resume_console(void)
 	console_suspended = 0;
 	release_console_sem();
 }
+#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */
 
 /**
  * acquire_console_sem - lock the console system for exclusive use.
-- 
cgit v1.2.1


From e1da95ae38afdcda83328300c4aed755d9fc01a6 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:57 -0700
Subject: [PATCH] suspend: make it possible to disable serial console suspend

Hack uart_suspend_port() and uart_resume_port() so that serial console
ports are not suspended if CONFIG_DISABLE_CONSOLE_SUSPEND is set.

This makes it possible to debug the suspend and resume routines of all
device drivers as well as the lowest-level swsusp code with the help of the
serial console.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/serial/serial_core.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 372e47f7d596..5f7ba1adb309 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -1929,6 +1929,13 @@ int uart_suspend_port(struct uart_driver *drv, struct uart_port *port)
 
 	mutex_lock(&state->mutex);
 
+#ifdef CONFIG_DISABLE_CONSOLE_SUSPEND
+	if (uart_console(port)) {
+		mutex_unlock(&state->mutex);
+		return 0;
+	}
+#endif
+
 	if (state->info && state->info->flags & UIF_INITIALIZED) {
 		const struct uart_ops *ops = port->ops;
 
@@ -1967,6 +1974,13 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port)
 
 	mutex_lock(&state->mutex);
 
+#ifdef CONFIG_DISABLE_CONSOLE_SUSPEND
+	if (uart_console(port)) {
+		mutex_unlock(&state->mutex);
+		return 0;
+	}
+#endif
+
 	uart_change_pm(state, 0);
 
 	/*
-- 
cgit v1.2.1


From 7d145aa3abf4d96c91f37c012facd5cfbb9010d1 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:57 -0700
Subject: [PATCH] i386: Detect clock skew during suspend

Detect the situations in which the time after a resume from disk would be
earlier than the time before the suspend and prevent them from happening on
i386.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: John Stultz <johnstul@us.ibm.com>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/time.c | 24 ++++++++++++++++++------
 1 file changed, 18 insertions(+), 6 deletions(-)

diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
index 6f333e7fb23c..1302e4ab3c4f 100644
--- a/arch/i386/kernel/time.c
+++ b/arch/i386/kernel/time.c
@@ -270,16 +270,19 @@ void notify_arch_cmos_timer(void)
 	mod_timer(&sync_cmos_timer, jiffies + 1);
 }
 
-static long clock_cmos_diff, sleep_start;
+static long clock_cmos_diff;
+static unsigned long sleep_start;
 
 static int timer_suspend(struct sys_device *dev, pm_message_t state)
 {
 	/*
 	 * Estimate time zone so that set_time can update the clock
 	 */
-	clock_cmos_diff = -get_cmos_time();
+	unsigned long ctime =  get_cmos_time();
+
+	clock_cmos_diff = -ctime;
 	clock_cmos_diff += get_seconds();
-	sleep_start = get_cmos_time();
+	sleep_start = ctime;
 	return 0;
 }
 
@@ -287,16 +290,25 @@ static int timer_resume(struct sys_device *dev)
 {
 	unsigned long flags;
 	unsigned long sec;
-	unsigned long sleep_length;
+	unsigned long ctime = get_cmos_time();
+	long sleep_length = (ctime - sleep_start) * HZ;
 	struct timespec ts;
+
+	if (sleep_length < 0) {
+		printk(KERN_WARNING "CMOS clock skew detected in timer resume!\n");
+		/* The time after the resume must not be earlier than the time
+		 * before the suspend or some nasty things will happen
+		 */
+		sleep_length = 0;
+		ctime = sleep_start;
+	}
 #ifdef CONFIG_HPET_TIMER
 	if (is_hpet_enabled())
 		hpet_reenable();
 #endif
 	setup_pit_timer();
-	sec = get_cmos_time() + clock_cmos_diff;
-	sleep_length = (get_cmos_time() - sleep_start) * HZ;
 
+	sec = ctime + clock_cmos_diff;
 	ts.tv_sec = sec;
 	ts.tv_nsec = 0;
 	do_settimeofday(&ts);
-- 
cgit v1.2.1


From c5c6ba4e08ab9c9e390a0f3a7d9a5c332f5cc6ef Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Mon, 25 Sep 2006 23:32:58 -0700
Subject: [PATCH] PM: Add pm_trace switch

Add the pm_trace attribute in /sys/power which has to be explicitly set to
one to really enable the "PM tracing" code compiled in when CONFIG_PM_TRACE
is set (which modifies the machine's CMOS clock in unpredictable ways).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/power/interface.txt | 15 +++++++++++++++
 include/linux/resume-trace.h      | 24 ++++++++++++++----------
 kernel/power/main.c               | 30 ++++++++++++++++++++++++++++++
 3 files changed, 59 insertions(+), 10 deletions(-)

diff --git a/Documentation/power/interface.txt b/Documentation/power/interface.txt
index 4117802af0f8..a66bec222b16 100644
--- a/Documentation/power/interface.txt
+++ b/Documentation/power/interface.txt
@@ -52,3 +52,18 @@ suspend image will be as small as possible.
 
 Reading from this file will display the current image size limit, which
 is set to 500 MB by default.
+
+/sys/power/pm_trace controls the code which saves the last PM event point in
+the RTC across reboots, so that you can debug a machine that just hangs
+during suspend (or more commonly, during resume).  Namely, the RTC is only
+used to save the last PM event point if this file contains '1'.  Initially it
+contains '0' which may be changed to '1' by writing a string representing a
+nonzero integer into it.
+
+To use this debugging feature you should attempt to suspend the machine, then
+reboot it and run
+
+	dmesg -s 1000000 | grep 'hash matches'
+
+CAUTION: Using it will cause your machine's real-time (CMOS) clock to be
+set to a random invalid time after a resume.
diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h
index a376bd4ade39..81e9299ca148 100644
--- a/include/linux/resume-trace.h
+++ b/include/linux/resume-trace.h
@@ -3,21 +3,25 @@
 
 #ifdef CONFIG_PM_TRACE
 
+extern int pm_trace_enabled;
+
 struct device;
 extern void set_trace_device(struct device *);
 extern void generate_resume_trace(void *tracedata, unsigned int user);
 
 #define TRACE_DEVICE(dev) set_trace_device(dev)
-#define TRACE_RESUME(user) do {				\
-	void *tracedata;				\
-	asm volatile("movl $1f,%0\n"			\
-		".section .tracedata,\"a\"\n"		\
-		"1:\t.word %c1\n"			\
-		"\t.long %c2\n"				\
-		".previous"				\
-		:"=r" (tracedata)			\
-		: "i" (__LINE__), "i" (__FILE__));	\
-	generate_resume_trace(tracedata, user);		\
+#define TRACE_RESUME(user) do {					\
+	if (pm_trace_enabled) {					\
+		void *tracedata;				\
+		asm volatile("movl $1f,%0\n"			\
+			".section .tracedata,\"a\"\n"		\
+			"1:\t.word %c1\n"			\
+			"\t.long %c2\n"				\
+			".previous"				\
+			:"=r" (tracedata)			\
+			: "i" (__LINE__), "i" (__FILE__));	\
+		generate_resume_trace(tracedata, user);		\
+	}							\
 } while (0)
 
 #else
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 4d403323a7bb..873228c71dab 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -17,6 +17,7 @@
 #include <linux/pm.h>
 #include <linux/console.h>
 #include <linux/cpu.h>
+#include <linux/resume-trace.h>
 
 #include "power.h"
 
@@ -281,10 +282,39 @@ static ssize_t state_store(struct subsystem * subsys, const char * buf, size_t n
 
 power_attr(state);
 
+#ifdef CONFIG_PM_TRACE
+int pm_trace_enabled;
+
+static ssize_t pm_trace_show(struct subsystem * subsys, char * buf)
+{
+	return sprintf(buf, "%d\n", pm_trace_enabled);
+}
+
+static ssize_t
+pm_trace_store(struct subsystem * subsys, const char * buf, size_t n)
+{
+	int val;
+
+	if (sscanf(buf, "%d", &val) == 1) {
+		pm_trace_enabled = !!val;
+		return n;
+	}
+	return -EINVAL;
+}
+
+power_attr(pm_trace);
+
+static struct attribute * g[] = {
+	&state_attr.attr,
+	&pm_trace_attr.attr,
+	NULL,
+};
+#else
 static struct attribute * g[] = {
 	&state_attr.attr,
 	NULL,
 };
+#endif /* CONFIG_PM_TRACE */
 
 static struct attribute_group attr_group = {
 	.attrs = g,
-- 
cgit v1.2.1


From 13c06be399902c9ebda08e092edb1614bb4a3761 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:32:59 -0700
Subject: [PATCH] uml: Use klibc setjmp/longjmp

This patch adds an implementation of setjmp and longjmp to UML, allowing
access to the inside of a jmpbuf without needing the access macros formerly
provided by libc.

The implementation is stolen from klibc.  I copy the relevant files into
arch/um.  I have another patch which avoids the copying, but requires klibc be
in the tree.

setjmp and longjmp users required some tweaking.  Includes of <setjmp.h> were
removed and includes of the UML longjmp.h were added where necessary.  There
are also replacements of siglongjmp with UML_LONGJMP which I somehow missed
earlier.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/longjmp.h                  |  5 ++-
 arch/um/include/sysdep-i386/archsetjmp.h   | 19 ++++++++++
 arch/um/include/sysdep-x86_64/archsetjmp.h | 21 +++++++++++
 arch/um/os-Linux/process.c                 |  1 -
 arch/um/os-Linux/skas/process.c            |  3 +-
 arch/um/os-Linux/sys-i386/registers.c      | 10 +++---
 arch/um/os-Linux/sys-x86_64/registers.c    | 10 +++---
 arch/um/os-Linux/trap.c                    |  1 -
 arch/um/os-Linux/uaccess.c                 |  3 +-
 arch/um/os-Linux/util.c                    |  5 ++-
 arch/um/sys-i386/Makefile                  |  2 +-
 arch/um/sys-i386/setjmp.S                  | 58 ++++++++++++++++++++++++++++++
 arch/um/sys-x86_64/Makefile                |  4 +--
 arch/um/sys-x86_64/setjmp.S                | 54 ++++++++++++++++++++++++++++
 14 files changed, 173 insertions(+), 23 deletions(-)
 create mode 100644 arch/um/include/sysdep-i386/archsetjmp.h
 create mode 100644 arch/um/include/sysdep-x86_64/archsetjmp.h
 create mode 100644 arch/um/sys-i386/setjmp.S
 create mode 100644 arch/um/sys-x86_64/setjmp.S

diff --git a/arch/um/include/longjmp.h b/arch/um/include/longjmp.h
index 1b5c0131a12e..e93c6d3e893b 100644
--- a/arch/um/include/longjmp.h
+++ b/arch/um/include/longjmp.h
@@ -1,9 +1,12 @@
 #ifndef __UML_LONGJMP_H
 #define __UML_LONGJMP_H
 
-#include <setjmp.h>
+#include "sysdep/archsetjmp.h"
 #include "os.h"
 
+extern int setjmp(jmp_buf);
+extern void longjmp(jmp_buf, int);
+
 #define UML_LONGJMP(buf, val) do { \
 	longjmp(*buf, val);	\
 } while(0)
diff --git a/arch/um/include/sysdep-i386/archsetjmp.h b/arch/um/include/sysdep-i386/archsetjmp.h
new file mode 100644
index 000000000000..ea1ba3d42aee
--- /dev/null
+++ b/arch/um/include/sysdep-i386/archsetjmp.h
@@ -0,0 +1,19 @@
+/*
+ * arch/i386/include/klibc/archsetjmp.h
+ */
+
+#ifndef _KLIBC_ARCHSETJMP_H
+#define _KLIBC_ARCHSETJMP_H
+
+struct __jmp_buf {
+	unsigned int __ebx;
+	unsigned int __esp;
+	unsigned int __ebp;
+	unsigned int __esi;
+	unsigned int __edi;
+	unsigned int __eip;
+};
+
+typedef struct __jmp_buf jmp_buf[1];
+
+#endif				/* _SETJMP_H */
diff --git a/arch/um/include/sysdep-x86_64/archsetjmp.h b/arch/um/include/sysdep-x86_64/archsetjmp.h
new file mode 100644
index 000000000000..454fc60aff6d
--- /dev/null
+++ b/arch/um/include/sysdep-x86_64/archsetjmp.h
@@ -0,0 +1,21 @@
+/*
+ * arch/x86_64/include/klibc/archsetjmp.h
+ */
+
+#ifndef _KLIBC_ARCHSETJMP_H
+#define _KLIBC_ARCHSETJMP_H
+
+struct __jmp_buf {
+	unsigned long __rbx;
+	unsigned long __rsp;
+	unsigned long __rbp;
+	unsigned long __r12;
+	unsigned long __r13;
+	unsigned long __r14;
+	unsigned long __r15;
+	unsigned long __rip;
+};
+
+typedef struct __jmp_buf jmp_buf[1];
+
+#endif				/* _SETJMP_H */
diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
index b98d3ca2cd1b..3afde92ad2c0 100644
--- a/arch/um/os-Linux/process.c
+++ b/arch/um/os-Linux/process.c
@@ -7,7 +7,6 @@
 #include <stdio.h>
 #include <errno.h>
 #include <signal.h>
-#include <setjmp.h>
 #include <linux/unistd.h>
 #include <sys/mman.h>
 #include <sys/wait.h>
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 7baf90fda58b..50418a5e7134 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -8,7 +8,6 @@
 #include <unistd.h>
 #include <errno.h>
 #include <signal.h>
-#include <setjmp.h>
 #include <sched.h>
 #include "ptrace_user.h"
 #include <sys/wait.h>
@@ -470,7 +469,7 @@ void thread_wait(void *sw, void *fb)
 	*switch_buf = &buf;
 	fork_buf = fb;
 	if(UML_SETJMP(&buf) == 0)
-		siglongjmp(*fork_buf, INIT_JMP_REMOVE_SIGSTACK);
+		UML_LONGJMP(fork_buf, INIT_JMP_REMOVE_SIGSTACK);
 }
 
 void switch_threads(void *me, void *next)
diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c
index 516f66dd87e3..1f90a2d71386 100644
--- a/arch/um/os-Linux/sys-i386/registers.c
+++ b/arch/um/os-Linux/sys-i386/registers.c
@@ -5,12 +5,12 @@
 
 #include <errno.h>
 #include <string.h>
-#include <setjmp.h>
 #include "sysdep/ptrace_user.h"
 #include "sysdep/ptrace.h"
 #include "uml-config.h"
 #include "skas_ptregs.h"
 #include "registers.h"
+#include "longjmp.h"
 #include "user.h"
 
 /* These are set once at boot time and not changed thereafter */
@@ -132,9 +132,9 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 
 void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
 {
-	struct __jmp_buf_tag *jmpbuf = buffer;
+	struct __jmp_buf *jmpbuf = buffer;
 
-	UPT_SET(uml_regs, EIP, jmpbuf->__jmpbuf[JB_PC]);
-	UPT_SET(uml_regs, UESP, jmpbuf->__jmpbuf[JB_SP]);
-	UPT_SET(uml_regs, EBP, jmpbuf->__jmpbuf[JB_BP]);
+	UPT_SET(uml_regs, EIP, jmpbuf->__eip);
+	UPT_SET(uml_regs, UESP, jmpbuf->__esp);
+	UPT_SET(uml_regs, EBP, jmpbuf->__ebp);
 }
diff --git a/arch/um/os-Linux/sys-x86_64/registers.c b/arch/um/os-Linux/sys-x86_64/registers.c
index becd898d9398..e730447d6c02 100644
--- a/arch/um/os-Linux/sys-x86_64/registers.c
+++ b/arch/um/os-Linux/sys-x86_64/registers.c
@@ -5,11 +5,11 @@
 
 #include <errno.h>
 #include <string.h>
-#include <setjmp.h>
 #include "ptrace_user.h"
 #include "uml-config.h"
 #include "skas_ptregs.h"
 #include "registers.h"
+#include "longjmp.h"
 #include "user.h"
 
 /* These are set once at boot time and not changed thereafter */
@@ -80,9 +80,9 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 
 void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
 {
-	struct __jmp_buf_tag *jmpbuf = buffer;
+	struct __jmp_buf *jmpbuf = buffer;
 
-	UPT_SET(uml_regs, RIP, jmpbuf->__jmpbuf[JB_PC]);
-	UPT_SET(uml_regs, RSP, jmpbuf->__jmpbuf[JB_RSP]);
-	UPT_SET(uml_regs, RBP, jmpbuf->__jmpbuf[JB_RBP]);
+	UPT_SET(uml_regs, RIP, jmpbuf->__rip);
+	UPT_SET(uml_regs, RSP, jmpbuf->__rsp);
+	UPT_SET(uml_regs, RBP, jmpbuf->__rbp);
 }
diff --git a/arch/um/os-Linux/trap.c b/arch/um/os-Linux/trap.c
index 90b29ae9af46..1df231a26244 100644
--- a/arch/um/os-Linux/trap.c
+++ b/arch/um/os-Linux/trap.c
@@ -5,7 +5,6 @@
 
 #include <stdlib.h>
 #include <signal.h>
-#include <setjmp.h>
 #include "kern_util.h"
 #include "user_util.h"
 #include "os.h"
diff --git a/arch/um/os-Linux/uaccess.c b/arch/um/os-Linux/uaccess.c
index 865f6a6a2590..bbb73a650370 100644
--- a/arch/um/os-Linux/uaccess.c
+++ b/arch/um/os-Linux/uaccess.c
@@ -4,8 +4,7 @@
  * Licensed under the GPL
  */
 
-#include <setjmp.h>
-#include <string.h>
+#include <stddef.h>
 #include "longjmp.h"
 
 unsigned long __do_user_copy(void *to, const void *from, int n,
diff --git a/arch/um/os-Linux/util.c b/arch/um/os-Linux/util.c
index c47a2a7ce70e..3f5b1514e8a7 100644
--- a/arch/um/os-Linux/util.c
+++ b/arch/um/os-Linux/util.c
@@ -7,7 +7,6 @@
 #include <stdlib.h>
 #include <unistd.h>
 #include <limits.h>
-#include <setjmp.h>
 #include <sys/mman.h>
 #include <sys/stat.h>
 #include <sys/utsname.h>
@@ -107,11 +106,11 @@ int setjmp_wrapper(void (*proc)(void *, void *), ...)
 	jmp_buf buf;
 	int n;
 
-	n = sigsetjmp(buf, 1);
+	n = UML_SETJMP(&buf);
 	if(n == 0){
 		va_start(args, proc);
 		(*proc)(&buf, &args);
 	}
 	va_end(args);
-	return(n);
+	return n;
 }
diff --git a/arch/um/sys-i386/Makefile b/arch/um/sys-i386/Makefile
index 374d61a19439..59cc70275754 100644
--- a/arch/um/sys-i386/Makefile
+++ b/arch/um/sys-i386/Makefile
@@ -1,5 +1,5 @@
 obj-y = bugs.o checksum.o delay.o fault.o ksyms.o ldt.o ptrace.o \
-	ptrace_user.o signal.o sigcontext.o syscalls.o sysrq.o \
+	ptrace_user.o setjmp.o signal.o sigcontext.o syscalls.o sysrq.o \
 	sys_call_table.o tls.o
 
 obj-$(CONFIG_MODE_SKAS) += stub.o stub_segv.o
diff --git a/arch/um/sys-i386/setjmp.S b/arch/um/sys-i386/setjmp.S
new file mode 100644
index 000000000000..b766792c9933
--- /dev/null
+++ b/arch/um/sys-i386/setjmp.S
@@ -0,0 +1,58 @@
+#
+# arch/i386/setjmp.S
+#
+# setjmp/longjmp for the i386 architecture
+#
+
+#
+# The jmp_buf is assumed to contain the following, in order:
+#	%ebx
+#	%esp
+#	%ebp
+#	%esi
+#	%edi
+#	<return address>
+#
+
+	.text
+	.align 4
+	.globl setjmp
+	.type setjmp, @function
+setjmp:
+#ifdef _REGPARM
+	movl %eax,%edx
+#else
+	movl 4(%esp),%edx
+#endif
+	popl %ecx			# Return address, and adjust the stack
+	xorl %eax,%eax			# Return value
+	movl %ebx,(%edx)
+	movl %esp,4(%edx)		# Post-return %esp!
+	pushl %ecx			# Make the call/return stack happy
+	movl %ebp,8(%edx)
+	movl %esi,12(%edx)
+	movl %edi,16(%edx)
+	movl %ecx,20(%edx)		# Return address
+	ret
+
+	.size setjmp,.-setjmp
+
+	.text
+	.align 4
+	.globl longjmp
+	.type longjmp, @function
+longjmp:
+#ifdef _REGPARM
+	xchgl %eax,%edx
+#else
+	movl 4(%esp),%edx		# jmp_ptr address
+	movl 8(%esp),%eax		# Return value
+#endif
+	movl (%edx),%ebx
+	movl 4(%edx),%esp
+	movl 8(%edx),%ebp
+	movl 12(%edx),%esi
+	movl 16(%edx),%edi
+	jmp *20(%edx)
+
+	.size longjmp,.-longjmp
diff --git a/arch/um/sys-x86_64/Makefile b/arch/um/sys-x86_64/Makefile
index c19794d435d6..f41768b8e25e 100644
--- a/arch/um/sys-x86_64/Makefile
+++ b/arch/um/sys-x86_64/Makefile
@@ -5,8 +5,8 @@
 #
 
 obj-y = bugs.o delay.o fault.o ldt.o mem.o ptrace.o ptrace_user.o \
-	sigcontext.o signal.o syscalls.o syscall_table.o sysrq.o ksyms.o \
-	tls.o
+	setjmp.o sigcontext.o signal.o syscalls.o syscall_table.o sysrq.o \
+	ksyms.o tls.o
 
 obj-$(CONFIG_MODE_SKAS) += stub.o stub_segv.o
 obj-$(CONFIG_MODULES) += um_module.o
diff --git a/arch/um/sys-x86_64/setjmp.S b/arch/um/sys-x86_64/setjmp.S
new file mode 100644
index 000000000000..45f547b4043e
--- /dev/null
+++ b/arch/um/sys-x86_64/setjmp.S
@@ -0,0 +1,54 @@
+#
+# arch/x86_64/setjmp.S
+#
+# setjmp/longjmp for the x86-64 architecture
+#
+
+#
+# The jmp_buf is assumed to contain the following, in order:
+#	%rbx
+#	%rsp (post-return)
+#	%rbp
+#	%r12
+#	%r13
+#	%r14
+#	%r15
+#	<return address>
+#
+
+	.text
+	.align 4
+	.globl setjmp
+	.type setjmp, @function
+setjmp:
+	pop  %rsi			# Return address, and adjust the stack
+	xorl %eax,%eax			# Return value
+	movq %rbx,(%rdi)
+	movq %rsp,8(%rdi)		# Post-return %rsp!
+	push %rsi			# Make the call/return stack happy
+	movq %rbp,16(%rdi)
+	movq %r12,24(%rdi)
+	movq %r13,32(%rdi)
+	movq %r14,40(%rdi)
+	movq %r15,48(%rdi)
+	movq %rsi,56(%rdi)		# Return address
+	ret
+
+	.size setjmp,.-setjmp
+
+	.text
+	.align 4
+	.globl longjmp
+	.type longjmp, @function
+longjmp:
+	movl %esi,%eax			# Return value (int)
+	movq (%rdi),%rbx
+	movq 8(%rdi),%rsp
+	movq 16(%rdi),%rbp
+	movq 24(%rdi),%r12
+	movq 32(%rdi),%r13
+	movq 40(%rdi),%r14
+	movq 48(%rdi),%r15
+	jmp *56(%rdi)
+
+	.size longjmp,.-longjmp
-- 
cgit v1.2.1


From 91b165c0594ab78c64f26d26e3174e6dfd60ed9d Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:00 -0700
Subject: [PATCH] uml: Use ARRAY_SIZE more assiduously

There were a bunch of missed ARRAY_SIZE opportunities.

Also, some formatting fixes in the affected areas of code.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/chan_kern.c     |  2 +-
 arch/um/drivers/mconsole_kern.c |  2 +-
 arch/um/drivers/mconsole_user.c |  7 ++++---
 arch/um/drivers/pcap_kern.c     |  2 +-
 arch/um/kernel/mem.c            |  3 ++-
 arch/um/kernel/reboot.c         | 13 +------------
 arch/um/kernel/tlb.c            |  5 +++--
 arch/um/os-Linux/mem.c          |  6 +++---
 arch/um/sys-i386/bugs.c         |  9 +++++----
 arch/um/sys-i386/ldt.c          |  3 +--
 arch/um/sys-i386/ptrace_user.c  |  5 +++--
 11 files changed, 25 insertions(+), 32 deletions(-)

diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c
index 7218c754505b..e82764f75e7f 100644
--- a/arch/um/drivers/chan_kern.c
+++ b/arch/um/drivers/chan_kern.c
@@ -544,7 +544,7 @@ static struct chan *parse_chan(struct line *line, char *str, int device,
 
 	ops = NULL;
 	data = NULL;
-	for(i = 0; i < sizeof(chan_table)/sizeof(chan_table[0]); i++){
+	for(i = 0; i < ARRAY_SIZE(chan_table); i++){
 		entry = &chan_table[i];
 		if(!strncmp(str, entry->key, strlen(entry->key))){
 			ops = entry->ops;
diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c
index b414522f7686..79610b5ce67e 100644
--- a/arch/um/drivers/mconsole_kern.c
+++ b/arch/um/drivers/mconsole_kern.c
@@ -497,7 +497,7 @@ static void mconsole_get_config(int (*get_config)(char *, char *, int,
 	}
 
 	error = NULL;
-	size = sizeof(default_buf)/sizeof(default_buf[0]);
+	size = ARRAY_SIZE(default_buf);
 	buf = default_buf;
 
 	while(1){
diff --git a/arch/um/drivers/mconsole_user.c b/arch/um/drivers/mconsole_user.c
index 9bfd405c3bd8..5b2f5fe9e426 100644
--- a/arch/um/drivers/mconsole_user.c
+++ b/arch/um/drivers/mconsole_user.c
@@ -16,6 +16,7 @@
 #include "user.h"
 #include "mconsole.h"
 #include "umid.h"
+#include "user_util.h"
 
 static struct mconsole_command commands[] = {
 	/* With uts namespaces, uts information becomes process-specific, so
@@ -65,14 +66,14 @@ static struct mconsole_command *mconsole_parse(struct mc_request *req)
 	struct mconsole_command *cmd;
 	int i;
 
-	for(i=0;i<sizeof(commands)/sizeof(commands[0]);i++){
+	for(i = 0; i < ARRAY_SIZE(commands); i++){
 		cmd = &commands[i];
 		if(!strncmp(req->request.data, cmd->command, 
 			    strlen(cmd->command))){
-			return(cmd);
+			return cmd;
 		}
 	}
-	return(NULL);
+	return NULL;
 }
 
 #define MIN(a,b) ((a)<(b) ? (a):(b))
diff --git a/arch/um/drivers/pcap_kern.c b/arch/um/drivers/pcap_kern.c
index 466ff2c2f918..4c767c7adb96 100644
--- a/arch/um/drivers/pcap_kern.c
+++ b/arch/um/drivers/pcap_kern.c
@@ -76,7 +76,7 @@ int pcap_setup(char *str, char **mac_out, void *data)
 	if(host_if != NULL)
 		init->host_if = host_if;
 
-	for(i = 0; i < sizeof(options)/sizeof(options[0]); i++){
+	for(i = 0; i < ARRAY_SIZE(options); i++){
 		if(options[i] == NULL)
 			continue;
 		if(!strcmp(options[i], "promisc"))
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index b1cd5c6e468b..93121c6d26e5 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -223,8 +223,9 @@ void paging_init(void)
 
 	empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE);
 	empty_bad_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE);
-	for(i=0;i<sizeof(zones_size)/sizeof(zones_size[0]);i++) 
+	for(i = 0; i < ARRAY_SIZE(zones_size); i++)
 		zones_size[i] = 0;
+
 	zones_size[ZONE_DMA] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT);
 #ifdef CONFIG_HIGHMEM
 	zones_size[ZONE_HIGHMEM] = highmem >> PAGE_SHIFT;
diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c
index 3ef73bf2e781..f602623644aa 100644
--- a/arch/um/kernel/reboot.c
+++ b/arch/um/kernel/reboot.c
@@ -22,7 +22,7 @@ static void kill_idlers(int me)
 	struct task_struct *p;
 	int i;
 
-	for(i = 0; i < sizeof(idle_threads)/sizeof(idle_threads[0]); i++){
+	for(i = 0; i < ARRAY_SIZE(idle_threads); i++){
 		p = idle_threads[i];
 		if((p != NULL) && (p->thread.mode.tt.extern_pid != me))
 			os_kill_process(p->thread.mode.tt.extern_pid, 0);
@@ -62,14 +62,3 @@ void machine_halt(void)
 {
 	machine_power_off();
 }
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c
index f5b0636f9ad7..cca330edf717 100644
--- a/arch/um/kernel/tlb.c
+++ b/arch/um/kernel/tlb.c
@@ -137,10 +137,11 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr,
         int r, w, x;
         struct host_vm_op ops[1];
         void *flush = NULL;
-        int op_index = -1, last_op = sizeof(ops) / sizeof(ops[0]) - 1;
+        int op_index = -1, last_op = ARRAY_SIZE(ops) - 1;
         int ret = 0;
 
-        if(mm == NULL) return;
+        if(mm == NULL)
+		return;
 
         ops[0].type = NONE;
         for(addr = start_addr; addr < end_addr && !ret;){
diff --git a/arch/um/os-Linux/mem.c b/arch/um/os-Linux/mem.c
index 560c8063c77c..b170b4704dc4 100644
--- a/arch/um/os-Linux/mem.c
+++ b/arch/um/os-Linux/mem.c
@@ -114,14 +114,14 @@ static void which_tmpdir(void)
 	}
 
 	while(1){
-		found = next(fd, buf, sizeof(buf) / sizeof(buf[0]), ' ');
+		found = next(fd, buf, ARRAY_SIZE(buf), ' ');
 		if(found != 1)
 			break;
 
 		if(!strncmp(buf, "/dev/shm", strlen("/dev/shm")))
 			goto found;
 
-		found = next(fd, buf, sizeof(buf) / sizeof(buf[0]), '\n');
+		found = next(fd, buf, ARRAY_SIZE(buf), '\n');
 		if(found != 1)
 			break;
 	}
@@ -135,7 +135,7 @@ err:
 	return;
 
 found:
-	found = next(fd, buf, sizeof(buf) / sizeof(buf[0]), ' ');
+	found = next(fd, buf, ARRAY_SIZE(buf), ' ');
 	if(found != 1)
 		goto err;
 
diff --git a/arch/um/sys-i386/bugs.c b/arch/um/sys-i386/bugs.c
index 41b0ab2fe830..f1bcd399ac90 100644
--- a/arch/um/sys-i386/bugs.c
+++ b/arch/um/sys-i386/bugs.c
@@ -13,6 +13,7 @@
 #include "sysdep/ptrace.h"
 #include "task.h"
 #include "os.h"
+#include "user_util.h"
 
 #define MAXTOKEN 64
 
@@ -104,17 +105,17 @@ int cpu_feature(char *what, char *buf, int len)
 static int check_cpu_flag(char *feature, int *have_it)
 {
 	char buf[MAXTOKEN], c;
-	int fd, len = sizeof(buf)/sizeof(buf[0]);
+	int fd, len = ARRAY_SIZE(buf);
 
 	printk("Checking for host processor %s support...", feature);
 	fd = os_open_file("/proc/cpuinfo", of_read(OPENFLAGS()), 0);
 	if(fd < 0){
 		printk("Couldn't open /proc/cpuinfo, err = %d\n", -fd);
-		return(0);
+		return 0;
 	}
 
 	*have_it = 0;
-	if(!find_cpuinfo_line(fd, "flags", buf, sizeof(buf) / sizeof(buf[0])))
+	if(!find_cpuinfo_line(fd, "flags", buf, ARRAY_SIZE(buf)))
 		goto out;
 
 	c = token(fd, buf, len - 1, ' ');
@@ -138,7 +139,7 @@ static int check_cpu_flag(char *feature, int *have_it)
 	if(*have_it == 0) printk("No\n");
 	else if(*have_it == 1) printk("Yes\n");
 	os_close_file(fd);
-	return(1);
+	return 1;
 }
 
 #if 0 /* This doesn't work in tt mode, plus it's causing compilation problems
diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c
index fe0877b3509c..69971b78beaf 100644
--- a/arch/um/sys-i386/ldt.c
+++ b/arch/um/sys-i386/ldt.c
@@ -424,9 +424,8 @@ void ldt_get_host_info(void)
 			size++;
 	}
 
-	if(size < sizeof(dummy_list)/sizeof(dummy_list[0])) {
+	if(size < ARRAY_SIZE(dummy_list))
 		host_ldt_entries = dummy_list;
-	}
 	else {
 		size = (size + 1) * sizeof(dummy_list[0]);
 		host_ldt_entries = (short *)kmalloc(size, GFP_KERNEL);
diff --git a/arch/um/sys-i386/ptrace_user.c b/arch/um/sys-i386/ptrace_user.c
index 40aa88531446..5f3cc6685820 100644
--- a/arch/um/sys-i386/ptrace_user.c
+++ b/arch/um/sys-i386/ptrace_user.c
@@ -15,6 +15,7 @@
 #include "user.h"
 #include "os.h"
 #include "uml-config.h"
+#include "user_util.h"
 
 int ptrace_getregs(long pid, unsigned long *regs_out)
 {
@@ -51,7 +52,7 @@ static void write_debugregs(int pid, unsigned long *regs)
 	int nregs, i;
 
 	dummy = NULL;
-	nregs = sizeof(dummy->u_debugreg)/sizeof(dummy->u_debugreg[0]);
+	nregs = ARRAY_SIZE(dummy->u_debugreg);
 	for(i = 0; i < nregs; i++){
 		if((i == 4) || (i == 5)) continue;
 		if(ptrace(PTRACE_POKEUSR, pid, &dummy->u_debugreg[i],
@@ -68,7 +69,7 @@ static void read_debugregs(int pid, unsigned long *regs)
 	int nregs, i;
 
 	dummy = NULL;
-	nregs = sizeof(dummy->u_debugreg)/sizeof(dummy->u_debugreg[0]);
+	nregs = ARRAY_SIZE(dummy->u_debugreg);
 	for(i = 0; i < nregs; i++){
 		regs[i] = ptrace(PTRACE_PEEKUSR, pid,
 				 &dummy->u_debugreg[i], 0);
-- 
cgit v1.2.1


From 8f80e9466e18288df7391c9d21532c4125ac9c62 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:01 -0700
Subject: [PATCH] uml: Fix stack alignment

Stack randomization needs to be conditional on the personality allowing it.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/process_kern.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
index f6a5a502120b..537895d68ad1 100644
--- a/arch/um/kernel/process_kern.c
+++ b/arch/um/kernel/process_kern.c
@@ -23,6 +23,7 @@
 #include "linux/proc_fs.h"
 #include "linux/ptrace.h"
 #include "linux/random.h"
+#include "linux/personality.h"
 #include "asm/unistd.h"
 #include "asm/mman.h"
 #include "asm/segment.h"
@@ -476,7 +477,7 @@ int singlestepping(void * t)
 #ifndef arch_align_stack
 unsigned long arch_align_stack(unsigned long sp)
 {
-	if (randomize_va_space)
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
 		sp -= get_random_int() % 8192;
 	return sp & ~0xf;
 }
-- 
cgit v1.2.1


From 5e1f65a67d76341795ea527d30bfdca03999d46b Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:01 -0700
Subject: [PATCH] uml: Whitespace fixes

arch/um/kernel/tlb.c had some pretty serious whitespace problems.  I also
fixed some returns.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/tlb.c | 367 +++++++++++++++++++++++++--------------------------
 1 file changed, 183 insertions(+), 184 deletions(-)

diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c
index cca330edf717..54a5ff25645a 100644
--- a/arch/um/kernel/tlb.c
+++ b/arch/um/kernel/tlb.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
  * Licensed under the GPL
  */
@@ -16,12 +16,12 @@
 #include "os.h"
 
 static int add_mmap(unsigned long virt, unsigned long phys, unsigned long len,
- 		    int r, int w, int x, struct host_vm_op *ops, int *index,
+		    int r, int w, int x, struct host_vm_op *ops, int *index,
 		    int last_filled, union mm_context *mmu, void **flush,
 		    int (*do_ops)(union mm_context *, struct host_vm_op *,
 				  int, int, void **))
 {
-        __u64 offset;
+	__u64 offset;
 	struct host_vm_op *last;
 	int fd, ret = 0;
 
@@ -89,7 +89,7 @@ static int add_munmap(unsigned long addr, unsigned long len,
 static int add_mprotect(unsigned long addr, unsigned long len, int r, int w,
 			int x, struct host_vm_op *ops, int *index,
 			int last_filled, union mm_context *mmu, void **flush,
- 			int (*do_ops)(union mm_context *, struct host_vm_op *,
+			int (*do_ops)(union mm_context *, struct host_vm_op *,
 				      int, int, void **))
 {
 	struct host_vm_op *last;
@@ -124,106 +124,105 @@ static int add_mprotect(unsigned long addr, unsigned long len, int r, int w,
 #define ADD_ROUND(n, inc) (((n) + (inc)) & ~((inc) - 1))
 
 void fix_range_common(struct mm_struct *mm, unsigned long start_addr,
-                      unsigned long end_addr, int force,
+		      unsigned long end_addr, int force,
 		      int (*do_ops)(union mm_context *, struct host_vm_op *,
 				    int, int, void **))
 {
-        pgd_t *npgd;
-        pud_t *npud;
-        pmd_t *npmd;
-        pte_t *npte;
-        union mm_context *mmu = &mm->context;
-        unsigned long addr, end;
-        int r, w, x;
-        struct host_vm_op ops[1];
-        void *flush = NULL;
-        int op_index = -1, last_op = ARRAY_SIZE(ops) - 1;
-        int ret = 0;
-
-        if(mm == NULL)
+	pgd_t *npgd;
+	pud_t *npud;
+	pmd_t *npmd;
+	pte_t *npte;
+	union mm_context *mmu = &mm->context;
+	unsigned long addr, end;
+	int r, w, x;
+	struct host_vm_op ops[1];
+	void *flush = NULL;
+	int op_index = -1, last_op = ARRAY_SIZE(ops) - 1;
+	int ret = 0;
+
+	if(mm == NULL)
 		return;
 
-        ops[0].type = NONE;
-        for(addr = start_addr; addr < end_addr && !ret;){
-                npgd = pgd_offset(mm, addr);
-                if(!pgd_present(*npgd)){
-                        end = ADD_ROUND(addr, PGDIR_SIZE);
-                        if(end > end_addr)
-                                end = end_addr;
-                        if(force || pgd_newpage(*npgd)){
-                                ret = add_munmap(addr, end - addr, ops,
-                                                 &op_index, last_op, mmu,
-                                                 &flush, do_ops);
-                                pgd_mkuptodate(*npgd);
-                        }
-                        addr = end;
-                        continue;
-                }
-
-                npud = pud_offset(npgd, addr);
-                if(!pud_present(*npud)){
-                        end = ADD_ROUND(addr, PUD_SIZE);
-                        if(end > end_addr)
-                                end = end_addr;
-                        if(force || pud_newpage(*npud)){
-                                ret = add_munmap(addr, end - addr, ops,
-                                                 &op_index, last_op, mmu,
-                                                 &flush, do_ops);
-                                pud_mkuptodate(*npud);
-                        }
-                        addr = end;
-                        continue;
-                }
-
-                npmd = pmd_offset(npud, addr);
-                if(!pmd_present(*npmd)){
-                        end = ADD_ROUND(addr, PMD_SIZE);
-                        if(end > end_addr)
-                                end = end_addr;
-                        if(force || pmd_newpage(*npmd)){
-                                ret = add_munmap(addr, end - addr, ops,
-                                                 &op_index, last_op, mmu,
-                                                 &flush, do_ops);
-                                pmd_mkuptodate(*npmd);
-                        }
-                        addr = end;
-                        continue;
-                }
-
-                npte = pte_offset_kernel(npmd, addr);
-                r = pte_read(*npte);
-                w = pte_write(*npte);
-                x = pte_exec(*npte);
+	ops[0].type = NONE;
+	for(addr = start_addr; addr < end_addr && !ret;){
+		npgd = pgd_offset(mm, addr);
+		if(!pgd_present(*npgd)){
+			end = ADD_ROUND(addr, PGDIR_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pgd_newpage(*npgd)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pgd_mkuptodate(*npgd);
+			}
+			addr = end;
+			continue;
+		}
+
+		npud = pud_offset(npgd, addr);
+		if(!pud_present(*npud)){
+			end = ADD_ROUND(addr, PUD_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pud_newpage(*npud)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pud_mkuptodate(*npud);
+			}
+			addr = end;
+			continue;
+		}
+
+		npmd = pmd_offset(npud, addr);
+		if(!pmd_present(*npmd)){
+			end = ADD_ROUND(addr, PMD_SIZE);
+			if(end > end_addr)
+				end = end_addr;
+			if(force || pmd_newpage(*npmd)){
+				ret = add_munmap(addr, end - addr, ops,
+						 &op_index, last_op, mmu,
+						 &flush, do_ops);
+				pmd_mkuptodate(*npmd);
+			}
+			addr = end;
+			continue;
+		}
+
+		npte = pte_offset_kernel(npmd, addr);
+		r = pte_read(*npte);
+		w = pte_write(*npte);
+		x = pte_exec(*npte);
 		if (!pte_young(*npte)) {
 			r = 0;
 			w = 0;
 		} else if (!pte_dirty(*npte)) {
 			w = 0;
 		}
-                if(force || pte_newpage(*npte)){
-                        if(pte_present(*npte))
-			  ret = add_mmap(addr,
-					 pte_val(*npte) & PAGE_MASK,
-					 PAGE_SIZE, r, w, x, ops,
-					 &op_index, last_op, mmu,
-					 &flush, do_ops);
+		if(force || pte_newpage(*npte)){
+			if(pte_present(*npte))
+				ret = add_mmap(addr,
+					       pte_val(*npte) & PAGE_MASK,
+					       PAGE_SIZE, r, w, x, ops,
+					       &op_index, last_op, mmu,
+					       &flush, do_ops);
 			else ret = add_munmap(addr, PAGE_SIZE, ops,
 					      &op_index, last_op, mmu,
 					      &flush, do_ops);
-                }
-                else if(pte_newprot(*npte))
+		}
+		else if(pte_newprot(*npte))
 			ret = add_mprotect(addr, PAGE_SIZE, r, w, x, ops,
 					   &op_index, last_op, mmu,
 					   &flush, do_ops);
 
-                *npte = pte_mkuptodate(*npte);
-                addr += PAGE_SIZE;
-        }
-
+		*npte = pte_mkuptodate(*npte);
+		addr += PAGE_SIZE;
+	}
 	if(!ret)
 		ret = (*do_ops)(mmu, ops, op_index, 1, &flush);
 
-	/* This is not an else because ret is modified above */
+/* This is not an else because ret is modified above */
 	if(ret) {
 		printk("fix_range_common: failed, killing current process\n");
 		force_sig(SIGKILL, current);
@@ -232,160 +231,160 @@ void fix_range_common(struct mm_struct *mm, unsigned long start_addr,
 
 int flush_tlb_kernel_range_common(unsigned long start, unsigned long end)
 {
-        struct mm_struct *mm;
-        pgd_t *pgd;
-        pud_t *pud;
-        pmd_t *pmd;
-        pte_t *pte;
-        unsigned long addr, last;
-        int updated = 0, err;
-
-        mm = &init_mm;
-        for(addr = start; addr < end;){
-                pgd = pgd_offset(mm, addr);
-                if(!pgd_present(*pgd)){
-                        last = ADD_ROUND(addr, PGDIR_SIZE);
-                        if(last > end)
-                                last = end;
-                        if(pgd_newpage(*pgd)){
-                                updated = 1;
-                                err = os_unmap_memory((void *) addr,
-                                                      last - addr);
-                                if(err < 0)
-                                        panic("munmap failed, errno = %d\n",
-                                              -err);
-                        }
-                        addr = last;
-                        continue;
-                }
-
-                pud = pud_offset(pgd, addr);
-                if(!pud_present(*pud)){
-                        last = ADD_ROUND(addr, PUD_SIZE);
-                        if(last > end)
-                                last = end;
-                        if(pud_newpage(*pud)){
-                                updated = 1;
-                                err = os_unmap_memory((void *) addr,
-                                                      last - addr);
-                                if(err < 0)
-                                        panic("munmap failed, errno = %d\n",
-                                              -err);
-                        }
-                        addr = last;
-                        continue;
-                }
-
-                pmd = pmd_offset(pud, addr);
-                if(!pmd_present(*pmd)){
-                        last = ADD_ROUND(addr, PMD_SIZE);
-                        if(last > end)
-                                last = end;
-                        if(pmd_newpage(*pmd)){
-                                updated = 1;
-                                err = os_unmap_memory((void *) addr,
-                                                      last - addr);
-                                if(err < 0)
-                                        panic("munmap failed, errno = %d\n",
-                                              -err);
-                        }
-                        addr = last;
-                        continue;
-                }
-
-                pte = pte_offset_kernel(pmd, addr);
-                if(!pte_present(*pte) || pte_newpage(*pte)){
-                        updated = 1;
-                        err = os_unmap_memory((void *) addr,
-                                              PAGE_SIZE);
-                        if(err < 0)
-                                panic("munmap failed, errno = %d\n",
-                                      -err);
-                        if(pte_present(*pte))
-                                map_memory(addr,
-                                           pte_val(*pte) & PAGE_MASK,
-                                           PAGE_SIZE, 1, 1, 1);
-                }
-                else if(pte_newprot(*pte)){
-                        updated = 1;
-                        os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1);
-                }
-                addr += PAGE_SIZE;
-        }
-        return(updated);
+	struct mm_struct *mm;
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long addr, last;
+	int updated = 0, err;
+
+	mm = &init_mm;
+	for(addr = start; addr < end;){
+		pgd = pgd_offset(mm, addr);
+		if(!pgd_present(*pgd)){
+			last = ADD_ROUND(addr, PGDIR_SIZE);
+			if(last > end)
+				last = end;
+			if(pgd_newpage(*pgd)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pud = pud_offset(pgd, addr);
+		if(!pud_present(*pud)){
+			last = ADD_ROUND(addr, PUD_SIZE);
+			if(last > end)
+				last = end;
+			if(pud_newpage(*pud)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pmd = pmd_offset(pud, addr);
+		if(!pmd_present(*pmd)){
+			last = ADD_ROUND(addr, PMD_SIZE);
+			if(last > end)
+				last = end;
+			if(pmd_newpage(*pmd)){
+				updated = 1;
+				err = os_unmap_memory((void *) addr,
+						      last - addr);
+				if(err < 0)
+					panic("munmap failed, errno = %d\n",
+					      -err);
+			}
+			addr = last;
+			continue;
+		}
+
+		pte = pte_offset_kernel(pmd, addr);
+		if(!pte_present(*pte) || pte_newpage(*pte)){
+			updated = 1;
+			err = os_unmap_memory((void *) addr,
+					      PAGE_SIZE);
+			if(err < 0)
+				panic("munmap failed, errno = %d\n",
+				      -err);
+			if(pte_present(*pte))
+				map_memory(addr,
+					   pte_val(*pte) & PAGE_MASK,
+					   PAGE_SIZE, 1, 1, 1);
+		}
+		else if(pte_newprot(*pte)){
+			updated = 1;
+			os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1);
+		}
+		addr += PAGE_SIZE;
+	}
+	return(updated);
 }
 
 pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address)
 {
-        return(pgd_offset(mm, address));
+	return(pgd_offset(mm, address));
 }
 
 pud_t *pud_offset_proc(pgd_t *pgd, unsigned long address)
 {
-        return(pud_offset(pgd, address));
+	return(pud_offset(pgd, address));
 }
 
 pmd_t *pmd_offset_proc(pud_t *pud, unsigned long address)
 {
-        return(pmd_offset(pud, address));
+	return(pmd_offset(pud, address));
 }
 
 pte_t *pte_offset_proc(pmd_t *pmd, unsigned long address)
 {
-        return(pte_offset_kernel(pmd, address));
+	return(pte_offset_kernel(pmd, address));
 }
 
 pte_t *addr_pte(struct task_struct *task, unsigned long addr)
 {
-        pgd_t *pgd = pgd_offset(task->mm, addr);
-        pud_t *pud = pud_offset(pgd, addr);
-        pmd_t *pmd = pmd_offset(pud, addr);
+	pgd_t *pgd = pgd_offset(task->mm, addr);
+	pud_t *pud = pud_offset(pgd, addr);
+	pmd_t *pmd = pmd_offset(pud, addr);
 
-        return(pte_offset_map(pmd, addr));
+	return(pte_offset_map(pmd, addr));
 }
 
 void flush_tlb_page(struct vm_area_struct *vma, unsigned long address)
 {
-        address &= PAGE_MASK;
-        flush_tlb_range(vma, address, address + PAGE_SIZE);
+	address &= PAGE_MASK;
+	flush_tlb_range(vma, address, address + PAGE_SIZE);
 }
 
 void flush_tlb_all(void)
 {
-        flush_tlb_mm(current->mm);
+	flush_tlb_mm(current->mm);
 }
 
 void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 {
-        CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt,
-                         flush_tlb_kernel_range_common, start, end);
+	CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt,
+			 flush_tlb_kernel_range_common, start, end);
 }
 
 void flush_tlb_kernel_vm(void)
 {
-        CHOOSE_MODE(flush_tlb_kernel_vm_tt(),
-                    flush_tlb_kernel_range_common(start_vm, end_vm));
+	CHOOSE_MODE(flush_tlb_kernel_vm_tt(),
+		    flush_tlb_kernel_range_common(start_vm, end_vm));
 }
 
 void __flush_tlb_one(unsigned long addr)
 {
-        CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr);
+	CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr);
 }
 
 void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 		     unsigned long end)
 {
-        CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start,
-                         end);
+	CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start,
+			 end);
 }
 
 void flush_tlb_mm(struct mm_struct *mm)
 {
-        CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm);
+	CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm);
 }
 
 void force_flush_all(void)
 {
-        CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas());
+	CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas());
 }
 
-- 
cgit v1.2.1


From 6b7aaad9ba4f2a059a70014be12a921eceebfc47 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:02 -0700
Subject: [PATCH] uml: Fix handling of failed execs of helpers

There were some bugs in handling failures to exec helper programs.  errno was
passed back from the child with the wrong sign.  It was also ignored.  In the
case where it mattered, the errno from the (successful) read in the parent was
used instead.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/os-Linux/helper.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/arch/um/os-Linux/helper.c b/arch/um/os-Linux/helper.c
index 6987d1d247a2..cd15b9df5b5c 100644
--- a/arch/um/os-Linux/helper.c
+++ b/arch/um/os-Linux/helper.c
@@ -42,7 +42,7 @@ static int helper_child(void *arg)
 	if(data->pre_exec != NULL)
 		(*data->pre_exec)(data->pre_data);
 	execvp(argv[0], argv);
-	errval = errno;
+	errval = -errno;
 	printk("helper_child - execve of '%s' failed - errno = %d\n", argv[0], errno);
 	os_write_file(data->fd, &errval, sizeof(errval));
 	kill(os_getpid(), SIGKILL);
@@ -62,7 +62,7 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv,
 		stack = *stack_out;
 	else stack = alloc_stack(0, __cant_sleep());
 	if(stack == 0)
-		return(-ENOMEM);
+		return -ENOMEM;
 
 	ret = os_pipe(fds, 1, 0);
 	if(ret < 0){
@@ -95,16 +95,16 @@ int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv,
 	/* Read the errno value from the child, if the exec failed, or get 0 if
 	 * the exec succeeded because the pipe fd was set as close-on-exec. */
 	n = os_read_file(fds[0], &ret, sizeof(ret));
-	if (n < 0) {
-		printk("run_helper : read on pipe failed, ret = %d\n", -n);
-		ret = n;
-		kill(pid, SIGKILL);
-		CATCH_EINTR(waitpid(pid, NULL, 0));
-	} else if(n != 0){
-		CATCH_EINTR(n = waitpid(pid, NULL, 0));
-		ret = -errno;
-	} else {
+	if(n == 0)
 		ret = pid;
+	else {
+		if(n < 0){
+			printk("run_helper : read on pipe failed, ret = %d\n",
+			       -n);
+			ret = n;
+			kill(pid, SIGKILL);
+		}
+		CATCH_EINTR(waitpid(pid, NULL, 0));
 	}
 
 out_close:
-- 
cgit v1.2.1


From 6edf428ed177e333863a8e5c37751a9ec176f241 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:03 -0700
Subject: [PATCH] uml: Improve SIGBUS diagnostics

UML can get a SIGBUS anywhere if the tmpfs mount being used for its memory
runs out of space.  This patch adds a printk before the panic to provide a
clue as to what likely went wrong.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/trap.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index ac70fa5a2e2a..e5eeaf2b6af1 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -227,9 +227,16 @@ void bad_segv(struct faultinfo fi, unsigned long ip)
 
 void relay_signal(int sig, union uml_pt_regs *regs)
 {
-	if(arch_handle_signal(sig, regs)) return;
-	if(!UPT_IS_USER(regs))
+	if(arch_handle_signal(sig, regs))
+		return;
+
+	if(!UPT_IS_USER(regs)){
+		if(sig == SIGBUS)
+			printk("Bus error - the /dev/shm or /tmp mount likely "
+			       "just ran out of space\n");
 		panic("Kernel mode signal %d", sig);
+	}
+
         current->thread.arch.faultinfo = *UPT_FAULTINFO(regs);
 	force_sig(sig, current);
 }
-- 
cgit v1.2.1


From 19bdf0409f25a85a45874a5a8da6f3e4edcf4a49 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:04 -0700
Subject: [PATCH] uml: SIGIO cleanups

- Various cleanups in the sigio code.

- Removed explicit zero-initializations of a few structures.

- Improved some error messages.

- An API change - there was an asymmetry between reactivate_fd calling
  maybe_sigio_broken, which goes through all the machinery of figuring out if
  a file descriptor supports SIGIO and applying the workaround to it if not,
  and deactivate_fd, which just turns off the descriptor.

  This is changed so that only activate_fd calls maybe_sigio_broken, when
  the descriptor is first seen.  reactivate_fd now calls add_sigio_fd, which
  is symmetric with ignore_sigio_fd.

  This removes a recursion which makes a critical section look more critical
  than it really was, obsoleting a big comment to that effect.  This requires
  keeping track of all descriptors which are getting the SIGIO treatment, not
  just the ones being polled at any given moment, so that reactivate_fd,
  through add_sigio_fd, doesn't try to tell the SIGIO thread about descriptors
  it doesn't care about.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h     |   1 +
 arch/um/kernel/irq.c     |  34 +++++-----------
 arch/um/os-Linux/sigio.c | 103 ++++++++++++++++++++++++++---------------------
 3 files changed, 67 insertions(+), 71 deletions(-)

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index 5316e8a4a4fd..c73dfa78f943 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -329,6 +329,7 @@ extern void os_set_ioignore(void);
 extern void init_irq_signals(int on_sigstack);
 
 /* sigio.c */
+extern int add_sigio_fd(int fd);
 extern int ignore_sigio_fd(int fd);
 extern void maybe_sigio_broken(int fd, int read);
 
diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index 589c69a75043..ce7f233fc490 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -142,19 +142,6 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 				     .events 		= events,
 				     .current_events 	= 0 } );
 
-	/* Critical section - locked by a spinlock because this stuff can
-	 * be changed from interrupt handlers.  The stuff above is done
-	 * outside the lock because it allocates memory.
-	 */
-
-	/* Actually, it only looks like it can be called from interrupt
-	 * context.  The culprit is reactivate_fd, which calls
-	 * maybe_sigio_broken, which calls write_sigio_workaround,
-	 * which calls activate_fd.  However, write_sigio_workaround should
-	 * only be called once, at boot time.  That would make it clear that
-	 * this is called only from process context, and can be locked with
-	 * a semaphore.
-	 */
 	spin_lock_irqsave(&irq_lock, flags);
 	for (irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next) {
 		if ((irq_fd->fd == fd) && (irq_fd->type == type)) {
@@ -165,7 +152,6 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 		}
 	}
 
-	/*-------------*/
 	if (type == IRQ_WRITE)
 		fd = -1;
 
@@ -198,7 +184,6 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 
 		spin_lock_irqsave(&irq_lock, flags);
 	}
-	/*-------------*/
 
 	*last_irq_ptr = new_fd;
 	last_irq_ptr = &new_fd->next;
@@ -210,14 +195,14 @@ int activate_fd(int irq, int fd, int type, void *dev_id)
 	 */
 	maybe_sigio_broken(fd, (type == IRQ_READ));
 
-	return(0);
+	return 0;
 
  out_unlock:
 	spin_unlock_irqrestore(&irq_lock, flags);
  out_kfree:
 	kfree(new_fd);
  out:
-	return(err);
+	return err;
 }
 
 static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg)
@@ -302,10 +287,7 @@ void reactivate_fd(int fd, int irqnum)
 	os_set_pollfd(i, irq->fd);
 	spin_unlock_irqrestore(&irq_lock, flags);
 
-	/* This calls activate_fd, so it has to be outside the critical
-	 * section.
-	 */
-	maybe_sigio_broken(fd, (irq->type == IRQ_READ));
+	add_sigio_fd(fd);
 }
 
 void deactivate_fd(int fd, int irqnum)
@@ -316,11 +298,15 @@ void deactivate_fd(int fd, int irqnum)
 
 	spin_lock_irqsave(&irq_lock, flags);
 	irq = find_irq_by_fd(fd, irqnum, &i);
-	if (irq == NULL)
-		goto out;
+	if(irq == NULL){
+		spin_unlock_irqrestore(&irq_lock, flags);
+		return;
+	}
+
 	os_set_pollfd(i, -1);
- out:
 	spin_unlock_irqrestore(&irq_lock, flags);
+
+	ignore_sigio_fd(fd);
 }
 
 int deactivate_all_fds(void)
diff --git a/arch/um/os-Linux/sigio.c b/arch/um/os-Linux/sigio.c
index 0ecac563c7b3..f6457765b17d 100644
--- a/arch/um/os-Linux/sigio.c
+++ b/arch/um/os-Linux/sigio.c
@@ -43,17 +43,9 @@ struct pollfds {
 /* Protected by sigio_lock().  Used by the sigio thread, but the UML thread
  * synchronizes with it.
  */
-static struct pollfds current_poll = {
-	.poll  		= NULL,
-	.size 		= 0,
-	.used 		= 0
-};
-
-static struct pollfds next_poll = {
-	.poll  		= NULL,
-	.size 		= 0,
-	.used 		= 0
-};
+static struct pollfds current_poll;
+static struct pollfds next_poll;
+static struct pollfds all_sigio_fds;
 
 static int write_sigio_thread(void *unused)
 {
@@ -78,7 +70,8 @@ static int write_sigio_thread(void *unused)
 				n = os_read_file(sigio_private[1], &c, sizeof(c));
 				if(n != sizeof(c))
 					printk("write_sigio_thread : "
-					       "read failed, err = %d\n", -n);
+					       "read on socket failed, "
+					       "err = %d\n", -n);
 				tmp = current_poll;
 				current_poll = next_poll;
 				next_poll = tmp;
@@ -93,35 +86,36 @@ static int write_sigio_thread(void *unused)
 
 			n = os_write_file(respond_fd, &c, sizeof(c));
 			if(n != sizeof(c))
-				printk("write_sigio_thread : write failed, "
-				       "err = %d\n", -n);
+				printk("write_sigio_thread : write on socket "
+				       "failed, err = %d\n", -n);
 		}
 	}
 
 	return 0;
 }
 
-static int need_poll(int n)
+static int need_poll(struct pollfds *polls, int n)
 {
-	if(n <= next_poll.size){
-		next_poll.used = n;
-		return(0);
+	if(n <= polls->size){
+		polls->used = n;
+		return 0;
 	}
-	kfree(next_poll.poll);
-	next_poll.poll = um_kmalloc_atomic(n * sizeof(struct pollfd));
-	if(next_poll.poll == NULL){
+	kfree(polls->poll);
+	polls->poll = um_kmalloc_atomic(n * sizeof(struct pollfd));
+	if(polls->poll == NULL){
 		printk("need_poll : failed to allocate new pollfds\n");
-		next_poll.size = 0;
-		next_poll.used = 0;
-		return(-1);
+		polls->size = 0;
+		polls->used = 0;
+		return -ENOMEM;
 	}
-	next_poll.size = n;
-	next_poll.used = n;
-	return(0);
+	polls->size = n;
+	polls->used = n;
+	return 0;
 }
 
 /* Must be called with sigio_lock held, because it's needed by the marked
- * critical section. */
+ * critical section.
+ */
 static void update_thread(void)
 {
 	unsigned long flags;
@@ -156,34 +150,39 @@ static void update_thread(void)
 	set_signals(flags);
 }
 
-static int add_sigio_fd(int fd, int read)
+int add_sigio_fd(int fd)
 {
-	int err = 0, i, n, events;
+	struct pollfd *p;
+	int err = 0, i, n;
 
 	sigio_lock();
+	for(i = 0; i < all_sigio_fds.used; i++){
+		if(all_sigio_fds.poll[i].fd == fd)
+			break;
+	}
+	if(i == all_sigio_fds.used)
+		goto out;
+
+	p = &all_sigio_fds.poll[i];
+
 	for(i = 0; i < current_poll.used; i++){
 		if(current_poll.poll[i].fd == fd)
 			goto out;
 	}
 
 	n = current_poll.used + 1;
-	err = need_poll(n);
+	err = need_poll(&next_poll, n);
 	if(err)
 		goto out;
 
 	for(i = 0; i < current_poll.used; i++)
 		next_poll.poll[i] = current_poll.poll[i];
 
-	if(read) events = POLLIN;
-	else events = POLLOUT;
-
-	next_poll.poll[n - 1] = ((struct pollfd) { .fd  	= fd,
-						   .events 	= events,
-						   .revents 	= 0 });
+	next_poll.poll[n - 1] = *p;
 	update_thread();
  out:
 	sigio_unlock();
-	return(err);
+	return err;
 }
 
 int ignore_sigio_fd(int fd)
@@ -205,18 +204,14 @@ int ignore_sigio_fd(int fd)
 	if(i == current_poll.used)
 		goto out;
 
-	err = need_poll(current_poll.used - 1);
+	err = need_poll(&next_poll, current_poll.used - 1);
 	if(err)
 		goto out;
 
 	for(i = 0; i < current_poll.used; i++){
 		p = &current_poll.poll[i];
-		if(p->fd != fd) next_poll.poll[n++] = current_poll.poll[i];
-	}
-	if(n == i){
-		printk("ignore_sigio_fd : fd %d not found\n", fd);
-		err = -1;
-		goto out;
+		if(p->fd != fd)
+			next_poll.poll[n++] = *p;
 	}
 
 	update_thread();
@@ -234,7 +229,7 @@ static struct pollfd *setup_initial_poll(int fd)
 		printk("setup_initial_poll : failed to allocate poll\n");
 		return NULL;
 	}
-	*p = ((struct pollfd) { .fd  	= fd,
+	*p = ((struct pollfd) { .fd		= fd,
 				.events 	= POLLIN,
 				.revents 	= 0 });
 	return p;
@@ -323,6 +318,8 @@ out_close1:
 
 void maybe_sigio_broken(int fd, int read)
 {
+	int err;
+
 	if(!isatty(fd))
 		return;
 
@@ -330,7 +327,19 @@ void maybe_sigio_broken(int fd, int read)
 		return;
 
 	write_sigio_workaround();
-	add_sigio_fd(fd, read);
+
+	sigio_lock();
+	err = need_poll(&all_sigio_fds, all_sigio_fds.used + 1);
+	if(err){
+		printk("maybe_sigio_broken - failed to add pollfd\n");
+		goto out;
+	}
+	all_sigio_fds.poll[all_sigio_fds.used++] =
+		((struct pollfd) { .fd  	= fd,
+				   .events 	= read ? POLLIN : POLLOUT,
+				   .revents 	= 0 });
+out:
+	sigio_unlock();
 }
 
 static void sigio_cleanup(void)
-- 
cgit v1.2.1


From 4b84c69b5f6c08a540e3683f1360a6cdef2806c7 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:04 -0700
Subject: [PATCH] uml: Move signal handlers to arch code

Have most signals go through an arch-provided handler which recovers the
sigcontext and then calls a generic handler.  This replaces the
ARCH_GET_SIGCONTEXT macro, which was somewhat fragile.  On x86_64, recovering
%rdx (which holds the sigcontext pointer) must be the first thing that
happens.  sig_handler duly invokes that first, but there is no guarantee that
I can see that instructions won't be reordered such that %rdx is used before
that.  Having the arch provide the handler seems much more robust.

Some signals in some parts of UML require their own handlers - these places
don't call set_handler any more.  They call sigaction or signal themselves.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/sysdep-i386/signal.h   | 27 ---------------------------
 arch/um/include/sysdep-x86_64/signal.h | 29 -----------------------------
 arch/um/os-Linux/irq.c                 |  2 +-
 arch/um/os-Linux/main.c                | 34 ++++++++++++++++++++++++++++------
 arch/um/os-Linux/process.c             | 12 +++++++++++-
 arch/um/os-Linux/signal.c              | 31 +++++++++++++------------------
 arch/um/os-Linux/skas/process.c        | 17 ++++++++++++++---
 arch/um/os-Linux/sys-i386/Makefile     |  2 +-
 arch/um/os-Linux/sys-i386/signal.c     | 15 +++++++++++++++
 arch/um/os-Linux/sys-x86_64/Makefile   |  2 +-
 arch/um/os-Linux/sys-x86_64/signal.c   | 16 ++++++++++++++++
 arch/um/os-Linux/time.c                |  4 ++--
 12 files changed, 102 insertions(+), 89 deletions(-)
 delete mode 100644 arch/um/include/sysdep-i386/signal.h
 delete mode 100644 arch/um/include/sysdep-x86_64/signal.h
 create mode 100644 arch/um/os-Linux/sys-i386/signal.c
 create mode 100644 arch/um/os-Linux/sys-x86_64/signal.c

diff --git a/arch/um/include/sysdep-i386/signal.h b/arch/um/include/sysdep-i386/signal.h
deleted file mode 100644
index 07518b162136..000000000000
--- a/arch/um/include/sysdep-i386/signal.h
+++ /dev/null
@@ -1,27 +0,0 @@
-/*
- * Copyright (C) 2004 PathScale, Inc
- * Licensed under the GPL
- */
-
-#ifndef __I386_SIGNAL_H_
-#define __I386_SIGNAL_H_
-
-#include <signal.h>
-
-#define ARCH_SIGHDLR_PARAM int sig
-
-#define ARCH_GET_SIGCONTEXT(sc, sig) \
-	do sc = (struct sigcontext *) (&sig + 1); while(0)
-
-#endif
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/include/sysdep-x86_64/signal.h b/arch/um/include/sysdep-x86_64/signal.h
deleted file mode 100644
index 6142897af3d1..000000000000
--- a/arch/um/include/sysdep-x86_64/signal.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * Copyright (C) 2004 PathScale, Inc
- * Licensed under the GPL
- */
-
-#ifndef __X86_64_SIGNAL_H_
-#define __X86_64_SIGNAL_H_
-
-#define ARCH_SIGHDLR_PARAM int sig
-
-#define ARCH_GET_SIGCONTEXT(sc, sig_addr) \
-	do { \
-		struct ucontext *__uc; \
-		asm("movq %%rdx, %0" : "=r" (__uc)); \
-		sc = (struct sigcontext *) &__uc->uc_mcontext; \
-	} while(0)
-
-#endif
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/os-Linux/irq.c b/arch/um/os-Linux/irq.c
index 7555bf9c33d9..a97206df5b52 100644
--- a/arch/um/os-Linux/irq.c
+++ b/arch/um/os-Linux/irq.c
@@ -132,7 +132,7 @@ void os_set_pollfd(int i, int fd)
 
 void os_set_ioignore(void)
 {
-	set_handler(SIGIO, SIG_IGN, 0, -1);
+	signal(SIGIO, SIG_IGN);
 }
 
 void init_irq_signals(int on_sigstack)
diff --git a/arch/um/os-Linux/main.c b/arch/um/os-Linux/main.c
index 90912aaca7aa..d1c5670787dc 100644
--- a/arch/um/os-Linux/main.c
+++ b/arch/um/os-Linux/main.c
@@ -67,13 +67,32 @@ static __init void do_uml_initcalls(void)
 
 static void last_ditch_exit(int sig)
 {
-	signal(SIGINT, SIG_DFL);
-	signal(SIGTERM, SIG_DFL);
-	signal(SIGHUP, SIG_DFL);
 	uml_cleanup();
 	exit(1);
 }
 
+static void install_fatal_handler(int sig)
+{
+	struct sigaction action;
+
+	/* All signals are enabled in this handler ... */
+	sigemptyset(&action.sa_mask);
+
+	/* ... including the signal being handled, plus we want the
+	 * handler reset to the default behavior, so that if an exit
+	 * handler is hanging for some reason, the UML will just die
+	 * after this signal is sent a second time.
+	 */
+	action.sa_flags = SA_RESETHAND | SA_NODEFER;
+	action.sa_restorer = NULL;
+	action.sa_handler = last_ditch_exit;
+	if(sigaction(sig, &action, NULL) < 0){
+		printf("failed to install handler for signal %d - errno = %d\n",
+		       errno);
+		exit(1);
+	}
+}
+
 #define UML_LIB_PATH	":/usr/lib/uml"
 
 static void setup_env_path(void)
@@ -158,9 +177,12 @@ int main(int argc, char **argv, char **envp)
 	}
 	new_argv[argc] = NULL;
 
-	set_handler(SIGINT, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1);
-	set_handler(SIGTERM, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1);
-	set_handler(SIGHUP, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1);
+	/* Allow these signals to bring down a UML if all other
+	 * methods of control fail.
+	 */
+	install_fatal_handler(SIGINT);
+	install_fatal_handler(SIGTERM);
+	install_fatal_handler(SIGHUP);
 
 	scan_elf_aux( envp);
 
diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
index 3afde92ad2c0..ff203625a4bd 100644
--- a/arch/um/os-Linux/process.c
+++ b/arch/um/os-Linux/process.c
@@ -246,7 +246,17 @@ void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int))
 		set_sigstack(sig_stack, pages * page_size());
 		flags = SA_ONSTACK;
 	}
-	if(usr1_handler) set_handler(SIGUSR1, usr1_handler, flags, -1);
+	if(usr1_handler){
+		struct sigaction sa;
+
+		sa.sa_handler = usr1_handler;
+		sigemptyset(&sa.sa_mask);
+		sa.sa_flags = flags;
+		sa.sa_restorer = NULL;
+		if(sigaction(SIGUSR1, &sa, NULL) < 0)
+			panic("init_new_thread_stack - sigaction failed - "
+			      "errno = %d\n", errno);
+	}
 }
 
 void init_new_thread_signals(void)
diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c
index 60e4faedf254..55b62e2b8f41 100644
--- a/arch/um/os-Linux/signal.c
+++ b/arch/um/os-Linux/signal.c
@@ -15,7 +15,6 @@
 #include "user.h"
 #include "signal_kern.h"
 #include "sysdep/sigcontext.h"
-#include "sysdep/signal.h"
 #include "sigcontext.h"
 #include "mode.h"
 #include "os.h"
@@ -38,18 +37,10 @@
 static int signals_enabled = 1;
 static int pending = 0;
 
-void sig_handler(ARCH_SIGHDLR_PARAM)
+void sig_handler(int sig, struct sigcontext *sc)
 {
-	struct sigcontext *sc;
 	int enabled;
 
-	/* Must be the first thing that this handler does - x86_64 stores
-	 * the sigcontext in %rdx, and we need to save it before it has a
-	 * chance to get trashed.
-	 */
-
-	ARCH_GET_SIGCONTEXT(sc, sig);
-
 	enabled = signals_enabled;
 	if(!enabled && (sig == SIGIO)){
 		pending |= SIGIO_MASK;
@@ -84,13 +75,10 @@ static void real_alarm_handler(int sig, struct sigcontext *sc)
 
 }
 
-void alarm_handler(ARCH_SIGHDLR_PARAM)
+void alarm_handler(int sig, struct sigcontext *sc)
 {
-	struct sigcontext *sc;
 	int enabled;
 
-	ARCH_GET_SIGCONTEXT(sc, sig);
-
 	enabled = signals_enabled;
 	if(!signals_enabled){
 		if(sig == SIGVTALRM)
@@ -126,6 +114,10 @@ void remove_sigstack(void)
 		panic("disabling signal stack failed, errno = %d\n", errno);
 }
 
+void (*handlers[_NSIG])(int sig, struct sigcontext *sc);
+
+extern void hard_handler(int sig);
+
 void set_handler(int sig, void (*handler)(int), int flags, ...)
 {
 	struct sigaction action;
@@ -133,13 +125,16 @@ void set_handler(int sig, void (*handler)(int), int flags, ...)
 	sigset_t sig_mask;
 	int mask;
 
-	va_start(ap, flags);
-	action.sa_handler = handler;
+	handlers[sig] = (void (*)(int, struct sigcontext *)) handler;
+	action.sa_handler = hard_handler;
+
 	sigemptyset(&action.sa_mask);
-	while((mask = va_arg(ap, int)) != -1){
+
+	va_start(ap, flags);
+	while((mask = va_arg(ap, int)) != -1)
 		sigaddset(&action.sa_mask, mask);
-	}
 	va_end(ap);
+
 	action.sa_flags = flags;
 	action.sa_restorer = NULL;
 	if(sigaction(sig, &action, NULL) < 0)
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 50418a5e7134..88ff0de95cd3 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -189,14 +189,25 @@ static int userspace_tramp(void *stack)
 		}
 	}
 	if(!ptrace_faultinfo && (stack != NULL)){
+		struct sigaction sa;
+
 		unsigned long v = UML_CONFIG_STUB_CODE +
 				  (unsigned long) stub_segv_handler -
 				  (unsigned long) &__syscall_stub_start;
 
 		set_sigstack((void *) UML_CONFIG_STUB_DATA, page_size());
-		set_handler(SIGSEGV, (void *) v, SA_ONSTACK,
-			    SIGIO, SIGWINCH, SIGALRM, SIGVTALRM,
-			    SIGUSR1, -1);
+		sigemptyset(&sa.sa_mask);
+		sigaddset(&sa.sa_mask, SIGIO);
+		sigaddset(&sa.sa_mask, SIGWINCH);
+		sigaddset(&sa.sa_mask, SIGALRM);
+		sigaddset(&sa.sa_mask, SIGVTALRM);
+		sigaddset(&sa.sa_mask, SIGUSR1);
+		sa.sa_flags = SA_ONSTACK;
+		sa.sa_handler = (void *) v;
+		sa.sa_restorer = NULL;
+		if(sigaction(SIGSEGV, &sa, NULL) < 0)
+			panic("userspace_tramp - setting SIGSEGV handler "
+			      "failed - errno = %d\n", errno);
 	}
 
 	os_stop_process(os_getpid());
diff --git a/arch/um/os-Linux/sys-i386/Makefile b/arch/um/os-Linux/sys-i386/Makefile
index b3213613c41c..37806621b25d 100644
--- a/arch/um/os-Linux/sys-i386/Makefile
+++ b/arch/um/os-Linux/sys-i386/Makefile
@@ -3,7 +3,7 @@
 # Licensed under the GPL
 #
 
-obj-$(CONFIG_MODE_SKAS) = registers.o tls.o
+obj-$(CONFIG_MODE_SKAS) = registers.o signal.o tls.o
 
 USER_OBJS := $(obj-y)
 
diff --git a/arch/um/os-Linux/sys-i386/signal.c b/arch/um/os-Linux/sys-i386/signal.c
new file mode 100644
index 000000000000..0d3eae518352
--- /dev/null
+++ b/arch/um/os-Linux/sys-i386/signal.c
@@ -0,0 +1,15 @@
+/*
+ * Copyright (C) 2006 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include <signal.h>
+
+extern void (*handlers[])(int sig, struct sigcontext *sc);
+
+void hard_handler(int sig)
+{
+	struct sigcontext *sc = (struct sigcontext *) (&sig + 1);
+
+	(*handlers[sig])(sig, sc);
+}
diff --git a/arch/um/os-Linux/sys-x86_64/Makefile b/arch/um/os-Linux/sys-x86_64/Makefile
index 340ef26f5944..f67842a7735b 100644
--- a/arch/um/os-Linux/sys-x86_64/Makefile
+++ b/arch/um/os-Linux/sys-x86_64/Makefile
@@ -3,7 +3,7 @@
 # Licensed under the GPL
 #
 
-obj-$(CONFIG_MODE_SKAS) = registers.o
+obj-$(CONFIG_MODE_SKAS) = registers.o signal.o
 
 USER_OBJS := $(obj-y)
 
diff --git a/arch/um/os-Linux/sys-x86_64/signal.c b/arch/um/os-Linux/sys-x86_64/signal.c
new file mode 100644
index 000000000000..3f369e5f976b
--- /dev/null
+++ b/arch/um/os-Linux/sys-x86_64/signal.c
@@ -0,0 +1,16 @@
+/*
+ * Copyright (C) 2006 Jeff Dike (jdike@addtoit.com)
+ * Licensed under the GPL
+ */
+
+#include <signal.h>
+
+extern void (*handlers[])(int sig, struct sigcontext *sc);
+
+void hard_handler(int sig)
+{
+	struct ucontext *uc;
+	asm("movq %%rdx, %0" : "=r" (uc));
+
+	(*handlers[sig])(sig, (struct sigcontext *) &uc->uc_mcontext);
+}
diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c
index 4ae73c0e5485..7f5ebbadca63 100644
--- a/arch/um/os-Linux/time.c
+++ b/arch/um/os-Linux/time.c
@@ -40,8 +40,8 @@ void disable_timer(void)
 		printk("disnable_timer - setitimer failed, errno = %d\n",
 		       errno);
 	/* If there are signals already queued, after unblocking ignore them */
-	set_handler(SIGALRM, SIG_IGN, 0, -1);
-	set_handler(SIGVTALRM, SIG_IGN, 0, -1);
+	signal(SIGALRM, SIG_IGN);
+	signal(SIGVTALRM, SIG_IGN);
 }
 
 void switch_timers(int to_real)
-- 
cgit v1.2.1


From 537ae946e808d0f22d660f7a3500832fe0c07d14 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:05 -0700
Subject: [PATCH] uml: timer cleanups

set_interval returns an error instead of panicing if setitimer fails.  Some of
its callers now check the return.

enable_timer is largely tt-mode-specific, so it is marked as such, and the
only skas-mode caller is made to call set-interval instead.

user_time_init was a no-value-added wrapper around set_interval, so it is
gone.

Since set_interval is now called from kernel code, callers no longer pass
ITIMER_* to it.  Instead, they pass a flag which is converted into ITIMER_REAL
or ITIMER_VIRTUAL.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h            |  4 +++-
 arch/um/kernel/time.c           |  7 +++++--
 arch/um/os-Linux/skas/process.c |  6 +++++-
 arch/um/os-Linux/time.c         | 18 +++++++++---------
 4 files changed, 22 insertions(+), 13 deletions(-)

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index c73dfa78f943..24fb6d8680e1 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -276,9 +276,11 @@ extern int setjmp_wrapper(void (*proc)(void *, void *), ...);
 
 extern void switch_timers(int to_real);
 extern void idle_sleep(int secs);
+extern int set_interval(int is_virtual);
+#ifdef CONFIG_MODE_TT
 extern void enable_timer(void);
+#endif
 extern void disable_timer(void);
-extern void user_time_init(void);
 extern void uml_idle_timer(void);
 extern unsigned long long os_nsecs(void);
 
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 552ca1cb9847..d1d799346e04 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -113,12 +113,15 @@ static void register_timer(void)
 
 	err = request_irq(TIMER_IRQ, um_timer, IRQF_DISABLED, "timer", NULL);
 	if(err != 0)
-		printk(KERN_ERR "timer_init : request_irq failed - "
+		printk(KERN_ERR "register_timer : request_irq failed - "
 		       "errno = %d\n", -err);
 
 	timer_irq_inited = 1;
 
-	user_time_init();
+	err = set_interval(1);
+	if(err != 0)
+		printk(KERN_ERR "register_timer : set_interval failed - "
+		       "errno = %d\n", -err);
 }
 
 extern void (*late_time_init)(void);
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 88ff0de95cd3..42e3d1ed802c 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -155,11 +155,15 @@ extern int __syscall_stub_start;
 static int userspace_tramp(void *stack)
 {
 	void *addr;
+	int err;
 
 	ptrace(PTRACE_TRACEME, 0, 0, 0);
 
 	init_new_thread_signals();
-	enable_timer();
+	err = set_interval(1);
+	if(err)
+		panic("userspace_tramp - setting timer failed, errno = %d\n",
+		      err);
 
 	if(!proc_mm){
 		/* This has a pte, but it can't be mapped in with the usual
diff --git a/arch/um/os-Linux/time.c b/arch/um/os-Linux/time.c
index 7f5ebbadca63..38be096e750f 100644
--- a/arch/um/os-Linux/time.c
+++ b/arch/um/os-Linux/time.c
@@ -17,20 +17,25 @@
 #include "kern_constants.h"
 #include "os.h"
 
-static void set_interval(int timer_type)
+int set_interval(int is_virtual)
 {
 	int usec = 1000000/hz();
+	int timer_type = is_virtual ? ITIMER_VIRTUAL : ITIMER_REAL;
 	struct itimerval interval = ((struct itimerval) { { 0, usec },
 							  { 0, usec } });
 
 	if(setitimer(timer_type, &interval, NULL) == -1)
-		panic("setitimer failed - errno = %d\n", errno);
+		return -errno;
+
+	return 0;
 }
 
+#ifdef CONFIG_MODE_TT
 void enable_timer(void)
 {
-	set_interval(ITIMER_VIRTUAL);
+	set_interval(1);
 }
+#endif
 
 void disable_timer(void)
 {
@@ -74,7 +79,7 @@ void uml_idle_timer(void)
 
 	set_handler(SIGALRM, (__sighandler_t) alarm_handler,
 		    SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, SIGVTALRM, -1);
-	set_interval(ITIMER_REAL);
+	set_interval(0);
 }
 #endif
 
@@ -94,8 +99,3 @@ void idle_sleep(int secs)
 	ts.tv_nsec = 0;
 	nanosleep(&ts, NULL);
 }
-
-void user_time_init(void)
-{
-	set_interval(ITIMER_VIRTUAL);
-}
-- 
cgit v1.2.1


From 602cc2418177a5b80f533f569e5a42c4495988c9 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:06 -0700
Subject: [PATCH] uml: Remove unused variable

timer_irq_inited was useless, so it is removed.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/kern_util.h | 1 -
 arch/um/kernel/time.c       | 5 -----
 arch/um/os-Linux/signal.c   | 7 -------
 3 files changed, 13 deletions(-)

diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h
index b98bdd8e052a..89e1dc835a5b 100644
--- a/arch/um/include/kern_util.h
+++ b/arch/um/include/kern_util.h
@@ -27,7 +27,6 @@ extern int ncpus;
 extern char *linux_prog;
 extern char *gdb_init;
 extern int kmalloc_ok;
-extern int timer_irq_inited;
 extern int jail;
 extern int nsyscalls;
 
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index d1d799346e04..2454bbd9555d 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -35,9 +35,6 @@ unsigned long long sched_clock(void)
 	return (unsigned long long)jiffies_64 * (1000000000 / HZ);
 }
 
-/* Changed at early boot */
-int timer_irq_inited = 0;
-
 static unsigned long long prev_nsecs;
 #ifdef CONFIG_UML_REAL_TIME_CLOCK
 static long long delta;   		/* Deviation per interval */
@@ -116,8 +113,6 @@ static void register_timer(void)
 		printk(KERN_ERR "register_timer : request_irq failed - "
 		       "errno = %d\n", -err);
 
-	timer_irq_inited = 1;
-
 	err = set_interval(1);
 	if(err != 0)
 		printk(KERN_ERR "register_timer : set_interval failed - "
diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c
index 55b62e2b8f41..6b81739279d1 100644
--- a/arch/um/os-Linux/signal.c
+++ b/arch/um/os-Linux/signal.c
@@ -55,15 +55,8 @@ void sig_handler(int sig, struct sigcontext *sc)
 	set_signals(enabled);
 }
 
-extern int timer_irq_inited;
-
 static void real_alarm_handler(int sig, struct sigcontext *sc)
 {
-	if(!timer_irq_inited){
-		signals_enabled = 1;
-		return;
-	}
-
 	if(sig == SIGALRM)
 		switch_timers(0);
 
-- 
cgit v1.2.1


From bf61f50d63b4d9e30d7a86a2d44bb300ae7c1dd4 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Mon, 25 Sep 2006 23:33:07 -0700
Subject: [PATCH] uml: clean our set_ether_mac

Clean set_ether_mac usage.  Maybe could also be removed, but surely it can't
be a global function taking a void* argument.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/net_kern.c | 14 ++++++--------
 arch/um/include/net_user.h |  1 -
 2 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 501f95675d89..4a7966b21931 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -31,6 +31,11 @@
 #include "irq_user.h"
 #include "irq_kern.h"
 
+static inline void set_ether_mac(struct net_device *dev, unsigned char *addr)
+{
+	memcpy(dev->dev_addr, addr, ETH_ALEN);
+}
+
 #define DRIVER_NAME "uml-netdev"
 
 static DEFINE_SPINLOCK(opened_lock);
@@ -242,7 +247,7 @@ static int uml_net_set_mac(struct net_device *dev, void *addr)
 	struct sockaddr *hwaddr = addr;
 
 	spin_lock(&lp->lock);
-	memcpy(dev->dev_addr, hwaddr->sa_data, ETH_ALEN);
+	set_ether_mac(dev, hwaddr->sa_data);
 	spin_unlock(&lp->lock);
 
 	return(0);
@@ -790,13 +795,6 @@ void dev_ip_addr(void *d, unsigned char *bin_buf)
 	memcpy(bin_buf, &in->ifa_address, sizeof(in->ifa_address));
 }
 
-void set_ether_mac(void *d, unsigned char *addr)
-{
-	struct net_device *dev = d;
-
-	memcpy(dev->dev_addr, addr, ETH_ALEN);	
-}
-
 struct sk_buff *ether_adjust_skb(struct sk_buff *skb, int extra)
 {
 	if((skb != NULL) && (skb_tailroom(skb) < extra)){
diff --git a/arch/um/include/net_user.h b/arch/um/include/net_user.h
index 800c403920bc..47ef7cb49a8e 100644
--- a/arch/um/include/net_user.h
+++ b/arch/um/include/net_user.h
@@ -26,7 +26,6 @@ struct net_user_info {
 
 extern void ether_user_init(void *data, void *dev);
 extern void dev_ip_addr(void *d, unsigned char *bin_buf);
-extern void set_ether_mac(void *d, unsigned char *addr);
 extern void iter_addresses(void *d, void (*cb)(unsigned char *, 
 					       unsigned char *, void *), 
 			   void *arg);
-- 
cgit v1.2.1


From 75e29b18d9a46bf3193278e92dc95609a8cca2ab Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:08 -0700
Subject: [PATCH] uml: stack usage reduction

The KSTK_* macros used an inordinate amount of stack.  In order to overcome
an impedance mismatch between their interface, which just returns a single
register value, and the interface of get_thread_regs, which took a full
pt_regs, the implementation created an on-stack pt_regs, filled it in, and
returned one field.  do_task_stat calls KSTK_* twice, resulting in two
local pt_regs, blowing out the stack.

This patch changes the interface (and name) of get_thread_regs to just
return a single register from a jmp_buf.

The include of archsetjmp.h" in registers.h to get the definition of
jmp_buf exposed a bogus include of <setjmp.h> in start_up.c.  <setjmp.h>
shouldn't be used anywhere any more since UML uses the klibc
setjmp/longjmp.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/registers.h             |  3 ++-
 arch/um/os-Linux/start_up.c             |  1 -
 arch/um/os-Linux/sys-i386/registers.c   | 15 +++++++++------
 arch/um/os-Linux/sys-x86_64/registers.c | 15 +++++++++------
 include/asm-um/processor-generic.h      |  4 +---
 5 files changed, 21 insertions(+), 17 deletions(-)

diff --git a/arch/um/include/registers.h b/arch/um/include/registers.h
index 83b688ca198f..f845b3629a6d 100644
--- a/arch/um/include/registers.h
+++ b/arch/um/include/registers.h
@@ -7,6 +7,7 @@
 #define __REGISTERS_H
 
 #include "sysdep/ptrace.h"
+#include "sysdep/archsetjmp.h"
 
 extern void init_thread_registers(union uml_pt_regs *to);
 extern int save_fp_registers(int pid, unsigned long *fp_regs);
@@ -15,6 +16,6 @@ extern void save_registers(int pid, union uml_pt_regs *regs);
 extern void restore_registers(int pid, union uml_pt_regs *regs);
 extern void init_registers(int pid);
 extern void get_safe_registers(unsigned long * regs, unsigned long * fp_regs);
-extern void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer);
+extern unsigned long get_thread_reg(int reg, jmp_buf *buf);
 
 #endif
diff --git a/arch/um/os-Linux/start_up.c b/arch/um/os-Linux/start_up.c
index 503148504009..7fe92680c7dd 100644
--- a/arch/um/os-Linux/start_up.c
+++ b/arch/um/os-Linux/start_up.c
@@ -14,7 +14,6 @@
 #include <sched.h>
 #include <fcntl.h>
 #include <errno.h>
-#include <setjmp.h>
 #include <sys/time.h>
 #include <sys/wait.h>
 #include <sys/mman.h>
diff --git a/arch/um/os-Linux/sys-i386/registers.c b/arch/um/os-Linux/sys-i386/registers.c
index 1f90a2d71386..7cd0369e02b3 100644
--- a/arch/um/os-Linux/sys-i386/registers.c
+++ b/arch/um/os-Linux/sys-i386/registers.c
@@ -130,11 +130,14 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 		       HOST_FP_SIZE * sizeof(unsigned long));
 }
 
-void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
+unsigned long get_thread_reg(int reg, jmp_buf *buf)
 {
-	struct __jmp_buf *jmpbuf = buffer;
-
-	UPT_SET(uml_regs, EIP, jmpbuf->__eip);
-	UPT_SET(uml_regs, UESP, jmpbuf->__esp);
-	UPT_SET(uml_regs, EBP, jmpbuf->__ebp);
+	switch(reg){
+	case EIP: return buf[0]->__eip;
+	case UESP: return buf[0]->__esp;
+	case EBP: return buf[0]->__ebp;
+	default:
+		printk("get_thread_regs - unknown register %d\n", reg);
+		return 0;
+	}
 }
diff --git a/arch/um/os-Linux/sys-x86_64/registers.c b/arch/um/os-Linux/sys-x86_64/registers.c
index e730447d6c02..cb8e8a263280 100644
--- a/arch/um/os-Linux/sys-x86_64/registers.c
+++ b/arch/um/os-Linux/sys-x86_64/registers.c
@@ -78,11 +78,14 @@ void get_safe_registers(unsigned long *regs, unsigned long *fp_regs)
 		       HOST_FP_SIZE * sizeof(unsigned long));
 }
 
-void get_thread_regs(union uml_pt_regs *uml_regs, void *buffer)
+unsigned long get_thread_reg(int reg, jmp_buf *buf)
 {
-	struct __jmp_buf *jmpbuf = buffer;
-
-	UPT_SET(uml_regs, RIP, jmpbuf->__rip);
-	UPT_SET(uml_regs, RSP, jmpbuf->__rsp);
-	UPT_SET(uml_regs, RBP, jmpbuf->__rbp);
+	switch(reg){
+	case RIP: return buf[0]->__rip;
+	case RSP: return buf[0]->__rsp;
+	case RBP: return buf[0]->__rbp;
+	default:
+		printk("get_thread_regs - unknown register %d\n", reg);
+		return 0;
+	}
 }
diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h
index 824c28896382..afa4fe1ca9f1 100644
--- a/include/asm-um/processor-generic.h
+++ b/include/asm-um/processor-generic.h
@@ -138,9 +138,7 @@ extern struct cpuinfo_um cpu_data[];
 
 #ifdef CONFIG_MODE_SKAS
 #define KSTK_REG(tsk, reg) \
-	({ union uml_pt_regs regs; \
-	   get_thread_regs(&regs, tsk->thread.mode.skas.switch_buf); \
-	   UPT_REG(&regs, reg); })
+	get_thread_reg(reg, tsk->thread.mode.skas.switch_buf)
 #else
 #define KSTK_REG(tsk, reg) (0xbadbabe)
 #endif
-- 
cgit v1.2.1


From b1fc0b1f21c4082d24d1f456a846b4fa7d16a70b Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Mon, 25 Sep 2006 23:33:08 -0700
Subject: [PATCH] UML: tty locking

Ensure current->signal->tty doesn't get freed during log_exec().

Signed-off-by: Alan Cox <alan@redhat.com>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/exec.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c
index fc38a6d5906d..0561c43b4685 100644
--- a/arch/um/kernel/exec.c
+++ b/arch/um/kernel/exec.c
@@ -41,9 +41,11 @@ static long execve1(char *file, char __user * __user *argv,
         long error;
 
 #ifdef CONFIG_TTY_LOG
-	task_lock(current);
+	mutex_lock(&tty_mutex);
+	task_lock(current);	/* FIXME:  is this needed ? */
 	log_exec(argv, current->signal->tty);
 	task_unlock(current);
+	mutex_unlock(&tty_mutex);
 #endif
         error = do_execve(file, argv, env, &current->thread.regs);
         if (error == 0){
-- 
cgit v1.2.1


From 70e0eb8ef143f3729065c504177413ffe165af22 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:09 -0700
Subject: [PATCH] Split i386 and x86_64 ptrace.h

The use of SEGMENT_RPL_MASK in the i386 ptrace.h introduced by
x86-allow-a-kernel-to-not-be-in-ring-0.patch broke the UML build, as UML
includes the underlying architecture's ptrace.h, but has no easy access to the
x86 segment definitions.

Rather than kludging around this, as in the past, this patch splits the
userspace-usable parts, which are the bits that UML needs, of ptrace.h into
ptrace-abi.h, which is included back into ptrace.h.  Thus, there is no net
effect on i386.

As a side-effect, this creates a ptrace header which is close to being usable
in /usr/include.

x86_64 is also treated in this way for consistency.  There was some trailing
whitespace there, which is cleaned up.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/Kbuild         |  1 +
 include/asm-i386/ptrace-abi.h   | 39 +++++++++++++++++++++++++++
 include/asm-i386/ptrace.h       | 35 +-----------------------
 include/asm-x86_64/Kbuild       |  1 +
 include/asm-x86_64/ptrace-abi.h | 51 +++++++++++++++++++++++++++++++++++
 include/asm-x86_64/ptrace.h     | 59 +++++------------------------------------
 6 files changed, 100 insertions(+), 86 deletions(-)
 create mode 100644 include/asm-i386/ptrace-abi.h
 create mode 100644 include/asm-x86_64/ptrace-abi.h

diff --git a/include/asm-i386/Kbuild b/include/asm-i386/Kbuild
index b75a348d0c1c..147e4ac1ebf0 100644
--- a/include/asm-i386/Kbuild
+++ b/include/asm-i386/Kbuild
@@ -3,6 +3,7 @@ include include/asm-generic/Kbuild.asm
 header-y += boot.h
 header-y += debugreg.h
 header-y += ldt.h
+header-y += ptrace-abi.h
 header-y += ucontext.h
 
 unifdef-y += mtrr.h
diff --git a/include/asm-i386/ptrace-abi.h b/include/asm-i386/ptrace-abi.h
new file mode 100644
index 000000000000..a44901817a26
--- /dev/null
+++ b/include/asm-i386/ptrace-abi.h
@@ -0,0 +1,39 @@
+#ifndef I386_PTRACE_ABI_H
+#define I386_PTRACE_ABI_H
+
+#define EBX 0
+#define ECX 1
+#define EDX 2
+#define ESI 3
+#define EDI 4
+#define EBP 5
+#define EAX 6
+#define DS 7
+#define ES 8
+#define FS 9
+#define GS 10
+#define ORIG_EAX 11
+#define EIP 12
+#define CS  13
+#define EFL 14
+#define UESP 15
+#define SS   16
+#define FRAME_SIZE 17
+
+/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
+#define PTRACE_GETREGS            12
+#define PTRACE_SETREGS            13
+#define PTRACE_GETFPREGS          14
+#define PTRACE_SETFPREGS          15
+#define PTRACE_GETFPXREGS         18
+#define PTRACE_SETFPXREGS         19
+
+#define PTRACE_OLDSETOPTIONS         21
+
+#define PTRACE_GET_THREAD_AREA    25
+#define PTRACE_SET_THREAD_AREA    26
+
+#define PTRACE_SYSEMU		  31
+#define PTRACE_SYSEMU_SINGLESTEP  32
+
+#endif
diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h
index f324c53b6f9a..1910880fcd40 100644
--- a/include/asm-i386/ptrace.h
+++ b/include/asm-i386/ptrace.h
@@ -1,24 +1,7 @@
 #ifndef _I386_PTRACE_H
 #define _I386_PTRACE_H
 
-#define EBX 0
-#define ECX 1
-#define EDX 2
-#define ESI 3
-#define EDI 4
-#define EBP 5
-#define EAX 6
-#define DS 7
-#define ES 8
-#define FS 9
-#define GS 10
-#define ORIG_EAX 11
-#define EIP 12
-#define CS  13
-#define EFL 14
-#define UESP 15
-#define SS   16
-#define FRAME_SIZE 17
+#include <asm/ptrace-abi.h>
 
 /* this struct defines the way the registers are stored on the 
    stack during a system call. */
@@ -41,22 +24,6 @@ struct pt_regs {
 	int  xss;
 };
 
-/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
-#define PTRACE_GETREGS            12
-#define PTRACE_SETREGS            13
-#define PTRACE_GETFPREGS          14
-#define PTRACE_SETFPREGS          15
-#define PTRACE_GETFPXREGS         18
-#define PTRACE_SETFPXREGS         19
-
-#define PTRACE_OLDSETOPTIONS         21
-
-#define PTRACE_GET_THREAD_AREA    25
-#define PTRACE_SET_THREAD_AREA    26
-
-#define PTRACE_SYSEMU		  31
-#define PTRACE_SYSEMU_SINGLESTEP  32
-
 #ifdef __KERNEL__
 
 #include <asm/vm86.h>
diff --git a/include/asm-x86_64/Kbuild b/include/asm-x86_64/Kbuild
index 40f2f13fe174..1ee9b07f3fe6 100644
--- a/include/asm-x86_64/Kbuild
+++ b/include/asm-x86_64/Kbuild
@@ -11,6 +11,7 @@ header-y += debugreg.h
 header-y += ldt.h
 header-y += msr.h
 header-y += prctl.h
+header-y += ptrace-abi.h
 header-y += setup.h
 header-y += sigcontext32.h
 header-y += ucontext.h
diff --git a/include/asm-x86_64/ptrace-abi.h b/include/asm-x86_64/ptrace-abi.h
new file mode 100644
index 000000000000..19184b0806b1
--- /dev/null
+++ b/include/asm-x86_64/ptrace-abi.h
@@ -0,0 +1,51 @@
+#ifndef _X86_64_PTRACE_ABI_H
+#define _X86_64_PTRACE_ABI_H
+
+#if defined(__ASSEMBLY__) || defined(__FRAME_OFFSETS)
+#define R15 0
+#define R14 8
+#define R13 16
+#define R12 24
+#define RBP 32
+#define RBX 40
+/* arguments: interrupts/non tracing syscalls only save upto here*/
+#define R11 48
+#define R10 56
+#define R9 64
+#define R8 72
+#define RAX 80
+#define RCX 88
+#define RDX 96
+#define RSI 104
+#define RDI 112
+#define ORIG_RAX 120       /* = ERROR */
+/* end of arguments */
+/* cpu exception frame or undefined in case of fast syscall. */
+#define RIP 128
+#define CS 136
+#define EFLAGS 144
+#define RSP 152
+#define SS 160
+#define ARGOFFSET R11
+#endif /* __ASSEMBLY__ */
+
+/* top of stack page */
+#define FRAME_SIZE 168
+
+#define PTRACE_OLDSETOPTIONS         21
+
+/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
+#define PTRACE_GETREGS            12
+#define PTRACE_SETREGS            13
+#define PTRACE_GETFPREGS          14
+#define PTRACE_SETFPREGS          15
+#define PTRACE_GETFPXREGS         18
+#define PTRACE_SETFPXREGS         19
+
+/* only useful for access 32bit programs */
+#define PTRACE_GET_THREAD_AREA    25
+#define PTRACE_SET_THREAD_AREA    26
+
+#define PTRACE_ARCH_PRCTL	  30	/* arch_prctl for child */
+
+#endif
diff --git a/include/asm-x86_64/ptrace.h b/include/asm-x86_64/ptrace.h
index ca6f15ff61d4..ab827dc381d7 100644
--- a/include/asm-x86_64/ptrace.h
+++ b/include/asm-x86_64/ptrace.h
@@ -1,40 +1,9 @@
 #ifndef _X86_64_PTRACE_H
 #define _X86_64_PTRACE_H
 
-#if defined(__ASSEMBLY__) || defined(__FRAME_OFFSETS) 
-#define R15 0
-#define R14 8
-#define R13 16
-#define R12 24
-#define RBP 32
-#define RBX 40
-/* arguments: interrupts/non tracing syscalls only save upto here*/
-#define R11 48
-#define R10 56	
-#define R9 64
-#define R8 72
-#define RAX 80
-#define RCX 88
-#define RDX 96
-#define RSI 104
-#define RDI 112
-#define ORIG_RAX 120       /* = ERROR */ 
-/* end of arguments */ 	
-/* cpu exception frame or undefined in case of fast syscall. */
-#define RIP 128
-#define CS 136
-#define EFLAGS 144
-#define RSP 152
-#define SS 160
-#define ARGOFFSET R11
-#endif /* __ASSEMBLY__ */
+#include <asm/ptrace-abi.h>
 
-/* top of stack page */ 
-#define FRAME_SIZE 168
-
-#define PTRACE_OLDSETOPTIONS         21
-
-#ifndef __ASSEMBLY__ 
+#ifndef __ASSEMBLY__
 
 struct pt_regs {
 	unsigned long r15;
@@ -45,7 +14,7 @@ struct pt_regs {
 	unsigned long rbx;
 /* arguments: non interrupts/non tracing syscalls only save upto here*/
  	unsigned long r11;
-	unsigned long r10;	
+	unsigned long r10;
 	unsigned long r9;
 	unsigned long r8;
 	unsigned long rax;
@@ -54,32 +23,18 @@ struct pt_regs {
 	unsigned long rsi;
 	unsigned long rdi;
 	unsigned long orig_rax;
-/* end of arguments */ 	
+/* end of arguments */
 /* cpu exception frame or undefined */
 	unsigned long rip;
 	unsigned long cs;
-	unsigned long eflags; 
-	unsigned long rsp; 
+	unsigned long eflags;
+	unsigned long rsp;
 	unsigned long ss;
-/* top of stack page */ 
+/* top of stack page */
 };
 
 #endif
 
-/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
-#define PTRACE_GETREGS            12
-#define PTRACE_SETREGS            13
-#define PTRACE_GETFPREGS          14
-#define PTRACE_SETFPREGS          15
-#define PTRACE_GETFPXREGS         18
-#define PTRACE_SETFPXREGS         19
-
-/* only useful for access 32bit programs */
-#define PTRACE_GET_THREAD_AREA    25
-#define PTRACE_SET_THREAD_AREA    26
-
-#define PTRACE_ARCH_PRCTL	  30	/* arch_prctl for child */
-
 #if defined(__KERNEL__) && !defined(__ASSEMBLY__) 
 #define user_mode(regs) (!!((regs)->cs & 3))
 #define user_mode_vm(regs) user_mode(regs)
-- 
cgit v1.2.1


From e8df8c3304cfc7d68e9a6a7b6df6c64783cd6991 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Mon, 25 Sep 2006 23:33:10 -0700
Subject: [PATCH] Make UML use ptrace-abi.h

Include the host architecture's ptrace-abi.h instead of ptrace.h.

There was some cpp mangling of names around the ptrace.h include to avoid
symbol clashes between UML and the host architecture.  Most of these can go
away.  The exception is struct pt_regs, which is convenient to have in
userspace, but must be renamed in order that UML can define its own.

ptrace-x86_64.h needed to have some now-obsolete cpp cruft and a declaration
removed.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-um/ptrace-generic.h | 14 +-------------
 include/asm-um/ptrace-x86_64.h  |  4 ----
 2 files changed, 1 insertion(+), 17 deletions(-)

diff --git a/include/asm-um/ptrace-generic.h b/include/asm-um/ptrace-generic.h
index a36f5371b36b..99c87c5ce994 100644
--- a/include/asm-um/ptrace-generic.h
+++ b/include/asm-um/ptrace-generic.h
@@ -8,19 +8,7 @@
 
 #ifndef __ASSEMBLY__
 
-
-#define pt_regs pt_regs_subarch
-#define show_regs show_regs_subarch
-#define send_sigtrap send_sigtrap_subarch
-
-#include "asm/arch/ptrace.h"
-
-#undef pt_regs
-#undef show_regs
-#undef send_sigtrap
-#undef user_mode
-#undef instruction_pointer
-
+#include "asm/arch/ptrace-abi.h"
 #include "sysdep/ptrace.h"
 
 struct pt_regs {
diff --git a/include/asm-um/ptrace-x86_64.h b/include/asm-um/ptrace-x86_64.h
index c894e68b1f96..2074483e6ca4 100644
--- a/include/asm-um/ptrace-x86_64.h
+++ b/include/asm-um/ptrace-x86_64.h
@@ -11,15 +11,11 @@
 #include "asm/errno.h"
 #include "asm/host_ldt.h"
 
-#define signal_fault signal_fault_x86_64
 #define __FRAME_OFFSETS /* Needed to get the R* macros */
 #include "asm/ptrace-generic.h"
-#undef signal_fault
 
 #define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64
 
-void signal_fault(struct pt_regs_subarch *regs, void *frame, char *where);
-
 #define FS_BASE (21 * sizeof(unsigned long))
 #define GS_BASE (22 * sizeof(unsigned long))
 #define DS (23 * sizeof(unsigned long))
-- 
cgit v1.2.1


From e8216dee838c09776680a6f1a2e54d81f3cdfa14 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 25 Sep 2006 23:33:11 -0700
Subject: [PATCH] s390: fix cmm kernel thread handling

Convert cmm's usage of kernel_thread to kthread_run.  Also create the
cmmthread at module load time, so it is possible to check if creation of
the thread fails.

In addition the cmmthread now gets terminated when the module gets unloaded
instead of leaving a stale kernel thread.  Also check the return values of
other registration functions at module load and handle their return values
appropriately.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/mm/cmm.c | 54 +++++++++++++++++++++++++++++++++++-------------------
 1 file changed, 35 insertions(+), 19 deletions(-)

diff --git a/arch/s390/mm/cmm.c b/arch/s390/mm/cmm.c
index f2e00df99bae..607f50ead1fd 100644
--- a/arch/s390/mm/cmm.c
+++ b/arch/s390/mm/cmm.c
@@ -16,6 +16,7 @@
 #include <linux/sysctl.h>
 #include <linux/ctype.h>
 #include <linux/swap.h>
+#include <linux/kthread.h>
 
 #include <asm/pgalloc.h>
 #include <asm/uaccess.h>
@@ -46,8 +47,7 @@ static struct cmm_page_array *cmm_page_list;
 static struct cmm_page_array *cmm_timed_page_list;
 static DEFINE_SPINLOCK(cmm_lock);
 
-static unsigned long cmm_thread_active;
-static struct work_struct cmm_thread_starter;
+static struct task_struct *cmm_thread_ptr;
 static wait_queue_head_t cmm_thread_wait;
 static struct timer_list cmm_timer;
 
@@ -143,14 +143,12 @@ cmm_thread(void *dummy)
 {
 	int rc;
 
-	daemonize("cmmthread");
 	while (1) {
 		rc = wait_event_interruptible(cmm_thread_wait,
 			(cmm_pages != cmm_pages_target ||
-			 cmm_timed_pages != cmm_timed_pages_target));
-		if (rc == -ERESTARTSYS) {
-			/* Got kill signal. End thread. */
-			clear_bit(0, &cmm_thread_active);
+			 cmm_timed_pages != cmm_timed_pages_target ||
+			 kthread_should_stop()));
+		if (kthread_should_stop() || rc == -ERESTARTSYS) {
 			cmm_pages_target = cmm_pages;
 			cmm_timed_pages_target = cmm_timed_pages;
 			break;
@@ -175,17 +173,9 @@ cmm_thread(void *dummy)
 	return 0;
 }
 
-static void
-cmm_start_thread(void)
-{
-	kernel_thread(cmm_thread, NULL, 0);
-}
-
 static void
 cmm_kick_thread(void)
 {
-	if (!test_and_set_bit(0, &cmm_thread_active))
-		schedule_work(&cmm_thread_starter);
 	wake_up(&cmm_thread_wait);
 }
 
@@ -429,22 +419,48 @@ struct ctl_table_header *cmm_sysctl_header;
 static int
 cmm_init (void)
 {
+	int rc = -ENOMEM;
+
 #ifdef CONFIG_CMM_PROC
 	cmm_sysctl_header = register_sysctl_table(cmm_dir_table, 1);
+	if (!cmm_sysctl_header)
+		goto out;
 #endif
 #ifdef CONFIG_CMM_IUCV
-	smsg_register_callback(SMSG_PREFIX, cmm_smsg_target);
+	rc = smsg_register_callback(SMSG_PREFIX, cmm_smsg_target);
+	if (rc < 0)
+		goto out_smsg;
 #endif
-	register_oom_notifier(&cmm_oom_nb);
-	INIT_WORK(&cmm_thread_starter, (void *) cmm_start_thread, NULL);
+	rc = register_oom_notifier(&cmm_oom_nb);
+	if (rc < 0)
+		goto out_oom_notify;
 	init_waitqueue_head(&cmm_thread_wait);
 	init_timer(&cmm_timer);
-	return 0;
+	cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread");
+	rc = IS_ERR(cmm_thread_ptr) ? PTR_ERR(cmm_thread_ptr) : 0;
+	if (!rc)
+		goto out;
+	/*
+	 * kthread_create failed. undo all the stuff from above again.
+	 */
+	unregister_oom_notifier(&cmm_oom_nb);
+
+out_oom_notify:
+#ifdef CONFIG_CMM_IUCV
+	smsg_unregister_callback(SMSG_PREFIX, cmm_smsg_target);
+out_smsg:
+#endif
+#ifdef CONFIG_CMM_PROC
+	unregister_sysctl_table(cmm_sysctl_header);
+#endif
+out:
+	return rc;
 }
 
 static void
 cmm_exit(void)
 {
+	kthread_stop(cmm_thread_ptr);
 	unregister_oom_notifier(&cmm_oom_nb);
 	cmm_free_pages(cmm_pages, &cmm_pages, &cmm_page_list);
 	cmm_free_pages(cmm_timed_pages, &cmm_timed_pages, &cmm_timed_page_list);
-- 
cgit v1.2.1


From 02f076aaa1478a91762de522ecb029efbc279690 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 26 Sep 2006 17:35:32 +0100
Subject: [PATCH] libata: refuse to register IRQless ports

We don't currently support pure polled operation so when we meet a BIOS
which forgot to assign an IRQ to a PCI device it all goes a little pear
shaped. Trap this case properly.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-core.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 753b0152afd1..b4abd6850367 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -5453,6 +5453,11 @@ int ata_device_add(const struct ata_probe_ent *ent)
 	int rc;
 
 	DPRINTK("ENTER\n");
+	
+	if (ent->irq == 0) {
+		dev_printk(KERN_ERR, dev, "is not available: No interrupt assigned.\n");
+		return 0;
+	}
 	/* alloc a container for our list of ATA ports (buses) */
 	host = kzalloc(sizeof(struct ata_host) +
 		       (ent->n_ports * sizeof(void *)), GFP_KERNEL);
-- 
cgit v1.2.1


From 4735ebedf37731160e3d3efc9fc9d4939c66fefa Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 26 Sep 2006 17:37:22 +0100
Subject: [PATCH] libata: tighten rules for legacy dependancies

The legacy and QDI drivers are ISA/VLB bus [we don't have a VLB define
but ISA will do nicely].

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
index 99837d932f36..3f4aa0c99ee4 100644
--- a/drivers/ata/Kconfig
+++ b/drivers/ata/Kconfig
@@ -311,7 +311,7 @@ config PATA_JMICRON
 
 config PATA_LEGACY
 	tristate "Legacy ISA PATA support (Experimental)"
-	depends on PCI && EXPERIMENTAL
+	depends on ISA && EXPERIMENTAL
 	help
 	  This option enables support for ISA/VLB bus legacy PATA
 	  ports and allows them to be accessed via the new ATA layer.
@@ -400,6 +400,7 @@ config PATA_PDC_OLD
 
 config PATA_QDI
 	tristate "QDI VLB PATA support"
+	depends on ISA
 	help
 	  Support for QDI 6500 and 6580 PATA controllers on VESA local bus.
 
-- 
cgit v1.2.1


From c961922b73dab429a759f560952fd4c3f60bd6b3 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 26 Sep 2006 17:53:38 +0100
Subject: [PATCH] libata-eh: Remove layering violation and duplication when
 handling absent ports

This removes the layering violation where drivers have to fiddle
directly with EH flags. Instead we now recognize -ENOENT means "no port"
and do the handling in the core code.

This also removes an instance of a call to disable the port, and an
identical printk from each driver doing this. Even better - future rule
changes will be in one place only.

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ata_piix.c      |  8 +++-----
 drivers/ata/libata-eh.c     |  6 +++++-
 drivers/ata/pata_amd.c      | 25 ++++++++-----------------
 drivers/ata/pata_artop.c    | 18 +++++++-----------
 drivers/ata/pata_atiixp.c   | 10 ++++------
 drivers/ata/pata_efar.c     | 10 ++++------
 drivers/ata/pata_jmicron.c  |  2 +-
 drivers/ata/pata_mpiix.c    | 11 ++++-------
 drivers/ata/pata_ns87410.c  |  7 ++-----
 drivers/ata/pata_oldpiix.c  |  9 +++------
 drivers/ata/pata_opti.c     | 10 ++++------
 drivers/ata/pata_optidma.c  | 10 ++++------
 drivers/ata/pata_pdc2027x.c |  8 +++-----
 drivers/ata/pata_sis.c      | 10 ++++------
 drivers/ata/pata_sl82c105.c |  9 +++------
 drivers/ata/pata_triflex.c  | 13 +++++--------
 drivers/ata/pata_via.c      |  9 +++------
 17 files changed, 67 insertions(+), 108 deletions(-)

diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c
index ffa111eea9da..5719704eb0ee 100644
--- a/drivers/ata/ata_piix.c
+++ b/drivers/ata/ata_piix.c
@@ -643,11 +643,9 @@ static int piix_pata_prereset(struct ata_port *ap)
 {
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
-	if (!pci_test_config_bits(pdev, &piix_enable_bits[ap->port_no])) {
-		ata_port_printk(ap, KERN_INFO, "port disabled. ignoring.\n");
-		ap->eh_context.i.action &= ~ATA_EH_RESET_MASK;
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &piix_enable_bits[ap->port_no]))
+		return -ENOENT;
+		
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c
index 3fa80f09f2ae..02b2b2787d9b 100644
--- a/drivers/ata/libata-eh.c
+++ b/drivers/ata/libata-eh.c
@@ -1515,7 +1515,11 @@ static int ata_eh_reset(struct ata_port *ap, int classify,
 	if (prereset) {
 		rc = prereset(ap);
 		if (rc) {
-			ata_port_printk(ap, KERN_ERR,
+			if (rc == -ENOENT) {
+				ata_port_printk(ap, KERN_DEBUG, "port disabled. ignoring.\n");
+				ap->eh_context.i.action &= ~ATA_EH_RESET_MASK;
+			} else
+				ata_port_printk(ap, KERN_ERR,
 					"prereset failed (errno=%d)\n", rc);
 			return rc;
 		}
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 3293cf9a7eb5..793a5e1cb4cb 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -25,7 +25,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_amd"
-#define DRV_VERSION "0.2.3"
+#define DRV_VERSION "0.2.4"
 
 /**
  *	timing_setup		-	shared timing computation and load
@@ -137,11 +137,8 @@ static int amd_pre_reset(struct ata_port *ap)
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 ata66;
 
-	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no]))
+		return -ENOENT;
 
 	pci_read_config_byte(pdev, 0x42, &ata66);
 	if (ata66 & bitmask[ap->port_no])
@@ -167,11 +164,9 @@ static int amd_early_pre_reset(struct ata_port *ap)
 		{ 0x40, 1, 0x01, 0x01 }
 	};
 
-	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &amd_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	/* No host side cable detection */
 	ap->cbl = ATA_CBL_PATA80;
 	return ata_std_prereset(ap);
@@ -262,12 +257,8 @@ static int nv_pre_reset(struct ata_port *ap) {
 	u8 ata66;
 	u16 udma;
 
-	if (!pci_test_config_bits(pdev, &nv_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
-
+	if (!pci_test_config_bits(pdev, &nv_enable_bits[ap->port_no]))
+		return -ENOENT;
 
 	pci_read_config_byte(pdev, 0x52, &ata66);
 	if (ata66 & bitmask[ap->port_no])
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index d6ef3bf1bac7..0f3b49176dc1 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -28,7 +28,7 @@
 #include <linux/ata.h>
 
 #define DRV_NAME	"pata_artop"
-#define DRV_VERSION	"0.4.1"
+#define DRV_VERSION	"0.4.2"
 
 /*
  *	The ARTOP has 33 Mhz and "over clocked" timing tables. Until we
@@ -47,11 +47,9 @@ static int artop6210_pre_reset(struct ata_port *ap)
 		{ 0x4AU, 1U, 0x04UL, 0x04UL },	/* port 1 */
 	};
 
-	if (!pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
@@ -90,11 +88,9 @@ static int artop6260_pre_reset(struct ata_port *ap)
 	u8 tmp;
 
 	/* Odd numbered device ids are the units with enable bits (the -R cards) */
-	if (pdev->device % 1 && !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (pdev->device % 1 && !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	pci_read_config_byte(pdev, 0x49, &tmp);
 	if (tmp & (1 >> ap->port_no))
 		ap->cbl = ATA_CBL_PATA40;
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 3f78a1e54a75..060d019fd30b 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -22,7 +22,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_atiixp"
-#define DRV_VERSION "0.4.2"
+#define DRV_VERSION "0.4.3"
 
 enum {
 	ATIIXP_IDE_PIO_TIMING	= 0x40,
@@ -41,11 +41,9 @@ static int atiixp_pre_reset(struct ata_port *ap)
 		{ 0x48, 1, 0x08, 0x00 }
 	};
 
-	if (!pci_test_config_bits(pdev, &atiixp_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &atiixp_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	ap->cbl = ATA_CBL_PATA80;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index c30bc181304f..92813827faa7 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -22,7 +22,7 @@
 #include <linux/ata.h>
 
 #define DRV_NAME	"pata_efar"
-#define DRV_VERSION	"0.4.1"
+#define DRV_VERSION	"0.4.2"
 
 /**
  *	efar_pre_reset	-	check for 40/80 pin
@@ -42,11 +42,9 @@ static int efar_pre_reset(struct ata_port *ap)
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u8 tmp;
 
-	if (!pci_test_config_bits(pdev, &efar_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &efar_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	pci_read_config_byte(pdev, 0x47, &tmp);
 	if (tmp & (2 >> ap->port_no))
 		ap->cbl = ATA_CBL_PATA40;
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 6832a643a9eb..3bdb6f2fa5ba 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -51,7 +51,7 @@ static int jmicron_pre_reset(struct ata_port *ap)
 	/* Check if our port is enabled */
 	pci_read_config_dword(pdev, 0x40, &control);
 	if ((control & port_mask) == 0)
-		return 0;
+		return -ENOENT;
 
 	/* There are two basic mappings. One has the two SATA ports merged
 	   as master/slave and the secondary as PATA, the other has only the
diff --git a/drivers/ata/pata_mpiix.c b/drivers/ata/pata_mpiix.c
index 1958c4ed09a8..3c65393c1f01 100644
--- a/drivers/ata/pata_mpiix.c
+++ b/drivers/ata/pata_mpiix.c
@@ -18,7 +18,7 @@
  * The driver conciously keeps this logic internally to avoid pushing quirky
  * PATA history into the clean libata layer.
  *
- * Thinkpad specific note: If you boot an MPIIX using thinkpad with a PCMCIA
+ * Thinkpad specific note: If you boot an MPIIX using a thinkpad with a PCMCIA
  * hard disk present this driver will not detect it. This is not a bug. In this
  * configuration the secondary port of the MPIIX is disabled and the addresses
  * are decoded by the PCMCIA bridge and therefore are for a generic IDE driver
@@ -35,7 +35,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_mpiix"
-#define DRV_VERSION "0.7.1"
+#define DRV_VERSION "0.7.2"
 
 enum {
 	IDETIM = 0x6C,		/* IDE control register */
@@ -54,11 +54,8 @@ static int mpiix_pre_reset(struct ata_port *ap)
 		{ 0x6F, 1, 0x80, 0x80 }
 	};
 
-	if (!pci_test_config_bits(pdev, &mpiix_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &mpiix_enable_bits[ap->port_no]))
+		return -ENOENT;
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index 93d6646d2954..f529fe4c29b7 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -45,11 +45,8 @@ static int ns87410_pre_reset(struct ata_port *ap)
 		{ 0x47, 1, 0x08, 0x08 }
 	};
 
-	if (!pci_test_config_bits(pdev, &ns87410_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &ns87410_enable_bits[ap->port_no]))
+		return -ENOENT;
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_oldpiix.c b/drivers/ata/pata_oldpiix.c
index 04c618a2664b..31a285ca88dc 100644
--- a/drivers/ata/pata_oldpiix.c
+++ b/drivers/ata/pata_oldpiix.c
@@ -25,7 +25,7 @@
 #include <linux/ata.h>
 
 #define DRV_NAME	"pata_oldpiix"
-#define DRV_VERSION	"0.5.1"
+#define DRV_VERSION	"0.5.2"
 
 /**
  *	oldpiix_pre_reset		-	probe begin
@@ -42,11 +42,8 @@ static int oldpiix_pre_reset(struct ata_port *ap)
 		{ 0x43U, 1U, 0x80UL, 0x80UL },	/* port 1 */
 	};
 
-	if (!pci_test_config_bits(pdev, &oldpiix_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &oldpiix_enable_bits[ap->port_no]))
+		return -ENOENT;
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index c3d01325e0e2..136d7a65feb3 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -34,7 +34,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_opti"
-#define DRV_VERSION "0.2.4"
+#define DRV_VERSION "0.2.5"
 
 enum {
 	READ_REG	= 0,	/* index of Read cycle timing register */
@@ -59,11 +59,9 @@ static int opti_pre_reset(struct ata_port *ap)
 		{ 0x40, 1, 0x08, 0x00 }
 	};
 
-	if (!pci_test_config_bits(pdev, &opti_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &opti_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index 177a455f4251..18ec8e72bbb6 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -33,7 +33,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_optidma"
-#define DRV_VERSION "0.2.1"
+#define DRV_VERSION "0.2.2"
 
 enum {
 	READ_REG	= 0,	/* index of Read cycle timing register */
@@ -59,11 +59,9 @@ static int optidma_pre_reset(struct ata_port *ap)
 		0x40, 1, 0x08, 0x00
 	};
 
-	if (ap->port_no && !pci_test_config_bits(pdev, &optidma_enable_bits)) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (ap->port_no && !pci_test_config_bits(pdev, &optidma_enable_bits))
+		return -ENOENT;
+
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_pdc2027x.c b/drivers/ata/pata_pdc2027x.c
index 31ab9c886209..bd4ed6734edc 100644
--- a/drivers/ata/pata_pdc2027x.c
+++ b/drivers/ata/pata_pdc2027x.c
@@ -36,7 +36,7 @@
 #include <asm/io.h>
 
 #define DRV_NAME	"pata_pdc2027x"
-#define DRV_VERSION	"0.74-ac3"
+#define DRV_VERSION	"0.74-ac5"
 #undef PDC_DEBUG
 
 #ifdef PDC_DEBUG
@@ -311,10 +311,8 @@ static inline int pdc2027x_port_enabled(struct ata_port *ap)
 static int pdc2027x_prereset(struct ata_port *ap)
 {
 	/* Check whether port enabled */
-	if (!pdc2027x_port_enabled(ap)) {
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pdc2027x_port_enabled(ap))
+		return -ENOENT;
 	pdc2027x_cbl_detect(ap);
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index 2e555168b431..fce74d9b2e67 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -34,7 +34,7 @@
 #include <linux/ata.h>
 
 #define DRV_NAME	"pata_sis"
-#define DRV_VERSION	"0.4.3"
+#define DRV_VERSION	"0.4.4"
 
 struct sis_chipset {
 	u16 device;			/* PCI host ID */
@@ -74,11 +74,9 @@ static int sis_133_pre_reset(struct ata_port *ap)
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 	u16 tmp;
 
-	if (!pci_test_config_bits(pdev, &sis_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &sis_enable_bits[ap->port_no]))
+		return -ENOENT;
+
 	/* The top bit of this register is the cable detect bit */
 	pci_read_config_word(pdev, 0x50 + 2 * ap->port_no, &tmp);
 	if (tmp & 0x8000)
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index f8499786917a..b3bc607f0293 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -19,7 +19,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_sl82c105"
-#define DRV_VERSION "0.2.2"
+#define DRV_VERSION "0.2.3"
 
 enum {
 	/*
@@ -49,11 +49,8 @@ static int sl82c105_pre_reset(struct ata_port *ap)
 	};
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
-	if (ap->port_no && !pci_test_config_bits(pdev, &sl82c105_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		dev_printk(KERN_INFO, &pdev->dev, "port disabled. ignoring.\n");
-		return 0;
-	}
+	if (ap->port_no && !pci_test_config_bits(pdev, &sl82c105_enable_bits[ap->port_no]))
+		return -ENOENT;
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 36f788728f3f..95adf92901fc 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -46,13 +46,13 @@
 #define DRV_VERSION "0.2.5"
 
 /**
- *	triflex_probe_init		-	probe begin
+ *	triflex_prereset		-	probe begin
  *	@ap: ATA port
  *
  *	Set up cable type and use generic probe init
  */
 
-static int triflex_probe_init(struct ata_port *ap)
+static int triflex_prereset(struct ata_port *ap)
 {
 	static const struct pci_bits triflex_enable_bits[] = {
 		{ 0x80, 1, 0x01, 0x01 },
@@ -61,11 +61,8 @@ static int triflex_probe_init(struct ata_port *ap)
 
 	struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
-	if (!pci_test_config_bits(pdev, &triflex_enable_bits[ap->port_no])) {
-		ata_port_disable(ap);
-		printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-		return 0;
-	}
+	if (!pci_test_config_bits(pdev, &triflex_enable_bits[ap->port_no]))
+		return -ENOENT;
 	ap->cbl = ATA_CBL_PATA40;
 	return ata_std_prereset(ap);
 }
@@ -74,7 +71,7 @@ static int triflex_probe_init(struct ata_port *ap)
 
 static void triflex_error_handler(struct ata_port *ap)
 {
-	ata_bmdma_drive_eh(ap, triflex_probe_init, ata_std_softreset, NULL, ata_std_postreset);
+	ata_bmdma_drive_eh(ap, triflex_prereset, ata_std_softreset, NULL, ata_std_postreset);
 }
 
 /**
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 1b2ff133b163..0e8b1303acd6 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -60,7 +60,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_via"
-#define DRV_VERSION "0.1.13"
+#define DRV_VERSION "0.1.14"
 
 /*
  *	The following comes directly from Vojtech Pavlik's ide/pci/via82cxxx
@@ -155,11 +155,8 @@ static int via_pre_reset(struct ata_port *ap)
 
 		struct pci_dev *pdev = to_pci_dev(ap->host->dev);
 
-		if (!pci_test_config_bits(pdev, &via_enable_bits[ap->port_no])) {
-			ata_port_disable(ap);
-			printk(KERN_INFO "ata%u: port disabled. ignoring.\n", ap->id);
-			return 0;
-		}
+		if (!pci_test_config_bits(pdev, &via_enable_bits[ap->port_no]))
+			return -ENOENT;
 	}
 
 	if ((config->flags & VIA_UDMA) >= VIA_UDMA_66)
-- 
cgit v1.2.1


From daac0acd597618b99bba5f8fff31f21e4f283154 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 26 Sep 2006 17:55:37 +0100
Subject: [PATCH] libata-sff: use our IRQ defines

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-sff.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 688bb55e197a..9222b6bb010a 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -881,7 +881,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 	probe_ent->private_data = port[0]->private_data;
 
 	if (port_mask & ATA_PORT_PRIMARY) {
-		probe_ent->irq = 14;
+		probe_ent->irq = ATA_PRIMARY_IRQ;
 		probe_ent->port[0].cmd_addr = ATA_PRIMARY_CMD;
 		probe_ent->port[0].altstatus_addr =
 		probe_ent->port[0].ctl_addr = ATA_PRIMARY_CTL;
@@ -896,7 +896,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 
 	if (port_mask & ATA_PORT_SECONDARY) {
 		if (probe_ent->irq)
-			probe_ent->irq2 = 15;
+			probe_ent->irq2 = ATA_SECONDARY_IRQ;
 		else
 			probe_ent->irq = 15;
 		probe_ent->port[1].cmd_addr = ATA_SECONDARY_CMD;
-- 
cgit v1.2.1


From 214ddde2f95037e129eff7e895869771719c7c1b Mon Sep 17 00:00:00 2001
From: bibo mao <bibo.mao@intel.com>
Date: Tue, 26 Sep 2006 11:20:37 -0700
Subject: [IA64] kprobe opcode 16 bytes alignment on IA64

On IA64 instruction opcode must be 16 bytes alignment, in kprobe structure
there is one element to save original instruction, currently saved opcode
is not statically allocated in kprobe structure, that can not assure
16 bytes alignment. This patch dynamically allocated kprobe instruction
opcode to assure 16 bytes alignment.

Signed-off-by: bibo mao <bibo.mao@intel.com>
Acked-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/kprobes.c | 55 ++++++++++++++++++++++++----------------------
 include/asm-ia64/kprobes.h |  9 ++++----
 2 files changed, 34 insertions(+), 30 deletions(-)

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 781960f80b6f..320cb7a91c9d 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -136,10 +136,8 @@ static void __kprobes update_kprobe_inst_flag(uint template, uint  slot,
 static int __kprobes unsupported_inst(uint template, uint  slot,
 				      uint major_opcode,
 				      unsigned long kprobe_inst,
-				      struct kprobe *p)
+				      unsigned long addr)
 {
-	unsigned long addr = (unsigned long)p->addr;
-
 	if (bundle_encoding[template][slot] == I) {
 		switch (major_opcode) {
 			case 0x0: //I_UNIT_MISC_OPCODE:
@@ -217,7 +215,7 @@ static void __kprobes prepare_break_inst(uint template, uint  slot,
 					 struct kprobe *p)
 {
 	unsigned long break_inst = BREAK_INST;
-	bundle_t *bundle = &p->ainsn.insn.bundle;
+	bundle_t *bundle = &p->opcode.bundle;
 
 	/*
 	 * Copy the original kprobe_inst qualifying predicate(qp)
@@ -423,11 +421,9 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	unsigned long *kprobe_addr = (unsigned long *)(addr & ~0xFULL);
 	unsigned long kprobe_inst=0;
 	unsigned int slot = addr & 0xf, template, major_opcode = 0;
-	bundle_t *bundle = &p->ainsn.insn.bundle;
-
-	memcpy(&p->opcode.bundle, kprobe_addr, sizeof(bundle_t));
-	memcpy(&p->ainsn.insn.bundle, kprobe_addr, sizeof(bundle_t));
+	bundle_t *bundle;
 
+	bundle = &((kprobe_opcode_t *)kprobe_addr)->bundle;
  	template = bundle->quad0.template;
 
 	if(valid_kprobe_addr(template, slot, addr))
@@ -440,20 +436,19 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p)
 	/* Get kprobe_inst and major_opcode from the bundle */
 	get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode);
 
-	if (unsupported_inst(template, slot, major_opcode, kprobe_inst, p))
+	if (unsupported_inst(template, slot, major_opcode, kprobe_inst, addr))
 			return -EINVAL;
 
-	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
-	return 0;
-}
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn)
+		return -ENOMEM;
+	memcpy(&p->opcode, kprobe_addr, sizeof(kprobe_opcode_t));
+	memcpy(p->ainsn.insn, kprobe_addr, sizeof(kprobe_opcode_t));
 
-void __kprobes flush_insn_slot(struct kprobe *p)
-{
-	unsigned long arm_addr;
+	prepare_break_inst(template, slot, major_opcode, kprobe_inst, p);
 
-	arm_addr = ((unsigned long)&p->opcode.bundle) & ~0xFULL;
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	return 0;
 }
 
 void __kprobes arch_arm_kprobe(struct kprobe *p)
@@ -461,9 +456,10 @@ void __kprobes arch_arm_kprobe(struct kprobe *p)
 	unsigned long addr = (unsigned long)p->addr;
 	unsigned long arm_addr = addr & ~0xFULL;
 
-	flush_insn_slot(p);
-	memcpy((char *)arm_addr, &p->ainsn.insn.bundle, sizeof(bundle_t));
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	flush_icache_range((unsigned long)p->ainsn.insn,
+			(unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t));
+	memcpy((char *)arm_addr, &p->opcode, sizeof(kprobe_opcode_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t));
 }
 
 void __kprobes arch_disarm_kprobe(struct kprobe *p)
@@ -471,11 +467,18 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 	unsigned long addr = (unsigned long)p->addr;
 	unsigned long arm_addr = addr & ~0xFULL;
 
-	/* p->opcode contains the original unaltered bundle */
-	memcpy((char *) arm_addr, (char *) &p->opcode.bundle, sizeof(bundle_t));
-	flush_icache_range(arm_addr, arm_addr + sizeof(bundle_t));
+	/* p->ainsn.insn contains the original unaltered kprobe_opcode_t */
+	memcpy((char *) arm_addr, (char *) p->ainsn.insn,
+					 sizeof(kprobe_opcode_t));
+	flush_icache_range(arm_addr, arm_addr + sizeof(kprobe_opcode_t));
 }
 
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	mutex_lock(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	mutex_unlock(&kprobe_mutex);
+}
 /*
  * We are resuming execution after a single step fault, so the pt_regs
  * structure reflects the register state after we executed the instruction
@@ -486,12 +489,12 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
  */
 static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs)
 {
-  	unsigned long bundle_addr = ((unsigned long) (&p->opcode.bundle)) & ~0xFULL;
+  	unsigned long bundle_addr = (unsigned long) (&p->ainsn.insn->bundle);
   	unsigned long resume_addr = (unsigned long)p->addr & ~0xFULL;
  	unsigned long template;
  	int slot = ((unsigned long)p->addr & 0xf);
 
-	template = p->opcode.bundle.quad0.template;
+	template = p->ainsn.insn->bundle.quad0.template;
 
  	if (slot == 1 && bundle_encoding[template][1] == L)
  		slot = 2;
@@ -553,7 +556,7 @@ turn_ss_off:
 
 static void __kprobes prepare_ss(struct kprobe *p, struct pt_regs *regs)
 {
-	unsigned long bundle_addr = (unsigned long) &p->opcode.bundle;
+	unsigned long bundle_addr = (unsigned long) &p->ainsn.insn->bundle;
 	unsigned long slot = (unsigned long)p->addr & 0xf;
 
 	/* single step inline if break instruction */
diff --git a/include/asm-ia64/kprobes.h b/include/asm-ia64/kprobes.h
index 938904910115..1b45b71c79b9 100644
--- a/include/asm-ia64/kprobes.h
+++ b/include/asm-ia64/kprobes.h
@@ -29,7 +29,8 @@
 #include <linux/percpu.h>
 #include <asm/break.h>
 
-#define MAX_INSN_SIZE   16
+#define __ARCH_WANT_KPROBES_INSN_SLOT
+#define MAX_INSN_SIZE   1
 #define BREAK_INST	(long)(__IA64_BREAK_KPROBE << 6)
 
 typedef union cmp_inst {
@@ -94,7 +95,7 @@ struct kprobe_ctlblk {
 #define IP_RELATIVE_PREDICT_OPCODE	(7)
 #define LONG_BRANCH_OPCODE		(0xC)
 #define LONG_CALL_OPCODE		(0xD)
-#define arch_remove_kprobe(p)		do {} while (0)
+#define flush_insn_slot(p)		do { } while (0)
 
 typedef struct kprobe_opcode {
 	bundle_t bundle;
@@ -108,7 +109,7 @@ struct fnptr {
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
 	/* copy of the instruction to be emulated */
-	kprobe_opcode_t insn;
+	kprobe_opcode_t *insn;
  #define INST_FLAG_FIX_RELATIVE_IP_ADDR		1
  #define INST_FLAG_FIX_BRANCH_REG		2
  #define INST_FLAG_BREAK_INST			4
@@ -125,6 +126,6 @@ static inline void jprobe_return(void)
 }
 extern void invalidate_stacked_regs(void);
 extern void flush_register_stack(void);
-extern void flush_insn_slot(struct kprobe *p);
+extern void arch_remove_kprobe(struct kprobe *p);
 
 #endif				/* _ASM_KPROBES_H */
-- 
cgit v1.2.1


From fd32cb3a9c9f9399421408e8734cd8a6d9d1a09f Mon Sep 17 00:00:00 2001
From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Date: Mon, 25 Sep 2006 16:32:20 -0700
Subject: [IA64] kprobes: fixup the pagefault exception caused by probehandlers

If the user-specified kprobe handler causes the page fault when accessing
user space address, fixup this fault since do_page_fault() should not
continue as the kprobe handler are run with preemption disabled.

Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/kprobes.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 320cb7a91c9d..169ec3a7156c 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -771,6 +771,12 @@ static int __kprobes kprobes_fault_handler(struct pt_regs *regs, int trapnr)
 		 */
 		if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr))
 			return 1;
+		/*
+		 * In case the user-specified fault handler returned
+		 * zero, try to fix up.
+		 */
+		if (ia64_done_with_exception(regs))
+			return 1;
 
 		/*
 		 * Let ia64_do_page_fault() fix it.
-- 
cgit v1.2.1


From dd562c05410e13e878a3ee0deb8ac06db2e132c7 Mon Sep 17 00:00:00 2001
From: Stephane Eranian <eranian@hpl.hp.com>
Date: Thu, 21 Sep 2006 10:35:44 -0700
Subject: [IA64] Add interface so modules can discover whether multithreading
 is on.

Add is_multithreading_enabled() to check whether multi-threading
is enabled independently of which cpu is currently online

Signed-off-by: stephane eranian <eranian@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smpboot.c | 24 ++++++++++++++++++++++++
 include/asm-ia64/smp.h     |  1 +
 2 files changed, 25 insertions(+)

diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 6203ed4ec8cf..f7d7f5668144 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -879,3 +879,27 @@ identify_siblings(struct cpuinfo_ia64 *c)
 	c->core_id = info.log1_cid;
 	c->thread_id = info.log1_tid;
 }
+
+/*
+ * returns non zero, if multi-threading is enabled
+ * on at least one physical package. Due to hotplug cpu
+ * and (maxcpus=), all threads may not necessarily be enabled
+ * even though the processor supports multi-threading.
+ */
+int is_multithreading_enabled(void)
+{
+	int i, j;
+
+	for_each_present_cpu(i) {
+		for_each_present_cpu(j) {
+			if (j == i)
+				continue;
+			if ((cpu_data(j)->socket_id == cpu_data(i)->socket_id)) {
+				if (cpu_data(j)->core_id == cpu_data(i)->core_id)
+					return 1;
+			}
+		}
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(is_multithreading_enabled);
diff --git a/include/asm-ia64/smp.h b/include/asm-ia64/smp.h
index 719ff309ce09..6533eb4e86ad 100644
--- a/include/asm-ia64/smp.h
+++ b/include/asm-ia64/smp.h
@@ -128,6 +128,7 @@ extern void smp_send_reschedule (int cpu);
 extern void lock_ipi_calllock(void);
 extern void unlock_ipi_calllock(void);
 extern void identify_siblings (struct cpuinfo_ia64 *);
+extern int is_multithreading_enabled(void);
 
 #else
 
-- 
cgit v1.2.1


From 35589a8fa8138244e7f2ef9317c440aa580c9335 Mon Sep 17 00:00:00 2001
From: Keshavamurthy Anil S <anil.s.keshavamurthy@intel.com>
Date: Tue, 26 Sep 2006 12:03:13 -0700
Subject: [IA64] Move perfmon tables from thread_struct to pfm_context

This patch renders thread_struct->pmcs[] and thread_struct->pmds[]
OBSOLETE. The actual table is moved to pfm_context structure which
saves space in thread_struct (in turn saving space in task_struct
which frees up more space for kernel stacks).

Signed-off-by: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/perfmon.c   | 113 +++++++++++++++++++------------------------
 include/asm-ia64/processor.h |  12 +----
 2 files changed, 50 insertions(+), 75 deletions(-)

diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 84a7e52f56f6..45000d5d0cfa 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -62,6 +62,9 @@
 
 #define PFM_INVALID_ACTIVATION	(~0UL)
 
+#define PFM_NUM_PMC_REGS	64	/* PMC save area for ctxsw */
+#define PFM_NUM_PMD_REGS	64	/* PMD save area for ctxsw */
+
 /*
  * depth of message queue
  */
@@ -296,14 +299,17 @@ typedef struct pfm_context {
 	unsigned long		ctx_reload_pmcs[4];	/* bitmask of force reload PMC on ctxsw in */
 	unsigned long		ctx_used_monitors[4];	/* bitmask of monitor PMC being used */
 
-	unsigned long		ctx_pmcs[IA64_NUM_PMC_REGS];	/*  saved copies of PMC values */
+	unsigned long		ctx_pmcs[PFM_NUM_PMC_REGS];	/*  saved copies of PMC values */
 
 	unsigned int		ctx_used_ibrs[1];		/* bitmask of used IBR (speedup ctxsw in) */
 	unsigned int		ctx_used_dbrs[1];		/* bitmask of used DBR (speedup ctxsw in) */
 	unsigned long		ctx_dbrs[IA64_NUM_DBG_REGS];	/* DBR values (cache) when not loaded */
 	unsigned long		ctx_ibrs[IA64_NUM_DBG_REGS];	/* IBR values (cache) when not loaded */
 
-	pfm_counter_t		ctx_pmds[IA64_NUM_PMD_REGS]; /* software state for PMDS */
+	pfm_counter_t		ctx_pmds[PFM_NUM_PMD_REGS]; /* software state for PMDS */
+
+	unsigned long		th_pmcs[PFM_NUM_PMC_REGS];	/* PMC thread save state */
+	unsigned long		th_pmds[PFM_NUM_PMD_REGS];	/* PMD thread save state */
 
 	u64			ctx_saved_psr_up;	/* only contains psr.up value */
 
@@ -867,7 +873,6 @@ static void
 pfm_mask_monitoring(struct task_struct *task)
 {
 	pfm_context_t *ctx = PFM_GET_CTX(task);
-	struct thread_struct *th = &task->thread;
 	unsigned long mask, val, ovfl_mask;
 	int i;
 
@@ -888,7 +893,7 @@ pfm_mask_monitoring(struct task_struct *task)
 	 * So in both cases, the live register contains the owner's
 	 * state. We can ONLY touch the PMU registers and NOT the PSR.
 	 *
-	 * As a consequence to this call, the thread->pmds[] array
+	 * As a consequence to this call, the ctx->th_pmds[] array
 	 * contains stale information which must be ignored
 	 * when context is reloaded AND monitoring is active (see
 	 * pfm_restart).
@@ -923,9 +928,9 @@ pfm_mask_monitoring(struct task_struct *task)
 	mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER;
 	for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) {
 		if ((mask & 0x1) == 0UL) continue;
-		ia64_set_pmc(i, th->pmcs[i] & ~0xfUL);
-		th->pmcs[i] &= ~0xfUL;
-		DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, th->pmcs[i]));
+		ia64_set_pmc(i, ctx->th_pmcs[i] & ~0xfUL);
+		ctx->th_pmcs[i] &= ~0xfUL;
+		DPRINT_ovfl(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i]));
 	}
 	/*
 	 * make all of this visible
@@ -942,7 +947,6 @@ static void
 pfm_restore_monitoring(struct task_struct *task)
 {
 	pfm_context_t *ctx = PFM_GET_CTX(task);
-	struct thread_struct *th = &task->thread;
 	unsigned long mask, ovfl_mask;
 	unsigned long psr, val;
 	int i, is_system;
@@ -1008,9 +1012,9 @@ pfm_restore_monitoring(struct task_struct *task)
 	mask = ctx->ctx_used_monitors[0] >> PMU_FIRST_COUNTER;
 	for(i= PMU_FIRST_COUNTER; mask; i++, mask>>=1) {
 		if ((mask & 0x1) == 0UL) continue;
-		th->pmcs[i] = ctx->ctx_pmcs[i];
-		ia64_set_pmc(i, th->pmcs[i]);
-		DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, th->pmcs[i]));
+		ctx->th_pmcs[i] = ctx->ctx_pmcs[i];
+		ia64_set_pmc(i, ctx->th_pmcs[i]);
+		DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, ctx->th_pmcs[i]));
 	}
 	ia64_srlz_d();
 
@@ -1069,7 +1073,6 @@ pfm_restore_pmds(unsigned long *pmds, unsigned long mask)
 static inline void
 pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 {
-	struct thread_struct *thread = &task->thread;
 	unsigned long ovfl_val = pmu_conf->ovfl_val;
 	unsigned long mask = ctx->ctx_all_pmds[0];
 	unsigned long val;
@@ -1091,11 +1094,11 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 			ctx->ctx_pmds[i].val = val & ~ovfl_val;
 			 val &= ovfl_val;
 		}
-		thread->pmds[i] = val;
+		ctx->th_pmds[i] = val;
 
 		DPRINT(("pmd[%d]=0x%lx soft_val=0x%lx\n",
 			i,
-			thread->pmds[i],
+			ctx->th_pmds[i],
 			ctx->ctx_pmds[i].val));
 	}
 }
@@ -1106,7 +1109,6 @@ pfm_copy_pmds(struct task_struct *task, pfm_context_t *ctx)
 static inline void
 pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx)
 {
-	struct thread_struct *thread = &task->thread;
 	unsigned long mask = ctx->ctx_all_pmcs[0];
 	int i;
 
@@ -1114,8 +1116,8 @@ pfm_copy_pmcs(struct task_struct *task, pfm_context_t *ctx)
 
 	for (i=0; mask; i++, mask>>=1) {
 		/* masking 0 with ovfl_val yields 0 */
-		thread->pmcs[i] = ctx->ctx_pmcs[i];
-		DPRINT(("pmc[%d]=0x%lx\n", i, thread->pmcs[i]));
+		ctx->th_pmcs[i] = ctx->ctx_pmcs[i];
+		DPRINT(("pmc[%d]=0x%lx\n", i, ctx->th_pmcs[i]));
 	}
 }
 
@@ -2859,7 +2861,6 @@ pfm_reset_regs(pfm_context_t *ctx, unsigned long *ovfl_regs, int is_long_reset)
 static int
 pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
 	unsigned long value, pmc_pm;
@@ -2880,7 +2881,6 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	if (state == PFM_CTX_ZOMBIE) return -EINVAL;
 
 	if (is_loaded) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3035,7 +3035,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 		 *
 		 * The value in ctx_pmcs[] can only be changed in pfm_write_pmcs().
 		 *
-		 * The value in thread->pmcs[] may be modified on overflow, i.e.,  when
+		 * The value in th_pmcs[] may be modified on overflow, i.e.,  when
 		 * monitoring needs to be stopped.
 		 */
 		if (is_monitor) CTX_USED_MONITOR(ctx, 1UL << cnum);
@@ -3049,7 +3049,7 @@ pfm_write_pmcs(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			/*
 			 * write thread state
 			 */
-			if (is_system == 0) thread->pmcs[cnum] = value;
+			if (is_system == 0) ctx->th_pmcs[cnum] = value;
 
 			/*
 			 * write hardware register if we can
@@ -3101,7 +3101,6 @@ error:
 static int
 pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
 	unsigned long value, hw_value, ovfl_mask;
@@ -3125,7 +3124,6 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	 * the owner of the local PMU.
 	 */
 	if (likely(is_loaded)) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3233,7 +3231,7 @@ pfm_write_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			/*
 		 	 * write thread state
 		 	 */
-			if (is_system == 0) thread->pmds[cnum] = hw_value;
+			if (is_system == 0) ctx->th_pmds[cnum] = hw_value;
 
 			/*
 			 * write hardware register if we can
@@ -3299,7 +3297,6 @@ abort_mission:
 static int
 pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 {
-	struct thread_struct *thread = NULL;
 	struct task_struct *task;
 	unsigned long val = 0UL, lval, ovfl_mask, sval;
 	pfarg_reg_t *req = (pfarg_reg_t *)arg;
@@ -3323,7 +3320,6 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	if (state == PFM_CTX_ZOMBIE) return -EINVAL;
 
 	if (likely(is_loaded)) {
-		thread = &task->thread;
 		/*
 		 * In system wide and when the context is loaded, access can only happen
 		 * when the caller is running on the CPU being monitored by the session.
@@ -3385,7 +3381,7 @@ pfm_read_pmds(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 			 * if context is zombie, then task does not exist anymore.
 			 * In this case, we use the full value saved in the context (pfm_flush_regs()).
 			 */
-			val = is_loaded ? thread->pmds[cnum] : 0UL;
+			val = is_loaded ? ctx->th_pmds[cnum] : 0UL;
 		}
 		rd_func = pmu_conf->pmd_desc[cnum].read_check;
 
@@ -4354,8 +4350,8 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 	pfm_copy_pmds(task, ctx);
 	pfm_copy_pmcs(task, ctx);
 
-	pmcs_source = thread->pmcs;
-	pmds_source = thread->pmds;
+	pmcs_source = ctx->th_pmcs;
+	pmds_source = ctx->th_pmds;
 
 	/*
 	 * always the case for system-wide
@@ -5864,14 +5860,12 @@ void
 pfm_save_regs(struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long flags;
 	u64 psr;
 
 
 	ctx = PFM_GET_CTX(task);
 	if (ctx == NULL) return;
-	t = &task->thread;
 
 	/*
  	 * we always come here with interrupts ALREADY disabled by
@@ -5929,19 +5923,19 @@ pfm_save_regs(struct task_struct *task)
 	 * guarantee we will be schedule at that same
 	 * CPU again.
 	 */
-	pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]);
+	pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]);
 
 	/*
 	 * save pmc0 ia64_srlz_d() done in pfm_save_pmds()
 	 * we will need it on the restore path to check
 	 * for pending overflow.
 	 */
-	t->pmcs[0] = ia64_get_pmc(0);
+	ctx->th_pmcs[0] = ia64_get_pmc(0);
 
 	/*
 	 * unfreeze PMU if had pending overflows
 	 */
-	if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
+	if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
 
 	/*
 	 * finally, allow context access.
@@ -5986,7 +5980,6 @@ static void
 pfm_lazy_save_regs (struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long flags;
 
 	{ u64 psr  = pfm_get_psr();
@@ -5994,7 +5987,6 @@ pfm_lazy_save_regs (struct task_struct *task)
 	}
 
 	ctx = PFM_GET_CTX(task);
-	t   = &task->thread;
 
 	/*
 	 * we need to mask PMU overflow here to
@@ -6019,19 +6011,19 @@ pfm_lazy_save_regs (struct task_struct *task)
 	/*
 	 * save all the pmds we use
 	 */
-	pfm_save_pmds(t->pmds, ctx->ctx_used_pmds[0]);
+	pfm_save_pmds(ctx->th_pmds, ctx->ctx_used_pmds[0]);
 
 	/*
 	 * save pmc0 ia64_srlz_d() done in pfm_save_pmds()
 	 * it is needed to check for pended overflow
 	 * on the restore path
 	 */
-	t->pmcs[0] = ia64_get_pmc(0);
+	ctx->th_pmcs[0] = ia64_get_pmc(0);
 
 	/*
 	 * unfreeze PMU if had pending overflows
 	 */
-	if (t->pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
+	if (ctx->th_pmcs[0] & ~0x1UL) pfm_unfreeze_pmu();
 
 	/*
 	 * now get can unmask PMU interrupts, they will
@@ -6050,7 +6042,6 @@ void
 pfm_load_regs (struct task_struct *task)
 {
 	pfm_context_t *ctx;
-	struct thread_struct *t;
 	unsigned long pmc_mask = 0UL, pmd_mask = 0UL;
 	unsigned long flags;
 	u64 psr, psr_up;
@@ -6061,11 +6052,10 @@ pfm_load_regs (struct task_struct *task)
 
 	BUG_ON(GET_PMU_OWNER());
 
-	t     = &task->thread;
 	/*
 	 * possible on unload
 	 */
-	if (unlikely((t->flags & IA64_THREAD_PM_VALID) == 0)) return;
+	if (unlikely((task->thread.flags & IA64_THREAD_PM_VALID) == 0)) return;
 
 	/*
  	 * we always come here with interrupts ALREADY disabled by
@@ -6147,21 +6137,21 @@ pfm_load_regs (struct task_struct *task)
 	 *
 	 * XXX: optimize here
 	 */
-	if (pmd_mask) pfm_restore_pmds(t->pmds, pmd_mask);
-	if (pmc_mask) pfm_restore_pmcs(t->pmcs, pmc_mask);
+	if (pmd_mask) pfm_restore_pmds(ctx->th_pmds, pmd_mask);
+	if (pmc_mask) pfm_restore_pmcs(ctx->th_pmcs, pmc_mask);
 
 	/*
 	 * check for pending overflow at the time the state
 	 * was saved.
 	 */
-	if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) {
+	if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) {
 		/*
 		 * reload pmc0 with the overflow information
 		 * On McKinley PMU, this will trigger a PMU interrupt
 		 */
-		ia64_set_pmc(0, t->pmcs[0]);
+		ia64_set_pmc(0, ctx->th_pmcs[0]);
 		ia64_srlz_d();
-		t->pmcs[0] = 0UL;
+		ctx->th_pmcs[0] = 0UL;
 
 		/*
 		 * will replay the PMU interrupt
@@ -6214,7 +6204,6 @@ pfm_load_regs (struct task_struct *task)
 void
 pfm_load_regs (struct task_struct *task)
 {
-	struct thread_struct *t;
 	pfm_context_t *ctx;
 	struct task_struct *owner;
 	unsigned long pmd_mask, pmc_mask;
@@ -6223,7 +6212,6 @@ pfm_load_regs (struct task_struct *task)
 
 	owner = GET_PMU_OWNER();
 	ctx   = PFM_GET_CTX(task);
-	t     = &task->thread;
 	psr   = pfm_get_psr();
 
 	BUG_ON(psr & (IA64_PSR_UP|IA64_PSR_PP));
@@ -6286,22 +6274,22 @@ pfm_load_regs (struct task_struct *task)
 	 */
 	pmc_mask = ctx->ctx_all_pmcs[0];
 
-	pfm_restore_pmds(t->pmds, pmd_mask);
-	pfm_restore_pmcs(t->pmcs, pmc_mask);
+	pfm_restore_pmds(ctx->th_pmds, pmd_mask);
+	pfm_restore_pmcs(ctx->th_pmcs, pmc_mask);
 
 	/*
 	 * check for pending overflow at the time the state
 	 * was saved.
 	 */
-	if (unlikely(PMC0_HAS_OVFL(t->pmcs[0]))) {
+	if (unlikely(PMC0_HAS_OVFL(ctx->th_pmcs[0]))) {
 		/*
 		 * reload pmc0 with the overflow information
 		 * On McKinley PMU, this will trigger a PMU interrupt
 		 */
-		ia64_set_pmc(0, t->pmcs[0]);
+		ia64_set_pmc(0, ctx->th_pmcs[0]);
 		ia64_srlz_d();
 
-		t->pmcs[0] = 0UL;
+		ctx->th_pmcs[0] = 0UL;
 
 		/*
 		 * will replay the PMU interrupt
@@ -6376,11 +6364,11 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 		 */
 		pfm_unfreeze_pmu();
 	} else {
-		pmc0 = task->thread.pmcs[0];
+		pmc0 = ctx->th_pmcs[0];
 		/*
 		 * clear whatever overflow status bits there were
 		 */
-		task->thread.pmcs[0] = 0;
+		ctx->th_pmcs[0] = 0;
 	}
 	ovfl_val = pmu_conf->ovfl_val;
 	/*
@@ -6401,7 +6389,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 		/*
 		 * can access PMU always true in system wide mode
 		 */
-		val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : task->thread.pmds[i];
+		val = pmd_val = can_access_pmu ? ia64_get_pmd(i) : ctx->th_pmds[i];
 
 		if (PMD_IS_COUNTING(i)) {
 			DPRINT(("[%d] pmd[%d] ctx_pmd=0x%lx hw_pmd=0x%lx\n",
@@ -6433,7 +6421,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 
 		DPRINT(("[%d] ctx_pmd[%d]=0x%lx  pmd_val=0x%lx\n", task->pid, i, val, pmd_val));
 
-		if (is_self) task->thread.pmds[i] = pmd_val;
+		if (is_self) ctx->th_pmds[i] = pmd_val;
 
 		ctx->ctx_pmds[i].val = val;
 	}
@@ -6677,7 +6665,7 @@ pfm_init(void)
 	       ffz(pmu_conf->ovfl_val));
 
 	/* sanity check */
-	if (pmu_conf->num_pmds >= IA64_NUM_PMD_REGS || pmu_conf->num_pmcs >= IA64_NUM_PMC_REGS) {
+	if (pmu_conf->num_pmds >= PFM_NUM_PMD_REGS || pmu_conf->num_pmcs >= PFM_NUM_PMC_REGS) {
 		printk(KERN_ERR "perfmon: not enough pmc/pmd, perfmon disabled\n");
 		pmu_conf = NULL;
 		return -1;
@@ -6752,7 +6740,6 @@ void
 dump_pmu_state(const char *from)
 {
 	struct task_struct *task;
-	struct thread_struct *t;
 	struct pt_regs *regs;
 	pfm_context_t *ctx;
 	unsigned long psr, dcr, info, flags;
@@ -6797,16 +6784,14 @@ dump_pmu_state(const char *from)
 	ia64_psr(regs)->up = 0;
 	ia64_psr(regs)->pp = 0;
 
-	t = &current->thread;
-
 	for (i=1; PMC_IS_LAST(i) == 0; i++) {
 		if (PMC_IS_IMPL(i) == 0) continue;
-		printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, t->pmcs[i]);
+		printk("->CPU%d pmc[%d]=0x%lx thread_pmc[%d]=0x%lx\n", this_cpu, i, ia64_get_pmc(i), i, ctx->th_pmcs[i]);
 	}
 
 	for (i=1; PMD_IS_LAST(i) == 0; i++) {
 		if (PMD_IS_IMPL(i) == 0) continue;
-		printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, t->pmds[i]);
+		printk("->CPU%d pmd[%d]=0x%lx thread_pmd[%d]=0x%lx\n", this_cpu, i, ia64_get_pmd(i), i, ctx->th_pmds[i]);
 	}
 
 	if (ctx) {
diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h
index ead68f8eb883..5830d36fd8e6 100644
--- a/include/asm-ia64/processor.h
+++ b/include/asm-ia64/processor.h
@@ -20,12 +20,6 @@
 #include <asm/ustack.h>
 
 #define IA64_NUM_DBG_REGS	8
-/*
- * Limits for PMC and PMD are set to less than maximum architected values
- * but should be sufficient for a while
- */
-#define IA64_NUM_PMC_REGS	64
-#define IA64_NUM_PMD_REGS	64
 
 #define DEFAULT_MAP_BASE	__IA64_UL_CONST(0x2000000000000000)
 #define DEFAULT_TASK_SIZE	__IA64_UL_CONST(0xa000000000000000)
@@ -263,13 +257,9 @@ struct thread_struct {
 # define INIT_THREAD_IA32
 #endif /* CONFIG_IA32_SUPPORT */
 #ifdef CONFIG_PERFMON
-	__u64 pmcs[IA64_NUM_PMC_REGS];
-	__u64 pmds[IA64_NUM_PMD_REGS];
 	void *pfm_context;		     /* pointer to detailed PMU context */
 	unsigned long pfm_needs_checking;    /* when >0, pending perfmon work on kernel exit */
-# define INIT_THREAD_PM		.pmcs =			{0UL, },  \
-				.pmds =			{0UL, },  \
-				.pfm_context =		NULL,     \
+# define INIT_THREAD_PM		.pfm_context =		NULL,     \
 				.pfm_needs_checking =	0UL,
 #else
 # define INIT_THREAD_PM
-- 
cgit v1.2.1


From fd1dfc6f0165e2ad426665e517103672d6832f90 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Tue, 5 Sep 2006 11:56:51 -0700
Subject: [IA64-SGI] Do not request DMA memory for BTE

The GFP_DMA option usually does nothing on SN2 since all of memory is in thei
DMA zone and the BTE has always been capable of addressing all of memory.
So there is no need to get memory from a restricted range of memory (which
is what GFP_DMA is for).

Remove useless __GFP_DMA option.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/bte.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c
index 27dee4584061..7f73ad4408aa 100644
--- a/arch/ia64/sn/kernel/bte.c
+++ b/arch/ia64/sn/kernel/bte.c
@@ -277,8 +277,7 @@ bte_result_t bte_unaligned_copy(u64 src, u64 dest, u64 len, u64 mode)
 	}
 
 	/* temporary buffer used during unaligned transfers */
-	bteBlock_unaligned = kmalloc(len + 3 * L1_CACHE_BYTES,
-				     GFP_KERNEL | GFP_DMA);
+	bteBlock_unaligned = kmalloc(len + 3 * L1_CACHE_BYTES, GFP_KERNEL);
 	if (bteBlock_unaligned == NULL) {
 		return BTEFAIL_NOTAVAIL;
 	}
-- 
cgit v1.2.1


From a192dc16000241dc02990a36b6830839b73c44de Mon Sep 17 00:00:00 2001
From: Jakub Jelinek <jakub@redhat.com>
Date: Tue, 26 Sep 2006 14:00:56 -0700
Subject: [IA64] Implement futex primitives

Implement futex_atomic_op_inuser() and futex_atomic_cmpxchg_inatomic()
on IA64 in order to fully support all futex functionality.

Signed-off-by: Jakub Jelinek <jakub@redhat.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/futex.h | 122 ++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 120 insertions(+), 2 deletions(-)

diff --git a/include/asm-ia64/futex.h b/include/asm-ia64/futex.h
index 6a332a9f099c..07d77f3a8cbe 100644
--- a/include/asm-ia64/futex.h
+++ b/include/asm-ia64/futex.h
@@ -1,6 +1,124 @@
 #ifndef _ASM_FUTEX_H
 #define _ASM_FUTEX_H
 
-#include <asm-generic/futex.h>
+#include <linux/futex.h>
+#include <asm/errno.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
 
-#endif
+#define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \
+do {									\
+	register unsigned long r8 __asm ("r8") = 0;			\
+	__asm__ __volatile__(						\
+		"	mf;;					\n"	\
+		"[1:] "	insn ";;				\n"	\
+		"	.xdata4 \"__ex_table\", 1b-., 2f-.	\n"	\
+		"[2:]"							\
+		: "+r" (r8), "=r" (oldval)				\
+		: "r" (uaddr), "r" (oparg)				\
+		: "memory");						\
+	ret = r8;							\
+} while (0)
+
+#define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \
+do {									\
+	register unsigned long r8 __asm ("r8") = 0;			\
+	int val, newval;						\
+	do {								\
+		__asm__ __volatile__(					\
+			"	mf;;				  \n"	\
+			"[1:]	ld4 %3=[%4];;			  \n"	\
+			"	mov %2=%3			  \n"	\
+				insn	";;			  \n"	\
+			"	mov ar.ccv=%2;;			  \n"	\
+			"[2:]	cmpxchg4.acq %1=[%4],%3,ar.ccv;;  \n"	\
+			"	.xdata4 \"__ex_table\", 1b-., 3f-.\n"	\
+			"	.xdata4 \"__ex_table\", 2b-., 3f-.\n"	\
+			"[3:]"						\
+			: "+r" (r8), "=r" (val), "=&r" (oldval),	\
+			   "=&r" (newval)				\
+			: "r" (uaddr), "r" (oparg)			\
+			: "memory");					\
+		if (unlikely (r8))					\
+			break;						\
+	} while (unlikely (val != oldval));				\
+	ret = r8;							\
+} while (0)
+
+static inline int
+futex_atomic_op_inuser (int encoded_op, int __user *uaddr)
+{
+	int op = (encoded_op >> 28) & 7;
+	int cmp = (encoded_op >> 24) & 15;
+	int oparg = (encoded_op << 8) >> 20;
+	int cmparg = (encoded_op << 20) >> 20;
+	int oldval = 0, ret;
+	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+		oparg = 1 << oparg;
+
+	if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+	inc_preempt_count();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_atomic_op1("xchg4 %1=[%2],%3", ret, oldval, uaddr,
+				   oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_atomic_op2("add %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_atomic_op2("or %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_atomic_op2("and %3=%3,%5", ret, oldval, uaddr,
+				   ~oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_atomic_op2("xor %3=%3,%5", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+
+	dec_preempt_count();
+
+	if (!ret) {
+		switch (cmp) {
+		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
+		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
+		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
+		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
+		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
+		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+		default: ret = -ENOSYS;
+		}
+	}
+	return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval)
+{
+	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+		return -EFAULT;
+
+	{
+		register unsigned long r8 __asm ("r8");
+		__asm__ __volatile__(
+			"	mf;;					\n"
+			"	mov ar.ccv=%3;;				\n"
+			"[1:]	cmpxchg4.acq %0=[%1],%2,ar.ccv		\n"
+			"	.xdata4 \"__ex_table\", 1b-., 2f-.	\n"
+			"[2:]"
+			: "=r" (r8)
+			: "r" (uaddr), "r" (newval),
+			  "rO" ((long) (unsigned) oldval)
+			: "memory");
+		return r8;
+	}
+}
+
+#endif /* _ASM_FUTEX_H */
-- 
cgit v1.2.1


From 5c55cd63a77a85f603c98c2171a8054ca34b6a9f Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Tue, 26 Sep 2006 14:04:42 -0700
Subject: Revert "[IA64] Unwire set/get_robust_list"

This reverts commit 2636255488484e04d6d54303d2b0ec30f7ef7e02.

Jakub Jelinek provided the missing futex_atomic_cmpxchg_inatomic()
function, so now it should be safe to re-enable these syscalls.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S  | 4 ++--
 include/asm-ia64/unistd.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index fef06571be99..12701cf32d99 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1605,8 +1605,8 @@ sys_call_table:
 	data8 sys_ni_syscall			// 1295 reserved for ppoll
 	data8 sys_unshare
 	data8 sys_splice
-	data8 sys_ni_syscall			// reserved for set_robust_list
-	data8 sys_ni_syscall			// reserved for get_robust_list
+	data8 sys_set_robust_list
+	data8 sys_get_robust_list
 	data8 sys_sync_file_range		// 1300
 	data8 sys_tee
 	data8 sys_vmsplice
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index f581662c5ab8..bb0eb727dcd0 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -286,7 +286,8 @@
 /* 1294, 1295 reserved for pselect/ppoll */
 #define __NR_unshare			1296
 #define __NR_splice			1297
-/* 1298, 1299 reserved for set_robust_list/get_robust_list */
+#define __NR_set_robust_list		1298
+#define __NR_get_robust_list		1299
 #define __NR_sync_file_range		1300
 #define __NR_tee			1301
 #define __NR_vmsplice			1302
-- 
cgit v1.2.1


From 68d0d7abcc4cfcf0a1eeb49785cb32ec05473e42 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 26 Sep 2006 22:24:31 +0100
Subject: [PATCH] pata_serverworks: correct PCI ID in cable detection table

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/pata_serverworks.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index af456113c55d..75cc5a22da0c 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -41,7 +41,7 @@
 #include <linux/libata.h>
 
 #define DRV_NAME "pata_serverworks"
-#define DRV_VERSION "0.3.6"
+#define DRV_VERSION "0.3.7"
 
 #define SVWKS_CSB5_REVISION_NEW	0x92 /* min PCI_REVISION_ID for UDMA5 (A2.0) */
 #define SVWKS_CSB6_REVISION	0xa0 /* min PCI_REVISION_ID for UDMA4 (A1.0) */
@@ -128,7 +128,7 @@ static struct sv_cable_table cable_detect[] = {
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_VENDOR_ID_DELL, dell_cable },
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE, PCI_VENDOR_ID_DELL, dell_cable },
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_VENDOR_ID_SUN,  sun_cable },
-	{ PCI_DEVICE_ID_SERVERWORKS_OSB4, PCI_ANY_ID, osb4_cable },
+	{ PCI_DEVICE_ID_SERVERWORKS_OSB4IDE, PCI_ANY_ID, osb4_cable },
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB5IDE, PCI_ANY_ID, csb_cable },
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE, PCI_ANY_ID, csb_cable },
 	{ PCI_DEVICE_ID_SERVERWORKS_CSB6IDE2, PCI_ANY_ID, csb_cable },
-- 
cgit v1.2.1


From efbf3f14204b77e6afc8c13571357173de9ab707 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Tue, 26 Sep 2006 17:10:53 -0400
Subject: [libata] pata_serverworks: fill in ->irq_clear hook

Required by libata, as it is called unconditionally.

Fixes an obvious oops.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/pata_serverworks.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index 75cc5a22da0c..d62051b8da05 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -356,6 +356,8 @@ static struct ata_port_operations serverworks_osb4_port_ops = {
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
 	.port_start	= ata_port_start,
 	.port_stop	= ata_port_stop,
 	.host_stop	= ata_host_stop
@@ -389,6 +391,8 @@ static struct ata_port_operations serverworks_csb_port_ops = {
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
+	.irq_clear	= ata_bmdma_irq_clear,
+
 	.port_start	= ata_port_start,
 	.port_stop	= ata_port_stop,
 	.host_stop	= ata_host_stop
-- 
cgit v1.2.1


From f5a3f3dc189485d607fbd42678cc23958acc0a6e Mon Sep 17 00:00:00 2001
From: Zou Nan hai <nanhai.zou@intel.com>
Date: Thu, 14 Sep 2006 08:25:15 +0800
Subject: [IA64] Make gp value point to Region 5 in mca handler

MCA dispatch code take physical address of GP passed from SAL, then call
DATA_PA_TO_VA twice on GP before call into C code.  The first time is
in ia64_set_kernel_register, the second time is in VIRTUAL_MODE_ENTER.
The gp is changed to a virtual address in region 7 because DATA_PA_TO_VA
is implemented by dep instruction.

However when notify blocks were called from MCA handler code, because
notify blocks are supported by callback function pointers, gp value
value was switched to region 5 again.

The patch set gp register to kernel gp of region 5 at entry of MCA
dispatch.

Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca_asm.S | 5 -----
 include/asm-ia64/mca_asm.h | 2 +-
 2 files changed, 1 insertion(+), 6 deletions(-)

diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S
index 96047491d1b9..ee3ff76c8a35 100644
--- a/arch/ia64/kernel/mca_asm.S
+++ b/arch/ia64/kernel/mca_asm.S
@@ -1025,18 +1025,13 @@ ia64_old_stack:
 
 ia64_set_kernel_registers:
 	add temp3=MCA_SP_OFFSET, r3
-	add temp4=MCA_SOS_OFFSET+SOS(OS_GP), r3
 	mov b0=r2		// save return address
 	GET_IA64_MCA_DATA(temp1)
 	;;
-	add temp4=temp4, temp1	// &struct ia64_sal_os_state.os_gp
 	add r12=temp1, temp3	// kernel stack pointer on MCA/INIT stack
 	add r13=temp1, r3	// set current to start of MCA/INIT stack
 	add r20=temp1, r3	// physical start of MCA/INIT stack
 	;;
-	ld8 r1=[temp4]		// OS GP from SAL OS state
-	;;
-	DATA_PA_TO_VA(r1,temp1)
 	DATA_PA_TO_VA(r12,temp2)
 	DATA_PA_TO_VA(r13,temp3)
 	;;
diff --git a/include/asm-ia64/mca_asm.h b/include/asm-ia64/mca_asm.h
index 27c9203d8ce3..76203f9a8718 100644
--- a/include/asm-ia64/mca_asm.h
+++ b/include/asm-ia64/mca_asm.h
@@ -197,9 +197,9 @@
 	movl	temp2 = start_addr;				\
 	;;							\
 	mov	cr.iip = temp2;					\
+	movl	gp = __gp					\
 	;;							\
 	DATA_PA_TO_VA(sp, temp1);				\
-	DATA_PA_TO_VA(gp, temp2);				\
 	srlz.i;							\
 	;;							\
 	nop	1;						\
-- 
cgit v1.2.1


From 709a6c1c07673a9f7879b5f7306dc8d723db93a2 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 13 Sep 2006 08:43:42 -0400
Subject: [IA64] show_mem() printk levels

Use the default sysrq printk level for printing show_mem() output both
for disconfig and contig versions. This is consistent with the printk
level used on other architectures (well ia32 at least).

Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/mm/contig.c    | 17 +++++++++--------
 arch/ia64/mm/discontig.c | 29 +++++++++++++++--------------
 2 files changed, 24 insertions(+), 22 deletions(-)

diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index e004143ba86b..537c2f3833ee 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -41,10 +41,11 @@ show_mem (void)
 	int i, total = 0, reserved = 0;
 	int shared = 0, cached = 0;
 
-	printk("Mem-info:\n");
+	printk(KERN_INFO "Mem-info:\n");
 	show_free_areas();
 
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+	printk(KERN_INFO "Free swap:       %6ldkB\n",
+	       nr_swap_pages<<(PAGE_SHIFT-10));
 	i = max_mapnr;
 	for (i = 0; i < max_mapnr; i++) {
 		if (!pfn_valid(i)) {
@@ -63,12 +64,12 @@ show_mem (void)
 		else if (page_count(mem_map + i))
 			shared += page_count(mem_map + i) - 1;
 	}
-	printk("%d pages of RAM\n", total);
-	printk("%d reserved pages\n", reserved);
-	printk("%d pages shared\n", shared);
-	printk("%d pages swap cached\n", cached);
-	printk("%ld pages in page table cache\n",
-		pgtable_quicklist_total_size());
+	printk(KERN_INFO "%d pages of RAM\n", total);
+	printk(KERN_INFO "%d reserved pages\n", reserved);
+	printk(KERN_INFO "%d pages shared\n", shared);
+	printk(KERN_INFO "%d pages swap cached\n", cached);
+	printk(KERN_INFO "%ld pages in page table cache\n",
+	       pgtable_quicklist_total_size());
 }
 
 /* physical address where the bootmem map is located */
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index d260bffa01ab..0ce247444f21 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -547,15 +547,16 @@ void show_mem(void)
 	unsigned long total_present = 0;
 	pg_data_t *pgdat;
 
-	printk("Mem-info:\n");
+	printk(KERN_INFO "Mem-info:\n");
 	show_free_areas();
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
+	printk(KERN_INFO "Free swap:       %6ldkB\n",
+	       nr_swap_pages<<(PAGE_SHIFT-10));
 	for_each_online_pgdat(pgdat) {
 		unsigned long present;
 		unsigned long flags;
 		int shared = 0, cached = 0, reserved = 0;
 
-		printk("Node ID: %d\n", pgdat->node_id);
+		printk(KERN_INFO "Node ID: %d\n", pgdat->node_id);
 		pgdat_resize_lock(pgdat, &flags);
 		present = pgdat->node_present_pages;
 		for(i = 0; i < pgdat->node_spanned_pages; i++) {
@@ -579,18 +580,18 @@ void show_mem(void)
 		total_reserved += reserved;
 		total_cached += cached;
 		total_shared += shared;
-		printk("\t%ld pages of RAM\n", present);
-		printk("\t%d reserved pages\n", reserved);
-		printk("\t%d pages shared\n", shared);
-		printk("\t%d pages swap cached\n", cached);
+		printk(KERN_INFO "\t%ld pages of RAM\n", present);
+		printk(KERN_INFO "\t%d reserved pages\n", reserved);
+		printk(KERN_INFO "\t%d pages shared\n", shared);
+		printk(KERN_INFO "\t%d pages swap cached\n", cached);
 	}
-	printk("%ld pages of RAM\n", total_present);
-	printk("%d reserved pages\n", total_reserved);
-	printk("%d pages shared\n", total_shared);
-	printk("%d pages swap cached\n", total_cached);
-	printk("Total of %ld pages in page table cache\n",
-		pgtable_quicklist_total_size());
-	printk("%d free buffer pages\n", nr_free_buffer_pages());
+	printk(KERN_INFO "%ld pages of RAM\n", total_present);
+	printk(KERN_INFO "%d reserved pages\n", total_reserved);
+	printk(KERN_INFO "%d pages shared\n", total_shared);
+	printk(KERN_INFO "%d pages swap cached\n", total_cached);
+	printk(KERN_INFO "Total of %ld pages in page table cache\n",
+	       pgtable_quicklist_total_size());
+	printk(KERN_INFO "%d free buffer pages\n", nr_free_buffer_pages());
 }
 
 /**
-- 
cgit v1.2.1


From fb4633019fd0a871cdf315e1f4082f4c6926af82 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Tue, 26 Sep 2006 17:16:32 -0400
Subject: [libata] One more s/15/ATA_SECONDARY_IRQ/ substitution

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/libata-sff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 9222b6bb010a..08b3a407473e 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -898,7 +898,7 @@ static struct ata_probe_ent *ata_pci_init_legacy_port(struct pci_dev *pdev,
 		if (probe_ent->irq)
 			probe_ent->irq2 = ATA_SECONDARY_IRQ;
 		else
-			probe_ent->irq = 15;
+			probe_ent->irq = ATA_SECONDARY_IRQ;
 		probe_ent->port[1].cmd_addr = ATA_SECONDARY_CMD;
 		probe_ent->port[1].altstatus_addr =
 		probe_ent->port[1].ctl_addr = ATA_SECONDARY_CTL;
-- 
cgit v1.2.1


From 816add4e986499145135c4014a7c8a8857f9f3c3 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Thu, 14 Sep 2006 07:10:30 -0400
Subject: [IA64] trim output of show_mem()

Cut the number of lines of memory info output per node from five
to one line.

Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/mm/discontig.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 0ce247444f21..6ea279766334 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -551,12 +551,12 @@ void show_mem(void)
 	show_free_areas();
 	printk(KERN_INFO "Free swap:       %6ldkB\n",
 	       nr_swap_pages<<(PAGE_SHIFT-10));
+	printk(KERN_INFO "Node memory in pages:\n");
 	for_each_online_pgdat(pgdat) {
 		unsigned long present;
 		unsigned long flags;
 		int shared = 0, cached = 0, reserved = 0;
 
-		printk(KERN_INFO "Node ID: %d\n", pgdat->node_id);
 		pgdat_resize_lock(pgdat, &flags);
 		present = pgdat->node_present_pages;
 		for(i = 0; i < pgdat->node_spanned_pages; i++) {
@@ -580,10 +580,9 @@ void show_mem(void)
 		total_reserved += reserved;
 		total_cached += cached;
 		total_shared += shared;
-		printk(KERN_INFO "\t%ld pages of RAM\n", present);
-		printk(KERN_INFO "\t%d reserved pages\n", reserved);
-		printk(KERN_INFO "\t%d pages shared\n", shared);
-		printk(KERN_INFO "\t%d pages swap cached\n", cached);
+		printk(KERN_INFO "Node %4d:  RAM: %11ld, rsvd: %8d, "
+		       "shrd: %10d, swpd: %10d\n", pgdat->node_id,
+		       present, reserved, shared, cached);
 	}
 	printk(KERN_INFO "%ld pages of RAM\n", total_present);
 	printk(KERN_INFO "%d reserved pages\n", total_reserved);
-- 
cgit v1.2.1


From eb2a2fd91f7c8a53b15063d6f08cf22b9a56cbfb Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Tue, 26 Sep 2006 23:23:45 +0200
Subject: [PATCH] Modularize generic HDLC

This patch enables building of individual WAN protocol support
routines (parts of generic HDLC) as separate modules.
All protocol-private definitions are moved from hdlc.h file
to protocol drivers. User-space interface and interface
between generic HDLC and underlying low-level HDLC drivers
are unchanged.

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/wan/Kconfig        |  12 +-
 drivers/net/wan/Makefile       |  19 +-
 drivers/net/wan/hdlc.c         | 368 ++++++++++++++++++++++++++++++++++++++
 drivers/net/wan/hdlc_cisco.c   | 198 ++++++++++++++-------
 drivers/net/wan/hdlc_fr.c      | 389 +++++++++++++++++++++++++----------------
 drivers/net/wan/hdlc_generic.c | 339 -----------------------------------
 drivers/net/wan/hdlc_ppp.c     |  77 ++++++--
 drivers/net/wan/hdlc_raw.c     |  50 +++++-
 drivers/net/wan/hdlc_raw_eth.c |  49 +++++-
 drivers/net/wan/hdlc_x25.c     |  54 ++++--
 include/linux/hdlc.h           | 201 +++++----------------
 include/linux/hdlc/ioctl.h     |  33 ++++
 12 files changed, 1012 insertions(+), 777 deletions(-)
 create mode 100644 drivers/net/wan/hdlc.c
 delete mode 100644 drivers/net/wan/hdlc_generic.c

diff --git a/drivers/net/wan/Kconfig b/drivers/net/wan/Kconfig
index 54b8e492ef97..58b7efbb0750 100644
--- a/drivers/net/wan/Kconfig
+++ b/drivers/net/wan/Kconfig
@@ -154,7 +154,7 @@ config HDLC
 	  If unsure, say N.
 
 config HDLC_RAW
-	bool "Raw HDLC support"
+	tristate "Raw HDLC support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting raw HDLC over WAN connections.
@@ -162,7 +162,7 @@ config HDLC_RAW
 	  If unsure, say N.
 
 config HDLC_RAW_ETH
-	bool "Raw HDLC Ethernet device support"
+	tristate "Raw HDLC Ethernet device support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting raw HDLC Ethernet device emulation
@@ -173,7 +173,7 @@ config HDLC_RAW_ETH
 	  If unsure, say N.
 
 config HDLC_CISCO
-	bool "Cisco HDLC support"
+	tristate "Cisco HDLC support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting Cisco HDLC over WAN connections.
@@ -181,7 +181,7 @@ config HDLC_CISCO
 	  If unsure, say N.
 
 config HDLC_FR
-	bool "Frame Relay support"
+	tristate "Frame Relay support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting Frame Relay over WAN connections.
@@ -189,7 +189,7 @@ config HDLC_FR
 	  If unsure, say N.
 
 config HDLC_PPP
-	bool "Synchronous Point-to-Point Protocol (PPP) support"
+	tristate "Synchronous Point-to-Point Protocol (PPP) support"
 	depends on HDLC
 	help
 	  Generic HDLC driver supporting PPP over WAN connections.
@@ -197,7 +197,7 @@ config HDLC_PPP
 	  If unsure, say N.
 
 config HDLC_X25
-	bool "X.25 protocol support"
+	tristate "X.25 protocol support"
 	depends on HDLC && (LAPB=m && HDLC=m || LAPB=y)
 	help
 	  Generic HDLC driver supporting X.25 over WAN connections.
diff --git a/drivers/net/wan/Makefile b/drivers/net/wan/Makefile
index 316ca6869d5e..83ec2c87ba3f 100644
--- a/drivers/net/wan/Makefile
+++ b/drivers/net/wan/Makefile
@@ -9,14 +9,13 @@ cyclomx-y                       := cycx_main.o
 cyclomx-$(CONFIG_CYCLOMX_X25)	+= cycx_x25.o
 cyclomx-objs			:= $(cyclomx-y)  
 
-hdlc-y				:= hdlc_generic.o
-hdlc-$(CONFIG_HDLC_RAW)		+= hdlc_raw.o
-hdlc-$(CONFIG_HDLC_RAW_ETH)	+= hdlc_raw_eth.o
-hdlc-$(CONFIG_HDLC_CISCO)	+= hdlc_cisco.o
-hdlc-$(CONFIG_HDLC_FR)		+= hdlc_fr.o
-hdlc-$(CONFIG_HDLC_PPP)		+= hdlc_ppp.o
-hdlc-$(CONFIG_HDLC_X25)		+= hdlc_x25.o
-hdlc-objs			:= $(hdlc-y)
+obj-$(CONFIG_HDLC)		+= hdlc.o
+obj-$(CONFIG_HDLC_RAW)		+= hdlc_raw.o
+obj-$(CONFIG_HDLC_RAW_ETH)	+= hdlc_raw_eth.o
+obj-$(CONFIG_HDLC_CISCO)	+= hdlc_cisco.o
+obj-$(CONFIG_HDLC_FR)		+= hdlc_fr.o
+obj-$(CONFIG_HDLC_PPP)		+= hdlc_ppp.o	syncppp.o
+obj-$(CONFIG_HDLC_X25)		+= hdlc_x25.o
 
 pc300-y				:= pc300_drv.o
 pc300-$(CONFIG_PC300_MLPPP)	+= pc300_tty.o
@@ -38,10 +37,6 @@ obj-$(CONFIG_CYCLADES_SYNC)	+= cycx_drv.o cyclomx.o
 obj-$(CONFIG_LAPBETHER)		+= lapbether.o
 obj-$(CONFIG_SBNI)		+= sbni.o
 obj-$(CONFIG_PC300)		+= pc300.o
-obj-$(CONFIG_HDLC)		+= hdlc.o
-ifeq ($(CONFIG_HDLC_PPP),y)
-  obj-$(CONFIG_HDLC)		+= syncppp.o
-endif
 obj-$(CONFIG_N2)		+= n2.o
 obj-$(CONFIG_C101)		+= c101.o
 obj-$(CONFIG_WANXL)		+= wanxl.o
diff --git a/drivers/net/wan/hdlc.c b/drivers/net/wan/hdlc.c
new file mode 100644
index 000000000000..db354e0edbe5
--- /dev/null
+++ b/drivers/net/wan/hdlc.c
@@ -0,0 +1,368 @@
+/*
+ * Generic HDLC support routines for Linux
+ *
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ *
+ * Currently supported:
+ *	* raw IP-in-HDLC
+ *	* Cisco HDLC
+ *	* Frame Relay with ANSI or CCITT LMI (both user and network side)
+ *	* PPP
+ *	* X.25
+ *
+ * Use sethdlc utility to set line parameters, protocol and PVCs
+ *
+ * How does it work:
+ * - proto->open(), close(), start(), stop() calls are serialized.
+ *   The order is: open, [ start, stop ... ] close ...
+ * - proto->start() and stop() are called with spin_lock_irq held.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/poll.h>
+#include <linux/errno.h>
+#include <linux/if_arp.h>
+#include <linux/init.h>
+#include <linux/skbuff.h>
+#include <linux/pkt_sched.h>
+#include <linux/inetdevice.h>
+#include <linux/lapb.h>
+#include <linux/rtnetlink.h>
+#include <linux/notifier.h>
+#include <linux/hdlc.h>
+
+
+static const char* version = "HDLC support module revision 1.20";
+
+#undef DEBUG_LINK
+
+static struct hdlc_proto *first_proto = NULL;
+
+
+static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
+{
+	if ((new_mtu < 68) || (new_mtu > HDLC_MAX_MTU))
+		return -EINVAL;
+	dev->mtu = new_mtu;
+	return 0;
+}
+
+
+
+static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
+{
+	return hdlc_stats(dev);
+}
+
+
+
+static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *p, struct net_device *orig_dev)
+{
+	struct hdlc_device_desc *desc = dev_to_desc(dev);
+	if (desc->netif_rx)
+		return desc->netif_rx(skb);
+
+	desc->stats.rx_dropped++; /* Shouldn't happen */
+	dev_kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+
+
+static inline void hdlc_proto_start(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+	if (hdlc->proto->start)
+		return hdlc->proto->start(dev);
+}
+
+
+
+static inline void hdlc_proto_stop(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+	if (hdlc->proto->stop)
+		return hdlc->proto->stop(dev);
+}
+
+
+
+static int hdlc_device_event(struct notifier_block *this, unsigned long event,
+			     void *ptr)
+{
+	struct net_device *dev = ptr;
+	hdlc_device *hdlc;
+	unsigned long flags;
+	int on;
+ 
+	if (dev->get_stats != hdlc_get_stats)
+		return NOTIFY_DONE; /* not an HDLC device */
+ 
+	if (event != NETDEV_CHANGE)
+		return NOTIFY_DONE; /* Only interrested in carrier changes */
+
+	on = netif_carrier_ok(dev);
+
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_device_event NETDEV_CHANGE, carrier %i\n",
+	       dev->name, on);
+#endif
+
+	hdlc = dev_to_hdlc(dev);
+	spin_lock_irqsave(&hdlc->state_lock, flags);
+
+	if (hdlc->carrier == on)
+		goto carrier_exit; /* no change in DCD line level */
+
+	hdlc->carrier = on;
+
+	if (!hdlc->open)
+		goto carrier_exit;
+
+	if (hdlc->carrier) {
+		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
+		hdlc_proto_start(dev);
+	} else {
+		printk(KERN_INFO "%s: Carrier lost\n", dev->name);
+		hdlc_proto_stop(dev);
+	}
+
+carrier_exit:
+	spin_unlock_irqrestore(&hdlc->state_lock, flags);
+	return NOTIFY_DONE;
+}
+
+
+
+/* Must be called by hardware driver when HDLC device is being opened */
+int hdlc_open(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_open() carrier %i open %i\n", dev->name,
+	       hdlc->carrier, hdlc->open);
+#endif
+
+	if (hdlc->proto == NULL)
+		return -ENOSYS;	/* no protocol attached */
+
+	if (hdlc->proto->open) {
+		int result = hdlc->proto->open(dev);
+		if (result)
+			return result;
+	}
+
+	spin_lock_irq(&hdlc->state_lock);
+
+	if (hdlc->carrier) {
+		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
+		hdlc_proto_start(dev);
+	} else
+		printk(KERN_INFO "%s: No carrier\n", dev->name);
+
+	hdlc->open = 1;
+
+	spin_unlock_irq(&hdlc->state_lock);
+	return 0;
+}
+
+
+
+/* Must be called by hardware driver when HDLC device is being closed */
+void hdlc_close(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+#ifdef DEBUG_LINK
+	printk(KERN_DEBUG "%s: hdlc_close() carrier %i open %i\n", dev->name,
+	       hdlc->carrier, hdlc->open);
+#endif
+
+	spin_lock_irq(&hdlc->state_lock);
+
+	hdlc->open = 0;
+	if (hdlc->carrier)
+		hdlc_proto_stop(dev);
+
+	spin_unlock_irq(&hdlc->state_lock);
+
+	if (hdlc->proto->close)
+		hdlc->proto->close(dev);
+}
+
+
+
+int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
+{
+	struct hdlc_proto *proto = first_proto;
+	int result;
+
+	if (cmd != SIOCWANDEV)
+		return -EINVAL;
+
+	if (dev_to_hdlc(dev)->proto) {
+		result = dev_to_hdlc(dev)->proto->ioctl(dev, ifr);
+		if (result != -EINVAL)
+			return result;
+	}
+
+	/* Not handled by currently attached protocol (if any) */
+
+	while (proto) {
+		if ((result = proto->ioctl(dev, ifr)) != -EINVAL)
+			return result;
+		proto = proto->next;
+	}
+	return -EINVAL;
+}
+
+void hdlc_setup(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+
+	dev->get_stats = hdlc_get_stats;
+	dev->change_mtu = hdlc_change_mtu;
+	dev->mtu = HDLC_MAX_MTU;
+
+	dev->type = ARPHRD_RAWHDLC;
+	dev->hard_header_len = 16;
+
+	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
+
+	hdlc->carrier = 1;
+	hdlc->open = 0;
+	spin_lock_init(&hdlc->state_lock);
+}
+
+struct net_device *alloc_hdlcdev(void *priv)
+{
+	struct net_device *dev;
+	dev = alloc_netdev(sizeof(struct hdlc_device_desc) +
+			   sizeof(hdlc_device), "hdlc%d", hdlc_setup);
+	if (dev)
+		dev_to_hdlc(dev)->priv = priv;
+	return dev;
+}
+
+void unregister_hdlc_device(struct net_device *dev)
+{
+	rtnl_lock();
+	unregister_netdevice(dev);
+	detach_hdlc_protocol(dev);
+	rtnl_unlock();
+}
+
+
+
+int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
+			 int (*rx)(struct sk_buff *skb), size_t size)
+{
+	detach_hdlc_protocol(dev);
+
+	if (!try_module_get(proto->module))
+		return -ENOSYS;
+
+	if (size)
+		if ((dev_to_hdlc(dev)->state = kmalloc(size,
+						       GFP_KERNEL)) == NULL) {
+			printk(KERN_WARNING "Memory squeeze on"
+			       " hdlc_proto_attach()\n");
+			module_put(proto->module);
+			return -ENOBUFS;
+		}
+	dev_to_hdlc(dev)->proto = proto;
+	dev_to_desc(dev)->netif_rx = rx;
+	return 0;
+}
+
+
+void detach_hdlc_protocol(struct net_device *dev)
+{
+	hdlc_device *hdlc = dev_to_hdlc(dev);
+
+	if (hdlc->proto) {
+		if (hdlc->proto->detach)
+			hdlc->proto->detach(dev);
+		module_put(hdlc->proto->module);
+		hdlc->proto = NULL;
+	}
+	kfree(hdlc->state);
+	hdlc->state = NULL;
+}
+
+
+void register_hdlc_protocol(struct hdlc_proto *proto)
+{
+	proto->next = first_proto;
+	first_proto = proto;
+}
+
+
+void unregister_hdlc_protocol(struct hdlc_proto *proto)
+{
+	struct hdlc_proto **p = &first_proto;
+	while (*p) {
+		if (*p == proto) {
+			*p = proto->next;
+			return;
+		}
+		p = &((*p)->next);
+	}
+}
+
+
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("HDLC support module");
+MODULE_LICENSE("GPL v2");
+
+EXPORT_SYMBOL(hdlc_open);
+EXPORT_SYMBOL(hdlc_close);
+EXPORT_SYMBOL(hdlc_ioctl);
+EXPORT_SYMBOL(hdlc_setup);
+EXPORT_SYMBOL(alloc_hdlcdev);
+EXPORT_SYMBOL(unregister_hdlc_device);
+EXPORT_SYMBOL(register_hdlc_protocol);
+EXPORT_SYMBOL(unregister_hdlc_protocol);
+EXPORT_SYMBOL(attach_hdlc_protocol);
+EXPORT_SYMBOL(detach_hdlc_protocol);
+
+static struct packet_type hdlc_packet_type = {
+	.type = __constant_htons(ETH_P_HDLC),
+	.func = hdlc_rcv,
+};
+
+
+static struct notifier_block hdlc_notifier = {
+        .notifier_call = hdlc_device_event,
+};
+
+
+static int __init hdlc_module_init(void)
+{
+	int result;
+
+	printk(KERN_INFO "%s\n", version);
+	if ((result = register_netdevice_notifier(&hdlc_notifier)) != 0)
+                return result;
+        dev_add_pack(&hdlc_packet_type);
+	return 0;
+}
+
+
+
+static void __exit hdlc_module_exit(void)
+{
+	dev_remove_pack(&hdlc_packet_type);
+	unregister_netdevice_notifier(&hdlc_notifier);
+}
+
+
+module_init(hdlc_module_init);
+module_exit(hdlc_module_exit);
diff --git a/drivers/net/wan/hdlc_cisco.c b/drivers/net/wan/hdlc_cisco.c
index f289daba0c7b..7ec2b2f9b7ee 100644
--- a/drivers/net/wan/hdlc_cisco.c
+++ b/drivers/net/wan/hdlc_cisco.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Cisco HDLC support
  *
- * Copyright (C) 2000 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 2000 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -34,17 +34,56 @@
 #define CISCO_KEEPALIVE_REQ	2	/* Cisco keepalive request */
 
 
+struct hdlc_header {
+	u8 address;
+	u8 control;
+	u16 protocol;
+}__attribute__ ((packed));
+
+
+struct cisco_packet {
+	u32 type;		/* code */
+	u32 par1;
+	u32 par2;
+	u16 rel;		/* reliability */
+	u32 time;
+}__attribute__ ((packed));
+#define	CISCO_PACKET_LEN	18
+#define	CISCO_BIG_PACKET_LEN	20
+
+
+struct cisco_state {
+	cisco_proto settings;
+
+	struct timer_list timer;
+	unsigned long last_poll;
+	int up;
+	int request_sent;
+	u32 txseq; /* TX sequence number */
+	u32 rxseq; /* RX sequence number */
+};
+
+
+static int cisco_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
+static inline struct cisco_state * state(hdlc_device *hdlc)
+{
+	return(struct cisco_state *)(hdlc->state);
+}
+
+
 static int cisco_hard_header(struct sk_buff *skb, struct net_device *dev,
 			     u16 type, void *daddr, void *saddr,
 			     unsigned int len)
 {
-	hdlc_header *data;
+	struct hdlc_header *data;
 #ifdef DEBUG_HARD_HEADER
 	printk(KERN_DEBUG "%s: cisco_hard_header called\n", dev->name);
 #endif
 
-	skb_push(skb, sizeof(hdlc_header));
-	data = (hdlc_header*)skb->data;
+	skb_push(skb, sizeof(struct hdlc_header));
+	data = (struct hdlc_header*)skb->data;
 	if (type == CISCO_KEEPALIVE)
 		data->address = CISCO_MULTICAST;
 	else
@@ -52,7 +91,7 @@ static int cisco_hard_header(struct sk_buff *skb, struct net_device *dev,
 	data->control = 0;
 	data->protocol = htons(type);
 
-	return sizeof(hdlc_header);
+	return sizeof(struct hdlc_header);
 }
 
 
@@ -61,9 +100,10 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 				 u32 par1, u32 par2)
 {
 	struct sk_buff *skb;
-	cisco_packet *data;
+	struct cisco_packet *data;
 
-	skb = dev_alloc_skb(sizeof(hdlc_header) + sizeof(cisco_packet));
+	skb = dev_alloc_skb(sizeof(struct hdlc_header) +
+			    sizeof(struct cisco_packet));
 	if (!skb) {
 		printk(KERN_WARNING
 		       "%s: Memory squeeze on cisco_keepalive_send()\n",
@@ -72,7 +112,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 	}
 	skb_reserve(skb, 4);
 	cisco_hard_header(skb, dev, CISCO_KEEPALIVE, NULL, NULL, 0);
-	data = (cisco_packet*)(skb->data + 4);
+	data = (struct cisco_packet*)(skb->data + 4);
 
 	data->type = htonl(type);
 	data->par1 = htonl(par1);
@@ -81,7 +121,7 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 	/* we will need do_div here if 1000 % HZ != 0 */
 	data->time = htonl((jiffies - INITIAL_JIFFIES) * (1000 / HZ));
 
-	skb_put(skb, sizeof(cisco_packet));
+	skb_put(skb, sizeof(struct cisco_packet));
 	skb->priority = TC_PRIO_CONTROL;
 	skb->dev = dev;
 	skb->nh.raw = skb->data;
@@ -93,9 +133,9 @@ static void cisco_keepalive_send(struct net_device *dev, u32 type,
 
 static __be16 cisco_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
-	hdlc_header *data = (hdlc_header*)skb->data;
+	struct hdlc_header *data = (struct hdlc_header*)skb->data;
 
-	if (skb->len < sizeof(hdlc_header))
+	if (skb->len < sizeof(struct hdlc_header))
 		return __constant_htons(ETH_P_HDLC);
 
 	if (data->address != CISCO_MULTICAST &&
@@ -106,7 +146,7 @@ static __be16 cisco_type_trans(struct sk_buff *skb, struct net_device *dev)
 	case __constant_htons(ETH_P_IP):
 	case __constant_htons(ETH_P_IPX):
 	case __constant_htons(ETH_P_IPV6):
-		skb_pull(skb, sizeof(hdlc_header));
+		skb_pull(skb, sizeof(struct hdlc_header));
 		return data->protocol;
 	default:
 		return __constant_htons(ETH_P_HDLC);
@@ -118,12 +158,12 @@ static int cisco_rx(struct sk_buff *skb)
 {
 	struct net_device *dev = skb->dev;
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	hdlc_header *data = (hdlc_header*)skb->data;
-	cisco_packet *cisco_data;
+	struct hdlc_header *data = (struct hdlc_header*)skb->data;
+	struct cisco_packet *cisco_data;
 	struct in_device *in_dev;
 	u32 addr, mask;
 
-	if (skb->len < sizeof(hdlc_header))
+	if (skb->len < sizeof(struct hdlc_header))
 		goto rx_error;
 
 	if (data->address != CISCO_MULTICAST &&
@@ -137,15 +177,17 @@ static int cisco_rx(struct sk_buff *skb)
 		return NET_RX_SUCCESS;
 
 	case CISCO_KEEPALIVE:
-		if (skb->len != sizeof(hdlc_header) + CISCO_PACKET_LEN &&
-		    skb->len != sizeof(hdlc_header) + CISCO_BIG_PACKET_LEN) {
-			printk(KERN_INFO "%s: Invalid length of Cisco "
-			       "control packet (%d bytes)\n",
-			       dev->name, skb->len);
+		if ((skb->len != sizeof(struct hdlc_header) +
+		     CISCO_PACKET_LEN) &&
+		    (skb->len != sizeof(struct hdlc_header) +
+		     CISCO_BIG_PACKET_LEN)) {
+			printk(KERN_INFO "%s: Invalid length of Cisco control"
+			       " packet (%d bytes)\n", dev->name, skb->len);
 			goto rx_error;
 		}
 
-		cisco_data = (cisco_packet*)(skb->data + sizeof(hdlc_header));
+		cisco_data = (struct cisco_packet*)(skb->data + sizeof
+						    (struct hdlc_header));
 
 		switch(ntohl (cisco_data->type)) {
 		case CISCO_ADDR_REQ: /* Stolen from syncppp.c :-) */
@@ -178,11 +220,11 @@ static int cisco_rx(struct sk_buff *skb)
 			goto rx_error;
 
 		case CISCO_KEEPALIVE_REQ:
-			hdlc->state.cisco.rxseq = ntohl(cisco_data->par1);
-			if (hdlc->state.cisco.request_sent &&
-			    ntohl(cisco_data->par2)==hdlc->state.cisco.txseq) {
-				hdlc->state.cisco.last_poll = jiffies;
-				if (!hdlc->state.cisco.up) {
+			state(hdlc)->rxseq = ntohl(cisco_data->par1);
+			if (state(hdlc)->request_sent &&
+			    ntohl(cisco_data->par2) == state(hdlc)->txseq) {
+				state(hdlc)->last_poll = jiffies;
+				if (!state(hdlc)->up) {
 					u32 sec, min, hrs, days;
 					sec = ntohl(cisco_data->time) / 1000;
 					min = sec / 60; sec -= min * 60;
@@ -193,7 +235,7 @@ static int cisco_rx(struct sk_buff *skb)
 					       dev->name, days, hrs,
 					       min, sec);
 					netif_dormant_off(dev);
-					hdlc->state.cisco.up = 1;
+					state(hdlc)->up = 1;
 				}
 			}
 
@@ -208,7 +250,7 @@ static int cisco_rx(struct sk_buff *skb)
 	return NET_RX_DROP;
 
  rx_error:
-	hdlc->stats.rx_errors++; /* Mark error */
+	dev_to_desc(dev)->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -220,23 +262,22 @@ static void cisco_timer(unsigned long arg)
 	struct net_device *dev = (struct net_device *)arg;
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 
-	if (hdlc->state.cisco.up &&
-	    time_after(jiffies, hdlc->state.cisco.last_poll +
-		       hdlc->state.cisco.settings.timeout * HZ)) {
-		hdlc->state.cisco.up = 0;
+	if (state(hdlc)->up &&
+	    time_after(jiffies, state(hdlc)->last_poll +
+		       state(hdlc)->settings.timeout * HZ)) {
+		state(hdlc)->up = 0;
 		printk(KERN_INFO "%s: Link down\n", dev->name);
 		netif_dormant_on(dev);
 	}
 
-	cisco_keepalive_send(dev, CISCO_KEEPALIVE_REQ,
-			     ++hdlc->state.cisco.txseq,
-			     hdlc->state.cisco.rxseq);
-	hdlc->state.cisco.request_sent = 1;
-	hdlc->state.cisco.timer.expires = jiffies +
-		hdlc->state.cisco.settings.interval * HZ;
-	hdlc->state.cisco.timer.function = cisco_timer;
-	hdlc->state.cisco.timer.data = arg;
-	add_timer(&hdlc->state.cisco.timer);
+	cisco_keepalive_send(dev, CISCO_KEEPALIVE_REQ, ++state(hdlc)->txseq,
+			     state(hdlc)->rxseq);
+	state(hdlc)->request_sent = 1;
+	state(hdlc)->timer.expires = jiffies +
+		state(hdlc)->settings.interval * HZ;
+	state(hdlc)->timer.function = cisco_timer;
+	state(hdlc)->timer.data = arg;
+	add_timer(&state(hdlc)->timer);
 }
 
 
@@ -244,15 +285,15 @@ static void cisco_timer(unsigned long arg)
 static void cisco_start(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	hdlc->state.cisco.up = 0;
-	hdlc->state.cisco.request_sent = 0;
-	hdlc->state.cisco.txseq = hdlc->state.cisco.rxseq = 0;
-
-	init_timer(&hdlc->state.cisco.timer);
-	hdlc->state.cisco.timer.expires = jiffies + HZ; /*First poll after 1s*/
-	hdlc->state.cisco.timer.function = cisco_timer;
-	hdlc->state.cisco.timer.data = (unsigned long)dev;
-	add_timer(&hdlc->state.cisco.timer);
+	state(hdlc)->up = 0;
+	state(hdlc)->request_sent = 0;
+	state(hdlc)->txseq = state(hdlc)->rxseq = 0;
+
+	init_timer(&state(hdlc)->timer);
+	state(hdlc)->timer.expires = jiffies + HZ; /*First poll after 1s*/
+	state(hdlc)->timer.function = cisco_timer;
+	state(hdlc)->timer.data = (unsigned long)dev;
+	add_timer(&state(hdlc)->timer);
 }
 
 
@@ -260,15 +301,24 @@ static void cisco_start(struct net_device *dev)
 static void cisco_stop(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	del_timer_sync(&hdlc->state.cisco.timer);
+	del_timer_sync(&state(hdlc)->timer);
 	netif_dormant_on(dev);
-	hdlc->state.cisco.up = 0;
-	hdlc->state.cisco.request_sent = 0;
+	state(hdlc)->up = 0;
+	state(hdlc)->request_sent = 0;
 }
 
 
 
-int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.start		= cisco_start,
+	.stop		= cisco_stop,
+	.type_trans	= cisco_type_trans,
+	.ioctl		= cisco_ioctl,
+	.module		= THIS_MODULE,
+};
+ 
+ 
+static int cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	cisco_proto __user *cisco_s = ifr->ifr_settings.ifs_ifsu.cisco;
 	const size_t size = sizeof(cisco_proto);
@@ -278,12 +328,14 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_CISCO;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(cisco_s, &hdlc->state.cisco.settings, size))
+		if (copy_to_user(cisco_s, &state(hdlc)->settings, size))
 			return -EFAULT;
 		return 0;
 
@@ -302,19 +354,15 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 			return -EINVAL;
 
 		result=hdlc->attach(dev, ENCODING_NRZ,PARITY_CRC16_PR1_CCITT);
-
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.cisco.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
+		result = attach_hdlc_protocol(dev, &proto, cisco_rx,
+					      sizeof(struct cisco_state));
+		if (result)
+			return result;
 
-		hdlc->proto.start = cisco_start;
-		hdlc->proto.stop = cisco_stop;
-		hdlc->proto.netif_rx = cisco_rx;
-		hdlc->proto.type_trans = cisco_type_trans;
-		hdlc->proto.id = IF_PROTO_CISCO;
+		memcpy(&state(hdlc)->settings, &new_settings, size);
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = cisco_hard_header;
 		dev->hard_header_cache = NULL;
@@ -327,3 +375,25 @@ int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Cisco HDLC protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_fr.c b/drivers/net/wan/hdlc_fr.c
index 7bb737bbdeb9..b45ab680d2d6 100644
--- a/drivers/net/wan/hdlc_fr.c
+++ b/drivers/net/wan/hdlc_fr.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Frame Relay support
  *
- * Copyright (C) 1999 - 2005 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -52,6 +52,8 @@
 #undef DEBUG_PKT
 #undef DEBUG_ECN
 #undef DEBUG_LINK
+#undef DEBUG_PROTO
+#undef DEBUG_PVC
 
 #define FR_UI			0x03
 #define FR_PAD			0x00
@@ -115,13 +117,53 @@ typedef struct {
 }__attribute__ ((packed)) fr_hdr;
 
 
+typedef struct pvc_device_struct {
+	struct net_device *frad;
+	struct net_device *main;
+	struct net_device *ether;	/* bridged Ethernet interface	*/
+	struct pvc_device_struct *next;	/* Sorted in ascending DLCI order */
+	int dlci;
+	int open_count;
+
+	struct {
+		unsigned int new: 1;
+		unsigned int active: 1;
+		unsigned int exist: 1;
+		unsigned int deleted: 1;
+		unsigned int fecn: 1;
+		unsigned int becn: 1;
+		unsigned int bandwidth;	/* Cisco LMI reporting only */
+	}state;
+}pvc_device;
+
+
+struct frad_state {
+	fr_proto settings;
+	pvc_device *first_pvc;
+	int dce_pvc_count;
+
+	struct timer_list timer;
+	unsigned long last_poll;
+	int reliable;
+	int dce_changed;
+	int request;
+	int fullrep_sent;
+	u32 last_errors; /* last errors bit list */
+	u8 n391cnt;
+	u8 txseq; /* TX sequence number */
+	u8 rxseq; /* RX sequence number */
+};
+
+
+static int fr_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
 static inline u16 q922_to_dlci(u8 *hdr)
 {
 	return ((hdr[0] & 0xFC) << 2) | ((hdr[1] & 0xF0) >> 4);
 }
 
 
-
 static inline void dlci_to_q922(u8 *hdr, u16 dlci)
 {
 	hdr[0] = (dlci >> 2) & 0xFC;
@@ -129,10 +171,21 @@ static inline void dlci_to_q922(u8 *hdr, u16 dlci)
 }
 
 
+static inline struct frad_state * state(hdlc_device *hdlc)
+{
+	return(struct frad_state *)(hdlc->state);
+}
+
+
+static __inline__ pvc_device* dev_to_pvc(struct net_device *dev)
+{
+	return dev->priv;
+}
+
 
 static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 {
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		if (pvc->dlci == dlci)
@@ -146,10 +199,10 @@ static inline pvc_device* find_pvc(hdlc_device *hdlc, u16 dlci)
 }
 
 
-static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
+static pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc, **pvc_p = &hdlc->state.fr.first_pvc;
+	pvc_device *pvc, **pvc_p = &state(hdlc)->first_pvc;
 
 	while (*pvc_p) {
 		if ((*pvc_p)->dlci == dlci)
@@ -160,12 +213,15 @@ static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 	}
 
 	pvc = kmalloc(sizeof(pvc_device), GFP_ATOMIC);
+#ifdef DEBUG_PVC
+	printk(KERN_DEBUG "add_pvc: allocated pvc %p, frad %p\n", pvc, dev);
+#endif
 	if (!pvc)
 		return NULL;
 
 	memset(pvc, 0, sizeof(pvc_device));
 	pvc->dlci = dlci;
-	pvc->master = dev;
+	pvc->frad = dev;
 	pvc->next = *pvc_p;	/* Put it in the chain */
 	*pvc_p = pvc;
 	return pvc;
@@ -174,7 +230,7 @@ static inline pvc_device* add_pvc(struct net_device *dev, u16 dlci)
 
 static inline int pvc_is_used(pvc_device *pvc)
 {
-	return pvc->main != NULL || pvc->ether != NULL;
+	return pvc->main || pvc->ether;
 }
 
 
@@ -200,11 +256,14 @@ static inline void pvc_carrier(int on, pvc_device *pvc)
 
 static inline void delete_unused_pvcs(hdlc_device *hdlc)
 {
-	pvc_device **pvc_p = &hdlc->state.fr.first_pvc;
+	pvc_device **pvc_p = &state(hdlc)->first_pvc;
 
 	while (*pvc_p) {
 		if (!pvc_is_used(*pvc_p)) {
 			pvc_device *pvc = *pvc_p;
+#ifdef DEBUG_PVC
+			printk(KERN_DEBUG "freeing unused pvc: %p\n", pvc);
+#endif
 			*pvc_p = pvc->next;
 			kfree(pvc);
 			continue;
@@ -295,16 +354,16 @@ static int pvc_open(struct net_device *dev)
 {
 	pvc_device *pvc = dev_to_pvc(dev);
 
-	if ((pvc->master->flags & IFF_UP) == 0)
-		return -EIO;  /* Master must be UP in order to activate PVC */
+	if ((pvc->frad->flags & IFF_UP) == 0)
+		return -EIO;  /* Frad must be UP in order to activate PVC */
 
 	if (pvc->open_count++ == 0) {
-		hdlc_device *hdlc = dev_to_hdlc(pvc->master);
-		if (hdlc->state.fr.settings.lmi == LMI_NONE)
-			pvc->state.active = netif_carrier_ok(pvc->master);
+		hdlc_device *hdlc = dev_to_hdlc(pvc->frad);
+		if (state(hdlc)->settings.lmi == LMI_NONE)
+			pvc->state.active = netif_carrier_ok(pvc->frad);
 
 		pvc_carrier(pvc->state.active, pvc);
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->dce_changed = 1;
 	}
 	return 0;
 }
@@ -316,12 +375,12 @@ static int pvc_close(struct net_device *dev)
 	pvc_device *pvc = dev_to_pvc(dev);
 
 	if (--pvc->open_count == 0) {
-		hdlc_device *hdlc = dev_to_hdlc(pvc->master);
-		if (hdlc->state.fr.settings.lmi == LMI_NONE)
+		hdlc_device *hdlc = dev_to_hdlc(pvc->frad);
+		if (state(hdlc)->settings.lmi == LMI_NONE)
 			pvc->state.active = 0;
 
-		if (hdlc->state.fr.settings.dce) {
-			hdlc->state.fr.dce_changed = 1;
+		if (state(hdlc)->settings.dce) {
+			state(hdlc)->dce_changed = 1;
 			pvc->state.active = 0;
 		}
 	}
@@ -348,7 +407,7 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		}
 
 		info.dlci = pvc->dlci;
-		memcpy(info.master, pvc->master->name, IFNAMSIZ);
+		memcpy(info.master, pvc->frad->name, IFNAMSIZ);
 		if (copy_to_user(ifr->ifr_settings.ifs_ifsu.fr_pvc_info,
 				 &info, sizeof(info)))
 			return -EFAULT;
@@ -361,7 +420,7 @@ static int pvc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 
 static inline struct net_device_stats *pvc_get_stats(struct net_device *dev)
 {
-	return netdev_priv(dev);
+	return &dev_to_desc(dev)->stats;
 }
 
 
@@ -393,7 +452,7 @@ static int pvc_xmit(struct sk_buff *skb, struct net_device *dev)
 			stats->tx_packets++;
 			if (pvc->state.fecn) /* TX Congestion counter */
 				stats->tx_compressed++;
-			skb->dev = pvc->master;
+			skb->dev = pvc->frad;
 			dev_queue_xmit(skb);
 			return 0;
 		}
@@ -419,7 +478,7 @@ static int pvc_change_mtu(struct net_device *dev, int new_mtu)
 static inline void fr_log_dlci_active(pvc_device *pvc)
 {
 	printk(KERN_INFO "%s: DLCI %d [%s%s%s]%s %s\n",
-	       pvc->master->name,
+	       pvc->frad->name,
 	       pvc->dlci,
 	       pvc->main ? pvc->main->name : "",
 	       pvc->main && pvc->ether ? " " : "",
@@ -438,21 +497,20 @@ static inline u8 fr_lmi_nextseq(u8 x)
 }
 
 
-
 static void fr_lmi_send(struct net_device *dev, int fullrep)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	struct sk_buff *skb;
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
-	int lmi = hdlc->state.fr.settings.lmi;
-	int dce = hdlc->state.fr.settings.dce;
+	pvc_device *pvc = state(hdlc)->first_pvc;
+	int lmi = state(hdlc)->settings.lmi;
+	int dce = state(hdlc)->settings.dce;
 	int len = lmi == LMI_ANSI ? LMI_ANSI_LENGTH : LMI_CCITT_CISCO_LENGTH;
 	int stat_len = (lmi == LMI_CISCO) ? 6 : 3;
 	u8 *data;
 	int i = 0;
 
 	if (dce && fullrep) {
-		len += hdlc->state.fr.dce_pvc_count * (2 + stat_len);
+		len += state(hdlc)->dce_pvc_count * (2 + stat_len);
 		if (len > HDLC_MAX_MRU) {
 			printk(KERN_WARNING "%s: Too many PVCs while sending "
 			       "LMI full report\n", dev->name);
@@ -486,8 +544,9 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 	data[i++] = fullrep ? LMI_FULLREP : LMI_INTEGRITY;
 	data[i++] = lmi == LMI_CCITT ? LMI_CCITT_ALIVE : LMI_ANSI_CISCO_ALIVE;
 	data[i++] = LMI_INTEG_LEN;
-	data[i++] = hdlc->state.fr.txseq =fr_lmi_nextseq(hdlc->state.fr.txseq);
-	data[i++] = hdlc->state.fr.rxseq;
+	data[i++] = state(hdlc)->txseq =
+		fr_lmi_nextseq(state(hdlc)->txseq);
+	data[i++] = state(hdlc)->rxseq;
 
 	if (dce && fullrep) {
 		while (pvc) {
@@ -496,7 +555,7 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 			data[i++] = stat_len;
 
 			/* LMI start/restart */
-			if (hdlc->state.fr.reliable && !pvc->state.exist) {
+			if (state(hdlc)->reliable && !pvc->state.exist) {
 				pvc->state.exist = pvc->state.new = 1;
 				fr_log_dlci_active(pvc);
 			}
@@ -541,15 +600,15 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)
 static void fr_set_link_state(int reliable, struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
-	hdlc->state.fr.reliable = reliable;
+	state(hdlc)->reliable = reliable;
 	if (reliable) {
 		netif_dormant_off(dev);
-		hdlc->state.fr.n391cnt = 0; /* Request full status */
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->n391cnt = 0; /* Request full status */
+		state(hdlc)->dce_changed = 1;
 
-		if (hdlc->state.fr.settings.lmi == LMI_NONE) {
+		if (state(hdlc)->settings.lmi == LMI_NONE) {
 			while (pvc) {	/* Activate all PVCs */
 				pvc_carrier(1, pvc);
 				pvc->state.exist = pvc->state.active = 1;
@@ -563,7 +622,7 @@ static void fr_set_link_state(int reliable, struct net_device *dev)
 			pvc_carrier(0, pvc);
 			pvc->state.exist = pvc->state.active = 0;
 			pvc->state.new = 0;
-			if (!hdlc->state.fr.settings.dce)
+			if (!state(hdlc)->settings.dce)
 				pvc->state.bandwidth = 0;
 			pvc = pvc->next;
 		}
@@ -571,7 +630,6 @@ static void fr_set_link_state(int reliable, struct net_device *dev)
 }
 
 
-
 static void fr_timer(unsigned long arg)
 {
 	struct net_device *dev = (struct net_device *)arg;
@@ -579,62 +637,61 @@ static void fr_timer(unsigned long arg)
 	int i, cnt = 0, reliable;
 	u32 list;
 
-	if (hdlc->state.fr.settings.dce) {
-		reliable = hdlc->state.fr.request &&
-			time_before(jiffies, hdlc->state.fr.last_poll +
-				    hdlc->state.fr.settings.t392 * HZ);
-		hdlc->state.fr.request = 0;
+	if (state(hdlc)->settings.dce) {
+		reliable = state(hdlc)->request &&
+			time_before(jiffies, state(hdlc)->last_poll +
+				    state(hdlc)->settings.t392 * HZ);
+		state(hdlc)->request = 0;
 	} else {
-		hdlc->state.fr.last_errors <<= 1; /* Shift the list */
-		if (hdlc->state.fr.request) {
-			if (hdlc->state.fr.reliable)
+		state(hdlc)->last_errors <<= 1; /* Shift the list */
+		if (state(hdlc)->request) {
+			if (state(hdlc)->reliable)
 				printk(KERN_INFO "%s: No LMI status reply "
 				       "received\n", dev->name);
-			hdlc->state.fr.last_errors |= 1;
+			state(hdlc)->last_errors |= 1;
 		}
 
-		list = hdlc->state.fr.last_errors;
-		for (i = 0; i < hdlc->state.fr.settings.n393; i++, list >>= 1)
+		list = state(hdlc)->last_errors;
+		for (i = 0; i < state(hdlc)->settings.n393; i++, list >>= 1)
 			cnt += (list & 1);	/* errors count */
 
-		reliable = (cnt < hdlc->state.fr.settings.n392);
+		reliable = (cnt < state(hdlc)->settings.n392);
 	}
 
-	if (hdlc->state.fr.reliable != reliable) {
+	if (state(hdlc)->reliable != reliable) {
 		printk(KERN_INFO "%s: Link %sreliable\n", dev->name,
 		       reliable ? "" : "un");
 		fr_set_link_state(reliable, dev);
 	}
 
-	if (hdlc->state.fr.settings.dce)
-		hdlc->state.fr.timer.expires = jiffies +
-			hdlc->state.fr.settings.t392 * HZ;
+	if (state(hdlc)->settings.dce)
+		state(hdlc)->timer.expires = jiffies +
+			state(hdlc)->settings.t392 * HZ;
 	else {
-		if (hdlc->state.fr.n391cnt)
-			hdlc->state.fr.n391cnt--;
+		if (state(hdlc)->n391cnt)
+			state(hdlc)->n391cnt--;
 
-		fr_lmi_send(dev, hdlc->state.fr.n391cnt == 0);
+		fr_lmi_send(dev, state(hdlc)->n391cnt == 0);
 
-		hdlc->state.fr.last_poll = jiffies;
-		hdlc->state.fr.request = 1;
-		hdlc->state.fr.timer.expires = jiffies +
-			hdlc->state.fr.settings.t391 * HZ;
+		state(hdlc)->last_poll = jiffies;
+		state(hdlc)->request = 1;
+		state(hdlc)->timer.expires = jiffies +
+			state(hdlc)->settings.t391 * HZ;
 	}
 
-	hdlc->state.fr.timer.function = fr_timer;
-	hdlc->state.fr.timer.data = arg;
-	add_timer(&hdlc->state.fr.timer);
+	state(hdlc)->timer.function = fr_timer;
+	state(hdlc)->timer.data = arg;
+	add_timer(&state(hdlc)->timer);
 }
 
 
-
 static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	pvc_device *pvc;
 	u8 rxseq, txseq;
-	int lmi = hdlc->state.fr.settings.lmi;
-	int dce = hdlc->state.fr.settings.dce;
+	int lmi = state(hdlc)->settings.lmi;
+	int dce = state(hdlc)->settings.dce;
 	int stat_len = (lmi == LMI_CISCO) ? 6 : 3, reptype, error, no_ram, i;
 
 	if (skb->len < (lmi == LMI_ANSI ? LMI_ANSI_LENGTH :
@@ -645,8 +702,8 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 
 	if (skb->data[3] != (lmi == LMI_CISCO ? NLPID_CISCO_LMI :
 			     NLPID_CCITT_ANSI_LMI)) {
-		printk(KERN_INFO "%s: Received non-LMI frame with LMI"
-		       " DLCI\n", dev->name);
+		printk(KERN_INFO "%s: Received non-LMI frame with LMI DLCI\n",
+		       dev->name);
 		return 1;
 	}
 
@@ -706,53 +763,53 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	}
 	i++;
 
-	hdlc->state.fr.rxseq = skb->data[i++]; /* TX sequence from peer */
+	state(hdlc)->rxseq = skb->data[i++]; /* TX sequence from peer */
 	rxseq = skb->data[i++];	/* Should confirm our sequence */
 
-	txseq = hdlc->state.fr.txseq;
+	txseq = state(hdlc)->txseq;
 
 	if (dce)
-		hdlc->state.fr.last_poll = jiffies;
+		state(hdlc)->last_poll = jiffies;
 
 	error = 0;
-	if (!hdlc->state.fr.reliable)
+	if (!state(hdlc)->reliable)
 		error = 1;
 
-	if (rxseq == 0 || rxseq != txseq) {
-		hdlc->state.fr.n391cnt = 0; /* Ask for full report next time */
+	if (rxseq == 0 || rxseq != txseq) { /* Ask for full report next time */
+		state(hdlc)->n391cnt = 0;
 		error = 1;
 	}
 
 	if (dce) {
-		if (hdlc->state.fr.fullrep_sent && !error) {
+		if (state(hdlc)->fullrep_sent && !error) {
 /* Stop sending full report - the last one has been confirmed by DTE */
-			hdlc->state.fr.fullrep_sent = 0;
-			pvc = hdlc->state.fr.first_pvc;
+			state(hdlc)->fullrep_sent = 0;
+			pvc = state(hdlc)->first_pvc;
 			while (pvc) {
 				if (pvc->state.new) {
 					pvc->state.new = 0;
 
 /* Tell DTE that new PVC is now active */
-					hdlc->state.fr.dce_changed = 1;
+					state(hdlc)->dce_changed = 1;
 				}
 				pvc = pvc->next;
 			}
 		}
 
-		if (hdlc->state.fr.dce_changed) {
+		if (state(hdlc)->dce_changed) {
 			reptype = LMI_FULLREP;
-			hdlc->state.fr.fullrep_sent = 1;
-			hdlc->state.fr.dce_changed = 0;
+			state(hdlc)->fullrep_sent = 1;
+			state(hdlc)->dce_changed = 0;
 		}
 
-		hdlc->state.fr.request = 1; /* got request */
+		state(hdlc)->request = 1; /* got request */
 		fr_lmi_send(dev, reptype == LMI_FULLREP ? 1 : 0);
 		return 0;
 	}
 
 	/* DTE */
 
-	hdlc->state.fr.request = 0; /* got response, no request pending */
+	state(hdlc)->request = 0; /* got response, no request pending */
 
 	if (error)
 		return 0;
@@ -760,7 +817,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	if (reptype != LMI_FULLREP)
 		return 0;
 
-	pvc = hdlc->state.fr.first_pvc;
+	pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		pvc->state.deleted = 1;
@@ -827,7 +884,7 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 		i += stat_len;
 	}
 
-	pvc = hdlc->state.fr.first_pvc;
+	pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {
 		if (pvc->state.deleted && pvc->state.exist) {
@@ -841,17 +898,16 @@ static int fr_lmi_recv(struct net_device *dev, struct sk_buff *skb)
 	}
 
 	/* Next full report after N391 polls */
-	hdlc->state.fr.n391cnt = hdlc->state.fr.settings.n391;
+	state(hdlc)->n391cnt = state(hdlc)->settings.n391;
 
 	return 0;
 }
 
 
-
 static int fr_rx(struct sk_buff *skb)
 {
-	struct net_device *ndev = skb->dev;
-	hdlc_device *hdlc = dev_to_hdlc(ndev);
+	struct net_device *frad = skb->dev;
+	hdlc_device *hdlc = dev_to_hdlc(frad);
 	fr_hdr *fh = (fr_hdr*)skb->data;
 	u8 *data = skb->data;
 	u16 dlci;
@@ -864,11 +920,11 @@ static int fr_rx(struct sk_buff *skb)
 	dlci = q922_to_dlci(skb->data);
 
 	if ((dlci == LMI_CCITT_ANSI_DLCI &&
-	     (hdlc->state.fr.settings.lmi == LMI_ANSI ||
-	      hdlc->state.fr.settings.lmi == LMI_CCITT)) ||
+	     (state(hdlc)->settings.lmi == LMI_ANSI ||
+	      state(hdlc)->settings.lmi == LMI_CCITT)) ||
 	    (dlci == LMI_CISCO_DLCI &&
-	     hdlc->state.fr.settings.lmi == LMI_CISCO)) {
-		if (fr_lmi_recv(ndev, skb))
+	     state(hdlc)->settings.lmi == LMI_CISCO)) {
+		if (fr_lmi_recv(frad, skb))
 			goto rx_error;
 		dev_kfree_skb_any(skb);
 		return NET_RX_SUCCESS;
@@ -878,7 +934,7 @@ static int fr_rx(struct sk_buff *skb)
 	if (!pvc) {
 #ifdef DEBUG_PKT
 		printk(KERN_INFO "%s: No PVC for received frame's DLCI %d\n",
-		       ndev->name, dlci);
+		       frad->name, dlci);
 #endif
 		dev_kfree_skb_any(skb);
 		return NET_RX_DROP;
@@ -886,7 +942,7 @@ static int fr_rx(struct sk_buff *skb)
 
 	if (pvc->state.fecn != fh->fecn) {
 #ifdef DEBUG_ECN
-		printk(KERN_DEBUG "%s: DLCI %d FECN O%s\n", ndev->name,
+		printk(KERN_DEBUG "%s: DLCI %d FECN O%s\n", frad->name,
 		       dlci, fh->fecn ? "N" : "FF");
 #endif
 		pvc->state.fecn ^= 1;
@@ -894,7 +950,7 @@ static int fr_rx(struct sk_buff *skb)
 
 	if (pvc->state.becn != fh->becn) {
 #ifdef DEBUG_ECN
-		printk(KERN_DEBUG "%s: DLCI %d BECN O%s\n", ndev->name,
+		printk(KERN_DEBUG "%s: DLCI %d BECN O%s\n", frad->name,
 		       dlci, fh->becn ? "N" : "FF");
 #endif
 		pvc->state.becn ^= 1;
@@ -902,7 +958,7 @@ static int fr_rx(struct sk_buff *skb)
 
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		dev_to_desc(frad)->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
@@ -938,13 +994,13 @@ static int fr_rx(struct sk_buff *skb)
 
 		default:
 			printk(KERN_INFO "%s: Unsupported protocol, OUI=%x "
-			       "PID=%x\n", ndev->name, oui, pid);
+			       "PID=%x\n", frad->name, oui, pid);
 			dev_kfree_skb_any(skb);
 			return NET_RX_DROP;
 		}
 	} else {
 		printk(KERN_INFO "%s: Unsupported protocol, NLPID=%x "
-		       "length = %i\n", ndev->name, data[3], skb->len);
+		       "length = %i\n", frad->name, data[3], skb->len);
 		dev_kfree_skb_any(skb);
 		return NET_RX_DROP;
 	}
@@ -964,7 +1020,7 @@ static int fr_rx(struct sk_buff *skb)
 	}
 
  rx_error:
-	hdlc->stats.rx_errors++; /* Mark error */
+	dev_to_desc(frad)->stats.rx_errors++; /* Mark error */
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
@@ -977,44 +1033,42 @@ static void fr_start(struct net_device *dev)
 #ifdef DEBUG_LINK
 	printk(KERN_DEBUG "fr_start\n");
 #endif
-	if (hdlc->state.fr.settings.lmi != LMI_NONE) {
-		hdlc->state.fr.reliable = 0;
-		hdlc->state.fr.dce_changed = 1;
-		hdlc->state.fr.request = 0;
-		hdlc->state.fr.fullrep_sent = 0;
-		hdlc->state.fr.last_errors = 0xFFFFFFFF;
-		hdlc->state.fr.n391cnt = 0;
-		hdlc->state.fr.txseq = hdlc->state.fr.rxseq = 0;
-
-		init_timer(&hdlc->state.fr.timer);
+	if (state(hdlc)->settings.lmi != LMI_NONE) {
+		state(hdlc)->reliable = 0;
+		state(hdlc)->dce_changed = 1;
+		state(hdlc)->request = 0;
+		state(hdlc)->fullrep_sent = 0;
+		state(hdlc)->last_errors = 0xFFFFFFFF;
+		state(hdlc)->n391cnt = 0;
+		state(hdlc)->txseq = state(hdlc)->rxseq = 0;
+
+		init_timer(&state(hdlc)->timer);
 		/* First poll after 1 s */
-		hdlc->state.fr.timer.expires = jiffies + HZ;
-		hdlc->state.fr.timer.function = fr_timer;
-		hdlc->state.fr.timer.data = (unsigned long)dev;
-		add_timer(&hdlc->state.fr.timer);
+		state(hdlc)->timer.expires = jiffies + HZ;
+		state(hdlc)->timer.function = fr_timer;
+		state(hdlc)->timer.data = (unsigned long)dev;
+		add_timer(&state(hdlc)->timer);
 	} else
 		fr_set_link_state(1, dev);
 }
 
 
-
 static void fr_stop(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 #ifdef DEBUG_LINK
 	printk(KERN_DEBUG "fr_stop\n");
 #endif
-	if (hdlc->state.fr.settings.lmi != LMI_NONE)
-		del_timer_sync(&hdlc->state.fr.timer);
+	if (state(hdlc)->settings.lmi != LMI_NONE)
+		del_timer_sync(&state(hdlc)->timer);
 	fr_set_link_state(0, dev);
 }
 
 
-
 static void fr_close(struct net_device *dev)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
-	pvc_device *pvc = hdlc->state.fr.first_pvc;
+	pvc_device *pvc = state(hdlc)->first_pvc;
 
 	while (pvc) {		/* Shutdown all PVCs for this FRAD */
 		if (pvc->main)
@@ -1025,7 +1079,8 @@ static void fr_close(struct net_device *dev)
 	}
 }
 
-static void dlci_setup(struct net_device *dev)
+
+static void pvc_setup(struct net_device *dev)
 {
 	dev->type = ARPHRD_DLCI;
 	dev->flags = IFF_POINTOPOINT;
@@ -1033,9 +1088,9 @@ static void dlci_setup(struct net_device *dev)
 	dev->addr_len = 2;
 }
 
-static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
+static int fr_add_pvc(struct net_device *frad, unsigned int dlci, int type)
 {
-	hdlc_device *hdlc = dev_to_hdlc(master);
+	hdlc_device *hdlc = dev_to_hdlc(frad);
 	pvc_device *pvc = NULL;
 	struct net_device *dev;
 	int result, used;
@@ -1044,9 +1099,9 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 	if (type == ARPHRD_ETHER)
 		prefix = "pvceth%d";
 
-	if ((pvc = add_pvc(master, dlci)) == NULL) {
+	if ((pvc = add_pvc(frad, dlci)) == NULL) {
 		printk(KERN_WARNING "%s: Memory squeeze on fr_add_pvc()\n",
-		       master->name);
+		       frad->name);
 		return -ENOBUFS;
 	}
 
@@ -1060,11 +1115,11 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 				   "pvceth%d", ether_setup);
 	else
 		dev = alloc_netdev(sizeof(struct net_device_stats),
-				   "pvc%d", dlci_setup);
+				   "pvc%d", pvc_setup);
 
 	if (!dev) {
 		printk(KERN_WARNING "%s: Memory squeeze on fr_pvc()\n",
-		       master->name);
+		       frad->name);
 		delete_unused_pvcs(hdlc);
 		return -ENOBUFS;
 	}
@@ -1102,8 +1157,8 @@ static int fr_add_pvc(struct net_device *master, unsigned int dlci, int type)
 	dev->destructor = free_netdev;
 	*get_dev_p(pvc, type) = dev;
 	if (!used) {
-		hdlc->state.fr.dce_changed = 1;
-		hdlc->state.fr.dce_pvc_count++;
+		state(hdlc)->dce_changed = 1;
+		state(hdlc)->dce_pvc_count++;
 	}
 	return 0;
 }
@@ -1128,8 +1183,8 @@ static int fr_del_pvc(hdlc_device *hdlc, unsigned int dlci, int type)
 	*get_dev_p(pvc, type) = NULL;
 
 	if (!pvc_is_used(pvc)) {
-		hdlc->state.fr.dce_pvc_count--;
-		hdlc->state.fr.dce_changed = 1;
+		state(hdlc)->dce_pvc_count--;
+		state(hdlc)->dce_changed = 1;
 	}
 	delete_unused_pvcs(hdlc);
 	return 0;
@@ -1137,14 +1192,13 @@ static int fr_del_pvc(hdlc_device *hdlc, unsigned int dlci, int type)
 
 
 
-static void fr_destroy(hdlc_device *hdlc)
+static void fr_destroy(struct net_device *frad)
 {
-	pvc_device *pvc;
-
-	pvc = hdlc->state.fr.first_pvc;
-	hdlc->state.fr.first_pvc = NULL; /* All PVCs destroyed */
-	hdlc->state.fr.dce_pvc_count = 0;
-	hdlc->state.fr.dce_changed = 1;
+	hdlc_device *hdlc = dev_to_hdlc(frad);
+	pvc_device *pvc = state(hdlc)->first_pvc;
+	state(hdlc)->first_pvc = NULL; /* All PVCs destroyed */
+	state(hdlc)->dce_pvc_count = 0;
+	state(hdlc)->dce_changed = 1;
 
 	while (pvc) {
 		pvc_device *next = pvc->next;
@@ -1161,8 +1215,17 @@ static void fr_destroy(hdlc_device *hdlc)
 }
 
 
+static struct hdlc_proto proto = {
+	.close		= fr_close,
+	.start		= fr_start,
+	.stop		= fr_stop,
+	.detach		= fr_destroy,
+	.ioctl		= fr_ioctl,
+	.module		= THIS_MODULE,
+};
+
 
-int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
+static int fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	fr_proto __user *fr_s = ifr->ifr_settings.ifs_ifsu.fr;
 	const size_t size = sizeof(fr_proto);
@@ -1173,12 +1236,14 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto) /* Different proto */
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_FR;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(fr_s, &hdlc->state.fr.settings, size))
+		if (copy_to_user(fr_s, &state(hdlc)->settings, size))
 			return -EFAULT;
 		return 0;
 
@@ -1213,20 +1278,16 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		if (hdlc->proto.id != IF_PROTO_FR) {
-			hdlc_proto_detach(hdlc);
-			hdlc->state.fr.first_pvc = NULL;
-			hdlc->state.fr.dce_pvc_count = 0;
+		if (dev_to_hdlc(dev)->proto != &proto) { /* Different proto */
+			result = attach_hdlc_protocol(dev, &proto, fr_rx,
+						      sizeof(struct frad_state));
+			if (result)
+				return result;
+			state(hdlc)->first_pvc = NULL;
+			state(hdlc)->dce_pvc_count = 0;
 		}
-		memcpy(&hdlc->state.fr.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.close = fr_close;
-		hdlc->proto.start = fr_start;
-		hdlc->proto.stop = fr_stop;
-		hdlc->proto.detach = fr_destroy;
-		hdlc->proto.netif_rx = fr_rx;
-		hdlc->proto.id = IF_PROTO_FR;
+		memcpy(&state(hdlc)->settings, &new_settings, size);
+
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_FRAD;
@@ -1238,6 +1299,9 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 	case IF_PROTO_FR_DEL_PVC:
 	case IF_PROTO_FR_ADD_ETH_PVC:
 	case IF_PROTO_FR_DEL_ETH_PVC:
+		if (dev_to_hdlc(dev)->proto != &proto) /* Different proto */
+			return -EINVAL;
+
 		if(!capable(CAP_NET_ADMIN))
 			return -EPERM;
 
@@ -1263,3 +1327,24 @@ int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Frame-Relay protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_generic.c b/drivers/net/wan/hdlc_generic.c
deleted file mode 100644
index 04ca1f7b6424..000000000000
--- a/drivers/net/wan/hdlc_generic.c
+++ /dev/null
@@ -1,339 +0,0 @@
-/*
- * Generic HDLC support routines for Linux
- *
- * Copyright (C) 1999 - 2005 Krzysztof Halasa <khc@pm.waw.pl>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2 of the GNU General Public License
- * as published by the Free Software Foundation.
- *
- * Currently supported:
- *	* raw IP-in-HDLC
- *	* Cisco HDLC
- *	* Frame Relay with ANSI or CCITT LMI (both user and network side)
- *	* PPP
- *	* X.25
- *
- * Use sethdlc utility to set line parameters, protocol and PVCs
- *
- * How does it work:
- * - proto.open(), close(), start(), stop() calls are serialized.
- *   The order is: open, [ start, stop ... ] close ...
- * - proto.start() and stop() are called with spin_lock_irq held.
- */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/poll.h>
-#include <linux/errno.h>
-#include <linux/if_arp.h>
-#include <linux/init.h>
-#include <linux/skbuff.h>
-#include <linux/pkt_sched.h>
-#include <linux/inetdevice.h>
-#include <linux/lapb.h>
-#include <linux/rtnetlink.h>
-#include <linux/notifier.h>
-#include <linux/hdlc.h>
-
-
-static const char* version = "HDLC support module revision 1.19";
-
-#undef DEBUG_LINK
-
-
-static int hdlc_change_mtu(struct net_device *dev, int new_mtu)
-{
-	if ((new_mtu < 68) || (new_mtu > HDLC_MAX_MTU))
-		return -EINVAL;
-	dev->mtu = new_mtu;
-	return 0;
-}
-
-
-
-static struct net_device_stats *hdlc_get_stats(struct net_device *dev)
-{
-	return hdlc_stats(dev);
-}
-
-
-
-static int hdlc_rcv(struct sk_buff *skb, struct net_device *dev,
-		    struct packet_type *p, struct net_device *orig_dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.netif_rx)
-		return hdlc->proto.netif_rx(skb);
-
-	hdlc->stats.rx_dropped++; /* Shouldn't happen */
-	dev_kfree_skb(skb);
-	return NET_RX_DROP;
-}
-
-
-
-static inline void hdlc_proto_start(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.start)
-		return hdlc->proto.start(dev);
-}
-
-
-
-static inline void hdlc_proto_stop(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	if (hdlc->proto.stop)
-		return hdlc->proto.stop(dev);
-}
-
-
-
-static int hdlc_device_event(struct notifier_block *this, unsigned long event,
-			     void *ptr)
-{
-	struct net_device *dev = ptr;
-	hdlc_device *hdlc;
-	unsigned long flags;
-	int on;
- 
-	if (dev->get_stats != hdlc_get_stats)
-		return NOTIFY_DONE; /* not an HDLC device */
- 
-	if (event != NETDEV_CHANGE)
-		return NOTIFY_DONE; /* Only interrested in carrier changes */
-
-	on = netif_carrier_ok(dev);
-
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "%s: hdlc_device_event NETDEV_CHANGE, carrier %i\n",
-	       dev->name, on);
-#endif
-
-	hdlc = dev_to_hdlc(dev);
-	spin_lock_irqsave(&hdlc->state_lock, flags);
-
-	if (hdlc->carrier == on)
-		goto carrier_exit; /* no change in DCD line level */
-
-	hdlc->carrier = on;
-
-	if (!hdlc->open)
-		goto carrier_exit;
-
-	if (hdlc->carrier) {
-		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
-		hdlc_proto_start(dev);
-	} else {
-		printk(KERN_INFO "%s: Carrier lost\n", dev->name);
-		hdlc_proto_stop(dev);
-	}
-
-carrier_exit:
-	spin_unlock_irqrestore(&hdlc->state_lock, flags);
-	return NOTIFY_DONE;
-}
-
-
-
-/* Must be called by hardware driver when HDLC device is being opened */
-int hdlc_open(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "hdlc_open() carrier %i open %i\n",
-	       hdlc->carrier, hdlc->open);
-#endif
-
-	if (hdlc->proto.id == -1)
-		return -ENOSYS;	/* no protocol attached */
-
-	if (hdlc->proto.open) {
-		int result = hdlc->proto.open(dev);
-		if (result)
-			return result;
-	}
-
-	spin_lock_irq(&hdlc->state_lock);
-
-	if (hdlc->carrier) {
-		printk(KERN_INFO "%s: Carrier detected\n", dev->name);
-		hdlc_proto_start(dev);
-	} else
-		printk(KERN_INFO "%s: No carrier\n", dev->name);
-
-	hdlc->open = 1;
-
-	spin_unlock_irq(&hdlc->state_lock);
-	return 0;
-}
-
-
-
-/* Must be called by hardware driver when HDLC device is being closed */
-void hdlc_close(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-#ifdef DEBUG_LINK
-	printk(KERN_DEBUG "hdlc_close() carrier %i open %i\n",
-	       hdlc->carrier, hdlc->open);
-#endif
-
-	spin_lock_irq(&hdlc->state_lock);
-
-	hdlc->open = 0;
-	if (hdlc->carrier)
-		hdlc_proto_stop(dev);
-
-	spin_unlock_irq(&hdlc->state_lock);
-
-	if (hdlc->proto.close)
-		hdlc->proto.close(dev);
-}
-
-
-
-#ifndef CONFIG_HDLC_RAW
-#define hdlc_raw_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_RAW_ETH
-#define hdlc_raw_eth_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_PPP
-#define hdlc_ppp_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_CISCO
-#define hdlc_cisco_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_FR
-#define hdlc_fr_ioctl(dev, ifr)		-ENOSYS
-#endif
-
-#ifndef CONFIG_HDLC_X25
-#define hdlc_x25_ioctl(dev, ifr)	-ENOSYS
-#endif
-
-
-int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-	unsigned int proto;
-
-	if (cmd != SIOCWANDEV)
-		return -EINVAL;
-
-	switch(ifr->ifr_settings.type) {
-	case IF_PROTO_HDLC:
-	case IF_PROTO_HDLC_ETH:
-	case IF_PROTO_PPP:
-	case IF_PROTO_CISCO:
-	case IF_PROTO_FR:
-	case IF_PROTO_X25:
-		proto = ifr->ifr_settings.type;
-		break;
-
-	default:
-		proto = hdlc->proto.id;
-	}
-
-	switch(proto) {
-	case IF_PROTO_HDLC:	return hdlc_raw_ioctl(dev, ifr);
-	case IF_PROTO_HDLC_ETH:	return hdlc_raw_eth_ioctl(dev, ifr);
-	case IF_PROTO_PPP:	return hdlc_ppp_ioctl(dev, ifr);
-	case IF_PROTO_CISCO:	return hdlc_cisco_ioctl(dev, ifr);
-	case IF_PROTO_FR:	return hdlc_fr_ioctl(dev, ifr);
-	case IF_PROTO_X25:	return hdlc_x25_ioctl(dev, ifr);
-	default:		return -EINVAL;
-	}
-}
-
-void hdlc_setup(struct net_device *dev)
-{
-	hdlc_device *hdlc = dev_to_hdlc(dev);
-
-	dev->get_stats = hdlc_get_stats;
-	dev->change_mtu = hdlc_change_mtu;
-	dev->mtu = HDLC_MAX_MTU;
-
-	dev->type = ARPHRD_RAWHDLC;
-	dev->hard_header_len = 16;
-
-	dev->flags = IFF_POINTOPOINT | IFF_NOARP;
-
-	hdlc->proto.id = -1;
-	hdlc->proto.detach = NULL;
-	hdlc->carrier = 1;
-	hdlc->open = 0;
-	spin_lock_init(&hdlc->state_lock);
-}
-
-struct net_device *alloc_hdlcdev(void *priv)
-{
-	struct net_device *dev;
-	dev = alloc_netdev(sizeof(hdlc_device), "hdlc%d", hdlc_setup);
-	if (dev)
-		dev_to_hdlc(dev)->priv = priv;
-	return dev;
-}
-
-void unregister_hdlc_device(struct net_device *dev)
-{
-	rtnl_lock();
-	hdlc_proto_detach(dev_to_hdlc(dev));
-	unregister_netdevice(dev);
-	rtnl_unlock();
-}
-
-
-
-MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
-MODULE_DESCRIPTION("HDLC support module");
-MODULE_LICENSE("GPL v2");
-
-EXPORT_SYMBOL(hdlc_open);
-EXPORT_SYMBOL(hdlc_close);
-EXPORT_SYMBOL(hdlc_ioctl);
-EXPORT_SYMBOL(hdlc_setup);
-EXPORT_SYMBOL(alloc_hdlcdev);
-EXPORT_SYMBOL(unregister_hdlc_device);
-
-static struct packet_type hdlc_packet_type = {
-	.type = __constant_htons(ETH_P_HDLC),
-	.func = hdlc_rcv,
-};
-
-
-static struct notifier_block hdlc_notifier = {
-        .notifier_call = hdlc_device_event,
-};
-
-
-static int __init hdlc_module_init(void)
-{
-	int result;
-
-	printk(KERN_INFO "%s\n", version);
-	if ((result = register_netdevice_notifier(&hdlc_notifier)) != 0)
-                return result;
-        dev_add_pack(&hdlc_packet_type);
-	return 0;
-}
-
-
-
-static void __exit hdlc_module_exit(void)
-{
-	dev_remove_pack(&hdlc_packet_type);
-	unregister_netdevice_notifier(&hdlc_notifier);
-}
-
-
-module_init(hdlc_module_init);
-module_exit(hdlc_module_exit);
diff --git a/drivers/net/wan/hdlc_ppp.c b/drivers/net/wan/hdlc_ppp.c
index fbaab5bf71eb..e9f717070fde 100644
--- a/drivers/net/wan/hdlc_ppp.c
+++ b/drivers/net/wan/hdlc_ppp.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * Point-to-point protocol support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -22,6 +22,21 @@
 #include <linux/lapb.h>
 #include <linux/rtnetlink.h>
 #include <linux/hdlc.h>
+#include <net/syncppp.h>
+
+struct ppp_state {
+	struct ppp_device pppdev;
+	struct ppp_device *syncppp_ptr;
+	int (*old_change_mtu)(struct net_device *dev, int new_mtu);
+};
+
+static int ppp_ioctl(struct net_device *dev, struct ifreq *ifr);
+
+
+static inline struct ppp_state* state(hdlc_device *hdlc)
+{
+	return(struct ppp_state *)(hdlc->state);
+}
 
 
 static int ppp_open(struct net_device *dev)
@@ -30,16 +45,16 @@ static int ppp_open(struct net_device *dev)
 	void *old_ioctl;
 	int result;
 
-	dev->priv = &hdlc->state.ppp.syncppp_ptr;
-	hdlc->state.ppp.syncppp_ptr = &hdlc->state.ppp.pppdev;
-	hdlc->state.ppp.pppdev.dev = dev;
+	dev->priv = &state(hdlc)->syncppp_ptr;
+	state(hdlc)->syncppp_ptr = &state(hdlc)->pppdev;
+	state(hdlc)->pppdev.dev = dev;
 
 	old_ioctl = dev->do_ioctl;
-	hdlc->state.ppp.old_change_mtu = dev->change_mtu;
-	sppp_attach(&hdlc->state.ppp.pppdev);
+	state(hdlc)->old_change_mtu = dev->change_mtu;
+	sppp_attach(&state(hdlc)->pppdev);
 	/* sppp_attach nukes them. We don't need syncppp's ioctl */
 	dev->do_ioctl = old_ioctl;
-	hdlc->state.ppp.pppdev.sppp.pp_flags &= ~PP_CISCO;
+	state(hdlc)->pppdev.sppp.pp_flags &= ~PP_CISCO;
 	dev->type = ARPHRD_PPP;
 	result = sppp_open(dev);
 	if (result) {
@@ -59,7 +74,7 @@ static void ppp_close(struct net_device *dev)
 	sppp_close(dev);
 	sppp_detach(dev);
 	dev->rebuild_header = NULL;
-	dev->change_mtu = hdlc->state.ppp.old_change_mtu;
+	dev->change_mtu = state(hdlc)->old_change_mtu;
 	dev->mtu = HDLC_MAX_MTU;
 	dev->hard_header_len = 16;
 }
@@ -73,13 +88,24 @@ static __be16 ppp_type_trans(struct sk_buff *skb, struct net_device *dev)
 
 
 
-int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.open		= ppp_open,
+	.close		= ppp_close,
+	.type_trans	= ppp_type_trans,
+	.ioctl		= ppp_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	int result;
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_PPP;
 		return 0; /* return protocol only, no settable parameters */
 
@@ -96,13 +122,10 @@ int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.open = ppp_open;
-		hdlc->proto.close = ppp_close;
-		hdlc->proto.type_trans = ppp_type_trans;
-		hdlc->proto.id = IF_PROTO_PPP;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(struct ppp_state));
+		if (result)
+			return result;
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_PPP;
@@ -113,3 +136,25 @@ int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("PPP protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_raw.c b/drivers/net/wan/hdlc_raw.c
index f15aa6ba77f1..fe3cae5c6b9d 100644
--- a/drivers/net/wan/hdlc_raw.c
+++ b/drivers/net/wan/hdlc_raw.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * HDLC support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -24,6 +24,8 @@
 #include <linux/hdlc.h>
 
 
+static int raw_ioctl(struct net_device *dev, struct ifreq *ifr);
+
 static __be16 raw_type_trans(struct sk_buff *skb, struct net_device *dev)
 {
 	return __constant_htons(ETH_P_IP);
@@ -31,7 +33,14 @@ static __be16 raw_type_trans(struct sk_buff *skb, struct net_device *dev)
 
 
 
-int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.type_trans	= raw_type_trans,
+	.ioctl		= raw_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	raw_hdlc_proto __user *raw_s = ifr->ifr_settings.ifs_ifsu.raw_hdlc;
 	const size_t size = sizeof(raw_hdlc_proto);
@@ -41,12 +50,14 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_HDLC;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(raw_s, &hdlc->state.raw_hdlc.settings, size))
+		if (copy_to_user(raw_s, hdlc->state, size))
 			return -EFAULT;
 		return 0;
 
@@ -71,12 +82,11 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.raw_hdlc.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.type_trans = raw_type_trans;
-		hdlc->proto.id = IF_PROTO_HDLC;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(raw_hdlc_proto));
+		if (result)
+			return result;
+		memcpy(hdlc->state, &new_settings, size);
 		dev->hard_start_xmit = hdlc->xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_RAWHDLC;
@@ -88,3 +98,25 @@ int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Raw HDLC protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_raw_eth.c b/drivers/net/wan/hdlc_raw_eth.c
index d1884987f94e..1a69a9aaa9b9 100644
--- a/drivers/net/wan/hdlc_raw_eth.c
+++ b/drivers/net/wan/hdlc_raw_eth.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * HDLC Ethernet emulation support
  *
- * Copyright (C) 2002-2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 2002-2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -25,6 +25,7 @@
 #include <linux/etherdevice.h>
 #include <linux/hdlc.h>
 
+static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr);
 
 static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 {
@@ -44,7 +45,14 @@ static int eth_tx(struct sk_buff *skb, struct net_device *dev)
 }
 
 
-int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
+static struct hdlc_proto proto = {
+	.type_trans	= eth_type_trans,
+	.ioctl		= raw_eth_ioctl,
+	.module		= THIS_MODULE,
+};
+
+
+static int raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	raw_hdlc_proto __user *raw_s = ifr->ifr_settings.ifs_ifsu.raw_hdlc;
 	const size_t size = sizeof(raw_hdlc_proto);
@@ -56,12 +64,14 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_HDLC_ETH;
 		if (ifr->ifr_settings.size < size) {
 			ifr->ifr_settings.size = size; /* data size wanted */
 			return -ENOBUFS;
 		}
-		if (copy_to_user(raw_s, &hdlc->state.raw_hdlc.settings, size))
+		if (copy_to_user(raw_s, hdlc->state, size))
 			return -EFAULT;
 		return 0;
 
@@ -86,12 +96,11 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memcpy(&hdlc->state.raw_hdlc.settings, &new_settings, size);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.type_trans = eth_type_trans;
-		hdlc->proto.id = IF_PROTO_HDLC_ETH;
+		result = attach_hdlc_protocol(dev, &proto, NULL,
+					      sizeof(raw_hdlc_proto));
+		if (result)
+			return result;
+		memcpy(hdlc->state, &new_settings, size);
 		dev->hard_start_xmit = eth_tx;
 		old_ch_mtu = dev->change_mtu;
 		old_qlen = dev->tx_queue_len;
@@ -106,3 +115,25 @@ int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("Ethernet encapsulation support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/net/wan/hdlc_x25.c b/drivers/net/wan/hdlc_x25.c
index a867fb411f89..e4bb9f8ad433 100644
--- a/drivers/net/wan/hdlc_x25.c
+++ b/drivers/net/wan/hdlc_x25.c
@@ -2,7 +2,7 @@
  * Generic HDLC support routines for Linux
  * X.25 support
  *
- * Copyright (C) 1999 - 2003 Krzysztof Halasa <khc@pm.waw.pl>
+ * Copyright (C) 1999 - 2006 Krzysztof Halasa <khc@pm.waw.pl>
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of version 2 of the GNU General Public License
@@ -25,6 +25,8 @@
 
 #include <net/x25device.h>
 
+static int x25_ioctl(struct net_device *dev, struct ifreq *ifr);
+
 /* These functions are callbacks called by LAPB layer */
 
 static void x25_connect_disconnect(struct net_device *dev, int reason, int code)
@@ -162,30 +164,39 @@ static void x25_close(struct net_device *dev)
 
 static int x25_rx(struct sk_buff *skb)
 {
-	hdlc_device *hdlc = dev_to_hdlc(skb->dev);
+	struct hdlc_device_desc *desc = dev_to_desc(skb->dev);
 
 	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL) {
-		hdlc->stats.rx_dropped++;
+		desc->stats.rx_dropped++;
 		return NET_RX_DROP;
 	}
 
 	if (lapb_data_received(skb->dev, skb) == LAPB_OK)
 		return NET_RX_SUCCESS;
 
-	hdlc->stats.rx_errors++;
+	desc->stats.rx_errors++;
 	dev_kfree_skb_any(skb);
 	return NET_RX_DROP;
 }
 
 
+static struct hdlc_proto proto = {
+	.open		= x25_open,
+	.close		= x25_close,
+	.ioctl		= x25_ioctl,
+	.module		= THIS_MODULE,
+};
+
 
-int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
+static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 {
 	hdlc_device *hdlc = dev_to_hdlc(dev);
 	int result;
 
 	switch (ifr->ifr_settings.type) {
 	case IF_GET_PROTO:
+		if (dev_to_hdlc(dev)->proto != &proto)
+			return -EINVAL;
 		ifr->ifr_settings.type = IF_PROTO_X25;
 		return 0; /* return protocol only, no settable parameters */
 
@@ -200,14 +211,9 @@ int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 		if (result)
 			return result;
 
-		hdlc_proto_detach(hdlc);
-		memset(&hdlc->proto, 0, sizeof(hdlc->proto));
-
-		hdlc->proto.open = x25_open;
-		hdlc->proto.close = x25_close;
-		hdlc->proto.netif_rx = x25_rx;
-		hdlc->proto.type_trans = NULL;
-		hdlc->proto.id = IF_PROTO_X25;
+		if ((result = attach_hdlc_protocol(dev, &proto,
+						   x25_rx, 0)) != 0)
+			return result;
 		dev->hard_start_xmit = x25_xmit;
 		dev->hard_header = NULL;
 		dev->type = ARPHRD_X25;
@@ -218,3 +224,25 @@ int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr)
 
 	return -EINVAL;
 }
+
+
+static int __init mod_init(void)
+{
+	register_hdlc_protocol(&proto);
+	return 0;
+}
+
+
+
+static void __exit mod_exit(void)
+{
+	unregister_hdlc_protocol(&proto);
+}
+
+
+module_init(mod_init);
+module_exit(mod_exit);
+
+MODULE_AUTHOR("Krzysztof Halasa <khc@pm.waw.pl>");
+MODULE_DESCRIPTION("X.25 protocol support for generic HDLC");
+MODULE_LICENSE("GPL v2");
diff --git a/include/linux/hdlc.h b/include/linux/hdlc.h
index d5ebbb29aeae..d4b333938f73 100644
--- a/include/linux/hdlc.h
+++ b/include/linux/hdlc.h
@@ -11,95 +11,46 @@
 #ifndef __HDLC_H
 #define __HDLC_H
 
-#define GENERIC_HDLC_VERSION 4	/* For synchronization with sethdlc utility */
-
-#define CLOCK_DEFAULT   0	/* Default setting */
-#define CLOCK_EXT	1	/* External TX and RX clock - DTE */
-#define CLOCK_INT	2	/* Internal TX and RX clock - DCE */
-#define CLOCK_TXINT	3	/* Internal TX and external RX clock */
-#define CLOCK_TXFROMRX	4	/* TX clock derived from external RX clock */
-
-
-#define ENCODING_DEFAULT	0 /* Default setting */
-#define ENCODING_NRZ		1
-#define ENCODING_NRZI		2
-#define ENCODING_FM_MARK	3
-#define ENCODING_FM_SPACE	4
-#define ENCODING_MANCHESTER	5
-
-
-#define PARITY_DEFAULT		0 /* Default setting */
-#define PARITY_NONE		1 /* No parity */
-#define PARITY_CRC16_PR0	2 /* CRC16, initial value 0x0000 */
-#define PARITY_CRC16_PR1	3 /* CRC16, initial value 0xFFFF */
-#define PARITY_CRC16_PR0_CCITT	4 /* CRC16, initial 0x0000, ITU-T version */
-#define PARITY_CRC16_PR1_CCITT	5 /* CRC16, initial 0xFFFF, ITU-T version */
-#define PARITY_CRC32_PR0_CCITT	6 /* CRC32, initial value 0x00000000 */
-#define PARITY_CRC32_PR1_CCITT	7 /* CRC32, initial value 0xFFFFFFFF */
-
-#define LMI_DEFAULT		0 /* Default setting */
-#define LMI_NONE		1 /* No LMI, all PVCs are static */
-#define LMI_ANSI		2 /* ANSI Annex D */
-#define LMI_CCITT		3 /* ITU-T Annex A */
-#define LMI_CISCO		4 /* The "original" LMI, aka Gang of Four */
 
 #define HDLC_MAX_MTU 1500	/* Ethernet 1500 bytes */
+#if 0
 #define HDLC_MAX_MRU (HDLC_MAX_MTU + 10 + 14 + 4) /* for ETH+VLAN over FR */
+#else
+#define HDLC_MAX_MRU 1600 /* as required for FR network */
+#endif
 
 
 #ifdef __KERNEL__
 
 #include <linux/skbuff.h>
 #include <linux/netdevice.h>
-#include <net/syncppp.h>
 #include <linux/hdlc/ioctl.h>
 
 
-typedef struct {		/* Used in Cisco and PPP mode */
-	u8 address;
-	u8 control;
-	u16 protocol;
-}__attribute__ ((packed)) hdlc_header;
-
-
-
-typedef struct {
-	u32 type;		/* code */
-	u32 par1;
-	u32 par2;
-	u16 rel;		/* reliability */
-	u32 time;
-}__attribute__ ((packed)) cisco_packet;
-#define	CISCO_PACKET_LEN	18
-#define	CISCO_BIG_PACKET_LEN	20
-
-
-
-typedef struct pvc_device_struct {
-	struct net_device *master;
-	struct net_device *main;
-	struct net_device *ether; /* bridged Ethernet interface */
-	struct pvc_device_struct *next;	/* Sorted in ascending DLCI order */
-	int dlci;
-	int open_count;
-
-	struct {
-		unsigned int new: 1;
-		unsigned int active: 1;
-		unsigned int exist: 1;
-		unsigned int deleted: 1;
-		unsigned int fecn: 1;
-		unsigned int becn: 1;
-		unsigned int bandwidth;	/* Cisco LMI reporting only */
-	}state;
-}pvc_device;
-
-
-
-typedef struct hdlc_device_struct {
-	/* To be initialized by hardware driver */
+/* Used by all network devices here, pointed to by netdev_priv(dev) */
+struct hdlc_device_desc {
+	int (*netif_rx)(struct sk_buff *skb);
 	struct net_device_stats stats;
-
+};
+
+/* This structure is a private property of HDLC protocols.
+   Hardware drivers have no interest here */
+
+struct hdlc_proto {
+	int (*open)(struct net_device *dev);
+	void (*close)(struct net_device *dev);
+	void (*start)(struct net_device *dev); /* if open & DCD */
+	void (*stop)(struct net_device *dev); /* if open & !DCD */
+	void (*detach)(struct net_device *dev);
+	int (*ioctl)(struct net_device *dev, struct ifreq *ifr);
+	unsigned short (*type_trans)(struct sk_buff *skb,
+				     struct net_device *dev);
+	struct module *module;
+	struct hdlc_proto *next; /* next protocol in the list */
+};
+
+
+typedef struct hdlc_device {
 	/* used by HDLC layer to take control over HDLC device from hw driver*/
 	int (*attach)(struct net_device *dev,
 		      unsigned short encoding, unsigned short parity);
@@ -107,82 +58,18 @@ typedef struct hdlc_device_struct {
 	/* hardware driver must handle this instead of dev->hard_start_xmit */
 	int (*xmit)(struct sk_buff *skb, struct net_device *dev);
 
-
 	/* Things below are for HDLC layer internal use only */
-	struct {
-		int (*open)(struct net_device *dev);
-		void (*close)(struct net_device *dev);
-
-		/* if open & DCD */
-		void (*start)(struct net_device *dev);
-		/* if open & !DCD */
-		void (*stop)(struct net_device *dev);
-
-		void (*detach)(struct hdlc_device_struct *hdlc);
-		int (*netif_rx)(struct sk_buff *skb);
-		unsigned short (*type_trans)(struct sk_buff *skb,
-					     struct net_device *dev);
-		int id;		/* IF_PROTO_HDLC/CISCO/FR/etc. */
-	}proto;
-
+	const struct hdlc_proto *proto;
 	int carrier;
 	int open;
 	spinlock_t state_lock;
-
-	union {
-		struct {
-			fr_proto settings;
-			pvc_device *first_pvc;
-			int dce_pvc_count;
-
-			struct timer_list timer;
-			unsigned long last_poll;
-			int reliable;
-			int dce_changed;
-			int request;
-			int fullrep_sent;
-			u32 last_errors; /* last errors bit list */
-			u8 n391cnt;
-			u8 txseq; /* TX sequence number */
-			u8 rxseq; /* RX sequence number */
-		}fr;
-
-		struct {
-			cisco_proto settings;
-
-			struct timer_list timer;
-			unsigned long last_poll;
-			int up;
-			int request_sent;
-			u32 txseq; /* TX sequence number */
-			u32 rxseq; /* RX sequence number */
-		}cisco;
-
-		struct {
-			raw_hdlc_proto settings;
-		}raw_hdlc;
-
-		struct {
-			struct ppp_device pppdev;
-			struct ppp_device *syncppp_ptr;
-			int (*old_change_mtu)(struct net_device *dev,
-					      int new_mtu);
-		}ppp;
-	}state;
+	void *state;
 	void *priv;
 }hdlc_device;
 
 
 
-int hdlc_raw_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_raw_eth_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_cisco_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_ppp_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_fr_ioctl(struct net_device *dev, struct ifreq *ifr);
-int hdlc_x25_ioctl(struct net_device *dev, struct ifreq *ifr);
-
-
-/* Exported from hdlc.o */
+/* Exported from hdlc module */
 
 /* Called by hardware driver when a user requests HDLC service */
 int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
@@ -191,17 +78,21 @@ int hdlc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd);
 #define register_hdlc_device(dev)	register_netdev(dev)
 void unregister_hdlc_device(struct net_device *dev);
 
+
+void register_hdlc_protocol(struct hdlc_proto *proto);
+void unregister_hdlc_protocol(struct hdlc_proto *proto);
+
 struct net_device *alloc_hdlcdev(void *priv);
 
-static __inline__ hdlc_device* dev_to_hdlc(struct net_device *dev)
+
+static __inline__ struct hdlc_device_desc* dev_to_desc(struct net_device *dev)
 {
 	return netdev_priv(dev);
 }
 
-
-static __inline__ pvc_device* dev_to_pvc(struct net_device *dev)
+static __inline__ hdlc_device* dev_to_hdlc(struct net_device *dev)
 {
-	return (pvc_device*)dev->priv;
+	return netdev_priv(dev) + sizeof(struct hdlc_device_desc);
 }
 
 
@@ -225,18 +116,14 @@ int hdlc_open(struct net_device *dev);
 /* Must be called by hardware driver when HDLC device is being closed */
 void hdlc_close(struct net_device *dev);
 
+int attach_hdlc_protocol(struct net_device *dev, struct hdlc_proto *proto,
+			 int (*rx)(struct sk_buff *skb), size_t size);
 /* May be used by hardware driver to gain control over HDLC device */
-static __inline__ void hdlc_proto_detach(hdlc_device *hdlc)
-{
-	if (hdlc->proto.detach)
-		hdlc->proto.detach(hdlc);
-	hdlc->proto.detach = NULL;
-}
-
+void detach_hdlc_protocol(struct net_device *dev);
 
 static __inline__ struct net_device_stats *hdlc_stats(struct net_device *dev)
 {
-	return &dev_to_hdlc(dev)->stats;
+	return &dev_to_desc(dev)->stats;
 }
 
 
@@ -248,8 +135,8 @@ static __inline__ __be16 hdlc_type_trans(struct sk_buff *skb,
 	skb->mac.raw  = skb->data;
 	skb->dev      = dev;
 
-	if (hdlc->proto.type_trans)
-		return hdlc->proto.type_trans(skb, dev);
+	if (hdlc->proto->type_trans)
+		return hdlc->proto->type_trans(skb, dev);
 	else
 		return htons(ETH_P_HDLC);
 }
diff --git a/include/linux/hdlc/ioctl.h b/include/linux/hdlc/ioctl.h
index 78430ba3ea69..583972364357 100644
--- a/include/linux/hdlc/ioctl.h
+++ b/include/linux/hdlc/ioctl.h
@@ -1,6 +1,39 @@
 #ifndef __HDLC_IOCTL_H__
 #define __HDLC_IOCTL_H__
 
+
+#define GENERIC_HDLC_VERSION 4	/* For synchronization with sethdlc utility */
+
+#define CLOCK_DEFAULT   0	/* Default setting */
+#define CLOCK_EXT	1	/* External TX and RX clock - DTE */
+#define CLOCK_INT	2	/* Internal TX and RX clock - DCE */
+#define CLOCK_TXINT	3	/* Internal TX and external RX clock */
+#define CLOCK_TXFROMRX	4	/* TX clock derived from external RX clock */
+
+
+#define ENCODING_DEFAULT	0 /* Default setting */
+#define ENCODING_NRZ		1
+#define ENCODING_NRZI		2
+#define ENCODING_FM_MARK	3
+#define ENCODING_FM_SPACE	4
+#define ENCODING_MANCHESTER	5
+
+
+#define PARITY_DEFAULT		0 /* Default setting */
+#define PARITY_NONE		1 /* No parity */
+#define PARITY_CRC16_PR0	2 /* CRC16, initial value 0x0000 */
+#define PARITY_CRC16_PR1	3 /* CRC16, initial value 0xFFFF */
+#define PARITY_CRC16_PR0_CCITT	4 /* CRC16, initial 0x0000, ITU-T version */
+#define PARITY_CRC16_PR1_CCITT	5 /* CRC16, initial 0xFFFF, ITU-T version */
+#define PARITY_CRC32_PR0_CCITT	6 /* CRC32, initial value 0x00000000 */
+#define PARITY_CRC32_PR1_CCITT	7 /* CRC32, initial value 0xFFFFFFFF */
+
+#define LMI_DEFAULT		0 /* Default setting */
+#define LMI_NONE		1 /* No LMI, all PVCs are static */
+#define LMI_ANSI		2 /* ANSI Annex D */
+#define LMI_CCITT		3 /* ITU-T Annex A */
+#define LMI_CISCO		4 /* The "original" LMI, aka Gang of Four */
+
 typedef struct { 
 	unsigned int clock_rate; /* bits per second */
 	unsigned int clock_type; /* internal, external, TX-internal etc. */
-- 
cgit v1.2.1


From 8aca23103c2ed2cf158adbe92f4f17ee69463d1a Mon Sep 17 00:00:00 2001
From: Krzysztof Halasa <khc@pm.waw.pl>
Date: Tue, 26 Sep 2006 23:24:16 +0200
Subject: [PATCH] Make PC300 WAN driver compile again

This patch removes accesses to the HDLC-internal data structures
from pc300 driver, thus enabling it to compile but breaking part
of its functionality.

Signed-off-by: Krzysztof Halasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/wan/pc300_drv.c | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)

diff --git a/drivers/net/wan/pc300_drv.c b/drivers/net/wan/pc300_drv.c
index 56e69403d178..8d9b959bf15b 100644
--- a/drivers/net/wan/pc300_drv.c
+++ b/drivers/net/wan/pc300_drv.c
@@ -2016,7 +2016,6 @@ static void sca_intr(pc300_t * card)
 			pc300ch_t *chan = &card->chan[ch];
 			pc300dev_t *d = &chan->d;
 			struct net_device *dev = d->dev;
-			hdlc_device *hdlc = dev_to_hdlc(dev);
 
 			spin_lock(&card->card_lock);
 
@@ -2049,8 +2048,8 @@ static void sca_intr(pc300_t * card)
 							}
 							cpc_net_rx(dev);
 							/* Discard invalid frames */
-							hdlc->stats.rx_errors++;
-							hdlc->stats.rx_over_errors++;
+							hdlc_stats(dev)->rx_errors++;
+							hdlc_stats(dev)->rx_over_errors++;
 							chan->rx_first_bd = 0;
 							chan->rx_last_bd = N_DMA_RX_BUF - 1;
 							rx_dma_start(card, ch);
@@ -2116,8 +2115,8 @@ static void sca_intr(pc300_t * card)
 										   card->hw.cpld_reg2) &
 								   ~ (CPLD_REG2_FALC_LED1 << (2 * ch)));
 						}
-						hdlc->stats.tx_errors++;
-						hdlc->stats.tx_fifo_errors++;
+						hdlc_stats(dev)->tx_errors++;
+						hdlc_stats(dev)->tx_fifo_errors++;
 						sca_tx_intr(d);
 					}
 				}
@@ -2534,7 +2533,6 @@ static int cpc_change_mtu(struct net_device *dev, int new_mtu)
 
 static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 {
-	hdlc_device *hdlc = dev_to_hdlc(dev);
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
@@ -2552,10 +2550,10 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		case SIOCGPC300CONF:
 #ifdef CONFIG_PC300_MLPPP
 			if (conf->proto != PC300_PROTO_MLPPP) {
-				conf->proto = hdlc->proto.id;
+				conf->proto = /* FIXME hdlc->proto.id */ 0;
 			}
 #else
-			conf->proto = hdlc->proto.id;
+			conf->proto = /* FIXME hdlc->proto.id */ 0;
 #endif
 			memcpy(&conf_aux.conf, conf, sizeof(pc300chconf_t));
 			memcpy(&conf_aux.hw, &card->hw, sizeof(pc300hw_t));
@@ -2588,12 +2586,12 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 					}
 				} else {
 					memcpy(conf, &conf_aux.conf, sizeof(pc300chconf_t));
-					hdlc->proto.id = conf->proto;
+					/* FIXME hdlc->proto.id = conf->proto; */
 				}
 			}
 #else
 			memcpy(conf, &conf_aux.conf, sizeof(pc300chconf_t));
-			hdlc->proto.id = conf->proto;
+			/* FIXME hdlc->proto.id = conf->proto; */
 #endif
 			return 0;
 		case SIOCGPC300STATUS:
@@ -2606,7 +2604,7 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 		case SIOCGPC300UTILSTATS:
 			{
 				if (!arg) {	/* clear statistics */
-					memset(&hdlc->stats, 0, sizeof(struct net_device_stats));
+					memset(hdlc_stats(dev), 0, sizeof(struct net_device_stats));
 					if (card->hw.type == PC300_TE) {
 						memset(&chan->falc, 0, sizeof(falc_t));
 					}
@@ -2617,7 +2615,7 @@ static int cpc_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 					pc300stats.hw_type = card->hw.type;
 					pc300stats.line_on = card->chan[ch].d.line_on;
 					pc300stats.line_off = card->chan[ch].d.line_off;
-					memcpy(&pc300stats.gen_stats, &hdlc->stats,
+					memcpy(&pc300stats.gen_stats, hdlc_stats(dev),
 					       sizeof(struct net_device_stats));
 					if (card->hw.type == PC300_TE)
 						memcpy(&pc300stats.te_stats,&chan->falc,sizeof(falc_t));
@@ -3147,7 +3145,6 @@ static void cpc_closech(pc300dev_t * d)
 
 int cpc_open(struct net_device *dev)
 {
-	hdlc_device *hdlc = dev_to_hdlc(dev);
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	struct ifreq ifr;
 	int result;
@@ -3156,12 +3153,14 @@ int cpc_open(struct net_device *dev)
 	printk("pc300: cpc_open");
 #endif
 
+#ifdef FIXME
 	if (hdlc->proto.id == IF_PROTO_PPP) {
 		d->if_ptr = &hdlc->state.ppp.pppdev;
 	}
+#endif
 
 	result = hdlc_open(dev);
-	if (hdlc->proto.id == IF_PROTO_PPP) {
+	if (/* FIXME hdlc->proto.id == IF_PROTO_PPP*/ 0) {
 		dev->priv = d;
 	}
 	if (result) {
@@ -3176,7 +3175,6 @@ int cpc_open(struct net_device *dev)
 
 static int cpc_close(struct net_device *dev)
 {
-	hdlc_device *hdlc = dev_to_hdlc(dev);
 	pc300dev_t *d = (pc300dev_t *) dev->priv;
 	pc300ch_t *chan = (pc300ch_t *) d->chan;
 	pc300_t *card = (pc300_t *) chan->card;
@@ -3193,7 +3191,7 @@ static int cpc_close(struct net_device *dev)
 	CPC_UNLOCK(card, flags);
 
 	hdlc_close(dev);
-	if (hdlc->proto.id == IF_PROTO_PPP) {
+	if (/* FIXME hdlc->proto.id == IF_PROTO_PPP*/ 0) {
 		d->if_ptr = NULL;
 	}
 #ifdef CONFIG_PC300_MLPPP
-- 
cgit v1.2.1


From 43ed3baf623410b3fa6ca14a9d3f6deca3493c56 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Tue, 26 Sep 2006 14:44:37 -0700
Subject: [IA64] printing support for MCA/INIT

Printing message to console from MCA/INIT handler is useful,
however doing oops_in_progress = 1 in them exactly makes
something in kernel wrong. Especially it sounds ugly if
system goes wrong after returning from recoverable MCA.

This patch adds ia64_mca_printk() function that collects
messages into temporary-not-so-large message buffer during
in MCA/INIT environment and print them out later, after
returning to normal context or when handlers determine to
down the system.

Also this print function is exported for use in extensional
MCA handler. It would be useful to describe detail about
recovery.

NOTE:
I don't think it is sane thing if temporary message buffer
is enlarged enough to hold whole stack dumps from INIT, so
buffering is disabled during stack dump from INIT-monarch
(= default_monarch_init_process). please fix it in future.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Acked-by: Russ Anderson <rja@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca.c     | 216 +++++++++++++++++++++++++++++++++++++++++----
 arch/ia64/kernel/mca_drv.c |  54 ++++++++----
 arch/ia64/kernel/mca_drv.h |   4 +
 arch/ia64/kernel/salinfo.c |   4 +
 4 files changed, 242 insertions(+), 36 deletions(-)

diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 2fbe4536fe18..98f3b26d7aff 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -54,6 +54,9 @@
  *
  * 2005-10-07 Keith Owens <kaos@sgi.com>
  *	      Add notify_die() hooks.
+ *
+ * 2006-09-15 Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
+ * 	      Add printing support for MCA/INIT.
  */
 #include <linux/types.h>
 #include <linux/init.h>
@@ -136,11 +139,175 @@ extern void salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe);
 
 static int mca_init __initdata;
 
+/*
+ * limited & delayed printing support for MCA/INIT handler
+ */
+
+#define mprintk(fmt...) ia64_mca_printk(fmt)
+
+#define MLOGBUF_SIZE (512+256*NR_CPUS)
+#define MLOGBUF_MSGMAX 256
+static char mlogbuf[MLOGBUF_SIZE];
+static DEFINE_SPINLOCK(mlogbuf_wlock);	/* mca context only */
+static DEFINE_SPINLOCK(mlogbuf_rlock);	/* normal context only */
+static unsigned long mlogbuf_start;
+static unsigned long mlogbuf_end;
+static unsigned int mlogbuf_finished = 0;
+static unsigned long mlogbuf_timestamp = 0;
+
+static int loglevel_save = -1;
+#define BREAK_LOGLEVEL(__console_loglevel)		\
+	oops_in_progress = 1;				\
+	if (loglevel_save < 0)				\
+		loglevel_save = __console_loglevel;	\
+	__console_loglevel = 15;
+
+#define RESTORE_LOGLEVEL(__console_loglevel)		\
+	if (loglevel_save >= 0) {			\
+		__console_loglevel = loglevel_save;	\
+		loglevel_save = -1;			\
+	}						\
+	mlogbuf_finished = 0;				\
+	oops_in_progress = 0;
+
+/*
+ * Push messages into buffer, print them later if not urgent.
+ */
+void ia64_mca_printk(const char *fmt, ...)
+{
+	va_list args;
+	int printed_len;
+	char temp_buf[MLOGBUF_MSGMAX];
+	char *p;
+
+	va_start(args, fmt);
+	printed_len = vscnprintf(temp_buf, sizeof(temp_buf), fmt, args);
+	va_end(args);
+
+	/* Copy the output into mlogbuf */
+	if (oops_in_progress) {
+		/* mlogbuf was abandoned, use printk directly instead. */
+		printk(temp_buf);
+	} else {
+		spin_lock(&mlogbuf_wlock);
+		for (p = temp_buf; *p; p++) {
+			unsigned long next = (mlogbuf_end + 1) % MLOGBUF_SIZE;
+			if (next != mlogbuf_start) {
+				mlogbuf[mlogbuf_end] = *p;
+				mlogbuf_end = next;
+			} else {
+				/* buffer full */
+				break;
+			}
+		}
+		mlogbuf[mlogbuf_end] = '\0';
+		spin_unlock(&mlogbuf_wlock);
+	}
+}
+EXPORT_SYMBOL(ia64_mca_printk);
+
+/*
+ * Print buffered messages.
+ *  NOTE: call this after returning normal context. (ex. from salinfod)
+ */
+void ia64_mlogbuf_dump(void)
+{
+	char temp_buf[MLOGBUF_MSGMAX];
+	char *p;
+	unsigned long index;
+	unsigned long flags;
+	unsigned int printed_len;
+
+	/* Get output from mlogbuf */
+	while (mlogbuf_start != mlogbuf_end) {
+		temp_buf[0] = '\0';
+		p = temp_buf;
+		printed_len = 0;
+
+		spin_lock_irqsave(&mlogbuf_rlock, flags);
+
+		index = mlogbuf_start;
+		while (index != mlogbuf_end) {
+			*p = mlogbuf[index];
+			index = (index + 1) % MLOGBUF_SIZE;
+			if (!*p)
+				break;
+			p++;
+			if (++printed_len >= MLOGBUF_MSGMAX - 1)
+				break;
+		}
+		*p = '\0';
+		if (temp_buf[0])
+			printk(temp_buf);
+		mlogbuf_start = index;
+
+		mlogbuf_timestamp = 0;
+		spin_unlock_irqrestore(&mlogbuf_rlock, flags);
+	}
+}
+EXPORT_SYMBOL(ia64_mlogbuf_dump);
+
+/*
+ * Call this if system is going to down or if immediate flushing messages to
+ * console is required. (ex. recovery was failed, crash dump is going to be
+ * invoked, long-wait rendezvous etc.)
+ *  NOTE: this should be called from monarch.
+ */
+static void ia64_mlogbuf_finish(int wait)
+{
+	BREAK_LOGLEVEL(console_loglevel);
+
+	spin_lock_init(&mlogbuf_rlock);
+	ia64_mlogbuf_dump();
+	printk(KERN_EMERG "mlogbuf_finish: printing switched to urgent mode, "
+		"MCA/INIT might be dodgy or fail.\n");
+
+	if (!wait)
+		return;
+
+	/* wait for console */
+	printk("Delaying for 5 seconds...\n");
+	udelay(5*1000000);
+
+	mlogbuf_finished = 1;
+}
+EXPORT_SYMBOL(ia64_mlogbuf_finish);
+
+/*
+ * Print buffered messages from INIT context.
+ */
+static void ia64_mlogbuf_dump_from_init(void)
+{
+	if (mlogbuf_finished)
+		return;
+
+	if (mlogbuf_timestamp && (mlogbuf_timestamp + 30*HZ > jiffies)) {
+		printk(KERN_ERR "INIT: mlogbuf_dump is interrupted by INIT "
+			" and the system seems to be messed up.\n");
+		ia64_mlogbuf_finish(0);
+		return;
+	}
+
+	if (!spin_trylock(&mlogbuf_rlock)) {
+		printk(KERN_ERR "INIT: mlogbuf_dump is interrupted by INIT. "
+			"Generated messages other than stack dump will be "
+			"buffered to mlogbuf and will be printed later.\n");
+		printk(KERN_ERR "INIT: If messages would not printed after "
+			"this INIT, wait 30sec and assert INIT again.\n");
+		if (!mlogbuf_timestamp)
+			mlogbuf_timestamp = jiffies;
+		return;
+	}
+	spin_unlock(&mlogbuf_rlock);
+	ia64_mlogbuf_dump();
+}
 
 static void inline
 ia64_mca_spin(const char *func)
 {
-	printk(KERN_EMERG "%s: spinning here, not returning to SAL\n", func);
+	if (monarch_cpu == smp_processor_id())
+		ia64_mlogbuf_finish(0);
+	mprintk(KERN_EMERG "%s: spinning here, not returning to SAL\n", func);
 	while (1)
 		cpu_relax();
 }
@@ -988,18 +1155,22 @@ ia64_wait_for_slaves(int monarch, const char *type)
 	}
 	if (!missing)
 		goto all_in;
-	printk(KERN_INFO "OS %s slave did not rendezvous on cpu", type);
+	/*
+	 * Maybe slave(s) dead. Print buffered messages immediately.
+	 */
+	ia64_mlogbuf_finish(0);
+	mprintk(KERN_INFO "OS %s slave did not rendezvous on cpu", type);
 	for_each_online_cpu(c) {
 		if (c == monarch)
 			continue;
 		if (ia64_mc_info.imi_rendez_checkin[c] == IA64_MCA_RENDEZ_CHECKIN_NOTDONE)
-			printk(" %d", c);
+			mprintk(" %d", c);
 	}
-	printk("\n");
+	mprintk("\n");
 	return;
 
 all_in:
-	printk(KERN_INFO "All OS %s slaves have reached rendezvous\n", type);
+	mprintk(KERN_INFO "All OS %s slaves have reached rendezvous\n", type);
 	return;
 }
 
@@ -1027,10 +1198,8 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw,
 	struct ia64_mca_notify_die nd =
 		{ .sos = sos, .monarch_cpu = &monarch_cpu };
 
-	oops_in_progress = 1;	/* FIXME: make printk NMI/MCA/INIT safe */
-	console_loglevel = 15;	/* make sure printks make it to console */
-	printk(KERN_INFO "Entered OS MCA handler. PSP=%lx cpu=%d monarch=%ld\n",
-		sos->proc_state_param, cpu, sos->monarch);
+	mprintk(KERN_INFO "Entered OS MCA handler. PSP=%lx cpu=%d "
+		"monarch=%ld\n", sos->proc_state_param, cpu, sos->monarch);
 
 	previous_current = ia64_mca_modify_original_stack(regs, sw, sos, "MCA");
 	monarch_cpu = cpu;
@@ -1066,6 +1235,9 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw,
 		rh->severity = sal_log_severity_corrected;
 		ia64_sal_clear_state_info(SAL_INFO_TYPE_MCA);
 		sos->os_status = IA64_MCA_CORRECTED;
+	} else {
+		/* Dump buffered message to console */
+		ia64_mlogbuf_finish(1);
 	}
 	if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, (long)&nd, 0, recover)
 			== NOTIFY_STOP)
@@ -1305,6 +1477,15 @@ default_monarch_init_process(struct notifier_block *self, unsigned long val, voi
 	struct task_struct *g, *t;
 	if (val != DIE_INIT_MONARCH_PROCESS)
 		return NOTIFY_DONE;
+
+	/*
+	 * FIXME: mlogbuf will brim over with INIT stack dumps.
+	 * To enable show_stack from INIT, we use oops_in_progress which should
+	 * be used in real oops. This would cause something wrong after INIT.
+	 */
+	BREAK_LOGLEVEL(console_loglevel);
+	ia64_mlogbuf_dump_from_init();
+
 	printk(KERN_ERR "Processes interrupted by INIT -");
 	for_each_online_cpu(c) {
 		struct ia64_sal_os_state *s;
@@ -1326,6 +1507,8 @@ default_monarch_init_process(struct notifier_block *self, unsigned long val, voi
 		} while_each_thread (g, t);
 		read_unlock(&tasklist_lock);
 	}
+	/* FIXME: This will not restore zapped printk locks. */
+	RESTORE_LOGLEVEL(console_loglevel);
 	return NOTIFY_DONE;
 }
 
@@ -1357,12 +1540,9 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 	struct ia64_mca_notify_die nd =
 		{ .sos = sos, .monarch_cpu = &monarch_cpu };
 
-	oops_in_progress = 1;	/* FIXME: make printk NMI/MCA/INIT safe */
-	console_loglevel = 15;	/* make sure printks make it to console */
-
 	(void) notify_die(DIE_INIT_ENTER, "INIT", regs, (long)&nd, 0, 0);
 
-	printk(KERN_INFO "Entered OS INIT handler. PSP=%lx cpu=%d monarch=%ld\n",
+	mprintk(KERN_INFO "Entered OS INIT handler. PSP=%lx cpu=%d monarch=%ld\n",
 		sos->proc_state_param, cpu, sos->monarch);
 	salinfo_log_wakeup(SAL_INFO_TYPE_INIT, NULL, 0, 0);
 
@@ -1375,7 +1555,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 	 * fix their proms and get their customers updated.
 	 */
 	if (!sos->monarch && atomic_add_return(1, &slaves) == num_online_cpus()) {
-		printk(KERN_WARNING "%s: Promoting cpu %d to monarch.\n",
+		mprintk(KERN_WARNING "%s: Promoting cpu %d to monarch.\n",
 		       __FUNCTION__, cpu);
 		atomic_dec(&slaves);
 		sos->monarch = 1;
@@ -1387,7 +1567,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 	 * fix their proms and get their customers updated.
 	 */
 	if (sos->monarch && atomic_add_return(1, &monarchs) > 1) {
-		printk(KERN_WARNING "%s: Demoting cpu %d to slave.\n",
+		mprintk(KERN_WARNING "%s: Demoting cpu %d to slave.\n",
 			       __FUNCTION__, cpu);
 		atomic_dec(&monarchs);
 		sos->monarch = 0;
@@ -1408,7 +1588,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 		if (notify_die(DIE_INIT_SLAVE_LEAVE, "INIT", regs, (long)&nd, 0, 0)
 				== NOTIFY_STOP)
 			ia64_mca_spin(__FUNCTION__);
-		printk("Slave on cpu %d returning to normal service.\n", cpu);
+		mprintk("Slave on cpu %d returning to normal service.\n", cpu);
 		set_curr_task(cpu, previous_current);
 		ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_NOTDONE;
 		atomic_dec(&slaves);
@@ -1426,7 +1606,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 	 * same serial line, the user will need some time to switch out of the BMC before
 	 * the dump begins.
 	 */
-	printk("Delaying for 5 seconds...\n");
+	mprintk("Delaying for 5 seconds...\n");
 	udelay(5*1000000);
 	ia64_wait_for_slaves(cpu, "INIT");
 	/* If nobody intercepts DIE_INIT_MONARCH_PROCESS then we drop through
@@ -1439,7 +1619,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw,
 	if (notify_die(DIE_INIT_MONARCH_LEAVE, "INIT", regs, (long)&nd, 0, 0)
 			== NOTIFY_STOP)
 		ia64_mca_spin(__FUNCTION__);
-	printk("\nINIT dump complete.  Monarch on cpu %d returning to normal service.\n", cpu);
+	mprintk("\nINIT dump complete.  Monarch on cpu %d returning to normal service.\n", cpu);
 	atomic_dec(&monarchs);
 	set_curr_task(cpu, previous_current);
 	monarch_cpu = -1;
diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c
index 8db6e0cedadc..a45009d2bc90 100644
--- a/arch/ia64/kernel/mca_drv.c
+++ b/arch/ia64/kernel/mca_drv.c
@@ -79,14 +79,30 @@ static int
 fatal_mca(const char *fmt, ...)
 {
 	va_list args;
+	char buf[256];
 
 	va_start(args, fmt);
-	vprintk(fmt, args);
+	vsnprintf(buf, sizeof(buf), fmt, args);
 	va_end(args);
+	ia64_mca_printk(KERN_ALERT "MCA: %s\n", buf);
 
 	return MCA_NOT_RECOVERED;
 }
 
+static int
+mca_recovered(const char *fmt, ...)
+{
+	va_list args;
+	char buf[256];
+
+	va_start(args, fmt);
+	vsnprintf(buf, sizeof(buf), fmt, args);
+	va_end(args);
+	ia64_mca_printk(KERN_INFO "MCA: %s\n", buf);
+
+	return MCA_RECOVERED;
+}
+
 /**
  * mca_page_isolate - isolate a poisoned page in order not to use it later
  * @paddr:	poisoned memory location
@@ -140,6 +156,7 @@ mca_page_isolate(unsigned long paddr)
 void
 mca_handler_bh(unsigned long paddr, void *iip, unsigned long ipsr)
 {
+	ia64_mlogbuf_dump();
 	printk(KERN_ERR "OS_MCA: process [cpu %d, pid: %d, uid: %d, "
 		"iip: %p, psr: 0x%lx,paddr: 0x%lx](%s) encounters MCA.\n",
 		raw_smp_processor_id(), current->pid, current->uid,
@@ -440,7 +457,7 @@ recover_from_read_error(slidx_table_t *slidx,
 
 	/* Is target address valid? */
 	if (!pbci->tv)
-		return fatal_mca(KERN_ALERT "MCA: target address not valid\n");
+		return fatal_mca("target address not valid");
 
 	/*
 	 * cpu read or memory-mapped io read
@@ -458,7 +475,7 @@ recover_from_read_error(slidx_table_t *slidx,
 
 	/* Is minstate valid? */
 	if (!peidx_bottom(peidx) || !(peidx_bottom(peidx)->valid.minstate))
-		return fatal_mca(KERN_ALERT "MCA: minstate not valid\n");
+		return fatal_mca("minstate not valid");
 	psr1 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_ipsr);
 	psr2 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_xpsr);
 
@@ -492,13 +509,14 @@ recover_from_read_error(slidx_table_t *slidx,
 			psr2->bn  = 1;
 			psr2->i  = 0;
 
-			return MCA_RECOVERED;
+			return mca_recovered("user memory corruption. "
+				"kill affected process - recovered.");
 		}
 
 	}
 
-	return fatal_mca(KERN_ALERT "MCA: kernel context not recovered,"
-			  " iip 0x%lx\n", pmsa->pmsa_iip);
+	return fatal_mca("kernel context not recovered, iip 0x%lx\n",
+			 pmsa->pmsa_iip);
 }
 
 /**
@@ -584,13 +602,13 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
 	 * The machine check is corrected.
 	 */
 	if (psp->cm == 1)
-		return MCA_RECOVERED;
+		return mca_recovered("machine check is already corrected.");
 
 	/*
 	 * The error was not contained.  Software must be reset.
 	 */
 	if (psp->us || psp->ci == 0)
-		return fatal_mca(KERN_ALERT "MCA: error not contained\n");
+		return fatal_mca("error not contained");
 
 	/*
 	 * The cache check and bus check bits have four possible states
@@ -601,22 +619,22 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
 	 *    1  1	Memory error, attempt recovery
 	 */
 	if (psp->bc == 0 || pbci == NULL)
-		return fatal_mca(KERN_ALERT "MCA: No bus check\n");
+		return fatal_mca("No bus check");
 
 	/*
 	 * Sorry, we cannot handle so many.
 	 */
 	if (peidx_bus_check_num(peidx) > 1)
-		return fatal_mca(KERN_ALERT "MCA: Too many bus checks\n");
+		return fatal_mca("Too many bus checks");
 	/*
 	 * Well, here is only one bus error.
 	 */
 	if (pbci->ib)
-		return fatal_mca(KERN_ALERT "MCA: Internal Bus error\n");
+		return fatal_mca("Internal Bus error");
 	if (pbci->cc)
-		return fatal_mca(KERN_ALERT "MCA: Cache-cache error\n");
+		return fatal_mca("Cache-cache error");
 	if (pbci->eb && pbci->bsi > 0)
-		return fatal_mca(KERN_ALERT "MCA: External bus check fatal status\n");
+		return fatal_mca("External bus check fatal status");
 
 	/*
 	 * This is a local MCA and estimated as recoverble external bus error.
@@ -628,7 +646,7 @@ recover_from_processor_error(int platform, slidx_table_t *slidx,
 	/*
 	 * On account of strange SAL error record, we cannot recover.
 	 */
-	return fatal_mca(KERN_ALERT "MCA: Strange SAL record\n");
+	return fatal_mca("Strange SAL record");
 }
 
 /**
@@ -657,10 +675,10 @@ mca_try_to_recover(void *rec, struct ia64_sal_os_state *sos)
 
 	 /* Now, OS can recover when there is one processor error section */
 	if (n_proc_err > 1)
-		return fatal_mca(KERN_ALERT "MCA: Too Many Errors\n");
+		return fatal_mca("Too Many Errors");
 	else if (n_proc_err == 0)
-		/* Weird SAL record ... We need not to recover */
-		return fatal_mca(KERN_ALERT "MCA: Weird SAL record\n");
+		/* Weird SAL record ... We can't do anything */
+		return fatal_mca("Weird SAL record");
 
 	/* Make index of processor error section */
 	mca_make_peidx((sal_log_processor_info_t*)
@@ -671,7 +689,7 @@ mca_try_to_recover(void *rec, struct ia64_sal_os_state *sos)
 
 	/* Check whether MCA is global or not */
 	if (is_mca_global(&peidx, &pbci, sos))
-		return fatal_mca(KERN_ALERT "MCA: global MCA\n");
+		return fatal_mca("global MCA");
 	
 	/* Try to recover a processor error */
 	return recover_from_processor_error(platform_err, &slidx, &peidx,
diff --git a/arch/ia64/kernel/mca_drv.h b/arch/ia64/kernel/mca_drv.h
index 31a2e52bb16f..c85e943ba5fd 100644
--- a/arch/ia64/kernel/mca_drv.h
+++ b/arch/ia64/kernel/mca_drv.h
@@ -118,3 +118,7 @@ struct mca_table_entry {
 
 extern const struct mca_table_entry *search_mca_tables (unsigned long addr);
 extern int mca_recover_range(unsigned long);
+extern void ia64_mca_printk(const char * fmt, ...)
+	 __attribute__ ((format (printf, 1, 2)));
+extern void ia64_mlogbuf_dump(void);
+
diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c
index 9065f0f01ba3..e63b8ca5344a 100644
--- a/arch/ia64/kernel/salinfo.c
+++ b/arch/ia64/kernel/salinfo.c
@@ -266,6 +266,7 @@ salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe)
 /* Check for outstanding MCA/INIT records every minute (arbitrary) */
 #define SALINFO_TIMER_DELAY (60*HZ)
 static struct timer_list salinfo_timer;
+extern void ia64_mlogbuf_dump(void);
 
 static void
 salinfo_timeout_check(struct salinfo_data *data)
@@ -283,6 +284,7 @@ salinfo_timeout_check(struct salinfo_data *data)
 static void
 salinfo_timeout (unsigned long arg)
 {
+	ia64_mlogbuf_dump();
 	salinfo_timeout_check(salinfo_data + SAL_INFO_TYPE_MCA);
 	salinfo_timeout_check(salinfo_data + SAL_INFO_TYPE_INIT);
 	salinfo_timer.expires = jiffies + SALINFO_TIMER_DELAY;
@@ -332,6 +334,8 @@ retry:
 	if (cpu == -1)
 		goto retry;
 
+	ia64_mlogbuf_dump();
+
 	/* for next read, start checking at next CPU */
 	data->cpu_check = cpu;
 	if (++data->cpu_check == NR_CPUS)
-- 
cgit v1.2.1


From 8f9e146732dcba5161dad3747ee73be1f8c13133 Mon Sep 17 00:00:00 2001
From: Russ Anderson <rja@sgi.com>
Date: Mon, 18 Sep 2006 18:37:15 -0500
Subject: [IA64] ar.fpsr not set on MCA/INIT kernel entry

When entering the kernel due to an MCA or INIT, ar.fpsr (ar40)
was not getting set to the kernel default value (remaining
at the user value).  The effect depends on the user setting
of ar.fpsr.  In the test case, the effect was addresses
printing with strange hex values.

Setting ar.fpsr in ia64_set_kernel_registers sets it for both
the MCA and INIT paths.  The user value of ar.fpsr is correctly
saved (in ia64_state_save) and restored (in ia64_state_restore).

Below is an example of output with very strange hex values.
Anyone know the value of hex 'g'?  :-)

Processes interrupted by INIT - 0 (cpu 14 task 0xdfffg55g7a4c6gA)

Signed-off-by: Russ Anderson (rja@sgi.com)
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca_asm.S | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S
index ee3ff76c8a35..c6b607c00dee 100644
--- a/arch/ia64/kernel/mca_asm.S
+++ b/arch/ia64/kernel/mca_asm.S
@@ -1062,6 +1062,10 @@ ia64_set_kernel_registers:
 	mov cr.itir=r18
 	mov cr.ifa=r13
 	mov r20=IA64_TR_CURRENT_STACK
+
+	movl r17=FPSR_DEFAULT
+	;;
+	mov.m ar.fpsr=r17			// set ar.fpsr to kernel default value
 	;;
 	itr.d dtr[r20]=r21
 	;;
-- 
cgit v1.2.1


From b29e7132b5a9f2496beed37beef7ba4d010afb2c Mon Sep 17 00:00:00 2001
From: Russ Anderson <rja@sgi.com>
Date: Tue, 26 Sep 2006 14:47:48 -0500
Subject: [IA64] PAL calls need physical mode, stacked

PAL_CACHE_READ and PAL_CACHE_WRITE need to be called in physical
mode with stacked registers.

Signed-off-by: Russ Anderson (rja@sgi.com)
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 include/asm-ia64/pal.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h
index d1587e4f5759..2c8fd92d0ece 100644
--- a/include/asm-ia64/pal.h
+++ b/include/asm-ia64/pal.h
@@ -964,7 +964,8 @@ static inline s64
 ia64_pal_cache_read (pal_cache_line_id_u_t line_id, u64 physical_addr)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_CACHE_READ, line_id.pclid_data, physical_addr, 0);
+	PAL_CALL_PHYS_STK(iprv, PAL_CACHE_READ, line_id.pclid_data,
+				physical_addr, 0);
 	return iprv.status;
 }
 
@@ -986,7 +987,8 @@ static inline s64
 ia64_pal_cache_write (pal_cache_line_id_u_t line_id, u64 physical_addr, u64 data)
 {
 	struct ia64_pal_retval iprv;
-	PAL_CALL(iprv, PAL_CACHE_WRITE, line_id.pclid_data, physical_addr, data);
+	PAL_CALL_PHYS_STK(iprv, PAL_CACHE_WRITE, line_id.pclid_data,
+				physical_addr, data);
 	return iprv.status;
 }
 
-- 
cgit v1.2.1


From ddb4f0df0424d174567a011a176782ffa4202071 Mon Sep 17 00:00:00 2001
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Date: Tue, 26 Sep 2006 15:27:56 -0700
Subject: [IA64] CMC/CPE: Reverse the order of fetching log and checking poll
 threshold

This patch reverses the order of fetching log from SAL and
checking poll threshold. This will fix following trivial issues:

- If SAL_GET_SATE_INFO is unbelievably slow (due to huge system
   or just its silly implementation) and if it takes more than
   1/5 sec, CMCI/CPEI will never switch to CMCP/CPEP.
- Assuming terrible flood of interrupt (continuous corrected
   errors let all CPUs enter to handler at once and bind them
   in it), CPUs will be serialized by IA64_LOG_LOCK(*).
   Now we check the poll threshold after the lock and log fetch,
   so we need to call SAL_GET_STATE_INFO (num_online_cpus() + 4)
   times in the worst case.
   if we can check the threshold before the lock, we can shut up
   interrupts quickly without waiting preceding log fetches, and
   the number of times will be reduced to (num_online_cpus()) in
   the same situation.

Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/mca.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)

diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c
index 98f3b26d7aff..bfbd8986153b 100644
--- a/arch/ia64/kernel/mca.c
+++ b/arch/ia64/kernel/mca.c
@@ -511,9 +511,6 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs)
 	/* SAL spec states this should run w/ interrupts enabled */
 	local_irq_enable();
 
-	/* Get the CPE error record and log it */
-	ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CPE);
-
 	spin_lock(&cpe_history_lock);
 	if (!cpe_poll_enabled && cpe_vector >= 0) {
 
@@ -542,7 +539,7 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs)
 			mod_timer(&cpe_poll_timer, jiffies + MIN_CPE_POLL_INTERVAL);
 
 			/* lock already released, get out now */
-			return IRQ_HANDLED;
+			goto out;
 		} else {
 			cpe_history[index++] = now;
 			if (index == CPE_HISTORY_LENGTH)
@@ -550,6 +547,10 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs)
 		}
 	}
 	spin_unlock(&cpe_history_lock);
+out:
+	/* Get the CPE error record and log it */
+	ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CPE);
+
 	return IRQ_HANDLED;
 }
 
@@ -1278,9 +1279,6 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs)
 	/* SAL spec states this should run w/ interrupts enabled */
 	local_irq_enable();
 
-	/* Get the CMC error record and log it */
-	ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CMC);
-
 	spin_lock(&cmc_history_lock);
 	if (!cmc_polling_enabled) {
 		int i, count = 1; /* we know 1 happened now */
@@ -1313,7 +1311,7 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs)
 			mod_timer(&cmc_poll_timer, jiffies + CMC_POLL_INTERVAL);
 
 			/* lock already released, get out now */
-			return IRQ_HANDLED;
+			goto out;
 		} else {
 			cmc_history[index++] = now;
 			if (index == CMC_HISTORY_LENGTH)
@@ -1321,6 +1319,10 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg, struct pt_regs *ptregs)
 		}
 	}
 	spin_unlock(&cmc_history_lock);
+out:
+	/* Get the CMC error record and log it */
+	ia64_mca_log_sal_error_record(SAL_INFO_TYPE_CMC);
+
 	return IRQ_HANDLED;
 }
 
-- 
cgit v1.2.1


From df8f0ec1a413ae610899f009e25dc9777881e149 Mon Sep 17 00:00:00 2001
From: Al Stone <ahs3@fc.hp.com>
Date: Sat, 12 Aug 2006 11:08:12 -0600
Subject: [IA64] minor reformatting to vmlinux.lds.S

Minor reformatting to vmlinux.lds.S to make it 80-column usable,
in accordance with Linux coding style.

Signed-off-by: Al Stone <ahs3@fc.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/vmlinux.lds.S | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S
index 5b0d5f64a9b1..b3b2e389d6b2 100644
--- a/arch/ia64/kernel/vmlinux.lds.S
+++ b/arch/ia64/kernel/vmlinux.lds.S
@@ -184,7 +184,9 @@ SECTIONS
 	  *(.data.gate)
 	  __stop_gate_section = .;
 	}
-  . = ALIGN(PAGE_SIZE);		/* make sure the gate page doesn't expose kernel data */
+  . = ALIGN(PAGE_SIZE);		/* make sure the gate page doesn't expose
+  				 * kernel data
+				 */
 
   .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET)
         { *(.data.read_mostly) }
@@ -202,7 +204,9 @@ SECTIONS
 		*(.data.percpu)
 		__per_cpu_end = .;
 	}
-  . = __phys_per_cpu_start + PERCPU_PAGE_SIZE;	/* ensure percpu data fits into percpu page size */
+  . = __phys_per_cpu_start + PERCPU_PAGE_SIZE;	/* ensure percpu data fits
+  						 * into percpu page size
+						 */
 
   data : { } :data
   .data : AT(ADDR(.data) - LOAD_OFFSET)
-- 
cgit v1.2.1


From 9455e4c9abf76fa02170743859b2ddbb484e7fdf Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sat, 1 Jul 2006 17:16:57 +0200
Subject: i2c-dev: Cleanups

i2c-dev: Cleanups

* We no more need to include platform_device.h.
* Delete the to_i2c_dev macro, which is no more used (and no more valid
  either.)
* Drop i2c_dev.minor. Now that the minor number always matches the i2c
  adapter number, this field is redundant with i2c_dev.adap->nr.
* Delete i2c_dev_get_by_adapter() which is now redundant with
  i2c_dev_get_by_minor() for the same reason.
* Drop the local variable dev in i2cdev_attach_adapter(), we can
  easily do without it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-dev.c | 33 ++++++++-------------------------
 1 file changed, 8 insertions(+), 25 deletions(-)

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index 58ccddd5c237..6e90dec02567 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -34,17 +34,14 @@
 #include <linux/init.h>
 #include <linux/i2c.h>
 #include <linux/i2c-dev.h>
-#include <linux/platform_device.h>
 #include <asm/uaccess.h>
 
 static struct i2c_client i2cdev_client_template;
 
 struct i2c_dev {
-	int minor;
 	struct i2c_adapter *adap;
 	struct class_device *class_dev;
 };
-#define to_i2c_dev(d) container_of(d, struct i2c_dev, class_dev)
 
 #define I2C_MINORS	256
 static struct i2c_dev *i2c_dev_array[I2C_MINORS];
@@ -60,18 +57,6 @@ static struct i2c_dev *i2c_dev_get_by_minor(unsigned index)
 	return i2c_dev;
 }
 
-static struct i2c_dev *i2c_dev_get_by_adapter(struct i2c_adapter *adap)
-{
-	struct i2c_dev *i2c_dev = NULL;
-
-	spin_lock(&i2c_dev_array_lock);
-	if ((i2c_dev_array[adap->nr]) &&
-	    (i2c_dev_array[adap->nr]->adap == adap))
-		i2c_dev = i2c_dev_array[adap->nr];
-	spin_unlock(&i2c_dev_array_lock);
-	return i2c_dev;
-}
-
 static struct i2c_dev *get_free_i2c_dev(struct i2c_adapter *adap)
 {
 	struct i2c_dev *i2c_dev;
@@ -86,7 +71,7 @@ static struct i2c_dev *get_free_i2c_dev(struct i2c_adapter *adap)
 		dev_err(&adap->dev, "i2c-dev already has a device assigned to this adapter\n");
 		goto error;
 	}
-	i2c_dev->minor = adap->nr;
+	i2c_dev->adap = adap;
 	i2c_dev_array[adap->nr] = i2c_dev;
 	spin_unlock(&i2c_dev_array_lock);
 	return i2c_dev;
@@ -98,7 +83,7 @@ error:
 static void return_i2c_dev(struct i2c_dev *i2c_dev)
 {
 	spin_lock(&i2c_dev_array_lock);
-	i2c_dev_array[i2c_dev->minor] = NULL;
+	i2c_dev_array[i2c_dev->adap->nr] = NULL;
 	spin_unlock(&i2c_dev_array_lock);
 }
 
@@ -415,21 +400,19 @@ static struct class *i2c_dev_class;
 static int i2cdev_attach_adapter(struct i2c_adapter *adap)
 {
 	struct i2c_dev *i2c_dev;
-	struct device *dev;
 
 	i2c_dev = get_free_i2c_dev(adap);
 	if (IS_ERR(i2c_dev))
 		return PTR_ERR(i2c_dev);
 
 	pr_debug("i2c-dev: adapter [%s] registered as minor %d\n",
-		 adap->name, i2c_dev->minor);
+		 adap->name, adap->nr);
 
 	/* register this i2c device with the driver core */
-	i2c_dev->adap = adap;
-	dev = &adap->dev;
 	i2c_dev->class_dev = class_device_create(i2c_dev_class, NULL,
-						 MKDEV(I2C_MAJOR, i2c_dev->minor),
-						 dev, "i2c-%d", i2c_dev->minor);
+						 MKDEV(I2C_MAJOR, adap->nr),
+						 &adap->dev, "i2c-%d",
+						 adap->nr);
 	if (!i2c_dev->class_dev)
 		goto error;
 	class_device_create_file(i2c_dev->class_dev, &class_device_attr_name);
@@ -444,12 +427,12 @@ static int i2cdev_detach_adapter(struct i2c_adapter *adap)
 {
 	struct i2c_dev *i2c_dev;
 
-	i2c_dev = i2c_dev_get_by_adapter(adap);
+	i2c_dev = i2c_dev_get_by_minor(adap->nr);
 	if (!i2c_dev)
 		return -ENODEV;
 
 	return_i2c_dev(i2c_dev);
-	class_device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, i2c_dev->minor));
+	class_device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
 	kfree(i2c_dev);
 
 	pr_debug("i2c-dev: adapter [%s] unregistered\n", adap->name);
-- 
cgit v1.2.1


From f3b3aadbbd66d8a020550b01b37d9b1ea559f2c3 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sat, 1 Jul 2006 17:17:38 +0200
Subject: i2c-dev: Use a list for data storage

i2c-dev: Use a list for data storage

Use a list instead of a static array for storing the i2c-dev data.
Given that most systems have less than 10 i2c busses, most of the
space was wasted, so this saves around 1 kB of memory (2 kB on 64-bit
archs.)

The drawback is that lookup was in O(1) and is now in O(N), but given
that the values of N are always small, I don't think this is a problem.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-dev.c | 45 +++++++++++++++++++++++++--------------------
 1 file changed, 25 insertions(+), 20 deletions(-)

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index 6e90dec02567..206e8052f90f 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -32,6 +32,7 @@
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
 #include <linux/init.h>
+#include <linux/list.h>
 #include <linux/i2c.h>
 #include <linux/i2c-dev.h>
 #include <asm/uaccess.h>
@@ -39,21 +40,27 @@
 static struct i2c_client i2cdev_client_template;
 
 struct i2c_dev {
+	struct list_head list;
 	struct i2c_adapter *adap;
 	struct class_device *class_dev;
 };
 
 #define I2C_MINORS	256
-static struct i2c_dev *i2c_dev_array[I2C_MINORS];
-static DEFINE_SPINLOCK(i2c_dev_array_lock);
+static LIST_HEAD(i2c_dev_list);
+static DEFINE_SPINLOCK(i2c_dev_list_lock);
 
 static struct i2c_dev *i2c_dev_get_by_minor(unsigned index)
 {
 	struct i2c_dev *i2c_dev;
 
-	spin_lock(&i2c_dev_array_lock);
-	i2c_dev = i2c_dev_array[index];
-	spin_unlock(&i2c_dev_array_lock);
+	spin_lock(&i2c_dev_list_lock);
+	list_for_each_entry(i2c_dev, &i2c_dev_list, list) {
+		if (i2c_dev->adap->nr == index)
+			goto found;
+	}
+	i2c_dev = NULL;
+found:
+	spin_unlock(&i2c_dev_list_lock);
 	return i2c_dev;
 }
 
@@ -61,30 +68,28 @@ static struct i2c_dev *get_free_i2c_dev(struct i2c_adapter *adap)
 {
 	struct i2c_dev *i2c_dev;
 
+	if (adap->nr >= I2C_MINORS) {
+		printk(KERN_ERR "i2c-dev: Out of device minors (%d)\n",
+		       adap->nr);
+		return ERR_PTR(-ENODEV);
+	}
+
 	i2c_dev = kzalloc(sizeof(*i2c_dev), GFP_KERNEL);
 	if (!i2c_dev)
 		return ERR_PTR(-ENOMEM);
-
-	spin_lock(&i2c_dev_array_lock);
-	if (i2c_dev_array[adap->nr]) {
-		spin_unlock(&i2c_dev_array_lock);
-		dev_err(&adap->dev, "i2c-dev already has a device assigned to this adapter\n");
-		goto error;
-	}
 	i2c_dev->adap = adap;
-	i2c_dev_array[adap->nr] = i2c_dev;
-	spin_unlock(&i2c_dev_array_lock);
+
+	spin_lock(&i2c_dev_list_lock);
+	list_add_tail(&i2c_dev->list, &i2c_dev_list);
+	spin_unlock(&i2c_dev_list_lock);
 	return i2c_dev;
-error:
-	kfree(i2c_dev);
-	return ERR_PTR(-ENODEV);
 }
 
 static void return_i2c_dev(struct i2c_dev *i2c_dev)
 {
-	spin_lock(&i2c_dev_array_lock);
-	i2c_dev_array[i2c_dev->adap->nr] = NULL;
-	spin_unlock(&i2c_dev_array_lock);
+	spin_lock(&i2c_dev_list_lock);
+	list_del(&i2c_dev->list);
+	spin_unlock(&i2c_dev_list_lock);
 }
 
 static ssize_t show_adapter_name(struct class_device *class_dev, char *buf)
-- 
cgit v1.2.1


From 22f76e744dc41096987c6df8270b5c249511cde5 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sat, 1 Jul 2006 17:20:57 +0200
Subject: i2c-dev: Drop the client template

i2c-dev: Drop the client template

Drop the i2c-dev client template. This saves about 360 bytes of
memory. I got the idea from a similar cleanup Hans-Frieder Vogt
made to i2c-nforce2 recently.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-dev.c | 13 ++++---------
 1 file changed, 4 insertions(+), 9 deletions(-)

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index 206e8052f90f..2ce083391292 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -37,7 +37,7 @@
 #include <linux/i2c-dev.h>
 #include <asm/uaccess.h>
 
-static struct i2c_client i2cdev_client_template;
+static struct i2c_driver i2cdev_driver;
 
 struct i2c_dev {
 	struct list_head list;
@@ -365,12 +365,13 @@ static int i2cdev_open(struct inode *inode, struct file *file)
 	if (!adap)
 		return -ENODEV;
 
-	client = kmalloc(sizeof(*client), GFP_KERNEL);
+	client = kzalloc(sizeof(*client), GFP_KERNEL);
 	if (!client) {
 		i2c_put_adapter(adap);
 		return -ENOMEM;
 	}
-	memcpy(client, &i2cdev_client_template, sizeof(*client));
+	snprintf(client->name, I2C_NAME_SIZE, "i2c-dev %d", adap->nr);
+	client->driver = &i2cdev_driver;
 
 	/* registered with adapter, passed as client to user */
 	client->adapter = adap;
@@ -459,12 +460,6 @@ static struct i2c_driver i2cdev_driver = {
 	.detach_client	= i2cdev_detach_client,
 };
 
-static struct i2c_client i2cdev_client_template = {
-	.name		= "I2C /dev entry",
-	.addr		= -1,
-	.driver		= &i2cdev_driver,
-};
-
 static int __init i2c_dev_init(void)
 {
 	int res;
-- 
cgit v1.2.1


From b119c6c952a086f74202ccda4b7ed72161bb6522 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Tue, 15 Aug 2006 18:26:30 +0200
Subject: i2c: __must_check fixes (core drivers)

i2c: __must_check fixes (core drivers)

Check for error on sysfs file creation.
Check for error on device registration.
Check for error on class device registration.

Greg, I am not familiar with completion, can you please tell me if I
need to take care of it in the error paths (as I did in this patch,
see /* Needed? */ comments), or if it isn't needed?

These patches were tested, including forced errors, so they should work
fine. But of course more testing can't hurt.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-isa.c | 29 +++++++++++++++++++++---
 drivers/i2c/i2c-core.c       | 53 +++++++++++++++++++++++++++++++++++---------
 2 files changed, 69 insertions(+), 13 deletions(-)

diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c
index c3e1d3e888d7..d7486e586068 100644
--- a/drivers/i2c/busses/i2c-isa.c
+++ b/drivers/i2c/busses/i2c-isa.c
@@ -125,6 +125,8 @@ int i2c_isa_del_driver(struct i2c_driver *driver)
 
 static int __init i2c_isa_init(void)
 {
+	int err;
+
 	mutex_init(&isa_adapter.clist_lock);
 	INIT_LIST_HEAD(&isa_adapter.clients);
 
@@ -133,8 +135,16 @@ static int __init i2c_isa_init(void)
 	sprintf(isa_adapter.dev.bus_id, "i2c-%d", isa_adapter.nr);
 	isa_adapter.dev.driver = &i2c_adapter_driver;
 	isa_adapter.dev.release = &i2c_adapter_dev_release;
-	device_register(&isa_adapter.dev);
-	device_create_file(&isa_adapter.dev, &dev_attr_name);
+	err = device_register(&isa_adapter.dev);
+	if (err) {
+		printk(KERN_ERR "i2c-isa: Failed to register device\n");
+		goto exit;
+	}
+	err = device_create_file(&isa_adapter.dev, &dev_attr_name);
+	if (err) {
+		printk(KERN_ERR "i2c-isa: Failed to create name file\n");
+		goto exit_unregister;
+	}
 
 	/* Add this adapter to the i2c_adapter class */
 	memset(&isa_adapter.class_dev, 0x00, sizeof(struct class_device));
@@ -142,11 +152,24 @@ static int __init i2c_isa_init(void)
 	isa_adapter.class_dev.class = &i2c_adapter_class;
 	strlcpy(isa_adapter.class_dev.class_id, isa_adapter.dev.bus_id,
 		BUS_ID_SIZE);
-	class_device_register(&isa_adapter.class_dev);
+	err = class_device_register(&isa_adapter.class_dev);
+	if (err) {
+		printk(KERN_ERR "i2c-isa: Failed to register class device\n");
+		goto exit_remove_name;
+	}
 
 	dev_dbg(&isa_adapter.dev, "%s registered\n", isa_adapter.name);
 
 	return 0;
+
+exit_remove_name:
+	device_remove_file(&isa_adapter.dev, &dev_attr_name);
+exit_unregister:
+	init_completion(&isa_adapter.dev_released); /* Needed? */
+	device_unregister(&isa_adapter.dev);
+	wait_for_completion(&isa_adapter.dev_released);
+exit:
+	return err;
 }
 
 static void __exit i2c_isa_exit(void)
diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 9cb277d6aa48..36e1214b0b1d 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -183,15 +183,21 @@ int i2c_add_adapter(struct i2c_adapter *adap)
 	sprintf(adap->dev.bus_id, "i2c-%d", adap->nr);
 	adap->dev.driver = &i2c_adapter_driver;
 	adap->dev.release = &i2c_adapter_dev_release;
-	device_register(&adap->dev);
-	device_create_file(&adap->dev, &dev_attr_name);
+	res = device_register(&adap->dev);
+	if (res)
+		goto out_list;
+	res = device_create_file(&adap->dev, &dev_attr_name);
+	if (res)
+		goto out_unregister;
 
 	/* Add this adapter to the i2c_adapter class */
 	memset(&adap->class_dev, 0x00, sizeof(struct class_device));
 	adap->class_dev.dev = &adap->dev;
 	adap->class_dev.class = &i2c_adapter_class;
 	strlcpy(adap->class_dev.class_id, adap->dev.bus_id, BUS_ID_SIZE);
-	class_device_register(&adap->class_dev);
+	res = class_device_register(&adap->class_dev);
+	if (res)
+		goto out_remove_name;
 
 	dev_dbg(&adap->dev, "adapter [%s] registered\n", adap->name);
 
@@ -206,6 +212,17 @@ int i2c_add_adapter(struct i2c_adapter *adap)
 out_unlock:
 	mutex_unlock(&core_lists);
 	return res;
+
+out_remove_name:
+	device_remove_file(&adap->dev, &dev_attr_name);
+out_unregister:
+	init_completion(&adap->dev_released); /* Needed? */
+	device_unregister(&adap->dev);
+	wait_for_completion(&adap->dev_released);
+out_list:
+	list_del(&adap->list);
+	idr_remove(&i2c_adapter_idr, adap->nr);
+	goto out_unlock;
 }
 
 
@@ -394,14 +411,14 @@ int i2c_check_addr(struct i2c_adapter *adapter, int addr)
 int i2c_attach_client(struct i2c_client *client)
 {
 	struct i2c_adapter *adapter = client->adapter;
+	int res = 0;
 
 	mutex_lock(&adapter->clist_lock);
 	if (__i2c_check_addr(client->adapter, client->addr)) {
-		mutex_unlock(&adapter->clist_lock);
-		return -EBUSY;
+		res = -EBUSY;
+		goto out_unlock;
 	}
 	list_add_tail(&client->list,&adapter->clients);
-	mutex_unlock(&adapter->clist_lock);
 	
 	if (adapter->client_register)  {
 		if (adapter->client_register(client))  {
@@ -422,10 +439,26 @@ int i2c_attach_client(struct i2c_client *client)
 		"%d-%04x", i2c_adapter_id(adapter), client->addr);
 	dev_dbg(&adapter->dev, "client [%s] registered with bus id %s\n",
 		client->name, client->dev.bus_id);
-	device_register(&client->dev);
-	device_create_file(&client->dev, &dev_attr_client_name);
-	
-	return 0;
+	res = device_register(&client->dev);
+	if (res)
+		goto out_list;
+	res = device_create_file(&client->dev, &dev_attr_client_name);
+	if (res)
+		goto out_unregister;
+
+out_unlock:
+	mutex_unlock(&adapter->clist_lock);
+	return res;
+
+out_unregister:
+	init_completion(&client->released); /* Needed? */
+	device_unregister(&client->dev);
+	wait_for_completion(&client->released);
+out_list:
+	list_del(&client->list);
+	dev_err(&adapter->dev, "Failed to attach i2c client %s at 0x%02x "
+		"(%d)\n", client->name, client->addr, res);
+	goto out_unlock;
 }
 
 
-- 
cgit v1.2.1


From defcb46ed4666095677c8f52904b9e328587a20d Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Tue, 15 Aug 2006 18:30:24 +0200
Subject: i2c: __must_check fixes, i2c-dev

i2c: __must_check fixes (i2c-dev)

Check for error on sysfs file creation.
Check for error on device creation.
Delete sysfs file on device destruction.

I couldn't test this one beyond compilation, as it applies on top of
another patch in Greg's tree [1] which breaks all my systems when I
apply it (my udev isn't recent enough.) Anyone with bleeding edge udev
is welcome to test and report.

[1] http://www.kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/i2c/i2c-dev-device.patch

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-dev.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index 2ce083391292..567fb05aeccc 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -406,6 +406,7 @@ static struct class *i2c_dev_class;
 static int i2cdev_attach_adapter(struct i2c_adapter *adap)
 {
 	struct i2c_dev *i2c_dev;
+	int res;
 
 	i2c_dev = get_free_i2c_dev(adap);
 	if (IS_ERR(i2c_dev))
@@ -419,14 +420,20 @@ static int i2cdev_attach_adapter(struct i2c_adapter *adap)
 						 MKDEV(I2C_MAJOR, adap->nr),
 						 &adap->dev, "i2c-%d",
 						 adap->nr);
-	if (!i2c_dev->class_dev)
+	if (!i2c_dev->class_dev) {
+		res = -ENODEV;
 		goto error;
-	class_device_create_file(i2c_dev->class_dev, &class_device_attr_name);
+	}
+	res = class_device_create_file(i2c_dev->class_dev, &class_device_attr_name);
+	if (res)
+		goto error_destroy;
 	return 0;
+error_destroy:
+	class_device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
 error:
 	return_i2c_dev(i2c_dev);
 	kfree(i2c_dev);
-	return -ENODEV;
+	return res;
 }
 
 static int i2cdev_detach_adapter(struct i2c_adapter *adap)
@@ -437,6 +444,7 @@ static int i2cdev_detach_adapter(struct i2c_adapter *adap)
 	if (!i2c_dev)
 		return -ENODEV;
 
+	class_device_remove_file(i2c_dev->class_dev, &class_device_attr_name);
 	return_i2c_dev(i2c_dev);
 	class_device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
 	kfree(i2c_dev);
-- 
cgit v1.2.1


From 3fd39687540bd5d0501f413c91461d3b9fcbb525 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:32:37 +0200
Subject: i2c-algo-sibyte: Cleanups

i2c-algo-sibyte: Cleanups

* Delete empty algo_control implementation.
* Simplify i2c_sibyte_del_bus.
* Delete empty module init and cleanup functions.
* Drop out-of-date #ifdef MODULE construct.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/algos/i2c-algo-sibyte.c | 31 +------------------------------
 1 file changed, 1 insertion(+), 30 deletions(-)

diff --git a/drivers/i2c/algos/i2c-algo-sibyte.c b/drivers/i2c/algos/i2c-algo-sibyte.c
index 32d41c6fac0f..16d666fa16f3 100644
--- a/drivers/i2c/algos/i2c-algo-sibyte.c
+++ b/drivers/i2c/algos/i2c-algo-sibyte.c
@@ -119,12 +119,6 @@ static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr,
         return 0;
 }
 
-static int algo_control(struct i2c_adapter *adapter, 
-	unsigned int cmd, unsigned long arg)
-{
-	return 0;
-}
-
 static u32 bit_func(struct i2c_adapter *adap)
 {
 	return (I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
@@ -136,7 +130,6 @@ static u32 bit_func(struct i2c_adapter *adap)
 
 static struct i2c_algorithm i2c_sibyte_algo = {
 	.smbus_xfer	= smbus_xfer,
-	.algo_control	= algo_control, /* ioctl */
 	.functionality	= bit_func,
 };
 
@@ -179,37 +172,15 @@ int i2c_sibyte_add_bus(struct i2c_adapter *i2c_adap, int speed)
 
 int i2c_sibyte_del_bus(struct i2c_adapter *adap)
 {
-	int res;
-
-	if ((res = i2c_del_adapter(adap)) < 0)
-		return res;
-
-	return 0;
-}
-
-int __init i2c_algo_sibyte_init (void)
-{
-	printk("i2c-algo-sibyte.o: i2c SiByte algorithm module\n");
-	return 0;
+	return i2c_del_adapter(adap);
 }
 
 
 EXPORT_SYMBOL(i2c_sibyte_add_bus);
 EXPORT_SYMBOL(i2c_sibyte_del_bus);
 
-#ifdef MODULE
 MODULE_AUTHOR("Kip Walker, Broadcom Corp.");
 MODULE_DESCRIPTION("SiByte I2C-Bus algorithm");
 module_param(bit_scan, int, 0);
 MODULE_PARM_DESC(bit_scan, "Scan for active chips on the bus");
 MODULE_LICENSE("GPL");
-
-int init_module(void) 
-{
-	return i2c_algo_sibyte_init();
-}
-
-void cleanup_module(void) 
-{
-}
-#endif
-- 
cgit v1.2.1


From 51c3711704b66986373408cbc0540abea43d2380 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:33:16 +0200
Subject: i2c-algo-sibyte: Merge into i2c-sibyte

i2c-algo-sibyte: Merge into i2c-sibyte

Merge i2c-algo-sibyte into i2c-sibyte, as this is a complete,
hardware-dependent SMBus implementation and not a reusable algorithm.

Perform some basic coding style cleanups while we're here (mainly
space-based indentation replaced by tabulations.)

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/algos/Kconfig           |   6 --
 drivers/i2c/algos/Makefile          |   1 -
 drivers/i2c/algos/i2c-algo-sibyte.c | 186 ------------------------------------
 drivers/i2c/busses/i2c-sibyte.c     | 158 +++++++++++++++++++++++++++++-
 include/linux/i2c-algo-sibyte.h     |  33 -------
 5 files changed, 155 insertions(+), 229 deletions(-)
 delete mode 100644 drivers/i2c/algos/i2c-algo-sibyte.c
 delete mode 100644 include/linux/i2c-algo-sibyte.h

diff --git a/drivers/i2c/algos/Kconfig b/drivers/i2c/algos/Kconfig
index 30408015d231..c034820615bb 100644
--- a/drivers/i2c/algos/Kconfig
+++ b/drivers/i2c/algos/Kconfig
@@ -53,12 +53,6 @@ config I2C_ALGO8XX
 	tristate "MPC8xx CPM I2C interface"
 	depends on 8xx && I2C
 
-config I2C_ALGO_SIBYTE
-	tristate "SiByte SMBus interface"
-	depends on SIBYTE_SB1xxx_SOC && I2C
-	help
-	  Supports the SiByte SOC on-chip I2C interfaces (2 channels).
-
 config I2C_ALGO_SGI
 	tristate "I2C SGI interfaces"
 	depends on I2C && (SGI_IP22 || SGI_IP32 || X86_VISWS)
diff --git a/drivers/i2c/algos/Makefile b/drivers/i2c/algos/Makefile
index 867fe1f67401..208be04a3dbd 100644
--- a/drivers/i2c/algos/Makefile
+++ b/drivers/i2c/algos/Makefile
@@ -6,7 +6,6 @@ obj-$(CONFIG_I2C_ALGOBIT)	+= i2c-algo-bit.o
 obj-$(CONFIG_I2C_ALGOPCF)	+= i2c-algo-pcf.o
 obj-$(CONFIG_I2C_ALGOPCA)	+= i2c-algo-pca.o
 obj-$(CONFIG_I2C_ALGOITE)	+= i2c-algo-ite.o
-obj-$(CONFIG_I2C_ALGO_SIBYTE)	+= i2c-algo-sibyte.o
 obj-$(CONFIG_I2C_ALGO_SGI)	+= i2c-algo-sgi.o
 
 ifeq ($(CONFIG_I2C_DEBUG_ALGO),y)
diff --git a/drivers/i2c/algos/i2c-algo-sibyte.c b/drivers/i2c/algos/i2c-algo-sibyte.c
deleted file mode 100644
index 16d666fa16f3..000000000000
--- a/drivers/i2c/algos/i2c-algo-sibyte.c
+++ /dev/null
@@ -1,186 +0,0 @@
-/* ------------------------------------------------------------------------- */
-/* i2c-algo-sibyte.c i2c driver algorithms for bit-shift adapters		     */
-/* ------------------------------------------------------------------------- */
-/*   Copyright (C) 2001,2002,2003 Broadcom Corporation
-     Copyright (C) 1995-2000 Simon G. Vogl
-
-    This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.		     */
-/* ------------------------------------------------------------------------- */
-
-/* With some changes from Kyösti Mälkki <kmalkki@cc.hut.fi> and even
-   Frodo Looijaard <frodol@dds.nl>.  */
-
-/* Ported for SiByte SOCs by Broadcom Corporation.  */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-
-#include <asm/io.h>
-#include <asm/sibyte/sb1250_regs.h>
-#include <asm/sibyte/sb1250_smbus.h>
-
-#include <linux/i2c.h>
-#include <linux/i2c-algo-sibyte.h>
-
-/* ----- global defines ----------------------------------------------- */
-#define SMB_CSR(a,r) ((long)(a->reg_base + r))
-
-/* ----- global variables ---------------------------------------------	*/
-
-/* module parameters:
- */
-static int bit_scan;	/* have a look at what's hanging 'round		*/
-
-
-static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr, 
-                      unsigned short flags, char read_write,
-                      u8 command, int size, union i2c_smbus_data * data)
-{
-	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
-        int data_bytes = 0;
-        int error;
-
-        while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
-                ;
-
-        switch (size) {
-        case I2C_SMBUS_QUICK:
-                csr_out32((V_SMB_ADDR(addr) | (read_write == I2C_SMBUS_READ ? M_SMB_QDATA : 0) |
-			   V_SMB_TT_QUICKCMD), SMB_CSR(adap, R_SMB_START));
-                break;
-        case I2C_SMBUS_BYTE:
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_RD1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 1;
-                } else {
-                        csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        case I2C_SMBUS_BYTE_DATA:
-                csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD1BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 1;
-                } else {
-                        csr_out32(V_SMB_LB(data->byte), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        case I2C_SMBUS_WORD_DATA:
-                csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
-                if (read_write == I2C_SMBUS_READ) {
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                        data_bytes = 2;
-                } else {
-                        csr_out32(V_SMB_LB(data->word & 0xff), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32(V_SMB_MB(data->word >> 8), SMB_CSR(adap, R_SMB_DATA));
-                        csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
-				  SMB_CSR(adap, R_SMB_START));
-                }
-                break;
-        default:
-                return -1;      /* XXXKW better error code? */
-        }
-
-        while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
-                ;
-
-        error = csr_in32(SMB_CSR(adap, R_SMB_STATUS));
-        if (error & M_SMB_ERROR) {
-                /* Clear error bit by writing a 1 */
-                csr_out32(M_SMB_ERROR, SMB_CSR(adap, R_SMB_STATUS));
-                return -1;      /* XXXKW better error code? */
-        }
-
-        if (data_bytes == 1)
-                data->byte = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xff;
-        if (data_bytes == 2)
-                data->word = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xffff;
-
-        return 0;
-}
-
-static u32 bit_func(struct i2c_adapter *adap)
-{
-	return (I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
-                I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA);
-}
-
-
-/* -----exported algorithm data: -------------------------------------	*/
-
-static struct i2c_algorithm i2c_sibyte_algo = {
-	.smbus_xfer	= smbus_xfer,
-	.functionality	= bit_func,
-};
-
-/* 
- * registering functions to load algorithms at runtime 
- */
-int i2c_sibyte_add_bus(struct i2c_adapter *i2c_adap, int speed)
-{
-	int i;
-	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
-
-	/* register new adapter to i2c module... */
-	i2c_adap->algo = &i2c_sibyte_algo;
-        
-        /* Set the frequency to 100 kHz */
-        csr_out32(speed, SMB_CSR(adap,R_SMB_FREQ));
-        csr_out32(0, SMB_CSR(adap,R_SMB_CONTROL));
-
-	/* scan bus */
-	if (bit_scan) {
-                union i2c_smbus_data data;
-                int rc;
-		printk(KERN_INFO " i2c-algo-sibyte.o: scanning bus %s.\n",
-		       i2c_adap->name);
-		for (i = 0x00; i < 0x7f; i++) {
-                        /* XXXKW is this a realistic probe? */
-                        rc = smbus_xfer(i2c_adap, i, 0, I2C_SMBUS_READ, 0,
-                                        I2C_SMBUS_BYTE_DATA, &data);
-			if (!rc) {
-				printk("(%02x)",i); 
-			} else 
-				printk("."); 
-		}
-		printk("\n");
-	}
-
-	return i2c_add_adapter(i2c_adap);
-}
-
-
-int i2c_sibyte_del_bus(struct i2c_adapter *adap)
-{
-	return i2c_del_adapter(adap);
-}
-
-
-EXPORT_SYMBOL(i2c_sibyte_add_bus);
-EXPORT_SYMBOL(i2c_sibyte_del_bus);
-
-MODULE_AUTHOR("Kip Walker, Broadcom Corp.");
-MODULE_DESCRIPTION("SiByte I2C-Bus algorithm");
-module_param(bit_scan, int, 0);
-MODULE_PARM_DESC(bit_scan, "Scan for active chips on the bus");
-MODULE_LICENSE("GPL");
diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index fa503ed9f86d..f516eb7a23f7 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -1,6 +1,7 @@
 /*
  * Copyright (C) 2004 Steven J. Hill
  * Copyright (C) 2001,2002,2003 Broadcom Corporation
+ * Copyright (C) 1995-2000 Simon G. Vogl
  *
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -17,11 +18,162 @@
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 
+#include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/i2c-algo-sibyte.h>
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <asm/io.h>
 #include <asm/sibyte/sb1250_regs.h>
 #include <asm/sibyte/sb1250_smbus.h>
 
+
+struct i2c_algo_sibyte_data {
+	void *data;		/* private data */
+	int   bus;		/* which bus */
+	void *reg_base;		/* CSR base */
+};
+
+/* ----- global defines ----------------------------------------------- */
+#define SMB_CSR(a,r) ((long)(a->reg_base + r))
+
+/* ----- global variables --------------------------------------------- */
+
+/* module parameters:
+ */
+static int bit_scan;	/* have a look at what's hanging 'round */
+module_param(bit_scan, int, 0);
+MODULE_PARM_DESC(bit_scan, "Scan for active chips on the bus");
+
+
+static int smbus_xfer(struct i2c_adapter *i2c_adap, u16 addr,
+		      unsigned short flags, char read_write,
+		      u8 command, int size, union i2c_smbus_data * data)
+{
+	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
+	int data_bytes = 0;
+	int error;
+
+	while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
+		;
+
+	switch (size) {
+	case I2C_SMBUS_QUICK:
+		csr_out32((V_SMB_ADDR(addr) |
+			   (read_write == I2C_SMBUS_READ ? M_SMB_QDATA : 0) |
+			   V_SMB_TT_QUICKCMD), SMB_CSR(adap, R_SMB_START));
+		break;
+	case I2C_SMBUS_BYTE:
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_RD1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 1;
+		} else {
+			csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	case I2C_SMBUS_BYTE_DATA:
+		csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD1BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 1;
+		} else {
+			csr_out32(V_SMB_LB(data->byte),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	case I2C_SMBUS_WORD_DATA:
+		csr_out32(V_SMB_CMD(command), SMB_CSR(adap, R_SMB_CMD));
+		if (read_write == I2C_SMBUS_READ) {
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_CMD_RD2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+			data_bytes = 2;
+		} else {
+			csr_out32(V_SMB_LB(data->word & 0xff),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32(V_SMB_MB(data->word >> 8),
+				  SMB_CSR(adap, R_SMB_DATA));
+			csr_out32((V_SMB_ADDR(addr) | V_SMB_TT_WR2BYTE),
+				  SMB_CSR(adap, R_SMB_START));
+		}
+		break;
+	default:
+		return -1;      /* XXXKW better error code? */
+	}
+
+	while (csr_in32(SMB_CSR(adap, R_SMB_STATUS)) & M_SMB_BUSY)
+		;
+
+	error = csr_in32(SMB_CSR(adap, R_SMB_STATUS));
+	if (error & M_SMB_ERROR) {
+		/* Clear error bit by writing a 1 */
+		csr_out32(M_SMB_ERROR, SMB_CSR(adap, R_SMB_STATUS));
+		return -1;      /* XXXKW better error code? */
+	}
+
+	if (data_bytes == 1)
+		data->byte = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xff;
+	if (data_bytes == 2)
+		data->word = csr_in32(SMB_CSR(adap, R_SMB_DATA)) & 0xffff;
+
+	return 0;
+}
+
+static u32 bit_func(struct i2c_adapter *adap)
+{
+	return (I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE |
+		I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA);
+}
+
+
+/* -----exported algorithm data: -------------------------------------	*/
+
+static struct i2c_algorithm i2c_sibyte_algo = {
+	.smbus_xfer	= smbus_xfer,
+	.functionality	= bit_func,
+};
+
+/*
+ * registering functions to load algorithms at runtime
+ */
+int i2c_sibyte_add_bus(struct i2c_adapter *i2c_adap, int speed)
+{
+	int i;
+	struct i2c_algo_sibyte_data *adap = i2c_adap->algo_data;
+
+	/* register new adapter to i2c module... */
+	i2c_adap->algo = &i2c_sibyte_algo;
+
+	/* Set the frequency to 100 kHz */
+	csr_out32(speed, SMB_CSR(adap,R_SMB_FREQ));
+	csr_out32(0, SMB_CSR(adap,R_SMB_CONTROL));
+
+	/* scan bus */
+	if (bit_scan) {
+		union i2c_smbus_data data;
+		int rc;
+		printk(KERN_INFO " i2c-algo-sibyte.o: scanning bus %s.\n",
+		       i2c_adap->name);
+		for (i = 0x00; i < 0x7f; i++) {
+			/* XXXKW is this a realistic probe? */
+			rc = smbus_xfer(i2c_adap, i, 0, I2C_SMBUS_READ, 0,
+					I2C_SMBUS_BYTE_DATA, &data);
+			if (!rc) {
+				printk("(%02x)",i);
+			} else
+				printk(".");
+		}
+		printk("\n");
+	}
+
+	return i2c_add_adapter(i2c_adap);
+}
+
+
 static struct i2c_algo_sibyte_data sibyte_board_data[2] = {
 	{ NULL, 0, (void *) (CKSEG1+A_SMB_BASE(0)) },
 	{ NULL, 1, (void *) (CKSEG1+A_SMB_BASE(1)) }
@@ -58,8 +210,8 @@ static int __init i2c_sibyte_init(void)
 
 static void __exit i2c_sibyte_exit(void)
 {
-	i2c_sibyte_del_bus(&sibyte_board_adapter[0]);
-	i2c_sibyte_del_bus(&sibyte_board_adapter[1]);
+	i2c_del_bus(&sibyte_board_adapter[0]);
+	i2c_del_bus(&sibyte_board_adapter[1]);
 }
 
 module_init(i2c_sibyte_init);
diff --git a/include/linux/i2c-algo-sibyte.h b/include/linux/i2c-algo-sibyte.h
deleted file mode 100644
index 03914ded8614..000000000000
--- a/include/linux/i2c-algo-sibyte.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/*
- * Copyright (C) 2001,2002,2003 Broadcom Corporation
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License
- * as published by the Free Software Foundation; either version 2
- * of the License, or (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
- */
-
-#ifndef I2C_ALGO_SIBYTE_H
-#define I2C_ALGO_SIBYTE_H 1
-
-#include <linux/i2c.h>
-
-struct i2c_algo_sibyte_data {
-	void *data;		/* private data */
-        int   bus;		/* which bus */
-        void *reg_base;		/* CSR base */
-};
-
-int i2c_sibyte_add_bus(struct i2c_adapter *, int speed);
-int i2c_sibyte_del_bus(struct i2c_adapter *);
-
-#endif /* I2C_ALGO_SIBYTE_H */
-- 
cgit v1.2.1


From 643bd3fbd9dc73ed3dc1e4f6980e6f15fdbb9bb6 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:34:05 +0200
Subject: i2c-sibyte: Kip Walker is gone

i2c-sibyte: Kip Walker is gone

Kip Walker no longer works at Broadcom, and his e-mail address there
bounces back, so let's drop it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-sibyte.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index f516eb7a23f7..96c9e7bdd9e3 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -217,6 +217,6 @@ static void __exit i2c_sibyte_exit(void)
 module_init(i2c_sibyte_init);
 module_exit(i2c_sibyte_exit);
 
-MODULE_AUTHOR("Kip Walker <kwalker@broadcom.com>, Steven J. Hill <sjhill@realitydiluted.com>");
+MODULE_AUTHOR("Kip Walker (Broadcom Corp.), Steven J. Hill <sjhill@realitydiluted.com>");
 MODULE_DESCRIPTION("SMBus adapter routines for SiByte boards");
 MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 8859942ede8154c1e90e3b0d1b60aecf0cfaa169 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen.puncer@ultra.si>
Date: Sun, 13 Aug 2006 23:35:40 +0200
Subject: i2c-au1550: Fix timeout problem

i2c-au1550: Fix timeout problem

Fix from Jordan Crouse:
If the transmit and recieve FIFOS are not empty, forceably flush them
rather then waiting for them to drain on their own.

This solves at least a problem reported by Clem Taylor:
http://www.linux-mips.org/archives/linux-mips/2006-05/msg00240.html
(1% of I2C transactions would timeout)

Signed-off-by: Domen Puncer <domen.puncer@ultra.si>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-au1550.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/i2c/busses/i2c-au1550.c b/drivers/i2c/busses/i2c-au1550.c
index d06edce03bf4..02a359ebc0f0 100644
--- a/drivers/i2c/busses/i2c-au1550.c
+++ b/drivers/i2c/busses/i2c-au1550.c
@@ -118,13 +118,19 @@ do_address(struct i2c_au1550_data *adap, unsigned int addr, int rd)
 
 	/* Reset the FIFOs, clear events.
 	*/
-	sp->psc_smbpcr = PSC_SMBPCR_DC;
+	stat = sp->psc_smbstat;
 	sp->psc_smbevnt = PSC_SMBEVNT_ALLCLR;
 	au_sync();
-	do {
-		stat = sp->psc_smbpcr;
+
+	if (!(stat & PSC_SMBSTAT_TE) || !(stat & PSC_SMBSTAT_RE)) {
+		sp->psc_smbpcr = PSC_SMBPCR_DC;
 		au_sync();
-	} while ((stat & PSC_SMBPCR_DC) != 0);
+		do {
+			stat = sp->psc_smbpcr;
+			au_sync();
+		} while ((stat & PSC_SMBPCR_DC) != 0);
+		udelay(50);
+	}
 
 	/* Write out the i2c chip address and specify operation
 	*/
-- 
cgit v1.2.1


From 6ed07134792655db12fddaafbf870425331a21f6 Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen.puncer@ultra.si>
Date: Sun, 13 Aug 2006 23:36:27 +0200
Subject: i2c-au1550: Add SMBus functionality flag

i2c-au1550: Add SMBus functionality flag

Add SMBus functionality flag, so we can use eeprom and similar
drivers.

Signed-off-by: Domen Puncer <domen.puncer@ultra.si>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-au1550.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-au1550.c b/drivers/i2c/busses/i2c-au1550.c
index 02a359ebc0f0..7140eb00587a 100644
--- a/drivers/i2c/busses/i2c-au1550.c
+++ b/drivers/i2c/busses/i2c-au1550.c
@@ -285,7 +285,7 @@ au1550_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msgs, int num)
 static u32
 au1550_func(struct i2c_adapter *adap)
 {
-	return I2C_FUNC_I2C;
+	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
 static struct i2c_algorithm au1550_algo = {
-- 
cgit v1.2.1


From a294de4e97a7a0cdeb949a3b2fbd4d81addc6d8c Mon Sep 17 00:00:00 2001
From: Domen Puncer <domen.puncer@ultra.si>
Date: Sun, 13 Aug 2006 23:37:13 +0200
Subject: i2c-au1550: Add I2C support for Au1200

i2c-au1550: Add I2C support for Au1200

Signed-off-by: Domen Puncer <domen.puncer@ultra.si>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/Kconfig      | 6 +++---
 drivers/i2c/busses/i2c-au1550.c | 3 +--
 2 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 884320e70403..404508ca7326 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -75,11 +75,11 @@ config I2C_AMD8111
 	  will be called i2c-amd8111.
 
 config I2C_AU1550
-	tristate "Au1550 SMBus interface"
-	depends on I2C && SOC_AU1550
+	tristate "Au1550/Au1200 SMBus interface"
+	depends on I2C && (SOC_AU1550 || SOC_AU1200)
 	help
 	  If you say yes to this option, support will be included for the
-	  Au1550 SMBus interface.
+	  Au1550 and Au1200 SMBus interface.
 
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-au1550.
diff --git a/drivers/i2c/busses/i2c-au1550.c b/drivers/i2c/busses/i2c-au1550.c
index 7140eb00587a..7a005ddfb11e 100644
--- a/drivers/i2c/busses/i2c-au1550.c
+++ b/drivers/i2c/busses/i2c-au1550.c
@@ -34,8 +34,7 @@
 #include <linux/errno.h>
 #include <linux/i2c.h>
 
-#include <asm/mach-au1x00/au1000.h>
-#include <asm/mach-pb1x00/pb1550.h>
+#include <asm/mach-au1x00/au1xxx.h>
 #include <asm/mach-au1x00/au1xxx_psc.h>
 
 #include "i2c-au1550.h"
-- 
cgit v1.2.1


From be53f9b2a08e647396ceaa004199ae4b032a9bd2 Mon Sep 17 00:00:00 2001
From: Arthur Othieno <apgo@patchbomb.org>
Date: Sun, 13 Aug 2006 23:39:11 +0200
Subject: i2c: Fix copy-n-paste in subsystem Kconfig

i2c: Fix copy-n-paste in subsystem Kconfig

We have:
  drivers/i2c/Kconfig:2:# Character device configuration
Which is obviously not true..

Signed-off-by: Arthur Othieno <apgo@patchbomb.org>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/i2c/Kconfig b/drivers/i2c/Kconfig
index 24383afdda76..11935f66fcd8 100644
--- a/drivers/i2c/Kconfig
+++ b/drivers/i2c/Kconfig
@@ -1,5 +1,5 @@
 #
-# Character device configuration
+# I2C subsystem configuration
 #
 
 menu "I2C support"
-- 
cgit v1.2.1


From 8241cb1401873e18bba746fd3f6c56ce4252a61f Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:40:09 +0200
Subject: i2c-matroxfb: Struct init conversion

i2c-matroxfb: Struct init conversion

Convert the struct i2c_algo_bit_data initialization to proper C99
style.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Petr Vandrovec <vandrove@vc.cvut.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/video/matrox/i2c-matroxfb.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/video/matrox/i2c-matroxfb.c b/drivers/video/matrox/i2c-matroxfb.c
index 57abbae5520f..9842042d2af5 100644
--- a/drivers/video/matrox/i2c-matroxfb.c
+++ b/drivers/video/matrox/i2c-matroxfb.c
@@ -95,12 +95,13 @@ static struct i2c_adapter matrox_i2c_adapter_template =
 
 static struct i2c_algo_bit_data matrox_i2c_algo_template =
 {
-	NULL,
-	matroxfb_gpio_setsda,
-	matroxfb_gpio_setscl,
-	matroxfb_gpio_getsda,
-	matroxfb_gpio_getscl,
-	10, 10, 100,
+	.setsda		= matroxfb_gpio_setsda,
+	.setscl		= matroxfb_gpio_setscl,
+	.getsda		= matroxfb_gpio_getsda,
+	.getscl		= matroxfb_gpio_getscl,
+	.udelay		= 10,
+	.mdelay		= 10,
+	.timeout	= 100,
 };
 
 static int i2c_bus_reg(struct i2c_bit_adapter* b, struct matrox_fb_info* minfo, 
-- 
cgit v1.2.1


From a0d9c63d3640bd4fc90a408e8334754ef44bcf48 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 27 Aug 2006 11:46:49 +0200
Subject: i2c-algo-bit: Discard the mdelay data struct member

i2c-algo-bit: Discard the mdelay data struct member

The i2c_algo_bit_data structure has an mdelay member, which is not
used by the algorithm code (the code has always been ifdef'd out.)
Let's discard it to save some code and memory.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Acked-by: Mauro Carvalho Chehab <mchehab@brturbo.com.br>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acorn/char/i2c.c                   |  1 -
 drivers/i2c/algos/i2c-algo-bit.c           |  4 ----
 drivers/i2c/busses/i2c-hydra.c             |  1 -
 drivers/i2c/busses/i2c-i810.c              |  2 --
 drivers/i2c/busses/i2c-ixp2000.c           |  1 -
 drivers/i2c/busses/i2c-ixp4xx.c            |  1 -
 drivers/i2c/busses/i2c-parport-light.c     |  1 -
 drivers/i2c/busses/i2c-parport.c           |  1 -
 drivers/i2c/busses/i2c-prosavage.c         |  1 -
 drivers/i2c/busses/i2c-savage4.c           |  1 -
 drivers/i2c/busses/i2c-via.c               |  1 -
 drivers/i2c/busses/i2c-voodoo3.c           |  2 --
 drivers/i2c/busses/scx200_i2c.c            | 12 ++++++------
 drivers/ieee1394/pcilynx.c                 |  1 -
 drivers/media/video/bt8xx/bttv-i2c.c       |  1 -
 drivers/media/video/cx88/cx88-i2c.c        |  1 -
 drivers/media/video/cx88/cx88-vp3054-i2c.c |  1 -
 drivers/media/video/zoran_card.c           |  1 -
 drivers/video/i810/i810-i2c.c              |  1 -
 drivers/video/matrox/i2c-matroxfb.c        |  1 -
 drivers/video/savage/savagefb-i2c.c        |  1 -
 include/linux/i2c-algo-bit.h               |  1 -
 22 files changed, 6 insertions(+), 32 deletions(-)

diff --git a/drivers/acorn/char/i2c.c b/drivers/acorn/char/i2c.c
index c26c08b36829..bdb9c8b78ed8 100644
--- a/drivers/acorn/char/i2c.c
+++ b/drivers/acorn/char/i2c.c
@@ -308,7 +308,6 @@ static struct i2c_algo_bit_data ioc_data = {
 	.getsda		= ioc_getsda,
 	.getscl		= ioc_getscl,
 	.udelay		= 80,
-	.mdelay		= 80,
 	.timeout	= 100
 };
 
diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c
index ab230c033f99..761df16838b4 100644
--- a/drivers/i2c/algos/i2c-algo-bit.c
+++ b/drivers/i2c/algos/i2c-algo-bit.c
@@ -354,10 +354,6 @@ static int sendbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
 			return (retval<0)? retval : -EFAULT;
 			        /* got a better one ?? */
 		}
-#if 0
-		/* from asm/delay.h */
-		__delay(adap->mdelay * (loops_per_sec / 1000) );
-#endif
 	}
 	return wrcount;
 }
diff --git a/drivers/i2c/busses/i2c-hydra.c b/drivers/i2c/busses/i2c-hydra.c
index e0cb3b0f92fa..457d48a0ab9d 100644
--- a/drivers/i2c/busses/i2c-hydra.c
+++ b/drivers/i2c/busses/i2c-hydra.c
@@ -99,7 +99,6 @@ static struct i2c_algo_bit_data hydra_bit_data = {
 	.getsda		= hydra_bit_getsda,
 	.getscl		= hydra_bit_getscl,
 	.udelay		= 5,
-	.mdelay		= 5,
 	.timeout	= HZ
 };
 
diff --git a/drivers/i2c/busses/i2c-i810.c b/drivers/i2c/busses/i2c-i810.c
index 748be30f2bae..b66fb6bb1870 100644
--- a/drivers/i2c/busses/i2c-i810.c
+++ b/drivers/i2c/busses/i2c-i810.c
@@ -166,7 +166,6 @@ static struct i2c_algo_bit_data i810_i2c_bit_data = {
 	.getsda		= bit_i810i2c_getsda,
 	.getscl		= bit_i810i2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT,
 };
 
@@ -182,7 +181,6 @@ static struct i2c_algo_bit_data i810_ddc_bit_data = {
 	.getsda		= bit_i810ddc_getsda,
 	.getscl		= bit_i810ddc_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT,
 };
 
diff --git a/drivers/i2c/busses/i2c-ixp2000.c b/drivers/i2c/busses/i2c-ixp2000.c
index cd6f45d186ab..dd3f4cd3aa68 100644
--- a/drivers/i2c/busses/i2c-ixp2000.c
+++ b/drivers/i2c/busses/i2c-ixp2000.c
@@ -114,7 +114,6 @@ static int ixp2000_i2c_probe(struct platform_device *plat_dev)
 	drv_data->algo_data.getsda = ixp2000_bit_getsda;
 	drv_data->algo_data.getscl = ixp2000_bit_getscl;
 	drv_data->algo_data.udelay = 6;
-	drv_data->algo_data.mdelay = 6;
 	drv_data->algo_data.timeout = 100;
 
 	drv_data->adapter.id = I2C_HW_B_IXP2000,
diff --git a/drivers/i2c/busses/i2c-ixp4xx.c b/drivers/i2c/busses/i2c-ixp4xx.c
index 2ed07112d683..ab573254a8aa 100644
--- a/drivers/i2c/busses/i2c-ixp4xx.c
+++ b/drivers/i2c/busses/i2c-ixp4xx.c
@@ -122,7 +122,6 @@ static int ixp4xx_i2c_probe(struct platform_device *plat_dev)
 	drv_data->algo_data.getsda = ixp4xx_bit_getsda;
 	drv_data->algo_data.getscl = ixp4xx_bit_getscl;
 	drv_data->algo_data.udelay = 10;
-	drv_data->algo_data.mdelay = 10;
 	drv_data->algo_data.timeout = 100;
 
 	drv_data->adapter.id = I2C_HW_B_IXP4XX;
diff --git a/drivers/i2c/busses/i2c-parport-light.c b/drivers/i2c/busses/i2c-parport-light.c
index e09ebbb2f9f0..5eb2bd294fd9 100644
--- a/drivers/i2c/busses/i2c-parport-light.c
+++ b/drivers/i2c/busses/i2c-parport-light.c
@@ -103,7 +103,6 @@ static struct i2c_algo_bit_data parport_algo_data = {
 	.getsda		= parport_getsda,
 	.getscl		= parport_getscl,
 	.udelay		= 50,
-	.mdelay		= 50,
 	.timeout	= HZ,
 }; 
 
diff --git a/drivers/i2c/busses/i2c-parport.c b/drivers/i2c/busses/i2c-parport.c
index 934bd55bae15..48a829431c7b 100644
--- a/drivers/i2c/busses/i2c-parport.c
+++ b/drivers/i2c/busses/i2c-parport.c
@@ -138,7 +138,6 @@ static struct i2c_algo_bit_data parport_algo_data = {
 	.getsda		= parport_getsda,
 	.getscl		= parport_getscl,
 	.udelay		= 60,
-	.mdelay		= 60,
 	.timeout	= HZ,
 }; 
 
diff --git a/drivers/i2c/busses/i2c-prosavage.c b/drivers/i2c/busses/i2c-prosavage.c
index 9479525892e3..7745e21874a8 100644
--- a/drivers/i2c/busses/i2c-prosavage.c
+++ b/drivers/i2c/busses/i2c-prosavage.c
@@ -180,7 +180,6 @@ static int i2c_register_bus(struct pci_dev *dev, struct s_i2c_bus *p, void __iom
 	p->algo.getsda	  = bit_s3via_getsda;
 	p->algo.getscl	  = bit_s3via_getscl;
 	p->algo.udelay	  = CYCLE_DELAY;
-	p->algo.mdelay	  = CYCLE_DELAY;
 	p->algo.timeout	  = TIMEOUT;
 	p->algo.data	  = p;
 	p->mmvga	  = mmvga;
diff --git a/drivers/i2c/busses/i2c-savage4.c b/drivers/i2c/busses/i2c-savage4.c
index 0c8518298e4d..209f47ea1750 100644
--- a/drivers/i2c/busses/i2c-savage4.c
+++ b/drivers/i2c/busses/i2c-savage4.c
@@ -140,7 +140,6 @@ static struct i2c_algo_bit_data sav_i2c_bit_data = {
 	.getsda		= bit_savi2c_getsda,
 	.getscl		= bit_savi2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
diff --git a/drivers/i2c/busses/i2c-via.c b/drivers/i2c/busses/i2c-via.c
index 484bbacfce6b..910e200ad500 100644
--- a/drivers/i2c/busses/i2c-via.c
+++ b/drivers/i2c/busses/i2c-via.c
@@ -81,7 +81,6 @@ static struct i2c_algo_bit_data bit_data = {
 	.getsda		= bit_via_getsda,
 	.getscl		= bit_via_getscl,
 	.udelay		= 5,
-	.mdelay		= 5,
 	.timeout	= HZ
 };
 
diff --git a/drivers/i2c/busses/i2c-voodoo3.c b/drivers/i2c/busses/i2c-voodoo3.c
index b675773b0cc1..6c8d25183382 100644
--- a/drivers/i2c/busses/i2c-voodoo3.c
+++ b/drivers/i2c/busses/i2c-voodoo3.c
@@ -160,7 +160,6 @@ static struct i2c_algo_bit_data voo_i2c_bit_data = {
 	.getsda		= bit_vooi2c_getsda,
 	.getscl		= bit_vooi2c_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
@@ -177,7 +176,6 @@ static struct i2c_algo_bit_data voo_ddc_bit_data = {
 	.getsda		= bit_vooddc_getsda,
 	.getscl		= bit_vooddc_getscl,
 	.udelay		= CYCLE_DELAY,
-	.mdelay		= CYCLE_DELAY,
 	.timeout	= TIMEOUT
 };
 
diff --git a/drivers/i2c/busses/scx200_i2c.c b/drivers/i2c/busses/scx200_i2c.c
index cb3ef5ac99fd..8b65a5cf8251 100644
--- a/drivers/i2c/busses/scx200_i2c.c
+++ b/drivers/i2c/busses/scx200_i2c.c
@@ -71,12 +71,12 @@ static int scx200_i2c_getsda(void *data)
  */
 
 static struct i2c_algo_bit_data scx200_i2c_data = {
-	NULL,
-	scx200_i2c_setsda,
-	scx200_i2c_setscl,
-	scx200_i2c_getsda,
-	scx200_i2c_getscl,
-	10, 10, 100,		/* waits, timeout */
+	.setsda		= scx200_i2c_setsda,
+	.setscl		= scx200_i2c_setscl,
+	.getsda		= scx200_i2c_getsda,
+	.getscl		= scx200_i2c_getscl,
+	.udelay		= 10,
+	.timeout	= 100,
 };
 
 static struct i2c_adapter scx200_i2c_ops = {
diff --git a/drivers/ieee1394/pcilynx.c b/drivers/ieee1394/pcilynx.c
index e6f41238f5e8..b4f146f2c951 100644
--- a/drivers/ieee1394/pcilynx.c
+++ b/drivers/ieee1394/pcilynx.c
@@ -137,7 +137,6 @@ static struct i2c_algo_bit_data bit_data = {
 	.getsda			= bit_getsda,
 	.getscl			= bit_getscl,
 	.udelay			= 5,
-	.mdelay			= 5,
 	.timeout		= 100,
 };
 
diff --git a/drivers/media/video/bt8xx/bttv-i2c.c b/drivers/media/video/bt8xx/bttv-i2c.c
index 4b562b386fcf..0dfbcc85ebb9 100644
--- a/drivers/media/video/bt8xx/bttv-i2c.c
+++ b/drivers/media/video/bt8xx/bttv-i2c.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data bttv_i2c_algo_bit_template = {
 	.getsda  = bttv_bit_getsda,
 	.getscl  = bttv_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/cx88/cx88-i2c.c b/drivers/media/video/cx88/cx88-i2c.c
index 70663805cc30..7bea34714861 100644
--- a/drivers/media/video/cx88/cx88-i2c.c
+++ b/drivers/media/video/cx88/cx88-i2c.c
@@ -155,7 +155,6 @@ static struct i2c_algo_bit_data cx8800_i2c_algo_template = {
 	.getsda  = cx8800_bit_getsda,
 	.getscl  = cx8800_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/cx88/cx88-vp3054-i2c.c b/drivers/media/video/cx88/cx88-vp3054-i2c.c
index 751a754a45e9..2b4f1970c7df 100644
--- a/drivers/media/video/cx88/cx88-vp3054-i2c.c
+++ b/drivers/media/video/cx88/cx88-vp3054-i2c.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data vp3054_i2c_algo_template = {
 	.getsda  = vp3054_bit_getsda,
 	.getscl  = vp3054_bit_getscl,
 	.udelay  = 16,
-	.mdelay  = 10,
 	.timeout = 200,
 };
 
diff --git a/drivers/media/video/zoran_card.c b/drivers/media/video/zoran_card.c
index f2249ed25273..29f59c36f001 100644
--- a/drivers/media/video/zoran_card.c
+++ b/drivers/media/video/zoran_card.c
@@ -820,7 +820,6 @@ static struct i2c_algo_bit_data zoran_i2c_bit_data_template = {
 	.getsda = zoran_i2c_getsda,
 	.getscl = zoran_i2c_getscl,
 	.udelay = 10,
-	.mdelay = 0,
 	.timeout = 100,
 };
 
diff --git a/drivers/video/i810/i810-i2c.c b/drivers/video/i810/i810-i2c.c
index c1f7b49975dd..7d06b38e80a0 100644
--- a/drivers/video/i810/i810-i2c.c
+++ b/drivers/video/i810/i810-i2c.c
@@ -98,7 +98,6 @@ static int i810_setup_i2c_bus(struct i810fb_i2c_chan *chan, const char *name)
 	chan->algo.getsda               = i810i2c_getsda;
 	chan->algo.getscl               = i810i2c_getscl;
 	chan->algo.udelay               = 10;
-	chan->algo.mdelay               = 10;
         chan->algo.timeout              = (HZ/2);
         chan->algo.data                 = chan;
 
diff --git a/drivers/video/matrox/i2c-matroxfb.c b/drivers/video/matrox/i2c-matroxfb.c
index 9842042d2af5..795c1a99a680 100644
--- a/drivers/video/matrox/i2c-matroxfb.c
+++ b/drivers/video/matrox/i2c-matroxfb.c
@@ -100,7 +100,6 @@ static struct i2c_algo_bit_data matrox_i2c_algo_template =
 	.getsda		= matroxfb_gpio_getsda,
 	.getscl		= matroxfb_gpio_getscl,
 	.udelay		= 10,
-	.mdelay		= 10,
 	.timeout	= 100,
 };
 
diff --git a/drivers/video/savage/savagefb-i2c.c b/drivers/video/savage/savagefb-i2c.c
index e83befd16d63..d7d810dbf0bd 100644
--- a/drivers/video/savage/savagefb-i2c.c
+++ b/drivers/video/savage/savagefb-i2c.c
@@ -148,7 +148,6 @@ static int savage_setup_i2c_bus(struct savagefb_i2c_chan *chan,
 		chan->adapter.algo_data		= &chan->algo;
 		chan->adapter.dev.parent	= &chan->par->pcidev->dev;
 		chan->algo.udelay		= 40;
-		chan->algo.mdelay               = 5;
 		chan->algo.timeout		= 20;
 		chan->algo.data 		= chan;
 
diff --git a/include/linux/i2c-algo-bit.h b/include/linux/i2c-algo-bit.h
index c0e7fab28ce3..c8f8df25c7e0 100644
--- a/include/linux/i2c-algo-bit.h
+++ b/include/linux/i2c-algo-bit.h
@@ -40,7 +40,6 @@ struct i2c_algo_bit_data {
 	/* local settings */
 	int udelay;		/* half-clock-cycle time in microsecs */
 				/* i.e. clock is (500 / udelay) KHz */
-	int mdelay;		/* in millisecs, unused */
 	int timeout;		/* in jiffies */
 };
 
-- 
cgit v1.2.1


From 010d442c4a295a73e90e93c5e42579cee61c5cc7 Mon Sep 17 00:00:00 2001
From: Komal Shah <komal_shah802003@yahoo.com>
Date: Sun, 13 Aug 2006 23:44:09 +0200
Subject: i2c: New bus driver for TI OMAP boards

i2c: New bus driver for TI OMAP boards

This patch adds I2C bus driver for various Texas
Instruments (TI) OMAP1/2 (http://www.ti.com/omap) series
based boards like OMAP1510/1610/1710/242x.

Signed-off-by: Komal Shah <komal_shah802003@yahoo.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/Kconfig    |  10 +
 drivers/i2c/busses/Makefile   |   1 +
 drivers/i2c/busses/i2c-omap.c | 676 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 687 insertions(+)
 create mode 100644 drivers/i2c/busses/i2c-omap.c

diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 404508ca7326..1df3bd97cb15 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -287,6 +287,16 @@ config I2C_OCORES
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-ocores.
 
+config I2C_OMAP
+	tristate "OMAP I2C adapter"
+	depends on I2C && ARCH_OMAP
+	default y if MACH_OMAP_H3 || MACH_OMAP_OSK
+	help
+	  If you say yes to this option, support will be included for the
+	  I2C interface on the Texas Instruments OMAP1/2 family of processors.
+	  Like OMAP1510/1610/1710/5912 and OMAP242x.
+	  For details see http://www.ti.com/omap.
+
 config I2C_PARPORT
 	tristate "Parallel port adapter"
 	depends on I2C && PARPORT
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index ac56df53155b..493c87289b62 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_I2C_MPC)		+= i2c-mpc.o
 obj-$(CONFIG_I2C_MV64XXX)	+= i2c-mv64xxx.o
 obj-$(CONFIG_I2C_NFORCE2)	+= i2c-nforce2.o
 obj-$(CONFIG_I2C_OCORES)	+= i2c-ocores.o
+obj-$(CONFIG_I2C_OMAP)		+= i2c-omap.o
 obj-$(CONFIG_I2C_PARPORT)	+= i2c-parport.o
 obj-$(CONFIG_I2C_PARPORT_LIGHT)	+= i2c-parport-light.o
 obj-$(CONFIG_I2C_PCA_ISA)	+= i2c-pca-isa.o
diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
new file mode 100644
index 000000000000..a129af991388
--- /dev/null
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -0,0 +1,676 @@
+/*
+ * TI OMAP I2C master mode driver
+ *
+ * Copyright (C) 2003 MontaVista Software, Inc.
+ * Copyright (C) 2004 Texas Instruments.
+ *
+ * Updated to work with multiple I2C interfaces on 24xx by
+ * Tony Lindgren <tony@atomide.com> and Imre Deak <imre.deak@nokia.com>
+ * Copyright (C) 2005 Nokia Corporation
+ *
+ * Cleaned up by Juha Yrjölä <juha.yrjola@nokia.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/completion.h>
+#include <linux/platform_device.h>
+#include <linux/clk.h>
+
+#include <asm/io.h>
+
+/* timeout waiting for the controller to respond */
+#define OMAP_I2C_TIMEOUT (msecs_to_jiffies(1000))
+
+#define OMAP_I2C_REV_REG		0x00
+#define OMAP_I2C_IE_REG			0x04
+#define OMAP_I2C_STAT_REG		0x08
+#define OMAP_I2C_IV_REG			0x0c
+#define OMAP_I2C_SYSS_REG		0x10
+#define OMAP_I2C_BUF_REG		0x14
+#define OMAP_I2C_CNT_REG		0x18
+#define OMAP_I2C_DATA_REG		0x1c
+#define OMAP_I2C_SYSC_REG		0x20
+#define OMAP_I2C_CON_REG		0x24
+#define OMAP_I2C_OA_REG			0x28
+#define OMAP_I2C_SA_REG			0x2c
+#define OMAP_I2C_PSC_REG		0x30
+#define OMAP_I2C_SCLL_REG		0x34
+#define OMAP_I2C_SCLH_REG		0x38
+#define OMAP_I2C_SYSTEST_REG		0x3c
+
+/* I2C Interrupt Enable Register (OMAP_I2C_IE): */
+#define OMAP_I2C_IE_XRDY	(1 << 4)	/* TX data ready int enable */
+#define OMAP_I2C_IE_RRDY	(1 << 3)	/* RX data ready int enable */
+#define OMAP_I2C_IE_ARDY	(1 << 2)	/* Access ready int enable */
+#define OMAP_I2C_IE_NACK	(1 << 1)	/* No ack interrupt enable */
+#define OMAP_I2C_IE_AL		(1 << 0)	/* Arbitration lost int ena */
+
+/* I2C Status Register (OMAP_I2C_STAT): */
+#define OMAP_I2C_STAT_SBD	(1 << 15)	/* Single byte data */
+#define OMAP_I2C_STAT_BB	(1 << 12)	/* Bus busy */
+#define OMAP_I2C_STAT_ROVR	(1 << 11)	/* Receive overrun */
+#define OMAP_I2C_STAT_XUDF	(1 << 10)	/* Transmit underflow */
+#define OMAP_I2C_STAT_AAS	(1 << 9)	/* Address as slave */
+#define OMAP_I2C_STAT_AD0	(1 << 8)	/* Address zero */
+#define OMAP_I2C_STAT_XRDY	(1 << 4)	/* Transmit data ready */
+#define OMAP_I2C_STAT_RRDY	(1 << 3)	/* Receive data ready */
+#define OMAP_I2C_STAT_ARDY	(1 << 2)	/* Register access ready */
+#define OMAP_I2C_STAT_NACK	(1 << 1)	/* No ack interrupt enable */
+#define OMAP_I2C_STAT_AL	(1 << 0)	/* Arbitration lost int ena */
+
+/* I2C Buffer Configuration Register (OMAP_I2C_BUF): */
+#define OMAP_I2C_BUF_RDMA_EN	(1 << 15)	/* RX DMA channel enable */
+#define OMAP_I2C_BUF_XDMA_EN	(1 << 7)	/* TX DMA channel enable */
+
+/* I2C Configuration Register (OMAP_I2C_CON): */
+#define OMAP_I2C_CON_EN		(1 << 15)	/* I2C module enable */
+#define OMAP_I2C_CON_BE		(1 << 14)	/* Big endian mode */
+#define OMAP_I2C_CON_STB	(1 << 11)	/* Start byte mode (master) */
+#define OMAP_I2C_CON_MST	(1 << 10)	/* Master/slave mode */
+#define OMAP_I2C_CON_TRX	(1 << 9)	/* TX/RX mode (master only) */
+#define OMAP_I2C_CON_XA		(1 << 8)	/* Expand address */
+#define OMAP_I2C_CON_RM		(1 << 2)	/* Repeat mode (master only) */
+#define OMAP_I2C_CON_STP	(1 << 1)	/* Stop cond (master only) */
+#define OMAP_I2C_CON_STT	(1 << 0)	/* Start condition (master) */
+
+/* I2C System Test Register (OMAP_I2C_SYSTEST): */
+#ifdef DEBUG
+#define OMAP_I2C_SYSTEST_ST_EN		(1 << 15)	/* System test enable */
+#define OMAP_I2C_SYSTEST_FREE		(1 << 14)	/* Free running mode */
+#define OMAP_I2C_SYSTEST_TMODE_MASK	(3 << 12)	/* Test mode select */
+#define OMAP_I2C_SYSTEST_TMODE_SHIFT	(12)		/* Test mode select */
+#define OMAP_I2C_SYSTEST_SCL_I		(1 << 3)	/* SCL line sense in */
+#define OMAP_I2C_SYSTEST_SCL_O		(1 << 2)	/* SCL line drive out */
+#define OMAP_I2C_SYSTEST_SDA_I		(1 << 1)	/* SDA line sense in */
+#define OMAP_I2C_SYSTEST_SDA_O		(1 << 0)	/* SDA line drive out */
+#endif
+
+/* I2C System Status register (OMAP_I2C_SYSS): */
+#define OMAP_I2C_SYSS_RDONE		(1 << 0)	/* Reset Done */
+
+/* I2C System Configuration Register (OMAP_I2C_SYSC): */
+#define OMAP_I2C_SYSC_SRST		(1 << 1)	/* Soft Reset */
+
+/* REVISIT: Use platform_data instead of module parameters */
+/* Fast Mode = 400 kHz, Standard = 100 kHz */
+static int clock = 100; /* Default: 100 kHz */
+module_param(clock, int, 0);
+MODULE_PARM_DESC(clock, "Set I2C clock in kHz: 400=fast mode (default == 100)");
+
+struct omap_i2c_dev {
+	struct device		*dev;
+	void __iomem		*base;		/* virtual */
+	int			irq;
+	struct clk		*iclk;		/* Interface clock */
+	struct clk		*fclk;		/* Functional clock */
+	struct completion	cmd_complete;
+	struct resource		*ioarea;
+	u16			cmd_err;
+	u8			*buf;
+	size_t			buf_len;
+	struct i2c_adapter	adapter;
+	unsigned		rev1:1;
+};
+
+static inline void omap_i2c_write_reg(struct omap_i2c_dev *i2c_dev,
+				      int reg, u16 val)
+{
+	__raw_writew(val, i2c_dev->base + reg);
+}
+
+static inline u16 omap_i2c_read_reg(struct omap_i2c_dev *i2c_dev, int reg)
+{
+	return __raw_readw(i2c_dev->base + reg);
+}
+
+static int omap_i2c_get_clocks(struct omap_i2c_dev *dev)
+{
+	if (cpu_is_omap16xx() || cpu_is_omap24xx()) {
+		dev->iclk = clk_get(dev->dev, "i2c_ick");
+		if (IS_ERR(dev->iclk)) {
+			dev->iclk = NULL;
+			return -ENODEV;
+		}
+	}
+
+	dev->fclk = clk_get(dev->dev, "i2c_fck");
+	if (IS_ERR(dev->fclk)) {
+		if (dev->iclk != NULL) {
+			clk_put(dev->iclk);
+			dev->iclk = NULL;
+		}
+		dev->fclk = NULL;
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static void omap_i2c_put_clocks(struct omap_i2c_dev *dev)
+{
+	clk_put(dev->fclk);
+	dev->fclk = NULL;
+	if (dev->iclk != NULL) {
+		clk_put(dev->iclk);
+		dev->iclk = NULL;
+	}
+}
+
+static void omap_i2c_enable_clocks(struct omap_i2c_dev *dev)
+{
+	if (dev->iclk != NULL)
+		clk_enable(dev->iclk);
+	clk_enable(dev->fclk);
+}
+
+static void omap_i2c_disable_clocks(struct omap_i2c_dev *dev)
+{
+	if (dev->iclk != NULL)
+		clk_disable(dev->iclk);
+	clk_disable(dev->fclk);
+}
+
+static int omap_i2c_init(struct omap_i2c_dev *dev)
+{
+	u16 psc = 0;
+	unsigned long fclk_rate = 12000000;
+	unsigned long timeout;
+
+	if (!dev->rev1) {
+		omap_i2c_write_reg(dev, OMAP_I2C_SYSC_REG, OMAP_I2C_SYSC_SRST);
+		/* For some reason we need to set the EN bit before the
+		 * reset done bit gets set. */
+		timeout = jiffies + OMAP_I2C_TIMEOUT;
+		omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_EN);
+		while (!(omap_i2c_read_reg(dev, OMAP_I2C_SYSS_REG) &
+			 OMAP_I2C_SYSS_RDONE)) {
+			if (time_after(jiffies, timeout)) {
+				dev_warn(dev->dev, "timeout waiting"
+						"for controller reset\n");
+				return -ETIMEDOUT;
+			}
+			msleep(1);
+		}
+	}
+	omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, 0);
+
+	if (cpu_class_is_omap1()) {
+		struct clk *armxor_ck;
+
+		armxor_ck = clk_get(NULL, "armxor_ck");
+		if (IS_ERR(armxor_ck))
+			dev_warn(dev->dev, "Could not get armxor_ck\n");
+		else {
+			fclk_rate = clk_get_rate(armxor_ck);
+			clk_put(armxor_ck);
+		}
+		/* TRM for 5912 says the I2C clock must be prescaled to be
+		 * between 7 - 12 MHz. The XOR input clock is typically
+		 * 12, 13 or 19.2 MHz. So we should have code that produces:
+		 *
+		 * XOR MHz	Divider		Prescaler
+		 * 12		1		0
+		 * 13		2		1
+		 * 19.2		2		1
+		 */
+		if (fclk_rate > 16000000)
+			psc = (fclk_rate + 8000000) / 12000000;
+	}
+
+	/* Setup clock prescaler to obtain approx 12MHz I2C module clock: */
+	omap_i2c_write_reg(dev, OMAP_I2C_PSC_REG, psc);
+
+	/* Program desired operating rate */
+	fclk_rate /= (psc + 1) * 1000;
+	if (psc > 2)
+		psc = 2;
+
+	omap_i2c_write_reg(dev, OMAP_I2C_SCLL_REG,
+			   fclk_rate / (clock * 2) - 7 + psc);
+	omap_i2c_write_reg(dev, OMAP_I2C_SCLH_REG,
+			   fclk_rate / (clock * 2) - 7 + psc);
+
+	/* Take the I2C module out of reset: */
+	omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_EN);
+
+	/* Enable interrupts */
+	omap_i2c_write_reg(dev, OMAP_I2C_IE_REG,
+			   (OMAP_I2C_IE_XRDY | OMAP_I2C_IE_RRDY |
+			    OMAP_I2C_IE_ARDY | OMAP_I2C_IE_NACK |
+			    OMAP_I2C_IE_AL));
+	return 0;
+}
+
+/*
+ * Waiting on Bus Busy
+ */
+static int omap_i2c_wait_for_bb(struct omap_i2c_dev *dev)
+{
+	unsigned long timeout;
+
+	timeout = jiffies + OMAP_I2C_TIMEOUT;
+	while (omap_i2c_read_reg(dev, OMAP_I2C_STAT_REG) & OMAP_I2C_STAT_BB) {
+		if (time_after(jiffies, timeout)) {
+			dev_warn(dev->dev, "timeout waiting for bus ready\n");
+			return -ETIMEDOUT;
+		}
+		msleep(1);
+	}
+
+	return 0;
+}
+
+/*
+ * Low level master read/write transaction.
+ */
+static int omap_i2c_xfer_msg(struct i2c_adapter *adap,
+			     struct i2c_msg *msg, int stop)
+{
+	struct omap_i2c_dev *dev = i2c_get_adapdata(adap);
+	int r;
+	u16 w;
+
+	dev_dbg(dev->dev, "addr: 0x%04x, len: %d, flags: 0x%x, stop: %d\n",
+		msg->addr, msg->len, msg->flags, stop);
+
+	if (msg->len == 0)
+		return -EINVAL;
+
+	omap_i2c_write_reg(dev, OMAP_I2C_SA_REG, msg->addr);
+
+	/* REVISIT: Could the STB bit of I2C_CON be used with probing? */
+	dev->buf = msg->buf;
+	dev->buf_len = msg->len;
+
+	omap_i2c_write_reg(dev, OMAP_I2C_CNT_REG, dev->buf_len);
+
+	init_completion(&dev->cmd_complete);
+	dev->cmd_err = 0;
+
+	w = OMAP_I2C_CON_EN | OMAP_I2C_CON_MST | OMAP_I2C_CON_STT;
+	if (msg->flags & I2C_M_TEN)
+		w |= OMAP_I2C_CON_XA;
+	if (!(msg->flags & I2C_M_RD))
+		w |= OMAP_I2C_CON_TRX;
+	if (stop)
+		w |= OMAP_I2C_CON_STP;
+	omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, w);
+
+	r = wait_for_completion_interruptible_timeout(&dev->cmd_complete,
+						      OMAP_I2C_TIMEOUT);
+	dev->buf_len = 0;
+	if (r < 0)
+		return r;
+	if (r == 0) {
+		dev_err(dev->dev, "controller timed out\n");
+		omap_i2c_init(dev);
+		return -ETIMEDOUT;
+	}
+
+	if (likely(!dev->cmd_err))
+		return 0;
+
+	/* We have an error */
+	if (dev->cmd_err & (OMAP_I2C_STAT_AL | OMAP_I2C_STAT_ROVR |
+			    OMAP_I2C_STAT_XUDF)) {
+		omap_i2c_init(dev);
+		return -EIO;
+	}
+
+	if (dev->cmd_err & OMAP_I2C_STAT_NACK) {
+		if (msg->flags & I2C_M_IGNORE_NAK)
+			return 0;
+		if (stop) {
+			w = omap_i2c_read_reg(dev, OMAP_I2C_CON_REG);
+			w |= OMAP_I2C_CON_STP;
+			omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, w);
+		}
+		return -EREMOTEIO;
+	}
+	return -EIO;
+}
+
+
+/*
+ * Prepare controller for a transaction and call omap_i2c_xfer_msg
+ * to do the work during IRQ processing.
+ */
+static int
+omap_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
+{
+	struct omap_i2c_dev *dev = i2c_get_adapdata(adap);
+	int i;
+	int r;
+
+	omap_i2c_enable_clocks(dev);
+
+	/* REVISIT: initialize and use adap->retries. This is an optional
+	 * feature */
+	if ((r = omap_i2c_wait_for_bb(dev)) < 0)
+		goto out;
+
+	for (i = 0; i < num; i++) {
+		r = omap_i2c_xfer_msg(adap, &msgs[i], (i == (num - 1)));
+		if (r != 0)
+			break;
+	}
+
+	if (r == 0)
+		r = num;
+out:
+	omap_i2c_disable_clocks(dev);
+	return r;
+}
+
+static u32
+omap_i2c_func(struct i2c_adapter *adap)
+{
+	return I2C_FUNC_I2C | (I2C_FUNC_SMBUS_EMUL & ~I2C_FUNC_SMBUS_QUICK);
+}
+
+static inline void
+omap_i2c_complete_cmd(struct omap_i2c_dev *dev, u16 err)
+{
+	dev->cmd_err |= err;
+	complete(&dev->cmd_complete);
+}
+
+static inline void
+omap_i2c_ack_stat(struct omap_i2c_dev *dev, u16 stat)
+{
+	omap_i2c_write_reg(dev, OMAP_I2C_STAT_REG, stat);
+}
+
+static irqreturn_t
+omap_i2c_rev1_isr(int this_irq, void *dev_id, struct pt_regs *regs)
+{
+	struct omap_i2c_dev *dev = dev_id;
+	u16 iv, w;
+
+	iv = omap_i2c_read_reg(dev, OMAP_I2C_IV_REG);
+	switch (iv) {
+	case 0x00:	/* None */
+		break;
+	case 0x01:	/* Arbitration lost */
+		dev_err(dev->dev, "Arbitration lost\n");
+		omap_i2c_complete_cmd(dev, OMAP_I2C_STAT_AL);
+		break;
+	case 0x02:	/* No acknowledgement */
+		omap_i2c_complete_cmd(dev, OMAP_I2C_STAT_NACK);
+		omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, OMAP_I2C_CON_STP);
+		break;
+	case 0x03:	/* Register access ready */
+		omap_i2c_complete_cmd(dev, 0);
+		break;
+	case 0x04:	/* Receive data ready */
+		if (dev->buf_len) {
+			w = omap_i2c_read_reg(dev, OMAP_I2C_DATA_REG);
+			*dev->buf++ = w;
+			dev->buf_len--;
+			if (dev->buf_len) {
+				*dev->buf++ = w >> 8;
+				dev->buf_len--;
+			}
+		} else
+			dev_err(dev->dev, "RRDY IRQ while no data requested\n");
+		break;
+	case 0x05:	/* Transmit data ready */
+		if (dev->buf_len) {
+			w = *dev->buf++;
+			dev->buf_len--;
+			if (dev->buf_len) {
+				w |= *dev->buf++ << 8;
+				dev->buf_len--;
+			}
+			omap_i2c_write_reg(dev, OMAP_I2C_DATA_REG, w);
+		} else
+			dev_err(dev->dev, "XRDY IRQ while no data to send\n");
+		break;
+	default:
+		return IRQ_NONE;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t
+omap_i2c_isr(int this_irq, void *dev_id, struct pt_regs *regs)
+{
+	struct omap_i2c_dev *dev = dev_id;
+	u16 bits;
+	u16 stat, w;
+	int count = 0;
+
+	bits = omap_i2c_read_reg(dev, OMAP_I2C_IE_REG);
+	while ((stat = (omap_i2c_read_reg(dev, OMAP_I2C_STAT_REG))) & bits) {
+		dev_dbg(dev->dev, "IRQ (ISR = 0x%04x)\n", stat);
+		if (count++ == 100) {
+			dev_warn(dev->dev, "Too much work in one IRQ\n");
+			break;
+		}
+
+		omap_i2c_write_reg(dev, OMAP_I2C_STAT_REG, stat);
+
+		if (stat & OMAP_I2C_STAT_ARDY) {
+			omap_i2c_complete_cmd(dev, 0);
+			continue;
+		}
+		if (stat & OMAP_I2C_STAT_RRDY) {
+			w = omap_i2c_read_reg(dev, OMAP_I2C_DATA_REG);
+			if (dev->buf_len) {
+				*dev->buf++ = w;
+				dev->buf_len--;
+				if (dev->buf_len) {
+					*dev->buf++ = w >> 8;
+					dev->buf_len--;
+				}
+			} else
+				dev_err(dev->dev, "RRDY IRQ while no data"
+						"requested\n");
+			omap_i2c_ack_stat(dev, OMAP_I2C_STAT_RRDY);
+			continue;
+		}
+		if (stat & OMAP_I2C_STAT_XRDY) {
+			w = 0;
+			if (dev->buf_len) {
+				w = *dev->buf++;
+				dev->buf_len--;
+				if (dev->buf_len) {
+					w |= *dev->buf++ << 8;
+					dev->buf_len--;
+				}
+			} else
+				dev_err(dev->dev, "XRDY IRQ while no"
+					"data to send\n");
+			omap_i2c_write_reg(dev, OMAP_I2C_DATA_REG, w);
+			omap_i2c_ack_stat(dev, OMAP_I2C_STAT_XRDY);
+			continue;
+		}
+		if (stat & OMAP_I2C_STAT_ROVR) {
+			dev_err(dev->dev, "Receive overrun\n");
+			dev->cmd_err |= OMAP_I2C_STAT_ROVR;
+		}
+		if (stat & OMAP_I2C_STAT_XUDF) {
+			dev_err(dev->dev, "Transmit overflow\n");
+			dev->cmd_err |= OMAP_I2C_STAT_XUDF;
+		}
+		if (stat & OMAP_I2C_STAT_NACK) {
+			omap_i2c_complete_cmd(dev, OMAP_I2C_STAT_NACK);
+			omap_i2c_write_reg(dev, OMAP_I2C_CON_REG,
+					   OMAP_I2C_CON_STP);
+		}
+		if (stat & OMAP_I2C_STAT_AL) {
+			dev_err(dev->dev, "Arbitration lost\n");
+			omap_i2c_complete_cmd(dev, OMAP_I2C_STAT_AL);
+		}
+	}
+
+	return count ? IRQ_HANDLED : IRQ_NONE;
+}
+
+static struct i2c_algorithm omap_i2c_algo = {
+	.master_xfer	= omap_i2c_xfer,
+	.functionality	= omap_i2c_func,
+};
+
+static int
+omap_i2c_probe(struct platform_device *pdev)
+{
+	struct omap_i2c_dev	*dev;
+	struct i2c_adapter	*adap;
+	struct resource		*mem, *irq, *ioarea;
+	int r;
+
+	/* NOTE: driver uses the static register mapping */
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!mem) {
+		dev_err(&pdev->dev, "no mem resource?\n");
+		return -ENODEV;
+	}
+	irq = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!irq) {
+		dev_err(&pdev->dev, "no irq resource?\n");
+		return -ENODEV;
+	}
+
+	ioarea = request_mem_region(mem->start, (mem->end - mem->start) + 1,
+			pdev->name);
+	if (!ioarea) {
+		dev_err(&pdev->dev, "I2C region already claimed\n");
+		return -EBUSY;
+	}
+
+	if (clock > 200)
+		clock = 400;	/* Fast mode */
+	else
+		clock = 100;	/* Standard mode */
+
+	dev = kzalloc(sizeof(struct omap_i2c_dev), GFP_KERNEL);
+	if (!dev) {
+		r = -ENOMEM;
+		goto err_release_region;
+	}
+
+	dev->dev = &pdev->dev;
+	dev->irq = irq->start;
+	dev->base = (void __iomem *) IO_ADDRESS(mem->start);
+	platform_set_drvdata(pdev, dev);
+
+	if ((r = omap_i2c_get_clocks(dev)) != 0)
+		goto err_free_mem;
+
+	omap_i2c_enable_clocks(dev);
+
+	if (cpu_is_omap15xx())
+		dev->rev1 = omap_i2c_read_reg(dev, OMAP_I2C_REV_REG) < 0x20;
+
+	/* reset ASAP, clearing any IRQs */
+	omap_i2c_init(dev);
+
+	r = request_irq(dev->irq, dev->rev1 ? omap_i2c_rev1_isr : omap_i2c_isr,
+			0, pdev->name, dev);
+
+	if (r) {
+		dev_err(dev->dev, "failure requesting irq %i\n", dev->irq);
+		goto err_unuse_clocks;
+	}
+	r = omap_i2c_read_reg(dev, OMAP_I2C_REV_REG) & 0xff;
+	dev_info(dev->dev, "bus %d rev%d.%d at %d kHz\n",
+		 pdev->id, r >> 4, r & 0xf, clock);
+
+	adap = &dev->adapter;
+	i2c_set_adapdata(adap, dev);
+	adap->owner = THIS_MODULE;
+	adap->class = I2C_CLASS_HWMON;
+	strncpy(adap->name, "OMAP I2C adapter", sizeof(adap->name));
+	adap->algo = &omap_i2c_algo;
+	adap->dev.parent = &pdev->dev;
+
+	/* i2c device drivers may be active on return from add_adapter() */
+	r = i2c_add_adapter(adap);
+	if (r) {
+		dev_err(dev->dev, "failure adding adapter\n");
+		goto err_free_irq;
+	}
+
+	omap_i2c_disable_clocks(dev);
+
+	return 0;
+
+err_free_irq:
+	free_irq(dev->irq, dev);
+err_unuse_clocks:
+	omap_i2c_disable_clocks(dev);
+	omap_i2c_put_clocks(dev);
+err_free_mem:
+	platform_set_drvdata(pdev, NULL);
+	kfree(dev);
+err_release_region:
+	omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, 0);
+	release_mem_region(mem->start, (mem->end - mem->start) + 1);
+
+	return r;
+}
+
+static int
+omap_i2c_remove(struct platform_device *pdev)
+{
+	struct omap_i2c_dev	*dev = platform_get_drvdata(pdev);
+	struct resource		*mem;
+
+	platform_set_drvdata(pdev, NULL);
+
+	free_irq(dev->irq, dev);
+	i2c_del_adapter(&dev->adapter);
+	omap_i2c_write_reg(dev, OMAP_I2C_CON_REG, 0);
+	omap_i2c_put_clocks(dev);
+	kfree(dev);
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	release_mem_region(mem->start, (mem->end - mem->start) + 1);
+	return 0;
+}
+
+static struct platform_driver omap_i2c_driver = {
+	.probe		= omap_i2c_probe,
+	.remove		= omap_i2c_remove,
+	.driver		= {
+		.name	= "i2c_omap",
+		.owner	= THIS_MODULE,
+	},
+};
+
+/* I2C may be needed to bring up other drivers */
+static int __init
+omap_i2c_init_driver(void)
+{
+	return platform_driver_register(&omap_i2c_driver);
+}
+subsys_initcall(omap_i2c_init_driver);
+
+static void __exit omap_i2c_exit_driver(void)
+{
+	platform_driver_unregister(&omap_i2c_driver);
+}
+module_exit(omap_i2c_exit_driver);
+
+MODULE_AUTHOR("MontaVista Software, Inc. (and others)");
+MODULE_DESCRIPTION("TI OMAP I2C bus adapter");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 6c805d2ce9d910ea915d7dbe4aed0a91f138be07 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:45:52 +0200
Subject: i2c: Plan i2c-isa for removal

i2c: Plan i2c-isa for removal

i2c-isa doesn't make much sense in the device driver model. Drivers
relying on it are better implemented as platform drivers. We must
wait for recent versions of libsensors (2.10.0 or later) to be widely
deployed beforehand, though. This move should also make it easier to
convert i2c-core to the device driver model.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/feature-removal-schedule.txt | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 611acc32fdf5..afe3d0197aca 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -321,3 +321,12 @@ Why:	The stacking of class devices makes these values misleading and
 Who:	Kay Sievers <kay.sievers@suse.de>
 
 ---------------------------
+
+What:	i2c-isa
+When:	December 2006
+Why:	i2c-isa is a non-sense and doesn't fit in the device driver
+	model. Drivers relying on it are better implemented as platform
+	drivers.
+Who:	Jean Delvare <khali@linux-fr.org>
+
+---------------------------
-- 
cgit v1.2.1


From 7a8d29cec7a53cf1a29dc5055aa9d1fa0f95830f Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 13 Aug 2006 23:46:44 +0200
Subject: i2c-stub: Chip address as a module parameter

i2c-stub: Chip address as a module parameter

Add a mandatory chip_addr parameter to i2c-stub. This parameter
defines to which chip address the driver will respond, instead of
reponding to all addresses as before. The idea is to prevent the
users from loading i2c-stub at random and being then confused by
the results of sensors-detect or other user-space tools.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mark M. Hoffman <mhoffman@lightlink.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/i2c/i2c-stub    | 15 +++++++++++++--
 drivers/i2c/busses/i2c-stub.c | 19 ++++++++++++++++++-
 2 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/Documentation/i2c/i2c-stub b/Documentation/i2c/i2c-stub
index d6dcb138abf5..9cc081e69764 100644
--- a/Documentation/i2c/i2c-stub
+++ b/Documentation/i2c/i2c-stub
@@ -6,9 +6,12 @@ This module is a very simple fake I2C/SMBus driver.  It implements four
 types of SMBus commands: write quick, (r/w) byte, (r/w) byte data, and
 (r/w) word data.
 
+You need to provide a chip address as a module parameter when loading
+this driver, which will then only react to SMBus commands to this address.
+
 No hardware is needed nor associated with this module.  It will accept write
-quick commands to all addresses; it will respond to the other commands (also
-to all addresses) by reading from or writing to an array in memory.  It will
+quick commands to one address; it will respond to the other commands (also
+to one address) by reading from or writing to an array in memory.  It will
 also spam the kernel logs for every command it handles.
 
 A pointer register with auto-increment is implemented for all byte
@@ -21,6 +24,11 @@ The typical use-case is like this:
 	3. load the target sensors chip driver module
 	4. observe its behavior in the kernel log
 
+PARAMETERS:
+
+int chip_addr:
+	The SMBus address to emulate a chip at.
+
 CAVEATS:
 
 There are independent arrays for byte/data and word/data commands.  Depending
@@ -33,6 +41,9 @@ If the hardware for your driver has banked registers (e.g. Winbond sensors
 chips) this module will not work well - although it could be extended to
 support that pretty easily.
 
+Only one chip address is supported - although this module could be
+extended to support more.
+
 If you spam it hard enough, printk can be lossy.  This module really wants
 something like relayfs.
 
diff --git a/drivers/i2c/busses/i2c-stub.c b/drivers/i2c/busses/i2c-stub.c
index 73f481e93a36..8cf374ddd989 100644
--- a/drivers/i2c/busses/i2c-stub.c
+++ b/drivers/i2c/busses/i2c-stub.c
@@ -27,6 +27,10 @@
 #include <linux/errno.h>
 #include <linux/i2c.h>
 
+static unsigned short chip_addr;
+module_param(chip_addr, ushort, S_IRUGO);
+MODULE_PARM_DESC(chip_addr, "Chip address (between 0x03 and 0x77)\n");
+
 static u8  stub_pointer;
 static u8  stub_bytes[256];
 static u16 stub_words[256];
@@ -37,6 +41,9 @@ static s32 stub_xfer(struct i2c_adapter * adap, u16 addr, unsigned short flags,
 {
 	s32 ret;
 
+	if (addr != chip_addr)
+		return -ENODEV;
+
 	switch (size) {
 
 	case I2C_SMBUS_QUICK:
@@ -122,7 +129,17 @@ static struct i2c_adapter stub_adapter = {
 
 static int __init i2c_stub_init(void)
 {
-	printk(KERN_INFO "i2c-stub loaded\n");
+	if (!chip_addr) {
+		printk(KERN_ERR "i2c-stub: Please specify a chip address\n");
+		return -ENODEV;
+	}
+	if (chip_addr < 0x03 || chip_addr > 0x77) {
+		printk(KERN_ERR "i2c-stub: Invalid chip address 0x%02x\n",
+		       chip_addr);
+		return -EINVAL;
+	}
+
+	printk(KERN_INFO "i2c-stub: Virtual chip at 0x%02x\n", chip_addr);
 	return i2c_add_adapter(&stub_adapter);
 }
 
-- 
cgit v1.2.1


From b32d20dc8b187e03605f091dbde9a78676a2a642 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:19:25 +0200
Subject: i2c-dev: attach/detach_adapter cleanups

i2c-dev: attach/detach_adapter cleanups

* Only print that an adapter was attached when it succeeds.
* i2c_dev == NULL on detach simply means that the attach failed
  before, this isn't an error per se.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-dev.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/i2c/i2c-dev.c b/drivers/i2c/i2c-dev.c
index 567fb05aeccc..3f869033ed70 100644
--- a/drivers/i2c/i2c-dev.c
+++ b/drivers/i2c/i2c-dev.c
@@ -412,9 +412,6 @@ static int i2cdev_attach_adapter(struct i2c_adapter *adap)
 	if (IS_ERR(i2c_dev))
 		return PTR_ERR(i2c_dev);
 
-	pr_debug("i2c-dev: adapter [%s] registered as minor %d\n",
-		 adap->name, adap->nr);
-
 	/* register this i2c device with the driver core */
 	i2c_dev->class_dev = class_device_create(i2c_dev_class, NULL,
 						 MKDEV(I2C_MAJOR, adap->nr),
@@ -427,6 +424,9 @@ static int i2cdev_attach_adapter(struct i2c_adapter *adap)
 	res = class_device_create_file(i2c_dev->class_dev, &class_device_attr_name);
 	if (res)
 		goto error_destroy;
+
+	pr_debug("i2c-dev: adapter [%s] registered as minor %d\n",
+		 adap->name, adap->nr);
 	return 0;
 error_destroy:
 	class_device_destroy(i2c_dev_class, MKDEV(I2C_MAJOR, adap->nr));
@@ -441,8 +441,8 @@ static int i2cdev_detach_adapter(struct i2c_adapter *adap)
 	struct i2c_dev *i2c_dev;
 
 	i2c_dev = i2c_dev_get_by_minor(adap->nr);
-	if (!i2c_dev)
-		return -ENODEV;
+	if (!i2c_dev) /* attach_adapter must have failed */
+		return 0;
 
 	class_device_remove_file(i2c_dev->class_dev, &class_device_attr_name);
 	return_i2c_dev(i2c_dev);
-- 
cgit v1.2.1


From 7d9db67febf67dd76329a9dd8f97cf4611a8ac2e Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:20:24 +0200
Subject: i2c: __must_check fixes (chip drivers)

i2c: __must_check fixes (chip drivers)

Check for error on sysfs file creation.
Delete sysfs files on device removal.

The approach taken for the most complex case (pcf8591) is similar to
what Mark M. Hoffman proposed for hardware monitoring chip drivers.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Ben Gardner <bgardner@wabtec.com>
Cc: Aurelien Jarno <aurelien@aurel32.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/chips/eeprom.c  |  8 ++++++-
 drivers/i2c/chips/max6875.c | 25 ++++++++++++++-----
 drivers/i2c/chips/pca9539.c | 11 +++++++--
 drivers/i2c/chips/pcf8574.c | 22 +++++++++++++----
 drivers/i2c/chips/pcf8591.c | 58 +++++++++++++++++++++++++++++++++++----------
 5 files changed, 97 insertions(+), 27 deletions(-)

diff --git a/drivers/i2c/chips/eeprom.c b/drivers/i2c/chips/eeprom.c
index 13c108269a6d..cec3a0c3894d 100644
--- a/drivers/i2c/chips/eeprom.c
+++ b/drivers/i2c/chips/eeprom.c
@@ -209,10 +209,14 @@ static int eeprom_detect(struct i2c_adapter *adapter, int address, int kind)
 	}
 
 	/* create the sysfs eeprom file */
-	sysfs_create_bin_file(&new_client->dev.kobj, &eeprom_attr);
+	err = sysfs_create_bin_file(&new_client->dev.kobj, &eeprom_attr);
+	if (err)
+		goto exit_detach;
 
 	return 0;
 
+exit_detach:
+	i2c_detach_client(new_client);
 exit_kfree:
 	kfree(data);
 exit:
@@ -223,6 +227,8 @@ static int eeprom_detach_client(struct i2c_client *client)
 {
 	int err;
 
+	sysfs_remove_bin_file(&client->dev.kobj, &eeprom_attr);
+
 	err = i2c_detach_client(client);
 	if (err)
 		return err;
diff --git a/drivers/i2c/chips/max6875.c b/drivers/i2c/chips/max6875.c
index 88d2ddee4490..76645c142977 100644
--- a/drivers/i2c/chips/max6875.c
+++ b/drivers/i2c/chips/max6875.c
@@ -199,8 +199,7 @@ static int max6875_detect(struct i2c_adapter *adapter, int address, int kind)
 	mutex_init(&data->update_lock);
 
 	/* Init fake client data */
-	/* set the client data to the i2c_client so that it will get freed */
-	i2c_set_clientdata(fake_client, fake_client);
+	i2c_set_clientdata(fake_client, NULL);
 	fake_client->addr = address | 1;
 	fake_client->adapter = adapter;
 	fake_client->driver = &max6875_driver;
@@ -214,13 +213,17 @@ static int max6875_detect(struct i2c_adapter *adapter, int address, int kind)
 		goto exit_kfree2;
 
 	if ((err = i2c_attach_client(fake_client)) != 0)
-		goto exit_detach;
+		goto exit_detach1;
 
-	sysfs_create_bin_file(&real_client->dev.kobj, &user_eeprom_attr);
+	err = sysfs_create_bin_file(&real_client->dev.kobj, &user_eeprom_attr);
+	if (err)
+		goto exit_detach2;
 
 	return 0;
 
-exit_detach:
+exit_detach2:
+	i2c_detach_client(fake_client);
+exit_detach1:
 	i2c_detach_client(real_client);
 exit_kfree2:
 	kfree(fake_client);
@@ -229,14 +232,24 @@ exit_kfree1:
 	return err;
 }
 
+/* Will be called for both the real client and the fake client */
 static int max6875_detach_client(struct i2c_client *client)
 {
 	int err;
+	struct max6875_data *data = i2c_get_clientdata(client);
+
+	/* data is NULL for the fake client */
+	if (data)
+		sysfs_remove_bin_file(&client->dev.kobj, &user_eeprom_attr);
 
 	err = i2c_detach_client(client);
 	if (err)
 		return err;
-	kfree(i2c_get_clientdata(client));
+
+	if (data)		/* real client */
+		kfree(data);
+	else			/* fake client */
+		kfree(client);
 	return 0;
 }
 
diff --git a/drivers/i2c/chips/pca9539.c b/drivers/i2c/chips/pca9539.c
index cb22280cdd27..f43c4e79b55e 100644
--- a/drivers/i2c/chips/pca9539.c
+++ b/drivers/i2c/chips/pca9539.c
@@ -148,11 +148,16 @@ static int pca9539_detect(struct i2c_adapter *adapter, int address, int kind)
 	if ((err = i2c_attach_client(new_client)))
 		goto exit_kfree;
 
-	/* Register sysfs hooks (don't care about failure) */
-	sysfs_create_group(&new_client->dev.kobj, &pca9539_defattr_group);
+	/* Register sysfs hooks */
+	err = sysfs_create_group(&new_client->dev.kobj,
+				 &pca9539_defattr_group);
+	if (err)
+		goto exit_detach;
 
 	return 0;
 
+exit_detach:
+	i2c_detach_client(new_client);
 exit_kfree:
 	kfree(data);
 exit:
@@ -163,6 +168,8 @@ static int pca9539_detach_client(struct i2c_client *client)
 {
 	int err;
 
+	sysfs_remove_group(&client->dev.kobj, &pca9539_defattr_group);
+
 	if ((err = i2c_detach_client(client)))
 		return err;
 
diff --git a/drivers/i2c/chips/pcf8574.c b/drivers/i2c/chips/pcf8574.c
index c3e6449c4481..32b25427eaba 100644
--- a/drivers/i2c/chips/pcf8574.c
+++ b/drivers/i2c/chips/pcf8574.c
@@ -105,6 +105,16 @@ static ssize_t set_write(struct device *dev, struct device_attribute *attr, cons
 
 static DEVICE_ATTR(write, S_IWUSR | S_IRUGO, show_write, set_write);
 
+static struct attribute *pcf8574_attributes[] = {
+	&dev_attr_read.attr,
+	&dev_attr_write.attr,
+	NULL
+};
+
+static const struct attribute_group pcf8574_attr_group = {
+	.attrs = pcf8574_attributes,
+};
+
 /*
  * Real code
  */
@@ -166,13 +176,13 @@ static int pcf8574_detect(struct i2c_adapter *adapter, int address, int kind)
 	pcf8574_init_client(new_client);
 
 	/* Register sysfs hooks */
-	device_create_file(&new_client->dev, &dev_attr_read);
-	device_create_file(&new_client->dev, &dev_attr_write);
+	err = sysfs_create_group(&new_client->dev.kobj, &pcf8574_attr_group);
+	if (err)
+		goto exit_detach;
 	return 0;
 
-/* OK, this is not exactly good programming practice, usually. But it is
-   very code-efficient in this case. */
-
+      exit_detach:
+	i2c_detach_client(new_client);
       exit_free:
 	kfree(data);
       exit:
@@ -183,6 +193,8 @@ static int pcf8574_detach_client(struct i2c_client *client)
 {
 	int err;
 
+	sysfs_remove_group(&client->dev.kobj, &pcf8574_attr_group);
+
 	if ((err = i2c_detach_client(client)))
 		return err;
 
diff --git a/drivers/i2c/chips/pcf8591.c b/drivers/i2c/chips/pcf8591.c
index 925a6b371fd2..4dc36376eb32 100644
--- a/drivers/i2c/chips/pcf8591.c
+++ b/drivers/i2c/chips/pcf8591.c
@@ -158,6 +158,28 @@ static ssize_t set_out0_enable(struct device *dev, struct device_attribute *attr
 static DEVICE_ATTR(out0_enable, S_IWUSR | S_IRUGO, 
 		   show_out0_enable, set_out0_enable);
 
+static struct attribute *pcf8591_attributes[] = {
+	&dev_attr_out0_enable.attr,
+	&dev_attr_out0_output.attr,
+	&dev_attr_in0_input.attr,
+	&dev_attr_in1_input.attr,
+	NULL
+};
+
+static const struct attribute_group pcf8591_attr_group = {
+	.attrs = pcf8591_attributes,
+};
+
+static struct attribute *pcf8591_attributes_opt[] = {
+	&dev_attr_in2_input.attr,
+	&dev_attr_in3_input.attr,
+	NULL
+};
+
+static const struct attribute_group pcf8591_attr_group_opt = {
+	.attrs = pcf8591_attributes_opt,
+};
+
 /*
  * Real code
  */
@@ -211,24 +233,31 @@ static int pcf8591_detect(struct i2c_adapter *adapter, int address, int kind)
 	pcf8591_init_client(new_client);
 
 	/* Register sysfs hooks */
-	device_create_file(&new_client->dev, &dev_attr_out0_enable);
-	device_create_file(&new_client->dev, &dev_attr_out0_output);
-	device_create_file(&new_client->dev, &dev_attr_in0_input);
-	device_create_file(&new_client->dev, &dev_attr_in1_input);
+	err = sysfs_create_group(&new_client->dev.kobj, &pcf8591_attr_group);
+	if (err)
+		goto exit_detach;
 
 	/* Register input2 if not in "two differential inputs" mode */
-	if (input_mode != 3 )
-		device_create_file(&new_client->dev, &dev_attr_in2_input);
-		
+	if (input_mode != 3) {
+		if ((err = device_create_file(&new_client->dev,
+					      &dev_attr_in2_input)))
+			goto exit_sysfs_remove;
+	}
+
 	/* Register input3 only in "four single ended inputs" mode */
-	if (input_mode == 0)
-		device_create_file(&new_client->dev, &dev_attr_in3_input);
-	
+	if (input_mode == 0) {
+		if ((err = device_create_file(&new_client->dev,
+					      &dev_attr_in3_input)))
+			goto exit_sysfs_remove;
+	}
+
 	return 0;
-	
-	/* OK, this is not exactly good programming practice, usually. But it is
-	   very code-efficient in this case. */
 
+exit_sysfs_remove:
+	sysfs_remove_group(&new_client->dev.kobj, &pcf8591_attr_group_opt);
+	sysfs_remove_group(&new_client->dev.kobj, &pcf8591_attr_group);
+exit_detach:
+	i2c_detach_client(new_client);
 exit_kfree:
 	kfree(data);
 exit:
@@ -239,6 +268,9 @@ static int pcf8591_detach_client(struct i2c_client *client)
 {
 	int err;
 
+	sysfs_remove_group(&client->dev.kobj, &pcf8591_attr_group_opt);
+	sysfs_remove_group(&client->dev.kobj, &pcf8591_attr_group);
+
 	if ((err = i2c_detach_client(client)))
 		return err;
 
-- 
cgit v1.2.1


From 8202632647278eba7223727dc442f49227c040d0 Mon Sep 17 00:00:00 2001
From: David Hubbard <david.c.hubbard@gmail.com>
Date: Sun, 3 Sep 2006 22:21:20 +0200
Subject: i2c-isa: Fail adding driver on attach_adapter error

i2c-isa: Fail adding driver on attach_adapter error

Signed-off-by: David Hubbard <david.c.hubbard@gmail.com>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-isa.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c
index d7486e586068..7a8d2c148a80 100644
--- a/drivers/i2c/busses/i2c-isa.c
+++ b/drivers/i2c/busses/i2c-isa.c
@@ -89,9 +89,14 @@ int i2c_isa_add_driver(struct i2c_driver *driver)
 	dev_dbg(&isa_adapter.dev, "Driver %s registered\n", driver->driver.name);
 
 	/* Now look for clients */
-	driver->attach_adapter(&isa_adapter);
-
-	return 0;
+	res = driver->attach_adapter(&isa_adapter);
+	if (res) {
+		dev_err(&isa_adapter.dev,
+			"Driver %s failed to attach adapter, unregistering\n",
+			driver->driver.name);
+		driver_unregister(&driver->driver);
+	}
+	return res;
 }
 
 int i2c_isa_del_driver(struct i2c_driver *driver)
-- 
cgit v1.2.1


From 7b288a018ac563f5babe0818f581d8f8a4fdcbfb Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:22:12 +0200
Subject: i2c-algo-bit: Cleanups

i2c-algo-bit: Cleanups

* Uninline long functions (saves around 1 kB or 15%)
* Refactor code in sclhi()
* Drop redundant udelay on repeated start

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/algos/i2c-algo-bit.c | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)

diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c
index 761df16838b4..48fe3a3b0b51 100644
--- a/drivers/i2c/algos/i2c-algo-bit.c
+++ b/drivers/i2c/algos/i2c-algo-bit.c
@@ -76,17 +76,15 @@ static inline void scllo(struct i2c_algo_bit_data *adap)
  * Raise scl line, and do checking for delays. This is necessary for slower
  * devices.
  */
-static inline int sclhi(struct i2c_algo_bit_data *adap)
+static int sclhi(struct i2c_algo_bit_data *adap)
 {
 	unsigned long start;
 
 	setscl(adap,1);
 
 	/* Not all adapters have scl sense line... */
-	if (adap->getscl == NULL ) {
-		udelay(adap->udelay);
-		return 0;
-	}
+	if (!adap->getscl)
+		goto done;
 
 	start=jiffies;
 	while (! getscl(adap) ) {	
@@ -101,6 +99,8 @@ static inline int sclhi(struct i2c_algo_bit_data *adap)
 		cond_resched();
 	}
 	DEBSTAT(printk(KERN_DEBUG "needed %ld jiffies\n", jiffies-start));
+
+done:
 	udelay(adap->udelay);
 	return 0;
 } 
@@ -121,7 +121,6 @@ static void i2c_repstart(struct i2c_algo_bit_data *adap)
 	DEBPROTO(printk(" Sr "));
 	setsda(adap,1);
 	sclhi(adap);
-	udelay(adap->udelay);
 	
 	sdalo(adap);
 	scllo(adap);
@@ -306,7 +305,7 @@ bailout:
  * 0 chip did not answer
  * -x transmission error
  */
-static inline int try_address(struct i2c_adapter *i2c_adap,
+static int try_address(struct i2c_adapter *i2c_adap,
 		       unsigned char addr, int retries)
 {
 	struct i2c_algo_bit_data *adap = i2c_adap->algo_data;
@@ -358,7 +357,7 @@ static int sendbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
 	return wrcount;
 }
 
-static inline int readbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
+static int readbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
 {
 	int inval;
 	int rdcount=0;   	/* counts bytes read */
@@ -408,7 +407,7 @@ static inline int readbytes(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
  * -x an error occurred (like: -EREMOTEIO if the device did not answer, or
  *	-ETIMEDOUT, for example if the lines are stuck...) 
  */
-static inline int bit_doAddress(struct i2c_adapter *i2c_adap, struct i2c_msg *msg) 
+static int bit_doAddress(struct i2c_adapter *i2c_adap, struct i2c_msg *msg)
 {
 	unsigned short flags = msg->flags;
 	unsigned short nak_ok = msg->flags & I2C_M_IGNORE_NAK;
-- 
cgit v1.2.1


From 9b4ccb86b4abe644ffd218720da2f942b6a20fc2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sun, 3 Sep 2006 22:22:50 +0200
Subject: i2c-algo-pcf: Discard the mdelay data struct member

i2c-algo-pcf: Discard the mdelay data struct member

Just as i2c-algo-bit, i2c-algo-pcf has an unused mdelay struct member,
which we can get rid of to spare some code and memory.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-elektor.c | 1 -
 include/linux/i2c-algo-pcf.h     | 1 -
 2 files changed, 2 deletions(-)

diff --git a/drivers/i2c/busses/i2c-elektor.c b/drivers/i2c/busses/i2c-elektor.c
index 59f8308c2356..caa8e5c8bfbb 100644
--- a/drivers/i2c/busses/i2c-elektor.c
+++ b/drivers/i2c/busses/i2c-elektor.c
@@ -196,7 +196,6 @@ static struct i2c_algo_pcf_data pcf_isa_data = {
 	.getclock   = pcf_isa_getclock,
 	.waitforpin = pcf_isa_waitforpin,
 	.udelay	    = 10,
-	.mdelay	    = 10,
 	.timeout    = 100,
 };
 
diff --git a/include/linux/i2c-algo-pcf.h b/include/linux/i2c-algo-pcf.h
index 18b0adf57a3d..9908f3fc4839 100644
--- a/include/linux/i2c-algo-pcf.h
+++ b/include/linux/i2c-algo-pcf.h
@@ -35,7 +35,6 @@ struct i2c_algo_pcf_data {
 
 	/* local settings */
 	int udelay;
-	int mdelay;
 	int timeout;
 };
 
-- 
cgit v1.2.1


From 7eff82c8b1511017ae605f0c99ac275a7e21b867 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:24:00 +0200
Subject: i2c-core: Drop useless bitmaskings

i2c-core: Drop useless bitmaskings

The code generated is exactly the same.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-core.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 36e1214b0b1d..88dd803174a1 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -901,7 +901,7 @@ s32 i2c_smbus_read_byte(struct i2c_client *client)
 	                   I2C_SMBUS_READ,0,I2C_SMBUS_BYTE, &data))
 		return -1;
 	else
-		return 0x0FF & data.byte;
+		return data.byte;
 }
 
 s32 i2c_smbus_write_byte(struct i2c_client *client, u8 value)
@@ -917,7 +917,7 @@ s32 i2c_smbus_read_byte_data(struct i2c_client *client, u8 command)
 	                   I2C_SMBUS_READ,command, I2C_SMBUS_BYTE_DATA,&data))
 		return -1;
 	else
-		return 0x0FF & data.byte;
+		return data.byte;
 }
 
 s32 i2c_smbus_write_byte_data(struct i2c_client *client, u8 command, u8 value)
@@ -936,7 +936,7 @@ s32 i2c_smbus_read_word_data(struct i2c_client *client, u8 command)
 	                   I2C_SMBUS_READ,command, I2C_SMBUS_WORD_DATA, &data))
 		return -1;
 	else
-		return 0x0FFFF & data.word;
+		return data.word;
 }
 
 s32 i2c_smbus_write_word_data(struct i2c_client *client, u8 command, u16 value)
@@ -1039,7 +1039,7 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr,
 		else {
 			msg[0].len=3;
 			msgbuf0[1] = data->word & 0xff;
-			msgbuf0[2] = (data->word >> 8) & 0xff;
+			msgbuf0[2] = data->word >> 8;
 		}
 		break;
 	case I2C_SMBUS_PROC_CALL:
@@ -1048,7 +1048,7 @@ static s32 i2c_smbus_xfer_emulated(struct i2c_adapter * adapter, u16 addr,
 		msg[0].len = 3;
 		msg[1].len = 2;
 		msgbuf0[1] = data->word & 0xff;
-		msgbuf0[2] = (data->word >> 8) & 0xff;
+		msgbuf0[2] = data->word >> 8;
 		break;
 	case I2C_SMBUS_BLOCK_DATA:
 		if (read_write == I2C_SMBUS_READ) {
-- 
cgit v1.2.1


From 114fd18397eb0eacf51ac784f7d5c929b8499715 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:25:04 +0200
Subject: i2c: Warn on i2c client creation failure

i2c: Warn on i2c client creation failure

Warn when an i2c client creation fails. If we don't, the user will
never know something wrong happened, as i2c client creation is
typically called through an attach_adapter callback, those return value
we currently ignore for technical reasons.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/i2c-core.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
index 88dd803174a1..01233f0f7771 100644
--- a/drivers/i2c/i2c-core.c
+++ b/drivers/i2c/i2c-core.c
@@ -707,11 +707,16 @@ static int i2c_probe_address(struct i2c_adapter *adapter, int addr, int kind,
 
 	/* Finally call the custom detection function */
 	err = found_proc(adapter, addr, kind);
-
 	/* -ENODEV can be returned if there is a chip at the given address
 	   but it isn't supported by this chip driver. We catch it here as
 	   this isn't an error. */
-	return (err == -ENODEV) ? 0 : err;
+	if (err == -ENODEV)
+		err = 0;
+
+	if (err)
+		dev_warn(&adapter->dev, "Client creation failed at 0x%x (%d)\n",
+			 addr, err);
+	return err;
 }
 
 int i2c_probe(struct i2c_adapter *adapter,
-- 
cgit v1.2.1


From c243353a90fae3a9a85d2bd79b1df06bb21c568a Mon Sep 17 00:00:00 2001
From: Rudolf Marek <r.marek@sh.cvut.cz>
Date: Sun, 3 Sep 2006 22:35:21 +0200
Subject: i2c-viapro: Add support for the VT8237A and VT8251

i2c-viapro: Add support for the VT8237A and VT8251

Documentation update included. Compile tested.

Signed-off-by: Rudolf Marek <r.marek@sh.cvut.cz>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/i2c/busses/i2c-viapro |  7 ++++++-
 drivers/i2c/busses/Kconfig          | 18 +++++++++---------
 drivers/i2c/busses/i2c-viapro.c     |  8 ++++++++
 3 files changed, 23 insertions(+), 10 deletions(-)

diff --git a/Documentation/i2c/busses/i2c-viapro b/Documentation/i2c/busses/i2c-viapro
index 16775663b9f5..25680346e0ac 100644
--- a/Documentation/i2c/busses/i2c-viapro
+++ b/Documentation/i2c/busses/i2c-viapro
@@ -7,9 +7,12 @@ Supported adapters:
   * VIA Technologies, Inc. VT82C686A/B
     Datasheet: Sometimes available at the VIA website
 
-  * VIA Technologies, Inc. VT8231, VT8233, VT8233A, VT8235, VT8237R
+  * VIA Technologies, Inc. VT8231, VT8233, VT8233A
     Datasheet: available on request from VIA
 
+  * VIA Technologies, Inc. VT8235, VT8237R, VT8237A, VT8251
+    Datasheet: available on request and under NDA from VIA
+
 Authors:
 	Kyösti Mälkki <kmalkki@cc.hut.fi>,
 	Mark D. Studebaker <mdsxyz123@yahoo.com>,
@@ -39,6 +42,8 @@ Your lspci -n listing must show one of these :
  device 1106:8235   (VT8231 function 4)
  device 1106:3177   (VT8235)
  device 1106:3227   (VT8237R)
+ device 1106:3337   (VT8237A)
+ device 1106:3287   (VT8251)
 
 If none of these show up, you should look in the BIOS for settings like
 enable ACPI / SMBus or even USB.
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 1df3bd97cb15..9e56c3989d68 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -492,19 +492,19 @@ config I2C_VIA
 	  will be called i2c-via.
 
 config I2C_VIAPRO
-	tristate "VIA 82C596/82C686/823x"
+	tristate "VIA 82C596/82C686/82xx"
 	depends on I2C && PCI
 	help
 	  If you say yes to this option, support will be included for the VIA
-	  82C596/82C686/823x I2C interfaces.  Specifically, the following 
+	  82C596/82C686/82xx I2C interfaces.  Specifically, the following
 	  chipsets are supported:
-	  82C596A/B
-	  82C686A/B
-	  8231
-	  8233
-	  8233A
-	  8235
-	  8237
+	    VT82C596A/B
+	    VT82C686A/B
+	    VT8231
+	    VT8233/A
+	    VT8235
+	    VT8237R/A
+	    VT8251
 
 	  This driver can also be built as a module.  If so, the module
 	  will be called i2c-viapro.
diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c
index 47e52bf2c5ec..1ccc0fbc5947 100644
--- a/drivers/i2c/busses/i2c-viapro.c
+++ b/drivers/i2c/busses/i2c-viapro.c
@@ -34,6 +34,8 @@
    VT8233A            0x3147             yes?
    VT8235             0x3177             yes
    VT8237R            0x3227             yes
+   VT8237A            0x3337             yes
+   VT8251             0x3287             yes
 
    Note: we assume there can only be one device, with one SMBus interface.
 */
@@ -381,7 +383,9 @@ found:
 	dev_dbg(&pdev->dev, "VT596_smba = 0x%X\n", vt596_smba);
 
 	switch (pdev->device) {
+	case PCI_DEVICE_ID_VIA_8251:
 	case PCI_DEVICE_ID_VIA_8237:
+	case PCI_DEVICE_ID_VIA_8237A:
 	case PCI_DEVICE_ID_VIA_8235:
 	case PCI_DEVICE_ID_VIA_8233A:
 	case PCI_DEVICE_ID_VIA_8233_0:
@@ -432,8 +436,12 @@ static struct pci_device_id vt596_ids[] = {
 	  .driver_data = SMBBA3 },
 	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237),
 	  .driver_data = SMBBA3 },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237A),
+	  .driver_data = SMBBA3 },
 	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231_4),
 	  .driver_data = SMBBA1 },
+	{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8251),
+	  .driver_data = SMBBA3 },
 	{ 0, }
 };
 
-- 
cgit v1.2.1


From 872188420997f7f7c1b968fd9bce6578e4c3d45f Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:36:14 +0200
Subject: i2c-isa: Restore driver owner

i2c-isa: Restore driver owner

Commit 2b48716d1d2f2edb1e7cbc5ecf1cb2cb39373e33 back in January
2006 was a bit overzealous. It removed .owner from all i2c drivers,
including i2c-isa ones, while they still need it.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hwmon/it87.c       | 1 +
 drivers/hwmon/lm78.c       | 1 +
 drivers/hwmon/pc87360.c    | 1 +
 drivers/hwmon/sis5595.c    | 1 +
 drivers/hwmon/smsc47b397.c | 1 +
 drivers/hwmon/smsc47m1.c   | 1 +
 drivers/hwmon/via686a.c    | 1 +
 drivers/hwmon/vt8231.c     | 1 +
 drivers/hwmon/w83627ehf.c  | 1 +
 drivers/hwmon/w83627hf.c   | 1 +
 drivers/hwmon/w83781d.c    | 1 +
 11 files changed, 11 insertions(+)

diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c
index 06df92b3ee49..b0ee57492228 100644
--- a/drivers/hwmon/it87.c
+++ b/drivers/hwmon/it87.c
@@ -243,6 +243,7 @@ static struct i2c_driver it87_driver = {
 
 static struct i2c_driver it87_isa_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "it87-isa",
 	},
 	.attach_adapter	= it87_isa_attach_adapter,
diff --git a/drivers/hwmon/lm78.c b/drivers/hwmon/lm78.c
index a6ce7abf8602..fa1715b9a996 100644
--- a/drivers/hwmon/lm78.c
+++ b/drivers/hwmon/lm78.c
@@ -175,6 +175,7 @@ static struct i2c_driver lm78_driver = {
 
 static struct i2c_driver lm78_isa_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "lm78-isa",
 	},
 	.attach_adapter	= lm78_isa_attach_adapter,
diff --git a/drivers/hwmon/pc87360.c b/drivers/hwmon/pc87360.c
index ae05e483a778..236f9f29c624 100644
--- a/drivers/hwmon/pc87360.c
+++ b/drivers/hwmon/pc87360.c
@@ -238,6 +238,7 @@ static struct pc87360_data *pc87360_update_device(struct device *dev);
 
 static struct i2c_driver pc87360_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "pc87360",
 	},
 	.attach_adapter	= pc87360_detect,
diff --git a/drivers/hwmon/sis5595.c b/drivers/hwmon/sis5595.c
index 063f71c5f07e..3783af4195bd 100644
--- a/drivers/hwmon/sis5595.c
+++ b/drivers/hwmon/sis5595.c
@@ -200,6 +200,7 @@ static void sis5595_init_client(struct i2c_client *client);
 
 static struct i2c_driver sis5595_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "sis5595",
 	},
 	.attach_adapter	= sis5595_detect,
diff --git a/drivers/hwmon/smsc47b397.c b/drivers/hwmon/smsc47b397.c
index b6086186d225..a85869393bab 100644
--- a/drivers/hwmon/smsc47b397.c
+++ b/drivers/hwmon/smsc47b397.c
@@ -228,6 +228,7 @@ static int smsc47b397_detect(struct i2c_adapter *adapter);
 
 static struct i2c_driver smsc47b397_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "smsc47b397",
 	},
 	.attach_adapter	= smsc47b397_detect,
diff --git a/drivers/hwmon/smsc47m1.c b/drivers/hwmon/smsc47m1.c
index 825e8f72698f..6c81b843d831 100644
--- a/drivers/hwmon/smsc47m1.c
+++ b/drivers/hwmon/smsc47m1.c
@@ -128,6 +128,7 @@ static struct smsc47m1_data *smsc47m1_update_device(struct device *dev,
 
 static struct i2c_driver smsc47m1_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "smsc47m1",
 	},
 	.attach_adapter	= smsc47m1_detect,
diff --git a/drivers/hwmon/via686a.c b/drivers/hwmon/via686a.c
index 166298f1f190..95ae056e5a94 100644
--- a/drivers/hwmon/via686a.c
+++ b/drivers/hwmon/via686a.c
@@ -574,6 +574,7 @@ static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL);
    smbus_driver and isa_driver, and clients could be of either kind */
 static struct i2c_driver via686a_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "via686a",
 	},
 	.attach_adapter	= via686a_detect,
diff --git a/drivers/hwmon/vt8231.c b/drivers/hwmon/vt8231.c
index 686f3deb3093..236ccf0e915d 100644
--- a/drivers/hwmon/vt8231.c
+++ b/drivers/hwmon/vt8231.c
@@ -587,6 +587,7 @@ static DEVICE_ATTR(alarms, S_IRUGO, show_alarms, NULL);
 
 static struct i2c_driver vt8231_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "vt8231",
 	},
 	.attach_adapter	= vt8231_detect,
diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c
index 40301bc6ce18..b21d6b9d7eac 100644
--- a/drivers/hwmon/w83627ehf.c
+++ b/drivers/hwmon/w83627ehf.c
@@ -903,6 +903,7 @@ static int w83627ehf_detach_client(struct i2c_client *client)
 
 static struct i2c_driver w83627ehf_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "w83627ehf",
 	},
 	.attach_adapter	= w83627ehf_detect,
diff --git a/drivers/hwmon/w83627hf.c b/drivers/hwmon/w83627hf.c
index 79368d53c363..30295028ea99 100644
--- a/drivers/hwmon/w83627hf.c
+++ b/drivers/hwmon/w83627hf.c
@@ -339,6 +339,7 @@ static void w83627hf_init_client(struct i2c_client *client);
 
 static struct i2c_driver w83627hf_driver = {
 	.driver = {
+		.owner	= THIS_MODULE,
 		.name	= "w83627hf",
 	},
 	.attach_adapter	= w83627hf_detect,
diff --git a/drivers/hwmon/w83781d.c b/drivers/hwmon/w83781d.c
index 7be469ed0f8f..95221b14e13a 100644
--- a/drivers/hwmon/w83781d.c
+++ b/drivers/hwmon/w83781d.c
@@ -288,6 +288,7 @@ static struct i2c_driver w83781d_driver = {
 
 static struct i2c_driver w83781d_isa_driver = {
 	.driver = {
+		.owner = THIS_MODULE,
 		.name = "w83781d-isa",
 	},
 	.attach_adapter = w83781d_isa_attach_adapter,
-- 
cgit v1.2.1


From af71ff690b92894f66ccede27f731150dc10d80d Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 3 Sep 2006 22:37:11 +0200
Subject: i2c: Let drivers constify i2c_algorithm data

i2c: Let drivers constify i2c_algorithm data

Let drivers constify I2C algorithm method operations tables,
moving them from ".data" to ".rodata".

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/i2c.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index eb0628a7ecc6..23ad1ee42a4c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -220,7 +220,7 @@ struct i2c_adapter {
 	struct module *owner;
 	unsigned int id;
 	unsigned int class;
-	struct i2c_algorithm *algo;/* the algorithm to access the bus	*/
+	const struct i2c_algorithm *algo; /* the algorithm to access the bus */
 	void *algo_data;
 
 	/* --- administration stuff. */
-- 
cgit v1.2.1


From 9e11a9fbfe48a2f65188aae64bf4f690e40ea2f4 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:38:52 +0200
Subject: i2c: Constify i2c_algorithm declarations, part 1

i2c: Constify i2c_algorithm declarations, part 1

Make struct i2c_algorithm declarations const in all i2c algorithm
drivers.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/algos/i2c-algo-bit.c | 2 +-
 drivers/i2c/algos/i2c-algo-pca.c | 2 +-
 drivers/i2c/algos/i2c-algo-pcf.c | 2 +-
 drivers/i2c/algos/i2c-algo-sgi.c | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/i2c/algos/i2c-algo-bit.c b/drivers/i2c/algos/i2c-algo-bit.c
index 48fe3a3b0b51..21c36bfb5e6b 100644
--- a/drivers/i2c/algos/i2c-algo-bit.c
+++ b/drivers/i2c/algos/i2c-algo-bit.c
@@ -512,7 +512,7 @@ static u32 bit_func(struct i2c_adapter *adap)
 
 /* -----exported algorithm data: -------------------------------------	*/
 
-static struct i2c_algorithm i2c_bit_algo = {
+static const struct i2c_algorithm i2c_bit_algo = {
 	.master_xfer	= bit_xfer,
 	.functionality	= bit_func,
 };
diff --git a/drivers/i2c/algos/i2c-algo-pca.c b/drivers/i2c/algos/i2c-algo-pca.c
index b88a6fcf7bd0..9081c9fbcd29 100644
--- a/drivers/i2c/algos/i2c-algo-pca.c
+++ b/drivers/i2c/algos/i2c-algo-pca.c
@@ -355,7 +355,7 @@ static int pca_init(struct i2c_algo_pca_data *adap)
 	return 0;
 }
 
-static struct i2c_algorithm pca_algo = {
+static const struct i2c_algorithm pca_algo = {
 	.master_xfer	= pca_xfer,
 	.functionality	= pca_func,
 };
diff --git a/drivers/i2c/algos/i2c-algo-pcf.c b/drivers/i2c/algos/i2c-algo-pcf.c
index 5b24930adb5a..3b2003398966 100644
--- a/drivers/i2c/algos/i2c-algo-pcf.c
+++ b/drivers/i2c/algos/i2c-algo-pcf.c
@@ -458,7 +458,7 @@ static u32 pcf_func(struct i2c_adapter *adap)
 
 /* -----exported algorithm data: -------------------------------------	*/
 
-static struct i2c_algorithm pcf_algo = {
+static const struct i2c_algorithm pcf_algo = {
 	.master_xfer	= pcf_xfer,
 	.functionality	= pcf_func,
 };
diff --git a/drivers/i2c/algos/i2c-algo-sgi.c b/drivers/i2c/algos/i2c-algo-sgi.c
index 932c4fa86c73..490d99997fd0 100644
--- a/drivers/i2c/algos/i2c-algo-sgi.c
+++ b/drivers/i2c/algos/i2c-algo-sgi.c
@@ -157,7 +157,7 @@ static u32 sgi_func(struct i2c_adapter *adap)
 	return I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm sgi_algo = {
+static const struct i2c_algorithm sgi_algo = {
 	.master_xfer	= sgi_xfer,
 	.functionality	= sgi_func,
 };
-- 
cgit v1.2.1


From 8f9082c5ce0e2c2f7ad0211b0c089f680d2efc11 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:39:46 +0200
Subject: i2c: Constify i2c_algorithm declarations, part 2

i2c: Constify i2c_algorithm declarations, part 2

Make struct i2c_algorithm declarations const in all i2c bus drivers
where it is possible.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/acpi/i2c_ec.c             | 2 +-
 drivers/i2c/busses/i2c-ali1535.c  | 2 +-
 drivers/i2c/busses/i2c-ali1563.c  | 2 +-
 drivers/i2c/busses/i2c-ali15x3.c  | 2 +-
 drivers/i2c/busses/i2c-amd756.c   | 2 +-
 drivers/i2c/busses/i2c-amd8111.c  | 2 +-
 drivers/i2c/busses/i2c-au1550.c   | 2 +-
 drivers/i2c/busses/i2c-i801.c     | 2 +-
 drivers/i2c/busses/i2c-ibm_iic.c  | 2 +-
 drivers/i2c/busses/i2c-iop3xx.c   | 2 +-
 drivers/i2c/busses/i2c-isa.c      | 2 +-
 drivers/i2c/busses/i2c-mpc.c      | 2 +-
 drivers/i2c/busses/i2c-mv64xxx.c  | 2 +-
 drivers/i2c/busses/i2c-nforce2.c  | 2 +-
 drivers/i2c/busses/i2c-ocores.c   | 2 +-
 drivers/i2c/busses/i2c-omap.c     | 2 +-
 drivers/i2c/busses/i2c-piix4.c    | 2 +-
 drivers/i2c/busses/i2c-powermac.c | 2 +-
 drivers/i2c/busses/i2c-pxa.c      | 2 +-
 drivers/i2c/busses/i2c-s3c2410.c  | 2 +-
 drivers/i2c/busses/i2c-sibyte.c   | 2 +-
 drivers/i2c/busses/i2c-sis5595.c  | 2 +-
 drivers/i2c/busses/i2c-sis630.c   | 2 +-
 drivers/i2c/busses/i2c-sis96x.c   | 2 +-
 drivers/i2c/busses/i2c-stub.c     | 2 +-
 drivers/i2c/busses/i2c-viapro.c   | 2 +-
 drivers/i2c/busses/scx200_acb.c   | 2 +-
 27 files changed, 27 insertions(+), 27 deletions(-)

diff --git a/drivers/acpi/i2c_ec.c b/drivers/acpi/i2c_ec.c
index 6809c283ec58..6342e612c203 100644
--- a/drivers/acpi/i2c_ec.c
+++ b/drivers/acpi/i2c_ec.c
@@ -293,7 +293,7 @@ static u32 acpi_ec_smb_func(struct i2c_adapter *adapter)
 		I2C_FUNC_SMBUS_I2C_BLOCK | I2C_FUNC_SMBUS_HWPEC_CALC);
 }
 
-static struct i2c_algorithm acpi_ec_smbus_algorithm = {
+static const struct i2c_algorithm acpi_ec_smbus_algorithm = {
 	.smbus_xfer = acpi_ec_smb_access,
 	.functionality = acpi_ec_smb_func,
 };
diff --git a/drivers/i2c/busses/i2c-ali1535.c b/drivers/i2c/busses/i2c-ali1535.c
index d3ef46aeeb3c..e75d339a3481 100644
--- a/drivers/i2c/busses/i2c-ali1535.c
+++ b/drivers/i2c/busses/i2c-ali1535.c
@@ -468,7 +468,7 @@ static u32 ali1535_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_BLOCK_DATA;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= ali1535_access,
 	.functionality	= ali1535_func,
 };
diff --git a/drivers/i2c/busses/i2c-ali1563.c b/drivers/i2c/busses/i2c-ali1563.c
index e6f63208fc4a..33fbb47100a3 100644
--- a/drivers/i2c/busses/i2c-ali1563.c
+++ b/drivers/i2c/busses/i2c-ali1563.c
@@ -367,7 +367,7 @@ static void ali1563_shutdown(struct pci_dev *dev)
 	release_region(ali1563_smba,ALI1563_SMB_IOSIZE);
 }
 
-static struct i2c_algorithm ali1563_algorithm = {
+static const struct i2c_algorithm ali1563_algorithm = {
 	.smbus_xfer	= ali1563_access,
 	.functionality	= ali1563_func,
 };
diff --git a/drivers/i2c/busses/i2c-ali15x3.c b/drivers/i2c/busses/i2c-ali15x3.c
index 7a5c0941dbc1..3f11b6e1a341 100644
--- a/drivers/i2c/busses/i2c-ali15x3.c
+++ b/drivers/i2c/busses/i2c-ali15x3.c
@@ -463,7 +463,7 @@ static u32 ali15x3_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_BLOCK_DATA;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= ali15x3_access,
 	.functionality	= ali15x3_func,
 };
diff --git a/drivers/i2c/busses/i2c-amd756.c b/drivers/i2c/busses/i2c-amd756.c
index 1750dedaf4b5..2d21afdc5b1c 100644
--- a/drivers/i2c/busses/i2c-amd756.c
+++ b/drivers/i2c/busses/i2c-amd756.c
@@ -294,7 +294,7 @@ static u32 amd756_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_BLOCK_DATA | I2C_FUNC_SMBUS_PROC_CALL;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= amd756_access,
 	.functionality	= amd756_func,
 };
diff --git a/drivers/i2c/busses/i2c-amd8111.c b/drivers/i2c/busses/i2c-amd8111.c
index e5ef560e686a..0fbc7186c91a 100644
--- a/drivers/i2c/busses/i2c-amd8111.c
+++ b/drivers/i2c/busses/i2c-amd8111.c
@@ -316,7 +316,7 @@ static u32 amd8111_func(struct i2c_adapter *adapter)
 		I2C_FUNC_SMBUS_I2C_BLOCK | I2C_FUNC_SMBUS_HWPEC_CALC;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer = amd8111_access,
 	.functionality = amd8111_func,
 };
diff --git a/drivers/i2c/busses/i2c-au1550.c b/drivers/i2c/busses/i2c-au1550.c
index 7a005ddfb11e..d7e7c359fc36 100644
--- a/drivers/i2c/busses/i2c-au1550.c
+++ b/drivers/i2c/busses/i2c-au1550.c
@@ -287,7 +287,7 @@ au1550_func(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm au1550_algo = {
+static const struct i2c_algorithm au1550_algo = {
 	.master_xfer	= au1550_xfer,
 	.functionality	= au1550_func,
 };
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index 7be1d0a3e8f8..bbb2fbee836f 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -434,7 +434,7 @@ static u32 i801_func(struct i2c_adapter *adapter)
 	     | (isich4 ? I2C_FUNC_SMBUS_HWPEC_CALC : 0);
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= i801_access,
 	.functionality	= i801_func,
 };
diff --git a/drivers/i2c/busses/i2c-ibm_iic.c b/drivers/i2c/busses/i2c-ibm_iic.c
index 0599bbd65d93..5bccb5d68318 100644
--- a/drivers/i2c/busses/i2c-ibm_iic.c
+++ b/drivers/i2c/busses/i2c-ibm_iic.c
@@ -625,7 +625,7 @@ static u32 iic_func(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL | I2C_FUNC_10BIT_ADDR;
 }
 
-static struct i2c_algorithm iic_algo = {
+static const struct i2c_algorithm iic_algo = {
 	.master_xfer 	= iic_xfer,
 	.functionality	= iic_func
 };
diff --git a/drivers/i2c/busses/i2c-iop3xx.c b/drivers/i2c/busses/i2c-iop3xx.c
index 48c56939c861..8e413150af37 100644
--- a/drivers/i2c/busses/i2c-iop3xx.c
+++ b/drivers/i2c/busses/i2c-iop3xx.c
@@ -401,7 +401,7 @@ iop3xx_i2c_func(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm iop3xx_i2c_algo = {
+static const struct i2c_algorithm iop3xx_i2c_algo = {
 	.master_xfer	= iop3xx_i2c_master_xfer,
 	.algo_control	= iop3xx_i2c_algo_control,
 	.functionality	= iop3xx_i2c_func,
diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c
index 7a8d2c148a80..4380653748a4 100644
--- a/drivers/i2c/busses/i2c-isa.c
+++ b/drivers/i2c/busses/i2c-isa.c
@@ -43,7 +43,7 @@
 static u32 isa_func(struct i2c_adapter *adapter);
 
 /* This is the actual algorithm we define */
-static struct i2c_algorithm isa_algorithm = {
+static const struct i2c_algorithm isa_algorithm = {
 	.functionality	= isa_func,
 };
 
diff --git a/drivers/i2c/busses/i2c-mpc.c b/drivers/i2c/busses/i2c-mpc.c
index 377ab40944b8..155a986de516 100644
--- a/drivers/i2c/busses/i2c-mpc.c
+++ b/drivers/i2c/busses/i2c-mpc.c
@@ -272,7 +272,7 @@ static u32 mpc_functionality(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm mpc_algo = {
+static const struct i2c_algorithm mpc_algo = {
 	.master_xfer = mpc_xfer,
 	.functionality = mpc_functionality,
 };
diff --git a/drivers/i2c/busses/i2c-mv64xxx.c b/drivers/i2c/busses/i2c-mv64xxx.c
index ac5cde1bbd2b..eacbaf745b64 100644
--- a/drivers/i2c/busses/i2c-mv64xxx.c
+++ b/drivers/i2c/busses/i2c-mv64xxx.c
@@ -431,7 +431,7 @@ mv64xxx_i2c_xfer(struct i2c_adapter *adap, struct i2c_msg msgs[], int num)
 	return num;
 }
 
-static struct i2c_algorithm mv64xxx_i2c_algo = {
+static const struct i2c_algorithm mv64xxx_i2c_algo = {
 	.master_xfer = mv64xxx_i2c_xfer,
 	.functionality = mv64xxx_i2c_functionality,
 };
diff --git a/drivers/i2c/busses/i2c-nforce2.c b/drivers/i2c/busses/i2c-nforce2.c
index 604b49e22df1..e0292e414ab2 100644
--- a/drivers/i2c/busses/i2c-nforce2.c
+++ b/drivers/i2c/busses/i2c-nforce2.c
@@ -109,7 +109,7 @@ static s32 nforce2_access(struct i2c_adapter *adap, u16 addr,
 static u32 nforce2_func(struct i2c_adapter *adapter);
 
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer = nforce2_access,
 	.functionality = nforce2_func,
 };
diff --git a/drivers/i2c/busses/i2c-ocores.c b/drivers/i2c/busses/i2c-ocores.c
index 592824087c49..952a28d485ce 100644
--- a/drivers/i2c/busses/i2c-ocores.c
+++ b/drivers/i2c/busses/i2c-ocores.c
@@ -199,7 +199,7 @@ static u32 ocores_func(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm ocores_algorithm = {
+static const struct i2c_algorithm ocores_algorithm = {
 	.master_xfer	= ocores_xfer,
 	.functionality	= ocores_func,
 };
diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index a129af991388..81d87d2c2a2d 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -526,7 +526,7 @@ omap_i2c_isr(int this_irq, void *dev_id, struct pt_regs *regs)
 	return count ? IRQ_HANDLED : IRQ_NONE;
 }
 
-static struct i2c_algorithm omap_i2c_algo = {
+static const struct i2c_algorithm omap_i2c_algo = {
 	.master_xfer	= omap_i2c_xfer,
 	.functionality	= omap_i2c_func,
 };
diff --git a/drivers/i2c/busses/i2c-piix4.c b/drivers/i2c/busses/i2c-piix4.c
index 8f2f65b793b9..30c7a1b38cbd 100644
--- a/drivers/i2c/busses/i2c-piix4.c
+++ b/drivers/i2c/busses/i2c-piix4.c
@@ -376,7 +376,7 @@ static u32 piix4_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_BLOCK_DATA;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= piix4_access,
 	.functionality	= piix4_func,
 };
diff --git a/drivers/i2c/busses/i2c-powermac.c b/drivers/i2c/busses/i2c-powermac.c
index d658d9107955..a508cb962d24 100644
--- a/drivers/i2c/busses/i2c-powermac.c
+++ b/drivers/i2c/busses/i2c-powermac.c
@@ -175,7 +175,7 @@ static u32 i2c_powermac_func(struct i2c_adapter * adapter)
 }
 
 /* For now, we only handle smbus */
-static struct i2c_algorithm i2c_powermac_algorithm = {
+static const struct i2c_algorithm i2c_powermac_algorithm = {
 	.smbus_xfer	= i2c_powermac_smbus_xfer,
 	.master_xfer	= i2c_powermac_master_xfer,
 	.functionality	= i2c_powermac_func,
diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c
index ee114b48face..cd4ad98ad517 100644
--- a/drivers/i2c/busses/i2c-pxa.c
+++ b/drivers/i2c/busses/i2c-pxa.c
@@ -926,7 +926,7 @@ static u32 i2c_pxa_functionality(struct i2c_adapter *adap)
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static struct i2c_algorithm i2c_pxa_algorithm = {
+static const struct i2c_algorithm i2c_pxa_algorithm = {
 	.master_xfer	= i2c_pxa_xfer,
 	.functionality	= i2c_pxa_functionality,
 };
diff --git a/drivers/i2c/busses/i2c-s3c2410.c b/drivers/i2c/busses/i2c-s3c2410.c
index 5d2950e91fc5..9ebe429a0a0f 100644
--- a/drivers/i2c/busses/i2c-s3c2410.c
+++ b/drivers/i2c/busses/i2c-s3c2410.c
@@ -566,7 +566,7 @@ static u32 s3c24xx_i2c_func(struct i2c_adapter *adap)
 
 /* i2c bus registration info */
 
-static struct i2c_algorithm s3c24xx_i2c_algorithm = {
+static const struct i2c_algorithm s3c24xx_i2c_algorithm = {
 	.master_xfer		= s3c24xx_i2c_xfer,
 	.functionality		= s3c24xx_i2c_func,
 };
diff --git a/drivers/i2c/busses/i2c-sibyte.c b/drivers/i2c/busses/i2c-sibyte.c
index 96c9e7bdd9e3..8f2b1f0deb81 100644
--- a/drivers/i2c/busses/i2c-sibyte.c
+++ b/drivers/i2c/busses/i2c-sibyte.c
@@ -132,7 +132,7 @@ static u32 bit_func(struct i2c_adapter *adap)
 
 /* -----exported algorithm data: -------------------------------------	*/
 
-static struct i2c_algorithm i2c_sibyte_algo = {
+static const struct i2c_algorithm i2c_sibyte_algo = {
 	.smbus_xfer	= smbus_xfer,
 	.functionality	= bit_func,
 };
diff --git a/drivers/i2c/busses/i2c-sis5595.c b/drivers/i2c/busses/i2c-sis5595.c
index b57ab74d23ec..38bbfd840b6b 100644
--- a/drivers/i2c/busses/i2c-sis5595.c
+++ b/drivers/i2c/busses/i2c-sis5595.c
@@ -358,7 +358,7 @@ static u32 sis5595_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_PROC_CALL;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= sis5595_access,
 	.functionality	= sis5595_func,
 };
diff --git a/drivers/i2c/busses/i2c-sis630.c b/drivers/i2c/busses/i2c-sis630.c
index acb75e282414..dec0bafb52ab 100644
--- a/drivers/i2c/busses/i2c-sis630.c
+++ b/drivers/i2c/busses/i2c-sis630.c
@@ -450,7 +450,7 @@ exit:
 }
 
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= sis630_access,
 	.functionality	= sis630_func,
 };
diff --git a/drivers/i2c/busses/i2c-sis96x.c b/drivers/i2c/busses/i2c-sis96x.c
index 1a73c0532fc7..7fd07fbac336 100644
--- a/drivers/i2c/busses/i2c-sis96x.c
+++ b/drivers/i2c/busses/i2c-sis96x.c
@@ -242,7 +242,7 @@ static u32 sis96x_func(struct i2c_adapter *adapter)
 	    I2C_FUNC_SMBUS_PROC_CALL;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= sis96x_access,
 	.functionality	= sis96x_func,
 };
diff --git a/drivers/i2c/busses/i2c-stub.c b/drivers/i2c/busses/i2c-stub.c
index 8cf374ddd989..a54adc50d162 100644
--- a/drivers/i2c/busses/i2c-stub.c
+++ b/drivers/i2c/busses/i2c-stub.c
@@ -115,7 +115,7 @@ static u32 stub_func(struct i2c_adapter *adapter)
 		I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.functionality	= stub_func,
 	.smbus_xfer	= stub_xfer,
 };
diff --git a/drivers/i2c/busses/i2c-viapro.c b/drivers/i2c/busses/i2c-viapro.c
index 1ccc0fbc5947..efc6bbf0cc0a 100644
--- a/drivers/i2c/busses/i2c-viapro.c
+++ b/drivers/i2c/busses/i2c-viapro.c
@@ -299,7 +299,7 @@ static u32 vt596_func(struct i2c_adapter *adapter)
 	return func;
 }
 
-static struct i2c_algorithm smbus_algorithm = {
+static const struct i2c_algorithm smbus_algorithm = {
 	.smbus_xfer	= vt596_access,
 	.functionality	= vt596_func,
 };
diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c
index eae9e81be375..32aab0d34ee9 100644
--- a/drivers/i2c/busses/scx200_acb.c
+++ b/drivers/i2c/busses/scx200_acb.c
@@ -383,7 +383,7 @@ static u32 scx200_acb_func(struct i2c_adapter *adapter)
 }
 
 /* For now, we only handle combined mode (smbus) */
-static struct i2c_algorithm scx200_acb_algorithm = {
+static const struct i2c_algorithm scx200_acb_algorithm = {
 	.smbus_xfer	= scx200_acb_smbus_xfer,
 	.functionality	= scx200_acb_func,
 };
-- 
cgit v1.2.1


From 6d3aae9d74221b00e2cbf50a353527e5a71a58ba Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 3 Sep 2006 22:41:08 +0200
Subject: i2c: Drop unimplemented slave functions

i2c: Drop unimplemented slave functions

Drop the function declarations for slave mode support of i2c adapters.
This was never implemented, and by the time it is I bet we will want
something different anyway.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/i2c.h | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/include/linux/i2c.h b/include/linux/i2c.h
index 23ad1ee42a4c..9b5d04768c2c 100644
--- a/include/linux/i2c.h
+++ b/include/linux/i2c.h
@@ -64,14 +64,6 @@ extern int i2c_master_recv(struct i2c_client *,char* ,int);
  */
 extern int i2c_transfer(struct i2c_adapter *adap, struct i2c_msg *msgs, int num);
 
-/*
- * Some adapter types (i.e. PCF 8584 based ones) may support slave behaviuor. 
- * This is not tested/implemented yet and will change in the future.
- */
-extern int i2c_slave_send(struct i2c_client *,char*,int);
-extern int i2c_slave_recv(struct i2c_client *,char*,int);
-
-
 
 /* This is the very generalized SMBus access routine. You probably do not
    want to use this, though; one of the functions below may be much easier,
@@ -201,10 +193,6 @@ struct i2c_algorithm {
 	                   unsigned short flags, char read_write,
 	                   u8 command, int size, union i2c_smbus_data * data);
 
-	/* --- these optional/future use for some adapter types.*/
-	int (*slave_send)(struct i2c_adapter *,char*,int);
-	int (*slave_recv)(struct i2c_adapter *,char*,int);
-
 	/* --- ioctl like call to set div. parameters. */
 	int (*algo_control)(struct i2c_adapter *, unsigned int, unsigned long);
 
-- 
cgit v1.2.1


From d33b6fba2c4350651f3f61ff2ab858a2f116e9a4 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Fri, 30 Jun 2006 02:31:24 -0700
Subject: Resources: insert identical resources above existing resources

If you have two resources which aree exactly the same size,
insert_resource() currently inserts the new one below the existing one.
This is wrong because there's no way to insert a resource of the same size
above an existing one.

I took this opportunity to rewrite the initial loop to be a for-loop
instead of a goto-loop and fix the documentation.

Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 kernel/resource.c | 32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/kernel/resource.c b/kernel/resource.c
index 46286434af80..9db38a1a7520 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -344,12 +344,11 @@ EXPORT_SYMBOL(allocate_resource);
  *
  * Returns 0 on success, -EBUSY if the resource can't be inserted.
  *
- * This function is equivalent of request_resource when no conflict
+ * This function is equivalent to request_resource when no conflict
  * happens. If a conflict happens, and the conflicting resources
  * entirely fit within the range of the new resource, then the new
- * resource is inserted and the conflicting resources become childs of
- * the new resource.  Otherwise the new resource becomes the child of
- * the conflicting resource
+ * resource is inserted and the conflicting resources become children of
+ * the new resource.
  */
 int insert_resource(struct resource *parent, struct resource *new)
 {
@@ -357,20 +356,21 @@ int insert_resource(struct resource *parent, struct resource *new)
 	struct resource *first, *next;
 
 	write_lock(&resource_lock);
- begin:
- 	result = 0;
-	first = __request_resource(parent, new);
-	if (!first)
-		goto out;
 
-	result = -EBUSY;
-	if (first == parent)
-		goto out;
+	for (;; parent = first) {
+	 	result = 0;
+		first = __request_resource(parent, new);
+		if (!first)
+			goto out;
 
-	/* Resource fully contained by the clashing resource? Recurse into it */
-	if (first->start <= new->start && first->end >= new->end) {
-		parent = first;
-		goto begin;
+		result = -EBUSY;
+		if (first == parent)
+			goto out;
+
+		if ((first->start > new->start) || (first->end < new->end))
+			break;
+		if ((first->start == new->start) && (first->end == new->end))
+			break;
 	}
 
 	for (next = first; ; next = next->sibling) {
-- 
cgit v1.2.1


From 3f79e107f72e8efa86cd2f21356692b712713b5c Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:54:56 -0400
Subject: MSI: Cleanup existing MSI quirks

Move MSI quirks in CONFIG_PCI_MSI, document why the serverworks quirk
does not simply set PCI_BUS_FLAGS_NO_MSI, and create a generic quirk
for other chipsets where setting PCI_BUS_FLAGS_NO_MSI is fine.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci.h    |  2 +-
 drivers/pci/quirks.c | 45 ++++++++++++++++++++++++++++++---------------
 2 files changed, 31 insertions(+), 16 deletions(-)

diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 08d58fc78ee1..6bf327db5c5e 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -42,7 +42,7 @@ extern void pci_remove_legacy_files(struct pci_bus *bus);
 /* Lock for read/write access to pci device and bus lists */
 extern struct rw_semaphore pci_bus_sem;
 
-#ifdef CONFIG_X86_IO_APIC
+#ifdef CONFIG_PCI_MSI
 extern int pci_msi_quirk;
 #else
 #define pci_msi_quirk 0
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index def78a2a7c15..8385b815ecb1 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -577,8 +577,6 @@ static void __init quirk_ioapic_rmw(struct pci_dev *dev)
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SI,	PCI_ANY_ID,			quirk_ioapic_rmw );
 
-int pci_msi_quirk;
-
 #define AMD8131_revA0        0x01
 #define AMD8131_revB0        0x11
 #define AMD8131_MISC         0x40
@@ -587,12 +585,6 @@ static void __init quirk_amd_8131_ioapic(struct pci_dev *dev)
 { 
         unsigned char revid, tmp;
         
-	if (dev->subordinate) {
-		printk(KERN_WARNING "PCI: MSI quirk detected. "
-		       "PCI_BUS_FLAGS_NO_MSI set for subordinate bus.\n");
-		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
-	}
-
         if (nr_ioapics == 0) 
                 return;
 
@@ -605,13 +597,6 @@ static void __init quirk_amd_8131_ioapic(struct pci_dev *dev)
         }
 } 
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_amd_8131_ioapic);
-
-static void __init quirk_svw_msi(struct pci_dev *dev)
-{
-	pci_msi_quirk = 1;
-	printk(KERN_WARNING "PCI: MSI quirk detected. pci_msi_quirk set.\n");
-}
-DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_GCNB_LE, quirk_svw_msi );
 #endif /* CONFIG_X86_IO_APIC */
 
 
@@ -1690,6 +1675,36 @@ static void __devinit quirk_nvidia_ck804_pcie_aer_ext_cap(struct pci_dev *dev)
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA,  PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
 			quirk_nvidia_ck804_pcie_aer_ext_cap);
 
+#ifdef CONFIG_PCI_MSI
+/* To disable MSI globally */
+int pci_msi_quirk;
+
+/* The Serverworks PCI-X chipset does not support MSI. We cannot easily rely
+ * on setting PCI_BUS_FLAGS_NO_MSI in its bus flags because there are actually
+ * some other busses controlled by the chipset even if Linux is not aware of it.
+ * Instead of setting the flag on all busses in the machine, simply disable MSI
+ * globally.
+ */
+static void __init quirk_svw_msi(struct pci_dev *dev)
+{
+	pci_msi_quirk = 1;
+	printk(KERN_WARNING "PCI: MSI quirk detected. pci_msi_quirk set.\n");
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_GCNB_LE, quirk_svw_msi);
+
+/* Disable MSI on chipsets that are known to not support it */
+static void __devinit quirk_disable_msi(struct pci_dev *dev)
+{
+	if (dev->subordinate) {
+		printk(KERN_WARNING "PCI: MSI quirk detected. "
+		       "PCI_BUS_FLAGS_NO_MSI set for %s subordinate bus.\n",
+		       pci_name(dev));
+		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+	}
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_disable_msi);
+#endif /* CONFIG_PCI_MSI */
+
 EXPORT_SYMBOL(pcie_mch_quirk);
 #ifdef CONFIG_HOTPLUG
 EXPORT_SYMBOL(pci_fixup_device);
-- 
cgit v1.2.1


From 24334a12533e9ac70dcb467ccd629f190afc5361 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:07 -0400
Subject: MSI: Factorize common code in pci_msi_supported()

pci_enable_msi() and pci_enable_msix() use the same code to detect
whether MSI might be enabled on this device. Factorize this code in
pci_msi_supported(). And improve the documentation about the fact
that only the root chipset must support MSI, but it is hard to
find the root bus so we check all parent busses MSI flags.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/msi.c | 51 +++++++++++++++++++++++++++++++--------------------
 1 file changed, 31 insertions(+), 20 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index a83c1f5735d6..008235947aa4 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -900,6 +900,33 @@ static int msix_capability_init(struct pci_dev *dev,
 	return 0;
 }
 
+/**
+ * pci_msi_supported - check whether MSI may be enabled on device
+ * @dev: pointer to the pci_dev data structure of MSI device function
+ *
+ * MSI must be globally enabled and supported by the device and its root
+ * bus. But, the root bus is not easy to find since some architectures
+ * have virtual busses on top of the PCI hierarchy (for instance the
+ * hypertransport bus), while the actual bus where MSI must be supported
+ * is below. So we test the MSI flag on all parent busses and assume
+ * that no quirk will ever set the NO_MSI flag on a non-root bus.
+ **/
+static
+int pci_msi_supported(struct pci_dev * dev)
+{
+	struct pci_bus *bus;
+
+	if (!pci_msi_enable || !dev || dev->no_msi)
+		return -EINVAL;
+
+	/* check MSI flags of all parent busses */
+	for (bus = dev->bus; bus; bus = bus->parent)
+		if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
+			return -EINVAL;
+
+	return 0;
+}
+
 /**
  * pci_enable_msi - configure device's MSI capability structure
  * @dev: pointer to the pci_dev data structure of MSI device function
@@ -912,19 +939,11 @@ static int msix_capability_init(struct pci_dev *dev,
  **/
 int pci_enable_msi(struct pci_dev* dev)
 {
-	struct pci_bus *bus;
-	int pos, temp, status = -EINVAL;
+	int pos, temp, status;
 	u16 control;
 
-	if (!pci_msi_enable || !dev)
- 		return status;
-
-	if (dev->no_msi)
-		return status;
-
-	for (bus = dev->bus; bus; bus = bus->parent)
-		if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
-			return -EINVAL;
+	if (pci_msi_supported(dev) < 0)
+		return -EINVAL;
 
 	temp = dev->irq;
 
@@ -1134,22 +1153,14 @@ static int reroute_msix_table(int head, struct msix_entry *entries, int *nvec)
  **/
 int pci_enable_msix(struct pci_dev* dev, struct msix_entry *entries, int nvec)
 {
-	struct pci_bus *bus;
 	int status, pos, nr_entries, free_vectors;
 	int i, j, temp;
 	u16 control;
 	unsigned long flags;
 
-	if (!pci_msi_enable || !dev || !entries)
+	if (!entries || pci_msi_supported(dev) < 0)
  		return -EINVAL;
 
-	if (dev->no_msi)
-		return -EINVAL;
-
-	for (bus = dev->bus; bus; bus = bus->parent)
-		if (bus->bus_flags & PCI_BUS_FLAGS_NO_MSI)
-			return -EINVAL;
-
 	status = msi_init();
 	if (status < 0)
 		return status;
-- 
cgit v1.2.1


From fe97064c2870e174a6ff4a93feb11a70c4b71cc5 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:15 -0400
Subject: MSI: Export the PCI_BUS_FLAGS_NO_MSI flag in sysfs

Export the PCI_BUS_FLAGS_NO_MSI flag of a PCI bus in the sysfs files
of its parent device and make it writable. Could be used to:
* disable MSI on a device which has not been blacklisted yet
* allow MSI when some setpci hacks enable MSI support (for instance
  on the ServerWorks HT2000 chipset where the MSI HT cap is disabled
  by default).
Architecture where some bus have no parent chipset cannot use this
strategy to change MSI support.

If the chipset does not have a subordinate bus, its 'bus_msi' file
is empty.

Also document and warn about the possible danger of changing the flag.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci-sysfs.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index fdefa7dcd156..010e01c4bd43 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -131,6 +131,46 @@ is_enabled_store(struct device *dev, struct device_attribute *attr,
 	return count;
 }
 
+static ssize_t
+msi_bus_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+
+	if (!pdev->subordinate)
+		return 0;
+
+	return sprintf (buf, "%u\n",
+			!(pdev->subordinate->bus_flags & PCI_BUS_FLAGS_NO_MSI));
+}
+
+static ssize_t
+msi_bus_store(struct device *dev, struct device_attribute *attr,
+	      const char *buf, size_t count)
+{
+	struct pci_dev *pdev = to_pci_dev(dev);
+
+	/* bad things may happen if the no_msi flag is changed
+	 * while some drivers are loaded */
+	if (!capable(CAP_SYS_ADMIN))
+		return count;
+
+	if (!pdev->subordinate)
+		return count;
+
+	if (*buf == '0') {
+		pdev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+		dev_warn(&pdev->dev, "forced subordinate bus to not support MSI,"
+			 " bad things could happen.\n");
+	}
+
+	if (*buf == '1') {
+		pdev->subordinate->bus_flags &= ~PCI_BUS_FLAGS_NO_MSI;
+		dev_warn(&pdev->dev, "forced subordinate bus to support MSI,"
+			 " bad things could happen.\n");
+	}
+
+	return count;
+}
 
 struct device_attribute pci_dev_attrs[] = {
 	__ATTR_RO(resource),
@@ -145,6 +185,7 @@ struct device_attribute pci_dev_attrs[] = {
 	__ATTR(enable, 0600, is_enabled_show, is_enabled_store),
 	__ATTR(broken_parity_status,(S_IRUGO|S_IWUSR),
 		broken_parity_status_show,broken_parity_status_store),
+	__ATTR(msi_bus, 0644, msi_bus_show, msi_bus_store),
 	__ATTR_NULL,
 };
 
-- 
cgit v1.2.1


From 46ff34633ed09f36ebc4b5c40ac37e592172df74 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:24 -0400
Subject: MSI: Rename PCI_CAP_ID_HT_IRQCONF into PCI_CAP_ID_HT

0x08 is the HT capability, while PCI_CAP_ID_HT_IRQCONF would be
the subtype 0x80 that mpic_scan_ht_pic() uses.
Rename PCI_CAP_ID_HT_IRQCONF into PCI_CAP_ID_HT.

And by the way, use it in the ipath driver instead of defining its
own HT_CAPABILITY_ID.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/powerpc/sysdev/mpic.c                  | 2 +-
 drivers/infiniband/hw/ipath/ipath_iba6110.c | 5 ++---
 include/linux/pci_regs.h                    | 2 +-
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index b604926401f5..723972bb5bd9 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -339,7 +339,7 @@ static void __init mpic_scan_ht_pic(struct mpic *mpic, u8 __iomem *devbase,
 	for (pos = readb(devbase + PCI_CAPABILITY_LIST); pos != 0;
 	     pos = readb(devbase + pos + PCI_CAP_LIST_NEXT)) {
 		u8 id = readb(devbase + pos + PCI_CAP_LIST_ID);
-		if (id == PCI_CAP_ID_HT_IRQCONF) {
+		if (id == PCI_CAP_ID_HT) {
 			id = readb(devbase + pos + 3);
 			if (id == 0x80)
 				break;
diff --git a/drivers/infiniband/hw/ipath/ipath_iba6110.c b/drivers/infiniband/hw/ipath/ipath_iba6110.c
index bf2455a6d562..5c9b509e40e4 100644
--- a/drivers/infiniband/hw/ipath/ipath_iba6110.c
+++ b/drivers/infiniband/hw/ipath/ipath_iba6110.c
@@ -742,7 +742,6 @@ static int ipath_setup_ht_reset(struct ipath_devdata *dd)
 	return 0;
 }
 
-#define HT_CAPABILITY_ID   0x08	/* HT capabilities not defined in kernel */
 #define HT_INTR_DISC_CONFIG  0x80	/* HT interrupt and discovery cap */
 #define HT_INTR_REG_INDEX    2	/* intconfig requires indirect accesses */
 
@@ -973,7 +972,7 @@ static int ipath_setup_ht_config(struct ipath_devdata *dd,
 	 * do this early, before we ever enable errors or hardware errors,
 	 * mostly to avoid causing the chip to enter freeze mode.
 	 */
-	pos = pci_find_capability(pdev, HT_CAPABILITY_ID);
+	pos = pci_find_capability(pdev, PCI_CAP_ID_HT);
 	if (!pos) {
 		ipath_dev_err(dd, "Couldn't find HyperTransport "
 			      "capability; no interrupts\n");
@@ -996,7 +995,7 @@ static int ipath_setup_ht_config(struct ipath_devdata *dd,
 		else if (cap_type == HT_INTR_DISC_CONFIG)
 			ihandler = set_int_handler(dd, pdev, pos);
 	} while ((pos = pci_find_next_capability(pdev, pos,
-						 HT_CAPABILITY_ID)));
+						 PCI_CAP_ID_HT)));
 
 	if (!ihandler) {
 		ipath_dev_err(dd, "Couldn't find interrupt handler in "
diff --git a/include/linux/pci_regs.h b/include/linux/pci_regs.h
index 96930cb5927c..7d0e26cba420 100644
--- a/include/linux/pci_regs.h
+++ b/include/linux/pci_regs.h
@@ -196,7 +196,7 @@
 #define  PCI_CAP_ID_MSI		0x05	/* Message Signalled Interrupts */
 #define  PCI_CAP_ID_CHSWP	0x06	/* CompactPCI HotSwap */
 #define  PCI_CAP_ID_PCIX	0x07	/* PCI-X */
-#define  PCI_CAP_ID_HT_IRQCONF	0x08	/* HyperTransport IRQ Configuration */
+#define  PCI_CAP_ID_HT		0x08	/* HyperTransport */
 #define  PCI_CAP_ID_VNDR	0x09	/* Vendor specific capability */
 #define  PCI_CAP_ID_SHPC 	0x0C	/* PCI Standard Hot-Plug Controller */
 #define  PCI_CAP_ID_EXP 	0x10	/* PCI Express */
-- 
cgit v1.2.1


From 6397c75cbc4d7dbc3d07278b57c82a47dafb21b5 Mon Sep 17 00:00:00 2001
From: Brice Goglin <brice@myri.com>
Date: Thu, 31 Aug 2006 01:55:32 -0400
Subject: MSI: Blacklist PCI-E chipsets depending on Hypertransport MSI
 capability

Introduce msi_ht_cap_enabled() to check the MSI capability in the
Hypertransport configuration space.
It is used in a generic quirk quirk_msi_ht_cap() to check whether
MSI is enabled on hypertransport chipset, and a nVidia specific quirk
quirk_nvidia_ck804_msi_ht_cap() where two 2 HT MSI mappings have to
be checked.
Both quirks set the PCI_BUS_FLAGS_NO_MSI bus flag when MSI is disabled.

Signed-off-by: Brice Goglin <brice@myri.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/quirks.c    | 59 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h |  1 +
 2 files changed, 60 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index 8385b815ecb1..08cd86a6dd66 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -1703,6 +1703,65 @@ static void __devinit quirk_disable_msi(struct pci_dev *dev)
 	}
 }
 DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE, quirk_disable_msi);
+
+/* Go through the list of Hypertransport capabilities and
+ * return 1 if a HT MSI capability is found and enabled */
+static int __devinit msi_ht_cap_enabled(struct pci_dev *dev)
+{
+	u8 pos;
+	int ttl;
+	for (pos = pci_find_capability(dev, PCI_CAP_ID_HT), ttl = 48;
+	     pos && ttl;
+	     pos = pci_find_next_capability(dev, pos, PCI_CAP_ID_HT), ttl--) {
+		u32 cap_hdr;
+		/* MSI mapping section according to Hypertransport spec */
+		if (pci_read_config_dword(dev, pos, &cap_hdr) == 0
+		    && (cap_hdr & 0xf8000000) == 0xa8000000 /* MSI mapping */) {
+			printk(KERN_INFO "PCI: Found HT MSI mapping on %s with capability %s\n",
+			       pci_name(dev), cap_hdr & 0x10000 ? "enabled" : "disabled");
+			return (cap_hdr & 0x10000) != 0; /* MSI mapping cap enabled */
+		}
+	}
+	return 0;
+}
+
+/* Check the hypertransport MSI mapping to know whether MSI is enabled or not */
+static void __devinit quirk_msi_ht_cap(struct pci_dev *dev)
+{
+	if (dev->subordinate && !msi_ht_cap_enabled(dev)) {
+		printk(KERN_WARNING "PCI: MSI quirk detected. "
+		       "MSI disabled on chipset %s.\n",
+		       pci_name(dev));
+		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+	}
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_SERVERWORKS, PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE,
+			quirk_msi_ht_cap);
+
+/* The nVidia CK804 chipset may have 2 HT MSI mappings.
+ * MSI are supported if the MSI capability set in any of these mappings.
+ */
+static void __devinit quirk_nvidia_ck804_msi_ht_cap(struct pci_dev *dev)
+{
+	struct pci_dev *pdev;
+
+	if (!dev->subordinate)
+		return;
+
+	/* check HT MSI cap on this chipset and the root one.
+	 * a single one having MSI is enough to be sure that MSI are supported.
+	 */
+	pdev = pci_find_slot(dev->bus->number, 0);
+	if (dev->subordinate && !msi_ht_cap_enabled(dev)
+	    && !msi_ht_cap_enabled(pdev)) {
+		printk(KERN_WARNING "PCI: MSI quirk detected. "
+		       "MSI disabled on chipset %s.\n",
+		       pci_name(dev));
+		dev->subordinate->bus_flags |= PCI_BUS_FLAGS_NO_MSI;
+	}
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_CK804_PCIE,
+			quirk_nvidia_ck804_msi_ht_cap);
 #endif /* CONFIG_PCI_MSI */
 
 EXPORT_SYMBOL(pcie_mch_quirk);
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 6a1e09834559..b9e263adebab 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1411,6 +1411,7 @@
 #define PCI_DEVICE_ID_SERVERWORKS_LE	  0x0009
 #define PCI_DEVICE_ID_SERVERWORKS_GCNB_LE 0x0017
 #define PCI_DEVICE_ID_SERVERWORKS_EPB	  0x0103
+#define PCI_DEVICE_ID_SERVERWORKS_HT2000_PCIE	0x0132
 #define PCI_DEVICE_ID_SERVERWORKS_OSB4	  0x0200
 #define PCI_DEVICE_ID_SERVERWORKS_CSB5	  0x0201
 #define PCI_DEVICE_ID_SERVERWORKS_CSB6    0x0203
-- 
cgit v1.2.1


From 20d516602c022997feb24a9f1a806fc986b9e4e8 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Sat, 8 Jul 2006 22:58:25 -0700
Subject: PCIE: check and return bus_register errors

Have pcie_port_bus_register() notice and return errors.
Mark it __must_check so that its caller(s) must check its return value.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pcie/portdrv.h      | 2 +-
 drivers/pci/pcie/portdrv_core.c | 5 +++--
 drivers/pci/pcie/portdrv_pci.c  | 9 +++++++--
 3 files changed, 11 insertions(+), 5 deletions(-)

diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h
index 1d317d22ee89..67fcd176babd 100644
--- a/drivers/pci/pcie/portdrv.h
+++ b/drivers/pci/pcie/portdrv.h
@@ -39,7 +39,7 @@ extern int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state);
 extern int pcie_port_device_resume(struct pci_dev *dev);
 #endif
 extern void pcie_port_device_remove(struct pci_dev *dev);
-extern void pcie_port_bus_register(void);
+extern int pcie_port_bus_register(void);
 extern void pcie_port_bus_unregister(void);
 
 #endif /* _PORTDRV_H_ */
diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c
index 55c662267868..cf9e810b4bf8 100644
--- a/drivers/pci/pcie/portdrv_core.c
+++ b/drivers/pci/pcie/portdrv_core.c
@@ -6,6 +6,7 @@
  * Copyright (C) Tom Long Nguyen (tom.l.nguyen@intel.com)
  */
 
+#include <linux/compiler.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/kernel.h>
@@ -402,9 +403,9 @@ void pcie_port_device_remove(struct pci_dev *dev)
 		pci_disable_msi(dev);
 }
 
-void pcie_port_bus_register(void)
+int __must_check pcie_port_bus_register(void)
 {
-	bus_register(&pcie_port_bus_type);
+	return bus_register(&pcie_port_bus_type);
 }
 
 void pcie_port_bus_unregister(void)
diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index 478d0d28f7ad..3284199ce396 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -129,12 +129,17 @@ static struct pci_driver pcie_portdrv = {
 
 static int __init pcie_portdrv_init(void)
 {
-	int retval = 0;
+	int retval;
 
-	pcie_port_bus_register();
+	retval = pcie_port_bus_register();
+	if (retval) {
+		printk(KERN_WARNING "PCIE: bus_register error: %d\n", retval);
+		goto out;
+	}
 	retval = pci_register_driver(&pcie_portdrv);
 	if (retval)
 		pcie_port_bus_unregister();
+ out:
 	return retval;
 }
 
-- 
cgit v1.2.1


From 47402400c6496dd114463deb3a2ba2d64055284e Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Date: Mon, 31 Jul 2006 15:15:18 +0800
Subject: PCI-Express AER implemetation: aer howto document

PCI-Express AER (Advanced Error Reporting) provides more robust error reporting.
The series of patches enable kernel support to AER.

The initial patches were written by Tom Long Nguyen. I ported them to the kernel
2.6.18-rc3. Many thanks to Rajesh Shah and Narayanan Chandramouli for their great
review comments and testing help.

Patch 1 consists of the pciaer-howto.txt document.

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/pcieaer-howto.txt | 253 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 253 insertions(+)
 create mode 100644 Documentation/pcieaer-howto.txt

diff --git a/Documentation/pcieaer-howto.txt b/Documentation/pcieaer-howto.txt
new file mode 100644
index 000000000000..16c251230c82
--- /dev/null
+++ b/Documentation/pcieaer-howto.txt
@@ -0,0 +1,253 @@
+   The PCI Express Advanced Error Reporting Driver Guide HOWTO
+		T. Long Nguyen	<tom.l.nguyen@intel.com>
+		Yanmin Zhang	<yanmin.zhang@intel.com>
+				07/29/2006
+
+
+1. Overview
+
+1.1 About this guide
+
+This guide describes the basics of the PCI Express Advanced Error
+Reporting (AER) driver and provides information on how to use it, as
+well as how to enable the drivers of endpoint devices to conform with
+PCI Express AER driver.
+
+1.2 Copyright © Intel Corporation 2006.
+
+1.3 What is the PCI Express AER Driver?
+
+PCI Express error signaling can occur on the PCI Express link itself
+or on behalf of transactions initiated on the link. PCI Express
+defines two error reporting paradigms: the baseline capability and
+the Advanced Error Reporting capability. The baseline capability is
+required of all PCI Express components providing a minimum defined
+set of error reporting requirements. Advanced Error Reporting
+capability is implemented with a PCI Express advanced error reporting
+extended capability structure providing more robust error reporting.
+
+The PCI Express AER driver provides the infrastructure to support PCI
+Express Advanced Error Reporting capability. The PCI Express AER
+driver provides three basic functions:
+
+-	Gathers the comprehensive error information if errors occurred.
+-	Reports error to the users.
+-	Performs error recovery actions.
+
+AER driver only attaches root ports which support PCI-Express AER
+capability.
+
+
+2. User Guide
+
+2.1 Include the PCI Express AER Root Driver into the Linux Kernel
+
+The PCI Express AER Root driver is a Root Port service driver attached
+to the PCI Express Port Bus driver. If a user wants to use it, the driver
+has to be compiled. Option CONFIG_PCIEAER supports this capability. It
+depends on CONFIG_PCIEPORTBUS, so pls. set CONFIG_PCIEPORTBUS=y and
+CONFIG_PCIEAER = y.
+
+2.2 Load PCI Express AER Root Driver
+There is a case where a system has AER support in BIOS. Enabling the AER
+Root driver and having AER support in BIOS may result unpredictable
+behavior. To avoid this conflict, a successful load of the AER Root driver
+requires ACPI _OSC support in the BIOS to allow the AER Root driver to
+request for native control of AER. See the PCI FW 3.0 Specification for
+details regarding OSC usage. Currently, lots of firmwares don't provide
+_OSC support while they use PCI Express. To support such firmwares,
+forceload, a parameter of type bool, could enable AER to continue to
+be initiated although firmwares have no _OSC support. To enable the
+walkaround, pls. add aerdriver.forceload=y to kernel boot parameter line
+when booting kernel. Note that forceload=n by default.
+
+2.3 AER error output
+When a PCI-E AER error is captured, an error message will be outputed to
+console. If it's a correctable error, it is outputed as a warning.
+Otherwise, it is printed as an error. So users could choose different
+log level to filter out correctable error messages.
+
+Below shows an example.
++------ PCI-Express Device Error -----+
+Error Severity          : Uncorrected (Fatal)
+PCIE Bus Error type     : Transaction Layer
+Unsupported Request     : First
+Requester ID            : 0500
+VendorID=8086h, DeviceID=0329h, Bus=05h, Device=00h, Function=00h
+TLB Header:
+04000001 00200a03 05010000 00050100
+
+In the example, 'Requester ID' means the ID of the device who sends
+the error message to root port. Pls. refer to pci express specs for
+other fields.
+
+
+3. Developer Guide
+
+To enable AER aware support requires a software driver to configure
+the AER capability structure within its device and to provide callbacks.
+
+To support AER better, developers need understand how AER does work
+firstly.
+
+PCI Express errors are classified into two types: correctable errors
+and uncorrectable errors. This classification is based on the impacts
+of those errors, which may result in degraded performance or function
+failure.
+
+Correctable errors pose no impacts on the functionality of the
+interface. The PCI Express protocol can recover without any software
+intervention or any loss of data. These errors are detected and
+corrected by hardware. Unlike correctable errors, uncorrectable
+errors impact functionality of the interface. Uncorrectable errors
+can cause a particular transaction or a particular PCI Express link
+to be unreliable. Depending on those error conditions, uncorrectable
+errors are further classified into non-fatal errors and fatal errors.
+Non-fatal errors cause the particular transaction to be unreliable,
+but the PCI Express link itself is fully functional. Fatal errors, on
+the other hand, cause the link to be unreliable.
+
+When AER is enabled, a PCI Express device will automatically send an
+error message to the PCIE root port above it when the device captures
+an error. The Root Port, upon receiving an error reporting message,
+internally processes and logs the error message in its PCI Express
+capability structure. Error information being logged includes storing
+the error reporting agent's requestor ID into the Error Source
+Identification Registers and setting the error bits of the Root Error
+Status Register accordingly. If AER error reporting is enabled in Root
+Error Command Register, the Root Port generates an interrupt if an
+error is detected.
+
+Note that the errors as described above are related to the PCI Express
+hierarchy and links. These errors do not include any device specific
+errors because device specific errors will still get sent directly to
+the device driver.
+
+3.1 Configure the AER capability structure
+
+AER aware drivers of PCI Express component need change the device
+control registers to enable AER. They also could change AER registers,
+including mask and severity registers. Helper function
+pci_enable_pcie_error_reporting could be used to enable AER. See
+section 3.3.
+
+3.2. Provide callbacks
+
+3.2.1 callback reset_link to reset pci express link
+
+This callback is used to reset the pci express physical link when a
+fatal error happens. The root port aer service driver provides a
+default reset_link function, but different upstream ports might
+have different specifications to reset pci express link, so all
+upstream ports should provide their own reset_link functions.
+
+In struct pcie_port_service_driver, a new pointer, reset_link, is
+added.
+
+pci_ers_result_t (*reset_link) (struct pci_dev *dev);
+
+Section 3.2.2.2 provides more detailed info on when to call
+reset_link.
+
+3.2.2 PCI error-recovery callbacks
+
+The PCI Express AER Root driver uses error callbacks to coordinate
+with downstream device drivers associated with a hierarchy in question
+when performing error recovery actions.
+
+Data struct pci_driver has a pointer, err_handler, to point to
+pci_error_handlers who consists of a couple of callback function
+pointers. AER driver follows the rules defined in
+pci-error-recovery.txt except pci express specific parts (e.g.
+reset_link). Pls. refer to pci-error-recovery.txt for detailed
+definitions of the callbacks.
+
+Below sections specify when to call the error callback functions.
+
+3.2.2.1 Correctable errors
+
+Correctable errors pose no impacts on the functionality of
+the interface. The PCI Express protocol can recover without any
+software intervention or any loss of data. These errors do not
+require any recovery actions. The AER driver clears the device's
+correctable error status register accordingly and logs these errors.
+
+3.2.2.2 Non-correctable (non-fatal and fatal) errors
+
+If an error message indicates a non-fatal error, performing link reset
+at upstream is not required. The AER driver calls error_detected(dev,
+pci_channel_io_normal) to all drivers associated within a hierarchy in
+question. for example,
+EndPoint<==>DownstreamPort B<==>UpstreamPort A<==>RootPort.
+If Upstream port A captures an AER error, the hierarchy consists of
+Downstream port B and EndPoint.
+
+A driver may return PCI_ERS_RESULT_CAN_RECOVER,
+PCI_ERS_RESULT_DISCONNECT, or PCI_ERS_RESULT_NEED_RESET, depending on
+whether it can recover or the AER driver calls mmio_enabled as next.
+
+If an error message indicates a fatal error, kernel will broadcast
+error_detected(dev, pci_channel_io_frozen) to all drivers within
+a hierarchy in question. Then, performing link reset at upstream is
+necessary. As different kinds of devices might use different approaches
+to reset link, AER port service driver is required to provide the
+function to reset link. Firstly, kernel looks for if the upstream
+component has an aer driver. If it has, kernel uses the reset_link
+callback of the aer driver. If the upstream component has no aer driver
+and the port is downstream port, we will use the aer driver of the
+root port who reports the AER error. As for upstream ports,
+they should provide their own aer service drivers with reset_link
+function. If error_detected returns PCI_ERS_RESULT_CAN_RECOVER and
+reset_link returns PCI_ERS_RESULT_RECOVERED, the error handling goes
+to mmio_enabled.
+
+3.3 helper functions
+
+3.3.1 int pci_find_aer_capability(struct pci_dev *dev);
+pci_find_aer_capability locates the PCI Express AER capability
+in the device configuration space. If the device doesn't support
+PCI-Express AER, the function returns 0.
+
+3.3.2 int pci_enable_pcie_error_reporting(struct pci_dev *dev);
+pci_enable_pcie_error_reporting enables the device to send error
+messages to root port when an error is detected. Note that devices
+don't enable the error reporting by default, so device drivers need
+call this function to enable it.
+
+3.3.3 int pci_disable_pcie_error_reporting(struct pci_dev *dev);
+pci_disable_pcie_error_reporting disables the device to send error
+messages to root port when an error is detected.
+
+3.3.4 int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev);
+pci_cleanup_aer_uncorrect_error_status cleanups the uncorrectable
+error status register.
+
+3.4 Frequent Asked Questions
+
+Q: What happens if a PCI Express device driver does not provide an
+error recovery handler (pci_driver->err_handler is equal to NULL)?
+
+A: The devices attached with the driver won't be recovered. If the
+error is fatal, kernel will print out warning messages. Please refer
+to section 3 for more information.
+
+Q: What happens if an upstream port service driver does not provide
+callback reset_link?
+
+A: Fatal error recovery will fail if the errors are reported by the
+upstream ports who are attached by the service driver.
+
+Q: How does this infrastructure deal with driver that is not PCI
+Express aware?
+
+A: This infrastructure calls the error callback functions of the
+driver when an error happens. But if the driver is not aware of
+PCI Express, the device might not report its own errors to root
+port.
+
+Q: What modifications will that driver need to make it compatible
+with the PCI Express AER Root driver?
+
+A: It could call the helper functions to enable AER in devices and
+cleanup uncorrectable status register. Pls. refer to section 3.3.
+
-- 
cgit v1.2.1


From 48408157ebf5b2c6dc1e04ba5d258012f6a7f356 Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Date: Mon, 31 Jul 2006 15:18:39 +0800
Subject: PCI-Express AER implemetation: export pcie_port_bus_type

Patch 2 exports pcie_port_bus_type.

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pcie/portdrv_bus.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/pci/pcie/portdrv_bus.c b/drivers/pci/pcie/portdrv_bus.c
index 3e84b501e6a4..3f0976868eda 100644
--- a/drivers/pci/pcie/portdrv_bus.c
+++ b/drivers/pci/pcie/portdrv_bus.c
@@ -24,6 +24,7 @@ struct bus_type pcie_port_bus_type = {
 	.suspend	= pcie_port_bus_suspend,
 	.resume		= pcie_port_bus_resume, 
 };
+EXPORT_SYMBOL_GPL(pcie_port_bus_type);
 
 static int pcie_port_bus_match(struct device *dev, struct device_driver *drv)
 {
-- 
cgit v1.2.1


From 6c2b374d74857e892080ee726184ec1d15e7d4e4 Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Date: Mon, 31 Jul 2006 15:21:33 +0800
Subject: PCI-Express AER implemetation: AER core and aerdriver

Patch 3 implements the core part of PCI-Express AER and aerdrv
port service driver.

When a root port service device is probed, the aerdrv will call
request_irq to register irq handler for AER error interrupt.

When a device sends an PCI-Express error message to the root port,
the root port will trigger an interrupt, by either MSI or IO-APIC,
then kernel would run the irq handler. The handler collects root
error status register and schedules a work. The work will call
the core part to process the error based on its type
(Correctable/non-fatal/fatal).

As for Correctable errors, the patch chooses to just clear the correctable
error status register of the device.

As for the non-fatal error, the patch follows generic PCI error handler
rules to call the error callback functions of the endpoint's driver. If
the device is a bridge, the patch chooses to broadcast the error to
downstream devices.

As for the fatal error, the patch resets the pci-express link and
follows generic PCI error handler rules to call the error callback
functions of the endpoint's driver. If the device is a bridge, the patch
chooses to broadcast the error to downstream devices.

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pcie/Kconfig               |   1 +
 drivers/pci/pcie/Makefile              |   3 +
 drivers/pci/pcie/aer/Kconfig           |  12 +
 drivers/pci/pcie/aer/Makefile          |   8 +
 drivers/pci/pcie/aer/aerdrv.c          | 346 +++++++++++++++
 drivers/pci/pcie/aer/aerdrv.h          | 125 ++++++
 drivers/pci/pcie/aer/aerdrv_acpi.c     |  68 +++
 drivers/pci/pcie/aer/aerdrv_core.c     | 757 +++++++++++++++++++++++++++++++++
 drivers/pci/pcie/aer/aerdrv_errprint.c | 248 +++++++++++
 include/linux/aer.h                    |  24 ++
 include/linux/pcieport_if.h            |   6 +
 11 files changed, 1598 insertions(+)
 create mode 100644 drivers/pci/pcie/aer/Kconfig
 create mode 100644 drivers/pci/pcie/aer/Makefile
 create mode 100644 drivers/pci/pcie/aer/aerdrv.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv.h
 create mode 100644 drivers/pci/pcie/aer/aerdrv_acpi.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv_core.c
 create mode 100644 drivers/pci/pcie/aer/aerdrv_errprint.c
 create mode 100644 include/linux/aer.h

diff --git a/drivers/pci/pcie/Kconfig b/drivers/pci/pcie/Kconfig
index 1012db8b8b2c..0ad92a8ad8b1 100644
--- a/drivers/pci/pcie/Kconfig
+++ b/drivers/pci/pcie/Kconfig
@@ -34,3 +34,4 @@ config HOTPLUG_PCI_PCIE_POLL_EVENT_MODE
 	   
 	  When in doubt, say N.
 
+source "drivers/pci/pcie/aer/Kconfig"
diff --git a/drivers/pci/pcie/Makefile b/drivers/pci/pcie/Makefile
index 984fa87283e3..e00fb99acf44 100644
--- a/drivers/pci/pcie/Makefile
+++ b/drivers/pci/pcie/Makefile
@@ -5,3 +5,6 @@
 pcieportdrv-y			:= portdrv_core.o portdrv_pci.o portdrv_bus.o
 
 obj-$(CONFIG_PCIEPORTBUS)	+= pcieportdrv.o
+
+# Build PCI Express AER if needed
+obj-$(CONFIG_PCIEAER)		+= aer/
diff --git a/drivers/pci/pcie/aer/Kconfig b/drivers/pci/pcie/aer/Kconfig
new file mode 100644
index 000000000000..3f37a60a6438
--- /dev/null
+++ b/drivers/pci/pcie/aer/Kconfig
@@ -0,0 +1,12 @@
+#
+# PCI Express Root Port Device AER Configuration
+#
+
+config PCIEAER
+	boolean "Root Port Advanced Error Reporting support"
+	depends on PCIEPORTBUS && ACPI
+	default y
+	help
+	  This enables PCI Express Root Port Advanced Error Reporting
+	  (AER) driver support. Error reporting messages sent to Root
+	  Port will be handled by PCI Express AER driver.
diff --git a/drivers/pci/pcie/aer/Makefile b/drivers/pci/pcie/aer/Makefile
new file mode 100644
index 000000000000..15a4f40d520b
--- /dev/null
+++ b/drivers/pci/pcie/aer/Makefile
@@ -0,0 +1,8 @@
+#
+# Makefile for PCI-Express Root Port Advanced Error Reporting Driver
+#
+
+obj-$(CONFIG_PCIEAER) += aerdriver.o
+
+aerdriver-objs := aerdrv_errprint.o aerdrv_core.o aerdrv.o aerdrv_acpi.o
+
diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c
new file mode 100644
index 000000000000..0d4ac027d53e
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv.c
@@ -0,0 +1,346 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * This file implements the AER root port service driver. The driver will
+ * register an irq handler. When root port triggers an AER interrupt, the irq
+ * handler will collect root port status and schedule a work.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/pcieport_if.h>
+
+#include "aerdrv.h"
+
+/*
+ * Version Information
+ */
+#define DRIVER_VERSION "v1.0"
+#define DRIVER_AUTHOR "tom.l.nguyen@intel.com"
+#define DRIVER_DESC "Root Port Advanced Error Reporting Driver"
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+static int __devinit aer_probe (struct pcie_device *dev,
+	const struct pcie_port_service_id *id );
+static void aer_remove(struct pcie_device *dev);
+static int aer_suspend(struct pcie_device *dev, pm_message_t state)
+{return 0;}
+static int aer_resume(struct pcie_device *dev) {return 0;}
+static pci_ers_result_t aer_error_detected(struct pci_dev *dev,
+	enum pci_channel_state error);
+static void aer_error_resume(struct pci_dev *dev);
+static pci_ers_result_t aer_root_reset(struct pci_dev *dev);
+
+/*
+ * PCI Express bus's AER Root service driver data structure
+ */
+static struct pcie_port_service_id aer_id[] = {
+	{
+	.vendor 	= PCI_ANY_ID,
+	.device 	= PCI_ANY_ID,
+	.port_type 	= PCIE_RC_PORT,
+	.service_type 	= PCIE_PORT_SERVICE_AER,
+	},
+	{ /* end: all zeroes */ }
+};
+
+static struct pci_error_handlers aer_error_handlers = {
+	.error_detected = aer_error_detected,
+	.resume = aer_error_resume,
+};
+
+static struct pcie_port_service_driver aerdrv = {
+	.name		= "aer",
+	.id_table	= &aer_id[0],
+
+	.probe		= aer_probe,
+	.remove		= aer_remove,
+
+	.suspend	= aer_suspend,
+	.resume		= aer_resume,
+
+	.err_handler	= &aer_error_handlers,
+
+	.reset_link	= aer_root_reset,
+};
+
+/**
+ * aer_irq - Root Port's ISR
+ * @irq: IRQ assigned to Root Port
+ * @context: pointer to Root Port data structure
+ * @r: pointer struct pt_regs
+ *
+ * Invoked when Root Port detects AER messages.
+ **/
+static irqreturn_t aer_irq(int irq, void *context, struct pt_regs * r)
+{
+	unsigned int status, id;
+	struct pcie_device *pdev = (struct pcie_device *)context;
+	struct aer_rpc *rpc = get_service_data(pdev);
+	int next_prod_idx;
+	unsigned long flags;
+	int pos;
+
+	pos = pci_find_aer_capability(pdev->port);
+	/*
+	 * Must lock access to Root Error Status Reg, Root Error ID Reg,
+	 * and Root error producer/consumer index
+	 */
+	spin_lock_irqsave(&rpc->e_lock, flags);
+
+	/* Read error status */
+	pci_read_config_dword(pdev->port, pos + PCI_ERR_ROOT_STATUS, &status);
+	if (!(status & ROOT_ERR_STATUS_MASKS)) {
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return IRQ_NONE;
+	}
+
+	/* Read error source and clear error status */
+	pci_read_config_dword(pdev->port, pos + PCI_ERR_ROOT_COR_SRC, &id);
+	pci_write_config_dword(pdev->port, pos + PCI_ERR_ROOT_STATUS, status);
+
+	/* Store error source for later DPC handler */
+	next_prod_idx = rpc->prod_idx + 1;
+	if (next_prod_idx == AER_ERROR_SOURCES_MAX)
+		next_prod_idx = 0;
+	if (next_prod_idx == rpc->cons_idx) {
+		/*
+		 * Error Storm Condition - possibly the same error occurred.
+		 * Drop the error.
+		 */
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return IRQ_HANDLED;
+	}
+	rpc->e_sources[rpc->prod_idx].status =  status;
+	rpc->e_sources[rpc->prod_idx].id = id;
+	rpc->prod_idx = next_prod_idx;
+	spin_unlock_irqrestore(&rpc->e_lock, flags);
+
+	/*  Invoke DPC handler */
+	schedule_work(&rpc->dpc_handler);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * aer_alloc_rpc - allocate Root Port data structure
+ * @dev: pointer to the pcie_dev data structure
+ *
+ * Invoked when Root Port's AER service is loaded.
+ **/
+static struct aer_rpc* aer_alloc_rpc(struct pcie_device *dev)
+{
+	struct aer_rpc *rpc;
+
+	if (!(rpc = (struct aer_rpc *)kmalloc(sizeof(struct aer_rpc),
+		GFP_KERNEL)))
+		return NULL;
+
+	memset(rpc, 0, sizeof(struct aer_rpc));
+	/*
+	 * Initialize Root lock access, e_lock, to Root Error Status Reg,
+	 * Root Error ID Reg, and Root error producer/consumer index.
+	 */
+	rpc->e_lock = SPIN_LOCK_UNLOCKED;
+
+	rpc->rpd = dev;
+	INIT_WORK(&rpc->dpc_handler, aer_isr, (void *)dev);
+	rpc->prod_idx = rpc->cons_idx = 0;
+	mutex_init(&rpc->rpc_mutex);
+	init_waitqueue_head(&rpc->wait_release);
+
+	/* Use PCIE bus function to store rpc into PCIE device */
+	set_service_data(dev, rpc);
+
+	return rpc;
+}
+
+/**
+ * aer_remove - clean up resources
+ * @dev: pointer to the pcie_dev data structure
+ *
+ * Invoked when PCI Express bus unloads or AER probe fails.
+ **/
+static void aer_remove(struct pcie_device *dev)
+{
+	struct aer_rpc *rpc = get_service_data(dev);
+
+	if (rpc) {
+		/* If register interrupt service, it must be free. */
+		if (rpc->isr)
+			free_irq(dev->irq, dev);
+
+		wait_event(rpc->wait_release, rpc->prod_idx == rpc->cons_idx);
+
+		aer_delete_rootport(rpc);
+		set_service_data(dev, NULL);
+	}
+}
+
+/**
+ * aer_probe - initialize resources
+ * @dev: pointer to the pcie_dev data structure
+ * @id: pointer to the service id data structure
+ *
+ * Invoked when PCI Express bus loads AER service driver.
+ **/
+static int __devinit aer_probe (struct pcie_device *dev,
+				const struct pcie_port_service_id *id )
+{
+	int status;
+	struct aer_rpc *rpc;
+	struct device *device = &dev->device;
+
+	/* Init */
+	if ((status = aer_init(dev)))
+		return status;
+
+	/* Alloc rpc data structure */
+	if (!(rpc = aer_alloc_rpc(dev))) {
+		printk(KERN_DEBUG "%s: Alloc rpc fails on PCIE device[%s]\n",
+			__FUNCTION__, device->bus_id);
+		aer_remove(dev);
+		return -ENOMEM;
+	}
+
+	/* Request IRQ ISR */
+	if ((status = request_irq(dev->irq, aer_irq, SA_SHIRQ, "aerdrv",
+				dev))) {
+		printk(KERN_DEBUG "%s: Request ISR fails on PCIE device[%s]\n",
+			__FUNCTION__, device->bus_id);
+		aer_remove(dev);
+		return status;
+	}
+
+	rpc->isr = 1;
+
+	aer_enable_rootport(rpc);
+
+	return status;
+}
+
+/**
+ * aer_root_reset - reset link on Root Port
+ * @dev: pointer to Root Port's pci_dev data structure
+ *
+ * Invoked by Port Bus driver when performing link reset at Root Port.
+ **/
+static pci_ers_result_t aer_root_reset(struct pci_dev *dev)
+{
+	u16 p2p_ctrl;
+	u32 status;
+	int pos;
+
+	pos = pci_find_aer_capability(dev);
+
+	/* Disable Root's interrupt in response to error messages */
+	pci_write_config_dword(dev, pos + PCI_ERR_ROOT_COMMAND, 0);
+
+	/* Assert Secondary Bus Reset */
+	pci_read_config_word(dev, PCI_BRIDGE_CONTROL, &p2p_ctrl);
+	p2p_ctrl |= PCI_CB_BRIDGE_CTL_CB_RESET;
+	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);
+
+	/* De-assert Secondary Bus Reset */
+	p2p_ctrl &= ~PCI_CB_BRIDGE_CTL_CB_RESET;
+	pci_write_config_word(dev, PCI_BRIDGE_CONTROL, p2p_ctrl);
+
+	/*
+	 * System software must wait for at least 100ms from the end
+	 * of a reset of one or more device before it is permitted
+	 * to issue Configuration Requests to those devices.
+	 */
+	msleep(200);
+	printk(KERN_DEBUG "Complete link reset at Root[%s]\n", dev->dev.bus_id);
+
+	/* Enable Root Port's interrupt in response to error messages */
+	pci_read_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, &status);
+	pci_write_config_dword(dev, pos + PCI_ERR_ROOT_STATUS, status);
+	pci_write_config_dword(dev,
+		pos + PCI_ERR_ROOT_COMMAND,
+		ROOT_PORT_INTR_ON_MESG_MASK);
+
+	return PCI_ERS_RESULT_RECOVERED;
+}
+
+/**
+ * aer_error_detected - update severity status
+ * @dev: pointer to Root Port's pci_dev data structure
+ * @error: error severity being notified by port bus
+ *
+ * Invoked by Port Bus driver during error recovery.
+ **/
+static pci_ers_result_t aer_error_detected(struct pci_dev *dev,
+			enum pci_channel_state error)
+{
+	/* Root Port has no impact. Always recovers. */
+	return PCI_ERS_RESULT_CAN_RECOVER;
+}
+
+/**
+ * aer_error_resume - clean up corresponding error status bits
+ * @dev: pointer to Root Port's pci_dev data structure
+ *
+ * Invoked by Port Bus driver during nonfatal recovery.
+ **/
+static void aer_error_resume(struct pci_dev *dev)
+{
+	int pos;
+	u32 status, mask;
+	u16 reg16;
+
+	/* Clean up Root device status */
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	pci_read_config_word(dev, pos + PCI_EXP_DEVSTA, &reg16);
+	pci_write_config_word(dev, pos + PCI_EXP_DEVSTA, reg16);
+
+	/* Clean AER Root Error Status */
+	pos = pci_find_aer_capability(dev);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, &mask);
+	if (dev->error_state == pci_channel_io_normal)
+		status &= ~mask; /* Clear corresponding nonfatal bits */
+	else
+		status &= mask; /* Clear corresponding fatal bits */
+	pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
+}
+
+/**
+ * aer_service_init - register AER root service driver
+ *
+ * Invoked when AER root service driver is loaded.
+ **/
+static int __init aer_service_init(void)
+{
+	return pcie_port_service_register(&aerdrv);
+}
+
+/**
+ * aer_service_exit - unregister AER root service driver
+ *
+ * Invoked when AER root service driver is unloaded.
+ **/
+static void __exit aer_service_exit(void)
+{
+	pcie_port_service_unregister(&aerdrv);
+}
+
+module_init(aer_service_init);
+module_exit(aer_service_exit);
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
new file mode 100644
index 000000000000..daf0cad88fc8
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -0,0 +1,125 @@
+/*
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#ifndef _AERDRV_H_
+#define _AERDRV_H_
+
+#include <linux/pcieport_if.h>
+#include <linux/aer.h>
+
+#define AER_NONFATAL			0
+#define AER_FATAL			1
+#define AER_CORRECTABLE			2
+#define AER_UNCORRECTABLE		4
+#define AER_ERROR_MASK			0x001fffff
+#define AER_ERROR(d)			(d & AER_ERROR_MASK)
+
+#define OSC_METHOD_RUN_SUCCESS		0
+#define OSC_METHOD_NOT_SUPPORTED	1
+#define OSC_METHOD_RUN_FAILURE		2
+
+/* Root Error Status Register Bits */
+#define ROOT_ERR_STATUS_MASKS			0x0f
+
+#define SYSTEM_ERROR_INTR_ON_MESG_MASK	(PCI_EXP_RTCTL_SECEE|	\
+					PCI_EXP_RTCTL_SENFEE|	\
+					PCI_EXP_RTCTL_SEFEE)
+#define ROOT_PORT_INTR_ON_MESG_MASK	(PCI_ERR_ROOT_CMD_COR_EN|	\
+					PCI_ERR_ROOT_CMD_NONFATAL_EN|	\
+					PCI_ERR_ROOT_CMD_FATAL_EN)
+#define ERR_COR_ID(d)			(d & 0xffff)
+#define ERR_UNCOR_ID(d)			(d >> 16)
+
+#define AER_SUCCESS			0
+#define AER_UNSUCCESS			1
+#define AER_ERROR_SOURCES_MAX		100
+
+#define AER_LOG_TLP_MASKS		(PCI_ERR_UNC_POISON_TLP|	\
+					PCI_ERR_UNC_ECRC|		\
+					PCI_ERR_UNC_UNSUP|		\
+					PCI_ERR_UNC_COMP_ABORT|		\
+					PCI_ERR_UNC_UNX_COMP|		\
+					PCI_ERR_UNC_MALF_TLP)
+
+/* AER Error Info Flags */
+#define AER_TLP_HEADER_VALID_FLAG	0x00000001
+#define AER_MULTI_ERROR_VALID_FLAG	0x00000002
+
+#define ERR_CORRECTABLE_ERROR_MASK	0x000031c1
+#define ERR_UNCORRECTABLE_ERROR_MASK	0x001ff010
+
+struct header_log_regs {
+	unsigned int dw0;
+	unsigned int dw1;
+	unsigned int dw2;
+	unsigned int dw3;
+};
+
+struct aer_err_info {
+	int severity;			/* 0:NONFATAL | 1:FATAL | 2:COR */
+	int flags;
+	unsigned int status;		/* COR/UNCOR Error Status */
+	struct header_log_regs tlp; 	/* TLP Header */
+};
+
+struct aer_err_source {
+	unsigned int status;
+	unsigned int id;
+};
+
+struct aer_rpc {
+	struct pcie_device *rpd;	/* Root Port device */
+	struct work_struct dpc_handler;
+	struct aer_err_source e_sources[AER_ERROR_SOURCES_MAX];
+	unsigned short prod_idx;	/* Error Producer Index */
+	unsigned short cons_idx;	/* Error Consumer Index */
+	int isr;
+	spinlock_t e_lock;		/*
+					 * Lock access to Error Status/ID Regs
+					 * and error producer/consumer index
+					 */
+	struct mutex rpc_mutex;		/*
+					 * only one thread could do
+					 * recovery on the same
+					 * root port hierachy
+					 */
+	wait_queue_head_t wait_release;
+};
+
+struct aer_broadcast_data {
+	enum pci_channel_state state;
+	enum pci_ers_result result;
+};
+
+static inline pci_ers_result_t merge_result(enum pci_ers_result orig,
+		enum pci_ers_result new)
+{
+	switch (orig) {
+	case PCI_ERS_RESULT_CAN_RECOVER:
+	case PCI_ERS_RESULT_RECOVERED:
+		orig = new;
+		break;
+	case PCI_ERS_RESULT_DISCONNECT:
+		if (new == PCI_ERS_RESULT_NEED_RESET)
+			orig = new;
+		break;
+	default:
+		break;
+	}
+
+	return orig;
+}
+
+extern struct bus_type pcie_port_bus_type;
+extern void aer_enable_rootport(struct aer_rpc *rpc);
+extern void aer_delete_rootport(struct aer_rpc *rpc);
+extern int aer_init(struct pcie_device *dev);
+extern void aer_isr(void *context);
+extern void aer_print_error(struct pci_dev *dev, struct aer_err_info *info);
+extern int aer_osc_setup(struct pci_dev *dev);
+
+#endif //_AERDRV_H_
diff --git a/drivers/pci/pcie/aer/aerdrv_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c
new file mode 100644
index 000000000000..fa68e89ebec9
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_acpi.c
@@ -0,0 +1,68 @@
+/*
+ * Access ACPI _OSC method
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+#include <linux/acpi.h>
+#include <linux/pci-acpi.h>
+#include <linux/delay.h>
+#include "aerdrv.h"
+
+/**
+ * aer_osc_setup - run ACPI _OSC method
+ *
+ * Return:
+ *	Zero if success. Nonzero for otherwise.
+ *
+ * Invoked when PCIE bus loads AER service driver. To avoid conflict with
+ * BIOS AER support requires BIOS to yield AER control to OS native driver.
+ **/
+int aer_osc_setup(struct pci_dev *dev)
+{
+	int retval = OSC_METHOD_RUN_SUCCESS;
+	acpi_status status;
+	acpi_handle handle = DEVICE_ACPI_HANDLE(&dev->dev);
+	struct pci_dev *pdev = dev;
+	struct pci_bus *parent;
+
+	while (!handle) {
+		if (!pdev || !pdev->bus->parent)
+			break;
+		parent = pdev->bus->parent;
+		if (!parent->self)
+			/* Parent must be a host bridge */
+			handle = acpi_get_pci_rootbridge_handle(
+					pci_domain_nr(parent),
+					parent->number);
+		else
+			handle = DEVICE_ACPI_HANDLE(
+					&(parent->self->dev));
+		pdev = parent->self;
+	}
+
+	if (!handle)
+		return OSC_METHOD_NOT_SUPPORTED;
+
+	pci_osc_support_set(OSC_EXT_PCI_CONFIG_SUPPORT);
+	status = pci_osc_control_set(handle, OSC_PCI_EXPRESS_AER_CONTROL |
+		OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL);
+	if (ACPI_FAILURE(status)) {
+		if (status == AE_SUPPORT)
+			retval = OSC_METHOD_NOT_SUPPORTED;
+	 	else
+			retval = OSC_METHOD_RUN_FAILURE;
+	}
+
+	return retval;
+}
+
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
new file mode 100644
index 000000000000..5591043acea7
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -0,0 +1,757 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv_core.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * This file implements the core part of PCI-Express AER. When an pci-express
+ * error is delivered, an error message will be collected and printed to
+ * console, then, an error recovery procedure will be executed by following
+ * the pci error recovery rules.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+#include <linux/acpi.h>
+#include <linux/pci-acpi.h>
+#include <linux/delay.h>
+#include "aerdrv.h"
+
+static int forceload;
+module_param(forceload, bool, 0);
+
+#define PCI_CFG_SPACE_SIZE	(0x100)
+int pci_find_aer_capability(struct pci_dev *dev)
+{
+	int pos;
+	u32 reg32 = 0;
+
+	/* Check if it's a pci-express device */
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return 0;
+
+	/* Check if it supports pci-express AER */
+	pos = PCI_CFG_SPACE_SIZE;
+	while (pos) {
+		if (pci_read_config_dword(dev, pos, &reg32))
+			return 0;
+
+		/* some broken boards return ~0 */
+		if (reg32 == 0xffffffff)
+			return 0;
+
+		if (PCI_EXT_CAP_ID(reg32) == PCI_EXT_CAP_ID_ERR)
+			break;
+
+		pos = reg32 >> 20;
+	}
+
+	return pos;
+}
+
+int pci_enable_pcie_error_reporting(struct pci_dev *dev)
+{
+	u16 reg16 = 0;
+	int pos;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 |
+		PCI_EXP_DEVCTL_CERE |
+		PCI_EXP_DEVCTL_NFERE |
+		PCI_EXP_DEVCTL_FERE |
+		PCI_EXP_DEVCTL_URRE;
+	pci_write_config_word(dev, pos+PCI_EXP_DEVCTL,
+			reg16);
+	return 0;
+}
+
+int pci_disable_pcie_error_reporting(struct pci_dev *dev)
+{
+	u16 reg16 = 0;
+	int pos;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_word(dev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 & ~(PCI_EXP_DEVCTL_CERE |
+			PCI_EXP_DEVCTL_NFERE |
+			PCI_EXP_DEVCTL_FERE |
+			PCI_EXP_DEVCTL_URRE);
+	pci_write_config_word(dev, pos+PCI_EXP_DEVCTL,
+			reg16);
+	return 0;
+}
+
+int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev)
+{
+	int pos;
+	u32 status, mask;
+
+	pos = pci_find_aer_capability(dev);
+	if (!pos)
+		return -EIO;
+
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, &status);
+	pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_SEVER, &mask);
+	if (dev->error_state == pci_channel_io_normal)
+		status &= ~mask; /* Clear corresponding nonfatal bits */
+	else
+		status &= mask; /* Clear corresponding fatal bits */
+	pci_write_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS, status);
+
+	return 0;
+}
+
+static int find_device_iter(struct device *device, void *data)
+{
+	struct pci_dev *dev;
+	u16 id = *(unsigned long *)data;
+	u8 secondary, subordinate, d_bus = id >> 8;
+
+	if (device->bus == &pci_bus_type) {
+		dev = to_pci_dev(device);
+		if (id == ((dev->bus->number << 8) | dev->devfn)) {
+			/*
+			 * Device ID match
+			 */
+			*(unsigned long*)data = (unsigned long)device;
+			return 1;
+		}
+
+		/*
+		 * If device is P2P, check if it is an upstream?
+		 */
+		if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
+			pci_read_config_byte(dev, PCI_SECONDARY_BUS,
+				&secondary);
+			pci_read_config_byte(dev, PCI_SUBORDINATE_BUS,
+				&subordinate);
+			if (d_bus >= secondary && d_bus <= subordinate) {
+				*(unsigned long*)data = (unsigned long)device;
+				return 1;
+			}
+		}
+	}
+
+	return 0;
+}
+
+/**
+ * find_source_device - search through device hierarchy for source device
+ * @p_dev: pointer to Root Port pci_dev data structure
+ * @id: device ID of agent who sends an error message to this Root Port
+ *
+ * Invoked when error is detected at the Root Port.
+ **/
+static struct device* find_source_device(struct pci_dev *parent, u16 id)
+{
+	struct pci_dev *dev = parent;
+	struct device *device;
+	unsigned long device_addr;
+	int status;
+
+	/* Is Root Port an agent that sends error message? */
+	if (id == ((dev->bus->number << 8) | dev->devfn))
+		return &dev->dev;
+
+	do {
+		device_addr = id;
+ 		if ((status = device_for_each_child(&dev->dev,
+			&device_addr, find_device_iter))) {
+			device = (struct device*)device_addr;
+			dev = to_pci_dev(device);
+			if (id == ((dev->bus->number << 8) | dev->devfn))
+				return device;
+		}
+ 	}while (status);
+
+	return NULL;
+}
+
+static void report_error_detected(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	dev->error_state = result_data->state;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->error_detected) {
+		if (result_data->state == pci_channel_io_frozen &&
+			!(dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)) {
+			/*
+			 * In case of fatal recovery, if one of down-
+			 * stream device has no driver. We might be
+			 * unable to recover because a later insmod
+			 * of a driver for this device is unaware of
+			 * its hw state.
+			 */
+			printk(KERN_DEBUG "Device ID[%s] has %s\n",
+					dev->dev.bus_id, (dev->driver) ?
+					"no AER-aware driver" : "no driver");
+		}
+		return;
+	}
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->error_detected(dev, result_data->state);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_mmio_enabled(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->mmio_enabled)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->mmio_enabled(dev);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_slot_reset(struct pci_dev *dev, void *data)
+{
+	pci_ers_result_t vote;
+	struct pci_error_handlers *err_handler;
+	struct aer_broadcast_data *result_data;
+	result_data = (struct aer_broadcast_data *) data;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->slot_reset)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	vote = err_handler->slot_reset(dev);
+	result_data->result = merge_result(result_data->result, vote);
+	return;
+}
+
+static void report_resume(struct pci_dev *dev, void *data)
+{
+	struct pci_error_handlers *err_handler;
+
+	dev->error_state = pci_channel_io_normal;
+
+	if (!dev->driver ||
+		!dev->driver->err_handler ||
+		!dev->driver->err_handler->slot_reset)
+		return;
+
+	err_handler = dev->driver->err_handler;
+	err_handler->resume(dev);
+	return;
+}
+
+/**
+ * broadcast_error_message - handle message broadcast to downstream drivers
+ * @device: pointer to from where in a hierarchy message is broadcasted down
+ * @api: callback to be broadcasted
+ * @state: error state
+ *
+ * Invoked during error recovery process. Once being invoked, the content
+ * of error severity will be broadcasted to all downstream drivers in a
+ * hierarchy in question.
+ **/
+static pci_ers_result_t broadcast_error_message(struct pci_dev *dev,
+	enum pci_channel_state state,
+	char *error_mesg,
+	void (*cb)(struct pci_dev *, void *))
+{
+	struct aer_broadcast_data result_data;
+
+	printk(KERN_DEBUG "Broadcast %s message\n", error_mesg);
+	result_data.state = state;
+	if (cb == report_error_detected)
+		result_data.result = PCI_ERS_RESULT_CAN_RECOVER;
+	else
+		result_data.result = PCI_ERS_RESULT_RECOVERED;
+
+	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE) {
+		/*
+		 * If the error is reported by a bridge, we think this error
+		 * is related to the downstream link of the bridge, so we
+		 * do error recovery on all subordinates of the bridge instead
+		 * of the bridge and clear the error status of the bridge.
+		 */
+		if (cb == report_error_detected)
+			dev->error_state = state;
+		pci_walk_bus(dev->subordinate, cb, &result_data);
+		if (cb == report_resume) {
+			pci_cleanup_aer_uncorrect_error_status(dev);
+			dev->error_state = pci_channel_io_normal;
+		}
+	}
+	else {
+		/*
+		 * If the error is reported by an end point, we think this
+		 * error is related to the upstream link of the end point.
+		 */
+		pci_walk_bus(dev->bus, cb, &result_data);
+	}
+
+	return result_data.result;
+}
+
+struct find_aer_service_data {
+	struct pcie_port_service_driver *aer_driver;
+	int is_downstream;
+};
+
+static int find_aer_service_iter(struct device *device, void *data)
+{
+	struct device_driver *driver;
+	struct pcie_port_service_driver *service_driver;
+	struct pcie_device *pcie_dev;
+	struct find_aer_service_data *result;
+
+	result = (struct find_aer_service_data *) data;
+
+	if (device->bus == &pcie_port_bus_type) {
+		pcie_dev = to_pcie_device(device);
+		if (pcie_dev->id.port_type == PCIE_SW_DOWNSTREAM_PORT)
+			result->is_downstream = 1;
+
+		driver = device->driver;
+		if (driver) {
+			service_driver = to_service_driver(driver);
+			if (service_driver->id_table->service_type ==
+					PCIE_PORT_SERVICE_AER) {
+				result->aer_driver = service_driver;
+				return 1;
+			}
+		}
+	}
+
+	return 0;
+}
+
+static void find_aer_service(struct pci_dev *dev,
+		struct find_aer_service_data *data)
+{
+	device_for_each_child(&dev->dev, data, find_aer_service_iter);
+}
+
+static pci_ers_result_t reset_link(struct pcie_device *aerdev,
+		struct pci_dev *dev)
+{
+	struct pci_dev *udev;
+	pci_ers_result_t status;
+	struct find_aer_service_data data;
+
+	if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE)
+		udev = dev;
+	else
+		udev= dev->bus->self;
+
+	data.is_downstream = 0;
+	data.aer_driver = NULL;
+	find_aer_service(udev, &data);
+
+	/*
+	 * Use the aer driver of the error agent firstly.
+	 * If it hasn't the aer driver, use the root port's
+	 */
+	if (!data.aer_driver || !data.aer_driver->reset_link) {
+		if (data.is_downstream &&
+			aerdev->device.driver &&
+			to_service_driver(aerdev->device.driver)->reset_link) {
+			data.aer_driver =
+				to_service_driver(aerdev->device.driver);
+		} else {
+			printk(KERN_DEBUG "No link-reset support to Device ID"
+				"[%s]\n",
+				dev->dev.bus_id);
+			return PCI_ERS_RESULT_DISCONNECT;
+		}
+	}
+
+	status = data.aer_driver->reset_link(udev);
+	if (status != PCI_ERS_RESULT_RECOVERED) {
+		printk(KERN_DEBUG "Link reset at upstream Device ID"
+			"[%s] failed\n",
+			udev->dev.bus_id);
+		return PCI_ERS_RESULT_DISCONNECT;
+	}
+
+	return status;
+}
+
+/**
+ * do_recovery - handle nonfatal/fatal error recovery process
+ * @aerdev: pointer to a pcie_device data structure of root port
+ * @dev: pointer to a pci_dev data structure of agent detecting an error
+ * @severity: error severity type
+ *
+ * Invoked when an error is nonfatal/fatal. Once being invoked, broadcast
+ * error detected message to all downstream drivers within a hierarchy in
+ * question and return the returned code.
+ **/
+static pci_ers_result_t do_recovery(struct pcie_device *aerdev,
+		struct pci_dev *dev,
+		int severity)
+{
+	pci_ers_result_t status, result = PCI_ERS_RESULT_RECOVERED;
+	enum pci_channel_state state;
+
+	if (severity == AER_FATAL)
+		state = pci_channel_io_frozen;
+	else
+		state = pci_channel_io_normal;
+
+	status = broadcast_error_message(dev,
+			state,
+			"error_detected",
+			report_error_detected);
+
+	if (severity == AER_FATAL) {
+		result = reset_link(aerdev, dev);
+		if (result != PCI_ERS_RESULT_RECOVERED) {
+			/* TODO: Should panic here? */
+			return result;
+		}
+	}
+
+	if (status == PCI_ERS_RESULT_CAN_RECOVER)
+		status = broadcast_error_message(dev,
+				state,
+				"mmio_enabled",
+				report_mmio_enabled);
+
+	if (status == PCI_ERS_RESULT_NEED_RESET) {
+		/*
+		 * TODO: Should call platform-specific
+		 * functions to reset slot before calling
+		 * drivers' slot_reset callbacks?
+		 */
+		status = broadcast_error_message(dev,
+				state,
+				"slot_reset",
+				report_slot_reset);
+	}
+
+	if (status == PCI_ERS_RESULT_RECOVERED)
+		broadcast_error_message(dev,
+				state,
+				"resume",
+				report_resume);
+
+	return status;
+}
+
+/**
+ * handle_error_source - handle logging error into an event log
+ * @aerdev: pointer to pcie_device data structure of the root port
+ * @dev: pointer to pci_dev data structure of error source device
+ * @info: comprehensive error information
+ *
+ * Invoked when an error being detected by Root Port.
+ **/
+static void handle_error_source(struct pcie_device * aerdev,
+	struct pci_dev *dev,
+	struct aer_err_info info)
+{
+	pci_ers_result_t status = 0;
+	int pos;
+
+	if (info.severity == AER_CORRECTABLE) {
+		/*
+		 * Correctable error does not need software intevention.
+		 * No need to go through error recovery process.
+		 */
+		pos = pci_find_aer_capability(dev);
+		if (pos)
+			pci_write_config_dword(dev, pos + PCI_ERR_COR_STATUS,
+					info.status);
+	} else {
+		status = do_recovery(aerdev, dev, info.severity);
+		if (status == PCI_ERS_RESULT_RECOVERED) {
+			printk(KERN_DEBUG "AER driver successfully recovered\n");
+		} else {
+			/* TODO: Should kernel panic here? */
+			printk(KERN_DEBUG "AER driver didn't recover\n");
+		}
+	}
+}
+
+/**
+ * aer_enable_rootport - enable Root Port's interrupts when receiving messages
+ * @rpc: pointer to a Root Port data structure
+ *
+ * Invoked when PCIE bus loads AER service driver.
+ **/
+void aer_enable_rootport(struct aer_rpc *rpc)
+{
+	struct pci_dev *pdev = rpc->rpd->port;
+	int pos, aer_pos;
+	u16 reg16;
+	u32 reg32;
+
+	pos = pci_find_capability(pdev, PCI_CAP_ID_EXP);
+	/* Clear PCIE Capability's Device Status */
+	pci_read_config_word(pdev, pos+PCI_EXP_DEVSTA, &reg16);
+	pci_write_config_word(pdev, pos+PCI_EXP_DEVSTA, reg16);
+
+	/* Disable system error generation in response to error messages */
+	pci_read_config_word(pdev, pos + PCI_EXP_RTCTL, &reg16);
+	reg16 &= ~(SYSTEM_ERROR_INTR_ON_MESG_MASK);
+	pci_write_config_word(pdev, pos + PCI_EXP_RTCTL, reg16);
+
+	aer_pos = pci_find_aer_capability(pdev);
+	/* Clear error status */
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_ROOT_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_ROOT_STATUS, reg32);
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_COR_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_COR_STATUS, reg32);
+	pci_read_config_dword(pdev, aer_pos + PCI_ERR_UNCOR_STATUS, &reg32);
+	pci_write_config_dword(pdev, aer_pos + PCI_ERR_UNCOR_STATUS, reg32);
+
+	/* Enable Root Port device reporting error itself */
+	pci_read_config_word(pdev, pos+PCI_EXP_DEVCTL, &reg16);
+	reg16 = reg16 |
+		PCI_EXP_DEVCTL_CERE |
+		PCI_EXP_DEVCTL_NFERE |
+		PCI_EXP_DEVCTL_FERE |
+		PCI_EXP_DEVCTL_URRE;
+	pci_write_config_word(pdev, pos+PCI_EXP_DEVCTL,
+		reg16);
+
+	/* Enable Root Port's interrupt in response to error messages */
+	pci_write_config_dword(pdev,
+		aer_pos + PCI_ERR_ROOT_COMMAND,
+		ROOT_PORT_INTR_ON_MESG_MASK);
+}
+
+/**
+ * disable_root_aer - disable Root Port's interrupts when receiving messages
+ * @rpc: pointer to a Root Port data structure
+ *
+ * Invoked when PCIE bus unloads AER service driver.
+ **/
+static void disable_root_aer(struct aer_rpc *rpc)
+{
+	struct pci_dev *pdev = rpc->rpd->port;
+	u32 reg32;
+	int pos;
+
+	pos = pci_find_aer_capability(pdev);
+	/* Disable Root's interrupt in response to error messages */
+	pci_write_config_dword(pdev, pos + PCI_ERR_ROOT_COMMAND, 0);
+
+	/* Clear Root's error status reg */
+	pci_read_config_dword(pdev, pos + PCI_ERR_ROOT_STATUS, &reg32);
+	pci_write_config_dword(pdev, pos + PCI_ERR_ROOT_STATUS, reg32);
+}
+
+/**
+ * get_e_source - retrieve an error source
+ * @rpc: pointer to the root port which holds an error
+ *
+ * Invoked by DPC handler to consume an error.
+ **/
+static struct aer_err_source* get_e_source(struct aer_rpc *rpc)
+{
+	struct aer_err_source *e_source;
+	unsigned long flags;
+
+	/* Lock access to Root error producer/consumer index */
+	spin_lock_irqsave(&rpc->e_lock, flags);
+	if (rpc->prod_idx == rpc->cons_idx) {
+		spin_unlock_irqrestore(&rpc->e_lock, flags);
+		return NULL;
+	}
+	e_source = &rpc->e_sources[rpc->cons_idx];
+	rpc->cons_idx++;
+	if (rpc->cons_idx == AER_ERROR_SOURCES_MAX)
+		rpc->cons_idx = 0;
+	spin_unlock_irqrestore(&rpc->e_lock, flags);
+
+	return e_source;
+}
+
+static int get_device_error_info(struct pci_dev *dev, struct aer_err_info *info)
+{
+	int pos;
+
+	pos = pci_find_aer_capability(dev);
+
+	/* The device might not support AER */
+	if (!pos)
+		return AER_SUCCESS;
+
+	if (info->severity == AER_CORRECTABLE) {
+		pci_read_config_dword(dev, pos + PCI_ERR_COR_STATUS,
+			&info->status);
+		if (!(info->status & ERR_CORRECTABLE_ERROR_MASK))
+			return AER_UNSUCCESS;
+	} else if (dev->hdr_type & PCI_HEADER_TYPE_BRIDGE ||
+		info->severity == AER_NONFATAL) {
+
+		/* Link is still healthy for IO reads */
+		pci_read_config_dword(dev, pos + PCI_ERR_UNCOR_STATUS,
+			&info->status);
+		if (!(info->status & ERR_UNCORRECTABLE_ERROR_MASK))
+			return AER_UNSUCCESS;
+
+		if (info->status & AER_LOG_TLP_MASKS) {
+			info->flags |= AER_TLP_HEADER_VALID_FLAG;
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG, &info->tlp.dw0);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 4, &info->tlp.dw1);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 8, &info->tlp.dw2);
+			pci_read_config_dword(dev,
+				pos + PCI_ERR_HEADER_LOG + 12, &info->tlp.dw3);
+		}
+	}
+
+	return AER_SUCCESS;
+}
+
+/**
+ * aer_isr_one_error - consume an error detected by root port
+ * @p_device: pointer to error root port service device
+ * @e_src: pointer to an error source
+ **/
+static void aer_isr_one_error(struct pcie_device *p_device,
+		struct aer_err_source *e_src)
+{
+	struct device *s_device;
+	struct aer_err_info e_info = {0, 0, 0,};
+	int i;
+	u16 id;
+
+	/*
+	 * There is a possibility that both correctable error and
+	 * uncorrectable error being logged. Report correctable error first.
+	 */
+	for (i = 1; i & ROOT_ERR_STATUS_MASKS ; i <<= 2) {
+		if (i > 4)
+			break;
+		if (!(e_src->status & i))
+			continue;
+
+		/* Init comprehensive error information */
+		if (i & PCI_ERR_ROOT_COR_RCV) {
+			id = ERR_COR_ID(e_src->id);
+			e_info.severity = AER_CORRECTABLE;
+		} else {
+			id = ERR_UNCOR_ID(e_src->id);
+			e_info.severity = ((e_src->status >> 6) & 1);
+		}
+		if (e_src->status &
+			(PCI_ERR_ROOT_MULTI_COR_RCV |
+			 PCI_ERR_ROOT_MULTI_UNCOR_RCV))
+			e_info.flags |= AER_MULTI_ERROR_VALID_FLAG;
+		if (!(s_device = find_source_device(p_device->port, id))) {
+			printk(KERN_DEBUG "%s->can't find device of ID%04x\n",
+				__FUNCTION__, id);
+			continue;
+		}
+		if (get_device_error_info(to_pci_dev(s_device), &e_info) ==
+				AER_SUCCESS) {
+			aer_print_error(to_pci_dev(s_device), &e_info);
+			handle_error_source(p_device,
+				to_pci_dev(s_device),
+				e_info);
+		}
+	}
+}
+
+/**
+ * aer_isr - consume errors detected by root port
+ * @context: pointer to a private data of pcie device
+ *
+ * Invoked, as DPC, when root port records new detected error
+ **/
+void aer_isr(void *context)
+{
+	struct pcie_device *p_device = (struct pcie_device *) context;
+	struct aer_rpc *rpc = get_service_data(p_device);
+	struct aer_err_source *e_src;
+
+	mutex_lock(&rpc->rpc_mutex);
+	e_src = get_e_source(rpc);
+	while (e_src) {
+		aer_isr_one_error(p_device, e_src);
+		e_src = get_e_source(rpc);
+	}
+	mutex_unlock(&rpc->rpc_mutex);
+
+	wake_up(&rpc->wait_release);
+}
+
+/**
+ * aer_delete_rootport - disable root port aer and delete service data
+ * @rpc: pointer to a root port device being deleted
+ *
+ * Invoked when AER service unloaded on a specific Root Port
+ **/
+void aer_delete_rootport(struct aer_rpc *rpc)
+{
+	/* Disable root port AER itself */
+	disable_root_aer(rpc);
+
+	kfree(rpc);
+}
+
+/**
+ * aer_init - provide AER initialization
+ * @dev: pointer to AER pcie device
+ *
+ * Invoked when AER service driver is loaded.
+ **/
+int aer_init(struct pcie_device *dev)
+{
+	int status;
+
+	/* Run _OSC Method */
+	status = aer_osc_setup(dev->port);
+
+	if(status != OSC_METHOD_RUN_SUCCESS) {
+		printk(KERN_DEBUG "%s: AER service init fails - %s\n",
+		__FUNCTION__,
+		(status == OSC_METHOD_NOT_SUPPORTED) ?
+			"No ACPI _OSC support" : "Run ACPI _OSC fails");
+
+		if (!forceload)
+			return status;
+	}
+
+	return AER_SUCCESS;
+}
+
+EXPORT_SYMBOL_GPL(pci_find_aer_capability);
+EXPORT_SYMBOL_GPL(pci_enable_pcie_error_reporting);
+EXPORT_SYMBOL_GPL(pci_disable_pcie_error_reporting);
+EXPORT_SYMBOL_GPL(pci_cleanup_aer_uncorrect_error_status);
+
diff --git a/drivers/pci/pcie/aer/aerdrv_errprint.c b/drivers/pci/pcie/aer/aerdrv_errprint.c
new file mode 100644
index 000000000000..3933d4f30e8c
--- /dev/null
+++ b/drivers/pci/pcie/aer/aerdrv_errprint.c
@@ -0,0 +1,248 @@
+/*
+ * drivers/pci/pcie/aer/aerdrv_errprint.c
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Format error messages and print them to console.
+ *
+ * Copyright (C) 2006 Intel Corp.
+ *	Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *	Zhang Yanmin (yanmin.zhang@intel.com)
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/pm.h>
+#include <linux/suspend.h>
+
+#include "aerdrv.h"
+
+#define AER_AGENT_RECEIVER		0
+#define AER_AGENT_REQUESTER		1
+#define AER_AGENT_COMPLETER		2
+#define AER_AGENT_TRANSMITTER		3
+
+#define AER_AGENT_REQUESTER_MASK	(PCI_ERR_UNC_COMP_TIME|	\
+					PCI_ERR_UNC_UNSUP)
+
+#define AER_AGENT_COMPLETER_MASK	PCI_ERR_UNC_COMP_ABORT
+
+#define AER_AGENT_TRANSMITTER_MASK(t, e) (e & (PCI_ERR_COR_REP_ROLL| \
+	((t == AER_CORRECTABLE) ? PCI_ERR_COR_REP_TIMER: 0)))
+
+#define AER_GET_AGENT(t, e)						\
+	((e & AER_AGENT_COMPLETER_MASK) ? AER_AGENT_COMPLETER :		\
+	(e & AER_AGENT_REQUESTER_MASK) ? AER_AGENT_REQUESTER :		\
+	(AER_AGENT_TRANSMITTER_MASK(t, e)) ? AER_AGENT_TRANSMITTER :	\
+	AER_AGENT_RECEIVER)
+
+#define AER_PHYSICAL_LAYER_ERROR_MASK	PCI_ERR_COR_RCVR
+#define AER_DATA_LINK_LAYER_ERROR_MASK(t, e)	\
+		(PCI_ERR_UNC_DLP|		\
+		PCI_ERR_COR_BAD_TLP| 		\
+		PCI_ERR_COR_BAD_DLLP|		\
+		PCI_ERR_COR_REP_ROLL| 		\
+		((t == AER_CORRECTABLE) ?	\
+		PCI_ERR_COR_REP_TIMER: 0))
+
+#define AER_PHYSICAL_LAYER_ERROR	0
+#define AER_DATA_LINK_LAYER_ERROR	1
+#define AER_TRANSACTION_LAYER_ERROR	2
+
+#define AER_GET_LAYER_ERROR(t, e)				\
+	((e & AER_PHYSICAL_LAYER_ERROR_MASK) ?			\
+	AER_PHYSICAL_LAYER_ERROR :				\
+	(e & AER_DATA_LINK_LAYER_ERROR_MASK(t, e)) ?		\
+		AER_DATA_LINK_LAYER_ERROR : 			\
+		AER_TRANSACTION_LAYER_ERROR)
+
+/*
+ * AER error strings
+ */
+static char* aer_error_severity_string[] = {
+	"Uncorrected (Non-Fatal)",
+	"Uncorrected (Fatal)",
+	"Corrected"
+};
+
+static char* aer_error_layer[] = {
+	"Physical Layer",
+	"Data Link Layer",
+	"Transaction Layer"
+};
+static char* aer_correctable_error_string[] = {
+	"Receiver Error        ",	/* Bit Position 0 	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Bad TLP               ",	/* Bit Position 6 	*/
+	"Bad DLLP              ",	/* Bit Position 7 	*/
+	"RELAY_NUM Rollover    ",	/* Bit Position 8 	*/
+	NULL,
+	NULL,
+	NULL,
+	"Replay Timer Timeout  ",	/* Bit Position 12 	*/
+	"Advisory Non-Fatal    ", 	/* Bit Position 13	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+};
+
+static char* aer_uncorrectable_error_string[] = {
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Data Link Protocol    ",	/* Bit Position 4	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	"Poisoned TLP          ",	/* Bit Position 12 	*/
+	"Flow Control Protocol ",	/* Bit Position 13	*/
+	"Completion Timeout    ",	/* Bit Position 14 	*/
+	"Completer Abort       ",	/* Bit Position 15 	*/
+	"Unexpected Completion ",	/* Bit Position 16	*/
+	"Receiver Overflow     ",	/* Bit Position 17	*/
+	"Malformed TLP         ",	/* Bit Position 18	*/
+	"ECRC                  ",	/* Bit Position 19	*/
+	"Unsupported Request   ",	/* Bit Position 20	*/
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+	NULL,
+};
+
+static char* aer_agent_string[] = {
+	"Receiver ID",
+	"Requester ID",
+	"Completer ID",
+	"Transmitter ID"
+};
+
+static char * aer_get_error_source_name(int severity,
+			unsigned int status,
+			char errmsg_buff[])
+{
+	int i;
+	char * errmsg = NULL;
+
+	for (i = 0; i < 32; i++) {
+		if (!(status & (1 << i)))
+			continue;
+
+		if (severity == AER_CORRECTABLE)
+			errmsg = aer_correctable_error_string[i];
+		else
+			errmsg = aer_uncorrectable_error_string[i];
+
+		if (!errmsg) {
+			sprintf(errmsg_buff, "Unknown Error Bit %2d  ", i);
+			errmsg = errmsg_buff;
+		}
+
+		break;
+	}
+
+	return errmsg;
+}
+
+static DEFINE_SPINLOCK(logbuf_lock);
+static char errmsg_buff[100];
+void aer_print_error(struct pci_dev *dev, struct aer_err_info *info)
+{
+	char * errmsg;
+	int err_layer, agent;
+	char * loglevel;
+
+	if (info->severity == AER_CORRECTABLE)
+		loglevel = KERN_WARNING;
+	else
+		loglevel = KERN_ERR;
+
+	printk("%s+------ PCI-Express Device Error ------+\n", loglevel);
+	printk("%sError Severity\t\t: %s\n", loglevel,
+		aer_error_severity_string[info->severity]);
+
+	if ( info->status == 0) {
+		printk("%sPCIE Bus Error type\t: (Unaccessible)\n", loglevel);
+		printk("%sUnaccessible Received\t: %s\n", loglevel,
+			info->flags & AER_MULTI_ERROR_VALID_FLAG ?
+				"Multiple" : "First");
+		printk("%sUnregistered Agent ID\t: %04x\n", loglevel,
+			(dev->bus->number << 8) | dev->devfn);
+	} else {
+		err_layer = AER_GET_LAYER_ERROR(info->severity, info->status);
+		printk("%sPCIE Bus Error type\t: %s\n", loglevel,
+			aer_error_layer[err_layer]);
+
+		spin_lock(&logbuf_lock);
+		errmsg = aer_get_error_source_name(info->severity,
+				info->status,
+				errmsg_buff);
+		printk("%s%s\t: %s\n", loglevel, errmsg,
+			info->flags & AER_MULTI_ERROR_VALID_FLAG ?
+				"Multiple" : "First");
+		spin_unlock(&logbuf_lock);
+
+		agent = AER_GET_AGENT(info->severity, info->status);
+		printk("%s%s\t\t: %04x\n", loglevel,
+			aer_agent_string[agent],
+			(dev->bus->number << 8) | dev->devfn);
+
+		printk("%sVendorID=%04xh, DeviceID=%04xh,"
+			" Bus=%02xh, Device=%02xh, Function=%02xh\n",
+			loglevel,
+			dev->vendor,
+			dev->device,
+			dev->bus->number,
+			PCI_SLOT(dev->devfn),
+			PCI_FUNC(dev->devfn));
+
+		if (info->flags & AER_TLP_HEADER_VALID_FLAG) {
+			unsigned char *tlp = (unsigned char *) &info->tlp;
+			printk("%sTLB Header:\n", loglevel);
+			printk("%s%02x%02x%02x%02x %02x%02x%02x%02x"
+				" %02x%02x%02x%02x %02x%02x%02x%02x\n",
+				loglevel,
+				*(tlp + 3), *(tlp + 2), *(tlp + 1), *tlp,
+				*(tlp + 7), *(tlp + 6), *(tlp + 5), *(tlp + 4),
+				*(tlp + 11), *(tlp + 10), *(tlp + 9),
+				*(tlp + 8), *(tlp + 15), *(tlp + 14),
+				*(tlp + 13), *(tlp + 12));
+		}
+	}
+}
+
diff --git a/include/linux/aer.h b/include/linux/aer.h
new file mode 100644
index 000000000000..402e178b38eb
--- /dev/null
+++ b/include/linux/aer.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (C) 2006 Intel Corp.
+ *     Tom Long Nguyen (tom.l.nguyen@intel.com)
+ *     Zhang Yanmin (yanmin.zhang@intel.com)
+ */
+
+#ifndef _AER_H_
+#define _AER_H_
+
+#if defined(CONFIG_PCIEAER)
+/* pci-e port driver needs this function to enable aer */
+extern int pci_enable_pcie_error_reporting(struct pci_dev *dev);
+extern int pci_find_aer_capability(struct pci_dev *dev);
+extern int pci_disable_pcie_error_reporting(struct pci_dev *dev);
+extern int pci_cleanup_aer_uncorrect_error_status(struct pci_dev *dev);
+#else
+#define pci_enable_pcie_error_reporting(dev)		do { } while (0)
+#define pci_find_aer_capability(dev)			do { } while (0)
+#define pci_disable_pcie_error_reporting(dev)		do { } while (0)
+#define pci_cleanup_aer_uncorrect_error_status(dev)	do { } while (0)
+#endif
+
+#endif //_AER_H_
+
diff --git a/include/linux/pcieport_if.h b/include/linux/pcieport_if.h
index b44e01a70914..6cd91e3f9820 100644
--- a/include/linux/pcieport_if.h
+++ b/include/linux/pcieport_if.h
@@ -62,6 +62,12 @@ struct pcie_port_service_driver {
 	int (*suspend) (struct pcie_device *dev, pm_message_t state);
 	int (*resume) (struct pcie_device *dev);
 
+	/* Service Error Recovery Handler */
+	struct pci_error_handlers *err_handler;
+
+	/* Link Reset Capability - AER service driver specific */
+	pci_ers_result_t (*reset_link) (struct pci_dev *dev);
+
 	const struct pcie_port_service_id *id_table;
 	struct device_driver driver;
 };
-- 
cgit v1.2.1


From 4bf3392e0bf55e5aabbd7bbdbc52cc58eb63f837 Mon Sep 17 00:00:00 2001
From: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Date: Mon, 31 Jul 2006 15:26:16 +0800
Subject: PCI-Express AER implemetation: pcie_portdrv error handler

Patch 4 implements error handlers for pcie_portdrv.

Signed-off-by: Zhang Yanmin <yanmin.zhang@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pcie/portdrv_pci.c | 196 ++++++++++++++++++++++++++++++++++++-----
 1 file changed, 175 insertions(+), 21 deletions(-)

diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index 3284199ce396..e4a2429986f0 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -14,8 +14,10 @@
 #include <linux/init.h>
 #include <linux/slab.h>
 #include <linux/pcieport_if.h>
+#include <linux/aer.h>
 
 #include "portdrv.h"
+#include "aer/aerdrv.h"
 
 /*
  * Version Information
@@ -30,6 +32,43 @@ MODULE_LICENSE("GPL");
 /* global data */
 static const char device_name[] = "pcieport-driver";
 
+static int pcie_portdrv_save_config(struct pci_dev *dev)
+{
+	return pci_save_state(dev);
+}
+
+#ifdef CONFIG_PM
+static int pcie_portdrv_restore_config(struct pci_dev *dev)
+{
+	int retval;
+
+	pci_restore_state(dev);
+	retval = pci_enable_device(dev);
+	if (retval)
+		return retval;
+	pci_set_master(dev);
+	return 0;
+}
+
+static int pcie_portdrv_suspend(struct pci_dev *dev, pm_message_t state)
+{
+	int ret = pcie_port_device_suspend(dev, state);
+
+	if (!ret)
+		ret = pcie_portdrv_save_config(dev);
+	return ret;
+}
+
+static int pcie_portdrv_resume(struct pci_dev *dev)
+{
+	pcie_portdrv_restore_config(dev);
+	return pcie_port_device_resume(dev);
+}
+#else
+#define pcie_portdrv_suspend NULL
+#define pcie_portdrv_resume NULL
+#endif
+
 /*
  * pcie_portdrv_probe - Probe PCI-Express port devices
  * @dev: PCI-Express port device being probed
@@ -61,6 +100,10 @@ static int __devinit pcie_portdrv_probe (struct pci_dev *dev,
 		return -ENOMEM;
 	}
 
+	pcie_portdrv_save_config(dev);
+
+	pci_enable_pcie_error_reporting(dev);
+
 	return 0;
 }
 
@@ -70,39 +113,143 @@ static void pcie_portdrv_remove (struct pci_dev *dev)
 	kfree(pci_get_drvdata(dev));
 }
 
-#ifdef CONFIG_PM
-static int pcie_portdrv_save_config(struct pci_dev *dev)
+static int error_detected_iter(struct device *device, void *data)
 {
-	return pci_save_state(dev);
+	struct pcie_device *pcie_device;
+	struct pcie_port_service_driver *driver;
+	struct aer_broadcast_data *result_data;
+	pci_ers_result_t status;
+
+	result_data = (struct aer_broadcast_data *) data;
+
+	if (device->bus == &pcie_port_bus_type && device->driver) {
+		driver = to_service_driver(device->driver);
+		if (!driver ||
+			!driver->err_handler ||
+			!driver->err_handler->error_detected)
+			return 0;
+
+		pcie_device = to_pcie_device(device);
+
+		/* Forward error detected message to service drivers */
+		status = driver->err_handler->error_detected(
+			pcie_device->port,
+			result_data->state);
+		result_data->result =
+			merge_result(result_data->result, status);
+	}
+
+	return 0;
 }
 
-static int pcie_portdrv_restore_config(struct pci_dev *dev)
+static pci_ers_result_t pcie_portdrv_error_detected(struct pci_dev *dev,
+					enum pci_channel_state error)
 {
-	int retval;
+	struct aer_broadcast_data result_data =
+			{error, PCI_ERS_RESULT_CAN_RECOVER};
+
+	device_for_each_child(&dev->dev, &result_data, error_detected_iter);
+
+	return result_data.result;
+}
+
+static int mmio_enabled_iter(struct device *device, void *data)
+{
+	struct pcie_device *pcie_device;
+	struct pcie_port_service_driver *driver;
+	pci_ers_result_t status, *result;
+
+	result = (pci_ers_result_t *) data;
+
+	if (device->bus == &pcie_port_bus_type && device->driver) {
+		driver = to_service_driver(device->driver);
+		if (driver &&
+			driver->err_handler &&
+			driver->err_handler->mmio_enabled) {
+			pcie_device = to_pcie_device(device);
+
+			/* Forward error message to service drivers */
+			status = driver->err_handler->mmio_enabled(
+					pcie_device->port);
+			*result = merge_result(*result, status);
+		}
+	}
 
-	pci_restore_state(dev);
-	retval = pci_enable_device(dev);
-	if (retval)
-		return retval;
-	pci_set_master(dev);
 	return 0;
 }
 
-static int pcie_portdrv_suspend (struct pci_dev *dev, pm_message_t state)
+static pci_ers_result_t pcie_portdrv_mmio_enabled(struct pci_dev *dev)
 {
-	int ret = pcie_port_device_suspend(dev, state);
+	pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED;
 
-	if (!ret)
-		ret = pcie_portdrv_save_config(dev);
-	return ret;
+	device_for_each_child(&dev->dev, &status, mmio_enabled_iter);
+	return status;
 }
 
-static int pcie_portdrv_resume (struct pci_dev *dev)
+static int slot_reset_iter(struct device *device, void *data)
 {
-	pcie_portdrv_restore_config(dev);
-	return pcie_port_device_resume(dev);
+	struct pcie_device *pcie_device;
+	struct pcie_port_service_driver *driver;
+	pci_ers_result_t status, *result;
+
+	result = (pci_ers_result_t *) data;
+
+	if (device->bus == &pcie_port_bus_type && device->driver) {
+		driver = to_service_driver(device->driver);
+		if (driver &&
+			driver->err_handler &&
+			driver->err_handler->slot_reset) {
+			pcie_device = to_pcie_device(device);
+
+			/* Forward error message to service drivers */
+			status = driver->err_handler->slot_reset(
+					pcie_device->port);
+			*result = merge_result(*result, status);
+		}
+	}
+
+	return 0;
+}
+
+static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev)
+{
+	pci_ers_result_t status;
+
+	/* If fatal, restore cfg space for possible link reset at upstream */
+	if (dev->error_state == pci_channel_io_frozen) {
+		pcie_portdrv_restore_config(dev);
+		pci_enable_pcie_error_reporting(dev);
+	}
+
+	device_for_each_child(&dev->dev, &status, slot_reset_iter);
+
+	return status;
+}
+
+static int resume_iter(struct device *device, void *data)
+{
+	struct pcie_device *pcie_device;
+	struct pcie_port_service_driver *driver;
+
+	if (device->bus == &pcie_port_bus_type && device->driver) {
+		driver = to_service_driver(device->driver);
+		if (driver &&
+			driver->err_handler &&
+			driver->err_handler->resume) {
+			pcie_device = to_pcie_device(device);
+
+			/* Forward error message to service drivers */
+			driver->err_handler->resume(pcie_device->port);
+		}
+	}
+
+	return 0;
+}
+
+static void pcie_portdrv_err_resume(struct pci_dev *dev)
+{
+	device_for_each_child(&dev->dev, NULL, resume_iter);
 }
-#endif
 
 /*
  * LINUX Device Driver Model
@@ -114,6 +261,13 @@ static const struct pci_device_id port_pci_ids[] = { {
 };
 MODULE_DEVICE_TABLE(pci, port_pci_ids);
 
+static struct pci_error_handlers pcie_portdrv_err_handler = {
+		.error_detected = pcie_portdrv_error_detected,
+		.mmio_enabled = pcie_portdrv_mmio_enabled,
+		.slot_reset = pcie_portdrv_slot_reset,
+		.resume = pcie_portdrv_err_resume,
+};
+
 static struct pci_driver pcie_portdrv = {
 	.name		= (char *)device_name,
 	.id_table	= &port_pci_ids[0],
@@ -121,10 +275,10 @@ static struct pci_driver pcie_portdrv = {
 	.probe		= pcie_portdrv_probe,
 	.remove		= pcie_portdrv_remove,
 
-#ifdef	CONFIG_PM
 	.suspend	= pcie_portdrv_suspend,
 	.resume		= pcie_portdrv_resume,
-#endif	/* PM */
+
+	.err_handler 	= &pcie_portdrv_err_handler,
 };
 
 static int __init pcie_portdrv_init(void)
-- 
cgit v1.2.1


From e1b95dc6b1cd02c3625ba3d1d770d095d6a4b313 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: SHPCHP: fix __must_check warnings

Cc: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/shpchp.h       | 2 +-
 drivers/pci/hotplug/shpchp_core.c  | 6 +++++-
 drivers/pci/hotplug/shpchp_sysfs.c | 4 ++--
 3 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/hotplug/shpchp.h b/drivers/pci/hotplug/shpchp.h
index 7208b95c6ee7..c7103ac5cd06 100644
--- a/drivers/pci/hotplug/shpchp.h
+++ b/drivers/pci/hotplug/shpchp.h
@@ -173,7 +173,7 @@ struct controller {
 #define msg_button_cancel	"PCI slot #%s - action canceled due to button press.\n"
 
 /* sysfs functions for the hotplug controller info */
-extern void shpchp_create_ctrl_files	(struct controller *ctrl);
+extern int __must_check shpchp_create_ctrl_files(struct controller *ctrl);
 
 extern int	shpchp_sysfs_enable_slot(struct slot *slot);
 extern int	shpchp_sysfs_disable_slot(struct slot *slot);
diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c
index a14e7de19846..235c18a22393 100644
--- a/drivers/pci/hotplug/shpchp_core.c
+++ b/drivers/pci/hotplug/shpchp_core.c
@@ -449,10 +449,14 @@ static int shpc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		ctrl->speed = PCI_SPEED_33MHz;
 	}
 
-	shpchp_create_ctrl_files(ctrl);
+	rc = shpchp_create_ctrl_files(ctrl);
+	if (rc)
+		goto err_cleanup_slots;
 
 	return 0;
 
+err_cleanup_slots:
+	cleanup_slots(ctrl);
 err_out_release_ctlr:
 	ctrl->hpc_ops->release_ctlr(ctrl);
 err_out_free_ctrl:
diff --git a/drivers/pci/hotplug/shpchp_sysfs.c b/drivers/pci/hotplug/shpchp_sysfs.c
index 620e1139e607..29fa9d26adae 100644
--- a/drivers/pci/hotplug/shpchp_sysfs.c
+++ b/drivers/pci/hotplug/shpchp_sysfs.c
@@ -91,9 +91,9 @@ static ssize_t show_ctrl (struct device *dev, struct device_attribute *attr, cha
 }
 static DEVICE_ATTR (ctrl, S_IRUGO, show_ctrl, NULL);
 
-void shpchp_create_ctrl_files (struct controller *ctrl)
+int __must_check shpchp_create_ctrl_files (struct controller *ctrl)
 {
-	device_create_file (&ctrl->pci_dev->dev, &dev_attr_ctrl);
+	return device_create_file (&ctrl->pci_dev->dev, &dev_attr_ctrl);
 }
 
 void shpchp_remove_ctrl_files(struct controller *ctrl)
-- 
cgit v1.2.1


From 660a0e8fdf85f30b1e5f6905a78361476094eb7c Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: PCI Hotplug: fix __must_check warnings

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/pci_hotplug.h      |   4 +-
 drivers/pci/hotplug/pci_hotplug_core.c | 157 ++++++++++++++++++++++++++-------
 2 files changed, 128 insertions(+), 33 deletions(-)

diff --git a/drivers/pci/hotplug/pci_hotplug.h b/drivers/pci/hotplug/pci_hotplug.h
index e929b7c11429..772523dc3860 100644
--- a/drivers/pci/hotplug/pci_hotplug.h
+++ b/drivers/pci/hotplug/pci_hotplug.h
@@ -172,8 +172,8 @@ struct hotplug_slot {
 
 extern int pci_hp_register		(struct hotplug_slot *slot);
 extern int pci_hp_deregister		(struct hotplug_slot *slot);
-extern int pci_hp_change_slot_info	(struct hotplug_slot *slot,
-					 struct hotplug_slot_info *info);
+extern int __must_check pci_hp_change_slot_info	(struct hotplug_slot *slot,
+						 struct hotplug_slot_info *info);
 extern struct subsystem pci_hotplug_slots_subsys;
 
 /* PCI Setting Record (Type 0) */
diff --git a/drivers/pci/hotplug/pci_hotplug_core.c b/drivers/pci/hotplug/pci_hotplug_core.c
index b7b378df89e3..e2823ea9c4ed 100644
--- a/drivers/pci/hotplug/pci_hotplug_core.c
+++ b/drivers/pci/hotplug/pci_hotplug_core.c
@@ -482,31 +482,95 @@ static int has_test_file (struct hotplug_slot *slot)
 
 static int fs_add_slot (struct hotplug_slot *slot)
 {
-	if (has_power_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_power.attr);
+	int retval = 0;
 
-	if (has_attention_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_attention.attr);
+	if (has_power_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj, &hotplug_slot_attr_power.attr);
+		if (retval)
+			goto exit_power;
+	}
 
-	if (has_latch_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_latch.attr);
+	if (has_attention_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_attention.attr);
+		if (retval)
+			goto exit_attention;
+	}
 
-	if (has_adapter_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
+	if (has_latch_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_latch.attr);
+		if (retval)
+			goto exit_latch;
+	}
 
-	if (has_address_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_address.attr);
+	if (has_adapter_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_presence.attr);
+		if (retval)
+			goto exit_adapter;
+	}
 
-	if (has_max_bus_speed_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
+	if (has_address_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_address.attr);
+		if (retval)
+			goto exit_address;
+	}
 
+	if (has_max_bus_speed_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_max_bus_speed.attr);
+		if (retval)
+			goto exit_max_speed;
+	}
+
+	if (has_cur_bus_speed_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_cur_bus_speed.attr);
+		if (retval)
+			goto exit_cur_speed;
+	}
+
+	if (has_test_file(slot) == 0) {
+		retval = sysfs_create_file(&slot->kobj,
+					   &hotplug_slot_attr_test.attr);
+		if (retval)
+			goto exit_test;
+	}
+
+	goto exit;
+
+exit_test:
 	if (has_cur_bus_speed_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr);
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr);
 
-	if (has_test_file(slot) == 0)
-		sysfs_create_file(&slot->kobj, &hotplug_slot_attr_test.attr);
+exit_cur_speed:
+	if (has_max_bus_speed_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
 
-	return 0;
+exit_max_speed:
+	if (has_address_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_address.attr);
+
+exit_address:
+	if (has_adapter_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
+
+exit_adapter:
+	if (has_latch_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_latch.attr);
+
+exit_latch:
+	if (has_attention_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_attention.attr);
+
+exit_attention:
+	if (has_power_file(slot) == 0)
+		sysfs_remove_file(&slot->kobj, &hotplug_slot_attr_power.attr);
+exit_power:
+exit:
+	return retval;
 }
 
 static void fs_remove_slot (struct hotplug_slot *slot)
@@ -626,8 +690,11 @@ int pci_hp_deregister (struct hotplug_slot *slot)
  *
  * Returns 0 if successful, anything else for an error.
  */
-int pci_hp_change_slot_info (struct hotplug_slot *slot, struct hotplug_slot_info *info)
+int __must_check pci_hp_change_slot_info(struct hotplug_slot *slot,
+					 struct hotplug_slot_info *info)
 {
+	int retval;
+
 	if ((slot == NULL) || (info == NULL))
 		return -ENODEV;
 
@@ -636,32 +703,60 @@ int pci_hp_change_slot_info (struct hotplug_slot *slot, struct hotplug_slot_info
 	* for the files referring to the fields that have now changed.
 	*/
 	if ((has_power_file(slot) == 0) &&
-	    (slot->info->power_status != info->power_status))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_power.attr);
+	    (slot->info->power_status != info->power_status)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_power.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_attention_file(slot) == 0) &&
-	    (slot->info->attention_status != info->attention_status))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_attention.attr);
+	    (slot->info->attention_status != info->attention_status)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_attention.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_latch_file(slot) == 0) &&
-	    (slot->info->latch_status != info->latch_status))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_latch.attr);
+	    (slot->info->latch_status != info->latch_status)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_latch.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_adapter_file(slot) == 0) &&
-	    (slot->info->adapter_status != info->adapter_status))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_presence.attr);
+	    (slot->info->adapter_status != info->adapter_status)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_presence.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_address_file(slot) == 0) &&
-	    (slot->info->address != info->address))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_address.attr);
+	    (slot->info->address != info->address)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_address.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_max_bus_speed_file(slot) == 0) &&
-	    (slot->info->max_bus_speed != info->max_bus_speed))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_max_bus_speed.attr);
+	    (slot->info->max_bus_speed != info->max_bus_speed)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_max_bus_speed.attr);
+		if (retval)
+			return retval;
+	}
 
 	if ((has_cur_bus_speed_file(slot) == 0) &&
-	    (slot->info->cur_bus_speed != info->cur_bus_speed))
-		sysfs_update_file(&slot->kobj, &hotplug_slot_attr_cur_bus_speed.attr);
+	    (slot->info->cur_bus_speed != info->cur_bus_speed)) {
+		retval = sysfs_update_file(&slot->kobj,
+					   &hotplug_slot_attr_cur_bus_speed.attr);
+		if (retval)
+			return retval;
+	}
 
 	memcpy (slot->info, info, sizeof (struct hotplug_slot_info));
 
-- 
cgit v1.2.1


From b19441af185559118e8247382ea4f2f76ebffc6d Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: PCI: fix __must_check warnings

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/bus.c                  |  22 ++++++--
 drivers/pci/hotplug/fakephp.c      |  18 ++++--
 drivers/pci/pci-driver.c           |   5 +-
 drivers/pci/pci-sysfs.c            | 112 ++++++++++++++++++++++++-------------
 drivers/pci/pcie/aer/aerdrv_core.c |   3 +-
 drivers/pci/pcie/portdrv_core.c    |   6 +-
 drivers/pci/pcie/portdrv_pci.c     |  16 ++++--
 drivers/pci/probe.c                |  16 +++++-
 include/linux/pci.h                |   2 +-
 9 files changed, 138 insertions(+), 62 deletions(-)

diff --git a/drivers/pci/bus.c b/drivers/pci/bus.c
index 5f7db9d2436e..aadaa3c8096b 100644
--- a/drivers/pci/bus.c
+++ b/drivers/pci/bus.c
@@ -77,9 +77,12 @@ pci_bus_alloc_resource(struct pci_bus *bus, struct resource *res,
  * This adds a single pci device to the global
  * device list and adds sysfs and procfs entries
  */
-void __devinit pci_bus_add_device(struct pci_dev *dev)
+int __devinit pci_bus_add_device(struct pci_dev *dev)
 {
-	device_add(&dev->dev);
+	int retval;
+	retval = device_add(&dev->dev);
+	if (retval)
+		return retval;
 
 	down_write(&pci_bus_sem);
 	list_add_tail(&dev->global_list, &pci_devices);
@@ -87,6 +90,7 @@ void __devinit pci_bus_add_device(struct pci_dev *dev)
 
 	pci_proc_attach_device(dev);
 	pci_create_sysfs_dev_files(dev);
+	return 0;
 }
 
 /**
@@ -104,6 +108,7 @@ void __devinit pci_bus_add_device(struct pci_dev *dev)
 void __devinit pci_bus_add_devices(struct pci_bus *bus)
 {
 	struct pci_dev *dev;
+	int retval;
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
 		/*
@@ -112,7 +117,9 @@ void __devinit pci_bus_add_devices(struct pci_bus *bus)
 		 */
 		if (!list_empty(&dev->global_list))
 			continue;
-		pci_bus_add_device(dev);
+		retval = pci_bus_add_device(dev);
+		if (retval)
+			dev_err(&dev->dev, "Error adding device, continuing\n");
 	}
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
@@ -129,10 +136,13 @@ void __devinit pci_bus_add_devices(struct pci_bus *bus)
 			       list_add_tail(&dev->subordinate->node,
 					       &dev->bus->children);
 			       up_write(&pci_bus_sem);
-		       }
+			}
 			pci_bus_add_devices(dev->subordinate);
-
-			sysfs_create_link(&dev->subordinate->class_dev.kobj, &dev->dev.kobj, "bridge");
+			retval = sysfs_create_link(&dev->subordinate->class_dev.kobj,
+						   &dev->dev.kobj, "bridge");
+			if (retval)
+				dev_err(&dev->dev, "Error creating sysfs "
+					"bridge symlink, continuing...\n");
 		}
 	}
 }
diff --git a/drivers/pci/hotplug/fakephp.c b/drivers/pci/hotplug/fakephp.c
index dd2b762777c4..05a4f0f90186 100644
--- a/drivers/pci/hotplug/fakephp.c
+++ b/drivers/pci/hotplug/fakephp.c
@@ -176,7 +176,9 @@ static void pci_rescan_slot(struct pci_dev *temp)
 	struct pci_bus *bus = temp->bus;
 	struct pci_dev *dev;
 	int func;
+	int retval;
 	u8 hdr_type;
+
 	if (!pci_read_config_byte(temp, PCI_HEADER_TYPE, &hdr_type)) {
 		temp->hdr_type = hdr_type & 0x7f;
 		if (!pci_find_slot(bus->number, temp->devfn)) {
@@ -185,8 +187,12 @@ static void pci_rescan_slot(struct pci_dev *temp)
 				dbg("New device on %s function %x:%x\n",
 					bus->name, temp->devfn >> 3,
 					temp->devfn & 7);
-				pci_bus_add_device(dev);
-				add_slot(dev);
+				retval = pci_bus_add_device(dev);
+				if (retval)
+					dev_err(&dev->dev, "error adding "
+						"device, continuing.\n");
+				else
+					add_slot(dev);
 			}
 		}
 		/* multifunction device? */
@@ -205,8 +211,12 @@ static void pci_rescan_slot(struct pci_dev *temp)
 					dbg("New device on %s function %x:%x\n",
 						bus->name, temp->devfn >> 3,
 						temp->devfn & 7);
-					pci_bus_add_device(dev);
-					add_slot(dev);
+					retval = pci_bus_add_device(dev);
+					if (retval)
+						dev_err(&dev->dev, "error adding "
+							"device, continuing.\n");
+					else
+						add_slot(dev);
 				}
 			}
 		}
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index d8ace1f90dd2..309629e03bae 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -56,6 +56,7 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count)
 		subdevice=PCI_ANY_ID, class=0, class_mask=0;
 	unsigned long driver_data=0;
 	int fields=0;
+	int retval = 0;
 
 	fields = sscanf(buf, "%x %x %x %x %x %x %lux",
 			&vendor, &device, &subvendor, &subdevice,
@@ -82,10 +83,12 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count)
 	spin_unlock(&pdrv->dynids.lock);
 
 	if (get_driver(&pdrv->driver)) {
-		driver_attach(&pdrv->driver);
+		retval = driver_attach(&pdrv->driver);
 		put_driver(&pdrv->driver);
 	}
 
+	if (retval)
+		return retval;
 	return count;
 }
 static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id);
diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c
index 010e01c4bd43..a1d2e979b17f 100644
--- a/drivers/pci/pci-sysfs.c
+++ b/drivers/pci/pci-sysfs.c
@@ -117,6 +117,7 @@ is_enabled_store(struct device *dev, struct device_attribute *attr,
 		const char *buf, size_t count)
 {
 	struct pci_dev *pdev = to_pci_dev(dev);
+	int retval = 0;
 
 	/* this can crash the machine when done on the "wrong" device */
 	if (!capable(CAP_SYS_ADMIN))
@@ -126,8 +127,10 @@ is_enabled_store(struct device *dev, struct device_attribute *attr,
 		pci_disable_device(pdev);
 
 	if (*buf == '1')
-		pci_enable_device(pdev);
+		retval = pci_enable_device(pdev);
 
+	if (retval)
+		return retval;
 	return count;
 }
 
@@ -425,16 +428,39 @@ pci_mmap_resource(struct kobject *kobj, struct bin_attribute *attr,
 	return pci_mmap_page_range(pdev, vma, mmap_type, 0);
 }
 
+/**
+ * pci_remove_resource_files - cleanup resource files
+ * @dev: dev to cleanup
+ *
+ * If we created resource files for @dev, remove them from sysfs and
+ * free their resources.
+ */
+static void
+pci_remove_resource_files(struct pci_dev *pdev)
+{
+	int i;
+
+	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
+		struct bin_attribute *res_attr;
+
+		res_attr = pdev->res_attr[i];
+		if (res_attr) {
+			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
+			kfree(res_attr);
+		}
+	}
+}
+
 /**
  * pci_create_resource_files - create resource files in sysfs for @dev
  * @dev: dev in question
  *
  * Walk the resources in @dev creating files for each resource available.
  */
-static void
-pci_create_resource_files(struct pci_dev *pdev)
+static int pci_create_resource_files(struct pci_dev *pdev)
 {
 	int i;
+	int retval;
 
 	/* Expose the PCI resources from this device as files */
 	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
@@ -457,35 +483,19 @@ pci_create_resource_files(struct pci_dev *pdev)
 			res_attr->size = pci_resource_len(pdev, i);
 			res_attr->mmap = pci_mmap_resource;
 			res_attr->private = &pdev->resource[i];
-			sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
-		}
-	}
-}
-
-/**
- * pci_remove_resource_files - cleanup resource files
- * @dev: dev to cleanup
- *
- * If we created resource files for @dev, remove them from sysfs and
- * free their resources.
- */
-static void
-pci_remove_resource_files(struct pci_dev *pdev)
-{
-	int i;
-
-	for (i = 0; i < PCI_ROM_RESOURCE; i++) {
-		struct bin_attribute *res_attr;
-
-		res_attr = pdev->res_attr[i];
-		if (res_attr) {
-			sysfs_remove_bin_file(&pdev->dev.kobj, res_attr);
-			kfree(res_attr);
+			retval = sysfs_create_bin_file(&pdev->dev.kobj, res_attr);
+			if (retval) {
+				pci_remove_resource_files(pdev);
+				return retval;
+			}
+		} else {
+			return -ENOMEM;
 		}
 	}
+	return 0;
 }
 #else /* !HAVE_PCI_MMAP */
-static inline void pci_create_resource_files(struct pci_dev *dev) { return; }
+static inline int pci_create_resource_files(struct pci_dev *dev) { return 0; }
 static inline void pci_remove_resource_files(struct pci_dev *dev) { return; }
 #endif /* HAVE_PCI_MMAP */
 
@@ -570,22 +580,27 @@ static struct bin_attribute pcie_config_attr = {
 	.write = pci_write_config,
 };
 
-int pci_create_sysfs_dev_files (struct pci_dev *pdev)
+int __must_check pci_create_sysfs_dev_files (struct pci_dev *pdev)
 {
+	struct bin_attribute *rom_attr = NULL;
+	int retval;
+
 	if (!sysfs_initialized)
 		return -EACCES;
 
 	if (pdev->cfg_size < 4096)
-		sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pci_config_attr);
 	else
-		sysfs_create_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+		retval = sysfs_create_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+	if (retval)
+		goto err;
 
-	pci_create_resource_files(pdev);
+	retval = pci_create_resource_files(pdev);
+	if (retval)
+		goto err_bin_file;
 
 	/* If the device has a ROM, try to expose it in sysfs. */
 	if (pci_resource_len(pdev, PCI_ROM_RESOURCE)) {
-		struct bin_attribute *rom_attr;
-		
 		rom_attr = kzalloc(sizeof(*rom_attr), GFP_ATOMIC);
 		if (rom_attr) {
 			pdev->rom_attr = rom_attr;
@@ -595,13 +610,28 @@ int pci_create_sysfs_dev_files (struct pci_dev *pdev)
 			rom_attr->attr.owner = THIS_MODULE;
 			rom_attr->read = pci_read_rom;
 			rom_attr->write = pci_write_rom;
-			sysfs_create_bin_file(&pdev->dev.kobj, rom_attr);
+			retval = sysfs_create_bin_file(&pdev->dev.kobj, rom_attr);
+			if (retval)
+				goto err_rom;
+		} else {
+			retval = -ENOMEM;
+			goto err_bin_file;
 		}
 	}
 	/* add platform-specific attributes */
 	pcibios_add_platform_entries(pdev);
-	
+
 	return 0;
+
+err_rom:
+	kfree(rom_attr);
+err_bin_file:
+	if (pdev->cfg_size < 4096)
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pci_config_attr);
+	else
+		sysfs_remove_bin_file(&pdev->dev.kobj, &pcie_config_attr);
+err:
+	return retval;
 }
 
 /**
@@ -630,10 +660,14 @@ void pci_remove_sysfs_dev_files(struct pci_dev *pdev)
 static int __init pci_sysfs_init(void)
 {
 	struct pci_dev *pdev = NULL;
-	
+	int retval;
+
 	sysfs_initialized = 1;
-	for_each_pci_dev(pdev)
-		pci_create_sysfs_dev_files(pdev);
+	for_each_pci_dev(pdev) {
+		retval = pci_create_sysfs_dev_files(pdev);
+		if (retval)
+			return retval;
+	}
 
 	return 0;
 }
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 5591043acea7..1c7e660d6535 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -357,7 +357,8 @@ static int find_aer_service_iter(struct device *device, void *data)
 static void find_aer_service(struct pci_dev *dev,
 		struct find_aer_service_data *data)
 {
-	device_for_each_child(&dev->dev, data, find_aer_service_iter);
+	int retval;
+	retval = device_for_each_child(&dev->dev, data, find_aer_service_iter);
 }
 
 static pci_ers_result_t reset_link(struct pcie_device *aerdev,
diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c
index cf9e810b4bf8..bd6615b4d40e 100644
--- a/drivers/pci/pcie/portdrv_core.c
+++ b/drivers/pci/pcie/portdrv_core.c
@@ -340,8 +340,7 @@ static int suspend_iter(struct device *dev, void *data)
 
 int pcie_port_device_suspend(struct pci_dev *dev, pm_message_t state)
 {
-	device_for_each_child(&dev->dev, &state, suspend_iter);
-	return 0;
+	return device_for_each_child(&dev->dev, &state, suspend_iter);
 }
 
 static int resume_iter(struct device *dev, void *data)
@@ -359,8 +358,7 @@ static int resume_iter(struct device *dev, void *data)
 
 int pcie_port_device_resume(struct pci_dev *dev)
 {
-	device_for_each_child(&dev->dev, NULL, resume_iter);
-	return 0;
+	return device_for_each_child(&dev->dev, NULL, resume_iter);
 }
 #endif
 
diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c
index e4a2429986f0..037690e08f5f 100644
--- a/drivers/pci/pcie/portdrv_pci.c
+++ b/drivers/pci/pcie/portdrv_pci.c
@@ -147,8 +147,10 @@ static pci_ers_result_t pcie_portdrv_error_detected(struct pci_dev *dev,
 {
 	struct aer_broadcast_data result_data =
 			{error, PCI_ERS_RESULT_CAN_RECOVER};
+	int retval;
 
-	device_for_each_child(&dev->dev, &result_data, error_detected_iter);
+	/* can not fail */
+	retval = device_for_each_child(&dev->dev, &result_data, error_detected_iter);
 
 	return result_data.result;
 }
@@ -181,8 +183,10 @@ static int mmio_enabled_iter(struct device *device, void *data)
 static pci_ers_result_t pcie_portdrv_mmio_enabled(struct pci_dev *dev)
 {
 	pci_ers_result_t status = PCI_ERS_RESULT_RECOVERED;
+	int retval;
 
-	device_for_each_child(&dev->dev, &status, mmio_enabled_iter);
+	/* get true return value from &status */
+	retval = device_for_each_child(&dev->dev, &status, mmio_enabled_iter);
 	return status;
 }
 
@@ -214,6 +218,7 @@ static int slot_reset_iter(struct device *device, void *data)
 static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev)
 {
 	pci_ers_result_t status;
+	int retval;
 
 	/* If fatal, restore cfg space for possible link reset at upstream */
 	if (dev->error_state == pci_channel_io_frozen) {
@@ -221,7 +226,8 @@ static pci_ers_result_t pcie_portdrv_slot_reset(struct pci_dev *dev)
 		pci_enable_pcie_error_reporting(dev);
 	}
 
-	device_for_each_child(&dev->dev, &status, slot_reset_iter);
+	/* get true return value from &status */
+	retval = device_for_each_child(&dev->dev, &status, slot_reset_iter);
 
 	return status;
 }
@@ -248,7 +254,9 @@ static int resume_iter(struct device *device, void *data)
 
 static void pcie_portdrv_err_resume(struct pci_dev *dev)
 {
-	device_for_each_child(&dev->dev, NULL, resume_iter);
+	int retval;
+	/* nothing to do with error value, if it ever happens */
+	retval = device_for_each_child(&dev->dev, NULL, resume_iter);
 }
 
 /*
diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c
index c5a58d1c6c1c..a3b0a5eb5054 100644
--- a/drivers/pci/probe.c
+++ b/drivers/pci/probe.c
@@ -339,6 +339,7 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 {
 	struct pci_bus *child;
 	int i;
+	int retval;
 
 	/*
 	 * Allocate a new bus, and inherit stuff from the parent..
@@ -356,8 +357,13 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 
 	child->class_dev.class = &pcibus_class;
 	sprintf(child->class_dev.class_id, "%04x:%02x", pci_domain_nr(child), busnr);
-	class_device_register(&child->class_dev);
-	class_device_create_file(&child->class_dev, &class_device_attr_cpuaffinity);
+	retval = class_device_register(&child->class_dev);
+	if (retval)
+		goto error_register;
+	retval = class_device_create_file(&child->class_dev,
+					  &class_device_attr_cpuaffinity);
+	if (retval)
+		goto error_file_create;
 
 	/*
 	 * Set up the primary, secondary and subordinate
@@ -375,6 +381,12 @@ pci_alloc_child_bus(struct pci_bus *parent, struct pci_dev *bridge, int busnr)
 	bridge->subordinate = child;
 
 	return child;
+
+error_file_create:
+	class_device_unregister(&child->class_dev);
+error_register:
+	kfree(child);
+	return NULL;
 }
 
 struct pci_bus * __devinit pci_add_new_bus(struct pci_bus *parent, struct pci_dev *dev, int busnr)
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 3ec72551ac31..c9bb7bee52c7 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -431,7 +431,7 @@ int pci_scan_slot(struct pci_bus *bus, int devfn);
 struct pci_dev * pci_scan_single_device(struct pci_bus *bus, int devfn);
 void pci_device_add(struct pci_dev *dev, struct pci_bus *bus);
 unsigned int pci_scan_child_bus(struct pci_bus *bus);
-void pci_bus_add_device(struct pci_dev *dev);
+int __must_check pci_bus_add_device(struct pci_dev *dev);
 void pci_read_bridge_bases(struct pci_bus *child);
 struct resource *pci_find_parent_resource(const struct pci_dev *dev, struct resource *res);
 int pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge);
-- 
cgit v1.2.1


From 50b0075520a0acba9cabab5203bbce918b966d9a Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Wed, 16 Aug 2006 17:42:18 +0100
Subject: PCI: Multiprobe sanitizer

There are numerous drivers that can use multithreaded probing but having
some kind of global flag as the way to control this makes migration to
threaded probing hard and since it enables it everywhere and is almost
as likely to cause serious pain as holding a clog dance in a minefield.

If we have a pci_driver multithread_probe flag to inherit you can turn
it on for one driver at a time.

From playing so far however I think we need a different model at the
device layer which serializes until the called probe function says "ok
you can start another one now". That would need some kind of flag and
semaphore plus a helper function.

Anyway in the absence of that this is a starting point to usefully play
with this stuff

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci-driver.c | 6 +++++-
 include/linux/pci.h      | 2 ++
 2 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c
index 309629e03bae..b1c0c707d96c 100644
--- a/drivers/pci/pci-driver.c
+++ b/drivers/pci/pci-driver.c
@@ -421,7 +421,11 @@ int __pci_register_driver(struct pci_driver *drv, struct module *owner)
 	drv->driver.bus = &pci_bus_type;
 	drv->driver.owner = owner;
 	drv->driver.kobj.ktype = &pci_driver_kobj_type;
-	drv->driver.multithread_probe = pci_multithread_probe;
+
+	if (pci_multithread_probe)
+		drv->driver.multithread_probe = pci_multithread_probe;
+	else
+		drv->driver.multithread_probe = drv->multithread_probe;
 
 	spin_lock_init(&drv->dynids.lock);
 	INIT_LIST_HEAD(&drv->dynids.list);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index c9bb7bee52c7..549d8410974b 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -356,6 +356,8 @@ struct pci_driver {
 	struct pci_error_handlers *err_handler;
 	struct device_driver	driver;
 	struct pci_dynids dynids;
+
+	int multithread_probe;
 };
 
 #define	to_pci_driver(drv) container_of(drv,struct pci_driver, driver)
-- 
cgit v1.2.1


From 6d47a5e4c3f8b6458002065d98a9cc6ff90fb597 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Mon, 14 Aug 2006 23:07:38 -0700
Subject: PCI: drivers/pci/hotplug/acpiphp_glue.c: make a function static

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index ae67a8f55ba1..be7e91662417 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -997,7 +997,7 @@ acpiphp_bus_add_out:
  * @handle: handle to acpi namespace
  *
  */
-int acpiphp_bus_trim(acpi_handle handle)
+static int acpiphp_bus_trim(acpi_handle handle)
 {
 	struct acpi_device *device;
 	int retval;
-- 
cgit v1.2.1


From b56a5a23bfecd9cac9187164a9d5f22d287c48b9 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Mon, 21 Aug 2006 16:22:22 +0300
Subject: PCI: Restore PCI Express capability registers after PM event

Restore PCI Express capability registers after PM event.
This includes maxumum MTU for PCI express and other vital data.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/pci.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 50 insertions(+)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 590f4e6f505d..a544997399b3 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -445,6 +445,51 @@ pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state)
 
 EXPORT_SYMBOL(pci_choose_state);
 
+static int pci_save_pcie_state(struct pci_dev *dev)
+{
+	int pos, i = 0;
+	struct pci_cap_saved_state *save_state;
+	u16 *cap;
+
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (pos <= 0)
+		return 0;
+
+	save_state = kzalloc(sizeof(*save_state) + sizeof(u16) * 4, GFP_KERNEL);
+	if (!save_state) {
+		dev_err(&dev->dev, "Out of memory in pci_save_pcie_state\n");
+		return -ENOMEM;
+	}
+	cap = (u16 *)&save_state->data[0];
+
+	pci_read_config_word(dev, pos + PCI_EXP_DEVCTL, &cap[i++]);
+	pci_read_config_word(dev, pos + PCI_EXP_LNKCTL, &cap[i++]);
+	pci_read_config_word(dev, pos + PCI_EXP_SLTCTL, &cap[i++]);
+	pci_read_config_word(dev, pos + PCI_EXP_RTCTL, &cap[i++]);
+	pci_add_saved_cap(dev, save_state);
+	return 0;
+}
+
+static void pci_restore_pcie_state(struct pci_dev *dev)
+{
+	int i = 0, pos;
+	struct pci_cap_saved_state *save_state;
+	u16 *cap;
+
+	save_state = pci_find_saved_cap(dev, PCI_CAP_ID_EXP);
+	pos = pci_find_capability(dev, PCI_CAP_ID_EXP);
+	if (!save_state || pos <= 0)
+		return;
+	cap = (u16 *)&save_state->data[0];
+
+	pci_write_config_word(dev, pos + PCI_EXP_DEVCTL, cap[i++]);
+	pci_write_config_word(dev, pos + PCI_EXP_LNKCTL, cap[i++]);
+	pci_write_config_word(dev, pos + PCI_EXP_SLTCTL, cap[i++]);
+	pci_write_config_word(dev, pos + PCI_EXP_RTCTL, cap[i++]);
+	pci_remove_saved_cap(save_state);
+	kfree(save_state);
+}
+
 /**
  * pci_save_state - save the PCI configuration space of a device before suspending
  * @dev: - PCI device that we're dealing with
@@ -460,6 +505,8 @@ pci_save_state(struct pci_dev *dev)
 		return i;
 	if ((i = pci_save_msix_state(dev)) != 0)
 		return i;
+	if ((i = pci_save_pcie_state(dev)) != 0)
+		return i;
 	return 0;
 }
 
@@ -473,6 +520,9 @@ pci_restore_state(struct pci_dev *dev)
 	int i;
 	int val;
 
+	/* PCI Express register must be restored first */
+	pci_restore_pcie_state(dev);
+
 	/*
 	 * The Base Address register should be programmed before the command
 	 * register(s)
-- 
cgit v1.2.1


From 753388c2e91c15c12550bb20dda05ce657cfdc3e Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Wed, 6 Sep 2006 16:47:28 +0900
Subject: PCI Hotplug: cleanup pcihp skeleton code.

Cleanup pcihp skeleton code.

Fix some typos and remove some unnecessary blank lines.

Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/pcihp_skeleton.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/hotplug/pcihp_skeleton.c b/drivers/pci/hotplug/pcihp_skeleton.c
index 8ad446605f75..2b9e10e38613 100644
--- a/drivers/pci/hotplug/pcihp_skeleton.c
+++ b/drivers/pci/hotplug/pcihp_skeleton.c
@@ -1,5 +1,5 @@
 /*
- * PCI Hot Plug Controller Skeleton Driver - 0.2
+ * PCI Hot Plug Controller Skeleton Driver - 0.3
  *
  * Copyright (C) 2001,2003 Greg Kroah-Hartman (greg@kroah.com)
  * Copyright (C) 2001,2003 IBM Corp.
@@ -21,7 +21,7 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  *
- * This driver is to be used as a skeleton driver to be show how to interface
+ * This driver is to be used as a skeleton driver to show how to interface
  * with the pci hotplug core easily.
  *
  * Send feedback to <greg@kroah.com>
@@ -58,8 +58,6 @@ static LIST_HEAD(slot_list);
 #define info(format, arg...) printk(KERN_INFO "%s: " format "\n", MY_NAME , ## arg)
 #define warn(format, arg...) printk(KERN_WARNING "%s: " format "\n", MY_NAME , ## arg)
 
-
-
 /* local variables */
 static int debug;
 static int num_slots;
@@ -109,7 +107,6 @@ static int enable_slot(struct hotplug_slot *hotplug_slot)
 	return retval;
 }
 
-
 static int disable_slot(struct hotplug_slot *hotplug_slot)
 {
 	struct slot *slot = hotplug_slot->private;
@@ -342,7 +339,7 @@ static int __init pcihp_skel_init(void)
 	info(DRIVER_DESC " version: " DRIVER_VERSION "\n");
 	/*
 	 * Do specific initialization stuff for your driver here
-	 * Like initializing your controller hardware (if any) and
+	 * like initializing your controller hardware (if any) and
 	 * determining the number of slots you have in the system
 	 * right now.
 	 */
-- 
cgit v1.2.1


From 287af2fbe902206fabd42ade4e94f77db900083e Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:12:16 -0700
Subject: acpiphp: set hpp values before starting devices

Currently acpiphp sets hpp values after starting devices, but
the values should be set before starting devices. This patch
fixes this bug.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index be7e91662417..768d0f0f450a 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -1074,9 +1074,9 @@ static int enable_device(struct acpiphp_slot *slot)
 
 	pci_bus_assign_resources(bus);
 	acpiphp_sanitize_bus(bus);
+	acpiphp_set_hpp_values(slot->bridge->handle, bus);
 	pci_enable_bridges(bus);
 	pci_bus_add_devices(bus);
-	acpiphp_set_hpp_values(slot->bridge->handle, bus);
 	acpiphp_configure_ioapics(slot->bridge->handle);
 
 	/* associate pci_dev to our representation */
-- 
cgit v1.2.1


From b99feebe597f7b8c566048e11dbbd2d6df9abc83 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:13:44 -0700
Subject: acpiphp: initialize ioapics before starting devices

Currently acpiphp initializes ioapics after starting devices,
but ioapics should be initialized before starting devices.
This patch fixes this bug.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 768d0f0f450a..7cc782fec70a 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -1075,9 +1075,9 @@ static int enable_device(struct acpiphp_slot *slot)
 	pci_bus_assign_resources(bus);
 	acpiphp_sanitize_bus(bus);
 	acpiphp_set_hpp_values(slot->bridge->handle, bus);
+	acpiphp_configure_ioapics(slot->bridge->handle);
 	pci_enable_bridges(bus);
 	pci_bus_add_devices(bus);
-	acpiphp_configure_ioapics(slot->bridge->handle);
 
 	/* associate pci_dev to our representation */
 	list_for_each (l, &slot->funcs) {
-- 
cgit v1.2.1


From 9b1d19ee86746618a8b43d2aaef8319c01af1514 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:15:10 -0700
Subject: acpiphp: do not initialize existing ioapics

Currently acpiphp initializes all ioapics under the bus on which
hot-add event occured. It also initializes already working ioapics.
This patch fixes this bug.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 7cc782fec70a..c1001ad81ad4 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -841,6 +841,7 @@ ioapic_add(acpi_handle handle, u32 lvl, void *context, void **rv)
 
 static int acpiphp_configure_ioapics(acpi_handle handle)
 {
+	ioapic_add(handle, 0, NULL, NULL);
 	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle,
 			    ACPI_UINT32_MAX, ioapic_add, NULL, NULL);
 	return 0;
@@ -1075,7 +1076,8 @@ static int enable_device(struct acpiphp_slot *slot)
 	pci_bus_assign_resources(bus);
 	acpiphp_sanitize_bus(bus);
 	acpiphp_set_hpp_values(slot->bridge->handle, bus);
-	acpiphp_configure_ioapics(slot->bridge->handle);
+	list_for_each_entry(func, &slot->funcs, sibling)
+		acpiphp_configure_ioapics(func->handle);
 	pci_enable_bridges(bus);
 	pci_bus_add_devices(bus);
 
-- 
cgit v1.2.1


From 24f8aa9b464b73e0553f092b747770940ee0ea54 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:16:36 -0700
Subject: PCI: add pci_stop_bus_device

This patch adds pci_stop_bus_device() which stops a PCI device (detach
the driver, remove from the global list and so on) and any children.
This is needed for ACPI based PCI-to-PCI bridge hot-remove, and it will
be also needed for ACPI based PCI root bridge hot-remove.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/remove.c | 37 ++++++++++++++++++++++++++++++++++++-
 include/linux/pci.h  |  1 +
 2 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/remove.c b/drivers/pci/remove.c
index 99ffbd478b29..430281b2e921 100644
--- a/drivers/pci/remove.c
+++ b/drivers/pci/remove.c
@@ -16,8 +16,11 @@ static void pci_free_resources(struct pci_dev *dev)
 	}
 }
 
-static void pci_destroy_dev(struct pci_dev *dev)
+static void pci_stop_dev(struct pci_dev *dev)
 {
+	if (!dev->global_list.next)
+		return;
+
 	if (!list_empty(&dev->global_list)) {
 		pci_proc_detach_device(dev);
 		pci_remove_sysfs_dev_files(dev);
@@ -27,6 +30,11 @@ static void pci_destroy_dev(struct pci_dev *dev)
 		dev->global_list.next = dev->global_list.prev = NULL;
 		up_write(&pci_bus_sem);
 	}
+}
+
+static void pci_destroy_dev(struct pci_dev *dev)
+{
+	pci_stop_dev(dev);
 
 	/* Remove the device from the device lists, and prevent any further
 	 * list accesses from this device */
@@ -119,5 +127,32 @@ void pci_remove_behind_bridge(struct pci_dev *dev)
 	}
 }
 
+static void pci_stop_bus_devices(struct pci_bus *bus)
+{
+	struct list_head *l, *n;
+
+	list_for_each_safe(l, n, &bus->devices) {
+		struct pci_dev *dev = pci_dev_b(l);
+		pci_stop_bus_device(dev);
+	}
+}
+
+/**
+ * pci_stop_bus_device - stop a PCI device and any children
+ * @dev: the device to stop
+ *
+ * Stop a PCI device (detach the driver, remove from the global list
+ * and so on). This also stop any subordinate buses and children in a
+ * depth-first manner.
+ */
+void pci_stop_bus_device(struct pci_dev *dev)
+{
+	if (dev->subordinate)
+		pci_stop_bus_devices(dev->subordinate);
+
+	pci_stop_dev(dev);
+}
+
 EXPORT_SYMBOL(pci_remove_bus_device);
 EXPORT_SYMBOL(pci_remove_behind_bridge);
+EXPORT_SYMBOL_GPL(pci_stop_bus_device);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 549d8410974b..5c3a4176eb64 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -441,6 +441,7 @@ extern struct pci_dev *pci_dev_get(struct pci_dev *dev);
 extern void pci_dev_put(struct pci_dev *dev);
 extern void pci_remove_bus(struct pci_bus *b);
 extern void pci_remove_bus_device(struct pci_dev *dev);
+extern void pci_stop_bus_device(struct pci_dev *dev);
 void pci_setup_cardbus(struct pci_bus *bus);
 
 /* Generic PCI functions exported to card drivers */
-- 
cgit v1.2.1


From 0dad3510ee82bcf8a380b81a2184a664a911ef9c Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:17:46 -0700
Subject: acpiphp: stop bus device before acpi_bus_trim

Contrary to PCI bridge hot-add, we need to follow the sequence below
for PCI bridge hot-removal.

  (1) Stop devices (detach drivers, remove from the global list, etc.)
  (2) Unbind ACPI node from the devices (remove the _PRT entries)
  (3) Remove devices (remove from the device list, etc.)

This patch fixes acpiphp driver to follow above sequence for P2P
bridge hot-removal.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index c1001ad81ad4..d36732cd4bad 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -1129,6 +1129,9 @@ static int disable_device(struct acpiphp_slot *slot)
 			func->bridge = NULL;
 		}
 
+		if (func->pci_dev)
+			pci_stop_bus_device(func->pci_dev);
+
 		acpiphp_bus_trim(func->handle);
 		/* try to remove anyway.
 		 * acpiphp_bus_add might have been failed */
-- 
cgit v1.2.1


From d5cdb67236dba94496de052c9f9f431e1fc658f4 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:19:00 -0700
Subject: acpiphp: disable bridges

Currently acpiphp calls pci_enable_device() against all
hot-added bridges, but acpiphp does not call pci_disable_device()
against them in hot-remove. So ioapic hot-remove would fail.
This patch fixes this issue.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp_glue.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index d36732cd4bad..712f02fb1cbb 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -1105,6 +1105,16 @@ static int enable_device(struct acpiphp_slot *slot)
 	return retval;
 }
 
+static void disable_bridges(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+		if (dev->subordinate) {
+			disable_bridges(dev->subordinate);
+			pci_disable_device(dev);
+		}
+	}
+}
 
 /**
  * disable_device - disable a slot
@@ -1129,8 +1139,13 @@ static int disable_device(struct acpiphp_slot *slot)
 			func->bridge = NULL;
 		}
 
-		if (func->pci_dev)
+		if (func->pci_dev) {
 			pci_stop_bus_device(func->pci_dev);
+			if (func->pci_dev->subordinate) {
+				disable_bridges(func->pci_dev->subordinate);
+				pci_disable_device(func->pci_dev);
+			}
+		}
 
 		acpiphp_bus_trim(func->handle);
 		/* try to remove anyway.
-- 
cgit v1.2.1


From 23186279658cea6d42a050400d3e79c56cb459b4 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:21:44 -0700
Subject: PCI: assign ioapic resource at hotplug

We need to assign resources to ioapics being hot-added. This patch
changes pbus_assign_resources_sorted() to assign resources if the
ioapic has no assigned resources.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/setup-bus.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/setup-bus.c b/drivers/pci/setup-bus.c
index 47c1071ad84e..54404917be9a 100644
--- a/drivers/pci/setup-bus.c
+++ b/drivers/pci/setup-bus.c
@@ -55,12 +55,19 @@ pbus_assign_resources_sorted(struct pci_bus *bus)
 	list_for_each_entry(dev, &bus->devices, bus_list) {
 		u16 class = dev->class >> 8;
 
-		/* Don't touch classless devices or host bridges or ioapics.  */
+		/* Don't touch classless devices or host bridges. */
 		if (class == PCI_CLASS_NOT_DEFINED ||
-		    class == PCI_CLASS_BRIDGE_HOST ||
-		    class == PCI_CLASS_SYSTEM_PIC)
+		    class == PCI_CLASS_BRIDGE_HOST)
 			continue;
 
+		/* Don't touch ioapics if it has the assigned resources. */
+		if (class == PCI_CLASS_SYSTEM_PIC) {
+			res = &dev->resource[0];
+			if (res[0].start || res[1].start || res[2].start ||
+			    res[3].start || res[4].start || res[5].start)
+				continue;
+		}
+
 		pdev_sort_resources(dev, &head);
 	}
 
-- 
cgit v1.2.1


From 600812ecead0da2e7b6f9e5f5aad68b1ad8ae581 Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:22:53 -0700
Subject: acpiphp: add support for ioapic hot-remove

This patch adds support for ioapics hot-remove.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/acpiphp.h      |   5 ++
 drivers/pci/hotplug/acpiphp_glue.c | 101 ++++++++++++++++++++++++++++++++-----
 2 files changed, 92 insertions(+), 14 deletions(-)

diff --git a/drivers/pci/hotplug/acpiphp.h b/drivers/pci/hotplug/acpiphp.h
index be104eced34c..7fff07e877c7 100644
--- a/drivers/pci/hotplug/acpiphp.h
+++ b/drivers/pci/hotplug/acpiphp.h
@@ -150,6 +150,11 @@ struct acpiphp_attention_info
 	struct module *owner;
 };
 
+struct acpiphp_ioapic {
+	struct pci_dev *dev;
+	u32 gsi_base;
+	struct list_head list;
+};
 
 /* PCI bus bridge HID */
 #define ACPI_PCI_HOST_HID		"PNP0A03"
diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c
index 712f02fb1cbb..83e8e4412de5 100644
--- a/drivers/pci/hotplug/acpiphp_glue.c
+++ b/drivers/pci/hotplug/acpiphp_glue.c
@@ -53,6 +53,8 @@
 #include "acpiphp.h"
 
 static LIST_HEAD(bridge_list);
+static LIST_HEAD(ioapic_list);
+static DEFINE_SPINLOCK(ioapic_list_lock);
 
 #define MY_NAME "acpiphp_glue"
 
@@ -797,6 +799,7 @@ ioapic_add(acpi_handle handle, u32 lvl, void *context, void **rv)
 	struct pci_dev *pdev;
 	u32 gsi_base;
 	u64 phys_addr;
+	struct acpiphp_ioapic *ioapic;
 
 	/* Evaluate _STA if present */
 	status = acpi_evaluate_integer(handle, "_STA", NULL, &sta);
@@ -811,30 +814,87 @@ ioapic_add(acpi_handle handle, u32 lvl, void *context, void **rv)
 	if (get_gsi_base(handle, &gsi_base))
 		return AE_OK;
 
+	ioapic = kmalloc(sizeof(*ioapic), GFP_KERNEL);
+	if (!ioapic)
+		return AE_NO_MEMORY;
+
 	pdev = get_apic_pci_info(handle);
 	if (!pdev)
-		return AE_OK;
+		goto exit_kfree;
 
-	if (pci_enable_device(pdev)) {
-		pci_dev_put(pdev);
-		return AE_OK;
-	}
+	if (pci_enable_device(pdev))
+		goto exit_pci_dev_put;
 
 	pci_set_master(pdev);
 
-	if (pci_request_region(pdev, 0, "I/O APIC(acpiphp)")) {
-		pci_disable_device(pdev);
-		pci_dev_put(pdev);
-		return AE_OK;
-	}
+	if (pci_request_region(pdev, 0, "I/O APIC(acpiphp)"))
+		goto exit_pci_disable_device;
 
 	phys_addr = pci_resource_start(pdev, 0);
-	if (acpi_register_ioapic(handle, phys_addr, gsi_base)) {
-		pci_release_region(pdev, 0);
-		pci_disable_device(pdev);
-		pci_dev_put(pdev);
+	if (acpi_register_ioapic(handle, phys_addr, gsi_base))
+		goto exit_pci_release_region;
+
+	ioapic->gsi_base = gsi_base;
+	ioapic->dev = pdev;
+	spin_lock(&ioapic_list_lock);
+	list_add_tail(&ioapic->list, &ioapic_list);
+	spin_unlock(&ioapic_list_lock);
+
+	return AE_OK;
+
+ exit_pci_release_region:
+	pci_release_region(pdev, 0);
+ exit_pci_disable_device:
+	pci_disable_device(pdev);
+ exit_pci_dev_put:
+	pci_dev_put(pdev);
+ exit_kfree:
+	kfree(ioapic);
+
+	return AE_OK;
+}
+
+static acpi_status
+ioapic_remove(acpi_handle handle, u32 lvl, void *context, void **rv)
+{
+	acpi_status status;
+	unsigned long sta;
+	acpi_handle tmp;
+	u32 gsi_base;
+	struct acpiphp_ioapic *pos, *n, *ioapic = NULL;
+
+	/* Evaluate _STA if present */
+	status = acpi_evaluate_integer(handle, "_STA", NULL, &sta);
+	if (ACPI_SUCCESS(status) && sta != ACPI_STA_ALL)
+		return AE_CTRL_DEPTH;
+
+	/* Scan only PCI bus scope */
+	status = acpi_get_handle(handle, "_HID", &tmp);
+	if (ACPI_SUCCESS(status))
+		return AE_CTRL_DEPTH;
+
+	if (get_gsi_base(handle, &gsi_base))
 		return AE_OK;
+
+	acpi_unregister_ioapic(handle, gsi_base);
+
+	spin_lock(&ioapic_list_lock);
+	list_for_each_entry_safe(pos, n, &ioapic_list, list) {
+		if (pos->gsi_base != gsi_base)
+			continue;
+		ioapic = pos;
+		list_del(&ioapic->list);
+		break;
 	}
+	spin_unlock(&ioapic_list_lock);
+
+	if (!ioapic)
+		return AE_OK;
+
+	pci_release_region(ioapic->dev, 0);
+	pci_disable_device(ioapic->dev);
+	pci_dev_put(ioapic->dev);
+	kfree(ioapic);
 
 	return AE_OK;
 }
@@ -847,6 +907,14 @@ static int acpiphp_configure_ioapics(acpi_handle handle)
 	return 0;
 }
 
+static int acpiphp_unconfigure_ioapics(acpi_handle handle)
+{
+	ioapic_remove(handle, 0, NULL, NULL);
+	acpi_walk_namespace(ACPI_TYPE_DEVICE, handle,
+			    ACPI_UINT32_MAX, ioapic_remove, NULL, NULL);
+	return 0;
+}
+
 static int power_on_slot(struct acpiphp_slot *slot)
 {
 	acpi_status status;
@@ -1146,7 +1214,12 @@ static int disable_device(struct acpiphp_slot *slot)
 				pci_disable_device(func->pci_dev);
 			}
 		}
+	}
+
+	list_for_each (l, &slot->funcs) {
+		func = list_entry(l, struct acpiphp_func, sibling);
 
+		acpiphp_unconfigure_ioapics(func->handle);
 		acpiphp_bus_trim(func->handle);
 		/* try to remove anyway.
 		 * acpiphp_bus_add might have been failed */
-- 
cgit v1.2.1


From aa4f63cad4f70a52adbabb66ac6c8b6072910fdf Mon Sep 17 00:00:00 2001
From: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Date: Tue, 12 Sep 2006 10:24:14 -0700
Subject: IA64: PCI: dont disable irq which is not enabled

This patch prevents pcibios_disable_device() from disabling interrupts
of devices which is not enabled.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: MUNEDA Takahiro <muneda.takahiro@jp.fujitsu.com>
Signed-off-by: Satoru Takeuchi <takeuchi_satoru@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/ia64/pci/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/ia64/pci/pci.c b/arch/ia64/pci/pci.c
index 60b45e79f080..15c7c670da39 100644
--- a/arch/ia64/pci/pci.c
+++ b/arch/ia64/pci/pci.c
@@ -562,7 +562,8 @@ pcibios_enable_device (struct pci_dev *dev, int mask)
 void
 pcibios_disable_device (struct pci_dev *dev)
 {
-	acpi_pci_irq_disable(dev);
+	if (dev->is_enabled)
+		acpi_pci_irq_disable(dev);
 }
 
 void
-- 
cgit v1.2.1


From c9d86d76c1cdd76d67292ab75643db66573ca7dd Mon Sep 17 00:00:00 2001
From: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Date: Tue, 19 Sep 2006 17:04:33 -0700
Subject: pciehp - fix wrong return value

This patch fixes the problem that trying to enable already enabled
slot disables the slot by returning the proper value from
pciehp_enable_slot()/pciehp_disable_slot().

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/hotplug/pciehp_ctrl.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c
index 33d198768356..41290a106bd8 100644
--- a/drivers/pci/hotplug/pciehp_ctrl.c
+++ b/drivers/pci/hotplug/pciehp_ctrl.c
@@ -762,14 +762,14 @@ int pciehp_enable_slot(struct slot *p_slot)
 	if (rc || !getstatus) {
 		info("%s: no adapter on slot(%x)\n", __FUNCTION__, p_slot->number);
 		mutex_unlock(&p_slot->ctrl->crit_sect);
-		return 1;
+		return -ENODEV;
 	}
 	if (MRL_SENS(p_slot->ctrl->ctrlcap)) {	
 		rc = p_slot->hpc_ops->get_latch_status(p_slot, &getstatus);
 		if (rc || getstatus) {
 			info("%s: latch open on slot(%x)\n", __FUNCTION__, p_slot->number);
 			mutex_unlock(&p_slot->ctrl->crit_sect);
-			return 1;
+			return -ENODEV;
 		}
 	}
 	
@@ -778,7 +778,7 @@ int pciehp_enable_slot(struct slot *p_slot)
 		if (rc || getstatus) {
 			info("%s: already enabled on slot(%x)\n", __FUNCTION__, p_slot->number);
 			mutex_unlock(&p_slot->ctrl->crit_sect);
-			return 1;
+			return -EINVAL;
 		}
 	}
 	mutex_unlock(&p_slot->ctrl->crit_sect);
@@ -813,7 +813,7 @@ int pciehp_disable_slot(struct slot *p_slot)
 		if (ret || !getstatus) {
 			info("%s: no adapter on slot(%x)\n", __FUNCTION__, p_slot->number);
 			mutex_unlock(&p_slot->ctrl->crit_sect);
-			return 1;
+			return -ENODEV;
 		}
 	}
 
@@ -822,7 +822,7 @@ int pciehp_disable_slot(struct slot *p_slot)
 		if (ret || getstatus) {
 			info("%s: latch open on slot(%x)\n", __FUNCTION__, p_slot->number);
 			mutex_unlock(&p_slot->ctrl->crit_sect);
-			return 1;
+			return -ENODEV;
 		}
 	}
 
@@ -831,7 +831,7 @@ int pciehp_disable_slot(struct slot *p_slot)
 		if (ret || !getstatus) {
 			info("%s: already disabled slot(%x)\n", __FUNCTION__, p_slot->number);
 			mutex_unlock(&p_slot->ctrl->crit_sect);
-			return 1;
+			return -EINVAL;
 		}
 	}
 
-- 
cgit v1.2.1


From 0c91c1a7012911e4b0180a1c1ae258e2b706f987 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:16:20 +0900
Subject: sh: Move smc37c93x.h for SystemH board use.

SystemH needs this header as well, not just 770x SE.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/se/770x/setup.c |   2 +-
 include/asm-sh/se/smc37c93x.h  | 190 -----------------------------------------
 include/asm-sh/smc37c93x.h     | 190 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 191 insertions(+), 191 deletions(-)
 delete mode 100644 include/asm-sh/se/smc37c93x.h
 create mode 100644 include/asm-sh/smc37c93x.h

diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index 7d1a071727cc..b6730ea57502 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -15,7 +15,7 @@
 #include <linux/ide.h>
 #include <asm/io.h>
 #include <asm/se/se.h>
-#include <asm/se/smc37c93x.h>
+#include <asm/smc37c93x.h>
 
 /*
  * Configure the Super I/O chip
diff --git a/include/asm-sh/se/smc37c93x.h b/include/asm-sh/se/smc37c93x.h
deleted file mode 100644
index 585da2a8fc45..000000000000
--- a/include/asm-sh/se/smc37c93x.h
+++ /dev/null
@@ -1,190 +0,0 @@
-#ifndef __ASM_SH_SMC37C93X_H
-#define __ASM_SH_SMC37C93X_H
-
-/*
- * linux/include/asm-sh/smc37c93x.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * SMSC 37C93x Super IO Chip support
- */
-
-/* Default base I/O address */
-#define FDC_PRIMARY_BASE	0x3f0
-#define IDE1_PRIMARY_BASE	0x1f0
-#define IDE1_SECONDARY_BASE	0x170
-#define PARPORT_PRIMARY_BASE	0x378
-#define COM1_PRIMARY_BASE	0x2f8
-#define COM2_PRIMARY_BASE	0x3f8
-#define RTC_PRIMARY_BASE	0x070
-#define KBC_PRIMARY_BASE	0x060
-#define AUXIO_PRIMARY_BASE	0x000	/* XXX */
-
-/* Logical device number */
-#define LDN_FDC			0
-#define LDN_IDE1		1
-#define LDN_IDE2		2
-#define LDN_PARPORT		3
-#define LDN_COM1		4
-#define LDN_COM2		5
-#define LDN_RTC			6
-#define LDN_KBC			7
-#define LDN_AUXIO		8
-
-/* Configuration port and key */
-#define CONFIG_PORT		0x3f0
-#define INDEX_PORT		CONFIG_PORT
-#define DATA_PORT		0x3f1
-#define CONFIG_ENTER		0x55
-#define CONFIG_EXIT		0xaa
-
-/* Configuration index */
-#define CURRENT_LDN_INDEX	0x07
-#define POWER_CONTROL_INDEX	0x22
-#define ACTIVATE_INDEX		0x30
-#define IO_BASE_HI_INDEX	0x60
-#define IO_BASE_LO_INDEX	0x61
-#define IRQ_SELECT_INDEX	0x70
-#define DMA_SELECT_INDEX	0x74
-
-#define GPIO46_INDEX		0xc6
-#define GPIO47_INDEX		0xc7
-
-/* UART stuff. Only for debugging.  */
-/* UART Register */
-
-#define UART_RBR	0x0	/* Receiver Buffer Register (Read Only) */
-#define UART_THR	0x0	/* Transmitter Holding Register (Write Only) */
-#define UART_IER	0x2	/* Interrupt Enable Register */
-#define UART_IIR	0x4	/* Interrupt Ident Register (Read Only) */
-#define UART_FCR	0x4	/* FIFO Control Register (Write Only) */
-#define UART_LCR	0x6	/* Line Control Register */
-#define UART_MCR	0x8	/* MODEM Control Register */
-#define UART_LSR	0xa	/* Line Status Register */
-#define UART_MSR	0xc	/* MODEM Status Register */
-#define UART_SCR	0xe	/* Scratch Register */
-#define UART_DLL	0x0	/* Divisor Latch (LS) */
-#define UART_DLM	0x2	/* Divisor Latch (MS) */
-
-#ifndef __ASSEMBLY__
-typedef struct uart_reg {
-	volatile __u16 rbr;
-	volatile __u16 ier;
-	volatile __u16 iir;
-	volatile __u16 lcr;
-	volatile __u16 mcr;
-	volatile __u16 lsr;
-	volatile __u16 msr;
-	volatile __u16 scr;
-} uart_reg;
-#endif /* ! __ASSEMBLY__ */
-
-/* Alias for Write Only Register */
-
-#define thr	rbr
-#define tcr	iir
-
-/* Alias for Divisor Latch Register */
-
-#define dll	rbr
-#define dlm	ier
-#define fcr	iir
-
-/* Interrupt Enable Register */
-
-#define IER_ERDAI	0x0100	/* Enable Received Data Available Interrupt */
-#define IER_ETHREI	0x0200	/* Enable Transmitter Holding Register Empty Interrupt */
-#define IER_ELSI	0x0400	/* Enable Receiver Line Status Interrupt */
-#define IER_EMSI	0x0800	/* Enable MODEM Status Interrupt */
-
-/* Interrupt Ident Register */
-
-#define IIR_IP		0x0100	/* "0" if Interrupt Pending */
-#define IIR_IIB0	0x0200	/* Interrupt ID Bit 0 */
-#define IIR_IIB1	0x0400	/* Interrupt ID Bit 1 */
-#define IIR_IIB2	0x0800	/* Interrupt ID Bit 2 */
-#define IIR_FIFO	0xc000	/* FIFOs enabled */
-
-/* FIFO Control Register */
-
-#define FCR_FEN		0x0100	/* FIFO enable */
-#define FCR_RFRES	0x0200	/* Receiver FIFO reset */
-#define FCR_TFRES	0x0400	/* Transmitter FIFO reset */
-#define FCR_DMA		0x0800	/* DMA mode select */
-#define FCR_RTL		0x4000	/* Receiver triger (LSB) */
-#define FCR_RTM		0x8000	/* Receiver triger (MSB) */
-
-/* Line Control Register */
-
-#define LCR_WLS0	0x0100	/* Word Length Select Bit 0 */
-#define LCR_WLS1	0x0200	/* Word Length Select Bit 1 */
-#define LCR_STB		0x0400	/* Number of Stop Bits */
-#define LCR_PEN		0x0800	/* Parity Enable */
-#define LCR_EPS		0x1000	/* Even Parity Select */
-#define LCR_SP		0x2000	/* Stick Parity */
-#define LCR_SB		0x4000	/* Set Break */
-#define LCR_DLAB	0x8000	/* Divisor Latch Access Bit */
-
-/* MODEM Control Register */
-
-#define MCR_DTR		0x0100	/* Data Terminal Ready */
-#define MCR_RTS		0x0200	/* Request to Send */
-#define MCR_OUT1	0x0400	/* Out 1 */
-#define MCR_IRQEN	0x0800	/* IRQ Enable */
-#define MCR_LOOP	0x1000	/* Loop */
-
-/* Line Status Register */
-
-#define LSR_DR		0x0100	/* Data Ready */
-#define LSR_OE		0x0200	/* Overrun Error */
-#define LSR_PE		0x0400	/* Parity Error */
-#define LSR_FE		0x0800	/* Framing Error */
-#define LSR_BI		0x1000	/* Break Interrupt */
-#define LSR_THRE	0x2000	/* Transmitter Holding Register Empty */
-#define LSR_TEMT	0x4000	/* Transmitter Empty */
-#define LSR_FIFOE	0x8000	/* Receiver FIFO error */
-
-/* MODEM Status Register */
-
-#define MSR_DCTS	0x0100	/* Delta Clear to Send */
-#define MSR_DDSR	0x0200	/* Delta Data Set Ready */
-#define MSR_TERI	0x0400	/* Trailing Edge Ring Indicator */
-#define MSR_DDCD	0x0800	/* Delta Data Carrier Detect */
-#define MSR_CTS		0x1000	/* Clear to Send */
-#define MSR_DSR		0x2000	/* Data Set Ready */
-#define MSR_RI		0x4000	/* Ring Indicator */
-#define MSR_DCD		0x8000	/* Data Carrier Detect */
-
-/* Baud Rate Divisor */
-
-#define UART_CLK	(1843200)	/* 1.8432 MHz */
-#define UART_BAUD(x)	(UART_CLK / (16 * (x)))
-
-/* RTC register definition */
-#define RTC_SECONDS             0
-#define RTC_SECONDS_ALARM       1
-#define RTC_MINUTES             2
-#define RTC_MINUTES_ALARM       3
-#define RTC_HOURS               4
-#define RTC_HOURS_ALARM         5
-#define RTC_DAY_OF_WEEK         6
-#define RTC_DAY_OF_MONTH        7
-#define RTC_MONTH               8
-#define RTC_YEAR                9
-#define RTC_FREQ_SELECT		10
-# define RTC_UIP 0x80
-# define RTC_DIV_CTL 0x70
-/* This RTC can work under 32.768KHz clock only.  */
-# define RTC_OSC_ENABLE 0x20
-# define RTC_OSC_DISABLE 0x00
-#define RTC_CONTROL     	11
-# define RTC_SET 0x80
-# define RTC_PIE 0x40
-# define RTC_AIE 0x20
-# define RTC_UIE 0x10
-# define RTC_SQWE 0x08
-# define RTC_DM_BINARY 0x04
-# define RTC_24H 0x02
-# define RTC_DST_EN 0x01
-
-#endif  /* __ASM_SH_SMC37C93X_H */
diff --git a/include/asm-sh/smc37c93x.h b/include/asm-sh/smc37c93x.h
new file mode 100644
index 000000000000..585da2a8fc45
--- /dev/null
+++ b/include/asm-sh/smc37c93x.h
@@ -0,0 +1,190 @@
+#ifndef __ASM_SH_SMC37C93X_H
+#define __ASM_SH_SMC37C93X_H
+
+/*
+ * linux/include/asm-sh/smc37c93x.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * SMSC 37C93x Super IO Chip support
+ */
+
+/* Default base I/O address */
+#define FDC_PRIMARY_BASE	0x3f0
+#define IDE1_PRIMARY_BASE	0x1f0
+#define IDE1_SECONDARY_BASE	0x170
+#define PARPORT_PRIMARY_BASE	0x378
+#define COM1_PRIMARY_BASE	0x2f8
+#define COM2_PRIMARY_BASE	0x3f8
+#define RTC_PRIMARY_BASE	0x070
+#define KBC_PRIMARY_BASE	0x060
+#define AUXIO_PRIMARY_BASE	0x000	/* XXX */
+
+/* Logical device number */
+#define LDN_FDC			0
+#define LDN_IDE1		1
+#define LDN_IDE2		2
+#define LDN_PARPORT		3
+#define LDN_COM1		4
+#define LDN_COM2		5
+#define LDN_RTC			6
+#define LDN_KBC			7
+#define LDN_AUXIO		8
+
+/* Configuration port and key */
+#define CONFIG_PORT		0x3f0
+#define INDEX_PORT		CONFIG_PORT
+#define DATA_PORT		0x3f1
+#define CONFIG_ENTER		0x55
+#define CONFIG_EXIT		0xaa
+
+/* Configuration index */
+#define CURRENT_LDN_INDEX	0x07
+#define POWER_CONTROL_INDEX	0x22
+#define ACTIVATE_INDEX		0x30
+#define IO_BASE_HI_INDEX	0x60
+#define IO_BASE_LO_INDEX	0x61
+#define IRQ_SELECT_INDEX	0x70
+#define DMA_SELECT_INDEX	0x74
+
+#define GPIO46_INDEX		0xc6
+#define GPIO47_INDEX		0xc7
+
+/* UART stuff. Only for debugging.  */
+/* UART Register */
+
+#define UART_RBR	0x0	/* Receiver Buffer Register (Read Only) */
+#define UART_THR	0x0	/* Transmitter Holding Register (Write Only) */
+#define UART_IER	0x2	/* Interrupt Enable Register */
+#define UART_IIR	0x4	/* Interrupt Ident Register (Read Only) */
+#define UART_FCR	0x4	/* FIFO Control Register (Write Only) */
+#define UART_LCR	0x6	/* Line Control Register */
+#define UART_MCR	0x8	/* MODEM Control Register */
+#define UART_LSR	0xa	/* Line Status Register */
+#define UART_MSR	0xc	/* MODEM Status Register */
+#define UART_SCR	0xe	/* Scratch Register */
+#define UART_DLL	0x0	/* Divisor Latch (LS) */
+#define UART_DLM	0x2	/* Divisor Latch (MS) */
+
+#ifndef __ASSEMBLY__
+typedef struct uart_reg {
+	volatile __u16 rbr;
+	volatile __u16 ier;
+	volatile __u16 iir;
+	volatile __u16 lcr;
+	volatile __u16 mcr;
+	volatile __u16 lsr;
+	volatile __u16 msr;
+	volatile __u16 scr;
+} uart_reg;
+#endif /* ! __ASSEMBLY__ */
+
+/* Alias for Write Only Register */
+
+#define thr	rbr
+#define tcr	iir
+
+/* Alias for Divisor Latch Register */
+
+#define dll	rbr
+#define dlm	ier
+#define fcr	iir
+
+/* Interrupt Enable Register */
+
+#define IER_ERDAI	0x0100	/* Enable Received Data Available Interrupt */
+#define IER_ETHREI	0x0200	/* Enable Transmitter Holding Register Empty Interrupt */
+#define IER_ELSI	0x0400	/* Enable Receiver Line Status Interrupt */
+#define IER_EMSI	0x0800	/* Enable MODEM Status Interrupt */
+
+/* Interrupt Ident Register */
+
+#define IIR_IP		0x0100	/* "0" if Interrupt Pending */
+#define IIR_IIB0	0x0200	/* Interrupt ID Bit 0 */
+#define IIR_IIB1	0x0400	/* Interrupt ID Bit 1 */
+#define IIR_IIB2	0x0800	/* Interrupt ID Bit 2 */
+#define IIR_FIFO	0xc000	/* FIFOs enabled */
+
+/* FIFO Control Register */
+
+#define FCR_FEN		0x0100	/* FIFO enable */
+#define FCR_RFRES	0x0200	/* Receiver FIFO reset */
+#define FCR_TFRES	0x0400	/* Transmitter FIFO reset */
+#define FCR_DMA		0x0800	/* DMA mode select */
+#define FCR_RTL		0x4000	/* Receiver triger (LSB) */
+#define FCR_RTM		0x8000	/* Receiver triger (MSB) */
+
+/* Line Control Register */
+
+#define LCR_WLS0	0x0100	/* Word Length Select Bit 0 */
+#define LCR_WLS1	0x0200	/* Word Length Select Bit 1 */
+#define LCR_STB		0x0400	/* Number of Stop Bits */
+#define LCR_PEN		0x0800	/* Parity Enable */
+#define LCR_EPS		0x1000	/* Even Parity Select */
+#define LCR_SP		0x2000	/* Stick Parity */
+#define LCR_SB		0x4000	/* Set Break */
+#define LCR_DLAB	0x8000	/* Divisor Latch Access Bit */
+
+/* MODEM Control Register */
+
+#define MCR_DTR		0x0100	/* Data Terminal Ready */
+#define MCR_RTS		0x0200	/* Request to Send */
+#define MCR_OUT1	0x0400	/* Out 1 */
+#define MCR_IRQEN	0x0800	/* IRQ Enable */
+#define MCR_LOOP	0x1000	/* Loop */
+
+/* Line Status Register */
+
+#define LSR_DR		0x0100	/* Data Ready */
+#define LSR_OE		0x0200	/* Overrun Error */
+#define LSR_PE		0x0400	/* Parity Error */
+#define LSR_FE		0x0800	/* Framing Error */
+#define LSR_BI		0x1000	/* Break Interrupt */
+#define LSR_THRE	0x2000	/* Transmitter Holding Register Empty */
+#define LSR_TEMT	0x4000	/* Transmitter Empty */
+#define LSR_FIFOE	0x8000	/* Receiver FIFO error */
+
+/* MODEM Status Register */
+
+#define MSR_DCTS	0x0100	/* Delta Clear to Send */
+#define MSR_DDSR	0x0200	/* Delta Data Set Ready */
+#define MSR_TERI	0x0400	/* Trailing Edge Ring Indicator */
+#define MSR_DDCD	0x0800	/* Delta Data Carrier Detect */
+#define MSR_CTS		0x1000	/* Clear to Send */
+#define MSR_DSR		0x2000	/* Data Set Ready */
+#define MSR_RI		0x4000	/* Ring Indicator */
+#define MSR_DCD		0x8000	/* Data Carrier Detect */
+
+/* Baud Rate Divisor */
+
+#define UART_CLK	(1843200)	/* 1.8432 MHz */
+#define UART_BAUD(x)	(UART_CLK / (16 * (x)))
+
+/* RTC register definition */
+#define RTC_SECONDS             0
+#define RTC_SECONDS_ALARM       1
+#define RTC_MINUTES             2
+#define RTC_MINUTES_ALARM       3
+#define RTC_HOURS               4
+#define RTC_HOURS_ALARM         5
+#define RTC_DAY_OF_WEEK         6
+#define RTC_DAY_OF_MONTH        7
+#define RTC_MONTH               8
+#define RTC_YEAR                9
+#define RTC_FREQ_SELECT		10
+# define RTC_UIP 0x80
+# define RTC_DIV_CTL 0x70
+/* This RTC can work under 32.768KHz clock only.  */
+# define RTC_OSC_ENABLE 0x20
+# define RTC_OSC_DISABLE 0x00
+#define RTC_CONTROL     	11
+# define RTC_SET 0x80
+# define RTC_PIE 0x40
+# define RTC_AIE 0x20
+# define RTC_UIE 0x10
+# define RTC_SQWE 0x08
+# define RTC_DM_BINARY 0x04
+# define RTC_24H 0x02
+# define RTC_DST_EN 0x01
+
+#endif  /* __ASM_SH_SMC37C93X_H */
-- 
cgit v1.2.1


From e4e3b5ccd77226c9c4dbb0737106b868dfc182d9 Mon Sep 17 00:00:00 2001
From: Tom Rini <trini@kernel.crashing.org>
Date: Wed, 27 Sep 2006 11:28:20 +0900
Subject: sh: Add a simple cmpxchg().

We didn't have one of these before, a simple implementation
borrowed from MIPS as well as the __HAVE_ARCH_CMPXCHG bits.

Signed-off-by: Tom Rini <trini@kernel.crashing.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 40 ++++++++++++++++++++++++++++++++++++++++
 1 file changed, 40 insertions(+)

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index ad35ad4958f4..e89728d405d8 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -6,6 +6,7 @@
  * Copyright (C) 2002 Paul Mundt
  */
 
+#include <asm/types.h>
 
 /*
  *	switch_to() should switch tasks to task nr n, first
@@ -260,6 +261,45 @@ static __inline__ unsigned long __xchg(unsigned long x, volatile void * ptr, int
 	return x;
 }
 
+static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
+	unsigned long new)
+{
+	__u32 retval;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	retval = *m;
+	if (retval == old)
+		*m = new;
+	local_irq_restore(flags);       /* implies memory barrier  */
+	return retval;
+}
+
+/* This function doesn't exist, so you'll get a linker error
+ * if something tries to do an invalid cmpxchg(). */
+extern void __cmpxchg_called_with_bad_pointer(void);
+
+#define __HAVE_ARCH_CMPXCHG 1
+
+static inline unsigned long __cmpxchg(volatile void * ptr, unsigned long old,
+		unsigned long new, int size)
+{
+	switch (size) {
+	case 4:
+		return __cmpxchg_u32(ptr, old, new);
+	}
+	__cmpxchg_called_with_bad_pointer();
+	return old;
+}
+
+#define cmpxchg(ptr,o,n)						 \
+  ({									 \
+     __typeof__(*(ptr)) _o_ = (o);					 \
+     __typeof__(*(ptr)) _n_ = (n);					 \
+     (__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_,		 \
+				    (unsigned long)_n_, sizeof(*(ptr))); \
+  })
+
 /* XXX
  * disable hlt during certain critical i/o operations
  */
-- 
cgit v1.2.1


From a252710fc5b63b24934905ca47ecf661702d7f00 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:29:55 +0900
Subject: sh: flush_cache_range() cleanup and optimizations.

flush_cache_range() wasn't page aligning the end of the range,
we can't assume that it will always be page aligned, and we
ended up getting unaligned faults in some rare call paths.

Additionally, we add a small optimization to just purge the
dcache entirely if the range is large enough that the page
table walking will take longer. We use an arbitrary value of
64 pages for the large range size, as per sh64.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c | 72 ++++++++++++++++++++++++++++++++------------------
 1 file changed, 46 insertions(+), 26 deletions(-)

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 524cea5b47f9..94c05d09c3f7 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -2,7 +2,7 @@
  * arch/sh/mm/cache-sh4.c
  *
  * Copyright (C) 1999, 2000, 2002  Niibe Yutaka
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001, 2002, 2003, 2004, 2005  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -25,6 +25,8 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
+extern void __flush_cache_4096(unsigned long addr, unsigned long phys,
+			       unsigned long exec_offset);
 extern void __flush_cache_4096_all(unsigned long start);
 static void __flush_cache_4096_all_ex(unsigned long start);
 extern void __flush_dcache_all(void);
@@ -112,9 +114,14 @@ static void __flush_dcache_all_ex(void)
 {
 	unsigned long addr, end_addr, entry_offset;
 
-	end_addr = CACHE_OC_ADDRESS_ARRAY + (cpu_data->dcache.sets << cpu_data->dcache.entry_shift) * cpu_data->dcache.ways;
+	end_addr = CACHE_OC_ADDRESS_ARRAY +
+		(cpu_data->dcache.sets << cpu_data->dcache.entry_shift) *
+		 cpu_data->dcache.ways;
+
 	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (addr = CACHE_OC_ADDRESS_ARRAY; addr < end_addr; addr += entry_offset) {
+	for (addr = CACHE_OC_ADDRESS_ARRAY;
+	     addr < end_addr;
+	     addr += entry_offset) {
 		ctrl_outl(0, addr);
 	}
 }
@@ -125,7 +132,8 @@ static void __flush_cache_4096_all_ex(unsigned long start)
 	int i;
 
 	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (i = 0; i < cpu_data->dcache.ways; i++, start += cpu_data->dcache.way_incr) {
+	for (i = 0; i < cpu_data->dcache.ways;
+	     i++, start += cpu_data->dcache.way_incr) {
 		for (addr = CACHE_OC_ADDRESS_ARRAY + start;
 		     addr < CACHE_OC_ADDRESS_ARRAY + 4096 + start;
 		     addr += entry_offset) {
@@ -153,14 +161,14 @@ void flush_icache_range(unsigned long start, unsigned long end)
 }
 
 /*
- * Write back the D-cache and purge the I-cache for signal trampoline. 
+ * Write back the D-cache and purge the I-cache for signal trampoline.
  * .. which happens to be the same behavior as flush_icache_range().
  * So, we simply flush out a line.
  */
 void flush_cache_sigtramp(unsigned long addr)
 {
 	unsigned long v, index;
-	unsigned long flags; 
+	unsigned long flags;
 	int i;
 
 	v = addr & ~(L1_CACHE_BYTES-1);
@@ -172,7 +180,8 @@ void flush_cache_sigtramp(unsigned long addr)
 
 	local_irq_save(flags);
 	jump_to_P2();
-	for(i = 0; i < cpu_data->icache.ways; i++, index += cpu_data->icache.way_incr)
+	for (i = 0; i < cpu_data->icache.ways;
+	     i++, index += cpu_data->icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
 	back_to_P1();
 	local_irq_restore(flags);
@@ -181,8 +190,7 @@ void flush_cache_sigtramp(unsigned long addr)
 static inline void flush_cache_4096(unsigned long start,
 				    unsigned long phys)
 {
-	unsigned long flags; 
-	extern void __flush_cache_4096(unsigned long addr, unsigned long phys, unsigned long exec_offset);
+	unsigned long flags;
 
 	/*
 	 * SH7751, SH7751R, and ST40 have no restriction to handle cache.
@@ -191,10 +199,12 @@ static inline void flush_cache_4096(unsigned long start,
 	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG)
 	   || start < CACHE_OC_ADDRESS_ARRAY) {
 		local_irq_save(flags);
-		__flush_cache_4096(start | SH_CACHE_ASSOC, P1SEGADDR(phys), 0x20000000);
+		__flush_cache_4096(start | SH_CACHE_ASSOC,
+				   P1SEGADDR(phys), 0x20000000);
 		local_irq_restore(flags);
 	} else {
-		__flush_cache_4096(start | SH_CACHE_ASSOC, P1SEGADDR(phys), 0);
+		__flush_cache_4096(start | SH_CACHE_ASSOC,
+				   P1SEGADDR(phys), 0);
 	}
 }
 
@@ -231,29 +241,22 @@ static inline void flush_icache_all(void)
 	local_irq_restore(flags);
 }
 
-void flush_cache_all(void)
+void flush_dcache_all(void)
 {
 	if (cpu_data->dcache.ways == 1)
 		__flush_dcache_all();
 	else
 		__flush_dcache_all_ex();
+}
+
+void flush_cache_all(void)
+{
+	flush_dcache_all();
 	flush_icache_all();
 }
 
 void flush_cache_mm(struct mm_struct *mm)
 {
-	/* Is there any good way? */
-	/* XXX: possibly call flush_cache_range for each vm area */
-	/* 
-	 * FIXME: Really, the optimal solution here would be able to flush out
-	 * individual lines created by the specified context, but this isn't
-	 * feasible for a number of architectures (such as MIPS, and some
-	 * SPARC) .. is this possible for SuperH?
-	 *
-	 * In the meantime, we'll just flush all of the caches.. this
-	 * seems to be the simplest way to avoid at least a few wasted
-	 * cache flushes. -Lethal
-	 */
 	flush_cache_all();
 }
 
@@ -301,13 +304,30 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 	unsigned long p = start & PAGE_MASK;
 	pgd_t *dir;
 	pmd_t *pmd;
+	pud_t *pud;
 	pte_t *pte;
 	pte_t entry;
 	unsigned long phys;
 	unsigned long d = 0;
 
+	/*
+	 * Don't bother with the lookup and alias check if we have a
+	 * wide range to cover, just blow away the dcache in its
+	 * entirety instead. -- PFM.
+	 */
+	if (((end - start) >> PAGE_SHIFT) >= 64) {
+		flush_dcache_all();
+
+		if (vma->vm_flags & VM_EXEC)
+			flush_icache_all();
+
+		return;
+	}
+
 	dir = pgd_offset(vma->vm_mm, p);
-	pmd = pmd_offset(dir, p);
+	pud = pud_offset(dir, p);
+	pmd = pmd_offset(pud, p);
+	end = PAGE_ALIGN(end);
 
 	do {
 		if (pmd_none(*pmd) || pmd_bad(*pmd)) {
@@ -322,7 +342,7 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 			if ((pte_val(entry) & _PAGE_PRESENT)) {
 				phys = pte_val(entry)&PTE_PHYS_MASK;
 				if ((p^phys) & CACHE_ALIAS) {
-					d |= 1 << ((p & CACHE_ALIAS)>>12); 
+					d |= 1 << ((p & CACHE_ALIAS)>>12);
 					d |= 1 << ((phys & CACHE_ALIAS)>>12);
 					if (d == 0x0f)
 						goto loop_exit;
-- 
cgit v1.2.1


From 765ae317ce030217af556dc46bec238dab15091e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:31:32 +0900
Subject: sh: Fixup some uninitialized spinlocks.

Fix use of uninitialized spinlocks, caught with spinlock debugging..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/semaphore.c | 2 +-
 arch/sh/kernel/traps.c     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sh/kernel/semaphore.c b/arch/sh/kernel/semaphore.c
index a3c24dcbf01d..184119eeae56 100644
--- a/arch/sh/kernel/semaphore.c
+++ b/arch/sh/kernel/semaphore.c
@@ -14,7 +14,7 @@
 #include <asm/semaphore.h>
 #include <asm/semaphore-helper.h>
 
-spinlock_t semaphore_wake_lock;
+DEFINE_SPINLOCK(semaphore_wake_lock);
 
 /*
  * Semaphores are implemented using a two-way counter:
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index d9db1180f770..21bef1b5f991 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -86,7 +86,7 @@ asmlinkage void do_##name(unsigned long r4, unsigned long r5,		\
 #define VMALLOC_OFFSET (8*1024*1024)
 #define MODULE_RANGE (8*1024*1024)
 
-spinlock_t die_lock;
+DEFINE_SPINLOCK(die_lock);
 
 void die(const char * str, struct pt_regs * regs, long err)
 {
-- 
cgit v1.2.1


From 1c5f8f85df96dd7c17d63eed2d7da18b3ad00b6d Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:36:10 +0900
Subject: sh: Move syscall table in to syscall.S.

Move the syscall table in to its own file, as per sh64. The entry.S
bits will end up being considerably different in the sh2/sh2a cases,
so this lets us keep things in sync somewhat..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile   |   2 +-
 arch/sh/kernel/entry.S    | 297 ------------------------------------------
 arch/sh/kernel/syscalls.S | 324 ++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 325 insertions(+), 298 deletions(-)
 create mode 100644 arch/sh/kernel/syscalls.S

diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index f05cd96f8867..2de82b66af4a 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -6,7 +6,7 @@ extra-y	:= head.o init_task.o vmlinux.lds
 
 obj-y	:= process.o signal.o entry.o traps.o irq.o \
 	ptrace.o setup.o time.o sys_sh.o semaphore.o \
-	io.o io_generic.o sh_ksyms.o
+	io.o io_generic.o sh_ksyms.o syscalls.o
 
 obj-y				+= cpu/ timers/
 
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 7dfd2ba75f7f..a3a717adbf43 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -838,300 +838,3 @@ ENTRY(exception_none)
 	rts
 	 nop
 
-	.data
-ENTRY(sys_call_table)
-	.long sys_ni_syscall	/* 0  -  old "setup()" system call*/
-	.long sys_exit
-	.long sys_fork
-	.long sys_read
-	.long sys_write
-	.long sys_open		/* 5 */
-	.long sys_close
-	.long sys_waitpid
-	.long sys_creat
-	.long sys_link
-	.long sys_unlink		/* 10 */
-	.long sys_execve
-	.long sys_chdir
-	.long sys_time
-	.long sys_mknod
-	.long sys_chmod		/* 15 */
-	.long sys_lchown16
-	.long sys_ni_syscall	/* old break syscall holder */
-	.long sys_stat
-	.long sys_lseek
-	.long sys_getpid		/* 20 */
-	.long sys_mount
-	.long sys_oldumount
-	.long sys_setuid16
-	.long sys_getuid16
-	.long sys_stime		/* 25 */
-	.long sys_ptrace
-	.long sys_alarm
-	.long sys_fstat
-	.long sys_pause
-	.long sys_utime		/* 30 */
-	.long sys_ni_syscall	/* old stty syscall holder */
-	.long sys_ni_syscall	/* old gtty syscall holder */
-	.long sys_access
-	.long sys_nice
-	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
-	.long sys_sync
-	.long sys_kill
-	.long sys_rename
-	.long sys_mkdir
-	.long sys_rmdir		/* 40 */
-	.long sys_dup
-	.long sys_pipe
-	.long sys_times
-	.long sys_ni_syscall	/* old prof syscall holder */
-	.long sys_brk		/* 45 */
-	.long sys_setgid16
-	.long sys_getgid16
-	.long sys_signal
-	.long sys_geteuid16
-	.long sys_getegid16	/* 50 */
-	.long sys_acct
-	.long sys_umount		/* recycled never used phys() */
-	.long sys_ni_syscall	/* old lock syscall holder */
-	.long sys_ioctl
-	.long sys_fcntl		/* 55 */
-	.long sys_ni_syscall	/* old mpx syscall holder */
-	.long sys_setpgid
-	.long sys_ni_syscall	/* old ulimit syscall holder */
-	.long sys_ni_syscall	/* sys_olduname */
-	.long sys_umask		/* 60 */
-	.long sys_chroot
-	.long sys_ustat
-	.long sys_dup2
-	.long sys_getppid
-	.long sys_getpgrp		/* 65 */
-	.long sys_setsid
-	.long sys_sigaction
-	.long sys_sgetmask
-	.long sys_ssetmask
-	.long sys_setreuid16	/* 70 */
-	.long sys_setregid16
-	.long sys_sigsuspend
-	.long sys_sigpending
-	.long sys_sethostname
-	.long sys_setrlimit	/* 75 */
-	.long sys_old_getrlimit
-	.long sys_getrusage
-	.long sys_gettimeofday
-	.long sys_settimeofday
-	.long sys_getgroups16	/* 80 */
-	.long sys_setgroups16
-	.long sys_ni_syscall	/* sys_oldselect */
-	.long sys_symlink
-	.long sys_lstat
-	.long sys_readlink		/* 85 */
-	.long sys_uselib
-	.long sys_swapon
-	.long sys_reboot
-	.long old_readdir
-	.long old_mmap		/* 90 */
-	.long sys_munmap
-	.long sys_truncate
-	.long sys_ftruncate
-	.long sys_fchmod
-	.long sys_fchown16		/* 95 */
-	.long sys_getpriority
-	.long sys_setpriority
-	.long sys_ni_syscall	/* old profil syscall holder */
-	.long sys_statfs
-	.long sys_fstatfs		/* 100 */
-	.long sys_ni_syscall	/* ioperm */
-	.long sys_socketcall
-	.long sys_syslog
-	.long sys_setitimer
-	.long sys_getitimer	/* 105 */
-	.long sys_newstat
-	.long sys_newlstat
-	.long sys_newfstat
-	.long sys_uname
-	.long sys_ni_syscall	/* 110 */ /* iopl */
-	.long sys_vhangup
-	.long sys_ni_syscall	/* idle */
-	.long sys_ni_syscall	/* vm86old */
-	.long sys_wait4
-	.long sys_swapoff		/* 115 */
-	.long sys_sysinfo
-	.long sys_ipc
-	.long sys_fsync
-	.long sys_sigreturn
-	.long sys_clone		/* 120 */
-	.long sys_setdomainname
-	.long sys_newuname
-	.long sys_ni_syscall	/* sys_modify_ldt */
-	.long sys_adjtimex
-	.long sys_mprotect		/* 125 */
-	.long sys_sigprocmask
-	.long sys_ni_syscall	/* old "create_module" */
-	.long sys_init_module
-	.long sys_delete_module
-	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
-	.long sys_quotactl
-	.long sys_getpgid
-	.long sys_fchdir
-	.long sys_bdflush
-	.long sys_sysfs		/* 135 */
-	.long sys_personality
-	.long sys_ni_syscall	/* for afs_syscall */
-	.long sys_setfsuid16
-	.long sys_setfsgid16
-	.long sys_llseek		/* 140 */
-	.long sys_getdents
-	.long sys_select
-	.long sys_flock
-	.long sys_msync
-	.long sys_readv		/* 145 */
-	.long sys_writev
-	.long sys_getsid
-	.long sys_fdatasync
-	.long sys_sysctl
-	.long sys_mlock		/* 150 */
-	.long sys_munlock
-	.long sys_mlockall
-	.long sys_munlockall
-	.long sys_sched_setparam
-	.long sys_sched_getparam   /* 155 */
-	.long sys_sched_setscheduler
-	.long sys_sched_getscheduler
-	.long sys_sched_yield
-	.long sys_sched_get_priority_max
-	.long sys_sched_get_priority_min  /* 160 */
-	.long sys_sched_rr_get_interval
-	.long sys_nanosleep
-	.long sys_mremap
-	.long sys_setresuid16
-	.long sys_getresuid16	/* 165 */
-	.long sys_ni_syscall	/* vm86 */
-	.long sys_ni_syscall	/* old "query_module" */
-	.long sys_poll
-	.long sys_nfsservctl
-	.long sys_setresgid16	/* 170 */
-	.long sys_getresgid16
-	.long sys_prctl
-	.long sys_rt_sigreturn
-	.long sys_rt_sigaction
-	.long sys_rt_sigprocmask	/* 175 */
-	.long sys_rt_sigpending
-	.long sys_rt_sigtimedwait
-	.long sys_rt_sigqueueinfo
-	.long sys_rt_sigsuspend
-	.long sys_pread_wrapper	   /* 180 */
-	.long sys_pwrite_wrapper
-	.long sys_chown16
-	.long sys_getcwd
-	.long sys_capget
-	.long sys_capset           /* 185 */
-	.long sys_sigaltstack
-	.long sys_sendfile
-	.long sys_ni_syscall	/* streams1 */
-	.long sys_ni_syscall	/* streams2 */
-	.long sys_vfork            /* 190 */
-	.long sys_getrlimit
-	.long sys_mmap2
-	.long sys_truncate64
-	.long sys_ftruncate64
-	.long sys_stat64		/* 195 */
-	.long sys_lstat64
-	.long sys_fstat64
-	.long sys_lchown
-	.long sys_getuid
-	.long sys_getgid		/* 200 */
-	.long sys_geteuid
-	.long sys_getegid
-	.long sys_setreuid
-	.long sys_setregid
-	.long sys_getgroups	/* 205 */
-	.long sys_setgroups
-	.long sys_fchown
-	.long sys_setresuid
-	.long sys_getresuid
-	.long sys_setresgid	/* 210 */
-	.long sys_getresgid
-	.long sys_chown
-	.long sys_setuid
-	.long sys_setgid
-	.long sys_setfsuid		/* 215 */
-	.long sys_setfsgid
-	.long sys_pivot_root
-	.long sys_mincore
-	.long sys_madvise
-	.long sys_getdents64	/* 220 */
-	.long sys_fcntl64
-	.long sys_ni_syscall	/* reserved for TUX */
-	.long sys_ni_syscall	/* Reserved for Security */
-	.long sys_gettid
-	.long sys_readahead	/* 225 */
-	.long sys_setxattr
-	.long sys_lsetxattr
-	.long sys_fsetxattr
-	.long sys_getxattr
-	.long sys_lgetxattr	/* 230 */
-	.long sys_fgetxattr
-	.long sys_listxattr
-	.long sys_llistxattr
-	.long sys_flistxattr
-	.long sys_removexattr	/* 235 */
-	.long sys_lremovexattr
-	.long sys_fremovexattr
-	.long sys_tkill
-	.long sys_sendfile64
-	.long sys_futex		/* 240 */
-	.long sys_sched_setaffinity
-	.long sys_sched_getaffinity
-	.long sys_ni_syscall
-	.long sys_ni_syscall
-	.long sys_io_setup	/* 245 */
-	.long sys_io_destroy
-	.long sys_io_getevents
-	.long sys_io_submit
-	.long sys_io_cancel
-	.long sys_fadvise64	/* 250 */
-	.long sys_ni_syscall
-	.long sys_exit_group
-	.long sys_lookup_dcookie
-	.long sys_epoll_create
-	.long sys_epoll_ctl	/* 255 */
-	.long sys_epoll_wait
- 	.long sys_remap_file_pages
- 	.long sys_set_tid_address
- 	.long sys_timer_create
- 	.long sys_timer_settime		/* 260 */
- 	.long sys_timer_gettime
- 	.long sys_timer_getoverrun
- 	.long sys_timer_delete
- 	.long sys_clock_settime
- 	.long sys_clock_gettime		/* 265 */
- 	.long sys_clock_getres
- 	.long sys_clock_nanosleep
-	.long sys_statfs64
-	.long sys_fstatfs64     
-	.long sys_tgkill		/* 270 */
-	.long sys_utimes
- 	.long sys_fadvise64_64_wrapper
-	.long sys_ni_syscall	/* Reserved for vserver */
-	.long sys_ni_syscall	/* Reserved for mbind */
-	.long sys_ni_syscall	/* 275 - get_mempolicy */
-	.long sys_ni_syscall	/* set_mempolicy */
-	.long sys_mq_open
-	.long sys_mq_unlink
-	.long sys_mq_timedsend
-	.long sys_mq_timedreceive       /* 280 */
-	.long sys_mq_notify
-	.long sys_mq_getsetattr
-	.long sys_ni_syscall	/* Reserved for kexec */
-	.long sys_waitid
-	.long sys_add_key		/* 285 */
-	.long sys_request_key
-	.long sys_keyctl
-	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init		/* 290 */
-	.long sys_inotify_add_watch
-	.long sys_inotify_rm_watch
-
-/* End of entry.S */
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
new file mode 100644
index 000000000000..f500304cbd69
--- /dev/null
+++ b/arch/sh/kernel/syscalls.S
@@ -0,0 +1,324 @@
+/*
+ * arch/sh/kernel/syscalls.S
+ *
+ * System call table for SuperH
+ *
+ *  Copyright (C) 1999, 2000, 2002  Niibe Yutaka
+ *  Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+#include <linux/sys.h>
+#include <linux/linkage.h>
+
+#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
+#define sys_nfsservctl		sys_ni_syscall
+#endif
+
+#if !defined(CONFIG_MMU)
+#define sys_madvise		sys_ni_syscall
+#define sys_readahead		sys_ni_syscall
+#define sys_mprotect		sys_ni_syscall
+#define sys_msync		sys_ni_syscall
+#define sys_mlock		sys_ni_syscall
+#define sys_munlock		sys_ni_syscall
+#define sys_mlockall		sys_ni_syscall
+#define sys_munlockall		sys_ni_syscall
+#define sys_mremap		sys_ni_syscall
+#define sys_mincore		sys_ni_syscall
+#define sys_remap_file_pages	sys_ni_syscall
+#endif
+
+	.data
+ENTRY(sys_call_table)
+	.long sys_ni_syscall	/* 0  -  old "setup()" system call*/
+	.long sys_exit
+	.long sys_fork
+	.long sys_read
+	.long sys_write
+	.long sys_open		/* 5 */
+	.long sys_close
+	.long sys_waitpid
+	.long sys_creat
+	.long sys_link
+	.long sys_unlink		/* 10 */
+	.long sys_execve
+	.long sys_chdir
+	.long sys_time
+	.long sys_mknod
+	.long sys_chmod		/* 15 */
+	.long sys_lchown16
+	.long sys_ni_syscall	/* old break syscall holder */
+	.long sys_stat
+	.long sys_lseek
+	.long sys_getpid		/* 20 */
+	.long sys_mount
+	.long sys_oldumount
+	.long sys_setuid16
+	.long sys_getuid16
+	.long sys_stime		/* 25 */
+	.long sys_ptrace
+	.long sys_alarm
+	.long sys_fstat
+	.long sys_pause
+	.long sys_utime		/* 30 */
+	.long sys_ni_syscall	/* old stty syscall holder */
+	.long sys_ni_syscall	/* old gtty syscall holder */
+	.long sys_access
+	.long sys_nice
+	.long sys_ni_syscall	/* 35 */		/* old ftime syscall holder */
+	.long sys_sync
+	.long sys_kill
+	.long sys_rename
+	.long sys_mkdir
+	.long sys_rmdir		/* 40 */
+	.long sys_dup
+	.long sys_pipe
+	.long sys_times
+	.long sys_ni_syscall	/* old prof syscall holder */
+	.long sys_brk		/* 45 */
+	.long sys_setgid16
+	.long sys_getgid16
+	.long sys_signal
+	.long sys_geteuid16
+	.long sys_getegid16	/* 50 */
+	.long sys_acct
+	.long sys_umount		/* recycled never used phys() */
+	.long sys_ni_syscall	/* old lock syscall holder */
+	.long sys_ioctl
+	.long sys_fcntl		/* 55 */
+	.long sys_ni_syscall	/* old mpx syscall holder */
+	.long sys_setpgid
+	.long sys_ni_syscall	/* old ulimit syscall holder */
+	.long sys_ni_syscall	/* sys_olduname */
+	.long sys_umask		/* 60 */
+	.long sys_chroot
+	.long sys_ustat
+	.long sys_dup2
+	.long sys_getppid
+	.long sys_getpgrp		/* 65 */
+	.long sys_setsid
+	.long sys_sigaction
+	.long sys_sgetmask
+	.long sys_ssetmask
+	.long sys_setreuid16	/* 70 */
+	.long sys_setregid16
+	.long sys_sigsuspend
+	.long sys_sigpending
+	.long sys_sethostname
+	.long sys_setrlimit	/* 75 */
+	.long sys_old_getrlimit
+	.long sys_getrusage
+	.long sys_gettimeofday
+	.long sys_settimeofday
+	.long sys_getgroups16	/* 80 */
+	.long sys_setgroups16
+	.long sys_ni_syscall	/* sys_oldselect */
+	.long sys_symlink
+	.long sys_lstat
+	.long sys_readlink		/* 85 */
+	.long sys_uselib
+	.long sys_swapon
+	.long sys_reboot
+	.long old_readdir
+	.long old_mmap		/* 90 */
+	.long sys_munmap
+	.long sys_truncate
+	.long sys_ftruncate
+	.long sys_fchmod
+	.long sys_fchown16		/* 95 */
+	.long sys_getpriority
+	.long sys_setpriority
+	.long sys_ni_syscall	/* old profil syscall holder */
+	.long sys_statfs
+	.long sys_fstatfs		/* 100 */
+	.long sys_ni_syscall	/* ioperm */
+	.long sys_socketcall
+	.long sys_syslog
+	.long sys_setitimer
+	.long sys_getitimer	/* 105 */
+	.long sys_newstat
+	.long sys_newlstat
+	.long sys_newfstat
+	.long sys_uname
+	.long sys_ni_syscall	/* 110 */ /* iopl */
+	.long sys_vhangup
+	.long sys_ni_syscall	/* idle */
+	.long sys_ni_syscall	/* vm86old */
+	.long sys_wait4
+	.long sys_swapoff		/* 115 */
+	.long sys_sysinfo
+	.long sys_ipc
+	.long sys_fsync
+	.long sys_sigreturn
+	.long sys_clone		/* 120 */
+	.long sys_setdomainname
+	.long sys_newuname
+	.long sys_ni_syscall	/* sys_modify_ldt */
+	.long sys_adjtimex
+	.long sys_mprotect		/* 125 */
+	.long sys_sigprocmask
+	.long sys_ni_syscall	/* old "create_module" */
+	.long sys_init_module
+	.long sys_delete_module
+	.long sys_ni_syscall	/* 130: old "get_kernel_syms" */
+	.long sys_quotactl
+	.long sys_getpgid
+	.long sys_fchdir
+	.long sys_bdflush
+	.long sys_sysfs		/* 135 */
+	.long sys_personality
+	.long sys_ni_syscall	/* for afs_syscall */
+	.long sys_setfsuid16
+	.long sys_setfsgid16
+	.long sys_llseek		/* 140 */
+	.long sys_getdents
+	.long sys_select
+	.long sys_flock
+	.long sys_msync
+	.long sys_readv		/* 145 */
+	.long sys_writev
+	.long sys_getsid
+	.long sys_fdatasync
+	.long sys_sysctl
+	.long sys_mlock		/* 150 */
+	.long sys_munlock
+	.long sys_mlockall
+	.long sys_munlockall
+	.long sys_sched_setparam
+	.long sys_sched_getparam   /* 155 */
+	.long sys_sched_setscheduler
+	.long sys_sched_getscheduler
+	.long sys_sched_yield
+	.long sys_sched_get_priority_max
+	.long sys_sched_get_priority_min  /* 160 */
+	.long sys_sched_rr_get_interval
+	.long sys_nanosleep
+	.long sys_mremap
+	.long sys_setresuid16
+	.long sys_getresuid16	/* 165 */
+	.long sys_ni_syscall	/* vm86 */
+	.long sys_ni_syscall	/* old "query_module" */
+	.long sys_poll
+	.long sys_nfsservctl
+	.long sys_setresgid16	/* 170 */
+	.long sys_getresgid16
+	.long sys_prctl
+	.long sys_rt_sigreturn
+	.long sys_rt_sigaction
+	.long sys_rt_sigprocmask	/* 175 */
+	.long sys_rt_sigpending
+	.long sys_rt_sigtimedwait
+	.long sys_rt_sigqueueinfo
+	.long sys_rt_sigsuspend
+	.long sys_pread_wrapper	   /* 180 */
+	.long sys_pwrite_wrapper
+	.long sys_chown16
+	.long sys_getcwd
+	.long sys_capget
+	.long sys_capset           /* 185 */
+	.long sys_sigaltstack
+	.long sys_sendfile
+	.long sys_ni_syscall	/* streams1 */
+	.long sys_ni_syscall	/* streams2 */
+	.long sys_vfork            /* 190 */
+	.long sys_getrlimit
+	.long sys_mmap2
+	.long sys_truncate64
+	.long sys_ftruncate64
+	.long sys_stat64		/* 195 */
+	.long sys_lstat64
+	.long sys_fstat64
+	.long sys_lchown
+	.long sys_getuid
+	.long sys_getgid		/* 200 */
+	.long sys_geteuid
+	.long sys_getegid
+	.long sys_setreuid
+	.long sys_setregid
+	.long sys_getgroups	/* 205 */
+	.long sys_setgroups
+	.long sys_fchown
+	.long sys_setresuid
+	.long sys_getresuid
+	.long sys_setresgid	/* 210 */
+	.long sys_getresgid
+	.long sys_chown
+	.long sys_setuid
+	.long sys_setgid
+	.long sys_setfsuid		/* 215 */
+	.long sys_setfsgid
+	.long sys_pivot_root
+	.long sys_mincore
+	.long sys_madvise
+	.long sys_getdents64	/* 220 */
+	.long sys_fcntl64
+	.long sys_ni_syscall	/* reserved for TUX */
+	.long sys_ni_syscall	/* Reserved for Security */
+	.long sys_gettid
+	.long sys_readahead	/* 225 */
+	.long sys_setxattr
+	.long sys_lsetxattr
+	.long sys_fsetxattr
+	.long sys_getxattr
+	.long sys_lgetxattr	/* 230 */
+	.long sys_fgetxattr
+	.long sys_listxattr
+	.long sys_llistxattr
+	.long sys_flistxattr
+	.long sys_removexattr	/* 235 */
+	.long sys_lremovexattr
+	.long sys_fremovexattr
+	.long sys_tkill
+	.long sys_sendfile64
+	.long sys_futex		/* 240 */
+	.long sys_sched_setaffinity
+	.long sys_sched_getaffinity
+	.long sys_ni_syscall
+	.long sys_ni_syscall
+	.long sys_io_setup	/* 245 */
+	.long sys_io_destroy
+	.long sys_io_getevents
+	.long sys_io_submit
+	.long sys_io_cancel
+	.long sys_fadvise64	/* 250 */
+	.long sys_ni_syscall
+	.long sys_exit_group
+	.long sys_lookup_dcookie
+	.long sys_epoll_create
+	.long sys_epoll_ctl	/* 255 */
+	.long sys_epoll_wait
+ 	.long sys_remap_file_pages
+ 	.long sys_set_tid_address
+ 	.long sys_timer_create
+ 	.long sys_timer_settime		/* 260 */
+ 	.long sys_timer_gettime
+ 	.long sys_timer_getoverrun
+ 	.long sys_timer_delete
+ 	.long sys_clock_settime
+ 	.long sys_clock_gettime		/* 265 */
+ 	.long sys_clock_getres
+ 	.long sys_clock_nanosleep
+	.long sys_statfs64
+	.long sys_fstatfs64
+	.long sys_tgkill		/* 270 */
+	.long sys_utimes
+ 	.long sys_fadvise64_64_wrapper
+	.long sys_ni_syscall	/* Reserved for vserver */
+	.long sys_ni_syscall	/* Reserved for mbind */
+	.long sys_ni_syscall	/* 275 - get_mempolicy */
+	.long sys_ni_syscall	/* set_mempolicy */
+	.long sys_mq_open
+	.long sys_mq_unlink
+	.long sys_mq_timedsend
+	.long sys_mq_timedreceive       /* 280 */
+	.long sys_mq_notify
+	.long sys_mq_getsetattr
+	.long sys_kexec_load
+	.long sys_waitid
+	.long sys_add_key		/* 285 */
+	.long sys_request_key
+	.long sys_keyctl
-- 
cgit v1.2.1


From 6ae5e8d7598ce01ac0e7dab1b89894c7386a2e61 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:37:33 +0900
Subject: sh: Fix kGDB NMI handling.

in_nmi shifted down a few labels, so we were inadvertently
clearing the lower byte of do_syscall_trace, badness ensues.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/entry.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index a3a717adbf43..1a05804caf37 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -644,7 +644,7 @@ skip_restore:
 	!
 #if defined(CONFIG_KGDB_NMI)
 	! Clear in_nmi
-	mov.l	4f, k0
+	mov.l	6f, k0
 	mov	#0, k1
 	mov.b	k1, @k0
 #endif
-- 
cgit v1.2.1


From 50e98e72e459e43b6b9a5449e35bb6fc54e21149 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:40:05 +0900
Subject: sh: Kill off the .stack section.

We had a special .stack section in the ld script that
was being used to position r15 initially. This is
nonsensical, as we can just use a THREAD_SIZE offset
from the init_thread_union instead (as every other arch
does).

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/head.S        | 2 +-
 arch/sh/kernel/vmlinux.lds.S | 2 --
 2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 9b9e6ef626ce..00cd4708ef46 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -69,7 +69,7 @@ ENTRY(_stext)
 
 	.balign 4
 1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
-2:	.long	stack
+2:	.long	init_thread_union+8192
 3:	.long	__bss_start
 4:	.long	_end
 5:	.long	start_kernel
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 95fdd9135fcf..3f3e1e0735f5 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -62,8 +62,6 @@ SECTIONS
 
   . = ALIGN(8192);		/* init_task */
   .data.init_task : { *(.data.init_task) }
-  /* stack */
-  .stack : { stack = .;  _stack = .; }
 
   . = ALIGN(4096);		/* Init code and data */
   __init_begin = .;
-- 
cgit v1.2.1


From a56d276c05a80ce727902076a3b4c6247705e2df Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 11:43:24 +0900
Subject: sh: Make hs7751rvoip/rts7751r2d use pm_power_off.

These were previously sprinkled in machine_power_off(),
though missed being updated when the rest of the boards
switched over.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/hs7751rvoip/setup.c | 13 +++++++++----
 arch/sh/boards/renesas/rts7751r2d/setup.c  |  7 +++++++
 2 files changed, 16 insertions(+), 4 deletions(-)

diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index 29fb5ff70fb5..7ce577822dc0 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -12,15 +12,14 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
 #include <linux/hdreg.h>
 #include <linux/ide.h>
+#include <linux/pm.h>
 #include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
 
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
-
 /* defined in mm/ioremap.c */
 extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
 
@@ -31,6 +30,11 @@ const char *get_system_type(void)
 	return "HS7751RVoIP";
 }
 
+static void hs7751rvoip_power_off(void)
+{
+	ctrl_outw(ctrl_inw(PA_OUTPORTR) & 0xffdf, PA_OUTPORTR);
+}
+
 /*
  * Initialize the board
  */
@@ -38,6 +42,7 @@ void __init platform_setup(void)
 {
 	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
 	ctrl_outb(0xf0, PA_OUTPORTR);
+	pm_power_off = hs7751rvoip_power_off;
 	debug_counter = 0;
 }
 
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index 2587fd1a0240..7953dde0b91c 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -10,6 +10,7 @@
  */
 
 #include <linux/init.h>
+#include <linux/pm.h>
 #include <asm/io.h>
 #include <asm/rts7751r2d/rts7751r2d.h>
 
@@ -20,6 +21,11 @@ const char *get_system_type(void)
 	return "RTS7751R2D";
 }
 
+static void rts7751r2d_power_off(void)
+{
+	ctrl_outw(0x0001, PA_POWOFF);
+}
+
 /*
  * Initialize the board
  */
@@ -27,5 +33,6 @@ void __init platform_setup(void)
 {
 	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 	ctrl_outw(0x0000, PA_OUTPORT);
+	pm_power_off = rts7751r2d_power_off;
 	debug_counter = 0;
 }
-- 
cgit v1.2.1


From e4c2cfee5d5cf3e4c16b423be23551aeddf2717b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 12:31:01 +0900
Subject: sh: Various cosmetic cleanups.

We had quite a bit of whitespace damage, clean most of it up..

Signed-off-by: Stuart Menefy <stuart.menefy@st.com>
Signed-off-by: Arthur Othieno <a.othieno@bluewin.ch>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/dreamcast/irq.c              |  8 ++--
 arch/sh/boards/dreamcast/rtc.c              | 18 ++++-----
 arch/sh/boards/dreamcast/setup.c            |  6 +--
 arch/sh/boards/renesas/edosk7705/Makefile   |  4 --
 arch/sh/boards/renesas/hs7751rvoip/Makefile |  4 --
 arch/sh/boards/renesas/hs7751rvoip/io.c     |  4 +-
 arch/sh/boards/renesas/rts7751r2d/Makefile  |  4 --
 arch/sh/boards/renesas/rts7751r2d/led.c     |  2 -
 arch/sh/boards/renesas/rts7751r2d/setup.c   |  3 --
 arch/sh/drivers/dma/dma-pvr2.c              |  5 +--
 arch/sh/kernel/cpu/clock.c                  |  2 +-
 arch/sh/kernel/cpu/sh4/sq.c                 |  3 +-
 arch/sh/kernel/entry.S                      | 18 ---------
 arch/sh/kernel/ptrace.c                     |  1 -
 arch/sh/kernel/sh_ksyms.c                   |  2 +
 drivers/char/watchdog/shwdt.c               | 58 ++++++++++++-----------------
 include/asm-sh/atomic.h                     |  1 +
 include/asm-sh/bitops.h                     | 16 ++++----
 include/asm-sh/checksum.h                   |  2 +
 include/asm-sh/fixmap.h                     |  2 +-
 include/asm-sh/watchdog.h                   |  3 --
 21 files changed, 60 insertions(+), 106 deletions(-)

diff --git a/arch/sh/boards/dreamcast/irq.c b/arch/sh/boards/dreamcast/irq.c
index b10a6b11c034..373a22e48ac4 100644
--- a/arch/sh/boards/dreamcast/irq.c
+++ b/arch/sh/boards/dreamcast/irq.c
@@ -26,10 +26,10 @@
    event.
 
    There are three 32-bit ESRs located at 0xa05f8900 - 0xa05f6908.  Event
-   types can be found in include/asm-sh/dc_sysasic.h.  There are three groups
-   of EMRs that parallel the ESRs.  Each EMR group corresponds to an IRQ, so
-   0xa05f6910 - 0xa05f6918 triggers IRQ 13, 0xa05f6920 - 0xa05f6928 triggers
-   IRQ 11, and 0xa05f6930 - 0xa05f6938 triggers IRQ 9.
+   types can be found in include/asm-sh/dreamcast/sysasic.h. There are three
+   groups of EMRs that parallel the ESRs.  Each EMR group corresponds to an
+   IRQ, so 0xa05f6910 - 0xa05f6918 triggers IRQ 13, 0xa05f6920 - 0xa05f6928
+   triggers IRQ 11, and 0xa05f6930 - 0xa05f6938 triggers IRQ 9.
 
    In the kernel, these events are mapped to virtual IRQs so that drivers can
    respond to them as they would a normal interrupt.  In order to keep this
diff --git a/arch/sh/boards/dreamcast/rtc.c b/arch/sh/boards/dreamcast/rtc.c
index 379de1629134..44420494dc22 100644
--- a/arch/sh/boards/dreamcast/rtc.c
+++ b/arch/sh/boards/dreamcast/rtc.c
@@ -1,4 +1,5 @@
-/* arch/sh/kernel/rtc-aica.c
+/*
+ * arch/sh/boards/dreamcast/rtc.c
  *
  * Dreamcast AICA RTC routines.
  *
@@ -10,15 +11,12 @@
  */
 
 #include <linux/time.h>
-
+#include <asm/rtc.h>
 #include <asm/io.h>
 
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
-
 /* The AICA RTC has an Epoch of 1/1/1950, so we must subtract 20 years (in
-   seconds to get the standard Unix Epoch when getting the time, and add 20
-   years when setting the time. */
+   seconds) to get the standard Unix Epoch when getting the time, and add
+   20 years when setting the time. */
 #define TWENTY_YEARS ((20 * 365LU + 5) * 86400)
 
 /* The AICA RTC is represented by a 32-bit seconds counter stored in 2 16-bit
@@ -32,7 +30,8 @@ extern int (*rtc_set_time)(const time_t);
  *
  * Grabs the current RTC seconds counter and adjusts it to the Unix Epoch.
  */
-void aica_rtc_gettimeofday(struct timespec *ts) {
+void aica_rtc_gettimeofday(struct timespec *ts)
+{
 	unsigned long val1, val2;
 
 	do {
@@ -55,7 +54,8 @@ void aica_rtc_gettimeofday(struct timespec *ts) {
  *
  * Adjusts the given @tv to the AICA Epoch and sets the RTC seconds counter.
  */
-int aica_rtc_settimeofday(const time_t secs) {
+int aica_rtc_settimeofday(const time_t secs)
+{
 	unsigned long val1, val2;
 	unsigned long adj = secs + TWENTY_YEARS;
 
diff --git a/arch/sh/boards/dreamcast/setup.c b/arch/sh/boards/dreamcast/setup.c
index 0027b80a2343..a00ac94beb05 100644
--- a/arch/sh/boards/dreamcast/setup.c
+++ b/arch/sh/boards/dreamcast/setup.c
@@ -25,18 +25,16 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
+#include <asm/rtc.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
 #include <asm/mach/sysasic.h>
 
 extern struct hw_interrupt_type systemasic_int;
-/* XXX: Move this into it's proper header. */
-extern void (*board_time_init)(void);
 extern void aica_time_init(void);
 extern int gapspci_init(void);
 extern int systemasic_irq_demux(int);
 
-void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, int);
+void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
 int dreamcast_consistent_free(struct device *, size_t, void *, dma_addr_t);
 
 const char *get_system_type(void)
diff --git a/arch/sh/boards/renesas/edosk7705/Makefile b/arch/sh/boards/renesas/edosk7705/Makefile
index 7fccbf2e4a1d..14bdd531f116 100644
--- a/arch/sh/boards/renesas/edosk7705/Makefile
+++ b/arch/sh/boards/renesas/edosk7705/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the EDOSK7705 specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := setup.o io.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
index e8b4109ace11..3ef8cbaa0b37 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ b/arch/sh/boards/renesas/hs7751rvoip/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the HS7751RVoIP specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := mach.o setup.o io.o irq.o led.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 3a1abfa2fefb..09fb77ffb835 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -167,7 +167,7 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
                 *(volatile unsigned char *)port = value;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned cjar *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
         	*(unsigned char *)PCI_IOMAP(port) = value;
@@ -181,7 +181,7 @@ void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
                 *(volatile unsigned char *)port = value;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned cjar *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
         	*(unsigned char *)PCI_IOMAP(port) = value;
diff --git a/arch/sh/boards/renesas/rts7751r2d/Makefile b/arch/sh/boards/renesas/rts7751r2d/Makefile
index daa53334bdc3..eee1ed6f5727 100644
--- a/arch/sh/boards/renesas/rts7751r2d/Makefile
+++ b/arch/sh/boards/renesas/rts7751r2d/Makefile
@@ -1,10 +1,6 @@
 #
 # Makefile for the RTS7751R2D specific parts of the kernel
 #
-# Note! Dependencies are done automagically by 'make dep', which also
-# removes any old dependencies. DON'T put your own dependencies here
-# unless it's something special (ie not a .c file).
-#
 
 obj-y	 := mach.o setup.o io.o irq.o led.o
 
diff --git a/arch/sh/boards/renesas/rts7751r2d/led.c b/arch/sh/boards/renesas/rts7751r2d/led.c
index 4d16de71fac1..cf35f90dfe03 100644
--- a/arch/sh/boards/renesas/rts7751r2d/led.c
+++ b/arch/sh/boards/renesas/rts7751r2d/led.c
@@ -12,8 +12,6 @@
 #include <asm/io.h>
 #include <asm/rts7751r2d/rts7751r2d.h>
 
-extern unsigned int debug_counter;
-
 #ifdef CONFIG_HEARTBEAT
 
 #include <linux/sched.h>
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index 7953dde0b91c..60907f574f34 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -14,8 +14,6 @@
 #include <asm/io.h>
 #include <asm/rts7751r2d/rts7751r2d.h>
 
-unsigned int debug_counter;
-
 const char *get_system_type(void)
 {
 	return "RTS7751R2D";
@@ -34,5 +32,4 @@ void __init platform_setup(void)
 	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
-	debug_counter = 0;
 }
diff --git a/arch/sh/drivers/dma/dma-pvr2.c b/arch/sh/drivers/dma/dma-pvr2.c
index 30a580aa7cbd..3b0b0f60bb3c 100644
--- a/arch/sh/drivers/dma/dma-pvr2.c
+++ b/arch/sh/drivers/dma/dma-pvr2.c
@@ -18,8 +18,8 @@
 #include <asm/dma.h>
 #include <asm/io.h>
 
-static unsigned int xfer_complete = 0;
-static int count = 0;
+static unsigned int xfer_complete;
+static int count;
 
 static irqreturn_t pvr2_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
@@ -107,4 +107,3 @@ module_exit(pvr2_dma_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>");
 MODULE_DESCRIPTION("NEC PowerVR 2 DMA driver");
 MODULE_LICENSE("GPL");
-
diff --git a/arch/sh/kernel/cpu/clock.c b/arch/sh/kernel/cpu/clock.c
index 97fa37f42b84..e7eef07536f5 100644
--- a/arch/sh/kernel/cpu/clock.c
+++ b/arch/sh/kernel/cpu/clock.c
@@ -225,7 +225,7 @@ int __init clk_init(void)
 {
 	int i, ret = 0;
 
-	BUG_ON(unlikely(!master_clk.rate));
+	BUG_ON(!master_clk.rate);
 
 	for (i = 0; i < ARRAY_SIZE(onchip_clocks); i++) {
 		struct clk *clk = onchip_clocks[i];
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index b09805f3ee23..00f6a3c8c43b 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -24,9 +24,10 @@
 #include <linux/proc_fs.h>
 #include <linux/miscdevice.h>
 #include <linux/vmalloc.h>
-
+#include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/page.h>
+#include <asm/cacheflush.h>
 #include <asm/mmu_context.h>
 #include <asm/cpu/sq.h>
 
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 1a05804caf37..306eb8d83de5 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -18,24 +18,6 @@
 #include <asm/cpu/mmu_context.h>
 #include <asm/unistd.h>
 
-#if !defined(CONFIG_NFSD) && !defined(CONFIG_NFSD_MODULE)
-#define sys_nfsservctl		sys_ni_syscall
-#endif
-
-#if !defined(CONFIG_MMU)
-#define sys_madvise		sys_ni_syscall
-#define sys_readahead		sys_ni_syscall
-#define sys_mprotect		sys_ni_syscall
-#define sys_msync		sys_ni_syscall
-#define sys_mlock		sys_ni_syscall
-#define sys_munlock		sys_ni_syscall
-#define sys_mlockall		sys_ni_syscall
-#define sys_munlockall		sys_ni_syscall
-#define sys_mremap		sys_ni_syscall
-#define sys_mincore		sys_ni_syscall
-#define sys_remap_file_pages	sys_ni_syscall
-#endif
-
 ! NOTE:
 ! GNU as (as of 2.9.1) changes bf/s into bt/s and bra, when the address
 ! to be jumped is too far, but it causes illegal slot exception.
diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c
index f7eebbde3291..04ca13a041c1 100644
--- a/arch/sh/kernel/ptrace.c
+++ b/arch/sh/kernel/ptrace.c
@@ -224,7 +224,6 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 
 	case PTRACE_SETDSPREGS: {
 		unsigned long dp;
-		int i;
 
 		ret = -EIO;
 		dp = ((unsigned long) child) + THREAD_SIZE -
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 245ed8f945e8..8cad80cec9a6 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -117,6 +117,8 @@ EXPORT_SYMBOL(synchronize_irq);
 #endif
 
 EXPORT_SYMBOL(csum_partial);
+#ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
+#endif
 EXPORT_SYMBOL(consistent_sync);
 EXPORT_SYMBOL(clear_page);
diff --git a/drivers/char/watchdog/shwdt.c b/drivers/char/watchdog/shwdt.c
index 1355038f1044..7c39f39effdf 100644
--- a/drivers/char/watchdog/shwdt.c
+++ b/drivers/char/watchdog/shwdt.c
@@ -125,7 +125,6 @@ static void sh_wdt_start(void)
 
 /**
  * 	sh_wdt_stop - Stop the Watchdog
- *
  * 	Stops the watchdog.
  */
 static void sh_wdt_stop(void)
@@ -141,22 +140,20 @@ static void sh_wdt_stop(void)
 
 /**
  * 	sh_wdt_keepalive - Keep the Userspace Watchdog Alive
- *
  * 	The Userspace watchdog got a KeepAlive: schedule the next heartbeat.
  */
-static void sh_wdt_keepalive(void)
+static inline void sh_wdt_keepalive(void)
 {
 	next_heartbeat = jiffies + (heartbeat * HZ);
 }
 
 /**
  * 	sh_wdt_set_heartbeat - Set the Userspace Watchdog heartbeat
- *
  * 	Set the Userspace Watchdog heartbeat
  */
 static int sh_wdt_set_heartbeat(int t)
 {
-	if ((t < 1) || (t > 3600)) /* arbitrary upper limit */
+	if (unlikely((t < 1) || (t > 3600))) /* arbitrary upper limit */
 		return -EINVAL;
 
 	heartbeat = t;
@@ -165,7 +162,6 @@ static int sh_wdt_set_heartbeat(int t)
 
 /**
  * 	sh_wdt_ping - Ping the Watchdog
- *
  *	@data: Unused
  *
  * 	Clears overflow bit, resets timer counter.
@@ -182,14 +178,13 @@ static void sh_wdt_ping(unsigned long data)
 		sh_wdt_write_cnt(0);
 
 		mod_timer(&timer, next_ping_period(clock_division_ratio));
-	} else {
-		printk(KERN_WARNING PFX "Heartbeat lost! Will not ping the watchdog\n");
-	}
+	} else
+		printk(KERN_WARNING PFX "Heartbeat lost! Will not ping "
+		       "the watchdog\n");
 }
 
 /**
  * 	sh_wdt_open - Open the Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  *
@@ -209,7 +204,6 @@ static int sh_wdt_open(struct inode *inode, struct file *file)
 
 /**
  * 	sh_wdt_close - Close the Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  *
@@ -220,7 +214,8 @@ static int sh_wdt_close(struct inode *inode, struct file *file)
 	if (shwdt_expect_close == 42) {
 		sh_wdt_stop();
 	} else {
-		printk(KERN_CRIT PFX "Unexpected close, not stopping watchdog!\n");
+		printk(KERN_CRIT PFX "Unexpected close, not "
+		       "stopping watchdog!\n");
 		sh_wdt_keepalive();
 	}
 
@@ -232,7 +227,6 @@ static int sh_wdt_close(struct inode *inode, struct file *file)
 
 /**
  * 	sh_wdt_write - Write to Device
- *
  * 	@file: file handle of device
  * 	@buf: buffer to write
  * 	@count: length of buffer
@@ -265,7 +259,6 @@ static ssize_t sh_wdt_write(struct file *file, const char *buf,
 
 /**
  * 	sh_wdt_ioctl - Query Device
- *
  * 	@inode: inode of device
  * 	@file: file handle of device
  * 	@cmd: watchdog command
@@ -326,7 +319,6 @@ static int sh_wdt_ioctl(struct inode *inode, struct file *file,
 
 /**
  * 	sh_wdt_notify_sys - Notifier Handler
- *
  * 	@this: notifier block
  * 	@code: notifier event
  * 	@unused: unused
@@ -337,9 +329,8 @@ static int sh_wdt_ioctl(struct inode *inode, struct file *file,
 static int sh_wdt_notify_sys(struct notifier_block *this,
 			     unsigned long code, void *unused)
 {
-	if (code == SYS_DOWN || code == SYS_HALT) {
+	if (code == SYS_DOWN || code == SYS_HALT)
 		sh_wdt_stop();
-	}
 
 	return NOTIFY_DONE;
 }
@@ -354,7 +345,8 @@ static const struct file_operations sh_wdt_fops = {
 };
 
 static struct watchdog_info sh_wdt_info = {
-	.options		= WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT | WDIOF_MAGICCLOSE,
+	.options		= WDIOF_KEEPALIVEPING | WDIOF_SETTIMEOUT |
+				  WDIOF_MAGICCLOSE,
 	.firmware_version	= 1,
 	.identity		= "SH WDT",
 };
@@ -371,7 +363,6 @@ static struct miscdevice sh_wdt_miscdev = {
 
 /**
  * 	sh_wdt_init - Initialize module
- *
  * 	Registers the device and notifier handler. Actual device
  * 	initialization is handled by sh_wdt_open().
  */
@@ -381,15 +372,15 @@ static int __init sh_wdt_init(void)
 
 	if ((clock_division_ratio < 0x5) || (clock_division_ratio > 0x7)) {
 		clock_division_ratio = WTCSR_CKS_4096;
-		printk(KERN_INFO PFX "clock_division_ratio value must be 0x5<=x<=0x7, using %d\n",
-			clock_division_ratio);
+		printk(KERN_INFO PFX "clock_division_ratio value must "
+		       "be 0x5<=x<=0x7, using %d\n", clock_division_ratio);
 	}
 
-	if (sh_wdt_set_heartbeat(heartbeat))
-	{
+	rc = sh_wdt_set_heartbeat(heartbeat);
+	if (unlikely(rc)) {
 		heartbeat = WATCHDOG_HEARTBEAT;
-		printk(KERN_INFO PFX "heartbeat value must be 1<=x<=3600, using %d\n",
-			heartbeat);
+		printk(KERN_INFO PFX "heartbeat value must "
+		       "be 1<=x<=3600, using %d\n", heartbeat);
 	}
 
 	init_timer(&timer);
@@ -397,15 +388,16 @@ static int __init sh_wdt_init(void)
 	timer.data = 0;
 
 	rc = register_reboot_notifier(&sh_wdt_notifier);
-	if (rc) {
-		printk(KERN_ERR PFX "Can't register reboot notifier (err=%d)\n", rc);
+	if (unlikely(rc)) {
+		printk(KERN_ERR PFX "Can't register reboot notifier (err=%d)\n",
+		       rc);
 		return rc;
 	}
 
 	rc = misc_register(&sh_wdt_miscdev);
-	if (rc) {
-		printk(KERN_ERR PFX "Can't register miscdev on minor=%d (err=%d)\n",
-			sh_wdt_miscdev.minor, rc);
+	if (unlikely(rc)) {
+		printk(KERN_ERR PFX "Can't register miscdev on "
+		       "minor=%d (err=%d)\n", sh_wdt_miscdev.minor, rc);
 		unregister_reboot_notifier(&sh_wdt_notifier);
 		return rc;
 	}
@@ -418,7 +410,6 @@ static int __init sh_wdt_init(void)
 
 /**
  * 	sh_wdt_exit - Deinitialize module
- *
  * 	Unregisters the device and notifier handler. Actual device
  * 	deinitialization is handled by sh_wdt_close().
  */
@@ -434,14 +425,13 @@ MODULE_LICENSE("GPL");
 MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
 
 module_param(clock_division_ratio, int, 0);
-MODULE_PARM_DESC(clock_division_ratio, "Clock division ratio. Valid ranges are from 0x5 (1.31ms) to 0x7 (5.25ms). Defaults to 0x7.");
+MODULE_PARM_DESC(clock_division_ratio, "Clock division ratio. Valid ranges are from 0x5 (1.31ms) to 0x7 (5.25ms). (default=" __MODULE_STRING(clock_division_ratio) ")");
 
 module_param(heartbeat, int, 0);
 MODULE_PARM_DESC(heartbeat, "Watchdog heartbeat in seconds. (1<=heartbeat<=3600, default=" __MODULE_STRING(WATCHDOG_HEARTBEAT) ")");
 
 module_param(nowayout, int, 0);
-MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=CONFIG_WATCHDOG_NOWAYOUT)");
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default=" __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
 
 module_init(sh_wdt_init);
 module_exit(sh_wdt_exit);
-
diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h
index fb627de217f2..049eb2dda6b6 100644
--- a/include/asm-sh/atomic.h
+++ b/include/asm-sh/atomic.h
@@ -14,6 +14,7 @@ typedef struct { volatile int counter; } atomic_t;
 #define atomic_read(v)		((v)->counter)
 #define atomic_set(v,i)		((v)->counter = (i))
 
+#include <linux/compiler.h>
 #include <asm/system.h>
 
 /*
diff --git a/include/asm-sh/bitops.h b/include/asm-sh/bitops.h
index e34f82508568..1c16792cee1d 100644
--- a/include/asm-sh/bitops.h
+++ b/include/asm-sh/bitops.h
@@ -6,7 +6,7 @@
 /* For __swab32 */
 #include <asm/byteorder.h>
 
-static __inline__ void set_bit(int nr, volatile void * addr)
+static inline void set_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -24,7 +24,7 @@ static __inline__ void set_bit(int nr, volatile void * addr)
  */
 #define smp_mb__before_clear_bit()	barrier()
 #define smp_mb__after_clear_bit()	barrier()
-static __inline__ void clear_bit(int nr, volatile void * addr)
+static inline void clear_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -37,7 +37,7 @@ static __inline__ void clear_bit(int nr, volatile void * addr)
 	local_irq_restore(flags);
 }
 
-static __inline__ void change_bit(int nr, volatile void * addr)
+static inline void change_bit(int nr, volatile void * addr)
 {
 	int	mask;
 	volatile unsigned int *a = addr;
@@ -50,7 +50,7 @@ static __inline__ void change_bit(int nr, volatile void * addr)
 	local_irq_restore(flags);
 }
 
-static __inline__ int test_and_set_bit(int nr, volatile void * addr)
+static inline int test_and_set_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -66,7 +66,7 @@ static __inline__ int test_and_set_bit(int nr, volatile void * addr)
 	return retval;
 }
 
-static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
+static inline int test_and_clear_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -82,7 +82,7 @@ static __inline__ int test_and_clear_bit(int nr, volatile void * addr)
 	return retval;
 }
 
-static __inline__ int test_and_change_bit(int nr, volatile void * addr)
+static inline int test_and_change_bit(int nr, volatile void * addr)
 {
 	int	mask, retval;
 	volatile unsigned int *a = addr;
@@ -100,7 +100,7 @@ static __inline__ int test_and_change_bit(int nr, volatile void * addr)
 
 #include <asm-generic/bitops/non-atomic.h>
 
-static __inline__ unsigned long ffz(unsigned long word)
+static inline unsigned long ffz(unsigned long word)
 {
 	unsigned long result;
 
@@ -120,7 +120,7 @@ static __inline__ unsigned long ffz(unsigned long word)
  *
  * Undefined if no bit exists, so code should check against 0 first.
  */
-static __inline__ unsigned long __ffs(unsigned long word)
+static inline unsigned long __ffs(unsigned long word)
 {
 	unsigned long result;
 
diff --git a/include/asm-sh/checksum.h b/include/asm-sh/checksum.h
index fa03b30c4269..08168afe6746 100644
--- a/include/asm-sh/checksum.h
+++ b/include/asm-sh/checksum.h
@@ -159,6 +159,7 @@ static __inline__ unsigned short ip_compute_csum(unsigned char * buff, int len)
 }
 
 #define _HAVE_ARCH_IPV6_CSUM
+#ifdef CONFIG_IPV6
 static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr,
 						     struct in6_addr *daddr,
 						     __u32 len,
@@ -194,6 +195,7 @@ static __inline__ unsigned short int csum_ipv6_magic(struct in6_addr *saddr,
 
 	return csum_fold(sum);
 }
+#endif
 
 /* 
  *	Copy and checksum to user
diff --git a/include/asm-sh/fixmap.h b/include/asm-sh/fixmap.h
index 412bccaa07e6..458e9fa59545 100644
--- a/include/asm-sh/fixmap.h
+++ b/include/asm-sh/fixmap.h
@@ -25,7 +25,7 @@
  * addresses. The point is to have a constant address at
  * compile time, but to set the physical address only
  * in the boot process. We allocate these special  addresses
- * from the end of virtual memory (0xfffff000) backwards.
+ * from the end of P3 backwards.
  * Also this lets us do fail-safe vmalloc(), we
  * can guarantee that these special addresses and
  * vmalloc()-ed addresses never overlap.
diff --git a/include/asm-sh/watchdog.h b/include/asm-sh/watchdog.h
index 09ca41972a11..d19ea62ef8c6 100644
--- a/include/asm-sh/watchdog.h
+++ b/include/asm-sh/watchdog.h
@@ -62,7 +62,6 @@
 
 /**
  * 	sh_wdt_read_cnt - Read from Counter
- *
  * 	Reads back the WTCNT value.
  */
 static inline __u8 sh_wdt_read_cnt(void)
@@ -72,7 +71,6 @@ static inline __u8 sh_wdt_read_cnt(void)
 
 /**
  *	sh_wdt_write_cnt - Write to Counter
- *
  *	@val: Value to write
  *
  *	Writes the given value @val to the lower byte of the timer counter.
@@ -95,7 +93,6 @@ static inline __u8 sh_wdt_read_csr(void)
 
 /**
  * 	sh_wdt_write_csr - Write to Control/Status Register
- *
  * 	@val: Value to write
  *
  * 	Writes the given value @val to the lower byte of the control/status
-- 
cgit v1.2.1


From 4bcac20a7a01d49dffb5e88a8140efa34927c383 Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 13:07:38 +0900
Subject: sh: hp6xx mach-type cleanups.

Some minor cleanups for the updated consolidated hp6xx
mach-type.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile         |  2 +-
 drivers/video/hitfb.c    | 15 +--------------
 sound/oss/sh_dac_audio.c | 34 +++++++++++++++-------------------
 3 files changed, 17 insertions(+), 34 deletions(-)

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index e467a450662b..66cf0ea6d38f 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -113,7 +113,7 @@ incdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se
 incdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se7751
 incdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)        := se7300
 incdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se73180
-incdir-$(CONFIG_SH_HP600)			:= hp6xx
+incdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 
 ifneq ($(machdir-y),)
 core-y				+= arch/sh/boards/$(machdir-y)/
diff --git a/drivers/video/hitfb.c b/drivers/video/hitfb.c
index 4cc6b454265e..c6ac0dcd8854 100644
--- a/drivers/video/hitfb.c
+++ b/drivers/video/hitfb.c
@@ -28,10 +28,8 @@
 #include <asm/io.h>
 #include <asm/hd64461/hd64461.h>
 
-#ifdef MACH_HP600
 #include <asm/cpu/dac.h>
 #include <asm/hp6xx/hp6xx.h>
-#endif
 
 #define	WIDTH 640
 
@@ -192,12 +190,6 @@ int hitfb_blank(int blank_mode, struct fb_info *info)
 	unsigned short v;
 
 	if (blank_mode) {
-#ifdef MACH_HP600
-		sh_dac_disable(DAC_LCD_BRIGHTNESS);
-		v = fb_readw(HD64461_GPBDR);
-		v |= HD64461_GPBDR_LCDOFF;
-		fb_writew(v, HD64461_GPBDR);
-#endif
 		v = fb_readw(HD64461_LDR1);
 		v &= ~HD64461_LDR1_DON;
 		fb_writew(v, HD64461_LDR1);
@@ -213,12 +205,7 @@ int hitfb_blank(int blank_mode, struct fb_info *info)
 		v = fb_readw(HD64461_STBCR);
 		v &= ~HD64461_STBCR_SLCDST;
 		fb_writew(v, HD64461_STBCR);
-#ifdef MACH_HP600
-		sh_dac_enable(DAC_LCD_BRIGHTNESS);
-		v = fb_readw(HD64461_GPBDR);
-		v &= ~HD64461_GPBDR_LCDOFF;
-		fb_writew(v, HD64461_GPBDR);
-#endif
+
 		v = fb_readw(HD64461_LDR1);
 		v |= HD64461_LDR1_DON;
 		fb_writew(v, HD64461_LDR1);
diff --git a/sound/oss/sh_dac_audio.c b/sound/oss/sh_dac_audio.c
index 7b168d85f4ab..ae7fccc510a1 100644
--- a/sound/oss/sh_dac_audio.c
+++ b/sound/oss/sh_dac_audio.c
@@ -6,18 +6,15 @@
 #include <linux/fs.h>
 #include <linux/sound.h>
 #include <linux/soundcard.h>
+#include <linux/interrupt.h>
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/irq.h>
 #include <asm/delay.h>
-#include <linux/interrupt.h>
-
 #include <asm/cpu/dac.h>
-
-#ifdef MACH_HP600
+#include <asm/machvec.h>
 #include <asm/hp6xx/hp6xx.h>
 #include <asm/hd64461/hd64461.h>
-#endif
 
 #define MODNAME "sh_dac_audio"
 
@@ -71,26 +68,25 @@ static void dac_audio_sync(void)
 
 static void dac_audio_start(void)
 {
-#ifdef MACH_HP600
-	u16 v;
-	v = inw(HD64461_GPADR);
-	v &= ~HD64461_GPADR_SPEAKER;
-	outw(v, HD64461_GPADR);
-#endif
+	if (mach_is_hp6xx()) {
+		u16 v = inw(HD64461_GPADR);
+		v &= ~HD64461_GPADR_SPEAKER;
+		outw(v, HD64461_GPADR);
+	}
+
 	sh_dac_enable(CONFIG_SOUND_SH_DAC_AUDIO_CHANNEL);
 	ctrl_outw(TMU1_TCR_INIT, TMU1_TCR);
 }
 static void dac_audio_stop(void)
 {
-#ifdef MACH_HP600
-	u16 v;
-#endif
 	dac_audio_stop_timer();
-#ifdef MACH_HP600
-	v = inw(HD64461_GPADR);
-	v |= HD64461_GPADR_SPEAKER;
-	outw(v, HD64461_GPADR);
-#endif
+
+	if (mach_is_hp6xx()) {
+		u16 v = inw(HD64461_GPADR);
+		v |= HD64461_GPADR_SPEAKER;
+		outw(v, HD64461_GPADR);
+	}
+
 	sh_dac_disable(CONFIG_SOUND_SH_DAC_AUDIO_CHANNEL);
 }
 
-- 
cgit v1.2.1


From 3f787fe2e077ecfe43ad9ad56f12a8e21cffafc3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:11:57 +0900
Subject: sh: hugetlb updates.

For some of the larger sizes we permitted spanning pages
across several PTEs, but this turned out to not be generally
useful. This reverts the sh hugetlbpage interface to something
more sensible using huge pages at single PTE granularity.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/hugetlbpage.c | 52 +++++++++++++++---------------------------------
 include/asm-sh/page.h    |  1 -
 2 files changed, 16 insertions(+), 37 deletions(-)

diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index 2a85bc15a412..329059d6b54a 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -26,61 +26,41 @@
 pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte = NULL;
 
 	pgd = pgd_offset(mm, addr);
 	if (pgd) {
-		pmd = pmd_alloc(mm, pgd, addr);
-		if (pmd)
-			pte = pte_alloc_map(mm, pmd, addr);
+		pud = pud_alloc(mm, pgd, addr);
+		if (pud) {
+			pmd = pmd_alloc(mm, pud, addr);
+			if (pmd)
+				pte = pte_alloc_map(mm, pmd, addr);
+		}
 	}
+
 	return pte;
 }
 
 pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte = NULL;
 
 	pgd = pgd_offset(mm, addr);
 	if (pgd) {
-		pmd = pmd_offset(pgd, addr);
-		if (pmd)
-			pte = pte_offset_map(pmd, addr);
-	}
-	return pte;
-}
-
-void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
-		     pte_t *ptep, pte_t entry)
-{
-	int i;
-
-	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		set_pte_at(mm, addr, ptep, entry);
-		ptep++;
-		addr += PAGE_SIZE;
-		pte_val(entry) += PAGE_SIZE;
+		pud = pud_offset(pgd, addr);
+		if (pud) {
+			pmd = pmd_offset(pud, addr);
+			if (pmd)
+				pte = pte_offset_map(pmd, addr);
+		}
 	}
-}
-
-pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
-			      pte_t *ptep)
-{
-	pte_t entry;
-	int i;
-
-	entry = *ptep;
 
-	for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
-		pte_clear(mm, addr, ptep);
-		addr += PAGE_SIZE;
-		ptep++;
-	}
-
-	return entry;
+	return pte;
 }
 
 struct page *follow_huge_addr(struct mm_struct *mm,
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 6f7eb8a3aba5..4811d410d123 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -30,7 +30,6 @@
 #define HPAGE_SIZE		(1UL << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE-1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT-PAGE_SHIFT)
-#define ARCH_HAS_SETCLEAR_HUGE_PTE
 #endif
 
 #ifdef __KERNEL__
-- 
cgit v1.2.1


From 5c930e8cbfaa9dfdf9fa082fd891c6e3ab6ef2fc Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:20:22 +0900
Subject: video: Disable vgacon for SuperH.

We don't support this on SH, so just disable it..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/video/console/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/console/Kconfig b/drivers/video/console/Kconfig
index 4444bef68fba..aa3935df852a 100644
--- a/drivers/video/console/Kconfig
+++ b/drivers/video/console/Kconfig
@@ -6,7 +6,7 @@ menu "Console display driver support"
 
 config VGA_CONSOLE
 	bool "VGA text console" if EMBEDDED || !X86
-	depends on !ARCH_ACORN && !ARCH_EBSA110 && !4xx && !8xx && !SPARC && !M68K && !PARISC && !FRV && !ARCH_VERSATILE
+	depends on !ARCH_ACORN && !ARCH_EBSA110 && !4xx && !8xx && !SPARC && !M68K && !PARISC && !FRV && !ARCH_VERSATILE && !SUPERH
 	default y
 	help
 	  Saying Y here will allow you to use Linux in text mode through a
-- 
cgit v1.2.1


From 3530570fd43632b60b00e5ea17519d2bd69d1434 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:28:23 +0900
Subject: sh: Kill off dead code for SE and SystemH boards.

Some of these have suffered some bitrot, and so there is
some degree of dead code that has been left sitting around,
clean it up..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/systemh/io.c |  2 +-
 arch/sh/boards/se/73180/io.c        |  6 +++++-
 arch/sh/boards/se/73180/irq.c       |  9 ---------
 arch/sh/boards/se/73180/led.c       | 15 +--------------
 arch/sh/boards/se/770x/Makefile     |  3 ++-
 arch/sh/boards/se/770x/led.c        | 17 +----------------
 arch/sh/boards/se/7751/Makefile     |  3 ++-
 arch/sh/boards/se/7751/led.c        | 15 ---------------
 8 files changed, 12 insertions(+), 58 deletions(-)

diff --git a/arch/sh/boards/renesas/systemh/io.c b/arch/sh/boards/renesas/systemh/io.c
index cf979011aa94..85511576d414 100644
--- a/arch/sh/boards/renesas/systemh/io.c
+++ b/arch/sh/boards/renesas/systemh/io.c
@@ -15,7 +15,7 @@
 #include <asm/io.h>
 
 #include <linux/pci.h>
-#include "../../drivers/pci/pci-sh7751.h"
+#include "../../../drivers/pci/pci-sh7751.h"
 
 /*
  * The 7751 SystemH Engine uses the built-in PCI controller (PCIC)
diff --git a/arch/sh/boards/se/73180/io.c b/arch/sh/boards/se/73180/io.c
index 755df5ac4a4e..72715575458b 100644
--- a/arch/sh/boards/se/73180/io.c
+++ b/arch/sh/boards/se/73180/io.c
@@ -99,6 +99,7 @@ bad_outb(struct iop *p, unsigned char value, unsigned long port)
 	badio(inw, port);
 }
 
+#ifdef CONFIG_SMC91X
 /* MSTLANEX01 LAN at 0xb400:0000 */
 static struct iop laniop = {
 	.start = 0x300,
@@ -110,6 +111,7 @@ static struct iop laniop = {
 	.outb = simple_outb,
 	.outw = simple_outw,
 };
+#endif
 
 /* NE2000 pc card NIC */
 static struct iop neiop = {
@@ -123,6 +125,7 @@ static struct iop neiop = {
 	.outw = simple_outw,
 };
 
+#ifdef CONFIG_IDE
 /* CF in CF slot */
 static struct iop cfiop = {
 	.base = 0xb0600000,
@@ -132,12 +135,13 @@ static struct iop cfiop = {
 	.outb = pcc_outb,
 	.outw = simple_outw,
 };
+#endif
 
 static __inline__ struct iop *
 port2iop(unsigned long port)
 {
 	if (0) ;
-#if defined(CONFIG_SMC91111)
+#if defined(CONFIG_SMC91X)
 	else if (laniop.check(&laniop, port))
 		return &laniop;
 #endif
diff --git a/arch/sh/boards/se/73180/irq.c b/arch/sh/boards/se/73180/irq.c
index 4344d0ef24aa..2c62b8ea350e 100644
--- a/arch/sh/boards/se/73180/irq.c
+++ b/arch/sh/boards/se/73180/irq.c
@@ -7,7 +7,6 @@
  * Modified for SH-Mobile SolutionEngine 73180 Support
  *              by YOSHII Takashi <yoshii-takashi@hitachi-ul.co.jp>
  *
- *
  */
 
 #include <linux/init.h>
@@ -16,14 +15,6 @@
 #include <asm/io.h>
 #include <asm/mach/se73180.h>
 
-static int
-intreq2irq(int i)
-{
-	if (i == 5)
-		return 10;
-	return 32 + 7 - i;
-}
-
 static int
 irq2intreq(int irq)
 {
diff --git a/arch/sh/boards/se/73180/led.c b/arch/sh/boards/se/73180/led.c
index 610439fde6ee..4b72e9a3ead9 100644
--- a/arch/sh/boards/se/73180/led.c
+++ b/arch/sh/boards/se/73180/led.c
@@ -14,21 +14,8 @@
 #include <linux/sched.h>
 #include <asm/mach/se73180.h>
 
-static void
-mach_led(int position, int value)
-{
-	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
-
-	if (value) {
-		*p |= (1 << LED_SHIFT);
-	} else {
-		*p &= ~(1 << LED_SHIFT);
-	}
-}
-
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
-void
-heartbeat_73180se(void)
+void heartbeat_73180se(void)
 {
 	static unsigned int cnt = 0, period = 0;
 	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
diff --git a/arch/sh/boards/se/770x/Makefile b/arch/sh/boards/se/770x/Makefile
index be89a73cc418..c96ed6933fe1 100644
--- a/arch/sh/boards/se/770x/Makefile
+++ b/arch/sh/boards/se/770x/Makefile
@@ -2,5 +2,6 @@
 # Makefile for the 770x SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
+obj-y	 := mach.o setup.o io.o irq.o
 
+obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/se/770x/led.c b/arch/sh/boards/se/770x/led.c
index 3cddbda025fc..73e9848387ea 100644
--- a/arch/sh/boards/se/770x/led.c
+++ b/arch/sh/boards/se/770x/led.c
@@ -9,22 +9,8 @@
  * This file contains Solution Engine specific LED code.
  */
 
-#include <asm/se/se.h>
-
-static void mach_led(int position, int value)
-{
-	volatile unsigned short* p = (volatile unsigned short*)PA_LED;
-
-	if (value) {
-		*p |= (1<<8);
-	} else {
-		*p &= ~(1<<8);
-	}
-}
-
-#ifdef CONFIG_HEARTBEAT
-
 #include <linux/sched.h>
+#include <asm/se/se.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
 void heartbeat_se(void)
@@ -64,4 +50,3 @@ void heartbeat_se(void)
 	*p = 1<<(bit+8);
 
 }
-#endif /* CONFIG_HEARTBEAT */
diff --git a/arch/sh/boards/se/7751/Makefile b/arch/sh/boards/se/7751/Makefile
index ce7ca247f84d..07bf17a4a19b 100644
--- a/arch/sh/boards/se/7751/Makefile
+++ b/arch/sh/boards/se/7751/Makefile
@@ -2,7 +2,8 @@
 # Makefile for the 7751 SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
+obj-y	 := mach.o setup.o io.o irq.o
 
 obj-$(CONFIG_PCI) += pci.o
+obj-$(CONFIG_HEARTBEAT) += led.o
 
diff --git a/arch/sh/boards/se/7751/led.c b/arch/sh/boards/se/7751/led.c
index a878726d3c7c..4405e26cf866 100644
--- a/arch/sh/boards/se/7751/led.c
+++ b/arch/sh/boards/se/7751/led.c
@@ -10,20 +10,6 @@
  */
 
 #include <asm/se7751/se7751.h>
-
-static void mach_led(int position, int value)
-{
-	volatile unsigned short* p = (volatile unsigned short*)PA_LED;
-
-	if (value) {
-		*p |= (1<<8);
-	} else {
-		*p &= ~(1<<8);
-	}
-}
-
-#ifdef CONFIG_HEARTBEAT
-
 #include <linux/sched.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
@@ -64,4 +50,3 @@ void heartbeat_7751se(void)
 	*p = 1<<(bit+8);
 
 }
-#endif /* CONFIG_HEARTBEAT */
-- 
cgit v1.2.1


From d95fb13c960ae19e9fd4a95807eb68fa20caf537 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:30:08 +0900
Subject: sh: Fixup TMU_TOCR definition for SH7300.

SH7300 has a different TMU_TOCR, make the TMU code work again.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh3/timer.h | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/include/asm-sh/cpu-sh3/timer.h b/include/asm-sh/cpu-sh3/timer.h
index 3d8e95e8d10c..2082ad956f21 100644
--- a/include/asm-sh/cpu-sh3/timer.h
+++ b/include/asm-sh/cpu-sh3/timer.h
@@ -23,6 +23,10 @@
  * ---------------------------------------------------------------------------
  */
 
+#if !defined(CONFIG_CPU_SUBTYPE_SH7727)
+#define TMU_TOCR	0xfffffe90	/* Byte access */
+#endif
+
 #if defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define TMU_TSTR	0xa412fe92	/* Byte access */
 
@@ -39,9 +43,6 @@
 #define TMU2_TCR	0xa412feb4	/* Word access */
 
 #else
-#if !defined(CONFIG_CPU_SUBTYPE_SH7727)
-#define TMU_TOCR	0xfffffe90	/* Byte access */
-#endif
 #define TMU_TSTR	0xfffffe92	/* Byte access */
 
 #define TMU0_TCOR	0xfffffe94	/* Long access */
-- 
cgit v1.2.1


From 6d75e650f1d0d59fd97c7629f0903ef18e8dfb7b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:42:57 +0900
Subject: sh: Move hd64461.h to a more sensible location.

With the I/O rework for hd64461 we're down to a single header,
so move it by itself and get rid of the directory.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/cchips/hd6446x/hd64461/io.c    |  11 +-
 arch/sh/cchips/hd6446x/hd64461/setup.c |   3 +-
 include/asm-sh/hd64461.h               | 203 +++++++++++++++++++++++++++++++++
 include/asm-sh/hd64461/hd64461.h       | 202 --------------------------------
 include/asm-sh/hd64461/io.h            |  43 -------
 include/asm-sh/hp6xx/io.h              |   2 +-
 6 files changed, 210 insertions(+), 254 deletions(-)
 create mode 100644 include/asm-sh/hd64461.h
 delete mode 100644 include/asm-sh/hd64461/hd64461.h
 delete mode 100644 include/asm-sh/hd64461/io.h

diff --git a/arch/sh/cchips/hd6446x/hd64461/io.c b/arch/sh/cchips/hd6446x/hd64461/io.c
index ac3062671db7..f77f18f2ba7f 100644
--- a/arch/sh/cchips/hd6446x/hd64461/io.c
+++ b/arch/sh/cchips/hd6446x/hd64461/io.c
@@ -1,11 +1,10 @@
 /*
- *	$Id: io.c,v 1.6 2004/03/16 00:07:50 lethal Exp $
  *	Copyright (C) 2000 YAEGASHI Takeshi
  *	Typical I/O routines for HD64461 system.
  */
 
 #include <asm/io.h>
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 #define MEM_BASE (CONFIG_HD64461_IOBASE - HD64461_STBCR)
 
@@ -144,13 +143,13 @@ void hd64461_outsl(unsigned long port, const void *buffer, unsigned long count)
 	while(count--) *addr=*buf++;
 }
 
-unsigned short hd64461_readw(unsigned long addr)
+unsigned short hd64461_readw(void __iomem *addr)
 {
-	return *(volatile unsigned short*)(MEM_BASE+addr);
+	return ctrl_inw(MEM_BASE+(unsigned long __force)addr);
 }
 
-void hd64461_writew(unsigned short b, unsigned long addr)
+void hd64461_writew(unsigned short b, void __iomem *addr)
 {
-	*(volatile unsigned short*)(MEM_BASE+addr) = b;
+	ctrl_outw(b, MEM_BASE+(unsigned long __force)addr);
 }
 
diff --git a/arch/sh/cchips/hd6446x/hd64461/setup.c b/arch/sh/cchips/hd6446x/hd64461/setup.c
index ad126016720f..6c961273a23a 100644
--- a/arch/sh/cchips/hd6446x/hd64461/setup.c
+++ b/arch/sh/cchips/hd6446x/hd64461/setup.c
@@ -14,8 +14,7 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 static void disable_hd64461_irq(unsigned int irq)
 {
diff --git a/include/asm-sh/hd64461.h b/include/asm-sh/hd64461.h
new file mode 100644
index 000000000000..0f2e2132cc35
--- /dev/null
+++ b/include/asm-sh/hd64461.h
@@ -0,0 +1,203 @@
+#ifndef __ASM_SH_HD64461
+#define __ASM_SH_HD64461
+/*
+ *	$Id: hd64461.h,v 1.5 2004/03/16 00:07:51 lethal Exp $
+ *	Copyright (C) 2000 YAEGASHI Takeshi
+ *	Hitachi HD64461 companion chip support
+ */
+
+/* Constants for PCMCIA mappings */
+#define HD64461_PCC_WINDOW	0x01000000
+
+#define HD64461_PCC0_BASE	0xb8000000	/* area 6 */
+#define HD64461_PCC0_ATTR	(HD64461_PCC0_BASE)
+#define HD64461_PCC0_COMM	(HD64461_PCC0_BASE+HD64461_PCC_WINDOW)
+#define HD64461_PCC0_IO		(HD64461_PCC0_BASE+2*HD64461_PCC_WINDOW)
+
+#define HD64461_PCC1_BASE	0xb4000000	/* area 5 */
+#define HD64461_PCC1_ATTR	(HD64461_PCC1_BASE)
+#define HD64461_PCC1_COMM	(HD64461_PCC1_BASE+HD64461_PCC_WINDOW)
+
+#define HD64461_STBCR	0x10000
+#define HD64461_STBCR_CKIO_STBY			0x2000
+#define HD64461_STBCR_SAFECKE_IST		0x1000
+#define HD64461_STBCR_SLCKE_IST			0x0800
+#define HD64461_STBCR_SAFECKE_OST		0x0400
+#define HD64461_STBCR_SLCKE_OST			0x0200
+#define HD64461_STBCR_SMIAST			0x0100
+#define HD64461_STBCR_SLCDST			0x0080
+#define HD64461_STBCR_SPC0ST			0x0040
+#define HD64461_STBCR_SPC1ST			0x0020
+#define HD64461_STBCR_SAFEST			0x0010
+#define HD64461_STBCR_STM0ST			0x0008
+#define HD64461_STBCR_STM1ST			0x0004
+#define HD64461_STBCR_SIRST				0x0002
+#define HD64461_STBCR_SURTST			0x0001
+
+#define HD64461_SYSCR	0x10002
+#define HD64461_SCPUCR	0x10004
+
+#define HD64461_LCDCBAR		0x11000
+#define HD64461_LCDCLOR		0x11002
+#define HD64461_LCDCCR		0x11004
+#define HD64461_LCDCCR_MOFF	0x80
+
+#define	HD64461_LDR1		0x11010
+#define	HD64461_LDR1_DON	0x01
+#define	HD64461_LDR1_DINV	0x80
+
+#define	HD64461_LDR2		0x11012
+#define	HD64461_LDHNCR		0x11014
+#define	HD64461_LDHNSR		0x11016
+#define HD64461_LDVNTR		0x11018
+#define HD64461_LDVNDR		0x1101a
+#define HD64461_LDVSPR		0x1101c
+#define HD64461_LDR3		0x1101e
+
+#define HD64461_CPTWAR		0x11030
+#define HD64461_CPTWDR		0x11032
+#define HD64461_CPTRAR		0x11034
+#define HD64461_CPTRDR		0x11036
+
+#define HD64461_GRDOR		0x11040
+#define HD64461_GRSCR		0x11042
+#define HD64461_GRCFGR		0x11044
+#define HD64461_GRCFGR_ACCSTATUS		0x10
+#define HD64461_GRCFGR_ACCRESET			0x08
+#define HD64461_GRCFGR_ACCSTART_BITBLT	0x06
+#define HD64461_GRCFGR_ACCSTART_LINE	0x04
+#define HD64461_GRCFGR_COLORDEPTH16		0x01
+
+#define HD64461_LNSARH		0x11046
+#define HD64461_LNSARL		0x11048
+#define HD64461_LNAXLR		0x1104a
+#define HD64461_LNDGR		0x1104c
+#define HD64461_LNAXR		0x1104e
+#define HD64461_LNERTR		0x11050
+#define HD64461_LNMDR		0x11052
+#define HD64461_BBTSSARH	0x11054
+#define HD64461_BBTSSARL	0x11056
+#define HD64461_BBTDSARH	0x11058
+#define HD64461_BBTDSARL	0x1105a
+#define HD64461_BBTDWR		0x1105c
+#define HD64461_BBTDHR		0x1105e
+#define HD64461_BBTPARH		0x11060
+#define HD64461_BBTPARL		0x11062
+#define HD64461_BBTMARH		0x11064
+#define HD64461_BBTMARL		0x11066
+#define HD64461_BBTROPR		0x11068
+#define HD64461_BBTMDR		0x1106a
+
+/* PC Card Controller Registers */
+#define HD64461_PCC0ISR         0x12000 /* socket 0 interface status */
+#define HD64461_PCC0GCR         0x12002 /* socket 0 general control */
+#define HD64461_PCC0CSCR        0x12004 /* socket 0 card status change */
+#define HD64461_PCC0CSCIER      0x12006 /* socket 0 card status change interrupt enable */
+#define HD64461_PCC0SCR         0x12008 /* socket 0 software control */
+#define HD64461_PCC1ISR         0x12010 /* socket 1 interface status */
+#define HD64461_PCC1GCR         0x12012 /* socket 1 general control */
+#define HD64461_PCC1CSCR        0x12014 /* socket 1 card status change */
+#define HD64461_PCC1CSCIER      0x12016 /* socket 1 card status change interrupt enable */
+#define HD64461_PCC1SCR         0x12018 /* socket 1 software control */
+
+/* PCC Interface Status Register */
+#define HD64461_PCCISR_READY		0x80	/* card ready */
+#define HD64461_PCCISR_MWP		0x40	/* card write-protected */
+#define HD64461_PCCISR_VS2		0x20	/* voltage select pin 2 */
+#define HD64461_PCCISR_VS1		0x10	/* voltage select pin 1 */
+#define HD64461_PCCISR_CD2		0x08	/* card detect 2 */
+#define HD64461_PCCISR_CD1		0x04	/* card detect 1 */
+#define HD64461_PCCISR_BVD2		0x02	/* battery 1 */
+#define HD64461_PCCISR_BVD1		0x01	/* battery 1 */
+
+#define HD64461_PCCISR_PCD_MASK		0x0c    /* card detect */
+#define HD64461_PCCISR_BVD_MASK		0x03    /* battery voltage */
+#define HD64461_PCCISR_BVD_BATGOOD	0x03    /* battery good */
+#define HD64461_PCCISR_BVD_BATWARN	0x01    /* battery low warning */
+#define HD64461_PCCISR_BVD_BATDEAD1	0x02    /* battery dead */
+#define HD64461_PCCISR_BVD_BATDEAD2	0x00    /* battery dead */
+
+/* PCC General Control Register */
+#define HD64461_PCCGCR_DRVE		0x80    /* output drive */
+#define HD64461_PCCGCR_PCCR		0x40    /* PC card reset */
+#define HD64461_PCCGCR_PCCT		0x20    /* PC card type, 1=IO&mem, 0=mem */
+#define HD64461_PCCGCR_VCC0		0x10    /* voltage control pin VCC0SEL0 */
+#define HD64461_PCCGCR_PMMOD		0x08    /* memory mode */
+#define HD64461_PCCGCR_PA25		0x04    /* pin A25 */
+#define HD64461_PCCGCR_PA24		0x02    /* pin A24 */
+#define HD64461_PCCGCR_REG		0x01    /* pin PCC0REG# */
+
+/* PCC Card Status Change Register */
+#define HD64461_PCCCSCR_SCDI		0x80    /* sw card detect intr */
+#define HD64461_PCCCSCR_SRV1		0x40    /* reserved */
+#define HD64461_PCCCSCR_IREQ		0x20    /* IREQ intr req */
+#define HD64461_PCCCSCR_SC		0x10    /* STSCHG (status change) pin */
+#define HD64461_PCCCSCR_CDC		0x08    /* CD (card detect) change */
+#define HD64461_PCCCSCR_RC		0x04    /* READY change */
+#define HD64461_PCCCSCR_BW		0x02    /* battery warning change */
+#define HD64461_PCCCSCR_BD		0x01    /* battery dead change */
+
+/* PCC Card Status Change Interrupt Enable Register */
+#define HD64461_PCCCSCIER_CRE		0x80    /* change reset enable */
+#define HD64461_PCCCSCIER_IREQE_MASK	0x60   /* IREQ enable */
+#define HD64461_PCCCSCIER_IREQE_DISABLED	0x00   /* IREQ disabled */
+#define HD64461_PCCCSCIER_IREQE_LEVEL	0x20   /* IREQ level-triggered */
+#define HD64461_PCCCSCIER_IREQE_FALLING	0x40   /* IREQ falling-edge-trig */
+#define HD64461_PCCCSCIER_IREQE_RISING	0x60   /* IREQ rising-edge-trig */
+
+#define HD64461_PCCCSCIER_SCE		0x10    /* status change enable */
+#define HD64461_PCCCSCIER_CDE		0x08    /* card detect change enable */
+#define HD64461_PCCCSCIER_RE		0x04    /* ready change enable */
+#define HD64461_PCCCSCIER_BWE		0x02    /* battery warn change enable */
+#define HD64461_PCCCSCIER_BDE		0x01    /* battery dead change enable*/
+
+/* PCC Software Control Register */
+#define HD64461_PCCSCR_VCC1		0x02	/* voltage control pin 1 */
+#define HD64461_PCCSCR_SWP		0x01    /* write protect */
+
+#define HD64461_P0OCR           0x1202a
+#define HD64461_P1OCR           0x1202c
+#define HD64461_PGCR            0x1202e
+
+#define HD64461_GPACR		0x14000
+#define HD64461_GPBCR		0x14002
+#define HD64461_GPCCR		0x14004
+#define HD64461_GPDCR		0x14006
+#define HD64461_GPADR		0x14010
+#define HD64461_GPBDR		0x14012
+#define HD64461_GPCDR		0x14014
+#define HD64461_GPDDR		0x14016
+#define HD64461_GPAICR		0x14020
+#define HD64461_GPBICR		0x14022
+#define HD64461_GPCICR		0x14024
+#define HD64461_GPDICR		0x14026
+#define HD64461_GPAISR		0x14040
+#define HD64461_GPBISR		0x14042
+#define HD64461_GPCISR		0x14044
+#define HD64461_GPDISR		0x14046
+
+#define HD64461_NIRR		0x15000
+#define HD64461_NIMR		0x15002
+
+#define HD64461_IRQBASE		OFFCHIP_IRQ_BASE
+#define HD64461_IRQ_NUM		16
+
+#define HD64461_IRQ_UART	(HD64461_IRQBASE+5)
+#define HD64461_IRQ_IRDA	(HD64461_IRQBASE+6)
+#define HD64461_IRQ_TMU1	(HD64461_IRQBASE+9)
+#define HD64461_IRQ_TMU0	(HD64461_IRQBASE+10)
+#define HD64461_IRQ_GPIO	(HD64461_IRQBASE+11)
+#define HD64461_IRQ_AFE		(HD64461_IRQBASE+12)
+#define HD64461_IRQ_PCC1	(HD64461_IRQBASE+13)
+#define HD64461_IRQ_PCC0	(HD64461_IRQBASE+14)
+
+#define __IO_PREFIX	hd64461
+#include <asm/io_generic.h>
+
+/* arch/sh/cchips/hd6446x/hd64461/setup.c */
+int hd64461_irq_demux(int irq);
+void hd64461_register_irq_demux(int irq,
+				int (*demux) (int irq, void *dev), void *dev);
+void hd64461_unregister_irq_demux(int irq);
+
+#endif
diff --git a/include/asm-sh/hd64461/hd64461.h b/include/asm-sh/hd64461/hd64461.h
deleted file mode 100644
index 87f13d24c630..000000000000
--- a/include/asm-sh/hd64461/hd64461.h
+++ /dev/null
@@ -1,202 +0,0 @@
-#ifndef __ASM_SH_HD64461
-#define __ASM_SH_HD64461
-/*
- *	$Id: hd64461.h,v 1.5 2004/03/16 00:07:51 lethal Exp $
- *	Copyright (C) 2000 YAEGASHI Takeshi
- *	Hitachi HD64461 companion chip support
- */
-
-/* Constants for PCMCIA mappings */
-#define HD64461_PCC_WINDOW	0x01000000
-
-#define HD64461_PCC0_BASE	0xb8000000	/* area 6 */
-#define HD64461_PCC0_ATTR	(HD64461_PCC0_BASE)
-#define HD64461_PCC0_COMM	(HD64461_PCC0_BASE+HD64461_PCC_WINDOW)
-#define HD64461_PCC0_IO		(HD64461_PCC0_BASE+2*HD64461_PCC_WINDOW)
-
-#define HD64461_PCC1_BASE	0xb4000000	/* area 5 */
-#define HD64461_PCC1_ATTR	(HD64461_PCC1_BASE)
-#define HD64461_PCC1_COMM	(HD64461_PCC1_BASE+HD64461_PCC_WINDOW)
-
-#define HD64461_STBCR	0x10000
-#define HD64461_STBCR_CKIO_STBY			0x2000
-#define HD64461_STBCR_SAFECKE_IST		0x1000
-#define HD64461_STBCR_SLCKE_IST			0x0800
-#define HD64461_STBCR_SAFECKE_OST		0x0400
-#define HD64461_STBCR_SLCKE_OST			0x0200
-#define HD64461_STBCR_SMIAST			0x0100
-#define HD64461_STBCR_SLCDST			0x0080
-#define HD64461_STBCR_SPC0ST			0x0040
-#define HD64461_STBCR_SPC1ST			0x0020
-#define HD64461_STBCR_SAFEST			0x0010
-#define HD64461_STBCR_STM0ST			0x0008
-#define HD64461_STBCR_STM1ST			0x0004
-#define HD64461_STBCR_SIRST				0x0002
-#define HD64461_STBCR_SURTST			0x0001
-
-#define HD64461_SYSCR	0x10002
-#define HD64461_SCPUCR	0x10004
-
-#define HD64461_LCDCBAR		0x11000
-#define HD64461_LCDCLOR		0x11002
-#define HD64461_LCDCCR		0x11004
-#define HD64461_LCDCCR_MOFF	0x80
-
-#define	HD64461_LDR1		0x11010
-#define	HD64461_LDR1_DON	0x01
-#define	HD64461_LDR1_DINV	0x80
-
-#define	HD64461_LDR2		0x11012
-#define	HD64461_LDHNCR		0x11014
-#define	HD64461_LDHNSR		0x11016
-#define HD64461_LDVNTR		0x11018
-#define HD64461_LDVNDR		0x1101a
-#define HD64461_LDVSPR		0x1101c
-#define HD64461_LDR3		0x1101e
-
-#define HD64461_CPTWAR		0x11030	
-#define HD64461_CPTWDR		0x11032
-#define HD64461_CPTRAR		0x11034	
-#define HD64461_CPTRDR		0x11036
-
-#define HD64461_GRDOR		0x11040
-#define HD64461_GRSCR		0x11042
-#define HD64461_GRCFGR		0x11044
-#define HD64461_GRCFGR_ACCSTATUS		0x10
-#define HD64461_GRCFGR_ACCRESET			0x08
-#define HD64461_GRCFGR_ACCSTART_BITBLT	0x06
-#define HD64461_GRCFGR_ACCSTART_LINE	0x04
-#define HD64461_GRCFGR_COLORDEPTH16		0x01
-
-#define HD64461_LNSARH		0x11046
-#define HD64461_LNSARL		0x11048
-#define HD64461_LNAXLR		0x1104a
-#define HD64461_LNDGR		0x1104c
-#define HD64461_LNAXR		0x1104e
-#define HD64461_LNERTR		0x11050
-#define HD64461_LNMDR		0x11052
-#define HD64461_BBTSSARH	0x11054
-#define HD64461_BBTSSARL	0x11056
-#define HD64461_BBTDSARH	0x11058
-#define HD64461_BBTDSARL	0x1105a
-#define HD64461_BBTDWR		0x1105c
-#define HD64461_BBTDHR		0x1105e
-#define HD64461_BBTPARH		0x11060
-#define HD64461_BBTPARL		0x11062
-#define HD64461_BBTMARH		0x11064
-#define HD64461_BBTMARL		0x11066
-#define HD64461_BBTROPR		0x11068
-#define HD64461_BBTMDR		0x1106a
-
-/* PC Card Controller Registers */
-#define HD64461_PCC0ISR         0x12000 /* socket 0 interface status */
-#define HD64461_PCC0GCR         0x12002 /* socket 0 general control */
-#define HD64461_PCC0CSCR        0x12004 /* socket 0 card status change */
-#define HD64461_PCC0CSCIER      0x12006 /* socket 0 card status change interrupt enable */
-#define HD64461_PCC0SCR         0x12008 /* socket 0 software control */
-#define HD64461_PCC1ISR         0x12010 /* socket 1 interface status */
-#define HD64461_PCC1GCR         0x12012 /* socket 1 general control */
-#define HD64461_PCC1CSCR        0x12014 /* socket 1 card status change */
-#define HD64461_PCC1CSCIER      0x12016 /* socket 1 card status change interrupt enable */
-#define HD64461_PCC1SCR         0x12018 /* socket 1 software control */
-
-/* PCC Interface Status Register */
-#define HD64461_PCCISR_READY		0x80	/* card ready */
-#define HD64461_PCCISR_MWP		0x40	/* card write-protected */
-#define HD64461_PCCISR_VS2		0x20	/* voltage select pin 2 */
-#define HD64461_PCCISR_VS1		0x10	/* voltage select pin 1 */
-#define HD64461_PCCISR_CD2		0x08	/* card detect 2 */
-#define HD64461_PCCISR_CD1		0x04	/* card detect 1 */
-#define HD64461_PCCISR_BVD2		0x02	/* battery 1 */
-#define HD64461_PCCISR_BVD1		0x01	/* battery 1 */
-
-#define HD64461_PCCISR_PCD_MASK		0x0c    /* card detect */
-#define HD64461_PCCISR_BVD_MASK	0x03    /* battery voltage */
-#define HD64461_PCCISR_BVD_BATGOOD	0x03    /* battery good */
-#define HD64461_PCCISR_BVD_BATWARN	0x01    /* battery low warning */
-#define HD64461_PCCISR_BVD_BATDEAD1	0x02    /* battery dead */
-#define HD64461_PCCISR_BVD_BATDEAD2	0x00    /* battery dead */
-
-/* PCC General Control Register */
-#define HD64461_PCCGCR_DRVE		0x80    /* output drive */
-#define HD64461_PCCGCR_PCCR		0x40    /* PC card reset */
-#define HD64461_PCCGCR_PCCT		0x20    /* PC card type, 1=IO&mem, 0=mem */
-#define HD64461_PCCGCR_VCC0		0x10    /* voltage control pin VCC0SEL0 */
-#define HD64461_PCCGCR_PMMOD		0x08    /* memory mode */
-#define HD64461_PCCGCR_PA25		0x04    /* pin A25 */
-#define HD64461_PCCGCR_PA24		0x02    /* pin A24 */
-#define HD64461_PCCGCR_REG		0x01    /* pin PCC0REG# */
-
-/* PCC Card Status Change Register */
-#define HD64461_PCCCSCR_SCDI		0x80    /* sw card detect intr */
-#define HD64461_PCCCSCR_SRV1		0x40    /* reserved */
-#define HD64461_PCCCSCR_IREQ		0x20    /* IREQ intr req */
-#define HD64461_PCCCSCR_SC		0x10    /* STSCHG (status change) pin */
-#define HD64461_PCCCSCR_CDC		0x08    /* CD (card detect) change */
-#define HD64461_PCCCSCR_RC		0x04    /* READY change */
-#define HD64461_PCCCSCR_BW		0x02    /* battery warning change */
-#define HD64461_PCCCSCR_BD		0x01    /* battery dead change */
-
-/* PCC Card Status Change Interrupt Enable Register */
-#define HD64461_PCCCSCIER_CRE		0x80    /* change reset enable */
-#define HD64461_PCCCSCIER_IREQE_MASK   	0x60   /* IREQ enable */
-#define HD64461_PCCCSCIER_IREQE_DISABLED	0x00   /* IREQ disabled */
-#define HD64461_PCCCSCIER_IREQE_LEVEL  	0x20   /* IREQ level-triggered */
-#define HD64461_PCCCSCIER_IREQE_FALLING	0x40   /* IREQ falling-edge-trig */
-#define HD64461_PCCCSCIER_IREQE_RISING 	0x60   /* IREQ rising-edge-trig */
-
-#define HD64461_PCCCSCIER_SCE		0x10    /* status change enable */
-#define HD64461_PCCCSCIER_CDE		0x08    /* card detect change enable */
-#define HD64461_PCCCSCIER_RE		0x04    /* ready change enable */
-#define HD64461_PCCCSCIER_BWE		0x02    /* battery warn change enable */
-#define HD64461_PCCCSCIER_BDE		0x01    /* battery dead change enable*/
-
-/* PCC Software Control Register */
-#define HD64461_PCCSCR_VCC1		0x02	/* voltage control pin 1 */
-#define HD64461_PCCSCR_SWP		0x01    /* write protect */
-
-
-#define HD64461_P0OCR           0x1202a
-#define HD64461_P1OCR           0x1202c
-#define HD64461_PGCR            0x1202e
-
-#define HD64461_GPACR		0x14000
-#define HD64461_GPBCR		0x14002
-#define HD64461_GPCCR		0x14004
-#define HD64461_GPDCR		0x14006
-#define HD64461_GPADR		0x14010
-#define HD64461_GPBDR		0x14012
-#define HD64461_GPCDR		0x14014
-#define HD64461_GPDDR		0x14016
-#define HD64461_GPAICR		0x14020
-#define HD64461_GPBICR		0x14022
-#define HD64461_GPCICR		0x14024
-#define HD64461_GPDICR		0x14026
-#define HD64461_GPAISR		0x14040
-#define HD64461_GPBISR		0x14042
-#define HD64461_GPCISR		0x14044
-#define HD64461_GPDISR		0x14046
-
-#define HD64461_NIRR		0x15000
-#define HD64461_NIMR		0x15002
-
-#ifndef CONFIG_HD64461_IOBASE
-#define CONFIG_HD64461_IOBASE	0xb0000000
-#endif
-#ifndef CONFIG_HD64461_IRQ
-#define CONFIG_HD64461_IRQ	36
-#endif
-
-#define HD64461_IRQBASE		OFFCHIP_IRQ_BASE
-#define HD64461_IRQ_NUM 	16
-
-#define HD64461_IRQ_UART    	(HD64461_IRQBASE+5)
-#define HD64461_IRQ_IRDA    	(HD64461_IRQBASE+6)
-#define HD64461_IRQ_TMU1   	(HD64461_IRQBASE+9)
-#define HD64461_IRQ_TMU0  	(HD64461_IRQBASE+10)
-#define HD64461_IRQ_GPIO    	(HD64461_IRQBASE+11)
-#define HD64461_IRQ_AFE     	(HD64461_IRQBASE+12)
-#define HD64461_IRQ_PCC1 	(HD64461_IRQBASE+13)
-#define HD64461_IRQ_PCC0 	(HD64461_IRQBASE+14)
-
-#endif
diff --git a/include/asm-sh/hd64461/io.h b/include/asm-sh/hd64461/io.h
deleted file mode 100644
index 67f2489088d9..000000000000
--- a/include/asm-sh/hd64461/io.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * include/asm-sh/io_hd64461.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an HD64461
- */
-
-#ifndef _ASM_SH_IO_HD64461_H
-#define _ASM_SH_IO_HD64461_H
-
-extern unsigned char hd64461_inb(unsigned long port);
-extern unsigned short hd64461_inw(unsigned long port);
-extern unsigned int hd64461_inl(unsigned long port);
-
-extern void hd64461_outb(unsigned char value, unsigned long port);
-extern void hd64461_outw(unsigned short value, unsigned long port);
-extern void hd64461_outl(unsigned int value, unsigned long port);
-
-extern unsigned char hd64461_inb_p(unsigned long port);
-extern void hd64461_outb_p(unsigned char value, unsigned long port);
-
-extern void hd64461_insb(unsigned long port, void *addr, unsigned long count);
-extern void hd64461_insw(unsigned long port, void *addr, unsigned long count);
-extern void hd64461_insl(unsigned long port, void *addr, unsigned long count);
-
-extern void hd64461_outsb(unsigned long port, const void *buffer, unsigned long count);
-extern void hd64461_outsw(unsigned long port, const void *buffer, unsigned long count);
-extern void hd64461_outsl(unsigned long port, const void *buffer, unsigned long count);
-
-extern unsigned short hd64461_readw(unsigned long addr);
-extern void hd64461_writew(unsigned short b, unsigned long addr);
-
-
-extern int hd64461_irq_demux(int irq);
-extern void hd64461_register_irq_demux(int irq,
-		int (*demux)(int irq, void *dev), void *dev);
-extern void hd64461_unregister_irq_demux(int irq);
-
-#endif /* _ASM_SH_IO_HD64461_H */
diff --git a/include/asm-sh/hp6xx/io.h b/include/asm-sh/hp6xx/io.h
index 731798003550..2044476ab199 100644
--- a/include/asm-sh/hp6xx/io.h
+++ b/include/asm-sh/hp6xx/io.h
@@ -4,7 +4,7 @@
 /*
  * Nothing special here.. just use the generic cchip io routines.
  */
-#include <asm/hd64461/io.h>
+#include <asm/hd64461.h>
 
 #endif /* __ASM_SH_HP6XX_IO_H */
 
-- 
cgit v1.2.1


From 048839dc548a5315b733993dfc7d082e1e848061 Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 13:47:22 +0900
Subject: video: hitfb suspend/resume and updates.

suspend/resume support for hitfb, as well as some other
minor cleanups.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/video/hitfb.c | 214 ++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 181 insertions(+), 33 deletions(-)

diff --git a/drivers/video/hitfb.c b/drivers/video/hitfb.c
index c6ac0dcd8854..3afb472763c0 100644
--- a/drivers/video/hitfb.c
+++ b/drivers/video/hitfb.c
@@ -4,7 +4,7 @@
  * (C) 1999 Mihai Spatar
  * (C) 2000 YAEGASHI Takeshi
  * (C) 2003, 2004 Paul Mundt
- * (C) 2003, 2004 Andriy Skulysh
+ * (C) 2003, 2004, 2006 Andriy Skulysh
  *
  *  This file is subject to the terms and conditions of the GNU General Public
  *  License. See the file COPYING in the main directory of this archive for
@@ -20,14 +20,14 @@
 #include <linux/slab.h>
 #include <linux/delay.h>
 #include <linux/init.h>
+#include <linux/platform_device.h>
 #include <linux/fb.h>
 
 #include <asm/machvec.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/io.h>
-#include <asm/hd64461/hd64461.h>
-
+#include <asm/hd64461.h>
 #include <asm/cpu/dac.h>
 #include <asm/hp6xx/hp6xx.h>
 
@@ -43,7 +43,6 @@ static struct fb_var_screeninfo hitfb_var __initdata = {
 static struct fb_fix_screeninfo hitfb_fix __initdata = {
 	.id		= "Hitachi HD64461",
 	.type		= FB_TYPE_PACKED_PIXELS,
-	.ypanstep	= 8,
 	.accel		= FB_ACCEL_NONE,
 };
 
@@ -71,26 +70,14 @@ static inline void hitfb_accel_set_dest(int truecolor, u16 dx, u16 dy,
 	if (truecolor)
 		saddr <<= 1;
 
-	fb_writew(width, HD64461_BBTDWR);
-	fb_writew(height, HD64461_BBTDHR);
+	fb_writew(width-1, HD64461_BBTDWR);
+	fb_writew(height-1, HD64461_BBTDHR);
 
 	fb_writew(saddr & 0xffff, HD64461_BBTDSARL);
 	fb_writew(saddr >> 16, HD64461_BBTDSARH);
 
 }
 
-static inline void hitfb_accel_solidfill(int truecolor, u16 dx, u16 dy,
-					 u16 width, u16 height, u16 color)
-{
-	hitfb_accel_set_dest(truecolor, dx, dy, width, height);
-
-	fb_writew(0x00f0, HD64461_BBTROPR);
-	fb_writew(16, HD64461_BBTMDR);
-	fb_writew(color, HD64461_GRSCR);
-
-	hitfb_accel_start(truecolor);
-}
-
 static inline void hitfb_accel_bitblt(int truecolor, u16 sx, u16 sy, u16 dx,
 				      u16 dy, u16 width, u16 height, u16 rop,
 				      u32 mask_addr)
@@ -98,6 +85,8 @@ static inline void hitfb_accel_bitblt(int truecolor, u16 sx, u16 sy, u16 dx,
 	u32 saddr, daddr;
 	u32 maddr = 0;
 
+	height--;
+	width--;
 	fb_writew(rop, HD64461_BBTROPR);
 	if ((sy < dy) || ((sy == dy) && (sx <= dx))) {
 		saddr = WIDTH * (sy + height) + sx + width;
@@ -144,6 +133,7 @@ static void hitfb_fillrect(struct fb_info *p, const struct fb_fillrect *rect)
 	if (rect->rop != ROP_COPY)
 		cfb_fillrect(p, rect);
 	else {
+		hitfb_accel_wait();
 		fb_writew(0x00f0, HD64461_BBTROPR);
 		fb_writew(16, HD64461_BBTMDR);
 
@@ -159,16 +149,15 @@ static void hitfb_fillrect(struct fb_info *p, const struct fb_fillrect *rect)
 					     rect->height);
 			hitfb_accel_start(0);
 		}
-		hitfb_accel_wait();
 	}
 }
 
 static void hitfb_copyarea(struct fb_info *p, const struct fb_copyarea *area)
 {
+	hitfb_accel_wait();
 	hitfb_accel_bitblt(p->var.bits_per_pixel == 16, area->sx, area->sy,
 			   area->dx, area->dy, area->width, area->height,
 			   0x00cc, 0);
-	hitfb_accel_wait();
 }
 
 static int hitfb_pan_display(struct fb_var_screeninfo *var,
@@ -180,7 +169,7 @@ static int hitfb_pan_display(struct fb_var_screeninfo *var,
 	if (xoffset != 0)
 		return -EINVAL;
 
-	fb_writew(yoffset, HD64461_LCDCBAR);
+	fb_writew((yoffset*info->fix.line_length)>>10, HD64461_LCDCBAR);
 
 	return 0;
 }
@@ -206,13 +195,17 @@ int hitfb_blank(int blank_mode, struct fb_info *info)
 		v &= ~HD64461_STBCR_SLCDST;
 		fb_writew(v, HD64461_STBCR);
 
+		v = fb_readw(HD64461_LCDCCR);
+		v &= ~(HD64461_LCDCCR_MOFF | HD64461_LCDCCR_STREQ);
+		fb_writew(v, HD64461_LCDCCR);
+
+		do {
+		    v = fb_readw(HD64461_LCDCCR);
+		} while(v&HD64461_LCDCCR_STBACK);
+
 		v = fb_readw(HD64461_LDR1);
 		v |= HD64461_LDR1_DON;
 		fb_writew(v, HD64461_LDR1);
-
-		v = fb_readw(HD64461_LCDCCR);
-		v &= ~HD64461_LCDCCR_MOFF;
-		fb_writew(v, HD64461_LCDCCR);
 	}
 	return 0;
 }
@@ -220,7 +213,7 @@ int hitfb_blank(int blank_mode, struct fb_info *info)
 static int hitfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 			   unsigned blue, unsigned transp, struct fb_info *info)
 {
-	if (regno >= info->cmap.len)
+	if (regno >= 256)
 		return 1;
 
 	switch (info->var.bits_per_pixel) {
@@ -231,6 +224,8 @@ static int hitfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 		fb_writew(blue >> 10, HD64461_CPTWDR);
 		break;
 	case 16:
+		if (regno >= 16)
+			return 1;
 		((u32 *) (info->pseudo_palette))[regno] =
 		    ((red & 0xf800)) |
 		    ((green & 0xfc00) >> 5) | ((blue & 0xf800) >> 11);
@@ -239,26 +234,113 @@ static int hitfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 	return 0;
 }
 
+static int hitfb_sync(struct fb_info *info)
+{
+	hitfb_accel_wait();
+
+	return 0;
+}
+
+static int hitfb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
+{
+	int maxy;
+
+	var->xres = info->var.xres;
+	var->xres_virtual = info->var.xres;
+	var->yres = info->var.yres;
+
+	if ((var->bits_per_pixel != 8) && (var->bits_per_pixel != 16))
+		var->bits_per_pixel = info->var.bits_per_pixel;
+
+	if (var->yres_virtual < var->yres)
+		var->yres_virtual = var->yres;
+
+	maxy = info->fix.smem_len / var->xres;
+
+	if (var->bits_per_pixel == 16)
+		maxy /= 2;
+
+	if (var->yres_virtual > maxy)
+		var->yres_virtual = maxy;
+
+	var->xoffset = 0;
+	var->yoffset = 0;
+
+	switch (var->bits_per_pixel) {
+	case 8:
+		var->red.offset = 0;
+		var->red.length = 8;
+		var->green.offset = 0;
+		var->green.length = 8;
+		var->blue.offset = 0;
+		var->blue.length = 8;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+	case 16:		/* RGB 565 */
+		var->red.offset = 11;
+		var->red.length = 5;
+		var->green.offset = 5;
+		var->green.length = 6;
+		var->blue.offset = 0;
+		var->blue.length = 5;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+	}
+
+	return 0;
+}
+
+static int hitfb_set_par(struct fb_info *info)
+{
+	unsigned short ldr3;
+
+	switch (info->var.bits_per_pixel) {
+	case 8:
+		info->fix.line_length = info->var.xres;
+		info->fix.visual = FB_VISUAL_PSEUDOCOLOR;
+		info->fix.ypanstep = 16;
+		break;
+	case 16:
+		info->fix.line_length = info->var.xres*2;
+		info->fix.visual = FB_VISUAL_TRUECOLOR;
+		info->fix.ypanstep = 8;
+		break;
+	}
+
+	fb_writew(info->fix.line_length, HD64461_LCDCLOR);
+	ldr3 = fb_readw(HD64461_LDR3);
+	ldr3 &= ~15;
+	ldr3 |= (info->var.bits_per_pixel == 8) ? 4 : 8;
+	fb_writew(ldr3, HD64461_LDR3);
+	return 0;
+}
+
 static struct fb_ops hitfb_ops = {
 	.owner		= THIS_MODULE,
+	.fb_check_var	= hitfb_check_var,
+	.fb_set_par		= hitfb_set_par,
 	.fb_setcolreg	= hitfb_setcolreg,
 	.fb_blank	= hitfb_blank,
+	.fb_sync	= hitfb_sync,
 	.fb_pan_display = hitfb_pan_display,
 	.fb_fillrect	= hitfb_fillrect,
 	.fb_copyarea	= hitfb_copyarea,
 	.fb_imageblit	= cfb_imageblit,
 };
 
-int __init hitfb_init(void)
+static int __init hitfb_probe(struct platform_device *dev)
 {
 	unsigned short lcdclor, ldr3, ldvndr;
-	int size;
 
 	if (fb_get_options("hitfb", NULL))
 		return -ENODEV;
 
+	hitfb_fix.mmio_start = CONFIG_HD64461_IOBASE+0x1000;
+	hitfb_fix.mmio_len = 0x1000;
 	hitfb_fix.smem_start = CONFIG_HD64461_IOBASE + 0x02000000;
-	hitfb_fix.smem_len = (MACH_HP690) ? 1024 * 1024 : 512 * 1024;
+	hitfb_fix.smem_len = 512 * 1024;
 
 	lcdclor = fb_readw(HD64461_LCDCLOR);
 	ldvndr = fb_readw(HD64461_LDVNDR);
@@ -308,12 +390,12 @@ int __init hitfb_init(void)
 	fb_info.var = hitfb_var;
 	fb_info.fix = hitfb_fix;
 	fb_info.pseudo_palette = pseudo_palette;
-	fb_info.flags = FBINFO_DEFAULT;
+	fb_info.flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN |
+		FBINFO_HWACCEL_FILLRECT | FBINFO_HWACCEL_COPYAREA;
 
 	fb_info.screen_base = (void *)hitfb_fix.smem_start;
 
-	size = (fb_info.var.bits_per_pixel == 8) ? 256 : 16;
-	fb_alloc_cmap(&fb_info.cmap, size, 0);
+	fb_alloc_cmap(&fb_info.cmap, 256, 0);
 
 	if (register_framebuffer(&fb_info) < 0)
 		return -EINVAL;
@@ -323,9 +405,75 @@ int __init hitfb_init(void)
 	return 0;
 }
 
+static int __devexit hitfb_remove(struct platform_device *dev)
+{
+	return unregister_framebuffer(&fb_info);
+}
+
+#ifdef CONFIG_PM
+static int hitfb_suspend(struct platform_device *dev, pm_message_t state)
+{
+	u16 v;
+
+	hitfb_blank(1,0);
+	v = fb_readw(HD64461_STBCR);
+	v |= HD64461_STBCR_SLCKE_IST;
+	fb_writew(v, HD64461_STBCR);
+
+	return 0;
+}
+
+static int hitfb_resume(struct platform_device *dev)
+{
+	u16 v;
+
+	v = fb_readw(HD64461_STBCR);
+	v &= ~HD64461_STBCR_SLCKE_OST;
+	msleep(100);
+	v = fb_readw(HD64461_STBCR);
+	v &= ~HD64461_STBCR_SLCKE_IST;
+	fb_writew(v, HD64461_STBCR);
+	hitfb_blank(0,0);
+
+	return 0;
+}
+#endif
+
+static struct platform_driver hitfb_driver = {
+	.probe		= hitfb_probe,
+	.remove		= __devexit_p(hitfb_remove),
+#ifdef CONFIG_PM
+	.suspend	= hitfb_suspend,
+	.resume		= hitfb_resume,
+#endif
+	.driver		= {
+		.name	= "hitfb",
+	},
+};
+
+static struct platform_device hitfb_device = {
+	.name	= "hitfb",
+	.id	= -1,
+};
+
+static int __init hitfb_init(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&hitfb_driver);
+	if (!ret) {
+		ret = platform_device_register(&hitfb_device);
+		if (ret)
+			platform_driver_unregister(&hitfb_driver);
+	}
+	return ret;
+}
+
+
 static void __exit hitfb_exit(void)
 {
-	unregister_framebuffer(&fb_info);
+	platform_device_unregister(&hitfb_device);
+	platform_driver_unregister(&hitfb_driver);
 }
 
 module_init(hitfb_init);
-- 
cgit v1.2.1


From 7e27b9b720e74f471f0f0880c56578d07206c0af Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 13:48:32 +0900
Subject: sound: SH DAC audio driver updates.

Update the SH DAC audio driver for the clock framework.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 sound/oss/sh_dac_audio.c | 26 +++++++++++++++++++-------
 1 file changed, 19 insertions(+), 7 deletions(-)

diff --git a/sound/oss/sh_dac_audio.c b/sound/oss/sh_dac_audio.c
index ae7fccc510a1..83ff8a71f716 100644
--- a/sound/oss/sh_dac_audio.c
+++ b/sound/oss/sh_dac_audio.c
@@ -1,3 +1,14 @@
+/*
+ * sound/oss/sh_dac_audio.c
+ *
+ * SH DAC based sound :(
+ *
+ *  Copyright (C) 2004,2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/sched.h>
@@ -11,10 +22,12 @@
 #include <asm/uaccess.h>
 #include <asm/irq.h>
 #include <asm/delay.h>
+#include <asm/clock.h>
 #include <asm/cpu/dac.h>
+#include <asm/cpu/timer.h>
 #include <asm/machvec.h>
 #include <asm/hp6xx/hp6xx.h>
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 #define MODNAME "sh_dac_audio"
 
@@ -23,11 +36,6 @@
 #define TMU1_TCR_INIT	0x0020	/* Clock/4, rising edge; interrupt on */
 #define TMU1_TSTR_INIT  0x02	/* Bit to turn on TMU1 */
 
-#define TMU_TSTR	0xfffffe92
-#define TMU1_TCOR	0xfffffea0
-#define TMU1_TCNT	0xfffffea4
-#define TMU1_TCR	0xfffffea8
-
 #define BUFFER_SIZE 48000
 
 static int rate;
@@ -87,14 +95,18 @@ static void dac_audio_stop(void)
 		outw(v, HD64461_GPADR);
 	}
 
+ 	sh_dac_output(0, CONFIG_SOUND_SH_DAC_AUDIO_CHANNEL);
 	sh_dac_disable(CONFIG_SOUND_SH_DAC_AUDIO_CHANNEL);
 }
 
 static void dac_audio_set_rate(void)
 {
 	unsigned long interval;
+ 	struct clk *clk;
 
-	interval = (current_cpu_data.module_clock / 4) / rate;
+ 	clk = clk_get("module_clk");
+ 	interval = (clk_get_rate(clk) / 4) / rate;
+ 	clk_put(clk);
 	ctrl_outl(interval, TMU1_TCOR);
 	ctrl_outl(interval, TMU1_TCNT);
 }
-- 
cgit v1.2.1


From e8fb67f8e05bb1f4c07c3585967cfc6d44822ab0 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 13:56:28 +0900
Subject: sh: HS7751RVoIP board updates.

Various cleanups for HS7751RVoIP. Mostly just getting
rid of the old mach.c and splitting codec configuration
in to its own Kconfig.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                             |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/Kconfig  | 12 ++++++
 arch/sh/boards/renesas/hs7751rvoip/Makefile |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/mach.c   | 54 -------------------------
 arch/sh/boards/renesas/hs7751rvoip/setup.c  | 63 +++++++++++++++++++++++------
 include/asm-sh/hs7751rvoip/hs7751rvoip.h    |  6 +++
 6 files changed, 71 insertions(+), 68 deletions(-)
 create mode 100644 arch/sh/boards/renesas/hs7751rvoip/Kconfig
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/mach.c

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 1a0db1d4c952..cbf0d527e991 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -400,7 +400,7 @@ config SH_TMU
 
 endmenu
 
-#source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
+source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
 
 #source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Kconfig b/arch/sh/boards/renesas/hs7751rvoip/Kconfig
new file mode 100644
index 000000000000..1743be477be5
--- /dev/null
+++ b/arch/sh/boards/renesas/hs7751rvoip/Kconfig
@@ -0,0 +1,12 @@
+if SH_HS7751RVOIP
+
+menu "HS7751RVoIP options"
+
+config HS7751RVOIP_CODEC
+	bool "Support VoIP Codec section"
+	help
+	  Selecting this option will support CODEC section.
+
+endmenu
+
+endif
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
index 3ef8cbaa0b37..bede2d54fd6c 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ b/arch/sh/boards/renesas/hs7751rvoip/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the HS7751RVoIP specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
+obj-y	 := setup.o io.o irq.o led.o
 
 obj-$(CONFIG_PCI) += pci.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/mach.c b/arch/sh/boards/renesas/hs7751rvoip/mach.c
deleted file mode 100644
index caf967f77c61..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/mach.c
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_hs7751rvoip.c
- *
- * Minor tweak of mach_se.c file to reference hs7751rvoip-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Renesas Technology sales HS7751RVoIP
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/irq.h>
-#include <asm/hs7751rvoip/io.h>
-
-extern void init_hs7751rvoip_IRQ(void);
-extern void *hs7751rvoip_ioremap(unsigned long, unsigned long);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_hs7751rvoip __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= hs7751rvoip_inb,
-	.mv_inw			= hs7751rvoip_inw,
-	.mv_inl			= hs7751rvoip_inl,
-	.mv_outb		= hs7751rvoip_outb,
-	.mv_outw		= hs7751rvoip_outw,
-	.mv_outl		= hs7751rvoip_outl,
-
-	.mv_inb_p		= hs7751rvoip_inb_p,
-	.mv_inw_p		= hs7751rvoip_inw,
-	.mv_inl_p		= hs7751rvoip_inl,
-	.mv_outb_p		= hs7751rvoip_outb_p,
-	.mv_outw_p		= hs7751rvoip_outw,
-	.mv_outl_p		= hs7751rvoip_outl,
-
-	.mv_insb		= hs7751rvoip_insb,
-	.mv_insw		= hs7751rvoip_insw,
-	.mv_insl		= hs7751rvoip_insl,
-	.mv_outsb		= hs7751rvoip_outsb,
-	.mv_outsw		= hs7751rvoip_outsw,
-	.mv_outsl		= hs7751rvoip_outsl,
-
-	.mv_ioremap		= hs7751rvoip_ioremap,
-	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
-	.mv_init_irq		= init_hs7751rvoip_IRQ,
-};
-ALIAS_MV(hs7751rvoip)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index 7ce577822dc0..813fc4d5862a 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -17,14 +17,55 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/pm.h>
-#include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
-
-/* defined in mm/ioremap.c */
-extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
+#include <asm/hs7751rvoip/io.h>
+#include <asm/io.h>
+#include <asm/machvec.h>
+#include <asm/rtc.h>
+#include <asm/irq.h>
 
 unsigned int debug_counter;
 
+static void __init hs7751rvoip_init_irq(void)
+{
+#if defined(CONFIG_HS7751RVOIP_CODEC)
+	make_ipr_irq(DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+	make_ipr_irq(DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+#endif
+
+	init_hs7751rvoip_IRQ();
+}
+
+struct sh_machine_vector mv_hs7751rvoip __initmv = {
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= hs7751rvoip_inb,
+	.mv_inw			= hs7751rvoip_inw,
+	.mv_inl			= hs7751rvoip_inl,
+	.mv_outb		= hs7751rvoip_outb,
+	.mv_outw		= hs7751rvoip_outw,
+	.mv_outl		= hs7751rvoip_outl,
+
+	.mv_inb_p		= hs7751rvoip_inb_p,
+	.mv_inw_p		= hs7751rvoip_inw,
+	.mv_inl_p		= hs7751rvoip_inl,
+	.mv_outb_p		= hs7751rvoip_outb_p,
+	.mv_outw_p		= hs7751rvoip_outw,
+	.mv_outl_p		= hs7751rvoip_outl,
+
+	.mv_insb		= hs7751rvoip_insb,
+	.mv_insw		= hs7751rvoip_insw,
+	.mv_insl		= hs7751rvoip_insl,
+	.mv_outsb		= hs7751rvoip_outsb,
+	.mv_outsw		= hs7751rvoip_outsw,
+	.mv_outsl		= hs7751rvoip_outsl,
+
+	.mv_ioremap		= hs7751rvoip_ioremap,
+	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
+	.mv_init_irq		= hs7751rvoip_init_irq,
+};
+ALIAS_MV(hs7751rvoip)
+
 const char *get_system_type(void)
 {
 	return "HS7751RVoIP";
@@ -51,16 +92,15 @@ void *area6_io8_base;
 void *area5_io16_base;
 void *area6_io16_base;
 
-int __init cf_init(void)
+static int __init hs7751rvoip_cf_init(void)
 {
 	pgprot_t prot;
-	unsigned long paddrbase, psize;
+	unsigned long paddrbase;
 
 	/* open I/O area window */
 	paddrbase = virt_to_phys((void *)(PA_AREA5_IO+0x00000800));
-	psize = PAGE_SIZE;
 	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_COM16);
-	area5_io16_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area5_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area5_io16_base) {
 		printk("allocate_cf_area : can't open CF I/O window!\n");
 		return -ENOMEM;
@@ -69,19 +109,18 @@ int __init cf_init(void)
 	/* XXX : do we need attribute and common-memory area also? */
 
 	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
-	psize = PAGE_SIZE;
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_COM8);
 #else
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO8);
 #endif
-	area6_io8_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area6_io8_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area6_io8_base) {
 		printk("allocate_cf_area : can't open CODEC I/O 8bit window!\n");
 		return -ENOMEM;
 	}
 	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
-	area6_io16_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	area6_io16_base = p3_ioremap(paddrbase, PAGE_SIZE, prot.pgprot);
 	if (!area6_io16_base) {
 		printk("allocate_cf_area : can't open CODEC I/O 16bit window!\n");
 		return -ENOMEM;
@@ -90,4 +129,4 @@ int __init cf_init(void)
 	return 0;
 }
 
-__initcall (cf_init);
+__initcall(hs7751rvoip_cf_init);
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 5f995f937a44..857cc2ebbc69 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -44,4 +44,10 @@
 #define	IRQ_RINGING	4		/* Ringing IRQ */
 #define	IRQ_CODEC	5		/* CODEC IRQ */
 
+/* arch/sh/boards/renesas/hs7751rvoip/irq.c */
+void init_hs7751rvoip_IRQ(void);
+
+/* arch/sh/boards/renesas/hs7751rvoip/io.c */
+void *hs7751rvoip_ioremap(unsigned long, unsigned long);
+
 #endif  /* __ASM_SH_RENESAS_HS7751RVOIP */
-- 
cgit v1.2.1


From 36efc35447154317f9ffc5163a1793b5f7ff3de1 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:02:09 +0900
Subject: sh: RTS7751R2D board updates.

More of the same, trivial cleanups, and moving options to their
own board-specific Kconfig.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                             |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/Makefile |  2 +-
 arch/sh/boards/renesas/hs7751rvoip/led.c    | 26 --------
 arch/sh/boards/renesas/rts7751r2d/Kconfig   | 12 ++++
 arch/sh/boards/renesas/rts7751r2d/io.c      | 95 ++++++++++++++++-------------
 arch/sh/boards/renesas/rts7751r2d/led.c     |  9 ---
 arch/sh/boards/renesas/rts7751r2d/mach.c    |  3 +-
 7 files changed, 68 insertions(+), 81 deletions(-)
 delete mode 100644 arch/sh/boards/renesas/hs7751rvoip/led.c
 create mode 100644 arch/sh/boards/renesas/rts7751r2d/Kconfig

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index cbf0d527e991..a66b9c44075f 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -402,7 +402,7 @@ endmenu
 
 source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
 
-#source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
+source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 
 config SH_PCLK_FREQ
 	int "Peripheral clock frequency (in Hz)"
diff --git a/arch/sh/boards/renesas/hs7751rvoip/Makefile b/arch/sh/boards/renesas/hs7751rvoip/Makefile
index bede2d54fd6c..e626377c55ee 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/Makefile
+++ b/arch/sh/boards/renesas/hs7751rvoip/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the HS7751RVoIP specific parts of the kernel
 #
 
-obj-y	 := setup.o io.o irq.o led.o
+obj-y	 := setup.o io.o irq.o
 
 obj-$(CONFIG_PCI) += pci.o
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/led.c b/arch/sh/boards/renesas/hs7751rvoip/led.c
deleted file mode 100644
index b6608fff9f38..000000000000
--- a/arch/sh/boards/renesas/hs7751rvoip/led.c
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * linux/arch/sh/kernel/setup_hs7751rvoip.c
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Renesas Technology Sales HS7751RVoIP Support.
- *
- * Modified for HS7751RVoIP by
- * Atom Create Engineering Co., Ltd. 2002.
- * Lineo uSolutions, Inc. 2003.
- */
-
-#include <asm/io.h>
-#include <asm/hs7751rvoip/hs7751rvoip.h>
-
-extern unsigned int debug_counter;
-
-void debug_led_disp(void)
-{
-	unsigned short value;
-
-	value = (unsigned char)debug_counter++;
-	ctrl_outb((0xf0|value), PA_OUTPORTR);
-	if (value == 0x0f)
-		debug_counter = 0;
-}
diff --git a/arch/sh/boards/renesas/rts7751r2d/Kconfig b/arch/sh/boards/renesas/rts7751r2d/Kconfig
new file mode 100644
index 000000000000..7780d1fb13ff
--- /dev/null
+++ b/arch/sh/boards/renesas/rts7751r2d/Kconfig
@@ -0,0 +1,12 @@
+if SH_RTS7751R2D
+
+menu "RTS7751R2D options"
+
+config RTS7751R2D_REV11
+	bool "RTS7751R2D Rev. 1.1 board support"
+	help
+	  Selecting this option will support version rev. 1.1.
+endmenu
+
+endif
+
diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c
index 123abbbc91e0..9e7fa726a86d 100644
--- a/arch/sh/boards/renesas/rts7751r2d/io.c
+++ b/arch/sh/boards/renesas/rts7751r2d/io.c
@@ -13,13 +13,11 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <asm/io.h>
-#include <asm/rts7751r2d/rts7751r2d.h>
-#include <asm/addrspace.h>
-
-#include <linux/module.h>
 #include <linux/pci.h>
 #include "../../../drivers/pci/pci-sh7751.h"
+#include <asm/rts7751r2d/rts7751r2d.h>
+#include <asm/io.h>
+#include <asm/addrspace.h>
 
 /*
  * The 7751R RTS7751R2D uses the built-in PCI controller (PCIC)
@@ -35,10 +33,6 @@
 
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -52,7 +46,7 @@ static inline unsigned long port2adr(unsigned int port)
 		else
 			return (PA_AREA5_IO + 0x1000 + ((port-0x1f0) << 1));
 	else
-		maybebadio(port2adr, (unsigned long)port);
+		maybebadio((unsigned long)port);
 
 	return port;
 }
@@ -138,13 +132,13 @@ unsigned char rts7751r2d_inb_p(unsigned long port)
 unsigned short rts7751r2d_inw(unsigned long port)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(inw, port);
+		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned short *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
 		return *(volatile unsigned short *)PCI_IOMAP(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 
 	return 0;
 }
@@ -152,13 +146,13 @@ unsigned short rts7751r2d_inw(unsigned long port)
 unsigned int rts7751r2d_inl(unsigned long port)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(inl, port);
+		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned long *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
 		return *(volatile unsigned long *)PCI_IOMAP(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 
 	return 0;
 }
@@ -191,51 +185,55 @@ void rts7751r2d_outb_p(unsigned char value, unsigned long port)
 void rts7751r2d_outw(unsigned short value, unsigned long port)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(outw, port);
+		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned short *)port = value;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
 		*(volatile unsigned short *)PCI_IOMAP(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void rts7751r2d_outl(unsigned int value, unsigned long port)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(outl, port);
+		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned long *)port = value;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
 		*(volatile unsigned long *)PCI_IOMAP(port) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void rts7751r2d_insb(unsigned long port, void *addr, unsigned long count)
 {
+	unsigned long a = (unsigned long)addr;
 	volatile __u8 *bp;
 	volatile __u16 *p;
-	unsigned char *s = addr;
 
 	if (CHECK_AX88796L_PORT(port)) {
 		p = (volatile unsigned short *)port88796l(port, 0);
-		while (count--) *s++ = *p & 0xff;
+		while (count--)
+			ctrl_outb(*p & 0xff, a++);
 	} else if (PXSEG(port))
-		while (count--) *s++ = *(volatile unsigned char *)port;
+		while (count--)
+			ctrl_outb(ctrl_inb(port), a++);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
 		bp = (__u8 *)PCI_IOMAP(port);
-		while (count--) *s++ = *bp;
+		while (count--)
+			ctrl_outb(*bp, a++);
 	} else {
 		p = (volatile unsigned short *)port2adr(port);
-		while (count--) *s++ = *p & 0xff;
+		while (count--)
+			ctrl_outb(*p & 0xff, a++);
 	}
 }
 
 void rts7751r2d_insw(unsigned long port, void *addr, unsigned long count)
 {
+	unsigned long a = (unsigned long)addr;
 	volatile __u16 *p;
-	__u16 *s = addr;
 
 	if (CHECK_AX88796L_PORT(port))
 		p = (volatile unsigned short *)port88796l(port, 1);
@@ -245,46 +243,53 @@ void rts7751r2d_insw(unsigned long port, void *addr, unsigned long count)
 		p = (volatile unsigned short *)PCI_IOMAP(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *s++ = *p;
+	while (count--)
+		ctrl_outw(*p, a++);
 }
 
 void rts7751r2d_insl(unsigned long port, void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(insl, port);
+		maybebadio(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
-		__u32 *s = addr;
+		unsigned long a = (unsigned long)addr;
 
-		while (count--) *s++ = *p;
+		while (count--) {
+			ctrl_outl(ctrl_inl(PCI_IOMAP(port)), a);
+			a += 4;
+		}
 	} else
-		maybebadio(insl, port);
+		maybebadio(port);
 }
 
 void rts7751r2d_outsb(unsigned long port, const void *addr, unsigned long count)
 {
+	unsigned long a = (unsigned long)addr;
 	volatile __u8 *bp;
 	volatile __u16 *p;
-	const __u8 *s = addr;
 
 	if (CHECK_AX88796L_PORT(port)) {
 		p = (volatile unsigned short *)port88796l(port, 0);
-		while (count--) *p = *s++;
+		while (count--)
+			*p = ctrl_inb(a++);
 	} else if (PXSEG(port))
-		while (count--) *(volatile unsigned char *)port = *s++;
+		while (count--)
+			ctrl_outb(a++, port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
 		bp = (__u8 *)PCI_IOMAP(port);
-		while (count--) *bp = *s++;
+		while (count--)
+			*bp = ctrl_inb(a++);
 	} else {
 		p = (volatile unsigned short *)port2adr(port);
-		while (count--) *p = *s++;
+		while (count--)
+			*p = ctrl_inb(a++);
 	}
 }
 
 void rts7751r2d_outsw(unsigned long port, const void *addr, unsigned long count)
 {
+	unsigned long a = (unsigned long)addr;
 	volatile __u16 *p;
-	const __u16 *s = addr;
 
 	if (CHECK_AX88796L_PORT(port))
 		p = (volatile unsigned short *)port88796l(port, 1);
@@ -294,20 +299,26 @@ void rts7751r2d_outsw(unsigned long port, const void *addr, unsigned long count)
 		p = (volatile unsigned short *)PCI_IOMAP(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *p = *s++;
+
+	while (count--) {
+		ctrl_outw(*p, a);
+		a += 2;
+	}
 }
 
 void rts7751r2d_outsl(unsigned long port, const void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
-		maybebadio(outsl, port);
+		maybebadio(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
-		const __u32 *s = addr;
+		unsigned long a = (unsigned long)addr;
 
-		while (count--) *p = *s++;
+		while (count--) {
+			ctrl_outl(ctrl_inl(a), PCI_IOMAP(port));
+			a += 4;
+		}
 	} else
-		maybebadio(outsl, port);
+		maybebadio(port);
 }
 
 void *rts7751r2d_ioremap(unsigned long offset, unsigned long size)
diff --git a/arch/sh/boards/renesas/rts7751r2d/led.c b/arch/sh/boards/renesas/rts7751r2d/led.c
index cf35f90dfe03..e14a13d12d4a 100644
--- a/arch/sh/boards/renesas/rts7751r2d/led.c
+++ b/arch/sh/boards/renesas/rts7751r2d/led.c
@@ -53,12 +53,3 @@ void rts7751r2d_led(unsigned short value)
 	ctrl_outw(value, PA_OUTPORT);
 }
 
-void debug_led_disp(void)
-{
-	unsigned short value;
-
-	value = (unsigned short)debug_counter++;
-	rts7751r2d_led(value);
-	if (value == 0xff)
-		debug_counter = 0;
-}
diff --git a/arch/sh/boards/renesas/rts7751r2d/mach.c b/arch/sh/boards/renesas/rts7751r2d/mach.c
index 5ed9e97ea197..175a93d726e8 100644
--- a/arch/sh/boards/renesas/rts7751r2d/mach.c
+++ b/arch/sh/boards/renesas/rts7751r2d/mach.c
@@ -15,7 +15,7 @@
 #include <asm/machvec.h>
 #include <asm/rtc.h>
 #include <asm/irq.h>
-#include <asm/rts7751r2d/io.h>
+#include <asm/mach/rts7751r2d.h>
 
 extern void heartbeat_rts7751r2d(void);
 extern void init_rts7751r2d_IRQ(void);
@@ -54,7 +54,6 @@ struct sh_machine_vector mv_rts7751r2d __initmv = {
 	.mv_outsl		= rts7751r2d_outsl,
 
 	.mv_ioremap		= rts7751r2d_ioremap,
-	.mv_isa_port2addr	= rts7751r2d_isa_port2addr,
 	.mv_init_irq		= init_rts7751r2d_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_rts7751r2d,
-- 
cgit v1.2.1


From fdfc74f9fcebdda14609159d5010b758a9409acf Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:05:52 +0900
Subject: sh: Support for SH-4A memory barriers.

SH-4A supports 'synco' as a barrier, sprinkle it around
the cache ops as necessary..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c  | 5 +++++
 include/asm-sh/system.h | 7 +++++++
 2 files changed, 12 insertions(+)

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 94c05d09c3f7..846b63d6f5e8 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -184,6 +184,7 @@ void flush_cache_sigtramp(unsigned long addr)
 	     i++, index += cpu_data->icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
 	back_to_P1();
+	wmb();
 	local_irq_restore(flags);
 }
 
@@ -223,6 +224,8 @@ void flush_dcache_page(struct page *page)
 		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x2000, phys);
 		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x3000, phys);
 	}
+
+	wmb();
 }
 
 static inline void flush_icache_all(void)
@@ -247,6 +250,7 @@ void flush_dcache_all(void)
 		__flush_dcache_all();
 	else
 		__flush_dcache_all_ex();
+	wmb();
 }
 
 void flush_cache_all(void)
@@ -377,5 +381,6 @@ void flush_icache_user_range(struct vm_area_struct *vma,
 			     struct page *page, unsigned long addr, int len)
 {
 	flush_cache_page(vma, addr, page_to_pfn(page));
+	mb();
 }
 
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index e89728d405d8..eb4902ed920a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -84,10 +84,17 @@ static __inline__ unsigned long tas(volatile int *m)
 
 extern void __xchg_called_with_bad_pointer(void);
 
+#ifdef CONFIG_CPU_SH4A
+#define mb()	__asm__ __volatile__ ("synco": : :"memory")
+#define rmb()	mb()
+#define wmb()	__asm__ __volatile__ ("synco": : :"memory")
+#define read_barrier_depends()	do { } while(0)
+#else
 #define mb()	__asm__ __volatile__ ("": : :"memory")
 #define rmb()	mb()
 #define wmb()	__asm__ __volatile__ ("": : :"memory")
 #define read_barrier_depends()	do { } while(0)
+#endif
 
 #ifdef CONFIG_SMP
 #define smp_mb()	mb()
-- 
cgit v1.2.1


From b638d0b921dc95229af0dfd09cd24850336a2f75 Mon Sep 17 00:00:00 2001
From: Richard Curnow <richard.curnow@st.com>
Date: Wed, 27 Sep 2006 14:09:26 +0900
Subject: sh: Optimized cache handling for SH-4/SH-4A caches.

This reworks some of the SH-4 cache handling code to more easily
accomodate newer-style caches (particularly for the > direct-mapped
case), as well as optimizing some of the old code.

Signed-off-by: Richard Curnow <richard.curnow@st.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c      |  16 +-
 arch/sh/kernel/cpu/sh4/probe.c |  11 +
 arch/sh/mm/cache-sh4.c         | 517 ++++++++++++++++++++++++++++++++++-------
 arch/sh/mm/clear_page.S        |  99 +-------
 include/asm-sh/cache.h         |  22 +-
 5 files changed, 475 insertions(+), 190 deletions(-)

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 868e68b28880..731dd61419dd 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -4,6 +4,7 @@
  * CPU init code
  *
  * Copyright (C) 2002, 2003  Paul Mundt
+ * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -51,7 +52,15 @@ static void __init cache_init(void)
 	ccr = ctrl_inl(CCR);
 
 	/*
-	 * If the cache is already enabled .. flush it.
+	 * At this point we don't know whether the cache is enabled or not - a
+	 * bootloader may have enabled it.  There are at least 2 things that
+	 * could be dirty in the cache at this point:
+	 * 1. kernel command line set up by boot loader
+	 * 2. spilled registers from the prolog of this function
+	 * => before re-initialising the cache, we must do a purge of the whole
+	 * cache out to memory for safety.  As long as nothing is spilled
+	 * during the loop to lines that have already been done, this is safe.
+	 * - RPC
 	 */
 	if (ccr & CCR_CACHE_ENABLE) {
 		unsigned long ways, waysize, addrstart;
@@ -98,6 +107,8 @@ static void __init cache_init(void)
 	/* Force EMODE if possible */
 	if (cpu_data->dcache.ways > 1)
 		flags |= CCR_CACHE_EMODE;
+	else
+		flags &= ~CCR_CACHE_EMODE;
 #endif
 
 #ifdef CONFIG_SH_WRITETHROUGH
@@ -112,6 +123,9 @@ static void __init cache_init(void)
 	/* Turn on OCRAM -- halve the OC */
 	flags |= CCR_CACHE_ORA;
 	cpu_data->dcache.sets >>= 1;
+
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
 #endif
 
 	ctrl_outl(flags, CCR);
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 42427b79697b..1208da8fe5db 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -113,6 +113,11 @@ int __init detect_cpu_and_cache_system(void)
 		break;
 	}
 
+#ifdef CONFIG_SH_DIRECT_MAPPED
+	cpu_data->icache.ways = 1;
+	cpu_data->dcache.ways = 1;
+#endif
+
 	/*
 	 * On anything that's not a direct-mapped cache, look to the CVR
 	 * for I/D-cache specifics.
@@ -125,6 +130,9 @@ int __init detect_cpu_and_cache_system(void)
 			(cpu_data->icache.way_incr - (1 << 5));
 	}
 
+	cpu_data->icache.way_size = cpu_data->icache.sets *
+				    cpu_data->icache.linesz;
+
 	if (cpu_data->dcache.ways > 1) {
 		size = sizes[(cvr >> 16) & 0xf];
 		cpu_data->dcache.way_incr	= (size >> 1);
@@ -133,6 +141,9 @@ int __init detect_cpu_and_cache_system(void)
 			(cpu_data->dcache.way_incr - (1 << 5));
 	}
 
+	cpu_data->dcache.way_size = cpu_data->dcache.sets *
+				    cpu_data->dcache.linesz;
+
 	return 0;
 }
 
diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 846b63d6f5e8..c036c2b4ac2b 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -25,28 +25,95 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
-extern void __flush_cache_4096(unsigned long addr, unsigned long phys,
+static void __flush_dcache_segment_1way(unsigned long start,
+					unsigned long extent);
+static void __flush_dcache_segment_2way(unsigned long start,
+					unsigned long extent);
+static void __flush_dcache_segment_4way(unsigned long start,
+					unsigned long extent);
+
+static void __flush_cache_4096(unsigned long addr, unsigned long phys,
 			       unsigned long exec_offset);
-extern void __flush_cache_4096_all(unsigned long start);
-static void __flush_cache_4096_all_ex(unsigned long start);
-extern void __flush_dcache_all(void);
-static void __flush_dcache_all_ex(void);
+
+/*
+ * This is initialised here to ensure that it is not placed in the BSS.  If
+ * that were to happen, note that cache_init gets called before the BSS is
+ * cleared, so this would get nulled out which would be hopeless.
+ */
+static void (*__flush_dcache_segment_fn)(unsigned long, unsigned long) =
+	(void (*)(unsigned long, unsigned long))0xdeadbeef;
+
+static void compute_alias(struct cache_info *c)
+{
+	c->alias_mask = ((c->sets - 1) << c->entry_shift) & ~(PAGE_SIZE - 1);
+	c->n_aliases = (c->alias_mask >> PAGE_SHIFT) + 1;
+}
+
+static void __init emit_cache_params(void)
+{
+	printk("PVR=%08x CVR=%08x PRR=%08x\n",
+		ctrl_inl(CCN_PVR),
+		ctrl_inl(CCN_CVR),
+		ctrl_inl(CCN_PRR));
+	printk("I-cache : n_ways=%d n_sets=%d way_incr=%d\n",
+		cpu_data->icache.ways,
+		cpu_data->icache.sets,
+		cpu_data->icache.way_incr);
+	printk("I-cache : entry_mask=0x%08x alias_mask=0x%08x n_aliases=%d\n",
+		cpu_data->icache.entry_mask,
+		cpu_data->icache.alias_mask,
+		cpu_data->icache.n_aliases);
+	printk("D-cache : n_ways=%d n_sets=%d way_incr=%d\n",
+		cpu_data->dcache.ways,
+		cpu_data->dcache.sets,
+		cpu_data->dcache.way_incr);
+	printk("D-cache : entry_mask=0x%08x alias_mask=0x%08x n_aliases=%d\n",
+		cpu_data->dcache.entry_mask,
+		cpu_data->dcache.alias_mask,
+		cpu_data->dcache.n_aliases);
+
+	if (!__flush_dcache_segment_fn)
+		panic("unknown number of cache ways\n");
+}
 
 /*
  * SH-4 has virtually indexed and physically tagged cache.
  */
 
-struct semaphore p3map_sem[4];
+/* Worst case assumed to be 64k cache, direct-mapped i.e. 4 synonym bits. */
+#define MAX_P3_SEMAPHORES 16
+
+struct semaphore p3map_sem[MAX_P3_SEMAPHORES];
 
 void __init p3_cache_init(void)
 {
-	if (remap_area_pages(P3SEG, 0, PAGE_SIZE*4, _PAGE_CACHABLE))
+	int i;
+
+	compute_alias(&cpu_data->icache);
+	compute_alias(&cpu_data->dcache);
+
+	switch (cpu_data->dcache.ways) {
+	case 1:
+		__flush_dcache_segment_fn = __flush_dcache_segment_1way;
+		break;
+	case 2:
+		__flush_dcache_segment_fn = __flush_dcache_segment_2way;
+		break;
+	case 4:
+		__flush_dcache_segment_fn = __flush_dcache_segment_4way;
+		break;
+	default:
+		__flush_dcache_segment_fn = NULL;
+		break;
+	}
+
+	emit_cache_params();
+
+	if (remap_area_pages(P3SEG, 0, PAGE_SIZE * 4, _PAGE_CACHABLE))
 		panic("%s failed.", __FUNCTION__);
 
-	sema_init (&p3map_sem[0], 1);
-	sema_init (&p3map_sem[1], 1);
-	sema_init (&p3map_sem[2], 1);
-	sema_init (&p3map_sem[3], 1);
+	for (i = 0; i < cpu_data->dcache.n_aliases; i++)
+		sema_init(&p3map_sem[i], 1);
 }
 
 /*
@@ -91,7 +158,6 @@ void __flush_purge_region(void *start, int size)
 	}
 }
 
-
 /*
  * No write back please
  */
@@ -110,46 +176,6 @@ void __flush_invalidate_region(void *start, int size)
 	}
 }
 
-static void __flush_dcache_all_ex(void)
-{
-	unsigned long addr, end_addr, entry_offset;
-
-	end_addr = CACHE_OC_ADDRESS_ARRAY +
-		(cpu_data->dcache.sets << cpu_data->dcache.entry_shift) *
-		 cpu_data->dcache.ways;
-
-	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (addr = CACHE_OC_ADDRESS_ARRAY;
-	     addr < end_addr;
-	     addr += entry_offset) {
-		ctrl_outl(0, addr);
-	}
-}
-
-static void __flush_cache_4096_all_ex(unsigned long start)
-{
-	unsigned long addr, entry_offset;
-	int i;
-
-	entry_offset = 1 << cpu_data->dcache.entry_shift;
-	for (i = 0; i < cpu_data->dcache.ways;
-	     i++, start += cpu_data->dcache.way_incr) {
-		for (addr = CACHE_OC_ADDRESS_ARRAY + start;
-		     addr < CACHE_OC_ADDRESS_ARRAY + 4096 + start;
-		     addr += entry_offset) {
-			ctrl_outl(0, addr);
-		}
-	}
-}
-
-void flush_cache_4096_all(unsigned long start)
-{
-	if (cpu_data->dcache.ways == 1)
-		__flush_cache_4096_all(start);
-	else
-		__flush_cache_4096_all_ex(start);
-}
-
 /*
  * Write back the range of D-cache, and purge the I-cache.
  *
@@ -180,9 +206,11 @@ void flush_cache_sigtramp(unsigned long addr)
 
 	local_irq_save(flags);
 	jump_to_P2();
+
 	for (i = 0; i < cpu_data->icache.ways;
 	     i++, index += cpu_data->icache.way_incr)
 		ctrl_outl(0, index);	/* Clear out Valid-bit */
+
 	back_to_P1();
 	wmb();
 	local_irq_restore(flags);
@@ -194,8 +222,8 @@ static inline void flush_cache_4096(unsigned long start,
 	unsigned long flags;
 
 	/*
-	 * SH7751, SH7751R, and ST40 have no restriction to handle cache.
-	 * (While SH7750 must do that at P2 area.)
+	 * All types of SH-4 require PC to be in P2 to operate on the I-cache.
+	 * Some types of SH-4 require PC to be in P2 to operate on the D-cache.
 	 */
 	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG)
 	   || start < CACHE_OC_ADDRESS_ARRAY) {
@@ -217,12 +245,13 @@ void flush_dcache_page(struct page *page)
 {
 	if (test_bit(PG_mapped, &page->flags)) {
 		unsigned long phys = PHYSADDR(page_address(page));
+		unsigned long addr = CACHE_OC_ADDRESS_ARRAY;
+		int i, n;
 
 		/* Loop all the D-cache */
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY,          phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x1000, phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x2000, phys);
-		flush_cache_4096(CACHE_OC_ADDRESS_ARRAY | 0x3000, phys);
+		n = cpu_data->dcache.n_aliases;
+		for (i = 0; i < n; i++, addr += PAGE_SIZE)
+			flush_cache_4096(addr, phys);
 	}
 
 	wmb();
@@ -246,10 +275,7 @@ static inline void flush_icache_all(void)
 
 void flush_dcache_all(void)
 {
-	if (cpu_data->dcache.ways == 1)
-		__flush_dcache_all();
-	else
-		__flush_dcache_all_ex();
+	(*__flush_dcache_segment_fn)(0UL, cpu_data->dcache.way_size);
 	wmb();
 }
 
@@ -261,6 +287,16 @@ void flush_cache_all(void)
 
 void flush_cache_mm(struct mm_struct *mm)
 {
+	/*
+	 * Note : (RPC) since the caches are physically tagged, the only point
+	 * of flush_cache_mm for SH-4 is to get rid of aliases from the
+	 * D-cache.  The assumption elsewhere, e.g. flush_cache_range, is that
+	 * lines can stay resident so long as the virtual address they were
+	 * accessed with (hence cache set) is in accord with the physical
+	 * address (i.e. tag).  It's no different here.  So I reckon we don't
+	 * need to flush the I-cache, since aliases don't matter for that.  We
+	 * should try that.
+	 */
 	flush_cache_all();
 }
 
@@ -273,24 +309,36 @@ void flush_cache_mm(struct mm_struct *mm)
 void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigned long pfn)
 {
 	unsigned long phys = pfn << PAGE_SHIFT;
+	unsigned int alias_mask;
+
+	alias_mask = cpu_data->dcache.alias_mask;
 
 	/* We only need to flush D-cache when we have alias */
-	if ((address^phys) & CACHE_ALIAS) {
+	if ((address^phys) & alias_mask) {
 		/* Loop 4K of the D-cache */
 		flush_cache_4096(
-			CACHE_OC_ADDRESS_ARRAY | (address & CACHE_ALIAS),
+			CACHE_OC_ADDRESS_ARRAY | (address & alias_mask),
 			phys);
 		/* Loop another 4K of the D-cache */
 		flush_cache_4096(
-			CACHE_OC_ADDRESS_ARRAY | (phys & CACHE_ALIAS),
+			CACHE_OC_ADDRESS_ARRAY | (phys & alias_mask),
 			phys);
 	}
 
-	if (vma->vm_flags & VM_EXEC)
-		/* Loop 4K (half) of the I-cache */
+	alias_mask = cpu_data->icache.alias_mask;
+	if (vma->vm_flags & VM_EXEC) {
+		/*
+		 * Evict entries from the portion of the cache from which code
+		 * may have been executed at this address (virtual).  There's
+		 * no need to evict from the portion corresponding to the
+		 * physical address as for the D-cache, because we know the
+		 * kernel has never executed the code through its identity
+		 * translation.
+		 */
 		flush_cache_4096(
-			CACHE_IC_ADDRESS_ARRAY | (address & 0x1000),
+			CACHE_IC_ADDRESS_ARRAY | (address & alias_mask),
 			phys);
+	}
 }
 
 /*
@@ -305,14 +353,28 @@ void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigne
 void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 		       unsigned long end)
 {
-	unsigned long p = start & PAGE_MASK;
+	unsigned long d = 0, p = start & PAGE_MASK;
+	unsigned long alias_mask = cpu_data->dcache.alias_mask;
+	unsigned long n_aliases = cpu_data->dcache.n_aliases;
+	unsigned long select_bit;
+	unsigned long all_aliases_mask;
+	unsigned long addr_offset;
+	unsigned long phys;
 	pgd_t *dir;
 	pmd_t *pmd;
 	pud_t *pud;
 	pte_t *pte;
 	pte_t entry;
-	unsigned long phys;
-	unsigned long d = 0;
+	int i;
+
+	/*
+	 * If cache is only 4k-per-way, there are never any 'aliases'.  Since
+	 * the cache is physically tagged, the data can just be left in there.
+	 */
+	if (n_aliases == 0)
+		return;
+
+	all_aliases_mask = (1 << n_aliases) - 1;
 
 	/*
 	 * Don't bother with the lookup and alias check if we have a
@@ -335,39 +397,52 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 
 	do {
 		if (pmd_none(*pmd) || pmd_bad(*pmd)) {
-			p &= ~((1 << PMD_SHIFT) -1);
+			p &= ~((1 << PMD_SHIFT) - 1);
 			p += (1 << PMD_SHIFT);
 			pmd++;
+
 			continue;
 		}
+
 		pte = pte_offset_kernel(pmd, p);
+
 		do {
 			entry = *pte;
+
 			if ((pte_val(entry) & _PAGE_PRESENT)) {
-				phys = pte_val(entry)&PTE_PHYS_MASK;
-				if ((p^phys) & CACHE_ALIAS) {
-					d |= 1 << ((p & CACHE_ALIAS)>>12);
-					d |= 1 << ((phys & CACHE_ALIAS)>>12);
-					if (d == 0x0f)
+				phys = pte_val(entry) & PTE_PHYS_MASK;
+
+				if ((p ^ phys) & alias_mask) {
+					d |= 1 << ((p & alias_mask) >> PAGE_SHIFT);
+					d |= 1 << ((phys & alias_mask) >> PAGE_SHIFT);
+
+					if (d == all_aliases_mask)
 						goto loop_exit;
 				}
 			}
+
 			pte++;
 			p += PAGE_SIZE;
 		} while (p < end && ((unsigned long)pte & ~PAGE_MASK));
 		pmd++;
 	} while (p < end);
- loop_exit:
-	if (d & 1)
-		flush_cache_4096_all(0);
-	if (d & 2)
-		flush_cache_4096_all(0x1000);
-	if (d & 4)
-		flush_cache_4096_all(0x2000);
-	if (d & 8)
-		flush_cache_4096_all(0x3000);
-	if (vma->vm_flags & VM_EXEC)
+
+loop_exit:
+	for (i = 0, select_bit = 0x1, addr_offset = 0x0; i < n_aliases;
+	     i++, select_bit <<= 1, addr_offset += PAGE_SIZE)
+		if (d & select_bit) {
+			(*__flush_dcache_segment_fn)(addr_offset, PAGE_SIZE);
+			wmb();
+		}
+
+	if (vma->vm_flags & VM_EXEC) {
+		/*
+		 * TODO: Is this required???  Need to look at how I-cache
+		 * coherency is assured when new programs are loaded to see if
+		 * this matters.
+		 */
 		flush_icache_all();
+	}
 }
 
 /*
@@ -384,3 +459,271 @@ void flush_icache_user_range(struct vm_area_struct *vma,
 	mb();
 }
 
+/**
+ * __flush_cache_4096
+ *
+ * @addr:  address in memory mapped cache array
+ * @phys:  P1 address to flush (has to match tags if addr has 'A' bit
+ *         set i.e. associative write)
+ * @exec_offset: set to 0x20000000 if flush has to be executed from P2
+ *               region else 0x0
+ *
+ * The offset into the cache array implied by 'addr' selects the
+ * 'colour' of the virtual address range that will be flushed.  The
+ * operation (purge/write-back) is selected by the lower 2 bits of
+ * 'phys'.
+ */
+static void __flush_cache_4096(unsigned long addr, unsigned long phys,
+			       unsigned long exec_offset)
+{
+	int way_count;
+	unsigned long base_addr = addr;
+	struct cache_info *dcache;
+	unsigned long way_incr;
+	unsigned long a, ea, p;
+	unsigned long temp_pc;
+
+	dcache = &cpu_data->dcache;
+	/* Write this way for better assembly. */
+	way_count = dcache->ways;
+	way_incr = dcache->way_incr;
+
+	/*
+	 * Apply exec_offset (i.e. branch to P2 if required.).
+	 *
+	 * FIXME:
+	 *
+	 *	If I write "=r" for the (temp_pc), it puts this in r6 hence
+	 *	trashing exec_offset before it's been added on - why?  Hence
+	 *	"=&r" as a 'workaround'
+	 */
+	asm volatile("mov.l 1f, %0\n\t"
+		     "add   %1, %0\n\t"
+		     "jmp   @%0\n\t"
+		     "nop\n\t"
+		     ".balign 4\n\t"
+		     "1:  .long 2f\n\t"
+		     "2:\n" : "=&r" (temp_pc) : "r" (exec_offset));
+
+	/*
+	 * We know there will be >=1 iteration, so write as do-while to avoid
+	 * pointless nead-of-loop check for 0 iterations.
+	 */
+	do {
+		ea = base_addr + PAGE_SIZE;
+		a = base_addr;
+		p = phys;
+
+		do {
+			*(volatile unsigned long *)a = p;
+			/*
+			 * Next line: intentionally not p+32, saves an add, p
+			 * will do since only the cache tag bits need to
+			 * match.
+			 */
+			*(volatile unsigned long *)(a+32) = p;
+			a += 64;
+			p += 64;
+		} while (a < ea);
+
+		base_addr += way_incr;
+	} while (--way_count != 0);
+}
+
+/*
+ * Break the 1, 2 and 4 way variants of this out into separate functions to
+ * avoid nearly all the overhead of having the conditional stuff in the function
+ * bodies (+ the 1 and 2 way cases avoid saving any registers too).
+ */
+static void __flush_dcache_segment_1way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/*
+	 * The previous code aligned base_addr to 16k, i.e. the way_size of all
+	 * existing SH-4 D-caches.  Whilst I don't see a need to have this
+	 * aligned to any better than the cache line size (which it will be
+	 * anyway by construction), let's align it to at least the way_size of
+	 * any existing or conceivable SH-4 D-cache.  -- RPC
+	 */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		a0 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "ocbi @%0" : : "r" (a0));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+	} while (a0 < a0e);
+}
+
+static void __flush_dcache_segment_2way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a1, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/* See comment under 1-way above */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a1 = a0 + way_incr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		a0 += linesz;
+		a1 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1" : :
+			     "r" (a0), "r" (a1));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+		a1 += linesz;
+	} while (a0 < a0e);
+}
+
+static void __flush_dcache_segment_4way(unsigned long start,
+					unsigned long extent_per_way)
+{
+	unsigned long orig_sr, sr_with_bl;
+	unsigned long base_addr;
+	unsigned long way_incr, linesz, way_size;
+	struct cache_info *dcache;
+	register unsigned long a0, a1, a2, a3, a0e;
+
+	asm volatile("stc sr, %0" : "=r" (orig_sr));
+	sr_with_bl = orig_sr | (1<<28);
+	base_addr = ((unsigned long)&empty_zero_page[0]);
+
+	/* See comment under 1-way above */
+	base_addr = ((base_addr >> 16) << 16);
+	base_addr |= start;
+
+	dcache = &cpu_data->dcache;
+	linesz = dcache->linesz;
+	way_incr = dcache->way_incr;
+	way_size = dcache->way_size;
+
+	a0 = base_addr;
+	a1 = a0 + way_incr;
+	a2 = a1 + way_incr;
+	a3 = a2 + way_incr;
+	a0e = base_addr + extent_per_way;
+	do {
+		asm volatile("ldc %0, sr" : : "r" (sr_with_bl));
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+		asm volatile("movca.l r0, @%0\n\t"
+			     "movca.l r0, @%1\n\t"
+			     "movca.l r0, @%2\n\t"
+			     "movca.l r0, @%3\n\t"
+			     "ocbi @%0\n\t"
+			     "ocbi @%1\n\t"
+			     "ocbi @%2\n\t"
+			     "ocbi @%3\n\t" : :
+			     "r" (a0), "r" (a1), "r" (a2), "r" (a3));
+		asm volatile("ldc %0, sr" : : "r" (orig_sr));
+		a0 += linesz;
+		a1 += linesz;
+		a2 += linesz;
+		a3 += linesz;
+	} while (a0 < a0e);
+}
+
diff --git a/arch/sh/mm/clear_page.S b/arch/sh/mm/clear_page.S
index 08acead7b2a1..7b96425ae270 100644
--- a/arch/sh/mm/clear_page.S
+++ b/arch/sh/mm/clear_page.S
@@ -193,102 +193,5 @@ ENTRY(__clear_user_page)
 	 nop
 .L4096:	.word	4096
 
-ENTRY(__flush_cache_4096)
-	mov.l	1f,r3
-	add	r6,r3
-	mov	r4,r0
-	mov	#64,r2
-	shll	r2
-	mov	#64,r6
-	jmp	@r3
-	 mov	#96,r7
-	.align	2
-1:	.long	2f
-2:
-	.rept	32
-	mov.l	r5,@r0
-	mov.l	r5,@(32,r0)
-	mov.l	r5,@(r0,r6)
-	mov.l	r5,@(r0,r7)
-	add	r2,r5
-	add	r2,r0
-	.endr
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	nop
-	rts
-	 nop
-
-ENTRY(__flush_dcache_all)
-	mov.l	2f,r0
-	mov.l	3f,r4
-	and	r0,r4		! r4 = (unsigned long)&empty_zero_page[0] & ~0xffffc000
-	stc	sr,r1		! save SR
-	mov.l	4f,r2
-	or	r1,r2
-	mov	#32,r3
-	shll2	r3
-1:
-	ldc	r2,sr		! set BL bit
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	ldc	r1,sr		! restore SR
-	dt	r3
-	bf/s	1b
-	 add	#32,r4
-
-	rts
-	 nop
-	.align	2
-2:	.long	0xffffc000
-3:	.long	empty_zero_page
-4:	.long	0x10000000	! BL bit
-
-/* __flush_cache_4096_all(unsigned long addr) */
-ENTRY(__flush_cache_4096_all)
-	mov.l	2f,r0
-	mov.l	3f,r2
-	and	r0,r2
-	or	r2,r4		! r4 = addr | (unsigned long)&empty_zero_page[0] & ~0x3fff
-	stc	sr,r1		! save SR
-	mov.l	4f,r2
-	or	r1,r2
-	mov	#32,r3
-1:
-	ldc	r2,sr		! set BL bit
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	add	#32,r4
-	movca.l	r0,@r4
-	ocbi	@r4
-	ldc	r1,sr		! restore SR
-	dt	r3
-	bf/s	1b
-	 add	#32,r4
-
-	rts
-	 nop
-	.align	2
-2:	.long	0xffffc000
-3:	.long	empty_zero_page
-4:	.long	0x10000000	! BL bit
 #endif
+
diff --git a/include/asm-sh/cache.h b/include/asm-sh/cache.h
index 656fdfe9e8b4..33f13367054b 100644
--- a/include/asm-sh/cache.h
+++ b/include/asm-sh/cache.h
@@ -23,15 +23,29 @@
 #define L1_CACHE_ALIGN(x)	(((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1))
 
 struct cache_info {
-	unsigned int ways;
-	unsigned int sets;
-	unsigned int linesz;
+	unsigned int ways;		/* Number of cache ways */
+	unsigned int sets;		/* Number of cache sets */
+	unsigned int linesz;		/* Cache line size (bytes) */
 
-	unsigned int way_incr;
+	unsigned int way_size;		/* sets * line size */
 
+	/*
+	 * way_incr is the address offset for accessing the next way
+	 * in memory mapped cache array ops.
+	 */
+	unsigned int way_incr;
 	unsigned int entry_shift;
 	unsigned int entry_mask;
 
+	/*
+	 * Compute a mask which selects the address bits which overlap between
+	 * 1. those used to select the cache set during indexing
+	 * 2. those in the physical page number.
+	 */
+	unsigned int alias_mask;
+
+	unsigned int n_aliases;		/* Number of aliases */
+
 	unsigned long flags;
 };
 
-- 
cgit v1.2.1


From 73388cc7c648861742e584a97fbffed16afc7dc3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:11:33 +0900
Subject: sh: Refactor PRR masking to catch newer SH7760 cuts.

Newer SH7760 cuts have a range of acceptable PRR values..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 1208da8fe5db..89986e70d041 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -99,9 +99,15 @@ int __init detect_cpu_and_cache_system(void)
 		break;
 	case 0x500 ... 0x501:
 		switch (prr) {
-		    case 0x10: cpu_data->type = CPU_SH7750R; break;
-		    case 0x11: cpu_data->type = CPU_SH7751R; break;
-		    case 0x50: cpu_data->type = CPU_SH7760;  break;
+		case 0x10:
+			cpu_data->type = CPU_SH7750R;
+			break;
+		case 0x11:
+			cpu_data->type = CPU_SH7751R;
+			break;
+		case 0x50 ... 0x5f:
+			cpu_data->type = CPU_SH7760;
+			break;
 		}
 
 		cpu_data->icache.ways = 2;
-- 
cgit v1.2.1


From e7be853df79fe8ae08ba7d933bd21e1dbb0db7bc Mon Sep 17 00:00:00 2001
From: Toshinobu Sugioka <sugioka@itonet.co.jp>
Date: Wed, 27 Sep 2006 14:13:14 +0900
Subject: sh: Fix a sign extension bug in memset().

Minor sign-extension bug in SH-specific memset()..

Signed-off-by: Toshinobu Sugioka <sugioka@itonet.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/lib/memset.S | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sh/lib/memset.S b/arch/sh/lib/memset.S
index 95670090680e..af91fe2b72a6 100644
--- a/arch/sh/lib/memset.S
+++ b/arch/sh/lib/memset.S
@@ -29,6 +29,7 @@ ENTRY(memset)
 	bf/s	1b
 	 mov.b	r5,@-r4
 2:				! make VVVV
+	extu.b	r5,r5
 	swap.b	r5,r0		!   V0
 	or	r0,r5		!   VV
 	swap.w	r5,r0		! VV00
-- 
cgit v1.2.1


From e86d6b66f5b38680746b2cb71186d90af17f150f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:20:54 +0900
Subject: sh: prefetch()/prefetchw() support.

SH-2/3/4 are able to prefetch, add support for it..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/processor.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index eeb0f48bb99e..e99aff706cf7 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -266,5 +266,18 @@ extern unsigned long get_wchan(struct task_struct *p);
 #define cpu_sleep()	__asm__ __volatile__ ("sleep" : : : "memory")
 #define cpu_relax()	barrier()
 
+#if defined(CONFIG_CPU_SH2A) || defined(CONFIG_CPU_SH3) || \
+    defined(CONFIG_CPU_SH4)
+#define PREFETCH_STRIDE		L1_CACHE_BYTES
+#define ARCH_HAS_PREFETCH
+#define ARCH_HAS_PREFETCHW
+static inline void prefetch(void *x)
+{
+	__asm__ __volatile__ ("pref @%0\n\t" : : "r" (x) : "memory");
+}
+
+#define prefetchw(x)	prefetch(x)
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_H */
-- 
cgit v1.2.1


From a80fd21e52cc09ff1e4d36de5781173a5b87b2dc Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:26:53 +0900
Subject: sh: earlyprintk= support and cleanups.

Allow multiple early printk consoles via earlyprintk=.

With this change earlyprintk is no longer enabled by default,
it must be specified on the kernel command line. Optionally
with ,keep to prevent unreg by tty_io.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/early_printk.c | 106 +++++++++++++++++++++++++++++++-----------
 arch/sh/kernel/setup.c        |  22 +++++++--
 include/asm-sh/setup.h        |   2 +
 3 files changed, 100 insertions(+), 30 deletions(-)

diff --git a/arch/sh/kernel/early_printk.c b/arch/sh/kernel/early_printk.c
index 1378db375e17..a00022722e9e 100644
--- a/arch/sh/kernel/early_printk.c
+++ b/arch/sh/kernel/early_printk.c
@@ -3,7 +3,7 @@
  *
  *  Copyright (C) 1999, 2000  Niibe Yutaka
  *  Copyright (C) 2002  M. R. Brown
- *  Copyright (C) 2004  Paul Mundt
+ *  Copyright (C) 2004 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -49,7 +49,7 @@ static int __init sh_console_setup(struct console *co, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console bios_console = {
 	.name		= "bios",
 	.write		= sh_console_write,
 	.setup		= sh_console_setup,
@@ -59,34 +59,43 @@ static struct console early_console = {
 #endif
 
 #ifdef CONFIG_EARLY_SCIF_CONSOLE
+#include <linux/serial_core.h>
+#include "../../../drivers/serial/sh-sci.h"
+
+#ifdef CONFIG_CPU_SH4
 #define SCIF_REG	0xffe80000
+#elif defined(CONFIG_CPU_SUBTYPE_SH72060)
+#define SCIF_REG	0xfffe9800
+#else
+#error "Undefined SCIF for this subtype"
+#endif
+
+static struct uart_port scif_port = {
+	.mapbase	= SCIF_REG,
+	.membase	= (char __iomem *)SCIF_REG,
+};
 
 static void scif_sercon_putc(int c)
 {
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x20)) ;
+	while (((sci_in(&scif_port, SCFDR) & 0x1f00 >> 8) == 16))
+		;
 
-	ctrl_outb(c, SCIF_REG + 12);
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0x9f), SCIF_REG + 0x10);
+	sci_out(&scif_port, SCxTDR, c);
+	sci_in(&scif_port, SCxSR);
+	sci_out(&scif_port, SCxSR, 0xf3 & ~(0x20 | 0x40));
+
+	while ((sci_in(&scif_port, SCxSR) & 0x40) == 0);
+		;
 
 	if (c == '\n')
 		scif_sercon_putc('\r');
 }
 
-static void scif_sercon_flush(void)
-{
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-
-	while (!(ctrl_inw(SCIF_REG + 0x10) & 0x40)) ;
-
-	ctrl_outw((ctrl_inw(SCIF_REG + 0x10) & 0xbf), SCIF_REG + 0x10);
-}
-
-static void scif_sercon_write(struct console *con, const char *s, unsigned count)
+static void scif_sercon_write(struct console *con, const char *s,
+			      unsigned count)
 {
 	while (count-- > 0)
 		scif_sercon_putc(*s++);
-
-	scif_sercon_flush();
 }
 
 static int __init scif_sercon_setup(struct console *con, char *options)
@@ -96,7 +105,7 @@ static int __init scif_sercon_setup(struct console *con, char *options)
 	return 0;
 }
 
-static struct console early_console = {
+static struct console scif_console = {
 	.name		= "sercon",
 	.write		= scif_sercon_write,
 	.setup		= scif_sercon_setup,
@@ -104,7 +113,7 @@ static struct console early_console = {
 	.index		= -1,
 };
 
-void scif_sercon_init(int baud)
+static void scif_sercon_init(int baud)
 {
 	ctrl_outw(0, SCIF_REG + 8);
 	ctrl_outw(0, SCIF_REG);
@@ -122,16 +131,61 @@ void scif_sercon_init(int baud)
 }
 #endif
 
-void __init enable_early_printk(void)
+/*
+ * Setup a default console, if more than one is compiled in, rely on the
+ * earlyprintk= parsing to give priority.
+ */
+static struct console *early_console =
+#ifdef CONFIG_SH_STANDARD_BIOS
+	&bios_console
+#elif defined(CONFIG_EARLY_SCIF_CONSOLE)
+	&scif_console
+#else
+	NULL
+#endif
+	;
+
+static int __initdata keep_early;
+
+int __init setup_early_printk(char *opt)
 {
-#ifdef CONFIG_EARLY_SCIF_CONSOLE
-	scif_sercon_init(115200);
+	char *space;
+	char buf[256];
+
+	strlcpy(buf, opt, sizeof(buf));
+	space = strchr(buf, ' ');
+	if (space)
+		*space = 0;
+
+	if (strstr(buf, "keep"))
+		keep_early = 1;
+
+#ifdef CONFIG_SH_STANDARD_BIOS
+	if (!strncmp(buf, "bios", 4))
+		early_console = &bios_console;
+#endif
+#if defined(CONFIG_EARLY_SCIF_CONSOLE)
+	if (!strncmp(buf, "serial", 6)) {
+		early_console = &scif_console;
+
+#ifdef CONFIG_CPU_SH4
+		scif_sercon_init(115200);
+#endif
+	}
 #endif
-	register_console(&early_console);
+
+	if (likely(early_console))
+		register_console(early_console);
+
+	return 1;
 }
+__setup("earlyprintk=", setup_early_printk);
 
-void disable_early_printk(void)
+void __init disable_early_printk(void)
 {
-	unregister_console(&early_console);
+	if (!keep_early) {
+		printk("disabling early console\n");
+		unregister_console(early_console);
+	} else
+		printk("keeping early console\n");
 }
-
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index e75189cb1db7..cff8d36f91b0 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -145,6 +145,24 @@ static inline void parse_cmdline (char ** cmdline_p, char mv_name[MV_NAME_SIZE],
 				memory_end = memory_start + mem_size;
 			}
 		}
+
+#ifdef CONFIG_EARLY_PRINTK
+		if (c == ' ' && !memcmp(from, "earlyprintk=", 12)) {
+			char *ep_end;
+
+			if (to != command_line)
+				to--;
+
+			from += 12;
+			ep_end = strchr(from, ' ');
+
+			setup_early_printk(from);
+			printk("early console enabled\n");
+
+			from = ep_end;
+		}
+#endif
+
 		if (c == ' ' && !memcmp(from, "sh_mv=", 6)) {
 			char* mv_end;
 			char* mv_comma;
@@ -245,11 +263,7 @@ void __init setup_arch(char **cmdline_p)
 	unsigned long bootmap_size;
 	unsigned long start_pfn, max_pfn, max_low_pfn;
 
-#ifdef CONFIG_EARLY_PRINTK
-	extern void enable_early_printk(void);
 
-	enable_early_printk();
-#endif
 #ifdef CONFIG_CMDLINE_BOOL
         strcpy(COMMAND_LINE, CONFIG_CMDLINE);
 #endif
diff --git a/include/asm-sh/setup.h b/include/asm-sh/setup.h
index d19de7c8df4e..34ca8a7f06ba 100644
--- a/include/asm-sh/setup.h
+++ b/include/asm-sh/setup.h
@@ -4,5 +4,7 @@
 
 #define COMMAND_LINE_SIZE 256
 
+int setup_early_printk(char *);
+
 #endif /* _SH_SETUP_H */
 #endif /* __KERNEL__ */
-- 
cgit v1.2.1


From 555ef1963029d19d2367acd09f6b9a6a0056f217 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:30:11 +0900
Subject: sh: Add SH7750S/SH7091 rules for SH7750 oprofile driver.

Update oprofile build rules for additional subtypes,
particularly SH7750S/SH7091.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/oprofile/Makefile | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/sh/oprofile/Makefile b/arch/sh/oprofile/Makefile
index 686738d4aa3c..1f25d9bb7538 100644
--- a/arch/sh/oprofile/Makefile
+++ b/arch/sh/oprofile/Makefile
@@ -7,7 +7,11 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \
 		timer_int.o )
 
 profdrvr-y				:= op_model_null.o
+
+# SH7750-style performance counters exist across 7750/7750S and 7091.
+profdrvr-$(CONFIG_CPU_SUBTYPE_SH7750S)	:= op_model_sh7750.o
 profdrvr-$(CONFIG_CPU_SUBTYPE_SH7750)	:= op_model_sh7750.o
+profdrvr-$(CONFIG_CPU_SUBTYPE_SH7091)	:= op_model_sh7750.o
 
 oprofile-y				:= $(DRIVER_OBJS) $(profdrvr-y)
 
-- 
cgit v1.2.1


From 5b19c9081fbd0882c936ec087bf9055a20251dec Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:31:40 +0900
Subject: sh: Support for SH7770/SH7780 CPU subtypes.

Merge support for SH7770 and SH7780 SH-4A subtypes.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/ex.S    | 164 +++++++++++++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/probe.c |  15 ++++
 arch/sh/kernel/cpu/sh4/sq.c    |   2 +
 arch/sh/mm/tlb-sh4.c           |   4 +
 include/asm-sh/bugs.h          |   4 +
 include/asm-sh/cpu-sh4/cache.h |   2 +
 6 files changed, 191 insertions(+)

diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S
index 26a27df06505..af5ecbddea55 100644
--- a/arch/sh/kernel/cpu/sh4/ex.S
+++ b/arch/sh/kernel/cpu/sh4/ex.S
@@ -72,6 +72,7 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 1110
 	.long	exception_error		
 	! Internal hardware
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	.long	do_IRQ	! TMU0 tuni0	/* 400 */
 	.long	do_IRQ	! TMU1 tuni1
 	.long	do_IRQ	! TMU2 tuni2
@@ -379,5 +380,168 @@ ENTRY(interrupt_table)
 	.long	exception_error			! 141 0x13a0
 	.long	exception_error			! 142 0x13c0
 	.long	exception_error			! 143 0x13e0
+#elif defined(CONFIG_CPU_SUBTYPE_SH7770)
+	.long	do_IRQ	!  50 0x840
+	.long	do_IRQ	!  51 0x860
+	.long	do_IRQ	!  52 0x880
+	.long	do_IRQ	!  53 0x8a0
+	.long	do_IRQ	!  54 0x8c0
+	.long	do_IRQ	!  55 0x8e0
+	.long	do_IRQ	!  56 0x900
+	.long	do_IRQ	!  57 0x920
+	.long	do_IRQ	!  58 0x940
+	.long	do_IRQ	!  59 0x960
+	.long	do_IRQ	!  60 0x980
+	.long	do_IRQ	!  61 0x9a0
+	.long	do_IRQ	!  62 0x9c0
+	.long	do_IRQ	!  63 0x9e0
+	.long	do_IRQ	!  64 0xa00
+	.long	do_IRQ	!  65 0xa20
+	.long	do_IRQ	!  66 0xa4d
+	.long	do_IRQ	!  67 0xa60
+	.long	do_IRQ	!  68 0xa80
+	.long	do_IRQ	!  69 0xaa0
+	.long	do_IRQ	!  70 0xac0
+	.long	do_IRQ	!  71 0xae0
+	.long	do_IRQ	!  72 0xb00
+	.long	do_IRQ	!  73 0xb20
+	.long	do_IRQ	!  74 0xb40
+	.long	do_IRQ	!  75 0xb60
+	.long	do_IRQ	!  76 0xb80
+	.long	do_IRQ	!  77 0xba0
+	.long	do_IRQ	!  78 0xbc0
+	.long	do_IRQ	!  79 0xbe0
+	.long	do_IRQ	!  80 0xc00
+	.long	do_IRQ	!  81 0xc20
+	.long	do_IRQ	!  82 0xc40
+	.long	do_IRQ	!  83 0xc60
+	.long	do_IRQ	!  84 0xc80
+	.long	do_IRQ	!  85 0xca0
+	.long	do_IRQ	!  86 0xcc0
+	.long	do_IRQ	!  87 0xce0
+	.long	do_IRQ	!  88 0xd00
+	.long	do_IRQ	!  89 0xd20
+	.long	do_IRQ	!  90 0xd40
+	.long	do_IRQ	!  91 0xd60
+	.long	do_IRQ	!  92 0xd80
+	.long	do_IRQ	!  93 0xda0
+	.long	do_IRQ	!  94 0xdc0
+	.long	do_IRQ	!  95 0xde0
+	.long	do_IRQ	!  96 0xe00
+	.long	do_IRQ	!  97 0xe20
+	.long	do_IRQ	!  98 0xe40
+	.long	do_IRQ	!  99 0xe60
+	.long	do_IRQ	! 100 0xe80
+	.long	do_IRQ	! 101 0xea0
+	.long	do_IRQ	! 102 0xec0
+	.long	do_IRQ	! 103 0xee0
+	.long	do_IRQ	! 104 0xf00
+	.long	do_IRQ	! 105 0xf20
+	.long	do_IRQ	! 106 0xf40
+	.long	do_IRQ	! 107 0xf60
+	.long	do_IRQ	! 108 0xf80
+#endif
+#else
+	.long	exception_error		/* 400 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! RTC	ati
+	.long	do_IRQ	!	pri
+	.long	do_IRQ	!	cui
+	.long	exception_error
+	.long	exception_error		/* 500 */
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! WDT	iti	/* 560 */
+	.long	do_IRQ	! TMU-ch0
+	.long	do_IRQ	! TMU-ch1
+	.long	do_IRQ	! TMU-ch2
+	.long	do_IRQ	! ticpi2	/* 5E0 */
+	.long	do_IRQ	! 32 Hitachi UDI	/* 600 */
+	.long	exception_error
+	.long	do_IRQ	! 34 DMAC dmte0
+	.long	do_IRQ	! 35	  dmte1
+	.long	do_IRQ	! 36	  dmte2
+	.long	do_IRQ	! 37	  dmte3
+	.long	do_IRQ	! 38	  dmae
+	.long	exception_error			! 39	/* 6E0 */
+	.long	do_IRQ	! 40 SCIF-ch0 eri		/* 700 */
+	.long	do_IRQ	! 41	      rxi
+	.long	do_IRQ	! 42	      bri
+	.long	do_IRQ	! 43	      txi
+	.long	do_IRQ	! 44 DMAC dmte4		/* 780 */
+	.long	do_IRQ	! 45	  dmte5
+	.long	do_IRQ	! 46	  dmte6
+	.long	do_IRQ	! 47	  dmte7		/* 7E0 */
+#if defined(CONFIG_SH_FPU)
+	.long	do_fpu_state_restore	! 48	/* 800 */
+	.long	do_fpu_state_restore	! 49	/* 820 */
+#else
+	.long	exception_error
+	.long	exception_error
+#endif
+	.long	exception_error			/* 840 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 56 CMT	/* 900 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! 60 HAC
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! PCI serr	/* A00 */
+	.long	do_IRQ	!     INTA
+	.long	do_IRQ	!     INTB
+	.long	do_IRQ	!     INTC
+	.long	do_IRQ	!     INTD
+	.long	do_IRQ	!     err
+	.long	do_IRQ	!     pwd3
+	.long	do_IRQ	!     pwd2
+	.long	do_IRQ	!     pwd1	/* B00 */
+	.long	do_IRQ	!     pwd0
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! SCIF-ch1 eri	/* B80 */
+	.long	do_IRQ	!	   rxi
+	.long	do_IRQ	!	   bri
+	.long	do_IRQ	!	   txi
+	.long	do_IRQ	! SIOF		/* C00 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! HSPI		/* C80 */
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! MMCIF	fatat	/* D00 */
+	.long	do_IRQ	!	tran
+	.long	do_IRQ	!	err
+	.long	do_IRQ	!	frdy
+	.long	do_IRQ	! DMAC dmint8	/* D80 */
+	.long	do_IRQ	!      dmint9
+	.long	do_IRQ	!      dmint10
+	.long	do_IRQ	!      dmint11
+	.long	do_IRQ	! TMU-ch3	/* E00 */
+	.long	do_IRQ	! TMU-ch4
+	.long	do_IRQ	! TMU-ch5
+	.long	exception_error
+	.long	do_IRQ	! SSI
+	.long	exception_error
+	.long	exception_error
+	.long	exception_error
+	.long	do_IRQ	! FLCTL	flste	/* F00 */
+	.long	do_IRQ	!	fltend
+	.long	do_IRQ	!	fltrq0
+	.long	do_IRQ	!	fltrq1
+	.long	do_IRQ	! GPIO gpioi0	/* F80 */
+	.long	do_IRQ	!      gpioi1
+	.long	do_IRQ	!      gpioi2
+	.long	do_IRQ	!      gpioi3
 #endif
 
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 89986e70d041..85ff48c1533a 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -78,6 +78,21 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->dcache.ways = 4;
 		cpu_data->flags &= ~CPU_HAS_FPU;
 		break;
+	case 0x2001:
+	case 0x2004:
+		cpu_data->type = CPU_SH7770;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
+	case 0x2006:
+	case 0x200A:
+		if (prr == 0x61)
+			cpu_data->type = CPU_SH7781;
+		else
+			cpu_data->type = CPU_SH7780;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
 		break;
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index 00f6a3c8c43b..32c93b781b51 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -166,7 +166,9 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
 	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
 
 	ptel = map->addr & PAGE_MASK;
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
+#endif
 
 	pgprot = pgprot_noncached(PAGE_KERNEL);
 
diff --git a/arch/sh/mm/tlb-sh4.c b/arch/sh/mm/tlb-sh4.c
index 115b1b6be40b..96e5fb0ac4fa 100644
--- a/arch/sh/mm/tlb-sh4.c
+++ b/arch/sh/mm/tlb-sh4.c
@@ -36,7 +36,9 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	unsigned long vpn;
 	struct page *page;
 	unsigned long pfn;
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	unsigned long ptea;
+#endif
 
 	/* Ptrace may call this routine. */
 	if (vma && current->active_mm != vma->vm_mm)
@@ -59,10 +61,12 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	ctrl_outl(vpn, MMU_PTEH);
 
 	pteval = pte_val(pte);
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 	/* Set PTEA register */
 	/* TODO: make this look less hacky */
 	ptea = ((pteval >> 28) & 0xe) | (pteval & 0x1);
 	ctrl_outl(ptea, MMU_PTEA);
+#endif
 
 	/* Set PTEL register */
 	pteval &= _PAGE_FLAGS_HARDWARE_MASK; /* drop software flags */
diff --git a/include/asm-sh/bugs.h b/include/asm-sh/bugs.h
index a6de3d06a3d9..b4000c8bf31b 100644
--- a/include/asm-sh/bugs.h
+++ b/include/asm-sh/bugs.h
@@ -32,6 +32,10 @@ static void __init check_bugs(void)
 	case CPU_SH7750 ... CPU_SH4_501:
 		*p++ = '4';
 		break;
+	case CPU_SH7770 ... CPU_SH7781:
+		*p++ = '4';
+		*p++ = 'a';
+		break;
 	default:
 		*p++ = '?';
 		*p++ = '!';
diff --git a/include/asm-sh/cpu-sh4/cache.h b/include/asm-sh/cpu-sh4/cache.h
index 1fe20359312c..6e9c7e6ee8e4 100644
--- a/include/asm-sh/cpu-sh4/cache.h
+++ b/include/asm-sh/cpu-sh4/cache.h
@@ -22,7 +22,9 @@
 #define CCR_CACHE_ICE	0x0100	/* Instruction Cache Enable */
 #define CCR_CACHE_ICI	0x0800	/* IC Invalidate */
 #define CCR_CACHE_IIX	0x8000	/* IC Index Enable */
+#ifndef CONFIG_CPU_SUBTYPE_SH7780
 #define CCR_CACHE_EMODE	0x80000000	/* EMODE Enable */
+#endif
 
 /* Default CCR setup: 8k+16k-byte cache,P1-wb,enable */
 #define CCR_CACHE_ENABLE	(CCR_CACHE_OCE|CCR_CACHE_ICE)
-- 
cgit v1.2.1


From ef9a1d4c0c383f75710f6adf2abb8cc264877e2c Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 14:32:57 +0900
Subject: sh: remove cpu_online() definition from <asm/smp.h>

It's defined in <linux/cpumask.h> and log is horribly flooded by
"redefined" messages.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/smp.h | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/include/asm-sh/smp.h b/include/asm-sh/smp.h
index f57c4fe9692a..71ecddf70db3 100644
--- a/include/asm-sh/smp.h
+++ b/include/asm-sh/smp.h
@@ -19,11 +19,6 @@
 #include <asm/atomic.h>
 #include <asm/current.h>
 
-extern cpumask_t cpu_online_map;
-extern cpumask_t cpu_possible_map;
-
-#define cpu_online(cpu)		cpu_isset(cpu, cpu_online_map)
-
 #define raw_smp_processor_id()	(current_thread_info()->cpu)
 
 /* I've no idea what the real meaning of this is */
-- 
cgit v1.2.1


From 75c92acdd5b19a5e3536ed670e1122d73c635b4a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:36:44 +0900
Subject: sh: Wire up new syscalls.

The syscall table has lagged behind a bit, wire up the new ones..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/syscalls.S | 28 ++++++++++++++++++++++++++++
 include/asm-sh/unistd.h   | 26 ++++++++++++++++++++++++--
 2 files changed, 52 insertions(+), 2 deletions(-)

diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
index f500304cbd69..ada61b0d9a73 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -322,3 +322,31 @@ ENTRY(sys_call_table)
 	.long sys_add_key		/* 285 */
 	.long sys_request_key
 	.long sys_keyctl
+	.long sys_ioprio_set
+	.long sys_ioprio_get
+	.long sys_inotify_init		/* 290 */
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_migrate_pages
+	.long sys_openat
+	.long sys_mkdirat		/* 295 */
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_fstatat64
+	.long sys_unlinkat		/* 300 */
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat
+	.long sys_fchmodat		/* 305 */
+	.long sys_faccessat
+	.long sys_ni_syscall		/* Reserved for pselect6 */
+	.long sys_ni_syscall		/* Reserved for ppoll */
+	.long sys_unshare
+	.long sys_set_robust_list	/* 310 */
+	.long sys_get_robust_list
+	.long sys_splice
+	.long sys_sync_file_range
+	.long sys_tee
+	.long sys_vmsplice		/* 315 */
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 76b5430cb458..49c6a6ee0ab2 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -300,9 +300,31 @@
 #define __NR_inotify_init	290
 #define __NR_inotify_add_watch	291
 #define __NR_inotify_rm_watch	292
+#define __NR_migrate_pages	293
+#define __NR_openat		294
+#define __NR_mkdirat		295
+#define __NR_mknodat		296
+#define __NR_fchownat		297
+#define __NR_futimesat		298
+#define __NR_fstatat64		299
+#define __NR_unlinkat		300
+#define __NR_renameat		301
+#define __NR_linkat		302
+#define __NR_symlinkat		303
+#define __NR_readlinkat		304
+#define __NR_fchmodat		305
+#define __NR_faccessat		305
+#define __NR_pselect6		307
+#define __NR_ppoll		308
+#define __NR_unshare		309
+#define __NR_set_robust_list	310
+#define __NR_get_robust_list	311
+#define __NR_splice		312
+#define __NR_sync_file_range	313
+#define __NR_tee		314
+#define __NR_vmsplice		315
 
-
-#define NR_syscalls 293
+#define NR_syscalls 316
 
 #ifdef __KERNEL__
 
-- 
cgit v1.2.1


From 8b395265f81817385f12e62f03f795efb732a445 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:38:02 +0900
Subject: sh: Fix fatal oops in copy_user_page() on sh4a (SH7780).

We had a pretty interesting oops happening, where copy_user_page()
was down()'ing p3map_sem[] with a bogus offset (particularly, an
offset that hadn't been initialized with sema_init(), due to the
mismatch between cpu_data->dcache.n_aliases and what was assumed
based off of the old CACHE_ALIAS value).

Luckily, spinlock debugging caught this for us, and so we drop
the old hardcoded CACHE_ALIAS for sh4 completely and rely on the
run-time probed cpu_data->dcache.alias_mask. This in turn gets
the p3map_sem[] index right, and everything works again.

While we're at it, also convert to 4-level page tables..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/pg-sh4.c                 | 24 ++++++++++++++----------
 include/asm-sh/cpu-sh4/cacheflush.h |  4 ----
 2 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/arch/sh/mm/pg-sh4.c b/arch/sh/mm/pg-sh4.c
index c776b60fc250..07371ed7a313 100644
--- a/arch/sh/mm/pg-sh4.c
+++ b/arch/sh/mm/pg-sh4.c
@@ -2,7 +2,7 @@
  * arch/sh/mm/pg-sh4.c
  *
  * Copyright (C) 1999, 2000, 2002  Niibe Yutaka
- * Copyright (C) 2002  Paul Mundt
+ * Copyright (C) 2002 - 2005  Paul Mundt
  *
  * Released under the terms of the GNU GPL v2.0.
  */
@@ -23,6 +23,8 @@
 
 extern struct semaphore p3map_sem[];
 
+#define CACHE_ALIAS (cpu_data->dcache.alias_mask)
+
 /*
  * clear_user_page
  * @to: P1 address
@@ -35,14 +37,15 @@ void clear_user_page(void *to, unsigned long address, struct page *page)
 	if (((address ^ (unsigned long)to) & CACHE_ALIAS) == 0)
 		clear_page(to);
 	else {
-		pgprot_t pgprot = __pgprot(_PAGE_PRESENT | 
+		pgprot_t pgprot = __pgprot(_PAGE_PRESENT |
 					   _PAGE_RW | _PAGE_CACHABLE |
-					   _PAGE_DIRTY | _PAGE_ACCESSED | 
+					   _PAGE_DIRTY | _PAGE_ACCESSED |
 					   _PAGE_HW_SHARED | _PAGE_FLAGS_HARD);
 		unsigned long phys_addr = PHYSADDR(to);
 		unsigned long p3_addr = P3SEG + (address & CACHE_ALIAS);
-		pgd_t *dir = pgd_offset_k(p3_addr);
-		pmd_t *pmd = pmd_offset(dir, p3_addr);
+		pgd_t *pgd = pgd_offset_k(p3_addr);
+		pud_t *pud = pud_offset(pgd, p3_addr);
+		pmd_t *pmd = pmd_offset(pud, p3_addr);
 		pte_t *pte = pte_offset_kernel(pmd, p3_addr);
 		pte_t entry;
 		unsigned long flags;
@@ -67,21 +70,22 @@ void clear_user_page(void *to, unsigned long address, struct page *page)
  * @address: U0 address to be mapped
  * @page: page (virt_to_page(to))
  */
-void copy_user_page(void *to, void *from, unsigned long address, 
+void copy_user_page(void *to, void *from, unsigned long address,
 		    struct page *page)
 {
 	__set_bit(PG_mapped, &page->flags);
 	if (((address ^ (unsigned long)to) & CACHE_ALIAS) == 0)
 		copy_page(to, from);
 	else {
-		pgprot_t pgprot = __pgprot(_PAGE_PRESENT | 
+		pgprot_t pgprot = __pgprot(_PAGE_PRESENT |
 					   _PAGE_RW | _PAGE_CACHABLE |
-					   _PAGE_DIRTY | _PAGE_ACCESSED | 
+					   _PAGE_DIRTY | _PAGE_ACCESSED |
 					   _PAGE_HW_SHARED | _PAGE_FLAGS_HARD);
 		unsigned long phys_addr = PHYSADDR(to);
 		unsigned long p3_addr = P3SEG + (address & CACHE_ALIAS);
-		pgd_t *dir = pgd_offset_k(p3_addr);
-		pmd_t *pmd = pmd_offset(dir, p3_addr);
+		pgd_t *pgd = pgd_offset_k(p3_addr);
+		pud_t *pud = pud_offset(pgd, p3_addr);
+		pmd_t *pmd = pmd_offset(pud, p3_addr);
 		pte_t *pte = pte_offset_kernel(pmd, p3_addr);
 		pte_t entry;
 		unsigned long flags;
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index f323567e085f..ea58c4c5944d 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -16,10 +16,6 @@
  *  caching; in which case they're only semi-broken),
  *  so we need them.
  */
-
-/* Page is 4K, OC size is 16K, there are four lines. */
-#define CACHE_ALIAS 0x00003000
-
 struct page;
 struct mm_struct;
 struct vm_area_struct;
-- 
cgit v1.2.1


From 24ab54cb49c099d691c68fdd1ac6a0c2f5177da4 Mon Sep 17 00:00:00 2001
From: Ollie Wild <aaw@rincewind.tv>
Date: Wed, 27 Sep 2006 14:46:24 +0900
Subject: sh: Fix TCP payload csum bug in csum_partial_copy_generic().

There's a bug in the Hitachi SuperH csum_partial_copy_generic()
implementation.  If the supplied length is 1 (and several alignment
conditions are met), the function immediately branches to label 4.
However, the assembly at label 4 expects the length to be stored in
register r2.  Since this has not occurred, subsequent behavior is
undefined.

This can cause bad payload checksums in TCP connections.

I've fixed the problem by initializing register r2 prior to the branch
instruction.

Signed-off-by: Ollie Wild <aaw@rincewind.tv>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/lib/checksum.S | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/sh/lib/checksum.S b/arch/sh/lib/checksum.S
index 7c50dfe68c07..cbdd0d40e545 100644
--- a/arch/sh/lib/checksum.S
+++ b/arch/sh/lib/checksum.S
@@ -202,8 +202,9 @@ ENTRY(csum_partial_copy_generic)
 	cmp/pz	r6		! Jump if we had at least two bytes.
 	bt/s	1f
 	 clrt
+	add	#2,r6		! r6 was < 2.	Deal with it.
 	bra	4f
-	 add	#2,r6		! r6 was < 2.	Deal with it.
+	 mov	r6,r2
 
 3:	! Handle different src and dest alignments.
 	! This is not common, so simple byte by byte copy will do.
-- 
cgit v1.2.1


From 634bf4f69b925950ddb09ef99ad7516a449a4333 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:48:09 +0900
Subject: sh: Fix libata build.

Drop virt_to_bus() from sg_dma_address() so libata builds.
While we're at it, move sg_dma_address() and sg_dma_len()
from pci.h to scatterlist.h.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pci.h         | 9 ---------
 include/asm-sh/scatterlist.h | 9 +++++++++
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 0a523c85b11c..18a109de0f2e 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -87,15 +87,6 @@ static inline void pcibios_penalize_isa_irq(int irq, int active)
  */
 #define pci_dac_dma_supported(pci_dev, mask) (0)
 
-/* These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns, or alternatively stop on the first sg_dma_len(sg) which
- * is 0.
- */
-#define sg_dma_address(sg)	(virt_to_bus((sg)->dma_address))
-#define sg_dma_len(sg)		((sg)->length)
-
 #ifdef CONFIG_PCI
 static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 					enum pci_dma_burst_strategy *strat,
diff --git a/include/asm-sh/scatterlist.h b/include/asm-sh/scatterlist.h
index 7b91df140022..d19e7cd3b023 100644
--- a/include/asm-sh/scatterlist.h
+++ b/include/asm-sh/scatterlist.h
@@ -10,4 +10,13 @@ struct scatterlist {
 
 #define ISA_DMA_THRESHOLD (0x1fffffff)
 
+/* These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns, or alternatively stop on the first sg_dma_len(sg) which
+ * is 0.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#define sg_dma_len(sg)		((sg)->length)
+
 #endif /* !(__ASM_SH_SCATTERLIST_H) */
-- 
cgit v1.2.1


From 94c0fa520cc169ccf661e9c03b5b95f74d1520b8 Mon Sep 17 00:00:00 2001
From: kogiidena <kogiidena@eggplant.ddo.jp>
Date: Wed, 27 Sep 2006 14:53:35 +0900
Subject: sh: landisk board support.

This adds support for the I-O DATA Landisk.

Signed-off-by: kogiidena <kogiidena@eggplant.ddo.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                        |    1 +
 arch/sh/boards/landisk/Makefile         |    7 +
 arch/sh/boards/landisk/io.c             |  322 ++++++++
 arch/sh/boards/landisk/irq.c            |  105 +++
 arch/sh/boards/landisk/landisk_pwb.c    |  348 ++++++++
 arch/sh/boards/landisk/rtc.c            |  102 +++
 arch/sh/boards/landisk/setup.c          |  218 +++++
 arch/sh/configs/landisk_defconfig       | 1373 +++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types                |    2 +-
 include/asm-sh/landisk/gio.h            |   45 +
 include/asm-sh/landisk/ide.h            |   14 +
 include/asm-sh/landisk/iodata_landisk.h |   78 ++
 12 files changed, 2614 insertions(+), 1 deletion(-)
 create mode 100644 arch/sh/boards/landisk/Makefile
 create mode 100644 arch/sh/boards/landisk/io.c
 create mode 100644 arch/sh/boards/landisk/irq.c
 create mode 100644 arch/sh/boards/landisk/landisk_pwb.c
 create mode 100644 arch/sh/boards/landisk/rtc.c
 create mode 100644 arch/sh/boards/landisk/setup.c
 create mode 100644 arch/sh/configs/landisk_defconfig
 create mode 100644 include/asm-sh/landisk/gio.h
 create mode 100644 include/asm-sh/landisk/ide.h
 create mode 100644 include/asm-sh/landisk/iodata_landisk.h

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 66cf0ea6d38f..e9112def48dd 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -105,6 +105,7 @@ machdir-$(CONFIG_SH_RTS7751R2D)			:= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
+machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/landisk/Makefile b/arch/sh/boards/landisk/Makefile
new file mode 100644
index 000000000000..60e75c6db0eb
--- /dev/null
+++ b/arch/sh/boards/landisk/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for I-O DATA DEVICE, INC. "LANDISK Series"
+#
+
+obj-y	 := setup.o io.o irq.o rtc.o landisk_pwb.o
+
+obj-$(CONFIG_PCI) += pci.o
diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
new file mode 100644
index 000000000000..1f1679af09d0
--- /dev/null
+++ b/arch/sh/boards/landisk/io.c
@@ -0,0 +1,322 @@
+/*
+ * arch/sh/boards/landisk/io.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for I-O Data Device, Inc. LANDISK.
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_landisk.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+/*
+ * modifed by kogiidena
+ * 2005.03.03
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/io.h>
+#include <asm/landisk/iodata_landisk.h>
+#include <asm/addrspace.h>
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include "../../drivers/pci/pci-sh7751.h"
+
+extern void *area5_io_base;	/* Area 5 I/O Base address */
+extern void *area6_io_base;	/* Area 6 I/O Base address */
+
+/*
+ * The 7751R LANDISK uses the built-in PCI controller (PCIC)
+ * of the 7751R processor, and has a SuperIO accessible via the PCI.
+ * The board also includes a PCMCIA controller on its memory bus,
+ * like the other Solution Engine boards.
+ */
+
+#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
+#define PCI_IO_AREA	SH7751_PCI_IO_BASE
+#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
+
+#define maybebadio(name,port) \
+  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
+	 #name, (port), (__u32) __builtin_return_address(0))
+
+static inline void delay(void)
+{
+	ctrl_inw(0xa0000000);
+}
+
+static inline unsigned long port2adr(unsigned int port)
+{
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		if (port == 0x3f6)
+			return ((unsigned long)area5_io_base + 0x2c);
+		else
+			return ((unsigned long)area5_io_base + PA_PIDE_OFFSET +
+				((port - 0x1f0) << 1));
+	else if ((0x170 <= port && port < 0x178) || port == 0x376)
+		if (port == 0x376)
+			return ((unsigned long)area6_io_base + 0x2c);
+		else
+			return ((unsigned long)area6_io_base + PA_SIDE_OFFSET +
+				((port - 0x170) << 1));
+	else
+		maybebadio(port2adr, (unsigned long)port);
+
+	return port;
+}
+
+/* In case someone configures the kernel w/o PCI support: in that */
+/* scenario, don't ever bother to check for PCI-window addresses */
+
+/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
+#if defined(CONFIG_PCI)
+#define CHECK_SH7751_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#else
+#define CHECK_SH_7751_PCIIO(port) (0)
+#endif
+
+/*
+ * General outline: remap really low stuff [eventually] to SuperIO,
+ * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
+ * is mapped through the PCI IO window.  Stuff with high bits (PXSEG)
+ * should be way beyond the window, and is used  w/o translation for
+ * compatibility.
+ */
+unsigned char landisk_inb(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else
+		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+}
+
+unsigned char landisk_inb_p(unsigned long port)
+{
+	unsigned char v;
+
+	if (PXSEG(port))
+		v = *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else
+		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+	delay();
+
+	return v;
+}
+
+unsigned short landisk_inw(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else
+		maybebadio(inw, port);
+
+	return 0;
+}
+
+unsigned int landisk_inl(unsigned long port)
+{
+	if (PXSEG(port))
+		return *(volatile unsigned long *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return *(volatile unsigned long *)PCI_IOMAP(port);
+	else
+		maybebadio(inl, port);
+
+	return 0;
+}
+
+void landisk_outb(unsigned char value, unsigned long port)
+{
+
+	if (PXSEG(port))
+		*(volatile unsigned char *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else
+		*(volatile unsigned short *)port2adr(port) = value;
+}
+
+void landisk_outb_p(unsigned char value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned char *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else
+		*(volatile unsigned short *)port2adr(port) = value;
+	delay();
+}
+
+void landisk_outw(unsigned short value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned short *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+	else
+		maybebadio(outw, port);
+}
+
+void landisk_outl(unsigned int value, unsigned long port)
+{
+	if (PXSEG(port))
+		*(volatile unsigned long *)port = value;
+	else if (CHECK_SH7751_PCIIO(port))
+		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+	else
+		maybebadio(outl, port);
+}
+
+void landisk_insb(unsigned long port, void *addr, unsigned long count)
+{
+	if (PXSEG(port))
+		while (count--)
+			*((unsigned char *)addr)++ =
+			    *(volatile unsigned char *)port;
+	else if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+
+		while (count--)
+			*((volatile unsigned char *)addr)++ = *bp;
+	} else {
+		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+
+		while (count--)
+			*((unsigned char *)addr)++ = *p;
+	}
+}
+
+void landisk_insw(unsigned long port, void *addr, unsigned long count)
+{
+	volatile __u16 *p;
+
+	if (PXSEG(port))
+		p = (volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else
+		p = (volatile unsigned short *)port2adr(port);
+	while (count--)
+		*((__u16 *) addr)++ = *p;
+}
+
+void landisk_insl(unsigned long port, void *addr, unsigned long count)
+{
+	if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+
+		while (count--)
+			*((__u32 *) addr)++ = *p;
+	} else
+		maybebadio(insl, port);
+}
+
+void landisk_outsb(unsigned long port, const void *addr, unsigned long count)
+{
+	if (PXSEG(port))
+		while (count--)
+			*(volatile unsigned char *)port =
+			    *((unsigned char *)addr)++;
+	else if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+
+		while (count--)
+			*bp = *((volatile unsigned char *)addr)++;
+	} else {
+		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+
+		while (count--)
+			*p = *((unsigned char *)addr)++;
+	}
+}
+
+void landisk_outsw(unsigned long port, const void *addr, unsigned long count)
+{
+	volatile __u16 *p;
+
+	if (PXSEG(port))
+		p = (volatile unsigned short *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else
+		p = (volatile unsigned short *)port2adr(port);
+	while (count--)
+		*p = *((__u16 *) addr)++;
+}
+
+void landisk_outsl(unsigned long port, const void *addr, unsigned long count)
+{
+	if (CHECK_SH7751_PCIIO(port)) {
+		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+
+		while (count--)
+			*p = *((__u32 *) addr)++;
+	} else
+		maybebadio(outsl, port);
+}
+
+/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
+/* already set up.  For a larger memory space, these would need to */
+/* reset PCIMBR as needed on a per-call basis...                   */
+
+unsigned char landisk_readb(unsigned long addr)
+{
+	return *(volatile unsigned char *)addr;
+}
+
+unsigned short landisk_readw(unsigned long addr)
+{
+	return *(volatile unsigned short *)addr;
+}
+
+unsigned int landisk_readl(unsigned long addr)
+{
+	return *(volatile unsigned long *)addr;
+}
+
+void landisk_writeb(unsigned char b, unsigned long addr)
+{
+	*(volatile unsigned char *)addr = b;
+}
+
+void landisk_writew(unsigned short b, unsigned long addr)
+{
+	*(volatile unsigned short *)addr = b;
+}
+
+void landisk_writel(unsigned int b, unsigned long addr)
+{
+	*(volatile unsigned long *)addr = b;
+}
+
+void *landisk_ioremap(unsigned long offset, unsigned long size)
+{
+	if (offset >= 0xfd000000)
+		return (void *)offset;
+	else
+		return (void *)P2SEGADDR(offset);
+}
+
+void landisk_iounmap(void *addr)
+{
+}
+
+/* Map ISA bus address to the real address. Only for PCMCIA.  */
+
+unsigned long landisk_isa_port2addr(unsigned long offset)
+{
+	return port2adr(offset);
+}
diff --git a/arch/sh/boards/landisk/irq.c b/arch/sh/boards/landisk/irq.c
new file mode 100644
index 000000000000..1dcc3feb7a44
--- /dev/null
+++ b/arch/sh/boards/landisk/irq.c
@@ -0,0 +1,105 @@
+/*
+ * arch/sh/boards/landisk/irq.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for I-O Data Device, Inc. LANDISK.
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_landisk.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+/*
+ * modified by kogiidena
+ * 2005.03.03
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/landisk/iodata_landisk.h>
+
+static void enable_landisk_irq(unsigned int irq);
+static void disable_landisk_irq(unsigned int irq);
+
+/* shutdown is same as "disable" */
+#define shutdown_landisk_irq disable_landisk_irq
+
+static void ack_landisk_irq(unsigned int irq);
+static void end_landisk_irq(unsigned int irq);
+
+static unsigned int startup_landisk_irq(unsigned int irq)
+{
+	enable_landisk_irq(irq);
+	return 0;		/* never anything pending */
+}
+
+static void disable_landisk_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned char val;
+	unsigned char mask = 0xff ^ (0x01 << (irq - 5));
+
+	/* Set the priority in IPR to 0 */
+	local_irq_save(flags);
+	val = ctrl_inb(PA_IMASK);
+	val &= mask;
+	ctrl_outb(val, PA_IMASK);
+	local_irq_restore(flags);
+}
+
+static void enable_landisk_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned char val;
+	unsigned char value = (0x01 << (irq - 5));
+
+	/* Set priority in IPR back to original value */
+	local_irq_save(flags);
+	val = ctrl_inb(PA_IMASK);
+	val |= value;
+	ctrl_outb(val, PA_IMASK);
+	local_irq_restore(flags);
+}
+
+static void ack_landisk_irq(unsigned int irq)
+{
+	disable_landisk_irq(irq);
+}
+
+static void end_landisk_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		enable_landisk_irq(irq);
+}
+
+static struct hw_interrupt_type landisk_irq_type = {
+	.typename = "LANDISK IRQ",
+	.startup = startup_landisk_irq,
+	.shutdown = shutdown_landisk_irq,
+	.enable = enable_landisk_irq,
+	.disable = disable_landisk_irq,
+	.ack = ack_landisk_irq,
+	.end = end_landisk_irq
+};
+
+static void make_landisk_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &landisk_irq_type;
+	disable_landisk_irq(irq);
+}
+
+/*
+ * Initialize IRQ setting
+ */
+void __init init_landisk_IRQ(void)
+{
+	int i;
+
+	for (i = 5; i < 14; i++)
+		make_landisk_irq(i);
+}
diff --git a/arch/sh/boards/landisk/landisk_pwb.c b/arch/sh/boards/landisk/landisk_pwb.c
new file mode 100644
index 000000000000..e75cb578a28b
--- /dev/null
+++ b/arch/sh/boards/landisk/landisk_pwb.c
@@ -0,0 +1,348 @@
+/*
+ * arch/sh/boards/landisk/landisk_pwb.c -- driver for the Power control switch.
+ *
+ * This driver will also support the I-O DATA Device, Inc. LANDISK Board.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copylight (C) 2002 Atom Create Engineering Co., Ltd.
+ *
+ * LED control drive function added by kogiidena
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/signal.h>
+#include <linux/major.h>
+#include <linux/poll.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/sched.h>
+#include <linux/timer.h>
+#include <linux/interrupt.h>
+
+#include <asm/system.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/uaccess.h>
+#include <asm/landisk/iodata_landisk.h>
+
+#define SHUTDOWN_BTN_MINOR	1	/* Shutdown button device minor no. */
+#define LED_MINOR	       21	/* LED minor no. */
+#define BTN_MINOR	       22	/* BUTTON minor no. */
+#define GIO_MINOR	       40	/* GIO minor no. */
+
+static int openCnt;
+static int openCntLED;
+static int openCntGio;
+static int openCntBtn;
+static int landisk_btn;
+static int landisk_btnctrlpid;
+/*
+ * Functions prototypes
+ */
+
+static int gio_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
+		     unsigned long arg);
+
+static int swdrv_open(struct inode *inode, struct file *filp)
+{
+	int minor;
+
+	minor = MINOR(inode->i_rdev);
+	filp->private_data = (void *)minor;
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		if (openCnt > 0) {
+			return -EALREADY;
+		} else {
+			openCnt++;
+			return 0;
+		}
+	} else if (minor == LED_MINOR) {
+		if (openCntLED > 0) {
+			return -EALREADY;
+		} else {
+			openCntLED++;
+			return 0;
+		}
+	} else if (minor == BTN_MINOR) {
+		if (openCntBtn > 0) {
+			return -EALREADY;
+		} else {
+			openCntBtn++;
+			return 0;
+		}
+	} else if (minor == GIO_MINOR) {
+		if (openCntGio > 0) {
+			return -EALREADY;
+		} else {
+			openCntGio++;
+			return 0;
+		}
+	}
+	return -ENOENT;
+
+}
+
+static int swdrv_close(struct inode *inode, struct file *filp)
+{
+	int minor;
+
+	minor = MINOR(inode->i_rdev);
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		openCnt--;
+	} else if (minor == LED_MINOR) {
+		openCntLED--;
+	} else if (minor == BTN_MINOR) {
+		openCntBtn--;
+	} else if (minor == GIO_MINOR) {
+		openCntGio--;
+	}
+	return 0;
+}
+
+static int swdrv_read(struct file *filp, char *buff, size_t count,
+		      loff_t * ppos)
+{
+	int minor;
+	minor = (int)(filp->private_data);
+
+	if (!access_ok(VERIFY_WRITE, (void *)buff, count))
+		return -EFAULT;
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		if (landisk_btn & 0x10) {
+			put_user(1, buff);
+			return 1;
+		} else {
+			return 0;
+		}
+	}
+	return 0;
+}
+
+static int swdrv_write(struct file *filp, const char *buff, size_t count,
+		       loff_t * ppos)
+{
+	int minor;
+	minor = (int)(filp->private_data);
+
+	if (minor == SHUTDOWN_BTN_MINOR) {
+		return count;
+	}
+	return count;
+}
+
+static irqreturn_t sw_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+{
+	landisk_btn = (0x0ff & (~ctrl_inb(PA_STATUS)));
+	disable_irq(IRQ_BUTTON);
+	disable_irq(IRQ_POWER);
+	ctrl_outb(0x00, PA_PWRINT_CLR);
+
+	if (landisk_btnctrlpid != 0) {
+		kill_proc(landisk_btnctrlpid, SIGUSR1, 1);
+		landisk_btnctrlpid = 0;
+	}
+
+	return IRQ_HANDLED;
+}
+
+static struct file_operations swdrv_fops = {
+	.read = swdrv_read,	/* read */
+	.write = swdrv_write,	/* write */
+	.open = swdrv_open,	/* open */
+	.release = swdrv_close,	/* release */
+	.ioctl = gio_ioctl,	/* ioctl */
+
+};
+
+static char banner[] __initdata =
+    KERN_INFO "LANDISK and USL-5P Button, LED and GIO driver initialized\n";
+
+int __init swdrv_init(void)
+{
+	int error;
+
+	printk("%s", banner);
+
+	openCnt = 0;
+	openCntLED = 0;
+	openCntBtn = 0;
+	openCntGio = 0;
+	landisk_btn = 0;
+	landisk_btnctrlpid = 0;
+
+	if ((error = register_chrdev(SHUTDOWN_BTN_MAJOR, "swdrv", &swdrv_fops))) {
+		printk(KERN_ERR
+		       "Button, LED and GIO driver:Couldn't register driver, error=%d\n",
+		       error);
+		return 1;
+	}
+
+	if (request_irq(IRQ_POWER, sw_interrupt, 0, "SHUTDOWNSWITCH", NULL)) {
+		printk(KERN_ERR "Unable to get IRQ 11.\n");
+		return 1;
+	}
+	if (request_irq(IRQ_BUTTON, sw_interrupt, 0, "USL-5P BUTTON", NULL)) {
+		printk(KERN_ERR "Unable to get IRQ 12.\n");
+		return 1;
+	}
+	ctrl_outb(0x00, PA_PWRINT_CLR);
+
+	return 0;
+}
+
+module_init(swdrv_init);
+
+/*
+ * gio driver
+ *
+ */
+
+#include <asm/landisk/gio.h>
+
+static int gio_ioctl(struct inode *inode, struct file *filp,
+		     unsigned int cmd, unsigned long arg)
+{
+	int minor;
+	unsigned int data, mask;
+	static unsigned int addr = 0;
+
+	minor = (int)(filp->private_data);
+
+	/* access control */
+	if (minor == GIO_MINOR) {
+		;
+	} else if (minor == LED_MINOR) {
+		if (((cmd & 0x0ff) >= 9) && ((cmd & 0x0ff) < 20)) {
+			;
+		} else {
+			return -EINVAL;
+		}
+	} else if (minor == BTN_MINOR) {
+		if (((cmd & 0x0ff) >= 20) && ((cmd & 0x0ff) < 30)) {
+			;
+		} else {
+			return -EINVAL;
+		}
+	} else {
+		return -EINVAL;
+	}
+
+	if (cmd & 0x01) {	/* write */
+		if (copy_from_user(&data, (int *)arg, sizeof(int))) {
+			return -EFAULT;
+		}
+	}
+
+	switch (cmd) {
+	case GIODRV_IOCSGIOSETADDR:	/* addres set */
+		addr = data;
+		break;
+
+	case GIODRV_IOCSGIODATA1:	/* write byte */
+		ctrl_outb((unsigned char)(0x0ff & data), addr);
+		break;
+
+	case GIODRV_IOCSGIODATA2:	/* write word */
+		if (addr & 0x01) {
+			return -EFAULT;
+		}
+		ctrl_outw((unsigned short int)(0x0ffff & data), addr);
+		break;
+
+	case GIODRV_IOCSGIODATA4:	/* write long */
+		if (addr & 0x03) {
+			return -EFAULT;
+		}
+		ctrl_outl(data, addr);
+		break;
+
+	case GIODRV_IOCGGIODATA1:	/* read byte */
+		data = ctrl_inb(addr);
+		break;
+
+	case GIODRV_IOCGGIODATA2:	/* read word */
+		if (addr & 0x01) {
+			return -EFAULT;
+		}
+		data = ctrl_inw(addr);
+		break;
+
+	case GIODRV_IOCGGIODATA4:	/* read long */
+		if (addr & 0x03) {
+			return -EFAULT;
+		}
+		data = ctrl_inl(addr);
+		break;
+	case GIODRV_IOCSGIO_LED:	/* write */
+		mask = ((data & 0x00ffffff) << 8)
+		    | ((data & 0x0000ffff) << 16)
+		    | ((data & 0x000000ff) << 24);
+		landisk_ledparam = data & (~mask);
+		if (landisk_arch == 0) {	/* arch == landisk */
+			landisk_ledparam &= 0x03030303;
+			mask = (~(landisk_ledparam >> 22)) & 0x000c;
+			landisk_ledparam |= mask;
+		} else {	                /* arch == usl-5p */
+			mask = (landisk_ledparam >> 24) & 0x0001;
+			landisk_ledparam |= mask;
+			landisk_ledparam &= 0x007f7f7f;
+		}
+		landisk_ledparam |= 0x80;
+		break;
+	case GIODRV_IOCGGIO_LED:	/* read */
+		data = landisk_ledparam;
+		if (landisk_arch == 0) {	/* arch == landisk */
+			data &= 0x03030303;
+		} else {	                /* arch == usl-5p */
+			;
+		}
+		data &= (~0x080);
+		break;
+	case GIODRV_IOCSGIO_BUZZER:	/* write */
+		landisk_buzzerparam = data;
+		landisk_ledparam |= 0x80;
+		break;
+	case GIODRV_IOCGGIO_LANDISK:	/* read */
+		data = landisk_arch & 0x01;
+		break;
+	case GIODRV_IOCGGIO_BTN:	/* read */
+		data = (0x0ff & ctrl_inb(PA_PWRINT_CLR));
+		data <<= 8;
+		data |= (0x0ff & ctrl_inb(PA_IMASK));
+		data <<= 8;
+		data |= (0x0ff & landisk_btn);
+		data <<= 8;
+		data |= (0x0ff & (~ctrl_inb(PA_STATUS)));
+		break;
+	case GIODRV_IOCSGIO_BTNPID:	/* write */
+		landisk_btnctrlpid = data;
+		landisk_btn = 0;
+		if (irq_desc[IRQ_BUTTON].depth) {
+			enable_irq(IRQ_BUTTON);
+		}
+		if (irq_desc[IRQ_POWER].depth) {
+			enable_irq(IRQ_POWER);
+		}
+		break;
+	case GIODRV_IOCGGIO_BTNPID:	/* read */
+		data = landisk_btnctrlpid;
+		break;
+	default:
+		return -EFAULT;
+		break;
+	}
+
+	if ((cmd & 0x01) == 0) {	/* read */
+		if (copy_to_user((int *)arg, &data, sizeof(int))) {
+			return -EFAULT;
+		}
+	}
+	return 0;
+}
diff --git a/arch/sh/boards/landisk/rtc.c b/arch/sh/boards/landisk/rtc.c
new file mode 100644
index 000000000000..d666aa63303f
--- /dev/null
+++ b/arch/sh/boards/landisk/rtc.c
@@ -0,0 +1,102 @@
+/*
+ * arch/sh/boards/landisk/rtc.c --  RTC support
+ *
+ *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
+ *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
+ */
+/*
+ * modifed by kogiidena
+ * 2005.09.16
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/time.h>
+#include <linux/delay.h>
+#include <linux/spinlock.h>
+
+#ifndef BCD_TO_BIN
+#define BCD_TO_BIN(val) ((val)=((val)&15) + ((val)>>4)*10)
+#endif
+
+#ifndef BIN_TO_BCD
+#define BIN_TO_BCD(val) ((val)=(((val)/10)<<4) + (val)%10)
+#endif
+
+extern void (*rtc_get_time) (struct timespec *);
+extern int (*rtc_set_time) (const time_t);
+extern spinlock_t rtc_lock;
+
+extern void
+rs5c313_set_cmos_time(unsigned int BCD_yr, unsigned int BCD_mon,
+		      unsigned int BCD_day, unsigned int BCD_hr,
+		      unsigned int BCD_min, unsigned int BCD_sec);
+
+extern unsigned long
+rs5c313_get_cmos_time(unsigned int *BCD_yr, unsigned int *BCD_mon,
+		      unsigned int *BCD_day, unsigned int *BCD_hr,
+		      unsigned int *BCD_min, unsigned int *BCD_sec);
+
+void landisk_rtc_gettimeofday(struct timespec *tv)
+{
+	unsigned int BCD_yr, BCD_mon, BCD_day, BCD_hr, BCD_min, BCD_sec;
+	unsigned long flags;
+
+	spin_lock_irqsave(&rtc_lock, flags);
+	tv->tv_sec = rs5c313_get_cmos_time
+	    (&BCD_yr, &BCD_mon, &BCD_day, &BCD_hr, &BCD_min, &BCD_sec);
+	tv->tv_nsec = 0;
+	spin_unlock_irqrestore(&rtc_lock, flags);
+}
+
+int landisk_rtc_settimeofday(const time_t secs)
+{
+	int retval = 0;
+	int real_seconds, real_minutes, cmos_minutes;
+	unsigned long flags;
+	unsigned long nowtime = secs;
+	unsigned int BCD_yr, BCD_mon, BCD_day, BCD_hr, BCD_min, BCD_sec;
+
+	spin_lock_irqsave(&rtc_lock, flags);
+
+	rs5c313_get_cmos_time
+	  (&BCD_yr, &BCD_mon, &BCD_day, &BCD_hr, &BCD_min, &BCD_sec);
+	cmos_minutes = BCD_min;
+	BCD_TO_BIN(cmos_minutes);
+
+	/*
+	 * since we're only adjusting minutes and seconds,
+	 * don't interfere with hour overflow. This avoids
+	 * messing with unknown time zones but requires your
+	 * RTC not to be off by more than 15 minutes
+	 */
+	real_seconds = nowtime % 60;
+	real_minutes = nowtime / 60;
+	if (((abs(real_minutes - cmos_minutes) + 15) / 30) & 1)
+		real_minutes += 30;	/* correct for half hour time zone */
+	real_minutes %= 60;
+
+	if (abs(real_minutes - cmos_minutes) < 30) {
+		BIN_TO_BCD(real_seconds);
+		BIN_TO_BCD(real_minutes);
+		rs5c313_set_cmos_time(BCD_yr, BCD_mon, BCD_day, BCD_hr,
+				      real_minutes, real_seconds);
+	} else {
+		printk(KERN_WARNING
+		       "set_rtc_time: can't update from %d to %d\n",
+		       cmos_minutes, real_minutes);
+		retval = -1;
+	}
+
+	spin_unlock_irqrestore(&rtc_lock, flags);
+	return retval;
+}
+
+
+void landisk_time_init(void)
+{
+	rtc_get_time = landisk_rtc_gettimeofday;
+	rtc_set_time = landisk_rtc_settimeofday;
+}
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
new file mode 100644
index 000000000000..0c60eaa10ba7
--- /dev/null
+++ b/arch/sh/boards/landisk/setup.c
@@ -0,0 +1,218 @@
+/*
+ * arch/sh/boards/landisk/setup.c
+ *
+ * Copyright (C) 2002 Paul Mundt
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * Setup code for an unknown machine (internal peripherials only)
+ */
+/*
+ * linux/arch/sh/kernel/setup_landisk.c
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * I-O DATA Device, Inc. LANDISK Support.
+ *
+ * Modified for LANDISK by
+ * Atom Create Engineering Co., Ltd. 2002.
+ */
+/*
+ * modifed by kogiidena
+ * 2005.09.16
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/pm.h>
+
+#include <linux/hdreg.h>
+#include <linux/ide.h>
+#include <linux/pci.h>
+
+#include <asm/machvec.h>
+#include <asm/rtc.h>
+#include <asm/machvec_init.h>
+#include <asm/io.h>
+#include <asm/landisk/iodata_landisk.h>
+#include <asm/landisk/io.h>
+
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+
+extern void (*board_time_init) (void);
+void landisk_time_init(void);
+extern void init_landisk_IRQ(void);
+
+int landisk_ledparam;
+int landisk_buzzerparam;
+int landisk_arch;
+
+/* defined in mm/ioremap.c */
+extern void *p3_ioremap(unsigned long phys_addr, unsigned long size,
+			unsigned long flags);
+
+/*
+ * Initialize the board
+ */
+
+const char *get_system_type(void)
+{
+	return "LANDISK";
+}
+
+static void landisk_power_off(void)
+{
+	ctrl_outb(0x01, PA_SHUTDOWN);
+}
+
+void check_usl5p(void)
+{
+	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+	unsigned char tmp1, tmp2;
+	tmp1 = *p;
+	*p = 0x40;
+	tmp2 = *p;
+	*p = tmp1;
+	landisk_arch = (tmp2 == 0x40) ? 1 : 0;
+	if (landisk_arch == 1) {	/* arch == usl-5p */
+		landisk_ledparam = 0x00000380;
+		landisk_ledparam |= (tmp1 & 0x07c);
+	} else {                        /* arch == landisk */
+		landisk_ledparam = 0x02000180;
+		landisk_ledparam |= 0x04;
+	}
+	return;
+}
+
+void __init platform_setup(void)
+{
+
+	landisk_buzzerparam = 0;
+	check_usl5p();
+
+	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+	board_time_init = landisk_time_init;
+	pm_power_off = landisk_power_off;
+
+}
+
+void *area5_io_base;
+void *area6_io_base;
+
+int __init cf_init(void)
+{
+	pgprot_t prot;
+	unsigned long paddrbase, psize;
+
+	/* open I/O area window */
+	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
+	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area5_io_base) {
+		printk("allocate_cf_area : can't open CF I/O window!\n");
+		return -ENOMEM;
+	}
+
+	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
+	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area6_io_base) {
+		printk("allocate_cf_area : can't open HDD I/O window!\n");
+		return -ENOMEM;
+	}
+
+	printk(KERN_INFO "Allocate Area5/6 success.\n");
+
+	/* XXX : do we need attribute and common-memory area also? */
+
+	return 0;
+}
+
+__initcall(cf_init);
+
+#include <linux/sched.h>
+
+/* Cycle the LED's in the clasic knightrider/Sun pattern */
+
+void heartbeat_landisk(void)
+{
+	static unsigned int cnt = 0, blink = 0x00, period = 25;
+	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+	char data;
+
+	if ((landisk_ledparam & 0x080) == 0) {
+		return;
+	}
+	cnt += 1;
+	if (cnt < period) {
+		return;
+	}
+	cnt = 0;
+	blink++;
+
+	data = (blink & 0x01) ? (landisk_ledparam >> 16) : 0;
+	data |= (blink & 0x02) ? (landisk_ledparam >> 8) : 0;
+	data |= landisk_ledparam;
+
+	/* buzzer */
+	if (landisk_buzzerparam & 0x1) {
+		data |= 0x80;
+	} else {
+		data &= 0x7f;
+	}
+	*p = data;
+
+	if (((landisk_ledparam & 0x007f7f00) == 0)
+	    && (landisk_buzzerparam == 0)) {
+		landisk_ledparam &= (~0x0080);
+	}
+	landisk_buzzerparam >>= 1;
+}
+
+/*
+ * The Machine Vector
+ */
+
+struct sh_machine_vector mv_landisk __initmv = {
+	.mv_nr_irqs = 72,
+	.mv_inb = landisk_inb,
+	.mv_inw = landisk_inw,
+	.mv_inl = landisk_inl,
+	.mv_outb = landisk_outb,
+	.mv_outw = landisk_outw,
+	.mv_outl = landisk_outl,
+	.mv_inb_p = landisk_inb_p,
+	.mv_inw_p = landisk_inw,
+	.mv_inl_p = landisk_inl,
+	.mv_outb_p = landisk_outb_p,
+	.mv_outw_p = landisk_outw,
+	.mv_outl_p = landisk_outl,
+	.mv_insb = landisk_insb,
+	.mv_insw = landisk_insw,
+	.mv_insl = landisk_insl,
+	.mv_outsb = landisk_outsb,
+	.mv_outsw = landisk_outsw,
+	.mv_outsl = landisk_outsl,
+	.mv_readb = landisk_readb,
+	.mv_readw = landisk_readw,
+	.mv_readl = landisk_readl,
+	.mv_writeb = landisk_writeb,
+	.mv_writew = landisk_writew,
+	.mv_writel = landisk_writel,
+	.mv_ioremap = landisk_ioremap,
+	.mv_iounmap = landisk_iounmap,
+	.mv_isa_port2addr = landisk_isa_port2addr,
+	.mv_init_irq = init_landisk_IRQ,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_landisk,
+#endif
+
+};
+
+ALIAS_MV(landisk)
diff --git a/arch/sh/configs/landisk_defconfig b/arch/sh/configs/landisk_defconfig
new file mode 100644
index 000000000000..6b43316d03cf
--- /dev/null
+++ b/arch/sh/configs/landisk_defconfig
@@ -0,0 +1,1373 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.13-sh
+# Sun Sep 11 13:00:46 2005
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_IOMAP=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+# CONFIG_IKCONFIG is not set
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_EMBEDDED is not set
+CONFIG_KALLSYMS=y
+CONFIG_KALLSYMS_EXTRA_PASS=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+CONFIG_SH_LANDISK=y
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+CONFIG_CPU_SUBTYPE_SH7751=y
+CONFIG_CPU_SUBTYPE_SH7751R=y
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x04000000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_FPU=y
+# CONFIG_SH_STORE_QUEUES is not set
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=33333333
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=4
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+CONFIG_HEARTBEAT=y
+
+#
+# Kernel features
+#
+CONFIG_KEXEC=y
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+# CONFIG_CMDLINE_BOOL is not set
+
+#
+# Bus options
+#
+CONFIG_ISA=y
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+CONFIG_PCCARD=y
+# CONFIG_PCMCIA_DEBUG is not set
+CONFIG_PCMCIA=y
+CONFIG_PCMCIA_LOAD_CIS=y
+CONFIG_PCMCIA_IOCTL=y
+CONFIG_CARDBUS=y
+
+#
+# PC-card bridges
+#
+CONFIG_YENTA=y
+# CONFIG_PD6729 is not set
+# CONFIG_I82092 is not set
+# CONFIG_I82365 is not set
+# CONFIG_TCIC is not set
+CONFIG_PCMCIA_PROBE=y
+CONFIG_PCCARD_NONSTATIC=y
+
+#
+# PCI Hotplug Support
+#
+# CONFIG_HOTPLUG_PCI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_MULTIPLE_TABLES is not set
+# CONFIG_IP_ROUTE_MULTIPATH is not set
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+
+#
+# IP: Virtual Server Configuration
+#
+# CONFIG_IP_VS is not set
+# CONFIG_IPV6 is not set
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+# CONFIG_NETFILTER_NETLINK is not set
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_CONNTRACK=m
+CONFIG_IP_NF_CT_ACCT=y
+CONFIG_IP_NF_CONNTRACK_MARK=y
+# CONFIG_IP_NF_CONNTRACK_EVENTS is not set
+# CONFIG_IP_NF_CT_PROTO_SCTP is not set
+CONFIG_IP_NF_FTP=m
+CONFIG_IP_NF_IRC=m
+# CONFIG_IP_NF_NETBIOS_NS is not set
+CONFIG_IP_NF_TFTP=m
+CONFIG_IP_NF_AMANDA=m
+# CONFIG_IP_NF_PPTP is not set
+CONFIG_IP_NF_QUEUE=m
+CONFIG_IP_NF_IPTABLES=m
+CONFIG_IP_NF_MATCH_LIMIT=m
+CONFIG_IP_NF_MATCH_IPRANGE=m
+CONFIG_IP_NF_MATCH_MAC=m
+CONFIG_IP_NF_MATCH_PKTTYPE=m
+CONFIG_IP_NF_MATCH_MARK=m
+CONFIG_IP_NF_MATCH_MULTIPORT=m
+CONFIG_IP_NF_MATCH_TOS=m
+CONFIG_IP_NF_MATCH_RECENT=m
+CONFIG_IP_NF_MATCH_ECN=m
+CONFIG_IP_NF_MATCH_DSCP=m
+CONFIG_IP_NF_MATCH_AH_ESP=m
+CONFIG_IP_NF_MATCH_LENGTH=m
+CONFIG_IP_NF_MATCH_TTL=m
+CONFIG_IP_NF_MATCH_TCPMSS=m
+CONFIG_IP_NF_MATCH_HELPER=m
+CONFIG_IP_NF_MATCH_STATE=m
+CONFIG_IP_NF_MATCH_CONNTRACK=m
+CONFIG_IP_NF_MATCH_OWNER=m
+CONFIG_IP_NF_MATCH_ADDRTYPE=m
+CONFIG_IP_NF_MATCH_REALM=m
+CONFIG_IP_NF_MATCH_SCTP=m
+# CONFIG_IP_NF_MATCH_DCCP is not set
+CONFIG_IP_NF_MATCH_COMMENT=m
+CONFIG_IP_NF_MATCH_CONNMARK=m
+# CONFIG_IP_NF_MATCH_CONNBYTES is not set
+CONFIG_IP_NF_MATCH_HASHLIMIT=m
+# CONFIG_IP_NF_MATCH_STRING is not set
+CONFIG_IP_NF_FILTER=m
+CONFIG_IP_NF_TARGET_REJECT=m
+CONFIG_IP_NF_TARGET_LOG=m
+CONFIG_IP_NF_TARGET_ULOG=m
+CONFIG_IP_NF_TARGET_TCPMSS=m
+# CONFIG_IP_NF_TARGET_NFQUEUE is not set
+CONFIG_IP_NF_NAT=m
+CONFIG_IP_NF_NAT_NEEDED=y
+CONFIG_IP_NF_TARGET_MASQUERADE=m
+CONFIG_IP_NF_TARGET_REDIRECT=m
+CONFIG_IP_NF_TARGET_NETMAP=m
+CONFIG_IP_NF_TARGET_SAME=m
+# CONFIG_IP_NF_NAT_SNMP_BASIC is not set
+CONFIG_IP_NF_NAT_IRC=m
+CONFIG_IP_NF_NAT_FTP=m
+CONFIG_IP_NF_NAT_TFTP=m
+CONFIG_IP_NF_NAT_AMANDA=m
+CONFIG_IP_NF_MANGLE=m
+CONFIG_IP_NF_TARGET_TOS=m
+CONFIG_IP_NF_TARGET_ECN=m
+CONFIG_IP_NF_TARGET_DSCP=m
+CONFIG_IP_NF_TARGET_MARK=m
+CONFIG_IP_NF_TARGET_CLASSIFY=m
+# CONFIG_IP_NF_TARGET_TTL is not set
+CONFIG_IP_NF_TARGET_CONNMARK=m
+# CONFIG_IP_NF_TARGET_CLUSTERIP is not set
+CONFIG_IP_NF_RAW=m
+CONFIG_IP_NF_TARGET_NOTRACK=m
+CONFIG_IP_NF_ARPTABLES=m
+CONFIG_IP_NF_ARPFILTER=m
+CONFIG_IP_NF_ARP_MANGLE=m
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+CONFIG_LLC=m
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+CONFIG_ATALK=m
+# CONFIG_DEV_APPLETALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+CONFIG_NET_CLS_ROUTE=y
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+# CONFIG_PNP is not set
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_LBD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+CONFIG_IDE=y
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=y
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+# CONFIG_BLK_DEV_IDE_SATA is not set
+CONFIG_BLK_DEV_IDEDISK=y
+# CONFIG_IDEDISK_MULTI_MODE is not set
+# CONFIG_BLK_DEV_IDECS is not set
+CONFIG_BLK_DEV_IDECD=y
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+CONFIG_BLK_DEV_IDESCSI=y
+# CONFIG_IDE_TASK_IOCTL is not set
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=y
+CONFIG_BLK_DEV_IDEPCI=y
+CONFIG_IDEPCI_SHARE_IRQ=y
+CONFIG_BLK_DEV_OFFBOARD=y
+CONFIG_BLK_DEV_GENERIC=y
+# CONFIG_BLK_DEV_OPTI621 is not set
+CONFIG_BLK_DEV_IDEDMA_PCI=y
+# CONFIG_BLK_DEV_IDEDMA_FORCED is not set
+CONFIG_IDEDMA_PCI_AUTO=y
+CONFIG_IDEDMA_ONLYDISK=y
+CONFIG_BLK_DEV_AEC62XX=y
+# CONFIG_BLK_DEV_ALI15X3 is not set
+# CONFIG_BLK_DEV_AMD74XX is not set
+# CONFIG_BLK_DEV_CMD64X is not set
+# CONFIG_BLK_DEV_TRIFLEX is not set
+# CONFIG_BLK_DEV_CY82C693 is not set
+# CONFIG_BLK_DEV_CS5520 is not set
+# CONFIG_BLK_DEV_CS5530 is not set
+# CONFIG_BLK_DEV_HPT34X is not set
+# CONFIG_BLK_DEV_HPT366 is not set
+# CONFIG_BLK_DEV_SC1200 is not set
+# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
+# CONFIG_BLK_DEV_NS87415 is not set
+# CONFIG_BLK_DEV_PDC202XX_OLD is not set
+# CONFIG_BLK_DEV_PDC202XX_NEW is not set
+# CONFIG_BLK_DEV_SVWKS is not set
+# CONFIG_BLK_DEV_SIIMAGE is not set
+# CONFIG_BLK_DEV_SLC90E66 is not set
+# CONFIG_BLK_DEV_TRM290 is not set
+# CONFIG_BLK_DEV_VIA82CXXX is not set
+CONFIG_IDE_SH=y
+# CONFIG_IDE_ARM is not set
+# CONFIG_IDE_CHIPSETS is not set
+CONFIG_BLK_DEV_IDEDMA=y
+# CONFIG_IDEDMA_IVB is not set
+CONFIG_IDEDMA_AUTO=y
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=y
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=y
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+# CONFIG_BLK_DEV_SR is not set
+# CONFIG_CHR_DEV_SG is not set
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+CONFIG_SCSI_MULTI_LUN=y
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AHA152X is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_SCSI_IN2000 is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_DTC3280 is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_GENERIC_NCR5380 is not set
+# CONFIG_SCSI_GENERIC_NCR5380_MMIO is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_NCR53C406A is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_PAS16 is not set
+# CONFIG_SCSI_PSI240I is not set
+# CONFIG_SCSI_QLOGIC_FAS is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=y
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_SYM53C416 is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_T128 is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# PCMCIA SCSI adapter support
+#
+# CONFIG_PCMCIA_AHA152X is not set
+# CONFIG_PCMCIA_FDOMAIN is not set
+# CONFIG_PCMCIA_NINJA_SCSI is not set
+# CONFIG_PCMCIA_QLOGIC is not set
+# CONFIG_PCMCIA_SYM53C500 is not set
+
+#
+# Old CD-ROM drivers (not SCSI, not IDE)
+#
+# CONFIG_CD_NO_IDESCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+CONFIG_MD=y
+CONFIG_BLK_DEV_MD=m
+CONFIG_MD_LINEAR=m
+CONFIG_MD_RAID0=m
+CONFIG_MD_RAID1=m
+# CONFIG_MD_RAID10 is not set
+# CONFIG_MD_RAID5 is not set
+# CONFIG_MD_RAID6 is not set
+# CONFIG_MD_MULTIPATH is not set
+# CONFIG_MD_FAULTY is not set
+# CONFIG_BLK_DEV_DM is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_NET_VENDOR_SMC is not set
+# CONFIG_SMC91X is not set
+# CONFIG_NET_VENDOR_RACAL is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_AT1700 is not set
+# CONFIG_DEPCA is not set
+# CONFIG_HP100 is not set
+# CONFIG_NET_ISA is not set
+# CONFIG_NE2000 is not set
+CONFIG_NET_PCI=y
+# CONFIG_PCNET32 is not set
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_AC3200 is not set
+# CONFIG_APRICOT is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_CS89x0 is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+CONFIG_8139CP=y
+# CONFIG_8139TOO is not set
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+# CONFIG_VIA_RHINE is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# PCMCIA network device support
+#
+# CONFIG_NET_PCMCIA is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+CONFIG_RS5C313_RTC=y
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+
+#
+# PCMCIA character devices
+#
+# CONFIG_SYNCLINK_CS is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+CONFIG_VIDEO_DEV=m
+
+#
+# Video For Linux
+#
+
+#
+# Video Adapters
+#
+# CONFIG_VIDEO_PMS is not set
+# CONFIG_VIDEO_CPIA is not set
+# CONFIG_VIDEO_STRADIS is not set
+# CONFIG_VIDEO_MXB is not set
+# CONFIG_VIDEO_DPC is not set
+# CONFIG_VIDEO_HEXIUM_ORION is not set
+# CONFIG_VIDEO_HEXIUM_GEMINI is not set
+
+#
+# Radio Adapters
+#
+# CONFIG_RADIO_CADET is not set
+# CONFIG_RADIO_RTRACK is not set
+# CONFIG_RADIO_RTRACK2 is not set
+# CONFIG_RADIO_AZTECH is not set
+# CONFIG_RADIO_GEMTEK is not set
+# CONFIG_RADIO_GEMTEK_PCI is not set
+# CONFIG_RADIO_MAXIRADIO is not set
+# CONFIG_RADIO_MAESTRO is not set
+# CONFIG_RADIO_SF16FMI is not set
+# CONFIG_RADIO_SF16FMR2 is not set
+# CONFIG_RADIO_TERRATEC is not set
+# CONFIG_RADIO_TRUST is not set
+# CONFIG_RADIO_TYPHOON is not set
+# CONFIG_RADIO_ZOLTRIX is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Console display driver support
+#
+# CONFIG_MDA_CONSOLE is not set
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FONT_8x16=y
+
+#
+# Sound
+#
+CONFIG_SOUND=m
+
+#
+# Advanced Linux Sound Architecture
+#
+# CONFIG_SND is not set
+
+#
+# Open Sound System
+#
+CONFIG_SOUND_PRIME=m
+# CONFIG_SOUND_BT878 is not set
+# CONFIG_SOUND_CMPCI is not set
+# CONFIG_SOUND_EMU10K1 is not set
+# CONFIG_SOUND_FUSION is not set
+# CONFIG_SOUND_CS4281 is not set
+# CONFIG_SOUND_ES1370 is not set
+# CONFIG_SOUND_ES1371 is not set
+# CONFIG_SOUND_ESSSOLO1 is not set
+# CONFIG_SOUND_MAESTRO is not set
+# CONFIG_SOUND_MAESTRO3 is not set
+# CONFIG_SOUND_ICH is not set
+# CONFIG_SOUND_SONICVIBES is not set
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+# CONFIG_SOUND_VIA82CXXX is not set
+# CONFIG_SOUND_ALI5455 is not set
+# CONFIG_SOUND_FORTE is not set
+# CONFIG_SOUND_RME96XX is not set
+# CONFIG_SOUND_AD1980 is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_BANDWIDTH is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
+CONFIG_USB_OHCI_HCD=y
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+
+#
+# USB Device Class drivers
+#
+CONFIG_OBSOLETE_OSS_USB_DRIVER=y
+CONFIG_USB_AUDIO=m
+# CONFIG_USB_BLUETOOTH_TTY is not set
+CONFIG_USB_MIDI=m
+# CONFIG_USB_ACM is not set
+CONFIG_USB_PRINTER=m
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=m
+# CONFIG_USB_STORAGE_DEBUG is not set
+CONFIG_USB_STORAGE_DATAFAB=y
+CONFIG_USB_STORAGE_FREECOM=y
+CONFIG_USB_STORAGE_ISD200=y
+CONFIG_USB_STORAGE_DPCM=y
+# CONFIG_USB_STORAGE_USBAT is not set
+CONFIG_USB_STORAGE_SDDR09=y
+CONFIG_USB_STORAGE_SDDR55=y
+CONFIG_USB_STORAGE_JUMPSHOT=y
+
+#
+# USB Input Devices
+#
+CONFIG_USB_HID=m
+CONFIG_USB_HIDINPUT=y
+# CONFIG_HID_FF is not set
+# CONFIG_USB_HIDDEV is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+# CONFIG_USB_AIPTEK is not set
+# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
+# CONFIG_USB_KBTAB is not set
+# CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
+# CONFIG_USB_EGALAX is not set
+# CONFIG_USB_YEALINK is not set
+# CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_KEYSPAN_REMOTE is not set
+# CONFIG_USB_APPLETOUCH is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB Multimedia devices
+#
+CONFIG_USB_DABUSB=m
+CONFIG_USB_VICAM=m
+CONFIG_USB_DSBR=m
+CONFIG_USB_IBMCAM=m
+CONFIG_USB_KONICAWC=m
+CONFIG_USB_OV511=m
+CONFIG_USB_SE401=m
+CONFIG_USB_SN9C102=m
+CONFIG_USB_STV680=m
+CONFIG_USB_PWC=m
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+CONFIG_USB_PEGASUS=m
+CONFIG_USB_RTL8150=m
+# CONFIG_USB_USBNET is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+CONFIG_USB_SERIAL=m
+# CONFIG_USB_SERIAL_GENERIC is not set
+# CONFIG_USB_SERIAL_AIRPRIME is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+# CONFIG_USB_SERIAL_CP2101 is not set
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+CONFIG_USB_SERIAL_FTDI_SIO=m
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+CONFIG_USB_SERIAL_PL2303=m
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OPTION is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+
+#
+# USB Miscellaneous drivers
+#
+CONFIG_USB_EMI62=m
+CONFIG_USB_EMI26=m
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGETKIT is not set
+# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+CONFIG_USB_SISUSBVGA=m
+CONFIG_USB_SISUSBVGA_CON=y
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+CONFIG_REISERFS_FS=y
+# CONFIG_REISERFS_CHECK is not set
+# CONFIG_REISERFS_PROC_INFO is not set
+# CONFIG_REISERFS_FS_XATTR is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+CONFIG_ROMFS_FS=y
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=m
+# CONFIG_JOLIET is not set
+# CONFIG_ZISOFS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=m
+# CONFIG_NTFS_DEBUG is not set
+CONFIG_NTFS_RW=y
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+# CONFIG_RELAYFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+CONFIG_UFS_FS=m
+CONFIG_UFS_FS_WRITE=y
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=m
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=m
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_LOCKD=m
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=m
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=m
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+CONFIG_SMB_FS=m
+# CONFIG_SMB_NLS_DEFAULT is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+CONFIG_NLS_CODEPAGE_932=y
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+# CONFIG_NLS_ISO8859_1 is not set
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_FRAME_POINTER is not set
+CONFIG_SH_STANDARD_BIOS=y
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_EARLY_PRINTK is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 182fe9092577..c503ca072e11 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -25,4 +25,4 @@ RTS7751R2D		SH_RTS7751R2D
 EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
-
+LANDISK			SH_LANDISK
diff --git a/include/asm-sh/landisk/gio.h b/include/asm-sh/landisk/gio.h
new file mode 100644
index 000000000000..3fce4c451a46
--- /dev/null
+++ b/include/asm-sh/landisk/gio.h
@@ -0,0 +1,45 @@
+#ifndef __ASM_SH_LANDISK_GIO_H
+#define __ASM_SH_LANDISK_GIO_H
+
+#include <linux/ioctl.h>
+
+/* version */
+#define VERSION_STR	"1.00"
+
+/* Driver name */
+#define GIO_DRIVER_NAME		"/dev/giodrv"
+
+/* Use 'k' as magic number */
+#define GIODRV_IOC_MAGIC  'k'
+
+#define GIODRV_IOCRESET    _IO(GIODRV_IOC_MAGIC, 0)
+/*
+ * S means "Set" through a ptr,
+ * T means "Tell" directly
+ * G means "Get" (to a pointed var)
+ * Q means "Query", response is on the return value
+ * X means "eXchange": G and S atomically
+ * H means "sHift": T and Q atomically
+ */
+#define GIODRV_IOCSGIODATA1   _IOW(GIODRV_IOC_MAGIC,  1, unsigned char *)
+#define GIODRV_IOCGGIODATA1   _IOR(GIODRV_IOC_MAGIC,  2, unsigned char *)
+#define GIODRV_IOCSGIODATA2   _IOW(GIODRV_IOC_MAGIC,  3, unsigned short *)
+#define GIODRV_IOCGGIODATA2   _IOR(GIODRV_IOC_MAGIC,  4, unsigned short *)
+#define GIODRV_IOCSGIODATA4   _IOW(GIODRV_IOC_MAGIC,  5, unsigned long *)
+#define GIODRV_IOCGGIODATA4   _IOR(GIODRV_IOC_MAGIC,  6, unsigned long *)
+#define GIODRV_IOCSGIOSETADDR _IOW(GIODRV_IOC_MAGIC,  7, unsigned long *)
+#define GIODRV_IOCHARDRESET   _IO(GIODRV_IOC_MAGIC, 8) /* debugging tool */
+
+#define GIODRV_IOCSGIO_LED    _IOW(GIODRV_IOC_MAGIC,  9, unsigned long *)
+#define GIODRV_IOCGGIO_LED    _IOR(GIODRV_IOC_MAGIC,  10, unsigned long *)
+#define GIODRV_IOCSGIO_BUZZER _IOW(GIODRV_IOC_MAGIC,  11, unsigned long *)
+#define GIODRV_IOCGGIO_LANDISK _IOR(GIODRV_IOC_MAGIC,  14, unsigned long *)
+#define GIODRV_IOCGGIO_BTN _IOR(GIODRV_IOC_MAGIC,  22, unsigned long *)
+#define GIODRV_IOCSGIO_BTNPID _IOW(GIODRV_IOC_MAGIC,  23, unsigned long *)
+#define GIODRV_IOCGGIO_BTNPID _IOR(GIODRV_IOC_MAGIC,  24, unsigned long *)
+
+#define GIODRV_IOC_MAXNR 8
+#define GIO_READ 0x00000000
+#define GIO_WRITE 0x00000001
+
+#endif /* __ASM_SH_LANDISK_GIO_H  */
diff --git a/include/asm-sh/landisk/ide.h b/include/asm-sh/landisk/ide.h
new file mode 100644
index 000000000000..6490e28415ed
--- /dev/null
+++ b/include/asm-sh/landisk/ide.h
@@ -0,0 +1,14 @@
+/*
+ * modifed by kogiidena
+ * 2005.03.03
+ */
+
+#ifndef __ASM_SH_LANDISK_IDE_H
+#define __ASM_SH_LANDISK_IDE_H
+
+/* Nothing to see here.. */
+#include <asm/landisk/iodata_landisk.h>
+#define IRQ_CFCARD	IRQ_FATA	/* CF Card IRQ */
+#define IRQ_PCMCIA	IRQ_ATA		/* PCMCIA IRQ */
+
+#endif /* __ASM_SH_LANDISK_IDE_H  */
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
new file mode 100644
index 000000000000..7189d3a36384
--- /dev/null
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -0,0 +1,78 @@
+#ifndef __ASM_SH_IODATA_LANDISK_H
+#define __ASM_SH_IODATA_LANDISK_H
+
+/*
+ * linux/include/asm-sh/landisk/iodata_landisk.h
+ *
+ * Copyright (C) 2000  Atom Create Engineering Co., Ltd.
+ *
+ * IO-DATA LANDISK support
+ */
+
+/* Box specific addresses.  */
+
+#define PA_USB		0xa4000000	/* USB Controller M66590 */
+
+#define PA_ATARST	0xb0000000	/* ATA/FATA Access Control Register */
+#define PA_LED		0xb0000001	/* LED Control Register */
+#define PA_STATUS	0xb0000002	/* Switch Status Register */
+#define PA_SHUTDOWN	0xb0000003	/* Shutdown Control Register */
+#define PA_PCIPME	0xb0000004	/* PCI PME Status Register */
+#define PA_IMASK	0xb0000005	/* Interrupt Mask Register */
+/* 2003.10.31 I-O DATA NSD NWG	add.	for shutdown port clear */
+#define PA_PWRINT_CLR	0xb0000006	/* Shutdown Interrupt clear Register */
+
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+#define PA_LCD_CLRDSP	0x00		/* LCD Clear Display Offset */
+#define PA_LCD_RTNHOME	0x00		/* LCD Return Home Offset */
+#define PA_LCD_ENTMODE	0x00		/* LCD Entry Mode Offset */
+#define PA_LCD_DSPCTL	0x00		/* LCD Display ON/OFF Control Offset */
+#define PA_LCD_FUNC	0x00		/* LCD Function Set Offset */
+#define PA_LCD_CGRAM	0x00		/* LCD Set CGRAM Address Offset */
+#define PA_LCD_DDRAM	0x00		/* LCD Set DDRAM Address Offset */
+#define PA_LCD_RDFLAG	0x01		/* LCD Read Busy Flag Offset */
+#define PA_LCD_WTDATA	0x02		/* LCD Write Datat to RAM Offset */
+#define PA_LCD_RDDATA	0x03		/* LCD Read Data from RAM Offset */
+#define PA_PIDE_OFFSET	0x40		/* CF IDE Offset */
+#define PA_SIDE_OFFSET	0x40		/* HDD IDE Offset */
+
+#define IRQ_PCIINTA	5		/* PCI INTA IRQ */
+#define IRQ_PCIINTB	6		/* PCI INTB IRQ */
+#define IRQ_PCIINDC	7		/* PCI INTC IRQ */
+#define IRQ_PCIINTD	8		/* PCI INTD IRQ */
+#define IRQ_ATA		9		/* ATA IRQ */
+#define IRQ_FATA	10		/* FATA IRQ */
+#define IRQ_POWER	11		/* Power Switch IRQ */
+#define IRQ_BUTTON	12		/* USL-5P Button IRQ */
+#define IRQ_FAULT	13		/* USL-5P Fault  IRQ */
+
+#define SHUTDOWN_BTN_MAJOR	99	/* Shutdown button device major no. */
+
+#define SHUTDOWN_LOOP_CNT	5	/* Shutdown button Detection loop */
+#define SHUTDOWN_DELAY		200	/* Shutdown button delay value(ms) */
+
+
+/* added by kogiidena */
+/*
+ *  landisk_ledparam
+ *
+ * led  ------10 -6543210 -6543210 -6543210
+ *     |000000..|0.......|0.......|U.......|
+ *     |  HARD  |fastblik| blink  |   on   |
+ *
+ *   led0: power       U:update flag
+ *   led1: error
+ *   led2: usb1
+ *   led3: usb2
+ *   led4: usb3
+ *   led5: usb4
+ *   led6: usb5
+ *
+ */
+extern int landisk_ledparam;    /* from setup.c */
+extern int landisk_buzzerparam; /* from setup.c */
+extern int landisk_arch;        /* from setup.c */
+
+#endif  /* __ASM_SH_IODATA_LANDISK_H */
+
-- 
cgit v1.2.1


From 749cf486920bf53f16e6a6889d9635a91ffb6c82 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:55:41 +0900
Subject: sh: Add flag for MMU PTEA capability.

Add CPU_HAS_PTEA, refactor some of the cpu flag settings.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c | 34 ++++++++++++++++++++++++----------
 arch/sh/kernel/cpu/sh4/sq.c    |  6 +++---
 arch/sh/mm/tlb-sh4.c           | 12 ++++--------
 3 files changed, 31 insertions(+), 21 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 85ff48c1533a..2a7707a81d8f 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -3,7 +3,7 @@
  *
  * CPU Subtype Probing for SH-4.
  *
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001 - 2005  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -53,9 +53,6 @@ int __init detect_cpu_and_cache_system(void)
 	cpu_data->dcache.ways		= 1;
 	cpu_data->dcache.linesz		= L1_CACHE_BYTES;
 
-	/* Set the FPU flag, virtually all SH-4's have one */
-	cpu_data->flags |= CPU_HAS_FPU;
-
 	/*
 	 * Probe the underlying processor version/revision and
 	 * adjust cpu_data setup accordingly.
@@ -63,26 +60,37 @@ int __init detect_cpu_and_cache_system(void)
 	switch (pvr) {
 	case 0x205:
 		cpu_data->type = CPU_SH7750;
-		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_FPU |
+				   CPU_HAS_PERF_COUNTER | CPU_HAS_PTEA;
 		break;
 	case 0x206:
 		cpu_data->type = CPU_SH7750S;
-		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_P2_FLUSH_BUG | CPU_HAS_FPU |
+				   CPU_HAS_PERF_COUNTER | CPU_HAS_PTEA;
 		break;
 	case 0x1100:
 		cpu_data->type = CPU_SH7751;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x2000:
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
-		cpu_data->flags &= ~CPU_HAS_FPU;
+
+		/*
+		 * XXX: Double check this, none of the SH-4A/SH-4AL processors
+		 * should have this, as it's essentially a legacy thing.
+		 */
+		cpu_data->flags |= CPU_HAS_PTEA;
 		break;
 	case 0x2001:
 	case 0x2004:
 		cpu_data->type = CPU_SH7770;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+
+		/* Same note as above applies here for PTEA */
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x2006:
 	case 0x200A:
@@ -90,27 +98,31 @@ int __init detect_cpu_and_cache_system(void)
 			cpu_data->type = CPU_SH7781;
 		else
 			cpu_data->type = CPU_SH7780;
+
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER;
 		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x8100:
 		cpu_data->type = CPU_ST40GX1;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x700:
 		cpu_data->type = CPU_SH4_501;
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
-
-		/* No FPU on the SH4-500 series.. */
-		cpu_data->flags &= ~CPU_HAS_FPU;
+		cpu_data->flags |= CPU_HAS_PTEA;
 		break;
 	case 0x600:
 		cpu_data->type = CPU_SH4_202;
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
 		break;
 	case 0x500 ... 0x501:
 		switch (prr) {
@@ -128,6 +140,8 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 2;
 		cpu_data->dcache.ways = 2;
 
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
+
 		break;
 	default:
 		cpu_data->type = CPU_SH_NONE;
diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index 32c93b781b51..b148966dd7c7 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -166,9 +166,9 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
 	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
 
 	ptel = map->addr & PAGE_MASK;
-#ifndef CONFIG_CPU_SUBTYPE_SH7780
-	ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
-#endif
+
+	if (cpu_data->flags & CPU_HAS_PTEA)
+		ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
 
 	pgprot = pgprot_noncached(PAGE_KERNEL);
 
diff --git a/arch/sh/mm/tlb-sh4.c b/arch/sh/mm/tlb-sh4.c
index 96e5fb0ac4fa..812b2d567de2 100644
--- a/arch/sh/mm/tlb-sh4.c
+++ b/arch/sh/mm/tlb-sh4.c
@@ -36,9 +36,6 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	unsigned long vpn;
 	struct page *page;
 	unsigned long pfn;
-#ifndef CONFIG_CPU_SUBTYPE_SH7780
-	unsigned long ptea;
-#endif
 
 	/* Ptrace may call this routine. */
 	if (vma && current->active_mm != vma->vm_mm)
@@ -61,12 +58,11 @@ void update_mmu_cache(struct vm_area_struct * vma,
 	ctrl_outl(vpn, MMU_PTEH);
 
 	pteval = pte_val(pte);
-#ifndef CONFIG_CPU_SUBTYPE_SH7780
+
 	/* Set PTEA register */
-	/* TODO: make this look less hacky */
-	ptea = ((pteval >> 28) & 0xe) | (pteval & 0x1);
-	ctrl_outl(ptea, MMU_PTEA);
-#endif
+	if (cpu_data->flags & CPU_HAS_PTEA)
+		/* TODO: make this look less hacky */
+		ctrl_outl(((pteval >> 28) & 0xe) | (pteval & 0x1), MMU_PTEA);
 
 	/* Set PTEL register */
 	pteval &= _PAGE_FLAGS_HARDWARE_MASK; /* drop software flags */
-- 
cgit v1.2.1


From 298476220d1f793ca0ac6c9e5dc817e1ad3e9851 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 14:57:44 +0900
Subject: sh: Add control register barriers.

Currently when making changes to control registers, we
typically need some time for changes to take effect (8
nops, generally).  However, for sh4a we simply need to
do an icbi..

This is a simple patch for implementing a general purpose
ctrl_barrier() which functions as a control register write
barrier. There's some additional documentation in the patch
itself, but it's pretty self explanatory.

There were also some places where we were not doing the
barrier, which didn't seem to have any adverse effects on
legacy parts, but certainly did on sh4a. It's safer to have
the barrier in place for legacy parts as well in these cases,
though this does make flush_tlb_all() more expensive (by an
order of 8 nops).  We can ifdef around the flush_tlb_all()
case for now if it's clear that all legacy parts won't have
a problem with this.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c       |  5 +++++
 arch/sh/mm/fault.c           |  5 +++--
 include/asm-sh/mmu_context.h |  7 +++----
 include/asm-sh/system.h      | 43 ++++++++++++++++++++++++++++++++++---------
 4 files changed, 45 insertions(+), 15 deletions(-)

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index c036c2b4ac2b..2203bd6aadb3 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -269,6 +269,11 @@ static inline void flush_icache_all(void)
 	ccr |= CCR_CACHE_ICI;
 	ctrl_outl(ccr, CCR);
 
+	/*
+	 * back_to_P1() will take care of the barrier for us, don't add
+	 * another one!
+	 */
+
 	back_to_P1();
 	local_irq_restore(flags);
 }
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 775f86cd3fe8..364181f27b79 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -369,12 +369,13 @@ void flush_tlb_all(void)
 	 * Flush all the TLB.
 	 *
 	 * Write to the MMU control register's bit:
-	 * 	TF-bit for SH-3, TI-bit for SH-4.
+	 *	TF-bit for SH-3, TI-bit for SH-4.
 	 *      It's same position, bit #2.
 	 */
 	local_irq_save(flags);
 	status = ctrl_inl(MMUCR);
-	status |= 0x04;		
+	status |= 0x04;
 	ctrl_outl(status, MMUCR);
+	ctrl_barrier();
 	local_irq_restore(flags);
 }
diff --git a/include/asm-sh/mmu_context.h b/include/asm-sh/mmu_context.h
index 6760d064bd02..87678ba8d6b6 100644
--- a/include/asm-sh/mmu_context.h
+++ b/include/asm-sh/mmu_context.h
@@ -174,9 +174,7 @@ static inline void enable_mmu(void)
 {
 	/* Enable MMU */
 	ctrl_outl(MMU_CONTROL_INIT, MMUCR);
-
-	/* The manual suggests doing some nops after turning on the MMU */
-	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop\n\t");
+	ctrl_barrier();
 
 	if (mmu_context_cache == NO_CONTEXT)
 		mmu_context_cache = MMU_CONTEXT_FIRST_VERSION;
@@ -191,7 +189,8 @@ static inline void disable_mmu(void)
 	cr = ctrl_inl(MMUCR);
 	cr &= ~MMU_CONTROL_INIT;
 	ctrl_outl(cr, MMUCR);
-	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop\n\t");
+
+	ctrl_barrier();
 }
 #else
 /*
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index eb4902ed920a..1630a5411e5f 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -67,8 +67,17 @@ static inline void sched_cacheflush(void)
 {
 }
 
-#define nop() __asm__ __volatile__ ("nop")
-
+#ifdef CONFIG_CPU_SH4A
+#define __icbi()			\
+{					\
+	unsigned long __addr;		\
+	__addr = 0xa8000000;		\
+	__asm__ __volatile__(		\
+		"icbi   %0\n\t"		\
+		: /* no output */	\
+		: "m" (__m(__addr)));	\
+}
+#endif
 
 #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
 
@@ -84,15 +93,31 @@ static __inline__ unsigned long tas(volatile int *m)
 
 extern void __xchg_called_with_bad_pointer(void);
 
+/*
+ * A brief note on ctrl_barrier(), the control register write barrier.
+ *
+ * Legacy SH cores typically require a sequence of 8 nops after
+ * modification of a control register in order for the changes to take
+ * effect. On newer cores (like the sh4a and sh5) this is accomplished
+ * with icbi.
+ *
+ * Also note that on sh4a in the icbi case we can forego a synco for the
+ * write barrier, as it's not necessary for control registers.
+ *
+ * Historically we have only done this type of barrier for the MMUCR, but
+ * it's also necessary for the CCR, so we make it generic here instead.
+ */
 #ifdef CONFIG_CPU_SH4A
-#define mb()	__asm__ __volatile__ ("synco": : :"memory")
-#define rmb()	mb()
-#define wmb()	__asm__ __volatile__ ("synco": : :"memory")
+#define mb()		__asm__ __volatile__ ("synco": : :"memory")
+#define rmb()		mb()
+#define wmb()		__asm__ __volatile__ ("synco": : :"memory")
+#define ctrl_barrier()	__icbi()
 #define read_barrier_depends()	do { } while(0)
 #else
-#define mb()	__asm__ __volatile__ ("": : :"memory")
-#define rmb()	mb()
-#define wmb()	__asm__ __volatile__ ("": : :"memory")
+#define mb()		__asm__ __volatile__ ("": : :"memory")
+#define rmb()		mb()
+#define wmb()		__asm__ __volatile__ ("": : :"memory")
+#define ctrl_barrier()	__asm__ __volatile__ ("nop;nop;nop;nop;nop;nop;nop;nop")
 #define read_barrier_depends()	do { } while(0)
 #endif
 
@@ -218,8 +243,8 @@ do {					\
 #define back_to_P1()					\
 do {							\
 	unsigned long __dummy;				\
+	ctrl_barrier();					\
 	__asm__ __volatile__(				\
-		"nop;nop;nop;nop;nop;nop;nop\n\t"	\
 		"mov.l	1f, %0\n\t"			\
 		"jmp	@%0\n\t"			\
 		" nop\n\t"				\
-- 
cgit v1.2.1


From b7e108ee63624176af85b97d4d80bef6fe099395 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:00:04 +0900
Subject: sh: BSS init bugfix and barrier in entry point.

A synco is needed before we jump to start_kernel().

While we're at it, also move the sh_cpu_init() jump until after
we've zeroed BSS, as this has caused some undesirable results
in sh_cpu_init().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/head.S | 36 +++++++++++++++++++++++++++++++-----
 1 file changed, 31 insertions(+), 5 deletions(-)

diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 00cd4708ef46..c5e363872b91 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -12,6 +12,17 @@
  */
 #include <linux/linkage.h>
 
+#ifdef CONFIG_CPU_SH4A
+#define SYNCO()		synco
+
+#define PREFI(label, reg)	\
+	mov.l	label, reg;	\
+	prefi	@reg
+#else
+#define SYNCO()
+#define PREFI(label, reg)
+#endif
+
 	.section	.empty_zero_page, "aw"
 ENTRY(empty_zero_page)
 	.long	1		/* MOUNT_ROOT_RDONLY */
@@ -42,6 +53,17 @@ ENTRY(_stext)
 	!			Initialize global interrupt mask
 	mov	#0, r0
 	ldc	r0, r6_bank
+
+	/*
+	 * Prefetch if possible to reduce cache miss penalty.
+	 *
+	 * We do this early on for SH-4A as a micro-optimization,
+	 * as later on we will have speculative execution enabled
+	 * and this will become less of an issue.
+	 */
+	PREFI(5f, r0)
+	PREFI(6f, r0)
+
 	!
 	mov.l	2f, r0
 	mov	r0, r15		! Set initial r15 (stack pointer)
@@ -49,11 +71,7 @@ ENTRY(_stext)
 	shll8	r1		! r1 = 8192
 	sub	r1, r0		!
 	ldc	r0, r7_bank	! ... and initial thread_info
-	!
-	!			Additional CPU initialization
-	mov.l	6f, r0
-	jsr	@r0
-	 nop
+
 	!			Clear BSS area
 	mov.l	3f, r1
 	add	#4, r1
@@ -62,6 +80,14 @@ ENTRY(_stext)
 9:	cmp/hs	r2, r1
 	bf/s	9b		! while (r1 < r2)
 	 mov.l	r0,@-r2
+
+	!			Additional CPU initialization
+	mov.l	6f, r0
+	jsr	@r0
+	 nop
+
+	SYNCO()			! Wait for pending instructions..
+
 	!			Start kernel
 	mov.l	5f, r0
 	jmp	@r0
-- 
cgit v1.2.1


From a09749dd86e9e93de10f12ab4ce4e90815b5650a Mon Sep 17 00:00:00 2001
From: Jamie Lenehan <lenehan@twibble.org>
Date: Wed, 27 Sep 2006 15:05:39 +0900
Subject: sh: Titan board support.

Add support for the titan board.

Signed-off-by: Jamie Lenehan <lenehan@twibble.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                |    1 +
 arch/sh/boards/titan/Makefile   |    5 +
 arch/sh/boards/titan/io.c       |  156 +++++
 arch/sh/boards/titan/setup.c    |   60 ++
 arch/sh/configs/titan_defconfig | 1367 +++++++++++++++++++++++++++++++++++++++
 arch/sh/drivers/pci/Makefile    |    1 +
 arch/sh/drivers/pci/ops-titan.c |   84 +++
 arch/sh/drivers/pci/pci.c       |   58 +-
 arch/sh/tools/mach-types        |    1 +
 include/asm-sh/titan.h          |   43 ++
 10 files changed, 1764 insertions(+), 12 deletions(-)
 create mode 100644 arch/sh/boards/titan/Makefile
 create mode 100644 arch/sh/boards/titan/io.c
 create mode 100644 arch/sh/boards/titan/setup.c
 create mode 100644 arch/sh/configs/titan_defconfig
 create mode 100644 arch/sh/drivers/pci/ops-titan.c
 create mode 100644 include/asm-sh/titan.h

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index e9112def48dd..55dcba2c06fc 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -106,6 +106,7 @@ machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
+machdir-$(CONFIG_SH_TITAN)			:= titan
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/titan/Makefile b/arch/sh/boards/titan/Makefile
new file mode 100644
index 000000000000..08d753700062
--- /dev/null
+++ b/arch/sh/boards/titan/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the Nimble Microsystems TITAN specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
new file mode 100644
index 000000000000..d66900c99a11
--- /dev/null
+++ b/arch/sh/boards/titan/io.c
@@ -0,0 +1,156 @@
+/*
+ *	I/O routines for Titan
+ */
+
+#include <linux/pci.h>
+#include <asm/machvec.h>
+#include <asm/addrspace.h>
+#include <asm/titan.h>
+#include <asm/io.h>
+#include "../../drivers/pci/pci-sh7751.h"
+
+#define PCIIOBR         (volatile long *)PCI_REG(SH7751_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
+#define PCI_IO_AREA     SH7751_PCI_IO_BASE
+#define PCI_MEM_AREA    SH7751_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)  (PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
+
+#if defined(CONFIG_PCI)
+#define CHECK_SH7751_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#define CHECK_SH7751_PCIMEMIO(port) \
+  ((port >= PCIBIOS_MIN_MEM) && (port < (PCIBIOS_MIN_MEM + SH7751_PCI_MEM_SIZE)))
+#else
+#define CHECK_SH7751_PCIIO(port) (0)
+#endif
+
+static inline void delay(void)
+{
+        ctrl_inw(0xa0000000);
+}
+
+static inline volatile u16 *port2adr(unsigned int port)
+{
+        maybebadio((unsigned long)port);
+        return (volatile u16*)port;
+}
+
+u8 titan_inb(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inb(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inb(PCI_IOMAP(port));
+        return ctrl_inw(port2adr(port)) & 0xff;
+}
+
+u8 titan_inb_p(unsigned long port)
+{
+        u8 v;
+
+        if (PXSEG(port))
+                v = ctrl_inb(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                v = ctrl_inb(PCI_IOMAP(port));
+        else
+                v = ctrl_inw(port2adr(port)) & 0xff;
+        delay();
+        return v;
+}
+
+u16 titan_inw(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inw(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inw(PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                return ctrl_inw(port2adr(port));
+        else
+                maybebadio(port);
+        return 0;
+}
+
+u32 titan_inl(unsigned long port)
+{
+        if (PXSEG(port))
+                return ctrl_inl(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return ctrl_inl(PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                return ctrl_inw(port2adr(port));
+        else
+                maybebadio(port);
+        return 0;
+}
+
+void titan_outb(u8 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outb(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outb(value, PCI_IOMAP(port));
+        else
+                ctrl_outw(value, port2adr(port));
+}
+
+void titan_outb_p(u8 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outb(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outb(value, PCI_IOMAP(port));
+        else
+                ctrl_outw(value, port2adr(port));
+        delay();
+}
+
+void titan_outw(u16 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outw(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outw(value, PCI_IOMAP(port));
+        else if (port >= 0x2000)
+                ctrl_outw(value, port2adr(port));
+        else
+                maybebadio(port);
+}
+
+void titan_outl(u32 value, unsigned long port)
+{
+        if (PXSEG(port))
+                ctrl_outl(value, port);
+        else if (CHECK_SH7751_PCIIO(port))
+                ctrl_outl(value, PCI_IOMAP(port));
+        else
+                maybebadio(port);
+}
+
+void titan_insl(unsigned long port, void *dst, unsigned long count)
+{
+        maybebadio(port);
+}
+
+void titan_outsl(unsigned long port, const void *src, unsigned long count)
+{
+        maybebadio(port);
+}
+
+void *titan_ioremap(unsigned long offset, unsigned long size) {
+	if (CHECK_SH7751_PCIIO(offset) || CHECK_SH7751_PCIMEMIO(offset))
+		return (void *)offset;
+}
+
+void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
+{
+	if (PXSEG(port))
+		return (void __iomem *)port;
+	else if (CHECK_SH7751_PCIIO(port))
+		return (void __iomem *)PCI_IOMAP(port);
+
+	return (void __iomem *)port2adr(port);
+}
+
+EXPORT_SYMBOL(titan_ioremap);
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
new file mode 100644
index 000000000000..6ac5c8d7b3fb
--- /dev/null
+++ b/arch/sh/boards/titan/setup.c
@@ -0,0 +1,60 @@
+/*
+ *	Setup for Titan
+ */
+
+#include <linux/init.h>
+#include <asm/irq.h>
+#include <asm/titan.h>
+#include <asm/io.h>
+
+extern void __init pcibios_init_platform(void);
+
+static void __init init_titan_irq(void)
+{
+	/* enable individual interrupt mode for externals */
+	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
+
+	make_ipr_irq( TITAN_IRQ_WAN,   IRL0_IPR_ADDR, IRL0_IPR_POS, IRL0_PRIORITY); /* PCIRQ0 */
+	make_ipr_irq( TITAN_IRQ_LAN,   IRL1_IPR_ADDR, IRL1_IPR_POS, IRL1_PRIORITY); /* PCIRQ1 */
+	make_ipr_irq( TITAN_IRQ_MPCIA, IRL2_IPR_ADDR, IRL2_IPR_POS, IRL2_PRIORITY); /* PCIRQ2 */
+	make_ipr_irq( TITAN_IRQ_USB,   IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); /* PCIRQ3 */
+}
+
+const char *get_system_type(void)
+{
+	return "Titan";
+}
+
+int __init platform_setup(void)
+{
+	printk("%s Platform Setup\n", get_system_type());
+	return 0;
+}
+
+struct sh_machine_vector mv_titan __initmv = {
+	.mv_nr_irqs =	NR_IRQS,
+
+	.mv_inb =	titan_inb,
+	.mv_inw =	titan_inw,
+	.mv_inl =	titan_inl,
+	.mv_outb =	titan_outb,
+	.mv_outw =	titan_outw,
+	.mv_outl =	titan_outl,
+
+	.mv_inb_p =	titan_inb_p,
+	.mv_inw_p =	titan_inw,
+	.mv_inl_p =	titan_inl,
+	.mv_outb_p =	titan_outb_p,
+	.mv_outw_p =	titan_outw,
+	.mv_outl_p =	titan_outl,
+
+	.mv_insl =	titan_insl,
+	.mv_outsl =	titan_outsl,
+
+	.mv_ioremap =	titan_ioremap,
+	.mv_ioport_map = titan_ioport_map,
+
+	.mv_init_irq =	init_titan_irq,
+	.mv_init_pci =	pcibios_init_platform,
+};
+ALIAS_MV(titan)
diff --git a/arch/sh/configs/titan_defconfig b/arch/sh/configs/titan_defconfig
new file mode 100644
index 000000000000..1db2904de9e5
--- /dev/null
+++ b/arch/sh/configs/titan_defconfig
@@ -0,0 +1,1367 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.14-sh
+# Wed Nov  9 00:35:56 2005
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_IOMAP=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_EMBEDDED is not set
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_EPOLL=y
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+CONFIG_SH_TITAN=y
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+CONFIG_CPU_SUBTYPE_SH7751=y
+CONFIG_CPU_SUBTYPE_SH7751R=y
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x08030000
+CONFIG_MEMORY_SIZE=0x7fd0000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_RTC=y
+CONFIG_SH_FPU=y
+# CONFIG_SH_STORE_QUEUES is not set
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=30000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=8
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x009e0000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttySC1,38400N81 root=/dev/nfs ip=:::::eth1:autoconf"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+#CONFIG_PCI_DEBUG is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_HOTPLUG_PCI_FAKE is not set
+# CONFIG_HOTPLUG_PCI_CPCI is not set
+# CONFIG_HOTPLUG_PCI_SHPC is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+# CONFIG_XFRM_USER is not set
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_MULTIPLE_TABLES=y
+# CONFIG_IP_ROUTE_FWMARK is not set
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_MULTIPATH_CACHED=y
+CONFIG_IP_ROUTE_MULTIPATH_RR=m
+CONFIG_IP_ROUTE_MULTIPATH_RANDOM=m
+CONFIG_IP_ROUTE_MULTIPATH_WRANDOM=m
+CONFIG_IP_ROUTE_MULTIPATH_DRR=m
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+CONFIG_NET_IPIP=m
+CONFIG_NET_IPGRE=m
+CONFIG_NET_IPGRE_BROADCAST=y
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+CONFIG_INET_AH=m
+CONFIG_INET_ESP=m
+CONFIG_INET_IPCOMP=m
+CONFIG_INET_TUNNEL=m
+CONFIG_INET_DIAG=m
+CONFIG_INET_TCP_DIAG=m
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+
+#
+# IP: Virtual Server Configuration
+#
+# CONFIG_IP_VS is not set
+CONFIG_IPV6=m
+CONFIG_IPV6_PRIVACY=y
+CONFIG_INET6_AH=m
+CONFIG_INET6_ESP=m
+CONFIG_INET6_IPCOMP=m
+CONFIG_INET6_TUNNEL=m
+CONFIG_IPV6_TUNNEL=m
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+CONFIG_BRIDGE_NETFILTER=y
+CONFIG_NETFILTER_NETLINK=m
+CONFIG_NETFILTER_NETLINK_QUEUE=m
+CONFIG_NETFILTER_NETLINK_LOG=m
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_CONNTRACK=m
+CONFIG_IP_NF_CT_ACCT=y
+CONFIG_IP_NF_CONNTRACK_MARK=y
+CONFIG_IP_NF_CONNTRACK_EVENTS=y
+CONFIG_IP_NF_CONNTRACK_NETLINK=m
+# CONFIG_IP_NF_CT_PROTO_SCTP is not set
+CONFIG_IP_NF_FTP=m
+CONFIG_IP_NF_IRC=m
+CONFIG_IP_NF_NETBIOS_NS=m
+CONFIG_IP_NF_TFTP=m
+# CONFIG_IP_NF_AMANDA is not set
+CONFIG_IP_NF_PPTP=m
+CONFIG_IP_NF_QUEUE=m
+CONFIG_IP_NF_IPTABLES=m
+CONFIG_IP_NF_MATCH_LIMIT=m
+CONFIG_IP_NF_MATCH_IPRANGE=m
+CONFIG_IP_NF_MATCH_MAC=m
+CONFIG_IP_NF_MATCH_PKTTYPE=m
+CONFIG_IP_NF_MATCH_MARK=m
+CONFIG_IP_NF_MATCH_MULTIPORT=m
+CONFIG_IP_NF_MATCH_TOS=m
+CONFIG_IP_NF_MATCH_RECENT=m
+CONFIG_IP_NF_MATCH_ECN=m
+CONFIG_IP_NF_MATCH_DSCP=m
+CONFIG_IP_NF_MATCH_AH_ESP=m
+CONFIG_IP_NF_MATCH_LENGTH=m
+CONFIG_IP_NF_MATCH_TTL=m
+CONFIG_IP_NF_MATCH_TCPMSS=m
+CONFIG_IP_NF_MATCH_HELPER=m
+CONFIG_IP_NF_MATCH_STATE=m
+CONFIG_IP_NF_MATCH_CONNTRACK=m
+CONFIG_IP_NF_MATCH_OWNER=m
+# CONFIG_IP_NF_MATCH_PHYSDEV is not set
+CONFIG_IP_NF_MATCH_ADDRTYPE=m
+CONFIG_IP_NF_MATCH_REALM=m
+# CONFIG_IP_NF_MATCH_SCTP is not set
+# CONFIG_IP_NF_MATCH_DCCP is not set
+CONFIG_IP_NF_MATCH_COMMENT=m
+CONFIG_IP_NF_MATCH_CONNMARK=m
+CONFIG_IP_NF_MATCH_CONNBYTES=m
+CONFIG_IP_NF_MATCH_HASHLIMIT=m
+CONFIG_IP_NF_MATCH_STRING=m
+CONFIG_IP_NF_FILTER=m
+CONFIG_IP_NF_TARGET_REJECT=m
+CONFIG_IP_NF_TARGET_LOG=m
+CONFIG_IP_NF_TARGET_ULOG=m
+CONFIG_IP_NF_TARGET_TCPMSS=m
+CONFIG_IP_NF_TARGET_NFQUEUE=m
+CONFIG_IP_NF_NAT=m
+CONFIG_IP_NF_NAT_NEEDED=y
+CONFIG_IP_NF_TARGET_MASQUERADE=m
+CONFIG_IP_NF_TARGET_REDIRECT=m
+CONFIG_IP_NF_TARGET_NETMAP=m
+CONFIG_IP_NF_TARGET_SAME=m
+CONFIG_IP_NF_NAT_SNMP_BASIC=m
+CONFIG_IP_NF_NAT_IRC=m
+CONFIG_IP_NF_NAT_FTP=m
+CONFIG_IP_NF_NAT_TFTP=m
+CONFIG_IP_NF_NAT_PPTP=m
+CONFIG_IP_NF_MANGLE=m
+CONFIG_IP_NF_TARGET_TOS=m
+CONFIG_IP_NF_TARGET_ECN=m
+CONFIG_IP_NF_TARGET_DSCP=m
+CONFIG_IP_NF_TARGET_MARK=m
+CONFIG_IP_NF_TARGET_CLASSIFY=m
+CONFIG_IP_NF_TARGET_TTL=m
+CONFIG_IP_NF_TARGET_CONNMARK=m
+CONFIG_IP_NF_TARGET_CLUSTERIP=m
+CONFIG_IP_NF_RAW=m
+CONFIG_IP_NF_TARGET_NOTRACK=m
+CONFIG_IP_NF_ARPTABLES=m
+CONFIG_IP_NF_ARPFILTER=m
+CONFIG_IP_NF_ARP_MANGLE=m
+
+#
+# IPv6: Netfilter Configuration (EXPERIMENTAL)
+#
+CONFIG_IP6_NF_QUEUE=m
+CONFIG_IP6_NF_IPTABLES=m
+CONFIG_IP6_NF_MATCH_LIMIT=m
+CONFIG_IP6_NF_MATCH_MAC=m
+CONFIG_IP6_NF_MATCH_RT=m
+CONFIG_IP6_NF_MATCH_OPTS=m
+CONFIG_IP6_NF_MATCH_FRAG=m
+CONFIG_IP6_NF_MATCH_HL=m
+CONFIG_IP6_NF_MATCH_MULTIPORT=m
+CONFIG_IP6_NF_MATCH_OWNER=m
+CONFIG_IP6_NF_MATCH_MARK=m
+CONFIG_IP6_NF_MATCH_IPV6HEADER=m
+CONFIG_IP6_NF_MATCH_AHESP=m
+CONFIG_IP6_NF_MATCH_LENGTH=m
+CONFIG_IP6_NF_MATCH_EUI64=m
+# CONFIG_IP6_NF_MATCH_PHYSDEV is not set
+CONFIG_IP6_NF_FILTER=m
+CONFIG_IP6_NF_TARGET_LOG=m
+CONFIG_IP6_NF_TARGET_REJECT=m
+CONFIG_IP6_NF_TARGET_NFQUEUE=m
+CONFIG_IP6_NF_MANGLE=m
+CONFIG_IP6_NF_TARGET_MARK=m
+CONFIG_IP6_NF_TARGET_HL=m
+CONFIG_IP6_NF_RAW=m
+
+#
+# Bridge: Netfilter Configuration
+#
+# CONFIG_BRIDGE_NF_EBTABLES is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+CONFIG_BRIDGE=m
+CONFIG_VLAN_8021Q=m
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_CLK_JIFFIES=y
+# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set
+# CONFIG_NET_SCH_CLK_CPU is not set
+CONFIG_NET_SCH_CBQ=m
+CONFIG_NET_SCH_HTB=m
+CONFIG_NET_SCH_HFSC=m
+CONFIG_NET_SCH_PRIO=m
+CONFIG_NET_SCH_RED=m
+CONFIG_NET_SCH_SFQ=m
+CONFIG_NET_SCH_TEQL=m
+CONFIG_NET_SCH_TBF=m
+CONFIG_NET_SCH_GRED=m
+CONFIG_NET_SCH_DSMARK=m
+CONFIG_NET_SCH_NETEM=m
+CONFIG_NET_SCH_INGRESS=m
+CONFIG_NET_QOS=y
+CONFIG_NET_ESTIMATOR=y
+CONFIG_NET_CLS=y
+CONFIG_NET_CLS_BASIC=m
+CONFIG_NET_CLS_TCINDEX=m
+CONFIG_NET_CLS_ROUTE4=m
+CONFIG_NET_CLS_ROUTE=y
+CONFIG_NET_CLS_FW=m
+CONFIG_NET_CLS_U32=m
+CONFIG_CLS_U32_PERF=y
+CONFIG_NET_CLS_IND=y
+CONFIG_CLS_U32_MARK=y
+CONFIG_NET_CLS_RSVP=m
+CONFIG_NET_CLS_RSVP6=m
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_STACK=32
+CONFIG_NET_EMATCH_CMP=m
+CONFIG_NET_EMATCH_NBYTE=m
+CONFIG_NET_EMATCH_U32=m
+CONFIG_NET_EMATCH_META=m
+CONFIG_NET_EMATCH_TEXT=m
+CONFIG_NET_CLS_ACT=y
+CONFIG_NET_ACT_POLICE=m
+CONFIG_NET_ACT_GACT=m
+CONFIG_GACT_PROB=y
+CONFIG_NET_ACT_MIRRED=m
+CONFIG_NET_ACT_IPT=m
+CONFIG_NET_ACT_PEDIT=m
+# CONFIG_NET_ACT_SIMP is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+CONFIG_CONNECTOR=m
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+CONFIG_BLK_SSFDC=y
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=m
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_INITRD=y
+# CONFIG_LBD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+CONFIG_IOSCHED_AS=y
+CONFIG_IOSCHED_DEADLINE=y
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=m
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=m
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+CONFIG_BLK_DEV_SR=m
+# CONFIG_BLK_DEV_SR_VENDOR is not set
+CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=m
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+CONFIG_TUN=m
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+CONFIG_PHYLIB=m
+CONFIG_PHYCONTROL=y
+
+#
+# MII PHY device drivers
+#
+CONFIG_MARVELL_PHY=m
+CONFIG_DAVICOM_PHY=m
+CONFIG_QSEMI_PHY=m
+CONFIG_LXT_PHY=m
+CONFIG_CICADA_PHY=m
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+CONFIG_CASSINI=m
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_SMC91X is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+# CONFIG_NE2000 is not set
+CONFIG_NET_PCI=y
+# CONFIG_PCNET32 is not set
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+# CONFIG_8139CP is not set
+CONFIG_8139TOO=y
+# CONFIG_8139TOO_PIO is not set
+CONFIG_8139TOO_TUNE_TWISTER=y
+# CONFIG_8139TOO_8129 is not set
+CONFIG_8139_OLD_RX_RESET=y
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+# CONFIG_VIA_RHINE is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+# CONFIG_E1000 is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+# CONFIG_R8169 is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+CONFIG_NET_RADIO=y
+
+#
+# Obsolete Wireless cards support (pre-802.11)
+#
+# CONFIG_STRIP is not set
+
+#
+# Wireless 802.11b ISA/PCI cards support
+#
+# CONFIG_HERMES is not set
+# CONFIG_ATMEL is not set
+
+#
+# Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support
+#
+CONFIG_PRISM54=m
+# CONFIG_HOSTAP is not set
+CONFIG_NET_WIRELESS=y
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+CONFIG_PPP=m
+CONFIG_PPP_MULTILINK=y
+CONFIG_PPP_FILTER=y
+CONFIG_PPP_ASYNC=m
+CONFIG_PPP_SYNC_TTY=m
+CONFIG_PPP_DEFLATE=m
+CONFIG_PPP_BSDCOMP=m
+CONFIG_PPPOE=m
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+CONFIG_USB=m
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+CONFIG_USB_DEVICEFS=y
+# CONFIG_USB_BANDWIDTH is not set
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+
+#
+# USB Host Controller Drivers
+#
+CONFIG_USB_EHCI_HCD=m
+# CONFIG_USB_EHCI_SPLIT_ISO is not set
+# CONFIG_USB_EHCI_ROOT_HUB_TT is not set
+# CONFIG_USB_ISP116X_HCD is not set
+CONFIG_USB_OHCI_HCD=m
+# CONFIG_USB_OHCI_BIG_ENDIAN is not set
+CONFIG_USB_OHCI_LITTLE_ENDIAN=y
+# CONFIG_USB_UHCI_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_BLUETOOTH_TTY is not set
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support' may also be needed; see USB_STORAGE Help for more information
+#
+CONFIG_USB_STORAGE=m
+# CONFIG_USB_STORAGE_DEBUG is not set
+# CONFIG_USB_STORAGE_DATAFAB is not set
+# CONFIG_USB_STORAGE_FREECOM is not set
+# CONFIG_USB_STORAGE_DPCM is not set
+# CONFIG_USB_STORAGE_USBAT is not set
+# CONFIG_USB_STORAGE_SDDR09 is not set
+# CONFIG_USB_STORAGE_SDDR55 is not set
+# CONFIG_USB_STORAGE_JUMPSHOT is not set
+
+#
+# USB Input Devices
+#
+# CONFIG_USB_HID is not set
+
+#
+# USB HID Boot Protocol drivers
+#
+# CONFIG_USB_KBD is not set
+# CONFIG_USB_MOUSE is not set
+# CONFIG_USB_AIPTEK is not set
+# CONFIG_USB_WACOM is not set
+# CONFIG_USB_ACECAD is not set
+# CONFIG_USB_KBTAB is not set
+# CONFIG_USB_POWERMATE is not set
+# CONFIG_USB_MTOUCH is not set
+# CONFIG_USB_ITMTOUCH is not set
+# CONFIG_USB_EGALAX is not set
+# CONFIG_USB_YEALINK is not set
+# CONFIG_USB_XPAD is not set
+# CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_KEYSPAN_REMOTE is not set
+# CONFIG_USB_APPLETOUCH is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+# CONFIG_USB_MICROTEK is not set
+
+#
+# USB Multimedia devices
+#
+# CONFIG_USB_DABUSB is not set
+
+#
+# Video4Linux support is needed for USB Multimedia device support
+#
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_USB_ZD1201 is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+CONFIG_USB_SERIAL=m
+CONFIG_USB_SERIAL_GENERIC=y
+# CONFIG_USB_SERIAL_AIRPRIME is not set
+# CONFIG_USB_SERIAL_BELKIN is not set
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
+# CONFIG_USB_SERIAL_DIGI_ACCELEPORT is not set
+# CONFIG_USB_SERIAL_CP2101 is not set
+# CONFIG_USB_SERIAL_CYPRESS_M8 is not set
+# CONFIG_USB_SERIAL_EMPEG is not set
+# CONFIG_USB_SERIAL_FTDI_SIO is not set
+# CONFIG_USB_SERIAL_VISOR is not set
+# CONFIG_USB_SERIAL_IPAQ is not set
+# CONFIG_USB_SERIAL_IR is not set
+# CONFIG_USB_SERIAL_EDGEPORT is not set
+# CONFIG_USB_SERIAL_EDGEPORT_TI is not set
+# CONFIG_USB_SERIAL_GARMIN is not set
+# CONFIG_USB_SERIAL_IPW is not set
+# CONFIG_USB_SERIAL_KEYSPAN_PDA is not set
+# CONFIG_USB_SERIAL_KEYSPAN is not set
+# CONFIG_USB_SERIAL_KLSI is not set
+# CONFIG_USB_SERIAL_KOBIL_SCT is not set
+# CONFIG_USB_SERIAL_MCT_U232 is not set
+CONFIG_USB_SERIAL_PL2303=m
+# CONFIG_USB_SERIAL_HP4X is not set
+# CONFIG_USB_SERIAL_SAFE is not set
+# CONFIG_USB_SERIAL_TI is not set
+# CONFIG_USB_SERIAL_CYBERJACK is not set
+# CONFIG_USB_SERIAL_XIRCOM is not set
+# CONFIG_USB_SERIAL_OMNINET is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGETKIT is not set
+# CONFIG_USB_PHIDGETSERVO is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TEST is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+CONFIG_EXT2_FS_XATTR=y
+CONFIG_EXT2_FS_POSIX_ACL=y
+CONFIG_EXT2_FS_SECURITY=y
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+CONFIG_EXT3_FS_POSIX_ACL=y
+CONFIG_EXT3_FS_SECURITY=y
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+CONFIG_REISERFS_FS=m
+# CONFIG_REISERFS_CHECK is not set
+# CONFIG_REISERFS_PROC_INFO is not set
+# CONFIG_REISERFS_FS_XATTR is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+CONFIG_FUSE_FS=m
+
+#
+# CD-ROM/DVD Filesystems
+#
+CONFIG_ISO9660_FS=m
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_ZISOFS_FS=m
+CONFIG_UDF_FS=m
+CONFIG_UDF_NLS=y
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=m
+CONFIG_MSDOS_FS=m
+CONFIG_VFAT_FS=m
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+CONFIG_RELAYFS_FS=m
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+# CONFIG_NFSD_V4 is not set
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+CONFIG_SMB_FS=m
+# CONFIG_SMB_NLS_DEFAULT is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+CONFIG_9P_FS=m
+
+#
+# Partition Types
+#
+CONFIG_PARTITION_ADVANCED=y
+# CONFIG_ACORN_PARTITION is not set
+# CONFIG_OSF_PARTITION is not set
+# CONFIG_AMIGA_PARTITION is not set
+# CONFIG_ATARI_PARTITION is not set
+# CONFIG_MAC_PARTITION is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_BSD_DISKLABEL is not set
+# CONFIG_MINIX_SUBPARTITION is not set
+# CONFIG_SOLARIS_X86_PARTITION is not set
+# CONFIG_UNIXWARE_DISKLABEL is not set
+# CONFIG_LDM_PARTITION is not set
+# CONFIG_SGI_PARTITION is not set
+# CONFIG_ULTRIX_PARTITION is not set
+# CONFIG_SUN_PARTITION is not set
+# CONFIG_EFI_PARTITION is not set
+
+#
+# Native Language Support
+#
+CONFIG_NLS=m
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=m
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+CONFIG_NLS_ASCII=m
+CONFIG_NLS_ISO8859_1=m
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+CONFIG_NLS_UTF8=m
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_DEBUG_KERNEL=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_LOG_BUF_SHIFT=16
+# CONFIG_DETECT_SOFTLOCKUP is not set
+# CONFIG_SCHEDSTATS is not set
+CONFIG_DEBUG_SLAB=y
+# CONFIG_DEBUG_SPINLOCK is not set
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_INFO is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_FRAME_POINTER is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+CONFIG_EARLY_SCIF_CONSOLE=y
+# CONFIG_EARLY_PRINTK is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_HMAC=y
+# CONFIG_CRYPTO_NULL is not set
+CONFIG_CRYPTO_MD4=m
+CONFIG_CRYPTO_MD5=m
+CONFIG_CRYPTO_SHA1=m
+CONFIG_CRYPTO_SHA256=m
+CONFIG_CRYPTO_SHA512=m
+CONFIG_CRYPTO_WP512=m
+CONFIG_CRYPTO_TGR192=m
+CONFIG_CRYPTO_DES=m
+CONFIG_CRYPTO_BLOWFISH=m
+CONFIG_CRYPTO_TWOFISH=m
+CONFIG_CRYPTO_SERPENT=m
+CONFIG_CRYPTO_AES=m
+CONFIG_CRYPTO_CAST5=m
+CONFIG_CRYPTO_CAST6=m
+CONFIG_CRYPTO_TEA=m
+CONFIG_CRYPTO_ARC4=m
+CONFIG_CRYPTO_KHAZAD=m
+CONFIG_CRYPTO_ANUBIS=m
+CONFIG_CRYPTO_DEFLATE=m
+CONFIG_CRYPTO_MICHAEL_MIC=m
+CONFIG_CRYPTO_CRC32C=m
+# CONFIG_CRYPTO_TEST is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+CONFIG_CRC_CCITT=m
+CONFIG_CRC16=m
+CONFIG_CRC32=y
+CONFIG_LIBCRC32C=m
+CONFIG_ZLIB_INFLATE=m
+CONFIG_ZLIB_DEFLATE=m
+CONFIG_TEXTSEARCH=y
+CONFIG_TEXTSEARCH_KMP=m
+CONFIG_TEXTSEARCH_BM=m
+CONFIG_TEXTSEARCH_FSM=m
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 365bc16a4a83..86d6d0660d4c 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -14,3 +14,4 @@ obj-$(CONFIG_SH_SECUREEDGE5410)		+= ops-snapgear.o
 obj-$(CONFIG_SH_BIGSUR)			+= ops-bigsur.o
 obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
+obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
new file mode 100644
index 000000000000..f76e4e963ac1
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -0,0 +1,84 @@
+/*
+ * arch/sh/drivers/pci/ops-titan.c
+ *
+ * Ported to new API by Paul Mundt <lethal@linux-sh.org>
+ *
+ * Modified from ops-snapgear.c written by  David McCullough
+ * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * PCI initialization for the Titan boards
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <asm/io.h>
+#include <asm/titan.h>
+#include "pci-sh7751.h"
+
+int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+{
+	int irq = -1;
+
+	switch (slot) {
+	case 0: irq = TITAN_IRQ_WAN;   break;	/* eth0 (WAN) */
+	case 1: irq = TITAN_IRQ_LAN;   break;	/* eth1 (LAN) */
+	case 2: irq = TITAN_IRQ_MPCIA; break;	/* mPCI A */
+	case 3: irq = TITAN_IRQ_MPCIB; break;	/* mPCI B */
+	case 4: irq = TITAN_IRQ_USB;   break;	/* USB */
+	default:
+		printk(KERN_INFO "PCI: Bad IRQ mapping request for slot %d\n", slot);
+		return -1;
+	}
+
+	printk("PCI: Mapping TITAN IRQ for slot %d, pin %c to irq %d\n",
+		slot, pin - 1 + 'A', irq);
+
+	return irq;
+}
+
+static struct resource sh7751_io_resource = {
+	.name	= "SH7751_IO",
+	.start	= SH7751_PCI_IO_BASE,
+	.end	= SH7751_PCI_IO_BASE + SH7751_PCI_IO_SIZE - 1,
+	.flags	= IORESOURCE_IO
+};
+
+static struct resource sh7751_mem_resource = {
+	.name	= "SH7751_mem",
+	.start	= SH7751_PCI_MEMORY_BASE,
+	.end	= SH7751_PCI_MEMORY_BASE + SH7751_PCI_MEM_SIZE - 1,
+	.flags	= IORESOURCE_MEM
+};
+
+extern struct pci_ops sh7751_pci_ops;
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+static struct sh7751_pci_address_map sh7751_pci_map = {
+	.window0	= {
+		.base	= SH7751_CS2_BASE_ADDR,
+		.size	= SH7751_MEM_REGION_SIZE*2,	/* cs2 and cs3 */
+	},
+
+	.window1	= {
+		.base	= SH7751_CS2_BASE_ADDR,
+		.size	= SH7751_MEM_REGION_SIZE*2,
+	},
+
+	.flags	= SH7751_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7751_pcic_init(&sh7751_pci_map);
+}
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 3d546ba329cf..1f5e23e8b163 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -1,21 +1,25 @@
-/* arch/sh/kernel/pci.c
- * $Id: pci.c,v 1.1 2003/08/24 19:15:45 lethal Exp $
+/*
+ * arch/sh/drivers/pci/pci.c
  *
  * Copyright (c) 2002 M. R. Brown  <mrbrown@linux-sh.org>
- * 
- * 
+ * Copyright (c) 2004, 2005 Paul Mundt  <lethal@linux-sh.org>
+ *
  * These functions are collected here to reduce duplication of common
  * code amongst the many platform-specific PCI support code files.
- * 
+ *
  * These routines require the following board-specific routines:
  * void pcibios_fixup_irqs();
  *
  * See include/asm-sh/pci.h for more information.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/kernel.h>
 #include <linux/pci.h>
 #include <linux/init.h>
+#include <asm/io.h>
 
 static int __init pcibios_init(void)
 {
@@ -26,9 +30,8 @@ static int __init pcibios_init(void)
 #ifdef CONFIG_PCI_AUTO
 	/* assign resources */
 	busno = 0;
-	for (p = board_pci_channels; p->pci_ops != NULL; p++) {
+	for (p = board_pci_channels; p->pci_ops != NULL; p++)
 		busno = pciauto_assign_resources(busno, p) + 1;
-	}
 #endif
 
 	/* scan the buses */
@@ -61,13 +64,17 @@ pcibios_update_resource(struct pci_dev *dev, struct resource *root,
 		new |= PCI_ROM_ADDRESS_ENABLE;
 		reg = dev->rom_base_reg;
 	} else {
-		/* Somebody might have asked allocation of a non-standard resource */
+		/*
+		 * Somebody might have asked allocation of a non-standard
+		 * resource
+		 */
 		return;
 	}
-	
+
 	pci_write_config_dword(dev, reg, new);
 	pci_read_config_dword(dev, reg, &check);
-	if ((new ^ check) & ((new & PCI_BASE_ADDRESS_SPACE_IO) ? PCI_BASE_ADDRESS_IO_MASK : PCI_BASE_ADDRESS_MEM_MASK)) {
+	if ((new ^ check) & ((new & PCI_BASE_ADDRESS_SPACE_IO) ?
+		PCI_BASE_ADDRESS_IO_MASK : PCI_BASE_ADDRESS_MEM_MASK)) {
 		printk(KERN_ERR "PCI: Error while updating region "
 		       "%s/%d (%08x != %08x)\n", pci_name(dev), resource,
 		       new, check);
@@ -145,7 +152,8 @@ void pcibios_set_master(struct pci_dev *dev)
 		lat = pcibios_max_latency;
 	else
 		return;
-	printk(KERN_INFO "PCI: Setting latency timer of device %s to %d\n", pci_name(dev), lat);
+	printk(KERN_INFO "PCI: Setting latency timer of device %s to %d\n",
+	       pci_name(dev), lat);
 	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
 }
 
@@ -153,3 +161,29 @@ void __init pcibios_update_irq(struct pci_dev *dev, int irq)
 {
 	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
 }
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
+{
+	unsigned long start = pci_resource_start(dev, bar);
+	unsigned long len = pci_resource_len(dev, bar);
+	unsigned long flags = pci_resource_flags(dev, bar);
+
+	if (!len || !start)
+		return NULL;
+	if (maxlen && len > maxlen)
+		len = maxlen;
+	if (flags & IORESOURCE_IO)
+		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM)
+		return ioremap(start, len);
+
+	return NULL;
+}
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	iounmap(addr);
+}
+
+EXPORT_SYMBOL(pci_iomap);
+EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index c503ca072e11..01ae394a6599 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -26,3 +26,4 @@ EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
 LANDISK			SH_LANDISK
+TITAN			SH_TITAN
diff --git a/include/asm-sh/titan.h b/include/asm-sh/titan.h
new file mode 100644
index 000000000000..270a4f4bc8a9
--- /dev/null
+++ b/include/asm-sh/titan.h
@@ -0,0 +1,43 @@
+/*
+ * Platform defintions for Titan
+ */
+
+#ifndef _ASM_SH_TITAN_TITAN_H
+#define _ASM_SH_TITAN_TITAN_H
+
+#define __IO_PREFIX titan
+#include <asm/io_generic.h>
+
+/* IRQ assignments */
+#define TITAN_IRQ_WAN		2	/* eth0 (WAN) */
+#define TITAN_IRQ_LAN		5	/* eth1 (LAN) */
+#define TITAN_IRQ_MPCIA		8	/* mPCI A */
+#define TITAN_IRQ_MPCIB		11	/* mPCI B */
+#define TITAN_IRQ_USB		11	/* USB */
+
+/*
+ * The external interrupt lines, these take up ints 0 - 15 inclusive
+ * depending on the priority for the interrupt.  In fact the priority
+ * is the interrupt :-)
+ */
+#define IRL0_IRQ	0
+#define IRL0_IPR_ADDR	INTC_IPRD
+#define IRL0_IPR_POS	3
+#define IRL0_PRIORITY	8
+
+#define IRL1_IRQ	1
+#define IRL1_IPR_ADDR	INTC_IPRD
+#define IRL1_IPR_POS	2
+#define IRL1_PRIORITY	8
+
+#define IRL2_IRQ	2
+#define IRL2_IPR_ADDR	INTC_IPRD
+#define IRL2_IPR_POS	1
+#define IRL2_PRIORITY	8
+
+#define IRL3_IRQ	3
+#define IRL3_IPR_ADDR	INTC_IPRD
+#define IRL3_IPR_POS	0
+#define IRL3_PRIORITY	8
+
+#endif
-- 
cgit v1.2.1


From 0c7b1df69c62209db19d1279dd882b37c04c5c2f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:08:07 +0900
Subject: sh: SH-4A Privileged Space Mapping Buffer (PMB) support.

Add support for 32-bit physical addressing through the SH-4A
Privileged Space Mapping Buffer (PMB).

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/Makefile  |   3 +-
 arch/sh/mm/pmb.c     | 269 +++++++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/mmu.h |  48 ++++++++-
 3 files changed, 318 insertions(+), 2 deletions(-)
 create mode 100644 arch/sh/mm/pmb.c

diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index 9489a1424644..f4e32b3d24dc 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -22,4 +22,5 @@ obj-$(CONFIG_CPU_SH4)	+= tlb-sh4.o ioremap.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
-obj-$(CONFIG_SH7705_CACHE_32KB) += cache-sh7705.o
+obj-$(CONFIG_SH7705_CACHE_32KB)	+= cache-sh7705.o
+obj-$(CONFIG_32BIT)		+= pmb.o
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
new file mode 100644
index 000000000000..ff5bde745647
--- /dev/null
+++ b/arch/sh/mm/pmb.c
@@ -0,0 +1,269 @@
+/*
+ * arch/sh/mm/pmb.c
+ *
+ * Privileged Space Mapping Buffer (PMB) Support.
+ *
+ * Copyright (C) 2005 Paul Mundt
+ *
+ * P1/P2 Section mapping definitions from map32.h, which was:
+ *
+ *	Copyright 2003 (c) Lineo Solutions,Inc.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/bitops.h>
+#include <linux/debugfs.h>
+#include <linux/fs.h>
+#include <linux/seq_file.h>
+#include <linux/err.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/mmu.h>
+#include <asm/io.h>
+
+#define NR_PMB_ENTRIES	16
+
+static kmem_cache_t *pmb_cache;
+static unsigned long pmb_map;
+
+static struct pmb_entry pmb_init_map[] = {
+	/* vpn         ppn         flags (ub/sz/c/wt) */
+
+	/* P1 Section Mappings */
+	{ 0x80000000, 0x00000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x84000000, 0x04000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x88000000, 0x08000000, PMB_SZ_128M | PMB_C, },
+	{ 0x90000000, 0x10000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x94000000, 0x14000000, PMB_SZ_64M  | PMB_C, },
+	{ 0x98000000, 0x18000000, PMB_SZ_64M  | PMB_C, },
+
+	/* P2 Section Mappings */
+	{ 0xa0000000, 0x00000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xa4000000, 0x04000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xa8000000, 0x08000000, PMB_UB | PMB_SZ_128M | PMB_WT, },
+	{ 0xb0000000, 0x10000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xb4000000, 0x14000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+	{ 0xb8000000, 0x18000000, PMB_UB | PMB_SZ_64M  | PMB_WT, },
+};
+
+static inline unsigned long mk_pmb_entry(unsigned int entry)
+{
+	return (entry & PMB_E_MASK) << PMB_E_SHIFT;
+}
+
+static inline unsigned long mk_pmb_addr(unsigned int entry)
+{
+	return mk_pmb_entry(entry) | PMB_ADDR;
+}
+
+static inline unsigned long mk_pmb_data(unsigned int entry)
+{
+	return mk_pmb_entry(entry) | PMB_DATA;
+}
+
+struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
+			    unsigned long flags)
+{
+	struct pmb_entry *pmbe;
+
+	pmbe = kmem_cache_alloc(pmb_cache, GFP_KERNEL);
+	if (!pmbe)
+		return ERR_PTR(-ENOMEM);
+
+	pmbe->vpn	= vpn;
+	pmbe->ppn	= ppn;
+	pmbe->flags	= flags;
+
+	return pmbe;
+}
+
+void pmb_free(struct pmb_entry *pmbe)
+{
+	kmem_cache_free(pmb_cache, pmbe);
+}
+
+/*
+ * Must be in P2 for __set_pmb_entry()
+ */
+int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
+		    unsigned long flags, int *entry)
+{
+	unsigned int pos = *entry;
+
+	if (unlikely(pos == PMB_NO_ENTRY))
+		pos = find_first_zero_bit(&pmb_map, NR_PMB_ENTRIES);
+
+repeat:
+	if (unlikely(pos > NR_PMB_ENTRIES))
+		return -ENOSPC;
+
+	if (test_and_set_bit(pos, &pmb_map)) {
+		pos = find_first_zero_bit(&pmb_map, NR_PMB_ENTRIES);
+		goto repeat;
+	}
+
+	ctrl_outl(vpn | PMB_V, mk_pmb_addr(pos));
+
+#ifdef CONFIG_SH_WRITETHROUGH
+	/*
+	 * When we are in 32-bit address extended mode, CCR.CB becomes
+	 * invalid, so care must be taken to manually adjust cacheable
+	 * translations.
+	 */
+	if (likely(flags & PMB_C))
+		flags |= PMB_WT;
+#endif
+
+	ctrl_outl(ppn | flags | PMB_V, mk_pmb_data(pos));
+
+	*entry = pos;
+
+	return 0;
+}
+
+void set_pmb_entry(struct pmb_entry *pmbe)
+{
+	jump_to_P2();
+	__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
+	back_to_P1();
+}
+
+void clear_pmb_entry(struct pmb_entry *pmbe)
+{
+	unsigned int entry = pmbe->entry;
+	unsigned long addr;
+
+	/*
+	 * Don't allow clearing of wired init entries, P1 or P2 access
+	 * without a corresponding mapping in the PMB will lead to reset
+	 * by the TLB.
+	 */
+	if (unlikely(entry < ARRAY_SIZE(pmb_init_map) ||
+		     entry >= NR_PMB_ENTRIES))
+		return;
+
+	jump_to_P2();
+
+	/* Clear V-bit */
+	addr = mk_pmb_addr(entry);
+	ctrl_outl(ctrl_inl(addr) & ~PMB_V, addr);
+
+	addr = mk_pmb_data(entry);
+	ctrl_outl(ctrl_inl(addr) & ~PMB_V, addr);
+
+	back_to_P1();
+
+	clear_bit(entry, &pmb_map);
+}
+
+static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
+{
+	memset(pmb, 0, sizeof(struct pmb_entry));
+
+	((struct pmb_entry *)pmb)->entry = PMB_NO_ENTRY;
+}
+
+static int __init pmb_init(void)
+{
+	unsigned int nr_entries = ARRAY_SIZE(pmb_init_map);
+	unsigned int entry;
+
+	BUG_ON(unlikely(nr_entries >= NR_PMB_ENTRIES));
+
+	pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry),
+				      0, 0, pmb_cache_ctor, NULL);
+	BUG_ON(!pmb_cache);
+
+	jump_to_P2();
+
+	/*
+	 * Ordering is important, P2 must be mapped in the PMB before we
+	 * can set PMB.SE, and P1 must be mapped before we jump back to
+	 * P1 space.
+	 */
+	for (entry = 0; entry < nr_entries; entry++) {
+		struct pmb_entry *pmbe = pmb_init_map + entry;
+
+		__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &entry);
+	}
+
+	ctrl_outl(0, PMB_IRMCR);
+
+	/* PMB.SE and UB[7] */
+	ctrl_outl((1 << 31) | (1 << 7), PMB_PASCR);
+
+	back_to_P1();
+
+	return 0;
+}
+
+arch_initcall(pmb_init);
+
+#ifdef CONFIG_DEBUG_FS
+static int pmb_seq_show(struct seq_file *file, void *iter)
+{
+	int i;
+
+	seq_printf(file, "V: Valid, C: Cacheable, WT: Write-Through\n"
+			 "CB: Copy-Back, B: Buffered, UB: Unbuffered\n");
+	seq_printf(file, "ety   vpn  ppn  size   flags\n");
+
+	for (i = 0; i < NR_PMB_ENTRIES; i++) {
+		unsigned long addr, data;
+		unsigned int size;
+		char *sz_str = NULL;
+
+		addr = ctrl_inl(mk_pmb_addr(i));
+		data = ctrl_inl(mk_pmb_data(i));
+
+		size = data & PMB_SZ_MASK;
+		sz_str = (size == PMB_SZ_16M)  ? " 16MB":
+			 (size == PMB_SZ_64M)  ? " 64MB":
+			 (size == PMB_SZ_128M) ? "128MB":
+					         "512MB";
+
+		/* 02: V 0x88 0x08 128MB C CB  B */
+		seq_printf(file, "%02d: %c 0x%02lx 0x%02lx %s %c %s %s\n",
+			   i, ((addr & PMB_V) && (data & PMB_V)) ? 'V' : ' ',
+			   (addr >> 24) & 0xff, (data >> 24) & 0xff,
+			   sz_str, (data & PMB_C) ? 'C' : ' ',
+			   (data & PMB_WT) ? "WT" : "CB",
+			   (data & PMB_UB) ? "UB" : " B");
+	}
+
+	return 0;
+}
+
+static int pmb_debugfs_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, pmb_seq_show, NULL);
+}
+
+static struct file_operations pmb_debugfs_fops = {
+	.owner		= THIS_MODULE,
+	.open		= pmb_debugfs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int __init pmb_debugfs_init(void)
+{
+	struct dentry *dentry;
+
+	dentry = debugfs_create_file("pmb", S_IFREG | S_IRUGO,
+				     NULL, NULL, &pmb_debugfs_fops);
+	if (IS_ERR(dentry))
+		return PTR_ERR(dentry);
+
+	return 0;
+}
+
+postcore_initcall(pmb_debugfs_init);
+#endif
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 72f07be6117f..91c884634276 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -25,5 +25,51 @@ typedef struct {
 typedef unsigned long mm_context_t;
 
 #endif /* CONFIG_MMU */
-#endif /* __MMH_H */
+
+/*
+ * Privileged Space Mapping Buffer (PMB) definitions
+ */
+#define PMB_PASCR		0xff000070
+#define PMB_IRMCR		0xff000078
+
+#define PMB_ADDR		0xf6100000
+#define PMB_DATA		0xf7100000
+#define PMB_ENTRY_MAX		16
+#define PMB_E_MASK		0x0000000f
+#define PMB_E_SHIFT		8
+
+#define PMB_SZ_16M		0x00000000
+#define PMB_SZ_64M		0x00000010
+#define PMB_SZ_128M		0x00000080
+#define PMB_SZ_512M		0x00000090
+#define PMB_SZ_MASK		PMB_SZ_512M
+#define PMB_C			0x00000008
+#define PMB_WT			0x00000001
+#define PMB_UB			0x00000200
+#define PMB_V			0x00000100
+
+#define PMB_NO_ENTRY		(-1)
+
+struct pmb_entry {
+	unsigned long vpn;
+	unsigned long ppn;
+	unsigned long flags;
+
+	/*
+	 * 0 .. NR_PMB_ENTRIES for specific entry selection, or
+	 * PMB_NO_ENTRY to search for a free one
+	 */
+	int entry;
+};
+
+/* arch/sh/mm/pmb.c */
+int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
+		    unsigned long flags, int *entry);
+void set_pmb_entry(struct pmb_entry *pmbe);
+void clear_pmb_entry(struct pmb_entry *pmbe);
+struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
+			    unsigned long flags);
+void pmb_free(struct pmb_entry *pmbe);
+
+#endif /* __MMU_H */
 
-- 
cgit v1.2.1


From 9359e757709a211040e4b0151eae69248e7c6eca Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:09:48 +0900
Subject: sh: export clear_user_page() for the modules that need it.

Some modules seem to need this, so we export it..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/sh_ksyms.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 8cad80cec9a6..8a6dd06fd071 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -100,6 +100,7 @@ EXPORT_SYMBOL(flush_cache_all);
 EXPORT_SYMBOL(flush_cache_range);
 EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
+EXPORT_SYMBOL(clear_user_page);
 #endif
 
 #if defined(CONFIG_SH7705_CACHE_32KB)
-- 
cgit v1.2.1


From e531dcc568ca326428158e9574b8560572b8ae28 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Tue, 26 Sep 2006 23:10:01 -0700
Subject: [SPARC64]: Update defconfig.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/defconfig | 38 ++++++++++++++++++++------------------
 1 file changed, 20 insertions(+), 18 deletions(-)

diff --git a/arch/sparc64/defconfig b/arch/sparc64/defconfig
index 51cf6027b701..0fbdaa5daa8c 100644
--- a/arch/sparc64/defconfig
+++ b/arch/sparc64/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
 # Linux kernel version: 2.6.18
-# Sat Sep 23 18:32:19 2006
+# Tue Sep 26 23:09:35 2006
 #
 CONFIG_SPARC=y
 CONFIG_SPARC64=y
@@ -141,6 +141,7 @@ CONFIG_SUN_AUXIO=y
 CONFIG_SUN_IO=y
 CONFIG_PCI=y
 CONFIG_PCI_DOMAINS=y
+# CONFIG_PCI_MULTITHREAD_PROBE is not set
 # CONFIG_PCI_DEBUG is not set
 CONFIG_SUN_OPENPROMFS=m
 CONFIG_SPARC32_COMPAT=y
@@ -194,21 +195,9 @@ CONFIG_INET_XFRM_MODE_TRANSPORT=y
 CONFIG_INET_XFRM_MODE_TUNNEL=y
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
-CONFIG_TCP_CONG_ADVANCED=y
-
-#
-# TCP congestion control
-#
-CONFIG_TCP_CONG_BIC=y
-CONFIG_TCP_CONG_CUBIC=m
-CONFIG_TCP_CONG_WESTWOOD=m
-CONFIG_TCP_CONG_HTCP=m
-CONFIG_TCP_CONG_HSTCP=m
-CONFIG_TCP_CONG_HYBLA=m
-CONFIG_TCP_CONG_VEGAS=m
-CONFIG_TCP_CONG_SCALABLE=m
-CONFIG_TCP_CONG_LP=m
-CONFIG_TCP_CONG_VENO=m
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
 CONFIG_IPV6=m
 CONFIG_IPV6_PRIVACY=y
 CONFIG_IPV6_ROUTER_PREF=y
@@ -247,6 +236,7 @@ CONFIG_IP_DCCP_TFRC_LIB=m
 # DCCP Kernel Hacking
 #
 # CONFIG_IP_DCCP_DEBUG is not set
+# CONFIG_NET_DCCPPROBE is not set
 
 #
 # SCTP Configuration (EXPERIMENTAL)
@@ -401,6 +391,7 @@ CONFIG_IDEDMA_AUTO=y
 #
 CONFIG_RAID_ATTRS=m
 CONFIG_SCSI=y
+CONFIG_SCSI_NETLINK=y
 CONFIG_SCSI_PROC_FS=y
 
 #
@@ -422,12 +413,13 @@ CONFIG_SCSI_CONSTANTS=y
 # CONFIG_SCSI_LOGGING is not set
 
 #
-# SCSI Transport Attributes
+# SCSI Transports
 #
 CONFIG_SCSI_SPI_ATTRS=y
 CONFIG_SCSI_FC_ATTRS=y
 CONFIG_SCSI_ISCSI_ATTRS=m
 # CONFIG_SCSI_SAS_ATTRS is not set
+# CONFIG_SCSI_SAS_LIBSAS is not set
 
 #
 # SCSI low-level drivers
@@ -440,16 +432,18 @@ CONFIG_ISCSI_TCP=m
 # CONFIG_SCSI_AIC7XXX is not set
 # CONFIG_SCSI_AIC7XXX_OLD is not set
 # CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_AIC94XX is not set
+# CONFIG_SCSI_ARCMSR is not set
 # CONFIG_MEGARAID_NEWGEN is not set
 # CONFIG_MEGARAID_LEGACY is not set
 # CONFIG_MEGARAID_SAS is not set
-# CONFIG_SCSI_SATA is not set
 # CONFIG_SCSI_HPTIOP is not set
 # CONFIG_SCSI_DMX3191D is not set
 # CONFIG_SCSI_FUTURE_DOMAIN is not set
 # CONFIG_SCSI_IPS is not set
 # CONFIG_SCSI_INITIO is not set
 # CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_STEX is not set
 # CONFIG_SCSI_SYM53C8XX_2 is not set
 # CONFIG_SCSI_IPR is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
@@ -461,6 +455,11 @@ CONFIG_ISCSI_TCP=m
 # CONFIG_SCSI_DEBUG is not set
 # CONFIG_SCSI_SUNESP is not set
 
+#
+# Serial ATA (prod) and Parallel ATA (experimental) drivers
+#
+# CONFIG_ATA is not set
+
 #
 # Multi-device support (RAID and LVM)
 #
@@ -575,6 +574,7 @@ CONFIG_E1000_NAPI=y
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=m
 CONFIG_BNX2=m
+# CONFIG_QLA3XXX is not set
 
 #
 # Ethernet (10000 Mbit)
@@ -1006,6 +1006,7 @@ CONFIG_SND_ALI5451=m
 # CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
 # CONFIG_SND_YMFPCI is not set
+# CONFIG_SND_AC97_POWER_SAVE is not set
 
 #
 # USB devices
@@ -1353,6 +1354,7 @@ CONFIG_KPROBES=y
 # Kernel hacking
 #
 CONFIG_PRINTK_TIME=y
+CONFIG_ENABLE_MUST_CHECK=y
 CONFIG_MAGIC_SYSRQ=y
 # CONFIG_UNUSED_SYMBOLS is not set
 CONFIG_DEBUG_KERNEL=y
-- 
cgit v1.2.1


From 26ff6c11ef38e08990c1e417c299246e6ab18ff7 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:13:36 +0900
Subject: sh: page table alloc cleanups and page fault optimizations.

Cleanup of page table allocators, using generic folded PMD and PUD
helpers. TLB flushing operations are moved to a more sensible spot.

The page fault handler is also optimized slightly, we no longer waste
cycles on IRQ disabling for flushing of the page from the ITLB, since
we're already under CLI protection by the initial exception handler.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/sys_sh.c             |   2 +-
 arch/sh/mm/Makefile                 |   2 +-
 arch/sh/mm/consistent.c             |   2 +
 arch/sh/mm/fault.c                  | 202 +++++++-----------------------------
 arch/sh/mm/init.c                   |  13 ++-
 arch/sh/mm/tlb-flush.c              | 132 +++++++++++++++++++++++
 include/asm-sh/cache.h              |   8 --
 include/asm-sh/cacheflush.h         |   1 +
 include/asm-sh/cpu-sh3/cacheflush.h |   8 +-
 include/asm-sh/cpu-sh4/cacheflush.h |  29 +++---
 include/asm-sh/page.h               |  12 +--
 include/asm-sh/pgalloc.h            |  37 +------
 include/asm-sh/pgtable.h            |  80 +++++++++-----
 13 files changed, 261 insertions(+), 267 deletions(-)
 create mode 100644 arch/sh/mm/tlb-flush.c

diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index 917b2f32f260..d8bcd8a22327 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -21,7 +21,7 @@
 #include <linux/mman.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
-
+#include <asm/cacheflush.h>
 #include <asm/uaccess.h>
 #include <asm/ipc.h>
 
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index f4e32b3d24dc..d90906367c5f 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -12,7 +12,7 @@ obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 
 mmu-y			:= fault-nommu.o tlb-nommu.o pg-nommu.o
-mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o
+mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o
 
 obj-y			+= $(mmu-y)
 
diff --git a/arch/sh/mm/consistent.c b/arch/sh/mm/consistent.c
index ee73e30263af..c81e6b67ad30 100644
--- a/arch/sh/mm/consistent.c
+++ b/arch/sh/mm/consistent.c
@@ -9,6 +9,8 @@
  */
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
+#include <asm/cacheflush.h>
+#include <asm/addrspace.h>
 #include <asm/io.h>
 
 void *consistent_alloc(gfp_t gfp, size_t size, dma_addr_t *handle)
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 364181f27b79..7a03ffe6dadd 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -1,33 +1,20 @@
-/* $Id: fault.c,v 1.14 2004/01/13 05:52:11 kkojima Exp $
+/*
+ * Page fault handler for SH with an MMU.
  *
- *  linux/arch/sh/mm/fault.c
  *  Copyright (C) 1999  Niibe Yutaka
  *  Copyright (C) 2003  Paul Mundt
  *
  *  Based on linux/arch/i386/mm/fault.c:
  *   Copyright (C) 1995  Linus Torvalds
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
-#include <linux/signal.h>
-#include <linux/sched.h>
 #include <linux/kernel.h>
-#include <linux/errno.h>
-#include <linux/string.h>
-#include <linux/types.h>
-#include <linux/ptrace.h>
-#include <linux/mman.h>
 #include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/interrupt.h>
-#include <linux/module.h>
-
 #include <asm/system.h>
-#include <asm/io.h>
-#include <asm/uaccess.h>
-#include <asm/pgalloc.h>
 #include <asm/mmu_context.h>
-#include <asm/cacheflush.h>
 #include <asm/kgdb.h>
 
 extern void die(const char *,struct pt_regs *,long);
@@ -187,14 +174,25 @@ do_sigbus:
 		goto no_context;
 }
 
+#ifdef CONFIG_SH_STORE_QUEUES
 /*
- * Called with interrupt disabled.
+ * This is a special case for the SH-4 store queues, as pages for this
+ * space still need to be faulted in before it's possible to flush the
+ * store queue cache for writeout to the remapped region.
+ */
+#define P3_ADDR_MAX		(P4SEG_STORE_QUE + 0x04000000)
+#else
+#define P3_ADDR_MAX		P4SEG
+#endif
+
+/*
+ * Called with interrupts disabled.
  */
 asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 			       unsigned long address)
 {
-	unsigned long addrmax = P4SEG;
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 	pte_t entry;
@@ -207,31 +205,36 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 		kgdb_bus_err_hook();
 #endif
 
-#ifdef CONFIG_SH_STORE_QUEUES
-	addrmax = P4SEG_STORE_QUE + 0x04000000;
-#endif
-
-	if (address >= P3SEG && address < addrmax) {
+	/*
+	 * We don't take page faults for P1, P2, and parts of P4, these
+	 * are always mapped, whether it be due to legacy behaviour in
+	 * 29-bit mode, or due to PMB configuration in 32-bit mode.
+	 */
+	if (address >= P3SEG && address < P3_ADDR_MAX)
 		pgd = pgd_offset_k(address);
-		mm = NULL;
-	} else if (address >= TASK_SIZE)
-		return 1;
-	else if (!(mm = current->mm))
-		return 1;
-	else
-		pgd = pgd_offset(mm, address);
+	else {
+		if (unlikely(address >= TASK_SIZE || !current->mm))
+			return 1;
+
+		pgd = pgd_offset(current->mm, address);
+	}
 
-	pmd = pmd_offset(pgd, address);
+	pud = pud_offset(pgd, address);
+	if (pud_none_or_clear_bad(pud))
+		return 1;
+	pmd = pmd_offset(pud, address);
 	if (pmd_none_or_clear_bad(pmd))
 		return 1;
+
 	if (mm)
 		pte = pte_offset_map_lock(mm, pmd, address, &ptl);
 	else
 		pte = pte_offset_kernel(pmd, address);
 
 	entry = *pte;
-	if (pte_none(entry) || pte_not_present(entry)
-	    || (writeaccess && !pte_write(entry)))
+	if (unlikely(pte_none(entry) || pte_not_present(entry)))
+		goto unlock;
+	if (unlikely(writeaccess && !pte_write(entry)))
 		goto unlock;
 
 	if (writeaccess)
@@ -243,13 +246,7 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 	 * ITLB is not affected by "ldtlb" instruction.
 	 * So, we need to flush the entry by ourselves.
 	 */
-
-	{
-		unsigned long flags;
-		local_irq_save(flags);
-		__flush_tlb_page(get_asid(), address&PAGE_MASK);
-		local_irq_restore(flags);
-	}
+	__flush_tlb_page(get_asid(), address & PAGE_MASK);
 #endif
 
 	set_pte(pte, entry);
@@ -260,122 +257,3 @@ unlock:
 		pte_unmap_unlock(pte, ptl);
 	return ret;
 }
-
-void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
-{
-	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
-		unsigned long flags;
-		unsigned long asid;
-		unsigned long saved_asid = MMU_NO_ASID;
-
-		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
-		page &= PAGE_MASK;
-
-		local_irq_save(flags);
-		if (vma->vm_mm != current->mm) {
-			saved_asid = get_asid();
-			set_asid(asid);
-		}
-		__flush_tlb_page(asid, page);
-		if (saved_asid != MMU_NO_ASID)
-			set_asid(saved_asid);
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
-		     unsigned long end)
-{
-	struct mm_struct *mm = vma->vm_mm;
-
-	if (mm->context != NO_CONTEXT) {
-		unsigned long flags;
-		int size;
-
-		local_irq_save(flags);
-		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
-		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-			mm->context = NO_CONTEXT;
-			if (mm == current->mm)
-				activate_context(mm);
-		} else {
-			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
-			unsigned long saved_asid = MMU_NO_ASID;
-
-			start &= PAGE_MASK;
-			end += (PAGE_SIZE - 1);
-			end &= PAGE_MASK;
-			if (mm != current->mm) {
-				saved_asid = get_asid();
-				set_asid(asid);
-			}
-			while (start < end) {
-				__flush_tlb_page(asid, start);
-				start += PAGE_SIZE;
-			}
-			if (saved_asid != MMU_NO_ASID)
-				set_asid(saved_asid);
-		}
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_kernel_range(unsigned long start, unsigned long end)
-{
-	unsigned long flags;
-	int size;
-
-	local_irq_save(flags);
-	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
-	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-		flush_tlb_all();
-	} else {
-		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
-		unsigned long saved_asid = get_asid();
-
-		start &= PAGE_MASK;
-		end += (PAGE_SIZE - 1);
-		end &= PAGE_MASK;
-		set_asid(asid);
-		while (start < end) {
-			__flush_tlb_page(asid, start);
-			start += PAGE_SIZE;
-		}
-		set_asid(saved_asid);
-	}
-	local_irq_restore(flags);
-}
-
-void flush_tlb_mm(struct mm_struct *mm)
-{
-	/* Invalidate all TLB of this process. */
-	/* Instead of invalidating each TLB, we get new MMU context. */
-	if (mm->context != NO_CONTEXT) {
-		unsigned long flags;
-
-		local_irq_save(flags);
-		mm->context = NO_CONTEXT;
-		if (mm == current->mm)
-			activate_context(mm);
-		local_irq_restore(flags);
-	}
-}
-
-void flush_tlb_all(void)
-{
-	unsigned long flags, status;
-
-	/*
-	 * Flush all the TLB.
-	 *
-	 * Write to the MMU control register's bit:
-	 *	TF-bit for SH-3, TI-bit for SH-4.
-	 *      It's same position, bit #2.
-	 */
-	local_irq_save(flags);
-	status = ctrl_inl(MMUCR);
-	status |= 0x04;
-	ctrl_outl(status, MMUCR);
-	ctrl_barrier();
-	local_irq_restore(flags);
-}
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 8ea27ca4b700..d1a979eab656 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -80,6 +80,7 @@ void show_mem(void)
 static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 {
 	pgd_t *pgd;
+	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 
@@ -89,7 +90,17 @@ static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 		return;
 	}
 
-	pmd = pmd_offset(pgd, addr);
+	pud = pud_offset(pgd, addr);
+	if (pud_none(*pud)) {
+		pmd = (pmd_t *)get_zeroed_page(GFP_ATOMIC);
+		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
+		if (pmd != pmd_offset(pud, 0)) {
+			pud_ERROR(*pud);
+			return;
+		}
+	}
+
+	pmd = pmd_offset(pud, addr);
 	if (pmd_none(*pmd)) {
 		pte = (pte_t *)get_zeroed_page(GFP_ATOMIC);
 		set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
diff --git a/arch/sh/mm/tlb-flush.c b/arch/sh/mm/tlb-flush.c
new file mode 100644
index 000000000000..fd7e42bcaa40
--- /dev/null
+++ b/arch/sh/mm/tlb-flush.c
@@ -0,0 +1,132 @@
+/*
+ * TLB flushing operations for SH with an MMU.
+ *
+ *  Copyright (C) 1999  Niibe Yutaka
+ *  Copyright (C) 2003  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/mm.h>
+#include <asm/mmu_context.h>
+#include <asm/tlbflush.h>
+
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
+{
+	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		unsigned long asid;
+		unsigned long saved_asid = MMU_NO_ASID;
+
+		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		page &= PAGE_MASK;
+
+		local_irq_save(flags);
+		if (vma->vm_mm != current->mm) {
+			saved_asid = get_asid();
+			set_asid(asid);
+		}
+		__flush_tlb_page(asid, page);
+		if (saved_asid != MMU_NO_ASID)
+			set_asid(saved_asid);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+		     unsigned long end)
+{
+	struct mm_struct *mm = vma->vm_mm;
+
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+		int size;
+
+		local_irq_save(flags);
+		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
+			mm->context = NO_CONTEXT;
+			if (mm == current->mm)
+				activate_context(mm);
+		} else {
+			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
+			unsigned long saved_asid = MMU_NO_ASID;
+
+			start &= PAGE_MASK;
+			end += (PAGE_SIZE - 1);
+			end &= PAGE_MASK;
+			if (mm != current->mm) {
+				saved_asid = get_asid();
+				set_asid(asid);
+			}
+			while (start < end) {
+				__flush_tlb_page(asid, start);
+				start += PAGE_SIZE;
+			}
+			if (saved_asid != MMU_NO_ASID)
+				set_asid(saved_asid);
+		}
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+	unsigned long flags;
+	int size;
+
+	local_irq_save(flags);
+	size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
+		flush_tlb_all();
+	} else {
+		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
+		unsigned long saved_asid = get_asid();
+
+		start &= PAGE_MASK;
+		end += (PAGE_SIZE - 1);
+		end &= PAGE_MASK;
+		set_asid(asid);
+		while (start < end) {
+			__flush_tlb_page(asid, start);
+			start += PAGE_SIZE;
+		}
+		set_asid(saved_asid);
+	}
+	local_irq_restore(flags);
+}
+
+void flush_tlb_mm(struct mm_struct *mm)
+{
+	/* Invalidate all TLB of this process. */
+	/* Instead of invalidating each TLB, we get new MMU context. */
+	if (mm->context != NO_CONTEXT) {
+		unsigned long flags;
+
+		local_irq_save(flags);
+		mm->context = NO_CONTEXT;
+		if (mm == current->mm)
+			activate_context(mm);
+		local_irq_restore(flags);
+	}
+}
+
+void flush_tlb_all(void)
+{
+	unsigned long flags, status;
+
+	/*
+	 * Flush all the TLB.
+	 *
+	 * Write to the MMU control register's bit:
+	 *	TF-bit for SH-3, TI-bit for SH-4.
+	 *      It's same position, bit #2.
+	 */
+	local_irq_save(flags);
+	status = ctrl_inl(MMUCR);
+	status |= 0x04;
+	ctrl_outl(status, MMUCR);
+	ctrl_barrier();
+	local_irq_restore(flags);
+}
diff --git a/include/asm-sh/cache.h b/include/asm-sh/cache.h
index 33f13367054b..e3a180cf5062 100644
--- a/include/asm-sh/cache.h
+++ b/include/asm-sh/cache.h
@@ -10,7 +10,6 @@
 #ifdef __KERNEL__
 
 #include <asm/cpu/cache.h>
-#include <asm/cpu/cacheflush.h>
 
 #define SH_CACHE_VALID		1
 #define SH_CACHE_UPDATED	2
@@ -49,12 +48,5 @@ struct cache_info {
 	unsigned long flags;
 };
 
-/* Flush (write-back only) a region (smaller than a page) */
-extern void __flush_wback_region(void *start, int size);
-/* Flush (write-back & invalidate) a region (smaller than a page) */
-extern void __flush_purge_region(void *start, int size);
-/* Flush (invalidate only) a region (smaller than a page) */
-extern void __flush_invalidate_region(void *start, int size);
-
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_CACHE_H */
diff --git a/include/asm-sh/cacheflush.h b/include/asm-sh/cacheflush.h
index 9dfb33edb008..92930b4a40d4 100644
--- a/include/asm-sh/cacheflush.h
+++ b/include/asm-sh/cacheflush.h
@@ -2,6 +2,7 @@
 #define __ASM_SH_CACHEFLUSH_H
 #ifdef __KERNEL__
 
+#include <linux/mm.h>
 #include <asm/cpu/cacheflush.h>
 
 /* Flush (write-back only) a region (smaller than a page) */
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index f51aed00c68f..db0cb071ea8e 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -10,7 +10,7 @@
 #ifndef __ASM_CPU_SH3_CACHEFLUSH_H
 #define __ASM_CPU_SH3_CACHEFLUSH_H
 
-/* 
+/*
  * Cache flushing:
  *
  *  - flush_cache_all() flushes entire cache
@@ -35,10 +35,6 @@
  /* 32KB cache, 4kb PAGE sizes need to check bit 12 */
 #define CACHE_ALIAS 0x00001000
 
-struct page;
-struct mm_struct;
-struct vm_area_struct;
-
 extern void flush_cache_all(void);
 extern void flush_cache_mm(struct mm_struct *mm);
 extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
@@ -79,8 +75,6 @@ extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 
 #define p3_cache_init()				do { } while (0)
 
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #endif
 
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index ea58c4c5944d..a95fc951aff6 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -16,30 +16,26 @@
  *  caching; in which case they're only semi-broken),
  *  so we need them.
  */
-struct page;
-struct mm_struct;
-struct vm_area_struct;
-
-extern void flush_cache_all(void);
-extern void flush_cache_mm(struct mm_struct *mm);
-extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
-			      unsigned long end);
-extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
-extern void flush_dcache_page(struct page *pg);
+void flush_cache_all(void);
+void flush_cache_mm(struct mm_struct *mm);
+void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+		       unsigned long end);
+void flush_cache_page(struct vm_area_struct *vma, unsigned long addr,
+		      unsigned long pfn);
+void flush_dcache_page(struct page *pg);
 
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void flush_cache_sigtramp(unsigned long addr);
-extern void flush_icache_user_range(struct vm_area_struct *vma,
-				    struct page *page, unsigned long addr,
-				    int len);
+void flush_icache_range(unsigned long start, unsigned long end);
+void flush_cache_sigtramp(unsigned long addr);
+void flush_icache_user_range(struct vm_area_struct *vma, struct page *page,
+			     unsigned long addr, int len);
 
 #define flush_icache_page(vma,pg)		do { } while (0)
 
 /* Initialization of P3 area for copy_user_page */
-extern void p3_cache_init(void);
+void p3_cache_init(void);
 
 #define PG_mapped	PG_arch_1
 
@@ -57,4 +53,3 @@ static inline int remap_area_pages(unsigned long addr, unsigned long phys_addr,
 }
 #endif /* CONFIG_MMU */
 #endif /* __ASM_CPU_SH4_CACHEFLUSH_H */
-
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 4811d410d123..51d7281a546a 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -41,7 +41,8 @@ extern void (*copy_page)(void *to, void *from);
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
 
-#if defined(CONFIG_SH7705_CACHE_32KB) && defined(CONFIG_MMU)
+#if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
+	defined(CONFIG_SH7705_CACHE_32KB))
 struct page;
 extern void clear_user_page(void *to, unsigned long address, struct page *pg);
 extern void copy_user_page(void *to, void *from, unsigned long address, struct page *pg);
@@ -50,29 +51,20 @@ extern void __copy_user_page(void *to, void *from, void *orig_to);
 #elif defined(CONFIG_CPU_SH2) || defined(CONFIG_CPU_SH3) || !defined(CONFIG_MMU)
 #define clear_user_page(page, vaddr, pg)	clear_page(page)
 #define copy_user_page(to, from, vaddr, pg)	copy_page(to, from)
-#elif defined(CONFIG_CPU_SH4)
-struct page;
-extern void clear_user_page(void *to, unsigned long address, struct page *pg);
-extern void copy_user_page(void *to, void *from, unsigned long address, struct page *pg);
-extern void __clear_user_page(void *to, void *orig_to);
-extern void __copy_user_page(void *to, void *from, void *orig_to);
 #endif
 
 /*
  * These are used to make use of C type-checking..
  */
 typedef struct { unsigned long pte; } pte_t;
-typedef struct { unsigned long pmd; } pmd_t;
 typedef struct { unsigned long pgd; } pgd_t;
 typedef struct { unsigned long pgprot; } pgprot_t;
 
 #define pte_val(x)	((x).pte)
-#define pmd_val(x)	((x).pmd)
 #define pgd_val(x)	((x).pgd)
 #define pgprot_val(x)	((x).pgprot)
 
 #define __pte(x) ((pte_t) { (x) } )
-#define __pmd(x) ((pmd_t) { (x) } )
 #define __pgd(x) ((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
diff --git a/include/asm-sh/pgalloc.h b/include/asm-sh/pgalloc.h
index f4f233f7a4f5..e841465ab4d2 100644
--- a/include/asm-sh/pgalloc.h
+++ b/include/asm-sh/pgalloc.h
@@ -1,15 +1,6 @@
 #ifndef __ASM_SH_PGALLOC_H
 #define __ASM_SH_PGALLOC_H
 
-#include <linux/threads.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-
-#define pgd_quicklist ((unsigned long *)0)
-#define pmd_quicklist ((unsigned long *)0)
-#define pte_quicklist ((unsigned long *)0)
-#define pgtable_cache_size 0L
-
 #define pmd_populate_kernel(mm, pmd, pte) \
 		set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
 
@@ -24,38 +15,24 @@ static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd,
  */
 static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-	unsigned int pgd_size = (USER_PTRS_PER_PGD * sizeof(pgd_t));
-	pgd_t *pgd = (pgd_t *)kmalloc(pgd_size, GFP_KERNEL);
-
-	if (pgd)
-		memset(pgd, 0, pgd_size);
-
-	return pgd;
+	return (pgd_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline void pgd_free(pgd_t *pgd)
 {
-	kfree(pgd);
+	free_page((unsigned long)pgd);
 }
 
 static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 					  unsigned long address)
 {
-	pte_t *pte;
-
-	pte = (pte_t *) __get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
-
-	return pte;
+	return (pte_t *)__get_free_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline struct page *pte_alloc_one(struct mm_struct *mm,
 					 unsigned long address)
 {
-	struct page *pte;
-
-   	pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
-
-	return pte;
+	return alloc_page(GFP_KERNEL | __GFP_REPEAT | __GFP_ZERO);
 }
 
 static inline void pte_free_kernel(pte_t *pte)
@@ -75,14 +52,8 @@ static inline void pte_free(struct page *pte)
  * inside the pgd, so has no extra memory associated with it.
  */
 
-#define pmd_alloc_one(mm, addr)		({ BUG(); ((pmd_t *)2); })
 #define pmd_free(x)			do { } while (0)
 #define __pmd_free_tlb(tlb,x)		do { } while (0)
-#define pgd_populate(mm, pmd, pte)	BUG()
 #define check_pgt_cache()		do { } while (0)
 
-#ifdef CONFIG_CPU_SH4
-#define PG_mapped			PG_arch_1
-#endif
-
 #endif /* __ASM_SH_PGALLOC_H */
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 40d41a78041e..9728b58f7c13 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -1,42 +1,42 @@
-#ifndef __ASM_SH_PGTABLE_H
-#define __ASM_SH_PGTABLE_H
-
-#include <asm-generic/4level-fixup.h>
-
 /*
+ * This file contains the functions and defines necessary to modify and
+ * use the SuperH page table tree.
+ *
  * Copyright (C) 1999 Niibe Yutaka
- * Copyright (C) 2002, 2003, 2004 Paul Mundt
+ * Copyright (C) 2002 - 2005 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License.  See the file "COPYING" in the main directory of this
+ * archive for more details.
  */
+#ifndef __ASM_SH_PGTABLE_H
+#define __ASM_SH_PGTABLE_H
 
-#include <asm/pgtable-2level.h>
+#include <asm-generic/pgtable-nopmd.h>
+#include <asm/page.h>
+
+#define PTRS_PER_PGD		1024
 
-/*
- * This file contains the functions and defines necessary to modify and use
- * the SuperH page table tree.
- */
 #ifndef __ASSEMBLY__
-#include <asm/processor.h>
 #include <asm/addrspace.h>
 #include <asm/fixmap.h>
-#include <linux/threads.h>
 
 extern pgd_t swapper_pg_dir[PTRS_PER_PGD];
 extern void paging_init(void);
 
-/*
- * Basically we have the same two-level (which is the logical three level
- * Linux page table layout folded) page tables as the i386.
- */
-
 /*
  * ZERO_PAGE is a global shared page that is always zero: used
  * for zero-mapped memory areas etc..
  */
-extern unsigned long empty_zero_page[1024];
+extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
 
 #endif /* !__ASSEMBLY__ */
 
+/* traditional two-level paging structure */
+#define PGDIR_SHIFT	22
+#define PTRS_PER_PMD	1
+#define PTRS_PER_PTE	1024
 #define PMD_SIZE	(1UL << PMD_SHIFT)
 #define PMD_MASK	(~(PMD_SIZE-1))
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
@@ -47,7 +47,6 @@ extern unsigned long empty_zero_page[1024];
 
 #define PTE_PHYS_MASK	0x1ffff000
 
-#ifndef __ASSEMBLY__
 /*
  * First 1MB map is used by fixed purpose.
  * Currently only 4-enty (16kB) is used (see arch/sh/mm/cache.c)
@@ -65,7 +64,7 @@ extern unsigned long empty_zero_page[1024];
 #define _PAGE_SZ1	0x080  /* SZ1-bit : Size of page (on SH-4) */
 #define _PAGE_PRESENT	0x100  /* V-bit   : page is valid */
 #define _PAGE_PROTNONE	0x200  /* software: if not present  */
-#define _PAGE_ACCESSED 	0x400  /* software: page referenced */
+#define _PAGE_ACCESSED	0x400  /* software: page referenced */
 #define _PAGE_U0_SHARED 0x800  /* software: page is shared in user space */
 
 #define	_PAGE_FILE	_PAGE_WT  /* software: pagecache or swap? */
@@ -83,7 +82,6 @@ extern unsigned long empty_zero_page[1024];
 #define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
 #define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
 
-
 /* Mask which drop software flags
  * We also drop WT bit since it is used for _PAGE_FILE
  * bit in this implementation.
@@ -115,6 +113,8 @@ extern unsigned long empty_zero_page[1024];
 #define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_SHARED)
 
+#ifndef __ASSEMBLY__
+
 #ifdef CONFIG_MMU
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_SHARED | _PAGE_FLAGS_HARD)
@@ -137,12 +137,13 @@ extern unsigned long empty_zero_page[1024];
 #define PAGE_KERNEL_PCC		__pgprot(0)
 #endif
 
+#endif /* __ASSEMBLY__ */
+
 /*
  * As i386 and MIPS, SuperH can't do page protection for execute, and
  * considers that the same as a read.  Also, write permissions imply
- * read permissions. This is the closest we can get..  
+ * read permissions. This is the closest we can get..
  */
-
 #define __P000	PAGE_NONE
 #define __P001	PAGE_READONLY
 #define __P010	PAGE_COPY
@@ -161,6 +162,26 @@ extern unsigned long empty_zero_page[1024];
 #define __S110	PAGE_SHARED
 #define __S111	PAGE_SHARED
 
+#ifndef __ASSEMBLY__
+
+/*
+ * Certain architectures need to do special things when PTEs
+ * within a page table are directly modified.  Thus, the following
+ * hook is made available.
+ */
+#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
+#define set_pte_at(mm,addr,ptep,pteval) set_pte(ptep,pteval)
+
+/*
+ * (pmds are folded into pgds so this doesn't get actually called,
+ * but the define is needed for a generic inline function.)
+ */
+#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
+
+#define pte_pfn(x)		((unsigned long)(((x).pte >> PAGE_SHIFT)))
+#define pfn_pte(pfn, prot)	__pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+#define pfn_pmd(pfn, prot)	__pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
+
 #define pte_none(x)	(!pte_val(x))
 #define pte_present(x)	(pte_val(x) & (_PAGE_PRESENT | _PAGE_PROTNONE))
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
@@ -171,7 +192,7 @@ extern unsigned long empty_zero_page[1024];
 #define	pmd_bad(x)	((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER)) != _KERNPG_TABLE)
 
 #define pages_to_mb(x)	((x) >> (20-PAGE_SHIFT))
-#define pte_page(x) 	phys_to_page(pte_val(x)&PTE_PHYS_MASK)
+#define pte_page(x)	phys_to_page(pte_val(x)&PTE_PHYS_MASK)
 
 /*
  * The following only work if pte_present() is true.
@@ -248,6 +269,11 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 #define pte_unmap(pte)		do { } while (0)
 #define pte_unmap_nested(pte)	do { } while (0)
 
+#define pte_ERROR(e) \
+	printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, pte_val(e))
+#define pgd_ERROR(e) \
+	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
+
 struct vm_area_struct;
 extern void update_mmu_cache(struct vm_area_struct * vma,
 			     unsigned long address, pte_t pte);
@@ -272,8 +298,6 @@ extern void update_mmu_cache(struct vm_area_struct * vma,
 
 typedef pte_t *pte_addr_t;
 
-#endif /* !__ASSEMBLY__ */
-
 #define kern_addr_valid(addr)	(1)
 
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
@@ -301,5 +325,7 @@ extern pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t
 
 #include <asm-generic/pgtable.h>
 
+#endif /* !__ASSEMBLY__ */
+
 #endif /* __ASM_SH_PAGE_H */
 
-- 
cgit v1.2.1


From 715a0ecc29c850d2b2f76e1803d3f22cd5a0ac0d Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 26 Sep 2006 23:14:21 -0700
Subject: [SPARC64]: Kill bogus check from bootmem_init().

There is an ancient and totally incorrect sanity check being
done on the ramdisk location.  The check assumes that the
kernel is always loaded to physical address zero, which is
wrong.  It was trying to validate the ramdisk value by saying that
if it fell within the kernel image address range it must be wrong.

Anyways, kill this because it actually creates problems.  The
'ramdisk_image' should always be adjusted down by KERNBASE.
SILO can easily put the ramdisk in a location which causes
this test to trigger, breaking things.

[ Based almost entirely upon a patch from Ben Collins. ]

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/mm/init.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index dcba4e6ab570..09cb7fccc03a 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -920,8 +920,7 @@ static unsigned long __init bootmem_init(unsigned long *pages_avail,
 	if (sparc_ramdisk_image || sparc_ramdisk_image64) {
 		unsigned long ramdisk_image = sparc_ramdisk_image ?
 			sparc_ramdisk_image : sparc_ramdisk_image64;
-		if (ramdisk_image >= (unsigned long)_end - 2 * PAGE_SIZE)
-			ramdisk_image -= KERNBASE;
+		ramdisk_image -= KERNBASE;
 		initrd_start = ramdisk_image + phys_base;
 		initrd_end = initrd_start + sparc_ramdisk_size;
 		if (initrd_end > end_of_phys_memory) {
-- 
cgit v1.2.1


From d7cdc9e8ac82c43fdcd4fde6b5b53d2dcba7f707 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:16:42 +0900
Subject: sh: ioremap() overhaul.

ioremap() overhaul. Add support for transparent PMB mapping, get rid of
p3_ioremap(), etc. Also drop ioremap() and iounmap() routines from the
machvec, as everyone can use the generic ioremap() API instead. For PCI
memory apertures and other special cases, use the pci_iomap() API, as
boards are already required to get the mapping right there.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/landisk/io.c                | 236 ++++++++++++----------------
 arch/sh/boards/landisk/setup.c             | 237 ++++++++++++-----------------
 arch/sh/boards/renesas/hs7751rvoip/io.c    |   9 --
 arch/sh/boards/renesas/hs7751rvoip/setup.c |   1 -
 arch/sh/boards/renesas/rts7751r2d/io.c     |   9 --
 arch/sh/boards/renesas/rts7751r2d/mach.c   |   2 -
 arch/sh/boards/titan/io.c                  |   9 +-
 arch/sh/boards/titan/setup.c               |   1 -
 arch/sh/drivers/pci/pci.c                  |  21 ++-
 arch/sh/kernel/cf-enabler.c                |   5 +-
 arch/sh/kernel/sh_ksyms.c                  |   1 -
 arch/sh/mm/Makefile                        |   7 +-
 arch/sh/mm/pmb.c                           | 145 +++++++++++++++++-
 include/asm-sh/landisk/iodata_landisk.h    |   3 +
 include/asm-sh/mmu.h                       |  11 +-
 15 files changed, 374 insertions(+), 323 deletions(-)

diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
index 1f1679af09d0..aa6b145c9e8f 100644
--- a/arch/sh/boards/landisk/io.c
+++ b/arch/sh/boards/landisk/io.c
@@ -17,9 +17,9 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <asm/io.h>
 #include <asm/landisk/iodata_landisk.h>
 #include <asm/addrspace.h>
+#include <asm/io.h>
 
 #include <linux/module.h>
 #include <linux/pci.h>
@@ -42,10 +42,6 @@ extern void *area6_io_base;	/* Area 6 I/O Base address */
 
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -66,7 +62,7 @@ static inline unsigned long port2adr(unsigned int port)
 			return ((unsigned long)area6_io_base + PA_SIDE_OFFSET +
 				((port - 0x170) << 1));
 	else
-		maybebadio(port2adr, (unsigned long)port);
+		maybebadio((unsigned long)port);
 
 	return port;
 }
@@ -89,234 +85,200 @@ static inline unsigned long port2adr(unsigned int port)
  * should be way beyond the window, and is used  w/o translation for
  * compatibility.
  */
-unsigned char landisk_inb(unsigned long port)
+u8 landisk_inb(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned char *)port;
+		return ctrl_inb(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
-	else
-		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+		return ctrl_inb(PCI_IOMAP(port));
+
+	return ctrl_inw(port2adr(port)) & 0xff;
 }
 
-unsigned char landisk_inb_p(unsigned long port)
+u8 landisk_inb_p(unsigned long port)
 {
-	unsigned char v;
+	u8 v;
 
 	if (PXSEG(port))
-		v = *(volatile unsigned char *)port;
+		v = ctrl_inb(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+		v = ctrl_inb(PCI_IOMAP(port));
 	else
-		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+		v = ctrl_inw(port2adr(port)) & 0xff;
+
 	delay();
 
 	return v;
 }
 
-unsigned short landisk_inw(unsigned long port)
+u16 landisk_inw(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned short *)port;
+		return ctrl_inw(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+		return ctrl_inw(PCI_IOMAP(port));
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 
 	return 0;
 }
 
-unsigned int landisk_inl(unsigned long port)
+u32 landisk_inl(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned long *)port;
+		return ctrl_inl(port);
 	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned long *)PCI_IOMAP(port);
+		return ctrl_inl(PCI_IOMAP(port));
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 
 	return 0;
 }
 
-void landisk_outb(unsigned char value, unsigned long port)
+void landisk_outb(u8 value, unsigned long port)
 {
-
 	if (PXSEG(port))
-		*(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
 }
 
-void landisk_outb_p(unsigned char value, unsigned long port)
+void landisk_outb_p(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
 	delay();
 }
 
-void landisk_outw(unsigned short value, unsigned long port)
+void landisk_outw(u16 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned short *)port = value;
+		ctrl_outw(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+		ctrl_outw(value, PCI_IOMAP(port));
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
-void landisk_outl(unsigned int value, unsigned long port)
+void landisk_outl(u32 value, unsigned long port)
 {
 	if (PXSEG(port))
-		*(volatile unsigned long *)port = value;
+		ctrl_outl(value, port);
 	else if (CHECK_SH7751_PCIIO(port))
-		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+		ctrl_outl(value, PCI_IOMAP(port));
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
-void landisk_insb(unsigned long port, void *addr, unsigned long count)
+void landisk_insb(unsigned long port, void *dst, unsigned long count)
 {
-	if (PXSEG(port))
-		while (count--)
-			*((unsigned char *)addr)++ =
-			    *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+        volatile u16 *p;
+        u8 *buf = dst;
 
-		while (count--)
-			*((volatile unsigned char *)addr)++ = *bp;
-	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+        if (PXSEG(port)) {
+                while (count--)
+                        *buf++ = *(volatile u8 *)port;
+	} else if (CHECK_SH7751_PCIIO(port)) {
+                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--)
-			*((unsigned char *)addr)++ = *p;
+                while (count--)
+                        *buf++ = *bp;
+	} else {
+                p = (volatile u16 *)port2adr(port);
+                while (count--)
+                        *buf++ = *p & 0xff;
 	}
 }
 
-void landisk_insw(unsigned long port, void *addr, unsigned long count)
+void landisk_insw(unsigned long port, void *dst, unsigned long count)
 {
-	volatile __u16 *p;
+        volatile u16 *p;
+        u16 *buf = dst;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
+		p = (volatile u16 *)port2adr(port);
 	while (count--)
-		*((__u16 *) addr)++ = *p;
+		*buf++ = *p;
 }
 
-void landisk_insl(unsigned long port, void *addr, unsigned long count)
+void landisk_insl(unsigned long port, void *dst, unsigned long count)
 {
+        u32 *buf = dst;
+
 	if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
+                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-		while (count--)
-			*((__u32 *) addr)++ = *p;
+                while (count--)
+                        *buf++ = *p;
 	} else
-		maybebadio(insl, port);
+		maybebadio(port);
 }
 
-void landisk_outsb(unsigned long port, const void *addr, unsigned long count)
+void landisk_outsb(unsigned long port, const void *src, unsigned long count)
 {
+        volatile u16 *p;
+        const u8 *buf = src;
+
 	if (PXSEG(port))
-		while (count--)
-			*(volatile unsigned char *)port =
-			    *((unsigned char *)addr)++;
+                while (count--)
+                        ctrl_outb(*buf++, port);
 	else if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u8 *bp = (__u8 *) PCI_IOMAP(port);
+                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--)
-			*bp = *((volatile unsigned char *)addr)++;
+                while (count--)
+                        *bp = *buf++;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
-
-		while (count--)
-			*p = *((unsigned char *)addr)++;
+                p = (volatile u16 *)port2adr(port);
+                while (count--)
+                        *p = *buf++;
 	}
 }
 
-void landisk_outsw(unsigned long port, const void *addr, unsigned long count)
+void landisk_outsw(unsigned long port, const void *src, unsigned long count)
 {
-	volatile __u16 *p;
+        volatile u16 *p;
+        const u16 *buf = src;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+                p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+                p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--)
-		*p = *((__u16 *) addr)++;
-}
+                p = (volatile u16 *)port2adr(port);
 
-void landisk_outsl(unsigned long port, const void *addr, unsigned long count)
-{
-	if (CHECK_SH7751_PCIIO(port)) {
-		volatile __u32 *p = (__u32 *) PCI_IOMAP(port);
-
-		while (count--)
-			*p = *((__u32 *) addr)++;
-	} else
-		maybebadio(outsl, port);
+        while (count--)
+                *p = *buf++;
 }
 
-/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
-/* already set up.  For a larger memory space, these would need to */
-/* reset PCIMBR as needed on a per-call basis...                   */
-
-unsigned char landisk_readb(unsigned long addr)
+void landisk_outsl(unsigned long port, const void *src, unsigned long count)
 {
-	return *(volatile unsigned char *)addr;
-}
+        const u32 *buf = src;
 
-unsigned short landisk_readw(unsigned long addr)
-{
-	return *(volatile unsigned short *)addr;
-}
-
-unsigned int landisk_readl(unsigned long addr)
-{
-	return *(volatile unsigned long *)addr;
-}
-
-void landisk_writeb(unsigned char b, unsigned long addr)
-{
-	*(volatile unsigned char *)addr = b;
-}
-
-void landisk_writew(unsigned short b, unsigned long addr)
-{
-	*(volatile unsigned short *)addr = b;
-}
-
-void landisk_writel(unsigned int b, unsigned long addr)
-{
-	*(volatile unsigned long *)addr = b;
-}
+	if (CHECK_SH7751_PCIIO(port)) {
+                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-void *landisk_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
+                while (count--)
+                        *p = *buf++;
+	} else
+		maybebadio(port);
 }
 
-void landisk_iounmap(void *addr)
+void __iomem *landisk_ioport_map(unsigned long port, unsigned int size)
 {
-}
+        if (PXSEG(port))
+                return (void __iomem *)port;
+        else if (CHECK_SH7751_PCIIO(port))
+                return (void __iomem *)PCI_IOMAP(port);
 
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-unsigned long landisk_isa_port2addr(unsigned long offset)
-{
-	return port2adr(offset);
+        return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
index 0c60eaa10ba7..3a795cfb1eda 100644
--- a/arch/sh/boards/landisk/setup.c
+++ b/arch/sh/boards/landisk/setup.c
@@ -1,157 +1,52 @@
 /*
  * arch/sh/boards/landisk/setup.c
  *
+ * Copyright (C) 2000 Kazumoto Kojima
  * Copyright (C) 2002 Paul Mundt
  *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Setup code for an unknown machine (internal peripherials only)
- */
-/*
- * linux/arch/sh/kernel/setup_landisk.c
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
  * I-O DATA Device, Inc. LANDISK Support.
  *
  * Modified for LANDISK by
  * Atom Create Engineering Co., Ltd. 2002.
- */
-/*
+ *
  * modifed by kogiidena
  * 2005.09.16
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/config.h>
 #include <linux/init.h>
-#include <linux/irq.h>
 #include <linux/pm.h>
-
-#include <linux/hdreg.h>
-#include <linux/ide.h>
-#include <linux/pci.h>
-
+#include <linux/mm.h>
 #include <asm/machvec.h>
 #include <asm/rtc.h>
-#include <asm/machvec_init.h>
-#include <asm/io.h>
 #include <asm/landisk/iodata_landisk.h>
-#include <asm/landisk/io.h>
-
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
+#include <asm/io.h>
 
-extern void (*board_time_init) (void);
 void landisk_time_init(void);
-extern void init_landisk_IRQ(void);
+void init_landisk_IRQ(void);
 
 int landisk_ledparam;
 int landisk_buzzerparam;
 int landisk_arch;
 
-/* defined in mm/ioremap.c */
-extern void *p3_ioremap(unsigned long phys_addr, unsigned long size,
-			unsigned long flags);
-
-/*
- * Initialize the board
- */
-
-const char *get_system_type(void)
-{
-	return "LANDISK";
-}
-
-static void landisk_power_off(void)
-{
-	ctrl_outb(0x01, PA_SHUTDOWN);
-}
-
-void check_usl5p(void)
-{
-	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
-	unsigned char tmp1, tmp2;
-	tmp1 = *p;
-	*p = 0x40;
-	tmp2 = *p;
-	*p = tmp1;
-	landisk_arch = (tmp2 == 0x40) ? 1 : 0;
-	if (landisk_arch == 1) {	/* arch == usl-5p */
-		landisk_ledparam = 0x00000380;
-		landisk_ledparam |= (tmp1 & 0x07c);
-	} else {                        /* arch == landisk */
-		landisk_ledparam = 0x02000180;
-		landisk_ledparam |= 0x04;
-	}
-	return;
-}
-
-void __init platform_setup(void)
-{
-
-	landisk_buzzerparam = 0;
-	check_usl5p();
-
-	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
-	board_time_init = landisk_time_init;
-	pm_power_off = landisk_power_off;
-
-}
-
-void *area5_io_base;
-void *area6_io_base;
-
-int __init cf_init(void)
-{
-	pgprot_t prot;
-	unsigned long paddrbase, psize;
-
-	/* open I/O area window */
-	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
-	psize = PAGE_SIZE;
-	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
-	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
-	if (!area5_io_base) {
-		printk("allocate_cf_area : can't open CF I/O window!\n");
-		return -ENOMEM;
-	}
-
-	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
-	psize = PAGE_SIZE;
-	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
-	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
-	if (!area6_io_base) {
-		printk("allocate_cf_area : can't open HDD I/O window!\n");
-		return -ENOMEM;
-	}
-
-	printk(KERN_INFO "Allocate Area5/6 success.\n");
-
-	/* XXX : do we need attribute and common-memory area also? */
-
-	return 0;
-}
-
-__initcall(cf_init);
-
-#include <linux/sched.h>
-
-/* Cycle the LED's in the clasic knightrider/Sun pattern */
-
-void heartbeat_landisk(void)
+/* cycle the led's in the clasic knightrider/sun pattern */
+static void heartbeat_landisk(void)
 {
 	static unsigned int cnt = 0, blink = 0x00, period = 25;
-	volatile unsigned char *p = (volatile unsigned char *)PA_LED;
+        volatile u8 *p = (volatile u8 *)PA_LED;
 	char data;
 
-	if ((landisk_ledparam & 0x080) == 0) {
+        if ((landisk_ledparam & 0x080) == 0)
 		return;
-	}
+
 	cnt += 1;
-	if (cnt < period) {
+
+        if (cnt < period)
 		return;
-	}
+
 	cnt = 0;
 	blink++;
 
@@ -167,17 +62,16 @@ void heartbeat_landisk(void)
 	}
 	*p = data;
 
-	if (((landisk_ledparam & 0x007f7f00) == 0)
-	    && (landisk_buzzerparam == 0)) {
+        if (((landisk_ledparam & 0x007f7f00) == 0) &&
+             (landisk_buzzerparam == 0))
 		landisk_ledparam &= (~0x0080);
-	}
+
 	landisk_buzzerparam >>= 1;
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_landisk __initmv = {
 	.mv_nr_irqs = 72,
 	.mv_inb = landisk_inb,
@@ -198,21 +92,88 @@ struct sh_machine_vector mv_landisk __initmv = {
 	.mv_outsb = landisk_outsb,
 	.mv_outsw = landisk_outsw,
 	.mv_outsl = landisk_outsl,
-	.mv_readb = landisk_readb,
-	.mv_readw = landisk_readw,
-	.mv_readl = landisk_readl,
-	.mv_writeb = landisk_writeb,
-	.mv_writew = landisk_writew,
-	.mv_writel = landisk_writel,
-	.mv_ioremap = landisk_ioremap,
-	.mv_iounmap = landisk_iounmap,
-	.mv_isa_port2addr = landisk_isa_port2addr,
+        .mv_ioport_map = landisk_ioport_map,
 	.mv_init_irq = init_landisk_IRQ,
-
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat = heartbeat_landisk,
 #endif
-
 };
-
 ALIAS_MV(landisk)
+
+const char *get_system_type(void)
+{
+        return "LANDISK";
+}
+
+static void landisk_power_off(void)
+{
+        ctrl_outb(0x01, PA_SHUTDOWN);
+}
+
+static void check_usl5p(void)
+{
+        volatile u8 *p = (volatile u8 *)PA_LED;
+        u8 tmp1, tmp2;
+
+        tmp1 = *p;
+        *p = 0x40;
+        tmp2 = *p;
+        *p = tmp1;
+
+        landisk_arch = (tmp2 == 0x40);
+        if (landisk_arch == 1) {
+                /* arch == usl-5p */
+                landisk_ledparam = 0x00000380;
+                landisk_ledparam |= (tmp1 & 0x07c);
+        } else {
+                /* arch == landisk */
+                landisk_ledparam = 0x02000180;
+                landisk_ledparam |= 0x04;
+        }
+}
+
+void __init platform_setup(void)
+{
+        landisk_buzzerparam = 0;
+        check_usl5p();
+
+        printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+        board_time_init = landisk_time_init;
+        pm_power_off = landisk_power_off;
+}
+
+void *area5_io_base;
+void *area6_io_base;
+
+static int __init landisk_cf_init(void)
+{
+	pgprot_t prot;
+	unsigned long paddrbase, psize;
+
+	/* open I/O area window */
+	paddrbase = virt_to_phys((void *)PA_AREA5_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(1, _PAGE_PCC_IO16);
+	area5_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area5_io_base) {
+		printk("allocate_cf_area : can't open CF I/O window!\n");
+		return -ENOMEM;
+	}
+
+	paddrbase = virt_to_phys((void *)PA_AREA6_IO);
+	psize = PAGE_SIZE;
+	prot = PAGE_KERNEL_PCC(0, _PAGE_PCC_IO16);
+	area6_io_base = p3_ioremap(paddrbase, psize, prot.pgprot);
+	if (!area6_io_base) {
+		printk("allocate_cf_area : can't open HDD I/O window!\n");
+		return -ENOMEM;
+	}
+
+	printk(KERN_INFO "Allocate Area5/6 success.\n");
+
+	/* XXX : do we need attribute and common-memory area also? */
+
+	return 0;
+}
+
+__initcall(landisk_cf_init);
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 09fb77ffb835..edecf107fc13 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -294,15 +294,6 @@ void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count
 		maybebadio(outsl, port);
 }
 
-void *hs7751rvoip_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
-}
-EXPORT_SYMBOL(hs7751rvoip_ioremap);
-
 unsigned long hs7751rvoip_isa_port2addr(unsigned long offset)
 {
 	return port2adr(offset);
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index 813fc4d5862a..aa51bda931f6 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -60,7 +60,6 @@ struct sh_machine_vector mv_hs7751rvoip __initmv = {
 	.mv_outsw		= hs7751rvoip_outsw,
 	.mv_outsl		= hs7751rvoip_outsl,
 
-	.mv_ioremap		= hs7751rvoip_ioremap,
 	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
 	.mv_init_irq		= hs7751rvoip_init_irq,
 };
diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c
index 9e7fa726a86d..8dc2a2e2e5df 100644
--- a/arch/sh/boards/renesas/rts7751r2d/io.c
+++ b/arch/sh/boards/renesas/rts7751r2d/io.c
@@ -321,15 +321,6 @@ void rts7751r2d_outsl(unsigned long port, const void *addr, unsigned long count)
 		maybebadio(port);
 }
 
-void *rts7751r2d_ioremap(unsigned long offset, unsigned long size)
-{
-	if (offset >= 0xfd000000)
-		return (void *)offset;
-	else
-		return (void *)P2SEGADDR(offset);
-}
-EXPORT_SYMBOL(rts7751r2d_ioremap);
-
 unsigned long rts7751r2d_isa_port2addr(unsigned long offset)
 {
 	return port2adr(offset);
diff --git a/arch/sh/boards/renesas/rts7751r2d/mach.c b/arch/sh/boards/renesas/rts7751r2d/mach.c
index 175a93d726e8..fe3e8735e9f8 100644
--- a/arch/sh/boards/renesas/rts7751r2d/mach.c
+++ b/arch/sh/boards/renesas/rts7751r2d/mach.c
@@ -19,7 +19,6 @@
 
 extern void heartbeat_rts7751r2d(void);
 extern void init_rts7751r2d_IRQ(void);
-extern void *rts7751r2d_ioremap(unsigned long, unsigned long);
 extern int rts7751r2d_irq_demux(int irq);
 
 extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
@@ -53,7 +52,6 @@ struct sh_machine_vector mv_rts7751r2d __initmv = {
 	.mv_outsw		= rts7751r2d_outsw,
 	.mv_outsl		= rts7751r2d_outsl,
 
-	.mv_ioremap		= rts7751r2d_ioremap,
 	.mv_init_irq		= init_rts7751r2d_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_rts7751r2d,
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index d66900c99a11..b886fd233a66 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -138,19 +138,12 @@ void titan_outsl(unsigned long port, const void *src, unsigned long count)
         maybebadio(port);
 }
 
-void *titan_ioremap(unsigned long offset, unsigned long size) {
-	if (CHECK_SH7751_PCIIO(offset) || CHECK_SH7751_PCIMEMIO(offset))
-		return (void *)offset;
-}
-
 void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
 {
-	if (PXSEG(port))
+	if (PXSEG(port) || CHECK_SH7751_PCIMEMIO(port))
 		return (void __iomem *)port;
 	else if (CHECK_SH7751_PCIIO(port))
 		return (void __iomem *)PCI_IOMAP(port);
 
 	return (void __iomem *)port2adr(port);
 }
-
-EXPORT_SYMBOL(titan_ioremap);
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
index 6ac5c8d7b3fb..c8b431c1d0fd 100644
--- a/arch/sh/boards/titan/setup.c
+++ b/arch/sh/boards/titan/setup.c
@@ -51,7 +51,6 @@ struct sh_machine_vector mv_titan __initmv = {
 	.mv_insl =	titan_insl,
 	.mv_outsl =	titan_outsl,
 
-	.mv_ioremap =	titan_ioremap,
 	.mv_ioport_map = titan_ioport_map,
 
 	.mv_init_irq =	init_titan_irq,
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 1f5e23e8b163..285dffd12bd8 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -2,7 +2,7 @@
  * arch/sh/drivers/pci/pci.c
  *
  * Copyright (c) 2002 M. R. Brown  <mrbrown@linux-sh.org>
- * Copyright (c) 2004, 2005 Paul Mundt  <lethal@linux-sh.org>
+ * Copyright (c) 2004 - 2006 Paul Mundt  <lethal@linux-sh.org>
  *
  * These functions are collected here to reduce duplication of common
  * code amongst the many platform-specific PCI support code files.
@@ -172,10 +172,23 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 		return NULL;
 	if (maxlen && len > maxlen)
 		len = maxlen;
-	if (flags & IORESOURCE_IO)
+
+	/*
+	 * Presently the IORESOURCE_MEM case is a bit special, most
+	 * SH7751 style PCI controllers have PCI memory at a fixed
+	 * location in the address space where no remapping is desired
+	 * (traditionally at 0xfd000000). Once this changes, the
+	 * IORESOURCE_MEM case will have to switch to using ioremap() and
+	 * more care will have to be taken to inhibit page table mapping
+	 * for legacy cores.
+	 *
+	 * For now everything wraps to ioport_map(), since boards that
+	 * have PCI will be able to check the address range properly on
+	 * their own.
+	 *			-- PFM.
+	 */
+	if (flags & (IORESOURCE_IO | IORESOURCE_MEM))
 		return ioport_map(start, len);
-	if (flags & IORESOURCE_MEM)
-		return ioremap(start, len);
 
 	return NULL;
 }
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index f1f9ab87f0b0..c9b823d1d073 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -41,9 +41,6 @@
 #define slot_no 1
 #endif
 
-/* defined in mm/ioremap.c */
-extern void * p3_ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags);
-
 /* use this pointer to access to directly connected compact flash io area*/
 void *cf_io_base;
 
@@ -62,7 +59,7 @@ static int __init allocate_cf_area(void)
 		return -ENOMEM;
 	}
 /*	printk("p3_ioremap(paddr=0x%08lx, psize=0x%08lx, prot=0x%08lx)=0x%08lx\n",
-	    	paddrbase, psize, prot.pgprot, cf_io_base);*/
+		paddrbase, psize, prot.pgprot, cf_io_base);*/
 
 	/* XXX : do we need attribute and common-memory area also? */
 
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 8a6dd06fd071..b73feafcd06e 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -27,7 +27,6 @@ EXPORT_SYMBOL(sh_mv);
 
 /* platform dependent support */
 EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(iounmap);
 EXPORT_SYMBOL(enable_irq);
 EXPORT_SYMBOL(disable_irq);
 EXPORT_SYMBOL(probe_irq_mask);
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index d90906367c5f..87a7c07265c0 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -12,13 +12,14 @@ obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 
 mmu-y			:= fault-nommu.o tlb-nommu.o pg-nommu.o
-mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o
+mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o	\
+			   ioremap.o
 
 obj-y			+= $(mmu-y)
 
 ifdef CONFIG_MMU
-obj-$(CONFIG_CPU_SH3)	+= tlb-sh3.o
-obj-$(CONFIG_CPU_SH4)	+= tlb-sh4.o ioremap.o
+obj-$(CONFIG_CPU_SH3)		+= tlb-sh3.o
+obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index ff5bde745647..819fd0faf022 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -3,7 +3,7 @@
  *
  * Privileged Space Mapping Buffer (PMB) Support.
  *
- * Copyright (C) 2005 Paul Mundt
+ * Copyright (C) 2005, 2006 Paul Mundt
  *
  * P1/P2 Section mapping definitions from map32.h, which was:
  *
@@ -24,6 +24,7 @@
 #include <linux/err.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
+#include <asm/pgtable.h>
 #include <asm/mmu.h>
 #include <asm/io.h>
 
@@ -127,11 +128,15 @@ repeat:
 	return 0;
 }
 
-void set_pmb_entry(struct pmb_entry *pmbe)
+int set_pmb_entry(struct pmb_entry *pmbe)
 {
+	int ret;
+
 	jump_to_P2();
-	__set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
+	ret = __set_pmb_entry(pmbe->vpn, pmbe->ppn, pmbe->flags, &pmbe->entry);
 	back_to_P1();
+
+	return ret;
 }
 
 void clear_pmb_entry(struct pmb_entry *pmbe)
@@ -162,11 +167,141 @@ void clear_pmb_entry(struct pmb_entry *pmbe)
 	clear_bit(entry, &pmb_map);
 }
 
+static DEFINE_SPINLOCK(pmb_list_lock);
+static struct pmb_entry *pmb_list;
+
+static inline void pmb_list_add(struct pmb_entry *pmbe)
+{
+	struct pmb_entry **p, *tmp;
+
+	p = &pmb_list;
+	while ((tmp = *p) != NULL)
+		p = &tmp->next;
+
+	pmbe->next = tmp;
+	*p = pmbe;
+}
+
+static inline void pmb_list_del(struct pmb_entry *pmbe)
+{
+	struct pmb_entry **p, *tmp;
+
+	for (p = &pmb_list; (tmp = *p); p = &tmp->next)
+		if (tmp == pmbe) {
+			*p = tmp->next;
+			return;
+		}
+}
+
+static struct {
+	unsigned long size;
+	int flag;
+} pmb_sizes[] = {
+	{ .size	= 0x20000000, .flag = PMB_SZ_512M, },
+	{ .size = 0x08000000, .flag = PMB_SZ_128M, },
+	{ .size = 0x04000000, .flag = PMB_SZ_64M,  },
+	{ .size = 0x01000000, .flag = PMB_SZ_16M,  },
+};
+
+long pmb_remap(unsigned long vaddr, unsigned long phys,
+	       unsigned long size, unsigned long flags)
+{
+	struct pmb_entry *pmbp;
+	unsigned long wanted;
+	int pmb_flags, i;
+
+	/* Convert typical pgprot value to the PMB equivalent */
+	if (flags & _PAGE_CACHABLE) {
+		if (flags & _PAGE_WT)
+			pmb_flags = PMB_WT;
+		else
+			pmb_flags = PMB_C;
+	} else
+		pmb_flags = PMB_WT | PMB_UB;
+
+	pmbp = NULL;
+	wanted = size;
+
+again:
+	for (i = 0; i < ARRAY_SIZE(pmb_sizes); i++) {
+		struct pmb_entry *pmbe;
+		int ret;
+
+		if (size < pmb_sizes[i].size)
+			continue;
+
+		pmbe = pmb_alloc(vaddr, phys, pmb_flags | pmb_sizes[i].flag);
+		if (IS_ERR(pmbe))
+			return PTR_ERR(pmbe);
+
+		ret = set_pmb_entry(pmbe);
+		if (ret != 0) {
+			pmb_free(pmbe);
+			return -EBUSY;
+		}
+
+		phys	+= pmb_sizes[i].size;
+		vaddr	+= pmb_sizes[i].size;
+		size	-= pmb_sizes[i].size;
+
+		/*
+		 * Link adjacent entries that span multiple PMB entries
+		 * for easier tear-down.
+		 */
+		if (likely(pmbp))
+			pmbp->link = pmbe;
+
+		pmbp = pmbe;
+	}
+
+	if (size >= 0x1000000)
+		goto again;
+
+	return wanted - size;
+}
+
+void pmb_unmap(unsigned long addr)
+{
+	struct pmb_entry **p, *pmbe;
+
+	for (p = &pmb_list; (pmbe = *p); p = &pmbe->next)
+		if (pmbe->vpn == addr)
+			break;
+
+	if (unlikely(!pmbe))
+		return;
+
+	WARN_ON(!test_bit(pmbe->entry, &pmb_map));
+
+	do {
+		struct pmb_entry *pmblink = pmbe;
+
+		clear_pmb_entry(pmbe);
+		pmbe = pmblink->link;
+
+		pmb_free(pmblink);
+	} while (pmbe);
+}
+
 static void pmb_cache_ctor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
 {
+	struct pmb_entry *pmbe = pmb;
+
 	memset(pmb, 0, sizeof(struct pmb_entry));
 
-	((struct pmb_entry *)pmb)->entry = PMB_NO_ENTRY;
+	spin_lock_irq(&pmb_list_lock);
+
+	pmbe->entry = PMB_NO_ENTRY;
+	pmb_list_add(pmbe);
+
+	spin_unlock_irq(&pmb_list_lock);
+}
+
+static void pmb_cache_dtor(void *pmb, kmem_cache_t *cachep, unsigned long flags)
+{
+	spin_lock_irq(&pmb_list_lock);
+	pmb_list_del(pmb);
+	spin_unlock_irq(&pmb_list_lock);
 }
 
 static int __init pmb_init(void)
@@ -177,7 +312,7 @@ static int __init pmb_init(void)
 	BUG_ON(unlikely(nr_entries >= NR_PMB_ENTRIES));
 
 	pmb_cache = kmem_cache_create("pmb", sizeof(struct pmb_entry),
-				      0, 0, pmb_cache_ctor, NULL);
+				      0, 0, pmb_cache_ctor, pmb_cache_dtor);
 	BUG_ON(!pmb_cache);
 
 	jump_to_P2();
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
index 7189d3a36384..9db3cdfe6776 100644
--- a/include/asm-sh/landisk/iodata_landisk.h
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -74,5 +74,8 @@ extern int landisk_ledparam;    /* from setup.c */
 extern int landisk_buzzerparam; /* from setup.c */
 extern int landisk_arch;        /* from setup.c */
 
+#define __IO_PREFIX landisk
+#include <asm/io_generic.h>
+
 #endif  /* __ASM_SH_IODATA_LANDISK_H */
 
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 91c884634276..ec09589fa6ca 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -50,6 +50,8 @@ typedef unsigned long mm_context_t;
 
 #define PMB_NO_ENTRY		(-1)
 
+struct pmb_entry;
+
 struct pmb_entry {
 	unsigned long vpn;
 	unsigned long ppn;
@@ -60,16 +62,23 @@ struct pmb_entry {
 	 * PMB_NO_ENTRY to search for a free one
 	 */
 	int entry;
+
+	struct pmb_entry *next;
+	/* Adjacent entry link for contiguous multi-entry mappings */
+	struct pmb_entry *link;
 };
 
 /* arch/sh/mm/pmb.c */
 int __set_pmb_entry(unsigned long vpn, unsigned long ppn,
 		    unsigned long flags, int *entry);
-void set_pmb_entry(struct pmb_entry *pmbe);
+int set_pmb_entry(struct pmb_entry *pmbe);
 void clear_pmb_entry(struct pmb_entry *pmbe);
 struct pmb_entry *pmb_alloc(unsigned long vpn, unsigned long ppn,
 			    unsigned long flags);
 void pmb_free(struct pmb_entry *pmbe);
+long pmb_remap(unsigned long virt, unsigned long phys,
+	       unsigned long size, unsigned long flags);
+void pmb_unmap(unsigned long addr);
 
 #endif /* __MMU_H */
 
-- 
cgit v1.2.1


From c470662854ff94d44bf8c192cefac3efa33db676 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:29:18 +0900
Subject: sh: Fixup SHMLBA definition for SH7705.

We need this set to something sensible anywhere were we have
an aliasing dcache..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/cpu-sh2/shmparam.h   | 16 -------------
 include/asm-sh/cpu-sh3/cacheflush.h | 48 ++++++++++++++++---------------------
 include/asm-sh/cpu-sh3/shmparam.h   | 16 -------------
 include/asm-sh/cpu-sh4/shmparam.h   | 19 ---------------
 include/asm-sh/shmparam.h           | 20 +++++++++++++---
 5 files changed, 38 insertions(+), 81 deletions(-)
 delete mode 100644 include/asm-sh/cpu-sh2/shmparam.h
 delete mode 100644 include/asm-sh/cpu-sh3/shmparam.h
 delete mode 100644 include/asm-sh/cpu-sh4/shmparam.h

diff --git a/include/asm-sh/cpu-sh2/shmparam.h b/include/asm-sh/cpu-sh2/shmparam.h
deleted file mode 100644
index 817c1821ee4b..000000000000
--- a/include/asm-sh/cpu-sh2/shmparam.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * include/asm-sh/cpu-sh2/shmparam.h
- *
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH2_SHMPARAM_H
-#define __ASM_CPU_SH2_SHMPARAM_H
-
-#define	SHMLBA PAGE_SIZE		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH2_SHMPARAM_H */
-
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index db0cb071ea8e..97f5a64c2ab8 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -35,47 +35,41 @@
  /* 32KB cache, 4kb PAGE sizes need to check bit 12 */
 #define CACHE_ALIAS 0x00001000
 
-extern void flush_cache_all(void);
-extern void flush_cache_mm(struct mm_struct *mm);
-extern void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
-                              unsigned long end);
-extern void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
-extern void flush_dcache_page(struct page *pg);
-extern void flush_icache_range(unsigned long start, unsigned long end);
-extern void flush_icache_page(struct vm_area_struct *vma, struct page *page);
-
-#define flush_dcache_mmap_lock(mapping)		do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
-
-/* SH3 has unified cache so no special action needed here */
-#define flush_cache_sigtramp(vaddr)		do { } while (0)
-#define flush_page_to_ram(page)			do { } while (0)
-#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
-
-#define p3_cache_init()				do { } while (0)
-
 #define PG_mapped	PG_arch_1
 
-/* We provide our own get_unmapped_area to avoid cache alias issue */
-#define HAVE_ARCH_UNMAPPED_AREA
-
+void flush_cache_all(void);
+void flush_cache_mm(struct mm_struct *mm);
+void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
+                              unsigned long end);
+void flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsigned long pfn);
+void flush_dcache_page(struct page *pg);
+void flush_icache_range(unsigned long start, unsigned long end);
+void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 #else
-
 #define flush_cache_all()			do { } while (0)
 #define flush_cache_mm(mm)			do { } while (0)
 #define flush_cache_range(vma, start, end)	do { } while (0)
 #define flush_cache_page(vma, vmaddr, pfn)	do { } while (0)
 #define flush_dcache_page(page)			do { } while (0)
-#define flush_dcache_mmap_lock(mapping)		do { } while (0)
-#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 #define flush_icache_range(start, end)		do { } while (0)
 #define flush_icache_page(vma,pg)		do { } while (0)
-#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
+#endif
+
+#define flush_dcache_mmap_lock(mapping)		do { } while (0)
+#define flush_dcache_mmap_unlock(mapping)	do { } while (0)
+
+/* SH3 has unified cache so no special action needed here */
 #define flush_cache_sigtramp(vaddr)		do { } while (0)
+#define flush_icache_user_range(vma,pg,adr,len)	do { } while (0)
 
 #define p3_cache_init()				do { } while (0)
 
-#endif
+/*
+ * We provide our own get_unmapped_area to avoid cache aliasing issues
+ * on SH7705 with a 32KB cache, and to page align addresses in the
+ * non-aliasing case.
+ */
+#define HAVE_ARCH_UNMAPPED_AREA
 
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
 
diff --git a/include/asm-sh/cpu-sh3/shmparam.h b/include/asm-sh/cpu-sh3/shmparam.h
deleted file mode 100644
index da5b5eec81ee..000000000000
--- a/include/asm-sh/cpu-sh3/shmparam.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- * include/asm-sh/cpu-sh3/shmparam.h
- *
- * Copyright (C) 1999 Niibe Yutaka
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH3_SHMPARAM_H
-#define __ASM_CPU_SH3_SHMPARAM_H
-
-#define	SHMLBA PAGE_SIZE		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH3_SHMPARAM_H */
-
diff --git a/include/asm-sh/cpu-sh4/shmparam.h b/include/asm-sh/cpu-sh4/shmparam.h
deleted file mode 100644
index a5a0aa9425fe..000000000000
--- a/include/asm-sh/cpu-sh4/shmparam.h
+++ /dev/null
@@ -1,19 +0,0 @@
-/*
- * include/asm-sh/cpu-sh4/shmparam.h
- *
- * Copyright (C) 1999 Niibe Yutaka
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_CPU_SH4_SHMPARAM_H
-#define __ASM_CPU_SH4_SHMPARAM_H
-
-/*
- * SH-4 has D-cache alias issue
- */
-#define	SHMLBA (PAGE_SIZE*4)		 /* attach addr a multiple of this */
-
-#endif /* __ASM_CPU_SH4_SHMPARAM_H */
-
diff --git a/include/asm-sh/shmparam.h b/include/asm-sh/shmparam.h
index 0a95604b9b66..ba1758d90106 100644
--- a/include/asm-sh/shmparam.h
+++ b/include/asm-sh/shmparam.h
@@ -1,8 +1,22 @@
+/*
+ * include/asm-sh/shmparam.h
+ *
+ * Copyright (C) 1999 Niibe Yutaka
+ * Copyright (C) 2006 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
 #ifndef __ASM_SH_SHMPARAM_H
 #define __ASM_SH_SHMPARAM_H
-#ifdef __KERNEL__
 
-#include <asm/cpu/shmparam.h>
+/*
+ * SH-4 and SH-3 7705 have an aliasing dcache. Bump this up to a sensible value
+ * for everyone, and work out the specifics from the probed cache descriptor.
+ */
+#define	SHMLBA	0x4000		 /* attach addr a multiple of this */
+
+#define __ARCH_FORCE_SHMLBA
 
-#endif /* __KERNEL__ */
 #endif /* __ASM_SH_SHMPARAM_H */
-- 
cgit v1.2.1


From f647d33f879d258de4ab2559975bd6eebda2033e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:30:24 +0900
Subject: sh: Fix split ptlock for user mappings in __do_page_fault().

There was a bug that got introduced when the split ptlock changes
went in where mm could be unintialized for user mappings, this
fixes it up..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/fault.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 7a03ffe6dadd..dc461d2bc183 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -210,10 +210,11 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 	 * are always mapped, whether it be due to legacy behaviour in
 	 * 29-bit mode, or due to PMB configuration in 32-bit mode.
 	 */
-	if (address >= P3SEG && address < P3_ADDR_MAX)
+	if (address >= P3SEG && address < P3_ADDR_MAX) {
 		pgd = pgd_offset_k(address);
-	else {
-		if (unlikely(address >= TASK_SIZE || !current->mm))
+		mm = NULL;
+	} else {
+		if (unlikely(address >= TASK_SIZE || !(mm = current->mm)))
 			return 1;
 
 		pgd = pgd_offset(current->mm, address);
-- 
cgit v1.2.1


From 373e68b5472d421cbd2703e7a77caf053f78c005 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:41:24 +0900
Subject: sh: Board updates for I/O routine rework.

This updates the various boards for some of the recent I/O routine
updates.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile                           |   3 -
 arch/sh/boards/renesas/hs7751rvoip/io.c    | 159 +++++++++++++++++------------
 arch/sh/boards/renesas/hs7751rvoip/setup.c |   5 +-
 arch/sh/boards/renesas/systemh/io.c        |  96 ++---------------
 arch/sh/boards/renesas/systemh/irq.c       |   2 +-
 arch/sh/boards/renesas/systemh/setup.c     |  14 +--
 arch/sh/boards/se/7300/io.c                |   8 +-
 arch/sh/boards/se/7300/irq.c               |   2 +-
 arch/sh/boards/se/7300/led.c               |  18 +---
 arch/sh/boards/se/7300/setup.c             |   3 +-
 arch/sh/boards/se/73180/setup.c            |   3 +-
 arch/sh/boards/se/770x/io.c                |  52 ++--------
 arch/sh/boards/se/770x/irq.c               |   2 +-
 arch/sh/boards/se/770x/led.c               |   2 +-
 arch/sh/boards/se/770x/mach.c              |   7 +-
 arch/sh/boards/se/770x/setup.c             |   2 +-
 arch/sh/boards/se/7751/io.c                |  26 ++---
 arch/sh/boards/se/7751/irq.c               |   2 +-
 arch/sh/boards/se/7751/led.c               |   3 +-
 arch/sh/boards/se/7751/mach.c              |   8 +-
 arch/sh/boards/se/7751/setup.c             |   6 +-
 arch/sh/boards/sh03/setup.c                |  25 ++---
 arch/sh/boards/snapgear/io.c               |  63 ++----------
 arch/sh/boards/snapgear/rtc.c              |   3 -
 arch/sh/boards/snapgear/setup.c            |  97 +-----------------
 arch/sh/boards/titan/io.c                  |   4 +-
 arch/sh/cchips/Kconfig                     |   6 +-
 arch/sh/kernel/cf-enabler.c                |   7 +-
 drivers/net/stnic.c                        |   2 +-
 include/asm-sh/hs7751rvoip/hs7751rvoip.h   |   3 +
 include/asm-sh/mc146818rtc.h               |   2 +-
 include/asm-sh/se.h                        |  80 +++++++++++++++
 include/asm-sh/se/io.h                     |  35 -------
 include/asm-sh/se/se.h                     |  77 --------------
 include/asm-sh/se7300.h                    |  64 ++++++++++++
 include/asm-sh/se7300/io.h                 |  29 ------
 include/asm-sh/se7300/se7300.h             |  61 -----------
 include/asm-sh/se73180.h                   |  65 ++++++++++++
 include/asm-sh/se73180/io.h                |  32 ------
 include/asm-sh/se73180/se73180.h           |  62 -----------
 include/asm-sh/se7751.h                    |  71 +++++++++++++
 include/asm-sh/se7751/io.h                 |  42 --------
 include/asm-sh/se7751/se7751.h             |  68 ------------
 include/asm-sh/sh03/io.h                   |  10 +-
 include/asm-sh/snapgear.h                  |  79 ++++++++++++++
 include/asm-sh/snapgear/io.h               |  92 -----------------
 include/asm-sh/systemh/7751systemh.h       |  68 ------------
 include/asm-sh/systemh/io.h                |  43 --------
 include/asm-sh/systemh7751.h               |  71 +++++++++++++
 49 files changed, 607 insertions(+), 1077 deletions(-)
 create mode 100644 include/asm-sh/se.h
 delete mode 100644 include/asm-sh/se/io.h
 delete mode 100644 include/asm-sh/se/se.h
 create mode 100644 include/asm-sh/se7300.h
 delete mode 100644 include/asm-sh/se7300/io.h
 delete mode 100644 include/asm-sh/se7300/se7300.h
 create mode 100644 include/asm-sh/se73180.h
 delete mode 100644 include/asm-sh/se73180/io.h
 delete mode 100644 include/asm-sh/se73180/se73180.h
 create mode 100644 include/asm-sh/se7751.h
 delete mode 100644 include/asm-sh/se7751/io.h
 delete mode 100644 include/asm-sh/se7751/se7751.h
 create mode 100644 include/asm-sh/snapgear.h
 delete mode 100644 include/asm-sh/snapgear/io.h
 delete mode 100644 include/asm-sh/systemh/7751systemh.h
 delete mode 100644 include/asm-sh/systemh/io.h
 create mode 100644 include/asm-sh/systemh7751.h

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 55dcba2c06fc..65676c33822e 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -111,10 +111,7 @@ machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
 
-incdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se
 incdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se7751
-incdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)        := se7300
-incdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se73180
 incdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 
 ifneq ($(machdir-y),)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index edecf107fc13..ecdce7ef6a34 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -21,10 +21,8 @@
 #include <linux/pci.h>
 #include "../../../drivers/pci/pci-sh7751.h"
 
-extern void *area5_io8_base;	/* Area 5 8bit I/O Base address */
 extern void *area6_io8_base;	/* Area 6 8bit I/O Base address */
 extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
-extern void *area6_io16_base;	/* Area 6 16bit I/O Base address */
 
 /*
  * The 7751R HS7751RVoIP uses the built-in PCI controller (PCIC)
@@ -37,16 +35,10 @@ extern void *area6_io16_base;	/* Area 6 16bit I/O Base address */
 #define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
 #define PCI_IO_AREA	SH7751_PCI_IO_BASE
 #define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 #define CODEC_IO_BASE	0x1000
-#endif
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
+#define CODEC_IOMAP(a)	((unsigned long)area6_io8_base + ((a) - CODEC_IO_BASE))
 
 static inline void delay(void)
 {
@@ -61,7 +53,7 @@ static inline unsigned long port2adr(unsigned int port)
 		else
 			return ((unsigned long)area5_io16_base + 0x800 + ((port-0x1f0) << 1));
 	else
-		maybebadio(port2adr, (unsigned long)port);
+		maybebadio((unsigned long)port);
 	return port;
 }
 
@@ -109,15 +101,15 @@ codec_port(unsigned long port)
 unsigned char hs7751rvoip_inb(unsigned long port)
 {
 	if (PXSEG(port))
-		return *(volatile unsigned char *)port;
+		return ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		return *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		return ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+		return ctrl_inb(PCI_IOMAP(port));
 	else
-		return (*(volatile unsigned short *)port2adr(port) & 0xff);
+		return ctrl_inw(port2adr(port)) & 0xff;
 }
 
 unsigned char hs7751rvoip_inb_p(unsigned long port)
@@ -125,15 +117,15 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 	unsigned char v;
 
         if (PXSEG(port))
-                v = *(volatile unsigned char *)port;
+		v = ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		v = *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		v = ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+		v = ctrl_inb(PCI_IOMAP(port));
 	else
-		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
+		v = ctrl_inw(port2adr(port)) & 0xff;
 	delay();
 	return v;
 }
@@ -141,22 +133,22 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 unsigned short hs7751rvoip_inw(unsigned long port)
 {
         if (PXSEG(port))
-                return *(volatile unsigned short *)port;
+		return ctrl_inw(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+		return ctrl_inw(PCI_IOMAP(port));
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
 unsigned int hs7751rvoip_inl(unsigned long port)
 {
         if (PXSEG(port))
-                return *(volatile unsigned long *)port;
+		return ctrl_inl(port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-                return *(volatile unsigned long *)PCI_IOMAP(port);
+		return ctrl_inl(PCI_IOMAP(port));
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -164,137 +156,168 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
 {
 
         if (PXSEG(port))
-                *(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		ctrl_outb(value, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outb(value, port2adr(port));
 }
 
 void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned char *)port = value;
+		ctrl_outb(value, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		*(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = value;
+		ctrl_outb(value, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned char *)PCI_IOMAP(port) = value;
+		ctrl_outb(value, PCI_IOMAP(port));
 	else
-		*(volatile unsigned short *)port2adr(port) = value;
+		ctrl_outw(value, port2adr(port));
+
 	delay();
 }
 
 void hs7751rvoip_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned short *)port = value;
+		ctrl_outw(value, port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*(unsigned short *)PCI_IOMAP(port) = value;
+		ctrl_outw(value, PCI_IOMAP(port));
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
-                *(volatile unsigned long *)port = value;
+		ctrl_outl(value, port);
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-        	*((unsigned long *)PCI_IOMAP(port)) = value;
+		ctrl_outl(value, PCI_IOMAP(port));
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_insb(unsigned long port, void *addr, unsigned long count)
 {
+	u8 *buf = addr;
+
 	if (PXSEG(port))
-		while (count--) *((unsigned char *) addr)++ = *(volatile unsigned char *)port;
+		while (count--)
+			*buf++ = ctrl_inb(port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		while (count--) *((unsigned char *) addr)++ = *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE));
+		while (count--)
+			*buf++ = ctrl_inb(CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u8 *bp = (__u8 *)PCI_IOMAP(port);
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--) *((volatile unsigned char *) addr)++ = *bp;
+		while (count--)
+			*buf++ = *bp;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+		volatile u16 *p = (volatile u16 *)port2adr(port);
 
-		while (count--) *((unsigned char *) addr)++ = *p & 0xff;
+		while (count--)
+			*buf++ = *p & 0xff;
 	}
 }
 
 void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 {
-	volatile __u16 *p;
+	volatile u16 *p;
+	u16 *buf = addr;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *((__u16 *) addr)++ = *p;
+		p = (volatile u16 *)port2adr(port);
+	while (count--)
+		*buf++ = *p;
 }
 
 void hs7751rvoip_insl(unsigned long port, void *addr, unsigned long count)
 {
+
 	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+		u32 *buf = addr;
 
-		while (count--) *((__u32 *) addr)++ = *p;
+		while (count--)
+			*buf++ = *p;
 	} else
-		maybebadio(insl, port);
+		maybebadio(port);
 }
 
 void hs7751rvoip_outsb(unsigned long port, const void *addr, unsigned long count)
 {
+	const u8 *buf = addr;
+
 	if (PXSEG(port))
-		while (count--) *(volatile unsigned char *)port = *((unsigned char *) addr)++;
+		while (count--)
+			ctrl_outb(*buf++, port);
 #if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
-		while (count--) *(volatile unsigned char *)((unsigned long)area6_io8_base+(port-CODEC_IO_BASE)) = *((unsigned char *) addr)++;
+		while (count--)
+			ctrl_outb(*buf++, CODEC_IOMAP(port));
 #endif
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u8 *bp = (__u8 *)PCI_IOMAP(port);
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
 
-		while (count--) *bp = *((volatile unsigned char *) addr)++;
+		while (count--)
+			*bp = *buf++;
 	} else {
-		volatile __u16 *p = (volatile unsigned short *)port2adr(port);
+		volatile u16 *p = (volatile u16 *)port2adr(port);
 
-		while (count--) *p = *((unsigned char *) addr)++;
+		while (count--)
+			*p = *buf++;
 	}
 }
 
 void hs7751rvoip_outsw(unsigned long port, const void *addr, unsigned long count)
 {
-	volatile __u16 *p;
+	volatile u16 *p;
+	const u16 *buf = addr;
 
 	if (PXSEG(port))
-		p = (volatile unsigned short *)port;
+		p = (volatile u16 *)port;
 	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+		p = (volatile u16 *)PCI_IOMAP(port);
 	else
-		p = (volatile unsigned short *)port2adr(port);
-	while (count--) *p = *((__u16 *) addr)++;
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*p = *buf++;
 }
 
 void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count)
 {
+	const u32 *buf = addr;
+
 	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile __u32 *p = (__u32 *)PCI_IOMAP(port);
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
 
-		while (count--) *p = *((__u32 *) addr)++;
+		while (count--)
+			*p = *buf++;
 	} else
-		maybebadio(outsl, port);
+		maybebadio(port);
 }
 
-unsigned long hs7751rvoip_isa_port2addr(unsigned long offset)
+void __iomem *hs7751rvoip_ioport_map(unsigned long port, unsigned int size)
 {
-	return port2adr(offset);
+        if (PXSEG(port))
+                return (void __iomem *)port;
+	else if (unlikely(codec_port(port) && (size == 1)))
+		return (void __iomem *)CODEC_IOMAP(port);
+        else if (CHECK_SH7751_PCIIO(port))
+                return (void __iomem *)PCI_IOMAP(port);
+
+        return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index aa51bda931f6..a2cbcc5d530f 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -17,9 +17,8 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/pm.h>
-#include <asm/hs7751rvoip/hs7751rvoip.h>
-#include <asm/hs7751rvoip/io.h>
 #include <asm/io.h>
+#include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/machvec.h>
 #include <asm/rtc.h>
 #include <asm/irq.h>
@@ -60,8 +59,8 @@ struct sh_machine_vector mv_hs7751rvoip __initmv = {
 	.mv_outsw		= hs7751rvoip_outsw,
 	.mv_outsl		= hs7751rvoip_outsl,
 
-	.mv_isa_port2addr	= hs7751rvoip_isa_port2addr,
 	.mv_init_irq		= hs7751rvoip_init_irq,
+	.mv_ioport_map		= hs7751rvoip_ioport_map,
 };
 ALIAS_MV(hs7751rvoip)
 
diff --git a/arch/sh/boards/renesas/systemh/io.c b/arch/sh/boards/renesas/systemh/io.c
index 85511576d414..0befd4f9894c 100644
--- a/arch/sh/boards/renesas/systemh/io.c
+++ b/arch/sh/boards/renesas/systemh/io.c
@@ -10,11 +10,10 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
-#include <asm/systemh/7751systemh.h>
+#include <linux/pci.h>
+#include <asm/systemh7751.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
-
-#include <linux/pci.h>
 #include "../../../drivers/pci/pci-sh7751.h"
 
 /*
@@ -31,11 +30,6 @@
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 #define ETHER_IOMAP(adr) (0xB3000000 + (adr)) /*map to 16bits access area
                                                 of smc lan chip*/
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -46,11 +40,7 @@ port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#if 0
-	else
-		return (volatile __u16 *) (PA_SUPERIO + (port << 1));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
@@ -111,7 +101,7 @@ unsigned short sh7751systemh_inw(unsigned long port)
 	else if (port <= 0x3F1)
 		return *(volatile unsigned int *)ETHER_IOMAP(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -126,7 +116,7 @@ unsigned int sh7751systemh_inl(unsigned long port)
 	else if (port <= 0x3F1)
 		return *(volatile unsigned int *)ETHER_IOMAP(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -167,7 +157,7 @@ void sh7751systemh_outw(unsigned short value, unsigned long port)
 	else if (port <= 0x3F1)
 		*(volatile unsigned short *)ETHER_IOMAP(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void sh7751systemh_outl(unsigned int value, unsigned long port)
@@ -177,7 +167,7 @@ void sh7751systemh_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
         	*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void sh7751systemh_insb(unsigned long port, void *addr, unsigned long count)
@@ -194,7 +184,7 @@ void sh7751systemh_insw(unsigned long port, void *addr, unsigned long count)
 
 void sh7751systemh_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void sh7751systemh_outsb(unsigned long port, const void *addr, unsigned long count)
@@ -211,73 +201,5 @@ void sh7751systemh_outsw(unsigned long port, const void *addr, unsigned long cou
 
 void sh7751systemh_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* For read/write calls, just copy generic (pass-thru); PCIMBR is  */
-/* already set up.  For a larger memory space, these would need to */
-/* reset PCIMBR as needed on a per-call basis...                   */
-
-unsigned char sh7751systemh_readb(unsigned long addr)
-{
-	return *(volatile unsigned char*)addr;
-}
-
-unsigned short sh7751systemh_readw(unsigned long addr)
-{
-	return *(volatile unsigned short*)addr;
-}
-
-unsigned int sh7751systemh_readl(unsigned long addr)
-{
-	return *(volatile unsigned long*)addr;
-}
-
-void sh7751systemh_writeb(unsigned char b, unsigned long addr)
-{
-	*(volatile unsigned char*)addr = b;
-}
-
-void sh7751systemh_writew(unsigned short b, unsigned long addr)
-{
-	*(volatile unsigned short*)addr = b;
-}
-
-void sh7751systemh_writel(unsigned int b, unsigned long addr)
-{
-        *(volatile unsigned long*)addr = b;
-}
-
-
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-#if 0
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-	return 0;
-}
-#endif
-
-unsigned long
-sh7751systemh_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/renesas/systemh/irq.c b/arch/sh/boards/renesas/systemh/irq.c
index 8372d967f601..53731a0abb8f 100644
--- a/arch/sh/boards/renesas/systemh/irq.c
+++ b/arch/sh/boards/renesas/systemh/irq.c
@@ -15,7 +15,7 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/mach/7751systemh.h>
+#include <asm/systemh7751.h>
 #include <asm/smc37c93x.h>
 
 /* address of external interrupt mask register
diff --git a/arch/sh/boards/renesas/systemh/setup.c b/arch/sh/boards/renesas/systemh/setup.c
index 826fa3d7669c..433c7c3f35e2 100644
--- a/arch/sh/boards/renesas/systemh/setup.c
+++ b/arch/sh/boards/renesas/systemh/setup.c
@@ -15,9 +15,8 @@
  * for more details.
  */
 #include <linux/init.h>
-#include <asm/mach/7751systemh.h>
-#include <asm/mach/io.h>
 #include <asm/machvec.h>
+#include <asm/systemh7751.h>
 
 extern void make_systemh_irq(unsigned int irq);
 
@@ -31,8 +30,6 @@ const char *get_system_type(void)
  */
 void __init init_7751systemh_IRQ(void)
 {
-/*  	make_ipr_irq(10, BCR_ILCRD, 1, 0x0f-10); LAN */
-/*  	make_ipr_irq(14, BCR_ILCRA, 2, 0x0f-4); */
 	make_systemh_irq(0xb);	/* Ethernet interrupt */
 }
 
@@ -60,15 +57,6 @@ struct sh_machine_vector mv_7751systemh __initmv = {
 	.mv_outsw		= sh7751systemh_outsw,
 	.mv_outsl		= sh7751systemh_outsl,
 
-	.mv_readb		= sh7751systemh_readb,
-	.mv_readw		= sh7751systemh_readw,
-	.mv_readl		= sh7751systemh_readl,
-	.mv_writeb		= sh7751systemh_writeb,
-	.mv_writew		= sh7751systemh_writew,
-	.mv_writel		= sh7751systemh_writel,
-
-	.mv_isa_port2addr	= sh7751systemh_isa_port2addr,
-
 	.mv_init_irq		= init_7751systemh_IRQ,
 };
 ALIAS_MV(7751systemh)
diff --git a/arch/sh/boards/se/7300/io.c b/arch/sh/boards/se/7300/io.c
index f449a94ddffd..8a03d7a52a7c 100644
--- a/arch/sh/boards/se/7300/io.c
+++ b/arch/sh/boards/se/7300/io.c
@@ -9,8 +9,8 @@
  */
 
 #include <linux/kernel.h>
-#include <asm/mach/se7300.h>
 #include <asm/io.h>
+#include <asm/se7300.h>
 
 #define badio(fn, a) panic("bad i/o operation %s for %08lx.", #fn, a)
 
@@ -99,6 +99,7 @@ bad_outb(struct iop *p, unsigned char value, unsigned long port)
 	badio(inw, port);
 }
 
+#ifdef CONFIG_SMC91X
 /* MSTLANEX01 LAN at 0xb400:0000 */
 static struct iop laniop = {
 	.start = 0x300,
@@ -110,6 +111,7 @@ static struct iop laniop = {
 	.outb = simple_outb,
 	.outw = simple_outw,
 };
+#endif
 
 /* NE2000 pc card NIC */
 static struct iop neiop = {
@@ -123,6 +125,7 @@ static struct iop neiop = {
 	.outw = simple_outw,
 };
 
+#ifdef CONFIG_IDE
 /* CF in CF slot */
 static struct iop cfiop = {
 	.base = 0xb0600000,
@@ -132,12 +135,13 @@ static struct iop cfiop = {
 	.outb = pcc_outb,
 	.outw = simple_outw,
 };
+#endif
 
 static __inline__ struct iop *
 port2iop(unsigned long port)
 {
 	if (0) ;
-#if defined(CONFIG_SMC91111)
+#if defined(CONFIG_SMC91X)
 	else if (laniop.check(&laniop, port))
 		return &laniop;
 #endif
diff --git a/arch/sh/boards/se/7300/irq.c b/arch/sh/boards/se/7300/irq.c
index 216a78d1a108..ad1034f98a29 100644
--- a/arch/sh/boards/se/7300/irq.c
+++ b/arch/sh/boards/se/7300/irq.c
@@ -11,7 +11,7 @@
 #include <linux/irq.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/mach/se7300.h>
+#include <asm/se7300.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/7300/led.c b/arch/sh/boards/se/7300/led.c
index ad51f0a9c1e3..4d03bb7774be 100644
--- a/arch/sh/boards/se/7300/led.c
+++ b/arch/sh/boards/se/7300/led.c
@@ -12,24 +12,10 @@
  */
 
 #include <linux/sched.h>
-#include <asm/mach/se7300.h>
-
-static void
-mach_led(int position, int value)
-{
-	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
-
-	if (value) {
-		*p |= (1 << 8);
-	} else {
-		*p &= ~(1 << 8);
-	}
-}
-
+#include <asm/se7300.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
-void
-heartbeat_7300se(void)
+void heartbeat_7300se(void)
 {
 	static unsigned int cnt = 0, period = 0;
 	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
diff --git a/arch/sh/boards/se/7300/setup.c b/arch/sh/boards/se/7300/setup.c
index ebcd98d4c081..bb7e1a189be8 100644
--- a/arch/sh/boards/se/7300/setup.c
+++ b/arch/sh/boards/se/7300/setup.c
@@ -9,8 +9,7 @@
 
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/mach/io.h>
+#include <asm/se7300.h>
 
 void heartbeat_7300se(void);
 void init_7300se_IRQ(void);
diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index cdb7b5f8d942..cddc7a2b858f 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -11,8 +11,7 @@
 
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/mach/io.h>
+#include <asm/se73180.h>
 
 void heartbeat_73180se(void);
 void init_73180se_IRQ(void);
diff --git a/arch/sh/boards/se/770x/io.c b/arch/sh/boards/se/770x/io.c
index 9a39ee963143..5102201c97ab 100644
--- a/arch/sh/boards/se/770x/io.c
+++ b/arch/sh/boards/se/770x/io.c
@@ -1,4 +1,4 @@
-/* $Id: io.c,v 1.5 2004/02/22 23:08:43 kkojima Exp $
+/* $Id: io.c,v 1.6 2006/01/04 17:53:54 lethal Exp $
  *
  * linux/arch/sh/kernel/io_se.c
  *
@@ -11,7 +11,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /* SH pcmcia io window base, start and end.  */
 int sh_pcic_io_wbase = 0xb8400000;
@@ -52,10 +52,6 @@ shifted_port(unsigned long port)
 		return 1;
 }
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 unsigned char se_inb(unsigned long port)
 {
 	if (sh_pcic_io_start <= port && port <= sh_pcic_io_stop)
@@ -86,13 +82,13 @@ unsigned short se_inw(unsigned long port)
 	    (sh_pcic_io_start <= port && port <= sh_pcic_io_stop))
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
 unsigned int se_inl(unsigned long port)
 {
-	maybebadio(inl, port);
+	maybebadio(port);
 	return 0;
 }
 
@@ -123,12 +119,12 @@ void se_outw(unsigned short value, unsigned long port)
 	    (sh_pcic_io_start <= port && port <= sh_pcic_io_stop))
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void se_outl(unsigned int value, unsigned long port)
 {
-	maybebadio(outl, port);
+	maybebadio(port);
 }
 
 void se_insb(unsigned long port, void *addr, unsigned long count)
@@ -159,7 +155,7 @@ void se_insw(unsigned long port, void *addr, unsigned long count)
 
 void se_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void se_outsb(unsigned long port, const void *addr, unsigned long count)
@@ -190,37 +186,5 @@ void se_outsw(unsigned long port, const void *addr, unsigned long count)
 
 void se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-#if 0
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-#endif
-	return 0;
-}
-
-unsigned long
-se_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/se/770x/irq.c b/arch/sh/boards/se/770x/irq.c
index 3e558716ce10..cff6700bbafd 100644
--- a/arch/sh/boards/se/770x/irq.c
+++ b/arch/sh/boards/se/770x/irq.c
@@ -11,7 +11,7 @@
 #include <linux/irq.h>
 #include <asm/irq.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/770x/led.c b/arch/sh/boards/se/770x/led.c
index 73e9848387ea..daf7b1ee786a 100644
--- a/arch/sh/boards/se/770x/led.c
+++ b/arch/sh/boards/se/770x/led.c
@@ -10,7 +10,7 @@
  */
 
 #include <linux/sched.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
 void heartbeat_se(void)
diff --git a/arch/sh/boards/se/770x/mach.c b/arch/sh/boards/se/770x/mach.c
index 6ec07bd3dcf1..e8968b71c353 100644
--- a/arch/sh/boards/se/770x/mach.c
+++ b/arch/sh/boards/se/770x/mach.c
@@ -13,12 +13,9 @@
 
 #include <asm/machvec.h>
 #include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/se/io.h>
+#include <asm/se.h>
 
 void heartbeat_se(void);
-void setup_se(void);
 void init_se_IRQ(void);
 
 /*
@@ -57,8 +54,6 @@ struct sh_machine_vector mv_se __initmv = {
 	.mv_outsw		= se_outsw,
 	.mv_outsl		= se_outsl,
 
-	.mv_isa_port2addr	= se_isa_port2addr,
-
 	.mv_init_irq		= init_se_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_se,
diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index b6730ea57502..c9f75272e751 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -14,7 +14,7 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 #include <asm/smc37c93x.h>
 
 /*
diff --git a/arch/sh/boards/se/7751/io.c b/arch/sh/boards/se/7751/io.c
index 99041b269261..0e8a3ba48316 100644
--- a/arch/sh/boards/se/7751/io.c
+++ b/arch/sh/boards/se/7751/io.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * linux/arch/sh/kernel/io_7751se.c
  *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
@@ -14,7 +14,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/io.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 #include <asm/addrspace.h>
 
 #include <linux/pci.h>
@@ -52,10 +52,6 @@ int sh_pcic_io_dummy;
 
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
@@ -66,11 +62,7 @@ port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#if 0
-	else
-		return (volatile __u16 *) (PA_SUPERIO + (port << 1));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
@@ -140,7 +132,7 @@ unsigned short sh7751se_inw(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -153,7 +145,7 @@ unsigned int sh7751se_inl(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -188,7 +180,7 @@ void sh7751se_outw(unsigned short value, unsigned long port)
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 void sh7751se_outl(unsigned int value, unsigned long port)
@@ -198,17 +190,17 @@ void sh7751se_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
         	*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void sh7751se_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
+	maybebadio(port);
 }
 
 /* Map ISA bus address to the real address. Only for PCMCIA.  */
diff --git a/arch/sh/boards/se/7751/irq.c b/arch/sh/boards/se/7751/irq.c
index bf6c023615df..c607b0a48479 100644
--- a/arch/sh/boards/se/7751/irq.c
+++ b/arch/sh/boards/se/7751/irq.c
@@ -12,7 +12,7 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <asm/irq.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 
 /*
  * Initialize IRQ setting
diff --git a/arch/sh/boards/se/7751/led.c b/arch/sh/boards/se/7751/led.c
index 4405e26cf866..ff0355dea81b 100644
--- a/arch/sh/boards/se/7751/led.c
+++ b/arch/sh/boards/se/7751/led.c
@@ -8,9 +8,8 @@
  *
  * This file contains Solution Engine specific LED code.
  */
-
-#include <asm/se7751/se7751.h>
 #include <linux/sched.h>
+#include <asm/se7751.h>
 
 /* Cycle the LED's in the clasic Knightrider/Sun pattern */
 void heartbeat_7751se(void)
diff --git a/arch/sh/boards/se/7751/mach.c b/arch/sh/boards/se/7751/mach.c
index 62d8d3e62590..1bb9047d863b 100644
--- a/arch/sh/boards/se/7751/mach.c
+++ b/arch/sh/boards/se/7751/mach.c
@@ -10,12 +10,8 @@
  */
 
 #include <linux/init.h>
-
 #include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/se7751/io.h>
+#include <asm/se7751.h>
 
 void heartbeat_7751se(void);
 void init_7751se_IRQ(void);
@@ -44,8 +40,6 @@ struct sh_machine_vector mv_7751se __initmv = {
 	.mv_insl		= sh7751se_insl,
 	.mv_outsl		= sh7751se_outsl,
 
-	.mv_isa_port2addr	= sh7751se_isa_port2addr,
-
 	.mv_init_irq		= init_7751se_IRQ,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat		= heartbeat_7751se,
diff --git a/arch/sh/boards/se/7751/setup.c b/arch/sh/boards/se/7751/setup.c
index 48dc5aee67d4..8b693105893c 100644
--- a/arch/sh/boards/se/7751/setup.c
+++ b/arch/sh/boards/se/7751/setup.c
@@ -1,4 +1,4 @@
-/* 
+/*
  * linux/arch/sh/kernel/setup_7751se.c
  *
  * Copyright (C) 2000  Kazumoto Kojima
@@ -11,11 +11,9 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-
-#include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
-#include <asm/se7751/se7751.h>
+#include <asm/se7751.h>
 
 #ifdef CONFIG_SH_KGDB
 #include <asm/kgdb.h>
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index 60290f8f289c..dab742a00c6f 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -7,22 +7,19 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-#include <linux/hdreg.h>
-#include <linux/ide.h>
 #include <asm/io.h>
+#include <asm/rtc.h>
 #include <asm/sh03/io.h>
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
 #include "../../drivers/pci/pci-sh7751.h"
 
-extern void (*board_time_init)(void);
-
 const char *get_system_type(void)
 {
 	return "Interface CTP/PCI-SH03)";
 }
 
-void init_sh03_IRQ(void)
+static void init_sh03_IRQ(void)
 {
 	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
 
@@ -34,24 +31,20 @@ void init_sh03_IRQ(void)
 
 extern void *cf_io_base;
 
-unsigned long sh03_isa_port2addr(unsigned long port)
+static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
 {
 	if (PXSEG(port))
-		return port;
+		return (void __iomem *)port;
 	/* CompactFlash (IDE) */
-	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6)) {
-		return (unsigned long)cf_io_base + port;
-	}
-        return port + SH7751_PCI_IO_BASE;
-}
+	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6))
+		return (void __iomem *)((unsigned long)cf_io_base + port);
 
-/*
- * The Machine Vector
- */
+        return (void __iomem *)(port + SH7751_PCI_IO_BASE);
+}
 
 struct sh_machine_vector mv_sh03 __initmv = {
 	.mv_nr_irqs		= 48,
-	.mv_isa_port2addr	= sh03_isa_port2addr,
+	.mv_ioport_map		= sh03_ioport_map,
 	.mv_init_irq		= init_sh03_IRQ,
 
 #ifdef CONFIG_HEARTBEAT
diff --git a/arch/sh/boards/snapgear/io.c b/arch/sh/boards/snapgear/io.c
index e2eb78fc381d..9f700b8392bb 100644
--- a/arch/sh/boards/snapgear/io.c
+++ b/arch/sh/boards/snapgear/io.c
@@ -28,39 +28,26 @@ unsigned short secureedge5410_ioport;
 /*
  * The SnapGear uses the built-in PCI controller (PCIC)
  * of the 7751 processor
- */ 
+ */
 
 #define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
 #define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
 #define PCI_IO_AREA	SH7751_PCI_IO_BASE
 #define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
 
-
 #define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 
-
-#define maybebadio(name,port) \
-  printk("bad PC-like io %s for port 0x%lx at 0x%08x\n", \
-	 #name, (port), (__u32) __builtin_return_address(0))
-
-
 static inline void delay(void)
 {
 	ctrl_inw(0xa0000000);
 }
 
-
 static inline volatile __u16 *port2adr(unsigned int port)
 {
-#if 0
-	if (port >= 0x2000)
-		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
-#endif
-	maybebadio(name,(unsigned long)port);
+	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
-
 /* In case someone configures the kernel w/o PCI support: in that */
 /* scenario, don't ever bother to check for PCI-window addresses */
 
@@ -115,7 +102,7 @@ unsigned short snapgear_inw(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inw, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -129,7 +116,7 @@ unsigned int snapgear_inl(unsigned long port)
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
-		maybebadio(inl, port);
+		maybebadio(port);
 	return 0;
 }
 
@@ -167,7 +154,7 @@ void snapgear_outw(unsigned short value, unsigned long port)
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
-		maybebadio(outw, port);
+		maybebadio(port);
 }
 
 
@@ -178,49 +165,15 @@ void snapgear_outl(unsigned int value, unsigned long port)
 	else if (CHECK_SH7751_PCIIO(port))
 		*((unsigned long*)PCI_IOMAP(port)) = value;
 	else
-		maybebadio(outl, port);
+		maybebadio(port);
 }
 
 void snapgear_insl(unsigned long port, void *addr, unsigned long count)
 {
-	maybebadio(insl, port);
+	maybebadio(port);
 }
 
 void snapgear_outsl(unsigned long port, const void *addr, unsigned long count)
 {
-	maybebadio(outsw, port);
-}
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-
-#if 0
-static int sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-#if 0
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-#endif
-	return 0;
-}
-#endif
-
-unsigned long snapgear_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
+	maybebadio(port);
 }
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index b71e009da35c..287bc4ed3ac7 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -17,10 +17,7 @@
 #include <linux/time.h>
 #include <linux/rtc.h>
 #include <linux/mc146818rtc.h>
-
 #include <asm/io.h>
-#include <asm/rtc.h>
-#include <asm/mc146818rtc.h>
 
 /****************************************************************************/
 
diff --git a/arch/sh/boards/snapgear/setup.c b/arch/sh/boards/snapgear/setup.c
index f1f7c70c9402..66ce32f8b13c 100644
--- a/arch/sh/boards/snapgear/setup.c
+++ b/arch/sh/boards/snapgear/setup.c
@@ -1,5 +1,4 @@
-/****************************************************************************/
-/* 
+/*
  * linux/arch/sh/boards/snapgear/setup.c
  *
  * Copyright (C) 2002  David McCullough <davidm@snapgear.com>
@@ -12,8 +11,6 @@
  *           Modified for 7751 Solution Engine by
  *           Ian da Silva and Jeremy Siegel, 2001.
  */
-/****************************************************************************/
-
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/interrupt.h>
@@ -21,14 +18,13 @@
 #include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/sched.h>
-
 #include <asm/machvec.h>
-#include <asm/mach/io.h>
+#include <asm/snapgear.h>
 #include <asm/irq.h>
 #include <asm/io.h>
+#include <asm/rtc.h>
 #include <asm/cpu/timer.h>
 
-extern void (*board_time_init)(void);
 extern void secureedge5410_rtc_init(void);
 extern void pcibios_init(void);
 
@@ -85,91 +81,6 @@ static void __init init_snapgear_IRQ(void)
 	make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY);
 }
 
-/****************************************************************************/
-/*
- *	Fast poll interrupt simulator.
- */
-
-/*
- * Leave all of the fast timer/fast poll stuff commented out for now, since
- * it's not clear whether it actually works or not. Since it wasn't being used
- * at all in 2.4, we'll assume it's not sane for 2.6 either.. -- PFM
- */
-#if 0
-#define FAST_POLL	1000
-//#define FAST_POLL_INTR
-
-#define FASTTIMER_IRQ   17
-#define FASTTIMER_IPR_ADDR  INTC_IPRA
-#define FASTTIMER_IPR_POS    2
-#define FASTTIMER_PRIORITY   3
-
-#ifdef FAST_POLL_INTR
-#define TMU1_TCR_INIT	0x0020
-#else
-#define TMU1_TCR_INIT	0
-#endif
-#define TMU_TSTR_INIT	1
-#define TMU1_TCR_CALIB	0x0000
-
-
-#ifdef FAST_POLL_INTR
-static void fast_timer_irq(int irq, void *dev_instance, struct pt_regs *regs)
-{
-	unsigned long timer_status;
-    timer_status = ctrl_inw(TMU1_TCR);
-	timer_status &= ~0x100;
-	ctrl_outw(timer_status, TMU1_TCR);
-}
-#endif
-
-/*
- * return the current ticks on the fast timer
- */
-
-unsigned long fast_timer_count(void)
-{
-	return(ctrl_inl(TMU1_TCNT));
-}
-
-/*
- * setup a fast timer for profiling etc etc
- */
-
-static void setup_fast_timer()
-{
-	unsigned long interval;
-
-#ifdef FAST_POLL_INTR
-	interval = (current_cpu_data.module_clock/4 + FAST_POLL/2) / FAST_POLL;
-
-	make_ipr_irq(FASTTIMER_IRQ, FASTTIMER_IPR_ADDR, FASTTIMER_IPR_POS,
-			FASTTIMER_PRIORITY);
-
-	printk("SnapGear: %dHz fast timer on IRQ %d\n",FAST_POLL,FASTTIMER_IRQ);
-
-	if (request_irq(FASTTIMER_IRQ, fast_timer_irq, 0, "SnapGear fast timer",
-			NULL) != 0)
-		printk("%s(%d): request_irq() failed?\n", __FILE__, __LINE__);
-#else
-	printk("SnapGear: fast timer running\n",FAST_POLL,FASTTIMER_IRQ);
-	interval = 0xffffffff;
-#endif
-
-	ctrl_outb(ctrl_inb(TMU_TSTR) & ~0x2, TMU_TSTR); /* disable timer 1 */
-	ctrl_outw(TMU1_TCR_INIT, TMU1_TCR);
-	ctrl_outl(interval, TMU1_TCOR);
-	ctrl_outl(interval, TMU1_TCNT);
-	ctrl_outb(ctrl_inb(TMU_TSTR) | 0x2, TMU_TSTR); /* enable timer 1 */
-
-	printk("Timer count 1 = 0x%x\n", fast_timer_count());
-	udelay(1000);
-	printk("Timer count 2 = 0x%x\n", fast_timer_count());
-}
-#endif
-
-/****************************************************************************/
-
 const char *get_system_type(void)
 {
 	return "SnapGear SecureEdge5410";
@@ -196,8 +107,6 @@ struct sh_machine_vector mv_snapgear __initmv = {
 	.mv_outw_p		= snapgear_outw,
 	.mv_outl_p		= snapgear_outl,
 
-	.mv_isa_port2addr	= snapgear_isa_port2addr,
-
 	.mv_init_irq		= init_snapgear_IRQ,
 };
 ALIAS_MV(snapgear)
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index b886fd233a66..48f3494f55b1 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -30,10 +30,10 @@ static inline void delay(void)
         ctrl_inw(0xa0000000);
 }
 
-static inline volatile u16 *port2adr(unsigned int port)
+static inline unsigned int port2adr(unsigned int port)
 {
         maybebadio((unsigned long)port);
-        return (volatile u16*)port;
+        return port;
 }
 
 u8 titan_inb(unsigned long port)
diff --git a/arch/sh/cchips/Kconfig b/arch/sh/cchips/Kconfig
index 155d139884c3..0582ca8346b6 100644
--- a/arch/sh/cchips/Kconfig
+++ b/arch/sh/cchips/Kconfig
@@ -65,6 +65,11 @@ config HD64461_IRQ
 
 	  Do not change this unless you know what you are doing.
 
+config HD64461_IOBASE
+	hex "HD64461 start address"
+	depends on HD64461
+	default "0xb0000000"
+
 config HD64461_ENABLER
 	bool "HD64461 PCMCIA enabler"
 	depends on HD64461
@@ -73,7 +78,6 @@ config HD64461_ENABLER
 	  via the HD64461 companion chip.
 	  Otherwise, say N.
 
-
 config HD64465_IOBASE
 	hex "HD64465 start address"
 	depends on HD64465
diff --git a/arch/sh/kernel/cf-enabler.c b/arch/sh/kernel/cf-enabler.c
index c9b823d1d073..3e5fa1e24df0 100644
--- a/arch/sh/kernel/cf-enabler.c
+++ b/arch/sh/kernel/cf-enabler.c
@@ -10,7 +10,8 @@
  */
 
 #include <linux/init.h>
-
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
 #include <asm/io.h>
 #include <asm/irq.h>
 
@@ -32,8 +33,6 @@
 /* SH4 can't access PCMCIA interface through P2 area.
  * we must remap it with appropreate attribute bit of the page set.
  * this part is based on Greg Banks' hd64465_ss.c implementation - Masahiro Abe */
-#include <linux/mm.h>
-#include <linux/vmalloc.h>
 
 #if defined(CONFIG_CF_AREA6)
 #define slot_no 0
@@ -84,7 +83,7 @@ static int __init cf_init_default(void)
 }
 
 #if defined(CONFIG_SH_SOLUTION_ENGINE)
-#include <asm/se/se.h>
+#include <asm/se.h>
 
 /*
  * SolutionEngine
diff --git a/drivers/net/stnic.c b/drivers/net/stnic.c
index 3fd7a4fee665..e6f90427160c 100644
--- a/drivers/net/stnic.c
+++ b/drivers/net/stnic.c
@@ -19,7 +19,7 @@
 
 #include <asm/system.h>
 #include <asm/io.h>
-#include <asm/se/se.h>
+#include <asm/se.h>
 #include <asm/machvec.h>
 #ifdef CONFIG_SH_STANDARD_BIOS
 #include <asm/sh_bios.h>
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 857cc2ebbc69..69faf0171473 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -44,6 +44,9 @@
 #define	IRQ_RINGING	4		/* Ringing IRQ */
 #define	IRQ_CODEC	5		/* CODEC IRQ */
 
+#define __IO_PREFIX	hs7751rvoip
+#include <asm/io_generic.h>
+
 /* arch/sh/boards/renesas/hs7751rvoip/irq.c */
 void init_hs7751rvoip_IRQ(void);
 
diff --git a/include/asm-sh/mc146818rtc.h b/include/asm-sh/mc146818rtc.h
index 1707cfb2915d..adc6e67c6b75 100644
--- a/include/asm-sh/mc146818rtc.h
+++ b/include/asm-sh/mc146818rtc.h
@@ -24,7 +24,7 @@
 #define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,b)
 
 #elif defined(CONFIG_SH_SECUREEDGE5410)
-#include <asm/snapgear/io.h>
+#include <asm/snapgear.h>
 
 #define RTC_PORT(n)             SECUREEDGE_IOPORT_ADDR
 #define CMOS_READ(addr)         secureedge5410_cmos_read(addr)
diff --git a/include/asm-sh/se.h b/include/asm-sh/se.h
new file mode 100644
index 000000000000..a1832154a3aa
--- /dev/null
+++ b/include/asm-sh/se.h
@@ -0,0 +1,80 @@
+#ifndef __ASM_SH_HITACHI_SE_H
+#define __ASM_SH_HITACHI_SE_H
+
+/*
+ * linux/include/asm-sh/hitachi_se.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SolutionEngine support
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_83902	0xb0000000	/* DP83902A */
+#define PA_83902_IF	0xb0040000	/* DP83902A remote io port */
+#define PA_83902_RST	0xb0080000	/* DP83902A reset port */
+
+#define PA_SUPERIO	0xb0400000	/* SMC37C935A super io chip */
+#define PA_DIPSW0	0xb0800000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb0800002	/* Dip switch 7,8 */
+#define PA_LED		0xb0c00000	/* LED */
+#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#define PA_BCR		0xb0e00000
+#else
+#define PA_BCR		0xb1400000	/* FPGA */
+#endif
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#define IRQ_STNIC   12
+#else
+#define IRQ_STNIC	10
+#endif
+
+#define __IO_PREFIX	se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE_H */
diff --git a/include/asm-sh/se/io.h b/include/asm-sh/se/io.h
deleted file mode 100644
index 9eeb86cd6cef..000000000000
--- a/include/asm-sh/se/io.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * include/asm-sh/io_se.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an Hitachi SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_SE_H
-#define _ASM_SH_IO_SE_H
-
-extern unsigned char se_inb(unsigned long port);
-extern unsigned short se_inw(unsigned long port);
-extern unsigned int se_inl(unsigned long port);
-
-extern void se_outb(unsigned char value, unsigned long port);
-extern void se_outw(unsigned short value, unsigned long port);
-extern void se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char se_inb_p(unsigned long port);
-extern void se_outb_p(unsigned char value, unsigned long port);
-
-extern void se_insb(unsigned long port, void *addr, unsigned long count);
-extern void se_insw(unsigned long port, void *addr, unsigned long count);
-extern void se_insl(unsigned long port, void *addr, unsigned long count);
-extern void se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long se_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_SE_H */
diff --git a/include/asm-sh/se/se.h b/include/asm-sh/se/se.h
deleted file mode 100644
index 791c5da0388a..000000000000
--- a/include/asm-sh/se/se.h
+++ /dev/null
@@ -1,77 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE_H
-#define __ASM_SH_HITACHI_SE_H
-
-/*
- * linux/include/asm-sh/hitachi_se.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SolutionEngine support
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_83902	0xb0000000	/* DP83902A */
-#define PA_83902_IF	0xb0040000	/* DP83902A remote io port */
-#define PA_83902_RST	0xb0080000	/* DP83902A reset port */
-
-#define PA_SUPERIO	0xb0400000	/* SMC37C935A super io chip */
-#define PA_DIPSW0	0xb0800000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb0800002	/* Dip switch 7,8 */
-#define PA_LED		0xb0c00000	/* LED */
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
-#define PA_BCR		0xb0e00000
-#else
-#define PA_BCR		0xb1400000	/* FPGA */
-#endif
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
-#define IRQ_STNIC   12
-#else
-#define IRQ_STNIC	10
-#endif
-
-#endif  /* __ASM_SH_HITACHI_SE_H */
diff --git a/include/asm-sh/se7300.h b/include/asm-sh/se7300.h
new file mode 100644
index 000000000000..4e24edccb30d
--- /dev/null
+++ b/include/asm-sh/se7300.h
@@ -0,0 +1,64 @@
+#ifndef __ASM_SH_HITACHI_SE7300_H
+#define __ASM_SH_HITACHI_SE7300_H
+
+/*
+ * linux/include/asm-sh/se/se7300.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 7300 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0800000	/* LED */
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
+#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
+#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh7300se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE7300_H */
diff --git a/include/asm-sh/se7300/io.h b/include/asm-sh/se7300/io.h
deleted file mode 100644
index c6af85529714..000000000000
--- a/include/asm-sh/se7300/io.h
+++ /dev/null
@@ -1,29 +0,0 @@
-/*
- * include/asm-sh/se7300/io.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- * IO functions for SH-Mobile(SH7300) SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_7300SE_H
-#define _ASM_SH_IO_7300SE_H
-
-extern unsigned char sh7300se_inb(unsigned long port);
-extern unsigned short sh7300se_inw(unsigned long port);
-extern unsigned int sh7300se_inl(unsigned long port);
-
-extern void sh7300se_outb(unsigned char value, unsigned long port);
-extern void sh7300se_outw(unsigned short value, unsigned long port);
-extern void sh7300se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7300se_inb_p(unsigned long port);
-extern void sh7300se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7300se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7300se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7300se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7300se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-#endif /* _ASM_SH_IO_7300SE_H */
diff --git a/include/asm-sh/se7300/se7300.h b/include/asm-sh/se7300/se7300.h
deleted file mode 100644
index 3ec1ded86c97..000000000000
--- a/include/asm-sh/se7300/se7300.h
+++ /dev/null
@@ -1,61 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE7300_H
-#define __ASM_SH_HITACHI_SE7300_H
-
-/*
- * linux/include/asm-sh/se/se7300.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- *
- * SH-Mobile SolutionEngine 7300 support
- */
-
-/* Box specific addresses.  */
-
-/* Area 0 */
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
-#define PA_FROM		0x00400000	/* Flash ROM */
-#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
-#define PA_SRAM		0x00800000	/* SRAM */
-#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
-/* Area 1 */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-/* Area 2 */
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-/* Area 3 */
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-/* Area 4 */
-#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
-#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-#define PA_LED		0xb0800000	/* LED */
-#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
-#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
-#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
-#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
-/* Area 5 */
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-/* Area 6 */
-#define PA_LCD1		0xb8000000
-#define PA_LCD2		0xb8800000
-
-#endif  /* __ASM_SH_HITACHI_SE7300_H */
diff --git a/include/asm-sh/se73180.h b/include/asm-sh/se73180.h
new file mode 100644
index 000000000000..3a4acb3e38a1
--- /dev/null
+++ b/include/asm-sh/se73180.h
@@ -0,0 +1,65 @@
+#ifndef __ASM_SH_HITACHI_SE73180_H
+#define __ASM_SH_HITACHI_SE73180_H
+
+/*
+ * include/asm-sh/se/se73180.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 73180 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0C00000	/* LED */
+#define LED_SHIFT       0
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
+#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
+#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh73180se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_SE73180_H */
diff --git a/include/asm-sh/se73180/io.h b/include/asm-sh/se73180/io.h
deleted file mode 100644
index c9cb1b9412c6..000000000000
--- a/include/asm-sh/se73180/io.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * include/asm-sh/se73180/io.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- * Based on include/asm-sh/se7300/io.h
- *
- * IO functions for SH-Mobile3(SH73180) SolutionEngine
- *
- */
-
-#ifndef _ASM_SH_IO_73180SE_H
-#define _ASM_SH_IO_73180SE_H
-
-extern unsigned char sh73180se_inb(unsigned long port);
-extern unsigned short sh73180se_inw(unsigned long port);
-extern unsigned int sh73180se_inl(unsigned long port);
-
-extern void sh73180se_outb(unsigned char value, unsigned long port);
-extern void sh73180se_outw(unsigned short value, unsigned long port);
-extern void sh73180se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh73180se_inb_p(unsigned long port);
-extern void sh73180se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh73180se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh73180se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh73180se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh73180se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-#endif /* _ASM_SH_IO_73180SE_H */
diff --git a/include/asm-sh/se73180/se73180.h b/include/asm-sh/se73180/se73180.h
deleted file mode 100644
index f5b93e39e768..000000000000
--- a/include/asm-sh/se73180/se73180.h
+++ /dev/null
@@ -1,62 +0,0 @@
-#ifndef __ASM_SH_HITACHI_SE73180_H
-#define __ASM_SH_HITACHI_SE73180_H
-
-/*
- * include/asm-sh/se/se73180.h
- *
- * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
- *
- * SH-Mobile SolutionEngine 73180 support
- */
-
-/* Box specific addresses.  */
-
-/* Area 0 */
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
-#define PA_FROM		0x00400000	/* Flash ROM */
-#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
-#define PA_SRAM		0x00800000	/* SRAM */
-#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
-/* Area 1 */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-/* Area 2 */
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-/* Area 3 */
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-/* Area 4 */
-#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
-#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
-#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-#define PA_LED		0xb0C00000	/* LED */
-#define LED_SHIFT       0
-#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
-#define PA_EPLD_MODESET	0xb0a00000	/* FPGA Mode set register */
-#define PA_EPLD_ST1	0xb0a80000	/* FPGA Interrupt status register1 */
-#define PA_EPLD_ST2	0xb0ac0000	/* FPGA Interrupt status register2 */
-/* Area 5 */
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-/* Area 6 */
-#define PA_LCD1		0xb8000000
-#define PA_LCD2		0xb8800000
-
-#endif  /* __ASM_SH_HITACHI_SE73180_H */
diff --git a/include/asm-sh/se7751.h b/include/asm-sh/se7751.h
new file mode 100644
index 000000000000..88cd379d9084
--- /dev/null
+++ b/include/asm-sh/se7751.h
@@ -0,0 +1,71 @@
+#ifndef __ASM_SH_HITACHI_7751SE_H
+#define __ASM_SH_HITACHI_7751SE_H
+
+/*
+ * linux/include/asm-sh/hitachi_7751se.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SolutionEngine support
+
+ * Modified for 7751 Solution Engine by
+ * Ian da Silva and Jeremy Siegel, 2001.
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
+#define PA_LED		0xba000000	/* LED */
+#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_MODE     (PA_MRSHPC + 4)
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#define IRQ_79C973	13
+
+#define __IO_PREFIX	sh7751se
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_HITACHI_7751SE_H */
diff --git a/include/asm-sh/se7751/io.h b/include/asm-sh/se7751/io.h
deleted file mode 100644
index 78d8f5744bc5..000000000000
--- a/include/asm-sh/se7751/io.h
+++ /dev/null
@@ -1,42 +0,0 @@
-/*
- * include/asm-sh/io_7751se.h
- *
- * Modified version of io_se.h for the 7751se-specific functions.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an Hitachi SolutionEngine
- */
-
-#ifndef _ASM_SH_IO_7751SE_H
-#define _ASM_SH_IO_7751SE_H
-
-extern unsigned char sh7751se_inb(unsigned long port);
-extern unsigned short sh7751se_inw(unsigned long port);
-extern unsigned int sh7751se_inl(unsigned long port);
-
-extern void sh7751se_outb(unsigned char value, unsigned long port);
-extern void sh7751se_outw(unsigned short value, unsigned long port);
-extern void sh7751se_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7751se_inb_p(unsigned long port);
-extern void sh7751se_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7751se_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7751se_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751se_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char sh7751se_readb(unsigned long addr);
-extern unsigned short sh7751se_readw(unsigned long addr);
-extern unsigned int sh7751se_readl(unsigned long addr);
-extern void sh7751se_writeb(unsigned char b, unsigned long addr);
-extern void sh7751se_writew(unsigned short b, unsigned long addr);
-extern void sh7751se_writel(unsigned int b, unsigned long addr);
-
-extern unsigned long sh7751se_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_7751SE_H */
diff --git a/include/asm-sh/se7751/se7751.h b/include/asm-sh/se7751/se7751.h
deleted file mode 100644
index 738e22bebdfb..000000000000
--- a/include/asm-sh/se7751/se7751.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef __ASM_SH_HITACHI_7751SE_H
-#define __ASM_SH_HITACHI_7751SE_H
-
-/*
- * linux/include/asm-sh/hitachi_7751se.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SolutionEngine support
-
- * Modified for 7751 Solution Engine by
- * Ian da Silva and Jeremy Siegel, 2001.
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
-#define PA_LED		0xba000000	/* LED */
-#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_MODE     (PA_MRSHPC + 4)
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#define IRQ_79C973	13
-
-#endif  /* __ASM_SH_HITACHI_7751SE_H */
diff --git a/include/asm-sh/sh03/io.h b/include/asm-sh/sh03/io.h
index 25792e9831ea..df3b187ef883 100644
--- a/include/asm-sh/sh03/io.h
+++ b/include/asm-sh/sh03/io.h
@@ -33,14 +33,6 @@
 #define IRL3_IPR_POS	0
 #define IRL3_PRIORITY	4
 
-
-extern unsigned long sh03_isa_port2addr(unsigned long offset);
-
-extern void setup_sh03(void);
-extern void init_sh03_IRQ(void);
-extern void heartbeat_sh03(void);
-
-extern void sh03_rtc_gettimeofday(struct timeval *tv);
-extern int sh03_rtc_settimeofday(const struct timeval *tv);
+void heartbeat_sh03(void);
 
 #endif /* _ASM_SH_IO_SH03_H */
diff --git a/include/asm-sh/snapgear.h b/include/asm-sh/snapgear.h
new file mode 100644
index 000000000000..6b5e4ddc073a
--- /dev/null
+++ b/include/asm-sh/snapgear.h
@@ -0,0 +1,79 @@
+/*
+ * include/asm-sh/snapgear/io.h
+ *
+ * Modified version of io_se.h for the snapgear-specific functions.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * IO functions for a SnapGear
+ */
+
+#ifndef _ASM_SH_IO_SNAPGEAR_H
+#define _ASM_SH_IO_SNAPGEAR_H
+
+#if defined(CONFIG_CPU_SH4)
+/*
+ * The external interrupt lines, these take up ints 0 - 15 inclusive
+ * depending on the priority for the interrupt.  In fact the priority
+ * is the interrupt :-)
+ */
+
+#define IRL0_IRQ		2
+#define IRL0_IPR_ADDR	INTC_IPRD
+#define IRL0_IPR_POS	3
+#define IRL0_PRIORITY	13
+
+#define IRL1_IRQ		5
+#define IRL1_IPR_ADDR	INTC_IPRD
+#define IRL1_IPR_POS	2
+#define IRL1_PRIORITY	10
+
+#define IRL2_IRQ		8
+#define IRL2_IPR_ADDR	INTC_IPRD
+#define IRL2_IPR_POS	1
+#define IRL2_PRIORITY	7
+
+#define IRL3_IRQ		11
+#define IRL3_IPR_ADDR	INTC_IPRD
+#define IRL3_IPR_POS	0
+#define IRL3_PRIORITY	4
+#endif
+
+#define __IO_PREFIX	snapgear
+#include <asm/io_generic.h>
+
+#ifdef CONFIG_SH_SECUREEDGE5410
+/*
+ * We need to remember what was written to the ioport as some bits
+ * are shared with other functions and you cannot read back what was
+ * written :-|
+ *
+ * Bit        Read                   Write
+ * -----------------------------------------------
+ * D0         DCD on ttySC1          power
+ * D1         Reset Switch           heatbeat
+ * D2         ttySC0 CTS (7100)      LAN
+ * D3         -                      WAN
+ * D4         ttySC0 DCD (7100)      CONSOLE
+ * D5         -                      ONLINE
+ * D6         -                      VPN
+ * D7         -                      DTR on ttySC1
+ * D8         -                      ttySC0 RTS (7100)
+ * D9         -                      ttySC0 DTR (7100)
+ * D10        -                      RTC SCLK
+ * D11        RTC DATA               RTC DATA
+ * D12        -                      RTS RESET
+ */
+
+#define SECUREEDGE_IOPORT_ADDR ((volatile short *) 0xb0000000)
+extern unsigned short secureedge5410_ioport;
+
+#define SECUREEDGE_WRITE_IOPORT(val, mask) (*SECUREEDGE_IOPORT_ADDR = \
+	 (secureedge5410_ioport = \
+			((secureedge5410_ioport & ~(mask)) | ((val) & (mask)))))
+#define SECUREEDGE_READ_IOPORT() \
+	 ((*SECUREEDGE_IOPORT_ADDR&0x0817) | (secureedge5410_ioport&~0x0817))
+#endif
+
+#endif /* _ASM_SH_IO_SNAPGEAR_H */
diff --git a/include/asm-sh/snapgear/io.h b/include/asm-sh/snapgear/io.h
deleted file mode 100644
index bfa97ac06280..000000000000
--- a/include/asm-sh/snapgear/io.h
+++ /dev/null
@@ -1,92 +0,0 @@
-/*
- * include/asm-sh/snapgear/io.h
- *
- * Modified version of io_se.h for the snapgear-specific functions.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for a SnapGear
- */
-
-#ifndef _ASM_SH_IO_SNAPGEAR_H
-#define _ASM_SH_IO_SNAPGEAR_H
-
-#if defined(CONFIG_CPU_SH4)
-/*
- * The external interrupt lines, these take up ints 0 - 15 inclusive
- * depending on the priority for the interrupt.  In fact the priority
- * is the interrupt :-)
- */
-
-#define IRL0_IRQ		2
-#define IRL0_IPR_ADDR	INTC_IPRD
-#define IRL0_IPR_POS	3
-#define IRL0_PRIORITY	13
-
-#define IRL1_IRQ		5
-#define IRL1_IPR_ADDR	INTC_IPRD
-#define IRL1_IPR_POS	2
-#define IRL1_PRIORITY	10
-
-#define IRL2_IRQ		8
-#define IRL2_IPR_ADDR	INTC_IPRD
-#define IRL2_IPR_POS	1
-#define IRL2_PRIORITY	7
-
-#define IRL3_IRQ		11
-#define IRL3_IPR_ADDR	INTC_IPRD
-#define IRL3_IPR_POS	0
-#define IRL3_PRIORITY	4
-#endif
-
-extern unsigned char snapgear_inb(unsigned long port);
-extern unsigned short snapgear_inw(unsigned long port);
-extern unsigned int snapgear_inl(unsigned long port);
-
-extern void snapgear_outb(unsigned char value, unsigned long port);
-extern void snapgear_outw(unsigned short value, unsigned long port);
-extern void snapgear_outl(unsigned int value, unsigned long port);
-
-extern unsigned char snapgear_inb_p(unsigned long port);
-extern void snapgear_outb_p(unsigned char value, unsigned long port);
-
-extern void snapgear_insl(unsigned long port, void *addr, unsigned long count);
-extern void snapgear_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long snapgear_isa_port2addr(unsigned long offset);
-
-#ifdef CONFIG_SH_SECUREEDGE5410
-/*
- * We need to remember what was written to the ioport as some bits
- * are shared with other functions and you cannot read back what was
- * written :-|
- *
- * Bit        Read                   Write
- * -----------------------------------------------
- * D0         DCD on ttySC1          power
- * D1         Reset Switch           heatbeat
- * D2         ttySC0 CTS (7100)      LAN
- * D3         -                      WAN
- * D4         ttySC0 DCD (7100)      CONSOLE
- * D5         -                      ONLINE
- * D6         -                      VPN
- * D7         -                      DTR on ttySC1
- * D8         -                      ttySC0 RTS (7100)
- * D9         -                      ttySC0 DTR (7100)
- * D10        -                      RTC SCLK
- * D11        RTC DATA               RTC DATA
- * D12        -                      RTS RESET
- */
-
- #define SECUREEDGE_IOPORT_ADDR ((volatile short *) 0xb0000000)
- extern unsigned short secureedge5410_ioport;
-
- #define SECUREEDGE_WRITE_IOPORT(val, mask) (*SECUREEDGE_IOPORT_ADDR = \
-		 (secureedge5410_ioport = \
-		 		((secureedge5410_ioport & ~(mask)) | ((val) & (mask)))))
- #define SECUREEDGE_READ_IOPORT() \
- 		 ((*SECUREEDGE_IOPORT_ADDR&0x0817) | (secureedge5410_ioport&~0x0817))
-#endif
-
-#endif /* _ASM_SH_IO_SNAPGEAR_H */
diff --git a/include/asm-sh/systemh/7751systemh.h b/include/asm-sh/systemh/7751systemh.h
deleted file mode 100644
index 4170531bdbd9..000000000000
--- a/include/asm-sh/systemh/7751systemh.h
+++ /dev/null
@@ -1,68 +0,0 @@
-#ifndef __ASM_SH_SYSTEMH_7751SYSTEMH_H
-#define __ASM_SH_SYSTEMH_7751SYSTEMH_H
-
-/*
- * linux/include/asm-sh/systemh/7751systemh.h
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
- * Hitachi SystemH support
-
- * Modified for 7751 SystemH by
- * Jonathan Short, 2002.
- */
-
-/* Box specific addresses.  */
-
-#define PA_ROM		0x00000000	/* EPROM */
-#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_FROM		0x01000000	/* EPROM */
-#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
-#define PA_EXT1		0x04000000
-#define PA_EXT1_SIZE	0x04000000
-#define PA_EXT2		0x08000000
-#define PA_EXT2_SIZE	0x04000000
-#define PA_SDRAM	0x0c000000
-#define PA_SDRAM_SIZE	0x04000000
-
-#define PA_EXT4		0x12000000
-#define PA_EXT4_SIZE	0x02000000
-#define PA_EXT5		0x14000000
-#define PA_EXT5_SIZE	0x04000000
-#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
-
-#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
-#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
-#define PA_LED		0xba000000	/* LED */
-#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
-
-#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
-#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
-#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
-#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
-#define MRSHPC_MODE     (PA_MRSHPC + 4)
-#define MRSHPC_OPTION   (PA_MRSHPC + 6)
-#define MRSHPC_CSR      (PA_MRSHPC + 8)
-#define MRSHPC_ISR      (PA_MRSHPC + 10)
-#define MRSHPC_ICR      (PA_MRSHPC + 12)
-#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
-#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
-#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
-#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
-#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
-#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
-#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
-#define MRSHPC_CDCR     (PA_MRSHPC + 28)
-#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
-
-#define BCR_ILCRA	(PA_BCR + 0)
-#define BCR_ILCRB	(PA_BCR + 2)
-#define BCR_ILCRC	(PA_BCR + 4)
-#define BCR_ILCRD	(PA_BCR + 6)
-#define BCR_ILCRE	(PA_BCR + 8)
-#define BCR_ILCRF	(PA_BCR + 10)
-#define BCR_ILCRG	(PA_BCR + 12)
-
-#define IRQ_79C973	13
-
-#endif  /* __ASM_SH_SYSTEMH_7751SYSTEMH_H */
diff --git a/include/asm-sh/systemh/io.h b/include/asm-sh/systemh/io.h
deleted file mode 100644
index 327849b49db8..000000000000
--- a/include/asm-sh/systemh/io.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/*
- * include/asm-sh/systemh/io.h
- *
- * Stupid I/O definitions for SystemH, cloned from SE7751.
- *
- * Copyright (C) 2003  Paul Mundt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- */
-#ifndef __ASM_SH_SYSTEMH_IO_H
-#define __ASM_SH_SYSTEMH_IO_H
-
-extern unsigned char sh7751systemh_inb(unsigned long port);
-extern unsigned short sh7751systemh_inw(unsigned long port);
-extern unsigned int sh7751systemh_inl(unsigned long port);
-
-extern void sh7751systemh_outb(unsigned char value, unsigned long port);
-extern void sh7751systemh_outw(unsigned short value, unsigned long port);
-extern void sh7751systemh_outl(unsigned int value, unsigned long port);
-
-extern unsigned char sh7751systemh_inb_p(unsigned long port);
-extern void sh7751systemh_outb_p(unsigned char value, unsigned long port);
-
-extern void sh7751systemh_insb(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_insw(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_insl(unsigned long port, void *addr, unsigned long count);
-extern void sh7751systemh_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751systemh_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void sh7751systemh_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char sh7751systemh_readb(unsigned long addr);
-extern unsigned short sh7751systemh_readw(unsigned long addr);
-extern unsigned int sh7751systemh_readl(unsigned long addr);
-extern void sh7751systemh_writeb(unsigned char b, unsigned long addr);
-extern void sh7751systemh_writew(unsigned short b, unsigned long addr);
-extern void sh7751systemh_writel(unsigned int b, unsigned long addr);
-
-extern unsigned long sh7751systemh_isa_port2addr(unsigned long offset);
-
-#endif /* __ASM_SH_SYSTEMH_IO_H */
-
diff --git a/include/asm-sh/systemh7751.h b/include/asm-sh/systemh7751.h
new file mode 100644
index 000000000000..b143bb2a2ca7
--- /dev/null
+++ b/include/asm-sh/systemh7751.h
@@ -0,0 +1,71 @@
+#ifndef __ASM_SH_SYSTEMH_7751SYSTEMH_H
+#define __ASM_SH_SYSTEMH_7751SYSTEMH_H
+
+/*
+ * linux/include/asm-sh/systemh/7751systemh.h
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Hitachi SystemH support
+
+ * Modified for 7751 SystemH by
+ * Jonathan Short, 2002.
+ */
+
+/* Box specific addresses.  */
+
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_FROM		0x01000000	/* EPROM */
+#define PA_FROM_SIZE	0x00400000	/* EPROM size 4M byte */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+
+#define PA_EXT4		0x12000000
+#define PA_EXT4_SIZE	0x02000000
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+#define PA_PCIC		0x18000000	/* MR-SHPC-01 PCMCIA */
+
+#define PA_DIPSW0	0xb9000000	/* Dip switch 5,6 */
+#define PA_DIPSW1	0xb9000002	/* Dip switch 7,8 */
+#define PA_LED		0xba000000	/* LED */
+#define	PA_BCR		0xbb000000	/* FPGA on the MS7751SE01 */
+
+#define PA_MRSHPC	0xb83fffe0	/* MR-SHPC-01 PCMCIA controler */
+#define PA_MRSHPC_MW1	0xb8400000	/* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2	0xb8500000	/* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO	0xb8600000	/* MR-SHPC-01 I/O window base */
+#define MRSHPC_MODE     (PA_MRSHPC + 4)
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+
+#define BCR_ILCRA	(PA_BCR + 0)
+#define BCR_ILCRB	(PA_BCR + 2)
+#define BCR_ILCRC	(PA_BCR + 4)
+#define BCR_ILCRD	(PA_BCR + 6)
+#define BCR_ILCRE	(PA_BCR + 8)
+#define BCR_ILCRF	(PA_BCR + 10)
+#define BCR_ILCRG	(PA_BCR + 12)
+
+#define IRQ_79C973	13
+
+#define __IO_PREFIX	sh7751systemh
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_SYSTEMH_7751SYSTEMH_H */
-- 
cgit v1.2.1


From d7c30c682a278abe1a52db83f69efec1a9d8f8c2 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:49:57 +0900
Subject: sh: Store Queue API rework.

Rewrite the store queue API for a per-cpu interface in the driver
model. The old miscdevice is dropped, due to TASK_SIZE limitations,
and no one was using it anyways.

Carve up and allocate store queue space with a bitmap, back sq
mapping objects with a slab cache, and let userspace worry about
its own prefetching.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/sq.c | 542 ++++++++++++++++++++------------------------
 include/asm-sh/cpu-sh4/sq.h |  23 +-
 2 files changed, 251 insertions(+), 314 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh4/sq.c b/arch/sh/kernel/cpu/sh4/sq.c
index b148966dd7c7..7bcc73f9b8df 100644
--- a/arch/sh/kernel/cpu/sh4/sq.c
+++ b/arch/sh/kernel/cpu/sh4/sq.c
@@ -1,50 +1,52 @@
 /*
- * arch/sh/kernel/cpu/sq.c
+ * arch/sh/kernel/cpu/sh4/sq.c
  *
  * General management API for SH-4 integrated Store Queues
  *
- * Copyright (C) 2001, 2002, 2003, 2004  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2001, 2002  M. R. Brown
  *
- * Some of this code has been adopted directly from the old arch/sh/mm/sq.c
- * hack that was part of the LinuxDC project. For all intents and purposes,
- * this is a completely new interface that really doesn't have much in common
- * with the old zone-based approach at all. In fact, it's only listed here for
- * general completeness.
- *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
 #include <linux/init.h>
+#include <linux/cpu.h>
+#include <linux/bitmap.h>
+#include <linux/sysdev.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/list.h>
-#include <linux/proc_fs.h>
-#include <linux/miscdevice.h>
 #include <linux/vmalloc.h>
 #include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/page.h>
 #include <asm/cacheflush.h>
-#include <asm/mmu_context.h>
 #include <asm/cpu/sq.h>
 
-static LIST_HEAD(sq_mapping_list);
+struct sq_mapping;
+
+struct sq_mapping {
+	const char *name;
+
+	unsigned long sq_addr;
+	unsigned long addr;
+	unsigned int size;
+
+	struct sq_mapping *next;
+};
+
+static struct sq_mapping *sq_mapping_list;
 static DEFINE_SPINLOCK(sq_mapping_lock);
+static kmem_cache_t *sq_cache;
+static unsigned long *sq_bitmap;
 
-/**
- * sq_flush - Flush (prefetch) the store queue cache
- * @addr: the store queue address to flush
- *
- * Executes a prefetch instruction on the specified store queue cache,
- * so that the cached data is written to physical memory.
- */
-inline void sq_flush(void *addr)
-{
-	__asm__ __volatile__ ("pref @%0" : : "r" (addr) : "memory");
-}
+#define store_queue_barrier()			\
+do {						\
+	(void)ctrl_inl(P4SEG_STORE_QUE);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 0);	\
+	ctrl_outl(0, P4SEG_STORE_QUE + 8);	\
+} while (0);
 
 /**
  * sq_flush_range - Flush (prefetch) a specific SQ range
@@ -57,154 +59,73 @@ inline void sq_flush(void *addr)
 void sq_flush_range(unsigned long start, unsigned int len)
 {
 	volatile unsigned long *sq = (unsigned long *)start;
-	unsigned long dummy;
 
 	/* Flush the queues */
 	for (len >>= 5; len--; sq += 8)
-		sq_flush((void *)sq);
+		prefetchw((void *)sq);
 
 	/* Wait for completion */
-	dummy = ctrl_inl(P4SEG_STORE_QUE);
-
-	ctrl_outl(0, P4SEG_STORE_QUE + 0);
-	ctrl_outl(0, P4SEG_STORE_QUE + 8);
+	store_queue_barrier();
 }
 
-static struct sq_mapping *__sq_alloc_mapping(unsigned long virt, unsigned long phys, unsigned long size, const char *name)
+static inline void sq_mapping_list_add(struct sq_mapping *map)
 {
-	struct sq_mapping *map;
-
-	if (virt + size > SQ_ADDRMAX)
-		return ERR_PTR(-ENOSPC);
+	struct sq_mapping **p, *tmp;
 
-	map = kmalloc(sizeof(struct sq_mapping), GFP_KERNEL);
-	if (!map)
-		return ERR_PTR(-ENOMEM);
+	spin_lock_irq(&sq_mapping_lock);
 
-	INIT_LIST_HEAD(&map->list);
+	p = &sq_mapping_list;
+	while ((tmp = *p) != NULL)
+		p = &tmp->next;
 
-	map->sq_addr	= virt;
-	map->addr	= phys;
-	map->size	= size + 1;
-	map->name	= name;
+	map->next = tmp;
+	*p = map;
 
-	list_add(&map->list, &sq_mapping_list);
-
-	return map;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-static unsigned long __sq_get_next_addr(void)
+static inline void sq_mapping_list_del(struct sq_mapping *map)
 {
-	if (!list_empty(&sq_mapping_list)) {
-		struct list_head *pos, *tmp;
-
-		/*
-		 * Read one off the list head, as it will have the highest
-		 * mapped allocation. Set the next one up right above it.
-		 *
-		 * This is somewhat sub-optimal, as we don't look at
-		 * gaps between allocations or anything lower then the
-		 * highest-level allocation.
-		 *
-		 * However, in the interest of performance and the general
-		 * lack of desire to do constant list rebalancing, we don't
-		 * worry about it.
-		 */
-		list_for_each_safe(pos, tmp, &sq_mapping_list) {
-			struct sq_mapping *entry;
-
-			entry = list_entry(pos, typeof(*entry), list);
-
-			return entry->sq_addr + entry->size;
+	struct sq_mapping **p, *tmp;
+
+	spin_lock_irq(&sq_mapping_lock);
+
+	for (p = &sq_mapping_list; (tmp = *p); p = &tmp->next)
+		if (tmp == map) {
+			*p = tmp->next;
+			break;
 		}
-	}
 
-	return P4SEG_STORE_QUE;
+	spin_unlock_irq(&sq_mapping_lock);
 }
 
-/**
- * __sq_remap - Perform a translation from the SQ to a phys addr
- * @map: sq mapping containing phys and store queue addresses.
- *
- * Maps the store queue address specified in the mapping to the physical
- * address specified in the mapping.
- */
-static struct sq_mapping *__sq_remap(struct sq_mapping *map)
+static int __sq_remap(struct sq_mapping *map, unsigned long flags)
 {
-	unsigned long flags, pteh, ptel;
+#if defined(CONFIG_MMU)
 	struct vm_struct *vma;
-	pgprot_t pgprot;
 
-	/*
-	 * Without an MMU (or with it turned off), this is much more
-	 * straightforward, as we can just load up each queue's QACR with
-	 * the physical address appropriately masked.
-	 */
-
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
-	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
-
-#ifdef CONFIG_MMU
-	/*
-	 * With an MMU on the other hand, things are slightly more involved.
-	 * Namely, we have to have a direct mapping between the SQ addr and
-	 * the associated physical address in the UTLB by way of setting up
-	 * a virt<->phys translation by hand. We do this by simply specifying
-	 * the SQ addr in UTLB.VPN and the associated physical address in
-	 * UTLB.PPN.
-	 *
-	 * Notably, even though this is a special case translation, and some
-	 * of the configuration bits are meaningless, we're still required
-	 * to have a valid ASID context in PTEH.
-	 *
-	 * We could also probably get by without explicitly setting PTEA, but
-	 * we do it here just for good measure.
-	 */
-	spin_lock_irqsave(&sq_mapping_lock, flags);
-
-	pteh = map->sq_addr;
-	ctrl_outl((pteh & MMU_VPN_MASK) | get_asid(), MMU_PTEH);
-
-	ptel = map->addr & PAGE_MASK;
-
-	if (cpu_data->flags & CPU_HAS_PTEA)
-		ctrl_outl(((ptel >> 28) & 0xe) | (ptel & 0x1), MMU_PTEA);
-
-	pgprot = pgprot_noncached(PAGE_KERNEL);
-
-	ptel &= _PAGE_FLAGS_HARDWARE_MASK;
-	ptel |= pgprot_val(pgprot);
-	ctrl_outl(ptel, MMU_PTEL);
-
-	__asm__ __volatile__ ("ldtlb" : : : "memory");
-
-	spin_unlock_irqrestore(&sq_mapping_lock, flags);
-
-	/*
-	 * Next, we need to map ourselves in the kernel page table, so that
-	 * future accesses after a TLB flush will be handled when we take a
-	 * page fault.
-	 *
-	 * Theoretically we could just do this directly and not worry about
-	 * setting up the translation by hand ahead of time, but for the
-	 * cases where we want a one-shot SQ mapping followed by a quick
-	 * writeout before we hit the TLB flush, we do it anyways. This way
-	 * we at least save ourselves the initial page fault overhead.
-	 */
 	vma = __get_vm_area(map->size, VM_ALLOC, map->sq_addr, SQ_ADDRMAX);
 	if (!vma)
-		return ERR_PTR(-ENOMEM);
+		return -ENOMEM;
 
 	vma->phys_addr = map->addr;
 
 	if (remap_area_pages((unsigned long)vma->addr, vma->phys_addr,
-			     map->size, pgprot_val(pgprot))) {
+			     map->size, flags)) {
 		vunmap(vma->addr);
-		return NULL;
+		return -EAGAIN;
 	}
-#endif /* CONFIG_MMU */
+#else
+	/*
+	 * Without an MMU (or with it turned off), this is much more
+	 * straightforward, as we can just load up each queue's QACR with
+	 * the physical address appropriately masked.
+	 */
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR0);
+	ctrl_outl(((map->addr >> 26) << 2) & 0x1c, SQ_QACR1);
+#endif
 
-	return map;
+	return 0;
 }
 
 /**
@@ -212,42 +133,65 @@ static struct sq_mapping *__sq_remap(struct sq_mapping *map)
  * @phys: Physical address of mapping.
  * @size: Length of mapping.
  * @name: User invoking mapping.
+ * @flags: Protection flags.
  *
  * Remaps the physical address @phys through the next available store queue
  * address of @size length. @name is logged at boot time as well as through
- * the procfs interface.
- *
- * A pre-allocated and filled sq_mapping pointer is returned, and must be
- * cleaned up with a call to sq_unmap() when the user is done with the
- * mapping.
+ * the sysfs interface.
  */
-struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *name)
+unsigned long sq_remap(unsigned long phys, unsigned int size,
+		       const char *name, unsigned long flags)
 {
 	struct sq_mapping *map;
-	unsigned long virt, end;
+	unsigned long end;
 	unsigned int psz;
+	int ret, page;
 
 	/* Don't allow wraparound or zero size */
 	end = phys + size - 1;
-	if (!size || end < phys)
-		return NULL;
+	if (unlikely(!size || end < phys))
+		return -EINVAL;
 	/* Don't allow anyone to remap normal memory.. */
-	if (phys < virt_to_phys(high_memory))
-		return NULL;
+	if (unlikely(phys < virt_to_phys(high_memory)))
+		return -EINVAL;
 
 	phys &= PAGE_MASK;
+	size = PAGE_ALIGN(end + 1) - phys;
+
+	map = kmem_cache_alloc(sq_cache, GFP_KERNEL);
+	if (unlikely(!map))
+		return -ENOMEM;
+
+	map->addr = phys;
+	map->size = size;
+	map->name = name;
+
+	page = bitmap_find_free_region(sq_bitmap, 0x04000000,
+				       get_order(map->size));
+	if (unlikely(page < 0)) {
+		ret = -ENOSPC;
+		goto out;
+	}
+
+	map->sq_addr = P4SEG_STORE_QUE + (page << PAGE_SHIFT);
+
+	ret = __sq_remap(map, flags);
+	if (unlikely(ret != 0))
+		goto out;
+
+	psz = (size + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
+	pr_info("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
+		likely(map->name) ? map->name : "???",
+		psz, psz == 1 ? " " : "s",
+		map->sq_addr, map->addr);
 
-	size  = PAGE_ALIGN(end + 1) - phys;
-	virt  = __sq_get_next_addr();
-	psz   = (size + (PAGE_SIZE - 1)) / PAGE_SIZE;
-	map   = __sq_alloc_mapping(virt, phys, size, name);
+	sq_mapping_list_add(map);
 
-	printk("sqremap: %15s  [%4d page%s]  va 0x%08lx   pa 0x%08lx\n",
-	       map->name ? map->name : "???",
-	       psz, psz == 1 ? " " : "s",
-	       map->sq_addr, map->addr);
+	return map->sq_addr;
 
-	return __sq_remap(map);
+out:
+	kmem_cache_free(sq_cache, map);
+	return ret;
 }
 
 /**
@@ -258,188 +202,198 @@ struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *n
  * sq_remap(). Also frees up the pte that was previously inserted into
  * the kernel page table and discards the UTLB translation.
  */
-void sq_unmap(struct sq_mapping *map)
+void sq_unmap(unsigned long vaddr)
 {
-	if (map->sq_addr > (unsigned long)high_memory)
-		vfree((void *)(map->sq_addr & PAGE_MASK));
+	struct sq_mapping **p, *map;
+	struct vm_struct *vma;
+	int page;
 
-	list_del(&map->list);
-	kfree(map);
-}
+	for (p = &sq_mapping_list; (map = *p); p = &map->next)
+		if (map->sq_addr == vaddr)
+			break;
 
-/**
- * sq_clear - Clear a store queue range
- * @addr: Address to start clearing from.
- * @len: Length to clear.
- *
- * A quick zero-fill implementation for clearing out memory that has been
- * remapped through the store queues.
- */
-void sq_clear(unsigned long addr, unsigned int len)
-{
-	int i;
+	if (unlikely(!map)) {
+		printk("%s: bad store queue address 0x%08lx\n",
+		       __FUNCTION__, vaddr);
+		return;
+	}
 
-	/* Clear out both queues linearly */
-	for (i = 0; i < 8; i++) {
-		ctrl_outl(0, addr + i + 0);
-		ctrl_outl(0, addr + i + 8);
+	page = (map->sq_addr - P4SEG_STORE_QUE) >> PAGE_SHIFT;
+	bitmap_release_region(sq_bitmap, page, get_order(map->size));
+
+#ifdef CONFIG_MMU
+	vma = remove_vm_area((void *)(map->sq_addr & PAGE_MASK));
+	if (!vma) {
+		printk(KERN_ERR "%s: bad address 0x%08lx\n",
+		       __FUNCTION__, map->sq_addr);
+		return;
 	}
+#endif
+
+	sq_mapping_list_del(map);
 
-	sq_flush_range(addr, len);
+	kmem_cache_free(sq_cache, map);
 }
 
-/**
- * sq_vma_unmap - Unmap a VMA range
- * @area: VMA containing range.
- * @addr: Start of range.
- * @len: Length of range.
+/*
+ * Needlessly complex sysfs interface. Unfortunately it doesn't seem like
+ * there is any other easy way to add things on a per-cpu basis without
+ * putting the directory entries somewhere stupid and having to create
+ * links in sysfs by hand back in to the per-cpu directories.
  *
- * Searches the sq_mapping_list for a mapping matching the sq addr @addr,
- * and subsequently frees up the entry. Further cleanup is done by generic
- * code.
+ * Some day we may want to have an additional abstraction per store
+ * queue, but considering the kobject hell we already have to deal with,
+ * it's simply not worth the trouble.
  */
-static void sq_vma_unmap(struct vm_area_struct *area,
-			 unsigned long addr, size_t len)
-{
-	struct list_head *pos, *tmp;
+static struct kobject *sq_kobject[NR_CPUS];
 
-	list_for_each_safe(pos, tmp, &sq_mapping_list) {
-		struct sq_mapping *entry;
+struct sq_sysfs_attr {
+	struct attribute attr;
+	ssize_t (*show)(char *buf);
+	ssize_t (*store)(const char *buf, size_t count);
+};
 
-		entry = list_entry(pos, typeof(*entry), list);
+#define to_sq_sysfs_attr(attr)	container_of(attr, struct sq_sysfs_attr, attr)
 
-		if (entry->sq_addr == addr) {
-			/*
-			 * We could probably get away without doing the tlb flush
-			 * here, as generic code should take care of most of this
-			 * when unmapping the rest of the VMA range for us. Leave
-			 * it in for added sanity for the time being..
-			 */
-			__flush_tlb_page(get_asid(), entry->sq_addr & PAGE_MASK);
+static ssize_t sq_sysfs_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-			list_del(&entry->list);
-			kfree(entry);
+	if (likely(sattr->show))
+		return sattr->show(buf);
 
-			return;
-		}
-	}
+	return -EIO;
 }
 
-/**
- * sq_vma_sync - Sync a VMA range
- * @area: VMA containing range.
- * @start: Start of range.
- * @len: Length of range.
- * @flags: Additional flags.
- *
- * Synchronizes an sq mapped range by flushing the store queue cache for
- * the duration of the mapping.
- *
- * Used internally for user mappings, which must use msync() to prefetch
- * the store queue cache.
- */
-static int sq_vma_sync(struct vm_area_struct *area,
-		       unsigned long start, size_t len, unsigned int flags)
+static ssize_t sq_sysfs_store(struct kobject *kobj, struct attribute *attr,
+			      const char *buf, size_t count)
 {
-	sq_flush_range(start, len);
+	struct sq_sysfs_attr *sattr = to_sq_sysfs_attr(attr);
 
-	return 0;
+	if (likely(sattr->store))
+		return sattr->store(buf, count);
+
+	return -EIO;
 }
 
-static struct vm_operations_struct sq_vma_ops = {
-	.unmap	= sq_vma_unmap,
-	.sync	= sq_vma_sync,
-};
+static ssize_t mapping_show(char *buf)
+{
+	struct sq_mapping **list, *entry;
+	char *p = buf;
 
-/**
- * sq_mmap - mmap() for /dev/cpu/sq
- * @file: unused.
- * @vma: VMA to remap.
- *
- * Remap the specified vma @vma through the store queues, and setup associated
- * information for the new mapping. Also build up the page tables for the new
- * area.
- */
-static int sq_mmap(struct file *file, struct vm_area_struct *vma)
+	for (list = &sq_mapping_list; (entry = *list); list = &entry->next)
+		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n",
+			     entry->sq_addr, entry->sq_addr + entry->size,
+			     entry->addr, entry->name);
+
+	return p - buf;
+}
+
+static ssize_t mapping_store(const char *buf, size_t count)
 {
-	unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
-	unsigned long size = vma->vm_end - vma->vm_start;
-	struct sq_mapping *map;
+	unsigned long base = 0, len = 0;
 
-	/*
-	 * We're not interested in any arbitrary virtual address that has
-	 * been stuck in the VMA, as we already know what addresses we
-	 * want. Save off the size, and reposition the VMA to begin at
-	 * the next available sq address.
-	 */
-	vma->vm_start = __sq_get_next_addr();
-	vma->vm_end   = vma->vm_start + size;
+	sscanf(buf, "%lx %lx", &base, &len);
+	if (!base)
+		return -EIO;
 
-	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	if (likely(len)) {
+		int ret = sq_remap(base, len, "Userspace",
+				   pgprot_val(PAGE_SHARED));
+		if (ret < 0)
+			return ret;
+	} else
+		sq_unmap(base);
 
-	vma->vm_flags |= VM_IO | VM_RESERVED;
+	return count;
+}
 
-	map = __sq_alloc_mapping(vma->vm_start, offset, size, "Userspace");
+static struct sq_sysfs_attr mapping_attr =
+	__ATTR(mapping, 0644, mapping_show, mapping_store);
 
-	if (io_remap_pfn_range(vma, map->sq_addr, map->addr >> PAGE_SHIFT,
-				size, vma->vm_page_prot))
-		return -EAGAIN;
+static struct attribute *sq_sysfs_attrs[] = {
+	&mapping_attr.attr,
+	NULL,
+};
 
-	vma->vm_ops = &sq_vma_ops;
+static struct sysfs_ops sq_sysfs_ops = {
+	.show	= sq_sysfs_show,
+	.store	= sq_sysfs_store,
+};
 
-	return 0;
-}
+static struct kobj_type ktype_percpu_entry = {
+	.sysfs_ops	= &sq_sysfs_ops,
+	.default_attrs	= sq_sysfs_attrs,
+};
 
-#ifdef CONFIG_PROC_FS
-static int sq_mapping_read_proc(char *buf, char **start, off_t off,
-				int len, int *eof, void *data)
+static int __devinit sq_sysdev_add(struct sys_device *sysdev)
 {
-	struct list_head *pos;
-	char *p = buf;
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj;
 
-	list_for_each_prev(pos, &sq_mapping_list) {
-		struct sq_mapping *entry;
+	sq_kobject[cpu] = kzalloc(sizeof(struct kobject), GFP_KERNEL);
+	if (unlikely(!sq_kobject[cpu]))
+		return -ENOMEM;
 
-		entry = list_entry(pos, typeof(*entry), list);
+	kobj = sq_kobject[cpu];
+	kobj->parent = &sysdev->kobj;
+	kobject_set_name(kobj, "%s", "sq");
+	kobj->ktype = &ktype_percpu_entry;
 
-		p += sprintf(p, "%08lx-%08lx [%08lx]: %s\n", entry->sq_addr,
-			     entry->sq_addr + entry->size - 1, entry->addr,
-			     entry->name);
-	}
-
-	return p - buf;
+	return kobject_register(kobj);
 }
-#endif
 
-static struct file_operations sq_fops = {
-	.owner		= THIS_MODULE,
-	.mmap		= sq_mmap,
-};
+static int __devexit sq_sysdev_remove(struct sys_device *sysdev)
+{
+	unsigned int cpu = sysdev->id;
+	struct kobject *kobj = sq_kobject[cpu];
 
-static struct miscdevice sq_dev = {
-	.minor		= STORE_QUEUE_MINOR,
-	.name		= "sq",
-	.fops		= &sq_fops,
+	kobject_unregister(kobj);
+	return 0;
+}
+
+static struct sysdev_driver sq_sysdev_driver = {
+	.add		= sq_sysdev_add,
+	.remove		= __devexit_p(sq_sysdev_remove),
 };
 
 static int __init sq_api_init(void)
 {
-	int ret;
+	unsigned int nr_pages = 0x04000000 >> PAGE_SHIFT;
+	unsigned int size = (nr_pages + (BITS_PER_LONG - 1)) / BITS_PER_LONG;
+	int ret = -ENOMEM;
+
 	printk(KERN_NOTICE "sq: Registering store queue API.\n");
 
-	create_proc_read_entry("sq_mapping", 0, 0, sq_mapping_read_proc, 0);
+	sq_cache = kmem_cache_create("store_queue_cache",
+				sizeof(struct sq_mapping), 0, 0,
+				NULL, NULL);
+	if (unlikely(!sq_cache))
+		return ret;
 
-	ret = misc_register(&sq_dev);
-	if (ret)
-		remove_proc_entry("sq_mapping", NULL);
+	sq_bitmap = kzalloc(size, GFP_KERNEL);
+	if (unlikely(!sq_bitmap))
+		goto out;
+
+	ret = sysdev_driver_register(&cpu_sysdev_class, &sq_sysdev_driver);
+	if (unlikely(ret != 0))
+		goto out;
+
+	return 0;
+
+out:
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 
 	return ret;
 }
 
 static void __exit sq_api_exit(void)
 {
-	misc_deregister(&sq_dev);
-	remove_proc_entry("sq_mapping", NULL);
+	sysdev_driver_unregister(&cpu_sysdev_class, &sq_sysdev_driver);
+	kfree(sq_bitmap);
+	kmem_cache_destroy(sq_cache);
 }
 
 module_init(sq_api_init);
@@ -448,11 +402,7 @@ module_exit(sq_api_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>, M. R. Brown <mrbrown@0xd6.org>");
 MODULE_DESCRIPTION("Simple API for SH-4 integrated Store Queues");
 MODULE_LICENSE("GPL");
-MODULE_ALIAS_MISCDEV(STORE_QUEUE_MINOR);
 
 EXPORT_SYMBOL(sq_remap);
 EXPORT_SYMBOL(sq_unmap);
-EXPORT_SYMBOL(sq_clear);
-EXPORT_SYMBOL(sq_flush);
 EXPORT_SYMBOL(sq_flush_range);
-
diff --git a/include/asm-sh/cpu-sh4/sq.h b/include/asm-sh/cpu-sh4/sq.h
index 366b09166d3b..586d6491816a 100644
--- a/include/asm-sh/cpu-sh4/sq.h
+++ b/include/asm-sh/cpu-sh4/sq.h
@@ -17,7 +17,7 @@
  * Store queues range from e0000000-e3fffffc, allowing approx. 64MB to be
  * mapped to any physical address space. Since data is written (and aligned)
  * to 32-byte boundaries, we need to be sure that all allocations are aligned.
- */ 
+ */
 #define SQ_SIZE                 32
 #define SQ_ALIGN_MASK           (~(SQ_SIZE - 1))
 #define SQ_ALIGN(addr)          (((addr)+SQ_SIZE-1) & SQ_ALIGN_MASK)
@@ -26,23 +26,10 @@
 #define SQ_QACR1		(P4SEG_REG_BASE  + 0x3c)
 #define SQ_ADDRMAX              (P4SEG_STORE_QUE + 0x04000000)
 
-struct sq_mapping {
-	const char *name;
-
-	unsigned long sq_addr;
-	unsigned long addr;
-	unsigned int size;
-
-	struct list_head list;
-};
-
 /* arch/sh/kernel/cpu/sh4/sq.c */
-extern struct sq_mapping *sq_remap(unsigned long phys, unsigned int size, const char *name);
-extern void sq_unmap(struct sq_mapping *map);
-
-extern void sq_clear(unsigned long addr, unsigned int len);
-extern void sq_flush(void *addr);
-extern void sq_flush_range(unsigned long start, unsigned int len);
+unsigned long sq_remap(unsigned long phys, unsigned int size,
+		       const char *name, unsigned long flags);
+void sq_unmap(unsigned long vaddr);
+void sq_flush_range(unsigned long start, unsigned int len);
 
 #endif /* __ASM_CPU_SH4_SQ_H */
-
-- 
cgit v1.2.1


From 5283ecb5ccbdb90d49fce6488d3944bba63a591c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 15:59:17 +0900
Subject: sh: Add support for R7780RP and R7780MP boards.

This adds support for the Renesas SH7780 development boards,
R7780RP and R7780MP.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                          |    2 +
 arch/sh/Makefile                         |    1 +
 arch/sh/boards/renesas/r7780rp/Kconfig   |   14 +
 arch/sh/boards/renesas/r7780rp/Makefile  |    6 +
 arch/sh/boards/renesas/r7780rp/io.c      |  338 +++++++++
 arch/sh/boards/renesas/r7780rp/irq.c     |  123 ++++
 arch/sh/boards/renesas/r7780rp/led.c     |   45 ++
 arch/sh/boards/renesas/r7780rp/setup.c   |  164 +++++
 arch/sh/configs/r7780rp_defconfig        | 1099 ++++++++++++++++++++++++++++++
 arch/sh/drivers/dma/Kconfig              |    3 +-
 arch/sh/drivers/dma/dma-sh.c             |    4 -
 arch/sh/drivers/pci/Makefile             |    2 +
 arch/sh/drivers/pci/fixups-r7780rp.c     |   39 ++
 arch/sh/drivers/pci/ops-r7780rp.c        |   77 +++
 arch/sh/drivers/pci/pci-auto.c           |   44 +-
 arch/sh/drivers/pci/pci-sh7751.c         |    4 +-
 arch/sh/drivers/pci/pci-sh7780.c         |  341 +++++++++
 arch/sh/drivers/pci/pci-sh7780.h         |  168 +++++
 arch/sh/kernel/cpu/irq/intc2.c           |    6 +-
 arch/sh/tools/mach-types                 |    2 +
 include/asm-sh/cpu-sh4/addrspace.h       |    3 +
 include/asm-sh/cpu-sh4/dma-sh7780.h      |   39 ++
 include/asm-sh/cpu-sh4/dma.h             |   11 +-
 include/asm-sh/hs7751rvoip/hs7751rvoip.h |    2 -
 include/asm-sh/irq-sh7780.h              |    5 -
 include/asm-sh/irq.h                     |    2 +-
 include/asm-sh/landisk/iodata_landisk.h  |    2 -
 include/asm-sh/r7780rp/ide.h             |    8 +
 include/asm-sh/r7780rp/r7780rp.h         |  177 +++++
 include/asm-sh/rts7751r2d/rts7751r2d.h   |    2 -
 30 files changed, 2684 insertions(+), 49 deletions(-)
 create mode 100644 arch/sh/boards/renesas/r7780rp/Kconfig
 create mode 100644 arch/sh/boards/renesas/r7780rp/Makefile
 create mode 100644 arch/sh/boards/renesas/r7780rp/io.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/irq.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/led.c
 create mode 100644 arch/sh/boards/renesas/r7780rp/setup.c
 create mode 100644 arch/sh/configs/r7780rp_defconfig
 create mode 100644 arch/sh/drivers/pci/fixups-r7780rp.c
 create mode 100644 arch/sh/drivers/pci/ops-r7780rp.c
 create mode 100644 arch/sh/drivers/pci/pci-sh7780.c
 create mode 100644 arch/sh/drivers/pci/pci-sh7780.h
 create mode 100644 include/asm-sh/cpu-sh4/dma-sh7780.h
 create mode 100644 include/asm-sh/r7780rp/ide.h
 create mode 100644 include/asm-sh/r7780rp/r7780rp.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index a66b9c44075f..fe982f0eba23 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -404,6 +404,8 @@ source "arch/sh/boards/renesas/hs7751rvoip/Kconfig"
 
 source "arch/sh/boards/renesas/rts7751r2d/Kconfig"
 
+source "arch/sh/boards/renesas/r7780rp/Kconfig"
+
 config SH_PCLK_FREQ
 	int "Peripheral clock frequency (in Hz)"
 	default "50000000" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7780
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 65676c33822e..c143510d71fc 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -104,6 +104,7 @@ machdir-$(CONFIG_SH_HS7751RVOIP)		:= renesas/hs7751rvoip
 machdir-$(CONFIG_SH_RTS7751R2D)			:= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
+machdir-$(CONFIG_SH_R7780RP)			:= renesas/r7780rp
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_TITAN)			:= titan
diff --git a/arch/sh/boards/renesas/r7780rp/Kconfig b/arch/sh/boards/renesas/r7780rp/Kconfig
new file mode 100644
index 000000000000..c26d9813d239
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/Kconfig
@@ -0,0 +1,14 @@
+if SH_R7780RP
+
+menu "R7780RP options"
+
+config SH_R7780MP
+	bool "R7780MP board support"
+	default y
+	help
+	  Selecting this option will enable support for the mass-production
+	  version of the R7780RP. If in doubt, say Y.
+
+endmenu
+
+endif
diff --git a/arch/sh/boards/renesas/r7780rp/Makefile b/arch/sh/boards/renesas/r7780rp/Makefile
new file mode 100644
index 000000000000..f1776d027978
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the R7780RP-1 specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o irq.o
+obj-$(CONFIG_HEARTBEAT)	+= led.o
diff --git a/arch/sh/boards/renesas/r7780rp/io.c b/arch/sh/boards/renesas/r7780rp/io.c
new file mode 100644
index 000000000000..f73ca3f0f5a1
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/io.c
@@ -0,0 +1,338 @@
+/*
+ * linux/arch/sh/kernel/io_r7780rp.c
+ *
+ * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
+ * Based largely on io_se.c.
+ *
+ * I/O routine for Renesas Solutions Highlander R7780RP-1
+ *
+ * Initial version only to support LAN access; some
+ * placeholder code from io_r7780rp.c left in with the
+ * expectation of later SuperIO and PCMCIA access.
+ */
+
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/r7780rp/r7780rp.h>
+#include <asm/addrspace.h>
+#include <asm/io.h>
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include "../../../drivers/pci/pci-sh7780.h"
+
+/*
+ * The 7780 R7780RP-1 uses the built-in PCI controller (PCIC)
+ * of the 7780 processor, and has a SuperIO accessible via the PCI.
+ * The board also includes a PCMCIA controller on its memory bus,
+ * like the other Solution Engine boards.
+ */
+
+#define	SH7780_PCIIOBR_MASK	0xFFFC0000	/* IO Space Mask */
+#define PCIIOBR		(volatile long *)PCI_REG(SH7780_PCIIOBR)
+#define PCIMBR          (volatile long *)PCI_REG(SH7780_PCIMBR)
+#define PCI_IO_AREA	SH7780_PCI_IO_BASE
+#define PCI_MEM_AREA	SH7780_PCI_CONFIG_BASE
+
+#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7780_PCIIOBR_MASK))
+
+static inline void delay(void)
+{
+	ctrl_inw(0xa0000000);
+}
+
+static inline unsigned long port2adr(unsigned int port)
+{
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		if (port == 0x3f6)
+			return (PA_AREA5_IO + 0x80c);
+		else
+			return (PA_AREA5_IO + 0x1000 + ((port-0x1f0) << 1));
+	else
+		maybebadio((unsigned long)port);
+
+	return port;
+}
+
+static inline unsigned long port88796l(unsigned int port, int flag)
+{
+	unsigned long addr;
+
+	if (flag)
+		addr = PA_AX88796L + ((port - AX88796L_IO_BASE) << 1);
+	else
+		addr = PA_AX88796L + ((port - AX88796L_IO_BASE) << 1) + 0x1000;
+
+	return addr;
+}
+
+/* The 7780 R7780RP-1 seems to have everything hooked */
+/* up pretty normally (nothing on high-bytes only...) so this */
+/* shouldn't be needed */
+static inline int shifted_port(unsigned long port)
+{
+	/* For IDE registers, value is not shifted */
+	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
+		return 0;
+	else
+		return 1;
+}
+
+/* In case someone configures the kernel w/o PCI support: in that */
+/* scenario, don't ever bother to check for PCI-window addresses */
+
+/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
+#if defined(CONFIG_PCI)
+#define CHECK_SH7780_PCIIO(port) \
+  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7780_PCI_IO_SIZE)))
+#else
+#define CHECK_SH7780_PCIIO(port) (0)
+#endif
+
+#if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
+#define CHECK_AX88796L_PORT(port) \
+  ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
+#else
+#define CHECK_AX88796L_PORT(port) (0)
+#endif
+
+/*
+ * General outline: remap really low stuff [eventually] to SuperIO,
+ * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
+ * is mapped through the PCI IO window.  Stuff with high bits (PXSEG)
+ * should be way beyond the window, and is used  w/o translation for
+ * compatibility.
+ */
+u8 r7780rp_inb(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		return ctrl_inw(port88796l(port, 0)) & 0xff;
+	else if (PXSEG(port))
+		return ctrl_inb(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inb(PCI_IOMAP(port));
+
+	return ctrl_inw(port2adr(port)) & 0xff;
+}
+
+u8 r7780rp_inb_p(unsigned long port)
+{
+	u8 v;
+
+	if (CHECK_AX88796L_PORT(port))
+		v = ctrl_inw(port88796l(port, 0)) & 0xff;
+	else if (PXSEG(port))
+		v = ctrl_inb(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		v = ctrl_inb(PCI_IOMAP(port));
+	else
+		v = ctrl_inw(port2adr(port)) & 0xff;
+
+	delay();
+
+	return v;
+}
+
+u16 r7780rp_inw(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		return ctrl_inw(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inw(PCI_IOMAP(port));
+	else
+		maybebadio(port);
+
+	return 0;
+}
+
+u32 r7780rp_inl(unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		return ctrl_inl(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return ctrl_inl(PCI_IOMAP(port));
+	else
+		maybebadio(port);
+
+	return 0;
+}
+
+void r7780rp_outb(u8 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		ctrl_outw(value, port88796l(port, 0));
+	else if (PXSEG(port))
+		ctrl_outb(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outb(value, PCI_IOMAP(port));
+	else
+		ctrl_outw(value, port2adr(port));
+}
+
+void r7780rp_outb_p(u8 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		ctrl_outw(value, port88796l(port, 0));
+	else if (PXSEG(port))
+		ctrl_outb(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outb(value, PCI_IOMAP(port));
+	else
+		ctrl_outw(value, port2adr(port));
+
+	delay();
+}
+
+void r7780rp_outw(u16 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		ctrl_outw(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outw(value, PCI_IOMAP(port));
+	else
+		maybebadio(port);
+}
+
+void r7780rp_outl(u32 value, unsigned long port)
+{
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (PXSEG(port))
+		ctrl_outl(value, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		ctrl_outl(value, PCI_IOMAP(port));
+	else
+		maybebadio(port);
+}
+
+void r7780rp_insb(unsigned long port, void *dst, unsigned long count)
+{
+	volatile u16 *p;
+	u8 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port)) {
+		p = (volatile u16 *)port88796l(port, 0);
+		while (count--)
+			*buf++ = *p & 0xff;
+	} else if (PXSEG(port)) {
+		while (count--)
+			*buf++ = *(volatile u8 *)port;
+	} else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+
+		while (count--)
+			*buf++ = *bp;
+	} else {
+		p = (volatile u16 *)port2adr(port);
+		while (count--)
+			*buf++ = *p & 0xff;
+	}
+}
+
+void r7780rp_insw(unsigned long port, void *dst, unsigned long count)
+{
+	volatile u16 *p;
+	u16 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port))
+		p = (volatile u16 *)port88796l(port, 1);
+	else if (PXSEG(port))
+		p = (volatile u16 *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		p = (volatile u16 *)PCI_IOMAP(port);
+	else
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*buf++ = *p;
+}
+
+void r7780rp_insl(unsigned long port, void *dst, unsigned long count)
+{
+	u32 *buf = dst;
+
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+
+		while (count--)
+			*buf++ = *p;
+	} else
+		maybebadio(port);
+}
+
+void r7780rp_outsb(unsigned long port, const void *src, unsigned long count)
+{
+	volatile u16 *p;
+	const u8 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port)) {
+		p = (volatile u16 *)port88796l(port, 0);
+		while (count--)
+			*p = *buf++;
+	} else if (PXSEG(port))
+		while (count--)
+			ctrl_outb(*buf++, port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+
+		while (count--)
+			*bp = *buf++;
+	} else {
+		p = (volatile u16 *)port2adr(port);
+		while (count--)
+			*p = *buf++;
+	}
+}
+
+void r7780rp_outsw(unsigned long port, const void *src, unsigned long count)
+{
+	volatile u16 *p;
+	const u16 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port))
+		p = (volatile u16 *)port88796l(port, 1);
+	else if (PXSEG(port))
+		p = (volatile u16 *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		p = (volatile u16 *)PCI_IOMAP(port);
+	else
+		p = (volatile u16 *)port2adr(port);
+
+	while (count--)
+		*p = *buf++;
+}
+
+void r7780rp_outsl(unsigned long port, const void *src, unsigned long count)
+{
+	const u32 *buf = src;
+
+	if (CHECK_AX88796L_PORT(port))
+		maybebadio(port);
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+
+		while (count--)
+			*p = *buf++;
+	} else
+		maybebadio(port);
+}
+
+void __iomem *r7780rp_ioport_map(unsigned long port, unsigned int size)
+{
+	if (CHECK_AX88796L_PORT(port))
+		return (void __iomem *)port88796l(port, size > 1);
+	else if (PXSEG(port))
+		return (void __iomem *)port;
+	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
+		return (void __iomem *)PCI_IOMAP(port);
+
+	return (void __iomem *)port2adr(port);
+}
diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c
new file mode 100644
index 000000000000..c7b9fe6e640d
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/irq.c
@@ -0,0 +1,123 @@
+/*
+ * linux/arch/sh/boards/renesas/r7780rp/irq.c
+ *
+ * Copyright (C) 2000  Kazumoto Kojima
+ *
+ * Renesas Solutions Highlander R7780RP-1 Support.
+ *
+ * Modified for R7780RP-1 by
+ * Atom Create Engineering Co., Ltd. 2002.
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/r7780rp/r7780rp.h>
+
+#ifdef CONFIG_SH_R7780MP
+static int mask_pos[] = {12, 11, 9, 14, 15, 8, 13, 6, 5, 4, 3, 2, 0, 0, 1, 0};
+#else
+static int mask_pos[] = {15, 14, 13, 12, 11, 10, 9, 8, 7, 5, 6, 4, 0, 1, 2, 0};
+#endif
+
+static void enable_r7780rp_irq(unsigned int irq);
+static void disable_r7780rp_irq(unsigned int irq);
+
+/* shutdown is same as "disable" */
+#define shutdown_r7780rp_irq disable_r7780rp_irq
+
+static void ack_r7780rp_irq(unsigned int irq);
+static void end_r7780rp_irq(unsigned int irq);
+
+static unsigned int startup_r7780rp_irq(unsigned int irq)
+{
+	enable_r7780rp_irq(irq);
+	return 0; /* never anything pending */
+}
+
+static void disable_r7780rp_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val;
+	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
+
+	/* Set the priority in IPR to 0 */
+	local_irq_save(flags);
+	val = ctrl_inw(IRLCNTR1);
+	val &= mask;
+	ctrl_outw(val, IRLCNTR1);
+	local_irq_restore(flags);
+}
+
+static void enable_r7780rp_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val;
+	unsigned short value = (0x0001 << mask_pos[irq]);
+
+	/* Set priority in IPR back to original value */
+	local_irq_save(flags);
+	val = ctrl_inw(IRLCNTR1);
+	val |= value;
+	ctrl_outw(val, IRLCNTR1);
+	local_irq_restore(flags);
+}
+
+static void ack_r7780rp_irq(unsigned int irq)
+{
+	disable_r7780rp_irq(irq);
+}
+
+static void end_r7780rp_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
+		enable_r7780rp_irq(irq);
+}
+
+static struct hw_interrupt_type r7780rp_irq_type = {
+	.typename = "R7780RP-IRQ",
+	.startup = startup_r7780rp_irq,
+	.shutdown = shutdown_r7780rp_irq,
+	.enable = enable_r7780rp_irq,
+	.disable = disable_r7780rp_irq,
+	.ack = ack_r7780rp_irq,
+	.end = end_r7780rp_irq,
+};
+
+static void make_r7780rp_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &r7780rp_irq_type;
+	disable_r7780rp_irq(irq);
+}
+
+/*
+ * Initialize IRQ setting
+ */
+void __init init_r7780rp_IRQ(void)
+{
+	int i;
+
+	/* IRL0=PCI Slot #A
+	 * IRL1=PCI Slot #B
+	 * IRL2=PCI Slot #C
+	 * IRL3=PCI Slot #D
+	 * IRL4=CF Card
+	 * IRL5=CF Card Insert
+	 * IRL6=M66596
+	 * IRL7=SD Card
+	 * IRL8=Touch Panel
+	 * IRL9=SCI
+	 * IRL10=Serial
+	 * IRL11=Extention #A
+	 * IRL11=Extention #B
+	 * IRL12=Debug LAN
+	 * IRL13=Push Switch
+	 * IRL14=ZiggBee IO
+	 */
+
+	for (i=0; i<15; i++)
+		make_r7780rp_irq(i);
+}
diff --git a/arch/sh/boards/renesas/r7780rp/led.c b/arch/sh/boards/renesas/r7780rp/led.c
new file mode 100644
index 000000000000..9f02766b6f53
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/led.c
@@ -0,0 +1,45 @@
+/*
+ * Copyright (C) Atom Create Engineering Co., Ltd.
+ *
+ * May be copied or modified under the terms of GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * This file contains Renesas Solutions HIGHLANDER R7780RP-1 specific LED code.
+ */
+
+#include <linux/config.h>
+#include <linux/sched.h>
+#include <asm/io.h>
+#include <asm/r7780rp/r7780rp.h>
+
+/* Cycle the LED's in the clasic Knightriger/Sun pattern */
+void heartbeat_r7780rp(void)
+{
+	static unsigned int cnt = 0, period = 0;
+	volatile unsigned short *p = (volatile unsigned short *)PA_OBLED;
+	static unsigned bit = 0, up = 1;
+	unsigned bit_pos[] = {2, 1, 0, 3, 6, 5, 4, 7};
+
+	cnt += 1;
+	if (cnt < period)
+		return;
+
+	cnt = 0;
+
+	/* Go through the points (roughly!):
+	 * f(0)=10, f(1)=16, f(2)=20, f(5)=35, f(int)->110
+	 */
+	period = 110 - ((300 << FSHIFT)/((avenrun[0]/5) + (3<<FSHIFT)));
+
+	*p = 1 << bit_pos[bit];
+	if (up)
+		if (bit == 7) {
+			bit--;
+			up = 0;
+		} else
+			bit++;
+	else if (bit == 0)
+		up = 1;
+	else
+		bit--;
+}
diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c
new file mode 100644
index 000000000000..0e4d5e1b541c
--- /dev/null
+++ b/arch/sh/boards/renesas/r7780rp/setup.c
@@ -0,0 +1,164 @@
+/*
+ * arch/sh/boards/renesas/r7780rp/setup.c
+ *
+ * Copyright (C) 2002 Atom Create Engineering Co., Ltd.
+ * Copyright (C) 2005, 2006 Paul Mundt
+ *
+ * Renesas Solutions Highlander R7780RP-1 Support.
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <asm/machvec.h>
+#include <asm/r7780rp/r7780rp.h>
+#include <asm/clock.h>
+#include <asm/io.h>
+
+extern void heartbeat_r7780rp(void);
+extern void init_r7780rp_IRQ(void);
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_r7780rp __initmv = {
+	.mv_nr_irqs		= 109,
+
+	.mv_inb			= r7780rp_inb,
+	.mv_inw			= r7780rp_inw,
+	.mv_inl			= r7780rp_inl,
+	.mv_outb		= r7780rp_outb,
+	.mv_outw		= r7780rp_outw,
+	.mv_outl		= r7780rp_outl,
+
+	.mv_inb_p		= r7780rp_inb_p,
+	.mv_inw_p		= r7780rp_inw,
+	.mv_inl_p		= r7780rp_inl,
+	.mv_outb_p		= r7780rp_outb_p,
+	.mv_outw_p		= r7780rp_outw,
+	.mv_outl_p		= r7780rp_outl,
+
+	.mv_insb		= r7780rp_insb,
+	.mv_insw		= r7780rp_insw,
+	.mv_insl		= r7780rp_insl,
+	.mv_outsb		= r7780rp_outsb,
+	.mv_outsw		= r7780rp_outsw,
+	.mv_outsl		= r7780rp_outsl,
+
+	.mv_ioport_map		= r7780rp_ioport_map,
+	.mv_init_irq		= init_r7780rp_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_r7780rp,
+#endif
+};
+ALIAS_MV(r7780rp)
+
+static struct resource m66596_usb_host_resources[] = {
+	[0] = {
+		.start	= 0xa4800000,
+		.end	= 0xa4ffffff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= 6,		/* irq number */
+		.end	= 6,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device m66596_usb_host_device = {
+	.name		= "m66596-hcd",
+	.id		= 0,
+	.dev = {
+		.dma_mask		= NULL,		/* don't use dma */
+		.coherent_dma_mask	= 0xffffffff,
+	},
+	.num_resources	= ARRAY_SIZE(m66596_usb_host_resources),
+	.resource	= m66596_usb_host_resources,
+};
+
+static struct platform_device *r7780rp_devices[] __initdata = {
+	&m66596_usb_host_device,
+};
+
+static int __init r7780rp_devices_setup(void)
+{
+	return platform_add_devices(r7780rp_devices,
+				    ARRAY_SIZE(r7780rp_devices));
+}
+__initcall(r7780rp_devices_setup);
+
+/*
+ * Platform specific clocks
+ */
+static void ivdr_clk_enable(struct clk *clk)
+{
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) | (1 << 8), PA_IVDRCTL);
+}
+
+static void ivdr_clk_disable(struct clk *clk)
+{
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) & ~(1 << 8), PA_IVDRCTL);
+}
+
+static struct clk_ops ivdr_clk_ops = {
+	.enable		= ivdr_clk_enable,
+	.disable	= ivdr_clk_disable,
+};
+
+static struct clk ivdr_clk = {
+	.name		= "ivdr_clk",
+	.ops		= &ivdr_clk_ops,
+};
+
+static struct clk *r7780rp_clocks[] = {
+	&ivdr_clk,
+};
+
+const char *get_system_type(void)
+{
+	return "Highlander R7780RP-1";
+}
+
+static void r7780rp_power_off(void)
+{
+#ifdef CONFIG_SH_R7780MP
+	ctrl_outw(0x0001, PA_POFF);
+#endif
+}
+
+/*
+ * Initialize the board
+ */
+void __init platform_setup(void)
+{
+	u16 ver = ctrl_inw(PA_VERREG);
+	int i;
+
+	printk(KERN_INFO "Renesas Solutions Highlander R7780RP-1 support.\n");
+
+	printk(KERN_INFO "Board version: %d (revision %d), "
+			 "FPGA version: %d (revision %d)\n",
+			 (ver >> 12) & 0xf, (ver >> 8) & 0xf,
+			 (ver >>  4) & 0xf, ver & 0xf);
+
+	/*
+	 * Enable the important clocks right away..
+	 */
+	for (i = 0; i < ARRAY_SIZE(r7780rp_clocks); i++) {
+		struct clk *clk = r7780rp_clocks[i];
+
+		clk_register(clk);
+		clk_enable(clk);
+	}
+
+	ctrl_outw(0x0000, PA_OBLED);	/* Clear LED. */
+#ifndef CONFIG_SH_R7780MP
+	ctrl_outw(0x0001, PA_SDPOW);	/* SD Power ON */
+#endif
+	ctrl_outw(ctrl_inw(PA_IVDRCTL) | 0x0100, PA_IVDRCTL);	/* Si13112 */
+
+	pm_power_off = r7780rp_power_off;
+}
diff --git a/arch/sh/configs/r7780rp_defconfig b/arch/sh/configs/r7780rp_defconfig
new file mode 100644
index 000000000000..d597fc571549
--- /dev/null
+++ b/arch/sh/configs/r7780rp_defconfig
@@ -0,0 +1,1099 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.15-sh
+# Sat Jan  7 19:47:53 2006
+#
+CONFIG_SUPERH=y
+CONFIG_UID16=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_CLEAN_COMPILE=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_LOCK_KERNEL=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SWAP=y
+CONFIG_SYSVIPC=y
+# CONFIG_POSIX_MQUEUE is not set
+CONFIG_BSD_PROCESS_ACCT=y
+# CONFIG_BSD_PROCESS_ACCT_V3 is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+CONFIG_HOTPLUG=y
+CONFIG_KOBJECT_UEVENT=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_ALL is not set
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+CONFIG_SHMEM=y
+CONFIG_CC_ALIGN_FUNCTIONS=0
+CONFIG_CC_ALIGN_LABELS=0
+CONFIG_CC_ALIGN_LOOPS=0
+CONFIG_CC_ALIGN_JUMPS=0
+# CONFIG_TINY_SHMEM is not set
+CONFIG_BASE_SMALL=0
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+CONFIG_OBSOLETE_MODPARM=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R77703DRP is not set
+CONFIG_SH_R7780RP=y
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+CONFIG_CPU_SH4A=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+CONFIG_CPU_SUBTYPE_SH7780=y
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_32BIT=y
+CONFIG_HUGETLB_PAGE_SIZE_64K=y
+# CONFIG_HUGETLB_PAGE_SIZE_1MB is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+CONFIG_MEMORY_START=0x08000000
+CONFIG_MEMORY_SIZE=0x08000000
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+CONFIG_SH_FPU=y
+CONFIG_SH_STORE_QUEUES=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ_BOOL=y
+CONFIG_SH_PCLK_FREQ=32000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+CONFIG_SH_DMA=y
+CONFIG_NR_ONCHIP_DMA_CHANNELS=6
+# CONFIG_NR_DMA_CHANNELS_BOOL is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+CONFIG_PREEMPT=y
+# CONFIG_SMP is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_INTC2_IRQ=y
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="mem=128M console=ttySC0,115200 root=/dev/hda1"
+
+#
+# Bus options
+#
+CONFIG_PCI=y
+CONFIG_SH_PCIDMA_NONCOHERENT=y
+CONFIG_PCI_AUTO=y
+CONFIG_PCI_AUTO_UPDATE_RESOURCES=y
+CONFIG_PCI_LEGACY_PROC=y
+# CONFIG_PCI_DEBUG is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_HOTPLUG_PCI_FAKE is not set
+# CONFIG_HOTPLUG_PCI_CPCI is not set
+# CONFIG_HOTPLUG_PCI_SHPC is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_ASK_IP_FIB_HASH=y
+# CONFIG_IP_FIB_TRIE is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_MULTIPLE_TABLES is not set
+# CONFIG_IP_ROUTE_MULTIPATH is not set
+# CONFIG_IP_ROUTE_VERBOSE is not set
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+# CONFIG_ATM is not set
+CONFIG_BRIDGE=m
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=m
+# CONFIG_DEBUG_DRIVER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+# CONFIG_MTD is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_CPQ_DA is not set
+# CONFIG_BLK_CPQ_CISS_DA is not set
+# CONFIG_BLK_DEV_DAC960 is not set
+# CONFIG_BLK_DEV_UMEM is not set
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_SX8 is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+CONFIG_IDE=m
+CONFIG_IDE_MAX_HWIFS=4
+CONFIG_BLK_DEV_IDE=m
+
+#
+# Please see Documentation/ide.txt for help/info on IDE drives
+#
+CONFIG_BLK_DEV_IDE_SATA=y
+CONFIG_BLK_DEV_IDEDISK=m
+CONFIG_IDEDISK_MULTI_MODE=y
+# CONFIG_BLK_DEV_IDECD is not set
+# CONFIG_BLK_DEV_IDETAPE is not set
+# CONFIG_BLK_DEV_IDEFLOPPY is not set
+CONFIG_BLK_DEV_IDESCSI=m
+# CONFIG_IDE_TASK_IOCTL is not set
+
+#
+# IDE chipset support/bugfixes
+#
+CONFIG_IDE_GENERIC=m
+CONFIG_BLK_DEV_IDEPCI=y
+CONFIG_IDEPCI_SHARE_IRQ=y
+# CONFIG_BLK_DEV_OFFBOARD is not set
+CONFIG_BLK_DEV_GENERIC=m
+# CONFIG_BLK_DEV_OPTI621 is not set
+CONFIG_BLK_DEV_IDEDMA_PCI=y
+# CONFIG_BLK_DEV_IDEDMA_FORCED is not set
+CONFIG_IDEDMA_PCI_AUTO=y
+# CONFIG_IDEDMA_ONLYDISK is not set
+CONFIG_BLK_DEV_AEC62XX=m
+# CONFIG_BLK_DEV_ALI15X3 is not set
+# CONFIG_BLK_DEV_AMD74XX is not set
+# CONFIG_BLK_DEV_CMD64X is not set
+# CONFIG_BLK_DEV_TRIFLEX is not set
+# CONFIG_BLK_DEV_CY82C693 is not set
+# CONFIG_BLK_DEV_CS5520 is not set
+# CONFIG_BLK_DEV_CS5530 is not set
+# CONFIG_BLK_DEV_HPT34X is not set
+# CONFIG_BLK_DEV_HPT366 is not set
+# CONFIG_BLK_DEV_SC1200 is not set
+# CONFIG_BLK_DEV_PIIX is not set
+# CONFIG_BLK_DEV_IT821X is not set
+# CONFIG_BLK_DEV_NS87415 is not set
+# CONFIG_BLK_DEV_PDC202XX_OLD is not set
+CONFIG_BLK_DEV_PDC202XX_NEW=m
+# CONFIG_PDC202XX_FORCE is not set
+# CONFIG_BLK_DEV_SVWKS is not set
+CONFIG_BLK_DEV_SIIMAGE=m
+# CONFIG_BLK_DEV_SLC90E66 is not set
+# CONFIG_BLK_DEV_TRM290 is not set
+# CONFIG_BLK_DEV_VIA82CXXX is not set
+CONFIG_IDE_SH=y
+# CONFIG_IDE_ARM is not set
+CONFIG_BLK_DEV_IDEDMA=y
+# CONFIG_IDEDMA_IVB is not set
+CONFIG_IDEDMA_AUTO=y
+# CONFIG_BLK_DEV_HD is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+CONFIG_SCSI=m
+CONFIG_SCSI_PROC_FS=y
+
+#
+# SCSI support type (disk, tape, CD-ROM)
+#
+CONFIG_BLK_DEV_SD=m
+# CONFIG_CHR_DEV_ST is not set
+# CONFIG_CHR_DEV_OSST is not set
+# CONFIG_BLK_DEV_SR is not set
+CONFIG_CHR_DEV_SG=m
+# CONFIG_CHR_DEV_SCH is not set
+
+#
+# Some SCSI devices (e.g. CD jukebox) support multiple LUNs
+#
+# CONFIG_SCSI_MULTI_LUN is not set
+# CONFIG_SCSI_CONSTANTS is not set
+# CONFIG_SCSI_LOGGING is not set
+
+#
+# SCSI Transport Attributes
+#
+# CONFIG_SCSI_SPI_ATTRS is not set
+# CONFIG_SCSI_FC_ATTRS is not set
+# CONFIG_SCSI_ISCSI_ATTRS is not set
+# CONFIG_SCSI_SAS_ATTRS is not set
+
+#
+# SCSI low-level drivers
+#
+# CONFIG_ISCSI_TCP is not set
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_3W_9XXX is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_MEGARAID_NEWGEN is not set
+# CONFIG_MEGARAID_LEGACY is not set
+# CONFIG_MEGARAID_SAS is not set
+# CONFIG_SCSI_SATA is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_IPS is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_IPR is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
+CONFIG_SCSI_QLA2XXX=m
+# CONFIG_SCSI_QLA21XX is not set
+# CONFIG_SCSI_QLA22XX is not set
+# CONFIG_SCSI_QLA2300 is not set
+# CONFIG_SCSI_QLA2322 is not set
+# CONFIG_SCSI_QLA6312 is not set
+# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_LPFC is not set
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+# CONFIG_FUSION_SPI is not set
+# CONFIG_FUSION_FC is not set
+# CONFIG_FUSION_SAS is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+# CONFIG_IEEE1394 is not set
+
+#
+# I2O device support
+#
+# CONFIG_I2O is not set
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# ARCnet devices
+#
+# CONFIG_ARCNET is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+# CONFIG_HAPPYMEAL is not set
+# CONFIG_SUNGEM is not set
+# CONFIG_CASSINI is not set
+# CONFIG_NET_VENDOR_3COM is not set
+# CONFIG_SMC91X is not set
+
+#
+# Tulip family network device support
+#
+# CONFIG_NET_TULIP is not set
+# CONFIG_HP100 is not set
+CONFIG_NE2000=y
+CONFIG_NET_PCI=y
+CONFIG_PCNET32=m
+# CONFIG_AMD8111_ETH is not set
+# CONFIG_ADAPTEC_STARFIRE is not set
+# CONFIG_B44 is not set
+# CONFIG_FORCEDETH is not set
+# CONFIG_DGRS is not set
+# CONFIG_EEPRO100 is not set
+# CONFIG_E100 is not set
+# CONFIG_FEALNX is not set
+# CONFIG_NATSEMI is not set
+# CONFIG_NE2K_PCI is not set
+CONFIG_8139CP=m
+CONFIG_8139TOO=m
+# CONFIG_8139TOO_PIO is not set
+# CONFIG_8139TOO_TUNE_TWISTER is not set
+CONFIG_8139TOO_8129=y
+# CONFIG_8139_OLD_RX_RESET is not set
+# CONFIG_SIS900 is not set
+# CONFIG_EPIC100 is not set
+# CONFIG_SUNDANCE is not set
+# CONFIG_TLAN is not set
+CONFIG_VIA_RHINE=m
+CONFIG_VIA_RHINE_MMIO=y
+
+#
+# Ethernet (1000 Mbit)
+#
+# CONFIG_ACENIC is not set
+# CONFIG_DL2K is not set
+CONFIG_E1000=m
+# CONFIG_E1000_NAPI is not set
+# CONFIG_NS83820 is not set
+# CONFIG_HAMACHI is not set
+# CONFIG_YELLOWFIN is not set
+CONFIG_R8169=y
+# CONFIG_R8169_NAPI is not set
+# CONFIG_SIS190 is not set
+# CONFIG_SKGE is not set
+# CONFIG_SK98LIN is not set
+# CONFIG_VIA_VELOCITY is not set
+# CONFIG_TIGON3 is not set
+# CONFIG_BNX2 is not set
+
+#
+# Ethernet (10000 Mbit)
+#
+# CONFIG_CHELSIO_T1 is not set
+# CONFIG_IXGB is not set
+# CONFIG_S2IO is not set
+
+#
+# Token Ring devices
+#
+# CONFIG_TR is not set
+
+#
+# Wireless LAN (non-hamradio)
+#
+CONFIG_NET_RADIO=y
+
+#
+# Obsolete Wireless cards support (pre-802.11)
+#
+# CONFIG_STRIP is not set
+
+#
+# Wireless 802.11b ISA/PCI cards support
+#
+CONFIG_HERMES=m
+# CONFIG_PLX_HERMES is not set
+# CONFIG_TMD_HERMES is not set
+# CONFIG_NORTEL_HERMES is not set
+# CONFIG_PCI_HERMES is not set
+# CONFIG_ATMEL is not set
+
+#
+# Prism GT/Duette 802.11(a/b/g) PCI/Cardbus support
+#
+CONFIG_PRISM54=m
+# CONFIG_HOSTAP is not set
+CONFIG_NET_WIRELESS=y
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_FDDI is not set
+# CONFIG_HIPPI is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NET_FC is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+CONFIG_INPUT_MOUSEDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
+CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+CONFIG_INPUT_KEYBOARD=y
+CONFIG_KEYBOARD_ATKBD=y
+# CONFIG_KEYBOARD_SUNKBD is not set
+# CONFIG_KEYBOARD_LKKBD is not set
+# CONFIG_KEYBOARD_XTKBD is not set
+# CONFIG_KEYBOARD_NEWTON is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+CONFIG_SERIO=y
+# CONFIG_SERIO_I8042 is not set
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_SERIO_PCIPS2 is not set
+CONFIG_SERIO_LIBPS2=y
+# CONFIG_SERIO_RAW is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+# CONFIG_APPLICOM is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_DRM is not set
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia Capabilities Port drivers
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+CONFIG_SOUND=m
+
+#
+# Advanced Linux Sound Architecture
+#
+# CONFIG_SND is not set
+
+#
+# Open Sound System
+#
+CONFIG_SOUND_PRIME=m
+# CONFIG_OBSOLETE_OSS_DRIVER is not set
+# CONFIG_SOUND_FUSION is not set
+# CONFIG_SOUND_ICH is not set
+# CONFIG_SOUND_TRIDENT is not set
+# CONFIG_SOUND_MSNDCLAS is not set
+# CONFIG_SOUND_MSNDPIN is not set
+
+#
+# USB support
+#
+CONFIG_USB_ARCH_HAS_HCD=y
+CONFIG_USB_ARCH_HAS_OHCI=y
+# CONFIG_USB is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# InfiniBand support
+#
+# CONFIG_INFINIBAND is not set
+
+#
+# SN Devices
+#
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT2_FS_XIP is not set
+CONFIG_EXT3_FS=y
+CONFIG_EXT3_FS_XATTR=y
+# CONFIG_EXT3_FS_POSIX_ACL is not set
+# CONFIG_EXT3_FS_SECURITY is not set
+CONFIG_JBD=y
+# CONFIG_JBD_DEBUG is not set
+CONFIG_FS_MBCACHE=y
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+CONFIG_FS_POSIX_ACL=y
+# CONFIG_XFS_FS is not set
+CONFIG_MINIX_FS=y
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+# CONFIG_QUOTA is not set
+CONFIG_DNOTIFY=y
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+CONFIG_FAT_FS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_FAT_DEFAULT_CODEPAGE=437
+CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
+CONFIG_NTFS_FS=y
+# CONFIG_NTFS_DEBUG is not set
+CONFIG_NTFS_RW=y
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+CONFIG_HUGETLBFS=y
+CONFIG_HUGETLB_PAGE=y
+CONFIG_RAMFS=y
+# CONFIG_RELAYFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+CONFIG_NFS_V4=y
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+CONFIG_NFSD_V3=y
+# CONFIG_NFSD_V3_ACL is not set
+CONFIG_NFSD_V4=y
+CONFIG_NFSD_TCP=y
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+CONFIG_SUNRPC_GSS=y
+CONFIG_RPCSEC_GSS_KRB5=y
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+CONFIG_NLS=y
+CONFIG_NLS_DEFAULT="iso8859-1"
+CONFIG_NLS_CODEPAGE_437=y
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+CONFIG_NLS_CODEPAGE_932=y
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+CONFIG_NLS_ISO8859_1=y
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_DEBUG_KERNEL=y
+# CONFIG_MAGIC_SYSRQ is not set
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_DETECT_SOFTLOCKUP=y
+# CONFIG_SCHEDSTATS is not set
+# CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_PREEMPT=y
+CONFIG_DEBUG_SPINLOCK=y
+# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
+# CONFIG_DEBUG_KOBJECT is not set
+# CONFIG_DEBUG_INFO is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_DEBUG_VM is not set
+CONFIG_FRAME_POINTER=y
+# CONFIG_RCU_TORTURE_TEST is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+CONFIG_CRYPTO=y
+CONFIG_CRYPTO_HMAC=y
+# CONFIG_CRYPTO_NULL is not set
+# CONFIG_CRYPTO_MD4 is not set
+CONFIG_CRYPTO_MD5=y
+# CONFIG_CRYPTO_SHA1 is not set
+# CONFIG_CRYPTO_SHA256 is not set
+# CONFIG_CRYPTO_SHA512 is not set
+# CONFIG_CRYPTO_WP512 is not set
+# CONFIG_CRYPTO_TGR192 is not set
+CONFIG_CRYPTO_DES=y
+# CONFIG_CRYPTO_BLOWFISH is not set
+# CONFIG_CRYPTO_TWOFISH is not set
+# CONFIG_CRYPTO_SERPENT is not set
+# CONFIG_CRYPTO_AES is not set
+# CONFIG_CRYPTO_CAST5 is not set
+# CONFIG_CRYPTO_CAST6 is not set
+# CONFIG_CRYPTO_TEA is not set
+# CONFIG_CRYPTO_ARC4 is not set
+# CONFIG_CRYPTO_KHAZAD is not set
+# CONFIG_CRYPTO_ANUBIS is not set
+# CONFIG_CRYPTO_DEFLATE is not set
+# CONFIG_CRYPTO_MICHAEL_MIC is not set
+# CONFIG_CRYPTO_CRC32C is not set
+# CONFIG_CRYPTO_TEST is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
diff --git a/arch/sh/drivers/dma/Kconfig b/arch/sh/drivers/dma/Kconfig
index 0f15216cd39d..defc13c37d48 100644
--- a/arch/sh/drivers/dma/Kconfig
+++ b/arch/sh/drivers/dma/Kconfig
@@ -11,6 +11,8 @@ config SH_DMA
 config NR_ONCHIP_DMA_CHANNELS
 	depends on SH_DMA
 	int "Number of on-chip DMAC channels"
+	default "8" if CPU_SUBTYPE_SH7750R || CPU_SUBTYPE_SH7751R
+	default "12" if CPU_SUBTYPE_SH7780
 	default "4"
 	help
 	  This allows you to specify the number of channels that the on-chip
@@ -52,4 +54,3 @@ config DMA_PAGE_OPS_CHANNEL
 	  are dual-address capable.
 
 endmenu
-
diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index e028a2d2a4ea..4428ee809651 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -11,14 +11,10 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-
 #include <linux/init.h>
-#include <linux/irq.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
 #include <asm/dreamcast/dma.h>
-#include <asm/signal.h>
-#include <asm/irq.h>
 #include <asm/dma.h>
 #include <asm/io.h>
 #include "dma-sh.h"
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 86d6d0660d4c..3d8078f1c05f 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_PCI_AUTO)			+= pci-auto.o
 
 obj-$(CONFIG_CPU_SUBTYPE_ST40STB1)	+= pci-st40.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o 
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o
 
 obj-$(CONFIG_SH_DREAMCAST)		+= ops-dreamcast.o fixups-dreamcast.o \
 					   dma-dreamcast.o
@@ -14,4 +15,5 @@ obj-$(CONFIG_SH_SECUREEDGE5410)		+= ops-snapgear.o
 obj-$(CONFIG_SH_BIGSUR)			+= ops-bigsur.o
 obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
+obj-$(CONFIG_SH_R7780RP)		+= ops-r7780rp.o fixups-r7780rp.o
 obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
diff --git a/arch/sh/drivers/pci/fixups-r7780rp.c b/arch/sh/drivers/pci/fixups-r7780rp.c
new file mode 100644
index 000000000000..b656b562ec99
--- /dev/null
+++ b/arch/sh/drivers/pci/fixups-r7780rp.c
@@ -0,0 +1,39 @@
+/*
+ * arch/sh/drivers/pci/fixups-r7780rp.c
+ *
+ * Highlander R7780RP-1 PCI fixups
+ *
+ * Copyright (C) 2003  Lineo uSolutions, Inc.
+ * Copyright (C) 2004  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include "pci-sh7780.h"
+#include <asm/io.h>
+
+int pci_fixup_pcic(void)
+{
+	outl(0x000043ff, PCI_REG(SH7780_PCIIMR));
+	outl(0x0000380f, PCI_REG(SH7780_PCIAINTM));
+
+	outl(0xfbb00047, PCI_REG(SH7780_PCICMD));
+	outl(0x00000000, PCI_REG(SH7780_PCIIBAR));
+
+	outl(0x00011912, PCI_REG(SH7780_PCISVID));
+	outl(0x08000000, PCI_REG(SH7780_PCICSCR0));
+	outl(0x0000001b, PCI_REG(SH7780_PCICSAR0));
+	outl(0xfd000000, PCI_REG(SH7780_PCICSCR1));
+	outl(0x0000000f, PCI_REG(SH7780_PCICSAR1));
+
+	outl(0xfd000000, PCI_REG(SH7780_PCIMBR0));
+	outl(0x00fc0000, PCI_REG(SH7780_PCIMBMR0));
+
+	/* Set IOBR for windows containing area specified in pci.h */
+	outl((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE-1)), PCI_REG(SH7780_PCIIOBR));
+	outl(((SH7780_PCI_IO_SIZE-1) & (7<<18)), PCI_REG(SH7780_PCIIOBMR));
+
+	return 0;
+}
+
diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c
new file mode 100644
index 000000000000..3254c4e917a9
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-r7780rp.c
@@ -0,0 +1,77 @@
+/*
+ * Author:  Ian DaSilva (idasilva@mvista.com)
+ *
+ * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ *
+ * PCI initialization for the Renesas SH7780 Highlander R7780RP-1 board
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include <linux/module.h>
+
+#include <asm/io.h>
+#include "pci-sh7780.h"
+#include <asm/r7780rp/r7780rp.h>
+
+int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+{
+        switch (slot) {
+	case 0: return IRQ_PCISLOT1;		/* PCI Interrupt #1 */
+	case 1: return IRQ_PCISLOT2;		/* PCI Interrupt #2 */
+	case 2: return IRQ_PCISLOT3;		/* PCI Interrupt #3 */
+	case 3: return IRQ_PCISLOT4;		/* PCI Interrupt E4 */
+	default:
+		printk("PCI: Bad IRQ mapping request for slot %d, func %d\n", slot, pin-1);
+		return -1;
+	}
+}
+
+static struct resource sh7780_io_resource = {
+	.name	= "SH7780_IO",
+	.start	= 0x2000,
+	.end	= 0x2000 + SH7780_PCI_IO_SIZE - 1,
+	.flags	= IORESOURCE_IO
+};
+
+static struct resource sh7780_mem_resource = {
+	.name	= "SH7780_mem",
+	.start	= SH7780_PCI_MEMORY_BASE,
+	.end	= SH7780_PCI_MEMORY_BASE + SH7780_PCI_MEM_SIZE - 1,
+	.flags	= IORESOURCE_MEM
+};
+
+extern struct pci_ops sh7780_pci_ops;
+
+struct pci_channel board_pci_channels[] = {
+	{ &sh7780_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
+	{ NULL, NULL, NULL, 0, 0 },
+};
+EXPORT_SYMBOL(board_pci_channels);
+
+static struct sh7780_pci_address_map sh7780_pci_map = {
+	.window0	= {
+		.base	= SH7780_CS2_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+
+	.window1	= {
+		.base	= SH7780_CS3_BASE_ADDR,
+		.size	= 0x04000000,
+	},
+
+	.flags	= SH7780_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7780_pcic_init(&sh7780_pci_map);
+}
+
diff --git a/arch/sh/drivers/pci/pci-auto.c b/arch/sh/drivers/pci/pci-auto.c
index 4cef4d1d8c84..d55e46618549 100644
--- a/arch/sh/drivers/pci/pci-auto.c
+++ b/arch/sh/drivers/pci/pci-auto.c
@@ -45,11 +45,11 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 
-#undef	DEBUG
-#ifdef 	DEBUG
+#define	DEBUG
+#ifdef	DEBUG
 #define	DBG(x...)	printk(x)
 #else
-#define	DBG(x...)	
+#define	DBG(x...)
 #endif
 
 /*
@@ -102,7 +102,7 @@ static u32 pciauto_upper_iospc;
 static u32 pciauto_lower_memspc;
 static u32 pciauto_upper_memspc;
 
-static void __init 
+static void __init
 pciauto_setup_bars(struct pci_channel *hose,
 		   int top_bus,
 		   int current_bus,
@@ -116,7 +116,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 	int found_mem64 = 0;
 
 	for (bar = PCI_BASE_ADDRESS_0; bar <= bar_limit; bar+=4) {
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
 		u32 bar_addr;
 
 		/* Read the old BAR value */
@@ -125,7 +124,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 					pci_devfn,
 					bar,
 					&bar_addr);
-#endif
 
 		/* Tickle the BAR and get the response */
 		early_write_config_dword(hose, top_bus,
@@ -140,8 +138,7 @@ pciauto_setup_bars(struct pci_channel *hose,
 					bar,
 					&bar_response);
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
-		/* 
+		/*
 		 * Write the old BAR value back out, only update the BAR
 		 * if we implicitly want resources to be updated, which
 		 * is done by the generic code further down. -- PFM.
@@ -151,7 +148,6 @@ pciauto_setup_bars(struct pci_channel *hose,
 					 pci_devfn,
 					 bar,
 					 bar_addr);
-#endif
 
 		/* If BAR is not implemented go to the next BAR */
 		if (!bar_response)
@@ -177,7 +173,7 @@ retry:
 			    PCI_BASE_ADDRESS_MEM_TYPE_64)
 				found_mem64 = 1;
 
-			addr_mask = PCI_BASE_ADDRESS_MEM_MASK;		
+			addr_mask = PCI_BASE_ADDRESS_MEM_MASK;
 			upper_limit = &pciauto_upper_memspc;
 			lower_limit = &pciauto_lower_memspc;
 			DBG("        Mem");
@@ -193,22 +189,22 @@ retry:
 		if ((bar_value + bar_size) > *upper_limit) {
 			if (bar_response & PCI_BASE_ADDRESS_SPACE) {
 				if (io_resource_inuse->child) {
-					io_resource_inuse = 
+					io_resource_inuse =
 						io_resource_inuse->child;
-					pciauto_lower_iospc = 
+					pciauto_lower_iospc =
 						io_resource_inuse->start;
-					pciauto_upper_iospc = 
+					pciauto_upper_iospc =
 						io_resource_inuse->end + 1;
 					goto retry;
 				}
 
 			} else {
 				if (mem_resource_inuse->child) {
-					mem_resource_inuse = 
+					mem_resource_inuse =
 						mem_resource_inuse->child;
-					pciauto_lower_memspc = 
+					pciauto_lower_memspc =
 						mem_resource_inuse->start;
-					pciauto_upper_memspc = 
+					pciauto_upper_memspc =
 						mem_resource_inuse->end + 1;
 					goto retry;
 				}
@@ -230,7 +226,7 @@ retry:
 		 * If we are a 64-bit decoder then increment to the
 		 * upper 32 bits of the bar and force it to locate
 		 * in the lower 4GB of memory.
-		 */ 
+		 */
 		if (found_mem64) {
 			bar += 4;
 			early_write_config_dword(hose, top_bus,
@@ -362,7 +358,7 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 {
 	u32 temp;
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D)
+#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D) && !defined(CONFIG_SH_R7780RP)
 	/*
 	 * [jsun] we always bump up baselines a little, so that if there
 	 * nothing behind P2P bridge, we don't wind up overlapping IO/MEM
@@ -396,7 +392,7 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 * configured by this routine to happily live behind a
 	 * P2P bridge in a system.
 	 */
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D)
+#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
 	pciauto_lower_memspc += 0x00400000;
 	pciauto_lower_iospc += 0x00004000;
 #endif
@@ -433,12 +429,12 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 	int devfn_stop = 0xff;
 
 	sub_bus = current_bus;
-	
+
 	if (hose->first_devfn)
 		devfn_start = hose->first_devfn;
 	if (hose->last_devfn)
 		devfn_stop = hose->last_devfn;
-	
+
 	for (pci_devfn=devfn_start; pci_devfn<devfn_stop; pci_devfn++) {
 
 		if (PCI_FUNC(pci_devfn) && !found_multi)
@@ -471,7 +467,7 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 		if ((pci_class >> 16) == PCI_CLASS_BRIDGE_PCI) {
 			DBG("        Bridge: primary=%.2x, secondary=%.2x\n",
 				current_bus, sub_bus + 1);
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D)
+#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
 			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_1);
 #endif
 			pciauto_prescan_setup_bridge(hose, top_bus, current_bus,
@@ -490,10 +486,10 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 			DBG("PCI Autoconfig: Found CardBus bridge, device %d function %d\n", PCI_SLOT(pci_devfn), PCI_FUNC(pci_devfn));
 			/* Place CardBus Socket/ExCA registers */
 			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_0);
- 
+
 			pciauto_prescan_setup_cardbus_bridge(hose, top_bus,
 					current_bus, pci_devfn, sub_bus);
- 
+
 			DBG("Scanning sub bus %.2x, I/O 0x%.8x, Mem 0x%.8x\n",
 				sub_bus + 1,
 				pciauto_lower_iospc, pciauto_lower_memspc);
diff --git a/arch/sh/drivers/pci/pci-sh7751.c b/arch/sh/drivers/pci/pci-sh7751.c
index 682f3dae305d..65093ec1b55e 100644
--- a/arch/sh/drivers/pci/pci-sh7751.c
+++ b/arch/sh/drivers/pci/pci-sh7751.c
@@ -223,7 +223,7 @@ static int __init __area_sdram_check(unsigned int area)
 
 	word = inl(SH7751_BCR1);
 	/* check BCR for SDRAM in area */
-	if(((word >> area) & 1) == 0) {
+	if (((word >> area) & 1) == 0) {
 		printk("PCI: Area %d is not configured for SDRAM. BCR1=0x%x\n",
 		       area, word);
 		return 0;
@@ -232,7 +232,7 @@ static int __init __area_sdram_check(unsigned int area)
 
 	word = (u16)inw(SH7751_BCR2);
 	/* check BCR2 for 32bit SDRAM interface*/
-	if(((word >> (area << 1)) & 0x3) != 0x3) {
+	if (((word >> (area << 1)) & 0x3) != 0x3) {
 		printk("PCI: Area %d is not 32 bit SDRAM. BCR2=0x%x\n",
 		       area, word);
 		return 0;
diff --git a/arch/sh/drivers/pci/pci-sh7780.c b/arch/sh/drivers/pci/pci-sh7780.c
new file mode 100644
index 000000000000..e09721330ac2
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh7780.c
@@ -0,0 +1,341 @@
+/*
+ *	Low-Level PCI Support for the SH7780
+ *
+ *  Dustin McIntire (dustin@sensoria.com)
+ *	Derived from arch/i386/kernel/pci-*.c which bore the message:
+ *	(c) 1999--2000 Martin Mares <mj@ucw.cz>
+ *
+ *  Ported to the new API by Paul Mundt <lethal@linux-sh.org>
+ *  With cleanup by Paul van Gool <pvangool@mimotech.com>
+ *
+ *  May be copied or modified under the terms of the GNU General Public
+ *  License.  See linux/COPYING for more information.
+ *
+ */
+
+#undef DEBUG
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/sched.h>
+#include <linux/ioport.h>
+#include <linux/errno.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+
+#include <asm/machvec.h>
+#include <asm/io.h>
+#include "pci-sh7780.h"
+
+static unsigned int pci_probe = PCI_PROBE_CONF1;
+extern int pci_fixup_pcic(void);
+
+/*
+ * Direct access to PCI hardware...
+ */
+
+#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
+
+/*
+ * Functions for accessing PCI configuration space with type 1 accesses
+ */
+static int sh7780_pci_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	unsigned long flags;
+	u32 data;
+
+	/*
+	 * PCIPDR may only be accessed as 32 bit words,
+	 * so we must do byte alignment by hand
+	 */
+	local_irq_save(flags);
+	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
+	data = inl(PCI_REG(SH7780_PCIPDR));
+	local_irq_restore(flags);
+
+	switch (size) {
+	case 1:
+		*val = (data >> ((where & 3) << 3)) & 0xff;
+		break;
+	case 2:
+		*val = (data >> ((where & 2) << 3)) & 0xffff;
+		break;
+	case 4:
+		*val = data;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/*
+ * Since SH7780 only does 32bit access we'll have to do a read,
+ * mask,write operation.
+ * We'll allow an odd byte offset, though it should be illegal.
+ */
+static int sh7780_pci_write(struct pci_bus *bus, unsigned int devfn,
+			    int where, int size, u32 val)
+{
+	unsigned long flags;
+	int shift;
+	u32 data;
+
+	local_irq_save(flags);
+	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
+	data = inl(PCI_REG(SH7780_PCIPDR));
+	local_irq_restore(flags);
+
+	switch (size) {
+	case 1:
+		shift = (where & 3) << 3;
+		data &= ~(0xff << shift);
+		data |= ((val & 0xff) << shift);
+		break;
+	case 2:
+		shift = (where & 2) << 3;
+		data &= ~(0xffff << shift);
+		data |= ((val & 0xffff) << shift);
+		break;
+	case 4:
+		data = val;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	outl(data, PCI_REG(SH7780_PCIPDR));
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+#undef CONFIG_CMD
+
+struct pci_ops sh7780_pci_ops = {
+	.read		= sh7780_pci_read,
+	.write		= sh7780_pci_write,
+};
+
+static int __init pci_check_direct(void)
+{
+	unsigned int tmp, id;
+
+	outl(0x00000001, SH7780_PCI_VCR2); /* Enable PCIC */
+
+	/* check for SH7780/SH7780R hardware */
+	id = inl(PCI_REG(SH7780_PCIVID));
+	if ((id != ((SH7780_DEVICE_ID << 16) | SH7780_VENDOR_ID)) &&
+	    (id != ((SH7781_DEVICE_ID << 16) | SH7780_VENDOR_ID))) {
+		printk(KERN_ERR "PCI: This is not an SH7780 (%x)\n", id);
+		return -ENODEV;
+	}
+
+	/*
+	 * Check if configuration works.
+	 */
+	if (pci_probe & PCI_PROBE_CONF1) {
+		tmp = inl(PCI_REG(SH7780_PCIPAR));
+		outl(0x80000000, PCI_REG(SH7780_PCIPAR));
+		if (inl(PCI_REG(SH7780_PCIPAR)) == 0x80000000) {
+			outl(tmp, PCI_REG(SH7780_PCIPAR));
+			printk(KERN_INFO "PCI: Using configuration type 1\n");
+			request_region(PCI_REG(SH7780_PCIPAR), 8, "PCI conf1");
+			return 0;
+		}
+		outl(tmp, PCI_REG(SH7780_PCIPAR));
+	}
+
+	pr_debug("PCI: pci_check_direct failed\n");
+	return -EINVAL;
+}
+
+/***************************************************************************************/
+
+/*
+ *  Handle bus scanning and fixups ....
+ */
+
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i=0; i<4; i++) {
+		struct resource *r = &d->resource[i];
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+/*
+ *  Called after each bus is probed, but before its children
+ *  are examined.
+ */
+
+void __init pcibios_fixup_bus(struct pci_bus *b)
+{
+	pci_read_bridge_bases(b);
+}
+
+/*
+ * Initialization. Try all known PCI access methods. Note that we support
+ * using both PCI BIOS and direct access: in such cases, we use I/O ports
+ * to access config space.
+ *
+ * Note that the platform specific initialization (BSC registers, and memory
+ * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
+ * exists and via the platform defined function pcibios_init_platform().
+ * See pci_bigsur.c for implementation;
+ *
+ * The BIOS version of the pci functions is not yet implemented but it is left
+ * in for completeness.  Currently an error will be genereated at compile time.
+ */
+
+static int __init sh7780_pci_init(void)
+{
+	int ret;
+
+	pr_debug("PCI: Starting intialization.\n");
+
+	/* Setup the INTC */
+	ctrl_outl(0x00200000, INTC_ICR0);	/* INTC SH-4 Mode */
+	ctrl_outl(0x00078000, INTC_INT2MSKCR);	/* enable PCIINTA - PCIINTD */
+	ctrl_outl(0x40000000, INTC_INTMSK1);	/* disable IRL4-7 Interrupt */
+	ctrl_outl(0x0000fffe, INTC_INTMSK2);	/* disable IRL4-7 Interrupt */
+	ctrl_outl(0x80000000, INTC_INTMSKCLR1);	/* enable IRL0-3 Interrupt */
+	ctrl_outl(0xfffe0000, INTC_INTMSKCLR2);	/* enable IRL0-3 Interrupt */
+
+	if ((ret = pci_check_direct()) != 0)
+		return ret;
+
+	return pcibios_init_platform();
+}
+
+core_initcall(sh7780_pci_init);
+
+int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
+{
+	u32 word;
+
+	/*
+	 * This code is unused for some boards as it is done in the
+	 * bootloader and doing it here means the MAC addresses loaded
+	 * by the bootloader get lost.
+	 */
+	if (!(map->flags & SH7780_PCIC_NO_RESET)) {
+		/* toggle PCI reset pin */
+		word = SH7780_PCICR_PREFIX | SH7780_PCICR_PRST;
+		outl(word,PCI_REG(SH7780_PCICR));
+		/* Wait for a long time... not 1 sec. but long enough */
+		mdelay(100);
+		word = SH7780_PCICR_PREFIX;
+		outl(word,PCI_REG(SH7780_PCICR));
+	}
+
+	/* set the command/status bits to:
+	 * Wait Cycle Control + Parity Enable + Bus Master +
+	 * Mem space enable
+	 */
+	outl(0x00000046, PCI_REG(SH7780_PCICMD));
+
+	/* define this host as the host bridge */
+	word = SH7780_PCI_HOST_BRIDGE << 24;
+	outl(word, PCI_REG(SH7780_PCIRID));
+
+	/* Set IO and Mem windows to local address
+	 * Make PCI and local address the same for easy 1 to 1 mapping
+	 * Window0 = map->window0.size @ non-cached area base = SDRAM
+	 * Window1 = map->window1.size @ cached area base = SDRAM
+	 */
+	word = ((map->window0.size - 1) & 0x1ff00001) | 0x01;
+	outl(0x07f00001, PCI_REG(SH7780_PCILSR0));
+	word = ((map->window1.size - 1) & 0x1ff00001) | 0x01;
+	outl(0x00000001, PCI_REG(SH7780_PCILSR1));
+	/* Set the values on window 0 PCI config registers */
+	word = P2SEGADDR(map->window0.base);
+	outl(0xa8000000, PCI_REG(SH7780_PCILAR0));
+	outl(0x08000000, PCI_REG(SH7780_PCIMBAR0));
+	/* Set the values on window 1 PCI config registers */
+	word = P2SEGADDR(map->window1.base);
+	outl(0x00000000, PCI_REG(SH7780_PCILAR1));
+	outl(0x00000000, PCI_REG(SH7780_PCIMBAR1));
+
+	/* Map IO space into PCI IO window
+	 * The IO window is 64K-PCIBIOS_MIN_IO in size
+	 * IO addresses will be translated to the
+	 * PCI IO window base address
+	 */
+	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
+	    (64*1024), SH7780_PCI_IO_BASE+PCIBIOS_MIN_IO);
+
+	/* NOTE: I'm ignoring the PCI error IRQs for now..
+	 * TODO: add support for the internal error interrupts and
+	 * DMA interrupts...
+	 */
+
+#ifdef CONFIG_SH_R7780RP
+	pci_fixup_pcic();
+#endif
+
+	/* SH7780 init done, set central function init complete */
+	/* use round robin mode to stop a device starving/overruning */
+	word = SH7780_PCICR_PREFIX | SH7780_PCICR_CFIN | /* SH7780_PCICR_ARBM |*/ SH7780_PCICR_FTO;
+	outl(word, PCI_REG(SH7780_PCICR));
+
+	return 1;
+}
+
+char * __init pcibios_setup(char *str)
+{
+	if (!strcmp(str, "off")) {
+		pci_probe = 0;
+		return NULL;
+	}
+
+	return str;
+}
+
+/*
+ *	IRQ functions
+ */
+static u8 __init sh7780_no_swizzle(struct pci_dev *dev, u8 *pin)
+{
+	/* no swizzling */
+	return PCI_SLOT(dev->devfn);
+}
+
+static int sh7780_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
+{
+	int irq = -1;
+
+	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
+	irq = pcibios_map_platform_irq(slot,pin);
+	if( irq < 0 ) {
+		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
+		return irq;
+	}
+
+	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
+
+	return irq;
+}
+
+void __init pcibios_fixup_irqs(void)
+{
+	pci_fixup_irqs(sh7780_no_swizzle, sh7780_pci_lookup_irq);
+}
+
diff --git a/arch/sh/drivers/pci/pci-sh7780.h b/arch/sh/drivers/pci/pci-sh7780.h
new file mode 100644
index 000000000000..750d5d7753a2
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh7780.h
@@ -0,0 +1,168 @@
+/*
+ *	Low-Level PCI Support for SH7780 targets
+ *
+ *  Dustin McIntire (dustin@sensoria.com) (c) 2001
+ *  Paul Mundt (lethal@linux-sh.org) (c) 2003
+ *
+ *  May be copied or modified under the terms of the GNU General Public
+ *  License.  See linux/COPYING for more information.
+ *
+ */
+
+#ifndef _PCI_SH7780_H_
+#define _PCI_SH7780_H_
+
+#include <linux/pci.h>
+
+/* set debug level 4=verbose...1=terse */
+//#define DEBUG_PCI 3
+#undef DEBUG_PCI
+
+#ifdef DEBUG_PCI
+#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
+#else
+#define PCIDBG(n, x...)
+#endif
+
+/* startup values */
+#define PCI_PROBE_BIOS		1
+#define PCI_PROBE_CONF1		2
+#define PCI_PROBE_CONF2		4
+#define PCI_NO_SORT		0x100
+#define PCI_BIOS_SORT		0x200
+#define PCI_NO_CHECKS		0x400
+#define PCI_ASSIGN_ROMS		0x1000
+#define PCI_BIOS_IRQ_SCAN	0x2000
+
+/* Platform Specific Values */
+#define SH7780_VENDOR_ID	0x1912
+#define SH7780_DEVICE_ID	0x0002
+#define SH7781_DEVICE_ID	0x0001
+
+/* SH7780 Control Registers */
+#define	SH7780_PCI_VCR0		0xFE000000
+#define	SH7780_PCI_VCR1		0xFE000004
+#define	SH7780_PCI_VCR2		0xFE000008
+
+/* SH7780 Specific Values */
+#define SH7780_PCI_CONFIG_BASE	0xFD000000	/* Config space base addr */
+#define SH7780_PCI_CONFIG_SIZE	0x01000000	/* Config space size */
+#define SH7780_PCI_MEMORY_BASE	0xFD000000	/* Memory space base addr */
+#define SH7780_PCI_MEM_SIZE	0x01000000	/* Size of Memory window */
+#if 1
+#define SH7780_PCI_IO_BASE	0xFE400000	/* IO space base address */
+#define SH7780_PCI_IO_SIZE	0x00400000	/* Size of IO window */
+#else
+#define SH7780_PCI_IO_BASE	0xFE200000	/* IO space base address */
+#define SH7780_PCI_IO_SIZE	0x00200000	/* Size of IO window */
+#endif
+
+#define SH7780_PCIREG_BASE	0xFE040000	/* PCI regs base address */
+#define PCI_REG(n)		(SH7780_PCIREG_BASE+n)
+
+/* SH7780 PCI Config Registers */
+#define SH7780_PCIVID		0x000		/* Vendor ID */
+#define SH7780_PCIDID		0x002		/* Device ID */
+#define SH7780_PCICMD		0x004		/* Command */
+#define SH7780_PCISTATUS	0x006		/* Status */
+#define SH7780_PCIRID		0x008		/* Revision ID */
+#define SH7780_PCIPIF		0x009		/* Program Interface */
+#define SH7780_PCISUB		0x00a		/* Sub class code */
+#define SH7780_PCIBCC		0x00b		/* Base class code */
+#define SH7780_PCICLS		0x00c		/* Cache line size */
+#define SH7780_PCILTM		0x00d		/* latency timer */
+#define SH7780_PCIHDR		0x00e		/* Header type */
+#define SH7780_PCIBIST		0x00f		/* BIST */
+#define SH7780_PCIIBAR		0x010		/* IO Base address */
+#define SH7780_PCIMBAR0		0x014		/* Memory base address0 */
+#define SH7780_PCIMBAR1		0x018		/* Memory base address1 */
+#define SH7780_PCISVID		0x02c		/* Sub system vendor ID */
+#define SH7780_PCISID		0x02e		/* Sub system ID */
+#define SH7780_PCICP		0x034
+#define SH7780_PCIINTLINE	0x03c		/* Interrupt line */
+#define SH7780_PCIINTPIN	0x03d		/* Interrupt pin */
+#define SH7780_PCIMINGNT	0x03e		/* Minumum grand */
+#define SH7780_PCIMAXLAT	0x03f		/* Maxmum latency */
+#define SH7780_PCICID		0x040
+#define SH7780_PCINIP		0x041
+#define SH7780_PCIPMC		0x042
+#define SH7780_PCIPMCSR		0x044
+#define SH7780_PCIPMCSR_BSE	0x046
+#define SH7780_PCICDD		0x047
+
+/* SH7780 PCI Local Registers */
+#define SH7780_PCICR		0x100		/* PCI Control Register */
+  #define SH7780_PCICR_PREFIX	0xA5000000	/* CR prefix for write */
+  #define SH7780_PCICR_PFCS	0x00000800	/* TRDY/IRDY Enable */
+  #define SH7780_PCICR_FTO	0x00000400	/* TRDY/IRDY Enable */
+  #define SH7780_PCICR_PFE	0x00000200	/* Target Read Single */
+  #define SH7780_PCICR_TBS	0x00000100	/* Target Byte Swap */
+  #define SH7780_PCICR_ARBM	0x00000040	/* PCI Arbitration Mode */
+  #define SH7780_PCICR_IOCS	0x00000004	/* INTA output assert */
+  #define SH7780_PCICR_PRST	0x00000002	/* PCI Reset Assert */
+  #define SH7780_PCICR_CFIN	0x00000001	/* Central Fun. Init Done */
+#define SH7780_PCILSR0		0x104		/* PCI Local Space Register0 */
+#define SH7780_PCILSR1		0x108		/* PCI Local Space Register1 */
+#define SH7780_PCILAR0		0x10C	/* PCI Local Address Register1 */
+#define SH7780_PCILAR1		0x110	/* PCI Local Address Register1 */
+#define SH7780_PCIIR		0x114		/* PCI Interrupt Register */
+#define SH7780_PCIIMR		0x118	/* PCI Interrupt Mask Register */
+#define SH7780_PCIAIR		0x11C		/* Error Address Register */
+#define SH7780_PCICIR		0x120	/* Error Command/Data Register */
+#define SH7780_PCIAINT		0x130		/* Arbiter Interrupt Register */
+#define SH7780_PCIAINTM		0x134		/* Arbiter Int. Mask Register */
+#define SH7780_PCIBMIR		0x138		/* Error Bus Master Register */
+#define SH7780_PCIPAR		0x1C0		/* PIO Address Register */
+#define SH7780_PCIPINT		0x1CC	/* Power Management Int. Register */
+#define SH7780_PCIPINTM		0x1D0	/* Power Management Mask Register */
+#define SH7780_PCIMBR0		0x1E0		/* Memory Bank0 Register */
+#define SH7780_PCIMBMR0		0x1E4		/* Memory Bank0 Mask Register */
+#define SH7780_PCIMBR1		0x1E8		/* Memory Bank1  Register */
+#define SH7780_PCIMBMR1		0x1EC		/* Memory Bank1 Mask Register */
+#define SH7780_PCIMBR2		0x1F0		/* Memory Bank2 Register */
+#define SH7780_PCIMBMR2		0x1F4		/* Memory Bank2 Mask Register */
+#define SH7780_PCIIOBR		0x1F8		/* Bank Register */
+#define SH7780_PCIIOBMR		0x1FC		/* Bank Mask Register */
+#define SH7780_PCICSCR0		0x210		/* Cache Snoop1 Cnt. Register */
+#define SH7780_PCICSCR1		0x214		/* Cache Snoop2 Cnt. Register */
+#define SH7780_PCICSAR0		0x218	/* Cache Snoop1 Addr. Register */
+#define SH7780_PCICSAR1		0x21C	/* Cache Snoop2 Addr. Register */
+#define SH7780_PCIPDR		0x220		/* Port IO Data Register */
+
+/* General Memory Config Addresses */
+#define SH7780_CS0_BASE_ADDR	0x0
+#define SH7780_MEM_REGION_SIZE	0x04000000
+#define SH7780_CS1_BASE_ADDR	(SH7780_CS0_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS2_BASE_ADDR	(SH7780_CS1_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS3_BASE_ADDR	(SH7780_CS2_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS4_BASE_ADDR	(SH7780_CS3_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS5_BASE_ADDR	(SH7780_CS4_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+#define SH7780_CS6_BASE_ADDR	(SH7780_CS5_BASE_ADDR + SH7780_MEM_REGION_SIZE)
+
+/* General PCI values */
+#define SH7780_PCI_HOST_BRIDGE	0x6
+
+/* Flags */
+#define SH7780_PCIC_NO_RESET	0x0001
+
+/* External functions defined per platform i.e. Big Sur, SE... (these could be routed
+ * through the machine vectors... */
+extern int pcibios_init_platform(void);
+extern int pcibios_map_platform_irq(u8 slot, u8 pin);
+
+struct sh7780_pci_address_space {
+	unsigned long base;
+	unsigned long size;
+};
+
+struct sh7780_pci_address_map {
+	struct sh7780_pci_address_space window0;
+	struct sh7780_pci_address_space window1;
+	unsigned long flags;
+};
+
+/* arch/sh/drivers/pci/pci-sh7780.c */
+extern int sh7780_pcic_init(struct sh7780_pci_address_map *map);
+
+#endif /* _PCI_SH7780_H_ */
+
diff --git a/arch/sh/kernel/cpu/irq/intc2.c b/arch/sh/kernel/cpu/irq/intc2.c
index 30064bf6e154..e30e4b7aa70e 100644
--- a/arch/sh/kernel/cpu/irq/intc2.c
+++ b/arch/sh/kernel/cpu/irq/intc2.c
@@ -241,9 +241,9 @@ static struct intc2_init {
 	/* 110-111 reserved/unused */
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 	{ TIMER_IRQ, 0, 24, 0, INTC_TMU0_MSK, 2},
-#ifdef CONFIG_SH_RTC
-	{ RTC_IRQ, 4, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
-#endif
+	{ 21, 1, 0, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 22, 1, 1, 0, INTC_RTC_MSK, TIMER_PRIORITY },
+	{ 23, 1, 2, 0, INTC_RTC_MSK, TIMER_PRIORITY },
 	{ SCIF0_ERI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_RXI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
 	{ SCIF0_BRI_IRQ, 8, 24, 0, INTC_SCIF0_MSK, SCIF0_PRIORITY },
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 01ae394a6599..1f7f2c255ca2 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -26,4 +26,6 @@ EDOSK7705		SH_EDOSK7705
 SH4202_MICRODEV		SH_SH4202_MICRODEV
 SH03			SH_SH03
 LANDISK			SH_LANDISK
+R7780RP			SH_R7780RP
+R7780MP			SH_R7780MP
 TITAN			SH_TITAN
diff --git a/include/asm-sh/cpu-sh4/addrspace.h b/include/asm-sh/cpu-sh4/addrspace.h
index 727634d886ce..bb2e1b03060c 100644
--- a/include/asm-sh/cpu-sh4/addrspace.h
+++ b/include/asm-sh/cpu-sh4/addrspace.h
@@ -22,5 +22,8 @@
 #define P4SEG_TLB_DATA	0xf7000000
 #define P4SEG_REG_BASE	0xff000000
 
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+
 #endif /* __ASM_CPU_SH4_ADDRSPACE_H */
 
diff --git a/include/asm-sh/cpu-sh4/dma-sh7780.h b/include/asm-sh/cpu-sh4/dma-sh7780.h
new file mode 100644
index 000000000000..6c90d28331b2
--- /dev/null
+++ b/include/asm-sh/cpu-sh4/dma-sh7780.h
@@ -0,0 +1,39 @@
+#ifndef __ASM_SH_CPU_SH4_DMA_SH7780_H
+#define __ASM_SH_CPU_SH4_DMA_SH7780_H
+
+#define REQ_HE	0x000000C0
+#define REQ_H	0x00000080
+#define REQ_LE	0x00000040
+#define TM_BURST 0x0000020
+#define TS_8	0x00000000
+#define TS_16	0x00000008
+#define TS_32	0x00000010
+#define TS_16BLK	0x00000018
+#define TS_32BLK	0x00100000
+
+/*
+ * The SuperH DMAC supports a number of transmit sizes, we list them here,
+ * with their respective values as they appear in the CHCR registers.
+ *
+ * Defaults to a 64-bit transfer size.
+ */
+enum {
+	XMIT_SZ_8BIT,
+	XMIT_SZ_16BIT,
+	XMIT_SZ_32BIT,
+	XMIT_SZ_128BIT,
+	XMIT_SZ_256BIT,
+};
+
+/*
+ * The DMA count is defined as the number of bytes to transfer.
+ */
+static unsigned int __attribute__ ((used)) ts_shift[] = {
+	[XMIT_SZ_8BIT]		= 0,
+	[XMIT_SZ_16BIT]		= 1,
+	[XMIT_SZ_32BIT]		= 2,
+	[XMIT_SZ_128BIT]	= 4,
+	[XMIT_SZ_256BIT]	= 5,
+};
+
+#endif /* __ASM_SH_CPU_SH4_DMA_SH7780_H */
diff --git a/include/asm-sh/cpu-sh4/dma.h b/include/asm-sh/cpu-sh4/dma.h
index 0dfe61f14802..3e4b3e6d80c0 100644
--- a/include/asm-sh/cpu-sh4/dma.h
+++ b/include/asm-sh/cpu-sh4/dma.h
@@ -1,11 +1,17 @@
 #ifndef __ASM_CPU_SH4_DMA_H
 #define __ASM_CPU_SH4_DMA_H
 
+#define DMAOR_INIT	( 0x8000 | DMAOR_DME )
+
 #ifdef CONFIG_CPU_SH4A
 #define SH_DMAC_BASE	0xfc808020
+
+#define CHCR_TS_MASK	0x18
+#define CHCR_TS_SHIFT	3
+
+#include <asm/cpu/dma-sh7780.h>
 #else
 #define SH_DMAC_BASE	0xffa00000
-#endif
 
 /* Definitions for the SuperH DMAC */
 #define TM_BURST	0x0000080
@@ -19,8 +25,6 @@
 
 #define DMAOR_COD	0x00000008
 
-#define DMAOR_INIT	( 0x8000 | DMAOR_DME )
-
 /*
  * The SuperH DMAC supports a number of transmit sizes, we list them here,
  * with their respective values as they appear in the CHCR registers.
@@ -45,5 +49,6 @@ static unsigned int ts_shift[] __attribute__ ((used)) = {
 	[XMIT_SZ_32BIT]		= 2,
 	[XMIT_SZ_256BIT]	= 5,
 };
+#endif
 
 #endif /* __ASM_CPU_SH4_DMA_H */
diff --git a/include/asm-sh/hs7751rvoip/hs7751rvoip.h b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
index 69faf0171473..c4cff9d33927 100644
--- a/include/asm-sh/hs7751rvoip/hs7751rvoip.h
+++ b/include/asm-sh/hs7751rvoip/hs7751rvoip.h
@@ -19,8 +19,6 @@
 #define PA_OUTPORTR	0xa400000e	/* Output Port Reguster */
 #define PA_VERREG	0xa4000014	/* FPGA Version Register */
 
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
 
 #define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
diff --git a/include/asm-sh/irq-sh7780.h b/include/asm-sh/irq-sh7780.h
index 7f90315cd830..895c5780e454 100644
--- a/include/asm-sh/irq-sh7780.h
+++ b/include/asm-sh/irq-sh7780.h
@@ -145,11 +145,6 @@
 #define	TMU_CH5_IPR_POS		1
 #define TMU_CH5_PRIORITY	2
 
-#define	RTC_IRQ		22
-#define	RTC_IPR_ADDR	INTC_INT2PRI1
-#define	RTC_IPR_POS	0
-#define	RTC_PRIORITY	TIMER_PRIORITY
-
 /* SCIF0 */
 #define SCIF0_ERI_IRQ	40
 #define SCIF0_RXI_IRQ	41
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 611e67cd0627..7e8455b1cb44 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -577,7 +577,7 @@ extern int ipr_irq_demux(int irq);
 #define NR_INTC2_IRQS	64
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 #define INTC2_BASE	0xffd40000
-#define INTC2_FIRST_IRQ	22
+#define INTC2_FIRST_IRQ	21
 #define INTC2_INTMSK_OFFSET	(0x38)
 #define INTC2_INTMSKCLR_OFFSET	(0x3c)
 #define NR_INTC2_IRQS	60
diff --git a/include/asm-sh/landisk/iodata_landisk.h b/include/asm-sh/landisk/iodata_landisk.h
index 9db3cdfe6776..c74d3c73f377 100644
--- a/include/asm-sh/landisk/iodata_landisk.h
+++ b/include/asm-sh/landisk/iodata_landisk.h
@@ -22,8 +22,6 @@
 /* 2003.10.31 I-O DATA NSD NWG	add.	for shutdown port clear */
 #define PA_PWRINT_CLR	0xb0000006	/* Shutdown Interrupt clear Register */
 
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_LCD_CLRDSP	0x00		/* LCD Clear Display Offset */
 #define PA_LCD_RTNHOME	0x00		/* LCD Return Home Offset */
 #define PA_LCD_ENTMODE	0x00		/* LCD Entry Mode Offset */
diff --git a/include/asm-sh/r7780rp/ide.h b/include/asm-sh/r7780rp/ide.h
new file mode 100644
index 000000000000..a1ed78e0f617
--- /dev/null
+++ b/include/asm-sh/r7780rp/ide.h
@@ -0,0 +1,8 @@
+#ifndef __ASM_SH_R7780RP_IDE_H
+#define __ASM_SH_R7780RP_IDE_H
+
+/* Nothing to see here.. */
+#include <asm/mach/r7780rp.h>
+
+#endif /* __ASM_SH_R7780RP_IDE_H */
+
diff --git a/include/asm-sh/r7780rp/r7780rp.h b/include/asm-sh/r7780rp/r7780rp.h
new file mode 100644
index 000000000000..f95d9dba31a2
--- /dev/null
+++ b/include/asm-sh/r7780rp/r7780rp.h
@@ -0,0 +1,177 @@
+#ifndef __ASM_SH_RENESAS_R7780RP_H
+#define __ASM_SH_RENESAS_R7780RP_H
+
+/*
+ * linux/include/asm-sh/r7780rp.h
+ *
+ * Copyright (C) 2000  Atom Create Engineering Co., Ltd.
+ *
+ * Renesas Solutions Highlander R7780RP support
+ */
+
+/* Box specific addresses.  */
+#if defined(CONFIG_SH_R7780MP)
+#define PA_BCR          0xa4000000      /* FPGA */
+#define PA_IRLMSK       (PA_BCR+0x0000) /* Interrupt Mask control */
+#define PA_IRLMON       (PA_BCR+0x0002) /* Interrupt Status control */
+#define PA_IRLPRI1      (PA_BCR+0x0004) /* Interrupt Priorty 1 */
+#define PA_IRLPRI2      (PA_BCR+0x0006) /* Interrupt Priorty 2 */
+#define PA_IRLPRI3      (PA_BCR+0x0008) /* Interrupt Priorty 3 */
+#define PA_IRLPRI4      (PA_BCR+0x000a) /* Interrupt Priorty 4 */
+#define PA_RSTCTL       (PA_BCR+0x000c) /* Reset Control */
+#define PA_PCIBD        (PA_BCR+0x000e) /* PCI Board detect control */
+#define PA_PCICD        (PA_BCR+0x0010) /* PCI Conector detect control */
+#define PA_EXTGIO       (PA_BCR+0x0016) /* Extension GPIO Control */
+#define PA_IVDRMON      (PA_BCR+0x0018) /* iVDR Moniter control */
+#define PA_IVDRCTL      (PA_BCR+0x001a) /* iVDR control */
+#define PA_OBLED        (PA_BCR+0x001c) /* On Board LED control */
+#define PA_OBSW         (PA_BCR+0x001e) /* On Board Switch control */
+#define PA_AUDIOSEL     (PA_BCR+0x0020) /* Sound Interface Select control */
+#define PA_EXTPLR       (PA_BCR+0x001e) /* Extention Pin Polarity control */
+#define PA_TPCTL        (PA_BCR+0x0100) /* Touch Panel Access control */
+#define PA_TPDCKCTL     (PA_BCR+0x0102) /* Touch Panel Access data control */
+#define PA_TPCTLCLR     (PA_BCR+0x0104) /* Touch Panel Access control */
+#define PA_TPXPOS       (PA_BCR+0x0106) /* Touch Panel X position control */
+#define PA_TPYPOS       (PA_BCR+0x0108) /* Touch Panel Y position control */
+#define PA_DBSW         (PA_BCR+0x0200) /* Debug Board Switch control */
+#define PA_CFCTL        (PA_BCR+0x0300) /* CF Timing control */
+#define PA_CFPOW        (PA_BCR+0x0302) /* CF Power control */
+#define PA_CFCDINTCLR   (PA_BCR+0x0304) /* CF Insert Interrupt clear */
+#define PA_SCSMR0       (PA_BCR+0x0400) /* SCIF0 Serial mode control */
+#define PA_SCBRR0       (PA_BCR+0x0404) /* SCIF0 Bit rate control */
+#define PA_SCSCR0       (PA_BCR+0x0408) /* SCIF0 Serial control */
+#define PA_SCFTDR0      (PA_BCR+0x040c) /* SCIF0 Send FIFO control */
+#define PA_SCFSR0       (PA_BCR+0x0410) /* SCIF0 Serial status control */
+#define PA_SCFRDR0      (PA_BCR+0x0414) /* SCIF0 Receive FIFO control */
+#define PA_SCFCR0       (PA_BCR+0x0418) /* SCIF0 FIFO control */
+#define PA_SCTFDR0      (PA_BCR+0x041c) /* SCIF0 Send FIFO data control */
+#define PA_SCRFDR0      (PA_BCR+0x0420) /* SCIF0 Receive FIFO data control */
+#define PA_SCSPTR0      (PA_BCR+0x0424) /* SCIF0 Serial Port control */
+#define PA_SCLSR0       (PA_BCR+0x0428) /* SCIF0 Line Status control */
+#define PA_SCRER0       (PA_BCR+0x042c) /* SCIF0 Serial Error control */
+#define PA_SCSMR1       (PA_BCR+0x0500) /* SCIF1 Serial mode control */
+#define PA_SCBRR1       (PA_BCR+0x0504) /* SCIF1 Bit rate control */
+#define PA_SCSCR1       (PA_BCR+0x0508) /* SCIF1 Serial control */
+#define PA_SCFTDR1      (PA_BCR+0x050c) /* SCIF1 Send FIFO control */
+#define PA_SCFSR1       (PA_BCR+0x0510) /* SCIF1 Serial status control */
+#define PA_SCFRDR1      (PA_BCR+0x0514) /* SCIF1 Receive FIFO control */
+#define PA_SCFCR1       (PA_BCR+0x0518) /* SCIF1 FIFO control */
+#define PA_SCTFDR1      (PA_BCR+0x051c) /* SCIF1 Send FIFO data control */
+#define PA_SCRFDR1      (PA_BCR+0x0520) /* SCIF1 Receive FIFO data control */
+#define PA_SCSPTR1      (PA_BCR+0x0524) /* SCIF1 Serial Port control */
+#define PA_SCLSR1       (PA_BCR+0x0528) /* SCIF1 Line Status control */
+#define PA_SCRER1       (PA_BCR+0x052c) /* SCIF1 Serial Error control */
+#define PA_ICCR         (PA_BCR+0x0600) /* Serial control */
+#define PA_SAR          (PA_BCR+0x0602) /* Serial Slave control */
+#define PA_MDR          (PA_BCR+0x0604) /* Serial Mode control */
+#define PA_ADR1         (PA_BCR+0x0606) /* Serial Address1 control */
+#define PA_DAR1         (PA_BCR+0x0646) /* Serial Data1 control */
+#define PA_VERREG       (PA_BCR+0x0700) /* FPGA Version Register */
+#define PA_POFF         (PA_BCR+0x0800) /* System Power Off control */
+#define PA_PMR          (PA_BCR+0x0900) /*  */
+
+#define PA_AX88796L     0xa4100400      /* AX88796L Area */
+#define PA_SC1602BSLB   0xa6000000      /* SC1602BSLB Area */
+#define PA_AREA5_IO     0xb4000000      /* Area 5 IO Memory */
+#define PA_AREA6_IO     0xb8000000      /* Area 6 IO Memory */
+#define PA_IDE_OFFSET   0x1f0           /* CF IDE Offset */
+#define AX88796L_IO_BASE        0x1000  /* AX88796L IO Base Address */
+
+#define IRLCNTR1        (PA_BCR + 0)    /* Interrupt Control Register1 */
+
+#define IRQ_PCISLOT1    65              /* PCI Slot #1 IRQ */
+#define IRQ_PCISLOT2    66              /* PCI Slot #2 IRQ */
+#define IRQ_PCISLOT3    67              /* PCI Slot #3 IRQ */
+#define IRQ_PCISLOT4    68              /* PCI Slot #4 IRQ */
+#define IRQ_CFCARD      1               /* CF Card IRQ */
+// #define IRQ_CFINST   0               /* CF Card Insert IRQ */
+#define IRQ_TP          2               /* Touch Panel IRQ */
+#define IRQ_SCI1        3               /* SCI1 IRQ */
+#define IRQ_SCI0        4               /* SCI0 IRQ */
+#define IRQ_2SERIAL     5               /* Serial IRQ */
+#define IRQ_RTC         6               /* RTC A / B IRQ */
+#define IRQ_EXTENTION6  7               /* EXT6n IRQ */
+#define IRQ_EXTENTION5  8               /* EXT5n IRQ */
+#define IRQ_EXTENTION4  9               /* EXT4n IRQ */
+#define IRQ_EXTENTION2  10              /* EXT2n IRQ */
+#define IRQ_EXTENTION1  11              /* EXT1n IRQ */
+#define IRQ_ONETH       13              /* On board Ethernet IRQ */
+#define IRQ_PSW         14              /* Push Switch IRQ */
+
+#else /* R7780RP */
+
+#define PA_BCR		0xa5000000	/* FPGA */
+#define	PA_IRLMSK	(PA_BCR+0x0000)	/* Interrupt Mask control */
+#define PA_IRLMON	(PA_BCR+0x0002)	/* Interrupt Status control */
+#define	PA_SDPOW	(PA_BCR+0x0004)	/* SD Power control */
+#define	PA_RSTCTL	(PA_BCR+0x0006)	/* Device Reset control */
+#define	PA_PCIBD	(PA_BCR+0x0008)	/* PCI Board detect control */
+#define	PA_PCICD	(PA_BCR+0x000a)	/* PCI Conector detect control */
+#define	PA_ZIGIO1	(PA_BCR+0x000c)	/* Zigbee IO control 1 */
+#define	PA_ZIGIO2	(PA_BCR+0x000e)	/* Zigbee IO control 2 */
+#define	PA_ZIGIO3	(PA_BCR+0x0010)	/* Zigbee IO control 3 */
+#define	PA_ZIGIO4	(PA_BCR+0x0012)	/* Zigbee IO control 4 */
+#define	PA_IVDRMON	(PA_BCR+0x0014)	/* iVDR Moniter control */
+#define	PA_IVDRCTL	(PA_BCR+0x0016)	/* iVDR control */
+#define PA_OBLED	(PA_BCR+0x0018)	/* On Board LED control */
+#define PA_OBSW		(PA_BCR+0x001a)	/* On Board Switch control */
+#define PA_AUDIOSEL	(PA_BCR+0x001c)	/* Sound Interface Select control */
+#define PA_EXTPLR	(PA_BCR+0x001e)	/* Extention Pin Polarity control */
+#define PA_TPCTL	(PA_BCR+0x0100)	/* Touch Panel Access control */
+#define PA_TPDCKCTL	(PA_BCR+0x0102)	/* Touch Panel Access data control */
+#define PA_TPCTLCLR	(PA_BCR+0x0104)	/* Touch Panel Access control */
+#define PA_TPXPOS	(PA_BCR+0x0106)	/* Touch Panel X position control */
+#define PA_TPYPOS	(PA_BCR+0x0108)	/* Touch Panel Y position control */
+#define PA_DBDET	(PA_BCR+0x0200)	/* Debug Board detect control */
+#define PA_DBDISPCTL	(PA_BCR+0x0202)	/* Debug Board Dot timing control */
+#define PA_DBSW		(PA_BCR+0x0204)	/* Debug Board Switch control */
+#define PA_CFCTL	(PA_BCR+0x0300)	/* CF Timing control */
+#define PA_CFPOW	(PA_BCR+0x0302)	/* CF Power control */
+#define PA_CFCDINTCLR	(PA_BCR+0x0304)	/* CF Insert Interrupt clear */
+#define PA_SCSMR	(PA_BCR+0x0400)	/* SCIF Serial mode control */
+#define PA_SCBRR	(PA_BCR+0x0402)	/* SCIF Bit rate control */
+#define PA_SCSCR	(PA_BCR+0x0404)	/* SCIF Serial control */
+#define PA_SCFDTR	(PA_BCR+0x0406)	/* SCIF Send FIFO control */
+#define PA_SCFSR	(PA_BCR+0x0408)	/* SCIF Serial status control */
+#define PA_SCFRDR	(PA_BCR+0x040a)	/* SCIF Receive FIFO control */
+#define PA_SCFCR	(PA_BCR+0x040c)	/* SCIF FIFO control */
+#define PA_SCFDR	(PA_BCR+0x040e)	/* SCIF FIFO data control */
+#define PA_SCLSR	(PA_BCR+0x0412)	/* SCIF Line Status control */
+#define PA_ICCR		(PA_BCR+0x0500)	/* Serial control */
+#define PA_SAR		(PA_BCR+0x0502)	/* Serial Slave control */
+#define PA_MDR		(PA_BCR+0x0504)	/* Serial Mode control */
+#define PA_ADR1		(PA_BCR+0x0506)	/* Serial Address1 control */
+#define PA_DAR1		(PA_BCR+0x0546)	/* Serial Data1 control */
+#define PA_VERREG	(PA_BCR+0x0600)	/* FPGA Version Register */
+
+#define PA_AX88796L	0xa5800400	/* AX88796L Area */
+#define PA_SC1602BSLB	0xa6000000	/* SC1602BSLB Area */
+#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
+#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
+#define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
+#define AX88796L_IO_BASE	0x1000	/* AX88796L IO Base Address */
+
+#define IRLCNTR1	(PA_BCR + 0)	/* Interrupt Control Register1 */
+
+#define IRQ_PCISLOT1	0		/* PCI Slot #1 IRQ */
+#define IRQ_PCISLOT2	1		/* PCI Slot #2 IRQ */
+#define IRQ_PCISLOT3	2		/* PCI Slot #3 IRQ */
+#define IRQ_PCISLOT4	3		/* PCI Slot #4 IRQ */
+#define IRQ_CFCARD	4		/* CF Card IRQ */
+#define IRQ_CFINST	5		/* CF Card Insert IRQ */
+#define IRQ_M66596	6		/* M66596 IRQ */
+#define IRQ_SDCARD	7		/* SD Card IRQ */
+#define IRQ_TUCHPANEL	8		/* Touch Panel IRQ */
+#define IRQ_SCI		9		/* SCI IRQ */
+#define IRQ_2SERIAL	10		/* Serial IRQ */
+#define	IRQ_EXTENTION	11		/* EXTn IRQ */
+#define IRQ_ONETH	12		/* On board Ethernet IRQ */
+#define IRQ_PSW		13		/* Push Switch IRQ */
+#define IRQ_ZIGBEE	14		/* Ziggbee IO IRQ */
+
+#endif  /* CONFIG_SH_R7780MP */
+
+#define __IO_PREFIX	r7780rp
+#include <asm/io_generic.h>
+
+#endif  /* __ASM_SH_RENESAS_R7780RP */
diff --git a/include/asm-sh/rts7751r2d/rts7751r2d.h b/include/asm-sh/rts7751r2d/rts7751r2d.h
index 4e09ba597e9a..b112ae221fd1 100644
--- a/include/asm-sh/rts7751r2d/rts7751r2d.h
+++ b/include/asm-sh/rts7751r2d/rts7751r2d.h
@@ -41,8 +41,6 @@
 
 #define PA_AX88796L	0xaa000400	/* AX88796L Area */
 #define PA_VOYAGER	0xab000000	/* VOYAGER GX Area */
-#define PA_AREA5_IO	0xb4000000	/* Area 5 IO Memory */
-#define PA_AREA6_IO	0xb8000000	/* Area 6 IO Memory */
 #define PA_IDE_OFFSET	0x1f0		/* CF IDE Offset */
 #define AX88796L_IO_BASE	0x1000	/* AX88796L IO Base Address */
 
-- 
cgit v1.2.1


From f1517494407b1f1ca0063a756cc30d75e96d433c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:01:12 +0900
Subject: sh: Cleanup and document register bank usage.

Initial register bank cleanup. Make SR.RB configurable, and add some
preliminary documentation on register bank usage within the kernel.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 Documentation/sh/register-banks.txt | 33 +++++++++++++++++++++++++++++++++
 include/asm-sh/system.h             | 17 ++++++++++++++++-
 include/asm-sh/thread_info.h        | 17 +++++++++++++++--
 3 files changed, 64 insertions(+), 3 deletions(-)
 create mode 100644 Documentation/sh/register-banks.txt

diff --git a/Documentation/sh/register-banks.txt b/Documentation/sh/register-banks.txt
new file mode 100644
index 000000000000..a6719f2f6594
--- /dev/null
+++ b/Documentation/sh/register-banks.txt
@@ -0,0 +1,33 @@
+	Notes on register bank usage in the kernel
+	==========================================
+
+Introduction
+------------
+
+The SH-3 and SH-4 CPU families traditionally include a single partial register
+bank (selected by SR.RB, only r0 ... r7 are banked), whereas other families
+may have more full-featured banking or simply no such capabilities at all.
+
+SR.RB banking
+-------------
+
+In the case of this type of banking, banked registers are mapped directly to
+r0 ... r7 if SR.RB is set to the bank we are interested in, otherwise ldc/stc
+can still be used to reference the banked registers (as r0_bank ... r7_bank)
+when in the context of another bank. The developer must keep the SR.RB value
+in mind when writing code that utilizes these banked registers, for obvious
+reasons. Userspace is also not able to poke at the bank1 values, so these can
+be used rather effectively as scratch registers by the kernel.
+
+Presently the kernel uses several of these registers.
+
+	- r0_bank, r1_bank (referenced as k0 and k1, used for scratch
+	  registers when doing exception handling).
+	- r2_bank (used to track the EXPEVT/INTEVT code)
+		- Used by do_IRQ() and friends for doing irq mapping based off
+		  of the interrupt exception vector jump table offset
+	- r6_bank (global interrupt mask)
+		- The SR.IMASK interrupt handler makes use of this to set the
+		  interrupt priority level (used by local_irq_enable())
+	- r7_bank (current)
+
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 1630a5411e5f..198d17e3069a 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -136,7 +136,8 @@ extern void __xchg_called_with_bad_pointer(void);
 #define set_mb(var, value) do { xchg(&var, value); } while (0)
 
 /* Interrupt Control */
-static __inline__ void local_irq_enable(void)
+#ifdef CONFIG_CPU_HAS_SR_RB
+static inline void local_irq_enable(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -149,6 +150,20 @@ static __inline__ void local_irq_enable(void)
 			     : "1" (~0x000000f0)
 			     : "memory");
 }
+#else
+static inline void local_irq_enable(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ (
+		"stc	sr, %0\n\t"
+		"and	%1, %0\n\t"
+		"ldc	%0, sr\n\t"
+		: "=&r" (__dummy0), "=r" (__dummy1)
+		: "1" (~0x000000f0)
+		: "memory");
+}
+#endif
 
 static __inline__ void local_irq_disable(void)
 {
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 7345350d98c0..f64dd803a014 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -48,16 +48,29 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
+#define THREAD_SIZE (2*PAGE_SIZE)
+
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
 	struct thread_info *ti;
+#ifdef CONFIG_CPU_HAS_SR_RB
 	__asm__("stc	r7_bank, %0" : "=r" (ti));
+#else
+	unsigned long __dummy;
+
+	__asm__ __volatile__ (
+		"mov	r15, %0\n\t"
+		"and	%1, %0\n\t"
+		: "=&r" (ti), "=r" (__dummy)
+		: "1" (~(THREAD_SIZE - 1))
+		: "memory");
+#endif
+
 	return ti;
 }
 
 /* thread information allocation */
-#define THREAD_SIZE (2*PAGE_SIZE)
 #define alloc_thread_info(ti) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
 #define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
 
@@ -65,7 +78,7 @@ static inline struct thread_info *current_thread_info(void)
 
 /* how to get the thread information struct from ASM */
 #define GET_THREAD_INFO(reg) \
-	stc	r7_bank, reg
+	stc     r7_bank, reg
 
 #endif
 
-- 
cgit v1.2.1


From d2b06a8b17f96b75fa1e8e7765cb900c99fd80fb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:03:25 +0900
Subject: video: Update pvr2fb for sq API changes.

With the store queue API rework, we need to change the in-kernel
users too.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/video/pvr2fb.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/video/pvr2fb.c b/drivers/video/pvr2fb.c
index 940ba2be55e9..78dc59a1751b 100644
--- a/drivers/video/pvr2fb.c
+++ b/drivers/video/pvr2fb.c
@@ -187,7 +187,7 @@ static short do_blank = 0;		/* (Un)Blank the screen */
 static unsigned int is_blanked = 0;		/* Is the screen blanked? */
 
 #ifdef CONFIG_SH_STORE_QUEUES
-static struct sq_mapping *pvr2fb_map;
+static unsigned long pvr2fb_map;
 #endif
 
 #ifdef CONFIG_SH_DMA
@@ -213,15 +213,17 @@ static irqreturn_t pvr2fb_interrupt(int irq, void *dev_id, struct pt_regs *fp);
 static int pvr2_init_cable(void);
 static int pvr2_get_param(const struct pvr2_params *p, const char *s,
                             int val, int size);
+#ifdef CONFIG_SH_DMA
 static ssize_t pvr2fb_write(struct file *file, const char *buf,
 			    size_t count, loff_t *ppos);
+#endif
 
 static struct fb_ops pvr2fb_ops = {
-	.owner 		= THIS_MODULE,
-	.fb_setcolreg 	= pvr2fb_setcolreg,
-	.fb_blank 	= pvr2fb_blank,
-	.fb_check_var 	= pvr2fb_check_var,
-	.fb_set_par 	= pvr2fb_set_par,
+	.owner		= THIS_MODULE,
+	.fb_setcolreg	= pvr2fb_setcolreg,
+	.fb_blank	= pvr2fb_blank,
+	.fb_check_var	= pvr2fb_check_var,
+	.fb_set_par	= pvr2fb_set_par,
 #ifdef CONFIG_SH_DMA
 	.fb_write	= pvr2fb_write,
 #endif
@@ -783,7 +785,7 @@ static int __init pvr2fb_common_init(void)
 		goto out_err;
 	}
 
-	fb_memset((unsigned long)fb_info->screen_base, 0, pvr2_fix.smem_len);
+	fb_memset(fb_info->screen_base, 0, pvr2_fix.smem_len);
 
 	pvr2_fix.ypanstep	= nopan  ? 0 : 1;
 	pvr2_fix.ywrapstep	= nowrap ? 0 : 1;
@@ -820,7 +822,7 @@ static int __init pvr2fb_common_init(void)
 	       modememused >> 10, (unsigned long)(fb_info->fix.smem_len >> 10));
 	printk("fb%d: Mode %dx%d-%d pitch = %ld cable: %s video output: %s\n", 
 	       fb_info->node, fb_info->var.xres, fb_info->var.yres,
-	       fb_info->var.bits_per_pixel, 
+	       fb_info->var.bits_per_pixel,
 	       get_line_length(fb_info->var.xres, fb_info->var.bits_per_pixel),
 	       (char *)pvr2_get_param(cables, NULL, cable_type, 3),
 	       (char *)pvr2_get_param(outputs, NULL, video_output, 3));
@@ -829,10 +831,10 @@ static int __init pvr2fb_common_init(void)
 	printk(KERN_NOTICE "fb%d: registering with SQ API\n", fb_info->node);
 
 	pvr2fb_map = sq_remap(fb_info->fix.smem_start, fb_info->fix.smem_len,
-			      fb_info->fix.id);
+			      fb_info->fix.id, pgprot_val(PAGE_SHARED));
 
 	printk(KERN_NOTICE "fb%d: Mapped video memory to SQ addr 0x%lx\n",
-	       fb_info->node, pvr2fb_map->sq_addr);
+	       fb_info->node, pvr2fb_map);
 #endif
 
 	return 0;
-- 
cgit v1.2.1


From bc8bff63bacea47561de34e04a17c79846ecfe91 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:04:51 +0900
Subject: sh: Drop incdir rule for SE7751.

No longer needed..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index c143510d71fc..50c5cc31c302 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -111,8 +111,6 @@ machdir-$(CONFIG_SH_TITAN)			:= titan
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
-
-incdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se7751
 incdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 
 ifneq ($(machdir-y),)
-- 
cgit v1.2.1


From 00b3aa3fc9bd827caaa859de90d9eba831b77d40 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:05:56 +0900
Subject: sh: xchg()/__xchg() always_inline fixes for gcc4.

Make __xchg() a macro, so that gcc 4.0 doesn't blow up thanks to
always_inline..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/system.h | 57 +++++++++++++++++++++++++++----------------------
 1 file changed, 32 insertions(+), 25 deletions(-)

diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 198d17e3069a..bd7dc0554b11 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -79,10 +79,8 @@ static inline void sched_cacheflush(void)
 }
 #endif
 
-#define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
-
 static __inline__ unsigned long tas(volatile int *m)
-{ /* #define tas(ptr) (xchg((ptr),1)) */
+{
 	unsigned long retval;
 
 	__asm__ __volatile__ ("tas.b	@%1\n\t"
@@ -91,8 +89,6 @@ static __inline__ unsigned long tas(volatile int *m)
 	return retval;
 }
 
-extern void __xchg_called_with_bad_pointer(void);
-
 /*
  * A brief note on ctrl_barrier(), the control register write barrier.
  *
@@ -220,17 +216,17 @@ static __inline__ void  local_irq_restore(unsigned long x)
 	}
 }
 #else
-#define local_irq_restore(x) do { 			\
+#define local_irq_restore(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();				\
+		local_irq_enable();			\
 } while (0)
 #endif
 
-#define really_restore_flags(x) do { 			\
+#define really_restore_flags(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();				\
+		local_irq_enable();			\
 	else						\
-		local_irq_disable();				\
+		local_irq_disable();			\
 } while (0)
 
 /*
@@ -272,7 +268,7 @@ do {							\
 /* For spinlocks etc */
 #define local_irq_save(x)	x = local_irq_save()
 
-static __inline__ unsigned long xchg_u32(volatile int * m, unsigned long val)
+static inline unsigned long xchg_u32(volatile u32 *m, unsigned long val)
 {
 	unsigned long flags, retval;
 
@@ -283,7 +279,7 @@ static __inline__ unsigned long xchg_u32(volatile int * m, unsigned long val)
 	return retval;
 }
 
-static __inline__ unsigned long xchg_u8(volatile unsigned char * m, unsigned long val)
+static inline unsigned long xchg_u8(volatile u8 *m, unsigned long val)
 {
 	unsigned long flags, retval;
 
@@ -294,19 +290,30 @@ static __inline__ unsigned long xchg_u8(volatile unsigned char * m, unsigned lon
 	return retval;
 }
 
-static __inline__ unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
-{
-	switch (size) {
-	case 4:
-		return xchg_u32(ptr, x);
-		break;
-	case 1:
-		return xchg_u8(ptr, x);
-		break;
-	}
-	__xchg_called_with_bad_pointer();
-	return x;
-}
+extern void __xchg_called_with_bad_pointer(void);
+
+#define __xchg(ptr, x, size)				\
+({							\
+	unsigned long __xchg__res;			\
+	volatile void *__xchg_ptr = (ptr);		\
+	switch (size) {					\
+	case 4:						\
+		__xchg__res = xchg_u32(__xchg_ptr, x);	\
+		break;					\
+	case 1:						\
+		__xchg__res = xchg_u8(__xchg_ptr, x);	\
+		break;					\
+	default:					\
+		__xchg_called_with_bad_pointer();	\
+		__xchg__res = x;			\
+		break;					\
+	}						\
+							\
+	__xchg__res;					\
+})
+
+#define xchg(ptr,x)	\
+	((__typeof__(*(ptr)))__xchg((ptr),(unsigned long)(x), sizeof(*(ptr))))
 
 static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
 	unsigned long new)
-- 
cgit v1.2.1


From ae31825e63e11188d96ad9c4d2d5d4e0fc798a75 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:08:24 +0900
Subject: sh: Make O= builds work again.

Some of the paths were a bit broken, fix it up.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Makefile | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 50c5cc31c302..859285d47a29 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -135,17 +135,14 @@ boot := arch/sh/boot
 
 CPPFLAGS_vmlinux.lds := -traditional
 
-ifneq ($(KBUILD_SRC),)
 incdir-prefix	:= $(srctree)/include/asm-sh/
-else
-incdir-prefix	:=
-endif
 
 #	Update machine arch and proc symlinks if something which affects
 #	them changed.  We use .arch and .mach to indicate when they were
 #	updated last, otherwise make uses the target directory mtime.
 
-include/asm-sh/.cpu: $(wildcard include/config/cpu/*.h) include/config/auto.conf
+include/asm-sh/.cpu: $(wildcard include/config/cpu/*.h) \
+		     include/config/auto.conf FORCE
 	@echo '  SYMLINK include/asm-sh/cpu -> include/asm-sh/$(cpuincdir-y)'
 	$(Q)if [ ! -d include/asm-sh ]; then mkdir -p include/asm-sh; fi
 	$(Q)ln -fsn $(incdir-prefix)$(cpuincdir-y) include/asm-sh/cpu
@@ -155,7 +152,8 @@ include/asm-sh/.cpu: $(wildcard include/config/cpu/*.h) include/config/auto.conf
 #	don't, just reference the parent directory so the semantics are
 #	kept roughly the same.
 
-include/asm-sh/.mach: $(wildcard include/config/sh/*.h) include/config/auto.conf
+include/asm-sh/.mach: $(wildcard include/config/sh/*.h) \
+		      include/config/auto.conf FORCE
 	@echo -n '  SYMLINK include/asm-sh/mach -> '
 	$(Q)if [ ! -d include/asm-sh ]; then mkdir -p include/asm-sh; fi
 	$(Q)if [ -d $(incdir-prefix)$(incdir-y) ]; then \
@@ -168,7 +166,7 @@ include/asm-sh/.mach: $(wildcard include/config/sh/*.h) include/config/auto.conf
 	fi
 	@touch $@
 
-archprepare: maketools include/asm-sh/.cpu include/asm-sh/.mach
+archprepare: include/asm-sh/.cpu include/asm-sh/.mach maketools
 
 PHONY += maketools FORCE
 maketools:  include/linux/version.h FORCE
-- 
cgit v1.2.1


From 848dd265947c976b6340027088c090ff7f0a0b8b Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 16:09:59 +0900
Subject: video: Update header location in hp680_bl.

Trivial build fix.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/video/backlight/hp680_bl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/backlight/hp680_bl.c b/drivers/video/backlight/hp680_bl.c
index ffc72ae3ada8..fe1488374f62 100644
--- a/drivers/video/backlight/hp680_bl.c
+++ b/drivers/video/backlight/hp680_bl.c
@@ -20,7 +20,7 @@
 
 #include <asm/cpu/dac.h>
 #include <asm/hp6xx/hp6xx.h>
-#include <asm/hd64461/hd64461.h>
+#include <asm/hd64461.h>
 
 #define HP680_MAX_INTENSITY 255
 #define HP680_DEFAULT_INTENSITY 10
@@ -163,6 +163,6 @@ static void __exit hp680bl_exit(void)
 module_init(hp680bl_init);
 module_exit(hp680bl_exit);
 
-MODULE_AUTHOR("Andriy Skulysh <askulysh@image.kiev.ua>");
+MODULE_AUTHOR("Andriy Skulysh <askulysh@gmail.com>");
 MODULE_DESCRIPTION("HP Jornada 680 Backlight Driver");
 MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From a328ff9a7e3109919f59ce1903a31f8b2e6339e5 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:14:54 +0900
Subject: sh: SE73180 updates for IRQ changes.

SE73180 can use the generic support, we just need to
wire up the IRQ demuxing.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/se/73180/setup.c   | 2 ++
 arch/sh/configs/se73180_defconfig | 1 +
 arch/sh/mm/Kconfig                | 2 +-
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index cddc7a2b858f..4daf53b1457f 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -12,6 +12,7 @@
 #include <linux/init.h>
 #include <asm/machvec.h>
 #include <asm/se73180.h>
+#include <asm/irq.h>
 
 void heartbeat_73180se(void);
 void init_73180se_IRQ(void);
@@ -50,6 +51,7 @@ struct sh_machine_vector mv_73180se __initmv = {
 	.mv_outsl = sh73180se_outsl,
 
 	.mv_init_irq = init_73180se_IRQ,
+	.mv_irq_demux = shmse_irq_demux,
 #ifdef CONFIG_HEARTBEAT
 	.mv_heartbeat = heartbeat_73180se,
 #endif
diff --git a/arch/sh/configs/se73180_defconfig b/arch/sh/configs/se73180_defconfig
index d217e44c89a6..fe19feb95ca9 100644
--- a/arch/sh/configs/se73180_defconfig
+++ b/arch/sh/configs/se73180_defconfig
@@ -78,6 +78,7 @@ CONFIG_SH_73180_SOLUTION_ENGINE=y
 # CONFIG_SH_SECUREEDGE5410 is not set
 # CONFIG_SH_HS7751RVOIP is not set
 # CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
 # CONFIG_SH_EDOSK7705 is not set
 # CONFIG_SH_SH4202_MICRODEV is not set
 # CONFIG_SH_UNKNOWN is not set
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index fb586b1cf8bb..9064dfc91e6d 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -20,7 +20,6 @@ config CPU_SH4
 config CPU_SH4A
 	bool
 	select CPU_SH4
-	select CPU_HAS_INTC2_IRQ
 
 config CPU_SUBTYPE_ST40
 	bool
@@ -144,6 +143,7 @@ config CPU_SUBTYPE_SH7770
 config CPU_SUBTYPE_SH7780
 	bool "Support SH7780 processor"
 	select CPU_SH4A
+	select CPU_HAS_INTC2_IRQ
 
 endmenu
 
-- 
cgit v1.2.1


From 801e045860586b6a4ca21b3269d111dfca7ef17b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:15:48 +0900
Subject: sh: Update new-machine.txt so it's more accurate.

This fell behind a bit, get it updated so the documentation
has something in common with reality.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 Documentation/sh/new-machine.txt | 128 ++++++++++++++++-----------------------
 1 file changed, 51 insertions(+), 77 deletions(-)

diff --git a/Documentation/sh/new-machine.txt b/Documentation/sh/new-machine.txt
index eb2dd2e6993b..73988e0d112b 100644
--- a/Documentation/sh/new-machine.txt
+++ b/Documentation/sh/new-machine.txt
@@ -41,11 +41,6 @@ Board-specific code:
         |
 	.. more boards here ...
 
-It should also be noted that each board is required to have some certain
-headers. At the time of this writing, io.h is the only thing that needs
-to be provided for each board, and can generally just reference generic
-functions (with the exception of isa_port2addr).
-
 Next, for companion chips:
 .
 `-- arch
@@ -104,12 +99,13 @@ and then populate that with sub-directories for each member of the family.
 Both the Solution Engine and the hp6xx boards are an example of this.
 
 After you have setup your new arch/sh/boards/ directory, remember that you
-also must add a directory in include/asm-sh for headers localized to this
-board. In order to interoperate seamlessly with the build system, it's best
-to have this directory the same as the arch/sh/boards/ directory name,
-though if your board is again part of a family, the build system has ways
-of dealing with this, and you can feel free to name the directory after
-the family member itself.
+should also add a directory in include/asm-sh for headers localized to this
+board (if there are going to be more than one). In order to interoperate
+seamlessly with the build system, it's best to have this directory the same
+as the arch/sh/boards/ directory name, though if your board is again part of
+a family, the build system has ways of dealing with this (via incdir-y
+overloading), and you can feel free to name the directory after the family
+member itself.
 
 There are a few things that each board is required to have, both in the
 arch/sh/boards and the include/asm-sh/ heirarchy. In order to better
@@ -122,6 +118,7 @@ might look something like:
  * arch/sh/boards/vapor/setup.c - Setup code for imaginary board
  */
 #include <linux/init.h>
+#include <asm/rtc.h> /* for board_time_init() */
 
 const char *get_system_type(void)
 {
@@ -152,79 +149,57 @@ int __init platform_setup(void)
 }
 
 Our new imaginary board will also have to tie into the machvec in order for it
-to be of any use. Currently the machvec is slowly on its way out, but is still
-required for the time being. As such, let us take a look at what needs to be
-done for the machvec assignment.
+to be of any use.
 
 machvec functions fall into a number of categories:
 
  - I/O functions to IO memory (inb etc) and PCI/main memory (readb etc).
- - I/O remapping functions (ioremap etc)
- - some initialisation functions
- - a 'heartbeat' function
- - some miscellaneous flags
-
-The tree can be built in two ways:
- - as a fully generic build. All drivers are linked in, and all functions
-   go through the machvec
- - as a machine specific build. In this case only the required drivers
-   will be linked in, and some macros may be redefined to not go through
-   the machvec where performance is important (in particular IO functions).
-
-There are three ways in which IO can be performed:
- - none at all. This is really only useful for the 'unknown' machine type,
-   which us designed to run on a machine about which we know nothing, and
-   so all all IO instructions do nothing.
- - fully custom. In this case all IO functions go to a machine specific
-   set of functions which can do what they like
- - a generic set of functions. These will cope with most situations,
-   and rely on a single function, mv_port2addr, which is called through the
-   machine vector, and converts an IO address into a memory address, which
-   can be read from/written to directly.
-
-Thus adding a new machine involves the following steps (I will assume I am
-adding a machine called vapor):
-
- - add a new file include/asm-sh/vapor/io.h which contains prototypes for
+ - I/O mapping functions (ioport_map, ioport_unmap, etc).
+ - a 'heartbeat' function.
+ - PCI and IRQ initialization routines.
+ - Consistent allocators (for boards that need special allocators,
+   particularly for allocating out of some board-specific SRAM for DMA
+   handles).
+
+There are machvec functions added and removed over time, so always be sure to
+consult include/asm-sh/machvec.h for the current state of the machvec.
+
+The kernel will automatically wrap in generic routines for undefined function
+pointers in the machvec at boot time, as machvec functions are referenced
+unconditionally throughout most of the tree. Some boards have incredibly
+sparse machvecs (such as the dreamcast and sh03), whereas others must define
+virtually everything (rts7751r2d).
+
+Adding a new machine is relatively trivial (using vapor as an example):
+
+If the board-specific definitions are quite minimalistic, as is the case for
+the vast majority of boards, simply having a single board-specific header is
+sufficient.
+
+ - add a new file include/asm-sh/vapor.h which contains prototypes for
    any machine specific IO functions prefixed with the machine name, for
    example vapor_inb. These will be needed when filling out the machine
    vector.
 
-   This is the minimum that is required, however there are ample
-   opportunities to optimise this. In particular, by making the prototypes
-   inline function definitions, it is possible to inline the function when
-   building machine specific versions. Note that the machine vector
-   functions will still be needed, so that a module built for a generic
-   setup can be loaded.
-
- - add a new file arch/sh/boards/vapor/mach.c. This contains the definition
-   of the machine vector. When building the machine specific version, this
-   will be the real machine vector (via an alias), while in the generic
-   version is used to initialise the machine vector, and then freed, by
-   making it initdata. This should be defined as:
-
-     struct sh_machine_vector mv_vapor __initmv = {
-       .mv_name = "vapor",
-     }
-     ALIAS_MV(vapor)
-
- - finally add a file arch/sh/boards/vapor/io.c, which contains
-   definitions of the machine specific io functions.
-
-A note about initialisation functions. Three initialisation functions are
-provided in the machine vector:
- - mv_arch_init - called very early on from setup_arch
- - mv_init_irq - called from init_IRQ, after the generic SH interrupt
-   initialisation
- - mv_init_pci - currently not used
-
-Any other remaining functions which need to be called at start up can be
-added to the list using the __initcalls macro (or module_init if the code
-can be built as a module). Many generic drivers probe to see if the device
-they are targeting is present, however this may not always be appropriate,
-so a flag can be added to the machine vector which will be set on those
-machines which have the hardware in question, reducing the probe to a
-single conditional.
+   Note that these prototypes are generated automatically by setting
+   __IO_PREFIX to something sensible. A typical example would be:
+
+	#define __IO_PREFIX vapor
+   	#include <asm/io_generic.h>
+
+   somewhere in the board-specific header. Any boards being ported that still
+   have a legacy io.h should remove it entirely and switch to the new model.
+
+ - Add machine vector definitions to the board's setup.c. At a bare minimum,
+   this must be defined as something like:
+
+	struct sh_machine_vector mv_vapor __initmv = {
+		.mv_name = "vapor",
+	};
+	ALIAS_MV(vapor)
+
+ - finally add a file arch/sh/boards/vapor/io.c, which contains definitions of
+   the machine specific io functions (if there are enough to warrant it).
 
 3. Hooking into the Build System
 ================================
@@ -303,4 +278,3 @@ which will in turn copy the defconfig for this board, run it through
 oldconfig (prompting you for any new options since the time of creation),
 and start you on your way to having a functional kernel for your new
 board.
-
-- 
cgit v1.2.1


From ef48e8e3498605351f91f195cc9af0ef981b0dde Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:17:17 +0900
Subject: sh: Free up some and document PTEL flags.

Drop _PAGE_SHARED/_PAGE_U0_SHARED and document Linux PTE encodings in
the PTEL value. Preserve the swap cache entry encoding semantics for
now, though it will need rework to free up _PAGE_WT from _PAGE_FILE.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/pgtable.h | 76 ++++++++++++++++++++++++++++--------------------
 1 file changed, 45 insertions(+), 31 deletions(-)

diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 9728b58f7c13..41c559d8ba87 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -54,20 +54,41 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define VMALLOC_START	(P3SEG+0x00100000)
 #define VMALLOC_END	(FIXADDR_START-2*PAGE_SIZE)
 
-#define	_PAGE_WT	0x001  /* WT-bit on SH-4, 0 on SH-3 */
-#define _PAGE_HW_SHARED	0x002  /* SH-bit  : page is shared among processes */
-#define _PAGE_DIRTY	0x004  /* D-bit   : page changed */
-#define _PAGE_CACHABLE	0x008  /* C-bit   : cachable */
-#define _PAGE_SZ0	0x010  /* SZ0-bit : Size of page */
-#define _PAGE_RW	0x020  /* PR0-bit : write access allowed */
-#define _PAGE_USER	0x040  /* PR1-bit : user space access allowed */
-#define _PAGE_SZ1	0x080  /* SZ1-bit : Size of page (on SH-4) */
-#define _PAGE_PRESENT	0x100  /* V-bit   : page is valid */
-#define _PAGE_PROTNONE	0x200  /* software: if not present  */
-#define _PAGE_ACCESSED	0x400  /* software: page referenced */
-#define _PAGE_U0_SHARED 0x800  /* software: page is shared in user space */
-
-#define	_PAGE_FILE	_PAGE_WT  /* software: pagecache or swap? */
+/*
+ * Linux PTEL encoding.
+ *
+ * Hardware and software bit definitions for the PTEL value:
+ *
+ * - Bits 0 and 7 are reserved on SH-3 (_PAGE_WT and _PAGE_SZ1 on SH-4).
+ *
+ * - Bit 1 is the SH-bit, but is unused on SH-3 due to an MMU bug (the
+ *   hardware PTEL value can't have the SH-bit set when MMUCR.IX is set,
+ *   which is the default in cpu-sh3/mmu_context.h:MMU_CONTROL_INIT).
+ *
+ *   In order to keep this relatively clean, do not use these for defining
+ *   SH-3 specific flags until all of the other unused bits have been
+ *   exhausted.
+ *
+ * - Bit 9 is reserved by everyone and used by _PAGE_PROTNONE.
+ *
+ * - Bits 10 and 11 are low bits of the PPN that are reserved on >= 4K pages.
+ *   Bit 10 is used for _PAGE_ACCESSED, bit 11 remains unused.
+ *
+ * - Bits 31, 30, and 29 remain unused by everyone and can be used for future
+ *   software flags, although care must be taken to update _PAGE_CLEAR_FLAGS.
+ */
+#define	_PAGE_WT	0x001		/* WT-bit on SH-4, 0 on SH-3 */
+#define _PAGE_HW_SHARED	0x002		/* SH-bit  : shared among processes */
+#define _PAGE_DIRTY	0x004		/* D-bit   : page changed */
+#define _PAGE_CACHABLE	0x008		/* C-bit   : cachable */
+#define _PAGE_SZ0	0x010		/* SZ0-bit : Size of page */
+#define _PAGE_RW	0x020		/* PR0-bit : write access allowed */
+#define _PAGE_USER	0x040		/* PR1-bit : user space access allowed */
+#define _PAGE_SZ1	0x080		/* SZ1-bit : Size of page (on SH-4) */
+#define _PAGE_PRESENT	0x100		/* V-bit   : page is valid */
+#define _PAGE_PROTNONE	0x200		/* software: if not present  */
+#define _PAGE_ACCESSED	0x400		/* software: page referenced */
+#define _PAGE_FILE	_PAGE_WT	/* software: pagecache or swap? */
 
 /* software: moves to PTEA.TC (Timing Control) */
 #define _PAGE_PCC_AREA5	0x00000000	/* use BSC registers for area5 */
@@ -82,22 +103,17 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define _PAGE_PCC_ATR8	0x60000000	/* Attribute Memory space, 8 bit bus */
 #define _PAGE_PCC_ATR16	0x60000001	/* Attribute Memory space, 6 bit bus */
 
-/* Mask which drop software flags
- * We also drop WT bit since it is used for _PAGE_FILE
- * bit in this implementation.
- */
-#define _PAGE_CLEAR_FLAGS	(_PAGE_WT | _PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_U0_SHARED)
-
-#if defined(CONFIG_CPU_SH3)
-/*
- * MMU on SH-3 has bug on SH-bit: We can't use it if MMUCR.IX=1.
- * Work around: Just drop SH-bit.
- */
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS | _PAGE_HW_SHARED))
+/* Mask which drops unused bits from the PTEL value */
+#ifdef CONFIG_CPU_SH3
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED| \
+				 _PAGE_FILE	| _PAGE_SZ1	| \
+				 _PAGE_HW_SHARED)
 #else
-#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+#define _PAGE_CLEAR_FLAGS	(_PAGE_PROTNONE | _PAGE_ACCESSED | _PAGE_FILE)
 #endif
 
+#define _PAGE_FLAGS_HARDWARE_MASK	(0x1fffffff & ~(_PAGE_CLEAR_FLAGS))
+
 /* Hardware flags: SZ0=1 (4k-byte) */
 #define _PAGE_FLAGS_HARD	_PAGE_SZ0
 
@@ -107,17 +123,15 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
 #define _PAGE_SZHUGE	(_PAGE_SZ0 | _PAGE_SZ1)
 #endif
 
-#define _PAGE_SHARED	_PAGE_U0_SHARED
-
 #define _PAGE_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define _KERNPG_TABLE	(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
-#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_SHARED)
+#define _PAGE_CHG_MASK	(PTE_MASK | _PAGE_ACCESSED | _PAGE_CACHABLE | _PAGE_DIRTY)
 
 #ifndef __ASSEMBLY__
 
 #ifdef CONFIG_MMU
 #define PAGE_NONE	__pgprot(_PAGE_PROTNONE | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
-#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_SHARED | _PAGE_FLAGS_HARD)
+#define PAGE_SHARED	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_CACHABLE |_PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_COPY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_READONLY	__pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_CACHABLE | _PAGE_ACCESSED | _PAGE_FLAGS_HARD)
 #define PAGE_KERNEL	__pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_CACHABLE | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_HW_SHARED | _PAGE_FLAGS_HARD)
-- 
cgit v1.2.1


From 3aa770e7972723f479122cf66b529017d2175289 Mon Sep 17 00:00:00 2001
From: Andriy Skulysh <askulysh@gmail.com>
Date: Wed, 27 Sep 2006 16:20:22 +0900
Subject: sh: APM/PM support.

This adds some simple PM stubs and the basic APM interfaces,
primarily for use by hp6xx, where the existing userland
expects it.

Signed-off-by: Andriy Skulysh <askulysh@gmail.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                            |  10 +
 arch/sh/boards/hp6xx/Makefile              |   5 +-
 arch/sh/boards/hp6xx/hp6xx_apm.c           | 123 +++++++
 arch/sh/boards/hp6xx/pm.c                  |  88 +++++
 arch/sh/boards/hp6xx/pm_wakeup.S           |  58 ++++
 arch/sh/boards/hp6xx/setup.c               |  14 +
 arch/sh/kernel/Makefile                    |   2 +
 arch/sh/kernel/apm.c                       | 539 +++++++++++++++++++++++++++++
 arch/sh/kernel/entry.S                     |   4 +-
 arch/sh/kernel/pm.c                        |  88 +++++
 arch/sh/kernel/sh_ksyms.c                  |   4 +
 arch/sh/kernel/time.c                      |  25 ++
 arch/sh/kernel/timers/timer-tmu.c          |  18 +-
 drivers/input/touchscreen/hp680_ts_input.c |  14 -
 include/asm-sh/apm.h                       |  46 +++
 include/asm-sh/cpu-sh3/freq.h              |   4 +
 include/asm-sh/hd64461.h                   |   7 +-
 include/asm-sh/hp6xx/hp6xx.h               |  53 ++-
 include/asm-sh/pm.h                        |  17 +
 include/asm-sh/system.h                    |  25 ++
 include/asm-sh/timer.h                     |   2 +
 21 files changed, 1124 insertions(+), 22 deletions(-)
 create mode 100644 arch/sh/boards/hp6xx/hp6xx_apm.c
 create mode 100644 arch/sh/boards/hp6xx/pm.c
 create mode 100644 arch/sh/boards/hp6xx/pm_wakeup.S
 create mode 100644 arch/sh/kernel/apm.c
 create mode 100644 arch/sh/kernel/pm.c
 create mode 100644 include/asm-sh/apm.h
 create mode 100644 include/asm-sh/pm.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index fe982f0eba23..d1d724447826 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -638,6 +638,16 @@ source "fs/Kconfig.binfmt"
 
 endmenu
 
+menu "Power management options (EXPERIMENTAL)"
+depends on EXPERIMENTAL
+
+source kernel/power/Kconfig
+
+config APM
+	bool "Advanced Power Management Emulation"
+	depends on PM
+endmenu
+
 source "net/Kconfig"
 
 source "drivers/Kconfig"
diff --git a/arch/sh/boards/hp6xx/Makefile b/arch/sh/boards/hp6xx/Makefile
index 927fe0aa5dfa..fdf98ffcf437 100644
--- a/arch/sh/boards/hp6xx/Makefile
+++ b/arch/sh/boards/hp6xx/Makefile
@@ -2,5 +2,8 @@
 # Makefile for the HP6xx specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o
+obj-y	 		:= mach.o setup.o
+obj-$(CONFIG_PM)	+= pm.o pm_wakeup.o
+obj-$(CONFIG_APM)	+= hp6xx_apm.o
+
 
diff --git a/arch/sh/boards/hp6xx/hp6xx_apm.c b/arch/sh/boards/hp6xx/hp6xx_apm.c
new file mode 100644
index 000000000000..ad0e712c29f6
--- /dev/null
+++ b/arch/sh/boards/hp6xx/hp6xx_apm.c
@@ -0,0 +1,123 @@
+/*
+ * bios-less APM driver for hp680
+ *
+ * Copyright 2005 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/apm_bios.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <asm/io.h>
+#include <asm/apm.h>
+#include <asm/adc.h>
+#include <asm/hp6xx/hp6xx.h>
+
+#define SH7709_PGDR			0xa400012c
+
+#define APM_CRITICAL			10
+#define APM_LOW				30
+
+#define HP680_BATTERY_MAX		875
+#define HP680_BATTERY_MIN		600
+#define HP680_BATTERY_AC_ON		900
+
+#define MODNAME "hp6x0_apm"
+
+static int hp6x0_apm_get_info(char *buf, char **start, off_t fpos, int length)
+{
+	u8 pgdr;
+	char *p;
+	int battery_status;
+	int battery_flag;
+	int ac_line_status;
+	int time_units = APM_BATTERY_LIFE_UNKNOWN;
+
+	int battery = adc_single(ADC_CHANNEL_BATTERY);
+	int backup = adc_single(ADC_CHANNEL_BACKUP);
+	int charging = adc_single(ADC_CHANNEL_CHARGE);
+	int percentage;
+
+	percentage = 100 * (battery - HP680_BATTERY_MIN) /
+			   (HP680_BATTERY_MAX - HP680_BATTERY_MIN);
+
+	ac_line_status = (battery > HP680_BATTERY_AC_ON) ?
+			 APM_AC_ONLINE : APM_AC_OFFLINE;
+
+	p = buf;
+
+	pgdr = ctrl_inb(SH7709_PGDR);
+	if (pgdr & PGDR_MAIN_BATTERY_OUT) {
+		battery_status = APM_BATTERY_STATUS_NOT_PRESENT;
+		battery_flag = 0x80;
+		percentage = -1;
+	} else if (charging < 8 ) {
+		battery_status = APM_BATTERY_STATUS_CHARGING;
+		battery_flag = 0x08;
+		ac_line_status = 0xff;
+	} else if (percentage <= APM_CRITICAL) {
+		battery_status = APM_BATTERY_STATUS_CRITICAL;
+		battery_flag = 0x04;
+	} else if (percentage <= APM_LOW) {
+		battery_status = APM_BATTERY_STATUS_LOW;
+		battery_flag = 0x02;
+	} else {
+		battery_status = APM_BATTERY_STATUS_HIGH;
+		battery_flag = 0x01;
+	}
+
+	p += sprintf(p, "1.0 1.2 0x%02x 0x%02x 0x%02x 0x%02x %d%% %d %s\n",
+		     APM_32_BIT_SUPPORT,
+		     ac_line_status,
+		     battery_status,
+		     battery_flag,
+		     percentage,
+		     time_units,
+		     "min");
+	p += sprintf(p, "bat=%d backup=%d charge=%d\n",
+		     battery, backup, charging);
+
+	return p - buf;
+}
+
+static irqreturn_t hp6x0_apm_interrupt(int irq, void *dev, struct pt_regs *regs)
+{
+	if (!apm_suspended)
+		apm_queue_event(APM_USER_SUSPEND);
+
+	return IRQ_HANDLED;
+}
+
+static int __init hp6x0_apm_init(void)
+{
+	int ret;
+
+	ret = request_irq(HP680_BTN_IRQ, hp6x0_apm_interrupt,
+			  SA_INTERRUPT, MODNAME, 0);
+	if (unlikely(ret < 0)) {
+		printk(KERN_ERR MODNAME ": IRQ %d request failed\n",
+		       HP680_BTN_IRQ);
+		return ret;
+	}
+
+	apm_get_info = hp6x0_apm_get_info;
+
+	return ret;
+}
+
+static void __exit hp6x0_apm_exit(void)
+{
+	free_irq(HP680_BTN_IRQ, 0);
+	apm_get_info = 0;
+}
+
+module_init(hp6x0_apm_init);
+module_exit(hp6x0_apm_exit);
+
+MODULE_AUTHOR("Adriy Skulysh");
+MODULE_DESCRIPTION("hp6xx Advanced Power Management");
+MODULE_LICENSE("GPL");
diff --git a/arch/sh/boards/hp6xx/pm.c b/arch/sh/boards/hp6xx/pm.c
new file mode 100644
index 000000000000..0e501bcbd7a9
--- /dev/null
+++ b/arch/sh/boards/hp6xx/pm.c
@@ -0,0 +1,88 @@
+/*
+ * hp6x0 Power Management Routines
+ *
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/suspend.h>
+#include <linux/errno.h>
+#include <linux/time.h>
+#include <asm/io.h>
+#include <asm/hd64461.h>
+#include <asm/hp6xx/hp6xx.h>
+#include <asm/cpu/dac.h>
+#include <asm/pm.h>
+
+#define STBCR		0xffffff82
+#define STBCR2		0xffffff88
+
+static int hp6x0_pm_enter(suspend_state_t state)
+{
+	u8 stbcr, stbcr2;
+#ifdef CONFIG_HD64461_ENABLER
+	u8 scr;
+	u16 hd64461_stbcr;
+#endif
+
+	if (state != PM_SUSPEND_MEM)
+		return -EINVAL;
+
+#ifdef CONFIG_HD64461_ENABLER
+	outb(0, HD64461_PCC1CSCIER);
+
+	scr = inb(HD64461_PCC1SCR);
+	scr |= HD64461_PCCSCR_VCC1;
+	outb(scr, HD64461_PCC1SCR);
+
+	hd64461_stbcr = inw(HD64461_STBCR);
+	hd64461_stbcr |= HD64461_STBCR_SPC1ST;
+	outw(hd64461_stbcr, HD64461_STBCR);
+#endif
+
+	ctrl_outb(0x1f, DACR);
+
+	stbcr = ctrl_inb(STBCR);
+	ctrl_outb(0x01, STBCR);
+
+	stbcr2 = ctrl_inb(STBCR2);
+	ctrl_outb(0x7f , STBCR2);
+
+	outw(0xf07f, HD64461_SCPUCR);
+
+	pm_enter();
+
+	outw(0, HD64461_SCPUCR);
+	ctrl_outb(stbcr, STBCR);
+	ctrl_outb(stbcr2, STBCR2);
+
+#ifdef CONFIG_HD64461_ENABLER
+	hd64461_stbcr = inw(HD64461_STBCR);
+	hd64461_stbcr &= ~HD64461_STBCR_SPC1ST;
+	outw(hd64461_stbcr, HD64461_STBCR);
+
+	outb(0x4c, HD64461_PCC1CSCIER);
+	outb(0x00, HD64461_PCC1CSCR);
+#endif
+
+	return 0;
+}
+
+/*
+ * Set to PM_DISK_FIRMWARE so we can quickly veto suspend-to-disk.
+ */
+static struct pm_ops hp6x0_pm_ops = {
+	.pm_disk_mode	= PM_DISK_FIRMWARE,
+	.enter		= hp6x0_pm_enter,
+};
+
+static int __init hp6x0_pm_init(void)
+{
+	pm_set_ops(&hp6x0_pm_ops);
+	return 0;
+}
+
+late_initcall(hp6x0_pm_init);
diff --git a/arch/sh/boards/hp6xx/pm_wakeup.S b/arch/sh/boards/hp6xx/pm_wakeup.S
new file mode 100644
index 000000000000..45e9bf0b9115
--- /dev/null
+++ b/arch/sh/boards/hp6xx/pm_wakeup.S
@@ -0,0 +1,58 @@
+/*
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/cpu/mmu_context.h>
+
+#define k0	r0
+#define k1	r1
+#define k2	r2
+#define k3	r3
+#define k4	r4
+
+/*
+ * Kernel mode register usage:
+ *	k0	scratch
+ *	k1	scratch
+ *	k2	scratch (Exception code)
+ *	k3	scratch (Return address)
+ *	k4	scratch
+ *	k5	reserved
+ *	k6	Global Interrupt Mask (0--15 << 4)
+ *	k7	CURRENT_THREAD_INFO (pointer to current thread info)
+ */
+
+ENTRY(wakeup_start)
+! clear STBY bit
+	mov	#-126, k2
+   	and	#127, k0
+	mov.b	k0, @k2
+! enable refresh
+	mov.l	5f, k1
+	mov.w	6f, k0
+  	mov.w	k0, @k1
+! jump to handler
+	mov.l	2f, k2
+	mov.l	3f, k3
+	mov.l	@k2, k2
+
+	mov.l	4f, k1
+	jmp	@k1
+	nop
+
+	.align	2
+1:	.long	EXPEVT
+2:	.long	INTEVT
+3:	.long	ret_from_irq
+4:	.long	handle_exception
+5:	.long	0xffffff68
+6:	.word	0x0524
+
+ENTRY(wakeup_end)
+	nop
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 71f315663cc9..5fc00f0727c4 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -15,6 +15,9 @@
 #include <asm/hp6xx/hp6xx.h>
 #include <asm/cpu/dac.h>
 
+#define	SCPCR	0xa4000116
+#define SCPDR	0xa4000136
+
 const char *get_system_type(void)
 {
 	return "HP6xx";
@@ -24,6 +27,7 @@ int __init platform_setup(void)
 {
 	u8 v8;
 	u16 v;
+
 	v = inw(HD64461_STBCR);
 	v |= HD64461_STBCR_SURTST | HD64461_STBCR_SIRST |
 	    HD64461_STBCR_STM1ST | HD64461_STBCR_STM0ST |
@@ -50,5 +54,15 @@ int __init platform_setup(void)
 	v8 &= ~DACR_DAE;
 	ctrl_outb(v8,DACR);
 
+	v8 = ctrl_inb(SCPDR);
+	v8 |= SCPDR_TS_SCAN_X | SCPDR_TS_SCAN_Y;
+	v8 &= ~SCPDR_TS_SCAN_ENABLE;
+	ctrl_outb(v8, SCPDR);
+
+	v = ctrl_inw(SCPCR);
+	v &= ~SCPCR_TS_MASK;
+	v |= SCPCR_TS_ENABLE;
+	ctrl_outw(v, SCPCR);
+
 	return 0;
 }
diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index 2de82b66af4a..0e2148f63e7e 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -18,3 +18,5 @@ obj-$(CONFIG_SH_CPU_FREQ)	+= cpufreq.o
 obj-$(CONFIG_MODULES)		+= module.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_KEXEC)		+= machine_kexec.o relocate_kernel.o
+obj-$(CONFIG_APM)		+= apm.o
+obj-$(CONFIG_PM)		+= pm.o
diff --git a/arch/sh/kernel/apm.c b/arch/sh/kernel/apm.c
new file mode 100644
index 000000000000..871e7d640002
--- /dev/null
+++ b/arch/sh/kernel/apm.c
@@ -0,0 +1,539 @@
+/*
+ * bios-less APM driver for hp680
+ *
+ * Copyright 2005 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * based on ARM APM driver by
+ *  Jamey Hicks <jamey@crl.dec.com>
+ *
+ * adapted from the APM BIOS driver for Linux by
+ *  Stephen Rothwell (sfr@linuxcare.com)
+ *
+ * APM 1.2 Reference:
+ *   Intel Corporation, Microsoft Corporation. Advanced Power Management
+ *   (APM) BIOS Interface Specification, Revision 1.2, February 1996.
+ *
+ * [This document is available from Microsoft at:
+ *    http://www.microsoft.com/hwdev/busbios/amp_12.htm]
+ */
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/timer.h>
+#include <linux/slab.h>
+#include <linux/proc_fs.h>
+#include <linux/miscdevice.h>
+#include <linux/apm_bios.h>
+#include <linux/pm.h>
+#include <linux/pm_legacy.h>
+#include <asm/apm.h>
+
+#define MODNAME "apm"
+
+/*
+ * The apm_bios device is one of the misc char devices.
+ * This is its minor number.
+ */
+#define APM_MINOR_DEV			134
+
+/*
+ * Maximum number of events stored
+ */
+#define APM_MAX_EVENTS			16
+
+struct apm_queue {
+	unsigned int		event_head;
+	unsigned int		event_tail;
+	apm_event_t		events[APM_MAX_EVENTS];
+};
+
+/*
+ * The per-file APM data
+ */
+struct apm_user {
+	struct list_head	list;
+
+	unsigned int		suser: 1;
+	unsigned int		writer: 1;
+	unsigned int		reader: 1;
+
+	int			suspend_result;
+	unsigned int		suspend_state;
+#define SUSPEND_NONE	0		/* no suspend pending */
+#define SUSPEND_PENDING	1		/* suspend pending read */
+#define SUSPEND_READ	2		/* suspend read, pending ack */
+#define SUSPEND_ACKED	3		/* suspend acked */
+#define SUSPEND_DONE	4		/* suspend completed */
+
+	struct apm_queue	queue;
+};
+
+/*
+ * Local variables
+ */
+static int suspends_pending;
+
+static DECLARE_WAIT_QUEUE_HEAD(apm_waitqueue);
+static DECLARE_WAIT_QUEUE_HEAD(apm_suspend_waitqueue);
+
+/*
+ * This is a list of everyone who has opened /dev/apm_bios
+ */
+static DECLARE_RWSEM(user_list_lock);
+static LIST_HEAD(apm_user_list);
+
+/*
+ * kapmd info.  kapmd provides us a process context to handle
+ * "APM" events within - specifically necessary if we're going
+ * to be suspending the system.
+ */
+static DECLARE_WAIT_QUEUE_HEAD(kapmd_wait);
+static DECLARE_COMPLETION(kapmd_exit);
+static DEFINE_SPINLOCK(kapmd_queue_lock);
+static struct apm_queue kapmd_queue;
+
+int apm_suspended;
+EXPORT_SYMBOL(apm_suspended);
+
+/* Platform-specific apm_read_proc(). */
+int (*apm_get_info)(char *buf, char **start, off_t fpos, int length);
+EXPORT_SYMBOL(apm_get_info);
+
+/*
+ * APM event queue management.
+ */
+static inline int queue_empty(struct apm_queue *q)
+{
+	return q->event_head == q->event_tail;
+}
+
+static inline apm_event_t queue_get_event(struct apm_queue *q)
+{
+	q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	return q->events[q->event_tail];
+}
+
+static void queue_add_event(struct apm_queue *q, apm_event_t event)
+{
+	q->event_head = (q->event_head + 1) % APM_MAX_EVENTS;
+	if (q->event_head == q->event_tail) {
+		static int notified;
+
+		if (notified++ == 0)
+			printk(KERN_ERR "apm: an event queue overflowed\n");
+
+		q->event_tail = (q->event_tail + 1) % APM_MAX_EVENTS;
+	}
+	q->events[q->event_head] = event;
+}
+
+static void queue_event_one_user(struct apm_user *as, apm_event_t event)
+{
+	if (as->suser && as->writer) {
+		switch (event) {
+		case APM_SYS_SUSPEND:
+		case APM_USER_SUSPEND:
+			/*
+			 * If this user already has a suspend pending,
+			 * don't queue another one.
+			 */
+			if (as->suspend_state != SUSPEND_NONE)
+				return;
+
+			as->suspend_state = SUSPEND_PENDING;
+			suspends_pending++;
+			break;
+		}
+	}
+	queue_add_event(&as->queue, event);
+}
+
+static void queue_event(apm_event_t event, struct apm_user *sender)
+{
+	struct apm_user *as;
+
+	down_read(&user_list_lock);
+
+	list_for_each_entry(as, &apm_user_list, list)
+		if (as != sender && as->reader)
+			queue_event_one_user(as, event);
+
+	up_read(&user_list_lock);
+	wake_up_interruptible(&apm_waitqueue);
+}
+
+/**
+ * apm_queue_event - queue an APM event for kapmd
+ * @event: APM event
+ *
+ * Queue an APM event for kapmd to process and ultimately take the
+ * appropriate action.  Only a subset of events are handled:
+ *   %APM_LOW_BATTERY
+ *   %APM_POWER_STATUS_CHANGE
+ *   %APM_USER_SUSPEND
+ *   %APM_SYS_SUSPEND
+ *   %APM_CRITICAL_SUSPEND
+ */
+void apm_queue_event(apm_event_t event)
+{
+	spin_lock_irq(&kapmd_queue_lock);
+	queue_add_event(&kapmd_queue, event);
+	spin_unlock_irq(&kapmd_queue_lock);
+
+	wake_up_interruptible(&kapmd_wait);
+}
+EXPORT_SYMBOL(apm_queue_event);
+
+static void apm_suspend(void)
+{
+	struct apm_user *as;
+	int err;
+
+	apm_suspended = 1;
+	err = pm_suspend(PM_SUSPEND_MEM);
+
+	/*
+	 * Anyone on the APM queues will think we're still suspended.
+	 * Send a message so everyone knows we're now awake again.
+	 */
+	queue_event(APM_NORMAL_RESUME, NULL);
+
+	/*
+	 * Finally, wake up anyone who is sleeping on the suspend.
+	 */
+	down_read(&user_list_lock);
+	list_for_each_entry(as, &apm_user_list, list) {
+		as->suspend_result = err;
+		as->suspend_state = SUSPEND_DONE;
+	}
+	up_read(&user_list_lock);
+
+	wake_up(&apm_suspend_waitqueue);
+	apm_suspended = 0;
+}
+
+static ssize_t apm_read(struct file *fp, char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct apm_user *as = fp->private_data;
+	apm_event_t event;
+	int i = count, ret = 0;
+
+	if (count < sizeof(apm_event_t))
+		return -EINVAL;
+
+	if (queue_empty(&as->queue) && fp->f_flags & O_NONBLOCK)
+		return -EAGAIN;
+
+	wait_event_interruptible(apm_waitqueue, !queue_empty(&as->queue));
+
+	while ((i >= sizeof(event)) && !queue_empty(&as->queue)) {
+		event = queue_get_event(&as->queue);
+
+		ret = -EFAULT;
+		if (copy_to_user(buf, &event, sizeof(event)))
+			break;
+
+		if (event == APM_SYS_SUSPEND || event == APM_USER_SUSPEND)
+			as->suspend_state = SUSPEND_READ;
+
+		buf += sizeof(event);
+		i -= sizeof(event);
+	}
+
+	if (i < count)
+		ret = count - i;
+
+	return ret;
+}
+
+static unsigned int apm_poll(struct file *fp, poll_table * wait)
+{
+	struct apm_user *as = fp->private_data;
+
+	poll_wait(fp, &apm_waitqueue, wait);
+	return queue_empty(&as->queue) ? 0 : POLLIN | POLLRDNORM;
+}
+
+/*
+ * apm_ioctl - handle APM ioctl
+ *
+ * APM_IOC_SUSPEND
+ *   This IOCTL is overloaded, and performs two functions.  It is used to:
+ *     - initiate a suspend
+ *     - acknowledge a suspend read from /dev/apm_bios.
+ *   Only when everyone who has opened /dev/apm_bios with write permission
+ *   has acknowledge does the actual suspend happen.
+ */
+static int
+apm_ioctl(struct inode * inode, struct file *filp, u_int cmd, u_long arg)
+{
+	struct apm_user *as = filp->private_data;
+	unsigned long flags;
+	int err = -EINVAL;
+
+	if (!as->suser || !as->writer)
+		return -EPERM;
+
+	switch (cmd) {
+	case APM_IOC_SUSPEND:
+		as->suspend_result = -EINTR;
+
+		if (as->suspend_state == SUSPEND_READ) {
+			/*
+			 * If we read a suspend command from /dev/apm_bios,
+			 * then the corresponding APM_IOC_SUSPEND ioctl is
+			 * interpreted as an acknowledge.
+			 */
+			as->suspend_state = SUSPEND_ACKED;
+			suspends_pending--;
+		} else {
+			/*
+			 * Otherwise it is a request to suspend the system.
+			 * Queue an event for all readers, and expect an
+			 * acknowledge from all writers who haven't already
+			 * acknowledged.
+			 */
+			queue_event(APM_USER_SUSPEND, as);
+		}
+
+		/*
+		 * If there are no further acknowledges required, suspend
+		 * the system.
+		 */
+		if (suspends_pending == 0)
+			apm_suspend();
+
+		/*
+		 * Wait for the suspend/resume to complete.  If there are
+		 * pending acknowledges, we wait here for them.
+		 *
+		 * Note that we need to ensure that the PM subsystem does
+		 * not kick us out of the wait when it suspends the threads.
+		 */
+		flags = current->flags;
+		current->flags |= PF_NOFREEZE;
+
+		/*
+		 * Note: do not allow a thread which is acking the suspend
+		 * to escape until the resume is complete.
+		 */
+		if (as->suspend_state == SUSPEND_ACKED)
+			wait_event(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+		else
+			wait_event_interruptible(apm_suspend_waitqueue,
+					 as->suspend_state == SUSPEND_DONE);
+
+		current->flags = flags;
+		err = as->suspend_result;
+		as->suspend_state = SUSPEND_NONE;
+		break;
+	}
+
+	return err;
+}
+
+static int apm_release(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as = filp->private_data;
+	filp->private_data = NULL;
+
+	down_write(&user_list_lock);
+	list_del(&as->list);
+	up_write(&user_list_lock);
+
+	/*
+	 * We are now unhooked from the chain.  As far as new
+	 * events are concerned, we no longer exist.  However, we
+	 * need to balance suspends_pending, which means the
+	 * possibility of sleeping.
+	 */
+	if (as->suspend_state != SUSPEND_NONE) {
+		suspends_pending -= 1;
+		if (suspends_pending == 0)
+			apm_suspend();
+	}
+
+	kfree(as);
+	return 0;
+}
+
+static int apm_open(struct inode * inode, struct file * filp)
+{
+	struct apm_user *as;
+
+	as = kzalloc(sizeof(*as), GFP_KERNEL);
+	if (as) {
+		/*
+		 * XXX - this is a tiny bit broken, when we consider BSD
+		 * process accounting. If the device is opened by root, we
+		 * instantly flag that we used superuser privs. Who knows,
+		 * we might close the device immediately without doing a
+		 * privileged operation -- cevans
+		 */
+		as->suser = capable(CAP_SYS_ADMIN);
+		as->writer = (filp->f_mode & FMODE_WRITE) == FMODE_WRITE;
+		as->reader = (filp->f_mode & FMODE_READ) == FMODE_READ;
+
+		down_write(&user_list_lock);
+		list_add(&as->list, &apm_user_list);
+		up_write(&user_list_lock);
+
+		filp->private_data = as;
+	}
+
+	return as ? 0 : -ENOMEM;
+}
+
+static struct file_operations apm_bios_fops = {
+	.owner		= THIS_MODULE,
+	.read		= apm_read,
+	.poll		= apm_poll,
+	.ioctl		= apm_ioctl,
+	.open		= apm_open,
+	.release	= apm_release,
+};
+
+static struct miscdevice apm_device = {
+	.minor		= APM_MINOR_DEV,
+	.name		= "apm_bios",
+	.fops		= &apm_bios_fops
+};
+
+
+#ifdef CONFIG_PROC_FS
+/*
+ * Arguments, with symbols from linux/apm_bios.h.
+ *
+ *   0) Linux driver version (this will change if format changes)
+ *   1) APM BIOS Version.  Usually 1.0, 1.1 or 1.2.
+ *   2) APM flags from APM Installation Check (0x00):
+ *	bit 0: APM_16_BIT_SUPPORT
+ *	bit 1: APM_32_BIT_SUPPORT
+ *	bit 2: APM_IDLE_SLOWS_CLOCK
+ *	bit 3: APM_BIOS_DISABLED
+ *	bit 4: APM_BIOS_DISENGAGED
+ *   3) AC line status
+ *	0x00: Off-line
+ *	0x01: On-line
+ *	0x02: On backup power (BIOS >= 1.1 only)
+ *	0xff: Unknown
+ *   4) Battery status
+ *	0x00: High
+ *	0x01: Low
+ *	0x02: Critical
+ *	0x03: Charging
+ *	0x04: Selected battery not present (BIOS >= 1.2 only)
+ *	0xff: Unknown
+ *   5) Battery flag
+ *	bit 0: High
+ *	bit 1: Low
+ *	bit 2: Critical
+ *	bit 3: Charging
+ *	bit 7: No system battery
+ *	0xff: Unknown
+ *   6) Remaining battery life (percentage of charge):
+ *	0-100: valid
+ *	-1: Unknown
+ *   7) Remaining battery life (time units):
+ *	Number of remaining minutes or seconds
+ *	-1: Unknown
+ *   8) min = minutes; sec = seconds
+ */
+static int apm_read_proc(char *buf, char **start, off_t fpos, int length)
+{
+	if (likely(apm_get_info))
+		return apm_get_info(buf, start, fpos, length);
+
+	return -EINVAL;
+}
+#endif
+
+static int kapmd(void *arg)
+{
+	daemonize("kapmd");
+	current->flags |= PF_NOFREEZE;
+
+	do {
+		apm_event_t event;
+
+		wait_event_interruptible(kapmd_wait,
+				!queue_empty(&kapmd_queue) || !pm_active);
+
+		if (!pm_active)
+			break;
+
+		spin_lock_irq(&kapmd_queue_lock);
+		event = 0;
+		if (!queue_empty(&kapmd_queue))
+			event = queue_get_event(&kapmd_queue);
+		spin_unlock_irq(&kapmd_queue_lock);
+
+		switch (event) {
+		case 0:
+			break;
+
+		case APM_LOW_BATTERY:
+		case APM_POWER_STATUS_CHANGE:
+			queue_event(event, NULL);
+			break;
+
+		case APM_USER_SUSPEND:
+		case APM_SYS_SUSPEND:
+			queue_event(event, NULL);
+			if (suspends_pending == 0)
+				apm_suspend();
+			break;
+
+		case APM_CRITICAL_SUSPEND:
+			apm_suspend();
+			break;
+		}
+	} while (1);
+
+	complete_and_exit(&kapmd_exit, 0);
+}
+
+static int __init apm_init(void)
+{
+	int ret;
+
+	pm_active = 1;
+
+	ret = kernel_thread(kapmd, NULL, CLONE_KERNEL);
+	if (unlikely(ret < 0)) {
+		pm_active = 0;
+		return ret;
+	}
+
+	create_proc_info_entry("apm", 0, NULL, apm_read_proc);
+
+	ret = misc_register(&apm_device);
+	if (unlikely(ret != 0)) {
+		remove_proc_entry("apm", NULL);
+
+		pm_active = 0;
+		wake_up(&kapmd_wait);
+		wait_for_completion(&kapmd_exit);
+	}
+
+	return ret;
+}
+
+static void __exit apm_exit(void)
+{
+	misc_deregister(&apm_device);
+	remove_proc_entry("apm", NULL);
+
+	pm_active = 0;
+	wake_up(&kapmd_wait);
+	wait_for_completion(&kapmd_exit);
+}
+
+module_init(apm_init);
+module_exit(apm_exit);
+
+MODULE_AUTHOR("Stephen Rothwell, Andriy Skulysh");
+MODULE_DESCRIPTION("Advanced Power Management");
+MODULE_LICENSE("GPL");
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 306eb8d83de5..b087d34dba35 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -308,7 +308,7 @@ ENTRY(exception_error)
 	.align	2
 ret_from_exception:
 	preempt_stop()
-ret_from_irq:
+ENTRY(ret_from_irq)
 	!
 	mov	#OFF_SR, r0
 	mov.l	@(r0,r15), r0	! get status register
@@ -704,7 +704,7 @@ interrupt:
 !
 !
 	.align	2
-handle_exception:
+ENTRY(handle_exception)
 	! Using k0, k1 for scratch registers (r0_bank1, r1_bank),
 	! save all registers onto stack.
 	!
diff --git a/arch/sh/kernel/pm.c b/arch/sh/kernel/pm.c
new file mode 100644
index 000000000000..10ab62c9aede
--- /dev/null
+++ b/arch/sh/kernel/pm.c
@@ -0,0 +1,88 @@
+/*
+ * Generic Power Management Routine
+ *
+ * Copyright (c) 2006 Andriy Skulysh <askulsyh@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License.
+ */
+#include <linux/suspend.h>
+#include <linux/delay.h>
+#include <linux/gfp.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+#include <asm/watchdog.h>
+#include <asm/pm.h>
+
+#define INTR_OFFSET	0x600
+
+#define STBCR		0xffffff82
+#define STBCR2		0xffffff88
+
+#define STBCR_STBY	0x80
+#define STBCR_MSTP2	0x04
+
+#define MCR		0xffffff68
+#define RTCNT		0xffffff70
+
+#define MCR_RMODE	2
+#define MCR_RFSH	4
+
+void pm_enter(void)
+{
+	u8 stbcr, csr;
+	u16 frqcr, mcr;
+	u32 vbr_new, vbr_old;
+
+	set_bl_bit();
+
+	/* set wdt */
+	csr = sh_wdt_read_csr();
+	csr &= ~WTCSR_TME;
+	csr |= WTCSR_CKS_4096;
+	sh_wdt_write_csr(csr);
+	csr = sh_wdt_read_csr();
+	sh_wdt_write_cnt(0);
+
+	/* disable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr &= ~(FRQCR_PLLEN | FRQCR_PSTBY);
+	ctrl_outw(frqcr, FRQCR);
+
+	/* enable standby */
+	stbcr = ctrl_inb(STBCR);
+	ctrl_outb(stbcr | STBCR_STBY | STBCR_MSTP2, STBCR);
+
+	/* set self-refresh */
+	mcr = ctrl_inw(MCR);
+	ctrl_outw(mcr & ~MCR_RFSH, MCR);
+
+	/* set interrupt handler */
+	asm volatile("stc vbr, %0" : "=r" (vbr_old));
+	vbr_new = get_zeroed_page(GFP_ATOMIC);
+	udelay(50);
+	memcpy((void*)(vbr_new + INTR_OFFSET),
+	       &wakeup_start, &wakeup_end - &wakeup_start);
+	asm volatile("ldc %0, vbr" : : "r" (vbr_new));
+
+	ctrl_outw(0, RTCNT);
+	ctrl_outw(mcr | MCR_RFSH | MCR_RMODE, MCR);
+
+	cpu_sleep();
+
+	asm volatile("ldc %0, vbr" : : "r" (vbr_old));
+
+	free_page(vbr_new);
+
+	/* enable PLL1 */
+	frqcr = ctrl_inw(FRQCR);
+	frqcr |= FRQCR_PSTBY;
+	ctrl_outw(frqcr, FRQCR);
+	udelay(50);
+	frqcr |= FRQCR_PLLEN;
+	ctrl_outw(frqcr, FRQCR);
+
+	ctrl_outb(stbcr, STBCR);
+
+	clear_bl_bit();
+}
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index b73feafcd06e..fd73ab0326e9 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -116,6 +116,10 @@ EXPORT_SYMBOL(__down_trylock);
 EXPORT_SYMBOL(synchronize_irq);
 #endif
 
+#ifdef CONFIG_PM
+EXPORT_SYMBOL(pm_suspend);
+#endif
+
 EXPORT_SYMBOL(csum_partial);
 #ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index a1589f85499d..c8db6ca4f9d1 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -143,8 +143,33 @@ void handle_timer_tick(struct pt_regs *regs)
 	}
 }
 
+#ifdef CONFIG_PM
+int timer_suspend(struct sys_device *dev, pm_message_t state)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->stop();
+
+	return 0;
+}
+
+int timer_resume(struct sys_device *dev)
+{
+	struct sys_timer *sys_timer = container_of(dev, struct sys_timer, dev);
+
+	sys_timer->ops->start();
+
+	return 0;
+}
+#else
+#define timer_suspend NULL
+#define timer_resume NULL
+#endif
+
 static struct sysdev_class timer_sysclass = {
 	set_kset_name("timer"),
+	.suspend = timer_suspend,
+	.resume	 = timer_resume,
 };
 
 static int __init timer_init_sysfs(void)
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index d4212add53b2..ea4bdf8cdf0d 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -188,6 +188,18 @@ static struct clk tmu0_clk = {
 	.ops		= &tmu_clk_ops,
 };
 
+static int tmu_timer_start(void)
+{
+	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	return 0;
+}
+
+static int tmu_timer_stop(void)
+{
+	ctrl_outb(0, TMU_TSTR);
+	return 0;
+}
+
 static int tmu_timer_init(void)
 {
 	unsigned long interval;
@@ -197,7 +209,7 @@ static int tmu_timer_init(void)
 	tmu0_clk.parent = clk_get("module_clk");
 
 	/* Start TMU0 */
-	ctrl_outb(0, TMU_TSTR);
+	tmu_timer_stop();
 #if !defined(CONFIG_CPU_SUBTYPE_SH7300) && !defined(CONFIG_CPU_SUBTYPE_SH7760)
 	ctrl_outb(TMU_TOCR_INIT, TMU_TOCR);
 #endif
@@ -211,13 +223,15 @@ static int tmu_timer_init(void)
 	ctrl_outl(interval, TMU0_TCOR);
 	ctrl_outl(interval, TMU0_TCNT);
 
-	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
+	tmu_timer_start();
 
 	return 0;
 }
 
 struct sys_timer_ops tmu_timer_ops = {
 	.init		= tmu_timer_init,
+	.start		= tmu_timer_start,
+	.stop		= tmu_timer_stop,
 	.get_frequency	= tmu_timer_get_frequency,
 	.get_offset	= tmu_timer_get_offset,
 };
diff --git a/drivers/input/touchscreen/hp680_ts_input.c b/drivers/input/touchscreen/hp680_ts_input.c
index fa97e0f79e7e..ee6c2f40cdf6 100644
--- a/drivers/input/touchscreen/hp680_ts_input.c
+++ b/drivers/input/touchscreen/hp680_ts_input.c
@@ -15,7 +15,6 @@
 #define HP680_TS_ABS_Y_MIN	80
 #define HP680_TS_ABS_Y_MAX	910
 
-#define	SCPCR	0xa4000116
 #define	PHDR	0xa400012e
 #define SCPDR	0xa4000136
 
@@ -77,19 +76,6 @@ static irqreturn_t hp680_ts_interrupt(int irq, void *dev, struct pt_regs *regs)
 
 static int __init hp680_ts_init(void)
 {
-	u8 scpdr;
-	u16 scpcr;
-
-	scpdr = ctrl_inb(SCPDR);
-	scpdr |= SCPDR_TS_SCAN_X | SCPDR_TS_SCAN_Y;
-	scpdr &= ~SCPDR_TS_SCAN_ENABLE;
-	ctrl_outb(scpdr, SCPDR);
-
-	scpcr = ctrl_inw(SCPCR);
-	scpcr &= ~SCPCR_TS_MASK;
-	scpcr |= SCPCR_TS_ENABLE;
-	ctrl_outw(scpcr, SCPCR);
-
 	hp680_ts_dev = input_allocate_device();
 	if (!hp680_ts_dev)
 		return -ENOMEM;
diff --git a/include/asm-sh/apm.h b/include/asm-sh/apm.h
new file mode 100644
index 000000000000..8b091e93651f
--- /dev/null
+++ b/include/asm-sh/apm.h
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2006 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ */
+
+#ifndef __ASM_SH_APM_H
+#define __ASM_SH_APM_H
+
+#define APM_AC_OFFLINE			0
+#define APM_AC_ONLINE			1
+#define APM_AC_BACKUP			2
+#define APM_AC_UNKNOWN			0xff
+
+#define APM_BATTERY_STATUS_HIGH		0
+#define APM_BATTERY_STATUS_LOW		1
+#define APM_BATTERY_STATUS_CRITICAL	2
+#define APM_BATTERY_STATUS_CHARGING	3
+#define APM_BATTERY_STATUS_NOT_PRESENT	4
+#define APM_BATTERY_STATUS_UNKNOWN	0xff
+
+#define APM_BATTERY_LIFE_UNKNOWN	0xFFFF
+#define APM_BATTERY_LIFE_MINUTES	0x8000
+#define APM_BATTERY_LIFE_VALUE_MASK	0x7FFF
+
+#define APM_BATTERY_FLAG_HIGH		(1 << 0)
+#define APM_BATTERY_FLAG_LOW		(1 << 1)
+#define APM_BATTERY_FLAG_CRITICAL	(1 << 2)
+#define APM_BATTERY_FLAG_CHARGING	(1 << 3)
+#define APM_BATTERY_FLAG_NOT_PRESENT	(1 << 7)
+#define APM_BATTERY_FLAG_UNKNOWN	0xff
+
+#define APM_UNITS_MINS			0
+#define APM_UNITS_SECS			1
+#define APM_UNITS_UNKNOWN		-1
+
+
+extern int (*apm_get_info)(char *buf, char **start, off_t fpos, int length);
+extern int apm_suspended;
+
+void apm_queue_event(apm_event_t event);
+
+#endif
diff --git a/include/asm-sh/cpu-sh3/freq.h b/include/asm-sh/cpu-sh3/freq.h
index b61b6e331df0..273f3229785c 100644
--- a/include/asm-sh/cpu-sh3/freq.h
+++ b/include/asm-sh/cpu-sh3/freq.h
@@ -18,5 +18,9 @@
 #define MIN_DIVISOR_NR		0
 #define MAX_DIVISOR_NR		4
 
+#define FRQCR_CKOEN	0x0100
+#define FRQCR_PLLEN	0x0080
+#define FRQCR_PSTBY	0x0040
+
 #endif /* __ASM_CPU_SH3_FREQ_H */
 
diff --git a/include/asm-sh/hd64461.h b/include/asm-sh/hd64461.h
index 0f2e2132cc35..27e5c34e2659 100644
--- a/include/asm-sh/hd64461.h
+++ b/include/asm-sh/hd64461.h
@@ -40,7 +40,12 @@
 #define HD64461_LCDCBAR		0x11000
 #define HD64461_LCDCLOR		0x11002
 #define HD64461_LCDCCR		0x11004
-#define HD64461_LCDCCR_MOFF	0x80
+#define HD64461_LCDCCR_STBACK	0x0400
+#define HD64461_LCDCCR_STREQ	0x0100
+#define HD64461_LCDCCR_MOFF	0x0080
+#define HD64461_LCDCCR_REFSEL	0x0040
+#define HD64461_LCDCCR_EPON	0x0020
+#define HD64461_LCDCCR_SPON	0x0010
 
 #define	HD64461_LDR1		0x11010
 #define	HD64461_LDR1_DON	0x01
diff --git a/include/asm-sh/hp6xx/hp6xx.h b/include/asm-sh/hp6xx/hp6xx.h
index a26247fd3d87..f35134c159dd 100644
--- a/include/asm-sh/hp6xx/hp6xx.h
+++ b/include/asm-sh/hp6xx/hp6xx.h
@@ -2,16 +2,33 @@
 #define __ASM_SH_HP6XX_H
 
 /*
- * Copyright (C) 2003  Andriy Skulysh
+ * Copyright (C) 2003, 2004, 2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
  */
 
-#define HP680_TS_IRQ IRQ3_IRQ
+#define HP680_BTN_IRQ		IRQ0_IRQ
+#define HP680_TS_IRQ		IRQ3_IRQ
+#define HP680_HD64461_IRQ	IRQ4_IRQ
 
 #define DAC_LCD_BRIGHTNESS	0
 #define DAC_SPEAKER_VOLUME	1
 
+#define PGDR_OPENED		0x01
+#define PGDR_MAIN_BATTERY_OUT	0x04
+#define PGDR_PLAY_BUTTON	0x08
+#define PGDR_REWIND_BUTTON	0x10
+#define PGDR_RECORD_BUTTON	0x20
+
 #define PHDR_TS_PEN_DOWN	0x08
 
+#define PJDR_LED_BLINK		0x02
+
+#define PKDR_LED_GREEN		0x10
+
 #define SCPDR_TS_SCAN_ENABLE	0x20
 #define SCPDR_TS_SCAN_Y		0x02
 #define SCPDR_TS_SCAN_X		0x01
@@ -21,11 +38,43 @@
 
 #define ADC_CHANNEL_TS_Y	1
 #define ADC_CHANNEL_TS_X	2
+#define ADC_CHANNEL_BATTERY	3
+#define ADC_CHANNEL_BACKUP	4
+#define ADC_CHANNEL_CHARGE	5
 
 #define HD64461_GPADR_SPEAKER	0x01
 #define HD64461_GPADR_PCMCIA0	(0x02|0x08)
+
 #define HD64461_GPBDR_LCDOFF	0x01
+#define HD64461_GPBDR_LCD_CONTRAST_MASK	0x78
 #define HD64461_GPBDR_LED_RED	0x80
 
+#include <asm/hd64461.h>
+#include <asm/io.h>
+
+#define PJDR	0xa4000130
+#define PKDR	0xa4000132
+
+static inline void hp6xx_led_red(int on)
+{
+	u16 v16;
+	v16 = ctrl_inw(CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+	if (on)
+	    ctrl_outw(v16 & (~HD64461_GPBDR_LED_RED), CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+	else
+	    ctrl_outw(v16 | HD64461_GPBDR_LED_RED, CONFIG_HD64461_IOBASE + HD64461_GPBDR - 0x10000);
+}
+
+static inline void hp6xx_led_green(int on)
+{
+	u8 v8;
+
+	v8 = ctrl_inb(PKDR);
+	if (on)
+	    ctrl_outb(v8 & (~PKDR_LED_GREEN), PKDR);
+	else
+	    ctrl_outb(v8 | PKDR_LED_GREEN, PKDR);
+}
+
 
 #endif /* __ASM_SH_HP6XX_H */
diff --git a/include/asm-sh/pm.h b/include/asm-sh/pm.h
new file mode 100644
index 000000000000..56fdbd6b1c94
--- /dev/null
+++ b/include/asm-sh/pm.h
@@ -0,0 +1,17 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright 2006 (c) Andriy Skulysh <askulysh@gmail.com>
+ *
+ */
+#ifndef __ASM_SH_PM_H
+#define __ASM_SH_PM_H
+
+extern u8 wakeup_start;
+extern u8 wakeup_end;
+
+void pm_enter(void);
+
+#endif
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index bd7dc0554b11..477422afeb0d 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -172,6 +172,31 @@ static __inline__ void local_irq_disable(void)
 			     : "memory");
 }
 
+static __inline__ void set_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ ("stc	sr, %0\n\t"
+			     "or	%2, %0\n\t"
+			     "and	%3, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy0), "=r" (__dummy1)
+			     : "r" (0x10000000), "r" (0xffffff0f)
+			     : "memory");
+}
+
+static __inline__ void clear_bl_bit(void)
+{
+	unsigned long __dummy0, __dummy1;
+
+	__asm__ __volatile__ ("stc	sr, %0\n\t"
+			     "and	%2, %0\n\t"
+			     "ldc	%0, sr"
+			     : "=&r" (__dummy0), "=r" (__dummy1)
+			     : "1" (~0x10000000)
+			     : "memory");
+}
+
 #define local_save_flags(x) \
 	__asm__("stc sr, %0; and #0xf0, %0" : "=&z" (x) :/**/: "memory" )
 
diff --git a/include/asm-sh/timer.h b/include/asm-sh/timer.h
index dd6579c0b04c..c7ab28095ba0 100644
--- a/include/asm-sh/timer.h
+++ b/include/asm-sh/timer.h
@@ -6,6 +6,8 @@
 
 struct sys_timer_ops {
 	int (*init)(void);
+	int (*start)(void);
+	int (*stop)(void);
 	unsigned long (*get_offset)(void);
 	unsigned long (*get_frequency)(void);
 };
-- 
cgit v1.2.1


From 2549b3222f588c188674aed0b9a0ef78bbba5c6c Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:22:33 +0900
Subject: sh: Use generic CONFIG_FRAME_POINTER.

We had our own version, which serves no purpose. Simply
hook SH in to the generic one.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug | 10 +---------
 lib/Kconfig.debug     |  2 +-
 2 files changed, 2 insertions(+), 10 deletions(-)

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 8fb31ab2c02c..75ed1be8a49e 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -32,6 +32,7 @@ config EARLY_PRINTK
 
 config KGDB
 	bool "Include KGDB kernel debugger"
+	select FRAME_POINTER
 	help
 	  Include in-kernel hooks for kgdb, the Linux kernel source level
 	  debugger.  See <http://kgdb.sourceforge.net/> for more information.
@@ -112,13 +113,4 @@ endchoice
 
 endmenu
 
-config FRAME_POINTER
-	bool "Compile the kernel with frame pointers"
-	default y if KGDB
-	help
-	  If you say Y here the resulting kernel image will be slightly larger
-	  and slower, but it will give very useful debugging information.
-	  If you don't debug the kernel, you can say N, but we may not be able
-	  to solve problems without frame pointers.
-
 endmenu
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index f1ac3184dc08..e7303fe54b01 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -322,7 +322,7 @@ config DEBUG_VM
 
 config FRAME_POINTER
 	bool "Compile the kernel with frame pointers"
-	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || AVR32)
+	depends on DEBUG_KERNEL && (X86 || CRIS || M68K || M68KNOMMU || FRV || UML || S390 || AVR32 || SUPERH)
 	default y if DEBUG_INFO && UML
 	help
 	  If you say Y here the resulting kernel image will be slightly larger
-- 
cgit v1.2.1


From 56e8d7b5786dc2f7d1f701500f8914fd2c52b111 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:24:55 +0900
Subject: sh: kgdb stub cleanups.

Some kgdb cleanup. Move hexchars/highhex/lowhex to the header, so it can
be reused by sh-sci. Also drop silly ctrl_inl/outl() overloading being
done by the kgdb stub.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/kgdb_stub.c | 33 +++++----------------------------
 include/asm-sh/kgdb.h      | 15 +++++++++++++++
 2 files changed, 20 insertions(+), 28 deletions(-)

diff --git a/arch/sh/kernel/kgdb_stub.c b/arch/sh/kernel/kgdb_stub.c
index 42638b92b51c..9c6315f0335d 100644
--- a/arch/sh/kernel/kgdb_stub.c
+++ b/arch/sh/kernel/kgdb_stub.c
@@ -101,16 +101,17 @@
 #include <linux/linkage.h>
 #include <linux/init.h>
 
+#ifdef CONFIG_SH_KGDB_CONSOLE
+#include <linux/console.h>
+#endif
+
 #include <asm/system.h>
 #include <asm/current.h>
 #include <asm/signal.h>
 #include <asm/pgtable.h>
 #include <asm/ptrace.h>
 #include <asm/kgdb.h>
-
-#ifdef CONFIG_SH_KGDB_CONSOLE
-#include <linux/console.h>
-#endif
+#include <asm/io.h>
 
 /* Function pointers for linkage */
 kgdb_debug_hook_t *kgdb_debug_hook;
@@ -240,7 +241,6 @@ static jmp_buf rem_com_env;
 /* Misc static */
 static int stepped_address;
 static short stepped_opcode;
-static const char hexchars[] = "0123456789abcdef";
 static char in_buffer[BUFMAX];
 static char out_buffer[OUTBUFMAX];
 
@@ -253,29 +253,6 @@ typedef unsigned char threadref[8];
 #define BUF_THREAD_ID_SIZE 16
 #endif
 
-/* Return addr as a real volatile address */
-static inline unsigned int ctrl_inl(const unsigned long addr)
-{
-	return *(volatile unsigned long *) addr;
-}
-
-/* Correctly set *addr using volatile */
-static inline void ctrl_outl(const unsigned int b, unsigned long addr)
-{
-	*(volatile unsigned long *) addr = b;
-}
-
-/* Get high hex bits */
-static char highhex(const int x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-/* Get low hex bits */
-static char lowhex(const int x)
-{
-	return hexchars[x & 0xf];
-}
 
 /* Convert ch to hex */
 static int hex(const char ch)
diff --git a/include/asm-sh/kgdb.h b/include/asm-sh/kgdb.h
index 1653ffb75fbe..7b26f53fe343 100644
--- a/include/asm-sh/kgdb.h
+++ b/include/asm-sh/kgdb.h
@@ -128,4 +128,19 @@ extern int     setjmp(jmp_buf __jmpb);
 #define KGDB_ASSERT(condition, message)
 #endif
 
+/* Taken from sh-stub.c of GDB 4.18 */
+static const char hexchars[] = "0123456789abcdef";
+
+/* Get high hex bits */
+static inline char highhex(const int x)
+{
+	return hexchars[(x >> 4) & 0xf];
+}
+
+/* Get low hex bits */
+static inline char lowhex(const int x)
+{
+	return hexchars[x & 0xf];
+}
+
 #endif
-- 
cgit v1.2.1


From e108b2ca2349f510ce7d7f910eda89f71d710d84 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:32:13 +0900
Subject: serial: Rework sh-sci for driver model.

sh-sci was turning in to an unmaintainable mess, especially with
regards to the port list. This cleans it up quite a bit, and
switches over to a platform device model where subtypes will
register their port list individually in their setup code.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/serial/sh-sci.c | 1146 +++++++++++++++++------------------------------
 drivers/serial/sh-sci.h |   90 ++--
 2 files changed, 467 insertions(+), 769 deletions(-)

diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c
index cbede06cac27..f336ba6778dd 100644
--- a/drivers/serial/sh-sci.c
+++ b/drivers/serial/sh-sci.c
@@ -3,7 +3,7 @@
  *
  * SuperH on-chip serial module support.  (SCI with no FIFO / with FIFO)
  *
- *  Copyright (C) 2002, 2003, 2004  Paul Mundt
+ *  Copyright (C) 2002 - 2006  Paul Mundt
  *
  * based off of the old drivers/char/sh-sci.c by:
  *
@@ -20,10 +20,9 @@
 
 #undef DEBUG
 
+#include <linux/config.h>
 #include <linux/module.h>
 #include <linux/errno.h>
-#include <linux/signal.h>
-#include <linux/sched.h>
 #include <linux/timer.h>
 #include <linux/interrupt.h>
 #include <linux/tty.h>
@@ -32,71 +31,77 @@
 #include <linux/major.h>
 #include <linux/string.h>
 #include <linux/sysrq.h>
-#include <linux/fcntl.h>
-#include <linux/ptrace.h>
 #include <linux/ioport.h>
 #include <linux/mm.h>
-#include <linux/slab.h>
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/console.h>
-#include <linux/bitops.h>
-#include <linux/generic_serial.h>
+#include <linux/platform_device.h>
 
 #ifdef CONFIG_CPU_FREQ
 #include <linux/notifier.h>
 #include <linux/cpufreq.h>
 #endif
 
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-
 #if defined(CONFIG_SUPERH) && !defined(CONFIG_SUPERH64)
 #include <asm/clock.h>
-#endif
-
-#ifdef CONFIG_SH_STANDARD_BIOS
 #include <asm/sh_bios.h>
+#include <asm/kgdb.h>
 #endif
 
+#include <asm/sci.h>
+
 #if defined(CONFIG_SERIAL_SH_SCI_CONSOLE) && defined(CONFIG_MAGIC_SYSRQ)
 #define SUPPORT_SYSRQ
 #endif
 
 #include "sh-sci.h"
 
-#ifdef CONFIG_SH_KGDB
-#include <asm/kgdb.h>
+struct sci_port {
+	struct uart_port	port;
+
+	/* Port type */
+	unsigned int		type;
+
+	/* Port IRQs: ERI, RXI, TXI, BRI (optional) */
+	unsigned int		irqs[SCIx_NR_IRQS]; 
+
+	/* Port pin configuration */
+	void			(*init_pins)(struct uart_port *port,
+					     unsigned int cflag);
 
-static int kgdb_get_char(struct sci_port *port);
-static void kgdb_put_char(struct sci_port *port, char c);
-static void kgdb_handle_error(struct sci_port *port);
+	/* Port enable callback */
+	void			(*enable)(struct uart_port *port);
+
+	/* Port disable callback */
+	void			(*disable)(struct uart_port *port);
+
+	/* Break timer */
+	struct timer_list	break_timer;
+	int			break_flag;
+};
+
+#ifdef CONFIG_SH_KGDB
 static struct sci_port *kgdb_sci_port;
-#endif /* CONFIG_SH_KGDB */
+#endif
 
 #ifdef CONFIG_SERIAL_SH_SCI_CONSOLE
-static struct sci_port *serial_console_port = 0;
-#endif /* CONFIG_SERIAL_SH_SCI_CONSOLE */
+static struct sci_port *serial_console_port;
+#endif
 
 /* Function prototypes */
 static void sci_stop_tx(struct uart_port *port);
-static void sci_start_tx(struct uart_port *port);
-static void sci_start_rx(struct uart_port *port, unsigned int tty_start);
-static void sci_stop_rx(struct uart_port *port);
-static int sci_request_irq(struct sci_port *port);
-static void sci_free_irq(struct sci_port *port);
-
-static struct sci_port sci_ports[];
-static struct uart_driver sci_uart_driver;
 
-#define SCI_NPORTS sci_uart_driver.nr
+#define SCI_NPORTS CONFIG_SERIAL_SH_SCI_NR_UARTS
 
-#if defined(CONFIG_SH_STANDARD_BIOS) || defined(CONFIG_SH_KGDB)
+static struct sci_port sci_ports[SCI_NPORTS];
+static struct uart_driver sci_uart_driver;
 
-static void handle_error(struct uart_port *port)
-{				/* Clear error flags */
+#if defined(CONFIG_SERIAL_SH_SCI_CONSOLE) && \
+    defined(CONFIG_SH_STANDARD_BIOS) || defined(CONFIG_SH_KGDB)
+static inline void handle_error(struct uart_port *port)
+{
+	/* Clear error flags */
 	sci_out(port, SCxSR, SCxSR_ERROR_CLEAR(port));
 }
 
@@ -106,8 +111,8 @@ static int get_char(struct uart_port *port)
 	unsigned short status;
 	int c;
 
-	local_irq_save(flags);
-        do {
+	spin_lock_irqsave(&port->lock, flags);
+	do {
 		status = sci_in(port, SCxSR);
 		if (status & SCxSR_ERRORS(port)) {
 			handle_error(port);
@@ -117,38 +122,19 @@ static int get_char(struct uart_port *port)
 	c = sci_in(port, SCxRDR);
 	sci_in(port, SCxSR);            /* Dummy read */
 	sci_out(port, SCxSR, SCxSR_RDxF_CLEAR(port));
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&port->lock, flags);
 
 	return c;
 }
-
-/* Taken from sh-stub.c of GDB 4.18 */
-static const char hexchars[] = "0123456789abcdef";
-
-static __inline__ char highhex(int  x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-static __inline__ char lowhex(int  x)
-{
-	return hexchars[x & 0xf];
-}
-
 #endif /* CONFIG_SH_STANDARD_BIOS || CONFIG_SH_KGDB */
 
-/*
- * Send the packet in buffer.  The host gets one chance to read it.
- * This routine does not wait for a positive acknowledge.
- */
-
-#ifdef CONFIG_SERIAL_SH_SCI_CONSOLE
+#if defined(CONFIG_SERIAL_SH_SCI_CONSOLE) || defined(CONFIG_SH_KGDB)
 static void put_char(struct uart_port *port, char c)
 {
 	unsigned long flags;
 	unsigned short status;
 
-	local_irq_save(flags);
+	spin_lock_irqsave(&port->lock, flags);
 
 	do {
 		status = sci_in(port, SCxSR);
@@ -158,9 +144,11 @@ static void put_char(struct uart_port *port, char c)
 	sci_in(port, SCxSR);            /* Dummy read */
 	sci_out(port, SCxSR, SCxSR_TDxE_CLEAR(port));
 
-	local_irq_restore(flags);
+	spin_unlock_irqrestore(&port->lock, flags);
 }
+#endif
 
+#ifdef CONFIG_SERIAL_SH_SCI_CONSOLE
 static void put_string(struct sci_port *sci_port, const char *buffer, int count)
 {
 	struct uart_port *port = &sci_port->port;
@@ -213,96 +201,28 @@ static void put_string(struct sci_port *sci_port, const char *buffer, int count)
 }
 #endif /* CONFIG_SERIAL_SH_SCI_CONSOLE */
 
-
 #ifdef CONFIG_SH_KGDB
-
-/* Is the SCI ready, ie is there a char waiting? */
-static int kgdb_is_char_ready(struct sci_port *port)
-{
-        unsigned short status = sci_in(port, SCxSR);
-
-        if (status & (SCxSR_ERRORS(port) | SCxSR_BRK(port)))
-                kgdb_handle_error(port);
-
-        return (status & SCxSR_RDxF(port));
-}
-
-/* Write a char */
-static void kgdb_put_char(struct sci_port *port, char c)
-{
-        unsigned short status;
-
-        do
-                status = sci_in(port, SCxSR);
-        while (!(status & SCxSR_TDxE(port)));
-
-        sci_out(port, SCxTDR, c);
-        sci_in(port, SCxSR);    /* Dummy read */
-        sci_out(port, SCxSR, SCxSR_TDxE_CLEAR(port));
-}
-
-/* Get a char if there is one, else ret -1 */
-static int kgdb_get_char(struct sci_port *port)
-{
-        int c;
-
-        if (kgdb_is_char_ready(port) == 0)
-                c = -1;
-        else {
-                c = sci_in(port, SCxRDR);
-                sci_in(port, SCxSR);    /* Dummy read */
-                sci_out(port, SCxSR, SCxSR_RDxF_CLEAR(port));
-        }
-
-        return c;
-}
-
-/* Called from kgdbstub.c to get a character, i.e. is blocking */
 static int kgdb_sci_getchar(void)
 {
-        volatile int c;
+        int c;
 
         /* Keep trying to read a character, this could be neater */
-        while ((c = kgdb_get_char(kgdb_sci_port)) < 0);
+        while ((c = get_char(kgdb_sci_port)) < 0)
+		cpu_relax();
 
         return c;
 }
 
-/* Called from kgdbstub.c to put a character, just a wrapper */
-static void kgdb_sci_putchar(int c)
-{
-
-        kgdb_put_char(kgdb_sci_port, c);
-}
-
-/* Clear any errors on the SCI */
-static void kgdb_handle_error(struct sci_port *port)
+static inline void kgdb_sci_putchar(int c)
 {
-        sci_out(port, SCxSR, SCxSR_ERROR_CLEAR(port));  /* Clear error flags */
+        put_char(kgdb_sci_port, c);
 }
-
-/* Breakpoint if there's a break sent on the serial port */
-static void kgdb_break_interrupt(int irq, void *ptr, struct pt_regs *regs)
-{
-        struct sci_port *port = ptr;
-        unsigned short status = sci_in(port, SCxSR);
-
-        if (status & SCxSR_BRK(port)) {
-
-                /* Break into the debugger if a break is detected */
-                BREAKPOINT();
-
-                /* Clear */
-                sci_out(port, SCxSR, SCxSR_BREAK_CLEAR(port));
-        }
-}
-
 #endif /* CONFIG_SH_KGDB */
 
 #if defined(__H8300S__)
 enum { sci_disable, sci_enable };
 
-static void h8300_sci_enable(struct uart_port* port, unsigned int ctrl)
+static void h8300_sci_config(struct uart_port* port, unsigned int ctrl)
 {
 	volatile unsigned char *mstpcrl=(volatile unsigned char *)MSTPCRL;
 	int ch = (port->mapbase  - SMR0) >> 3;
@@ -314,32 +234,66 @@ static void h8300_sci_enable(struct uart_port* port, unsigned int ctrl)
 		*mstpcrl &= ~mask;
 	}
 }
+
+static inline void h8300_sci_enable(struct uart_port *port)
+{
+	h8300_sci_config(port, sci_enable);
+}
+
+static inline void h8300_sci_disable(struct uart_port *port)
+{
+	h8300_sci_config(port, sci_disable);
+}
 #endif
 
-#if defined(SCI_ONLY) || defined(SCI_AND_SCIF)
-#if defined(__H8300H__) || defined(__H8300S__)
+#if defined(SCI_ONLY) || defined(SCI_AND_SCIF) && \
+    defined(__H8300H__) || defined(__H8300S__)
 static void sci_init_pins_sci(struct uart_port* port, unsigned int cflag)
 {
 	int ch = (port->mapbase - SMR0) >> 3;
 
 	/* set DDR regs */
-	H8300_GPIO_DDR(h8300_sci_pins[ch].port,h8300_sci_pins[ch].rx,H8300_GPIO_INPUT);
-	H8300_GPIO_DDR(h8300_sci_pins[ch].port,h8300_sci_pins[ch].tx,H8300_GPIO_OUTPUT);
+	H8300_GPIO_DDR(h8300_sci_pins[ch].port,
+		       h8300_sci_pins[ch].rx,
+		       H8300_GPIO_INPUT);
+	H8300_GPIO_DDR(h8300_sci_pins[ch].port,
+		       h8300_sci_pins[ch].tx,
+		       H8300_GPIO_OUTPUT);
+
 	/* tx mark output*/
 	H8300_SCI_DR(ch) |= h8300_sci_pins[ch].tx;
 }
+#else
+#define sci_init_pins_sci NULL
+#endif
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709)
+static void sci_init_pins_irda(struct uart_port *port, unsigned int cflag)
+{
+	unsigned int fcr_val = 0;
+
+	if (cflag & CRTSCTS)
+		fcr_val |= SCFCR_MCE;
+
+	sci_out(port, SCFCR, fcr_val);
+}
+#else
+#define sci_init_pins_irda NULL
 #endif
+
+#ifdef SCI_ONLY
+#define sci_init_pins_scif NULL
 #endif
 
 #if defined(SCIF_ONLY) || defined(SCI_AND_SCIF)
-#if defined(CONFIG_CPU_SUBTYPE_SH7300)
+#if defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7710)
 /* SH7300 doesn't use RTS/CTS */
 static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 {
 	sci_out(port, SCFCR, 0);
 }
 #elif defined(CONFIG_CPU_SH3)
-/* For SH7705, SH7707, SH7709, SH7709A, SH7729 */
+/* For SH7705, SH7706, SH7707, SH7709, SH7709A, SH7729 */
 static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 {
 	unsigned int fcr_val = 0;
@@ -366,20 +320,7 @@ static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 
 	sci_out(port, SCFCR, fcr_val);
 }
-
-#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709)
-static void sci_init_pins_irda(struct uart_port *port, unsigned int cflag)
-{
-	unsigned int fcr_val = 0;
-
-	if (cflag & CRTSCTS)
-		fcr_val |= SCFCR_MCE;
-
-	sci_out(port, SCFCR, fcr_val);
-}
-#endif
 #else
-
 /* For SH7750 */
 static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 {
@@ -388,7 +329,9 @@ static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 	if (cflag & CRTSCTS) {
 		fcr_val |= SCFCR_MCE;
 	} else {
-#ifdef CONFIG_CPU_SUBTYPE_SH7780
+#ifdef CONFIG_CPU_SUBTYPE_SH7343
+		/* Nothing */
+#elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 		ctrl_outw(0x0080, SCSPTR0); /* Set RTS = 1 */
 #else
 		ctrl_outw(0x0080, SCSPTR2); /* Set RTS = 1 */
@@ -396,10 +339,41 @@ static void sci_init_pins_scif(struct uart_port *port, unsigned int cflag)
 	}
 	sci_out(port, SCFCR, fcr_val);
 }
+#endif
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7760) || defined(CONFIG_CPU_SUBTYPE_SH7780)
+static inline int scif_txroom(struct uart_port *port)
+{
+	return SCIF_TXROOM_MAX - (sci_in(port, SCTFDR) & 0x7f);
+}
+
+static inline int scif_rxroom(struct uart_port *port)
+{
+	return sci_in(port, SCRFDR) & 0x7f;
+}
+#else
+static inline int scif_txroom(struct uart_port *port)
+{
+	return SCIF_TXROOM_MAX - (sci_in(port, SCFDR) >> 8);
+}
 
+static inline int scif_rxroom(struct uart_port *port)
+{
+	return sci_in(port, SCFDR) & SCIF_RFDC_MASK;
+}
 #endif
 #endif /* SCIF_ONLY || SCI_AND_SCIF */
 
+static inline int sci_txroom(struct uart_port *port)
+{
+	return ((sci_in(port, SCxSR) & SCI_TDRE) != 0);
+}
+
+static inline int sci_rxroom(struct uart_port *port)
+{
+	return ((sci_in(port, SCxSR) & SCxSR_RDxF(port)) != 0);
+}
+
 /* ********************************************************************** *
  *                   the interrupt related routines                       *
  * ********************************************************************** */
@@ -408,14 +382,12 @@ static void sci_transmit_chars(struct uart_port *port)
 {
 	struct circ_buf *xmit = &port->info->xmit;
 	unsigned int stopped = uart_tx_stopped(port);
-	unsigned long flags;
 	unsigned short status;
 	unsigned short ctrl;
-	int count, txroom;
+	int count;
 
 	status = sci_in(port, SCxSR);
 	if (!(status & SCxSR_TDxE(port))) {
-		local_irq_save(flags);
 		ctrl = sci_in(port, SCSCR);
 		if (uart_circ_empty(xmit)) {
 			ctrl &= ~SCI_CTRL_FLAGS_TIE;
@@ -423,25 +395,15 @@ static void sci_transmit_chars(struct uart_port *port)
 			ctrl |= SCI_CTRL_FLAGS_TIE;
 		}
 		sci_out(port, SCSCR, ctrl);
-		local_irq_restore(flags);
 		return;
 	}
 
-#if !defined(SCI_ONLY)
-	if (port->type == PORT_SCIF) {
-#if defined(CONFIG_CPU_SUBTYPE_SH7760) || defined(CONFIG_CPU_SUBTYPE_SH7780)
-		txroom = SCIF_TXROOM_MAX - (sci_in(port, SCTFDR) & 0x7f);
-#else
-		txroom = SCIF_TXROOM_MAX - (sci_in(port, SCFDR)>>8);
-#endif
-	} else {
-		txroom = (sci_in(port, SCxSR) & SCI_TDRE)?1:0;
-	}
-#else
-	txroom = (sci_in(port, SCxSR) & SCI_TDRE)?1:0;
+#ifndef SCI_ONLY
+	if (port->type == PORT_SCIF)
+		count = scif_txroom(port);
+	else
 #endif
-
-	count = txroom;
+		count = sci_txroom(port);
 
 	do {
 		unsigned char c;
@@ -468,7 +430,6 @@ static void sci_transmit_chars(struct uart_port *port)
 	if (uart_circ_empty(xmit)) {
 		sci_stop_tx(port);
 	} else {
-		local_irq_save(flags);
 		ctrl = sci_in(port, SCSCR);
 
 #if !defined(SCI_ONLY)
@@ -480,7 +441,6 @@ static void sci_transmit_chars(struct uart_port *port)
 
 		ctrl |= SCI_CTRL_FLAGS_TIE;
 		sci_out(port, SCSCR, ctrl);
-		local_irq_restore(flags);
 	}
 }
 
@@ -490,6 +450,7 @@ static void sci_transmit_chars(struct uart_port *port)
 static inline void sci_receive_chars(struct uart_port *port,
 				     struct pt_regs *regs)
 {
+	struct sci_port *sci_port = (struct sci_port *)port;
 	struct tty_struct *tty = port->info->tty;
 	int i, count, copied = 0;
 	unsigned short status;
@@ -501,18 +462,11 @@ static inline void sci_receive_chars(struct uart_port *port,
 
 	while (1) {
 #if !defined(SCI_ONLY)
-		if (port->type == PORT_SCIF) {
-#if defined(CONFIG_CPU_SUBTYPE_SH7760) || defined(CONFIG_CPU_SUBTYPE_SH7780)
-			count = sci_in(port, SCRFDR) & 0x7f;
-#else
-			count = sci_in(port, SCFDR)&SCIF_RFDC_MASK ;
-#endif
-		} else {
-			count = (sci_in(port, SCxSR)&SCxSR_RDxF(port))?1:0;
-		}
-#else
-		count = (sci_in(port, SCxSR)&SCxSR_RDxF(port))?1:0;
+		if (port->type == PORT_SCIF)
+			count = scif_rxroom(port);
+		else
 #endif
+			count = sci_rxroom(port);
 
 		/* Don't copy more bytes than there is room for in the buffer */
 		count = tty_buffer_request_room(tty, count);
@@ -523,11 +477,10 @@ static inline void sci_receive_chars(struct uart_port *port,
 
 		if (port->type == PORT_SCI) {
 			char c = sci_in(port, SCxRDR);
-                       if(((struct sci_port *)port)->break_flag
-			    || uart_handle_sysrq_char(port, c, regs)) {
+			if (uart_handle_sysrq_char(port, c, regs) || sci_port->break_flag)
 				count = 0;
-			} else {
-			    tty_insert_flip_char(tty, c, TTY_NORMAL);
+			else {
+				tty_insert_flip_char(tty, c, TTY_NORMAL);
 			}
 		} else {
 			for (i=0; i<count; i++) {
@@ -535,15 +488,17 @@ static inline void sci_receive_chars(struct uart_port *port,
 				status = sci_in(port, SCxSR);
 #if defined(CONFIG_CPU_SH3)
 				/* Skip "chars" during break */
-				if (((struct sci_port *)port)->break_flag) {
+				if (sci_port->break_flag) {
 					if ((c == 0) &&
 					    (status & SCxSR_FER(port))) {
 						count--; i--;
 						continue;
 					}
+
 					/* Nonzero => end-of-break */
 					pr_debug("scif: debounce<%02x>\n", c);
-					((struct sci_port *)port)->break_flag = 0;
+					sci_port->break_flag = 0;
+
 					if (STEPFN(c)) {
 						count--; i--;
 						continue;
@@ -600,15 +555,17 @@ static void sci_schedule_break_timer(struct sci_port *port)
 /* Ensure that two consecutive samples find the break over. */
 static void sci_break_timer(unsigned long data)
 {
-    struct sci_port * port = (struct sci_port *)data;
-	if(sci_rxd_in(&port->port) == 0) {
+	struct sci_port *port = (struct sci_port *)data;
+
+	if (sci_rxd_in(&port->port) == 0) {
 		port->break_flag = 1;
-	    sci_schedule_break_timer(port);
-	} else if(port->break_flag == 1){
+		sci_schedule_break_timer(port);
+	} else if (port->break_flag == 1) {
 		/* break is over. */
 		port->break_flag = 2;
-	    sci_schedule_break_timer(port);
-	} else port->break_flag = 0;
+		sci_schedule_break_timer(port);
+	} else
+		port->break_flag = 0;
 }
 
 static inline int sci_handle_errors(struct uart_port *port)
@@ -617,40 +574,41 @@ static inline int sci_handle_errors(struct uart_port *port)
 	unsigned short status = sci_in(port, SCxSR);
 	struct tty_struct *tty = port->info->tty;
 
-	if (status&SCxSR_ORER(port)) {
+	if (status & SCxSR_ORER(port)) {
 		/* overrun error */
-		if(tty_insert_flip_char(tty, 0, TTY_OVERRUN))
+		if (tty_insert_flip_char(tty, 0, TTY_OVERRUN))
 			copied++;
 		pr_debug("sci: overrun error\n");
 	}
 
-	if (status&SCxSR_FER(port)) {
+	if (status & SCxSR_FER(port)) {
 		if (sci_rxd_in(port) == 0) {
 			/* Notify of BREAK */
-			struct sci_port * sci_port = (struct sci_port *)port;
-			if(!sci_port->break_flag) {
-	                	sci_port->break_flag = 1;
-	                	sci_schedule_break_timer((struct sci_port *)port);
+			struct sci_port *sci_port = (struct sci_port *)port;
+
+			if (!sci_port->break_flag) {
+				sci_port->break_flag = 1;
+				sci_schedule_break_timer(sci_port);
+
 				/* Do sysrq handling. */
-				if(uart_handle_break(port))
+				if (uart_handle_break(port))
 					return 0;
 			        pr_debug("sci: BREAK detected\n");
-			        if(tty_insert_flip_char(tty, 0, TTY_BREAK))
+				if (tty_insert_flip_char(tty, 0, TTY_BREAK))
 				        copied++;
                        }
-		}
-		else {
+		} else {
 			/* frame error */
-			if(tty_insert_flip_char(tty, 0, TTY_FRAME))
+			if (tty_insert_flip_char(tty, 0, TTY_FRAME))
 				copied++;
 			pr_debug("sci: frame error\n");
 		}
 	}
 
-	if (status&SCxSR_PER(port)) {
-		if(tty_insert_flip_char(tty, 0, TTY_PARITY))
-			copied++;
+	if (status & SCxSR_PER(port)) {
 		/* parity error */
+		if (tty_insert_flip_char(tty, 0, TTY_PARITY))
+			copied++;
 		pr_debug("sci: parity error\n");
 	}
 
@@ -673,7 +631,7 @@ static inline int sci_handle_breaks(struct uart_port *port)
 		s->break_flag = 1;
 #endif
 		/* Notify of BREAK */
-		if(tty_insert_flip_char(tty, 0, TTY_BREAK))
+		if (tty_insert_flip_char(tty, 0, TTY_BREAK))
 			copied++;
 		pr_debug("sci: BREAK detected\n");
 	}
@@ -682,7 +640,7 @@ static inline int sci_handle_breaks(struct uart_port *port)
 	/* XXX: Handle SCIF overrun error */
 	if (port->type == PORT_SCIF && (sci_in(port, SCLSR) & SCIF_ORER) != 0) {
 		sci_out(port, SCLSR, 0);
-		if(tty_insert_flip_char(tty, 0, TTY_OVERRUN)) {
+		if (tty_insert_flip_char(tty, 0, TTY_OVERRUN)) {
 			copied++;
 			pr_debug("sci: overrun error\n");
 		}
@@ -691,13 +649,12 @@ static inline int sci_handle_breaks(struct uart_port *port)
 
 	if (copied)
 		tty_flip_buffer_push(tty);
+
 	return copied;
 }
 
-static irqreturn_t sci_rx_interrupt(int irq, void *ptr, struct pt_regs *regs)
+static irqreturn_t sci_rx_interrupt(int irq, void *port, struct pt_regs *regs)
 {
-	struct uart_port *port = ptr;
-
 	/* I think sci_receive_chars has to be called irrespective
 	 * of whether the I_IXOFF is set, otherwise, how is the interrupt
 	 * to be disabled?
@@ -711,7 +668,9 @@ static irqreturn_t sci_tx_interrupt(int irq, void *ptr, struct pt_regs *regs)
 {
 	struct uart_port *port = ptr;
 
+	spin_lock_irq(&port->lock);
 	sci_transmit_chars(port);
+	spin_unlock_irq(&port->lock);
 
 	return IRQ_HANDLED;
 }
@@ -755,6 +714,12 @@ static irqreturn_t sci_br_interrupt(int irq, void *ptr, struct pt_regs *regs)
 
 	/* Handle BREAKs */
 	sci_handle_breaks(port);
+
+#ifdef CONFIG_SH_KGDB
+	/* Break into the debugger if a break is detected */
+	BREAKPOINT();
+#endif
+
 	sci_out(port, SCxSR, SCxSR_BREAK_CLEAR(port));
 
 	return IRQ_HANDLED;
@@ -769,16 +734,16 @@ static irqreturn_t sci_mpxed_interrupt(int irq, void *ptr, struct pt_regs *regs)
         scr_status = sci_in(port,SCSCR);
 
 	/* Tx Interrupt */
-        if ((ssr_status&0x0020) && (scr_status&0x0080))
+        if ((ssr_status & 0x0020) && (scr_status & 0x0080))
                 sci_tx_interrupt(irq, ptr, regs);
 	/* Rx Interrupt */
-        if ((ssr_status&0x0002) && (scr_status&0x0040))
+        if ((ssr_status & 0x0002) && (scr_status & 0x0040))
                 sci_rx_interrupt(irq, ptr, regs);
 	/* Error Interrupt */
-        if ((ssr_status&0x0080) && (scr_status&0x0400))
+        if ((ssr_status & 0x0080) && (scr_status & 0x0400))
                 sci_er_interrupt(irq, ptr, regs);
 	/* Break Interrupt */
-        if ((ssr_status&0x0010) && (scr_status&0x0200))
+        if ((ssr_status & 0x0010) && (scr_status & 0x0200))
                 sci_br_interrupt(irq, ptr, regs);
 
 	return IRQ_HANDLED;
@@ -789,7 +754,8 @@ static irqreturn_t sci_mpxed_interrupt(int irq, void *ptr, struct pt_regs *regs)
  * Here we define a transistion notifier so that we can update all of our
  * ports' baud rate when the peripheral clock changes.
  */
-static int sci_notifier(struct notifier_block *self, unsigned long phase, void *p)
+static int sci_notifier(struct notifier_block *self,
+			unsigned long phase, void *p)
 {
 	struct cpufreq_freqs *freqs = p;
 	int i;
@@ -816,8 +782,9 @@ static int sci_notifier(struct notifier_block *self, unsigned long phase, void *
 			clk_put(clk);
 		}
 
-		printk("%s: got a postchange notification for cpu %d (old %d, new %d)\n",
-				__FUNCTION__, freqs->cpu, freqs->old, freqs->new);
+		printk(KERN_INFO "%s: got a postchange notification "
+		       "for cpu %d (old %d, new %d)\n",
+		       __FUNCTION__, freqs->cpu, freqs->old, freqs->new);
 	}
 
 	return NOTIFY_OK;
@@ -841,8 +808,9 @@ static int sci_request_irq(struct sci_port *port)
 			printk(KERN_ERR "sci: Cannot allocate irq.(IRQ=0)\n");
 			return -ENODEV;
 		}
-		if (request_irq(port->irqs[0], sci_mpxed_interrupt, IRQF_DISABLED,
-				"sci", port)) {
+
+		if (request_irq(port->irqs[0], sci_mpxed_interrupt,
+				SA_INTERRUPT, "sci", port)) {
 			printk(KERN_ERR "sci: Cannot allocate irq.\n");
 			return -ENODEV;
 		}
@@ -850,8 +818,8 @@ static int sci_request_irq(struct sci_port *port)
 		for (i = 0; i < ARRAY_SIZE(handlers); i++) {
 			if (!port->irqs[i])
 				continue;
-			if (request_irq(port->irqs[i], handlers[i], IRQF_DISABLED,
-					desc[i], port)) {
+			if (request_irq(port->irqs[i], handlers[i],
+					SA_INTERRUPT, desc[i], port)) {
 				printk(KERN_ERR "sci: Cannot allocate irq.\n");
 				return -ENODEV;
 			}
@@ -903,50 +871,42 @@ static unsigned int sci_get_mctrl(struct uart_port *port)
 
 static void sci_start_tx(struct uart_port *port)
 {
-	struct sci_port *s = &sci_ports[port->line];
+	unsigned short ctrl;
 
-	disable_irq(s->irqs[SCIx_TXI_IRQ]);
-	sci_transmit_chars(port);
-	enable_irq(s->irqs[SCIx_TXI_IRQ]);
+	/* Set TIE (Transmit Interrupt Enable) bit in SCSCR */
+	ctrl = sci_in(port, SCSCR);
+	ctrl |= SCI_CTRL_FLAGS_TIE;
+	sci_out(port, SCSCR, ctrl);
 }
 
 static void sci_stop_tx(struct uart_port *port)
 {
-	unsigned long flags;
 	unsigned short ctrl;
 
 	/* Clear TIE (Transmit Interrupt Enable) bit in SCSCR */
-	local_irq_save(flags);
 	ctrl = sci_in(port, SCSCR);
 	ctrl &= ~SCI_CTRL_FLAGS_TIE;
 	sci_out(port, SCSCR, ctrl);
-	local_irq_restore(flags);
 }
 
 static void sci_start_rx(struct uart_port *port, unsigned int tty_start)
 {
-	unsigned long flags;
 	unsigned short ctrl;
 
 	/* Set RIE (Receive Interrupt Enable) bit in SCSCR */
-	local_irq_save(flags);
 	ctrl = sci_in(port, SCSCR);
 	ctrl |= SCI_CTRL_FLAGS_RIE | SCI_CTRL_FLAGS_REIE;
 	sci_out(port, SCSCR, ctrl);
-	local_irq_restore(flags);
 }
 
 static void sci_stop_rx(struct uart_port *port)
 {
-	unsigned long flags;
 	unsigned short ctrl;
 
 	/* Clear RIE (Receive Interrupt Enable) bit in SCSCR */
-	local_irq_save(flags);
 	ctrl = sci_in(port, SCSCR);
 	ctrl &= ~(SCI_CTRL_FLAGS_RIE | SCI_CTRL_FLAGS_REIE);
 	sci_out(port, SCSCR, ctrl);
-	local_irq_restore(flags);
 }
 
 static void sci_enable_ms(struct uart_port *port)
@@ -963,9 +923,8 @@ static int sci_startup(struct uart_port *port)
 {
 	struct sci_port *s = &sci_ports[port->line];
 
-#if defined(__H8300S__)
-	h8300_sci_enable(port, sci_enable);
-#endif
+	if (s->enable)
+		s->enable(port);
 
 	sci_request_irq(s);
 	sci_start_tx(port);
@@ -982,9 +941,8 @@ static void sci_shutdown(struct uart_port *port)
 	sci_stop_tx(port);
 	sci_free_irq(s);
 
-#if defined(__H8300S__)
-	h8300_sci_enable(port, sci_disable);
-#endif
+	if (s->disable)
+		s->disable(port);
 }
 
 static void sci_set_termios(struct uart_port *port, struct termios *termios,
@@ -997,6 +955,23 @@ static void sci_set_termios(struct uart_port *port, struct termios *termios,
 
 	baud = uart_get_baud_rate(port, termios, old, 0, port->uartclk/16);
 
+	switch (baud) {
+		case 0:
+			t = -1;
+			break;
+		default:
+		{
+#if defined(CONFIG_SUPERH) && !defined(CONFIG_SUPERH64)
+			struct clk *clk = clk_get("module_clk");
+			t = SCBRR_VALUE(baud, clk_get_rate(clk));
+			clk_put(clk);
+#else
+			t = SCBRR_VALUE(baud);
+#endif
+		}
+			break;
+	}
+
 	spin_lock_irqsave(&port->lock, flags);
 
 	do {
@@ -1006,9 +981,8 @@ static void sci_set_termios(struct uart_port *port, struct termios *termios,
 	sci_out(port, SCSCR, 0x00);	/* TE=0, RE=0, CKE1=0 */
 
 #if !defined(SCI_ONLY)
-	if (port->type == PORT_SCIF) {
+	if (port->type == PORT_SCIF)
 		sci_out(port, SCFCR, SCFCR_RFRST | SCFCR_TFRST);
-	}
 #endif
 
 	smr_val = sci_in(port, SCSMR) & 3;
@@ -1025,23 +999,6 @@ static void sci_set_termios(struct uart_port *port, struct termios *termios,
 
 	sci_out(port, SCSMR, smr_val);
 
-	switch (baud) {
-		case 0:
-			t = -1;
-			break;
-		default:
-		{
-#if defined(CONFIG_SUPERH) && !defined(CONFIG_SUPERH64)
-			struct clk *clk = clk_get("module_clk");
-			t = SCBRR_VALUE(baud, clk_get_rate(clk));
-			clk_put(clk);
-#else
-			t = SCBRR_VALUE(baud);
-#endif
-		}
-			break;
-	}
-
 	if (t > 0) {
 		if(t >= 256) {
 			sci_out(port, SCSMR, (sci_in(port, SCSMR) & ~3) | 1);
@@ -1092,11 +1049,23 @@ static void sci_config_port(struct uart_port *port, int flags)
 
 	port->type = s->type;
 
+	switch (port->type) {
+	case PORT_SCI:
+		s->init_pins = sci_init_pins_sci;
+		break;
+	case PORT_SCIF:
+		s->init_pins = sci_init_pins_scif;
+		break;
+	case PORT_IRDA:
+		s->init_pins = sci_init_pins_irda;
+		break;
+	}
+
 #if defined(CONFIG_CPU_SUBTYPE_SH5_101) || defined(CONFIG_CPU_SUBTYPE_SH5_103)
 	if (port->mapbase == 0)
 		port->mapbase = onchip_remap(SCIF_ADDR_SH5, 1024, "SCIF");
 
-	port->membase = (void *)port->mapbase;
+	port->membase = (void __iomem *)port->mapbase;
 #endif
 }
 
@@ -1132,412 +1101,61 @@ static struct uart_ops sci_uart_ops = {
 	.verify_port	= sci_verify_port,
 };
 
-static struct sci_port sci_ports[] = {
-#if defined(CONFIG_CPU_SUBTYPE_SH7708)
-	{
-		.port	= {
-			.membase	= (void *)0xfffffe80,
-			.mapbase	= 0xfffffe80,
-			.iotype		= UPIO_MEM,
-			.irq		= 25,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCI,
-		.irqs		= SCI_IRQS,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH7705)
-	{
-		.port	= {
-			.membase	= (void *)SCIF0,
-			.mapbase	= SCIF0,
-			.iotype		= UPIO_MEM,
-			.irq		= 55,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH3_IRDA_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port	= {
-			.membase	= (void *)SCIF2,
-			.mapbase	= SCIF2,
-			.iotype		= UPIO_MEM,
-			.irq		= 59,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH3_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	}
-#elif defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709)
-	{
-		.port	= {
-			.membase	= (void *)0xfffffe80,
-			.mapbase	= 0xfffffe80,
-			.iotype		= UPIO_MEM,
-			.irq		= 25,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCI,
-		.irqs		= SCI_IRQS,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xa4000150,
-			.mapbase	= 0xa4000150,
-			.iotype		= UPIO_MEM,
-			.irq		= 59,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH3_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xa4000140,
-			.mapbase	= 0xa4000140,
-			.iotype		= UPIO_MEM,
-			.irq		= 55,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 2,
-		},
-		.type		= PORT_IRDA,
-		.irqs		= SH3_IRDA_IRQS,
-		.init_pins	= sci_init_pins_irda,
-	}
-#elif defined(CONFIG_CPU_SUBTYPE_SH7300)
-	{
-		.port	= {
-			.membase	= (void *)0xA4430000,
-			.mapbase	= 0xA4430000,
-			.iotype		= UPIO_MEM,
-			.irq		= 25,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7300_SCIF0_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH73180)
-	{
-		.port	= {
-			.membase	= (void *)0xffe00000,
-			.mapbase	= 0xffe00000,
-			.iotype		= UPIO_MEM,
-			.irq		= 25,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH73180_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH4_202)
-	{
-		.port	= {
-			.membase	= (void *)0xffe80000,
-			.mapbase	= 0xffe80000,
-			.iotype		= UPIO_MEM,
-			.irq		= 43,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH4_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH7750) || defined(CONFIG_CPU_SUBTYPE_SH7751)
-	{
-		.port	= {
-			.membase	= (void *)0xffe00000,
-			.mapbase	= 0xffe00000,
-			.iotype		= UPIO_MEM,
-			.irq		= 25,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCI,
-		.irqs		= SCI_IRQS,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xffe80000,
-			.mapbase	= 0xffe80000,
-			.iotype		= UPIO_MEM,
-			.irq		= 43,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH4_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH7760)
-	{
-		.port	= {
-			.membase	= (void *)0xfe600000,
-			.mapbase	= 0xfe600000,
-			.iotype		= UPIO_MEM,
-			.irq		= 55,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7760_SCIF0_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xfe610000,
-			.mapbase	= 0xfe610000,
-			.iotype		= UPIO_MEM,
-			.irq		= 75,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7760_SCIF1_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xfe620000,
-			.mapbase	= 0xfe620000,
-			.iotype		= UPIO_MEM,
-			.irq		= 79,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 2,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7760_SCIF2_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_ST40STB1)
-	{
-		.port	= {
-			.membase	= (void *)0xffe00000,
-			.mapbase	= 0xffe00000,
-			.iotype		= UPIO_MEM,
-			.irq		= 26,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= STB1_SCIF1_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0xffe80000,
-			.mapbase	= 0xffe80000,
-			.iotype		= UPIO_MEM,
-			.irq		= 43,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH4_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH5_101) || defined(CONFIG_CPU_SUBTYPE_SH5_103)
-	{
-		.port	= {
-			.iotype		= UPIO_MEM,
-			.irq		= 42,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH5_SCIF_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_H83007) || defined(CONFIG_H83068)
-	{
-		.port	= {
-			.membase	= (void *)0x00ffffb0,
-			.mapbase	= 0x00ffffb0,
-			.iotype		= UPIO_MEM,
-			.irq		= 54,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8300H_SCI_IRQS0,
-		.init_pins	= sci_init_pins_sci,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0x00ffffb8,
-			.mapbase	= 0x00ffffb8,
-			.iotype		= UPIO_MEM,
-			.irq		= 58,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8300H_SCI_IRQS1,
-		.init_pins	= sci_init_pins_sci,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0x00ffffc0,
-			.mapbase	= 0x00ffffc0,
-			.iotype		= UPIO_MEM,
-			.irq		= 62,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 2,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8300H_SCI_IRQS2,
-		.init_pins	= sci_init_pins_sci,
-	},
-#elif defined(CONFIG_H8S2678)
-	{
-		.port	= {
-			.membase	= (void *)0x00ffff78,
-			.mapbase	= 0x00ffff78,
-			.iotype		= UPIO_MEM,
-			.irq		= 90,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8S_SCI_IRQS0,
-		.init_pins	= sci_init_pins_sci,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0x00ffff80,
-			.mapbase	= 0x00ffff80,
-			.iotype		= UPIO_MEM,
-			.irq		= 94,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8S_SCI_IRQS1,
-		.init_pins	= sci_init_pins_sci,
-	},
-	{
-		.port	= {
-			.membase	= (void *)0x00ffff88,
-			.mapbase	= 0x00ffff88,
-			.iotype		= UPIO_MEM,
-			.irq		= 98,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 2,
-		},
-		.type		= PORT_SCI,
-		.irqs		= H8S_SCI_IRQS2,
-		.init_pins	= sci_init_pins_sci,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH7770)
-	{
-		.port   = {
-			.membase	= (void *)0xff923000,
-			.mapbase	= 0xff923000,
-			.iotype		= UPIO_MEM,
-			.irq		= 61,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7770_SCIF0_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port   = {
-			.membase	= (void *)0xff924000,
-			.mapbase	= 0xff924000,
-			.iotype		= UPIO_MEM,
-			.irq		= 62,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7770_SCIF1_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port   = {
-			.membase	= (void *)0xff925000,
-			.mapbase	= 0xff925000,
-			.iotype		= UPIO_MEM,
-			.irq		= 63,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 2,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7770_SCIF2_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-#elif defined(CONFIG_CPU_SUBTYPE_SH7780)
-	{
-		.port   = {
-			.membase	= (void *)0xffe00000,
-			.mapbase	= 0xffe00000,
-			.iotype		= UPIO_MEM,
-			.irq		= 43,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 0,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7780_SCIF0_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
-	{
-		.port   = {
-			.membase	= (void *)0xffe10000,
-			.mapbase	= 0xffe10000,
-			.iotype		= UPIO_MEM,
-			.irq		= 79,
-			.ops		= &sci_uart_ops,
-			.flags		= UPF_BOOT_AUTOCONF,
-			.line		= 1,
-		},
-		.type		= PORT_SCIF,
-		.irqs		= SH7780_SCIF1_IRQS,
-		.init_pins	= sci_init_pins_scif,
-	},
+static void __init sci_init_ports(void)
+{
+	static int first = 1;
+	int i;
+
+	if (!first)
+		return;
+
+	first = 0;
+
+	for (i = 0; i < SCI_NPORTS; i++) {
+		sci_ports[i].port.ops		= &sci_uart_ops;
+		sci_ports[i].port.iotype	= UPIO_MEM;
+		sci_ports[i].port.line		= i;
+		sci_ports[i].port.fifosize	= 1;
+
+#if defined(__H8300H__) || defined(__H8300S__)
+#ifdef __H8300S__
+		sci_ports[i].enable	= h8300_sci_enable;
+		sci_ports[i].disable	= h8300_sci_disable;
+#endif
+		sci_ports[i].port.uartclk = CONFIG_CPU_CLOCK;
+#elif defined(CONFIG_SUPERH64)
+		sci_ports[i].port.uartclk = current_cpu_data.module_clock * 16;
 #else
-#error "CPU subtype not defined"
+		/*
+		 * XXX: We should use a proper SCI/SCIF clock
+		 */
+		{
+			struct clk *clk = clk_get("module_clk");
+			sci_ports[i].port.uartclk = clk_get_rate(clk) * 16;
+			clk_put(clk);
+		}
 #endif
-};
+
+		sci_ports[i].break_timer.data = (unsigned long)&sci_ports[i];
+		sci_ports[i].break_timer.function = sci_break_timer;
+
+		init_timer(&sci_ports[i].break_timer);
+	}
+}
+
+int __init early_sci_setup(struct uart_port *port)
+{
+	if (unlikely(port->line > SCI_NPORTS))
+		return -ENODEV;
+
+	sci_init_ports();
+
+	sci_ports[port->line].port.membase	= port->membase;
+	sci_ports[port->line].port.mapbase	= port->mapbase;
+	sci_ports[port->line].port.type		= port->type;
+
+	return 0;
+}
 
 #ifdef CONFIG_SERIAL_SH_SCI_CONSOLE
 /*
@@ -1559,34 +1177,38 @@ static int __init serial_console_setup(struct console *co, char *options)
 	int flow = 'n';
 	int ret;
 
+	/*
+	 * Check whether an invalid uart number has been specified, and
+	 * if so, search for the first available port that does have
+	 * console support.
+	 */
+	if (co->index >= SCI_NPORTS)
+		co->index = 0;
+
 	serial_console_port = &sci_ports[co->index];
 	port = &serial_console_port->port;
-	port->type = serial_console_port->type;
-
-#ifdef CONFIG_SUPERH64
-	/* This is especially needed on sh64 to remap the SCIF */
-	sci_config_port(port, 0);
-#endif
 
 	/*
-	 * We need to set the initial uartclk here, since otherwise it will
-	 * only ever be setup at sci_init() time.
+	 * Also need to check port->type, we don't actually have any
+	 * UPIO_PORT ports, but uart_report_port() handily misreports
+	 * it anyways if we don't have a port available by the time this is
+	 * called.
 	 */
-#if defined(__H8300H__) || defined(__H8300S__)
-	port->uartclk = CONFIG_CPU_CLOCK;
+	if (!port->type)
+		return -ENODEV;
+	if (!port->membase || !port->mapbase)
+		return -ENODEV;
+
+	spin_lock_init(&port->lock);
+
+	port->type = serial_console_port->type;
+
+	if (port->flags & UPF_IOREMAP)
+		sci_config_port(port, 0);
+
+	if (serial_console_port->enable)
+		serial_console_port->enable(port);
 
-#if defined(__H8300S__)
-	h8300_sci_enable(port, sci_enable);
-#endif
-#elif defined(CONFIG_SUPERH64)
-	port->uartclk = current_cpu_data.module_clock * 16;
-#else
-	{
-		struct clk *clk = clk_get("module_clk");
-		port->uartclk = clk_get_rate(clk) * 16;
-		clk_put(clk);
-	}
-#endif
 	if (options)
 		uart_parse_options(options, &baud, &parity, &bits, &flow);
 
@@ -1604,17 +1226,17 @@ static struct console serial_console = {
 	.device		= uart_console_device,
 	.write		= serial_console_write,
 	.setup		= serial_console_setup,
-	.flags		= CON_PRINTBUFFER,
+	.flags		= CON_PRINTBUFFER, 
 	.index		= -1,
 	.data		= &sci_uart_driver,
 };
 
 static int __init sci_console_init(void)
 {
+	sci_init_ports();
 	register_console(&serial_console);
 	return 0;
 }
-
 console_initcall(sci_console_init);
 #endif /* CONFIG_SERIAL_SH_SCI_CONSOLE */
 
@@ -1649,6 +1271,8 @@ int __init kgdb_console_setup(struct console *co, char *options)
 	int parity = 'n';
 	int flow = 'n';
 
+	spin_lock_init(&port->lock);
+
 	if (co->index != kgdb_portnum)
 		co->index = kgdb_portnum;
 
@@ -1677,10 +1301,10 @@ static struct console kgdb_console = {
 /* Register the KGDB console so we get messages (d'oh!) */
 static int __init kgdb_console_init(void)
 {
+	sci_init_ports();
 	register_console(&kgdb_console);
 	return 0;
 }
-
 console_initcall(kgdb_console_init);
 #endif /* CONFIG_SH_KGDB_CONSOLE */
 
@@ -1701,60 +1325,132 @@ static struct uart_driver sci_uart_driver = {
 	.dev_name	= "ttySC",
 	.major		= SCI_MAJOR,
 	.minor		= SCI_MINOR_START,
+	.nr		= SCI_NPORTS,
 	.cons		= SCI_CONSOLE,
 };
 
-static int __init sci_init(void)
+/*
+ * Register a set of serial devices attached to a platform device.  The
+ * list is terminated with a zero flags entry, which means we expect
+ * all entries to have at least UPF_BOOT_AUTOCONF set. Platforms that need
+ * remapping (such as sh64) should also set UPF_IOREMAP.
+ */
+static int __devinit sci_probe(struct platform_device *dev)
 {
-	int chan, ret;
+	struct plat_sci_port *p = dev->dev.platform_data;
+	int i;
 
-	printk("%s", banner);
+	for (i = 0; p && p->flags != 0 && i < SCI_NPORTS; p++, i++) {
+		struct sci_port *sciport = &sci_ports[i];
 
-	sci_uart_driver.nr = ARRAY_SIZE(sci_ports);
+		sciport->port.mapbase	= p->mapbase;
 
-	ret = uart_register_driver(&sci_uart_driver);
-	if (ret == 0) {
-		for (chan = 0; chan < SCI_NPORTS; chan++) {
-			struct sci_port *sciport = &sci_ports[chan];
+		/*
+		 * For the simple (and majority of) cases where we don't need
+		 * to do any remapping, just cast the cookie directly.
+		 */
+		if (p->mapbase && !p->membase && !(p->flags & UPF_IOREMAP))
+			p->membase = (void __iomem *)p->mapbase;
 
-#if defined(__H8300H__) || defined(__H8300S__)
-			sciport->port.uartclk = CONFIG_CPU_CLOCK;
-#elif defined(CONFIG_SUPERH64)
-			sciport->port.uartclk = current_cpu_data.module_clock * 16;
-#else
-			struct clk *clk = clk_get("module_clk");
-			sciport->port.uartclk = clk_get_rate(clk) * 16;
-			clk_put(clk);
-#endif
-			uart_add_one_port(&sci_uart_driver, &sciport->port);
-			sciport->break_timer.data = (unsigned long)sciport;
-			sciport->break_timer.function = sci_break_timer;
-			init_timer(&sciport->break_timer);
-		}
+		sciport->port.membase	= p->membase;
+
+		sciport->port.irq	= p->irqs[SCIx_TXI_IRQ];
+		sciport->port.flags	= p->flags;
+		sciport->port.dev	= &dev->dev;
+
+		sciport->type		= sciport->port.type = p->type;
+
+		memcpy(&sciport->irqs, &p->irqs, sizeof(p->irqs));
+
+		uart_add_one_port(&sci_uart_driver, &sciport->port);
 	}
 
 #ifdef CONFIG_CPU_FREQ
 	cpufreq_register_notifier(&sci_nb, CPUFREQ_TRANSITION_NOTIFIER);
-	printk("sci: CPU frequency notifier registered\n");
+	dev_info(&dev->dev, "sci: CPU frequency notifier registered\n");
 #endif
 
 #ifdef CONFIG_SH_STANDARD_BIOS
 	sh_bios_gdb_detach();
 #endif
 
-	return ret;
+	return 0;
 }
 
-static void __exit sci_exit(void)
+static int __devexit sci_remove(struct platform_device *dev)
+{
+	int i;
+
+	for (i = 0; i < SCI_NPORTS; i++)
+		uart_remove_one_port(&sci_uart_driver, &sci_ports[i].port);
+
+	return 0;
+}
+
+static int sci_suspend(struct platform_device *dev, pm_message_t state)
 {
-	int chan;
+	int i;
+
+	for (i = 0; i < SCI_NPORTS; i++) {
+		struct sci_port *p = &sci_ports[i];
+
+		if (p->type != PORT_UNKNOWN && p->port.dev == &dev->dev)
+			uart_suspend_port(&sci_uart_driver, &p->port);
+	}
 
-	for (chan = 0; chan < SCI_NPORTS; chan++)
-		uart_remove_one_port(&sci_uart_driver, &sci_ports[chan].port);
+	return 0;
+}
 
+static int sci_resume(struct platform_device *dev)
+{
+	int i;
+
+	for (i = 0; i < SCI_NPORTS; i++) {
+		struct sci_port *p = &sci_ports[i];
+
+		if (p->type != PORT_UNKNOWN && p->port.dev == &dev->dev)
+			uart_resume_port(&sci_uart_driver, &p->port);
+	}
+
+	return 0;
+}
+
+static struct platform_driver sci_driver = {
+	.probe		= sci_probe,
+	.remove		= __devexit_p(sci_remove),
+	.suspend	= sci_suspend,
+	.resume		= sci_resume,
+	.driver		= {
+		.name	= "sh-sci",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init sci_init(void)
+{
+	int ret;
+
+	printk(banner);
+
+	sci_init_ports();
+
+	ret = uart_register_driver(&sci_uart_driver);
+	if (likely(ret == 0)) {
+		ret = platform_driver_register(&sci_driver);
+		if (unlikely(ret))
+			uart_unregister_driver(&sci_uart_driver);
+	}
+
+	return ret;
+}
+
+static void __exit sci_exit(void)
+{
+	platform_driver_unregister(&sci_driver);
 	uart_unregister_driver(&sci_uart_driver);
 }
 
 module_init(sci_init);
 module_exit(sci_exit);
 
+MODULE_LICENSE("GPL");
diff --git a/drivers/serial/sh-sci.h b/drivers/serial/sh-sci.h
index ab320fa3237c..28643c4dc850 100644
--- a/drivers/serial/sh-sci.h
+++ b/drivers/serial/sh-sci.h
@@ -10,7 +10,9 @@
  *  Modified to support SH7300(SH-Mobile) SCIF. Takashi Kusuda (Jun 2003).
  *  Modified to support H8/300 Series Yoshinori Sato (Feb 2004).
  */
+#include <linux/config.h>
 #include <linux/serial_core.h>
+#include <asm/io.h>
 
 #if defined(__H8300H__) || defined(__H8300S__)
 #include <asm/gpio.h>
@@ -22,40 +24,13 @@
 #endif
 #endif
 
-/* Offsets into the sci_port->irqs array */
-#define SCIx_ERI_IRQ 0
-#define SCIx_RXI_IRQ 1
-#define SCIx_TXI_IRQ 2
-
-/*                     ERI, RXI, TXI, BRI */
-#define SCI_IRQS      { 23,  24,  25,   0 }
-#define SH3_SCIF_IRQS { 56,  57,  59,  58 }
-#define SH3_IRDA_IRQS { 52,  53,  55,  54 }
-#define SH4_SCIF_IRQS { 40,  41,  43,  42 }
-#define STB1_SCIF1_IRQS {23, 24,  26,  25 }
-#define SH7760_SCIF0_IRQS { 52, 53, 55, 54 }
-#define SH7760_SCIF1_IRQS { 72, 73, 75, 74 }
-#define SH7760_SCIF2_IRQS { 76, 77, 79, 78 }
-#define SH7300_SCIF0_IRQS {80,  80,  80,  80 }
-#define SH73180_SCIF_IRQS {80,  81,  83,  82 }
-#define H8300H_SCI_IRQS0 {52, 53, 54,   0 }
-#define H8300H_SCI_IRQS1 {56, 57, 58,   0 }
-#define H8300H_SCI_IRQS2 {60, 61, 62,   0 }
-#define H8S_SCI_IRQS0 {88, 89, 90,   0 }
-#define H8S_SCI_IRQS1 {92, 93, 94,   0 }
-#define H8S_SCI_IRQS2 {96, 97, 98,   0 }
-#define SH5_SCIF_IRQS {39, 40, 42,   0 }
-#define	SH7770_SCIF0_IRQS {61, 61, 61, 61 }
-#define	SH7770_SCIF1_IRQS {62, 62, 62, 62 }
-#define	SH7770_SCIF2_IRQS {63, 63, 63, 63 }
-#define	SH7780_SCIF0_IRQS {40, 41, 43, 42 }
-#define	SH7780_SCIF1_IRQS {76, 77, 79, 78 }
-
 #if defined(CONFIG_CPU_SUBTYPE_SH7708)
 # define SCSPTR 0xffffff7c /* 8 bit */
 # define SCSCR_INIT(port)          0x30 /* TIE=0,RIE=0,TE=1,RE=1 */
 # define SCI_ONLY
-#elif defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706)
 # define SCPCR  0xA4000116 /* 16 bit SCI and SCIF */
 # define SCPDR  0xA4000136 /* 8  bit SCI and SCIF */
 # define SCSCR_INIT(port)          0x30 /* TIE=0,RIE=0,TE=1,RE=1 */
@@ -99,12 +74,23 @@
 # define SCPDR  0xA4050136        /* 16 bit SCIF */
 # define SCSCR_INIT(port)  0x0030 /* TIE=0,RIE=0,TE=1,RE=1 */
 # define SCIF_ONLY
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+# define SCSPTR0 0xA4400000	  /* 16 bit SCIF */
+# define SCSCR_INIT(port)  0x0030 /* TIE=0,RIE=0,TE=1,RE=1 */
+# define SCIF_ONLY
 #elif defined(CONFIG_CPU_SUBTYPE_SH73180)
 # define SCPDR  0xA4050138        /* 16 bit SCIF */
 # define SCSPTR2 SCPDR
 # define SCIF_ORER 0x0001   /* overrun error bit */
 # define SCSCR_INIT(port)  0x0038 /* TIE=0,RIE=0,TE=1,RE=1 */
 # define SCIF_ONLY
+#elif defined(CONFIG_CPU_SUBTYPE_SH7343)
+# define SCSPTR0 0xffe00010	/* 16 bit SCIF */
+# define SCSPTR1 0xffe10010	/* 16 bit SCIF */
+# define SCSPTR2 0xffe20010	/* 16 bit SCIF */
+# define SCSPTR3 0xffe30010	/* 16 bit SCIF */
+# define SCSCR_INIT(port) 0x32	/* TIE=0,RIE=0,TE=1,RE=1,REIE=0,CKE=1 */
+# define SCIF_ONLY
 #elif defined(CONFIG_CPU_SUBTYPE_SH4_202)
 # define SCSPTR2 0xffe80020 /* 16 bit SCIF */
 # define SCIF_ORER 0x0001   /* overrun error bit */
@@ -145,7 +131,7 @@
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 # define SCSPTR0	0xffe00024	/* 16 bit SCIF */
 # define SCSPTR1	0xffe10024	/* 16 bit SCIF */
-# define SCIF_OPER	0x0001		/* Overrun error bit */
+# define SCIF_ORER	0x0001		/* Overrun error bit */
 # define SCSCR_INIT(port)	0x3a	/* TIE=0,RIE=0,TE=1,RE=1,REIE=1 */
 # define SCIF_ONLY
 #else
@@ -273,15 +259,6 @@
  */
 #define SCI_EVENT_WRITE_WAKEUP	0
 
-struct sci_port {
-	struct uart_port port;
-	int type;
-	unsigned char irqs[4]; /* ERI, RXI, TXI, BRI */
-	void (*init_pins)(struct uart_port *port, unsigned int cflag);
-	int break_flag;
-	struct timer_list break_timer;
-};
-
 #define SCI_IN(size, offset)					\
   unsigned int addr = port->mapbase + (offset);			\
   if ((size) == 8) {						\
@@ -336,7 +313,9 @@ struct sci_port {
   }
 
 #ifdef CONFIG_CPU_SH3
-#if defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define SCIF_FNS(name, scif_offset, scif_size) \
   CPU_SCIF_FNS(name, scif_offset, scif_size)
 #else
@@ -362,7 +341,9 @@ struct sci_port {
   CPU_SCIF_FNS(name, sh4_scif_offset, sh4_scif_size)
 #endif
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7710)
 SCIF_FNS(SCSMR,  0x00, 16)
 SCIF_FNS(SCBRR,  0x04,  8)
 SCIF_FNS(SCSCR,  0x08, 16)
@@ -447,7 +428,9 @@ static inline int sci_rxd_in(struct uart_port *port)
 		return ctrl_inb(SCSPTR)&0x01 ? 1 : 0; /* SCI */
 	return 1;
 }
-#elif defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709)
+#elif defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706)
 static inline int sci_rxd_in(struct uart_port *port)
 {
 	if (port->mapbase == 0xfffffe80)
@@ -467,6 +450,13 @@ static inline int sci_rxd_in(struct uart_port *port)
 		return ctrl_inb(SCPDR)&0x10 ? 1 : 0; /* SCIF */
 	return 1;
 }
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+static inline int sci_rxd_in(struct uart_port *port)
+{
+	if (port->mapbase == SCSPTR0)
+		return ctrl_inw(SCSPTR0 + 0x10) & 0x01 ? 1 : 0;
+	return 1;
+}
 #elif defined(CONFIG_CPU_SUBTYPE_SH7750) || \
       defined(CONFIG_CPU_SUBTYPE_SH7751) || \
       defined(CONFIG_CPU_SUBTYPE_SH4_202)
@@ -504,6 +494,19 @@ static inline int sci_rxd_in(struct uart_port *port)
 {
 	return ctrl_inb(SCPDR)&0x01 ? 1 : 0; /* SCIF0 */
 }
+#elif defined(CONFIG_CPU_SUBTYPE_SH7343)
+static inline int sci_rxd_in(struct uart_port *port)
+{
+	if (port->mapbase == 0xffe00000)
+		return ctrl_inw(SCSPTR0) & 0x0001 ? 1 : 0; /* SCIF */
+	if (port->mapbase == 0xffe10000)
+		return ctrl_inw(SCSPTR1) & 0x0001 ? 1 : 0; /* SCIF */
+	if (port->mapbase == 0xffe20000)
+		return ctrl_inw(SCSPTR2) & 0x0001 ? 1 : 0; /* SCIF */
+	if (port->mapbase == 0xffe30000)
+		return ctrl_inw(SCSPTR3) & 0x0001 ? 1 : 0; /* SCIF */
+	return 1;
+}
 #elif defined(CONFIG_CPU_SUBTYPE_ST40STB1)
 static inline int sci_rxd_in(struct uart_port *port)
 {
@@ -587,4 +590,3 @@ static inline int sci_rxd_in(struct uart_port *port)
 #else /* Generic SH */
 #define SCBRR_VALUE(bps, clk) ((clk+16*bps)/(32*bps)-1)
 #endif
-
-- 
cgit v1.2.1


From ae1f19aeb715098ac06323f279383f3843429d97 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Wed, 27 Sep 2006 03:42:37 -0400
Subject: [libata] sata_mv: fix oops by filling in missing hook

Only two of three ata_port_operations structs had a ->data_xfer member,
which led to, uh, a lack of data xfer.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/sata_mv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c
index fdce6e07ecd2..c01496df4a99 100644
--- a/drivers/ata/sata_mv.c
+++ b/drivers/ata/sata_mv.c
@@ -463,6 +463,7 @@ static const struct ata_port_operations mv_iie_ops = {
 
 	.qc_prep		= mv_qc_prep_iie,
 	.qc_issue		= mv_qc_issue,
+	.data_xfer		= ata_mmio_data_xfer,
 
 	.eng_timeout		= mv_eng_timeout,
 
-- 
cgit v1.2.1


From 959f85f8a3223c116bbe95dd8a9b207790b5d4d3 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:43:28 +0900
Subject: sh: Consolidated SH7751/SH7780 PCI support.

This cleans up quite a lot of the PCI mess that we
currently have, and attempts to consolidate the
duplication in the SH7780 and SH7751 PCI controllers.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/landisk/Makefile         |   2 -
 arch/sh/boards/landisk/io.c             | 100 ++++------
 arch/sh/boards/mpc1211/setup.c          |  15 +-
 arch/sh/boards/renesas/hs7751rvoip/io.c | 116 ++++--------
 arch/sh/boards/renesas/r7780rp/io.c     | 103 ++++------
 arch/sh/boards/renesas/rts7751r2d/io.c  |  89 ++++-----
 arch/sh/boards/renesas/systemh/io.c     |  67 ++-----
 arch/sh/boards/se/770x/io.c             |  11 +-
 arch/sh/boards/se/7751/io.c             | 145 +++-----------
 arch/sh/boards/sh03/setup.c             |   6 +-
 arch/sh/boards/snapgear/io.c            |  84 +++------
 arch/sh/boards/titan/io.c               |  65 +++----
 arch/sh/cchips/hd6446x/hd64461/io.c     |   9 +-
 arch/sh/drivers/pci/Makefile            |   5 +-
 arch/sh/drivers/pci/fixups-dreamcast.c  |  38 +---
 arch/sh/drivers/pci/fixups-r7780rp.c    |  38 ++--
 arch/sh/drivers/pci/fixups-rts7751r2d.c |  24 +--
 arch/sh/drivers/pci/fixups-sh03.c       |  38 +---
 arch/sh/drivers/pci/ops-bigsur.c        |  18 +-
 arch/sh/drivers/pci/ops-landisk.c       |  68 +++++++
 arch/sh/drivers/pci/ops-r7780rp.c       |  18 +-
 arch/sh/drivers/pci/ops-rts7751r2d.c    |  13 +-
 arch/sh/drivers/pci/ops-sh4.c           | 164 ++++++++++++++++
 arch/sh/drivers/pci/ops-snapgear.c      |  21 +--
 arch/sh/drivers/pci/ops-titan.c         |  16 +-
 arch/sh/drivers/pci/pci-auto.c          |  10 -
 arch/sh/drivers/pci/pci-sh4.h           | 180 ++++++++++++++++++
 arch/sh/drivers/pci/pci-sh7751.c        | 322 ++++++--------------------------
 arch/sh/drivers/pci/pci-sh7751.h        | 174 +----------------
 arch/sh/drivers/pci/pci-sh7780.c        | 270 ++++----------------------
 arch/sh/drivers/pci/pci-sh7780.h        |  94 +---------
 arch/sh/drivers/pci/pci-st40.c          |  19 --
 arch/sh/drivers/pci/pci.c               |  40 +++-
 arch/sh/kernel/cpu/sh4/Makefile         |   3 +
 arch/sh/kernel/cpu/sh4/setup-sh7780.c   |  79 ++++++++
 include/asm-sh/io.h                     |   5 +
 include/asm-sh/pci.h                    |  35 +++-
 37 files changed, 984 insertions(+), 1520 deletions(-)
 create mode 100644 arch/sh/drivers/pci/ops-landisk.c
 create mode 100644 arch/sh/drivers/pci/ops-sh4.c
 create mode 100644 arch/sh/drivers/pci/pci-sh4.h
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7780.c

diff --git a/arch/sh/boards/landisk/Makefile b/arch/sh/boards/landisk/Makefile
index 60e75c6db0eb..89e4beb2ad47 100644
--- a/arch/sh/boards/landisk/Makefile
+++ b/arch/sh/boards/landisk/Makefile
@@ -3,5 +3,3 @@
 #
 
 obj-y	 := setup.o io.o irq.o rtc.o landisk_pwb.o
-
-obj-$(CONFIG_PCI) += pci.o
diff --git a/arch/sh/boards/landisk/io.c b/arch/sh/boards/landisk/io.c
index aa6b145c9e8f..92498b4947d5 100644
--- a/arch/sh/boards/landisk/io.c
+++ b/arch/sh/boards/landisk/io.c
@@ -14,39 +14,16 @@
  * modifed by kogiidena
  * 2005.03.03
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/pci.h>
 #include <asm/landisk/iodata_landisk.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
 extern void *area5_io_base;	/* Area 5 I/O Base address */
 extern void *area6_io_base;	/* Area 6 I/O Base address */
 
-/*
- * The 7751R LANDISK uses the built-in PCI controller (PCIC)
- * of the 7751R processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -67,17 +44,6 @@ static inline unsigned long port2adr(unsigned int port)
 	return port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH_7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -89,8 +55,8 @@ u8 landisk_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inb(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inb(pci_ioaddr(port));
 
 	return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -101,12 +67,12 @@ u8 landisk_inb_p(unsigned long port)
 
 	if (PXSEG(port))
 		v = ctrl_inb(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
 
-	delay();
+	ctrl_delay();
 
 	return v;
 }
@@ -115,8 +81,8 @@ u16 landisk_inw(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -127,8 +93,8 @@ u32 landisk_inl(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7751_PCIIO(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -139,8 +105,8 @@ void landisk_outb(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 }
@@ -149,19 +115,19 @@ void landisk_outb_p(u8 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
-	delay();
+	ctrl_delay();
 }
 
 void landisk_outw(u16 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -170,8 +136,8 @@ void landisk_outl(u32 value, unsigned long port)
 {
 	if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7751_PCIIO(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -184,8 +150,8 @@ void landisk_insb(unsigned long port, void *dst, unsigned long count)
         if (PXSEG(port)) {
                 while (count--)
                         *buf++ = *(volatile u8 *)port;
-	} else if (CHECK_SH7751_PCIIO(port)) {
-                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	} else if (is_pci_ioaddr(port)) {
+                volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
                 while (count--)
                         *buf++ = *bp;
@@ -203,8 +169,8 @@ void landisk_insw(unsigned long port, void *dst, unsigned long count)
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 	while (count--)
@@ -215,8 +181,8 @@ void landisk_insl(unsigned long port, void *dst, unsigned long count)
 {
         u32 *buf = dst;
 
-	if (CHECK_SH7751_PCIIO(port)) {
-                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port)) {
+                volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
                 while (count--)
                         *buf++ = *p;
@@ -232,8 +198,8 @@ void landisk_outsb(unsigned long port, const void *src, unsigned long count)
 	if (PXSEG(port))
                 while (count--)
                         ctrl_outb(*buf++, port);
-	else if (CHECK_SH7751_PCIIO(port)) {
-                volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port)) {
+                volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
                 while (count--)
                         *bp = *buf++;
@@ -251,8 +217,8 @@ void landisk_outsw(unsigned long port, const void *src, unsigned long count)
 
 	if (PXSEG(port))
                 p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                p = (volatile u16 *)pci_ioaddr(port);
 	else
                 p = (volatile u16 *)port2adr(port);
 
@@ -264,8 +230,8 @@ void landisk_outsl(unsigned long port, const void *src, unsigned long count)
 {
         const u32 *buf = src;
 
-	if (CHECK_SH7751_PCIIO(port)) {
-                volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port)) {
+                volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
                 while (count--)
                         *p = *buf++;
@@ -277,8 +243,8 @@ void __iomem *landisk_ioport_map(unsigned long port, unsigned int size)
 {
         if (PXSEG(port))
                 return (void __iomem *)port;
-        else if (CHECK_SH7751_PCIIO(port))
-                return (void __iomem *)PCI_IOMAP(port);
+        else if (is_pci_ioaddr(port))
+                return (void __iomem *)pci_ioaddr(port);
 
         return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c
index 2bfb221cc35c..a8c5180ae219 100644
--- a/arch/sh/boards/mpc1211/setup.c
+++ b/arch/sh/boards/mpc1211/setup.c
@@ -255,23 +255,12 @@ void __init init_mpc1211_IRQ(void)
 	}
 }
 
-/*
-  Initialize the board
-*/
-
-
-static void delay (void)
-{
-	volatile unsigned short tmp;
-	tmp = *(volatile unsigned short *) 0xa0000000;
-}
-
-static void delay1000 (void)
+static void delay1000(void)
 {
 	int i;
 
 	for (i=0; i<1000; i++)
-		delay ();
+		ctrl_delay();
 }
 
 static int put_smb_blk(unsigned char *p, int address, int command, int no)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index ecdce7ef6a34..8c26550ca2e4 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -13,14 +13,11 @@
 
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/module.h>
 #include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/addrspace.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
 extern void *area6_io8_base;	/* Area 6 8bit I/O Base address */
 extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
 
@@ -31,27 +28,17 @@ extern void *area5_io16_base;	/* Area 5 16bit I/O Base address */
  * like the other Solution Engine boards.
  */
 
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
 #define CODEC_IO_BASE	0x1000
 #define CODEC_IOMAP(a)	((unsigned long)area6_io8_base + ((a) - CODEC_IO_BASE))
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
 		if (port == 0x3f6)
 			return ((unsigned long)area5_io16_base + 0x0c);
 		else
-			return ((unsigned long)area5_io16_base + 0x800 + ((port-0x1f0) << 1));
+			return ((unsigned long)area5_io16_base + 0x800 +
+				((port-0x1f0) << 1));
 	else
 		maybebadio((unsigned long)port);
 	return port;
@@ -70,25 +57,10 @@ static inline int shifted_port(unsigned long port)
 }
 
 #if defined(CONFIG_HS7751RVOIP_CODEC)
-static inline int
-codec_port(unsigned long port)
-{
-	if (CODEC_IO_BASE <= port && port < (CODEC_IO_BASE+0x20))
-		return 1;
-	else
-		return 0;
-}
-#endif
-
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
+#define codec_port(port)	\
+	((CODEC_IO_BASE <= (port)) && ((port) < (CODEC_IO_BASE + 0x20)))
 #else
-#define CHECK_SH7751_PCIIO(port) (0)
+#define codec_port(port)	(0)
 #endif
 
 /*
@@ -102,12 +74,10 @@ unsigned char hs7751rvoip_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		return ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inb(pci_ioaddr(port));
 	else
 		return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -118,15 +88,13 @@ unsigned char hs7751rvoip_inb_p(unsigned long port)
 
         if (PXSEG(port))
 		v = ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		v = ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -134,8 +102,8 @@ unsigned short hs7751rvoip_inw(unsigned long port)
 {
         if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 	return 0;
@@ -145,8 +113,8 @@ unsigned int hs7751rvoip_inl(unsigned long port)
 {
         if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 	return 0;
@@ -157,12 +125,10 @@ void hs7751rvoip_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
 		ctrl_outb(value, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		ctrl_outb(value, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outb(value, port2adr(port));
 }
@@ -171,24 +137,22 @@ void hs7751rvoip_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outb(value, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		ctrl_outb(value, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 
-	delay();
+	ctrl_delay();
 }
 
 void hs7751rvoip_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -197,8 +161,8 @@ void hs7751rvoip_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -210,13 +174,11 @@ void hs7751rvoip_insb(unsigned long port, void *addr, unsigned long count)
 	if (PXSEG(port))
 		while (count--)
 			*buf++ = ctrl_inb(port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		while (count--)
 			*buf++ = ctrl_inb(CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *bp;
@@ -235,8 +197,8 @@ void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 	while (count--)
@@ -246,8 +208,8 @@ void hs7751rvoip_insw(unsigned long port, void *addr, unsigned long count)
 void hs7751rvoip_insl(unsigned long port, void *addr, unsigned long count)
 {
 
-	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 		u32 *buf = addr;
 
 		while (count--)
@@ -263,13 +225,11 @@ void hs7751rvoip_outsb(unsigned long port, const void *addr, unsigned long count
 	if (PXSEG(port))
 		while (count--)
 			ctrl_outb(*buf++, port);
-#if defined(CONFIG_HS7751RVOIP_CODEC)
 	else if (codec_port(port))
 		while (count--)
 			ctrl_outb(*buf++, CODEC_IOMAP(port));
-#endif
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*bp = *buf++;
@@ -288,8 +248,8 @@ void hs7751rvoip_outsw(unsigned long port, const void *addr, unsigned long count
 
 	if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -301,8 +261,8 @@ void hs7751rvoip_outsl(unsigned long port, const void *addr, unsigned long count
 {
 	const u32 *buf = addr;
 
-	if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*p = *buf++;
@@ -316,8 +276,8 @@ void __iomem *hs7751rvoip_ioport_map(unsigned long port, unsigned int size)
                 return (void __iomem *)port;
 	else if (unlikely(codec_port(port) && (size == 1)))
 		return (void __iomem *)CODEC_IOMAP(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return (void __iomem *)PCI_IOMAP(port);
+        else if (is_pci_ioaddr(port))
+                return (void __iomem *)pci_ioaddr(port);
 
         return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/r7780rp/io.c b/arch/sh/boards/renesas/r7780rp/io.c
index f73ca3f0f5a1..db92d6e6ae99 100644
--- a/arch/sh/boards/renesas/r7780rp/io.c
+++ b/arch/sh/boards/renesas/r7780rp/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_r7780rp.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,37 +8,13 @@
  * placeholder code from io_r7780rp.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
+#include <linux/pci.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <asm/r7780rp/r7780rp.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
 
-#include <linux/module.h>
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7780.h"
-
-/*
- * The 7780 R7780RP-1 uses the built-in PCI controller (PCIC)
- * of the 7780 processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */
-
-#define	SH7780_PCIIOBR_MASK	0xFFFC0000	/* IO Space Mask */
-#define PCIIOBR		(volatile long *)PCI_REG(SH7780_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7780_PCIMBR)
-#define PCI_IO_AREA	SH7780_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7780_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7780_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -78,17 +52,6 @@ static inline int shifted_port(unsigned long port)
 		return 1;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7780_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7780_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7780_PCIIO(port) (0)
-#endif
-
 #if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
 #define CHECK_AX88796L_PORT(port) \
   ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
@@ -109,8 +72,8 @@ u8 r7780rp_inb(unsigned long port)
 		return ctrl_inw(port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		return ctrl_inb(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inb(pci_ioaddr(port));
 
 	return ctrl_inw(port2adr(port)) & 0xff;
 }
@@ -123,12 +86,12 @@ u8 r7780rp_inb_p(unsigned long port)
 		v = ctrl_inw(port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		v = ctrl_inb(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		v = ctrl_inb(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = ctrl_inb(pci_ioaddr(port));
 	else
 		v = ctrl_inw(port2adr(port)) & 0xff;
 
-	delay();
+	ctrl_delay();
 
 	return v;
 }
@@ -139,8 +102,8 @@ u16 r7780rp_inw(unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		return ctrl_inw(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inw(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inw(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -153,8 +116,8 @@ u32 r7780rp_inl(unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		return ctrl_inl(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return ctrl_inl(PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return ctrl_inl(pci_ioaddr(port));
 	else
 		maybebadio(port);
 
@@ -167,8 +130,8 @@ void r7780rp_outb(u8 value, unsigned long port)
 		ctrl_outw(value, port88796l(port, 0));
 	else if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 }
@@ -179,12 +142,12 @@ void r7780rp_outb_p(u8 value, unsigned long port)
 		ctrl_outw(value, port88796l(port, 0));
 	else if (PXSEG(port))
 		ctrl_outb(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outb(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outb(value, pci_ioaddr(port));
 	else
 		ctrl_outw(value, port2adr(port));
 
-	delay();
+	ctrl_delay();
 }
 
 void r7780rp_outw(u16 value, unsigned long port)
@@ -193,8 +156,8 @@ void r7780rp_outw(u16 value, unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		ctrl_outw(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outw(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outw(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -205,8 +168,8 @@ void r7780rp_outl(u32 value, unsigned long port)
 		maybebadio(port);
 	else if (PXSEG(port))
 		ctrl_outl(value, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		ctrl_outl(value, PCI_IOMAP(port));
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		ctrl_outl(value, pci_ioaddr(port));
 	else
 		maybebadio(port);
 }
@@ -223,8 +186,8 @@ void r7780rp_insb(unsigned long port, void *dst, unsigned long count)
 	} else if (PXSEG(port)) {
 		while (count--)
 			*buf++ = *(volatile u8 *)port;
-	} else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	} else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *bp;
@@ -244,8 +207,8 @@ void r7780rp_insw(unsigned long port, void *dst, unsigned long count)
 		p = (volatile u16 *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -259,8 +222,8 @@ void r7780rp_insl(unsigned long port, void *dst, unsigned long count)
 
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*buf++ = *p;
@@ -280,8 +243,8 @@ void r7780rp_outsb(unsigned long port, const void *src, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(*buf++, port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u8 *bp = (volatile u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u8 *bp = (volatile u8 *)pci_ioaddr(port);
 
 		while (count--)
 			*bp = *buf++;
@@ -301,8 +264,8 @@ void r7780rp_outsw(unsigned long port, const void *src, unsigned long count)
 		p = (volatile u16 *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile u16 *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		p = (volatile u16 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile u16 *)pci_ioaddr(port);
 	else
 		p = (volatile u16 *)port2adr(port);
 
@@ -316,8 +279,8 @@ void r7780rp_outsl(unsigned long port, const void *src, unsigned long count)
 
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port)) {
-		volatile u32 *p = (volatile u32 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		volatile u32 *p = (volatile u32 *)pci_ioaddr(port);
 
 		while (count--)
 			*p = *buf++;
@@ -331,8 +294,8 @@ void __iomem *r7780rp_ioport_map(unsigned long port, unsigned int size)
 		return (void __iomem *)port88796l(port, size > 1);
 	else if (PXSEG(port))
 		return (void __iomem *)port;
-	else if (CHECK_SH7780_PCIIO(port) || shifted_port(port))
-		return (void __iomem *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return (void __iomem *)pci_ioaddr(port);
 
 	return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/boards/renesas/rts7751r2d/io.c b/arch/sh/boards/renesas/rts7751r2d/io.c
index 8dc2a2e2e5df..135aa0b5e62d 100644
--- a/arch/sh/boards/renesas/rts7751r2d/io.c
+++ b/arch/sh/boards/renesas/rts7751r2d/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_rts7751r2d.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,11 +8,9 @@
  * placeholder code from io_rts7751r2d.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
 #include <asm/rts7751r2d/rts7751r2d.h>
 #include <asm/io.h>
 #include <asm/addrspace.h>
@@ -26,18 +22,6 @@
  * like the other Solution Engine boards.
  */
 
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline unsigned long port2adr(unsigned int port)
 {
 	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
@@ -75,17 +59,6 @@ static inline int shifted_port(unsigned long port)
 		return 1;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 #if defined(CONFIG_NE2000) || defined(CONFIG_NE2000_MODULE)
 #define CHECK_AX88796L_PORT(port) \
   ((port >= AX88796L_IO_BASE) && (port < (AX88796L_IO_BASE+0x20)))
@@ -106,8 +79,8 @@ unsigned char rts7751r2d_inb(unsigned long port)
 		return (*(volatile unsigned short *)port88796l(port, 0)) & 0xff;
 	else if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
 		return (*(volatile unsigned short *)port2adr(port) & 0xff);
 }
@@ -120,11 +93,12 @@ unsigned char rts7751r2d_inb_p(unsigned long port)
 		v = (*(volatile unsigned short *)port88796l(port, 0)) & 0xff;
         else if (PXSEG(port))
 		v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
 		v = (*(volatile unsigned short *)port2adr(port) & 0xff);
-	delay();
+
+	ctrl_delay();
 
 	return v;
 }
@@ -135,8 +109,8 @@ unsigned short rts7751r2d_inw(unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned short *)pci_ioaddr(port);
 	else
 		maybebadio(port);
 
@@ -149,8 +123,8 @@ unsigned int rts7751r2d_inl(unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		return *(volatile unsigned long *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		return *(volatile unsigned long *)pci_ioaddr(port);
 	else
 		maybebadio(port);
 
@@ -163,8 +137,8 @@ void rts7751r2d_outb(unsigned char value, unsigned long port)
 		*((volatile unsigned short *)port88796l(port, 0)) = value;
         else if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned char *)pci_ioaddr(port) = value;
 	else
 		*(volatile unsigned short *)port2adr(port) = value;
 }
@@ -175,11 +149,12 @@ void rts7751r2d_outb_p(unsigned char value, unsigned long port)
 		*((volatile unsigned short *)port88796l(port, 0)) = value;
         else if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned char *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned char *)pci_ioaddr(port) = value;
 	else
 		*(volatile unsigned short *)port2adr(port) = value;
-	delay();
+
+	ctrl_delay();
 }
 
 void rts7751r2d_outw(unsigned short value, unsigned long port)
@@ -188,8 +163,8 @@ void rts7751r2d_outw(unsigned short value, unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned short *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned short *)pci_ioaddr(port) = value;
 	else
 		maybebadio(port);
 }
@@ -200,8 +175,8 @@ void rts7751r2d_outl(unsigned int value, unsigned long port)
 		maybebadio(port);
         else if (PXSEG(port))
 		*(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		*(volatile unsigned long *)PCI_IOMAP(port) = value;
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		*(volatile unsigned long *)pci_ioaddr(port) = value;
 	else
 		maybebadio(port);
 }
@@ -219,8 +194,8 @@ void rts7751r2d_insb(unsigned long port, void *addr, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(ctrl_inb(port), a++);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		bp = (__u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		bp = (__u8 *)pci_ioaddr(port);
 		while (count--)
 			ctrl_outb(*bp, a++);
 	} else {
@@ -239,8 +214,8 @@ void rts7751r2d_insw(unsigned long port, void *addr, unsigned long count)
 		p = (volatile unsigned short *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile unsigned short *)pci_ioaddr(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
 	while (count--)
@@ -251,11 +226,11 @@ void rts7751r2d_insl(unsigned long port, void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
 		unsigned long a = (unsigned long)addr;
 
 		while (count--) {
-			ctrl_outl(ctrl_inl(PCI_IOMAP(port)), a);
+			ctrl_outl(ctrl_inl(pci_ioaddr(port)), a);
 			a += 4;
 		}
 	} else
@@ -275,8 +250,8 @@ void rts7751r2d_outsb(unsigned long port, const void *addr, unsigned long count)
 	} else if (PXSEG(port))
 		while (count--)
 			ctrl_outb(a++, port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
-		bp = (__u8 *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
+		bp = (__u8 *)pci_ioaddr(port);
 		while (count--)
 			*bp = ctrl_inb(a++);
 	} else {
@@ -295,8 +270,8 @@ void rts7751r2d_outsw(unsigned long port, const void *addr, unsigned long count)
 		p = (volatile unsigned short *)port88796l(port, 1);
 	else if (PXSEG(port))
 		p = (volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port))
-		p = (volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port) || shifted_port(port))
+		p = (volatile unsigned short *)pci_ioaddr(port);
 	else
 		p = (volatile unsigned short *)port2adr(port);
 
@@ -310,11 +285,11 @@ void rts7751r2d_outsl(unsigned long port, const void *addr, unsigned long count)
 {
 	if (CHECK_AX88796L_PORT(port))
 		maybebadio(port);
-	else if (CHECK_SH7751_PCIIO(port) || shifted_port(port)) {
+	else if (is_pci_ioaddr(port) || shifted_port(port)) {
 		unsigned long a = (unsigned long)addr;
 
 		while (count--) {
-			ctrl_outl(ctrl_inl(a), PCI_IOMAP(port));
+			ctrl_outl(ctrl_inl(a), pci_ioaddr(port));
 			a += 4;
 		}
 	} else
diff --git a/arch/sh/boards/renesas/systemh/io.c b/arch/sh/boards/renesas/systemh/io.c
index 0befd4f9894c..cde6e5d192c4 100644
--- a/arch/sh/boards/renesas/systemh/io.c
+++ b/arch/sh/boards/renesas/systemh/io.c
@@ -5,36 +5,16 @@
  * Based largely on io_se.c.
  *
  * I/O routine for Hitachi 7751 Systemh.
- *
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <asm/systemh7751.h>
 #include <asm/addrspace.h>
 #include <asm/io.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
-/*
- * The 7751 SystemH Engine uses the built-in PCI controller (PCIC)
- * of the 7751 processor, and has a SuperIO accessible on its memory
- * bus.
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
 
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
 #define ETHER_IOMAP(adr) (0xB3000000 + (adr)) /*map to 16bits access area
                                                 of smc lan chip*/
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline volatile __u16 *
 port2adr(unsigned int port)
 {
@@ -44,17 +24,6 @@ port2adr(unsigned int port)
 	return (volatile __u16*)port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -66,8 +35,8 @@ unsigned char sh7751systemh_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else if (port <= 0x3F1)
 		return *(volatile unsigned char *)ETHER_IOMAP(port);
 	else
@@ -80,13 +49,13 @@ unsigned char sh7751systemh_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                v = *(volatile unsigned char *)pci_ioaddr(port);
 	else if (port <= 0x3F1)
 		v = *(volatile unsigned char *)ETHER_IOMAP(port);
 	else
 		v = (*port2adr(port))&0xff;
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -94,8 +63,8 @@ unsigned short sh7751systemh_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else if (port <= 0x3F1)
@@ -109,8 +78,8 @@ unsigned int sh7751systemh_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else if (port <= 0x3F1)
@@ -125,8 +94,8 @@ void sh7751systemh_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else if (port <= 0x3F1)
 		*(volatile unsigned char *)ETHER_IOMAP(port) = value;
 	else
@@ -137,21 +106,21 @@ void sh7751systemh_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else if (port <= 0x3F1)
 		*(volatile unsigned char *)ETHER_IOMAP(port) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void sh7751systemh_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else if (port <= 0x3F1)
@@ -164,8 +133,8 @@ void sh7751systemh_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
diff --git a/arch/sh/boards/se/770x/io.c b/arch/sh/boards/se/770x/io.c
index 5102201c97ab..9941949331ab 100644
--- a/arch/sh/boards/se/770x/io.c
+++ b/arch/sh/boards/se/770x/io.c
@@ -1,4 +1,4 @@
-/* $Id: io.c,v 1.6 2006/01/04 17:53:54 lethal Exp $
+/* $Id: io.c,v 1.7 2006/02/05 21:55:29 lethal Exp $
  *
  * linux/arch/sh/kernel/io_se.c
  *
@@ -20,11 +20,6 @@ int sh_pcic_io_stop;
 int sh_pcic_io_type;
 int sh_pcic_io_dummy;
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 /* MS7750 requires special versions of in*, out* routines, since
    PC-like io ports are located at upper half byte of 16-bit word which
    can be accessed only with 16-bit wide.  */
@@ -72,7 +67,7 @@ unsigned char se_inb_p(unsigned long port)
 		v = (*port2adr(port) >> 8); 
 	else
 		v = (*port2adr(port))&0xff; 
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -110,7 +105,7 @@ void se_outb_p(unsigned char value, unsigned long port)
 		*(port2adr(port)) = value << 8;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void se_outw(unsigned short value, unsigned long port)
diff --git a/arch/sh/boards/se/7751/io.c b/arch/sh/boards/se/7751/io.c
index 0e8a3ba48316..e8d846cec89d 100644
--- a/arch/sh/boards/se/7751/io.c
+++ b/arch/sh/boards/se/7751/io.c
@@ -1,6 +1,4 @@
 /*
- * linux/arch/sh/kernel/io_7751se.c
- *
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
  *
@@ -10,55 +8,14 @@
  * placeholder code from io_se.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/se7751.h>
 #include <asm/addrspace.h>
 
-#include <linux/pci.h>
-#include "../../../drivers/pci/pci-sh7751.h"
-
-#if 0
-/******************************************************************
- * Variables from io_se.c, related to PCMCIA (not PCI); we're not
- * compiling them in, and have removed references from functions
- * which follow.  [Many checked for IO ports in the range bounded
- * by sh_pcic_io_start/stop, and used sh_pcic_io_wbase as offset.
- * As start/stop are uninitialized, only port 0x0 would match?]
- * When used, remember to adjust names to avoid clash with io_se?
- *****************************************************************/
-/* SH pcmcia io window base, start and end.  */
-int sh_pcic_io_wbase = 0xb8400000;
-int sh_pcic_io_start;
-int sh_pcic_io_stop;
-int sh_pcic_io_type;
-int sh_pcic_io_dummy;
-/*************************************************************/
-#endif
-
-/*
- * The 7751 Solution Engine uses the built-in PCI controller (PCIC)
- * of the 7751 processor, and has a SuperIO accessible via the PCI.
- * The board also includes a PCMCIA controller on its memory bus,
- * like the other Solution Engine boards.
- */ 
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
-static inline volatile __u16 *
-port2adr(unsigned int port)
+static inline volatile u16 *port2adr(unsigned int port)
 {
 	if (port >= 0x2000)
 		return (volatile __u16 *) (PA_MRSHPC + (port - 0x2000));
@@ -66,32 +23,6 @@ port2adr(unsigned int port)
 	return (volatile __u16*)port;
 }
 
-#if 0
-/* The 7751 Solution Engine seems to have everything hooked */
-/* up pretty normally (nothing on high-bytes only...) so this */
-/* shouldn't be needed */
-static inline int
-shifted_port(unsigned long port)
-{
-	/* For IDE registers, value is not shifted */
-	if ((0x1f0 <= port && port < 0x1f8) || port == 0x3f6)
-		return 0;
-	else
-		return 1;
-}
-#endif
-
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -103,10 +34,10 @@ unsigned char sh7751se_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		return (*port2adr(port))&0xff; 
+		return (*port2adr(port)) & 0xff;
 }
 
 unsigned char sh7751se_inb_p(unsigned long port)
@@ -115,11 +46,11 @@ unsigned char sh7751se_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		v = (*port2adr(port))&0xff; 
-	delay();
+		v = (*port2adr(port)) & 0xff;
+	ctrl_delay();
 	return v;
 }
 
@@ -127,8 +58,8 @@ unsigned short sh7751se_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -140,8 +71,8 @@ unsigned int sh7751se_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-                return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+                return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -154,8 +85,8 @@ void sh7751se_outb(unsigned char value, unsigned long port)
 
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
 }
@@ -164,19 +95,19 @@ void sh7751se_outb_p(unsigned char value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
 void sh7751se_outw(unsigned short value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
@@ -187,8 +118,8 @@ void sh7751se_outl(unsigned int value, unsigned long port)
 {
         if (PXSEG(port))
                 *(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-        	*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
@@ -202,35 +133,3 @@ void sh7751se_outsl(unsigned long port, const void *addr, unsigned long count)
 {
 	maybebadio(port);
 }
-
-/* Map ISA bus address to the real address. Only for PCMCIA.  */
-
-/* ISA page descriptor.  */
-static __u32 sh_isa_memmap[256];
-
-#if 0
-static int
-sh_isa_mmap(__u32 start, __u32 length, __u32 offset)
-{
-	int idx;
-
-	if (start >= 0x100000 || (start & 0xfff) || (length != 0x1000))
-		return -1;
-
-	idx = start >> 12;
-	sh_isa_memmap[idx] = 0xb8000000 + (offset &~ 0xfff);
-	printk("sh_isa_mmap: start %x len %x offset %x (idx %x paddr %x)\n",
-	       start, length, offset, idx, sh_isa_memmap[idx]);
-	return 0;
-}
-#endif
-
-unsigned long
-sh7751se_isa_port2addr(unsigned long offset)
-{
-	int idx;
-
-	idx = (offset >> 12) & 0xff;
-	offset &= 0xfff;
-	return sh_isa_memmap[idx] + offset;
-}
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index dab742a00c6f..7d31d6aed8a5 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -12,11 +12,10 @@
 #include <asm/sh03/io.h>
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
-#include "../../drivers/pci/pci-sh7751.h"
 
 const char *get_system_type(void)
 {
-	return "Interface CTP/PCI-SH03)";
+	return "Interface (CTP/PCI-SH03)";
 }
 
 static void init_sh03_IRQ(void)
@@ -39,7 +38,7 @@ static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
 	if (((port >= 0x1f0) && (port <= 0x1f7)) || (port == 0x3f6))
 		return (void __iomem *)((unsigned long)cf_io_base + port);
 
-        return (void __iomem *)(port + SH7751_PCI_IO_BASE);
+        return (void __iomem *)(port + PCI_IO_BASE);
 }
 
 struct sh_machine_vector mv_sh03 __initmv = {
@@ -51,7 +50,6 @@ struct sh_machine_vector mv_sh03 __initmv = {
 	.mv_heartbeat		= heartbeat_sh03,
 #endif
 };
-
 ALIAS_MV(sh03)
 
 /* arch/sh/boards/sh03/rtc.c */
diff --git a/arch/sh/boards/snapgear/io.c b/arch/sh/boards/snapgear/io.c
index 9f700b8392bb..0f4824264557 100644
--- a/arch/sh/boards/snapgear/io.c
+++ b/arch/sh/boards/snapgear/io.c
@@ -1,6 +1,4 @@
-/* 
- * linux/arch/sh/kernel/io_7751se.c
- *
+/*
  * Copyright (C) 2002  David McCullough <davidm@snapgear.com>
  * Copyright (C) 2001  Ian da Silva, Jeremy Siegel
  * Based largely on io_se.c.
@@ -11,54 +9,22 @@
  * placeholder code from io_se.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/addrspace.h>
 
-#include <asm/pci.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
 #ifdef CONFIG_SH_SECUREEDGE5410
 unsigned short secureedge5410_ioport;
 #endif
 
-/*
- * The SnapGear uses the built-in PCI controller (PCIC)
- * of the 7751 processor
- */
-
-#define PCIIOBR		(volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA	SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA	SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)	(PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 static inline volatile __u16 *port2adr(unsigned int port)
 {
 	maybebadio((unsigned long)port);
 	return (volatile __u16*)port;
 }
 
-/* In case someone configures the kernel w/o PCI support: in that */
-/* scenario, don't ever bother to check for PCI-window addresses */
-
-/* NOTE: WINDOW CHECK MAY BE A BIT OFF, HIGH PCIBIOS_MIN_IO WRAPS? */
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
 /*
  * General outline: remap really low stuff [eventually] to SuperIO,
  * stuff in PCI IO space (at or above window at pci.h:PCIBIOS_MIN_IO)
@@ -66,39 +32,36 @@ static inline volatile __u16 *port2adr(unsigned int port)
  * should be way beyond the window, and is used  w/o translation for
  * compatibility.
  */
-
 unsigned char snapgear_inb(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		return (*port2adr(port))&0xff; 
+		return (*port2adr(port)) & 0xff;
 }
 
-
 unsigned char snapgear_inb_p(unsigned long port)
 {
 	unsigned char v;
 
 	if (PXSEG(port))
 		v = *(volatile unsigned char *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		v = *(volatile unsigned char *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		v = *(volatile unsigned char *)pci_ioaddr(port);
 	else
-		v = (*port2adr(port))&0xff; 
-	delay();
+		v = (*port2adr(port))&0xff;
+	ctrl_delay();
 	return v;
 }
 
-
 unsigned short snapgear_inw(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned short *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned short *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned short *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -106,13 +69,12 @@ unsigned short snapgear_inw(unsigned long port)
 	return 0;
 }
 
-
 unsigned int snapgear_inl(unsigned long port)
 {
 	if (PXSEG(port))
 		return *(volatile unsigned long *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return *(volatile unsigned int *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return *(volatile unsigned int *)pci_ioaddr(port);
 	else if (port >= 0x2000)
 		return *port2adr(port);
 	else
@@ -120,50 +82,46 @@ unsigned int snapgear_inl(unsigned long port)
 	return 0;
 }
 
-
 void snapgear_outb(unsigned char value, unsigned long port)
 {
 
 	if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
 }
 
-
 void snapgear_outb_p(unsigned char value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned char *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned char*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned char*)pci_ioaddr(port)) = value;
 	else
 		*(port2adr(port)) = value;
-	delay();
+	ctrl_delay();
 }
 
-
 void snapgear_outw(unsigned short value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned short *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned short *)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned short *)pci_ioaddr(port)) = value;
 	else if (port >= 0x2000)
 		*port2adr(port) = value;
 	else
 		maybebadio(port);
 }
 
-
 void snapgear_outl(unsigned int value, unsigned long port)
 {
 	if (PXSEG(port))
 		*(volatile unsigned long *)port = value;
-	else if (CHECK_SH7751_PCIIO(port))
-		*((unsigned long*)PCI_IOMAP(port)) = value;
+	else if (is_pci_ioaddr(port))
+		*((unsigned long*)pci_ioaddr(port)) = value;
 	else
 		maybebadio(port);
 }
diff --git a/arch/sh/boards/titan/io.c b/arch/sh/boards/titan/io.c
index 48f3494f55b1..4730c1dd697d 100644
--- a/arch/sh/boards/titan/io.c
+++ b/arch/sh/boards/titan/io.c
@@ -1,34 +1,11 @@
 /*
  *	I/O routines for Titan
  */
-
 #include <linux/pci.h>
 #include <asm/machvec.h>
 #include <asm/addrspace.h>
 #include <asm/titan.h>
 #include <asm/io.h>
-#include "../../drivers/pci/pci-sh7751.h"
-
-#define PCIIOBR         (volatile long *)PCI_REG(SH7751_PCIIOBR)
-#define PCIMBR          (volatile long *)PCI_REG(SH7751_PCIMBR)
-#define PCI_IO_AREA     SH7751_PCI_IO_BASE
-#define PCI_MEM_AREA    SH7751_PCI_CONFIG_BASE
-
-#define PCI_IOMAP(adr)  (PCI_IO_AREA + (adr & ~SH7751_PCIIOBR_MASK))
-
-#if defined(CONFIG_PCI)
-#define CHECK_SH7751_PCIIO(port) \
-  ((port >= PCIBIOS_MIN_IO) && (port < (PCIBIOS_MIN_IO + SH7751_PCI_IO_SIZE)))
-#define CHECK_SH7751_PCIMEMIO(port) \
-  ((port >= PCIBIOS_MIN_MEM) && (port < (PCIBIOS_MIN_MEM + SH7751_PCI_MEM_SIZE)))
-#else
-#define CHECK_SH7751_PCIIO(port) (0)
-#endif
-
-static inline void delay(void)
-{
-        ctrl_inw(0xa0000000);
-}
 
 static inline unsigned int port2adr(unsigned int port)
 {
@@ -40,8 +17,8 @@ u8 titan_inb(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inb(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inb(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inb(pci_ioaddr(port));
         return ctrl_inw(port2adr(port)) & 0xff;
 }
 
@@ -51,11 +28,11 @@ u8 titan_inb_p(unsigned long port)
 
         if (PXSEG(port))
                 v = ctrl_inb(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                v = ctrl_inb(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                v = ctrl_inb(pci_ioaddr(port));
         else
                 v = ctrl_inw(port2adr(port)) & 0xff;
-        delay();
+        ctrl_delay();
         return v;
 }
 
@@ -63,8 +40,8 @@ u16 titan_inw(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inw(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inw(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inw(pci_ioaddr(port));
         else if (port >= 0x2000)
                 return ctrl_inw(port2adr(port));
         else
@@ -76,8 +53,8 @@ u32 titan_inl(unsigned long port)
 {
         if (PXSEG(port))
                 return ctrl_inl(port);
-        else if (CHECK_SH7751_PCIIO(port))
-                return ctrl_inl(PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                return ctrl_inl(pci_ioaddr(port));
         else if (port >= 0x2000)
                 return ctrl_inw(port2adr(port));
         else
@@ -89,8 +66,8 @@ void titan_outb(u8 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outb(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outb(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outb(value, pci_ioaddr(port));
         else
                 ctrl_outw(value, port2adr(port));
 }
@@ -99,19 +76,19 @@ void titan_outb_p(u8 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outb(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outb(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outb(value, pci_ioaddr(port));
         else
                 ctrl_outw(value, port2adr(port));
-        delay();
+        ctrl_delay();
 }
 
 void titan_outw(u16 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outw(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outw(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outw(value, pci_ioaddr(port));
         else if (port >= 0x2000)
                 ctrl_outw(value, port2adr(port));
         else
@@ -122,8 +99,8 @@ void titan_outl(u32 value, unsigned long port)
 {
         if (PXSEG(port))
                 ctrl_outl(value, port);
-        else if (CHECK_SH7751_PCIIO(port))
-                ctrl_outl(value, PCI_IOMAP(port));
+        else if (is_pci_ioaddr(port))
+                ctrl_outl(value, pci_ioaddr(port));
         else
                 maybebadio(port);
 }
@@ -140,10 +117,10 @@ void titan_outsl(unsigned long port, const void *src, unsigned long count)
 
 void __iomem *titan_ioport_map(unsigned long port, unsigned int size)
 {
-	if (PXSEG(port) || CHECK_SH7751_PCIMEMIO(port))
+	if (PXSEG(port) || is_pci_memaddr(port))
 		return (void __iomem *)port;
-	else if (CHECK_SH7751_PCIIO(port))
-		return (void __iomem *)PCI_IOMAP(port);
+	else if (is_pci_ioaddr(port))
+		return (void __iomem *)pci_ioaddr(port);
 
 	return (void __iomem *)port2adr(port);
 }
diff --git a/arch/sh/cchips/hd6446x/hd64461/io.c b/arch/sh/cchips/hd6446x/hd64461/io.c
index f77f18f2ba7f..7909a1b7b512 100644
--- a/arch/sh/cchips/hd6446x/hd64461/io.c
+++ b/arch/sh/cchips/hd6446x/hd64461/io.c
@@ -53,11 +53,6 @@ static __inline__ unsigned long PORT2ADDR(unsigned long port)
 	return 0xa0000000 + (port & 0x1fffffff);
 }
 
-static inline void delay(void)
-{
-	ctrl_inw(0xa0000000);
-}
-
 unsigned char hd64461_inb(unsigned long port)
 {
 	return *(volatile unsigned char*)PORT2ADDR(port);
@@ -66,7 +61,7 @@ unsigned char hd64461_inb(unsigned long port)
 unsigned char hd64461_inb_p(unsigned long port)
 {
 	unsigned long v = *(volatile unsigned char*)PORT2ADDR(port);
-	delay();
+	ctrl_delay();
 	return v;
 }
 
@@ -88,7 +83,7 @@ void hd64461_outb(unsigned char b, unsigned long port)
 void hd64461_outb_p(unsigned char b, unsigned long port)
 {
 	*(volatile unsigned char*)PORT2ADDR(port) = b;
-	delay();
+	ctrl_delay();
 }
 
 void hd64461_outw(unsigned short b, unsigned long port)
diff --git a/arch/sh/drivers/pci/Makefile b/arch/sh/drivers/pci/Makefile
index 3d8078f1c05f..9e00cb8a39e9 100644
--- a/arch/sh/drivers/pci/Makefile
+++ b/arch/sh/drivers/pci/Makefile
@@ -6,8 +6,8 @@ obj-y					+= pci.o
 obj-$(CONFIG_PCI_AUTO)			+= pci-auto.o
 
 obj-$(CONFIG_CPU_SUBTYPE_ST40STB1)	+= pci-st40.o
-obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o 
-obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= pci-sh7751.o ops-sh4.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= pci-sh7780.o ops-sh4.o
 
 obj-$(CONFIG_SH_DREAMCAST)		+= ops-dreamcast.o fixups-dreamcast.o \
 					   dma-dreamcast.o
@@ -17,3 +17,4 @@ obj-$(CONFIG_SH_RTS7751R2D)		+= ops-rts7751r2d.o fixups-rts7751r2d.o
 obj-$(CONFIG_SH_SH03)			+= ops-sh03.o fixups-sh03.o
 obj-$(CONFIG_SH_R7780RP)		+= ops-r7780rp.o fixups-r7780rp.o
 obj-$(CONFIG_SH_TITAN)			+= ops-titan.o
+obj-$(CONFIG_SH_LANDISK)		+= ops-landisk.o
diff --git a/arch/sh/drivers/pci/fixups-dreamcast.c b/arch/sh/drivers/pci/fixups-dreamcast.c
index 63b1c6f4b8d2..c0af5f7ef414 100644
--- a/arch/sh/drivers/pci/fixups-dreamcast.c
+++ b/arch/sh/drivers/pci/fixups-dreamcast.c
@@ -4,7 +4,7 @@
  * PCI fixups for the Sega Dreamcast
  *
  * Copyright (C) 2001, 2002  M. R. Brown
- * Copyright (C) 2002, 2003  Paul Mundt
+ * Copyright (C) 2002, 2003, 2006  Paul Mundt
  *
  * This file originally bore the message (with enclosed-$):
  *	Id: pci.c,v 1.3 2003/05/04 19:29:46 lethal Exp
@@ -45,36 +45,16 @@ static void __init gapspci_fixup_resources(struct pci_dev *dev)
 		printk("PCI: Failed resource fixup\n");
 	}
 }
-
 DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, gapspci_fixup_resources);
 
-void __init pcibios_fixup_bus(struct pci_bus *bus)
+int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
 {
-	/* 
-	 * We don't have any sub bus to fix up, and this is a rather
-	 * stupid place to put general device fixups. Don't do it.
-	 * Use the pcibios_fixups table or suffer the consequences.
+	/*
+	 * The interrupt routing semantics here are quite trivial.
+	 *
+	 * We basically only support one interrupt, so we only bother
+	 * updating a device's interrupt line with this single shared
+	 * interrupt. Keeps routing quite simple, doesn't it?
 	 */
+	return GAPSPCI_IRQ;
 }
-
-void __init pcibios_fixup_irqs(void)
-{
-	struct pci_dev *dev = 0;
-
-	for_each_pci_dev(dev) {
-		/*
-		 * The interrupt routing semantics here are quite trivial.
-		 *
-		 * We basically only support one interrupt, so we only bother
-		 * updating a device's interrupt line with this single shared
-		 * interrupt. Keeps routing quite simple, doesn't it?
-		 */
-		printk(KERN_NOTICE "PCI: Fixing up IRQ routing for device %s\n",
-		       pci_name(dev));
-
-		dev->irq = GAPSPCI_IRQ;
-
-		pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq);
-	}
-}
-
diff --git a/arch/sh/drivers/pci/fixups-r7780rp.c b/arch/sh/drivers/pci/fixups-r7780rp.c
index b656b562ec99..3e321df65d22 100644
--- a/arch/sh/drivers/pci/fixups-r7780rp.c
+++ b/arch/sh/drivers/pci/fixups-r7780rp.c
@@ -4,36 +4,42 @@
  * Highlander R7780RP-1 PCI fixups
  *
  * Copyright (C) 2003  Lineo uSolutions, Inc.
- * Copyright (C) 2004  Paul Mundt
+ * Copyright (C) 2004 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-#include "pci-sh7780.h"
+#include <linux/pci.h>
+#include "pci-sh4.h"
 #include <asm/io.h>
 
 int pci_fixup_pcic(void)
 {
-	outl(0x000043ff, PCI_REG(SH7780_PCIIMR));
-	outl(0x0000380f, PCI_REG(SH7780_PCIAINTM));
+	pci_write_reg(0x000043ff, SH4_PCIINTM);
+	pci_write_reg(0x0000380f, SH4_PCIAINTM);
 
-	outl(0xfbb00047, PCI_REG(SH7780_PCICMD));
-	outl(0x00000000, PCI_REG(SH7780_PCIIBAR));
+	pci_write_reg(0xfbb00047, SH7780_PCICMD);
+	pci_write_reg(0x00000000, SH7780_PCIIBAR);
 
-	outl(0x00011912, PCI_REG(SH7780_PCISVID));
-	outl(0x08000000, PCI_REG(SH7780_PCICSCR0));
-	outl(0x0000001b, PCI_REG(SH7780_PCICSAR0));
-	outl(0xfd000000, PCI_REG(SH7780_PCICSCR1));
-	outl(0x0000000f, PCI_REG(SH7780_PCICSAR1));
+	pci_write_reg(0x00011912, SH7780_PCISVID);
+	pci_write_reg(0x08000000, SH7780_PCICSCR0);
+	pci_write_reg(0x0000001b, SH7780_PCICSAR0);
+	pci_write_reg(0xfd000000, SH7780_PCICSCR1);
+	pci_write_reg(0x0000000f, SH7780_PCICSAR1);
 
-	outl(0xfd000000, PCI_REG(SH7780_PCIMBR0));
-	outl(0x00fc0000, PCI_REG(SH7780_PCIMBMR0));
+	pci_write_reg(0xfd000000, SH7780_PCIMBR0);
+	pci_write_reg(0x00fc0000, SH7780_PCIMBMR0);
+
+#ifdef CONFIG_32BIT
+	pci_write_reg(0xc0000000, SH7780_PCIMBR2);
+	pci_write_reg(0x20000000 - SH7780_PCI_IO_SIZE, SH7780_PCIMBMR2);
+#endif
 
 	/* Set IOBR for windows containing area specified in pci.h */
-	outl((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE-1)), PCI_REG(SH7780_PCIIOBR));
-	outl(((SH7780_PCI_IO_SIZE-1) & (7<<18)), PCI_REG(SH7780_PCIIOBMR));
+	pci_write_reg((PCIBIOS_MIN_IO & ~(SH7780_PCI_IO_SIZE - 1)),
+		      SH7780_PCIIOBR);
+	pci_write_reg(((SH7780_PCI_IO_SIZE-1) & (7<<18)), SH7780_PCIIOBMR);
 
 	return 0;
 }
-
diff --git a/arch/sh/drivers/pci/fixups-rts7751r2d.c b/arch/sh/drivers/pci/fixups-rts7751r2d.c
index 0c590fc7a081..e72ceb560d5b 100644
--- a/arch/sh/drivers/pci/fixups-rts7751r2d.c
+++ b/arch/sh/drivers/pci/fixups-rts7751r2d.c
@@ -10,8 +10,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-#include "pci-sh7751.h"
-#include <asm/io.h>
+#include "pci-sh4.h"
 
 #define PCIMCR_MRSET_OFF	0xBFFFFFFF
 #define PCIMCR_RFSH_OFF		0xFFFFFFFB
@@ -22,22 +21,23 @@ int pci_fixup_pcic(void)
 
 	bcr1 = inl(SH7751_BCR1);
 	bcr1 |= 0x40080000;	/* Enable Bit 19 BREQEN, set PCIC to slave */
-	outl(bcr1, PCI_REG(SH7751_PCIBCR1));
+	pci_write_reg(bcr1, SH4_PCIBCR1);
 
 	/* Enable all interrupts, so we known what to fix */
-	outl(0x0000c3ff, PCI_REG(SH7751_PCIINTM));
-	outl(0x0000380f, PCI_REG(SH7751_PCIAINTM));
+	pci_write_reg(0x0000c3ff, SH4_PCIINTM);
+	pci_write_reg(0x0000380f, SH4_PCIAINTM);
 
-	outl(0xfb900047, PCI_REG(SH7751_PCICONF1));
-	outl(0xab000001, PCI_REG(SH7751_PCICONF4));
+	pci_write_reg(0xfb900047, SH7751_PCICONF1);
+	pci_write_reg(0xab000001, SH7751_PCICONF4);
 
 	mcr = inl(SH7751_MCR);
 	mcr = (mcr & PCIMCR_MRSET_OFF) & PCIMCR_RFSH_OFF;
-	outl(mcr, PCI_REG(SH7751_PCIMCR));
+	pci_write_reg(mcr, SH4_PCIMCR);
+
+	pci_write_reg(0x0c000000, SH7751_PCICONF5);
+	pci_write_reg(0xd0000000, SH7751_PCICONF6);
+	pci_write_reg(0x0c000000, SH4_PCILAR0);
+	pci_write_reg(0x00000000, SH4_PCILAR1);
 
-	outl(0x0c000000, PCI_REG(SH7751_PCICONF5));
-	outl(0xd0000000, PCI_REG(SH7751_PCICONF6));
-	outl(0x0c000000, PCI_REG(SH7751_PCILAR0));
-	outl(0x00000000, PCI_REG(SH7751_PCILAR1));
 	return 0;
 }
diff --git a/arch/sh/drivers/pci/fixups-sh03.c b/arch/sh/drivers/pci/fixups-sh03.c
index 57ac26c2171f..2e8a18b7ee53 100644
--- a/arch/sh/drivers/pci/fixups-sh03.c
+++ b/arch/sh/drivers/pci/fixups-sh03.c
@@ -3,11 +3,7 @@
 #include <linux/types.h>
 #include <linux/pci.h>
 
-/*
- * 	IRQ functions
- */
-
-int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
+int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
 {
 	int irq;
 
@@ -17,8 +13,9 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
 		case 8: return 5;	/* eth1       */
 		case 6: return 2;	/* PCI bridge */
 		default:
-                	printk("PCI: Bad IRQ mapping request for slot %d\n", slot);
-                	return 2;
+			printk(KERN_ERR "PCI: Bad IRQ mapping request "
+					"for slot %d\n", slot);
+			return 2;
 		}
 	} else {
 		switch (pin) {
@@ -32,30 +29,3 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin, struct pci_dev *dev)
 	}
 	return irq;
 }
-
-static u8 __init sh03_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh03_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot, pin, dev);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh03_no_swizzle, sh03_pci_lookup_irq);
-}
diff --git a/arch/sh/drivers/pci/ops-bigsur.c b/arch/sh/drivers/pci/ops-bigsur.c
index ae82c6ca05e5..5da501bd77b5 100644
--- a/arch/sh/drivers/pci/ops-bigsur.c
+++ b/arch/sh/drivers/pci/ops-bigsur.c
@@ -10,15 +10,12 @@
  *
  * PCI initialization for the Hitachi Big Sur Evaluation Board
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <linux/pci.h>
-
 #include <asm/io.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 #include <asm/bigsur/bigsur.h>
 
 #define BIGSUR_PCI_IO	0x4000
@@ -41,11 +38,11 @@ static struct resource sh7751_mem_resource = {
 extern struct pci_ops sh7751_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ 0, }
 };
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS3_BASE_ADDR,
 		.size	= BIGSUR_LSR0_SIZE,
@@ -58,7 +55,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 };
 
 /*
- * Initialize the Big Sur PCI interface 
+ * Initialize the Big Sur PCI interface
  * Setup hardware to be Central Funtion
  * Copy the BSR regs to the PCI interface
  * Setup PCI windows into local RAM
@@ -68,15 +65,15 @@ int __init pcibios_init_platform(void)
 	return sh7751_pcic_init(&sh7751_pci_map);
 }
 
-int pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
-	/* 
+	/*
 	 * The Big Sur can be used in a CPCI chassis, but the SH7751 PCI
 	 * interface is on the wrong end of the board so that it can also
 	 * support a V320 CPI interface chip...  Therefor the IRQ mapping is
 	 * somewhat use dependent... I'l assume a linear map for now, i.e.
 	 * INTA=slot0,pin0... INTD=slot3,pin0...
-	 */ 
+	 */
 	int irq = (slot + pin-1) % 4 + BIGSUR_SH7751_PCI_IRQ_BASE;
 
 	PCIDBG(2, "PCI: Mapping Big Sur IRQ for slot %d, pin %c to irq %d\n",
@@ -84,4 +81,3 @@ int pcibios_map_platform_irq(u8 slot, u8 pin)
 
 	return irq;
 }
-
diff --git a/arch/sh/drivers/pci/ops-landisk.c b/arch/sh/drivers/pci/ops-landisk.c
new file mode 100644
index 000000000000..ada301c21fe7
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-landisk.c
@@ -0,0 +1,68 @@
+/*
+ * arch/sh/drivers/pci/ops-landisk.c
+ *
+ * PCI initialization for the I-O DATA Device, Inc. LANDISK board
+ *
+ * Copyright (C) 2006 kogiidena
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ */
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/pci.h>
+#include "pci-sh4.h"
+
+static struct resource sh7751_io_resource = {
+	.name = "SH7751 IO",
+	.start = 0x4000,
+	.end = 0x4000 + SH7751_PCI_IO_SIZE - 1,
+	.flags = IORESOURCE_IO
+};
+
+static struct resource sh7751_mem_resource = {
+	.name = "SH7751 mem",
+	.start = SH7751_PCI_MEMORY_BASE,
+	.end = SH7751_PCI_MEMORY_BASE + SH7751_PCI_MEM_SIZE - 1,
+	.flags = IORESOURCE_MEM
+};
+
+struct pci_channel board_pci_channels[] = {
+	{&sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0x3ff},
+	{NULL, NULL, NULL, 0, 0},
+};
+
+static struct sh4_pci_address_map sh7751_pci_map = {
+	.window0 = {
+		.base	= SH7751_CS3_BASE_ADDR,
+		.size	= (64 << 20),	/* 64MB */
+	},
+
+	.flags = SH4_PCIC_NO_RESET,
+};
+
+int __init pcibios_init_platform(void)
+{
+	return sh7751_pcic_init(&sh7751_pci_map);
+}
+
+int pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
+{
+	/*
+	 * slot0: pin1-4 = irq5,6,7,8
+	 * slot1: pin1-4 = irq6,7,8,5
+	 * slot2: pin1-4 = irq7,8,5,6
+	 * slot3: pin1-4 = irq8,5,6,7
+	 */
+	int irq = ((slot + pin - 1) & 0x3) + 5;
+
+	if ((slot | (pin - 1)) > 0x3) {
+		printk("PCI: Bad IRQ mapping request for slot %d pin %c\n",
+		       slot, pin - 1 + 'A');
+		return -1;
+	}
+	return irq;
+}
diff --git a/arch/sh/drivers/pci/ops-r7780rp.c b/arch/sh/drivers/pci/ops-r7780rp.c
index 3254c4e917a9..554d5ed2c586 100644
--- a/arch/sh/drivers/pci/ops-r7780rp.c
+++ b/arch/sh/drivers/pci/ops-r7780rp.c
@@ -15,13 +15,11 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/pci.h>
-#include <linux/module.h>
-
-#include <asm/io.h>
-#include "pci-sh7780.h"
 #include <asm/r7780rp/r7780rp.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
         switch (slot) {
 	case 0: return IRQ_PCISLOT1;		/* PCI Interrupt #1 */
@@ -29,7 +27,8 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin)
 	case 2: return IRQ_PCISLOT3;		/* PCI Interrupt #3 */
 	case 3: return IRQ_PCISLOT4;		/* PCI Interrupt E4 */
 	default:
-		printk("PCI: Bad IRQ mapping request for slot %d, func %d\n", slot, pin-1);
+		printk(KERN_ERR "PCI: Bad IRQ mapping "
+		       "request for slot %d, func %d\n", slot, pin-1);
 		return -1;
 	}
 }
@@ -51,12 +50,12 @@ static struct resource sh7780_mem_resource = {
 extern struct pci_ops sh7780_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7780_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7780_io_resource, &sh7780_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7780_pci_address_map sh7780_pci_map = {
+static struct sh4_pci_address_map sh7780_pci_map = {
 	.window0	= {
 		.base	= SH7780_CS2_BASE_ADDR,
 		.size	= 0x04000000,
@@ -67,11 +66,10 @@ static struct sh7780_pci_address_map sh7780_pci_map = {
 		.size	= 0x04000000,
 	},
 
-	.flags	= SH7780_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
 {
 	return sh7780_pcic_init(&sh7780_pci_map);
 }
-
diff --git a/arch/sh/drivers/pci/ops-rts7751r2d.c b/arch/sh/drivers/pci/ops-rts7751r2d.c
index 83171d10141a..88f44e245424 100644
--- a/arch/sh/drivers/pci/ops-rts7751r2d.c
+++ b/arch/sh/drivers/pci/ops-rts7751r2d.c
@@ -17,12 +17,11 @@
 #include <linux/delay.h>
 #include <linux/pci.h>
 #include <linux/module.h>
-
-#include <asm/io.h>
-#include "pci-sh7751.h"
 #include <asm/rts7751r2d/rts7751r2d.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
         switch (slot) {
 	case 0: return IRQ_PCISLOT1;	/* PCI Extend slot #1 */
@@ -52,12 +51,12 @@ static struct resource sh7751_mem_resource = {
 extern struct pci_ops sh7751_pci_ops;
 
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS3_BASE_ADDR,
 		.size	= 0x04000000,
@@ -68,7 +67,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= 0x00000000,	/* Unused */
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
diff --git a/arch/sh/drivers/pci/ops-sh4.c b/arch/sh/drivers/pci/ops-sh4.c
new file mode 100644
index 000000000000..2d4371009a5e
--- /dev/null
+++ b/arch/sh/drivers/pci/ops-sh4.c
@@ -0,0 +1,164 @@
+/*
+ * Generic SH-4 / SH-4A PCIC operations (SH7751, SH7780).
+ *
+ * Copyright (C) 2002 - 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License v2. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/pci.h>
+#include <asm/addrspace.h>
+#include <asm/io.h>
+#include "pci-sh4.h"
+
+/*
+ * Direct access to PCI hardware...
+ */
+#define CONFIG_CMD(bus, devfn, where) \
+	P1SEGADDR((bus->number << 16) | (devfn << 8) | (where & ~3))
+
+static DEFINE_SPINLOCK(sh4_pci_lock);
+
+/*
+ * Functions for accessing PCI configuration space with type 1 accesses
+ */
+static int sh4_pci_read(struct pci_bus *bus, unsigned int devfn,
+			   int where, int size, u32 *val)
+{
+	unsigned long flags;
+	u32 data;
+
+	/*
+	 * PCIPDR may only be accessed as 32 bit words,
+	 * so we must do byte alignment by hand
+	 */
+	spin_lock_irqsave(&sh4_pci_lock, flags);
+	pci_write_reg(CONFIG_CMD(bus, devfn, where), SH4_PCIPAR);
+	data = pci_read_reg(SH4_PCIPDR);
+	spin_unlock_irqrestore(&sh4_pci_lock, flags);
+
+	switch (size) {
+	case 1:
+		*val = (data >> ((where & 3) << 3)) & 0xff;
+		break;
+	case 2:
+		*val = (data >> ((where & 2) << 3)) & 0xffff;
+		break;
+	case 4:
+		*val = data;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+/*
+ * Since SH4 only does 32bit access we'll have to do a read,
+ * mask,write operation.
+ * We'll allow an odd byte offset, though it should be illegal.
+ */
+static int sh4_pci_write(struct pci_bus *bus, unsigned int devfn,
+			 int where, int size, u32 val)
+{
+	unsigned long flags;
+	int shift;
+	u32 data;
+
+	spin_lock_irqsave(&sh4_pci_lock, flags);
+	pci_write_reg(CONFIG_CMD(bus, devfn, where), SH4_PCIPAR);
+	data = pci_read_reg(SH4_PCIPDR);
+	spin_unlock_irqrestore(&sh4_pci_lock, flags);
+
+	switch (size) {
+	case 1:
+		shift = (where & 3) << 3;
+		data &= ~(0xff << shift);
+		data |= ((val & 0xff) << shift);
+		break;
+	case 2:
+		shift = (where & 2) << 3;
+		data &= ~(0xffff << shift);
+		data |= ((val & 0xffff) << shift);
+		break;
+	case 4:
+		data = val;
+		break;
+	default:
+		return PCIBIOS_FUNC_NOT_SUPPORTED;
+	}
+
+	pci_write_reg(data, SH4_PCIPDR);
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+struct pci_ops sh4_pci_ops = {
+	.read		= sh4_pci_read,
+	.write		= sh4_pci_write,
+};
+
+/*
+ * Not really related to pci_ops, but it's common and not worth shoving
+ * somewhere else for now..
+ */
+static unsigned int pci_probe = PCI_PROBE_CONF1;
+
+int __init sh4_pci_check_direct(void)
+{
+	/*
+	 * Check if configuration works.
+	 */
+	if (pci_probe & PCI_PROBE_CONF1) {
+		unsigned int tmp = pci_read_reg(SH4_PCIPAR);
+
+		pci_write_reg(P1SEG, SH4_PCIPAR);
+
+		if (pci_read_reg(SH4_PCIPAR) == P1SEG) {
+			pci_write_reg(tmp, SH4_PCIPAR);
+			printk(KERN_INFO "PCI: Using configuration type 1\n");
+			request_region(PCI_REG(SH4_PCIPAR), 8, "PCI conf1");
+
+			return 0;
+		}
+
+		pci_write_reg(tmp, SH4_PCIPAR);
+	}
+
+	pr_debug("PCI: pci_check_direct failed\n");
+	return -EINVAL;
+}
+
+/* Handle generic fixups */
+static void __init pci_fixup_ide_bases(struct pci_dev *d)
+{
+	int i;
+
+	/*
+	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
+	 */
+	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
+		return;
+	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
+	for(i = 0; i < 4; i++) {
+		struct resource *r = &d->resource[i];
+
+		if ((r->start & ~0x80) == 0x374) {
+			r->start |= 2;
+			r->end = r->start;
+		}
+	}
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
+
+char * __init pcibios_setup(char *str)
+{
+	if (!strcmp(str, "off")) {
+		pci_probe = 0;
+		return NULL;
+	}
+
+	return str;
+}
diff --git a/arch/sh/drivers/pci/ops-snapgear.c b/arch/sh/drivers/pci/ops-snapgear.c
index 3cbd14dd28fe..53dd893d4e54 100644
--- a/arch/sh/drivers/pci/ops-snapgear.c
+++ b/arch/sh/drivers/pci/ops-snapgear.c
@@ -2,7 +2,7 @@
  * arch/sh/drivers/pci/ops-snapgear.c
  *
  * Author:  David McCullough <davidm@snapgear.com>
- * 
+ *
  * Ported to new API by Paul Mundt <lethal@linux-sh.org>
  *
  * Highly leveraged from pci-bigsur.c, written by Dustin McIntire.
@@ -12,15 +12,11 @@
  *
  * PCI initialization for the SnapGear boards
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <linux/pci.h>
-
-#include <asm/io.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 
 #define SNAPGEAR_PCI_IO		0x4000
 #define SNAPGEAR_PCI_MEM	0xfd000000
@@ -43,14 +39,12 @@ static struct resource sh7751_mem_resource = {
 	.flags		= IORESOURCE_MEM,
 };
 
-extern struct pci_ops sh7751_pci_ops;
-
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ 0, }
 };
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS2_BASE_ADDR,
 		.size	= SNAPGEAR_LSR0_SIZE,
@@ -61,11 +55,11 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= SNAPGEAR_LSR1_SIZE,
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 /*
- * Initialize the SnapGear PCI interface 
+ * Initialize the SnapGear PCI interface
  * Setup hardware to be Central Funtion
  * Copy the BSR regs to the PCI interface
  * Setup PCI windows into local RAM
@@ -75,7 +69,7 @@ int __init pcibios_init_platform(void)
 	return sh7751_pcic_init(&sh7751_pci_map);
 }
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
 	int irq = -1;
 
@@ -98,4 +92,3 @@ void __init pcibios_fixup(void)
 {
 	/* Nothing to fixup .. */
 }
-
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
index f76e4e963ac1..9c8b2027c35d 100644
--- a/arch/sh/drivers/pci/ops-titan.c
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -16,12 +16,11 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
-#include <linux/delay.h>
 #include <asm/io.h>
 #include <asm/titan.h>
-#include "pci-sh7751.h"
+#include "pci-sh4.h"
 
-int __init pcibios_map_platform_irq(u8 slot, u8 pin)
+int __init pcibios_map_platform_irq(struct pci_dev *pdev, u8 slot, u8 pin)
 {
 	int irq = -1;
 
@@ -32,7 +31,8 @@ int __init pcibios_map_platform_irq(u8 slot, u8 pin)
 	case 3: irq = TITAN_IRQ_MPCIB; break;	/* mPCI B */
 	case 4: irq = TITAN_IRQ_USB;   break;	/* USB */
 	default:
-		printk(KERN_INFO "PCI: Bad IRQ mapping request for slot %d\n", slot);
+		printk(KERN_INFO "PCI: Bad IRQ mapping "
+				 "request for slot %d\n", slot);
 		return -1;
 	}
 
@@ -56,15 +56,13 @@ static struct resource sh7751_mem_resource = {
 	.flags	= IORESOURCE_MEM
 };
 
-extern struct pci_ops sh7751_pci_ops;
-
 struct pci_channel board_pci_channels[] = {
-	{ &sh7751_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
+	{ &sh4_pci_ops, &sh7751_io_resource, &sh7751_mem_resource, 0, 0xff },
 	{ NULL, NULL, NULL, 0, 0 },
 };
 EXPORT_SYMBOL(board_pci_channels);
 
-static struct sh7751_pci_address_map sh7751_pci_map = {
+static struct sh4_pci_address_map sh7751_pci_map = {
 	.window0	= {
 		.base	= SH7751_CS2_BASE_ADDR,
 		.size	= SH7751_MEM_REGION_SIZE*2,	/* cs2 and cs3 */
@@ -75,7 +73,7 @@ static struct sh7751_pci_address_map sh7751_pci_map = {
 		.size	= SH7751_MEM_REGION_SIZE*2,
 	},
 
-	.flags	= SH7751_PCIC_NO_RESET,
+	.flags	= SH4_PCIC_NO_RESET,
 };
 
 int __init pcibios_init_platform(void)
diff --git a/arch/sh/drivers/pci/pci-auto.c b/arch/sh/drivers/pci/pci-auto.c
index d55e46618549..ecf16344f94a 100644
--- a/arch/sh/drivers/pci/pci-auto.c
+++ b/arch/sh/drivers/pci/pci-auto.c
@@ -358,7 +358,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 {
 	u32 temp;
 
-#if !defined(CONFIG_SH_HS7751RVOIP) && !defined(CONFIG_SH_RTS7751R2D) && !defined(CONFIG_SH_R7780RP)
 	/*
 	 * [jsun] we always bump up baselines a little, so that if there
 	 * nothing behind P2P bridge, we don't wind up overlapping IO/MEM
@@ -366,7 +365,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 */
 	pciauto_lower_memspc += 1;
 	pciauto_lower_iospc += 1;
-#endif
 
 	/*
 	 * Configure subordinate bus number.  The PCI subsystem
@@ -392,11 +390,6 @@ pciauto_postscan_setup_cardbus_bridge(struct pci_channel *hose,
 	 * configured by this routine to happily live behind a
 	 * P2P bridge in a system.
 	 */
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
-	pciauto_lower_memspc += 0x00400000;
-	pciauto_lower_iospc += 0x00004000;
-#endif
-
 	/* Align memory and I/O to 4KB and 4 byte boundaries. */
 	pciauto_lower_memspc = (pciauto_lower_memspc + (0x1000 - 1))
 		& ~(0x1000 - 1);
@@ -467,9 +460,6 @@ pciauto_bus_scan(struct pci_channel *hose, int top_bus, int current_bus)
 		if ((pci_class >> 16) == PCI_CLASS_BRIDGE_PCI) {
 			DBG("        Bridge: primary=%.2x, secondary=%.2x\n",
 				current_bus, sub_bus + 1);
-#if defined(CONFIG_SH_HS7751RVOIP) || defined(CONFIG_SH_RTS7751R2D) || defined(CONFIG_SH_R7780RP)
-			pciauto_setup_bars(hose, top_bus, current_bus, pci_devfn, PCI_BASE_ADDRESS_1);
-#endif
 			pciauto_prescan_setup_bridge(hose, top_bus, current_bus,
 						     pci_devfn, sub_bus);
 			DBG("Scanning sub bus %.2x, I/O 0x%.8x, Mem 0x%.8x\n",
diff --git a/arch/sh/drivers/pci/pci-sh4.h b/arch/sh/drivers/pci/pci-sh4.h
new file mode 100644
index 000000000000..5a61d6041f2c
--- /dev/null
+++ b/arch/sh/drivers/pci/pci-sh4.h
@@ -0,0 +1,180 @@
+#ifndef __PCI_SH4_H
+#define __PCI_SH4_H
+
+#ifdef CONFIG_CPU_SUBTYPE_SH7780
+#include "pci-sh7780.h"
+#else
+#include "pci-sh7751.h"
+#endif
+
+#include <asm/io.h>
+
+/* startup values */
+#define PCI_PROBE_BIOS		1
+#define PCI_PROBE_CONF1		2
+#define PCI_PROBE_CONF2		4
+#define PCI_NO_SORT		0x100
+#define PCI_BIOS_SORT		0x200
+#define PCI_NO_CHECKS		0x400
+#define PCI_ASSIGN_ROMS		0x1000
+#define PCI_BIOS_IRQ_SCAN	0x2000
+
+#define SH4_PCICR		0x100		/* PCI Control Register */
+  #define SH4_PCICR_PREFIX	  0xA5000000	/* CR prefix for write */
+  #define SH4_PCICR_FTO		  0x00000400	/* TRDY/IRDY Enable */
+  #define SH4_PCICR_TRSB	  0x00000200	/* Target Read Single */
+  #define SH4_PCICR_BSWP	  0x00000100	/* Target Byte Swap */
+  #define SH4_PCICR_PLUP	  0x00000080	/* Enable PCI Pullup */
+  #define SH4_PCICR_ARBM	  0x00000040	/* PCI Arbitration Mode */
+  #define SH4_PCICR_MD		  0x00000030	/* MD9 and MD10 status */
+  #define SH4_PCICR_SERR	  0x00000008	/* SERR output assert */
+  #define SH4_PCICR_INTA	  0x00000004	/* INTA output assert */
+  #define SH4_PCICR_PRST	  0x00000002	/* PCI Reset Assert */
+  #define SH4_PCICR_CFIN	  0x00000001	/* Central Fun. Init Done */
+#define SH4_PCILSR0		0x104		/* PCI Local Space Register0 */
+#define SH4_PCILSR1		0x108		/* PCI Local Space Register1 */
+#define SH4_PCILAR0		0x10C		/* PCI Local Addr Register1 */
+#define SH4_PCILAR1		0x110		/* PCI Local Addr Register1 */
+#define SH4_PCIINT		0x114		/* PCI Interrupt Register */
+  #define SH4_PCIINT_MLCK	  0x00008000	/* Master Lock Error */
+  #define SH4_PCIINT_TABT	  0x00004000	/* Target Abort Error */
+  #define SH4_PCIINT_TRET	  0x00000200	/* Target Retry Error */
+  #define SH4_PCIINT_MFDE	  0x00000100	/* Master Func. Disable Error */
+  #define SH4_PCIINT_PRTY	  0x00000080	/* Address Parity Error */
+  #define SH4_PCIINT_SERR	  0x00000040	/* SERR Detection Error */
+  #define SH4_PCIINT_TWDP	  0x00000020	/* Tgt. Write Parity Error */
+  #define SH4_PCIINT_TRDP	  0x00000010	/* Tgt. Read Parity Err Det. */
+  #define SH4_PCIINT_MTABT	  0x00000008	/* Master-Tgt. Abort Error */
+  #define SH4_PCIINT_MMABT	  0x00000004	/* Master-Master Abort Error */
+  #define SH4_PCIINT_MWPD	  0x00000002	/* Master Write PERR Detect */
+  #define SH4_PCIINT_MRPD	  0x00000001	/* Master Read PERR Detect */
+#define SH4_PCIINTM		0x118		/* PCI Interrupt Mask */
+#define SH4_PCIALR		0x11C		/* Error Address Register */
+#define SH4_PCICLR		0x120		/* Error Command/Data */
+  #define SH4_PCICLR_MPIO	  0x80000000
+  #define SH4_PCICLR_MDMA0	  0x40000000	/* DMA0 Transfer Error */
+  #define SH4_PCICLR_MDMA1	  0x20000000	/* DMA1 Transfer Error */
+  #define SH4_PCICLR_MDMA2	  0x10000000	/* DMA2 Transfer Error */
+  #define SH4_PCICLR_MDMA3	  0x08000000	/* DMA3 Transfer Error */
+  #define SH4_PCICLR_TGT	  0x04000000	/* Target Transfer Error */
+  #define SH4_PCICLR_CMDL	  0x0000000F	/* PCI Command at Error */
+#define SH4_PCIAINT		0x130		/* Arbiter Interrupt Register */
+  #define SH4_PCIAINT_MBKN	  0x00002000	/* Master Broken Interrupt */
+  #define SH4_PCIAINT_TBTO	  0x00001000	/* Target Bus Time Out */
+  #define SH4_PCIAINT_MBTO	  0x00001000	/* Master Bus Time Out */
+  #define SH4_PCIAINT_TABT	  0x00000008	/* Target Abort */
+  #define SH4_PCIAINT_MABT	  0x00000004	/* Master Abort */
+  #define SH4_PCIAINT_RDPE	  0x00000002	/* Read Data Parity Error */
+  #define SH4_PCIAINT_WDPE	  0x00000001	/* Write Data Parity Error */
+#define SH4_PCIAINTM            0x134		/* Arbiter Int. Mask Register */
+#define SH4_PCIBMLR		0x138		/* Error Bus Master Register */
+  #define SH4_PCIBMLR_REQ4	  0x00000010	/* REQ4 bus master at error */
+  #define SH4_PCIBMLR_REQ3	  0x00000008	/* REQ3 bus master at error */
+  #define SH4_PCIBMLR_REQ2	  0x00000004	/* REQ2 bus master at error */
+  #define SH4_PCIBMLR_REQ1	  0x00000002	/* REQ1 bus master at error */
+  #define SH4_PCIBMLR_REQ0	  0x00000001	/* REQ0 bus master at error */
+#define SH4_PCIDMABT		0x140		/* DMA Transfer Arb. Register */
+  #define SH4_PCIDMABT_RRBN	  0x00000001	/* DMA Arbitor Round-Robin */
+#define SH4_PCIDPA0		0x180		/* DMA0 Transfer Addr. */
+#define SH4_PCIDLA0		0x184		/* DMA0 Local Addr. */
+#define SH4_PCIDTC0		0x188		/* DMA0 Transfer Cnt. */
+#define SH4_PCIDCR0		0x18C		/* DMA0 Control Register */
+  #define SH4_PCIDCR_ALGN	  0x00000600	/* DMA Alignment Mode */
+  #define SH4_PCIDCR_MAST	  0x00000100	/* DMA Termination Type */
+  #define SH4_PCIDCR_INTM	  0x00000080	/* DMA Interrupt Done Mask*/
+  #define SH4_PCIDCR_INTS	  0x00000040	/* DMA Interrupt Done Status */
+  #define SH4_PCIDCR_LHLD	  0x00000020	/* Local Address Control */
+  #define SH4_PCIDCR_PHLD	  0x00000010	/* PCI Address Control*/
+  #define SH4_PCIDCR_IOSEL	  0x00000008	/* PCI Address Space Type */
+  #define SH4_PCIDCR_DIR	  0x00000004	/* DMA Transfer Direction */
+  #define SH4_PCIDCR_STOP	  0x00000002	/* Force DMA Stop */
+  #define SH4_PCIDCR_STRT	  0x00000001	/* DMA Start */
+#define SH4_PCIDPA1		0x190		/* DMA1 Transfer Addr. */
+#define SH4_PCIDLA1		0x194		/* DMA1 Local Addr. */
+#define SH4_PCIDTC1		0x198		/* DMA1 Transfer Cnt. */
+#define SH4_PCIDCR1		0x19C		/* DMA1 Control Register */
+#define SH4_PCIDPA2		0x1A0		/* DMA2 Transfer Addr. */
+#define SH4_PCIDLA2		0x1A4		/* DMA2 Local Addr. */
+#define SH4_PCIDTC2		0x1A8		/* DMA2 Transfer Cnt. */
+#define SH4_PCIDCR2		0x1AC		/* DMA2 Control Register */
+#define SH4_PCIDPA3		0x1B0		/* DMA3 Transfer Addr. */
+#define SH4_PCIDLA3		0x1B4		/* DMA3 Local Addr. */
+#define SH4_PCIDTC3		0x1B8		/* DMA3 Transfer Cnt. */
+#define SH4_PCIDCR3		0x1BC		/* DMA3 Control Register */
+#define SH4_PCIPAR		0x1C0		/* PIO Address Register */
+  #define SH4_PCIPAR_CFGEN	  0x80000000	/* Configuration Enable */
+  #define SH4_PCIPAR_BUSNO	  0x00FF0000	/* Config. Bus Number */
+  #define SH4_PCIPAR_DEVNO	  0x0000FF00	/* Config. Device Number */
+  #define SH4_PCIPAR_REGAD	  0x000000FC	/* Register Address Number */
+#define SH4_PCIMBR		0x1C4		/* Memory Base Address */
+  #define SH4_PCIMBR_MASK	  0xFF000000	/* Memory Space Mask */
+  #define SH4_PCIMBR_LOCK	  0x00000001	/* Lock Memory Space */
+#define SH4_PCIIOBR		0x1C8		/* I/O Base Address Register */
+  #define SH4_PCIIOBR_MASK	  0xFFFC0000	/* IO Space Mask */
+  #define SH4_PCIIOBR_LOCK	  0x00000001	/* Lock IO Space */
+#define SH4_PCIPINT		0x1CC		/* Power Mgmnt Int. Register */
+  #define SH4_PCIPINT_D3	  0x00000002	/* D3 Pwr Mgmt. Interrupt */
+  #define SH4_PCIPINT_D0	  0x00000001	/* D0 Pwr Mgmt. Interrupt */
+#define SH4_PCIPINTM		0x1D0		/* Power Mgmnt Mask Register */
+#define SH4_PCICLKR		0x1D4		/* Clock Ctrl. Register */
+  #define SH4_PCICLKR_PCSTP	  0x00000002	/* PCI Clock Stop */
+  #define SH4_PCICLKR_BCSTP	  0x00000001	/* BCLK Clock Stop */
+/* For definitions of BCR, MCR see ... */
+#define SH4_PCIBCR1		0x1E0		/* Memory BCR1 Register */
+  #define SH4_PCIMBR0		SH4_PCIBCR1
+#define SH4_PCIBCR2		0x1E4		/* Memory BCR2 Register */
+  #define SH4_PCIMBMR0		SH4_PCIBCR2
+#define SH4_PCIWCR1		0x1E8		/* Wait Control 1 Register */
+#define SH4_PCIWCR2		0x1EC		/* Wait Control 2 Register */
+#define SH4_PCIWCR3		0x1F0		/* Wait Control 3 Register */
+  #define SH4_PCIMBR2		SH4_PCIWCR3
+#define SH4_PCIMCR		0x1F4		/* Memory Control Register */
+#define SH4_PCIBCR3		0x1f8		/* Memory BCR3 Register */
+#define SH4_PCIPCTR             0x200		/* Port Control Register */
+  #define SH4_PCIPCTR_P2EN	  0x000400000	/* Port 2 Enable */
+  #define SH4_PCIPCTR_P1EN	  0x000200000	/* Port 1 Enable */
+  #define SH4_PCIPCTR_P0EN	  0x000100000	/* Port 0 Enable */
+  #define SH4_PCIPCTR_P2UP	  0x000000020	/* Port2 Pull Up Enable */
+  #define SH4_PCIPCTR_P2IO	  0x000000010	/* Port2 Output Enable */
+  #define SH4_PCIPCTR_P1UP	  0x000000008	/* Port1 Pull Up Enable */
+  #define SH4_PCIPCTR_P1IO	  0x000000004	/* Port1 Output Enable */
+  #define SH4_PCIPCTR_P0UP	  0x000000002	/* Port0 Pull Up Enable */
+  #define SH4_PCIPCTR_P0IO	  0x000000001	/* Port0 Output Enable */
+#define SH4_PCIPDTR		0x204		/* Port Data Register */
+  #define SH4_PCIPDTR_PB5	  0x000000020	/* Port 5 Enable */
+  #define SH4_PCIPDTR_PB4	  0x000000010	/* Port 4 Enable */
+  #define SH4_PCIPDTR_PB3	  0x000000008	/* Port 3 Enable */
+  #define SH4_PCIPDTR_PB2	  0x000000004	/* Port 2 Enable */
+  #define SH4_PCIPDTR_PB1	  0x000000002	/* Port 1 Enable */
+  #define SH4_PCIPDTR_PB0	  0x000000001	/* Port 0 Enable */
+#define SH4_PCIPDR		0x220		/* Port IO Data Register */
+
+/* Flags */
+#define SH4_PCIC_NO_RESET	0x0001
+
+/* arch/sh/kernel/drivers/pci/ops-sh4.c */
+extern struct pci_ops sh4_pci_ops;
+int sh4_pci_check_direct(void);
+int pci_fixup_pcic(void);
+
+struct sh4_pci_address_space {
+	unsigned long base;
+	unsigned long size;
+};
+
+struct sh4_pci_address_map {
+	struct sh4_pci_address_space window0;
+	struct sh4_pci_address_space window1;
+	unsigned long flags;
+};
+
+static inline void pci_write_reg(unsigned long val, unsigned long reg)
+{
+	outl(val, PCI_REG(reg));
+}
+
+static inline unsigned long pci_read_reg(unsigned long reg)
+{
+	return inl(PCI_REG(reg));
+}
+#endif /* __PCI_SH4_H */
diff --git a/arch/sh/drivers/pci/pci-sh7751.c b/arch/sh/drivers/pci/pci-sh7751.c
index 65093ec1b55e..dbe837884983 100644
--- a/arch/sh/drivers/pci/pci-sh7751.c
+++ b/arch/sh/drivers/pci/pci-sh7751.c
@@ -15,180 +15,14 @@
 
 #undef DEBUG
 
-#include <linux/types.h>
-#include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/pci.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
+#include <linux/types.h>
 #include <linux/errno.h>
-#include <linux/irq.h>
 #include <linux/delay.h>
-
-#include <asm/machvec.h>
+#include "pci-sh4.h"
+#include <asm/addrspace.h>
 #include <asm/io.h>
-#include "pci-sh7751.h"
-
-static unsigned int pci_probe = PCI_PROBE_CONF1;
-extern int pci_fixup_pcic(void);
-
-void pcibios_fixup_irqs(void) __attribute__ ((weak));
-
-/*
- * Direct access to PCI hardware...
- */
-
-#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
-
-/*
- * Functions for accessing PCI configuration space with type 1 accesses
- */
-static int sh7751_pci_read(struct pci_bus *bus, unsigned int devfn,
-			   int where, int size, u32 *val)
-{
-	unsigned long flags;
-	u32 data;
-
-	/* 
-	 * PCIPDR may only be accessed as 32 bit words, 
-	 * so we must do byte alignment by hand 
-	 */
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus,devfn,where), PCI_REG(SH7751_PCIPAR));
-	data = inl(PCI_REG(SH7751_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-	case 2:
-		*val = (data >> ((where & 2) << 3)) & 0xffff;
-		break;
-	case 4:
-		*val = data;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-/* 
- * Since SH7751 only does 32bit access we'll have to do a read,
- * mask,write operation.
- * We'll allow an odd byte offset, though it should be illegal.
- */ 
-static int sh7751_pci_write(struct pci_bus *bus, unsigned int devfn,
-			    int where, int size, u32 val)
-{
-	unsigned long flags;
-	int shift;
-	u32 data;
-
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus,devfn,where), PCI_REG(SH7751_PCIPAR));
-	data = inl(PCI_REG(SH7751_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		shift = (where & 3) << 3;
-		data &= ~(0xff << shift);
-		data |= ((val & 0xff) << shift);
-		break;
-	case 2:
-		shift = (where & 2) << 3;
-		data &= ~(0xffff << shift);
-		data |= ((val & 0xffff) << shift);
-		break;
-	case 4:
-		data = val;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	outl(data, PCI_REG(SH7751_PCIPDR));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-#undef CONFIG_CMD
-
-struct pci_ops sh7751_pci_ops = {
-	.read 		= sh7751_pci_read,
-	.write		= sh7751_pci_write,
-};
-
-static int __init pci_check_direct(void)
-{
-	unsigned int tmp, id;
-
-	/* check for SH7751/SH7751R hardware */
-	id = inl(SH7751_PCIREG_BASE+SH7751_PCICONF0);
-	if (id != ((SH7751_DEVICE_ID << 16) | SH7751_VENDOR_ID) &&
-	    id != ((SH7751R_DEVICE_ID << 16) | SH7751_VENDOR_ID)) {
-		pr_debug("PCI: This is not an SH7751(R) (%x)\n", id);
-		return -ENODEV;
-	}
-
-	/*
-	 * Check if configuration works.
-	 */
-	if (pci_probe & PCI_PROBE_CONF1) {
-		tmp = inl (PCI_REG(SH7751_PCIPAR));
-		outl (0x80000000, PCI_REG(SH7751_PCIPAR));
-		if (inl (PCI_REG(SH7751_PCIPAR)) == 0x80000000) {
-			outl (tmp, PCI_REG(SH7751_PCIPAR));
-			printk(KERN_INFO "PCI: Using configuration type 1\n");
-			request_region(PCI_REG(SH7751_PCIPAR), 8, "PCI conf1");
-			return 0;
-		}
-		outl (tmp, PCI_REG(SH7751_PCIPAR));
-	}
-
-	pr_debug("PCI: pci_check_direct failed\n");
-	return -EINVAL;
-}
-
-/***************************************************************************************/
-
-/*
- *  Handle bus scanning and fixups ....
- */
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-/*
- *  Called after each bus is probed, but before its children
- *  are examined.
- */
-
-void __init pcibios_fixup_bus(struct pci_bus *b)
-{
-	pci_read_bridge_bases(b);
-}
 
 /*
  * Initialization. Try all known PCI access methods. Note that we support
@@ -196,25 +30,29 @@ void __init pcibios_fixup_bus(struct pci_bus *b)
  * to access config space.
  * 
  * Note that the platform specific initialization (BSC registers, and memory
- * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
- * exitst and via the platform defined function pcibios_init_platform().  
- * See pci_bigsur.c for implementation;
- * 
- * The BIOS version of the pci functions is not yet implemented but it is left
- * in for completeness.  Currently an error will be genereated at compile time. 
+ * space mapping) will be called via the platform defined function
+ * pcibios_init_platform().
  */
-
 static int __init sh7751_pci_init(void)
 {
+	unsigned int id;
 	int ret;
 
 	pr_debug("PCI: Starting intialization.\n");
-	if ((ret = pci_check_direct()) != 0)
+
+	/* check for SH7751/SH7751R hardware */
+	id = pci_read_reg(SH7751_PCICONF0);
+	if (id != ((SH7751_DEVICE_ID << 16) | SH7751_VENDOR_ID) &&
+	    id != ((SH7751R_DEVICE_ID << 16) | SH7751_VENDOR_ID)) {
+		pr_debug("PCI: This is not an SH7751(R) (%x)\n", id);
+		return -ENODEV;
+	}
+
+	if ((ret = sh4_pci_check_direct()) != 0)
 		return ret;
 
 	return pcibios_init_platform();
 }
-
 subsys_initcall(sh7751_pci_init);
 
 static int __init __area_sdram_check(unsigned int area)
@@ -228,7 +66,7 @@ static int __init __area_sdram_check(unsigned int area)
 		       area, word);
 		return 0;
 	}
-	outl(word, PCI_REG(SH7751_PCIBCR1));
+	pci_write_reg(word, SH4_PCIBCR1);
 
 	word = (u16)inw(SH7751_BCR2);
 	/* check BCR2 for 32bit SDRAM interface*/
@@ -237,12 +75,12 @@ static int __init __area_sdram_check(unsigned int area)
 		       area, word);
 		return 0;
 	}
-	outl(word, PCI_REG(SH7751_PCIBCR2));
+	pci_write_reg(word, SH4_PCIBCR2);
 
 	return 1;
 }
 
-int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
+int __init sh7751_pcic_init(struct sh4_pci_address_map *map)
 {
 	u32 reg;
 	u32 word;
@@ -251,39 +89,39 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 	reg = inl(SH7751_BCR1);
 	reg |= 0x80000;
 	outl(reg, SH7751_BCR1);
-	
+
 	/* Turn the clocks back on (not done in reset)*/
-	outl(0, PCI_REG(SH7751_PCICLKR));
+	pci_write_reg(0, SH4_PCICLKR);
 	/* Clear Powerdown IRQ's (not done in reset) */
-	word = SH7751_PCIPINT_D3 | SH7751_PCIPINT_D0;
-	outl(word, PCI_REG(SH7751_PCIPINT));
+	word = SH4_PCIPINT_D3 | SH4_PCIPINT_D0;
+	pci_write_reg(word, SH4_PCIPINT);
 
 	/*
 	 * This code is unused for some boards as it is done in the
 	 * bootloader and doing it here means the MAC addresses loaded
 	 * by the bootloader get lost.
 	 */
-	if (!(map->flags & SH7751_PCIC_NO_RESET)) {
+	if (!(map->flags & SH4_PCIC_NO_RESET)) {
 		/* toggle PCI reset pin */
-		word = SH7751_PCICR_PREFIX | SH7751_PCICR_PRST;
-		outl(word,PCI_REG(SH7751_PCICR));
+		word = SH4_PCICR_PREFIX | SH4_PCICR_PRST;
+		pci_write_reg(word, SH4_PCICR);
 		/* Wait for a long time... not 1 sec. but long enough */
 		mdelay(100);
-		word = SH7751_PCICR_PREFIX;
-		outl(word,PCI_REG(SH7751_PCICR));
+		word = SH4_PCICR_PREFIX;
+		pci_write_reg(word, SH4_PCICR);
 	}
-	
+
 	/* set the command/status bits to:
 	 * Wait Cycle Control + Parity Enable + Bus Master +
 	 * Mem space enable
 	 */
 	word = SH7751_PCICONF1_WCC | SH7751_PCICONF1_PER | 
 	       SH7751_PCICONF1_BUM | SH7751_PCICONF1_MES;
-	outl(word, PCI_REG(SH7751_PCICONF1));
+	pci_write_reg(word, SH7751_PCICONF1);
 
 	/* define this host as the host bridge */
-	word = SH7751_PCI_HOST_BRIDGE << 24;
-	outl(word, PCI_REG(SH7751_PCICONF2));
+	word = PCI_BASE_CLASS_BRIDGE << 24;
+	pci_write_reg(word, SH7751_PCICONF2);
 
 	/* Set IO and Mem windows to local address 
 	 * Make PCI and local address the same for easy 1 to 1 mapping 
@@ -291,46 +129,49 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 	 * Window1 = map->window1.size @ cached area base = SDRAM 
 	 */
 	word = map->window0.size - 1;
-	outl(word, PCI_REG(SH7751_PCILSR0));
+	pci_write_reg(word, SH4_PCILSR0);
 	word = map->window1.size - 1;
-	outl(word, PCI_REG(SH7751_PCILSR1));
+	pci_write_reg(word, SH4_PCILSR1);
 	/* Set the values on window 0 PCI config registers */
 	word = P2SEGADDR(map->window0.base);
-	outl(word, PCI_REG(SH7751_PCILAR0));
-	outl(word, PCI_REG(SH7751_PCICONF5));
+	pci_write_reg(word, SH4_PCILAR0);
+	pci_write_reg(word, SH7751_PCICONF5);
 	/* Set the values on window 1 PCI config registers */
 	word =  PHYSADDR(map->window1.base);
-	outl(word, PCI_REG(SH7751_PCILAR1));
-	outl(word, PCI_REG(SH7751_PCICONF6));
+	pci_write_reg(word, SH4_PCILAR1);
+	pci_write_reg(word, SH7751_PCICONF6);
 
-	/* Set the local 16MB PCI memory space window to 
+	/* Set the local 16MB PCI memory space window to
 	 * the lowest PCI mapped address
 	 */
-	word = PCIBIOS_MIN_MEM & SH7751_PCIMBR_MASK;
-	PCIDBG(2,"PCI: Setting upper bits of Memory window to 0x%x\n", word);
-	outl(word , PCI_REG(SH7751_PCIMBR));
+	word = PCIBIOS_MIN_MEM & SH4_PCIMBR_MASK;
+	pr_debug("PCI: Setting upper bits of Memory window to 0x%x\n", word);
+	pci_write_reg(word , SH4_PCIMBR);
 
 	/* Map IO space into PCI IO window
 	 * The IO window is 64K-PCIBIOS_MIN_IO in size
-	 * IO addresses will be translated to the 
+	 * IO addresses will be translated to the
 	 * PCI IO window base address
 	 */
-	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
-	    (64*1024), SH7751_PCI_IO_BASE+PCIBIOS_MIN_IO);
+	pr_debug("PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n",
+		 PCIBIOS_MIN_IO, (64 << 10),
+		 SH4_PCI_IO_BASE + PCIBIOS_MIN_IO);
 
-	/* 
+	/*
 	 * XXX: For now, leave this board-specific. In the event we have other
 	 * boards that need to do similar work, this can be wrapped.
 	 */
 #ifdef CONFIG_SH_BIGSUR
-	bigsur_port_map(PCIBIOS_MIN_IO, (64*1024), SH7751_PCI_IO_BASE+PCIBIOS_MIN_IO,0);
+	bigsur_port_map(PCIBIOS_MIN_IO, (64 << 10),
+			SH4_PCI_IO_BASE + PCIBIOS_MIN_IO, 0);
 #endif
 
-	/* Make sure the MSB's of IO window are set to access PCI space correctly */
-	word = PCIBIOS_MIN_IO & SH7751_PCIIOBR_MASK;
-	PCIDBG(2,"PCI: Setting upper bits of IO window to 0x%x\n", word);
-	outl(word, PCI_REG(SH7751_PCIIOBR));
-	
+	/* Make sure the MSB's of IO window are set to access PCI space
+	 * correctly */
+	word = PCIBIOS_MIN_IO & SH4_PCIIOBR_MASK;
+	pr_debug("PCI: Setting upper bits of IO window to 0x%x\n", word);
+	pci_write_reg(word, SH4_PCIIOBR);
+
 	/* Set PCI WCRx, BCRx's, copy from BSC locations */
 
 	/* check BCR for SDRAM in specified area */
@@ -349,13 +190,13 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 
 	/* configure the wait control registers */
 	word = inl(SH7751_WCR1);
-	outl(word, PCI_REG(SH7751_PCIWCR1));
+	pci_write_reg(word, SH4_PCIWCR1);
 	word = inl(SH7751_WCR2);
-	outl(word, PCI_REG(SH7751_PCIWCR2));
+	pci_write_reg(word, SH4_PCIWCR2);
 	word = inl(SH7751_WCR3);
-	outl(word, PCI_REG(SH7751_PCIWCR3));
+	pci_write_reg(word, SH4_PCIWCR3);
 	word = inl(SH7751_MCR);
-	outl(word, PCI_REG(SH7751_PCIMCR));
+	pci_write_reg(word, SH4_PCIMCR);
 
 	/* NOTE: I'm ignoring the PCI error IRQs for now..
 	 * TODO: add support for the internal error interrupts and
@@ -368,49 +209,8 @@ int __init sh7751_pcic_init(struct sh7751_pci_address_map *map)
 
 	/* SH7751 init done, set central function init complete */
 	/* use round robin mode to stop a device starving/overruning */
-	word = SH7751_PCICR_PREFIX | SH7751_PCICR_CFIN | SH7751_PCICR_ARBM;
-	outl(word,PCI_REG(SH7751_PCICR)); 
+	word = SH4_PCICR_PREFIX | SH4_PCICR_CFIN | SH4_PCICR_ARBM;
+	pci_write_reg(word, SH4_PCICR);
 
 	return 1;
 }
-
-char * __init pcibios_setup(char *str)
-{
-	if (!strcmp(str, "off")) {
-		pci_probe = 0;
-		return NULL;
-	}
-
-	return str;
-}
-
-/* 
- * 	IRQ functions 
- */
-static u8 __init sh7751_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh7751_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot,pin);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh7751_no_swizzle, sh7751_pci_lookup_irq);
-}
-
diff --git a/arch/sh/drivers/pci/pci-sh7751.h b/arch/sh/drivers/pci/pci-sh7751.h
index 1fee5cae10d1..68e3cb5e6bec 100644
--- a/arch/sh/drivers/pci/pci-sh7751.h
+++ b/arch/sh/drivers/pci/pci-sh7751.h
@@ -3,7 +3,7 @@
  *
  *  Dustin McIntire (dustin@sensoria.com) (c) 2001
  *  Paul Mundt (lethal@linux-sh.org) (c) 2003
- *	
+ *
  *  May be copied or modified under the terms of the GNU General Public
  *  License.  See linux/COPYING for more information.
  *
@@ -12,28 +12,6 @@
 #ifndef _PCI_SH7751_H_
 #define _PCI_SH7751_H_
 
-#include <linux/pci.h>
-
-/* set debug level 4=verbose...1=terse */
-//#define DEBUG_PCI 3
-#undef DEBUG_PCI
-
-#ifdef DEBUG_PCI
-#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
-#else
-#define PCIDBG(n, x...)
-#endif
-
-/* startup values */
-#define PCI_PROBE_BIOS 1
-#define PCI_PROBE_CONF1 2
-#define PCI_PROBE_CONF2 4
-#define PCI_NO_SORT 0x100
-#define PCI_BIOS_SORT 0x200
-#define PCI_NO_CHECKS 0x400
-#define PCI_ASSIGN_ROMS 0x1000
-#define PCI_BIOS_IRQ_SCAN 0x2000
-
 /* Platform Specific Values */
 #define SH7751_VENDOR_ID             0x1054
 #define SH7751_DEVICE_ID             0x3505
@@ -128,131 +106,6 @@
   #define SH7751_PCICONF17_PMEN      0x00010000  /* PME Enable */
   #define SH7751_PCICONF17_PWST      0x00000003  /* Power State */
 /* SH7715 Internal PCI Registers */
-#define SH7751_PCICR               0x100         /* PCI Control Register */
-  #define SH7751_PCICR_PREFIX        0xA5000000  /* CR prefix for write */
-  #define SH7751_PCICR_TRSB          0x00000200  /* Target Read Single */
-  #define SH7751_PCICR_BSWP          0x00000100  /* Target Byte Swap */
-  #define SH7751_PCICR_PLUP          0x00000080  /* Enable PCI Pullup */
-  #define SH7751_PCICR_ARBM          0x00000040  /* PCI Arbitration Mode */
-  #define SH7751_PCICR_MD            0x00000030  /* MD9 and MD10 status */
-  #define SH7751_PCICR_SERR          0x00000008  /* SERR output assert */
-  #define SH7751_PCICR_INTA          0x00000004  /* INTA output assert */
-  #define SH7751_PCICR_PRST          0x00000002  /* PCI Reset Assert */
-  #define SH7751_PCICR_CFIN          0x00000001  /* Central Fun. Init Done */
-#define SH7751_PCILSR0             0x104         /* PCI Local Space Register0 */
-#define SH7751_PCILSR1             0x108         /* PCI Local Space Register1 */
-#define SH7751_PCILAR0             0x10C         /* PCI Local Address Register1 */
-#define SH7751_PCILAR1             0x110         /* PCI Local Address Register1 */
-#define SH7751_PCIINT              0x114         /* PCI Interrupt Register */
-  #define SH7751_PCIINT_MLCK         0x00008000  /* Master Lock Error */
-  #define SH7751_PCIINT_TABT         0x00004000  /* Target Abort Error */
-  #define SH7751_PCIINT_TRET         0x00000200  /* Target Retry Error */
-  #define SH7751_PCIINT_MFDE         0x00000100  /* Master Func. Disable Error */
-  #define SH7751_PCIINT_PRTY         0x00000080  /* Address Parity Error */
-  #define SH7751_PCIINT_SERR         0x00000040  /* SERR Detection Error */
-  #define SH7751_PCIINT_TWDP         0x00000020  /* Tgt. Write Parity Error */
-  #define SH7751_PCIINT_TRDP         0x00000010  /* Tgt. Read Parity Error Det. */
-  #define SH7751_PCIINT_MTABT        0x00000008  /* Master-Tgt. Abort Error */
-  #define SH7751_PCIINT_MMABT        0x00000004  /* Master-Master Abort Error */
-  #define SH7751_PCIINT_MWPD         0x00000002  /* Master Write PERR Detect */
-  #define SH7751_PCIINT_MRPD         0x00000002  /* Master Read PERR Detect */
-#define SH7751_PCIINTM             0x118         /* PCI Interrupt Mask Register */
-#define SH7751_PCIALR              0x11C         /* Error Address Register */
-#define SH7751_PCICLR              0x120         /* Error Command/Data Register */
-  #define SH7751_PCICLR_MPIO         0x80000000  /* Error Command/Data Register */
-  #define SH7751_PCICLR_MDMA0        0x40000000  /* DMA0 Transfer Error */
-  #define SH7751_PCICLR_MDMA1        0x20000000  /* DMA1 Transfer Error */
-  #define SH7751_PCICLR_MDMA2        0x10000000  /* DMA2 Transfer Error */
-  #define SH7751_PCICLR_MDMA3        0x08000000  /* DMA3 Transfer Error */
-  #define SH7751_PCICLR_TGT          0x04000000  /* Target Transfer Error */
-  #define SH7751_PCICLR_CMDL         0x0000000F  /* PCI Command at Error */
-#define SH7751_PCIAINT             0x130         /* Arbiter Interrupt Register */
-  #define SH7751_PCIAINT_MBKN        0x00002000  /* Master Broken Interrupt */
-  #define SH7751_PCIAINT_TBTO        0x00001000  /* Target Bus Time Out */
-  #define SH7751_PCIAINT_MBTO        0x00001000  /* Master Bus Time Out */
-  #define SH7751_PCIAINT_TABT        0x00000008  /* Target Abort */
-  #define SH7751_PCIAINT_MABT        0x00000004  /* Master Abort */
-  #define SH7751_PCIAINT_RDPE        0x00000002  /* Read Data Parity Error */
-  #define SH7751_PCIAINT_WDPE        0x00000002  /* Write Data Parity Error */
-#define SH7751_PCIAINTM            0x134         /* Arbiter Int. Mask Register */
-#define SH7751_PCIBMLR             0x138         /* Error Bus Master Register */
-  #define SH7751_PCIBMLR_REQ4        0x00000010  /* REQ4 bus master at error */
-  #define SH7751_PCIBMLR_REQ3        0x00000008  /* REQ3 bus master at error */
-  #define SH7751_PCIBMLR_REQ2        0x00000004  /* REQ2 bus master at error */
-  #define SH7751_PCIBMLR_REQ1        0x00000002  /* REQ1 bus master at error */
-  #define SH7751_PCIBMLR_REQ0        0x00000001  /* REQ0 bus master at error */
-#define SH7751_PCIDMABT            0x140         /* DMA Transfer Arb. Register */
-  #define SH7751_PCIDMABT_RRBN       0x00000001  /* DMA Arbitor Round-Robin */
-#define SH7751_PCIDPA0             0x180         /* DMA0 Transfer Addr. Register */
-#define SH7751_PCIDLA0             0x184         /* DMA0 Local Addr. Register */
-#define SH7751_PCIDTC0             0x188         /* DMA0 Transfer Cnt. Register */
-#define SH7751_PCIDCR0             0x18C         /* DMA0 Control Register */
-  #define SH7751_PCIDCR_ALGN         0x00000600  /* DMA Alignment Mode */
-  #define SH7751_PCIDCR_MAST         0x00000100  /* DMA Termination Type */
-  #define SH7751_PCIDCR_INTM         0x00000080  /* DMA Interrupt Done Mask*/
-  #define SH7751_PCIDCR_INTS         0x00000040  /* DMA Interrupt Done Status */
-  #define SH7751_PCIDCR_LHLD         0x00000020  /* Local Address Control */
-  #define SH7751_PCIDCR_PHLD         0x00000010  /* PCI Address Control*/
-  #define SH7751_PCIDCR_IOSEL        0x00000008  /* PCI Address Space Type */
-  #define SH7751_PCIDCR_DIR          0x00000004  /* DMA Transfer Direction */
-  #define SH7751_PCIDCR_STOP         0x00000002  /* Force DMA Stop */
-  #define SH7751_PCIDCR_STRT         0x00000001  /* DMA Start */
-#define SH7751_PCIDPA1             0x190         /* DMA1 Transfer Addr. Register */
-#define SH7751_PCIDLA1             0x194         /* DMA1 Local Addr. Register */
-#define SH7751_PCIDTC1             0x198         /* DMA1 Transfer Cnt. Register */
-#define SH7751_PCIDCR1             0x19C         /* DMA1 Control Register */
-#define SH7751_PCIDPA2             0x1A0         /* DMA2 Transfer Addr. Register */
-#define SH7751_PCIDLA2             0x1A4         /* DMA2 Local Addr. Register */
-#define SH7751_PCIDTC2             0x1A8         /* DMA2 Transfer Cnt. Register */
-#define SH7751_PCIDCR2             0x1AC         /* DMA2 Control Register */
-#define SH7751_PCIDPA3             0x1B0         /* DMA3 Transfer Addr. Register */
-#define SH7751_PCIDLA3             0x1B4         /* DMA3 Local Addr. Register */
-#define SH7751_PCIDTC3             0x1B8         /* DMA3 Transfer Cnt. Register */
-#define SH7751_PCIDCR3             0x1BC         /* DMA3 Control Register */
-#define SH7751_PCIPAR              0x1C0         /* PIO Address Register */
-  #define SH7751_PCIPAR_CFGEN        0x80000000  /* Configuration Enable */
-  #define SH7751_PCIPAR_BUSNO        0x00FF0000  /* Config. Bus Number */
-  #define SH7751_PCIPAR_DEVNO        0x0000FF00  /* Config. Device Number */
-  #define SH7751_PCIPAR_REGAD        0x000000FC  /* Register Address Number */
-#define SH7751_PCIMBR              0x1C4         /* Memory Base Address Register */
-  #define SH7751_PCIMBR_MASK         0xFF000000  /* Memory Space Mask */
-  #define SH7751_PCIMBR_LOCK         0x00000001  /* Lock Memory Space */
-#define SH7751_PCIIOBR             0x1C8         /* I/O Base Address Register */
-  #define SH7751_PCIIOBR_MASK         0xFFFC0000 /* IO Space Mask */
-  #define SH7751_PCIIOBR_LOCK         0x00000001 /* Lock IO Space */
-#define SH7751_PCIPINT             0x1CC         /* Power Mgmnt Int. Register */
-  #define SH7751_PCIPINT_D3           0x00000002 /* D3 Pwr Mgmt. Interrupt */
-  #define SH7751_PCIPINT_D0           0x00000001 /* D0 Pwr Mgmt. Interrupt */  
-#define SH7751_PCIPINTM            0x1D0         /* Power Mgmnt Mask Register */
-#define SH7751_PCICLKR             0x1D4         /* Clock Ctrl. Register */
-  #define SH7751_PCICLKR_PCSTP        0x00000002 /* PCI Clock Stop */
-  #define SH7751_PCICLKR_BCSTP        0x00000002 /* BCLK Clock Stop */
-/* For definitions of BCR, MCR see ... */
-#define SH7751_PCIBCR1             0x1E0         /* Memory BCR1 Register */
-#define SH7751_PCIBCR2             0x1E4         /* Memory BCR2 Register */
-#define SH7751_PCIWCR1             0x1E8         /* Wait Control 1 Register */
-#define SH7751_PCIWCR2             0x1EC         /* Wait Control 2 Register */
-#define SH7751_PCIWCR3             0x1F0         /* Wait Control 3 Register */
-#define SH7751_PCIMCR              0x1F4         /* Memory Control Register */
-#define SH7751_PCIBCR3		   0x1f8	 /* Memory BCR3 Register */
-#define SH7751_PCIPCTR             0x200         /* Port Control Register */
-  #define SH7751_PCIPCTR_P2EN        0x000400000 /* Port 2 Enable */
-  #define SH7751_PCIPCTR_P1EN        0x000200000 /* Port 1 Enable */
-  #define SH7751_PCIPCTR_P0EN        0x000100000 /* Port 0 Enable */
-  #define SH7751_PCIPCTR_P2UP        0x000000020 /* Port2 Pull Up Enable */
-  #define SH7751_PCIPCTR_P2IO        0x000000010 /* Port2 Output Enable */
-  #define SH7751_PCIPCTR_P1UP        0x000000008 /* Port1 Pull Up Enable */
-  #define SH7751_PCIPCTR_P1IO        0x000000004 /* Port1 Output Enable */
-  #define SH7751_PCIPCTR_P0UP        0x000000002 /* Port0 Pull Up Enable */
-  #define SH7751_PCIPCTR_P0IO        0x000000001 /* Port0 Output Enable */
-#define SH7751_PCIPDTR             0x204         /* Port Data Register */
-  #define SH7751_PCIPDTR_PB5         0x000000020 /* Port 5 Enable */
-  #define SH7751_PCIPDTR_PB4         0x000000010 /* Port 4 Enable */
-  #define SH7751_PCIPDTR_PB3         0x000000008 /* Port 3 Enable */
-  #define SH7751_PCIPDTR_PB2         0x000000004 /* Port 2 Enable */
-  #define SH7751_PCIPDTR_PB1         0x000000002 /* Port 1 Enable */
-  #define SH7751_PCIPDTR_PB0         0x000000001 /* Port 0 Enable */
-#define SH7751_PCIPDR              0x220         /* Port IO Data Register */
 
 /* Memory Control Registers */
 #define SH7751_BCR1                0xFF800000    /* Memory BCR1 Register */
@@ -274,30 +127,9 @@
 #define SH7751_CS5_BASE_ADDR       (SH7751_CS4_BASE_ADDR + SH7751_MEM_REGION_SIZE)
 #define SH7751_CS6_BASE_ADDR       (SH7751_CS5_BASE_ADDR + SH7751_MEM_REGION_SIZE)
 
-/* General PCI values */
-#define SH7751_PCI_HOST_BRIDGE		0x6
-
-/* Flags */
-#define SH7751_PCIC_NO_RESET	0x0001
-
-/* External functions defined per platform i.e. Big Sur, SE... (these could be routed 
- * through the machine vectors... */
-extern int pcibios_init_platform(void);
-extern int pcibios_map_platform_irq(u8 slot, u8 pin);
-
-struct sh7751_pci_address_space {
-	unsigned long base;
-	unsigned long size;
-};
-
-struct sh7751_pci_address_map {
-	struct sh7751_pci_address_space window0;
-	struct sh7751_pci_address_space window1;
-	unsigned long flags;
-};
+struct sh4_pci_address_map;
 
 /* arch/sh/drivers/pci/pci-sh7751.c */
-extern int sh7751_pcic_init(struct sh7751_pci_address_map *map);
+int sh7751_pcic_init(struct sh4_pci_address_map *map);
 
 #endif /* _PCI_SH7751_H_ */
-
diff --git a/arch/sh/drivers/pci/pci-sh7780.c b/arch/sh/drivers/pci/pci-sh7780.c
index e09721330ac2..bd3064a82087 100644
--- a/arch/sh/drivers/pci/pci-sh7780.c
+++ b/arch/sh/drivers/pci/pci-sh7780.c
@@ -20,197 +20,36 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/pci.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
 #include <linux/errno.h>
-#include <linux/irq.h>
 #include <linux/delay.h>
-
-#include <asm/machvec.h>
-#include <asm/io.h>
-#include "pci-sh7780.h"
-
-static unsigned int pci_probe = PCI_PROBE_CONF1;
-extern int pci_fixup_pcic(void);
-
-/*
- * Direct access to PCI hardware...
- */
-
-#define CONFIG_CMD(bus, devfn, where) (0x80000000 | (bus->number << 16) | (devfn << 8) | (where & ~3))
-
-/*
- * Functions for accessing PCI configuration space with type 1 accesses
- */
-static int sh7780_pci_read(struct pci_bus *bus, unsigned int devfn,
-			   int where, int size, u32 *val)
-{
-	unsigned long flags;
-	u32 data;
-
-	/*
-	 * PCIPDR may only be accessed as 32 bit words,
-	 * so we must do byte alignment by hand
-	 */
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
-	data = inl(PCI_REG(SH7780_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-	case 2:
-		*val = (data >> ((where & 2) << 3)) & 0xffff;
-		break;
-	case 4:
-		*val = data;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	return PCIBIOS_SUCCESSFUL;
-}
+#include "pci-sh4.h"
 
 /*
- * Since SH7780 only does 32bit access we'll have to do a read,
- * mask,write operation.
- * We'll allow an odd byte offset, though it should be illegal.
+ * Initialization. Try all known PCI access methods. Note that we support
+ * using both PCI BIOS and direct access: in such cases, we use I/O ports
+ * to access config space.
+ *
+ * Note that the platform specific initialization (BSC registers, and memory
+ * space mapping) will be called via the platform defined function
+ * pcibios_init_platform().
  */
-static int sh7780_pci_write(struct pci_bus *bus, unsigned int devfn,
-			    int where, int size, u32 val)
+static int __init sh7780_pci_init(void)
 {
-	unsigned long flags;
-	int shift;
-	u32 data;
-
-	local_irq_save(flags);
-	outl(CONFIG_CMD(bus, devfn, where), PCI_REG(SH7780_PCIPAR));
-	data = inl(PCI_REG(SH7780_PCIPDR));
-	local_irq_restore(flags);
-
-	switch (size) {
-	case 1:
-		shift = (where & 3) << 3;
-		data &= ~(0xff << shift);
-		data |= ((val & 0xff) << shift);
-		break;
-	case 2:
-		shift = (where & 2) << 3;
-		data &= ~(0xffff << shift);
-		data |= ((val & 0xffff) << shift);
-		break;
-	case 4:
-		data = val;
-		break;
-	default:
-		return PCIBIOS_FUNC_NOT_SUPPORTED;
-	}
-
-	outl(data, PCI_REG(SH7780_PCIPDR));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-#undef CONFIG_CMD
-
-struct pci_ops sh7780_pci_ops = {
-	.read		= sh7780_pci_read,
-	.write		= sh7780_pci_write,
-};
+	unsigned int id;
+	int ret;
 
-static int __init pci_check_direct(void)
-{
-	unsigned int tmp, id;
+	pr_debug("PCI: Starting intialization.\n");
 
 	outl(0x00000001, SH7780_PCI_VCR2); /* Enable PCIC */
 
 	/* check for SH7780/SH7780R hardware */
-	id = inl(PCI_REG(SH7780_PCIVID));
+	id = pci_read_reg(SH7780_PCIVID);
 	if ((id != ((SH7780_DEVICE_ID << 16) | SH7780_VENDOR_ID)) &&
 	    (id != ((SH7781_DEVICE_ID << 16) | SH7780_VENDOR_ID))) {
 		printk(KERN_ERR "PCI: This is not an SH7780 (%x)\n", id);
 		return -ENODEV;
 	}
 
-	/*
-	 * Check if configuration works.
-	 */
-	if (pci_probe & PCI_PROBE_CONF1) {
-		tmp = inl(PCI_REG(SH7780_PCIPAR));
-		outl(0x80000000, PCI_REG(SH7780_PCIPAR));
-		if (inl(PCI_REG(SH7780_PCIPAR)) == 0x80000000) {
-			outl(tmp, PCI_REG(SH7780_PCIPAR));
-			printk(KERN_INFO "PCI: Using configuration type 1\n");
-			request_region(PCI_REG(SH7780_PCIPAR), 8, "PCI conf1");
-			return 0;
-		}
-		outl(tmp, PCI_REG(SH7780_PCIPAR));
-	}
-
-	pr_debug("PCI: pci_check_direct failed\n");
-	return -EINVAL;
-}
-
-/***************************************************************************************/
-
-/*
- *  Handle bus scanning and fixups ....
- */
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	pr_debug("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-/*
- *  Called after each bus is probed, but before its children
- *  are examined.
- */
-
-void __init pcibios_fixup_bus(struct pci_bus *b)
-{
-	pci_read_bridge_bases(b);
-}
-
-/*
- * Initialization. Try all known PCI access methods. Note that we support
- * using both PCI BIOS and direct access: in such cases, we use I/O ports
- * to access config space.
- *
- * Note that the platform specific initialization (BSC registers, and memory
- * space mapping) will be called via the machine vectors (sh_mv.mv_pci_init()) if it
- * exists and via the platform defined function pcibios_init_platform().
- * See pci_bigsur.c for implementation;
- *
- * The BIOS version of the pci functions is not yet implemented but it is left
- * in for completeness.  Currently an error will be genereated at compile time.
- */
-
-static int __init sh7780_pci_init(void)
-{
-	int ret;
-
-	pr_debug("PCI: Starting intialization.\n");
-
 	/* Setup the INTC */
 	ctrl_outl(0x00200000, INTC_ICR0);	/* INTC SH-4 Mode */
 	ctrl_outl(0x00078000, INTC_INT2MSKCR);	/* enable PCIINTA - PCIINTD */
@@ -219,15 +58,14 @@ static int __init sh7780_pci_init(void)
 	ctrl_outl(0x80000000, INTC_INTMSKCLR1);	/* enable IRL0-3 Interrupt */
 	ctrl_outl(0xfffe0000, INTC_INTMSKCLR2);	/* enable IRL0-3 Interrupt */
 
-	if ((ret = pci_check_direct()) != 0)
+	if ((ret = sh4_pci_check_direct()) != 0)
 		return ret;
 
 	return pcibios_init_platform();
 }
-
 core_initcall(sh7780_pci_init);
 
-int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
+int __init sh7780_pcic_init(struct sh4_pci_address_map *map)
 {
 	u32 word;
 
@@ -236,25 +74,25 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 	 * bootloader and doing it here means the MAC addresses loaded
 	 * by the bootloader get lost.
 	 */
-	if (!(map->flags & SH7780_PCIC_NO_RESET)) {
+	if (!(map->flags & SH4_PCIC_NO_RESET)) {
 		/* toggle PCI reset pin */
-		word = SH7780_PCICR_PREFIX | SH7780_PCICR_PRST;
-		outl(word,PCI_REG(SH7780_PCICR));
+		word = SH4_PCICR_PREFIX | SH4_PCICR_PRST;
+		pci_write_reg(word, SH4_PCICR);
 		/* Wait for a long time... not 1 sec. but long enough */
 		mdelay(100);
-		word = SH7780_PCICR_PREFIX;
-		outl(word,PCI_REG(SH7780_PCICR));
+		word = SH4_PCICR_PREFIX;
+		pci_write_reg(word, SH4_PCICR);
 	}
 
 	/* set the command/status bits to:
 	 * Wait Cycle Control + Parity Enable + Bus Master +
 	 * Mem space enable
 	 */
-	outl(0x00000046, PCI_REG(SH7780_PCICMD));
+	pci_write_reg(0x00000046, SH7780_PCICMD);
 
 	/* define this host as the host bridge */
-	word = SH7780_PCI_HOST_BRIDGE << 24;
-	outl(word, PCI_REG(SH7780_PCIRID));
+	word = PCI_BASE_CLASS_BRIDGE << 24;
+	pci_write_reg(word, SH7780_PCIRID);
 
 	/* Set IO and Mem windows to local address
 	 * Make PCI and local address the same for easy 1 to 1 mapping
@@ -262,25 +100,26 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 	 * Window1 = map->window1.size @ cached area base = SDRAM
 	 */
 	word = ((map->window0.size - 1) & 0x1ff00001) | 0x01;
-	outl(0x07f00001, PCI_REG(SH7780_PCILSR0));
+	pci_write_reg(0x07f00001, SH4_PCILSR0);
 	word = ((map->window1.size - 1) & 0x1ff00001) | 0x01;
-	outl(0x00000001, PCI_REG(SH7780_PCILSR1));
+	pci_write_reg(0x00000001, SH4_PCILSR1);
 	/* Set the values on window 0 PCI config registers */
 	word = P2SEGADDR(map->window0.base);
-	outl(0xa8000000, PCI_REG(SH7780_PCILAR0));
-	outl(0x08000000, PCI_REG(SH7780_PCIMBAR0));
+	pci_write_reg(0xa8000000, SH4_PCILAR0);
+	pci_write_reg(0x08000000, SH7780_PCIMBAR0);
 	/* Set the values on window 1 PCI config registers */
 	word = P2SEGADDR(map->window1.base);
-	outl(0x00000000, PCI_REG(SH7780_PCILAR1));
-	outl(0x00000000, PCI_REG(SH7780_PCIMBAR1));
+	pci_write_reg(0x00000000, SH4_PCILAR1);
+	pci_write_reg(0x00000000, SH7780_PCIMBAR1);
 
 	/* Map IO space into PCI IO window
 	 * The IO window is 64K-PCIBIOS_MIN_IO in size
 	 * IO addresses will be translated to the
 	 * PCI IO window base address
 	 */
-	PCIDBG(3,"PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n", PCIBIOS_MIN_IO,
-	    (64*1024), SH7780_PCI_IO_BASE+PCIBIOS_MIN_IO);
+	pr_debug("PCI: Mapping IO address 0x%x - 0x%x to base 0x%x\n",
+		 PCIBIOS_MIN_IO, (64 << 10),
+		 SH7780_PCI_IO_BASE + PCIBIOS_MIN_IO);
 
 	/* NOTE: I'm ignoring the PCI error IRQs for now..
 	 * TODO: add support for the internal error interrupts and
@@ -293,49 +132,8 @@ int __init sh7780_pcic_init(struct sh7780_pci_address_map *map)
 
 	/* SH7780 init done, set central function init complete */
 	/* use round robin mode to stop a device starving/overruning */
-	word = SH7780_PCICR_PREFIX | SH7780_PCICR_CFIN | /* SH7780_PCICR_ARBM |*/ SH7780_PCICR_FTO;
-	outl(word, PCI_REG(SH7780_PCICR));
+	word = SH4_PCICR_PREFIX | SH4_PCICR_CFIN | SH4_PCICR_FTO;
+	pci_write_reg(word, SH4_PCICR);
 
 	return 1;
 }
-
-char * __init pcibios_setup(char *str)
-{
-	if (!strcmp(str, "off")) {
-		pci_probe = 0;
-		return NULL;
-	}
-
-	return str;
-}
-
-/*
- *	IRQ functions
- */
-static u8 __init sh7780_no_swizzle(struct pci_dev *dev, u8 *pin)
-{
-	/* no swizzling */
-	return PCI_SLOT(dev->devfn);
-}
-
-static int sh7780_pci_lookup_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	int irq = -1;
-
-	/* now lookup the actual IRQ on a platform specific basis (pci-'platform'.c) */
-	irq = pcibios_map_platform_irq(slot,pin);
-	if( irq < 0 ) {
-		pr_debug("PCI: Error mapping IRQ on device %s\n", pci_name(dev));
-		return irq;
-	}
-
-	pr_debug("Setting IRQ for slot %s to %d\n", pci_name(dev), irq);
-
-	return irq;
-}
-
-void __init pcibios_fixup_irqs(void)
-{
-	pci_fixup_irqs(sh7780_no_swizzle, sh7780_pci_lookup_irq);
-}
-
diff --git a/arch/sh/drivers/pci/pci-sh7780.h b/arch/sh/drivers/pci/pci-sh7780.h
index 750d5d7753a2..f02d2180a4bc 100644
--- a/arch/sh/drivers/pci/pci-sh7780.h
+++ b/arch/sh/drivers/pci/pci-sh7780.h
@@ -12,28 +12,6 @@
 #ifndef _PCI_SH7780_H_
 #define _PCI_SH7780_H_
 
-#include <linux/pci.h>
-
-/* set debug level 4=verbose...1=terse */
-//#define DEBUG_PCI 3
-#undef DEBUG_PCI
-
-#ifdef DEBUG_PCI
-#define PCIDBG(n, x...) { if(DEBUG_PCI>=n) printk(x); }
-#else
-#define PCIDBG(n, x...)
-#endif
-
-/* startup values */
-#define PCI_PROBE_BIOS		1
-#define PCI_PROBE_CONF1		2
-#define PCI_PROBE_CONF2		4
-#define PCI_NO_SORT		0x100
-#define PCI_BIOS_SORT		0x200
-#define PCI_NO_CHECKS		0x400
-#define PCI_ASSIGN_ROMS		0x1000
-#define PCI_BIOS_IRQ_SCAN	0x2000
-
 /* Platform Specific Values */
 #define SH7780_VENDOR_ID	0x1912
 #define SH7780_DEVICE_ID	0x0002
@@ -47,15 +25,12 @@
 /* SH7780 Specific Values */
 #define SH7780_PCI_CONFIG_BASE	0xFD000000	/* Config space base addr */
 #define SH7780_PCI_CONFIG_SIZE	0x01000000	/* Config space size */
+
 #define SH7780_PCI_MEMORY_BASE	0xFD000000	/* Memory space base addr */
 #define SH7780_PCI_MEM_SIZE	0x01000000	/* Size of Memory window */
-#if 1
+
 #define SH7780_PCI_IO_BASE	0xFE400000	/* IO space base address */
 #define SH7780_PCI_IO_SIZE	0x00400000	/* Size of IO window */
-#else
-#define SH7780_PCI_IO_BASE	0xFE200000	/* IO space base address */
-#define SH7780_PCI_IO_SIZE	0x00200000	/* Size of IO window */
-#endif
 
 #define SH7780_PCIREG_BASE	0xFE040000	/* PCI regs base address */
 #define PCI_REG(n)		(SH7780_PCIREG_BASE+n)
@@ -90,44 +65,16 @@
 #define SH7780_PCIPMCSR_BSE	0x046
 #define SH7780_PCICDD		0x047
 
-/* SH7780 PCI Local Registers */
-#define SH7780_PCICR		0x100		/* PCI Control Register */
-  #define SH7780_PCICR_PREFIX	0xA5000000	/* CR prefix for write */
-  #define SH7780_PCICR_PFCS	0x00000800	/* TRDY/IRDY Enable */
-  #define SH7780_PCICR_FTO	0x00000400	/* TRDY/IRDY Enable */
-  #define SH7780_PCICR_PFE	0x00000200	/* Target Read Single */
-  #define SH7780_PCICR_TBS	0x00000100	/* Target Byte Swap */
-  #define SH7780_PCICR_ARBM	0x00000040	/* PCI Arbitration Mode */
-  #define SH7780_PCICR_IOCS	0x00000004	/* INTA output assert */
-  #define SH7780_PCICR_PRST	0x00000002	/* PCI Reset Assert */
-  #define SH7780_PCICR_CFIN	0x00000001	/* Central Fun. Init Done */
-#define SH7780_PCILSR0		0x104		/* PCI Local Space Register0 */
-#define SH7780_PCILSR1		0x108		/* PCI Local Space Register1 */
-#define SH7780_PCILAR0		0x10C	/* PCI Local Address Register1 */
-#define SH7780_PCILAR1		0x110	/* PCI Local Address Register1 */
-#define SH7780_PCIIR		0x114		/* PCI Interrupt Register */
-#define SH7780_PCIIMR		0x118	/* PCI Interrupt Mask Register */
-#define SH7780_PCIAIR		0x11C		/* Error Address Register */
-#define SH7780_PCICIR		0x120	/* Error Command/Data Register */
-#define SH7780_PCIAINT		0x130		/* Arbiter Interrupt Register */
-#define SH7780_PCIAINTM		0x134		/* Arbiter Int. Mask Register */
-#define SH7780_PCIBMIR		0x138		/* Error Bus Master Register */
-#define SH7780_PCIPAR		0x1C0		/* PIO Address Register */
-#define SH7780_PCIPINT		0x1CC	/* Power Management Int. Register */
-#define SH7780_PCIPINTM		0x1D0	/* Power Management Mask Register */
-#define SH7780_PCIMBR0		0x1E0		/* Memory Bank0 Register */
-#define SH7780_PCIMBMR0		0x1E4		/* Memory Bank0 Mask Register */
-#define SH7780_PCIMBR1		0x1E8		/* Memory Bank1  Register */
-#define SH7780_PCIMBMR1		0x1EC		/* Memory Bank1 Mask Register */
-#define SH7780_PCIMBR2		0x1F0		/* Memory Bank2 Register */
-#define SH7780_PCIMBMR2		0x1F4		/* Memory Bank2 Mask Register */
-#define SH7780_PCIIOBR		0x1F8		/* Bank Register */
-#define SH7780_PCIIOBMR		0x1FC		/* Bank Mask Register */
+#define SH7780_PCIMBR0		0x1E0
+#define SH7780_PCIMBMR0		0x1E4
+#define SH7780_PCIMBR2		0x1F0
+#define SH7780_PCIMBMR2		0x1F4
+#define SH7780_PCIIOBR		0x1F8
+#define SH7780_PCIIOBMR		0x1FC
 #define SH7780_PCICSCR0		0x210		/* Cache Snoop1 Cnt. Register */
 #define SH7780_PCICSCR1		0x214		/* Cache Snoop2 Cnt. Register */
 #define SH7780_PCICSAR0		0x218	/* Cache Snoop1 Addr. Register */
 #define SH7780_PCICSAR1		0x21C	/* Cache Snoop2 Addr. Register */
-#define SH7780_PCIPDR		0x220		/* Port IO Data Register */
 
 /* General Memory Config Addresses */
 #define SH7780_CS0_BASE_ADDR	0x0
@@ -139,30 +86,9 @@
 #define SH7780_CS5_BASE_ADDR	(SH7780_CS4_BASE_ADDR + SH7780_MEM_REGION_SIZE)
 #define SH7780_CS6_BASE_ADDR	(SH7780_CS5_BASE_ADDR + SH7780_MEM_REGION_SIZE)
 
-/* General PCI values */
-#define SH7780_PCI_HOST_BRIDGE	0x6
-
-/* Flags */
-#define SH7780_PCIC_NO_RESET	0x0001
-
-/* External functions defined per platform i.e. Big Sur, SE... (these could be routed
- * through the machine vectors... */
-extern int pcibios_init_platform(void);
-extern int pcibios_map_platform_irq(u8 slot, u8 pin);
-
-struct sh7780_pci_address_space {
-	unsigned long base;
-	unsigned long size;
-};
-
-struct sh7780_pci_address_map {
-	struct sh7780_pci_address_space window0;
-	struct sh7780_pci_address_space window1;
-	unsigned long flags;
-};
+struct sh4_pci_address_map;
 
 /* arch/sh/drivers/pci/pci-sh7780.c */
-extern int sh7780_pcic_init(struct sh7780_pci_address_map *map);
+int sh7780_pcic_init(struct sh4_pci_address_map *map);
 
 #endif /* _PCI_SH7780_H_ */
-
diff --git a/arch/sh/drivers/pci/pci-st40.c b/arch/sh/drivers/pci/pci-st40.c
index 7c81b8b65bb5..4ab5ea6b35fb 100644
--- a/arch/sh/drivers/pci/pci-st40.c
+++ b/arch/sh/drivers/pci/pci-st40.c
@@ -70,12 +70,6 @@
 static void pci_set_rbar_region(unsigned int region,     unsigned long localAddr,
 			 unsigned long pciOffset, unsigned long regionSize);
 
-/*
- * The pcibios_map_platform_irq function is defined in the appropriate
- * board specific code and referenced here
- */
-extern int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin);
-
 static __init void SetPCIPLL(void)
 {
 	{
@@ -422,13 +416,6 @@ struct pci_ops st40pci_config_ops = {
 /* Everything hangs off this */
 static struct pci_bus *pci_root_bus;
 
-
-static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
-{
-	return PCI_SLOT(dev->devfn);
-}
-
-
 static int __init pcibios_init(void)
 {
 	extern unsigned long memory_start, memory_end;
@@ -465,17 +452,11 @@ static int __init pcibios_init(void)
 	/* ok, do the scan man */
 	pci_root_bus = pci_scan_bus(0, &st40pci_config_ops, NULL);
 	pci_assign_unassigned_resources();
-	pci_fixup_irqs(no_swizzle, pcibios_map_platform_irq);
 
 	return 0;
 }
-
 subsys_initcall(pcibios_init);
 
-void __init pcibios_fixup_bus(struct pci_bus *bus)
-{
-}
-
 /*
  * Publish a region of local address space over the PCI bus
  * to other devices.
diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 285dffd12bd8..7377a8a8e161 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -21,6 +21,26 @@
 #include <linux/init.h>
 #include <asm/io.h>
 
+static inline u8 bridge_swizzle(u8 pin, u8 slot)
+{
+	return (((pin - 1) + slot) % 4) + 1;
+}
+
+static u8 __init simple_swizzle(struct pci_dev *dev, u8 *pinp)
+{
+	u8 pin = *pinp;
+
+	while (dev->bus->parent) {
+		pin = bridge_swizzle(pin, PCI_SLOT(dev->devfn));
+		/* Move up the chain of bridges. */
+		dev = dev->bus->self;
+	}
+	*pinp = pin;
+
+	/* The slot is the slot of the last bridge. */
+	return PCI_SLOT(dev->devfn);
+}
+
 static int __init pcibios_init(void)
 {
 	struct pci_channel *p;
@@ -36,19 +56,26 @@ static int __init pcibios_init(void)
 
 	/* scan the buses */
 	busno = 0;
-	for (p= board_pci_channels; p->pci_ops != NULL; p++) {
+	for (p = board_pci_channels; p->pci_ops != NULL; p++) {
 		bus = pci_scan_bus(busno, p->pci_ops, p);
-		busno = bus->subordinate+1;
+		busno = bus->subordinate + 1;
 	}
 
-	/* board-specific fixups */
-	pcibios_fixup_irqs();
+	pci_fixup_irqs(simple_swizzle, pcibios_map_platform_irq);
 
 	return 0;
 }
-
 subsys_initcall(pcibios_init);
 
+/*
+ *  Called after each bus is probed, but before its children
+ *  are examined.
+ */
+void __init pcibios_fixup_bus(struct pci_bus *bus)
+{
+	pci_read_bridge_bases(bus);
+}
+
 void
 pcibios_update_resource(struct pci_dev *dev, struct resource *root,
 			struct resource *res, int resource)
@@ -192,11 +219,10 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 
 	return NULL;
 }
+EXPORT_SYMBOL(pci_iomap);
 
 void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
 {
 	iounmap(addr);
 }
-
-EXPORT_SYMBOL(pci_iomap);
 EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 3d5cafc71ae3..9611fab03b55 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -7,6 +7,9 @@ obj-y	:= ex.o probe.o
 obj-$(CONFIG_SH_FPU)                    += fpu.o
 obj-$(CONFIG_SH_STORE_QUEUES)		+= sq.o
 
+# CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
+
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH4)			:= clock-sh4.o
 clock-$(CONFIG_CPU_SUBTYPE_SH73180)	:= clock-sh73180.o
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7780.c b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
new file mode 100644
index 000000000000..72493f259edc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7780.c
@@ -0,0 +1,79 @@
+/*
+ * SH7780 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct resource rtc_resources[] = {
+	[0] = {
+		.start	= 0xffe80000,
+		.end	= 0xffe80000 + 0x58 - 1,
+		.flags	= IORESOURCE_IO,
+	},
+	[1] = {
+		/* Period IRQ */
+		.start	= 21,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[2] = {
+		/* Carry IRQ */
+		.start	= 22,
+		.flags	= IORESOURCE_IRQ,
+	},
+	[3] = {
+		/* Alarm IRQ */
+		.start	= 23,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device rtc_device = {
+	.name		= "sh-rtc",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(rtc_resources),
+	.resource	= rtc_resources,
+};
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.mapbase	= 0xffe10000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7780_devices[] __initdata = {
+	&rtc_device,
+	&sci_device,
+};
+
+static int __init sh7780_devices_setup(void)
+{
+	return platform_add_devices(sh7780_devices,
+				    ARRAY_SIZE(sh7780_devices));
+}
+__initcall(sh7780_devices_setup);
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 894e64b2d5f0..27dba653cbe7 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -209,6 +209,11 @@ static inline void ctrl_outl(unsigned int b, unsigned long addr)
         *(volatile unsigned long*)addr = b;
 }
 
+static inline void ctrl_delay(void)
+{
+	ctrl_inw(P2SEG);
+}
+
 #define IO_SPACE_LIMIT 0xffffffff
 
 /*
diff --git a/include/asm-sh/pci.h b/include/asm-sh/pci.h
index 18a109de0f2e..6ccc948fe216 100644
--- a/include/asm-sh/pci.h
+++ b/include/asm-sh/pci.h
@@ -32,6 +32,34 @@ extern struct pci_channel board_pci_channels[];
 #define PCIBIOS_MIN_IO		board_pci_channels->io_resource->start
 #define PCIBIOS_MIN_MEM		board_pci_channels->mem_resource->start
 
+/*
+ * I/O routine helpers
+ */
+#ifdef CONFIG_CPU_SUBTYPE_SH7780
+#define PCI_IO_AREA		0xFE400000
+#define PCI_IO_SIZE		0x00400000
+#else
+#define PCI_IO_AREA		0xFE240000
+#define PCI_IO_SIZE		0X00040000
+#endif
+
+#define PCI_MEM_SIZE		0x01000000
+
+#define SH4_PCIIOBR_MASK	0xFFFC0000
+#define pci_ioaddr(addr)	(PCI_IO_AREA + (addr & ~SH4_PCIIOBR_MASK))
+
+#if defined(CONFIG_PCI)
+#define is_pci_ioaddr(port)		\
+	(((port) >= PCIBIOS_MIN_IO) &&	\
+	 ((port) < (PCIBIOS_MIN_IO + PCI_IO_SIZE)))
+#define is_pci_memaddr(port)		\
+	(((port) >= PCIBIOS_MIN_MEM) &&	\
+	 ((port) < (PCIBIOS_MIN_MEM + PCI_MEM_SIZE)))
+#else
+#define is_pci_ioaddr(port)	(0)
+#define is_pci_memaddr(port)	(0)
+#endif
+
 struct pci_dev;
 
 extern void pcibios_set_master(struct pci_dev *dev);
@@ -98,11 +126,12 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 #endif
 
 /* Board-specific fixup routines. */
-extern void pcibios_fixup(void);
-extern void pcibios_fixup_irqs(void);
+void pcibios_fixup(void);
+int pcibios_init_platform(void);
+int pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin);
 
 #ifdef CONFIG_PCI_AUTO
-extern int pciauto_assign_resources(int busno, struct pci_channel *hose);
+int pciauto_assign_resources(int busno, struct pci_channel *hose);
 #endif
 
 static inline void pcibios_add_platform_entries(struct pci_dev *dev)
-- 
cgit v1.2.1


From a3e61d50dc82475ebca3ff8b18c174c02c5ff511 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:45:22 +0900
Subject: sh: Inhibit mapping PCI apertures through page tables.

Inhibit mapping through page tables in __ioremap() for PCI memory
apertures on SH7751 and SH7780-style PCI controllers, translation is
not possible for these areas. For other users that map a small window
in P1/P2 space, ioremap() traps that already, and should never make
it to __ioremap().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/drivers/pci/pci.c | 20 +++++++++-----------
 arch/sh/mm/ioremap.c      | 17 ++++++++++++++++-
 2 files changed, 25 insertions(+), 12 deletions(-)

diff --git a/arch/sh/drivers/pci/pci.c b/arch/sh/drivers/pci/pci.c
index 7377a8a8e161..d439336d2e18 100644
--- a/arch/sh/drivers/pci/pci.c
+++ b/arch/sh/drivers/pci/pci.c
@@ -195,7 +195,7 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 	unsigned long len = pci_resource_len(dev, bar);
 	unsigned long flags = pci_resource_flags(dev, bar);
 
-	if (!len || !start)
+	if (unlikely(!len || !start))
 		return NULL;
 	if (maxlen && len > maxlen)
 		len = maxlen;
@@ -204,18 +204,16 @@ void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
 	 * Presently the IORESOURCE_MEM case is a bit special, most
 	 * SH7751 style PCI controllers have PCI memory at a fixed
 	 * location in the address space where no remapping is desired
-	 * (traditionally at 0xfd000000). Once this changes, the
-	 * IORESOURCE_MEM case will have to switch to using ioremap() and
-	 * more care will have to be taken to inhibit page table mapping
-	 * for legacy cores.
-	 *
-	 * For now everything wraps to ioport_map(), since boards that
-	 * have PCI will be able to check the address range properly on
-	 * their own.
-	 *			-- PFM.
+	 * (typically at 0xfd000000, but is_pci_memaddr() will know
+	 * best). With the IORESOURCE_MEM case more care has to be taken
+	 * to inhibit page table mapping for legacy cores, but this is
+	 * punted off to __ioremap().
+	 *					-- PFM.
 	 */
-	if (flags & (IORESOURCE_IO | IORESOURCE_MEM))
+	if (flags & IORESOURCE_IO)
 		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM)
+		return ioremap(start, len);
 
 	return NULL;
 }
diff --git a/arch/sh/mm/ioremap.c b/arch/sh/mm/ioremap.c
index 96fa4a999e2a..a9fe80cfc233 100644
--- a/arch/sh/mm/ioremap.c
+++ b/arch/sh/mm/ioremap.c
@@ -15,6 +15,7 @@
 #include <linux/vmalloc.h>
 #include <linux/module.h>
 #include <linux/mm.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/page.h>
 #include <asm/pgalloc.h>
@@ -134,6 +135,20 @@ void __iomem *__ioremap(unsigned long phys_addr, unsigned long size,
 	if (phys_addr >= 0xA0000 && last_addr < 0x100000)
 		return (void __iomem *)phys_to_virt(phys_addr);
 
+	/*
+	 * If we're on an SH7751 or SH7780 PCI controller, PCI memory is
+	 * mapped at the end of the address space (typically 0xfd000000)
+	 * in a non-translatable area, so mapping through page tables for
+	 * this area is not only pointless, but also fundamentally
+	 * broken. Just return the physical address instead.
+	 *
+	 * For boards that map a small PCI memory aperture somewhere in
+	 * P1/P2 space, ioremap() will already do the right thing,
+	 * and we'll never get this far.
+	 */
+	if (is_pci_memaddr(phys_addr) && is_pci_memaddr(last_addr))
+		return (void __iomem *)phys_addr;
+
 	/*
 	 * Don't allow anybody to remap normal RAM that we're using..
 	 */
@@ -192,7 +207,7 @@ void __iounmap(void __iomem *addr)
 	unsigned long vaddr = (unsigned long __force)addr;
 	struct vm_struct *p;
 
-	if (PXSEG(vaddr) < P3SEG)
+	if (PXSEG(vaddr) < P3SEG || is_pci_memaddr(vaddr))
 		return;
 
 #ifdef CONFIG_32BIT
-- 
cgit v1.2.1


From ade2b3f6ed199dc4a7a0be53edf20f9399023640 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:48:46 +0900
Subject: sh: VoyagerGX cleanups and 8250 UART support.

This adds the VoyagerGX UART to the RTS7751R2D setup
code, and cleans up a few build issues.

Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/rts7751r2d/setup.c | 77 +++++++++++++++++++++++++++----
 arch/sh/cchips/voyagergx/irq.c            |  3 +-
 arch/sh/cchips/voyagergx/setup.c          |  2 +-
 3 files changed, 71 insertions(+), 11 deletions(-)

diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index 60907f574f34..f6ce7021feba 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -1,18 +1,76 @@
 /*
- * linux/arch/sh/kernel/setup_rts7751r2d.c
- *
- * Copyright (C) 2000  Kazumoto Kojima
- *
  * Renesas Technology Sales RTS7751R2D Support.
  *
- * Modified for RTS7751R2D by
- * Atom Create Engineering Co., Ltd. 2002.
+ * Copyright (C) 2002 Atom Create Engineering Co., Ltd.
+ * Copyright (C) 2004 - 2006 Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
  */
-
 #include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/serial_8250.h>
 #include <linux/pm.h>
 #include <asm/io.h>
-#include <asm/rts7751r2d/rts7751r2d.h>
+#include <asm/mach/rts7751r2d.h>
+#include <asm/mach/voyagergx_reg.h>
+
+static struct plat_serial8250_port uart_platform_data[] = {
+	{
+		.membase	= (void *)VOYAGER_UART_BASE,
+		.mapbase	= VOYAGER_UART_BASE,
+		.iotype		= UPIO_MEM,
+		.irq		= VOYAGER_UART0_IRQ,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.regshift	= 2,
+		.uartclk	= (9600 * 16),
+	}, {
+		.flags		= 0,
+	},
+};
+
+static void __init voyagergx_serial_init(void)
+{
+	unsigned long val;
+
+	/*
+	 * GPIO Control
+	 */
+	val = inl(GPIO_MUX_HIGH);
+	val |= 0x00001fe0;
+	outl(val, GPIO_MUX_HIGH);
+
+	/*
+	 * Power Mode Gate
+	 */
+	val = inl(POWER_MODE0_GATE);
+	val |= (POWER_MODE0_GATE_U0 | POWER_MODE0_GATE_U1);
+	outl(val, POWER_MODE0_GATE);
+
+	val = inl(POWER_MODE1_GATE);
+	val |= (POWER_MODE1_GATE_U0 | POWER_MODE1_GATE_U1);
+	outl(val, POWER_MODE1_GATE);
+}
+
+static struct platform_device uart_device = {
+	.name		= "serial8250",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= uart_platform_data,
+	},
+};
+
+static struct platform_device *rts7751r2d_devices[] __initdata = {
+	&uart_device,
+};
+
+static int __init rts7751r2d_devices_setup(void)
+{
+	return platform_add_devices(rts7751r2d_devices,
+				    ARRAY_SIZE(rts7751r2d_devices));
+}
+__initcall(rts7751r2d_devices_setup);
 
 const char *get_system_type(void)
 {
@@ -32,4 +90,7 @@ void __init platform_setup(void)
 	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
+
+	voyagergx_serial_init();
+
 }
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index 0dc1fb8f9687..036e8f900013 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -32,8 +32,7 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/rts7751r2d/rts7751r2d.h>
-#include <asm/rts7751r2d/voyagergx_reg.h>
+#include <asm/mach/voyagergx_reg.h>
 
 static void disable_voyagergx_irq(unsigned int irq)
 {
diff --git a/arch/sh/cchips/voyagergx/setup.c b/arch/sh/cchips/voyagergx/setup.c
index 139ca88ac9e6..3a4cfaa9a214 100644
--- a/arch/sh/cchips/voyagergx/setup.c
+++ b/arch/sh/cchips/voyagergx/setup.c
@@ -13,7 +13,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <asm/io.h>
-#include <asm/rts7751r2d/voyagergx_reg.h>
+#include <asm/mach/voyagergx_reg.h>
 
 static int __init setup_voyagergx(void)
 {
-- 
cgit v1.2.1


From e2d1864da5bfa419a108f42c0615f69432b1b876 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:53:40 +0900
Subject: sh: G2 DMA IRQ and residue sampling.

This fixes a long-standing FIXME for G2 DMA, where we finally
wire up the IRQ handler and allow for sampling remaining bytes
while in-flight.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/drivers/dma/dma-g2.c | 54 +++++++++++++++++++++++++++++++++-----------
 1 file changed, 41 insertions(+), 13 deletions(-)

diff --git a/arch/sh/drivers/dma/dma-g2.c b/arch/sh/drivers/dma/dma-g2.c
index 0f866f8789f0..9cb070924180 100644
--- a/arch/sh/drivers/dma/dma-g2.c
+++ b/arch/sh/drivers/dma/dma-g2.c
@@ -3,7 +3,7 @@
  *
  * G2 bus DMA support
  *
- * Copyright (C) 2003, 2004  Paul Mundt
+ * Copyright (C) 2003 - 2006  Paul Mundt
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
@@ -13,7 +13,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/interrupt.h>
-
+#include <asm/cacheflush.h>
 #include <asm/mach/sysasic.h>
 #include <asm/mach/dma.h>
 #include <asm/dma.h>
@@ -47,17 +47,31 @@ struct g2_dma_info {
 
 static volatile struct g2_dma_info *g2_dma = (volatile struct g2_dma_info *)0xa05f7800;
 
+#define g2_bytes_remaining(i) \
+	((g2_dma->channel[i].size - \
+	  g2_dma->status[i].size) & 0x0fffffff)
+
 static irqreturn_t g2_dma_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
-	/* FIXME: Do some meaningful completion work here.. */
-	return IRQ_HANDLED;
-}
+	int i;
 
-static struct irqaction g2_dma_irq = {
-	.name		= "g2 DMA handler",
-	.handler	= g2_dma_interrupt,
-	.flags		= IRQF_DISABLED,
-};
+	for (i = 0; i < G2_NR_DMA_CHANNELS; i++) {
+		if (g2_dma->status[i].status & 0x20000000) {
+			unsigned int bytes = g2_bytes_remaining(i);
+
+			if (likely(bytes == 0)) {
+				struct dma_info *info = dev_id;
+				struct dma_channel *chan = info->channels + i;
+
+				wake_up(&chan->wait_queue);
+
+				return IRQ_HANDLED;
+			}
+		}
+	}
+
+	return IRQ_NONE;
+}
 
 static int g2_enable_dma(struct dma_channel *chan)
 {
@@ -135,8 +149,14 @@ static int g2_xfer_dma(struct dma_channel *chan)
 	return 0;
 }
 
+static int g2_get_residue(struct dma_channel *chan)
+{
+	return g2_bytes_remaining(chan->chan);
+}
+
 static struct dma_ops g2_dma_ops = {
 	.xfer		= g2_xfer_dma,
+	.get_residue	= g2_get_residue,
 };
 
 static struct dma_info g2_dma_info = {
@@ -148,13 +168,22 @@ static struct dma_info g2_dma_info = {
 
 static int __init g2_dma_init(void)
 {
-	setup_irq(HW_EVENT_G2_DMA, &g2_dma_irq);
+	int ret;
+
+	ret = request_irq(HW_EVENT_G2_DMA, g2_dma_interrupt, IRQF_DISABLED,
+			  "g2 DMA handler", &g2_dma_info);
+	if (unlikely(ret))
+		return -EINVAL;
 
 	/* Magic */
 	g2_dma->wait_state	= 27;
 	g2_dma->magic		= 0x4659404f;
 
-	return register_dmac(&g2_dma_info);
+	ret = register_dmac(&g2_dma_info);
+	if (unlikely(ret != 0))
+		free_irq(HW_EVENT_G2_DMA, 0);
+
+	return ret;
 }
 
 static void __exit g2_dma_exit(void)
@@ -169,4 +198,3 @@ module_exit(g2_dma_exit);
 MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>");
 MODULE_DESCRIPTION("G2 bus DMA driver");
 MODULE_LICENSE("GPL");
-
-- 
cgit v1.2.1


From 9e3043c091819729ecf4fc5063d0a2d0954dfd7f Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 16:55:24 +0900
Subject: sh: Fixup TEI IRQ requests in request_dma().

If a channel is not TEI capable, don't try to request_irq()..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/drivers/dma/dma-sh.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index 4428ee809651..40a480d20aaf 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -82,6 +82,9 @@ static int sh_dmac_request_dma(struct dma_channel *chan)
 {
 	char name[32];
 
+	if (unlikely(!chan->flags & DMA_TEI_CAPABLE))
+		return 0;
+
 	snprintf(name, sizeof(name), "DMAC Transfer End (Channel %d)",
 		 chan->chan);
 
@@ -255,7 +258,7 @@ static int __init sh_dmac_init(void)
 #ifdef CONFIG_CPU_SH4
 	make_ipr_irq(DMAE_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
 	i = request_irq(DMAE_IRQ, dma_err, IRQF_DISABLED, "DMAC Address Error", 0);
-	if (i < 0)
+	if (unlikely(i < 0))
 		return i;
 #endif
 
@@ -270,7 +273,7 @@ static int __init sh_dmac_init(void)
 	 * been set.
 	 */
 	i = dmaor_reset();
-	if (i < 0)
+	if (unlikely(i != 0))
 		return i;
 
 	return register_dmac(info);
-- 
cgit v1.2.1


From 0f08f338083cc1d68788ccbccc44bd0502fc57ae Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:03:56 +0900
Subject: sh: More cosmetic cleanups and trivial fixes.

Nothing exciting here, just trivial fixes..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/drivers/dma/dma-sh.c    | 10 ++++++----
 arch/sh/drivers/pci/ops-titan.c |  1 +
 arch/sh/kernel/setup.c          |  2 +-
 arch/sh/kernel/sh_ksyms.c       | 15 ---------------
 arch/sh/mm/cache-sh7705.c       | 19 ++++++-------------
 arch/sh/mm/fault.c              | 13 ++++++++-----
 arch/sh/mm/pmb.c                |  4 ----
 include/asm-sh/dma.h            |  1 +
 include/asm-sh/irq.h            |  5 +++++
 include/asm-sh/kexec.h          |  2 ++
 include/asm-sh/system.h         | 36 +++++-------------------------------
 11 files changed, 35 insertions(+), 73 deletions(-)

diff --git a/arch/sh/drivers/dma/dma-sh.c b/arch/sh/drivers/dma/dma-sh.c
index 40a480d20aaf..cbbe8bce3d67 100644
--- a/arch/sh/drivers/dma/dma-sh.c
+++ b/arch/sh/drivers/dma/dma-sh.c
@@ -80,21 +80,23 @@ static irqreturn_t dma_tei(int irq, void *dev_id, struct pt_regs *regs)
 
 static int sh_dmac_request_dma(struct dma_channel *chan)
 {
-	char name[32];
-
 	if (unlikely(!chan->flags & DMA_TEI_CAPABLE))
 		return 0;
 
-	snprintf(name, sizeof(name), "DMAC Transfer End (Channel %d)",
+	chan->name = kzalloc(32, GFP_KERNEL);
+	if (unlikely(chan->name == NULL))
+		return -ENOMEM;
+	snprintf(chan->name, 32, "DMAC Transfer End (Channel %d)",
 		 chan->chan);
 
 	return request_irq(get_dmte_irq(chan->chan), dma_tei,
-			   IRQF_DISABLED, name, chan);
+			   IRQF_DISABLED, chan->name, chan);
 }
 
 static void sh_dmac_free_dma(struct dma_channel *chan)
 {
 	free_irq(get_dmte_irq(chan->chan), chan);
+	kfree(chan->name);
 }
 
 static void
diff --git a/arch/sh/drivers/pci/ops-titan.c b/arch/sh/drivers/pci/ops-titan.c
index 9c8b2027c35d..c6097bcd97fd 100644
--- a/arch/sh/drivers/pci/ops-titan.c
+++ b/arch/sh/drivers/pci/ops-titan.c
@@ -16,6 +16,7 @@
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/init.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/titan.h>
 #include "pci-sh4.h"
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index cff8d36f91b0..4afdec071700 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -507,7 +507,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	 * unified cache on the SH-2 and SH-3, as well as the harvard
 	 * style cache on the SH-4.
 	 */
-	if (test_bit(SH_CACHE_COMBINED, &(boot_cpu_data.icache.flags))) {
+	if (boot_cpu_data.icache.flags & SH_CACHE_COMBINED) {
 		seq_printf(m, "unified\n");
 		show_cacheinfo(m, "cache", boot_cpu_data.icache);
 	} else {
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index fd73ab0326e9..7f3a42244cab 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -27,20 +27,11 @@ EXPORT_SYMBOL(sh_mv);
 
 /* platform dependent support */
 EXPORT_SYMBOL(dump_fpu);
-EXPORT_SYMBOL(enable_irq);
-EXPORT_SYMBOL(disable_irq);
-EXPORT_SYMBOL(probe_irq_mask);
 EXPORT_SYMBOL(kernel_thread);
-EXPORT_SYMBOL(disable_irq_nosync);
 EXPORT_SYMBOL(irq_desc);
 EXPORT_SYMBOL(no_irq_type);
 
-EXPORT_SYMBOL(strstr);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strnlen);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strncat);
 
 /* PCI exports */
 #ifdef CONFIG_PCI
@@ -51,13 +42,8 @@ EXPORT_SYMBOL(pci_free_consistent);
 /* mem exports */
 EXPORT_SYMBOL(memchr);
 EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(memcpy_fromio);
-EXPORT_SYMBOL(memcpy_toio);
 EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(memset_io);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memcmp);
-EXPORT_SYMBOL(memscan);
 EXPORT_SYMBOL(__copy_user);
 EXPORT_SYMBOL(boot_cpu_data);
 
@@ -124,5 +110,4 @@ EXPORT_SYMBOL(csum_partial);
 #ifdef CONFIG_IPV6
 EXPORT_SYMBOL(csum_ipv6_magic);
 #endif
-EXPORT_SYMBOL(consistent_sync);
 EXPORT_SYMBOL(clear_page);
diff --git a/arch/sh/mm/cache-sh7705.c b/arch/sh/mm/cache-sh7705.c
index bf94eedb0a8e..045abdf078f5 100644
--- a/arch/sh/mm/cache-sh7705.c
+++ b/arch/sh/mm/cache-sh7705.c
@@ -9,7 +9,6 @@
  * for more details.
  *
  */
-
 #include <linux/init.h>
 #include <linux/mman.h>
 #include <linux/mm.h>
@@ -25,14 +24,10 @@
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
-/* The 32KB cache on the SH7705 suffers from the same synonym problem
- * as SH4 CPUs */
-
-#define __pte_offset(address) \
-		((address >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
-#define pte_offset(dir, address) ((pte_t *) pmd_page_vaddr(*(dir)) + \
-		__pte_offset(address))
-
+/*
+ * The 32KB cache on the SH7705 suffers from the same synonym problem
+ * as SH4 CPUs
+ */
 static inline void cache_wback_all(void)
 {
 	unsigned long ways, waysize, addrstart;
@@ -73,7 +68,6 @@ void flush_icache_range(unsigned long start, unsigned long end)
 	__flush_wback_region((void *)start, end - start);
 }
 
-
 /*
  * Writeback&Invalidate the D-cache of the page
  */
@@ -128,7 +122,6 @@ static void __flush_dcache_page(unsigned long phys)
 	local_irq_restore(flags);
 }
 
-
 /*
  * Write back & invalidate the D-cache of the page.
  * (To avoid "alias" issues)
@@ -186,7 +179,8 @@ void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
  *
  * ADDRESS: Virtual Address (U0 address)
  */
-void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigned long pfn)
+void flush_cache_page(struct vm_area_struct *vma, unsigned long address,
+		      unsigned long pfn)
 {
 	__flush_dcache_page(pfn << PAGE_SHIFT);
 }
@@ -203,4 +197,3 @@ void flush_icache_page(struct vm_area_struct *vma, struct page *page)
 {
 	__flush_purge_region(page_address(page), PAGE_SIZE);
 }
-
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index dc461d2bc183..c69fd603226a 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -13,6 +13,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/hardirq.h>
+#include <linux/kprobes.h>
 #include <asm/system.h>
 #include <asm/mmu_context.h>
 #include <asm/kgdb.h>
@@ -188,15 +190,16 @@ do_sigbus:
 /*
  * Called with interrupts disabled.
  */
-asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
-			       unsigned long address)
+asmlinkage int __kprobes __do_page_fault(struct pt_regs *regs,
+					 unsigned long writeaccess,
+					 unsigned long address)
 {
 	pgd_t *pgd;
 	pud_t *pud;
 	pmd_t *pmd;
 	pte_t *pte;
 	pte_t entry;
-	struct mm_struct *mm;
+	struct mm_struct *mm = current->mm;
 	spinlock_t *ptl;
 	int ret = 1;
 
@@ -214,10 +217,10 @@ asmlinkage int __do_page_fault(struct pt_regs *regs, unsigned long writeaccess,
 		pgd = pgd_offset_k(address);
 		mm = NULL;
 	} else {
-		if (unlikely(address >= TASK_SIZE || !(mm = current->mm)))
+		if (unlikely(address >= TASK_SIZE || !mm))
 			return 1;
 
-		pgd = pgd_offset(current->mm, address);
+		pgd = pgd_offset(mm, address);
 	}
 
 	pud = pud_offset(pgd, address);
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index 819fd0faf022..92e745341e4d 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -337,10 +337,8 @@ static int __init pmb_init(void)
 
 	return 0;
 }
-
 arch_initcall(pmb_init);
 
-#ifdef CONFIG_DEBUG_FS
 static int pmb_seq_show(struct seq_file *file, void *iter)
 {
 	int i;
@@ -399,6 +397,4 @@ static int __init pmb_debugfs_init(void)
 
 	return 0;
 }
-
 postcore_initcall(pmb_debugfs_init);
-#endif
diff --git a/include/asm-sh/dma.h b/include/asm-sh/dma.h
index e62a6d0ed932..d9daa028689f 100644
--- a/include/asm-sh/dma.h
+++ b/include/asm-sh/dma.h
@@ -89,6 +89,7 @@ struct dma_channel {
 	wait_queue_head_t wait_queue;
 
 	struct sys_device dev;
+	char *name;
 };
 
 struct dma_info {
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 7e8455b1cb44..648102e9236f 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -334,6 +334,11 @@ extern void enable_irq(unsigned int);
 extern void make_maskreg_irq(unsigned int irq);
 extern unsigned short *irq_mask_register;
 
+/*
+ * PINT IRQs
+ */
+void init_IRQ_pint(void);
+
 /*
  * Function for "on chip support modules".
  */
diff --git a/include/asm-sh/kexec.h b/include/asm-sh/kexec.h
index 9dfe59f6fcb5..a5f85e9e428d 100644
--- a/include/asm-sh/kexec.h
+++ b/include/asm-sh/kexec.h
@@ -23,6 +23,8 @@
 /* The native architecture */
 #define KEXEC_ARCH KEXEC_ARCH_SH
 
+#define MAX_NOTE_BYTES 1024
+
 #ifndef __ASSEMBLY__
 
 extern void machine_shutdown(void);
diff --git a/include/asm-sh/system.h b/include/asm-sh/system.h
index 477422afeb0d..6c1f8fde5ac4 100644
--- a/include/asm-sh/system.h
+++ b/include/asm-sh/system.h
@@ -79,7 +79,7 @@ static inline void sched_cacheflush(void)
 }
 #endif
 
-static __inline__ unsigned long tas(volatile int *m)
+static inline unsigned long tas(volatile int *m)
 {
 	unsigned long retval;
 
@@ -161,7 +161,7 @@ static inline void local_irq_enable(void)
 }
 #endif
 
-static __inline__ void local_irq_disable(void)
+static inline void local_irq_disable(void)
 {
 	unsigned long __dummy;
 	__asm__ __volatile__("stc	sr, %0\n\t"
@@ -172,7 +172,7 @@ static __inline__ void local_irq_disable(void)
 			     : "memory");
 }
 
-static __inline__ void set_bl_bit(void)
+static inline void set_bl_bit(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -185,7 +185,7 @@ static __inline__ void set_bl_bit(void)
 			     : "memory");
 }
 
-static __inline__ void clear_bl_bit(void)
+static inline void clear_bl_bit(void)
 {
 	unsigned long __dummy0, __dummy1;
 
@@ -207,7 +207,7 @@ static __inline__ void clear_bl_bit(void)
 	(flags != 0);			\
 })
 
-static __inline__ unsigned long local_irq_save(void)
+static inline unsigned long local_irq_save(void)
 {
 	unsigned long flags, __dummy;
 
@@ -223,36 +223,10 @@ static __inline__ unsigned long local_irq_save(void)
 	return flags;
 }
 
-#ifdef DEBUG_CLI_STI
-static __inline__ void  local_irq_restore(unsigned long x)
-{
-	if ((x & 0x000000f0) != 0x000000f0)
-		local_irq_enable();
-	else {
-		unsigned long flags;
-		local_save_flags(flags);
-
-		if (flags == 0) {
-			extern void dump_stack(void);
-			printk(KERN_ERR "BUG!\n");
-			dump_stack();
-			local_irq_disable();
-		}
-	}
-}
-#else
 #define local_irq_restore(x) do {			\
 	if ((x & 0x000000f0) != 0x000000f0)		\
 		local_irq_enable();			\
 } while (0)
-#endif
-
-#define really_restore_flags(x) do {			\
-	if ((x & 0x000000f0) != 0x000000f0)		\
-		local_irq_enable();			\
-	else						\
-		local_irq_disable();			\
-} while (0)
 
 /*
  * Jump to P2 area.
-- 
cgit v1.2.1


From 2991be725260d6fec11691a6138b9d71de949956 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:07:07 +0900
Subject: sh: Fixup __strnlen_user() behaviour.

Drop TIF_USERSPACE and add addr_limit to the thread_info struct.
Subsequently, use that for address checking in strnlen_user() to
ward off bogus -EFAULTs.

Make __strnlen_user() return 0 on exception, rather than -EFAULT.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/process.c     |  3 ++
 include/asm-sh/processor.h   |  4 +++
 include/asm-sh/thread_info.h |  4 +--
 include/asm-sh/uaccess.h     | 65 +++++++++-----------------------------------
 4 files changed, 22 insertions(+), 54 deletions(-)

diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index f2031314cb2b..4a32550fd7c6 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -263,6 +263,7 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 		unsigned long unused,
 		struct task_struct *p, struct pt_regs *regs)
 {
+	struct thread_info *ti = task_thread_info(p);
 	struct pt_regs *childregs;
 #if defined(CONFIG_SH_FPU)
 	struct task_struct *tsk = current;
@@ -277,8 +278,10 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 
 	if (user_mode(regs)) {
 		childregs->regs[15] = usp;
+		ti->addr_limit = USER_DS;
 	} else {
 		childregs->regs[15] = (unsigned long)task_stack_page(p) + THREAD_SIZE;
+		ti->addr_limit = KERNEL_DS;
 	}
         if (clone_flags & CLONE_SETTLS) {
 		childregs->gbr = childregs->regs[0];
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index e99aff706cf7..a22732007dd4 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -149,6 +149,10 @@ struct thread_struct {
 	union sh_fpu_union fpu;
 };
 
+typedef struct {
+	unsigned long seg;
+} mm_segment_t;
+
 /* Count of active tasks with UBC settings */
 extern int ubc_usercnt;
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index f64dd803a014..b986a1914472 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -21,6 +21,7 @@ struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	__u32			cpu;
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
+	mm_segment_t		addr_limit;	/* thread address space */
 	struct restart_block	restart_block;
 	__u8			supervisor_stack[0];
 };
@@ -40,6 +41,7 @@ struct thread_info {
 	.flags		= 0,			\
 	.cpu		= 0,			\
 	.preempt_count	= 1,			\
+	.addr_limit	= KERNEL_DS,		\
 	.restart_block	= {			\
 		.fn = do_no_restart_syscall,	\
 	},					\
@@ -95,7 +97,6 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
-#define TIF_USERSPACE		31	/* true if FS sets userspace */
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
@@ -103,7 +104,6 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
-#define _TIF_USERSPACE		(1<<TIF_USERSPACE)
 
 #define _TIF_WORK_MASK		0x000000FE	/* work to do on interrupt/exception return */
 #define _TIF_ALLWORK_MASK	0x000000FF	/* work to do on any return to u-space */
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 2cb01861e7c5..6c0014dd2ef7 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -16,21 +16,9 @@
 #include <linux/errno.h>
 #include <linux/sched.h>
 
-/*
- * NOTE: Macro/functions in this file depends on threads_info.h implementation.
- * Assumes:
- * TI_FLAGS == 8
- * TIF_USERSPACE == 31
- * USER_ADDR_LIMIT == 0x80000000
- */
-
 #define VERIFY_READ    0
 #define VERIFY_WRITE   1
 
-typedef struct {
-	unsigned int is_user_space;
-} mm_segment_t;
-
 /*
  * The fs value determines whether argument validity checking should be
  * performed or not.  If get_fs() == USER_DS, checking is performed, with
@@ -40,12 +28,14 @@ typedef struct {
  */
 
 #define MAKE_MM_SEG(s)	((mm_segment_t) { (s) })
-#define segment_eq(a,b)	((a).is_user_space == (b).is_user_space)
 
-#define USER_ADDR_LIMIT	0x80000000
+#define KERNEL_DS	MAKE_MM_SEG(0xFFFFFFFFUL)
+#define USER_DS		MAKE_MM_SEG(PAGE_OFFSET)
+
+#define segment_eq(a,b)	((a).seg == (b).seg)
 
-#define KERNEL_DS	MAKE_MM_SEG(0)
-#define USER_DS		MAKE_MM_SEG(1)
+#define __addr_ok(addr) \
+	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
 
 #define get_ds()	(KERNEL_DS)
 
@@ -76,31 +66,8 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 	return ((addr >= memory_start) && ((addr + size) < memory_end));
 }
 #else /* CONFIG_MMU */
-static inline mm_segment_t get_fs(void)
-{
-	return MAKE_MM_SEG(test_thread_flag(TIF_USERSPACE));
-}
-
-static inline void set_fs(mm_segment_t s)
-{
-	unsigned long ti, flag;
-	__asm__ __volatile__(
-		"stc	r7_bank, %0\n\t"
-		"mov.l	@(8,%0), %1\n\t"
-		"shal	%1\n\t"
-		"cmp/pl	%2\n\t"
-		"rotcr	%1\n\t"
-		"mov.l	%1, @(8,%0)"
-		: "=&r" (ti), "=&r" (flag)
-		: "r" (s.is_user_space)
-		: "t");
-/****
-	if (s.is_user_space)
-		set_thread_flag(TIF_USERSPACE);
-	else
-		clear_thread_flag(TIF_USERSPACE);
-****/
-}
+#define get_fs()	(current_thread_info()->addr_limit)
+#define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
 /*
  * __access_ok: Check if address with size is OK or not.
@@ -108,7 +75,7 @@ static inline void set_fs(mm_segment_t s)
  * We do three checks:
  * (1) is it user space? 
  * (2) addr + size --> carry?
- * (3) addr + size >= 0x80000000  (USER_ADDR_LIMIT)
+ * (3) addr + size >= 0x80000000  (PAGE_OFFSET)
  *
  * (1) (2) (3) | RESULT
  *  0   0   0  |  ok
@@ -541,7 +508,7 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 		"3:\n\t"
 		"mov.l	4f, %1\n\t"
 		"jmp	@%1\n\t"
-		" mov	%5, %0\n"
+		" mov	#0, %0\n"
 		".balign 4\n"
 		"4:	.long 2b\n"
 		".previous\n"
@@ -550,26 +517,20 @@ static __inline__ long __strnlen_user(const char __user *__s, long __n)
 		"	.long 1b,3b\n"
 		".previous"
 		: "=z" (res), "=&r" (__dummy)
-		: "0" (0), "r" (__s), "r" (__n), "i" (-EFAULT)
+		: "0" (0), "r" (__s), "r" (__n)
 		: "t");
 	return res;
 }
 
 static __inline__ long strnlen_user(const char __user *s, long n)
 {
-	if (!access_ok(VERIFY_READ, s, n))
+	if (!__addr_ok(s))
 		return 0;
 	else
 		return __strnlen_user(s, n);
 }
 
-static __inline__ long strlen_user(const char __user *s)
-{
-	if (!access_ok(VERIFY_READ, s, 0))
-		return 0;
-	else
-		return __strnlen_user(s, ~0UL >> 1);
-}
+#define strlen_user(str)	strnlen_user(str, ~0UL >> 1)
 
 /*
  * The exception table consists of pairs of addresses: the first is the
-- 
cgit v1.2.1


From af514ca7d27b31e3c278e1331f0ebdb3ad385a90 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:11:32 +0900
Subject: sh: Rename rtc_get/set_time() to avoid RTC_CLASS conflict.

We have a clash with RTC_CLASS over these names, so we
change them..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/dreamcast/rtc.c    |  4 ++--
 arch/sh/boards/landisk/rtc.c      | 17 ++++-------------
 arch/sh/boards/mpc1211/rtc.c      |  4 ++--
 arch/sh/boards/sh03/rtc.c         |  9 ++++-----
 arch/sh/boards/snapgear/rtc.c     |  8 ++++----
 arch/sh/kernel/time.c             | 14 +++++++-------
 arch/sh/kernel/timers/timer-tmu.c |  6 +++---
 include/asm-sh/rtc.h              |  4 ++--
 8 files changed, 28 insertions(+), 38 deletions(-)

diff --git a/arch/sh/boards/dreamcast/rtc.c b/arch/sh/boards/dreamcast/rtc.c
index 44420494dc22..b3a876a3b859 100644
--- a/arch/sh/boards/dreamcast/rtc.c
+++ b/arch/sh/boards/dreamcast/rtc.c
@@ -75,7 +75,7 @@ int aica_rtc_settimeofday(const time_t secs)
 
 void aica_time_init(void)
 {
-	rtc_get_time = aica_rtc_gettimeofday;
-	rtc_set_time = aica_rtc_settimeofday;
+	rtc_sh_get_time = aica_rtc_gettimeofday;
+	rtc_sh_set_time = aica_rtc_settimeofday;
 }
 
diff --git a/arch/sh/boards/landisk/rtc.c b/arch/sh/boards/landisk/rtc.c
index d666aa63303f..35ba726a0979 100644
--- a/arch/sh/boards/landisk/rtc.c
+++ b/arch/sh/boards/landisk/rtc.c
@@ -16,17 +16,9 @@
 #include <linux/time.h>
 #include <linux/delay.h>
 #include <linux/spinlock.h>
+#include <linux/bcd.h>
+#include <asm/rtc.h>
 
-#ifndef BCD_TO_BIN
-#define BCD_TO_BIN(val) ((val)=((val)&15) + ((val)>>4)*10)
-#endif
-
-#ifndef BIN_TO_BCD
-#define BIN_TO_BCD(val) ((val)=(((val)/10)<<4) + (val)%10)
-#endif
-
-extern void (*rtc_get_time) (struct timespec *);
-extern int (*rtc_set_time) (const time_t);
 extern spinlock_t rtc_lock;
 
 extern void
@@ -94,9 +86,8 @@ int landisk_rtc_settimeofday(const time_t secs)
 	return retval;
 }
 
-
 void landisk_time_init(void)
 {
-	rtc_get_time = landisk_rtc_gettimeofday;
-	rtc_set_time = landisk_rtc_settimeofday;
+	rtc_sh_get_time = landisk_rtc_gettimeofday;
+	rtc_sh_set_time = landisk_rtc_settimeofday;
 }
diff --git a/arch/sh/boards/mpc1211/rtc.c b/arch/sh/boards/mpc1211/rtc.c
index a76c655dceee..03b123a4bba4 100644
--- a/arch/sh/boards/mpc1211/rtc.c
+++ b/arch/sh/boards/mpc1211/rtc.c
@@ -130,7 +130,7 @@ int mpc1211_rtc_settimeofday(const struct timeval *tv)
 
 void mpc1211_time_init(void)
 {
-	rtc_get_time = mpc1211_rtc_gettimeofday;
-	rtc_set_time = mpc1211_rtc_settimeofday;
+	rtc_sh_get_time = mpc1211_rtc_gettimeofday;
+	rtc_sh_set_time = mpc1211_rtc_settimeofday;
 }
 
diff --git a/arch/sh/boards/sh03/rtc.c b/arch/sh/boards/sh03/rtc.c
index d609863cfe53..0a9266bb51c5 100644
--- a/arch/sh/boards/sh03/rtc.c
+++ b/arch/sh/boards/sh03/rtc.c
@@ -10,9 +10,10 @@
 #include <linux/sched.h>
 #include <linux/time.h>
 #include <linux/bcd.h>
-#include <asm/io.h>
 #include <linux/rtc.h>
 #include <linux/spinlock.h>
+#include <asm/io.h>
+#include <asm/rtc.h>
 
 #define RTC_BASE	0xb0000000
 #define RTC_SEC1	(RTC_BASE + 0)
@@ -34,8 +35,6 @@
 #define RTC_BUSY	1
 #define RTC_STOP	2
 
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
 extern spinlock_t rtc_lock;
 
 unsigned long get_cmos_time(void)
@@ -128,6 +127,6 @@ int sh03_rtc_settimeofday(const time_t secs)
 
 void sh03_time_init(void)
 {
-	rtc_get_time = sh03_rtc_gettimeofday;
-	rtc_set_time = sh03_rtc_settimeofday;
+	rtc_sh_get_time = sh03_rtc_gettimeofday;
+	rtc_sh_set_time = sh03_rtc_settimeofday;
 }
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index 287bc4ed3ac7..722479deb55a 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -165,11 +165,11 @@ void __init secureedge5410_rtc_init(void)
 		}
 
 	if (use_ds1302) {
-		rtc_get_time = snapgear_rtc_gettimeofday;
-		rtc_set_time = snapgear_rtc_settimeofday;
+		rtc_sh_get_time = snapgear_rtc_gettimeofday;
+		rtc_sh_set_time = snapgear_rtc_settimeofday;
 	} else {
-		rtc_get_time = sh_rtc_gettimeofday;
-		rtc_set_time = sh_rtc_settimeofday;
+		rtc_sh_get_time = sh_rtc_gettimeofday;
+		rtc_sh_set_time = sh_rtc_settimeofday;
 	}
 		
 	printk("SnapGear RTC: using %s rtc.\n", use_ds1302 ? "ds1302" : "internal");
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index c8db6ca4f9d1..8acd70bffe76 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -29,11 +29,11 @@ EXPORT_SYMBOL(rtc_lock);
 /* XXX: Can we initialize this in a routine somewhere?  Dreamcast doesn't want
  * these routines anywhere... */
 #ifdef CONFIG_SH_RTC
-void (*rtc_get_time)(struct timespec *) = sh_rtc_gettimeofday;
-int (*rtc_set_time)(const time_t) = sh_rtc_settimeofday;
+void (*rtc_sh_get_time)(struct timespec *) = sh_rtc_gettimeofday;
+int (*rtc_sh_set_time)(const time_t) = sh_rtc_settimeofday;
 #else
-void (*rtc_get_time)(struct timespec *);
-int (*rtc_set_time)(const time_t);
+void (*rtc_sh_get_time)(struct timespec *);
+int (*rtc_sh_set_time)(const time_t);
 #endif
 
 /*
@@ -135,7 +135,7 @@ void handle_timer_tick(struct pt_regs *regs)
 	    xtime.tv_sec > last_rtc_update + 660 &&
 	    (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 &&
 	    (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) {
-		if (rtc_set_time(xtime.tv_sec) == 0)
+		if (rtc_sh_set_time(xtime.tv_sec) == 0)
 			last_rtc_update = xtime.tv_sec;
 		else
 			/* do it again in 60s */
@@ -193,8 +193,8 @@ void __init time_init(void)
 
 	clk_init();
 
-	if (rtc_get_time) {
-		rtc_get_time(&xtime);
+	if (rtc_sh_get_time) {
+		rtc_sh_get_time(&xtime);
 	} else {
 		xtime.tv_sec = mktime(2000, 1, 1, 0, 0, 0);
 		xtime.tv_nsec = 0;
diff --git a/arch/sh/kernel/timers/timer-tmu.c b/arch/sh/kernel/timers/timer-tmu.c
index ea4bdf8cdf0d..205816fcf0da 100644
--- a/arch/sh/kernel/timers/timer-tmu.c
+++ b/arch/sh/kernel/timers/timer-tmu.c
@@ -132,17 +132,17 @@ static unsigned long tmu_timer_get_frequency(void)
 	ctrl_outl(0xffffffff, TMU0_TCOR);
 	ctrl_outl(0xffffffff, TMU0_TCNT);
 
-	rtc_get_time(&ts2);
+	rtc_sh_get_time(&ts2);
 
 	do {
-		rtc_get_time(&ts1);
+		rtc_sh_get_time(&ts1);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	/* actually start the timer */
 	ctrl_outb(TMU_TSTR_INIT, TMU_TSTR);
 
 	do {
-		rtc_get_time(&ts2);
+		rtc_sh_get_time(&ts2);
 	} while (ts1.tv_nsec == ts2.tv_nsec && ts1.tv_sec == ts2.tv_sec);
 
 	freq = 0xffffffff - ctrl_inl(TMU0_TCNT);
diff --git a/include/asm-sh/rtc.h b/include/asm-sh/rtc.h
index cea9cdf9b925..4c7be859c9c0 100644
--- a/include/asm-sh/rtc.h
+++ b/include/asm-sh/rtc.h
@@ -8,8 +8,8 @@
 extern void sh_rtc_gettimeofday(struct timespec *ts);
 extern int sh_rtc_settimeofday(const time_t secs);
 extern void (*board_time_init)(void);
-extern void (*rtc_get_time)(struct timespec *);
-extern int (*rtc_set_time)(const time_t);
+extern void (*rtc_sh_get_time)(struct timespec *);
+extern int (*rtc_sh_set_time)(const time_t);
 
 /* RCR1 Bits */
 #define RCR1_CF		0x80	/* Carry Flag             */
-- 
cgit v1.2.1


From 317a6104a99f87c0b35c0d9f19ec23ee7429b33e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:13:19 +0900
Subject: rtc: New RTC driver for SuperH On-Chip RTC.

This replaces the old SH RTC driver, and allows us to
clean quite a lot of things up on the board-specific
side.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/rtc/Kconfig  |  10 ++
 drivers/rtc/Makefile |   1 +
 drivers/rtc/rtc-sh.c | 467 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 478 insertions(+)
 create mode 100644 drivers/rtc/rtc-sh.c

diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index 7ff1d88094b6..33a7b720539b 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -238,6 +238,16 @@ config RTC_DRV_SA1100
 	  To compile this driver as a module, choose M here: the
 	  module will be called rtc-sa1100.
 
+config RTC_DRV_SH
+	tristate "SuperH On-Chip RTC"
+	depends on RTC_CLASS && SUPERH
+	help
+	  Say Y here to enable support for the on-chip RTC found in
+	  most SuperH processors.
+
+ 	  To compile this driver as a module, choose M here: the
+	  module will be called rtc-sh.
+
 config RTC_DRV_VR41XX
 	tristate "NEC VR41XX"
 	depends on RTC_CLASS && CPU_VR41XX
diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
index bbcfb09d81d9..e72d467ab214 100644
--- a/drivers/rtc/Makefile
+++ b/drivers/rtc/Makefile
@@ -31,3 +31,4 @@ obj-$(CONFIG_RTC_DRV_PL031)	+= rtc-pl031.o
 obj-$(CONFIG_RTC_DRV_MAX6902)	+= rtc-max6902.o
 obj-$(CONFIG_RTC_DRV_V3020)	+= rtc-v3020.o
 obj-$(CONFIG_RTC_DRV_AT91)	+= rtc-at91.o
+obj-$(CONFIG_RTC_DRV_SH)	+= rtc-sh.o
diff --git a/drivers/rtc/rtc-sh.c b/drivers/rtc/rtc-sh.c
new file mode 100644
index 000000000000..d2ce0c8bb8f3
--- /dev/null
+++ b/drivers/rtc/rtc-sh.c
@@ -0,0 +1,467 @@
+/*
+ * SuperH On-Chip RTC Support
+ *
+ * Copyright (C) 2006  Paul Mundt
+ *
+ * Based on the old arch/sh/kernel/cpu/rtc.c by:
+ *
+ *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
+ *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/bcd.h>
+#include <linux/rtc.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/seq_file.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <asm/io.h>
+
+#ifdef CONFIG_CPU_SH3
+#define rtc_reg_size		sizeof(u16)
+#define RTC_BIT_INVERTED	0	/* No bug on SH7708, SH7709A */
+#elif defined(CONFIG_CPU_SH4)
+#define rtc_reg_size		sizeof(u32)
+#define RTC_BIT_INVERTED	0x40	/* bug on SH7750, SH7750S */
+#endif
+
+#define RTC_REG(r)	((r) * rtc_reg_size)
+
+#define R64CNT  	RTC_REG(0)
+#define RSECCNT 	RTC_REG(1)
+#define RMINCNT 	RTC_REG(2)
+#define RHRCNT  	RTC_REG(3)
+#define RWKCNT  	RTC_REG(4)
+#define RDAYCNT 	RTC_REG(5)
+#define RMONCNT 	RTC_REG(6)
+#define RYRCNT  	RTC_REG(7)
+#define RSECAR  	RTC_REG(8)
+#define RMINAR  	RTC_REG(9)
+#define RHRAR   	RTC_REG(10)
+#define RWKAR   	RTC_REG(11)
+#define RDAYAR  	RTC_REG(12)
+#define RMONAR  	RTC_REG(13)
+#define RCR1    	RTC_REG(14)
+#define RCR2    	RTC_REG(15)
+
+/* RCR1 Bits */
+#define RCR1_CF		0x80	/* Carry Flag             */
+#define RCR1_CIE	0x10	/* Carry Interrupt Enable */
+#define RCR1_AIE	0x08	/* Alarm Interrupt Enable */
+#define RCR1_AF		0x01	/* Alarm Flag             */
+
+/* RCR2 Bits */
+#define RCR2_PEF	0x80	/* PEriodic interrupt Flag */
+#define RCR2_PESMASK	0x70	/* Periodic interrupt Set  */
+#define RCR2_RTCEN	0x08	/* ENable RTC              */
+#define RCR2_ADJ	0x04	/* ADJustment (30-second)  */
+#define RCR2_RESET	0x02	/* Reset bit               */
+#define RCR2_START	0x01	/* Start bit               */
+
+struct sh_rtc {
+	void __iomem *regbase;
+	unsigned long regsize;
+	struct resource *res;
+	unsigned int alarm_irq, periodic_irq, carry_irq;
+	struct rtc_device *rtc_dev;
+	spinlock_t lock;
+};
+
+static irqreturn_t sh_rtc_interrupt(int irq, void *id, struct pt_regs *regs)
+{
+	struct platform_device *pdev = id;
+	struct sh_rtc *rtc = platform_get_drvdata(pdev);
+	unsigned int tmp, events = 0;
+
+	spin_lock(&rtc->lock);
+
+	tmp = readb(rtc->regbase + RCR1);
+
+	if (tmp & RCR1_AF)
+		events |= RTC_AF | RTC_IRQF;
+
+	tmp &= ~(RCR1_CF | RCR1_AF);
+
+	writeb(tmp, rtc->regbase + RCR1);
+
+	rtc_update_irq(&rtc->rtc_dev->class_dev, 1, events);
+
+	spin_unlock(&rtc->lock);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t sh_rtc_periodic(int irq, void *id, struct pt_regs *regs)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(id);
+
+	spin_lock(&rtc->lock);
+
+	rtc_update_irq(&rtc->rtc_dev->class_dev, 1, RTC_PF | RTC_IRQF);
+
+	spin_unlock(&rtc->lock);
+
+	return IRQ_HANDLED;
+}
+
+static inline void sh_rtc_setpie(struct device *dev, unsigned int enable)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(dev);
+	unsigned int tmp;
+
+	spin_lock_irq(&rtc->lock);
+
+	tmp = readb(rtc->regbase + RCR2);
+
+	if (enable) {
+		tmp &= ~RCR2_PESMASK;
+		tmp |= RCR2_PEF | (2 << 4);
+	} else
+		tmp &= ~(RCR2_PESMASK | RCR2_PEF);
+
+	writeb(tmp, rtc->regbase + RCR2);
+
+	spin_unlock_irq(&rtc->lock);
+}
+
+static inline void sh_rtc_setaie(struct device *dev, unsigned int enable)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(dev);
+	unsigned int tmp;
+
+	spin_lock_irq(&rtc->lock);
+
+	tmp = readb(rtc->regbase + RCR1);
+
+	if (enable)
+		tmp |= RCR1_AIE;
+	else
+		tmp &= ~RCR1_AIE;
+
+	writeb(tmp, rtc->regbase + RCR1);
+
+	spin_unlock_irq(&rtc->lock);
+}
+
+static int sh_rtc_open(struct device *dev)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(dev);
+	unsigned int tmp;
+	int ret;
+
+	tmp = readb(rtc->regbase + RCR1);
+	tmp &= ~RCR1_CF;
+	tmp |= RCR1_CIE;
+	writeb(tmp, rtc->regbase + RCR1);
+
+	ret = request_irq(rtc->periodic_irq, sh_rtc_periodic, SA_INTERRUPT,
+			  "sh-rtc period", dev);
+	if (unlikely(ret)) {
+		dev_err(dev, "request period IRQ failed with %d, IRQ %d\n",
+			ret, rtc->periodic_irq);
+		return ret;
+	}
+
+	ret = request_irq(rtc->carry_irq, sh_rtc_interrupt, SA_INTERRUPT,
+			  "sh-rtc carry", dev);
+	if (unlikely(ret)) {
+		dev_err(dev, "request carry IRQ failed with %d, IRQ %d\n",
+			ret, rtc->carry_irq);
+		free_irq(rtc->periodic_irq, dev);
+		goto err_bad_carry;
+	}
+
+	ret = request_irq(rtc->alarm_irq, sh_rtc_interrupt, SA_INTERRUPT,
+			  "sh-rtc alarm", dev);
+	if (unlikely(ret)) {
+		dev_err(dev, "request alarm IRQ failed with %d, IRQ %d\n",
+			ret, rtc->alarm_irq);
+		goto err_bad_alarm;
+	}
+
+	return 0;
+
+err_bad_alarm:
+	free_irq(rtc->carry_irq, dev);
+err_bad_carry:
+	free_irq(rtc->periodic_irq, dev);
+
+	return ret;
+}
+
+static void sh_rtc_release(struct device *dev)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(dev);
+
+	sh_rtc_setpie(dev, 0);
+
+	free_irq(rtc->periodic_irq, dev);
+	free_irq(rtc->carry_irq, dev);
+	free_irq(rtc->alarm_irq, dev);
+}
+
+static int sh_rtc_proc(struct device *dev, struct seq_file *seq)
+{
+	struct sh_rtc *rtc = dev_get_drvdata(dev);
+	unsigned int tmp;
+
+	tmp = readb(rtc->regbase + RCR1);
+	seq_printf(seq, "alarm_IRQ\t: %s\n",
+		   (tmp & RCR1_AIE) ? "yes" : "no");
+	seq_printf(seq, "carry_IRQ\t: %s\n",
+		   (tmp & RCR1_CIE) ? "yes" : "no");
+
+	tmp = readb(rtc->regbase + RCR2);
+	seq_printf(seq, "periodic_IRQ\t: %s\n",
+		   (tmp & RCR2_PEF) ? "yes" : "no");
+
+	return 0;
+}
+
+static int sh_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
+{
+	unsigned int ret = -ENOIOCTLCMD;
+
+	switch (cmd) {
+	case RTC_PIE_OFF:
+	case RTC_PIE_ON:
+		sh_rtc_setpie(dev, cmd == RTC_PIE_ON);
+		ret = 0;
+		break;
+	case RTC_AIE_OFF:
+	case RTC_AIE_ON:
+		sh_rtc_setaie(dev, cmd == RTC_AIE_ON);
+		ret = 0;
+		break;
+	}
+
+	return ret;
+}
+
+static int sh_rtc_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct sh_rtc *rtc = platform_get_drvdata(pdev);
+	unsigned int sec128, sec2, yr, yr100, cf_bit;
+
+	do {
+		unsigned int tmp;
+
+		spin_lock_irq(&rtc->lock);
+
+		tmp = readb(rtc->regbase + RCR1);
+		tmp &= ~RCR1_CF; /* Clear CF-bit */
+		tmp |= RCR1_CIE;
+		writeb(tmp, rtc->regbase + RCR1);
+
+		sec128 = readb(rtc->regbase + R64CNT);
+
+		tm->tm_sec	= BCD2BIN(readb(rtc->regbase + RSECCNT));
+		tm->tm_min	= BCD2BIN(readb(rtc->regbase + RMINCNT));
+		tm->tm_hour	= BCD2BIN(readb(rtc->regbase + RHRCNT));
+		tm->tm_wday	= BCD2BIN(readb(rtc->regbase + RWKCNT));
+		tm->tm_mday	= BCD2BIN(readb(rtc->regbase + RDAYCNT));
+		tm->tm_mon	= BCD2BIN(readb(rtc->regbase + RMONCNT));
+
+#if defined(CONFIG_CPU_SH4)
+		yr  = readw(rtc->regbase + RYRCNT);
+		yr100 = BCD2BIN(yr >> 8);
+		yr &= 0xff;
+#else
+		yr  = readb(rtc->regbase + RYRCNT);
+		yr100 = BCD2BIN((yr == 0x99) ? 0x19 : 0x20);
+#endif
+
+		tm->tm_year = (yr100 * 100 + BCD2BIN(yr)) - 1900;
+
+		sec2 = readb(rtc->regbase + R64CNT);
+		cf_bit = readb(rtc->regbase + RCR1) & RCR1_CF;
+
+		spin_unlock_irq(&rtc->lock);
+	} while (cf_bit != 0 || ((sec128 ^ sec2) & RTC_BIT_INVERTED) != 0);
+
+#if RTC_BIT_INVERTED != 0
+	if ((sec128 & RTC_BIT_INVERTED))
+		tm->tm_sec--;
+#endif
+
+	dev_dbg(&dev, "%s: tm is secs=%d, mins=%d, hours=%d, "
+		"mday=%d, mon=%d, year=%d, wday=%d\n",
+		__FUNCTION__,
+		tm->tm_sec, tm->tm_min, tm->tm_hour,
+		tm->tm_mday, tm->tm_mon, tm->tm_year, tm->tm_wday);
+
+	if (rtc_valid_tm(tm) < 0)
+		dev_err(dev, "invalid date\n");
+
+	return 0;
+}
+
+static int sh_rtc_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct sh_rtc *rtc = platform_get_drvdata(pdev);
+	unsigned int tmp;
+	int year;
+
+	spin_lock_irq(&rtc->lock);
+
+	/* Reset pre-scaler & stop RTC */
+	tmp = readb(rtc->regbase + RCR2);
+	tmp |= RCR2_RESET;
+	writeb(tmp, rtc->regbase + RCR2);
+
+	writeb(BIN2BCD(tm->tm_sec),  rtc->regbase + RSECCNT);
+	writeb(BIN2BCD(tm->tm_min),  rtc->regbase + RMINCNT);
+	writeb(BIN2BCD(tm->tm_hour), rtc->regbase + RHRCNT);
+	writeb(BIN2BCD(tm->tm_wday), rtc->regbase + RWKCNT);
+	writeb(BIN2BCD(tm->tm_mday), rtc->regbase + RDAYCNT);
+	writeb(BIN2BCD(tm->tm_mon),  rtc->regbase + RMONCNT);
+
+#ifdef CONFIG_CPU_SH3
+	year = tm->tm_year % 100;
+	writeb(BIN2BCD(year), rtc->regbase + RYRCNT);
+#else
+	year = (BIN2BCD((tm->tm_year + 1900) / 100) << 8) |
+		BIN2BCD(tm->tm_year % 100);
+	writew(year, rtc->regbase + RYRCNT);
+#endif
+
+	/* Start RTC */
+	tmp = readb(rtc->regbase + RCR2);
+	tmp &= ~RCR2_RESET;
+	tmp |= RCR2_RTCEN | RCR2_START;
+	writeb(tmp, rtc->regbase + RCR2);
+
+	spin_unlock_irq(&rtc->lock);
+
+	return 0;
+}
+
+static struct rtc_class_ops sh_rtc_ops = {
+	.open		= sh_rtc_open,
+	.release	= sh_rtc_release,
+	.ioctl		= sh_rtc_ioctl,
+	.read_time	= sh_rtc_read_time,
+	.set_time	= sh_rtc_set_time,
+	.proc		= sh_rtc_proc,
+};
+
+static int __devinit sh_rtc_probe(struct platform_device *pdev)
+{
+	struct sh_rtc *rtc;
+	struct resource *res;
+	int ret = -ENOENT;
+
+	rtc = kzalloc(sizeof(struct sh_rtc), GFP_KERNEL);
+	if (unlikely(!rtc))
+		return -ENOMEM;
+
+	spin_lock_init(&rtc->lock);
+
+	rtc->periodic_irq = platform_get_irq(pdev, 0);
+	if (unlikely(rtc->periodic_irq < 0)) {
+		dev_err(&pdev->dev, "No IRQ for period\n");
+		goto err_badres;
+	}
+
+	rtc->carry_irq = platform_get_irq(pdev, 1);
+	if (unlikely(rtc->carry_irq < 0)) {
+		dev_err(&pdev->dev, "No IRQ for carry\n");
+		goto err_badres;
+	}
+
+	rtc->alarm_irq = platform_get_irq(pdev, 2);
+	if (unlikely(rtc->alarm_irq < 0)) {
+		dev_err(&pdev->dev, "No IRQ for alarm\n");
+		goto err_badres;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_IO, 0);
+	if (unlikely(res == NULL)) {
+		dev_err(&pdev->dev, "No IO resource\n");
+		goto err_badres;
+	}
+
+	rtc->regsize = res->end - res->start + 1;
+
+	rtc->res = request_mem_region(res->start, rtc->regsize, pdev->name);
+	if (unlikely(!rtc->res)) {
+		ret = -EBUSY;
+		goto err_badres;
+	}
+
+	rtc->regbase = (void __iomem *)rtc->res->start;
+	if (unlikely(!rtc->regbase)) {
+		ret = -EINVAL;
+		goto err_badmap;
+	}
+
+	rtc->rtc_dev = rtc_device_register("sh", &pdev->dev,
+					   &sh_rtc_ops, THIS_MODULE);
+	if (IS_ERR(rtc)) {
+		ret = PTR_ERR(rtc->rtc_dev);
+		goto err_badmap;
+	}
+
+	platform_set_drvdata(pdev, rtc);
+
+	return 0;
+
+err_badmap:
+	release_resource(rtc->res);
+err_badres:
+	kfree(rtc);
+
+	return ret;
+}
+
+static int __devexit sh_rtc_remove(struct platform_device *pdev)
+{
+	struct sh_rtc *rtc = platform_get_drvdata(pdev);
+
+	if (likely(rtc->rtc_dev))
+		rtc_device_unregister(rtc->rtc_dev);
+
+	sh_rtc_setpie(&pdev->dev, 0);
+	sh_rtc_setaie(&pdev->dev, 0);
+
+	release_resource(rtc->res);
+
+	platform_set_drvdata(pdev, NULL);
+
+	kfree(rtc);
+
+	return 0;
+}
+static struct platform_driver sh_rtc_platform_driver = {
+	.driver		= {
+		.name	= "sh-rtc",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= sh_rtc_probe,
+	.remove		= __devexit_p(sh_rtc_remove),
+};
+
+static int __init sh_rtc_init(void)
+{
+	return platform_driver_register(&sh_rtc_platform_driver);
+}
+
+static void __exit sh_rtc_exit(void)
+{
+	platform_driver_unregister(&sh_rtc_platform_driver);
+}
+
+module_init(sh_rtc_init);
+module_exit(sh_rtc_exit);
+
+MODULE_DESCRIPTION("SuperH on-chip RTC driver");
+MODULE_AUTHOR("Paul Mundt <lethal@linux-sh.org>");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 4b565680d16300acab0ff167e24f0ea289a6bd5d Mon Sep 17 00:00:00 2001
From: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Date: Wed, 27 Sep 2006 17:15:32 +0900
Subject: sh: math-emu support

This implements initial math-emu support, aimed primarily at SH-3.

Signed-off-by: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig              |   9 +
 arch/sh/Makefile             |   1 +
 arch/sh/kernel/traps.c       | 164 +++++++++---
 arch/sh/math-emu/Makefile    |   1 +
 arch/sh/math-emu/math.c      | 624 +++++++++++++++++++++++++++++++++++++++++++
 arch/sh/math-emu/sfp-util.h  |  72 +++++
 include/asm-sh/sfp-machine.h |  86 ++++++
 7 files changed, 918 insertions(+), 39 deletions(-)
 create mode 100644 arch/sh/math-emu/Makefile
 create mode 100644 arch/sh/math-emu/math.c
 create mode 100644 arch/sh/math-emu/sfp-util.h
 create mode 100644 include/asm-sh/sfp-machine.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d1d724447826..07d6d699e9c2 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -339,6 +339,15 @@ config SH_FPU
 
 	  This option must be set in order to enable the FPU.
 
+config SH_FPU_EMU
+	bool "FPU emulation support"
+	depends on !SH_FPU && EXPERIMENTAL
+	default n
+	help
+	  Selecting this option will enable support for software FPU emulation.
+	  Most SH-3 users will want to say Y here, whereas most SH-4 users will
+	  want to say N.
+
 config SH_DSP
 	bool "DSP support"
 	depends on !CPU_SH4
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 859285d47a29..13b688b13e1c 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -79,6 +79,7 @@ head-y := arch/sh/kernel/head.o arch/sh/kernel/init_task.o
 LIBGCC := $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 core-y				+= arch/sh/kernel/ arch/sh/mm/
+core-$(CONFIG_SH_FPU_EMU)	+= arch/sh/math-emu/
 
 # Boards
 machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index 21bef1b5f991..95c810b3c97e 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -36,40 +36,15 @@
 
 #ifdef CONFIG_SH_KGDB
 #include <asm/kgdb.h>
-#define CHK_REMOTE_DEBUG(regs)                                               \
-{                                                                            \
-  if ((kgdb_debug_hook != (kgdb_debug_hook_t *) NULL) && (!user_mode(regs))) \
-  {                                                                          \
-    (*kgdb_debug_hook)(regs);                                                \
-  }                                                                          \
+#define CHK_REMOTE_DEBUG(regs)                 	\
+{       					\
+	if (kgdb_debug_hook && !user_mode(regs))\
+		(*kgdb_debug_hook)(regs);       \
 }
 #else
 #define CHK_REMOTE_DEBUG(regs)
 #endif
 
-#define DO_ERROR(trapnr, signr, str, name, tsk)				\
-asmlinkage void do_##name(unsigned long r4, unsigned long r5,		\
-			  unsigned long r6, unsigned long r7,		\
-			  struct pt_regs regs)				\
-{									\
-	unsigned long error_code;					\
- 									\
-	/* Check if it's a DSP instruction */				\
- 	if (is_dsp_inst(&regs)) {					\
-		/* Enable DSP mode, and restart instruction. */		\
-		regs.sr |= SR_DSP;					\
-		return;							\
-	}								\
-									\
-	asm volatile("stc	r2_bank, %0": "=r" (error_code));	\
-	local_irq_enable();						\
-	tsk->thread.error_code = error_code;				\
-	tsk->thread.trap_no = trapnr;					\
-        CHK_REMOTE_DEBUG(&regs);					\
-	force_sig(signr, tsk);						\
-	die_if_no_fixup(str,&regs,error_code);				\
-}
-
 #ifdef CONFIG_CPU_SH2
 #define TRAP_RESERVED_INST	4
 #define TRAP_ILLEGAL_SLOT_INST	6
@@ -575,8 +550,117 @@ int is_dsp_inst(struct pt_regs *regs)
 #define is_dsp_inst(regs)	(0)
 #endif /* CONFIG_SH_DSP */
 
-DO_ERROR(TRAP_RESERVED_INST, SIGILL, "reserved instruction", reserved_inst, current)
-DO_ERROR(TRAP_ILLEGAL_SLOT_INST, SIGILL, "illegal slot instruction", illegal_slot_inst, current)
+extern int do_fpu_inst(unsigned short, struct pt_regs*);
+
+asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+	int err;
+
+	get_user(inst, (unsigned short*)regs.pc);
+
+	err = do_fpu_inst(inst, &regs);
+	if (!err) {
+		regs.pc += 2;
+		return;
+	}
+	/* not a FPU inst. */
+#endif
+
+#ifdef CONFIG_SH_DSP
+	/* Check if it's a DSP instruction */
+ 	if (is_dsp_inst(&regs)) {
+		/* Enable DSP mode, and restart instruction. */
+		regs.sr |= SR_DSP;
+		return;
+	}
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("reserved instruction", &regs, error_code);
+}
+
+#ifdef CONFIG_SH_FPU_EMU
+static int emulate_branch(unsigned short inst, struct pt_regs* regs)
+{
+	/*
+	 * bfs: 8fxx: PC+=d*2+4;
+	 * bts: 8dxx: PC+=d*2+4;
+	 * bra: axxx: PC+=D*2+4;
+	 * bsr: bxxx: PC+=D*2+4  after PR=PC+4;
+	 * braf:0x23: PC+=Rn*2+4;
+	 * bsrf:0x03: PC+=Rn*2+4 after PR=PC+4;
+	 * jmp: 4x2b: PC=Rn;
+	 * jsr: 4x0b: PC=Rn      after PR=PC+4;
+	 * rts: 000b: PC=PR;
+	 */
+	if ((inst & 0xfd00) == 0x8d00) {
+		regs->pc += SH_PC_8BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xe000) == 0xa000) {
+		regs->pc += SH_PC_12BIT_OFFSET(inst);
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x0003) {
+		regs->pc += regs->regs[(inst & 0x0f00) >> 8] + 4;
+		return 0;
+	}
+
+	if ((inst & 0xf0df) == 0x400b) {
+		regs->pc = regs->regs[(inst & 0x0f00) >> 8];
+		return 0;
+	}
+
+	if ((inst & 0xffff) == 0x000b) {
+		regs->pc = regs->pr;
+		return 0;
+	}
+
+	return 1;
+}
+#endif
+
+asmlinkage void do_illegal_slot_inst(unsigned long r4, unsigned long r5,
+				unsigned long r6, unsigned long r7,
+				struct pt_regs regs)
+{
+	unsigned long error_code;
+	struct task_struct *tsk = current;
+#ifdef CONFIG_SH_FPU_EMU
+	unsigned short inst;
+
+	get_user(inst, (unsigned short *)regs.pc + 1);
+	if (!do_fpu_inst(inst, &regs)) {
+		get_user(inst, (unsigned short *)regs.pc);
+		if (!emulate_branch(inst, &regs))
+			return;
+		/* fault in branch.*/
+	}
+	/* not a FPU inst. */
+#endif
+
+	asm volatile("stc	r2_bank, %0": "=r" (error_code));
+	local_irq_enable();
+	tsk->thread.error_code = error_code;
+	tsk->thread.trap_no = TRAP_RESERVED_INST;
+	CHK_REMOTE_DEBUG(&regs);
+	force_sig(SIGILL, tsk);
+	die_if_no_fixup("illegal slot instruction", &regs, error_code);
+}
 
 asmlinkage void do_exception_error(unsigned long r4, unsigned long r5,
 				   unsigned long r6, unsigned long r7,
@@ -634,14 +718,16 @@ void __init trap_init(void)
 	exception_handling_table[TRAP_ILLEGAL_SLOT_INST]
 		= (void *)do_illegal_slot_inst;
 
-#ifdef CONFIG_CPU_SH4
-	if (!(cpu_data->flags & CPU_HAS_FPU)) {
-		/* For SH-4 lacking an FPU, treat floating point instructions
-		   as reserved. */
-		/* entry 64 corresponds to EXPEVT=0x800 */
-		exception_handling_table[64] = (void *)do_reserved_inst;
-		exception_handling_table[65] = (void *)do_illegal_slot_inst;
-	}
+#if defined(CONFIG_CPU_SH4) && !defined(CONFIG_SH_FPU) || \
+    defined(CONFIG_SH_FPU_EMU)
+	/*
+	 * For SH-4 lacking an FPU, treat floating point instructions as
+	 * reserved. They'll be handled in the math-emu case, or faulted on
+	 * otherwise.
+	 */
+	/* entry 64 corresponds to EXPEVT=0x800 */
+	exception_handling_table[64] = (void *)do_reserved_inst;
+	exception_handling_table[65] = (void *)do_illegal_slot_inst;
 #endif
 		
 	/* Setup VBR for boot cpu */
diff --git a/arch/sh/math-emu/Makefile b/arch/sh/math-emu/Makefile
new file mode 100644
index 000000000000..638b342c781a
--- /dev/null
+++ b/arch/sh/math-emu/Makefile
@@ -0,0 +1 @@
+obj-y	:= math.o
diff --git a/arch/sh/math-emu/math.c b/arch/sh/math-emu/math.c
new file mode 100644
index 000000000000..26b6046814fd
--- /dev/null
+++ b/arch/sh/math-emu/math.c
@@ -0,0 +1,624 @@
+/*
+ * arch/sh/math-emu/math.c
+ *
+ * Copyright (C) 2006 Takashi YOSHII <takasi-y@ops.dti.ne.jp>
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+
+#include <asm/system.h>
+#include <asm/uaccess.h>
+#include <asm/processor.h>
+#include <asm/io.h>
+
+#include "sfp-util.h"
+#include <math-emu/soft-fp.h>
+#include <math-emu/single.h>
+#include <math-emu/double.h>
+
+#define	FPUL		(fregs->fpul)
+#define FPSCR		(fregs->fpscr)
+#define FPSCR_RM	(FPSCR&3)
+#define FPSCR_DN	((FPSCR>>18)&1)
+#define FPSCR_PR	((FPSCR>>19)&1)
+#define FPSCR_SZ	((FPSCR>>20)&1)
+#define FPSCR_FR	((FPSCR>>21)&1)
+#define FPSCR_MASK	0x003fffffUL
+
+#define BANK(n)	(n^(FPSCR_FR?16:0))
+#define FR	((unsigned long*)(fregs->fp_regs))
+#define FR0	(FR[BANK(0)])
+#define FRn	(FR[BANK(n)])
+#define FRm	(FR[BANK(m)])
+#define DR	((unsigned long long*)(fregs->fp_regs))
+#define DRn	(DR[BANK(n)/2])
+#define DRm	(DR[BANK(m)/2])
+
+#define XREG(n)	(n^16)
+#define XFn	(FR[BANK(XREG(n))])
+#define XFm	(FR[BANK(XREG(m))])
+#define XDn	(DR[BANK(XREG(n))/2])
+#define XDm	(DR[BANK(XREG(m))/2])
+
+#define R0	(regs->regs[0])
+#define Rn	(regs->regs[n])
+#define Rm	(regs->regs[m])
+
+#define WRITE(d,a)	({if(put_user(d, (typeof (d)*)a)) return -EFAULT;})
+#define READ(d,a)	({if(get_user(d, (typeof (d)*)a)) return -EFAULT;})
+
+#define PACK_S(r,f)	FP_PACK_SP(&r,f)
+#define UNPACK_S(f,r)	FP_UNPACK_SP(f,&r)
+#define PACK_D(r,f) \
+	{u32 t[2]; FP_PACK_DP(t,f); ((u32*)&r)[0]=t[1]; ((u32*)&r)[1]=t[0];}
+#define UNPACK_D(f,r) \
+	{u32 t[2]; t[0]=((u32*)&r)[1]; t[1]=((u32*)&r)[0]; FP_UNPACK_DP(f,t);}
+
+// 2 args instructions.
+#define BOTH_PRmn(op,x) \
+	FP_DECL_EX; if(FPSCR_PR) op(D,x,DRm,DRn); else op(S,x,FRm,FRn);
+
+#define CMP_X(SZ,R,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_CMP_##SZ(R, Fn, Fm, 2); }while(0)
+#define EQ_X(SZ,R,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_CMP_EQ_##SZ(R, Fn, Fm); }while(0)
+#define CMP(OP) ({ int r; BOTH_PRmn(OP##_X,r); r; })
+
+static int
+fcmp_gt(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	if (CMP(CMP) > 0)
+		regs->sr |= 1;
+	else
+		regs->sr &= ~1;
+
+	return 0;
+}
+
+static int
+fcmp_eq(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	if (CMP(CMP /*EQ*/) == 0)
+		regs->sr |= 1;
+	else
+		regs->sr &= ~1;
+	return 0;
+}
+
+#define ARITH_X(SZ,OP,M,N) do{ \
+	FP_DECL_##SZ(Fm); FP_DECL_##SZ(Fn); FP_DECL_##SZ(Fr); \
+	UNPACK_##SZ(Fm, M); UNPACK_##SZ(Fn, N); \
+	FP_##OP##_##SZ(Fr, Fn, Fm); \
+	PACK_##SZ(N, Fr); }while(0)
+
+static int
+fadd(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, ADD);
+	return 0;
+}
+
+static int
+fsub(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, SUB);
+	return 0;
+}
+
+static int
+fmul(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, MUL);
+	return 0;
+}
+
+static int
+fdiv(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	BOTH_PRmn(ARITH_X, DIV);
+	return 0;
+}
+
+static int
+fmac(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_S(Fr);
+	FP_DECL_S(Ft);
+	FP_DECL_S(F0);
+	FP_DECL_S(Fm);
+	FP_DECL_S(Fn);
+	UNPACK_S(F0, FR0);
+	UNPACK_S(Fm, FRm);
+	UNPACK_S(Fn, FRn);
+	FP_MUL_S(Ft, Fm, F0);
+	FP_ADD_S(Fr, Fn, Ft);
+	PACK_S(FRn, Fr);
+	return 0;
+}
+
+// to process fmov's extention (odd n for DR access XD).
+#define FMOV_EXT(x) if(x&1) x+=16-1
+
+static int
+fmov_idx_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + R0 + 4);
+		n++;
+		READ(FRn, Rm + R0);
+	} else {
+		READ(FRn, Rm + R0);
+	}
+
+	return 0;
+}
+
+static int
+fmov_mem_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + 4);
+		n++;
+		READ(FRn, Rm);
+	} else {
+		READ(FRn, Rm);
+	}
+
+	return 0;
+}
+
+static int
+fmov_inc_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(n);
+		READ(FRn, Rm + 4);
+		n++;
+		READ(FRn, Rm);
+		Rm += 8;
+	} else {
+		READ(FRn, Rm);
+		Rm += 4;
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_idx(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		WRITE(FRm, Rn + R0 + 4);
+		m++;
+		WRITE(FRm, Rn + R0);
+	} else {
+		WRITE(FRm, Rn + R0);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_mem(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		WRITE(FRm, Rn + 4);
+		m++;
+		WRITE(FRm, Rn);
+	} else {
+		WRITE(FRm, Rn);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_dec(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		Rn -= 8;
+		WRITE(FRm, Rn + 4);
+		m++;
+		WRITE(FRm, Rn);
+	} else {
+		Rn -= 4;
+		WRITE(FRm, Rn);
+	}
+
+	return 0;
+}
+
+static int
+fmov_reg_reg(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m,
+	     int n)
+{
+	if (FPSCR_SZ) {
+		FMOV_EXT(m);
+		FMOV_EXT(n);
+		DRn = DRm;
+	} else {
+		FRn = FRm;
+	}
+
+	return 0;
+}
+
+static int
+fnop_mn(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int m, int n)
+{
+	return -EINVAL;
+}
+
+// 1 arg instructions.
+#define NOTYETn(i) static int i(struct sh_fpu_soft_struct *fregs, int n) \
+	{ printk( #i " not yet done.\n"); return 0; }
+
+NOTYETn(ftrv)
+NOTYETn(fsqrt)
+NOTYETn(fipr)
+NOTYETn(fsca)
+NOTYETn(fsrra)
+
+#define EMU_FLOAT_X(SZ,N) do { \
+	FP_DECL_##SZ(Fn); \
+	FP_FROM_INT_##SZ(Fn, FPUL, 32, int); \
+	PACK_##SZ(N, Fn); }while(0)
+static int ffloat(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+
+	if (FPSCR_PR)
+		EMU_FLOAT_X(D, DRn);
+	else
+		EMU_FLOAT_X(S, FRn);
+
+	return 0;
+}
+
+#define EMU_FTRC_X(SZ,N) do { \
+	FP_DECL_##SZ(Fn); \
+	UNPACK_##SZ(Fn, N); \
+	FP_TO_INT_##SZ(FPUL, Fn, 32, 1); }while(0)
+static int ftrc(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+
+	if (FPSCR_PR)
+		EMU_FTRC_X(D, DRn);
+	else
+		EMU_FTRC_X(S, FRn);
+
+	return 0;
+}
+
+static int fcnvsd(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_S(Fn);
+	FP_DECL_D(Fr);
+	UNPACK_S(Fn, FPUL);
+	FP_CONV(D, S, 2, 1, Fr, Fn);
+	PACK_D(DRn, Fr);
+	return 0;
+}
+
+static int fcnvds(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FP_DECL_EX;
+	FP_DECL_D(Fn);
+	FP_DECL_S(Fr);
+	UNPACK_D(Fn, DRn);
+	FP_CONV(S, D, 1, 2, Fr, Fn);
+	PACK_S(FPUL, Fr);
+	return 0;
+}
+
+static int fxchg(struct sh_fpu_soft_struct *fregs, int flag)
+{
+	FPSCR ^= flag;
+	return 0;
+}
+
+static int fsts(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = FPUL;
+	return 0;
+}
+
+static int flds(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FPUL = FRn;
+	return 0;
+}
+
+static int fneg(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn ^= (1 << (_FP_W_TYPE_SIZE - 1));
+	return 0;
+}
+
+static int fabs(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn &= ~(1 << (_FP_W_TYPE_SIZE - 1));
+	return 0;
+}
+
+static int fld0(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = 0;
+	return 0;
+}
+
+static int fld1(struct sh_fpu_soft_struct *fregs, int n)
+{
+	FRn = (_FP_EXPBIAS_S << (_FP_FRACBITS_S - 1));
+	return 0;
+}
+
+static int fnop_n(struct sh_fpu_soft_struct *fregs, int n)
+{
+	return -EINVAL;
+}
+
+/// Instruction decoders.
+
+static int id_fxfd(struct sh_fpu_soft_struct *, int);
+static int id_fnxd(struct sh_fpu_soft_struct *, struct pt_regs *, int, int);
+
+static int (*fnxd[])(struct sh_fpu_soft_struct *, int) = {
+	fsts, flds, ffloat, ftrc, fneg, fabs, fsqrt, fsrra,
+	fld0, fld1, fcnvsd, fcnvds, fnop_n, fnop_n, fipr, id_fxfd
+};
+
+static int (*fnmx[])(struct sh_fpu_soft_struct *, struct pt_regs *, int, int) = {
+	fadd, fsub, fmul, fdiv, fcmp_eq, fcmp_gt, fmov_idx_reg, fmov_reg_idx,
+	fmov_mem_reg, fmov_inc_reg, fmov_reg_mem, fmov_reg_dec,
+	fmov_reg_reg, id_fnxd, fmac, fnop_mn};
+
+static int id_fxfd(struct sh_fpu_soft_struct *fregs, int x)
+{
+	const int flag[] = { FPSCR_SZ, FPSCR_PR, FPSCR_FR, 0 };
+	switch (x & 3) {
+	case 3:
+		fxchg(fregs, flag[x >> 2]);
+		break;
+	case 1:
+		ftrv(fregs, x - 1);
+		break;
+	default:
+		fsca(fregs, x);
+	}
+	return 0;
+}
+
+static int
+id_fnxd(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, int x, int n)
+{
+	return (fnxd[x])(fregs, n);
+}
+
+static int
+id_fnmx(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, u16 code)
+{
+	int n = (code >> 8) & 0xf, m = (code >> 4) & 0xf, x = code & 0xf;
+	return (fnmx[x])(fregs, regs, m, n);
+}
+
+static int
+id_sys(struct sh_fpu_soft_struct *fregs, struct pt_regs *regs, u16 code)
+{
+	int n = ((code >> 8) & 0xf);
+	unsigned long *reg = (code & 0x0010) ? &FPUL : &FPSCR;
+
+	switch (code & 0xf0ff) {
+	case 0x005a:
+	case 0x006a:
+		Rn = *reg;
+		break;
+	case 0x405a:
+	case 0x406a:
+		*reg = Rn;
+		break;
+	case 0x4052:
+	case 0x4062:
+		Rn -= 4;
+		WRITE(*reg, Rn);
+		break;
+	case 0x4056:
+	case 0x4066:
+		READ(*reg, Rn);
+		Rn += 4;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int fpu_emulate(u16 code, struct sh_fpu_soft_struct *fregs, struct pt_regs *regs)
+{
+	if ((code & 0xf000) == 0xf000)
+		return id_fnmx(fregs, regs, code);
+	else
+		return id_sys(fregs, regs, code);
+}
+
+/**
+ *	denormal_to_double - Given denormalized float number,
+ *	                     store double float
+ *
+ *	@fpu: Pointer to sh_fpu_hard structure
+ *	@n: Index to FP register
+ */
+static void denormal_to_double(struct sh_fpu_hard_struct *fpu, int n)
+{
+	unsigned long du, dl;
+	unsigned long x = fpu->fpul;
+	int exp = 1023 - 126;
+
+	if (x != 0 && (x & 0x7f800000) == 0) {
+		du = (x & 0x80000000);
+		while ((x & 0x00800000) == 0) {
+			x <<= 1;
+			exp--;
+		}
+		x &= 0x007fffff;
+		du |= (exp << 20) | (x >> 3);
+		dl = x << 29;
+
+		fpu->fp_regs[n] = du;
+		fpu->fp_regs[n+1] = dl;
+	}
+}
+
+/**
+ *	ieee_fpe_handler - Handle denormalized number exception
+ *
+ *	@regs: Pointer to register structure
+ *
+ *	Returns 1 when it's handled (should not cause exception).
+ */
+static int ieee_fpe_handler(struct pt_regs *regs)
+{
+	unsigned short insn = *(unsigned short *)regs->pc;
+	unsigned short finsn;
+	unsigned long nextpc;
+	int nib[4] = {
+		(insn >> 12) & 0xf,
+		(insn >> 8) & 0xf,
+		(insn >> 4) & 0xf,
+		insn & 0xf};
+
+	if (nib[0] == 0xb ||
+	    (nib[0] == 0x4 && nib[2] == 0x0 && nib[3] == 0xb)) /* bsr & jsr */
+		regs->pr = regs->pc + 4;
+
+	if (nib[0] == 0xa || nib[0] == 0xb) { /* bra & bsr */
+		nextpc = regs->pc + 4 + ((short) ((insn & 0xfff) << 4) >> 3);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xd) { /* bt/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		else
+			nextpc = regs->pc + 4;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x8 && nib[1] == 0xf) { /* bf/s */
+		if (regs->sr & 1)
+			nextpc = regs->pc + 4;
+		else
+			nextpc = regs->pc + 4 + ((char) (insn & 0xff) << 1);
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x4 && nib[3] == 0xb &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* jmp & jsr */
+		nextpc = regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (nib[0] == 0x0 && nib[3] == 0x3 &&
+		 (nib[2] == 0x0 || nib[2] == 0x2)) { /* braf & bsrf */
+		nextpc = regs->pc + 4 + regs->regs[nib[1]];
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else if (insn == 0x000b) { /* rts */
+		nextpc = regs->pr;
+		finsn = *(unsigned short *) (regs->pc + 2);
+	} else {
+		nextpc = regs->pc + 2;
+		finsn = insn;
+	}
+
+	if ((finsn & 0xf1ff) == 0xf0ad) { /* fcnvsd */
+		struct task_struct *tsk = current;
+
+		if ((tsk->thread.fpu.hard.fpscr & (1 << 17))) {
+			/* FPU error */
+			denormal_to_double (&tsk->thread.fpu.hard,
+					    (finsn >> 8) & 0xf);
+			tsk->thread.fpu.hard.fpscr &=
+				~(FPSCR_CAUSE_MASK | FPSCR_FLAG_MASK);
+			set_tsk_thread_flag(tsk, TIF_USEDFPU);
+		} else {
+			tsk->thread.trap_no = 11;
+			tsk->thread.error_code = 0;
+			force_sig(SIGFPE, tsk);
+		}
+
+		regs->pc = nextpc;
+		return 1;
+	}
+
+	return 0;
+}
+
+asmlinkage void do_fpu_error(unsigned long r4, unsigned long r5,
+			     unsigned long r6, unsigned long r7,
+			     struct pt_regs regs)
+{
+	struct task_struct *tsk = current;
+
+	if (ieee_fpe_handler (&regs))
+		return;
+
+	regs.pc += 2;
+	tsk->thread.trap_no = 11;
+	tsk->thread.error_code = 0;
+	force_sig(SIGFPE, tsk);
+}
+
+/**
+ * fpu_init - Initialize FPU registers
+ * @fpu: Pointer to software emulated FPU registers.
+ */
+static void fpu_init(struct sh_fpu_soft_struct *fpu)
+{
+	int i;
+
+	fpu->fpscr = FPSCR_INIT;
+	fpu->fpul = 0;
+
+	for (i = 0; i < 16; i++) {
+		fpu->fp_regs[i] = 0;
+		fpu->xfp_regs[i]= 0;
+	}
+}
+
+/**
+ * do_fpu_inst - Handle reserved instructions for FPU emulation
+ * @inst: instruction code.
+ * @regs: registers on stack.
+ */
+int do_fpu_inst(unsigned short inst, struct pt_regs *regs)
+{
+	struct task_struct *tsk = current;
+	struct sh_fpu_soft_struct *fpu = &(tsk->thread.fpu.soft);
+
+	if (!test_tsk_thread_flag(tsk, TIF_USEDFPU)) {
+		/* initialize once. */
+		fpu_init(fpu);
+		set_tsk_thread_flag(tsk, TIF_USEDFPU);
+	}
+
+	return fpu_emulate(inst, fpu, regs);
+}
diff --git a/arch/sh/math-emu/sfp-util.h b/arch/sh/math-emu/sfp-util.h
new file mode 100644
index 000000000000..8ae1bd310ad0
--- /dev/null
+++ b/arch/sh/math-emu/sfp-util.h
@@ -0,0 +1,72 @@
+/*
+ * These are copied from glibc/stdlib/longlong.h
+ */
+
+#define add_ssaaaa(sh, sl, ah, al, bh, bl) \
+  do {                                                                  \
+    UWtype __x;                                                         \
+    __x = (al) + (bl);                                                  \
+    (sh) = (ah) + (bh) + (__x < (al));                                  \
+    (sl) = __x;                                                         \
+  } while (0)
+
+#define sub_ddmmss(sh, sl, ah, al, bh, bl) \
+  do {                                                                  \
+    UWtype __x;                                                         \
+    __x = (al) - (bl);                                                  \
+    (sh) = (ah) - (bh) - (__x > (al));                                  \
+    (sl) = __x;                                                         \
+  } while (0)
+
+#define umul_ppmm(w1, w0, u, v) \
+  __asm__ ("dmulu.l %2,%3\n\tsts    macl,%1\n\tsts  mach,%0"	\
+	: "=r" ((u32)(w1)), "=r" ((u32)(w0))	\
+	:  "r" ((u32)(u)),   "r" ((u32)(v))	\
+	: "macl", "mach")
+
+#define __ll_B ((UWtype) 1 << (W_TYPE_SIZE / 2))
+#define __ll_lowpart(t) ((UWtype) (t) & (__ll_B - 1))
+#define __ll_highpart(t) ((UWtype) (t) >> (W_TYPE_SIZE / 2))
+
+#define udiv_qrnnd(q, r, n1, n0, d) \
+  do {									\
+    UWtype __d1, __d0, __q1, __q0;					\
+    UWtype __r1, __r0, __m;						\
+    __d1 = __ll_highpart (d);						\
+    __d0 = __ll_lowpart (d);						\
+									\
+    __r1 = (n1) % __d1;							\
+    __q1 = (n1) / __d1;							\
+    __m = (UWtype) __q1 * __d0;						\
+    __r1 = __r1 * __ll_B | __ll_highpart (n0);				\
+    if (__r1 < __m)							\
+      {									\
+	__q1--, __r1 += (d);						\
+	if (__r1 >= (d)) /* i.e. we didn't get carry when adding to __r1 */\
+	  if (__r1 < __m)						\
+	    __q1--, __r1 += (d);					\
+      }									\
+    __r1 -= __m;							\
+									\
+    __r0 = __r1 % __d1;							\
+    __q0 = __r1 / __d1;							\
+    __m = (UWtype) __q0 * __d0;						\
+    __r0 = __r0 * __ll_B | __ll_lowpart (n0);				\
+    if (__r0 < __m)							\
+      {									\
+	__q0--, __r0 += (d);						\
+	if (__r0 >= (d))						\
+	  if (__r0 < __m)						\
+	    __q0--, __r0 += (d);					\
+      }									\
+    __r0 -= __m;							\
+									\
+    (q) = (UWtype) __q1 * __ll_B | __q0;				\
+    (r) = __r0;								\
+  } while (0)
+
+#define abort()	return 0
+
+#define __BYTE_ORDER __LITTLE_ENDIAN
+
+
diff --git a/include/asm-sh/sfp-machine.h b/include/asm-sh/sfp-machine.h
new file mode 100644
index 000000000000..8a6399a8cfe0
--- /dev/null
+++ b/include/asm-sh/sfp-machine.h
@@ -0,0 +1,86 @@
+/* Machine-dependent software floating-point definitions.
+   SuperH kernel version.
+   Copyright (C) 1997,1998,1999 Free Software Foundation, Inc.
+   This file is part of the GNU C Library.
+   Contributed by Richard Henderson (rth@cygnus.com),
+		  Jakub Jelinek (jj@ultra.linux.cz),
+		  David S. Miller (davem@redhat.com) and
+		  Peter Maydell (pmaydell@chiark.greenend.org.uk).
+
+   The GNU C Library is free software; you can redistribute it and/or
+   modify it under the terms of the GNU Library General Public License as
+   published by the Free Software Foundation; either version 2 of the
+   License, or (at your option) any later version.
+
+   The GNU C Library is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Library General Public License for more details.
+
+   You should have received a copy of the GNU Library General Public
+   License along with the GNU C Library; see the file COPYING.LIB.  If
+   not, write to the Free Software Foundation, Inc.,
+   59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
+
+#ifndef _SFP_MACHINE_H
+#define _SFP_MACHINE_H
+
+#include <linux/config.h>
+
+#define _FP_W_TYPE_SIZE		32
+#define _FP_W_TYPE		unsigned long
+#define _FP_WS_TYPE		signed long
+#define _FP_I_TYPE		long
+
+#define _FP_MUL_MEAT_S(R,X,Y)					\
+  _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_S,R,X,Y,umul_ppmm)
+#define _FP_MUL_MEAT_D(R,X,Y)					\
+  _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm)
+#define _FP_MUL_MEAT_Q(R,X,Y)					\
+  _FP_MUL_MEAT_4_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm)
+
+#define _FP_DIV_MEAT_S(R,X,Y)	_FP_DIV_MEAT_1_udiv(S,R,X,Y)
+#define _FP_DIV_MEAT_D(R,X,Y)	_FP_DIV_MEAT_2_udiv(D,R,X,Y)
+#define _FP_DIV_MEAT_Q(R,X,Y)	_FP_DIV_MEAT_4_udiv(Q,R,X,Y)
+
+#define _FP_NANFRAC_S		((_FP_QNANBIT_S << 1) - 1)
+#define _FP_NANFRAC_D		((_FP_QNANBIT_D << 1) - 1), -1
+#define _FP_NANFRAC_Q		((_FP_QNANBIT_Q << 1) - 1), -1, -1, -1
+#define _FP_NANSIGN_S		0
+#define _FP_NANSIGN_D		0
+#define _FP_NANSIGN_Q		0
+
+#define _FP_KEEPNANFRACP 1
+
+/*
+ * If one NaN is signaling and the other is not,
+ * we choose that one, otherwise we choose X.
+ */
+#define _FP_CHOOSENAN(fs, wc, R, X, Y, OP)                      \
+  do {                                                          \
+    if ((_FP_FRAC_HIGH_RAW_##fs(X) & _FP_QNANBIT_##fs)          \
+        && !(_FP_FRAC_HIGH_RAW_##fs(Y) & _FP_QNANBIT_##fs))     \
+      {                                                         \
+        R##_s = Y##_s;                                          \
+        _FP_FRAC_COPY_##wc(R,Y);                                \
+      }                                                         \
+    else                                                        \
+      {                                                         \
+        R##_s = X##_s;                                          \
+        _FP_FRAC_COPY_##wc(R,X);                                \
+      }                                                         \
+    R##_c = FP_CLS_NAN;                                         \
+  } while (0)
+
+//#define FP_ROUNDMODE		FPSCR_RM
+#define FP_DENORM_ZERO		1/*FPSCR_DN*/
+
+/* Exception flags. */
+#define FP_EX_INVALID		(1<<4)
+#define FP_EX_DIVZERO		(1<<3)
+#define FP_EX_OVERFLOW		(1<<2)
+#define FP_EX_UNDERFLOW		(1<<1)
+#define FP_EX_INEXACT		(1<<0)
+
+#endif
+
-- 
cgit v1.2.1


From adf1890b0cd63f754b2171b73e4d845c0950d407 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:17:27 +0900
Subject: sh: Move voyagergx_reg.h to a more sensible place.

Other boards require this as well, so move it out of the
rts7751r2d directory.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/renesas/rts7751r2d/setup.c |   2 +-
 arch/sh/cchips/voyagergx/irq.c            |   2 +-
 arch/sh/cchips/voyagergx/setup.c          |   2 +-
 include/asm-sh/rts7751r2d/voyagergx_reg.h | 313 ------------------------------
 include/asm-sh/voyagergx.h                | 313 ++++++++++++++++++++++++++++++
 5 files changed, 316 insertions(+), 316 deletions(-)
 delete mode 100644 include/asm-sh/rts7751r2d/voyagergx_reg.h
 create mode 100644 include/asm-sh/voyagergx.h

diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index f6ce7021feba..bb56c018f2ee 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -14,7 +14,7 @@
 #include <linux/pm.h>
 #include <asm/io.h>
 #include <asm/mach/rts7751r2d.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static struct plat_serial8250_port uart_platform_data[] = {
 	{
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index 036e8f900013..4cfa7c990553 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -32,7 +32,7 @@
 
 #include <asm/io.h>
 #include <asm/irq.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static void disable_voyagergx_irq(unsigned int irq)
 {
diff --git a/arch/sh/cchips/voyagergx/setup.c b/arch/sh/cchips/voyagergx/setup.c
index 3a4cfaa9a214..66b2fedd7ad9 100644
--- a/arch/sh/cchips/voyagergx/setup.c
+++ b/arch/sh/cchips/voyagergx/setup.c
@@ -13,7 +13,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <asm/io.h>
-#include <asm/mach/voyagergx_reg.h>
+#include <asm/voyagergx.h>
 
 static int __init setup_voyagergx(void)
 {
diff --git a/include/asm-sh/rts7751r2d/voyagergx_reg.h b/include/asm-sh/rts7751r2d/voyagergx_reg.h
deleted file mode 100644
index f031b5d6cf54..000000000000
--- a/include/asm-sh/rts7751r2d/voyagergx_reg.h
+++ /dev/null
@@ -1,313 +0,0 @@
-/* -------------------------------------------------------------------- */
-/* voyagergx_reg.h                                                      */
-/* -------------------------------------------------------------------- */
-/*  This program is free software; you can redistribute it and/or modify
-    it under the terms of the GNU General Public License as published by
-    the Free Software Foundation; either version 2 of the License, or
-    (at your option) any later version.
-
-    This program is distributed in the hope that it will be useful,
-    but WITHOUT ANY WARRANTY; without even the implied warranty of
-    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-    GNU General Public License for more details.
-
-    You should have received a copy of the GNU General Public License
-    along with this program; if not, write to the Free Software
-    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
-
-    Copyright 2003 (c) Lineo uSolutions,Inc.
-*/
-/* -------------------------------------------------------------------- */
-
-#ifndef _VOYAGER_GX_REG_H
-#define _VOYAGER_GX_REG_H
-
-#define VOYAGER_BASE			0xb3e00000
-#define VOYAGER_USBH_BASE		(0x40000 + VOYAGER_BASE)
-#define VOYAGER_UART_BASE		(0x30000 + VOYAGER_BASE)
-#define	VOYAGER_AC97_BASE		(0xa0000 + VOYAGER_BASE)
-
-#define VOYAGER_IRQ_NUM			32
-#define VOYAGER_IRQ_BASE		50
-#define VOYAGER_USBH_IRQ		VOYAGER_IRQ_BASE + 6
-#define VOYAGER_8051_IRQ		VOYAGER_IRQ_BASE + 10
-#define VOYAGER_UART0_IRQ		VOYAGER_IRQ_BASE + 12
-#define VOYAGER_UART1_IRQ		VOYAGER_IRQ_BASE + 13
-#define	VOYAGER_AC97_IRQ		VOYAGER_IRQ_BASE + 17
-
-/* ----- MISC controle  register ------------------------------ */
-#define MISC_CTRL			(0x000004 + VOYAGER_BASE)
-#define MISC_CTRL_USBCLK_48		(3 << 28)
-#define MISC_CTRL_USBCLK_96		(2 << 28)
-#define MISC_CTRL_USBCLK_CRYSTAL	(1 << 28)
-
-/* ----- GPIO[31:0] register --------------------------------- */
-#define GPIO_MUX_LOW			(0x000008 + VOYAGER_BASE)
-#define GPIO_MUX_LOW_AC97		0x1F000000
-#define GPIO_MUX_LOW_8051		0x0000ffff
-#define GPIO_MUX_LOW_PWM		(1 << 29)
-
-/* ----- GPIO[63:32] register --------------------------------- */
-#define GPIO_MUX_HIGH			(0x00000C + VOYAGER_BASE)
-
-/* ----- DRAM controle  register ------------------------------- */
-#define DRAM_CTRL			(0x000010 + VOYAGER_BASE)
-#define DRAM_CTRL_EMBEDDED		(1 << 31)
-#define DRAM_CTRL_CPU_BURST_1		(0 << 28)
-#define DRAM_CTRL_CPU_BURST_2		(1 << 28)
-#define DRAM_CTRL_CPU_BURST_4		(2 << 28)
-#define DRAM_CTRL_CPU_BURST_8		(3 << 28)
-#define DRAM_CTRL_CPU_CAS_LATENCY	(1 << 27)
-#define DRAM_CTRL_CPU_SIZE_2		(0 << 24)
-#define DRAM_CTRL_CPU_SIZE_4		(1 << 24)
-#define DRAM_CTRL_CPU_SIZE_64		(4 << 24)
-#define DRAM_CTRL_CPU_SIZE_32		(5 << 24)
-#define DRAM_CTRL_CPU_SIZE_16		(6 << 24)
-#define DRAM_CTRL_CPU_SIZE_8		(7 << 24)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_1024	(0 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_512	(2 << 22)
-#define DRAM_CTRL_CPU_COLUMN_SIZE_256	(3 << 22)
-#define DRAM_CTRL_CPU_ACTIVE_PRECHARGE	(1 << 21)
-#define DRAM_CTRL_CPU_RESET		(1 << 20)
-#define DRAM_CTRL_CPU_BANKS		(1 << 19)
-#define DRAM_CTRL_CPU_WRITE_PRECHARGE	(1 << 18)
-#define DRAM_CTRL_BLOCK_WRITE		(1 << 17)
-#define DRAM_CTRL_REFRESH_COMMAND	(1 << 16)
-#define DRAM_CTRL_SIZE_4		(0 << 13)
-#define DRAM_CTRL_SIZE_8		(1 << 13)
-#define DRAM_CTRL_SIZE_16		(2 << 13)
-#define DRAM_CTRL_SIZE_32		(3 << 13)
-#define DRAM_CTRL_SIZE_64		(4 << 13)
-#define DRAM_CTRL_SIZE_2		(5 << 13)
-#define DRAM_CTRL_COLUMN_SIZE_256	(0 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_512	(2 << 11)
-#define DRAM_CTRL_COLUMN_SIZE_1024	(3 << 11)
-#define DRAM_CTRL_BLOCK_WRITE_TIME	(1 << 10)
-#define DRAM_CTRL_BLOCK_WRITE_PRECHARGE	(1 << 9)
-#define DRAM_CTRL_ACTIVE_PRECHARGE	(1 << 8)
-#define DRAM_CTRL_RESET			(1 << 7)
-#define DRAM_CTRL_REMAIN_ACTIVE		(1 << 6)
-#define DRAM_CTRL_BANKS			(1 << 1)
-#define DRAM_CTRL_WRITE_PRECHARGE	(1 << 0)
-
-/* ----- Arvitration control register -------------------------- */
-#define ARBITRATION_CTRL		(0x000014 + VOYAGER_BASE)
-#define ARBITRATION_CTRL_CPUMEM		(1 << 29)
-#define ARBITRATION_CTRL_INTMEM		(1 << 28)
-#define ARBITRATION_CTRL_USB_OFF	(0 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_1	(1 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_2	(2 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_3	(3 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_4	(4 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_5	(5 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_6	(6 << 24)
-#define ARBITRATION_CTRL_USB_PRIORITY_7	(7 << 24)
-#define ARBITRATION_CTRL_PANEL_OFF	(0 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_1	(1 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_2	(2 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_3	(3 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_4	(4 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_5	(5 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_6	(6 << 20)
-#define ARBITRATION_CTRL_PANEL_PRIORITY_7	(7 << 20)
-#define ARBITRATION_CTRL_ZVPORT_OFF	(0 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_1	(1 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_2	(2 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_3	(3 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_4	(4 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_5	(5 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_6	(6 << 16)
-#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_7	(7 << 16)
-#define ARBITRATION_CTRL_CMD_INTPR_OFF	(0 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_1	(1 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_2	(2 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_3	(3 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_4	(4 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_5	(5 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_6	(6 << 12)
-#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_7	(7 << 12)
-#define ARBITRATION_CTRL_DMA_OFF	(0 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_1	(1 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_2	(2 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_3	(3 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_4	(4 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_5	(5 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_6	(6 << 8)
-#define ARBITRATION_CTRL_DMA_PRIORITY_7	(7 << 8)
-#define ARBITRATION_CTRL_VIDEO_OFF	(0 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_1	(1 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_2	(2 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_3	(3 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_4	(4 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_5	(5 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_6	(6 << 4)
-#define ARBITRATION_CTRL_VIDEO_PRIORITY_7	(7 << 4)
-#define ARBITRATION_CTRL_CRT_OFF	(0 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_1	(1 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_2	(2 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_3	(3 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_4	(4 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_5	(5 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_6	(6 << 0)
-#define ARBITRATION_CTRL_CRT_PRIORITY_7	(7 << 0)
-
-/* ----- Command list status register -------------------------- */
-#define CMD_INTPR_STATUS		(0x000024 + VOYAGER_BASE)
-
-/* ----- Interrupt status register ----------------------------- */
-#define INT_STATUS			(0x00002c + VOYAGER_BASE)
-#define INT_STATUS_UH			(1 << 6)
-#define INT_STATUS_MC			(1 << 10)
-#define INT_STATUS_U0			(1 << 12)
-#define INT_STATUS_U1			(1 << 13)
-#define	INT_STATUS_AC			(1 << 17)
-
-/* ----- Interrupt mask register ------------------------------ */
-#define VOYAGER_INT_MASK		(0x000030 + VOYAGER_BASE)
-#define VOYAGER_INT_MASK_AC		(1 << 17)
-
-/* ----- Current Gate register ---------------------------------*/
-#define CURRENT_GATE			(0x000038 + VOYAGER_BASE)
-
-/* ----- Power mode 0 gate register --------------------------- */
-#define POWER_MODE0_GATE		(0x000040 + VOYAGER_BASE)
-#define POWER_MODE0_GATE_G		(1 << 6)
-#define POWER_MODE0_GATE_U0		(1 << 7)
-#define POWER_MODE0_GATE_U1		(1 << 8)
-#define POWER_MODE0_GATE_UH		(1 << 11)
-#define	POWER_MODE0_GATE_AC		(1 << 18)
-
-/* ----- Power mode 1 gate register --------------------------- */
-#define POWER_MODE1_GATE		(0x000048 + VOYAGER_BASE)
-#define POWER_MODE1_GATE_G		(1 << 6)
-#define POWER_MODE1_GATE_U0		(1 << 7)
-#define POWER_MODE1_GATE_U1		(1 << 8)
-#define POWER_MODE1_GATE_UH		(1 << 11)
-#define	POWER_MODE1_GATE_AC		(1 << 18)
-
-/* ----- Power mode 0 clock register -------------------------- */
-#define POWER_MODE0_CLOCK		(0x000044 + VOYAGER_BASE)
-
-/* ----- Power mode 1 clock register -------------------------- */
-#define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
-
-/* ----- Power mode controll register ------------------------- */
-#define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
-
-/* ----- Miscellaneous Timing register ------------------------ */
-#define SYSTEM_DRAM_CTRL		(0x000068 + VOYAGER_BASE)
-
-/* ----- PWM register ------------------------------------------*/
-#define PWM_0				(0x010020 + VOYAGER_BASE)
-#define PWM_0_HC(x)			(((x)&0x0fff)<<20)
-#define PWM_0_LC(x)			(((x)&0x0fff)<<8 )
-#define PWM_0_CLK_DEV(x)		(((x)&0x000f)<<4 )
-#define PWM_0_EN			(1<<0)
-
-/* ----- I2C register ----------------------------------------- */
-#define I2C_BYTECOUNT			(0x010040 + VOYAGER_BASE)
-#define I2C_CONTROL			(0x010041 + VOYAGER_BASE)
-#define I2C_STATUS			(0x010042 + VOYAGER_BASE)
-#define I2C_RESET			(0x010042 + VOYAGER_BASE)
-#define I2C_SADDRESS			(0x010043 + VOYAGER_BASE)
-#define I2C_DATA			(0x010044 + VOYAGER_BASE)
-
-/* ----- Controle register bits ----------------------------------------- */
-#define I2C_CONTROL_E			(1 << 0)
-#define I2C_CONTROL_MODE		(1 << 1)
-#define I2C_CONTROL_STATUS		(1 << 2)
-#define I2C_CONTROL_INT			(1 << 4)
-#define I2C_CONTROL_INTACK		(1 << 5)
-#define I2C_CONTROL_REPEAT		(1 << 6)
-
-/* ----- Status register bits ----------------------------------------- */
-#define I2C_STATUS_BUSY			(1 << 0)
-#define I2C_STATUS_ACK			(1 << 1)
-#define I2C_STATUS_ERROR		(1 << 2)
-#define I2C_STATUS_COMPLETE		(1 << 3)
-
-/* ----- Reset register  ---------------------------------------------- */
-#define I2C_RESET_ERROR			(1 << 2)
-
-/* ----- transmission frequencies ------------------------------------- */
-#define I2C_SADDRESS_SELECT		(1 << 0)
-
-/* ----- Display Controll register ----------------------------------------- */
-#define PANEL_DISPLAY_CTRL		(0x080000 + VOYAGER_BASE)
-#define PANEL_DISPLAY_CTRL_BIAS         (1<<26)
-#define PANEL_PAN_CTRL			(0x080004 + VOYAGER_BASE)
-#define PANEL_COLOR_KEY			(0x080008 + VOYAGER_BASE)
-#define PANEL_FB_ADDRESS		(0x08000C + VOYAGER_BASE)
-#define PANEL_FB_WIDTH			(0x080010 + VOYAGER_BASE)
-#define PANEL_WINDOW_WIDTH		(0x080014 + VOYAGER_BASE)
-#define PANEL_WINDOW_HEIGHT		(0x080018 + VOYAGER_BASE)
-#define PANEL_PLANE_TL			(0x08001C + VOYAGER_BASE)
-#define PANEL_PLANE_BR			(0x080020 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_TOTAL		(0x080024 + VOYAGER_BASE)
-#define PANEL_HORIZONTAL_SYNC		(0x080028 + VOYAGER_BASE)
-#define PANEL_VERTICAL_TOTAL		(0x08002C + VOYAGER_BASE)
-#define PANEL_VERTICAL_SYNC		(0x080030 + VOYAGER_BASE)
-#define PANEL_CURRENT_LINE		(0x080034 + VOYAGER_BASE)
-#define VIDEO_DISPLAY_CTRL		(0x080040 + VOYAGER_BASE)
-#define VIDEO_FB_0_ADDRESS		(0x080044 + VOYAGER_BASE)
-#define VIDEO_FB_WIDTH			(0x080048 + VOYAGER_BASE)
-#define VIDEO_FB_0_LAST_ADDRESS		(0x08004C + VOYAGER_BASE)
-#define VIDEO_PLANE_TL			(0x080050 + VOYAGER_BASE)
-#define VIDEO_PLANE_BR			(0x080054 + VOYAGER_BASE)
-#define VIDEO_SCALE			(0x080058 + VOYAGER_BASE)
-#define VIDEO_INITIAL_SCALE		(0x08005C + VOYAGER_BASE)
-#define VIDEO_YUV_CONSTANTS		(0x080060 + VOYAGER_BASE)
-#define VIDEO_FB_1_ADDRESS		(0x080064 + VOYAGER_BASE)
-#define VIDEO_FB_1_LAST_ADDRESS		(0x080068 + VOYAGER_BASE)
-#define VIDEO_ALPHA_DISPLAY_CTRL	(0x080080 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_ADDRESS		(0x080084 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_WIDTH		(0x080088 + VOYAGER_BASE)
-#define VIDEO_ALPHA_FB_LAST_ADDRESS	(0x08008C + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_TL		(0x080090 + VOYAGER_BASE)
-#define VIDEO_ALPHA_PLANE_BR		(0x080094 + VOYAGER_BASE)
-#define VIDEO_ALPHA_SCALE		(0x080098 + VOYAGER_BASE)
-#define VIDEO_ALPHA_INITIAL_SCALE	(0x08009C + VOYAGER_BASE)
-#define VIDEO_ALPHA_CHROMA_KEY		(0x0800A0 + VOYAGER_BASE)
-#define PANEL_HWC_ADDRESS		(0x0800F0 + VOYAGER_BASE)
-#define PANEL_HWC_LOCATION		(0x0800F4 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_12		(0x0800F8 + VOYAGER_BASE)
-#define PANEL_HWC_COLOR_3		(0x0800FC + VOYAGER_BASE)
-#define ALPHA_DISPLAY_CTRL		(0x080100 + VOYAGER_BASE)
-#define ALPHA_FB_ADDRESS		(0x080104 + VOYAGER_BASE)
-#define ALPHA_FB_WIDTH			(0x080108 + VOYAGER_BASE)
-#define ALPHA_PLANE_TL			(0x08010C + VOYAGER_BASE)
-#define ALPHA_PLANE_BR			(0x080110 + VOYAGER_BASE)
-#define ALPHA_CHROMA_KEY		(0x080114 + VOYAGER_BASE)
-#define CRT_DISPLAY_CTRL		(0x080200 + VOYAGER_BASE)
-#define CRT_FB_ADDRESS			(0x080204 + VOYAGER_BASE)
-#define CRT_FB_WIDTH			(0x080208 + VOYAGER_BASE)
-#define CRT_HORIZONTAL_TOTAL		(0x08020C + VOYAGER_BASE)
-#define CRT_HORIZONTAL_SYNC		(0x080210 + VOYAGER_BASE)
-#define CRT_VERTICAL_TOTAL		(0x080214 + VOYAGER_BASE)
-#define CRT_VERTICAL_SYNC		(0x080218 + VOYAGER_BASE)
-#define CRT_SIGNATURE_ANALYZER		(0x08021C + VOYAGER_BASE)
-#define CRT_CURRENT_LINE		(0x080220 + VOYAGER_BASE)
-#define CRT_MONITOR_DETECT		(0x080224 + VOYAGER_BASE)
-#define CRT_HWC_ADDRESS			(0x080230 + VOYAGER_BASE)
-#define CRT_HWC_LOCATION		(0x080234 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_12		(0x080238 + VOYAGER_BASE)
-#define CRT_HWC_COLOR_3			(0x08023C + VOYAGER_BASE)
-#define CRT_PALETTE_RAM			(0x080400 + VOYAGER_BASE)
-#define PANEL_PALETTE_RAM		(0x080800 + VOYAGER_BASE)
-#define VIDEO_PALETTE_RAM		(0x080C00 + VOYAGER_BASE)
-
-/* ----- 8051 Controle register ----------------------------------------- */
-#define VOYAGER_8051_BASE		(0x000c0000 + VOYAGER_BASE)
-#define VOYAGER_8051_RESET		(0x000b0000 + VOYAGER_BASE)
-#define VOYAGER_8051_SELECT		(0x000b0004 + VOYAGER_BASE)
-#define VOYAGER_8051_CPU_INT		(0x000b000c + VOYAGER_BASE)
-
-/* ----- AC97 Controle register ----------------------------------------- */
-#define AC97_TX_SLOT0			(0x00000000 + VOYAGER_AC97_BASE)
-#define AC97_CONTROL_STATUS		(0x00000080 + VOYAGER_AC97_BASE)
-#define AC97C_READ			(1 << 19)
-#define AC97C_WD_BIT			(1 << 2)
-#define AC97C_INDEX_MASK		0x7f
-/* -------------------------------------------------------------------- */
-
-#endif /* _VOYAGER_GX_REG_H */
diff --git a/include/asm-sh/voyagergx.h b/include/asm-sh/voyagergx.h
new file mode 100644
index 000000000000..99b0807d1c9f
--- /dev/null
+++ b/include/asm-sh/voyagergx.h
@@ -0,0 +1,313 @@
+/* -------------------------------------------------------------------- */
+/* voyagergx.h	                                                      */
+/* -------------------------------------------------------------------- */
+/*  This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+    Copyright 2003 (c) Lineo uSolutions,Inc.
+*/
+/* -------------------------------------------------------------------- */
+
+#ifndef _VOYAGER_GX_REG_H
+#define _VOYAGER_GX_REG_H
+
+#define VOYAGER_BASE			0xb3e00000
+#define VOYAGER_USBH_BASE		(0x40000 + VOYAGER_BASE)
+#define VOYAGER_UART_BASE		(0x30000 + VOYAGER_BASE)
+#define	VOYAGER_AC97_BASE		(0xa0000 + VOYAGER_BASE)
+
+#define VOYAGER_IRQ_NUM			32
+#define VOYAGER_IRQ_BASE		50
+#define VOYAGER_USBH_IRQ		VOYAGER_IRQ_BASE + 6
+#define VOYAGER_8051_IRQ		VOYAGER_IRQ_BASE + 10
+#define VOYAGER_UART0_IRQ		VOYAGER_IRQ_BASE + 12
+#define VOYAGER_UART1_IRQ		VOYAGER_IRQ_BASE + 13
+#define	VOYAGER_AC97_IRQ		VOYAGER_IRQ_BASE + 17
+
+/* ----- MISC controle  register ------------------------------ */
+#define MISC_CTRL			(0x000004 + VOYAGER_BASE)
+#define MISC_CTRL_USBCLK_48		(3 << 28)
+#define MISC_CTRL_USBCLK_96		(2 << 28)
+#define MISC_CTRL_USBCLK_CRYSTAL	(1 << 28)
+
+/* ----- GPIO[31:0] register --------------------------------- */
+#define GPIO_MUX_LOW			(0x000008 + VOYAGER_BASE)
+#define GPIO_MUX_LOW_AC97		0x1F000000
+#define GPIO_MUX_LOW_8051		0x0000ffff
+#define GPIO_MUX_LOW_PWM		(1 << 29)
+
+/* ----- GPIO[63:32] register --------------------------------- */
+#define GPIO_MUX_HIGH			(0x00000C + VOYAGER_BASE)
+
+/* ----- DRAM controle  register ------------------------------- */
+#define DRAM_CTRL			(0x000010 + VOYAGER_BASE)
+#define DRAM_CTRL_EMBEDDED		(1 << 31)
+#define DRAM_CTRL_CPU_BURST_1		(0 << 28)
+#define DRAM_CTRL_CPU_BURST_2		(1 << 28)
+#define DRAM_CTRL_CPU_BURST_4		(2 << 28)
+#define DRAM_CTRL_CPU_BURST_8		(3 << 28)
+#define DRAM_CTRL_CPU_CAS_LATENCY	(1 << 27)
+#define DRAM_CTRL_CPU_SIZE_2		(0 << 24)
+#define DRAM_CTRL_CPU_SIZE_4		(1 << 24)
+#define DRAM_CTRL_CPU_SIZE_64		(4 << 24)
+#define DRAM_CTRL_CPU_SIZE_32		(5 << 24)
+#define DRAM_CTRL_CPU_SIZE_16		(6 << 24)
+#define DRAM_CTRL_CPU_SIZE_8		(7 << 24)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_1024	(0 << 22)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_512	(2 << 22)
+#define DRAM_CTRL_CPU_COLUMN_SIZE_256	(3 << 22)
+#define DRAM_CTRL_CPU_ACTIVE_PRECHARGE	(1 << 21)
+#define DRAM_CTRL_CPU_RESET		(1 << 20)
+#define DRAM_CTRL_CPU_BANKS		(1 << 19)
+#define DRAM_CTRL_CPU_WRITE_PRECHARGE	(1 << 18)
+#define DRAM_CTRL_BLOCK_WRITE		(1 << 17)
+#define DRAM_CTRL_REFRESH_COMMAND	(1 << 16)
+#define DRAM_CTRL_SIZE_4		(0 << 13)
+#define DRAM_CTRL_SIZE_8		(1 << 13)
+#define DRAM_CTRL_SIZE_16		(2 << 13)
+#define DRAM_CTRL_SIZE_32		(3 << 13)
+#define DRAM_CTRL_SIZE_64		(4 << 13)
+#define DRAM_CTRL_SIZE_2		(5 << 13)
+#define DRAM_CTRL_COLUMN_SIZE_256	(0 << 11)
+#define DRAM_CTRL_COLUMN_SIZE_512	(2 << 11)
+#define DRAM_CTRL_COLUMN_SIZE_1024	(3 << 11)
+#define DRAM_CTRL_BLOCK_WRITE_TIME	(1 << 10)
+#define DRAM_CTRL_BLOCK_WRITE_PRECHARGE	(1 << 9)
+#define DRAM_CTRL_ACTIVE_PRECHARGE	(1 << 8)
+#define DRAM_CTRL_RESET			(1 << 7)
+#define DRAM_CTRL_REMAIN_ACTIVE		(1 << 6)
+#define DRAM_CTRL_BANKS			(1 << 1)
+#define DRAM_CTRL_WRITE_PRECHARGE	(1 << 0)
+
+/* ----- Arvitration control register -------------------------- */
+#define ARBITRATION_CTRL		(0x000014 + VOYAGER_BASE)
+#define ARBITRATION_CTRL_CPUMEM		(1 << 29)
+#define ARBITRATION_CTRL_INTMEM		(1 << 28)
+#define ARBITRATION_CTRL_USB_OFF	(0 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_1	(1 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_2	(2 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_3	(3 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_4	(4 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_5	(5 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_6	(6 << 24)
+#define ARBITRATION_CTRL_USB_PRIORITY_7	(7 << 24)
+#define ARBITRATION_CTRL_PANEL_OFF	(0 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_1	(1 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_2	(2 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_3	(3 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_4	(4 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_5	(5 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_6	(6 << 20)
+#define ARBITRATION_CTRL_PANEL_PRIORITY_7	(7 << 20)
+#define ARBITRATION_CTRL_ZVPORT_OFF	(0 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_1	(1 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_2	(2 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_3	(3 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_4	(4 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_5	(5 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_6	(6 << 16)
+#define ARBITRATION_CTRL_ZVPORTL_PRIORITY_7	(7 << 16)
+#define ARBITRATION_CTRL_CMD_INTPR_OFF	(0 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_1	(1 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_2	(2 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_3	(3 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_4	(4 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_5	(5 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_6	(6 << 12)
+#define ARBITRATION_CTRL_CMD_INTPR_PRIORITY_7	(7 << 12)
+#define ARBITRATION_CTRL_DMA_OFF	(0 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_1	(1 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_2	(2 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_3	(3 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_4	(4 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_5	(5 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_6	(6 << 8)
+#define ARBITRATION_CTRL_DMA_PRIORITY_7	(7 << 8)
+#define ARBITRATION_CTRL_VIDEO_OFF	(0 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_1	(1 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_2	(2 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_3	(3 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_4	(4 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_5	(5 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_6	(6 << 4)
+#define ARBITRATION_CTRL_VIDEO_PRIORITY_7	(7 << 4)
+#define ARBITRATION_CTRL_CRT_OFF	(0 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_1	(1 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_2	(2 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_3	(3 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_4	(4 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_5	(5 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_6	(6 << 0)
+#define ARBITRATION_CTRL_CRT_PRIORITY_7	(7 << 0)
+
+/* ----- Command list status register -------------------------- */
+#define CMD_INTPR_STATUS		(0x000024 + VOYAGER_BASE)
+
+/* ----- Interrupt status register ----------------------------- */
+#define INT_STATUS			(0x00002c + VOYAGER_BASE)
+#define INT_STATUS_UH			(1 << 6)
+#define INT_STATUS_MC			(1 << 10)
+#define INT_STATUS_U0			(1 << 12)
+#define INT_STATUS_U1			(1 << 13)
+#define	INT_STATUS_AC			(1 << 17)
+
+/* ----- Interrupt mask register ------------------------------ */
+#define VOYAGER_INT_MASK		(0x000030 + VOYAGER_BASE)
+#define VOYAGER_INT_MASK_AC		(1 << 17)
+
+/* ----- Current Gate register ---------------------------------*/
+#define CURRENT_GATE			(0x000038 + VOYAGER_BASE)
+
+/* ----- Power mode 0 gate register --------------------------- */
+#define POWER_MODE0_GATE		(0x000040 + VOYAGER_BASE)
+#define POWER_MODE0_GATE_G		(1 << 6)
+#define POWER_MODE0_GATE_U0		(1 << 7)
+#define POWER_MODE0_GATE_U1		(1 << 8)
+#define POWER_MODE0_GATE_UH		(1 << 11)
+#define	POWER_MODE0_GATE_AC		(1 << 18)
+
+/* ----- Power mode 1 gate register --------------------------- */
+#define POWER_MODE1_GATE		(0x000048 + VOYAGER_BASE)
+#define POWER_MODE1_GATE_G		(1 << 6)
+#define POWER_MODE1_GATE_U0		(1 << 7)
+#define POWER_MODE1_GATE_U1		(1 << 8)
+#define POWER_MODE1_GATE_UH		(1 << 11)
+#define	POWER_MODE1_GATE_AC		(1 << 18)
+
+/* ----- Power mode 0 clock register -------------------------- */
+#define POWER_MODE0_CLOCK		(0x000044 + VOYAGER_BASE)
+
+/* ----- Power mode 1 clock register -------------------------- */
+#define POWER_MODE1_CLOCK		(0x00004C + VOYAGER_BASE)
+
+/* ----- Power mode controll register ------------------------- */
+#define POWER_MODE_CTRL			(0x000054 + VOYAGER_BASE)
+
+/* ----- Miscellaneous Timing register ------------------------ */
+#define SYSTEM_DRAM_CTRL		(0x000068 + VOYAGER_BASE)
+
+/* ----- PWM register ------------------------------------------*/
+#define PWM_0				(0x010020 + VOYAGER_BASE)
+#define PWM_0_HC(x)			(((x)&0x0fff)<<20)
+#define PWM_0_LC(x)			(((x)&0x0fff)<<8 )
+#define PWM_0_CLK_DEV(x)		(((x)&0x000f)<<4 )
+#define PWM_0_EN			(1<<0)
+
+/* ----- I2C register ----------------------------------------- */
+#define I2C_BYTECOUNT			(0x010040 + VOYAGER_BASE)
+#define I2C_CONTROL			(0x010041 + VOYAGER_BASE)
+#define I2C_STATUS			(0x010042 + VOYAGER_BASE)
+#define I2C_RESET			(0x010042 + VOYAGER_BASE)
+#define I2C_SADDRESS			(0x010043 + VOYAGER_BASE)
+#define I2C_DATA			(0x010044 + VOYAGER_BASE)
+
+/* ----- Controle register bits ----------------------------------------- */
+#define I2C_CONTROL_E			(1 << 0)
+#define I2C_CONTROL_MODE		(1 << 1)
+#define I2C_CONTROL_STATUS		(1 << 2)
+#define I2C_CONTROL_INT			(1 << 4)
+#define I2C_CONTROL_INTACK		(1 << 5)
+#define I2C_CONTROL_REPEAT		(1 << 6)
+
+/* ----- Status register bits ----------------------------------------- */
+#define I2C_STATUS_BUSY			(1 << 0)
+#define I2C_STATUS_ACK			(1 << 1)
+#define I2C_STATUS_ERROR		(1 << 2)
+#define I2C_STATUS_COMPLETE		(1 << 3)
+
+/* ----- Reset register  ---------------------------------------------- */
+#define I2C_RESET_ERROR			(1 << 2)
+
+/* ----- transmission frequencies ------------------------------------- */
+#define I2C_SADDRESS_SELECT		(1 << 0)
+
+/* ----- Display Controll register ----------------------------------------- */
+#define PANEL_DISPLAY_CTRL		(0x080000 + VOYAGER_BASE)
+#define PANEL_DISPLAY_CTRL_BIAS         (1<<26)
+#define PANEL_PAN_CTRL			(0x080004 + VOYAGER_BASE)
+#define PANEL_COLOR_KEY			(0x080008 + VOYAGER_BASE)
+#define PANEL_FB_ADDRESS		(0x08000C + VOYAGER_BASE)
+#define PANEL_FB_WIDTH			(0x080010 + VOYAGER_BASE)
+#define PANEL_WINDOW_WIDTH		(0x080014 + VOYAGER_BASE)
+#define PANEL_WINDOW_HEIGHT		(0x080018 + VOYAGER_BASE)
+#define PANEL_PLANE_TL			(0x08001C + VOYAGER_BASE)
+#define PANEL_PLANE_BR			(0x080020 + VOYAGER_BASE)
+#define PANEL_HORIZONTAL_TOTAL		(0x080024 + VOYAGER_BASE)
+#define PANEL_HORIZONTAL_SYNC		(0x080028 + VOYAGER_BASE)
+#define PANEL_VERTICAL_TOTAL		(0x08002C + VOYAGER_BASE)
+#define PANEL_VERTICAL_SYNC		(0x080030 + VOYAGER_BASE)
+#define PANEL_CURRENT_LINE		(0x080034 + VOYAGER_BASE)
+#define VIDEO_DISPLAY_CTRL		(0x080040 + VOYAGER_BASE)
+#define VIDEO_FB_0_ADDRESS		(0x080044 + VOYAGER_BASE)
+#define VIDEO_FB_WIDTH			(0x080048 + VOYAGER_BASE)
+#define VIDEO_FB_0_LAST_ADDRESS		(0x08004C + VOYAGER_BASE)
+#define VIDEO_PLANE_TL			(0x080050 + VOYAGER_BASE)
+#define VIDEO_PLANE_BR			(0x080054 + VOYAGER_BASE)
+#define VIDEO_SCALE			(0x080058 + VOYAGER_BASE)
+#define VIDEO_INITIAL_SCALE		(0x08005C + VOYAGER_BASE)
+#define VIDEO_YUV_CONSTANTS		(0x080060 + VOYAGER_BASE)
+#define VIDEO_FB_1_ADDRESS		(0x080064 + VOYAGER_BASE)
+#define VIDEO_FB_1_LAST_ADDRESS		(0x080068 + VOYAGER_BASE)
+#define VIDEO_ALPHA_DISPLAY_CTRL	(0x080080 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_ADDRESS		(0x080084 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_WIDTH		(0x080088 + VOYAGER_BASE)
+#define VIDEO_ALPHA_FB_LAST_ADDRESS	(0x08008C + VOYAGER_BASE)
+#define VIDEO_ALPHA_PLANE_TL		(0x080090 + VOYAGER_BASE)
+#define VIDEO_ALPHA_PLANE_BR		(0x080094 + VOYAGER_BASE)
+#define VIDEO_ALPHA_SCALE		(0x080098 + VOYAGER_BASE)
+#define VIDEO_ALPHA_INITIAL_SCALE	(0x08009C + VOYAGER_BASE)
+#define VIDEO_ALPHA_CHROMA_KEY		(0x0800A0 + VOYAGER_BASE)
+#define PANEL_HWC_ADDRESS		(0x0800F0 + VOYAGER_BASE)
+#define PANEL_HWC_LOCATION		(0x0800F4 + VOYAGER_BASE)
+#define PANEL_HWC_COLOR_12		(0x0800F8 + VOYAGER_BASE)
+#define PANEL_HWC_COLOR_3		(0x0800FC + VOYAGER_BASE)
+#define ALPHA_DISPLAY_CTRL		(0x080100 + VOYAGER_BASE)
+#define ALPHA_FB_ADDRESS		(0x080104 + VOYAGER_BASE)
+#define ALPHA_FB_WIDTH			(0x080108 + VOYAGER_BASE)
+#define ALPHA_PLANE_TL			(0x08010C + VOYAGER_BASE)
+#define ALPHA_PLANE_BR			(0x080110 + VOYAGER_BASE)
+#define ALPHA_CHROMA_KEY		(0x080114 + VOYAGER_BASE)
+#define CRT_DISPLAY_CTRL		(0x080200 + VOYAGER_BASE)
+#define CRT_FB_ADDRESS			(0x080204 + VOYAGER_BASE)
+#define CRT_FB_WIDTH			(0x080208 + VOYAGER_BASE)
+#define CRT_HORIZONTAL_TOTAL		(0x08020C + VOYAGER_BASE)
+#define CRT_HORIZONTAL_SYNC		(0x080210 + VOYAGER_BASE)
+#define CRT_VERTICAL_TOTAL		(0x080214 + VOYAGER_BASE)
+#define CRT_VERTICAL_SYNC		(0x080218 + VOYAGER_BASE)
+#define CRT_SIGNATURE_ANALYZER		(0x08021C + VOYAGER_BASE)
+#define CRT_CURRENT_LINE		(0x080220 + VOYAGER_BASE)
+#define CRT_MONITOR_DETECT		(0x080224 + VOYAGER_BASE)
+#define CRT_HWC_ADDRESS			(0x080230 + VOYAGER_BASE)
+#define CRT_HWC_LOCATION		(0x080234 + VOYAGER_BASE)
+#define CRT_HWC_COLOR_12		(0x080238 + VOYAGER_BASE)
+#define CRT_HWC_COLOR_3			(0x08023C + VOYAGER_BASE)
+#define CRT_PALETTE_RAM			(0x080400 + VOYAGER_BASE)
+#define PANEL_PALETTE_RAM		(0x080800 + VOYAGER_BASE)
+#define VIDEO_PALETTE_RAM		(0x080C00 + VOYAGER_BASE)
+
+/* ----- 8051 Controle register ----------------------------------------- */
+#define VOYAGER_8051_BASE		(0x000c0000 + VOYAGER_BASE)
+#define VOYAGER_8051_RESET		(0x000b0000 + VOYAGER_BASE)
+#define VOYAGER_8051_SELECT		(0x000b0004 + VOYAGER_BASE)
+#define VOYAGER_8051_CPU_INT		(0x000b000c + VOYAGER_BASE)
+
+/* ----- AC97 Controle register ----------------------------------------- */
+#define AC97_TX_SLOT0			(0x00000000 + VOYAGER_AC97_BASE)
+#define AC97_CONTROL_STATUS		(0x00000080 + VOYAGER_AC97_BASE)
+#define AC97C_READ			(1 << 19)
+#define AC97C_WD_BIT			(1 << 2)
+#define AC97C_INDEX_MASK		0x7f
+/* -------------------------------------------------------------------- */
+
+#endif /* _VOYAGER_GX_REG_H */
-- 
cgit v1.2.1


From e7f93a355c7e32c26eab8910cf53b7506bb046c5 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:19:13 +0900
Subject: sh: Make PAGE_OFFSET configurable.

nommu needs to be able to shift PAGE_OFFSET, so we switch it to a
non-user-visible CONFIG_PAGE_OFFSET and use that in the few places
where it matters.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  | 26 --------------------------
 arch/sh/boot/compressed/Makefile |  3 ++-
 arch/sh/kernel/vmlinux.lds.S     |  2 +-
 arch/sh/mm/Kconfig               | 31 +++++++++++++++++++++++++++++++
 include/asm-sh/page.h            |  2 +-
 5 files changed, 35 insertions(+), 29 deletions(-)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 07d6d699e9c2..035df7fb2237 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -238,32 +238,6 @@ endchoice
 
 source "arch/sh/mm/Kconfig"
 
-config MEMORY_START
-	hex "Physical memory start address"
-	default "0x08000000"
-	---help---
-	  Computers built with Hitachi SuperH processors always
-	  map the ROM starting at address zero.  But the processor
-	  does not specify the range that RAM takes.
-
-	  The physical memory (RAM) start address will be automatically
-	  set to 08000000. Other platforms, such as the Solution Engine
-	  boards typically map RAM at 0C000000.
-
-	  Tweak this only when porting to a new machine which does not
-	  already have a defconfig. Changing it from the known correct
-	  value on any of the known systems will only lead to disaster.
-
-config MEMORY_SIZE
-	hex "Physical memory size"
-	default "0x00400000"
-	help
-	  This sets the default memory size assumed by your SH kernel. It can
-	  be overridden as normal by the 'mem=' argument on the kernel command
-	  line. If unsure, consult your board specifications or just leave it
-	  as 0x00400000 which was the default value before this became
-	  configurable.
-
 config CF_ENABLER
 	bool "Compact Flash Enabler support"
 	depends on SH_ADX || SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_CAT68701 || SH_SH03
diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 75a6876bf6c6..7074267c01b7 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -18,9 +18,10 @@ endif
 # Assign dummy values if these 2 variables are not defined,
 # in order to suppress error message.
 #
+CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
-IMAGE_OFFSET := $(shell printf "0x%8x" $$[0x80000000+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
+IMAGE_OFFSET := $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
 
 LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup -T $(obj)/../../kernel/vmlinux.lds
 
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 3f3e1e0735f5..eb860c51c697 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -13,7 +13,7 @@ OUTPUT_ARCH(sh)
 ENTRY(_start)
 SECTIONS
 {
-  . = 0x80000000 + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
+  . = CONFIG_PAGE_OFFSET + CONFIG_MEMORY_START + CONFIG_ZERO_PAGE_OFFSET;
   _text = .;			/* Text and read-only data */
   text = .;			/* Text and read-only data */
   .empty_zero_page : {
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index 9064dfc91e6d..f25903aa607b 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -161,6 +161,37 @@ config MMU
 	  turning this off will boot the kernel on these machines with the
 	  MMU implicitly switched off.
 
+config PAGE_OFFSET
+	hex
+	default "0x80000000" if MMU
+	default "0x00000000"
+
+config MEMORY_START
+	hex "Physical memory start address"
+	default "0x08000000"
+	---help---
+	  Computers built with Hitachi SuperH processors always
+	  map the ROM starting at address zero.  But the processor
+	  does not specify the range that RAM takes.
+
+	  The physical memory (RAM) start address will be automatically
+	  set to 08000000. Other platforms, such as the Solution Engine
+	  boards typically map RAM at 0C000000.
+
+	  Tweak this only when porting to a new machine which does not
+	  already have a defconfig. Changing it from the known correct
+	  value on any of the known systems will only lead to disaster.
+
+config MEMORY_SIZE
+	hex "Physical memory size"
+	default "0x00400000"
+	help
+	  This sets the default memory size assumed by your SH kernel. It can
+	  be overridden as normal by the 'mem=' argument on the kernel command
+	  line. If unsure, consult your board specifications or just leave it
+	  as 0x00400000 which was the default value before this became
+	  configurable.
+
 config 32BIT
 	bool "Support 32-bit physical addressing through PMB"
 	depends on CPU_SH4A
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 51d7281a546a..1b3cfd165a66 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -84,7 +84,7 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #define __MEMORY_START		CONFIG_MEMORY_START
 #define __MEMORY_SIZE		CONFIG_MEMORY_SIZE
 
-#define PAGE_OFFSET		(0x80000000UL)
+#define PAGE_OFFSET		CONFIG_PAGE_OFFSET
 #define __pa(x)			((unsigned long)(x)-PAGE_OFFSET)
 #define __va(x)			((void *)((unsigned long)(x)+PAGE_OFFSET))
 
-- 
cgit v1.2.1


From e96636ccfa373a00a0ee0558e1971baa7856d8b5 Mon Sep 17 00:00:00 2001
From: Yoshinori Sato <ysato@users.sourceforge.jp>
Date: Wed, 27 Sep 2006 17:21:02 +0900
Subject: sh: Various nommu fixes.

This fixes up some of the various outstanding nommu bugs on
SH.

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boot/compressed/Makefile | 10 ++++++++--
 arch/sh/mm/init.c                |  5 +++++
 arch/sh/mm/pg-nommu.c            | 17 +++++++++--------
 include/asm-sh/addrspace.h       |  8 ++++++++
 include/asm-sh/flat.h            |  2 +-
 include/asm-sh/mmu.h             | 13 +------------
 include/asm-sh/page.h            |  5 +++++
 include/asm-sh/uaccess.h         | 13 +++++++++++++
 8 files changed, 50 insertions(+), 23 deletions(-)

diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 7074267c01b7..903470429cb4 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -21,11 +21,17 @@ endif
 CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
-IMAGE_OFFSET := $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)+$(CONFIG_MEMORY_START)+$(CONFIG_BOOT_LINK_OFFSET)])
+
+IMAGE_OFFSET	:= $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)  + \
+					     $(CONFIG_MEMORY_START) + \
+					     $(CONFIG_BOOT_LINK_OFFSET)])
+
+LIBGCC	:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
 LDFLAGS_vmlinux := -Ttext $(IMAGE_OFFSET) -e startup -T $(obj)/../../kernel/vmlinux.lds
 
-$(obj)/vmlinux: $(OBJECTS) $(obj)/piggy.o FORCE
+
+$(obj)/vmlinux: $(OBJECTS) $(obj)/piggy.o $(LIBGCC) FORCE
 	$(call if_changed,ld)
 	@:
 
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index d1a979eab656..2c6dc3d8df2c 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -248,8 +248,13 @@ void __init mem_init(void)
 	 * Setup wrappers for copy/clear_page(), these will get overridden
 	 * later in the boot process if a better method is available.
 	 */
+#ifdef CONFIG_MMU
 	copy_page = copy_page_slow;
 	clear_page = clear_page_slow;
+#else
+	copy_page = copy_page_nommu;
+	clear_page = clear_page_nommu;
+#endif
 
 	/* this will put all low memory onto the freelists */
 	totalram_pages += free_all_bootmem_node(NODE_DATA(0));
diff --git a/arch/sh/mm/pg-nommu.c b/arch/sh/mm/pg-nommu.c
index 8f9165a4e333..d15221beaa16 100644
--- a/arch/sh/mm/pg-nommu.c
+++ b/arch/sh/mm/pg-nommu.c
@@ -14,23 +14,24 @@
 #include <linux/string.h>
 #include <asm/page.h>
 
-static void copy_page_nommu(void *to, void *from)
+void copy_page_nommu(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
 
-static void clear_page_nommu(void *to)
+void clear_page_nommu(void *to)
 {
 	memset(to, 0, PAGE_SIZE);
 }
 
-static int __init pg_nommu_init(void)
+__kernel_size_t __copy_user(void *to, const void *from, __kernel_size_t n)
 {
-	copy_page = copy_page_nommu;
-	clear_page = clear_page_nommu;
-
+	memcpy(to, from, n);
 	return 0;
 }
 
-subsys_initcall(pg_nommu_init);
-
+__kernel_size_t __clear_user(void *to, __kernel_size_t n)
+{
+	memset(to, 0, n);
+	return 0;
+}
diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index 720afc11c2ca..4207368267b1 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -14,11 +14,19 @@
 #include <asm/cpu/addrspace.h>
 
 /* Memory segments (32bit Privileged mode addresses)  */
+#ifdef CONFIG_MMU
 #define P0SEG		0x00000000
 #define P1SEG		0x80000000
 #define P2SEG		0xa0000000
 #define P3SEG		0xc0000000
 #define P4SEG		0xe0000000
+#else
+#define P0SEG		0x00000000
+#define P1SEG		0x00000000
+#define P2SEG		0x20000000
+#define P3SEG		0x40000000
+#define P4SEG 		0x80000000
+#endif
 
 /* Returns the privileged segment base of a given address  */
 #define PXSEG(a)	(((unsigned long)(a)) & 0xe0000000)
diff --git a/include/asm-sh/flat.h b/include/asm-sh/flat.h
index f29072e1c87e..0d5cc04ab005 100644
--- a/include/asm-sh/flat.h
+++ b/include/asm-sh/flat.h
@@ -13,7 +13,7 @@
 #define __ASM_SH_FLAT_H
 
 #define	flat_stack_align(sp)			/* nothing needed */
-#define	flat_argvp_envp_on_stack()		1
+#define	flat_argvp_envp_on_stack()		0
 #define	flat_old_ram_flag(flags)		(flags)
 #define	flat_reloc_valid(reloc, size)		((reloc) <= (size))
 #define	flat_get_addr_from_rp(rp, relval, flags)	get_unaligned(rp)
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index ec09589fa6ca..6383dc84501e 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -3,19 +3,8 @@
 
 #if !defined(CONFIG_MMU)
 
-struct mm_rblock_struct {
-	int	size;
-	int	refcount;
-	void	*kblock;
-};
-
-struct mm_tblock_struct {
-	struct mm_rblock_struct *rblock;
-	struct mm_tblock_struct *next;
-};
-
 typedef struct {
-	struct mm_tblock_struct tblock;
+	struct vm_list_struct	*vmlist;
 	unsigned long		end_brk;
 } mm_context_t;
 
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 1b3cfd165a66..e9135532d00c 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -38,8 +38,13 @@
 extern void (*clear_page)(void *to);
 extern void (*copy_page)(void *to, void *from);
 
+#ifdef CONFIG_MMU
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
+#else
+extern void clear_page_nommu(void *to);
+extern void copy_page_nommu(void *to, void *from);
+#endif
 
 #if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
 	defined(CONFIG_SH7705_CACHE_32KB))
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 6c0014dd2ef7..5c3b00c2f107 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -168,6 +168,7 @@ do {								\
 	__gu_err;						\
 })
 
+#ifdef CONFIG_MMU
 #define __get_user_check(x,ptr,size)				\
 ({								\
 	long __gu_err, __gu_val;				\
@@ -257,6 +258,18 @@ __asm__("stc	r7_bank, %1\n\t"		\
 	: "r" (addr)				\
 	: "t");					\
 })
+#else /* CONFIG_MMU */
+#define __get_user_check(x,ptr,size)					\
+({									\
+	long __gu_err, __gu_val;					\
+	if (__access_ok((unsigned long)(ptr), (size))) {		\
+		__get_user_size(__gu_val, (ptr), (size), __gu_err);	\
+		(x) = (__typeof__(*(ptr)))__gu_val;			\
+	} else								\
+		__gu_err = -EFAULT;					\
+	__gu_err;							\
+})
+#endif
 
 #define __get_user_asm(x, addr, err, insn) \
 ({ \
-- 
cgit v1.2.1


From 0b8929354cdeddb17e81bfda903812c9adfd0b67 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:22:49 +0900
Subject: sh: __NR_restart_syscall support.

This implements support for __NR_restart_syscall.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/entry.S    |  2 ++
 arch/sh/kernel/signal.c   | 16 ++++++++++------
 arch/sh/kernel/syscalls.S |  2 +-
 3 files changed, 13 insertions(+), 7 deletions(-)

diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index b087d34dba35..fcbf50d4ee10 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -376,6 +376,7 @@ work_notifysig:
 	bt/s	restore_all
 	 mov	r15, r4
 	mov	#0, r5
+	mov	r12, r6		! set arg2(save_r0)
 	mov.l	2f, r1
 	mova	restore_all, r0
 	jmp	@r1
@@ -534,6 +535,7 @@ syscall_call:
 	mov.l	@r9, r8
 	jsr	@r8	    	! jump to specific syscall handler
 	 nop
+	mov.l	@(OFF_R0,r15), r12		! save r0
 	mov.l	r0, @(OFF_R0,r15)		! save the return value
 	!
 syscall_exit:
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index b475c4d2405f..eb6a3b97e46c 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -33,7 +33,8 @@
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
+asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset,
+			 unsigned int save_r0);
 
 /*
  * Atomically swap in the new signal mask, and wait for a signal.
@@ -56,7 +57,7 @@ sys_sigsuspend(old_sigset_t mask,
 	while (1) {
 		current->state = TASK_INTERRUPTIBLE;
 		schedule();
-		if (do_signal(&regs, &saveset))
+		if (do_signal(&regs, &saveset, regs.regs[0]))
 			return -EINTR;
 	}
 }
@@ -85,7 +86,7 @@ sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
 	while (1) {
 		current->state = TASK_INTERRUPTIBLE;
 		schedule();
-		if (do_signal(&regs, &saveset))
+		if (do_signal(&regs, &saveset, regs.regs[0]))
 			return -EINTR;
 	}
 }
@@ -563,7 +564,7 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
  * the kernel can handle, and then we build all the user-level signal handling
  * stack-frames in one go after that.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
+int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
 {
 	siginfo_t info;
 	int signr;
@@ -597,9 +598,12 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
 		/* Restart the system call - no handlers present */
 		if (regs->regs[0] == -ERESTARTNOHAND ||
 		    regs->regs[0] == -ERESTARTSYS ||
-		    regs->regs[0] == -ERESTARTNOINTR ||
-		    regs->regs[0] == -ERESTART_RESTARTBLOCK) {
+		    regs->regs[0] == -ERESTARTNOINTR) {
+		    	regs->regs[0] = save_r0;
 			regs->pc -= 2;
+		} else if (regs->regs[0] == -ERESTART_RESTARTBLOCK) {
+			regs->pc -= 2;
+			regs->regs[3] = __NR_restart_syscall;
 		}
 	}
 	return 0;
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
index ada61b0d9a73..ea23b213c77f 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -34,7 +34,7 @@
 
 	.data
 ENTRY(sys_call_table)
-	.long sys_ni_syscall	/* 0  -  old "setup()" system call*/
+	.long sys_restart_syscall	/* 0  -  old "setup()" system call*/
 	.long sys_exit
 	.long sys_fork
 	.long sys_read
-- 
cgit v1.2.1


From a2d1a5fae6296c2a3ac1aaa982c95464c46c0585 Mon Sep 17 00:00:00 2001
From: Yoshinori Sato <ysato@users.sourceforge.jp>
Date: Wed, 27 Sep 2006 17:25:07 +0900
Subject: sh: __addr_ok() and other misc nommu fixups.

A few more outstanding nommu fixups..

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boot/compressed/Makefile |  6 +++---
 arch/sh/kernel/process.c         | 13 +++++++++----
 arch/sh/kernel/sh_ksyms.c        | 10 ++++------
 arch/sh/kernel/sys_sh.c          |  2 +-
 arch/sh/mm/Kconfig               |  2 +-
 arch/sh/mm/Makefile              |  4 ++--
 include/asm-sh/addrspace.h       |  4 ++--
 include/asm-sh/io.h              |  5 +++++
 include/asm-sh/uaccess.h         |  9 ++++++---
 9 files changed, 33 insertions(+), 22 deletions(-)

diff --git a/arch/sh/boot/compressed/Makefile b/arch/sh/boot/compressed/Makefile
index 903470429cb4..e5f443790079 100644
--- a/arch/sh/boot/compressed/Makefile
+++ b/arch/sh/boot/compressed/Makefile
@@ -22,9 +22,9 @@ CONFIG_PAGE_OFFSET	?= 0x80000000
 CONFIG_MEMORY_START     ?= 0x0c000000
 CONFIG_BOOT_LINK_OFFSET ?= 0x00800000
 
-IMAGE_OFFSET	:= $(shell printf "0x%8x" $$[$(CONFIG_PAGE_OFFSET)  + \
-					     $(CONFIG_MEMORY_START) + \
-					     $(CONFIG_BOOT_LINK_OFFSET)])
+IMAGE_OFFSET	:= $(shell printf "0x%08x" $$[$(CONFIG_PAGE_OFFSET)  + \
+					      $(CONFIG_MEMORY_START) + \
+					      $(CONFIG_BOOT_LINK_OFFSET)])
 
 LIBGCC	:= $(shell $(CC) $(CFLAGS) -print-libgcc-file-name)
 
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 4a32550fd7c6..c4aa687ba26a 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -302,9 +302,11 @@ ubc_set_tracing(int asid, unsigned long pc)
 {
 	ctrl_outl(pc, UBC_BARA);
 
+#ifdef CONFIG_MMU
 	/* We don't have any ASID settings for the SH-2! */
 	if (cpu_data->type != CPU_SH7604)
 		ctrl_outb(asid, UBC_BASRA);
+#endif
 
 	ctrl_outl(0, UBC_BAMRA);
 
@@ -347,6 +349,7 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	}
 #endif
 
+#ifdef CONFIG_MMU
 	/*
 	 * Restore the kernel mode register
 	 *   	k7 (r7_bank1)
@@ -354,19 +357,21 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	asm volatile("ldc	%0, r7_bank"
 		     : /* no output */
 		     : "r" (task_thread_info(next)));
+#endif
 
-#ifdef CONFIG_MMU
 	/* If no tasks are using the UBC, we're done */
 	if (ubc_usercnt == 0)
 		/* If no tasks are using the UBC, we're done */;
 	else if (next->thread.ubc_pc && next->mm) {
-		ubc_set_tracing(next->mm->context & MMU_CONTEXT_ASID_MASK,
-				next->thread.ubc_pc);
+		int asid = 0;
+#ifdef CONFIG_MMU
+		asid |= next->mm->context & MMU_CONTEXT_ASID_MASK;
+#endif
+		ubc_set_tracing(asid, next->thread.ubc_pc);
 	} else {
 		ctrl_outw(0, UBC_BBRA);
 		ctrl_outw(0, UBC_BBRB);
 	}
-#endif
 
 	return prev;
 }
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index 7f3a42244cab..bf59d73415d7 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -79,20 +79,18 @@ EXPORT_SYMBOL(strcpy);
 DECLARE_EXPORT(__movstr_i4_even);
 DECLARE_EXPORT(__movstr_i4_odd);
 DECLARE_EXPORT(__movstrSI12_i4);
+#endif
 
+#if defined(CONFIG_CPU_SH4) || defined(CONFIG_SH7705_CACHE_32KB)
 /* needed by some modules */
 EXPORT_SYMBOL(flush_cache_all);
 EXPORT_SYMBOL(flush_cache_range);
 EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
-EXPORT_SYMBOL(clear_user_page);
 #endif
 
-#if defined(CONFIG_SH7705_CACHE_32KB)
-EXPORT_SYMBOL(flush_cache_all);
-EXPORT_SYMBOL(flush_cache_range);
-EXPORT_SYMBOL(flush_dcache_page);
-EXPORT_SYMBOL(__flush_purge_region);
+#ifdef CONFIG_MMU
+EXPORT_SYMBOL(clear_user_page);
 #endif
 
 EXPORT_SYMBOL(flush_tlb_page);
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index d8bcd8a22327..0ee7bf4cb238 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -44,7 +44,7 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	return error;
 }
 
-#if defined(HAVE_ARCH_UNMAPPED_AREA)
+#if defined(HAVE_ARCH_UNMAPPED_AREA) && defined(CONFIG_MMU)
 /*
  * To avoid cache alias, we map the shard page with same color.
  */
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index f25903aa607b..bed697c0dc19 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -194,7 +194,7 @@ config MEMORY_SIZE
 
 config 32BIT
 	bool "Support 32-bit physical addressing through PMB"
-	depends on CPU_SH4A
+	depends on CPU_SH4A && MMU
 	default y
 	help
 	  If you say Y here, physical addressing will be extended to
diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index 87a7c07265c0..da37d86e65eb 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -6,7 +6,7 @@ obj-y			:= init.o extable.o consistent.o
 
 obj-$(CONFIG_CPU_SH2)	+= cache-sh2.o
 obj-$(CONFIG_CPU_SH3)	+= cache-sh3.o
-obj-$(CONFIG_CPU_SH4)	+= cache-sh4.o pg-sh4.o
+obj-$(CONFIG_CPU_SH4)	+= cache-sh4.o
 
 obj-$(CONFIG_DMA_PAGE_OPS)	+= pg-dma.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
@@ -19,7 +19,7 @@ obj-y			+= $(mmu-y)
 
 ifdef CONFIG_MMU
 obj-$(CONFIG_CPU_SH3)		+= tlb-sh3.o
-obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o
+obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o pg-sh4.o
 obj-$(CONFIG_SH7705_CACHE_32KB) += pg-sh7705.o
 endif
 
diff --git a/include/asm-sh/addrspace.h b/include/asm-sh/addrspace.h
index 4207368267b1..b860218e402e 100644
--- a/include/asm-sh/addrspace.h
+++ b/include/asm-sh/addrspace.h
@@ -14,7 +14,7 @@
 #include <asm/cpu/addrspace.h>
 
 /* Memory segments (32bit Privileged mode addresses)  */
-#ifdef CONFIG_MMU
+#ifndef CONFIG_CPU_SH2A
 #define P0SEG		0x00000000
 #define P1SEG		0x80000000
 #define P2SEG		0xa0000000
@@ -24,7 +24,7 @@
 #define P0SEG		0x00000000
 #define P1SEG		0x00000000
 #define P2SEG		0x20000000
-#define P3SEG		0x40000000
+#define P3SEG		0x00000000
 #define P4SEG 		0x80000000
 #endif
 
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 27dba653cbe7..377160b86295 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -216,6 +216,7 @@ static inline void ctrl_delay(void)
 
 #define IO_SPACE_LIMIT 0xffffffff
 
+#ifdef CONFIG_MMU
 /*
  * Change virtual addresses to physical addresses and vv.
  * These are trivial on the 1:1 Linux/SuperH mapping
@@ -229,6 +230,10 @@ static inline void *phys_to_virt(unsigned long address)
 {
 	return (void *)P1SEGADDR(address);
 }
+#else
+#define phys_to_virt(address)	((void *)(address))
+#define virt_to_phys(address)	((unsigned long)(address))
+#endif
 
 #define virt_to_bus virt_to_phys
 #define bus_to_virt phys_to_virt
diff --git a/include/asm-sh/uaccess.h b/include/asm-sh/uaccess.h
index 5c3b00c2f107..5c49ed6715f2 100644
--- a/include/asm-sh/uaccess.h
+++ b/include/asm-sh/uaccess.h
@@ -34,12 +34,12 @@
 
 #define segment_eq(a,b)	((a).seg == (b).seg)
 
-#define __addr_ok(addr) \
-	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
-
 #define get_ds()	(KERNEL_DS)
 
 #if !defined(CONFIG_MMU)
+/* NOMMU is always true */
+#define __addr_ok(addr) (1)
+
 static inline mm_segment_t get_fs(void)
 {
 	return USER_DS;
@@ -66,6 +66,9 @@ static inline int __access_ok(unsigned long addr, unsigned long size)
 	return ((addr >= memory_start) && ((addr + size) < memory_end));
 }
 #else /* CONFIG_MMU */
+#define __addr_ok(addr) \
+	((unsigned long)(addr) < (current_thread_info()->addr_limit.seg))
+
 #define get_fs()	(current_thread_info()->addr_limit)
 #define set_fs(x)	(current_thread_info()->addr_limit = (x))
 
-- 
cgit v1.2.1


From 9f23e7e94f7083d9705b595cbd6b30972be6fbbb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:27:00 +0900
Subject: sh: pselect6 and ppoll, along with signal trampoline rework.

This implements support for ppoll() and pselect6()..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/entry.S       |   8 +--
 arch/sh/kernel/signal.c      | 137 ++++++++++++++++++++-----------------------
 arch/sh/kernel/syscalls.S    |  31 +++++-----
 include/asm-sh/thread_info.h |   2 +
 include/asm-sh/unistd.h      |  66 +++++++++++----------
 5 files changed, 121 insertions(+), 123 deletions(-)

diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index fcbf50d4ee10..fd5fe2349f20 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -371,12 +371,12 @@ work_pending:
 	! r8: current_thread_info
 	! t:  result of "tst	#_TIF_NEED_RESCHED, r0"
 	bf/s	work_resched
-	 tst	#_TIF_SIGPENDING, r0
+	 tst	#(_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK), r0
 work_notifysig:
 	bt/s	restore_all
 	 mov	r15, r4
-	mov	#0, r5
-	mov	r12, r6		! set arg2(save_r0)
+	mov	r12, r5		! set arg1(save_r0)
+	mov	r0, r6
 	mov.l	2f, r1
 	mova	restore_all, r0
 	jmp	@r1
@@ -414,7 +414,7 @@ work_resched:
 
 	.align	2
 1:	.long	schedule
-2:	.long	do_signal
+2:	.long	do_notify_resume
 
 	.align	2
 syscall_exit_work:
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index eb6a3b97e46c..2f1c9545b49f 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -29,13 +29,10 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#define DEBUG_SIG 0
+#undef DEBUG
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset,
-			 unsigned int save_r0);
-
 /*
  * Atomically swap in the new signal mask, and wait for a signal.
  */
@@ -44,51 +41,17 @@ sys_sigsuspend(old_sigset_t mask,
 	       unsigned long r5, unsigned long r6, unsigned long r7,
 	       struct pt_regs regs)
 {
-	sigset_t saveset;
-
 	mask &= _BLOCKABLE;
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	siginitset(&current->blocked, mask);
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset, regs.regs[0]))
-			return -EINTR;
-	}
-}
-
-asmlinkage int
-sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
-		  unsigned long r6, unsigned long r7,
-		  struct pt_regs regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	regs.regs[0] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(&regs, &saveset, regs.regs[0]))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage int 
@@ -349,7 +312,7 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 	return (void __user *)((sp - frame_size) & -8ul);
 }
 
-static void setup_frame(int sig, struct k_sigaction *ka,
+static int setup_frame(int sig, struct k_sigaction *ka,
 			sigset_t *set, struct pt_regs *regs)
 {
 	struct sigframe __user *frame;
@@ -369,10 +332,9 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
 
-	if (_NSIG_WORDS > 1) {
+	if (_NSIG_WORDS > 1)
 		err |= __copy_to_user(frame->extramask, &set->sig[1],
 				      sizeof(frame->extramask));
-	}
 
 	/* Set up to return from userspace.  If provided, use a stub
 	   already in userspace.  */
@@ -403,21 +365,22 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
+static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 			   sigset_t *set, struct pt_regs *regs)
 {
 	struct rt_sigframe __user *frame;
@@ -478,28 +441,31 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 
 	set_fs(USER_DS);
 
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
+	pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
+		 current->comm, current->pid, frame, regs->pc, regs->pr);
 
 	flush_cache_sigtramp(regs->pr);
+
 	if ((-regs->pr & (L1_CACHE_BYTES-1)) < sizeof(frame->retcode))
 		flush_cache_sigtramp(regs->pr + L1_CACHE_BYTES);
-	return;
+
+	return 0;
 
 give_sigsegv:
 	force_sigsegv(sig, current);
+	return -EFAULT;
 }
 
 /*
  * OK, we're invoking a handler
  */
 
-static void
+static int
 handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 	      sigset_t *oldset, struct pt_regs *regs)
 {
+	int ret;
+
 	/* Are we from a system call? */
 	if (regs->tra >= 0) {
 		/* If so, check system call restarting.. */
@@ -540,19 +506,23 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
 
 	/* Set up the stack frame */
 	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
+		ret = setup_rt_frame(sig, ka, info, oldset, regs);
 	else
-		setup_frame(sig, ka, oldset, regs);
+		ret = setup_frame(sig, ka, oldset, regs);
 
 	if (ka->sa.sa_flags & SA_ONESHOT)
 		ka->sa.sa_handler = SIG_DFL;
 
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked,sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
+	if (ret == 0) {
+		spin_lock_irq(&current->sighand->siglock);
+		sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
+		if (!(ka->sa.sa_flags & SA_NODEFER))
+			sigaddset(&current->blocked,sig);
+		recalc_sigpending();
+		spin_unlock_irq(&current->sighand->siglock);
+	}
+
+	return ret;
 }
 
 /*
@@ -564,11 +534,12 @@ handle_signal(unsigned long sig, struct k_sigaction *ka, siginfo_t *info,
  * the kernel can handle, and then we build all the user-level signal handling
  * stack-frames in one go after that.
  */
-int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
+static void do_signal(struct pt_regs *regs, unsigned int save_r0)
 {
 	siginfo_t info;
 	int signr;
 	struct k_sigaction ka;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which
@@ -577,19 +548,27 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
 	if (try_to_freeze())
 		goto no_signal;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
 		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &ka, &info, oldset, regs);
-		return 1;
+		if (handle_signal(signr, &ka, &info, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
 	}
 
  no_signal:
@@ -606,5 +585,19 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset, unsigned int save_r0)
 			regs->regs[3] = __NR_restart_syscall;
 		}
 	}
-	return 0;
+
+	/* if there's no signal to deliver, we just put the saved sigmask
+	 * back */
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+}
+
+asmlinkage void do_notify_resume(struct pt_regs *regs, unsigned int save_r0,
+				 __u32 thread_info_flags)
+{
+	/* deal with pending signal delivery */
+	if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK))
+		do_signal(regs, save_r0);
 }
diff --git a/arch/sh/kernel/syscalls.S b/arch/sh/kernel/syscalls.S
index ea23b213c77f..768334e95075 100644
--- a/arch/sh/kernel/syscalls.S
+++ b/arch/sh/kernel/syscalls.S
@@ -319,34 +319,35 @@ ENTRY(sys_call_table)
 	.long sys_mq_getsetattr
 	.long sys_kexec_load
 	.long sys_waitid
-	.long sys_add_key		/* 285 */
+	.long sys_ni_syscall		/* 285 */
+	.long sys_add_key
 	.long sys_request_key
 	.long sys_keyctl
 	.long sys_ioprio_set
-	.long sys_ioprio_get
-	.long sys_inotify_init		/* 290 */
+	.long sys_ioprio_get		/* 290 */
+	.long sys_inotify_init
 	.long sys_inotify_add_watch
 	.long sys_inotify_rm_watch
 	.long sys_migrate_pages
-	.long sys_openat
-	.long sys_mkdirat		/* 295 */
+	.long sys_openat		/* 295 */
+	.long sys_mkdirat
 	.long sys_mknodat
 	.long sys_fchownat
 	.long sys_futimesat
-	.long sys_fstatat64
-	.long sys_unlinkat		/* 300 */
+	.long sys_fstatat64		/* 300 */
+	.long sys_unlinkat
 	.long sys_renameat
 	.long sys_linkat
 	.long sys_symlinkat
-	.long sys_readlinkat
-	.long sys_fchmodat		/* 305 */
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
 	.long sys_faccessat
-	.long sys_ni_syscall		/* Reserved for pselect6 */
-	.long sys_ni_syscall		/* Reserved for ppoll */
-	.long sys_unshare
-	.long sys_set_robust_list	/* 310 */
+	.long sys_pselect6
+	.long sys_ppoll
+	.long sys_unshare		/* 310 */
+	.long sys_set_robust_list
 	.long sys_get_robust_list
 	.long sys_splice
 	.long sys_sync_file_range
-	.long sys_tee
-	.long sys_vmsplice		/* 315 */
+	.long sys_tee			/* 315 */
+	.long sys_vmsplice
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index b986a1914472..5eb874b065a6 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -94,6 +94,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_NOTIFY_RESUME	1	/* resumption notification requested */
 #define TIF_SIGPENDING		2	/* signal pending */
 #define TIF_NEED_RESCHED	3	/* rescheduling necessary */
+#define TIF_RESTORE_SIGMASK	4	/* restore signal mask in do_signal() */
 #define TIF_USEDFPU		16	/* FPU was used by this task this quantum (SMP) */
 #define TIF_POLLING_NRFLAG	17	/* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_MEMDIE		18
@@ -102,6 +103,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
 #define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
+#define _TIF_RESTORE_SIGMASK	(1<<TIF_RESTORE_SIGMASK)
 #define _TIF_USEDFPU		(1<<TIF_USEDFPU)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 49c6a6ee0ab2..080e5729f17d 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -292,39 +292,40 @@
 #define __NR_mq_getsetattr      (__NR_mq_open+5)
 #define __NR_kexec_load		283
 #define __NR_waitid		284
-#define __NR_add_key		285
-#define __NR_request_key	286
-#define __NR_keyctl		287
-#define __NR_ioprio_set		288
-#define __NR_ioprio_get		289
-#define __NR_inotify_init	290
-#define __NR_inotify_add_watch	291
-#define __NR_inotify_rm_watch	292
-#define __NR_migrate_pages	293
-#define __NR_openat		294
-#define __NR_mkdirat		295
-#define __NR_mknodat		296
-#define __NR_fchownat		297
-#define __NR_futimesat		298
-#define __NR_fstatat64		299
-#define __NR_unlinkat		300
-#define __NR_renameat		301
-#define __NR_linkat		302
-#define __NR_symlinkat		303
-#define __NR_readlinkat		304
-#define __NR_fchmodat		305
-#define __NR_faccessat		305
-#define __NR_pselect6		307
-#define __NR_ppoll		308
-#define __NR_unshare		309
-#define __NR_set_robust_list	310
-#define __NR_get_robust_list	311
-#define __NR_splice		312
-#define __NR_sync_file_range	313
-#define __NR_tee		314
-#define __NR_vmsplice		315
+/* #define __NR_sys_setaltroot	285 */
+#define __NR_add_key		286
+#define __NR_request_key	287
+#define __NR_keyctl		288
+#define __NR_ioprio_set		289
+#define __NR_ioprio_get		290
+#define __NR_inotify_init	291
+#define __NR_inotify_add_watch	292
+#define __NR_inotify_rm_watch	293
+#define __NR_migrate_pages	294
+#define __NR_openat		295
+#define __NR_mkdirat		296
+#define __NR_mknodat		297
+#define __NR_fchownat		298
+#define __NR_futimesat		299
+#define __NR_newfstatat		300
+#define __NR_unlinkat		301
+#define __NR_renameat		302
+#define __NR_linkat		303
+#define __NR_symlinkat		304
+#define __NR_readlinkat		305
+#define __NR_fchmodat		306
+#define __NR_faccessat		307
+#define __NR_pselect6		308
+#define __NR_ppoll		309
+#define __NR_unshare		310
+#define __NR_set_robust_list	311
+#define __NR_get_robust_list	312
+#define __NR_splice		313
+#define __NR_sync_file_range	314
+#define __NR_tee		315
+#define __NR_vmsplice		316
 
-#define NR_syscalls 316
+#define NR_syscalls 317
 
 #ifdef __KERNEL__
 
@@ -466,6 +467,7 @@ __syscall_return(type,__sc0); \
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
+#define __ARCH_WANT_SYS_RT_SIGSUSPEND
 
 #ifdef __KERNEL_SYSCALLS__
 
-- 
cgit v1.2.1


From 237b98f61d02499ee248ae33cad616d950c15f49 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:28:20 +0900
Subject: sh: sem2mutex conversion for clock framework.

Simple sem2mutex conversion.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/clock.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/arch/sh/kernel/cpu/clock.c b/arch/sh/kernel/cpu/clock.c
index e7eef07536f5..51ec64cdf348 100644
--- a/arch/sh/kernel/cpu/clock.c
+++ b/arch/sh/kernel/cpu/clock.c
@@ -1,7 +1,7 @@
 /*
  * arch/sh/kernel/cpu/clock.c - SuperH clock framework
  *
- *  Copyright (C) 2005  Paul Mundt
+ *  Copyright (C) 2005, 2006  Paul Mundt
  *
  * This clock framework is derived from the OMAP version by:
  *
@@ -15,6 +15,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/module.h>
+#include <linux/mutex.h>
 #include <linux/list.h>
 #include <linux/kref.h>
 #include <linux/seq_file.h>
@@ -24,7 +25,7 @@
 
 static LIST_HEAD(clock_list);
 static DEFINE_SPINLOCK(clock_lock);
-static DECLARE_MUTEX(clock_list_sem);
+static DEFINE_MUTEX(clock_list_sem);
 
 /*
  * Each subtype is expected to define the init routines for these clocks,
@@ -140,21 +141,21 @@ void clk_disable(struct clk *clk)
 
 int clk_register(struct clk *clk)
 {
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 
 	list_add(&clk->node, &clock_list);
 	kref_init(&clk->kref);
 
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 
 	return 0;
 }
 
 void clk_unregister(struct clk *clk)
 {
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 	list_del(&clk->node);
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 }
 
 inline unsigned long clk_get_rate(struct clk *clk)
@@ -198,14 +199,14 @@ struct clk *clk_get(const char *id)
 {
 	struct clk *p, *clk = ERR_PTR(-ENOENT);
 
-	down(&clock_list_sem);
+	mutex_lock(&clock_list_sem);
 	list_for_each_entry(p, &clock_list, node) {
 		if (strcmp(id, p->name) == 0 && try_module_get(p->owner)) {
 			clk = p;
 			break;
 		}
 	}
-	up(&clock_list_sem);
+	mutex_unlock(&clock_list_sem);
 
 	return clk;
 }
-- 
cgit v1.2.1


From 7dec62e96b38e6c62490ea5dc6939dd8d680a8b8 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:30:35 +0900
Subject: sh: Add setup code for various CPU subtypes.

This adds some simple setup code for most of the CPU subtypes,
primarily simple platform device registration.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh3/Makefile        |  8 +++++
 arch/sh/kernel/cpu/sh3/setup-sh7300.c  | 43 +++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh3/setup-sh7705.c  | 48 ++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh3/setup-sh7708.c  | 43 +++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh3/setup-sh7709.c  | 53 ++++++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/Makefile        |  6 ++++
 arch/sh/kernel/cpu/sh4/setup-sh4-202.c | 43 +++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/setup-sh73180.c | 43 +++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/setup-sh7750.c  | 48 ++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/setup-sh7760.c  | 53 ++++++++++++++++++++++++++++++++++
 arch/sh/kernel/cpu/sh4/setup-sh7770.c  | 53 ++++++++++++++++++++++++++++++++++
 11 files changed, 441 insertions(+)
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7300.c
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7705.c
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7708.c
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7709.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh4-202.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh73180.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7750.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7760.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7770.c

diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index b54dbb9a0c86..1b292ae16f07 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -4,6 +4,14 @@
 
 obj-y	:= ex.o probe.o
 
+# CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7705)	+= setup-sh7705.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7707)	+= setup-sh7709.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7708)	+= setup-sh7708.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7709)	+= setup-sh7709.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7300)	+= setup-sh7300.o
+
+# Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH3)			:= clock-sh3.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7300)	:= clock-sh7300.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7705)	:= clock-sh7705.o
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7300.c b/arch/sh/kernel/cpu/sh3/setup-sh7300.c
new file mode 100644
index 000000000000..ab4d204bfba5
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7300.c
@@ -0,0 +1,43 @@
+/*
+ * SH7300 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4430000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 80, 80, 80, 80 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7300_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7300_devices_setup(void)
+{
+	return platform_add_devices(sh7300_devices,
+				    ARRAY_SIZE(sh7300_devices));
+}
+__initcall(sh7300_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7705.c b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
new file mode 100644
index 000000000000..a8e41c5241fa
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7705.c
@@ -0,0 +1,48 @@
+/*
+ * SH7705 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4400000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.mapbase	= 0xa4410000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 56, 57, 59, 58 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7705_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7705_devices_setup(void)
+{
+	return platform_add_devices(sh7705_devices,
+				    ARRAY_SIZE(sh7705_devices));
+}
+__initcall(sh7705_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7708.c b/arch/sh/kernel/cpu/sh3/setup-sh7708.c
new file mode 100644
index 000000000000..f933723911ca
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7708.c
@@ -0,0 +1,43 @@
+/*
+ * SH7708 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfffffe80,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7708_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7708_devices_setup(void)
+{
+	return platform_add_devices(sh7708_devices,
+				    ARRAY_SIZE(sh7708_devices));
+}
+__initcall(sh7708_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7709.c b/arch/sh/kernel/cpu/sh3/setup-sh7709.c
new file mode 100644
index 000000000000..ff43ef2a1f0c
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7709.c
@@ -0,0 +1,53 @@
+/*
+ * SH7707/SH7709 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfffffe80,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.mapbase	= 0xa4000150,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 56, 57, 59, 58 },
+	}, {
+		.mapbase	= 0xa4000140,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_IRDA,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7709_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7709_devices_setup(void)
+{
+	return platform_add_devices(sh7709_devices,
+				    ARRAY_SIZE(sh7709_devices));
+}
+__initcall(sh7709_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 9611fab03b55..48946d54a2cd 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -8,7 +8,13 @@ obj-$(CONFIG_SH_FPU)                    += fpu.o
 obj-$(CONFIG_SH_STORE_QUEUES)		+= sq.o
 
 # CPU subtype setup
+obj-$(CONFIG_CPU_SUBTYPE_SH7750)	+= setup-sh7750.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7751)	+= setup-sh7750.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7760)	+= setup-sh7760.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7770)	+= setup-sh7770.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
+obj-$(CONFIG_CPU_SUBTYPE_SH73180)	+= setup-sh73180.o
+obj-$(CONFIG_CPU_SUBTYPE_SH4_202)	+= setup-sh4-202.o
 
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH4)			:= clock-sh4.o
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh4-202.c b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c
new file mode 100644
index 000000000000..6e4e96541358
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh4-202.c
@@ -0,0 +1,43 @@
+/*
+ * SH4-202 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh4202_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh4202_devices_setup(void)
+{
+	return platform_add_devices(sh4202_devices,
+				    ARRAY_SIZE(sh4202_devices));
+}
+__initcall(sh4202_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh73180.c b/arch/sh/kernel/cpu/sh4/setup-sh73180.c
new file mode 100644
index 000000000000..cc9ea1e2e5df
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh73180.c
@@ -0,0 +1,43 @@
+/*
+ * SH73180 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 80, 81, 83, 82 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh73180_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh73180_devices_setup(void)
+{
+	return platform_add_devices(sh73180_devices,
+				    ARRAY_SIZE(sh73180_devices));
+}
+__initcall(sh73180_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7750.c b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
new file mode 100644
index 000000000000..50812d57c1c1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7750.c
@@ -0,0 +1,48 @@
+/*
+ * SH7750/SH7751 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCI,
+		.irqs		= { 23, 24, 25, 0 },
+	}, {
+		.mapbase	= 0xffe80000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 40, 41, 43, 42 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7750_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7750_devices_setup(void)
+{
+	return platform_add_devices(sh7750_devices,
+				    ARRAY_SIZE(sh7750_devices));
+}
+__initcall(sh7750_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7760.c b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
new file mode 100644
index 000000000000..97f1c9af35d6
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7760.c
@@ -0,0 +1,53 @@
+/*
+ * SH7760 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xfe600000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.mapbase	= 0xfe610000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 72, 73, 75, 74 },
+	}, {
+		.mapbase	= 0xfe620000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 76, 77, 79, 78 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7760_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7760_devices_setup(void)
+{
+	return platform_add_devices(sh7760_devices,
+				    ARRAY_SIZE(sh7760_devices));
+}
+__initcall(sh7760_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7770.c b/arch/sh/kernel/cpu/sh4/setup-sh7770.c
new file mode 100644
index 000000000000..6a04cc5f5aca
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7770.c
@@ -0,0 +1,53 @@
+/*
+ * SH7770 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xff923000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 61, 61, 61, 61 },
+	}, {
+		.mapbase	= 0xff924000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 62, 62, 62, 62 },
+	}, {
+		.mapbase	= 0xff925000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 63, 63, 63, 63 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7770_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7770_devices_setup(void)
+{
+	return platform_add_devices(sh7770_devices,
+				    ARRAY_SIZE(sh7770_devices));
+}
+__initcall(sh7770_devices_setup);
-- 
cgit v1.2.1


From ecd9561687a0952a96a0a705f618e59cb6f3189b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:32:30 +0900
Subject: serial: Add SERIAL_SH_SCI_NR_UARTS for sh-sci.

sh-sci needs to be able to define its number of ports to
support, we do this with a config option, like most other
ports do.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/serial/Kconfig |  9 +++++++--
 include/asm-sh/sci.h   | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 41 insertions(+), 2 deletions(-)
 create mode 100644 include/asm-sh/sci.h

diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 5b48ac22c9c5..261eaa442953 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -642,12 +642,17 @@ config V850E_UART_CONSOLE
 	select SERIAL_CORE_CONSOLE
 
 config SERIAL_SH_SCI
-	tristate "SH SCI(F) serial port support"
+	tristate "SuperH SCI(F) serial port support"
 	depends on SUPERH || H8300
 	select SERIAL_CORE
 
+config SERIAL_SH_SCI_NR_UARTS
+	int "Maximum number of SCI(F) serial ports"
+	depends on SERIAL_SH_SCI
+	default "2"
+
 config SERIAL_SH_SCI_CONSOLE
-	bool "Support for console on SH SCI(F)"
+	bool "Support for console on SuperH SCI(F)"
 	depends on SERIAL_SH_SCI=y
 	select SERIAL_CORE_CONSOLE
 
diff --git a/include/asm-sh/sci.h b/include/asm-sh/sci.h
new file mode 100644
index 000000000000..52e73660c129
--- /dev/null
+++ b/include/asm-sh/sci.h
@@ -0,0 +1,34 @@
+#ifndef __ASM_SH_SCI_H
+#define __ASM_SH_SCI_H
+
+#include <linux/serial_core.h>
+
+/*
+ * Generic header for SuperH SCI(F)
+ *
+ * Do not place SH-specific parts in here, sh64 and h8300 depend on this too.
+ */
+
+/* Offsets into the sci_port->irqs array */
+enum {
+	SCIx_ERI_IRQ,
+	SCIx_RXI_IRQ,
+	SCIx_TXI_IRQ,
+	SCIx_BRI_IRQ,
+	SCIx_NR_IRQS,
+};
+
+/*
+ * Platform device specific platform_data struct
+ */
+struct plat_sci_port {
+	void __iomem	*membase;		/* io cookie */
+	unsigned long	mapbase;		/* resource base */
+	unsigned int	irqs[SCIx_NR_IRQS];	/* ERI, RXI, TXI, BRI */
+	unsigned int	type;			/* SCI / SCIF / IRDA */
+	upf_t		flags;			/* UPF_* flags */
+};
+
+int early_sci_setup(struct uart_port *port);
+
+#endif /* __ASM_SH_SCI_H */
-- 
cgit v1.2.1


From e5723e0eeb2dc16629e86d66785024ead9169000 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:38:11 +0900
Subject: sh: Add support for SH7706/SH7710/SH7343 CPUs.

This adds support for the aforementioned CPU subtypes, and cleans
up some build issues encountered as a result.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                         |  12 +-
 arch/sh/Makefile                        |   7 +-
 arch/sh/boards/hp6xx/setup.c            |   5 +-
 arch/sh/boards/renesas/hs7751rvoip/io.c |   2 +-
 arch/sh/kernel/cpu/irq/ipr.c            |   2 +
 arch/sh/kernel/cpu/sh3/Makefile         |   5 +-
 arch/sh/kernel/cpu/sh3/clock-sh7706.c   |  84 +++++++++
 arch/sh/kernel/cpu/sh3/ex.S             |  54 +++++-
 arch/sh/kernel/cpu/sh3/probe.c          |   6 +
 arch/sh/kernel/cpu/sh3/setup-sh7710.c   |  43 +++++
 arch/sh/kernel/cpu/sh4/Makefile         |   1 +
 arch/sh/kernel/cpu/sh4/ex.S             |  12 +-
 arch/sh/kernel/cpu/sh4/probe.c          |  17 +-
 arch/sh/kernel/cpu/sh4/setup-sh7343.c   |  43 +++++
 arch/sh/kernel/process.c                |   2 +-
 arch/sh/kernel/setup.c                  |  31 ++--
 arch/sh/kernel/sh_ksyms.c               |   3 +-
 arch/sh/mm/Kconfig                      |  30 ++-
 include/asm-sh/cpu-sh3/cache.h          |   4 +-
 include/asm-sh/cpu-sh3/mmu_context.h    |   8 +-
 include/asm-sh/cpu-sh3/timer.h          |   1 +
 include/asm-sh/cpu-sh3/ubc.h            |  15 +-
 include/asm-sh/irq-sh73180.h            |   2 -
 include/asm-sh/irq-sh7343.h             | 317 ++++++++++++++++++++++++++++++++
 include/asm-sh/irq.h                    | 119 +++++++++++-
 include/asm-sh/processor.h              |   8 +-
 26 files changed, 767 insertions(+), 66 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/sh3/clock-sh7706.c
 create mode 100644 arch/sh/kernel/cpu/sh3/setup-sh7710.c
 create mode 100644 arch/sh/kernel/cpu/sh4/setup-sh7343.c
 create mode 100644 include/asm-sh/irq-sh7343.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 035df7fb2237..557e7232141c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -324,12 +324,11 @@ config SH_FPU_EMU
 
 config SH_DSP
 	bool "DSP support"
-	depends on !CPU_SH4
-	default y
+	default y if SH4AL_DSP || !CPU_SH4
+	default n
 	help
 	  Selecting this option will enable support for SH processors that
-	  have DSP units (ie, SH2-DSP and SH3-DSP). It is safe to say Y here
-	  by default, as the existance of the DSP will be probed at runtime.
+	  have DSP units (ie, SH2-DSP, SH3-DSP, and SH4AL-DSP).
 
 	  This option must be set in order to enable the DSP.
 
@@ -393,8 +392,9 @@ config SH_PCLK_FREQ
 	int "Peripheral clock frequency (in Hz)"
 	default "50000000" if CPU_SUBTYPE_SH7750 || CPU_SUBTYPE_SH7780
 	default "60000000" if CPU_SUBTYPE_SH7751
-	default "33333333" if CPU_SUBTYPE_SH7300 || CPU_SUBTYPE_SH7770 || CPU_SUBTYPE_SH7760
-	default "27000000" if CPU_SUBTYPE_SH73180
+	default "33333333" if CPU_SUBTYPE_SH7300 || CPU_SUBTYPE_SH7770 || \
+			      CPU_SUBTYPE_SH7760
+	default "27000000" if CPU_SUBTYPE_SH73180 || CPU_SUBTYPE_SH7343
 	default "66000000" if CPU_SUBTYPE_SH4_202
 	help
 	  This option is used to specify the peripheral clock frequency.
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 13b688b13e1c..d118e48ea1ba 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -18,11 +18,13 @@ cflags-y				:= -mb
 cflags-$(CONFIG_CPU_LITTLE_ENDIAN)	:= -ml
 
 isa-y					:= any
+isa-$(CONFIG_SH_DSP)			:= sh
 isa-$(CONFIG_CPU_SH2)			:= sh2
+isa-$(CONFIG_CPU_SH2A)			:= sh2a
 isa-$(CONFIG_CPU_SH3)			:= sh3
 isa-$(CONFIG_CPU_SH4)			:= sh4
 isa-$(CONFIG_CPU_SH4A)			:= sh4a
-isa-$(CONFIG_CPU_SH2A)			:= sh2a
+isa-$(CONFIG_CPU_SH4AL_DSP)		:= sh4al
 
 isa-$(CONFIG_SH_DSP)			:= $(isa-y)-dsp
 
@@ -30,9 +32,11 @@ ifndef CONFIG_MMU
 isa-y			:= $(isa-y)-nommu
 endif
 
+ifndef CONFIG_SH_DSP
 ifndef CONFIG_SH_FPU
 isa-y			:= $(isa-y)-nofpu
 endif
+endif
 
 cflags-y	+= $(call as-option,-Wa$(comma)-isa=$(isa-y),)
 
@@ -188,4 +192,3 @@ CLEAN_FILES += include/asm-sh/machtypes.h
 define archhelp
 	@echo '  zImage 	           - Compressed kernel image (arch/sh/boot/zImage)'
 endef
-
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 5fc00f0727c4..629016bec809 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -8,10 +8,11 @@
  *
  * Setup code for an HP680  (internal peripherials only)
  */
-
+#include <linux/types.h>
 #include <linux/init.h>
-#include <asm/io.h>
 #include <asm/hd64461.h>
+#include <asm/io.h>
+#include <asm/irq.h>
 #include <asm/hp6xx/hp6xx.h>
 #include <asm/cpu/dac.h>
 
diff --git a/arch/sh/boards/renesas/hs7751rvoip/io.c b/arch/sh/boards/renesas/hs7751rvoip/io.c
index 8c26550ca2e4..9ea1136b219b 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/io.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/io.c
@@ -10,10 +10,10 @@
  * placeholder code from io_hs7751rvoip.c left in with the
  * expectation of later SuperIO and PCMCIA access.
  */
-
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/module.h>
+#include <linux/pci.h>
 #include <asm/io.h>
 #include <asm/hs7751rvoip/hs7751rvoip.h>
 #include <asm/addrspace.h>
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 0f545941fb4f..d2b715a04c7e 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -89,6 +89,7 @@ static void mask_and_ack_ipr(unsigned int irq)
 	disable_ipr_irq(irq);
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/* This is needed when we use edge triggered setting */
 	/* XXX: Is it really needed? */
@@ -162,6 +163,7 @@ void __init init_IRQ(void)
 #endif
 
 #if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
 	/*
 	 * Initialize the Interrupt Controller (INTC)
diff --git a/arch/sh/kernel/cpu/sh3/Makefile b/arch/sh/kernel/cpu/sh3/Makefile
index 1b292ae16f07..58d3815695ff 100644
--- a/arch/sh/kernel/cpu/sh3/Makefile
+++ b/arch/sh/kernel/cpu/sh3/Makefile
@@ -6,16 +6,19 @@ obj-y	:= ex.o probe.o
 
 # CPU subtype setup
 obj-$(CONFIG_CPU_SUBTYPE_SH7705)	+= setup-sh7705.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7706)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7707)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7708)	+= setup-sh7708.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7709)	+= setup-sh7709.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7300)	+= setup-sh7300.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7710)	+= setup-sh7710.o
 
 # Primary on-chip clocks (common)
 clock-$(CONFIG_CPU_SH3)			:= clock-sh3.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7300)	:= clock-sh7300.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7705)	:= clock-sh7705.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7706)	:= clock-sh7706.o
 clock-$(CONFIG_CPU_SUBTYPE_SH7709)	:= clock-sh7709.o
+clock-$(CONFIG_CPU_SUBTYPE_SH7710)	:= clock-sh7300.o
 
 obj-y	+= $(clock-y)
-
diff --git a/arch/sh/kernel/cpu/sh3/clock-sh7706.c b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
new file mode 100644
index 000000000000..0cf96f9833bc
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/clock-sh7706.c
@@ -0,0 +1,84 @@
+/*
+ * arch/sh/kernel/cpu/sh3/clock-sh7706.c
+ *
+ * SH7706 support for the clock framework
+ *
+ *  Copyright (C) 2006  Takashi YOSHII
+ *
+ * Based on arch/sh/kernel/cpu/sh3/clock-sh7709.c
+ *  Copyright (C) 2005  Andriy Skulysh
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <asm/clock.h>
+#include <asm/freq.h>
+#include <asm/io.h>
+
+static int stc_multipliers[] = { 1, 2, 4, 1, 3, 6, 1, 1 };
+static int ifc_divisors[]    = { 1, 2, 4, 1, 3, 1, 1, 1 };
+static int pfc_divisors[]    = { 1, 2, 4, 1, 3, 6, 1, 1 };
+
+static void master_clk_init(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate *= pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_master_clk_ops = {
+	.init		= master_clk_init,
+};
+
+static void module_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x2000) >> 11) | (frqcr & 0x0003);
+
+	clk->rate = clk->parent->rate / pfc_divisors[idx];
+}
+
+static struct clk_ops sh7706_module_clk_ops = {
+	.recalc		= module_clk_recalc,
+};
+
+static void bus_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x8000) >> 13) | ((frqcr & 0x0030) >> 4);
+
+	clk->rate = clk->parent->rate / stc_multipliers[idx];
+}
+
+static struct clk_ops sh7706_bus_clk_ops = {
+	.recalc		= bus_clk_recalc,
+};
+
+static void cpu_clk_recalc(struct clk *clk)
+{
+	int frqcr = ctrl_inw(FRQCR);
+	int idx = ((frqcr & 0x4000) >> 12) | ((frqcr & 0x000c) >> 2);
+
+	clk->rate = clk->parent->rate / ifc_divisors[idx];
+}
+
+static struct clk_ops sh7706_cpu_clk_ops = {
+	.recalc		= cpu_clk_recalc,
+};
+
+static struct clk_ops *sh7706_clk_ops[] = {
+	&sh7706_master_clk_ops,
+	&sh7706_module_clk_ops,
+	&sh7706_bus_clk_ops,
+	&sh7706_cpu_clk_ops,
+};
+
+void __init arch_init_clk_ops(struct clk_ops **ops, int idx)
+{
+	if (idx < ARRAY_SIZE(sh7706_clk_ops))
+		*ops = sh7706_clk_ops[idx];
+}
diff --git a/arch/sh/kernel/cpu/sh3/ex.S b/arch/sh/kernel/cpu/sh3/ex.S
index cc04e9e239ff..44daf44833f9 100644
--- a/arch/sh/kernel/cpu/sh3/ex.S
+++ b/arch/sh/kernel/cpu/sh3/ex.S
@@ -84,8 +84,12 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	!      rovi
 	.long	do_IRQ			
 	.long	do_IRQ			/* 5E0 */
-#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
-     defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if  defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7706) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+     defined(CONFIG_CPU_SUBTYPE_SH7710)
 	.long	do_IRQ	! 32 IRQ  irq0	/* 600 */
 	.long	do_IRQ	! 33      irq1
 	.long	do_IRQ	! 34      irq2
@@ -147,6 +151,51 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 62 PCC  pcc0i
 	.long   do_IRQ	! 63      pcc1i	/* 9E0 */
 #endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+	.long   exception_none	! 61 	/* 9A0 */
+	.long   exception_none	! 62
+	.long   exception_none	! 63
+	.long   exception_none	! 64	/* A00 */
+	.long   exception_none	! 65
+	.long   exception_none	! 66
+	.long   exception_none	! 67
+	.long   exception_none	! 68
+	.long   exception_none	! 69
+	.long   exception_none	! 70
+	.long   exception_none	! 71
+	.long   exception_none	! 72	/* B00 */
+	.long   exception_none	! 73
+	.long   exception_none	! 74
+	.long   exception_none	! 75
+	.long   do_IRQ	! 76 DMAC2 dei4	/* B80 */
+	.long   do_IRQ	! 77 DMAC2 dei5
+	.long   exception_none	! 78
+	.long   do_IRQ	! 79 IPSEC ipseci /* BE0 */
+	.long   do_IRQ	! 80 EDMAC eint0 /* C00 */
+	.long   do_IRQ	! 81 EDMAC eint1
+	.long   do_IRQ	! 82 EDMAC eint2
+	.long   exception_none	! 83	/* C60 */
+	.long   exception_none	! 84
+	.long   exception_none	! 85
+	.long   exception_none	! 86
+	.long   exception_none	! 87
+	.long   exception_none	! 88	/* D00 */
+	.long   exception_none	! 89
+	.long   exception_none	! 90
+	.long   exception_none	! 91
+	.long   exception_none	! 92
+	.long   exception_none	! 93
+	.long   exception_none	! 94
+	.long   exception_none	! 95
+	.long   do_IRQ	! 96 SIOF eri0	/* E00 */
+	.long   do_IRQ	! 97      txi0
+	.long   do_IRQ	! 98      rxi0
+	.long   do_IRQ	! 99      cci0
+	.long   do_IRQ	! 100     eri1	/* E80 */
+	.long   do_IRQ	! 101     txi1
+	.long   do_IRQ	! 102     rxi2
+	.long   do_IRQ	! 103     cci3
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7300)
 	.long   do_IRQ	! 64
 	.long   do_IRQ	! 65
@@ -195,4 +244,3 @@ ENTRY(interrupt_table)
 	.long   do_IRQ	! 108
 #endif
 #endif
-
diff --git a/arch/sh/kernel/cpu/sh3/probe.c b/arch/sh/kernel/cpu/sh3/probe.c
index 5cdc88638601..e67098836290 100644
--- a/arch/sh/kernel/cpu/sh3/probe.c
+++ b/arch/sh/kernel/cpu/sh3/probe.c
@@ -72,6 +72,12 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->dcache.sets		= 256;
 		cpu_data->type = CPU_SH7729;
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7706)
+		cpu_data->type = CPU_SH7706;
+#endif
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+		cpu_data->type = CPU_SH7710;
+#endif
 #if defined(CONFIG_CPU_SUBTYPE_SH7705)
 		cpu_data->type = CPU_SH7705;
 
diff --git a/arch/sh/kernel/cpu/sh3/setup-sh7710.c b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
new file mode 100644
index 000000000000..895f99ee6a95
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh3/setup-sh7710.c
@@ -0,0 +1,43 @@
+/*
+ * SH7710 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xa4400000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 52, 53, 55, 54 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7710_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7710_devices_setup(void)
+{
+	return platform_add_devices(sh7710_devices,
+				    ARRAY_SIZE(sh7710_devices));
+}
+__initcall(sh7710_devices_setup);
diff --git a/arch/sh/kernel/cpu/sh4/Makefile b/arch/sh/kernel/cpu/sh4/Makefile
index 48946d54a2cd..8dbf3895ece7 100644
--- a/arch/sh/kernel/cpu/sh4/Makefile
+++ b/arch/sh/kernel/cpu/sh4/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_CPU_SUBTYPE_SH7760)	+= setup-sh7760.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7770)	+= setup-sh7770.o
 obj-$(CONFIG_CPU_SUBTYPE_SH7780)	+= setup-sh7780.o
 obj-$(CONFIG_CPU_SUBTYPE_SH73180)	+= setup-sh73180.o
+obj-$(CONFIG_CPU_SUBTYPE_SH7343)	+= setup-sh7343.o
 obj-$(CONFIG_CPU_SUBTYPE_SH4_202)	+= setup-sh4-202.o
 
 # Primary on-chip clocks (common)
diff --git a/arch/sh/kernel/cpu/sh4/ex.S b/arch/sh/kernel/cpu/sh4/ex.S
index af5ecbddea55..7146893a6cca 100644
--- a/arch/sh/kernel/cpu/sh4/ex.S
+++ b/arch/sh/kernel/cpu/sh4/ex.S
@@ -123,6 +123,13 @@ ENTRY(interrupt_table)
 	.long	do_IRQ	! 45      dmte5
 	.long	do_IRQ	! 46      dmte6
 	.long	do_IRQ	! 47      dmte7		/* 7E0 */
+#elif defined(CONFIG_CPU_SUBTYPE_SH7343)
+	.long	do_IRQ	! 44 IIC1 ali		/* 780 */
+	.long	do_IRQ	! 45      tacki
+	.long	do_IRQ	! 46      waiti
+	.long	do_IRQ	! 47      dtei		/* 7E0 */
+	.long	do_IRQ	! 48 DMAC dei0		/* 800 */
+	.long	do_IRQ	! 49      dei1		/* 820 */
 #else
 	.long	exception_error			! 44	/* 780 */
 	.long	exception_error			! 45
@@ -132,7 +139,8 @@ ENTRY(interrupt_table)
 #if defined(CONFIG_SH_FPU)
 	.long	do_fpu_state_restore	! 48	/* 800 */
 	.long	do_fpu_state_restore	! 49	/* 820 */
-#else
+#elif !defined(CONFIG_CPU_SUBTYPE_SH7343) && \
+      !defined(CONFIG_CPU_SUBTYPE_SH73180)
 	.long	exception_error
 	.long	exception_error
 #endif
@@ -225,7 +233,7 @@ ENTRY(interrupt_table)
 	.long	exception_error
 	.long	do_IRQ	! ADC	adi
 	.long	do_IRQ	! CMT	cmti	/* FA0 */
-#elif defined(CONFIG_CPU_SUBTYPE_SH73180)
+#elif defined(CONFIG_CPU_SUBTYPE_SH73180) || defined(CONFIG_CPU_SUBTYPE_SH7343)
 	.long	do_IRQ	!  50 0x840
 	.long	do_IRQ	!  51 0x860
 	.long	do_IRQ	!  52 0x880
diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 2a7707a81d8f..6e8a2b5268e8 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -3,7 +3,7 @@
  *
  * CPU Subtype Probing for SH-4.
  *
- * Copyright (C) 2001 - 2005  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
@@ -76,12 +76,6 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
-
-		/*
-		 * XXX: Double check this, none of the SH-4A/SH-4AL processors
-		 * should have this, as it's essentially a legacy thing.
-		 */
-		cpu_data->flags |= CPU_HAS_PTEA;
 		break;
 	case 0x2001:
 	case 0x2004:
@@ -89,8 +83,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		/* Same note as above applies here for PTEA */
-		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
+		cpu_data->flags |= CPU_HAS_FPU;
 		break;
 	case 0x2006:
 	case 0x200A:
@@ -104,6 +97,12 @@ int __init detect_cpu_and_cache_system(void)
 
 		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER;
 		break;
+	case 0x3000:
+	case 0x3003:
+		cpu_data->type = CPU_SH7343;
+		cpu_data->icache.ways = 4;
+		cpu_data->dcache.ways = 4;
+		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
 		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PTEA;
diff --git a/arch/sh/kernel/cpu/sh4/setup-sh7343.c b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
new file mode 100644
index 000000000000..91d61cf91ba1
--- /dev/null
+++ b/arch/sh/kernel/cpu/sh4/setup-sh7343.c
@@ -0,0 +1,43 @@
+/*
+ * SH7343 Setup
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/platform_device.h>
+#include <linux/init.h>
+#include <linux/serial.h>
+#include <asm/sci.h>
+
+static struct plat_sci_port sci_platform_data[] = {
+	{
+		.mapbase	= 0xffe00000,
+		.flags		= UPF_BOOT_AUTOCONF,
+		.type		= PORT_SCIF,
+		.irqs		= { 80, 81, 83, 82 },
+	}, {
+		.flags = 0,
+	}
+};
+
+static struct platform_device sci_device = {
+	.name		= "sh-sci",
+	.id		= -1,
+	.dev		= {
+		.platform_data	= sci_platform_data,
+	},
+};
+
+static struct platform_device *sh7343_devices[] __initdata = {
+	&sci_device,
+};
+
+static int __init sh7343_devices_setup(void)
+{
+	return platform_add_devices(sh7343_devices,
+				    ARRAY_SIZE(sh7343_devices));
+}
+__initcall(sh7343_devices_setup);
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index c4aa687ba26a..2167746e88f1 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -310,7 +310,7 @@ ubc_set_tracing(int asid, unsigned long pc)
 
 	ctrl_outl(0, UBC_BAMRA);
 
-	if (cpu_data->type == CPU_SH7729) {
+	if (cpu_data->type == CPU_SH7729 || cpu_data->type == CPU_SH7710) {
 		ctrl_outw(BBR_INST | BBR_READ | BBR_CPU, UBC_BBRA);
 		ctrl_outl(BRCR_PCBA | BRCR_PCTE, UBC_BRCR);
 	} else {
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 4afdec071700..86ef17fe48b5 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -424,25 +424,18 @@ static int __init topology_init(void)
 subsys_initcall(topology_init);
 
 static const char *cpu_name[] = {
-	[CPU_SH7604]	= "SH7604",
-	[CPU_SH7705]	= "SH7705",
-	[CPU_SH7708]	= "SH7708",
-	[CPU_SH7729]	= "SH7729",
-	[CPU_SH7300]	= "SH7300",
-	[CPU_SH7750]	= "SH7750",
-	[CPU_SH7750S]	= "SH7750S",
-	[CPU_SH7750R]	= "SH7750R",
-	[CPU_SH7751]	= "SH7751",
-	[CPU_SH7751R]	= "SH7751R",
-	[CPU_SH7760]	= "SH7760",
-	[CPU_SH73180]	= "SH73180",
-	[CPU_ST40RA]	= "ST40RA",
-	[CPU_ST40GX1]	= "ST40GX1",
-	[CPU_SH4_202]	= "SH4-202",
-	[CPU_SH4_501]	= "SH4-501",
-	[CPU_SH7770]	= "SH7770",
-	[CPU_SH7780]	= "SH7780",
-	[CPU_SH7781]	= "SH7781",
+	[CPU_SH7604]	= "SH7604",	[CPU_SH7300]	= "SH7300",
+	[CPU_SH7705]	= "SH7705",	[CPU_SH7706]	= "SH7706",
+	[CPU_SH7707]	= "SH7707",	[CPU_SH7708]	= "SH7708",
+	[CPU_SH7709]	= "SH7709",	[CPU_SH7710]	= "SH7710",
+	[CPU_SH7729]	= "SH7729",	[CPU_SH7750]	= "SH7750",
+	[CPU_SH7750S]	= "SH7750S",	[CPU_SH7750R]	= "SH7750R",
+	[CPU_SH7751]	= "SH7751",	[CPU_SH7751R]	= "SH7751R",
+	[CPU_SH7760]	= "SH7760",	[CPU_SH73180]	= "SH73180",
+	[CPU_ST40RA]	= "ST40RA",	[CPU_ST40GX1]	= "ST40GX1",
+	[CPU_SH4_202]	= "SH4-202",	[CPU_SH4_501]	= "SH4-501",
+	[CPU_SH7770]	= "SH7770",	[CPU_SH7780]	= "SH7780",
+	[CPU_SH7781]	= "SH7781",	[CPU_SH7343]	= "SH7343",
 	[CPU_SH_NONE]	= "Unknown"
 };
 
diff --git a/arch/sh/kernel/sh_ksyms.c b/arch/sh/kernel/sh_ksyms.c
index bf59d73415d7..d3cbfa2ad4a7 100644
--- a/arch/sh/kernel/sh_ksyms.c
+++ b/arch/sh/kernel/sh_ksyms.c
@@ -89,7 +89,8 @@ EXPORT_SYMBOL(flush_dcache_page);
 EXPORT_SYMBOL(__flush_purge_region);
 #endif
 
-#ifdef CONFIG_MMU
+#if defined(CONFIG_MMU) && (defined(CONFIG_CPU_SH4) || \
+	defined(CONFIG_SH7705_CACHE_32KB))
 EXPORT_SYMBOL(clear_user_page);
 #endif
 
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index bed697c0dc19..b445d02075e8 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -21,6 +21,10 @@ config CPU_SH4A
 	bool
 	select CPU_SH4
 
+config CPU_SH4AL_DSP
+	bool
+	select CPU_SH4A
+
 config CPU_SUBTYPE_ST40
 	bool
 	select CPU_SH4
@@ -47,6 +51,12 @@ config CPU_SUBTYPE_SH7705
 	select CPU_SH3
 	select CPU_HAS_PINT_IRQ
 
+config CPU_SUBTYPE_SH7706
+	bool "Support SH7706 processor"
+	select CPU_SH3
+	help
+	  Select SH7706 if you have a 133 Mhz SH-3 HD6417706 CPU.
+
 config CPU_SUBTYPE_SH7707
 	bool "Support SH7707 processor"
 	select CPU_SH3
@@ -68,6 +78,12 @@ config CPU_SUBTYPE_SH7709
 	help
 	  Select SH7709 if you have a  80 Mhz SH-3 HD6417709 CPU.
 
+config CPU_SUBTYPE_SH7710
+	bool "Support SH7710 processor"
+	select CPU_SH3
+	help
+	  Select SH7710 if you have a SH3-DSP SH7710 CPU.
+
 comment "SH-4 Processor Support"
 
 config CPU_SUBTYPE_SH7750
@@ -132,10 +148,6 @@ config CPU_SUBTYPE_ST40GX1
 
 comment "SH-4A Processor Support"
 
-config CPU_SUBTYPE_SH73180
-	bool "Support SH73180 processor"
-	select CPU_SH4A
-
 config CPU_SUBTYPE_SH7770
 	bool "Support SH7770 processor"
 	select CPU_SH4A
@@ -145,6 +157,16 @@ config CPU_SUBTYPE_SH7780
 	select CPU_SH4A
 	select CPU_HAS_INTC2_IRQ
 
+comment "SH4AL-DSP Processor Support"
+
+config CPU_SUBTYPE_SH73180
+	bool "Support SH73180 processor"
+	select CPU_SH4AL_DSP
+
+config CPU_SUBTYPE_SH7343
+	bool "Support SH7343 processor"
+	select CPU_SH4AL_DSP
+
 endmenu
 
 menu "Memory management options"
diff --git a/include/asm-sh/cpu-sh3/cache.h b/include/asm-sh/cpu-sh3/cache.h
index 406aa8d9b947..ffe08d2813f9 100644
--- a/include/asm-sh/cpu-sh3/cache.h
+++ b/include/asm-sh/cpu-sh3/cache.h
@@ -26,12 +26,10 @@
 #define CCR_CACHE_ENABLE	CCR_CACHE_CE
 #define CCR_CACHE_INVALIDATE	CCR_CACHE_CF
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7705) || defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define CCR3	0xa40000b4
 #define CCR_CACHE_16KB  0x00010000
 #define CCR_CACHE_32KB	0x00020000
 #endif
 
-
 #endif /* __ASM_CPU_SH3_CACHE_H */
-
diff --git a/include/asm-sh/cpu-sh3/mmu_context.h b/include/asm-sh/cpu-sh3/mmu_context.h
index a844ea0965b6..bccb7ddb438b 100644
--- a/include/asm-sh/cpu-sh3/mmu_context.h
+++ b/include/asm-sh/cpu-sh3/mmu_context.h
@@ -27,8 +27,12 @@
 #define TRA	0xffffffd0
 #define EXPEVT	0xffffffd4
 
-#if defined(CONFIG_CPU_SUBTYPE_SH7707) || defined(CONFIG_CPU_SUBTYPE_SH7709) || \
-    defined(CONFIG_CPU_SUBTYPE_SH7300) || defined(CONFIG_CPU_SUBTYPE_SH7705)
+#if defined(CONFIG_CPU_SUBTYPE_SH7707) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7706) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7300) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+    defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define INTEVT	0xa4000000	/* INTEVTE2(0xa4000000) */
 #else
 #define INTEVT	0xffffffd8
diff --git a/include/asm-sh/cpu-sh3/timer.h b/include/asm-sh/cpu-sh3/timer.h
index 2082ad956f21..b2394cf76f49 100644
--- a/include/asm-sh/cpu-sh3/timer.h
+++ b/include/asm-sh/cpu-sh3/timer.h
@@ -20,6 +20,7 @@
  *	SH7710
  *	SH7720
  *	SH7300
+ *	SH7710
  * ---------------------------------------------------------------------------
  */
 
diff --git a/include/asm-sh/cpu-sh3/ubc.h b/include/asm-sh/cpu-sh3/ubc.h
index 0f809dec4e17..9d308cbe9b29 100644
--- a/include/asm-sh/cpu-sh3/ubc.h
+++ b/include/asm-sh/cpu-sh3/ubc.h
@@ -11,6 +11,19 @@
 #ifndef __ASM_CPU_SH3_UBC_H
 #define __ASM_CPU_SH3_UBC_H
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+#define UBC_BARA		0xa4ffffb0
+#define UBC_BAMRA		0xa4ffffb4
+#define UBC_BBRA		0xa4ffffb8
+#define UBC_BASRA		0xffffffe4
+#define UBC_BARB		0xa4ffffa0
+#define UBC_BAMRB		0xa4ffffa4
+#define UBC_BBRB		0xa4ffffa8
+#define UBC_BASRB		0xffffffe8
+#define UBC_BDRB		0xa4ffff90
+#define UBC_BDMRB		0xa4ffff94
+#define UBC_BRCR		0xa4ffff98
+#else
 #define UBC_BARA                0xffffffb0
 #define UBC_BAMRA               0xffffffb4
 #define UBC_BBRA                0xffffffb8
@@ -22,6 +35,6 @@
 #define UBC_BDRB                0xffffff90
 #define UBC_BDMRB               0xffffff94
 #define UBC_BRCR                0xffffff98
+#endif
 
 #endif /* __ASM_CPU_SH3_UBC_H */
-
diff --git a/include/asm-sh/irq-sh73180.h b/include/asm-sh/irq-sh73180.h
index d705252be260..b28af9a69d72 100644
--- a/include/asm-sh/irq-sh73180.h
+++ b/include/asm-sh/irq-sh73180.h
@@ -311,6 +311,4 @@
 #define IRQ6_PRIORITY	1
 #define IRQ7_PRIORITY	1
 
-int shmse_irq_demux(int irq);
-
 #endif /* __ASM_SH_IRQ_SH73180_H */
diff --git a/include/asm-sh/irq-sh7343.h b/include/asm-sh/irq-sh7343.h
new file mode 100644
index 000000000000..5d15419b53b0
--- /dev/null
+++ b/include/asm-sh/irq-sh7343.h
@@ -0,0 +1,317 @@
+#ifndef __ASM_SH_IRQ_SH7343_H
+#define __ASM_SH_IRQ_SH7343_H
+
+/*
+ * linux/include/asm-sh/irq-sh7343.h
+ *
+ * Copyright (C) 2006 Kenati Technologies Inc.
+ * Andre Mccurdy <andre@kenati.com>
+ * Ranjit Deshpande <ranjit@kenati.com>
+ */
+
+#undef INTC_IPRA
+#undef INTC_IPRB
+#undef INTC_IPRC
+#undef INTC_IPRD
+
+#undef DMTE0_IRQ
+#undef DMTE1_IRQ
+#undef DMTE2_IRQ
+#undef DMTE3_IRQ
+#undef DMTE4_IRQ
+#undef DMTE5_IRQ
+#undef DMTE6_IRQ
+#undef DMTE7_IRQ
+#undef DMAE_IRQ
+#undef DMA_IPR_ADDR
+#undef DMA_IPR_POS
+#undef DMA_PRIORITY
+
+#undef INTC_IMCR0
+#undef INTC_IMCR1
+#undef INTC_IMCR2
+#undef INTC_IMCR3
+#undef INTC_IMCR4
+#undef INTC_IMCR5
+#undef INTC_IMCR6
+#undef INTC_IMCR7
+#undef INTC_IMCR8
+#undef INTC_IMCR9
+#undef INTC_IMCR10
+
+
+#define INTC_IPRA  	0xA4080000UL
+#define INTC_IPRB  	0xA4080004UL
+#define INTC_IPRC  	0xA4080008UL
+#define INTC_IPRD  	0xA408000CUL
+#define INTC_IPRE  	0xA4080010UL
+#define INTC_IPRF  	0xA4080014UL
+#define INTC_IPRG  	0xA4080018UL
+#define INTC_IPRH  	0xA408001CUL
+#define INTC_IPRI  	0xA4080020UL
+#define INTC_IPRJ  	0xA4080024UL
+#define INTC_IPRK  	0xA4080028UL
+#define INTC_IPRL  	0xA408002CUL
+
+#define INTC_IMR0	0xA4080080UL
+#define INTC_IMR1	0xA4080084UL
+#define INTC_IMR2	0xA4080088UL
+#define INTC_IMR3	0xA408008CUL
+#define INTC_IMR4	0xA4080090UL
+#define INTC_IMR5	0xA4080094UL
+#define INTC_IMR6	0xA4080098UL
+#define INTC_IMR7	0xA408009CUL
+#define INTC_IMR8	0xA40800A0UL
+#define INTC_IMR9	0xA40800A4UL
+#define INTC_IMR10	0xA40800A8UL
+#define INTC_IMR11	0xA40800ACUL
+
+#define INTC_IMCR0	0xA40800C0UL
+#define INTC_IMCR1	0xA40800C4UL
+#define INTC_IMCR2	0xA40800C8UL
+#define INTC_IMCR3	0xA40800CCUL
+#define INTC_IMCR4	0xA40800D0UL
+#define INTC_IMCR5	0xA40800D4UL
+#define INTC_IMCR6	0xA40800D8UL
+#define INTC_IMCR7	0xA40800DCUL
+#define INTC_IMCR8	0xA40800E0UL
+#define INTC_IMCR9	0xA40800E4UL
+#define INTC_IMCR10	0xA40800E8UL
+#define INTC_IMCR11	0xA40800ECUL
+
+#define INTC_ICR0	0xA4140000UL
+#define INTC_ICR1	0xA414001CUL
+
+#define INTMSK0		0xa4140044
+#define INTMSKCLR0	0xa4140064
+#define INTC_INTPRI0	0xa4140010
+
+/*
+  NOTE:
+
+  *_IRQ = (INTEVT2 - 0x200)/0x20
+*/
+
+/* TMU0 */
+#define TMU0_IRQ	16
+#define TMU0_IPR_ADDR	INTC_IPRA
+#define TMU0_IPR_POS	 3
+#define TMU0_PRIORITY	 2
+
+#define TIMER_IRQ       16
+#define TIMER_IPR_ADDR  INTC_IPRA
+#define TIMER_IPR_POS    3
+#define TIMER_PRIORITY   2
+
+/* TMU1 */
+#define TMU1_IRQ	17
+#define TMU1_IPR_ADDR	INTC_IPRA
+#define TMU1_IPR_POS	 2
+#define TMU1_PRIORITY	 2
+
+/* TMU2 */
+#define TMU2_IRQ	18
+#define TMU2_IPR_ADDR	INTC_IPRA
+#define TMU2_IPR_POS	 1
+#define TMU2_PRIORITY	 2
+
+/* LCDC */
+#define LCDC_IRQ	28
+#define LCDC_IPR_ADDR	INTC_IPRB
+#define LCDC_IPR_POS	 2
+#define LCDC_PRIORITY	 2
+
+/* VIO (Video I/O) */
+#define CEU_IRQ		52
+#define BEU_IRQ		53
+#define VEU_IRQ		54
+#define VOU_IRQ		55
+#define VIO_IPR_ADDR	INTC_IPRE
+#define VIO_IPR_POS	 2
+#define VIO_PRIORITY	 2
+
+/* MFI (Multi Functional Interface) */
+#define MFI_IRQ		56
+#define MFI_IPR_ADDR	INTC_IPRE
+#define MFI_IPR_POS	 1
+#define MFI_PRIORITY	 2
+
+/* VPU (Video Processing Unit) */
+#define VPU_IRQ		60
+#define VPU_IPR_ADDR	INTC_IPRE
+#define VPU_IPR_POS	 0
+#define VPU_PRIORITY	 2
+
+/* 3DG */
+#define TDG_IRQ		63
+#define TDG_IPR_ADDR	INTC_IPRJ
+#define TDG_IPR_POS	 2
+#define TDG_PRIORITY	 2
+
+/* DMAC(1) */
+#define DMTE0_IRQ	48
+#define DMTE1_IRQ	49
+#define DMTE2_IRQ	50
+#define DMTE3_IRQ	51
+#define DMA1_IPR_ADDR	INTC_IPRE
+#define DMA1_IPR_POS	3
+#define DMA1_PRIORITY	7
+
+/* DMAC(2) */
+#define DMTE4_IRQ	76
+#define DMTE5_IRQ	77
+#define DMA2_IPR_ADDR	INTC_IPRF
+#define DMA2_IPR_POS	2
+#define DMA2_PRIORITY	7
+
+/* SCIF0 */
+#define SCIF_ERI_IRQ	80
+#define SCIF_RXI_IRQ	81
+#define SCIF_BRI_IRQ	82
+#define SCIF_TXI_IRQ	83
+#define SCIF_IPR_ADDR	INTC_IPRG
+#define SCIF_IPR_POS	3
+#define SCIF_PRIORITY	3
+
+/* SIOF0 */
+#define SIOF0_IRQ	84
+#define SIOF0_IPR_ADDR	INTC_IPRH
+#define SIOF0_IPR_POS	3
+#define SIOF0_PRIORITY	3
+
+/* FLCTL (Flash Memory Controller) */
+#define FLSTE_IRQ	92
+#define FLTEND_IRQ	93
+#define FLTRQ0_IRQ	94
+#define FLTRQ1_IRQ	95
+#define FLCTL_IPR_ADDR	INTC_IPRH
+#define FLCTL_IPR_POS	1
+#define FLCTL_PRIORITY	3
+
+/* IIC(0) (IIC Bus Interface) */
+#define IIC0_ALI_IRQ	96
+#define IIC0_TACKI_IRQ	97
+#define IIC0_WAITI_IRQ	98
+#define IIC0_DTEI_IRQ	99
+#define IIC0_IPR_ADDR	INTC_IPRH
+#define IIC0_IPR_POS	0
+#define IIC0_PRIORITY	3
+
+/* IIC(1) (IIC Bus Interface) */
+#define IIC1_ALI_IRQ	44
+#define IIC1_TACKI_IRQ	45
+#define IIC1_WAITI_IRQ	46
+#define IIC1_DTEI_IRQ	47
+#define IIC1_IPR_ADDR	INTC_IPRI
+#define IIC1_IPR_POS	0
+#define IIC1_PRIORITY	3
+
+/* SIO0 */
+#define SIO0_IRQ	88
+#define SIO0_IPR_ADDR	INTC_IPRI
+#define SIO0_IPR_POS	3
+#define SIO0_PRIORITY	3
+
+/* SDHI */
+#define SDHI_SDHII0_IRQ	100
+#define SDHI_SDHII1_IRQ	101
+#define SDHI_SDHII2_IRQ	102
+#define SDHI_SDHII3_IRQ	103
+#define SDHI_IPR_ADDR	INTC_IPRK
+#define SDHI_IPR_POS	0
+#define SDHI_PRIORITY	3
+
+/* SIU (Sound Interface Unit) */
+#define SIU_IRQ		108
+#define SIU_IPR_ADDR	INTC_IPRJ
+#define SIU_IPR_POS	1
+#define SIU_PRIORITY	3
+
+#define PORT_PACR	0xA4050100UL
+#define PORT_PBCR	0xA4050102UL
+#define PORT_PCCR	0xA4050104UL
+#define PORT_PDCR	0xA4050106UL
+#define PORT_PECR	0xA4050108UL
+#define PORT_PFCR	0xA405010AUL
+#define PORT_PGCR	0xA405010CUL
+#define PORT_PHCR	0xA405010EUL
+#define PORT_PJCR	0xA4050110UL
+#define PORT_PKCR	0xA4050112UL
+#define PORT_PLCR	0xA4050114UL
+#define PORT_SCPCR	0xA4050116UL
+#define PORT_PMCR	0xA4050118UL
+#define PORT_PNCR	0xA405011AUL
+#define PORT_PQCR	0xA405011CUL
+#define PORT_PRCR	0xA405011EUL
+#define PORT_PTCR	0xA405014CUL
+#define PORT_PUCR	0xA405014EUL
+#define PORT_PVCR	0xA4050150UL
+
+#define PORT_PSELA	0xA4050140UL
+#define PORT_PSELB	0xA4050142UL
+#define PORT_PSELC	0xA4050144UL
+#define PORT_PSELE	0xA4050158UL
+
+#define PORT_HIZCRA	0xA4050146UL
+#define PORT_HIZCRB	0xA4050148UL
+#define PORT_DRVCR	0xA405014AUL
+
+#define PORT_PADR  	0xA4050120UL
+#define PORT_PBDR  	0xA4050122UL
+#define PORT_PCDR  	0xA4050124UL
+#define PORT_PDDR  	0xA4050126UL
+#define PORT_PEDR  	0xA4050128UL
+#define PORT_PFDR  	0xA405012AUL
+#define PORT_PGDR  	0xA405012CUL
+#define PORT_PHDR  	0xA405012EUL
+#define PORT_PJDR  	0xA4050130UL
+#define PORT_PKDR  	0xA4050132UL
+#define PORT_PLDR  	0xA4050134UL
+#define PORT_SCPDR  	0xA4050136UL
+#define PORT_PMDR  	0xA4050138UL
+#define PORT_PNDR  	0xA405013AUL
+#define PORT_PQDR  	0xA405013CUL
+#define PORT_PRDR  	0xA405013EUL
+#define PORT_PTDR  	0xA405016CUL
+#define PORT_PUDR  	0xA405016EUL
+#define PORT_PVDR  	0xA4050170UL
+
+#define IRQ0_IRQ	32
+#define IRQ1_IRQ	33
+#define IRQ2_IRQ	34
+#define IRQ3_IRQ	35
+#define IRQ4_IRQ	36
+#define IRQ5_IRQ	37
+#define IRQ6_IRQ	38
+#define IRQ7_IRQ	39
+
+#define INTPRI00	0xA4140010UL
+
+#define IRQ0_IPR_ADDR	INTPRI00
+#define IRQ1_IPR_ADDR	INTPRI00
+#define IRQ2_IPR_ADDR	INTPRI00
+#define IRQ3_IPR_ADDR	INTPRI00
+#define IRQ4_IPR_ADDR	INTPRI00
+#define IRQ5_IPR_ADDR	INTPRI00
+#define IRQ6_IPR_ADDR	INTPRI00
+#define IRQ7_IPR_ADDR	INTPRI00
+
+#define IRQ0_IPR_POS	7
+#define IRQ1_IPR_POS	6
+#define IRQ2_IPR_POS	5
+#define IRQ3_IPR_POS	4
+#define IRQ4_IPR_POS	3
+#define IRQ5_IPR_POS	2
+#define IRQ6_IPR_POS	1
+#define IRQ7_IPR_POS	0
+
+#define IRQ0_PRIORITY	1
+#define IRQ1_PRIORITY	1
+#define IRQ2_PRIORITY	1
+#define IRQ3_PRIORITY	1
+#define IRQ4_PRIORITY	1
+#define IRQ5_PRIORITY	1
+#define IRQ6_PRIORITY	1
+#define IRQ7_PRIORITY	1
+
+#endif /* __ASM_SH_IRQ_SH7343_H */
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 648102e9236f..00886f9adb4d 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -192,7 +192,7 @@
 
 #if defined (CONFIG_CPU_SUBTYPE_SH7707) || defined (CONFIG_CPU_SUBTYPE_SH7708) || \
     defined (CONFIG_CPU_SUBTYPE_SH7709) || defined (CONFIG_CPU_SUBTYPE_SH7750) || \
-    defined (CONFIG_CPU_SUBTYPE_SH7751)
+    defined (CONFIG_CPU_SUBTYPE_SH7751) || defined (CONFIG_CPU_SUBTYPE_SH7706)
 #define SCI_ERI_IRQ	23
 #define SCI_RXI_IRQ	24
 #define SCI_TXI_IRQ	25
@@ -207,6 +207,7 @@
 #define SCIF0_IPR_POS	3
 #define SCIF0_PRIORITY	3
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7707) || \
       defined(CONFIG_CPU_SUBTYPE_SH7709)
 #define SCIF_ERI_IRQ	56
@@ -261,9 +262,12 @@
 #elif defined(CONFIG_CPU_SUBTYPE_SH7708)
 # define ONCHIP_NR_IRQS 32
 #elif defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7705)
 # define ONCHIP_NR_IRQS 64	// Actually 61
 # define PINT_NR_IRQS   16
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+# define ONCHIP_NR_IRQS 104
 #elif defined(CONFIG_CPU_SUBTYPE_SH7750)
 # define ONCHIP_NR_IRQS 48	// Actually 44
 #elif defined(CONFIG_CPU_SUBTYPE_SH7751)
@@ -275,7 +279,8 @@
 #elif defined(CONFIG_CPU_SUBTYPE_ST40STB1)
 # define ONCHIP_NR_IRQS 144
 #elif defined(CONFIG_CPU_SUBTYPE_SH7300) || \
-      defined(CONFIG_CPU_SUBTYPE_SH73180)
+      defined(CONFIG_CPU_SUBTYPE_SH73180) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7343)
 # define ONCHIP_NR_IRQS 109
 #elif defined(CONFIG_CPU_SUBTYPE_SH7780)
 # define ONCHIP_NR_IRQS 111
@@ -476,8 +481,10 @@ extern int ipr_irq_demux(int irq);
 
 #define INTC_ICR	0xfffffee0UL
 #elif defined(CONFIG_CPU_SUBTYPE_SH7705) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7706) || \
       defined(CONFIG_CPU_SUBTYPE_SH7707) || \
-      defined(CONFIG_CPU_SUBTYPE_SH7709)
+      defined(CONFIG_CPU_SUBTYPE_SH7709) || \
+      defined(CONFIG_CPU_SUBTYPE_SH7710)
 #define INTC_IRR0	0xa4000004UL
 #define INTC_IRR1	0xa4000006UL
 #define INTC_IRR2	0xa4000008UL
@@ -496,8 +503,105 @@ extern int ipr_irq_demux(int irq);
 #define INTC_IPRF	0xa4080000UL
 #define INTC_IPRG	0xa4080002UL
 #define INTC_IPRH	0xa4080004UL
-#endif
+#elif defined(CONFIG_CPU_SUBTYPE_SH7710)
+/* Interrupt Controller Registers */
+#undef INTC_IPRA
+#undef INTC_IPRB
+#define INTC_IPRA  	0xA414FEE2UL
+#define INTC_IPRB  	0xA414FEE4UL
+#define INTC_IPRF  	0xA4080000UL
+#define INTC_IPRG  	0xA4080002UL
+#define INTC_IPRH  	0xA4080004UL
+#define INTC_IPRI  	0xA4080006UL
+
+#undef INTC_ICR0
+#undef INTC_ICR1
+#define INTC_ICR0	0xA414FEE0UL
+#define INTC_ICR1	0xA4140010UL
+
+#define INTC_IRR0	0xa4000004UL
+#define INTC_IRR1	0xa4000006UL
+#define INTC_IRR2	0xa4000008UL
+#define INTC_IRR3	0xa400000AUL
+#define INTC_IRR4	0xa400000CUL
+#define INTC_IRR5	0xa4080020UL
+#define INTC_IRR7	0xa4080024UL
+#define INTC_IRR8	0xa4080026UL
+
+/* Interrupt numbers */
+#define TIMER2_IRQ      18
+#define TIMER2_IPR_ADDR INTC_IPRA
+#define TIMER2_IPR_POS   1
+#define TIMER2_PRIORITY  2
 
+/* WDT */
+#define WDT_IRQ		27
+#define WDT_IPR_ADDR	INTC_IPRB
+#define WDT_IPR_POS	 3
+#define WDT_PRIORITY	 2
+
+#define SCIF0_ERI_IRQ	52
+#define SCIF0_RXI_IRQ	53
+#define SCIF0_BRI_IRQ	54
+#define SCIF0_TXI_IRQ	55
+#define SCIF0_IPR_ADDR	INTC_IPRE
+#define SCIF0_IPR_POS	2
+#define SCIF0_PRIORITY	3
+
+#define DMTE4_IRQ	76
+#define DMTE5_IRQ	77
+#define DMA2_IPR_ADDR	INTC_IPRF
+#define DMA2_IPR_POS	2
+#define DMA2_PRIORITY	7
+
+#define IPSEC_IRQ	79
+#define IPSEC_IPR_ADDR	INTC_IPRF
+#define IPSEC_IPR_POS	3
+#define IPSEC_PRIORITY	3
+
+/* EDMAC */
+#define EDMAC0_IRQ	80
+#define EDMAC0_IPR_ADDR	INTC_IPRG
+#define EDMAC0_IPR_POS	3
+#define EDMAC0_PRIORITY	3
+
+#define EDMAC1_IRQ	81
+#define EDMAC1_IPR_ADDR	INTC_IPRG
+#define EDMAC1_IPR_POS	2
+#define EDMAC1_PRIORITY	3
+
+#define EDMAC2_IRQ	82
+#define EDMAC2_IPR_ADDR	INTC_IPRG
+#define EDMAC2_IPR_POS	1
+#define EDMAC2_PRIORITY	3
+
+/* SIOF */
+#define SIOF0_ERI_IRQ	96
+#define SIOF0_TXI_IRQ	97
+#define SIOF0_RXI_IRQ	98
+#define SIOF0_CCI_IRQ	99
+#define SIOF0_IPR_ADDR	INTC_IPRH
+#define SIOF0_IPR_POS	0
+#define SIOF0_PRIORITY	7
+
+#define SIOF1_ERI_IRQ	100
+#define SIOF1_TXI_IRQ	101
+#define SIOF1_RXI_IRQ	102
+#define SIOF1_CCI_IRQ	103
+#define SIOF1_IPR_ADDR	INTC_IPRI
+#define SIOF1_IPR_POS	1
+#define SIOF1_PRIORITY	7
+#endif /* CONFIG_CPU_SUBTYPE_SH7710 */
+
+#if defined(CONFIG_CPU_SUBTYPE_SH7710)
+#define PORT_PACR	0xa4050100UL
+#define PORT_PBCR	0xa4050102UL
+#define PORT_PCCR	0xa4050104UL
+#define PORT_PETCR	0xa4050106UL
+#define PORT_PADR  	0xa4050120UL
+#define PORT_PBDR  	0xa4050122UL
+#define PORT_PCDR  	0xa4050124UL
+#else
 #define PORT_PACR	0xa4000100UL
 #define PORT_PBCR	0xa4000102UL
 #define PORT_PCCR	0xa4000104UL
@@ -506,6 +610,7 @@ extern int ipr_irq_demux(int irq);
 #define PORT_PBDR  	0xa4000122UL
 #define PORT_PCDR  	0xa4000124UL
 #define PORT_PFDR  	0xa400012aUL
+#endif
 
 #define IRQ0_IRQ	32
 #define IRQ1_IRQ	33
@@ -599,6 +704,8 @@ void intc2_add_clear_irq(int irq, int (*fn)(int));
 
 #endif
 
+extern int shmse_irq_demux(int irq);
+
 static inline int generic_irq_demux(int irq)
 {
 	return irq;
@@ -614,4 +721,8 @@ static inline int generic_irq_demux(int irq)
 #include <asm/irq-sh73180.h>
 #endif
 
+#if defined(CONFIG_CPU_SUBTYPE_SH7343)
+#include <asm/irq-sh7343.h>
+#endif
+
 #endif /* __ASM_SH_IRQ_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index a22732007dd4..3b3ef4f2bf31 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -38,13 +38,15 @@ enum cpu_type {
 	CPU_SH7604,
 
 	/* SH-3 types */
-	CPU_SH7705, CPU_SH7707,  CPU_SH7708, CPU_SH7708S, CPU_SH7708R,
-	CPU_SH7709, CPU_SH7709A, CPU_SH7729, CPU_SH7300,
+	CPU_SH7705, CPU_SH7706, CPU_SH7707,
+	CPU_SH7708, CPU_SH7708S, CPU_SH7708R,
+	CPU_SH7709, CPU_SH7709A, CPU_SH7710,
+	CPU_SH7729, CPU_SH7300,
 
 	/* SH-4 types */
 	CPU_SH7750, CPU_SH7750S, CPU_SH7750R, CPU_SH7751, CPU_SH7751R,
 	CPU_SH7760, CPU_ST40RA, CPU_ST40GX1, CPU_SH4_202, CPU_SH4_501,
-	CPU_SH73180, CPU_SH7770, CPU_SH7780, CPU_SH7781,
+	CPU_SH73180, CPU_SH7343, CPU_SH7770, CPU_SH7780, CPU_SH7781,
 
 	/* Unknown subtype */
 	CPU_SH_NONE
-- 
cgit v1.2.1


From 51e22e7a05c1c6f2e38ac7459d3404e32e543b75 Mon Sep 17 00:00:00 2001
From: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Date: Wed, 27 Sep 2006 17:41:31 +0900
Subject: sh: SHMIN board support.

This adds support for the SHMIN SH7706 board.

Signed-off-by: Takashi YOSHII <takasi-y@ops.dti.ne.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                 |   8 +-
 arch/sh/Makefile                |   1 +
 arch/sh/boards/shmin/Makefile   |   5 +
 arch/sh/boards/shmin/setup.c    |  49 +++
 arch/sh/configs/shmin_defconfig | 827 ++++++++++++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types        |   1 +
 include/asm-sh/shmin/shmin.h    |   9 +
 7 files changed, 899 insertions(+), 1 deletion(-)
 create mode 100644 arch/sh/boards/shmin/Makefile
 create mode 100644 arch/sh/boards/shmin/setup.c
 create mode 100644 arch/sh/configs/shmin_defconfig
 create mode 100644 include/asm-sh/shmin/shmin.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 557e7232141c..5bec3af164e5 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -222,6 +222,12 @@ config SH_TITAN
 	  Select Titan if you are configuring for a Nimble Microsystems
 	  NetEngine NP51R.
 
+config SH_SHMIN
+	bool "SHMIN"
+	select CPU_SUBTYPE_SH7706
+	help
+	  Select SHMIN if configureing for the SHMIN board
+
 config SH_UNKNOWN
 	bool "BareCPU"
 	help
@@ -295,7 +301,7 @@ config SH_RTC
 	bool
 	depends on !SH_DREAMCAST && !SH_SATURN && !SH_7300_SOLUTION_ENGINE && \
 		   !SH_73180_SOLUTION_ENGINE && !SH_LANDISK && \
-		   !SH_R7780RP
+		   !SH_R7780RP && !SH_SHMIN
 	default y
 	help
 	  Selecting this option will allow the Linux kernel to emulate
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index d118e48ea1ba..8ad0a286ff8f 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -113,6 +113,7 @@ machdir-$(CONFIG_SH_R7780RP)			:= renesas/r7780rp
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_TITAN)			:= titan
+machdir-$(CONFIG_SH_SHMIN)			:= shmin
 machdir-$(CONFIG_SH_UNKNOWN)			:= unknown
 
 incdir-y			:= $(notdir $(machdir-y))
diff --git a/arch/sh/boards/shmin/Makefile b/arch/sh/boards/shmin/Makefile
new file mode 100644
index 000000000000..3190cc72430e
--- /dev/null
+++ b/arch/sh/boards/shmin/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the SHMIN board.
+#
+
+obj-y	 := setup.o
diff --git a/arch/sh/boards/shmin/setup.c b/arch/sh/boards/shmin/setup.c
new file mode 100644
index 000000000000..a950ec4d765d
--- /dev/null
+++ b/arch/sh/boards/shmin/setup.c
@@ -0,0 +1,49 @@
+/*
+ * arch/sh/boards/shmin/setup.c
+ *
+ * Copyright (C) 2006 Takashi YOSHII
+ *
+ * SHMIN Support.
+ */
+#include <linux/init.h>
+#include <asm/machvec.h>
+#include <asm/shmin/shmin.h>
+#include <asm/clock.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+
+#define PFC_PHCR	0xa400010e
+
+const char *get_system_type(void)
+{
+	return "SHMIN";
+}
+
+static void __init init_shmin_irq(void)
+{
+	ctrl_outw(0x2a00, PFC_PHCR);	// IRQ0-3=IRQ
+	ctrl_outw(0x0aaa, INTC_ICR1);	// IRQ0-3=IRQ-mode,Low-active.
+}
+
+void __init platform_setup(void)
+{
+}
+
+static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
+{
+	static int dummy;
+
+	if ((port & ~0x1f) == SHMIN_NE_BASE)
+		return (void __iomem *)(SHMIN_IO_BASE + port);
+
+	dummy = 0;
+
+	return &dummy;
+
+}
+
+struct sh_machine_vector mv_shmin __initmv = {
+	.mv_init_irq	= init_shmin_irq,
+	.mv_ioport_map	= shmin_ioport_map,
+};
+ALIAS_MV(shmin)
diff --git a/arch/sh/configs/shmin_defconfig b/arch/sh/configs/shmin_defconfig
new file mode 100644
index 000000000000..382b3bd3963b
--- /dev/null
+++ b/arch/sh/configs/shmin_defconfig
@@ -0,0 +1,827 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.17
+# Wed Aug  2 01:45:03 2006
+#
+CONFIG_SUPERH=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_PTRACE=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SWAP is not set
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_SYSCTL is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_UID16 is not set
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+# CONFIG_BUG is not set
+# CONFIG_ELF_CORE is not set
+# CONFIG_BASE_FULL is not set
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SHMEM is not set
+# CONFIG_SLAB is not set
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=1
+CONFIG_SLOB=y
+CONFIG_OBSOLETE_INTERMODULE=y
+
+#
+# Loadable module support
+#
+# CONFIG_MODULES is not set
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+# CONFIG_LSF is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7709_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_STB1_HARP is not set
+# CONFIG_SH_STB1_OVERDRIVE is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_CQREEK is not set
+# CONFIG_SH_DMIDA is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_CAT68701 is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_SH2000 is not set
+# CONFIG_SH_ADX is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+CONFIG_SH_SHMIN=y
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH3=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+CONFIG_CPU_SUBTYPE_SH7706=y
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x00800000
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_SH_FPU_EMU is not set
+# CONFIG_SH_DSP is not set
+# CONFIG_SH_ADC is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ=32000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+# CONFIG_SH_DMA is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_KEXEC is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00210000
+# CONFIG_UBC_WAKEUP is not set
+CONFIG_CMDLINE_BOOL=y
+CONFIG_CMDLINE="console=ttySC1,115200 root=1f01 mtdparts=phys_mapped_flash:64k(firm)ro,-(sys) netdev=34,0x300,eth0 "
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options (EXPERIMENTAL)
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+# CONFIG_PACKET is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_FW_LOADER is not set
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+CONFIG_MTD_CMDLINE_PARTS=y
+
+#
+# User Modules And Translation Layers
+#
+# CONFIG_MTD_CHAR is not set
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+CONFIG_MTD_JEDECPROBE=y
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+# CONFIG_MTD_RAM is not set
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+CONFIG_MTD_PHYSMAP=y
+CONFIG_MTD_PHYSMAP_START=0xa0000000
+CONFIG_MTD_PHYSMAP_LEN=0x80000
+CONFIG_MTD_PHYSMAP_BANKWIDTH=1
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# OneNAND Flash Device Drivers
+#
+# CONFIG_MTD_ONENAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+CONFIG_BLK_DEV_LOOP=y
+# CONFIG_BLK_DEV_CRYPTOLOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_STNIC is not set
+# CONFIG_SMC91X is not set
+CONFIG_NE2000=y
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+# CONFIG_SYSFS is not set
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+CONFIG_CRAMFS=y
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_UNWIND_INFO is not set
+CONFIG_SH_STANDARD_BIOS=y
+CONFIG_EARLY_PRINTK=y
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 1f7f2c255ca2..ffd218083876 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -29,3 +29,4 @@ LANDISK			SH_LANDISK
 R7780RP			SH_R7780RP
 R7780MP			SH_R7780MP
 TITAN			SH_TITAN
+SHMIN			SH_SHMIN
diff --git a/include/asm-sh/shmin/shmin.h b/include/asm-sh/shmin/shmin.h
new file mode 100644
index 000000000000..36ba138a81fb
--- /dev/null
+++ b/include/asm-sh/shmin/shmin.h
@@ -0,0 +1,9 @@
+#ifndef __ASM_SH_SHMIN_H
+#define __ASM_SH_SHMIN_H
+
+#define SHMIN_IO_BASE 0xb0000000UL
+
+#define SHMIN_NE_IRQ IRQ2_IRQ
+#define SHMIN_NE_BASE 0x300
+
+#endif
-- 
cgit v1.2.1


From 91550f715b7f7707b5ab5b9b0cd455bda8505954 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:45:01 +0900
Subject: sh: Kill off the rest of the legacy rtc mess.

With the new RTC class driver, we can get rid of most of the
old left over cruft.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig               |  13 ----
 arch/sh/boards/snapgear/rtc.c |  27 +------
 arch/sh/kernel/cpu/Makefile   |   1 -
 arch/sh/kernel/cpu/irq/ipr.c  |   2 +-
 arch/sh/kernel/cpu/rtc.c      | 128 --------------------------------
 arch/sh/kernel/time.c         |  41 +++++-----
 drivers/char/Kconfig          |   2 +-
 include/asm-sh/mc146818rtc.h  | 169 ------------------------------------------
 include/asm-sh/rtc.h          |  21 ------
 9 files changed, 23 insertions(+), 381 deletions(-)
 delete mode 100644 arch/sh/kernel/cpu/rtc.c

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 5bec3af164e5..c7e3596dcde1 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -296,19 +296,6 @@ config CPU_LITTLE_ENDIAN
 	  endian byte order. These modes require different kernels. Say Y if
 	  your machine is little endian, N if it's a big endian machine.
 
-# The SH7750 RTC module is disabled in the Dreamcast
-config SH_RTC
-	bool
-	depends on !SH_DREAMCAST && !SH_SATURN && !SH_7300_SOLUTION_ENGINE && \
-		   !SH_73180_SOLUTION_ENGINE && !SH_LANDISK && \
-		   !SH_R7780RP && !SH_SHMIN
-	default y
-	help
-	  Selecting this option will allow the Linux kernel to emulate
-	  PC's RTC.
-
-	  If unsure, say N.
-
 config SH_FPU
 	bool "FPU support"
 	depends on !CPU_SH3
diff --git a/arch/sh/boards/snapgear/rtc.c b/arch/sh/boards/snapgear/rtc.c
index 722479deb55a..1659fdd6695a 100644
--- a/arch/sh/boards/snapgear/rtc.c
+++ b/arch/sh/boards/snapgear/rtc.c
@@ -19,9 +19,7 @@
 #include <linux/mc146818rtc.h>
 #include <asm/io.h>
 
-/****************************************************************************/
-
-static int use_ds1302 = 0;
+static int use_ds1302;
 
 /****************************************************************************/
 /*
@@ -79,10 +77,6 @@ static unsigned int ds1302_readbyte(unsigned int addr)
 	unsigned int	val;
 	unsigned long	flags;
 
-#if 0
-	printk("SnapGear RTC: ds1302_readbyte(addr=%x)\n", addr);
-#endif
-
 	local_irq_save(flags);
 	set_dirp(get_dirp() | RTC_RESET | RTC_IODATA | RTC_SCLK);
 	set_dp(get_dp() & ~(RTC_RESET | RTC_IODATA | RTC_SCLK));
@@ -101,10 +95,6 @@ static void ds1302_writebyte(unsigned int addr, unsigned int val)
 {
 	unsigned long	flags;
 
-#if 0
-	printk("SnapGear RTC: ds1302_writebyte(addr=%x)\n", addr);
-#endif
-
 	local_irq_save(flags);
 	set_dirp(get_dirp() | RTC_RESET | RTC_IODATA | RTC_SCLK);
 	set_dp(get_dp() & ~(RTC_RESET | RTC_IODATA | RTC_SCLK));
@@ -167,9 +157,6 @@ void __init secureedge5410_rtc_init(void)
 	if (use_ds1302) {
 		rtc_sh_get_time = snapgear_rtc_gettimeofday;
 		rtc_sh_set_time = snapgear_rtc_settimeofday;
-	} else {
-		rtc_sh_get_time = sh_rtc_gettimeofday;
-		rtc_sh_set_time = sh_rtc_settimeofday;
 	}
 		
 	printk("SnapGear RTC: using %s rtc.\n", use_ds1302 ? "ds1302" : "internal");
@@ -184,10 +171,8 @@ void snapgear_rtc_gettimeofday(struct timespec *ts)
 {
 	unsigned int sec, min, hr, day, mon, yr;
 
-	if (!use_ds1302) {
-		sh_rtc_gettimeofday(ts);
+	if (!use_ds1302)
 		return;
-	}
 
  	sec = bcd2int(ds1302_readbyte(RTC_ADDR_SEC));
  	min = bcd2int(ds1302_readbyte(RTC_ADDR_MIN));
@@ -228,7 +213,7 @@ int snapgear_rtc_settimeofday(const time_t secs)
 	unsigned long nowtime;
 
 	if (!use_ds1302)
-		return sh_rtc_settimeofday(secs);
+		return 0;
 
 /*
  *	This is called direct from the kernel timer handling code.
@@ -237,10 +222,6 @@ int snapgear_rtc_settimeofday(const time_t secs)
 
 	nowtime = secs;
 
-#if 1
-	printk("SnapGear RTC: snapgear_rtc_settimeofday(nowtime=%ld)\n", nowtime);
-#endif
-
 	/* STOP RTC */
 	ds1302_writebyte(RTC_ADDR_SEC, ds1302_readbyte(RTC_ADDR_SEC) | 0x80);
 
@@ -326,5 +307,3 @@ void secureedge5410_cmos_write(unsigned char val, int addr)
 	default:                                                      break;
 	}
 }
-
-/****************************************************************************/
diff --git a/arch/sh/kernel/cpu/Makefile b/arch/sh/kernel/cpu/Makefile
index 59d5b748752f..fb5dac069382 100644
--- a/arch/sh/kernel/cpu/Makefile
+++ b/arch/sh/kernel/cpu/Makefile
@@ -8,6 +8,5 @@ obj-$(CONFIG_CPU_SH2)		+= sh2/
 obj-$(CONFIG_CPU_SH3)		+= sh3/
 obj-$(CONFIG_CPU_SH4)		+= sh4/
 
-obj-$(CONFIG_SH_RTC)		+= rtc.o
 obj-$(CONFIG_UBC_WAKEUP)	+= ubc.o
 obj-$(CONFIG_SH_ADC)		+= adc.o
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index d2b715a04c7e..00edee971bcb 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -124,7 +124,7 @@ void __init init_IRQ(void)
 #ifndef CONFIG_CPU_SUBTYPE_SH7780
 	make_ipr_irq(TIMER_IRQ, TIMER_IPR_ADDR, TIMER_IPR_POS, TIMER_PRIORITY);
 	make_ipr_irq(TIMER1_IRQ, TIMER1_IPR_ADDR, TIMER1_IPR_POS, TIMER1_PRIORITY);
-#if defined(CONFIG_SH_RTC)
+#ifdef RTC_IRQ
 	make_ipr_irq(RTC_IRQ, RTC_IPR_ADDR, RTC_IPR_POS, RTC_PRIORITY);
 #endif
 
diff --git a/arch/sh/kernel/cpu/rtc.c b/arch/sh/kernel/cpu/rtc.c
deleted file mode 100644
index 4304cf75cfa2..000000000000
--- a/arch/sh/kernel/cpu/rtc.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * linux/arch/sh/kernel/rtc.c -- SH3 / SH4 on-chip RTC support
- *
- *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
- */
-
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/sched.h>
-#include <linux/time.h>
-#include <linux/bcd.h>
-#include <asm/io.h>
-#include <asm/rtc.h>
-
-void sh_rtc_gettimeofday(struct timespec *ts)
-{
-	unsigned int sec128, sec, sec2, min, hr, wk, day, mon, yr, yr100, cf_bit;
-	unsigned long flags;
-
- again:
-	do {
-		local_irq_save(flags);
-		ctrl_outb(0, RCR1);  /* Clear CF-bit */
-		sec128 = ctrl_inb(R64CNT);
-		sec = ctrl_inb(RSECCNT);
-		min = ctrl_inb(RMINCNT);
-		hr  = ctrl_inb(RHRCNT);
-		wk  = ctrl_inb(RWKCNT);
-		day = ctrl_inb(RDAYCNT);
-		mon = ctrl_inb(RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		yr  = ctrl_inw(RYRCNT);
-		yr100 = (yr >> 8);
-		yr &= 0xff;
-#else
-		yr  = ctrl_inb(RYRCNT);
-		yr100 = (yr == 0x99) ? 0x19 : 0x20;
-#endif
-		sec2 = ctrl_inb(R64CNT);
-		cf_bit = ctrl_inb(RCR1) & RCR1_CF;
-		local_irq_restore(flags);
-	} while (cf_bit != 0 || ((sec128 ^ sec2) & RTC_BIT_INVERTED) != 0);
-
-	BCD_TO_BIN(yr100);
-	BCD_TO_BIN(yr);
-	BCD_TO_BIN(mon);
-	BCD_TO_BIN(day);
-	BCD_TO_BIN(hr);
-	BCD_TO_BIN(min);
-	BCD_TO_BIN(sec);
-
-	if (yr > 99 || mon < 1 || mon > 12 || day > 31 || day < 1 ||
-	    hr > 23 || min > 59 || sec > 59) {
-		printk(KERN_ERR
-		       "SH RTC: invalid value, resetting to 1 Jan 2000\n");
-		local_irq_save(flags);
-		ctrl_outb(RCR2_RESET, RCR2);  /* Reset & Stop */
-		ctrl_outb(0, RSECCNT);
-		ctrl_outb(0, RMINCNT);
-		ctrl_outb(0, RHRCNT);
-		ctrl_outb(6, RWKCNT);
-		ctrl_outb(1, RDAYCNT);
-		ctrl_outb(1, RMONCNT);
-#if defined(CONFIG_CPU_SH4)
-		ctrl_outw(0x2000, RYRCNT);
-#else
-		ctrl_outb(0, RYRCNT);
-#endif
-		ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start */
-		goto again;
-	}
-
-#if RTC_BIT_INVERTED != 0
-	if ((sec128 & RTC_BIT_INVERTED))
-		sec--;
-#endif
-
-	ts->tv_sec = mktime(yr100 * 100 + yr, mon, day, hr, min, sec);
-	ts->tv_nsec = ((sec128 * 1000000) / 128) * 1000;
-}
-
-/*
- * Changed to only care about tv_sec, and not the full timespec struct
- * (i.e. tv_nsec).  It can easily be switched to timespec for future cpus
- * that support setting usec or nsec RTC values.
- */
-int sh_rtc_settimeofday(const time_t secs)
-{
-	int retval = 0;
-	int real_seconds, real_minutes, cmos_minutes;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ctrl_outb(RCR2_RESET, RCR2);  /* Reset pre-scaler & stop RTC */
-
-	cmos_minutes = ctrl_inb(RMINCNT);
-	BCD_TO_BIN(cmos_minutes);
-
-	/*
-	 * since we're only adjusting minutes and seconds,
-	 * don't interfere with hour overflow. This avoids
-	 * messing with unknown time zones but requires your
-	 * RTC not to be off by more than 15 minutes
-	 */
-	real_seconds = secs % 60;
-	real_minutes = secs / 60;
-	if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
-		real_minutes += 30;	/* correct for half hour time zone */
-	real_minutes %= 60;
-
-	if (abs(real_minutes - cmos_minutes) < 30) {
-		BIN_TO_BCD(real_seconds);
-		BIN_TO_BCD(real_minutes);
-		ctrl_outb(real_seconds, RSECCNT);
-		ctrl_outb(real_minutes, RMINCNT);
-	} else {
-		printk(KERN_WARNING
-		       "set_rtc_time: can't update from %d to %d\n",
-		       cmos_minutes, real_minutes);
-		retval = -1;
-	}
-
-	ctrl_outb(RCR2_RTCEN|RCR2_START, RCR2);  /* Start RTC */
-	local_irq_restore(flags);
-
-	return retval;
-}
diff --git a/arch/sh/kernel/time.c b/arch/sh/kernel/time.c
index 8acd70bffe76..149d9713eddf 100644
--- a/arch/sh/kernel/time.c
+++ b/arch/sh/kernel/time.c
@@ -3,13 +3,12 @@
  *
  *  Copyright (C) 1999  Tetsuya Okada & Niibe Yutaka
  *  Copyright (C) 2000  Philipp Rumpf <prumpf@tux.org>
- *  Copyright (C) 2002, 2003, 2004, 2005  Paul Mundt
+ *  Copyright (C) 2002 - 2006  Paul Mundt
  *  Copyright (C) 2002  M. R. Brown  <mrbrown@linux-sh.org>
  *
  *  Some code taken from i386 version.
  *    Copyright (C) 1991, 1992, 1995  Linus Torvalds
  */
-
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
@@ -26,15 +25,20 @@ struct sys_timer *sys_timer;
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL(rtc_lock);
 
-/* XXX: Can we initialize this in a routine somewhere?  Dreamcast doesn't want
- * these routines anywhere... */
-#ifdef CONFIG_SH_RTC
-void (*rtc_sh_get_time)(struct timespec *) = sh_rtc_gettimeofday;
-int (*rtc_sh_set_time)(const time_t) = sh_rtc_settimeofday;
-#else
-void (*rtc_sh_get_time)(struct timespec *);
-int (*rtc_sh_set_time)(const time_t);
-#endif
+/* Dummy RTC ops */
+static void null_rtc_get_time(struct timespec *tv)
+{
+	tv->tv_sec = mktime(2000, 1, 1, 0, 0, 0);
+	tv->tv_nsec = 0;
+}
+
+static int null_rtc_set_time(const time_t secs)
+{
+	return 0;
+}
+
+void (*rtc_sh_get_time)(struct timespec *) = null_rtc_get_time;
+int (*rtc_sh_set_time)(const time_t) = null_rtc_set_time;
 
 /*
  * Scheduler clock - returns current time in nanosec units.
@@ -70,7 +74,6 @@ void do_gettimeofday(struct timeval *tv)
 	tv->tv_sec = sec;
 	tv->tv_usec = usec;
 }
-
 EXPORT_SYMBOL(do_gettimeofday);
 
 int do_settimeofday(struct timespec *tv)
@@ -103,7 +106,6 @@ int do_settimeofday(struct timespec *tv)
 
 	return 0;
 }
-
 EXPORT_SYMBOL(do_settimeofday);
 
 /* last time the RTC clock got updated */
@@ -181,7 +183,6 @@ static int __init timer_init_sysfs(void)
 	sys_timer->dev.cls = &timer_sysclass;
 	return sysdev_register(&sys_timer->dev);
 }
-
 device_initcall(timer_init_sysfs);
 
 void (*board_time_init)(void);
@@ -193,15 +194,9 @@ void __init time_init(void)
 
 	clk_init();
 
-	if (rtc_sh_get_time) {
-		rtc_sh_get_time(&xtime);
-	} else {
-		xtime.tv_sec = mktime(2000, 1, 1, 0, 0, 0);
-		xtime.tv_nsec = 0;
-	}
-
-        set_normalized_timespec(&wall_to_monotonic,
-                                -xtime.tv_sec, -xtime.tv_nsec);
+	rtc_sh_get_time(&xtime);
+	set_normalized_timespec(&wall_to_monotonic,
+				-xtime.tv_sec, -xtime.tv_nsec);
 
 	/*
 	 * Find the timer to use as the system timer, it will be
diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 52ea94b891f5..7f0d323b4725 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -739,7 +739,7 @@ config NVRAM
 
 config RTC
 	tristate "Enhanced Real Time Clock Support"
-	depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV && !ARM
+	depends on !PPC && !PARISC && !IA64 && !M68K && (!SPARC || PCI) && !FRV && !ARM && !SUPERH
 	---help---
 	  If you say Y here and create a character special file /dev/rtc with
 	  major number 10 and minor number 135 using mknod ("man mknod"), you
diff --git a/include/asm-sh/mc146818rtc.h b/include/asm-sh/mc146818rtc.h
index adc6e67c6b75..0aee96a97330 100644
--- a/include/asm-sh/mc146818rtc.h
+++ b/include/asm-sh/mc146818rtc.h
@@ -4,173 +4,4 @@
 #ifndef _ASM_MC146818RTC_H
 #define _ASM_MC146818RTC_H
 
-#ifdef CONFIG_SH_MPC1211
-#undef  _ASM_MC146818RTC_H
-#undef  RTC_IRQ
-#include <asm/mpc1211/mc146818rtc.h>
-#else
-
-#include <asm/rtc.h>
-
-#define RTC_ALWAYS_BCD	1
-
-/* FIXME:RTC Interrupt feature is not implemented yet. */
-#undef  RTC_IRQ
-#define RTC_IRQ		0
-
-#if defined(CONFIG_CPU_SH3)
-#define RTC_PORT(n)		(R64CNT+(n)*2)
-#define CMOS_READ(addr)		__CMOS_READ(addr,b)
-#define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,b)
-
-#elif defined(CONFIG_SH_SECUREEDGE5410)
-#include <asm/snapgear.h>
-
-#define RTC_PORT(n)             SECUREEDGE_IOPORT_ADDR
-#define CMOS_READ(addr)         secureedge5410_cmos_read(addr)
-#define CMOS_WRITE(val,addr)    secureedge5410_cmos_write(val,addr)
-extern unsigned char secureedge5410_cmos_read(int addr);
-extern void secureedge5410_cmos_write(unsigned char val, int addr);
-
-#elif defined(CONFIG_CPU_SH4)
-#define RTC_PORT(n)		(R64CNT+(n)*4)
-#define CMOS_READ(addr)		__CMOS_READ(addr,w)
-#define CMOS_WRITE(val,addr)	__CMOS_WRITE(val,addr,w)
-#endif
-
-#define __CMOS_READ(addr, s) ({						\
-	unsigned char val=0, rcr1, rcr2, r64cnt, retry;			\
-	switch(addr) {							\
-		case RTC_SECONDS:					\
-			val = ctrl_inb(RSECCNT);			\
-			break;						\
-		case RTC_SECONDS_ALARM:					\
-			val = ctrl_inb(RSECAR);				\
-			break;						\
-		case RTC_MINUTES:					\
-			val = ctrl_inb(RMINCNT);			\
-			break;						\
-		case RTC_MINUTES_ALARM:					\
-			val = ctrl_inb(RMINAR);				\
-			break;						\
-		case RTC_HOURS:						\
-			val = ctrl_inb(RHRCNT);				\
-			break;						\
-		case RTC_HOURS_ALARM:					\
-			val = ctrl_inb(RHRAR);				\
-			break;						\
-		case RTC_DAY_OF_WEEK:					\
-			val = ctrl_inb(RWKCNT);				\
-			break;						\
-		case RTC_DAY_OF_MONTH:					\
-			val = ctrl_inb(RDAYCNT);			\
-			break;						\
-		case RTC_MONTH:						\
-			val = ctrl_inb(RMONCNT);			\
-			break;						\
-		case RTC_YEAR:						\
-			val = ctrl_in##s(RYRCNT);			\
-			break;						\
-		case RTC_REG_A: /* RTC_FREQ_SELECT */			\
-			rcr2 = ctrl_inb(RCR2);				\
-			val = (rcr2 & RCR2_PESMASK) >> 4;		\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr1 = (rcr1 & (RCR1_CIE | RCR1_AIE)) | RCR1_AF;\
-			retry = 0;					\
-			do {						\
-				ctrl_outb(rcr1, RCR1); /* clear CF */	\
-				r64cnt = ctrl_inb(R64CNT);		\
-			} while((ctrl_inb(RCR1) & RCR1_CF) && retry++ < 1000);\
-			r64cnt ^= RTC_BIT_INVERTED;			\
-			if(r64cnt == 0x7f || r64cnt == 0)		\
-				val |= RTC_UIP;				\
-			break;						\
-		case RTC_REG_B:	/* RTC_CONTROL */			\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			if(rcr1 & RCR1_CIE)	val |= RTC_UIE;		\
-			if(rcr1 & RCR1_AIE)	val |= RTC_AIE;		\
-			if(rcr2 & RCR2_PESMASK)	val |= RTC_PIE;		\
-			if(!(rcr2 & RCR2_START))val |= RTC_SET;		\
-			val |= RTC_24H;					\
-			break;						\
-		case RTC_REG_C:	/* RTC_INTR_FLAGS */			\
-			rcr1 = ctrl_inb(RCR1);				\
-			rcr1 &= ~(RCR1_CF | RCR1_AF);			\
-			ctrl_outb(rcr1, RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			rcr2 &= ~RCR2_PEF;				\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_D:	/* RTC_VALID */				\
-			/* Always valid ... */				\
-			val = RTC_VRT;					\
-			break;						\
-		default:						\
-			break;						\
-	}								\
-	val;								\
-})
-
-#define __CMOS_WRITE(val, addr, s) ({					\
-	unsigned char rcr1,rcr2;					\
-	switch(addr) {							\
-		case RTC_SECONDS:					\
-			ctrl_outb(val, RSECCNT);			\
-			break;						\
-		case RTC_SECONDS_ALARM:					\
-			ctrl_outb(val, RSECAR);				\
-			break;						\
-		case RTC_MINUTES:					\
-			ctrl_outb(val, RMINCNT);			\
-			break;						\
-		case RTC_MINUTES_ALARM:					\
-			ctrl_outb(val, RMINAR);				\
-			break;						\
-		case RTC_HOURS:						\
-			ctrl_outb(val, RHRCNT);				\
-			break;						\
-		case RTC_HOURS_ALARM:					\
-			ctrl_outb(val, RHRAR);				\
-			break;						\
-		case RTC_DAY_OF_WEEK:					\
-			ctrl_outb(val, RWKCNT);				\
-			break;						\
-		case RTC_DAY_OF_MONTH:					\
-			ctrl_outb(val, RDAYCNT);			\
-			break;						\
-		case RTC_MONTH:						\
-			ctrl_outb(val, RMONCNT);			\
-			break;						\
-		case RTC_YEAR:						\
-			ctrl_out##s((ctrl_in##s(RYRCNT) & 0xff00) | (val & 0xff), RYRCNT);\
-			break;						\
-		case RTC_REG_A: /* RTC_FREQ_SELECT */			\
-			rcr2 = ctrl_inb(RCR2);				\
-			if((val & RTC_DIV_CTL) == RTC_DIV_RESET2)	\
-				rcr2 |= RCR2_RESET;			\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_B:	/* RTC_CONTROL */			\
-			rcr1 = (ctrl_inb(RCR1) & 0x99) | RCR1_AF;	\
-			if(val & RTC_AIE) rcr1 |= RCR1_AIE;		\
-			else              rcr1 &= ~RCR1_AIE;		\
-			if(val & RTC_UIE) rcr1 |= RCR1_CIE;		\
-			else              rcr1 &= ~RCR1_CIE;		\
-			ctrl_outb(rcr1, RCR1);				\
-			rcr2 = ctrl_inb(RCR2);				\
-			if(val & RTC_SET) rcr2 &= ~RCR2_START;		\
-			else              rcr2 |= RCR2_START;		\
-			ctrl_outb(rcr2, RCR2);				\
-			break;						\
-		case RTC_REG_C:	/* RTC_INTR_FLAGS */			\
-			break;						\
-		case RTC_REG_D:	/* RTC_VALID */				\
-			break;						\
-		default:						\
-			break;						\
-	}								\
-})
-
-#endif /* CONFIG_SH_MPC1211 */
 #endif /* _ASM_MC146818RTC_H */
diff --git a/include/asm-sh/rtc.h b/include/asm-sh/rtc.h
index 4c7be859c9c0..91aacc96151b 100644
--- a/include/asm-sh/rtc.h
+++ b/include/asm-sh/rtc.h
@@ -1,29 +1,8 @@
 #ifndef _ASM_RTC_H
 #define _ASM_RTC_H
-#ifdef __KERNEL__
 
-#include <asm/machvec.h>
-#include <asm/cpu/rtc.h>
-
-extern void sh_rtc_gettimeofday(struct timespec *ts);
-extern int sh_rtc_settimeofday(const time_t secs);
 extern void (*board_time_init)(void);
 extern void (*rtc_sh_get_time)(struct timespec *);
 extern int (*rtc_sh_set_time)(const time_t);
 
-/* RCR1 Bits */
-#define RCR1_CF		0x80	/* Carry Flag             */
-#define RCR1_CIE	0x10	/* Carry Interrupt Enable */
-#define RCR1_AIE	0x08	/* Alarm Interrupt Enable */
-#define RCR1_AF		0x01	/* Alarm Flag             */
-
-/* RCR2 Bits */
-#define RCR2_PEF	0x80	/* PEriodic interrupt Flag */
-#define RCR2_PESMASK	0x70	/* Periodic interrupt Set  */
-#define RCR2_RTCEN	0x08	/* ENable RTC              */
-#define RCR2_ADJ	0x04	/* ADJustment (30-second)  */
-#define RCR2_RESET	0x02	/* Reset bit               */
-#define RCR2_START	0x01	/* Start bit               */
-
-#endif /* __KERNEL__ */
 #endif /* _ASM_RTC_H */
-- 
cgit v1.2.1


From c7afb7e5cbc4baa781ec82731fc9fe9039efee22 Mon Sep 17 00:00:00 2001
From: Nobuhiro Iwamatsu <hemamu@t-base.ne.jp>
Date: Wed, 27 Sep 2006 17:50:03 +0900
Subject: sh: Fix memcpy() build error on sh4eb.

A trivial bug breaking the build on sh4eb.

Signed-off-by: Nobuhiro Iwamatsu <hemamu@t-base.ne.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/lib/memcpy-sh4.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/sh/lib/memcpy-sh4.S b/arch/sh/lib/memcpy-sh4.S
index db6b736537ad..560bc17eebdd 100644
--- a/arch/sh/lib/memcpy-sh4.S
+++ b/arch/sh/lib/memcpy-sh4.S
@@ -727,8 +727,8 @@ ENTRY(memcpy)
 	mov.l	@(0x04,r5), r11	!  18 LS (latency=2)
 	xtrct	r9, r8		!  48 EX
 
-	mov.w	@(0x02,r5), r12	!  18 LS (latency=2)
-	xtrct	r10, r9		!  48 EX
+	mov.l   @(0x00,r5), r12 !  18 LS (latency=2)
+    	xtrct	r10, r9		!  48 EX
 
 	movca.l	r0,@r1		!  40 LS (latency=3-7)
 	add	#-0x1c, r1	!  50 EX
-- 
cgit v1.2.1


From 15f57a29a19ad0dbb468363cb617b06f71f6de92 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:51:01 +0900
Subject: sh: Add support for cacheline poking through debugfs.

A simple debugging aid for easier visibility of the respective
cachelines.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/Makefile        |   4 ++
 arch/sh/mm/cache-debugfs.c | 147 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 151 insertions(+)
 create mode 100644 arch/sh/mm/cache-debugfs.c

diff --git a/arch/sh/mm/Makefile b/arch/sh/mm/Makefile
index da37d86e65eb..3ffd7f68c0a2 100644
--- a/arch/sh/mm/Makefile
+++ b/arch/sh/mm/Makefile
@@ -17,6 +17,10 @@ mmu-$(CONFIG_MMU)	:= fault.o clear_page.o copy_page.o tlb-flush.o	\
 
 obj-y			+= $(mmu-y)
 
+ifdef CONFIG_DEBUG_FS
+obj-$(CONFIG_CPU_SH4)		+= cache-debugfs.o
+endif
+
 ifdef CONFIG_MMU
 obj-$(CONFIG_CPU_SH3)		+= tlb-sh3.o
 obj-$(CONFIG_CPU_SH4)		+= tlb-sh4.o pg-sh4.o
diff --git a/arch/sh/mm/cache-debugfs.c b/arch/sh/mm/cache-debugfs.c
new file mode 100644
index 000000000000..a22d914e4d15
--- /dev/null
+++ b/arch/sh/mm/cache-debugfs.c
@@ -0,0 +1,147 @@
+/*
+ * debugfs ops for the L1 cache
+ *
+ *  Copyright (C) 2006  Paul Mundt
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/cache.h>
+#include <asm/io.h>
+
+enum cache_type {
+	CACHE_TYPE_ICACHE,
+	CACHE_TYPE_DCACHE,
+	CACHE_TYPE_UNIFIED,
+};
+
+static int cache_seq_show(struct seq_file *file, void *iter)
+{
+	unsigned int cache_type = (unsigned int)file->private;
+	struct cache_info *cache;
+	unsigned int waysize, way, cache_size;
+	unsigned long ccr, base;
+	static unsigned long addrstart = 0;
+
+	/*
+	 * Go uncached immediately so we don't skew the results any
+	 * more than we already are..
+	 */
+	jump_to_P2();
+
+	ccr = ctrl_inl(CCR);
+	if ((ccr & CCR_CACHE_ENABLE) == 0) {
+		back_to_P1();
+
+		seq_printf(file, "disabled\n");
+		return 0;
+	}
+
+	if (cache_type == CACHE_TYPE_DCACHE) {
+		base = CACHE_OC_ADDRESS_ARRAY;
+		cache = &cpu_data->dcache;
+	} else {
+		base = CACHE_IC_ADDRESS_ARRAY;
+		cache = &cpu_data->icache;
+	}
+
+	/*
+	 * Due to the amount of data written out (depending on the cache size),
+	 * we may be iterated over multiple times. In this case, keep track of
+	 * the entry position in addrstart, and rewind it when we've hit the
+	 * end of the cache.
+	 *
+	 * Likewise, the same code is used for multiple caches, so care must
+	 * be taken for bouncing addrstart back and forth so the appropriate
+	 * cache is hit.
+	 */
+	cache_size = cache->ways * cache->sets * cache->linesz;
+	if (((addrstart & 0xff000000) != base) ||
+	     (addrstart & 0x00ffffff) > cache_size)
+		addrstart = base;
+
+	waysize = cache->sets;
+
+	/*
+	 * If the OC is already in RAM mode, we only have
+	 * half of the entries to consider..
+	 */
+	if ((ccr & CCR_CACHE_ORA) && cache_type == CACHE_TYPE_DCACHE)
+		waysize >>= 1;
+
+	waysize <<= cache->entry_shift;
+
+	for (way = 0; way < cache->ways; way++) {
+		unsigned long addr;
+		unsigned int line;
+
+		seq_printf(file, "-----------------------------------------\n");
+		seq_printf(file, "Way %d\n", way);
+		seq_printf(file, "-----------------------------------------\n");
+
+		for (addr = addrstart, line = 0;
+		     addr < addrstart + waysize;
+		     addr += cache->linesz, line++) {
+			unsigned long data = ctrl_inl(addr);
+
+			/* Check the V bit, ignore invalid cachelines */
+			if ((data & 1) == 0)
+				continue;
+
+			/* U: Dirty, cache tag is 10 bits up */
+			seq_printf(file, "%3d: %c 0x%lx\n",
+				   line, data & 2 ? 'U' : ' ',
+				   data & 0x1ffffc00);
+		}
+
+		addrstart += cache->way_incr;
+	}
+
+	back_to_P1();
+
+	return 0;
+}
+
+static int cache_debugfs_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cache_seq_show, inode->u.generic_ip);
+}
+
+static struct file_operations cache_debugfs_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cache_debugfs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release,
+};
+
+static int __init cache_debugfs_init(void)
+{
+	struct dentry *dcache_dentry, *icache_dentry;
+
+	dcache_dentry = debugfs_create_file("dcache", S_IRUSR, NULL,
+					    (unsigned int *)CACHE_TYPE_DCACHE,
+					    &cache_debugfs_fops);
+	if (IS_ERR(dcache_dentry))
+		return PTR_ERR(dcache_dentry);
+
+	icache_dentry = debugfs_create_file("icache", S_IRUSR, NULL,
+					    (unsigned int *)CACHE_TYPE_ICACHE,
+					    &cache_debugfs_fops);
+	if (IS_ERR(icache_dentry)) {
+		debugfs_remove(dcache_dentry);
+		return PTR_ERR(icache_dentry);
+	}
+
+	return 0;
+}
+module_init(cache_debugfs_init);
+
+MODULE_LICENSE("GPL v2");
-- 
cgit v1.2.1


From 781125ca58dfbd47635cfc0e408f1f9d7e10b227 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:52:19 +0900
Subject: sh: New atomic ops for SH-4A movli.l/movco.l

SH-4A implements LL/SC instructions, so we implement a simple
set of atomic operations using these.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/atomic.h | 105 +++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 96 insertions(+), 9 deletions(-)

diff --git a/include/asm-sh/atomic.h b/include/asm-sh/atomic.h
index 049eb2dda6b6..8bdc1ba56f73 100644
--- a/include/asm-sh/atomic.h
+++ b/include/asm-sh/atomic.h
@@ -22,49 +22,110 @@ typedef struct { volatile int counter; } atomic_t;
  * forward to code at the end of this object's .text section, then
  * branch back to restart the operation.
  */
-
-static __inline__ void atomic_add(int i, atomic_t * v)
+static inline void atomic_add(int i, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_add	\n"
+"	add	%2, %0				\n"
+"	movco.l	%0, @%3				\n"
+"	bf	1b				\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v += i;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ void atomic_sub(int i, atomic_t *v)
+static inline void atomic_sub(int i, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_sub	\n"
+"	sub	%2, %0				\n"
+"	movco.l	%0, @%3				\n"
+"	bf	1b				\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v -= i;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ int atomic_add_return(int i, atomic_t * v)
+/*
+ * SH-4A note:
+ *
+ * We basically get atomic_xxx_return() for free compared with
+ * atomic_xxx(). movli.l/movco.l require r0 due to the instruction
+ * encoding, so the retval is automatically set without having to
+ * do any special work.
+ */
+static inline int atomic_add_return(int i, atomic_t *v)
 {
-	unsigned long temp, flags;
+	unsigned long temp;
+
+#ifdef CONFIG_CPU_SH4A
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_add_return	\n"
+"	add	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+"	synco						\n"
+	: "=&z" (temp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
+	unsigned long flags;
 
 	local_irq_save(flags);
 	temp = *(long *)v;
 	temp += i;
 	*(long *)v = temp;
 	local_irq_restore(flags);
+#endif
 
 	return temp;
 }
 
 #define atomic_add_negative(a, v)	(atomic_add_return((a), (v)) < 0)
 
-static __inline__ int atomic_sub_return(int i, atomic_t * v)
+static inline int atomic_sub_return(int i, atomic_t *v)
 {
-	unsigned long temp, flags;
+	unsigned long temp;
+
+#ifdef CONFIG_CPU_SH4A
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_sub_return	\n"
+"	sub	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+"	synco						\n"
+	: "=&z" (temp), "=r" (&v->counter)
+	: "r" (i), "r" (&v->counter)
+	: "t");
+#else
+	unsigned long flags;
 
 	local_irq_save(flags);
 	temp = *(long *)v;
 	temp -= i;
 	*(long *)v = temp;
 	local_irq_restore(flags);
+#endif
 
 	return temp;
 }
@@ -119,22 +180,48 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u)
 }
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
-static __inline__ void atomic_clear_mask(unsigned int mask, atomic_t *v)
+static inline void atomic_clear_mask(unsigned int mask, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_clear_mask	\n"
+"	and	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (~mask), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v &= ~mask;
 	local_irq_restore(flags);
+#endif
 }
 
-static __inline__ void atomic_set_mask(unsigned int mask, atomic_t *v)
+static inline void atomic_set_mask(unsigned int mask, atomic_t *v)
 {
+#ifdef CONFIG_CPU_SH4A
+	unsigned long tmp;
+
+	__asm__ __volatile__ (
+"1:	movli.l @%3, %0		! atomic_set_mask	\n"
+"	or	%2, %0					\n"
+"	movco.l	%0, @%3					\n"
+"	bf	1b					\n"
+	: "=&z" (tmp), "=r" (&v->counter)
+	: "r" (mask), "r" (&v->counter)
+	: "t");
+#else
 	unsigned long flags;
 
 	local_irq_save(flags);
 	*(long *)v |= mask;
 	local_irq_restore(flags);
+#endif
 }
 
 /* Atomic operations are already serializing on SH */
-- 
cgit v1.2.1


From f118420be83c3ce7888fe1d42db84af495da9edb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 17:53:59 +0900
Subject: watchdog: Add a simple mmap() stub for shwdt.

In some applications people have expressed a need for an mmap() method,
so we implement a simple stub for this that maps back a page with the
counter in it.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 drivers/char/watchdog/Kconfig |  8 +++++++
 drivers/char/watchdog/shwdt.c | 52 ++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 59 insertions(+), 1 deletion(-)

diff --git a/drivers/char/watchdog/Kconfig b/drivers/char/watchdog/Kconfig
index fff89c2d88fd..f114d7b5bb2a 100644
--- a/drivers/char/watchdog/Kconfig
+++ b/drivers/char/watchdog/Kconfig
@@ -510,6 +510,14 @@ config SH_WDT
 	  To compile this driver as a module, choose M here: the
 	  module will be called shwdt.
 
+config SH_WDT_MMAP
+	bool "Allow mmap of SH WDT"
+	default n
+	depends on SH_WDT
+	help
+	  If you say Y here, user applications will be able to mmap the
+	  WDT/CPG registers.
+#
 # SPARC64 Architecture
 
 config WATCHDOG_CP1XXX
diff --git a/drivers/char/watchdog/shwdt.c b/drivers/char/watchdog/shwdt.c
index 7c39f39effdf..e5b8c64f1d65 100644
--- a/drivers/char/watchdog/shwdt.c
+++ b/drivers/char/watchdog/shwdt.c
@@ -27,7 +27,7 @@
 #include <linux/notifier.h>
 #include <linux/ioport.h>
 #include <linux/fs.h>
-
+#include <linux/mm.h>
 #include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/watchdog.h>
@@ -257,6 +257,55 @@ static ssize_t sh_wdt_write(struct file *file, const char *buf,
 	return count;
 }
 
+/**
+ * 	sh_wdt_mmap - map WDT/CPG registers into userspace
+ * 	@file: file structure for the device
+ * 	@vma: VMA to map the registers into
+ *
+ * 	A simple mmap() implementation for the corner cases where the counter
+ * 	needs to be mapped in userspace directly. Due to the relatively small
+ * 	size of the area, neighbouring registers not necessarily tied to the
+ * 	CPG will also be accessible through the register page, so this remains
+ * 	configurable for users that really know what they're doing.
+ *
+ *	Additionaly, the register page maps in the CPG register base relative
+ *	to the nearest page-aligned boundary, which requires that userspace do
+ *	the appropriate CPU subtype math for calculating the page offset for
+ *	the counter value.
+ */
+static int sh_wdt_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	int ret = -ENOSYS;
+
+#ifdef CONFIG_SH_WDT_MMAP
+	unsigned long addr;
+
+	/* Only support the simple cases where we map in a register page. */
+	if (((vma->vm_end - vma->vm_start) != PAGE_SIZE) || vma->vm_pgoff)
+		return -EINVAL;
+
+	/*
+	 * Pick WTCNT as the start, it's usually the first register after the
+	 * FRQCR, and neither one are generally page-aligned out of the box.
+	 */
+	addr = WTCNT & ~(PAGE_SIZE - 1);
+
+	vma->vm_flags |= VM_IO;
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+
+	if (io_remap_pfn_range(vma, vma->vm_start, addr >> PAGE_SHIFT,
+			       PAGE_SIZE, vma->vm_page_prot)) {
+		printk(KERN_ERR PFX "%s: io_remap_pfn_range failed\n",
+		       __FUNCTION__);
+		return -EAGAIN;
+	}
+
+	ret = 0;
+#endif
+
+	return ret;
+}
+
 /**
  * 	sh_wdt_ioctl - Query Device
  * 	@inode: inode of device
@@ -342,6 +391,7 @@ static const struct file_operations sh_wdt_fops = {
 	.ioctl		= sh_wdt_ioctl,
 	.open		= sh_wdt_open,
 	.release	= sh_wdt_close,
+	.mmap		= sh_wdt_mmap,
 };
 
 static struct watchdog_info sh_wdt_info = {
-- 
cgit v1.2.1


From 5a4053b23262afefa748e1e4c439931d4c27693b Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:00:19 +0900
Subject: sh: Kill off dead boards.

None of these have been maintained in years, and no one seems to
be interested in doing so, so just get rid of them.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                      |  43 +--
 arch/sh/Makefile                     |   7 -
 arch/sh/boards/adx/Makefile          |   6 -
 arch/sh/boards/adx/irq.c             |  31 --
 arch/sh/boards/adx/irq_maskreg.c     | 106 -------
 arch/sh/boards/adx/setup.c           |  56 ----
 arch/sh/boards/cat68701/Makefile     |   6 -
 arch/sh/boards/cat68701/irq.c        |  28 --
 arch/sh/boards/cat68701/setup.c      |  85 -----
 arch/sh/boards/cqreek/Makefile       |   6 -
 arch/sh/boards/cqreek/irq.c          | 128 --------
 arch/sh/boards/cqreek/setup.c        | 100 ------
 arch/sh/boards/dmida/Makefile        |   7 -
 arch/sh/boards/dmida/mach.c          |  59 ----
 arch/sh/boards/harp/Makefile         |   8 -
 arch/sh/boards/harp/irq.c            | 147 ---------
 arch/sh/boards/harp/led.c            |  51 ---
 arch/sh/boards/harp/mach.c           |  62 ----
 arch/sh/boards/harp/pcidma.c         |  42 ---
 arch/sh/boards/harp/setup.c          |  90 ------
 arch/sh/boards/overdrive/Makefile    |   8 -
 arch/sh/boards/overdrive/fpga.c      | 133 --------
 arch/sh/boards/overdrive/galileo.c   | 587 -----------------------------------
 arch/sh/boards/overdrive/io.c        | 172 ----------
 arch/sh/boards/overdrive/irq.c       | 191 ------------
 arch/sh/boards/overdrive/led.c       |  58 ----
 arch/sh/boards/overdrive/mach.c      |  62 ----
 arch/sh/boards/overdrive/pcidma.c    |  46 ---
 arch/sh/boards/overdrive/setup.c     |  36 ---
 arch/sh/boards/sh2000/Makefile       |   6 -
 arch/sh/boards/sh2000/setup.c        |  70 -----
 arch/sh/tools/mach-types             |   2 -
 include/asm-sh/adx/io.h              |  86 -----
 include/asm-sh/cat68701/io.h         |  22 --
 include/asm-sh/cqreek/cqreek.h       |  27 --
 include/asm-sh/dmida/io.h            |  10 -
 include/asm-sh/harp/harp.h           |  43 ---
 include/asm-sh/harp/io.h             |  10 -
 include/asm-sh/overdrive/fpga.h      |  15 -
 include/asm-sh/overdrive/gt64111.h   | 109 -------
 include/asm-sh/overdrive/io.h        |  39 ---
 include/asm-sh/overdrive/overdrive.h |  88 ------
 include/asm-sh/sh2000/sh2000.h       |   8 -
 43 files changed, 4 insertions(+), 2892 deletions(-)
 delete mode 100644 arch/sh/boards/adx/Makefile
 delete mode 100644 arch/sh/boards/adx/irq.c
 delete mode 100644 arch/sh/boards/adx/irq_maskreg.c
 delete mode 100644 arch/sh/boards/adx/setup.c
 delete mode 100644 arch/sh/boards/cat68701/Makefile
 delete mode 100644 arch/sh/boards/cat68701/irq.c
 delete mode 100644 arch/sh/boards/cat68701/setup.c
 delete mode 100644 arch/sh/boards/cqreek/Makefile
 delete mode 100644 arch/sh/boards/cqreek/irq.c
 delete mode 100644 arch/sh/boards/cqreek/setup.c
 delete mode 100644 arch/sh/boards/dmida/Makefile
 delete mode 100644 arch/sh/boards/dmida/mach.c
 delete mode 100644 arch/sh/boards/harp/Makefile
 delete mode 100644 arch/sh/boards/harp/irq.c
 delete mode 100644 arch/sh/boards/harp/led.c
 delete mode 100644 arch/sh/boards/harp/mach.c
 delete mode 100644 arch/sh/boards/harp/pcidma.c
 delete mode 100644 arch/sh/boards/harp/setup.c
 delete mode 100644 arch/sh/boards/overdrive/Makefile
 delete mode 100644 arch/sh/boards/overdrive/fpga.c
 delete mode 100644 arch/sh/boards/overdrive/galileo.c
 delete mode 100644 arch/sh/boards/overdrive/io.c
 delete mode 100644 arch/sh/boards/overdrive/irq.c
 delete mode 100644 arch/sh/boards/overdrive/led.c
 delete mode 100644 arch/sh/boards/overdrive/mach.c
 delete mode 100644 arch/sh/boards/overdrive/pcidma.c
 delete mode 100644 arch/sh/boards/overdrive/setup.c
 delete mode 100644 arch/sh/boards/sh2000/Makefile
 delete mode 100644 arch/sh/boards/sh2000/setup.c
 delete mode 100644 include/asm-sh/adx/io.h
 delete mode 100644 include/asm-sh/cat68701/io.h
 delete mode 100644 include/asm-sh/cqreek/cqreek.h
 delete mode 100644 include/asm-sh/dmida/io.h
 delete mode 100644 include/asm-sh/harp/harp.h
 delete mode 100644 include/asm-sh/harp/io.h
 delete mode 100644 include/asm-sh/overdrive/fpga.h
 delete mode 100644 include/asm-sh/overdrive/gt64111.h
 delete mode 100644 include/asm-sh/overdrive/io.h
 delete mode 100644 include/asm-sh/overdrive/overdrive.h
 delete mode 100644 include/asm-sh/sh2000/sh2000.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index c7e3596dcde1..ddf2cc56868c 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -89,12 +89,6 @@ config SH_7751_SYSTEMH
 	  Select SystemH if you are configuring for a Renesas SystemH
 	  7751R evaluation board.
 
-config SH_STB1_HARP
-	bool "STB1_Harp"
-
-config SH_STB1_OVERDRIVE
-	bool "STB1_Overdrive"
-
 config SH_HP6XX
 	bool "HP6XX"
 	help
@@ -102,19 +96,6 @@ config SH_HP6XX
 	  More information (hardware only) at
 	  <http://www.hp.com/jornada/>.
 
-config SH_CQREEK
-	bool "CqREEK"
-	help
-	  Select CqREEK if configuring for a CqREEK SH7708 or SH7750.
-	  More information at
-	  <http://sources.redhat.com/ecos/hardware.html#SuperH>.
-
-config SH_DMIDA
-	bool "DMIDA"
-	help
-	  Select DMIDA if configuring for a DataMyte 4000 Industrial
-	  Digital Assistant. More information at <http://www.dmida.com/>.
-
 config SH_EC3104
 	bool "EC3104"
 	help
@@ -136,25 +117,9 @@ config SH_DREAMCAST
 	  <http://www.m17n.org/linux-sh/dreamcast/>.  There is a
 	  Dreamcast project is at <http://linuxdc.sourceforge.net/>.
 
-config SH_CAT68701
-	bool "CAT68701"
-
 config SH_BIGSUR
 	bool "BigSur"
 
-config SH_SH2000
-	bool "SH2000"
-	select CPU_SUBTYPE_SH7709
-	help
-	  SH-2000 is a single-board computer based around SH7709A chip
-	  intended for embedded applications.
-	  It has an Ethernet interface (CS8900A), direct connected
-	  Compact Flash socket, three serial ports and PC-104 bus.
-	  More information at <http://sh2000.sh-linux.org>.
-
-config SH_ADX
-	bool "ADX"
-
 config SH_MPC1211
 	bool "Interface MPC1211"
 	help
@@ -246,7 +211,7 @@ source "arch/sh/mm/Kconfig"
 
 config CF_ENABLER
 	bool "Compact Flash Enabler support"
-	depends on SH_ADX || SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_CAT68701 || SH_SH03
+	depends on SH_SOLUTION_ENGINE || SH_UNKNOWN || SH_SH03
 	---help---
 	  Compact Flash is a small, removable mass storage device introduced
 	  in 1994 originally as a PCMCIA device.  If you say `Y' here, you
@@ -274,7 +239,7 @@ config CF_AREA5
 	  - "Area5" if CompactFlash is connected to Area 5 (0x14000000)
 	  - "Area6" if it is connected to Area 6 (0x18000000)
 
-	  "Area6" will work for most boards. For ADX, select "Area5".
+	  "Area6" will work for most boards.
 
 config CF_AREA6
 	bool "Area6"
@@ -418,8 +383,8 @@ source "arch/sh/cchips/Kconfig"
 
 config HEARTBEAT
 	bool "Heartbeat LED"
-	depends on SH_MPC1211 || SH_SH03 || SH_CAT68701 || \
-		   SH_STB1_HARP || SH_STB1_OVERDRIVE || SH_BIGSUR || \
+	depends on SH_MPC1211 || SH_SH03 || \
+		   SH_BIGSUR || \
 		   SH_7751_SOLUTION_ENGINE || SH_7300_SOLUTION_ENGINE || \
 		   SH_73180_SOLUTION_ENGINE || SH_SOLUTION_ENGINE || \
 		   SH_RTS7751R2D || SH_SH4202_MICRODEV || SH_LANDISK
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 8ad0a286ff8f..d0168844c3c7 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -90,18 +90,11 @@ machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
 machdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se/7751
 machdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)	:= se/7300
 machdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se/73180
-machdir-$(CONFIG_SH_STB1_HARP)			:= harp
-machdir-$(CONFIG_SH_STB1_OVERDRIVE)		:= overdrive
 machdir-$(CONFIG_SH_HP6XX)			:= hp6xx
-machdir-$(CONFIG_SH_CQREEK)			:= cqreek
-machdir-$(CONFIG_SH_DMIDA)			:= dmida
 machdir-$(CONFIG_SH_EC3104)			:= ec3104
 machdir-$(CONFIG_SH_SATURN)			:= saturn
 machdir-$(CONFIG_SH_DREAMCAST)			:= dreamcast
-machdir-$(CONFIG_SH_CAT68701)			:= cat68701
 machdir-$(CONFIG_SH_BIGSUR)			:= bigsur
-machdir-$(CONFIG_SH_SH2000)			:= sh2000
-machdir-$(CONFIG_SH_ADX)			:= adx
 machdir-$(CONFIG_SH_MPC1211)			:= mpc1211
 machdir-$(CONFIG_SH_SH03)			:= sh03
 machdir-$(CONFIG_SH_SECUREEDGE5410)		:= snapgear
diff --git a/arch/sh/boards/adx/Makefile b/arch/sh/boards/adx/Makefile
deleted file mode 100644
index 5b1c531b3991..000000000000
--- a/arch/sh/boards/adx/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for ADX boards
-#
-
-obj-y	 := setup.o irq.o irq_maskreq.o
-
diff --git a/arch/sh/boards/adx/irq.c b/arch/sh/boards/adx/irq.c
deleted file mode 100644
index c6ca409dff98..000000000000
--- a/arch/sh/boards/adx/irq.c
+++ /dev/null
@@ -1,31 +0,0 @@
-/*
- * linux/arch/sh/boards/adx/irq.c
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * I/O routine and setup routines for A&D ADX Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/irq.h>
-
-void init_adx_IRQ(void)
-{
-        int i;
-
-/*      printk("init_adx_IRQ()\n");*/
-        /* setup irq_mask_register */
-        irq_mask_register = (unsigned short *)0xa6000008;
-
-        /* cover all external interrupt area by maskreg_irq_type
-         * (Actually, irq15 doesn't exist)
-         */
-        for (i = 0; i < 16; i++) {
-                make_maskreg_irq(i);
-                disable_irq(i);
-        }
-}
diff --git a/arch/sh/boards/adx/irq_maskreg.c b/arch/sh/boards/adx/irq_maskreg.c
deleted file mode 100644
index 4b2abe5eb165..000000000000
--- a/arch/sh/boards/adx/irq_maskreg.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * linux/arch/sh/kernel/irq_maskreg.c
- *
- * Copyright (C) 2001 A&D Co., Ltd. <http://www.aandd.co.jp>
- *
- * This file may be copied or modified under the terms of the GNU
- * General Public License.  See linux/COPYING for more information.
- *
- * Interrupt handling for Simple external interrupt mask register
- *
- * This is for the machine which have single 16 bit register
- * for masking external IRQ individually.
- * Each bit of the register is for masking each interrupt.  
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/machvec.h>
-
-/* address of external interrupt mask register
- * address must be set prior to use these (maybe in init_XXX_irq())
- * XXX : is it better to use .config than specifying it in code? */
-unsigned short *irq_mask_register = 0;
-
-/* forward declaration */
-static unsigned int startup_maskreg_irq(unsigned int irq);
-static void shutdown_maskreg_irq(unsigned int irq);
-static void enable_maskreg_irq(unsigned int irq);
-static void disable_maskreg_irq(unsigned int irq);
-static void mask_and_ack_maskreg(unsigned int);
-static void end_maskreg_irq(unsigned int irq);
-
-/* hw_interrupt_type */
-static struct hw_interrupt_type maskreg_irq_type = {
-	.typename = " Mask Register",
-	.startup = startup_maskreg_irq,
-	.shutdown = shutdown_maskreg_irq,
-	.enable = enable_maskreg_irq,
-	.disable = disable_maskreg_irq,
-	.ack = mask_and_ack_maskreg,
-	.end = end_maskreg_irq
-};
-
-/* actual implementatin */
-static unsigned int startup_maskreg_irq(unsigned int irq)
-{ 
-	enable_maskreg_irq(irq);
-	return 0; /* never anything pending */
-}
-
-static void shutdown_maskreg_irq(unsigned int irq)
-{
-	disable_maskreg_irq(irq);
-}
-
-static void disable_maskreg_irq(unsigned int irq)
-{
-	if (irq_mask_register) {
-		unsigned long flags;
-		unsigned short val, mask = 0x01 << irq;
-
-		/* Set "irq"th bit */
-		local_irq_save(flags);
-		val = ctrl_inw((unsigned long)irq_mask_register);
-		val |= mask;
-		ctrl_outw(val, (unsigned long)irq_mask_register);
-		local_irq_restore(flags);
-	}
-}
-
-static void enable_maskreg_irq(unsigned int irq)
-{
-	if (irq_mask_register) {
-		unsigned long flags;
-		unsigned short val, mask = ~(0x01 << irq);
-
-		/* Clear "irq"th bit */
-		local_irq_save(flags);
-		val = ctrl_inw((unsigned long)irq_mask_register);
-		val &= mask;
-		ctrl_outw(val, (unsigned long)irq_mask_register);
-		local_irq_restore(flags);
-	}
-}
-
-static void mask_and_ack_maskreg(unsigned int irq)
-{
-	disable_maskreg_irq(irq);
-}
-
-static void end_maskreg_irq(unsigned int irq)
-{
-	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-		enable_maskreg_irq(irq);
-}
-
-void make_maskreg_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &maskreg_irq_type;
-	disable_maskreg_irq(irq);
-}
diff --git a/arch/sh/boards/adx/setup.c b/arch/sh/boards/adx/setup.c
deleted file mode 100644
index 4938d9592343..000000000000
--- a/arch/sh/boards/adx/setup.c
+++ /dev/null
@@ -1,56 +0,0 @@
-/* 
- * linux/arch/sh/board/adx/setup.c
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * I/O routine and setup routines for A&D ADX Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/machvec.h>
-#include <linux/module.h>
-
-extern void init_adx_IRQ(void);
-extern void *cf_io_base;
-
-const char *get_system_type(void)
-{
-	return "A&D ADX";
-}
-
-unsigned long adx_isa_port2addr(unsigned long offset)
-{
-	/* CompactFlash (IDE) */
-	if (((offset >= 0x1f0) && (offset <= 0x1f7)) || (offset == 0x3f6)) {
-		return (unsigned long)cf_io_base + offset;
-	}
-
-	/* eth0 */
-	if ((offset >= 0x300) && (offset <= 0x30f)) {
-		return 0xa5000000 + offset;	/* COMM BOARD (AREA1) */
-	}
-
-	return offset + 0xb0000000; /* IOBUS (AREA 4)*/
-}
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_adx __initmv = {
-	.mv_nr_irqs		= 48,
-	.mv_isa_port2addr	= adx_isa_port2addr,
-	.mv_init_irq		= init_adx_IRQ,
-};
-ALIAS_MV(adx)
-
-int __init platform_setup(void)
-{
-	/* Nothing to see here .. */
-	return 0;
-}
-
diff --git a/arch/sh/boards/cat68701/Makefile b/arch/sh/boards/cat68701/Makefile
deleted file mode 100644
index 52c1de0a6dfd..000000000000
--- a/arch/sh/boards/cat68701/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the CAT-68701 specific parts of the kernel
-#
-
-obj-y	 := setup.o irq.o
-
diff --git a/arch/sh/boards/cat68701/irq.c b/arch/sh/boards/cat68701/irq.c
deleted file mode 100644
index f9a6d185fb8b..000000000000
--- a/arch/sh/boards/cat68701/irq.c
+++ /dev/null
@@ -1,28 +0,0 @@
-/*
- * linux/arch/sh/boards/cat68701/irq.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *               2001  Yutaro Ebihara
- *
- * Setup routines for A-ONE Corp CAT-68701 SH7708 Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/irq.h>
-
-int cat68701_irq_demux(int irq)
-{
-        if(irq==13) return 14;
-        if(irq==7)  return 10;
-        return irq;
-}
-
-void init_cat68701_IRQ()
-{
-        make_imask_irq(10);
-        make_imask_irq(14);
-}
diff --git a/arch/sh/boards/cat68701/setup.c b/arch/sh/boards/cat68701/setup.c
deleted file mode 100644
index 90e5175df227..000000000000
--- a/arch/sh/boards/cat68701/setup.c
+++ /dev/null
@@ -1,85 +0,0 @@
-/* 
- * linux/arch/sh/boards/cat68701/setup.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *               2001  Yutaro Ebihara
- *
- * Setup routines for A-ONE Corp CAT-68701 SH7708 Board
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- */
-
-#include <asm/io.h>
-#include <asm/machvec.h>
-#include <asm/mach/io.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/sched.h>
-
-const char *get_system_type(void)
-{
-	return "CAT-68701";
-}
-
-#ifdef CONFIG_HEARTBEAT
-void heartbeat_cat68701()
-{
-        static unsigned int cnt = 0, period = 0 , bit = 0;
-        cnt += 1;
-        if (cnt < period) {
-                return;
-        }
-        cnt = 0;
-
-        /* Go through the points (roughly!):
-         * f(0)=10, f(1)=16, f(2)=20, f(5)=35,f(inf)->110
-         */
-        period = 110 - ( (300<<FSHIFT)/
-                         ((avenrun[0]/5) + (3<<FSHIFT)) );
-
-	if(bit){ bit=0; }else{ bit=1; }
-	outw(bit<<15,0x3fe);
-}
-#endif /* CONFIG_HEARTBEAT */
-
-unsigned long cat68701_isa_port2addr(unsigned long offset)
-{
-	/* CompactFlash (IDE) */
-	if (((offset >= 0x1f0) && (offset <= 0x1f7)) || (offset==0x3f6))
-		return 0xba000000 + offset;
-
-	/* INPUT PORT */
-	if ((offset >= 0x3fc) && (offset <= 0x3fd))
-		return 0xb4007000 + offset;
-
-	/* OUTPUT PORT */
-	if ((offset >= 0x3fe) && (offset <= 0x3ff))
-		return 0xb4007400 + offset;
-
-	return offset + 0xb4000000; /* other I/O (EREA 5)*/
-}
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_cat68701 __initmv = {
-	.mv_nr_irqs		= 32,
-	.mv_isa_port2addr	= cat68701_isa_port2addr,
-	.mv_irq_demux		= cat68701_irq_demux,
-
-	.mv_init_irq		= init_cat68701_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_cat68701,
-#endif
-};
-ALIAS_MV(cat68701)
-
-int __init platform_setup(void)
-{
-	/* dummy read erea5 (CS8900A) */
-}
-
diff --git a/arch/sh/boards/cqreek/Makefile b/arch/sh/boards/cqreek/Makefile
deleted file mode 100644
index 1a788a85eba3..000000000000
--- a/arch/sh/boards/cqreek/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the CqREEK specific parts of the kernel
-#
-
-obj-y	 := setup.o irq.o
-
diff --git a/arch/sh/boards/cqreek/irq.c b/arch/sh/boards/cqreek/irq.c
deleted file mode 100644
index 2955adc52310..000000000000
--- a/arch/sh/boards/cqreek/irq.c
+++ /dev/null
@@ -1,128 +0,0 @@
-/* $Id: irq.c,v 1.1.2.4 2002/11/04 20:33:56 lethal Exp $
- *
- * arch/sh/boards/cqreek/irq.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *
- * CqREEK IDE/ISA Bridge Support.
- *
- */
-
-#include <linux/irq.h>
-#include <linux/init.h>
-
-#include <asm/cqreek/cqreek.h>
-#include <asm/io.h>
-#include <asm/io_generic.h>
-#include <asm/irq.h>
-#include <asm/machvec.h>
-#include <asm/machvec_init.h>
-#include <asm/rtc.h>
-
-struct cqreek_irq_data {
-	unsigned short mask_port;	/* Port of Interrupt Mask Register */
-	unsigned short stat_port;	/* Port of Interrupt Status Register */
-	unsigned short bit;		/* Value of the bit */
-};
-static struct cqreek_irq_data cqreek_irq_data[NR_IRQS];
-
-static void disable_cqreek_irq(unsigned int irq)
-{
-	unsigned long flags;
-	unsigned short mask;
-	unsigned short mask_port = cqreek_irq_data[irq].mask_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	local_irq_save(flags);
-	/* Disable IRQ */
-	mask = inw(mask_port) & ~bit;
-	outw_p(mask, mask_port);
-	local_irq_restore(flags);
-}
-
-static void enable_cqreek_irq(unsigned int irq)
-{
-	unsigned long flags;
-	unsigned short mask;
-	unsigned short mask_port = cqreek_irq_data[irq].mask_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	local_irq_save(flags);
-	/* Enable IRQ */
-	mask = inw(mask_port) | bit;
-	outw_p(mask, mask_port);
-	local_irq_restore(flags);
-}
-
-static void mask_and_ack_cqreek(unsigned int irq)
-{
-	unsigned short stat_port = cqreek_irq_data[irq].stat_port;
-	unsigned short bit = cqreek_irq_data[irq].bit;
-
-	disable_cqreek_irq(irq);
-	/* Clear IRQ (it might be edge IRQ) */
-	inw(stat_port);
-	outw_p(bit, stat_port);
-}
-
-static void end_cqreek_irq(unsigned int irq)
-{
-	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-		enable_cqreek_irq(irq);
-}
-
-static unsigned int startup_cqreek_irq(unsigned int irq)
-{ 
-	enable_cqreek_irq(irq);
-	return 0;
-}
-
-static void shutdown_cqreek_irq(unsigned int irq)
-{
-	disable_cqreek_irq(irq);
-}
-
-static struct hw_interrupt_type cqreek_irq_type = {
-	.typename = "CqREEK-IRQ",
-	.startup = startup_cqreek_irq,
-	.shutdown = shutdown_cqreek_irq,
-	.enable = enable_cqreek_irq,
-	.disable = disable_cqreek_irq,
-	.ack = mask_and_ack_cqreek,
-	.end = end_cqreek_irq
-};
-
-int cqreek_has_ide, cqreek_has_isa;
-
-/* XXX: This is just for test for my NE2000 ISA board
-   What we really need is virtualized IRQ and demultiplexer like HP600 port */
-void __init init_cqreek_IRQ(void)
-{
-	if (cqreek_has_ide) {
-		cqreek_irq_data[14].mask_port = BRIDGE_IDE_INTR_MASK;
-		cqreek_irq_data[14].stat_port = BRIDGE_IDE_INTR_STAT;
-		cqreek_irq_data[14].bit = 1;
-
-		irq_desc[14].chip = &cqreek_irq_type;
-		irq_desc[14].status = IRQ_DISABLED;
-		irq_desc[14].action = 0;
-		irq_desc[14].depth = 1;
-
-		disable_cqreek_irq(14);
-	}
-
-	if (cqreek_has_isa) {
-		cqreek_irq_data[10].mask_port = BRIDGE_ISA_INTR_MASK;
-		cqreek_irq_data[10].stat_port = BRIDGE_ISA_INTR_STAT;
-		cqreek_irq_data[10].bit = (1 << 10);
-
-		/* XXX: Err... we may need demultiplexer for ISA irq... */
-		irq_desc[10].chip = &cqreek_irq_type;
-		irq_desc[10].status = IRQ_DISABLED;
-		irq_desc[10].action = 0;
-		irq_desc[10].depth = 1;
-
-		disable_cqreek_irq(10);
-	}
-}
-
diff --git a/arch/sh/boards/cqreek/setup.c b/arch/sh/boards/cqreek/setup.c
deleted file mode 100644
index eff4ed93599f..000000000000
--- a/arch/sh/boards/cqreek/setup.c
+++ /dev/null
@@ -1,100 +0,0 @@
-/* $Id: setup.c,v 1.5 2003/08/04 01:51:58 lethal Exp $
- *
- * arch/sh/kernel/setup_cqreek.c
- *
- * Copyright (C) 2000  Niibe Yutaka
- *
- * CqREEK IDE/ISA Bridge Support.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/mach/cqreek.h>
-#include <asm/machvec.h>
-#include <asm/io.h>
-#include <asm/io_generic.h>
-#include <asm/irq.h>
-#include <asm/rtc.h>
-
-#define IDE_OFFSET 0xA4000000UL
-#define ISA_OFFSET 0xA4A00000UL
-
-const char *get_system_type(void)
-{
-	return "CqREEK";
-}
-
-static unsigned long cqreek_port2addr(unsigned long port)
-{
-	if (0x0000<=port && port<=0x0040)
-		return IDE_OFFSET + port;
-	if ((0x01f0<=port && port<=0x01f7) || port == 0x03f6)
-		return IDE_OFFSET + port;
-
-	return ISA_OFFSET + port;
-}
-
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_cqreek __initmv = {
-#if defined(CONFIG_CPU_SH4)
-	.mv_nr_irqs		= 48,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
-	.mv_nr_irqs		= 32,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
-	.mv_nr_irqs		= 61,
-#endif
-
-	.mv_init_irq		= init_cqreek_IRQ,
-
-	.mv_isa_port2addr	= cqreek_port2addr,
-};
-ALIAS_MV(cqreek)
-
-/*
- * Initialize the board
- */
-void __init platform_setup(void)
-{
-	int i;
-/* udelay is not available at setup time yet... */
-#define DELAY() do {for (i=0; i<10000; i++) ctrl_inw(0xa0000000);} while(0)
-
-	if ((inw (BRIDGE_FEATURE) & 1)) { /* We have IDE interface */
-		outw_p(0, BRIDGE_IDE_INTR_LVL);
-		outw_p(0, BRIDGE_IDE_INTR_MASK);
-
-		outw_p(0, BRIDGE_IDE_CTRL);
-		DELAY();
-
-		outw_p(0x8000, BRIDGE_IDE_CTRL);
-		DELAY();
-
-		outw_p(0xffff, BRIDGE_IDE_INTR_STAT); /* Clear interrupt status */
-		outw_p(0x0f-14, BRIDGE_IDE_INTR_LVL); /* Use 14 IPR */
-		outw_p(1, BRIDGE_IDE_INTR_MASK); /* Enable interrupt */
-		cqreek_has_ide=1;
-	}
-
-	if ((inw (BRIDGE_FEATURE) & 2)) { /* We have ISA interface */
-		outw_p(0, BRIDGE_ISA_INTR_LVL);
-		outw_p(0, BRIDGE_ISA_INTR_MASK);
-
-		outw_p(0, BRIDGE_ISA_CTRL);
-		DELAY();
-		outw_p(0x8000, BRIDGE_ISA_CTRL);
-		DELAY();
-
-		outw_p(0xffff, BRIDGE_ISA_INTR_STAT); /* Clear interrupt status */
-		outw_p(0x0f-10, BRIDGE_ISA_INTR_LVL); /* Use 10 IPR */
-		outw_p(0xfff8, BRIDGE_ISA_INTR_MASK); /* Enable interrupt */
-		cqreek_has_isa=1;
-	}
-
-	printk(KERN_INFO "CqREEK Setup (IDE=%d, ISA=%d)...done\n", cqreek_has_ide, cqreek_has_isa);
-}
-
diff --git a/arch/sh/boards/dmida/Makefile b/arch/sh/boards/dmida/Makefile
deleted file mode 100644
index 75999aa0a2d9..000000000000
--- a/arch/sh/boards/dmida/Makefile
+++ /dev/null
@@ -1,7 +0,0 @@
-#
-# Makefile for the DataMyte Industrial Digital Assistant(tm) specific parts
-# of the kernel
-#
-
-obj-y	 := mach.o
-
diff --git a/arch/sh/boards/dmida/mach.c b/arch/sh/boards/dmida/mach.c
deleted file mode 100644
index d03a25f989c2..000000000000
--- a/arch/sh/boards/dmida/mach.c
+++ /dev/null
@@ -1,59 +0,0 @@
-/*
- * linux/arch/sh/boards/dmida/mach.c
- *
- * by Greg Banks <gbanks@pocketpenguins.com>
- * (c) 2000 PocketPenguins Inc
- *
- * Derived from mach_hp600.c, which bore the message:
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the DataMyte Industrial Digital Assistant(tm).
- * See http://www.dmida.com
- *
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/io.h>
-#include <asm/hd64465/hd64465.h>
-#include <asm/irq.h>
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_dmida __initmv = {
-	.mv_nr_irqs		= HD64465_IRQ_BASE+HD64465_IRQ_NUM,
-
-	.mv_inb			= hd64465_inb,
-	.mv_inw			= hd64465_inw,
-	.mv_inl			= hd64465_inl,
-	.mv_outb		= hd64465_outb,
-	.mv_outw		= hd64465_outw,
-	.mv_outl		= hd64465_outl,
-
-	.mv_inb_p		= hd64465_inb_p,
-	.mv_inw_p		= hd64465_inw,
-	.mv_inl_p		= hd64465_inl,
-	.mv_outb_p		= hd64465_outb_p,
-	.mv_outw_p		= hd64465_outw,
-	.mv_outl_p		= hd64465_outl,
-
-	.mv_insb		= hd64465_insb,
-	.mv_insw		= hd64465_insw,
-	.mv_insl		= hd64465_insl,
-	.mv_outsb		= hd64465_outsb,
-	.mv_outsw		= hd64465_outsw,
-	.mv_outsl		= hd64465_outsl,
-
-	.mv_irq_demux		= hd64465_irq_demux,
-};
-ALIAS_MV(dmida)
-
diff --git a/arch/sh/boards/harp/Makefile b/arch/sh/boards/harp/Makefile
deleted file mode 100644
index eb753d31812e..000000000000
--- a/arch/sh/boards/harp/Makefile
+++ /dev/null
@@ -1,8 +0,0 @@
-#
-# Makefile for STMicroelectronics board specific parts of the kernel
-#
-
-obj-y := irq.o setup.o mach.o led.o
-
-obj-$(CONFIG_PCI) += pcidma.o
-
diff --git a/arch/sh/boards/harp/irq.c b/arch/sh/boards/harp/irq.c
deleted file mode 100644
index 96bb41c9fc55..000000000000
--- a/arch/sh/boards/harp/irq.c
+++ /dev/null
@@ -1,147 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Looks after interrupts on the HARP board.
- *
- * Bases on the IPR irq system
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-
-#define NUM_EXTERNAL_IRQS 16
-
-// Early versions of the STB1 Overdrive required this nasty frig
-//#define INVERT_INTMASK_WRITES
-
-static void enable_harp_irq(unsigned int irq);
-static void disable_harp_irq(unsigned int irq);
-
-/* shutdown is same as "disable" */
-#define shutdown_harp_irq disable_harp_irq
-
-static void mask_and_ack_harp(unsigned int);
-static void end_harp_irq(unsigned int irq);
-
-static unsigned int startup_harp_irq(unsigned int irq)
-{
-	enable_harp_irq(irq);
-	return 0;		/* never anything pending */
-}
-
-static struct hw_interrupt_type harp_irq_type = {
-	.typename = "Harp-IRQ",
-	.startup = startup_harp_irq,
-	.shutdown = shutdown_harp_irq,
-	.enable = enable_harp_irq,
-	.disable = disable_harp_irq,
-	.ack = mask_and_ack_harp,
-	.end = end_harp_irq
-};
-
-static void disable_harp_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	unsigned maskReg;
-	unsigned mask;
-	int pri;
-
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	pri = 15 - irq;
-
-	if (pri < 8) {
-		maskReg = EPLD_INTMASK0;
-	} else {
-		maskReg = EPLD_INTMASK1;
-		pri -= 8;
-	}
-
-	local_irq_save(flags);
-	mask = ctrl_inl(maskReg);
-	mask &= (~(1 << pri));
-#if defined(INVERT_INTMASK_WRITES)
-	mask ^= 0xff;
-#endif
-	ctrl_outl(mask, maskReg);
-	local_irq_restore(flags);
-}
-
-static void enable_harp_irq(unsigned int irq)
-{
-	unsigned flags;
-	unsigned maskReg;
-	unsigned mask;
-	int pri;
-
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	pri = 15 - irq;
-
-	if (pri < 8) {
-		maskReg = EPLD_INTMASK0;
-	} else {
-		maskReg = EPLD_INTMASK1;
-		pri -= 8;
-	}
-
-	local_irq_save(flags);
-	mask = ctrl_inl(maskReg);
-
-
-	mask |= (1 << pri);
-
-#if defined(INVERT_INTMASK_WRITES)
-	mask ^= 0xff;
-#endif
-	ctrl_outl(mask, maskReg);
-
-	local_irq_restore(flags);
-}
-
-/* This functions sets the desired irq handler to be an overdrive type */
-static void __init make_harp_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &harp_irq_type;
-	disable_harp_irq(irq);
-}
-
-static void mask_and_ack_harp(unsigned int irq)
-{
-	disable_harp_irq(irq);
-}
-
-static void end_harp_irq(unsigned int irq)
-{
-	enable_harp_irq(irq);
-}
-
-void __init init_harp_irq(void)
-{
-	int i;
-
-#if !defined(INVERT_INTMASK_WRITES)
-	// On the harp these are set to enable an interrupt
-	ctrl_outl(0x00, EPLD_INTMASK0);
-	ctrl_outl(0x00, EPLD_INTMASK1);
-#else
-	// On the Overdrive the data is inverted before being stored in the reg
-	ctrl_outl(0xff, EPLD_INTMASK0);
-	ctrl_outl(0xff, EPLD_INTMASK1);
-#endif
-
-	for (i = 0; i < NUM_EXTERNAL_IRQS; i++) {
-		make_harp_irq(i);
-	}
-}
diff --git a/arch/sh/boards/harp/led.c b/arch/sh/boards/harp/led.c
deleted file mode 100644
index aeb7b392b190..000000000000
--- a/arch/sh/boards/harp/led.c
+++ /dev/null
@@ -1,51 +0,0 @@
-/*
- * linux/arch/sh/stboards/led.c
- *
- * Copyright (C) 2000 Stuart Menefy <stuart.menefy@st.com>
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * This file contains ST40STB1 HARP and compatible code.
- */
-
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-/* Harp: Flash LD10 (front pannel) connected to EPLD (IC8) */
-/* Overdrive: Flash LD1 (front panel) connected to EPLD (IC4) */
-/* Works for HARP and overdrive */
-static void mach_led(int position, int value)
-{
-	if (value) {
-		ctrl_outl(EPLD_LED_ON, EPLD_LED);
-	} else {
-		ctrl_outl(EPLD_LED_OFF, EPLD_LED);
-	}
-}
-
-#ifdef CONFIG_HEARTBEAT
-
-#include <linux/sched.h>
-
-/* acts like an actual heart beat -- ie thump-thump-pause... */
-void heartbeat_harp(void)
-{
-	static unsigned cnt = 0, period = 0, dist = 0;
-
-	if (cnt == 0 || cnt == dist)
-		mach_led( -1, 1);
-	else if (cnt == 7 || cnt == dist+7)
-		mach_led( -1, 0);
-
-	if (++cnt > period) {
-		cnt = 0;
-		/* The hyperbolic function below modifies the heartbeat period
-		 * length in dependency of the current (5min) load. It goes
-		 * through the points f(0)=126, f(1)=86, f(5)=51,
-		 * f(inf)->30. */
-		period = ((672<<FSHIFT)/(5*avenrun[0]+(7<<FSHIFT))) + 30;
-		dist = period / 4;
-	}
-}
-#endif
diff --git a/arch/sh/boards/harp/mach.c b/arch/sh/boards/harp/mach.c
deleted file mode 100644
index a946dd1674ca..000000000000
--- a/arch/sh/boards/harp/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/boards/harp/mach.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the STMicroelectronics STB1 HARP and compatible boards
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-#include <asm/hd64465/io.h>
-#include <asm/hd64465/hd64465.h>
-
-void setup_harp(void);
-void init_harp_irq(void);
-void heartbeat_harp(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_harp __initmv = {
-	.mv_nr_irqs		= 89 + HD64465_IRQ_NUM,
-
-	.mv_inb			= hd64465_inb,
-	.mv_inw			= hd64465_inw,
-	.mv_inl			= hd64465_inl,
-	.mv_outb		= hd64465_outb,
-	.mv_outw		= hd64465_outw,
-	.mv_outl		= hd64465_outl,
-
-	.mv_inb_p		= hd64465_inb_p,
-	.mv_inw_p		= hd64465_inw,
-	.mv_inl_p		= hd64465_inl,
-	.mv_outb_p		= hd64465_outb_p,
-	.mv_outw_p		= hd64465_outw,
-	.mv_outl_p		= hd64465_outl,
-
-	.mv_insb		= hd64465_insb,
-	.mv_insw		= hd64465_insw,
-	.mv_insl		= hd64465_insl,
-	.mv_outsb		= hd64465_outsb,
-	.mv_outsw		= hd64465_outsw,
-	.mv_outsl		= hd64465_outsl,
-
-        .mv_isa_port2addr       = hd64465_isa_port2addr,
-
-#ifdef CONFIG_PCI
-	.mv_init_irq		= init_harp_irq,
-#endif
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_harp,
-#endif
-};
-
-ALIAS_MV(harp)
diff --git a/arch/sh/boards/harp/pcidma.c b/arch/sh/boards/harp/pcidma.c
deleted file mode 100644
index 475311390fd6..000000000000
--- a/arch/sh/boards/harp/pcidma.c
+++ /dev/null
@@ -1,42 +0,0 @@
-/* 
- * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Dynamic DMA mapping support.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-#include <asm/addrspace.h>
-
-
-void *pci_alloc_consistent(struct pci_dev *hwdev, size_t size,
-			   dma_addr_t * dma_handle)
-{
-	void *ret;
-	int gfp = GFP_ATOMIC;
-
-	ret = (void *) __get_free_pages(gfp, get_order(size));
-
-	if (ret != NULL) {
-	        /* Is it neccessary to do the memset? */
-		memset(ret, 0, size);
-		*dma_handle = virt_to_bus(ret);
-	}
-	/* We must flush the cache before we pass it on to the device */
-	flush_cache_all();
-	return  P2SEGADDR(ret);
-}
-
-void pci_free_consistent(struct pci_dev *hwdev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
-{
-        unsigned long p1addr=P1SEGADDR((unsigned long)vaddr);
-
-	free_pages(p1addr, get_order(size));
-}
diff --git a/arch/sh/boards/harp/setup.c b/arch/sh/boards/harp/setup.c
deleted file mode 100644
index 886e450ab63e..000000000000
--- a/arch/sh/boards/harp/setup.c
+++ /dev/null
@@ -1,90 +0,0 @@
-/*
- * arch/sh/stboard/setup.c
- *
- * Copyright (C) 2001 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * STMicroelectronics ST40STB1 HARP and compatible support.
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <asm/io.h>
-#include <asm/harp/harp.h>
-
-const char *get_system_type(void)
-{
-	return "STB1 Harp";
-}
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-#ifdef CONFIG_SH_STB1_HARP
-	unsigned long ic8_version, ic36_version;
-
-	ic8_version = ctrl_inl(EPLD_REVID2);
-	ic36_version = ctrl_inl(EPLD_REVID1);
-
-        printk("STMicroelectronics STB1 HARP initialisaton\n");
-        printk("EPLD versions: IC8: %d.%02d, IC36: %d.%02d\n",
-               (ic8_version >> 4) & 0xf, ic8_version & 0xf,
-               (ic36_version >> 4) & 0xf, ic36_version & 0xf);
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-	unsigned long version;
-
-	version = ctrl_inl(EPLD_REVID);
-
-        printk("STMicroelectronics STB1 Overdrive initialisaton\n");
-        printk("EPLD version: %d.%02d\n",
-	       (version >> 4) & 0xf, version & 0xf);
-#else
-#error Undefined machine
-#endif
- 
-        /* Currently all STB1 chips have problems with the sleep instruction,
-         * so disable it here.
-         */
-	disable_hlt();
-
-	return 0;
-}
-
-/*
- * pcibios_map_platform_irq
- *
- * This is board specific and returns the IRQ for a given PCI device.
- * It is used by the PCI code (arch/sh/kernel/st40_pci*)
- *
- */
-
-#define HARP_PCI_IRQ    1
-#define HARP_BRIDGE_IRQ 2
-#define OVERDRIVE_SLOT0_IRQ 0
-
-
-int __init pcibios_map_platform_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	switch (slot) {
-#ifdef CONFIG_SH_STB1_HARP
-	case 2:		/*This is the PCI slot on the */
-		return HARP_PCI_IRQ;
-	case 1:		/* this is the bridge */
-		return HARP_BRIDGE_IRQ;
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-	case 1:
-	case 2:
-	case 3:
-		return slot - 1;
-#else
-#error Unknown board
-#endif
-	default:
-		return -1;
-	}
-}
-
diff --git a/arch/sh/boards/overdrive/Makefile b/arch/sh/boards/overdrive/Makefile
deleted file mode 100644
index 245f03baf762..000000000000
--- a/arch/sh/boards/overdrive/Makefile
+++ /dev/null
@@ -1,8 +0,0 @@
-#
-# Makefile for the STMicroelectronics Overdrive specific parts of the kernel
-#
-
-obj-y	 := mach.o setup.o io.o irq.o led.o
-
-obj-$(CONFIG_PCI) += fpga.o galileo.o pcidma.o
-
diff --git a/arch/sh/boards/overdrive/fpga.c b/arch/sh/boards/overdrive/fpga.c
deleted file mode 100644
index 956c23901228..000000000000
--- a/arch/sh/boards/overdrive/fpga.c
+++ /dev/null
@@ -1,133 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file handles programming up the Altera Flex10K that interfaces to
- * the Galileo, and does the PS/2 keyboard and mouse
- *
- */
-
-
-#include <linux/kernel.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/init.h>
-#include <linux/errno.h>
-#include <linux/pci.h>
-#include <linux/delay.h>
-
-
-#include <asm/overdriver/gt64111.h>
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/fpga.h>
-
-#define FPGA_NotConfigHigh()  (*FPGA_ControlReg) = (*FPGA_ControlReg) | ENABLE_FPGA_BIT
-#define FPGA_NotConfigLow()   (*FPGA_ControlReg) = (*FPGA_ControlReg) & RESET_FPGA_MASK
-
-/* I need to find out what (if any) the real delay factor here is */
-/* The delay is definately not critical */
-#define long_delay() {int i;for(i=0;i<10000;i++);}
-#define short_delay() {int i;for(i=0;i<100;i++);}
-
-static void __init program_overdrive_fpga(const unsigned char *fpgacode,
-					  int size)
-{
-	int timeout = 0;
-	int i, j;
-	unsigned char b;
-	static volatile unsigned char *FPGA_ControlReg =
-	    (volatile unsigned char *) (OVERDRIVE_CTRL);
-	static volatile unsigned char *FPGA_ProgramReg =
-	    (volatile unsigned char *) (FPGA_DCLK_ADDRESS);
-
-	printk("FPGA:  Commencing FPGA Programming\n");
-
-	/* The PCI reset but MUST be low when programming the FPGA !!! */
-	b = (*FPGA_ControlReg) & RESET_PCI_MASK;
-
-	(*FPGA_ControlReg) = b;
-
-	/* Prepare FPGA to program */
-
-	FPGA_NotConfigHigh();
-	long_delay();
-
-	FPGA_NotConfigLow();
-	short_delay();
-
-	while ((*FPGA_ProgramReg & FPGA_NOT_STATUS) != 0) {
-		printk("FPGA:  Waiting for NotStatus to go Low ... \n");
-	}
-
-	FPGA_NotConfigHigh();
-
-	/* Wait for FPGA "ready to be programmed" signal */
-	printk("FPGA:  Waiting for NotStatus to go high (FPGA ready)... \n");
-
-	for (timeout = 0;
-	     (((*FPGA_ProgramReg & FPGA_NOT_STATUS) == 0)
-	      && (timeout < FPGA_TIMEOUT)); timeout++);
-
-	/* Check if timeout condition occured - i.e. an error */
-
-	if (timeout == FPGA_TIMEOUT) {
-		printk
-		    ("FPGA:  Failed to program - Timeout waiting for notSTATUS to go high\n");
-		return;
-	}
-
-	printk("FPGA:  Copying data to FPGA ... %d bytes\n", size);
-
-	/* Copy array to FPGA - bit at a time */
-
-	for (i = 0; i < size; i++) {
-		volatile unsigned w = 0;
-
-		for (j = 0; j < 8; j++) {
-			*FPGA_ProgramReg = (fpgacode[i] >> j) & 0x01;
-			short_delay();
-		}
-		if ((i & 0x3ff) == 0) {
-			printk(".");
-		}
-	}
-
-	/* Waiting for CONFDONE to go high - means the program is complete  */
-
-	for (timeout = 0;
-	     (((*FPGA_ProgramReg & FPGA_CONFDONE) == 0)
-	      && (timeout < FPGA_TIMEOUT)); timeout++) {
-
-		*FPGA_ProgramReg = 0x0;
-		long_delay();
-	}
-
-	if (timeout == FPGA_TIMEOUT) {
-		printk
-		    ("FPGA:  Failed to program - Timeout waiting for CONFDONE to go high\n");
-		return;
-	} else {		/* Clock another 10 times - gets the device into a working state      */
-		for (i = 0; i < 10; i++) {
-			*FPGA_ProgramReg = 0x0;
-			short_delay();
-		}
-	}
-
-	printk("FPGA:  Programming complete\n");
-}
-
-
-static const unsigned char __init fpgacode[] = {
-#include "./overdrive.ttf"	/* Code from maxplus2 compiler */
-	, 0, 0
-};
-
-
-int __init init_overdrive_fpga(void)
-{
-	program_overdrive_fpga(fpgacode, sizeof(fpgacode));
-
-	return 0;
-}
diff --git a/arch/sh/boards/overdrive/galileo.c b/arch/sh/boards/overdrive/galileo.c
deleted file mode 100644
index 29e48971bba0..000000000000
--- a/arch/sh/boards/overdrive/galileo.c
+++ /dev/null
@@ -1,587 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file contains the PCI routines required for the Galileo GT6411 
- * PCI bridge as used on the Orion and Overdrive boards.
- *
- */
-
-#include <linux/kernel.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/init.h>
-#include <linux/errno.h>
-#include <linux/pci.h>
-#include <linux/delay.h>
-#include <linux/types.h>
-#include <linux/ioport.h>
-
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/gt64111.h>
-
-
-/* After boot, we shift the Galileo registers so that they appear 
- * in BANK6, along with IO space. This means we can have one contingous
- * lump of PCI address space without these registers appearing in the 
- * middle of them 
- */
-
-#define GT64111_BASE_ADDRESS  0xbb000000
-#define GT64111_IO_BASE_ADDRESS  0x1000
-/* The GT64111 registers appear at this address to the SH4 after reset */
-#define RESET_GT64111_BASE_ADDRESS           0xb4000000
-
-/* Macros used to access the Galileo registers */
-#define RESET_GT64111_REG(x) (RESET_GT64111_BASE_ADDRESS+x)
-#define GT64111_REG(x) (GT64111_BASE_ADDRESS+x)
-
-#define RESET_GT_WRITE(x,v) writel((v),RESET_GT64111_REG(x))
-
-#define RESET_GT_READ(x) readl(RESET_GT64111_REG(x))
-
-#define GT_WRITE(x,v) writel((v),GT64111_REG(x))
-#define GT_WRITE_BYTE(x,v) writeb((v),GT64111_REG(x))
-#define GT_WRITE_SHORT(x,v) writew((v),GT64111_REG(x))
-
-#define GT_READ(x)    readl(GT64111_REG(x))
-#define GT_READ_BYTE(x)  readb(GT64111_REG(x))
-#define GT_READ_SHORT(x) readw(GT64111_REG(x))
-
-
-/* Where the various SH banks start at */
-#define SH_BANK4_ADR 0xb0000000
-#define SH_BANK5_ADR 0xb4000000
-#define SH_BANK6_ADR 0xb8000000
-
-/* Masks out everything but lines 28,27,26 */
-#define BANK_SELECT_MASK 0x1c000000
-
-#define SH4_TO_BANK(x) ( (x) & BANK_SELECT_MASK)
-
-/* 
- * Masks used for address conversaion. Bank 6 is used for IO and 
- * has all the address bits zeroed by the FPGA. Special case this
- */
-#define MEMORY_BANK_MASK 0x1fffffff
-#define IO_BANK_MASK  0x03ffffff
-
-/* Mark bank 6 as the bank used for IO. You can change this in the FPGA code
- * if you want 
- */
-#define IO_BANK_ADR PCI_GTIO_BASE
-
-/* Will select the correct mask to apply depending on the SH$ address */
-#define SELECT_BANK_MASK(x) \
-   ( (SH4_TO_BANK(x)==SH4_TO_BANK(IO_BANK_ADR)) ? IO_BANK_MASK : MEMORY_BANK_MASK)
-
-/* Converts between PCI space and P2 region */
-#define SH4_TO_PCI(x) ((x)&SELECT_BANK_MASK(x))
-
-/* Various macros for figuring out what to stick in the Galileo registers. 
- * You *really* don't want to figure this stuff out by hand, you always get
- * it wrong
- */
-#define GT_MEM_LO_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>21)&0x7ff)
-#define GT_MEM_HI_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>21)&0x7f)
-#define GT_MEM_SUB_ADR(x) ((((unsigned)((x)&SELECT_BANK_MASK(x)))>>20)&0xff)
-
-#define PROGRAM_HI_LO(block,a,s) \
-    GT_WRITE(block##_LO_DEC_ADR,GT_MEM_LO_ADR(a));\
-    GT_WRITE(block##_HI_DEC_ADR,GT_MEM_HI_ADR(a+s-1))
-
-#define PROGRAM_SUB_HI_LO(block,a,s) \
-    GT_WRITE(block##_LO_DEC_ADR,GT_MEM_SUB_ADR(a));\
-    GT_WRITE(block##_HI_DEC_ADR,GT_MEM_SUB_ADR(a+s-1))
-
-/* We need to set the size, and the offset register */
-
-#define GT_BAR_MASK(x) ((x)&~0xfff)
-
-/* Macro to set up the BAR in the Galileo. Essentially used for the DRAM */
-#define PROGRAM_GT_BAR(block,a,s) \
-  GT_WRITE(PCI_##block##_BANK_SIZE,GT_BAR_MASK((s-1)));\
-  write_config_to_galileo(PCI_CONFIG_##block##_BASE_ADR,\
-			     GT_BAR_MASK(a))
-
-#define DISABLE_GT_BAR(block) \
-  GT_WRITE(PCI_##block##_BANK_SIZE,0),\
-  GT_CONFIG_WRITE(PCI_CONFIG_##block##_BASE_ADR,\
-    0x80000000)
-
-/* Macros to disable things we are not going to use */
-#define DISABLE_DECODE(x) GT_WRITE(x##_LO_DEC_ADR,0x7ff);\
-                          GT_WRITE(x##_HI_DEC_ADR,0x00)
-
-#define DISABLE_SUB_DECODE(x) GT_WRITE(x##_LO_DEC_ADR,0xff);\
-                              GT_WRITE(x##_HI_DEC_ADR,0x00)
-
-static void __init reset_pci(void)
-{
-	/* Set RESET_PCI bit high */
-	writeb(readb(OVERDRIVE_CTRL) | ENABLE_PCI_BIT, OVERDRIVE_CTRL);
-	udelay(250);
-
-	/* Set RESET_PCI bit low */
-	writeb(readb(OVERDRIVE_CTRL) & RESET_PCI_MASK, OVERDRIVE_CTRL);
-	udelay(250);
-
-	writeb(readb(OVERDRIVE_CTRL) | ENABLE_PCI_BIT, OVERDRIVE_CTRL);
-	udelay(250);
-}
-
-static int write_config_to_galileo(int where, u32 val);
-#define GT_CONFIG_WRITE(where,val) write_config_to_galileo(where,val)
-
-#define ENABLE_PCI_DRAM
-
-
-#ifdef TEST_DRAM
-/* Test function to check out if the PCI DRAM is working OK */
-static int  /* __init */ test_dram(unsigned *base, unsigned size)
-{
-	unsigned *p = base;
-	unsigned *end = (unsigned *) (((unsigned) base) + size);
-	unsigned w;
-
-	for (p = base; p < end; p++) {
-		*p = 0xffffffff;
-		if (*p != 0xffffffff) {
-			printk("AAARGH -write failed!!! at %p is %x\n", p,
-			       *p);
-			return 0;
-		}
-		*p = 0x0;
-		if (*p != 0x0) {
-			printk("AAARGH -write failed!!!\n");
-			return 0;
-		}
-	}
-
-	for (p = base; p < end; p++) {
-		*p = (unsigned) p;
-		if (*p != (unsigned) p) {
-			printk("Failed at 0x%p, actually is 0x%x\n", p,
-			       *p);
-			return 0;
-		}
-	}
-
-	for (p = base; p < end; p++) {
-		w = ((unsigned) p & 0xffff0000);
-		*p = w | (w >> 16);
-	}
-
-	for (p = base; p < end; p++) {
-		w = ((unsigned) p & 0xffff0000);
-		w |= (w >> 16);
-		if (*p != w) {
-			printk
-			    ("Failed at 0x%p, should be 0x%x actually is 0x%x\n",
-			     p, w, *p);
-			return 0;
-		}
-	}
-
-	return 1;
-}
-#endif
-
-
-/* Function to set up and initialise the galileo. This sets up the BARS,
- * maps the DRAM into the address space etc,etc
- */
-int __init galileo_init(void)
-{
-	reset_pci();
-
-	/* Now shift the galileo regs into this block */
-	RESET_GT_WRITE(INTERNAL_SPACE_DEC,
-		       GT_MEM_LO_ADR(GT64111_BASE_ADDRESS));
-
-	/* Should have a sanity check here, that you can read back  at the new
-	 * address what you just wrote 
-	 */
-
-	/* Disable decode for all regions */
-	DISABLE_DECODE(RAS10);
-	DISABLE_DECODE(RAS32);
-	DISABLE_DECODE(CS20);
-	DISABLE_DECODE(CS3);
-	DISABLE_DECODE(PCI_IO);
-	DISABLE_DECODE(PCI_MEM0);
-	DISABLE_DECODE(PCI_MEM1);
-
-	/* Disable all BARS */
-	GT_WRITE(BAR_ENABLE_ADR, 0x1ff);
-	DISABLE_GT_BAR(RAS10);
-	DISABLE_GT_BAR(RAS32);
-	DISABLE_GT_BAR(CS20);
-	DISABLE_GT_BAR(CS3);
-
-	/* Tell the BAR where the IO registers now are */
-	GT_CONFIG_WRITE(PCI_CONFIG_INT_REG_IO_ADR,GT_BAR_MASK(
-					    (GT64111_IO_BASE_ADDRESS &
-					     IO_BANK_MASK)));
-	/* set up a 112 Mb decode */
-	PROGRAM_HI_LO(PCI_MEM0, SH_BANK4_ADR, 112 * 1024 * 1024);
-
-	/* Set up a 32 MB io space decode */
-	PROGRAM_HI_LO(PCI_IO, IO_BANK_ADR, 32 * 1024 * 1024);
-
-#ifdef ENABLE_PCI_DRAM
-	/* Program up the DRAM configuration - there is DRAM only in bank 0 */
-	/* Now set up the DRAM decode */
-	PROGRAM_HI_LO(RAS10, PCI_DRAM_BASE, PCI_DRAM_SIZE);
-	/* And the sub decode */
-	PROGRAM_SUB_HI_LO(RAS0, PCI_DRAM_BASE, PCI_DRAM_SIZE);
-
-	DISABLE_SUB_DECODE(RAS1);
-
-	/* Set refresh rate */
-	GT_WRITE(DRAM_BANK0_PARMS, 0x3f);
-	GT_WRITE(DRAM_CFG, 0x100);
-
-	/* we have to lob off the top bits rememeber!! */
-	PROGRAM_GT_BAR(RAS10, SH4_TO_PCI(PCI_DRAM_BASE), PCI_DRAM_SIZE);
-
-#endif
-
-	/* We are only interested in decoding RAS10 and the Galileo's internal 
-	 * registers (as IO) on the PCI bus
-	 */
-#ifdef ENABLE_PCI_DRAM
-	GT_WRITE(BAR_ENABLE_ADR, (~((1 << 8) | (1 << 3))) & 0x1ff);
-#else
-	GT_WRITE(BAR_ENABLE_ADR, (~(1 << 3)) & 0x1ff);
-#endif
-
-	/* Change the class code to host bridge, it actually powers up 
-	 * as a memory controller
-         */
-	GT_CONFIG_WRITE(8, 0x06000011);
-
-	/* Allow the galileo to master the PCI bus */
-	GT_CONFIG_WRITE(PCI_COMMAND,
-			PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER |
-			PCI_COMMAND_IO);
-
-
-#if 0
-        printk("Testing PCI DRAM - ");
-	if(test_dram(PCI_DRAM_BASE,PCI_DRAM_SIZE)) {
-		printk("Passed\n");
-	}else {
-		printk("FAILED\n");
-	}
-#endif
-	return 0;
-
-}
-
-
-#define SET_CONFIG_BITS(bus,devfn,where)\
-  ((1<<31) | ((bus) << 16) | ((devfn) << 8) | ((where) & ~3))
-
-#define CONFIG_CMD(dev, where) SET_CONFIG_BITS((dev)->bus->number,(dev)->devfn,where)
-
-/* This write to the galileo config registers, unlike the functions below, can
- * be used before the PCI subsystem has started up
- */
-static int __init write_config_to_galileo(int where, u32 val)
-{
-	GT_WRITE(PCI_CFG_ADR, SET_CONFIG_BITS(0, 0, where));
-
-	GT_WRITE(PCI_CFG_DATA, val);
-	return 0;
-}
-
-/* We exclude the galileo and slot 31, the galileo because I don't know how to stop
- * the setup code shagging up the setup I have done on it, and 31 because the whole
- * thing locks up if you try to access that slot (which doesn't exist of course anyway
- */
-
-#define EXCLUDED_DEV(dev) ((dev->bus->number==0) && ((PCI_SLOT(dev->devfn)==0) || (PCI_SLOT(dev->devfn) == 31)))
-
-static int galileo_read_config_byte(struct pci_dev *dev, int where,
-				    u8 * val)
-{
-
-        
-	/* I suspect this doesn't work because this drives a special cycle ? */
-	if (EXCLUDED_DEV(dev)) {
-		*val = 0xff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-	/* Start the config cycle */
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	/* Read back the result */
-	*val = GT_READ_BYTE(PCI_CFG_DATA + (where & 3));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_read_config_word(struct pci_dev *dev, int where,
-				    u16 * val)
-{
-
-        if (EXCLUDED_DEV(dev)) {
-		*val = 0xffff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	*val = GT_READ_SHORT(PCI_CFG_DATA + (where & 2));
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_read_config_dword(struct pci_dev *dev, int where,
-				     u32 * val)
-{
-	if (EXCLUDED_DEV(dev)) {
-		*val = 0xffffffff;
-		return PCIBIOS_SUCCESSFUL;
-	}
-
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-	*val = GT_READ(PCI_CFG_DATA);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int galileo_write_config_byte(struct pci_dev *dev, int where,
-				     u8 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE_BYTE(PCI_CFG_DATA + (where & 3), val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-
-static int galileo_write_config_word(struct pci_dev *dev, int where,
-				     u16 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE_SHORT(PCI_CFG_DATA + (where & 2), val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int galileo_write_config_dword(struct pci_dev *dev, int where,
-				      u32 val)
-{
-	GT_WRITE(PCI_CFG_ADR, CONFIG_CMD(dev, where));
-
-	GT_WRITE(PCI_CFG_DATA, val);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops pci_config_ops = {
-	galileo_read_config_byte,
-	galileo_read_config_word,
-	galileo_read_config_dword,
-	galileo_write_config_byte,
-	galileo_write_config_word,
-	galileo_write_config_dword
-};
-
-
-/* Everything hangs off this */
-static struct pci_bus *pci_root_bus;
-
-
-static u8 __init no_swizzle(struct pci_dev *dev, u8 * pin)
-{
-	return PCI_SLOT(dev->devfn);
-}
-
-static int __init map_od_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	/* Slot 1: Galileo 
-	 * Slot 2: PCI Slot 1
-	 * Slot 3: PCI Slot 2
-	 * Slot 4: ESS
-	 */
-	switch (slot) {
-	case 2: 
-		return OVERDRIVE_PCI_IRQ1;
-	case 3:
-		/* Note this assumes you have a hacked card in slot 2 */
-		return OVERDRIVE_PCI_IRQ2;
-	case 4:
-		return OVERDRIVE_ESS_IRQ;
-	default:
-		/* printk("PCI: Unexpected IRQ mapping request for slot %d\n", slot); */
-		return -1;
-	}
-}
-
-
-
-void __init
-pcibios_fixup_pbus_ranges(struct pci_bus *bus, struct pbus_set_ranges_data *ranges)
-{
-        ranges->io_start -= bus->resource[0]->start;
-        ranges->io_end -= bus->resource[0]->start;
-        ranges->mem_start -= bus->resource[1]->start;
-        ranges->mem_end -= bus->resource[1]->start;
-}                                                                                
-
-static void __init pci_fixup_ide_bases(struct pci_dev *d)
-{
-	int i;
-
-	/*
-	 * PCI IDE controllers use non-standard I/O port decoding, respect it.
-	 */
-	if ((d->class >> 8) != PCI_CLASS_STORAGE_IDE)
-		return;
-	printk("PCI: IDE base address fixup for %s\n", pci_name(d));
-	for(i=0; i<4; i++) {
-		struct resource *r = &d->resource[i];
-		if ((r->start & ~0x80) == 0x374) {
-			r->start |= 2;
-			r->end = r->start;
-		}
-	}
-}
-DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, pci_fixup_ide_bases);
-
-void __init pcibios_init(void)
-{
-	static struct resource galio,galmem;
-
-        /* Allocate the registers used by the Galileo */
-        galio.flags = IORESOURCE_IO;
-        galio.name  = "Galileo GT64011";
-        galmem.flags = IORESOURCE_MEM|IORESOURCE_PREFETCH;
-        galmem.name  = "Galileo GT64011 DRAM";
-
-        allocate_resource(&ioport_resource, &galio, 256,
-		    GT64111_IO_BASE_ADDRESS,GT64111_IO_BASE_ADDRESS+256, 256, NULL, NULL);
-        allocate_resource(&iomem_resource, &galmem,PCI_DRAM_SIZE,
-		    PHYSADDR(PCI_DRAM_BASE), PHYSADDR(PCI_DRAM_BASE)+PCI_DRAM_SIZE, 
-			     PCI_DRAM_SIZE, NULL, NULL);
-
-  	/* ok, do the scan man */
-	pci_root_bus = pci_scan_bus(0, &pci_config_ops, NULL);
-
-        pci_assign_unassigned_resources();
-	pci_fixup_irqs(no_swizzle, map_od_irq);
-
-#ifdef TEST_DRAM
-        printk("Testing PCI DRAM - ");
-	if(test_dram(PCI_DRAM_BASE,PCI_DRAM_SIZE)) {
-		printk("Passed\n");
-	}else {
-		printk("FAILED\n");
-	}
-#endif
-
-}
-
-char * __init pcibios_setup(char *str)
-{
-	return str;
-}
-
-
-
-int pcibios_enable_device(struct pci_dev *dev)
-{
-
-	u16 cmd, old_cmd;
-	int idx;
-	struct resource *r;
-
-	pci_read_config_word(dev, PCI_COMMAND, &cmd);
-	old_cmd = cmd;
-	for (idx = 0; idx < 6; idx++) {
-		r = dev->resource + idx;
-		if (!r->start && r->end) {
-			printk(KERN_ERR
-			       "PCI: Device %s not available because"
-			       " of resource collisions\n",
-			       pci_name(dev));
-			return -EINVAL;
-		}
-		if (r->flags & IORESOURCE_IO)
-			cmd |= PCI_COMMAND_IO;
-		if (r->flags & IORESOURCE_MEM)
-			cmd |= PCI_COMMAND_MEMORY;
-	}
-	if (cmd != old_cmd) {
-		printk("PCI: enabling device %s (%04x -> %04x)\n",
-		       pci_name(dev), old_cmd, cmd);
-		pci_write_config_word(dev, PCI_COMMAND, cmd);
-	}
-	return 0;
-
-}
-
-/* We should do some optimisation work here I think. Ok for now though */
-void __init pcibios_fixup_bus(struct pci_bus *bus)
-{
-
-}
-
-void pcibios_align_resource(void *data, struct resource *res,
-			    resource_size_t size)
-{
-}
-
-void __init pcibios_update_resource(struct pci_dev *dev, struct resource *root,
-			     struct resource *res, int resource)
-{
-
-	unsigned long where, size;
-	u32 reg;
-	
-
-	printk("PCI: Assigning %3s %08lx to %s\n",
-	       res->flags & IORESOURCE_IO ? "IO" : "MEM",
-	       res->start, dev->name);
-
-	where = PCI_BASE_ADDRESS_0 + resource * 4;
-	size = res->end - res->start;
-
-	pci_read_config_dword(dev, where, &reg);
-	reg = (reg & size) | (((u32) (res->start - root->start)) & ~size);
-	pci_write_config_dword(dev, where, reg);
-}
-
-
-void __init pcibios_update_irq(struct pci_dev *dev, int irq)
-{
-	printk("PCI: Assigning IRQ %02d to %s\n", irq, dev->name);
-	pci_write_config_byte(dev, PCI_INTERRUPT_LINE, irq);
-}
-
-/*
- *  If we set up a device for bus mastering, we need to check the latency
- *  timer as certain crappy BIOSes forget to set it properly.
- */
-unsigned int pcibios_max_latency = 255;
-
-void pcibios_set_master(struct pci_dev *dev)
-{
-	u8 lat;
-	pci_read_config_byte(dev, PCI_LATENCY_TIMER, &lat);
-	if (lat < 16)
-		lat = (64 <= pcibios_max_latency) ? 64 : pcibios_max_latency;
-	else if (lat > pcibios_max_latency)
-		lat = pcibios_max_latency;
-	else
-		return;
-	printk("PCI: Setting latency timer of device %s to %d\n", pci_name(dev), lat);
-	pci_write_config_byte(dev, PCI_LATENCY_TIMER, lat);
-}
diff --git a/arch/sh/boards/overdrive/io.c b/arch/sh/boards/overdrive/io.c
deleted file mode 100644
index 4671b6b047bb..000000000000
--- a/arch/sh/boards/overdrive/io.c
+++ /dev/null
@@ -1,172 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * This file contains the I/O routines for use on the overdrive board
- *
- */
-
-#include <linux/types.h>
-#include <linux/delay.h>
-#include <asm/processor.h>
-#include <asm/io.h>
-#include <asm/addrspace.h>
-
-#include <asm/overdrive/overdrive.h>
-
-/*
- * readX/writeX() are used to access memory mapped devices. On some
- * architectures the memory mapped IO stuff needs to be accessed
- * differently. On the SuperH architecture, we just read/write the
- * memory location directly.
- */
-
-#define dprintk(x...)
-
-/* Translates an IO address to where it is mapped in memory */
-
-#define io_addr(x) (((unsigned)(x))|PCI_GTIO_BASE)
-
-unsigned char od_inb(unsigned long port)
-{
-dprintk("od_inb(%x)\n", port);
-	return readb(io_addr(port)) & 0xff;
-}
-
-
-unsigned short od_inw(unsigned long port)
-{
-dprintk("od_inw(%x)\n", port);
-	return readw(io_addr(port)) & 0xffff;
-}
-
-unsigned int od_inl(unsigned long port)
-{
-dprintk("od_inl(%x)\n", port);
-	return readl(io_addr(port));
-}
-
-void od_outb(unsigned char value, unsigned long port)
-{
-dprintk("od_outb(%x, %x)\n", value, port);
-	writeb(value, io_addr(port));
-}
-
-void od_outw(unsigned short value, unsigned long port)
-{
-dprintk("od_outw(%x, %x)\n", value, port);
-	writew(value, io_addr(port));
-}
-
-void od_outl(unsigned int value, unsigned long port)
-{
-dprintk("od_outl(%x, %x)\n", value, port);
-	writel(value, io_addr(port));
-}
-
-/* This is horrible at the moment - needs more work to do something sensible */
-#define IO_DELAY() udelay(10)
-
-#define OUT_DELAY(x,type) \
-void od_out##x##_p(unsigned type value,unsigned long port){out##x(value,port);IO_DELAY();}
-
-#define IN_DELAY(x,type) \
-unsigned type od_in##x##_p(unsigned long port) {unsigned type tmp=in##x(port);IO_DELAY();return tmp;}
-
-
-OUT_DELAY(b,char)
-OUT_DELAY(w,short)
-OUT_DELAY(l,int)
-
-IN_DELAY(b,char)
-IN_DELAY(w,short)
-IN_DELAY(l,int)
-
-
-/*  Now for the string version of these functions */
-void od_outsb(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p++) {
-		outb(*p, port);
-	}
-}
-
-
-void od_insb(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p++) {
-		*p = inb(port);
-	}
-}
-
-/* For the 16 and 32 bit string functions, we have to worry about alignment.
- * The SH does not do unaligned accesses, so we have to read as bytes and
- * then write as a word or dword. 
- * This can be optimised a lot more, especially in the case where the data
- * is aligned
- */
-
-void od_outsw(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned short tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 2) {
-		tmp = (*p) | ((*(p + 1)) << 8);
-		outw(tmp, port);
-	}
-}
-
-
-void od_insw(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned short tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 2) {
-		tmp = inw(port);
-		p[0] = tmp & 0xff;
-		p[1] = (tmp >> 8) & 0xff;
-	}
-}
-
-
-void od_outsl(unsigned long port, const void *addr, unsigned long count)
-{
-	int i;
-	unsigned tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 4) {
-		tmp = (*p) | ((*(p + 1)) << 8) | ((*(p + 2)) << 16) |
-		      ((*(p + 3)) << 24);
-		outl(tmp, port);
-	}
-}
-
-
-void od_insl(unsigned long port, void *addr, unsigned long count)
-{
-	int i;
-	unsigned tmp;
-	unsigned char *p = (unsigned char *) addr;
-
-	for (i = 0; i < count; i++, p += 4) {
-		tmp = inl(port);
-		p[0] = tmp & 0xff;
-		p[1] = (tmp >> 8) & 0xff;
-		p[2] = (tmp >> 16) & 0xff;
-		p[3] = (tmp >> 24) & 0xff;
-
-	}
-}
diff --git a/arch/sh/boards/overdrive/irq.c b/arch/sh/boards/overdrive/irq.c
deleted file mode 100644
index 5d730c70389e..000000000000
--- a/arch/sh/boards/overdrive/irq.c
+++ /dev/null
@@ -1,191 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Looks after interrupts on the overdrive board.
- *
- * Bases on the IPR irq system
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/system.h>
-#include <asm/io.h>
-
-#include <asm/overdrive/overdrive.h>
-
-struct od_data {
-	int overdrive_irq;
-	int irq_mask;
-};
-
-#define NUM_EXTERNAL_IRQS 16
-#define EXTERNAL_IRQ_NOT_IN_USE (-1)
-#define EXTERNAL_IRQ_NOT_ASSIGNED (-1)
-
-/*
- * This table is used to determine what to program into the FPGA's CT register
- * for the specified Linux IRQ.
- *
- * The irq_mask gives the interrupt number from the PCI board (PCI_Int(6:0))
- * but is one greater than that because the because the FPGA treats 0
- * as disabled, a value of 1 asserts PCI_Int0, and so on.
- *
- * The overdrive_irq specifies which of the eight interrupt sources generates
- * that interrupt, and but is multiplied by four to give the bit offset into
- * the CT register.
- *
- * The seven interrupts levels (SH4 IRL's) we have available here is hardwired
- * by the EPLD. The assignments here of which PCI interrupt generates each
- * level is arbitary.
- */
-static struct od_data od_data_table[NUM_EXTERNAL_IRQS] = {
-	/*    overdrive_irq       , irq_mask */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 0 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 7},	/* 1 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 6},	/* 2 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 3 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 5},	/* 4 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 5 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 6 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 4},	/* 7 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 8 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 9 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 3},	/* 10 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 2},	/* 11 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 12 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, 1},	/* 13 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE},	/* 14 */
-	{EXTERNAL_IRQ_NOT_ASSIGNED, EXTERNAL_IRQ_NOT_IN_USE}	/* 15 */
-};
-
-static void set_od_data(int overdrive_irq, int irq)
-{
-	if (irq >= NUM_EXTERNAL_IRQS || irq < 0)
-		return;
-	od_data_table[irq].overdrive_irq = overdrive_irq << 2;
-}
-
-static void enable_od_irq(unsigned int irq);
-void disable_od_irq(unsigned int irq);
-
-/* shutdown is same as "disable" */
-#define shutdown_od_irq disable_od_irq
-
-static void mask_and_ack_od(unsigned int);
-static void end_od_irq(unsigned int irq);
-
-static unsigned int startup_od_irq(unsigned int irq)
-{
-	enable_od_irq(irq);
-	return 0;		/* never anything pending */
-}
-
-static struct hw_interrupt_type od_irq_type = {
-	.typename = "Overdrive-IRQ",
-	.startup = startup_od_irq,
-	.shutdown = shutdown_od_irq,
-	.enable = enable_od_irq,
-	.disable = disable_od_irq,
-	.ack = mask_and_ack_od,
-	.end = end_od_irq
-};
-
-static void disable_od_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	int overdrive_irq;
-	unsigned mask;
-
-	/* Not a valid interrupt */
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-        /* Is is necessary to use a cli here? Would a spinlock not be 
-         * mroe efficient?
-         */
-	local_irq_save(flags);
-	overdrive_irq = od_data_table[irq].overdrive_irq;
-	if (overdrive_irq != EXTERNAL_IRQ_NOT_ASSIGNED) {
-		mask = ~(0x7 << overdrive_irq);
-		val = ctrl_inl(OVERDRIVE_INT_CT);
-		val &= mask;
-		ctrl_outl(val, OVERDRIVE_INT_CT);
-	}
-	local_irq_restore(flags);
-}
-
-static void enable_od_irq(unsigned int irq)
-{
-	unsigned val, flags;
-	int overdrive_irq;
-	unsigned mask;
-
-	/* Not a valid interrupt */
-	if (irq < 0 || irq >= NUM_EXTERNAL_IRQS)
-		return;
-
-	/* Set priority in OD back to original value */
-	local_irq_save(flags);
-	/* This one is not in use currently */
-	overdrive_irq = od_data_table[irq].overdrive_irq;
-	if (overdrive_irq != EXTERNAL_IRQ_NOT_ASSIGNED) {
-		val = ctrl_inl(OVERDRIVE_INT_CT);
-		mask = ~(0x7 << overdrive_irq);
-		val &= mask;
-		mask = od_data_table[irq].irq_mask << overdrive_irq;
-		val |= mask;
-		ctrl_outl(val, OVERDRIVE_INT_CT);
-	}
-	local_irq_restore(flags);
-}
-
-
-
-/* this functions sets the desired irq handler to be an overdrive type */
-static void __init make_od_irq(unsigned int irq)
-{
-	disable_irq_nosync(irq);
-	irq_desc[irq].chip = &od_irq_type;
-	disable_od_irq(irq);
-}
-
-
-static void mask_and_ack_od(unsigned int irq)
-{
-	disable_od_irq(irq);
-}
-
-static void end_od_irq(unsigned int irq)
-{
-	enable_od_irq(irq);
-}
-
-void __init init_overdrive_irq(void)
-{
-	int i;
-
-	/* Disable all interrupts */
-	ctrl_outl(0, OVERDRIVE_INT_CT);
-
-	/* Update interrupt pin mode to use encoded interrupts */
-	i = ctrl_inw(INTC_ICR);
-	i &= ~INTC_ICR_IRLM;
-	ctrl_outw(i, INTC_ICR);
-
-	for (i = 0; i < NUM_EXTERNAL_IRQS; i++) {
-		if (od_data_table[i].irq_mask != EXTERNAL_IRQ_NOT_IN_USE) {
-			make_od_irq(i);
-		} else if (i != 15) {	// Cannot use imask on level 15
-			make_imask_irq(i);
-		}
-	}
-
-	/* Set up the interrupts */
-	set_od_data(OVERDRIVE_PCI_INTA, OVERDRIVE_PCI_IRQ1);
-	set_od_data(OVERDRIVE_PCI_INTB, OVERDRIVE_PCI_IRQ2);
-	set_od_data(OVERDRIVE_AUDIO_INT, OVERDRIVE_ESS_IRQ);
-}
diff --git a/arch/sh/boards/overdrive/led.c b/arch/sh/boards/overdrive/led.c
deleted file mode 100644
index 860d7f204a4e..000000000000
--- a/arch/sh/boards/overdrive/led.c
+++ /dev/null
@@ -1,58 +0,0 @@
-/*
- * linux/arch/sh/overdrive/led.c
- *
- * Copyright (C) 1999 Stuart Menefy <stuart.menefy@st.com>
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * This file contains an Overdrive specific LED feature.
- */
-
-#include <asm/system.h>
-#include <asm/io.h>
-#include <asm/overdrive/overdrive.h>
-
-static void mach_led(int position, int value)
-{
-	unsigned long flags;
-	unsigned long reg;
-
-	local_irq_save(flags);
-	
-	reg = readl(OVERDRIVE_CTRL);
-	if (value) {
-		reg |= (1<<3);
-	} else {
-		reg &= ~(1<<3);
-	}
-	writel(reg, OVERDRIVE_CTRL);
-
-	local_irq_restore(flags);
-}
-
-#ifdef CONFIG_HEARTBEAT
-
-#include <linux/sched.h>
-
-/* acts like an actual heart beat -- ie thump-thump-pause... */
-void heartbeat_od(void)
-{
-	static unsigned cnt = 0, period = 0, dist = 0;
-
-	if (cnt == 0 || cnt == dist)
-		mach_led( -1, 1);
-	else if (cnt == 7 || cnt == dist+7)
-		mach_led( -1, 0);
-
-	if (++cnt > period) {
-		cnt = 0;
-		/* The hyperbolic function below modifies the heartbeat period
-		 * length in dependency of the current (5min) load. It goes
-		 * through the points f(0)=126, f(1)=86, f(5)=51,
-		 * f(inf)->30. */
-		period = ((672<<FSHIFT)/(5*avenrun[0]+(7<<FSHIFT))) + 30;
-		dist = period / 4;
-	}
-}
-#endif /* CONFIG_HEARTBEAT */
diff --git a/arch/sh/boards/overdrive/mach.c b/arch/sh/boards/overdrive/mach.c
deleted file mode 100644
index 2834a03ae477..000000000000
--- a/arch/sh/boards/overdrive/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/overdrive/mach.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the STMicroelectronics Overdrive
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/machvec_init.h>
-
-#include <asm/io_unknown.h>
-#include <asm/io_generic.h>
-#include <asm/overdrive/io.h>
-
-void heartbeat_od(void);
-void init_overdrive_irq(void);
-void galileo_pcibios_init(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_od __initmv = {
-	.mv_nr_irqs		= 48,
-
-	.mv_inb			= od_inb,
-	.mv_inw			= od_inw,
-	.mv_inl			= od_inl,
-	.mv_outb		= od_outb,
-	.mv_outw		= od_outw,
-	.mv_outl		= od_outl,
-
-	.mv_inb_p		= od_inb_p,
-	.mv_inw_p		= od_inw_p,
-	.mv_inl_p		= od_inl_p,
-	.mv_outb_p		= od_outb_p,
-	.mv_outw_p		= od_outw_p,
-	.mv_outl_p		= od_outl_p,
-
-	.mv_insb		= od_insb,
-	.mv_insw		= od_insw,
-	.mv_insl		= od_insl,
-	.mv_outsb		= od_outsb,
-	.mv_outsw		= od_outsw,
-	.mv_outsl		= od_outsl,
-
-#ifdef CONFIG_PCI
-	.mv_init_irq		= init_overdrive_irq,
-#endif
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_od,
-#endif
-};
-
-ALIAS_MV(od)
diff --git a/arch/sh/boards/overdrive/pcidma.c b/arch/sh/boards/overdrive/pcidma.c
deleted file mode 100644
index 1c9bfeda00b7..000000000000
--- a/arch/sh/boards/overdrive/pcidma.c
+++ /dev/null
@@ -1,46 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Dynamic DMA mapping support.
- *
- * On the overdrive, we can only DMA from memory behind the PCI bus!
- * this means that all DMA'able memory must come from there. 
- * this restriction will not apply to later boards.
- */
-
-#include <linux/types.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/pci.h>
-#include <asm/io.h>
-
-void *pci_alloc_consistent(struct pci_dev *hwdev, size_t size,
-			   dma_addr_t * dma_handle)
-{
-	void *ret;
-	int gfp = GFP_ATOMIC;
-
-        printk("BUG: pci_alloc_consistent() called - not yet supported\n");
-	/* We ALWAYS need DMA memory on the overdrive hardware,
-	 * due to it's extreme weirdness
-	 * Need to flush the cache here as well, since the memory
-	 * can still be seen through the cache!
-	 */
-	gfp |= GFP_DMA;
-	ret = (void *) __get_free_pages(gfp, get_order(size));
-
-	if (ret != NULL) {
-		memset(ret, 0, size);
-		*dma_handle = virt_to_bus(ret);
-	}
-	return ret;
-}
-
-void pci_free_consistent(struct pci_dev *hwdev, size_t size,
-			 void *vaddr, dma_addr_t dma_handle)
-{
-	free_pages((unsigned long) vaddr, get_order(size));
-}
diff --git a/arch/sh/boards/overdrive/setup.c b/arch/sh/boards/overdrive/setup.c
deleted file mode 100644
index a3a7744c2047..000000000000
--- a/arch/sh/boards/overdrive/setup.c
+++ /dev/null
@@ -1,36 +0,0 @@
-/*
- * arch/sh/overdrive/setup.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * STMicroelectronics Overdrive Support.
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <asm/io.h>
-
-#include <asm/overdrive/overdrive.h>
-#include <asm/overdrive/fpga.h>
-
-const char *get_system_type(void)
-{
-	return "SH7750 Overdrive";
-}
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-#ifdef CONFIG_PCI
-	init_overdrive_fpga();
-	galileo_init();
-#endif
-
-        /* Enable RS232 receive buffers */
-	writel(0x1e, OVERDRIVE_CTRL);
-}
diff --git a/arch/sh/boards/sh2000/Makefile b/arch/sh/boards/sh2000/Makefile
deleted file mode 100644
index 05d390c3599c..000000000000
--- a/arch/sh/boards/sh2000/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# Makefile for the SH2000 specific parts of the kernel
-#
-
-obj-y	 := setup.o
-
diff --git a/arch/sh/boards/sh2000/setup.c b/arch/sh/boards/sh2000/setup.c
deleted file mode 100644
index 2fe6a11765e9..000000000000
--- a/arch/sh/boards/sh2000/setup.c
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * linux/arch/sh/kernel/setup_sh2000.c
- *
- * Copyright (C) 2001  SUGIOKA Tochinobu
- *
- * SH-2000 Support.
- *
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/io.h>
-#include <asm/machvec.h>
-#include <asm/mach/sh2000.h>
-
-#define CF_CIS_BASE	0xb4200000
-
-#define PORT_PECR	0xa4000108
-#define PORT_PHCR	0xa400010E
-#define	PORT_ICR1	0xa4000010
-#define	PORT_IRR0	0xa4000004
-
-#define IDE_OFFSET	0xb6200000
-#define NIC_OFFSET	0xb6000000
-#define EXTBUS_OFFSET	0xba000000
-
-
-const char *get_system_type(void)
-{
-	return "sh2000";
-}
-
-static unsigned long sh2000_isa_port2addr(unsigned long offset)
-{
-	if((offset & ~7) == 0x1f0 || offset == 0x3f6)
-		return IDE_OFFSET + offset;
-	else if((offset & ~0x1f) == 0x300)
-		return NIC_OFFSET + offset;
-	return EXTBUS_OFFSET + offset;
-}
-
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_sh2000 __initmv = {
-        .mv_nr_irqs		= 80,
-        .mv_isa_port2addr	= sh2000_isa_port2addr,
-};
-ALIAS_MV(sh2000)
-
-/*
- * Initialize the board
- */
-int __init platform_setup(void)
-{
-	/* XXX: RTC setting comes here */
-
-	/* These should be done by BIOS/IPL ... */
-	/* Enable nCE2A, nCE2B output */
-	ctrl_outw(ctrl_inw(PORT_PECR) & ~0xf00, PORT_PECR);
-	/* Enable the Compact Flash card, and set the level interrupt */
-	ctrl_outw(0x0042, CF_CIS_BASE+0x0200);
-	/* Enable interrupt */
-	ctrl_outw(ctrl_inw(PORT_PHCR) & ~0x03f3, PORT_PHCR);
-	ctrl_outw(1, PORT_ICR1);
-	ctrl_outw(ctrl_inw(PORT_IRR0) & ~0xff3f, PORT_IRR0);
-	printk(KERN_INFO "SH-2000 Setup...done\n");
-	return 0;
-}
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index ffd218083876..b30e784b3a31 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -13,11 +13,9 @@ SE			SH_SOLUTION_ENGINE
 HP6XX			SH_HP6XX
 HD64461			HD64461
 HD64465			HD64465
-SH2000			SH_SH2000
 SATURN			SH_SATURN
 DREAMCAST		SH_DREAMCAST
 BIGSUR			SH_BIGSUR
-ADX			SH_ADX
 MPC1211			SH_MPC1211
 SNAPGEAR		SH_SECUREEDGE5410
 HS7751RVOIP		SH_HS7751RVOIP
diff --git a/include/asm-sh/adx/io.h b/include/asm-sh/adx/io.h
deleted file mode 100644
index ab1225f1d557..000000000000
--- a/include/asm-sh/adx/io.h
+++ /dev/null
@@ -1,86 +0,0 @@
-/*
- * include/asm-sh/io_adx.h
- *
- * Copyright (C) 2001 A&D Co., Ltd.
- *
- * This file may be copied or modified under the terms of the GNU
- * General Public License.  See linux/COPYING for more information.
- *
- * IO functions for an A&D ADX Board
- */
-
-#ifndef _ASM_SH_IO_ADX_H
-#define _ASM_SH_IO_ADX_H
-
-#include <asm/io_generic.h>
-
-extern unsigned char adx_inb(unsigned long port);
-extern unsigned short adx_inw(unsigned long port);
-extern unsigned int adx_inl(unsigned long port);
-
-extern void adx_outb(unsigned char value, unsigned long port);
-extern void adx_outw(unsigned short value, unsigned long port);
-extern void adx_outl(unsigned int value, unsigned long port);
-
-extern unsigned char adx_inb_p(unsigned long port);
-extern void adx_outb_p(unsigned char value, unsigned long port);
-
-extern void adx_insb(unsigned long port, void *addr, unsigned long count);
-extern void adx_insw(unsigned long port, void *addr, unsigned long count);
-extern void adx_insl(unsigned long port, void *addr, unsigned long count);
-extern void adx_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void adx_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void adx_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned char adx_readb(unsigned long addr);
-extern unsigned short adx_readw(unsigned long addr);
-extern unsigned int adx_readl(unsigned long addr);
-extern void adx_writeb(unsigned char b, unsigned long addr);
-extern void adx_writew(unsigned short b, unsigned long addr);
-extern void adx_writel(unsigned int b, unsigned long addr);
-
-extern void * adx_ioremap(unsigned long offset, unsigned long size);
-extern void adx_iounmap(void *addr);
-
-extern unsigned long adx_isa_port2addr(unsigned long offset);
-
-extern void setup_adx(void);
-extern void init_adx_IRQ(void);
-
-#ifdef __WANT_IO_DEF
-
-#define __inb		adx_inb
-#define __inw		adx_inw
-#define __inl		adx_inl
-#define __outb		adx_outb
-#define __outw		adx_outw
-#define __outl		adx_outl
-
-#define __inb_p		adx_inb_p
-#define __inw_p		adx_inw
-#define __inl_p		adx_inl
-#define __outb_p	adx_outb_p
-#define __outw_p	adx_outw
-#define __outl_p	adx_outl
-
-#define __insb		adx_insb
-#define __insw		adx_insw
-#define __insl		adx_insl
-#define __outsb		adx_outsb
-#define __outsw		adx_outsw
-#define __outsl		adx_outsl
-
-#define __readb		adx_readb
-#define __readw		adx_readw
-#define __readl		adx_readl
-#define __writeb	adx_writeb
-#define __writew	adx_writew
-#define __writel	adx_writel
-
-#define __isa_port2addr	adx_isa_port2addr
-#define __ioremap	adx_ioremap
-#define __iounmap	adx_iounmap
-
-#endif
-
-#endif /* _ASM_SH_IO_AANDD_H */
diff --git a/include/asm-sh/cat68701/io.h b/include/asm-sh/cat68701/io.h
deleted file mode 100644
index 753b8466ad11..000000000000
--- a/include/asm-sh/cat68701/io.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * include/asm-sh/io_cat68701.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *           2001 Yutarou Ebihar (ebihara@si-linux.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an AONE Corp. CAT-68701 SH7708 Borad
- */
-
-#ifndef _ASM_SH_IO_CAT68701_H
-#define _ASM_SH_IO_CAT68701_H
-
-extern unsigned long cat68701_isa_port2addr(unsigned long offset);
-extern int cat68701_irq_demux(int irq);
-
-extern void init_cat68701_IRQ(void);
-extern void heartbeat_cat68701(void);
-
-#endif /* _ASM_SH_IO_CAT68701_H */
diff --git a/include/asm-sh/cqreek/cqreek.h b/include/asm-sh/cqreek/cqreek.h
deleted file mode 100644
index 09aecc06693e..000000000000
--- a/include/asm-sh/cqreek/cqreek.h
+++ /dev/null
@@ -1,27 +0,0 @@
-#ifndef __ASM_SH_CQREEK_CQREEK_H
-#define __ASM_SH_CQREEK_CQREEK_H
-
-#define BRIDGE_FEATURE		0x0002
-
-#define BRIDGE_IDE_CTRL		0x0018
-#define BRIDGE_IDE_INTR_LVL    	0x001A
-#define BRIDGE_IDE_INTR_MASK	0x001C
-#define BRIDGE_IDE_INTR_STAT	0x001E
-
-#define BRIDGE_ISA_CTRL		0x0028
-#define BRIDGE_ISA_INTR_LVL    	0x002A
-#define BRIDGE_ISA_INTR_MASK	0x002C
-#define BRIDGE_ISA_INTR_STAT	0x002E
-
-/* arch/sh/boards/cqreek/setup.c */
-extern void setup_cqreek(void);
-
-/* arch/sh/boards/cqreek/irq.c */
-extern int cqreek_has_ide, cqreek_has_isa;
-extern void init_cqreek_IRQ(void);
-
-/* arch/sh/boards/cqreek/io.c */
-extern unsigned long cqreek_port2addr(unsigned long port);
-
-#endif /* __ASM_SH_CQREEK_CQREEK_H */
-
diff --git a/include/asm-sh/dmida/io.h b/include/asm-sh/dmida/io.h
deleted file mode 100644
index 21bd416c01c3..000000000000
--- a/include/asm-sh/dmida/io.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef __ASM_SH_DMIDA_IO_H
-#define __ASM_SH_DMIDA_IO_H
-
-/*
- * Nothing special here.. just use the generic cchip io routines.
- */
-#include <asm/hd64465/io.h>
-
-#endif /* __ASM_SH_DMIDA_IO_H */
-
diff --git a/include/asm-sh/harp/harp.h b/include/asm-sh/harp/harp.h
deleted file mode 100644
index b2fbcfae9940..000000000000
--- a/include/asm-sh/harp/harp.h
+++ /dev/null
@@ -1,43 +0,0 @@
-/* 
- * Copyright (C) 2001 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- * Defintions applicable to the STMicroelectronics ST40STB1 HARP and
- * compatible boards.
- */
-
-#if defined(CONFIG_SH_STB1_HARP)
-
-#define EPLD_BASE     0xa0800000
-
-#define EPLD_LED      (EPLD_BASE+0x000c0000)
-#define EPLD_INTSTAT0 (EPLD_BASE+0x00200000)
-#define EPLD_INTSTAT1 (EPLD_BASE+0x00240000)
-#define EPLD_INTMASK0 (EPLD_BASE+0x00280000)
-#define EPLD_INTMASK1 (EPLD_BASE+0x002c0000)
-#define EPLD_PAGEADDR (EPLD_BASE+0x00300000)
-#define EPLD_REVID1   (EPLD_BASE+0x00380000)
-#define EPLD_REVID2   (EPLD_BASE+0x003c0000)
-
-#define EPLD_LED_ON  1
-#define EPLD_LED_OFF 0
-
-#elif defined(CONFIG_SH_STB1_OVERDRIVE)
-
-#define EPLD_BASE     0xa7000000
-
-#define EPLD_REVID    (EPLD_BASE+0x00000000)
-#define EPLD_LED      (EPLD_BASE+0x00040000)
-#define EPLD_INTMASK0 (EPLD_BASE+0x001c0000)
-#define EPLD_INTMASK1 (EPLD_BASE+0x00200000)
-#define EPLD_INTSTAT0 (EPLD_BASE+0x00240000)
-#define EPLD_INTSTAT1 (EPLD_BASE+0x00280000)
-
-#define EPLD_LED_ON  0
-#define EPLD_LED_OFF 1
-
-#else
-#error Unknown board
-#endif
diff --git a/include/asm-sh/harp/io.h b/include/asm-sh/harp/io.h
deleted file mode 100644
index 68f39e0b39de..000000000000
--- a/include/asm-sh/harp/io.h
+++ /dev/null
@@ -1,10 +0,0 @@
-#ifndef __ASM_SH_HARP_IO_H
-#define __ASM_SH_HARP_IO_H
-
-/*
- * Nothing special here.. just use the generic cchip io routines.
- */
-#include <asm/hd64465/io.h>
-
-#endif /* __ASM_SH_HARP_IO_H */
-
diff --git a/include/asm-sh/overdrive/fpga.h b/include/asm-sh/overdrive/fpga.h
deleted file mode 100644
index 1cd87992c124..000000000000
--- a/include/asm-sh/overdrive/fpga.h
+++ /dev/null
@@ -1,15 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- */
-
-#ifndef __FPGA_OD_H__
-#define __FPGA_OD_H__
-
-/* This routine will program up the fpga which interfaces to the galileo */
-int init_overdrive_fpga(void);
-
-#endif
diff --git a/include/asm-sh/overdrive/gt64111.h b/include/asm-sh/overdrive/gt64111.h
deleted file mode 100644
index 01d58bc13a44..000000000000
--- a/include/asm-sh/overdrive/gt64111.h
+++ /dev/null
@@ -1,109 +0,0 @@
-#ifndef _GT64111_H_
-#define _GT64111_H_
-
-#define MASTER_INTERFACE         0x0
-#define RAS10_LO_DEC_ADR         0x8        
-#define RAS10_HI_DEC_ADR         0x10
-#define RAS32_LO_DEC_ADR         0x18
-#define RAS32_HI_DEC_ADR         0x20
-#define CS20_LO_DEC_ADR          0x28
-#define CS20_HI_DEC_ADR          0x30
-#define CS3_LO_DEC_ADR           0x38
-#define CS3_HI_DEC_ADR           0x40
-#define PCI_IO_LO_DEC_ADR        0x48
-#define PCI_IO_HI_DEC_ADR        0x50
-#define PCI_MEM0_LO_DEC_ADR      0x58
-#define PCI_MEM0_HI_DEC_ADR      0x60
-#define INTERNAL_SPACE_DEC       0x68
-#define BUS_ERR_ADR_LO_CPU       0x70
-#define READONLY0                0x78
-#define PCI_MEM1_LO_DEC_ADR      0x80
-#define PCI_MEM1_HI_DEC_ADR      0x88
-#define RAS0_LO_DEC_ADR          0x400   
-#define RAS0_HI_DEC_ADR          0x404
-#define RAS1_LO_DEC_ADR          0x408
-#define RAS1_HI_DEC_ADR          0x40c
-#define RAS2_LO_DEC_ADR          0x410
-#define RAS2_HI_DEC_ADR          0x414
-#define RAS3_LO_DEC_ADR          0x418
-#define RAS3_HI_DEC_ADR          0x41c
-#define DEV_CS0_LO_DEC_ADR       0x420
-#define DEV_CS0_HI_DEC_ADR       0x424
-#define DEV_CS1_LO_DEC_ADR       0x428
-#define DEV_CS1_HI_DEC_ADR       0x42c
-#define DEV_CS2_LO_DEC_ADR       0x430
-#define DEV_CS2_HI_DEC_ADR       0x434
-#define DEV_CS3_LO_DEC_ADR       0x438
-#define DEV_CS3_HI_DEC_ADR       0x43c
-#define DEV_BOOTCS_LO_DEC_ADR    0x440
-#define DEV_BOOTCS_HI_DEC_ADR    0x444
-#define DEV_ADR_DEC_ERR          0x470
-#define DRAM_CFG                 0x448   
-#define DRAM_BANK0_PARMS         0x44c   
-#define DRAM_BANK1_PARMS         0x450
-#define DRAM_BANK2_PARMS         0x454
-#define DRAM_BANK3_PARMS         0x458
-#define DEV_BANK0_PARMS          0x45c
-#define DEV_BANK1_PARMS          0x460
-#define DEV_BANK2_PARMS          0x464
-#define DEV_BANK3_PARMS          0x468
-#define DEV_BOOT_BANK_PARMS      0x46c
-#define CH0_DMA_BYTECOUNT        0x800
-#define CH1_DMA_BYTECOUNT        0x804
-#define CH2_DMA_BYTECOUNT        0x808
-#define CH3_DMA_BYTECOUNT        0x80c
-#define CH0_DMA_SRC_ADR          0x810
-#define CH1_DMA_SRC_ADR          0x814
-#define CH2_DMA_SRC_ADR          0x818
-#define CH3_DMA_SRC_ADR          0x81c
-#define CH0_DMA_DST_ADR          0x820
-#define CH1_DMA_DST_ADR          0x824
-#define CH2_DMA_DST_ADR          0x828
-#define CH3_DMA_DST_ADR          0x82c
-#define CH0_NEXT_REC_PTR         0x830
-#define CH1_NEXT_REC_PTR         0x834
-#define CH2_NEXT_REC_PTR         0x838
-#define CH3_NEXT_REC_PTR         0x83c
-#define CH0_CTRL                 0x840
-#define CH1_CTRL                 0x844
-#define CH2_CTRL                 0x848
-#define CH3_CTRL                 0x84c
-#define DMA_ARBITER              0x860
-#define TIMER0                   0x850
-#define TIMER1                   0x854
-#define TIMER2                   0x858
-#define TIMER3                   0x85c
-#define TIMER_CTRL               0x864
-#define PCI_CMD                  0xc00
-#define PCI_TIMEOUT              0xc04
-#define PCI_RAS10_BANK_SIZE      0xc08
-#define PCI_RAS32_BANK_SIZE      0xc0c
-#define PCI_CS20_BANK_SIZE       0xc10
-#define PCI_CS3_BANK_SIZE        0xc14
-#define PCI_SERRMASK             0xc28
-#define PCI_INTACK               0xc34
-#define PCI_BAR_EN               0xc3c
-#define PCI_CFG_ADR              0xcf8
-#define PCI_CFG_DATA             0xcfc
-#define PCI_INTCAUSE             0xc18
-#define PCI_MAST_MASK            0xc1c
-#define PCI_PCIMASK              0xc24
-#define BAR_ENABLE_ADR           0xc3c
-
-/* These are config registers, accessible via PCI space */
-#define PCI_CONFIG_RAS10_BASE_ADR   0x010
-#define PCI_CONFIG_RAS32_BASE_ADR   0x014
-#define PCI_CONFIG_CS20_BASE_ADR    0x018
-#define PCI_CONFIG_CS3_BASE_ADR     0x01c
-#define PCI_CONFIG_INT_REG_MM_ADR   0x020
-#define PCI_CONFIG_INT_REG_IO_ADR   0x024
-#define PCI_CONFIG_BOARD_VENDOR     0x02c
-#define PCI_CONFIG_ROM_ADR          0x030
-#define PCI_CONFIG_INT_PIN_LINE     0x03c
-
-
-
-
-
-#endif
-
diff --git a/include/asm-sh/overdrive/io.h b/include/asm-sh/overdrive/io.h
deleted file mode 100644
index 0dba700e9643..000000000000
--- a/include/asm-sh/overdrive/io.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * include/asm-sh/io_od.h
- *
- * Copyright 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * IO functions for an STMicroelectronics Overdrive
- */
-
-#ifndef _ASM_SH_IO_OD_H
-#define _ASM_SH_IO_OD_H
-
-extern unsigned char od_inb(unsigned long port);
-extern unsigned short od_inw(unsigned long port);
-extern unsigned int od_inl(unsigned long port);
-
-extern void od_outb(unsigned char value, unsigned long port);
-extern void od_outw(unsigned short value, unsigned long port);
-extern void od_outl(unsigned int value, unsigned long port);
-
-extern unsigned char od_inb_p(unsigned long port);
-extern unsigned short od_inw_p(unsigned long port);
-extern unsigned int od_inl_p(unsigned long port);
-extern void od_outb_p(unsigned char value, unsigned long port);
-extern void od_outw_p(unsigned short value, unsigned long port);
-extern void od_outl_p(unsigned int value, unsigned long port);
-
-extern void od_insb(unsigned long port, void *addr, unsigned long count);
-extern void od_insw(unsigned long port, void *addr, unsigned long count);
-extern void od_insl(unsigned long port, void *addr, unsigned long count);
-extern void od_outsb(unsigned long port, const void *addr, unsigned long count);
-extern void od_outsw(unsigned long port, const void *addr, unsigned long count);
-extern void od_outsl(unsigned long port, const void *addr, unsigned long count);
-
-extern unsigned long od_isa_port2addr(unsigned long offset);
-
-#endif /* _ASM_SH_IO_OD_H */
diff --git a/include/asm-sh/overdrive/overdrive.h b/include/asm-sh/overdrive/overdrive.h
deleted file mode 100644
index fc746c244f83..000000000000
--- a/include/asm-sh/overdrive/overdrive.h
+++ /dev/null
@@ -1,88 +0,0 @@
-/* 
- * Copyright (C) 2000 David J. Mckay (david.mckay@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.                            
- *
- */
-
-
-#ifndef __OVERDRIVE_H__
-#define __OVERDRIVE_H__
-
-#define OVERDRIVE_INT_CT 0xa3a00000
-#define OVERDRIVE_INT_DT 0xa3b00000
-
-#define OVERDRIVE_CTRL    0xa3000000
-
-/* Shoving all these bits into the same register is not a good idea. 
- * As soon as I get a spare moment, I'll change the FPGA and put each 
- * bit in a separate register
- */
-
-#define VALID_CTRL_BITS		          0x1f
-
-#define ENABLE_RS232_MASK	  	  0x1e
-#define DISABLE_RS232_BIT		  0x01
-
-#define ENABLE_NMI_MASK			  0x1d
-#define DISABLE_NMI_BIT			  0x02
-
-#define RESET_PCI_MASK			  0x1b
-#define ENABLE_PCI_BIT			  0x04
-
-#define ENABLE_LED_MASK			  0x17
-#define DISABLE_LED_BIT			  0x08
-
-#define RESET_FPGA_MASK			  0x0f
-#define ENABLE_FPGA_BIT			  0x10
-
-
-#define FPGA_DCLK_ADDRESS           0xA3C00000
-
-#define FPGA_DATA        0x01	/*   W */
-#define FPGA_CONFDONE    0x02	/* R   */
-#define FPGA_NOT_STATUS  0x04	/* R   */
-#define FPGA_INITDONE    0x08	/* R   */
-
-#define FPGA_TIMEOUT     100000
-
-
-/* Interrupts for the overdrive. Note that these numbers have 
- * nothing to do with the actual IRQ numbers they appear on, 
- * this is all programmable. This is simply the position in the 
- * INT_CT register.
- */
-
-#define OVERDRIVE_PCI_INTA              0
-#define OVERDRIVE_PCI_INTB              1
-#define OVERDRIVE_PCI_INTC              2
-#define OVERDRIVE_PCI_INTD              3
-#define OVERDRIVE_GALILEO_INT           4
-#define OVERDRIVE_GALILEO_LOCAL_INT     5
-#define OVERDRIVE_AUDIO_INT             6
-#define OVERDRIVE_KEYBOARD_INT          7
-
-/* Which Linux IRQ should we assign to each interrupt source? */
-#define OVERDRIVE_PCI_IRQ1              2
-#ifdef CONFIG_HACKED_NE2K
-#define OVERDRIVE_PCI_IRQ2              7
-#else
-#define OVERDRIVE_PCI_IRQ2              2
-#undef OVERDRIVE_PCI_INTB 
-#define OVERDRIVE_PCI_INTB OVERDRIVE_PCI_INTA
-
-#endif
-
-/* Put the ESS solo audio chip on IRQ 4 */
-#define OVERDRIVE_ESS_IRQ               4
-
-/* Where the memory behind the PCI bus appears */
-#define PCI_DRAM_BASE   0xb7000000
-#define PCI_DRAM_SIZE (16*1024*1024)
-#define PCI_DRAM_FINISH (PCI_DRAM_BASE+PCI_DRAM_SIZE-1)
-
-/* Where the IO region appears in the memory */
-#define PCI_GTIO_BASE   0xb8000000
-
-#endif
diff --git a/include/asm-sh/sh2000/sh2000.h b/include/asm-sh/sh2000/sh2000.h
deleted file mode 100644
index 8d547324d59a..000000000000
--- a/include/asm-sh/sh2000/sh2000.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef __ASM_SH_SH2000_SH2000_H
-#define __ASM_SH_SH2000_SH2000_H
-
-/* arch/sh/boards/sh2000/setup.c */
-extern int setup_sh2000(void);
-
-#endif /* __ASM_SH_SH2000_SH2000_H */
-
-- 
cgit v1.2.1


From ba463937ef75bceaf3943edf01f849257c68623a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:01:16 +0900
Subject: sh: maskreg IRQ support.

Formerly implemented by ADX, we can use this generically,
so move it over.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  |  3 ++
 arch/sh/kernel/cpu/irq/Makefile  |  5 +-
 arch/sh/kernel/cpu/irq/maskreg.c | 99 ++++++++++++++++++++++++++++++++++++++++
 3 files changed, 105 insertions(+), 2 deletions(-)
 create mode 100644 arch/sh/kernel/cpu/irq/maskreg.c

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index ddf2cc56868c..d90fb893bea9 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -313,6 +313,9 @@ config CPU_HAS_INTEVT
 config CPU_HAS_PINT_IRQ
 	bool
 
+config CPU_HAS_MASKREG_IRQ
+	bool
+
 config CPU_HAS_INTC2_IRQ
 	bool
 
diff --git a/arch/sh/kernel/cpu/irq/Makefile b/arch/sh/kernel/cpu/irq/Makefile
index e3cccea15e1d..1c034c283f59 100644
--- a/arch/sh/kernel/cpu/irq/Makefile
+++ b/arch/sh/kernel/cpu/irq/Makefile
@@ -3,5 +3,6 @@
 #
 obj-y	+= ipr.o imask.o
 
-obj-$(CONFIG_CPU_HAS_PINT_IRQ)	+= pint.o
-obj-$(CONFIG_CPU_HAS_INTC2_IRQ)	+= intc2.o
+obj-$(CONFIG_CPU_HAS_PINT_IRQ)		+= pint.o
+obj-$(CONFIG_CPU_HAS_MASKREG_IRQ)	+= maskreg.o
+obj-$(CONFIG_CPU_HAS_INTC2_IRQ)		+= intc2.o
diff --git a/arch/sh/kernel/cpu/irq/maskreg.c b/arch/sh/kernel/cpu/irq/maskreg.c
new file mode 100644
index 000000000000..1cc0de15e4a6
--- /dev/null
+++ b/arch/sh/kernel/cpu/irq/maskreg.c
@@ -0,0 +1,99 @@
+/*
+ * Interrupt handling for Simple external interrupt mask register
+ *
+ * Copyright (C) 2001 A&D Co., Ltd. <http://www.aandd.co.jp>
+ *
+ * This is for the machine which have single 16 bit register
+ * for masking external IRQ individually.
+ * Each bit of the register is for masking each interrupt.
+ *
+ * This file may be copied or modified under the terms of the GNU
+ * General Public License.  See linux/COPYING for more information.
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <asm/system.h>
+#include <asm/io.h>
+
+/* address of external interrupt mask register */
+unsigned long irq_mask_register;
+
+/* forward declaration */
+static unsigned int startup_maskreg_irq(unsigned int irq);
+static void shutdown_maskreg_irq(unsigned int irq);
+static void enable_maskreg_irq(unsigned int irq);
+static void disable_maskreg_irq(unsigned int irq);
+static void mask_and_ack_maskreg(unsigned int);
+static void end_maskreg_irq(unsigned int irq);
+
+/* hw_interrupt_type */
+static struct hw_interrupt_type maskreg_irq_type = {
+	.typename = "Mask Register",
+	.startup = startup_maskreg_irq,
+	.shutdown = shutdown_maskreg_irq,
+	.enable = enable_maskreg_irq,
+	.disable = disable_maskreg_irq,
+	.ack = mask_and_ack_maskreg,
+	.end = end_maskreg_irq
+};
+
+/* actual implementatin */
+static unsigned int startup_maskreg_irq(unsigned int irq)
+{
+	enable_maskreg_irq(irq);
+	return 0; /* never anything pending */
+}
+
+static void shutdown_maskreg_irq(unsigned int irq)
+{
+	disable_maskreg_irq(irq);
+}
+
+static void disable_maskreg_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val, mask = 0x01 << irq;
+
+	BUG_ON(!irq_mask_register);
+
+	/* Set "irq"th bit */
+	local_irq_save(flags);
+	val = ctrl_inw(irq_mask_register);
+	val |= mask;
+	ctrl_outw(val, irq_mask_register);
+	local_irq_restore(flags);
+}
+
+static void enable_maskreg_irq(unsigned int irq)
+{
+	unsigned long flags;
+	unsigned short val, mask = ~(0x01 << irq);
+
+	BUG_ON(!irq_mask_register);
+
+	/* Clear "irq"th bit */
+	local_irq_save(flags);
+	val = ctrl_inw(irq_mask_register);
+	val &= mask;
+	ctrl_outw(val, irq_mask_register);
+	local_irq_restore(flags);
+}
+
+static void mask_and_ack_maskreg(unsigned int irq)
+{
+	disable_maskreg_irq(irq);
+}
+
+static void end_maskreg_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
+		enable_maskreg_irq(irq);
+}
+
+void make_maskreg_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &maskreg_irq_type;
+	disable_maskreg_irq(irq);
+}
-- 
cgit v1.2.1


From 8599cf059209de22dd3be16816b90f1aad10c74a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:03:34 +0900
Subject: sh: Cleanup IRQ disabling for hardirq handlers.

The generic hardirq layer already takes care of a lot of the
appropriate locking and disabling for us, no need to duplicate
it in the handlers..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/bigsur/irq.c              | 47 ++++++++++++--------------------
 arch/sh/boards/dreamcast/irq.c           |  7 -----
 arch/sh/boards/landisk/irq.c             |  6 ----
 arch/sh/boards/mpc1211/setup.c           | 14 ----------
 arch/sh/boards/renesas/hs7751rvoip/irq.c |  6 ----
 arch/sh/boards/renesas/r7780rp/irq.c     |  6 ----
 arch/sh/boards/renesas/rts7751r2d/irq.c  |  6 ----
 arch/sh/boards/renesas/systemh/irq.c     |  8 ------
 arch/sh/boards/superh/microdev/irq.c     | 39 ++++++++------------------
 arch/sh/cchips/hd6446x/hd64461/setup.c   |  7 -----
 arch/sh/cchips/hd6446x/hd64465/setup.c   |  6 ----
 arch/sh/cchips/voyagergx/irq.c           | 20 ++++----------
 arch/sh/kernel/cpu/irq/ipr.c             |  8 ++----
 arch/sh/kernel/cpu/irq/maskreg.c         |  6 ----
 arch/sh/kernel/cpu/irq/pint.c            |  8 ++----
 15 files changed, 38 insertions(+), 156 deletions(-)

diff --git a/arch/sh/boards/bigsur/irq.c b/arch/sh/boards/bigsur/irq.c
index ac946a2201c7..1ab04da36382 100644
--- a/arch/sh/boards/bigsur/irq.c
+++ b/arch/sh/boards/bigsur/irq.c
@@ -19,6 +19,7 @@
  * IRQ functions for a Hitachi Big Sur Evaluation Board.
  *
  */
+#undef DEBUG
 
 #include <linux/sched.h>
 #include <linux/module.h>
@@ -41,10 +42,8 @@
 #undef BIGSUR_DEBUG
 
 #ifdef BIGSUR_DEBUG
-#define DPRINTK(args...)        printk(args)
 #define DIPRINTK(n, args...)    if (BIGSUR_DEBUG>(n)) printk(args)
 #else
-#define DPRINTK(args...)
 #define DIPRINTK(n, args...)
 #endif /* BIGSUR_DEBUG */
 
@@ -60,45 +59,39 @@ extern int hd64465_irq_demux(int irq);
 /* Level 1 IRQ routines */
 static void disable_bigsur_l1irq(unsigned int irq)
 {
-        unsigned long flags;
         unsigned char mask;
         unsigned int mask_port = ((irq - BIGSUR_IRQ_LOW)/8) ? BIGSUR_IRLMR1 : BIGSUR_IRLMR0;
         unsigned char bit =  (1 << ((irq - MGATE_IRQ_LOW)%8) );
 
         if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH) {
-                DPRINTK("Disable L1 IRQ %d\n", irq);
+                pr_debug("Disable L1 IRQ %d\n", irq);
                 DIPRINTK(2,"disable_bigsur_l1irq: IMR=0x%08x mask=0x%x\n",
                         mask_port, bit);
-                local_irq_save(flags);
 
                 /* Disable IRQ - set mask bit */
                 mask = inb(mask_port) | bit;
                 outb(mask, mask_port);
-                local_irq_restore(flags);
                 return;
         }
-        DPRINTK("disable_bigsur_l1irq: Invalid IRQ %d\n", irq);
+        pr_debug("disable_bigsur_l1irq: Invalid IRQ %d\n", irq);
 }
 
 static void enable_bigsur_l1irq(unsigned int irq)
 {
-        unsigned long flags;
         unsigned char mask;
         unsigned int mask_port = ((irq - BIGSUR_IRQ_LOW)/8) ? BIGSUR_IRLMR1 : BIGSUR_IRLMR0;
         unsigned char bit =  (1 << ((irq - MGATE_IRQ_LOW)%8) );
 
         if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH) {
-                DPRINTK("Enable L1 IRQ %d\n", irq);
+                pr_debug("Enable L1 IRQ %d\n", irq);
                 DIPRINTK(2,"enable_bigsur_l1irq: IMR=0x%08x mask=0x%x\n",
                         mask_port, bit);
-                local_irq_save(flags);
                 /* Enable L1 IRQ - clear mask bit */
                 mask = inb(mask_port) & ~bit;
                 outb(mask, mask_port);
-                local_irq_restore(flags);
                 return;
         }
-        DPRINTK("enable_bigsur_l1irq: Invalid IRQ %d\n", irq);
+        pr_debug("enable_bigsur_l1irq: Invalid IRQ %d\n", irq);
 }
 
 
@@ -126,51 +119,45 @@ static const u32 imr_offset = BIGSUR_IMR0 - BIGSUR_IMR1;
 /* Level 2 IRQ routines */
 static void disable_bigsur_l2irq(unsigned int irq)
 {
-        unsigned long flags;
         unsigned char mask;
         unsigned char bit = 1 << ((irq-BIGSUR_2NDLVL_IRQ_LOW)%8);
         unsigned int mask_port = imr_base - REG_NUM(irq)*imr_offset;
 
-    if(irq >= BIGSUR_2NDLVL_IRQ_LOW && irq < BIGSUR_2NDLVL_IRQ_HIGH) {
-                DPRINTK("Disable L2 IRQ %d\n", irq);
+	if(irq >= BIGSUR_2NDLVL_IRQ_LOW && irq < BIGSUR_2NDLVL_IRQ_HIGH) {
+                pr_debug("Disable L2 IRQ %d\n", irq);
                 DIPRINTK(2,"disable_bigsur_l2irq: IMR=0x%08x mask=0x%x\n",
                         mask_port, bit);
-                local_irq_save(flags);
 
                 /* Disable L2 IRQ - set mask bit */
                 mask = inb(mask_port) | bit;
                 outb(mask, mask_port);
-                local_irq_restore(flags);
                 return;
         }
-        DPRINTK("disable_bigsur_l2irq: Invalid IRQ %d\n", irq);
+        pr_debug("disable_bigsur_l2irq: Invalid IRQ %d\n", irq);
 }
 
 static void enable_bigsur_l2irq(unsigned int irq)
 {
-        unsigned long flags;
         unsigned char mask;
         unsigned char bit = 1 << ((irq-BIGSUR_2NDLVL_IRQ_LOW)%8);
         unsigned int mask_port = imr_base - REG_NUM(irq)*imr_offset;
 
-    if(irq >= BIGSUR_2NDLVL_IRQ_LOW && irq < BIGSUR_2NDLVL_IRQ_HIGH) {
-                DPRINTK("Enable L2 IRQ %d\n", irq);
+	if(irq >= BIGSUR_2NDLVL_IRQ_LOW && irq < BIGSUR_2NDLVL_IRQ_HIGH) {
+                pr_debug("Enable L2 IRQ %d\n", irq);
                 DIPRINTK(2,"enable_bigsur_l2irq: IMR=0x%08x mask=0x%x\n",
                         mask_port, bit);
-                local_irq_save(flags);
 
                 /* Enable L2 IRQ - clear mask bit */
                 mask = inb(mask_port) & ~bit;
                 outb(mask, mask_port);
-                local_irq_restore(flags);
                 return;
         }
-        DPRINTK("enable_bigsur_l2irq: Invalid IRQ %d\n", irq);
+        pr_debug("enable_bigsur_l2irq: Invalid IRQ %d\n", irq);
 }
 
 static void mask_and_ack_bigsur(unsigned int irq)
 {
-        DPRINTK("mask_and_ack_bigsur IRQ %d\n", irq);
+        pr_debug("mask_and_ack_bigsur IRQ %d\n", irq);
         if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH)
                 disable_bigsur_l1irq(irq);
         else
@@ -179,7 +166,7 @@ static void mask_and_ack_bigsur(unsigned int irq)
 
 static void end_bigsur_irq(unsigned int irq)
 {
-        DPRINTK("end_bigsur_irq IRQ %d\n", irq);
+        pr_debug("end_bigsur_irq IRQ %d\n", irq);
         if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS))) {
                 if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH)
                         enable_bigsur_l1irq(irq);
@@ -193,7 +180,7 @@ static unsigned int startup_bigsur_irq(unsigned int irq)
         u8 mask;
         u32 reg;
 
-        DPRINTK("startup_bigsur_irq IRQ %d\n", irq);
+        pr_debug("startup_bigsur_irq IRQ %d\n", irq);
 
         if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH) {
                 /* Enable the L1 IRQ */
@@ -218,7 +205,7 @@ static unsigned int startup_bigsur_irq(unsigned int irq)
 
 static void shutdown_bigsur_irq(unsigned int irq)
 {
-        DPRINTK("shutdown_bigsur_irq IRQ %d\n", irq);
+        pr_debug("shutdown_bigsur_irq IRQ %d\n", irq);
         if(irq >= BIGSUR_IRQ_LOW && irq < BIGSUR_IRQ_HIGH)
                 disable_bigsur_l1irq(irq);
         else
@@ -260,7 +247,7 @@ static void make_bigsur_l1isr(unsigned int irq) {
                 disable_bigsur_l1irq(irq);
                 return;
         }
-        DPRINTK("make_bigsur_l1isr: bad irq, %d\n", irq);
+        pr_debug("make_bigsur_l1isr: bad irq, %d\n", irq);
         return;
 }
 
@@ -277,7 +264,7 @@ static void make_bigsur_l2isr(unsigned int irq) {
                 disable_bigsur_l2irq(irq);
                 return;
         }
-        DPRINTK("make_bigsur_l2isr: bad irq, %d\n", irq);
+        pr_debug("make_bigsur_l2isr: bad irq, %d\n", irq);
         return;
 }
 
diff --git a/arch/sh/boards/dreamcast/irq.c b/arch/sh/boards/dreamcast/irq.c
index 373a22e48ac4..5bf01f86c20c 100644
--- a/arch/sh/boards/dreamcast/irq.c
+++ b/arch/sh/boards/dreamcast/irq.c
@@ -10,7 +10,6 @@
  */
 
 #include <linux/irq.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/dreamcast/sysasic.h>
@@ -57,29 +56,23 @@
 /* Disable the hardware event by masking its bit in its EMR */
 static inline void disable_systemasic_irq(unsigned int irq)
 {
-        unsigned long flags;
         __u32 emr = EMR_BASE + (LEVEL(irq) << 4) + (LEVEL(irq) << 2);
         __u32 mask;
 
-        local_irq_save(flags);
         mask = inl(emr);
         mask &= ~(1 << EVENT_BIT(irq));
         outl(mask, emr);
-        local_irq_restore(flags);
 }
 
 /* Enable the hardware event by setting its bit in its EMR */
 static inline void enable_systemasic_irq(unsigned int irq)
 {
-        unsigned long flags;
         __u32 emr = EMR_BASE + (LEVEL(irq) << 4) + (LEVEL(irq) << 2);
         __u32 mask;
 
-        local_irq_save(flags);
         mask = inl(emr);
         mask |= (1 << EVENT_BIT(irq));
         outl(mask, emr);
-        local_irq_restore(flags);
 }
 
 /* Acknowledge a hardware event by writing its bit back to its ESR */
diff --git a/arch/sh/boards/landisk/irq.c b/arch/sh/boards/landisk/irq.c
index 1dcc3feb7a44..a006d6443225 100644
--- a/arch/sh/boards/landisk/irq.c
+++ b/arch/sh/boards/landisk/irq.c
@@ -39,30 +39,24 @@ static unsigned int startup_landisk_irq(unsigned int irq)
 
 static void disable_landisk_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned char val;
 	unsigned char mask = 0xff ^ (0x01 << (irq - 5));
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inb(PA_IMASK);
 	val &= mask;
 	ctrl_outb(val, PA_IMASK);
-	local_irq_restore(flags);
 }
 
 static void enable_landisk_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned char val;
 	unsigned char value = (0x01 << (irq - 5));
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inb(PA_IMASK);
 	val |= value;
 	ctrl_outb(val, PA_IMASK);
-	local_irq_restore(flags);
 }
 
 static void ack_landisk_irq(unsigned int irq)
diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c
index a8c5180ae219..e9c8ff221dda 100644
--- a/arch/sh/boards/mpc1211/setup.c
+++ b/arch/sh/boards/mpc1211/setup.c
@@ -80,9 +80,6 @@ volatile unsigned long irq_err_count;
 
 static void disable_mpc1211_irq(unsigned int irq)
 {
-	unsigned long flags;
-
-	save_and_cli(flags);
 	if( irq < 8) {
 		m_irq_mask |= (1 << irq);
 		outb(m_irq_mask,I8259_M_MR);
@@ -90,16 +87,11 @@ static void disable_mpc1211_irq(unsigned int irq)
 		s_irq_mask |= (1 << (irq - 8));
 		outb(s_irq_mask,I8259_S_MR);
 	}
-	restore_flags(flags);
 
 }
 
 static void enable_mpc1211_irq(unsigned int irq)
 {
-	unsigned long flags;
-
-	save_and_cli(flags);
-
 	if( irq < 8) {
 		m_irq_mask &= ~(1 << irq);
 		outb(m_irq_mask,I8259_M_MR);
@@ -107,7 +99,6 @@ static void enable_mpc1211_irq(unsigned int irq)
 		s_irq_mask &= ~(1 << (irq - 8));
 		outb(s_irq_mask,I8259_S_MR);
 	}
-	restore_flags(flags);
 }
 
 static inline int mpc1211_irq_real(unsigned int irq)
@@ -131,10 +122,6 @@ static inline int mpc1211_irq_real(unsigned int irq)
 
 static void mask_and_ack_mpc1211(unsigned int irq)
 {
-	unsigned long flags;
-
-	save_and_cli(flags);
-
 	if(irq < 8) {
 		if(m_irq_mask & (1<<irq)){
 		  if(!mpc1211_irq_real(irq)){
@@ -162,7 +149,6 @@ static void mask_and_ack_mpc1211(unsigned int irq)
 		outb(0x60+(irq-8),I8259_S_CR); 	/* EOI */
 		outb(0x60+2,I8259_M_CR);
 	}
-	restore_flags(flags);
 }
 
 static void end_mpc1211_irq(unsigned int irq)
diff --git a/arch/sh/boards/renesas/hs7751rvoip/irq.c b/arch/sh/boards/renesas/hs7751rvoip/irq.c
index 705b7ddcb0d2..c617b188258a 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/irq.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/irq.c
@@ -35,30 +35,24 @@ static unsigned int startup_hs7751rvoip_irq(unsigned int irq)
 
 static void disable_hs7751rvoip_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR3);
 	val &= mask;
 	ctrl_outw(val, IRLCNTR3);
-	local_irq_restore(flags);
 }
 
 static void enable_hs7751rvoip_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short value = (0x0001 << mask_pos[irq]);
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR3);
 	val |= value;
 	ctrl_outw(val, IRLCNTR3);
-	local_irq_restore(flags);
 }
 
 static void ack_hs7751rvoip_irq(unsigned int irq)
diff --git a/arch/sh/boards/renesas/r7780rp/irq.c b/arch/sh/boards/renesas/r7780rp/irq.c
index c7b9fe6e640d..61d5e5d3c294 100644
--- a/arch/sh/boards/renesas/r7780rp/irq.c
+++ b/arch/sh/boards/renesas/r7780rp/irq.c
@@ -39,30 +39,24 @@ static unsigned int startup_r7780rp_irq(unsigned int irq)
 
 static void disable_r7780rp_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR1);
 	val &= mask;
 	ctrl_outw(val, IRLCNTR1);
-	local_irq_restore(flags);
 }
 
 static void enable_r7780rp_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short value = (0x0001 << mask_pos[irq]);
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR1);
 	val |= value;
 	ctrl_outw(val, IRLCNTR1);
-	local_irq_restore(flags);
 }
 
 static void ack_r7780rp_irq(unsigned int irq)
diff --git a/arch/sh/boards/renesas/rts7751r2d/irq.c b/arch/sh/boards/renesas/rts7751r2d/irq.c
index 154535440bbf..c915e7a3693a 100644
--- a/arch/sh/boards/renesas/rts7751r2d/irq.c
+++ b/arch/sh/boards/renesas/rts7751r2d/irq.c
@@ -41,30 +41,24 @@ static unsigned int startup_rts7751r2d_irq(unsigned int irq)
 
 static void disable_rts7751r2d_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short mask = 0xffff ^ (0x0001 << mask_pos[irq]);
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR1);
 	val &= mask;
 	ctrl_outw(val, IRLCNTR1);
-	local_irq_restore(flags);
 }
 
 static void enable_rts7751r2d_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val;
 	unsigned short value = (0x0001 << mask_pos[irq]);
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inw(IRLCNTR1);
 	val |= value;
 	ctrl_outw(val, IRLCNTR1);
-	local_irq_restore(flags);
 }
 
 int rts7751r2d_irq_demux(int irq)
diff --git a/arch/sh/boards/renesas/systemh/irq.c b/arch/sh/boards/renesas/systemh/irq.c
index 53731a0abb8f..8d016dae2333 100644
--- a/arch/sh/boards/renesas/systemh/irq.c
+++ b/arch/sh/boards/renesas/systemh/irq.c
@@ -57,12 +57,9 @@ static void shutdown_systemh_irq(unsigned int irq)
 static void disable_systemh_irq(unsigned int irq)
 {
 	if (systemh_irq_mask_register) {
-		unsigned long flags;
 		unsigned long val, mask = 0x01 << 1;
 
 		/* Clear the "irq"th bit in the mask and set it in the request */
-		local_irq_save(flags);
-
 		val = ctrl_inl((unsigned long)systemh_irq_mask_register);
 		val &= ~mask;
 		ctrl_outl(val, (unsigned long)systemh_irq_mask_register);
@@ -70,23 +67,18 @@ static void disable_systemh_irq(unsigned int irq)
 		val = ctrl_inl((unsigned long)systemh_irq_request_register);
 		val |= mask;
 		ctrl_outl(val, (unsigned long)systemh_irq_request_register);
-
-		local_irq_restore(flags);
 	}
 }
 
 static void enable_systemh_irq(unsigned int irq)
 {
 	if (systemh_irq_mask_register) {
-		unsigned long flags;
 		unsigned long val, mask = 0x01 << 1;
 
 		/* Set "irq"th bit in the mask register */
-		local_irq_save(flags);
 		val = ctrl_inl((unsigned long)systemh_irq_mask_register);
 		val |= mask;
 		ctrl_outl(val, (unsigned long)systemh_irq_mask_register);
-		local_irq_restore(flags);
 	}
 }
 
diff --git a/arch/sh/boards/superh/microdev/irq.c b/arch/sh/boards/superh/microdev/irq.c
index 236398fbc083..8c64baa30364 100644
--- a/arch/sh/boards/superh/microdev/irq.c
+++ b/arch/sh/boards/superh/microdev/irq.c
@@ -11,14 +11,12 @@
 
 #include <linux/init.h>
 #include <linux/irq.h>
-
 #include <asm/system.h>
 #include <asm/io.h>
 #include <asm/microdev.h>
 
 #define NUM_EXTERNAL_IRQS 16	/* IRL0 .. IRL15 */
 
-
 static const struct {
 	unsigned char fpgaIrq;
 	unsigned char mapped;
@@ -93,53 +91,42 @@ static struct hw_interrupt_type microdev_irq_type = {
 
 static void disable_microdev_irq(unsigned int irq)
 {
-	unsigned int flags;
 	unsigned int fpgaIrq;
 
-	if (irq >= NUM_EXTERNAL_IRQS) return;
-	if (!fpgaIrqTable[irq].mapped) return;
+	if (irq >= NUM_EXTERNAL_IRQS)
+		return;
+	if (!fpgaIrqTable[irq].mapped)
+		return;
 
 	fpgaIrq = fpgaIrqTable[irq].fpgaIrq;
 
-		/* disable interrupts */
-	local_irq_save(flags);
-
-		/* disable interupts on the FPGA INTC register */
+	/* disable interupts on the FPGA INTC register */
 	ctrl_outl(MICRODEV_FPGA_INTC_MASK(fpgaIrq), MICRODEV_FPGA_INTDSB_REG);
-
-		/* restore interrupts */
-	local_irq_restore(flags);
 }
 
 static void enable_microdev_irq(unsigned int irq)
 {
 	unsigned long priorityReg, priorities, pri;
-	unsigned int flags;
 	unsigned int fpgaIrq;
 
-
-	if (irq >= NUM_EXTERNAL_IRQS) return;
-	if (!fpgaIrqTable[irq].mapped) return;
+	if (unlikely(irq >= NUM_EXTERNAL_IRQS))
+		return;
+	if (unlikely(!fpgaIrqTable[irq].mapped))
+		return;
 
 	pri = 15 - irq;
 
 	fpgaIrq = fpgaIrqTable[irq].fpgaIrq;
 	priorityReg = MICRODEV_FPGA_INTPRI_REG(fpgaIrq);
 
-		/* disable interrupts */
-	local_irq_save(flags);
-
-		/* set priority for the interrupt */
+	/* set priority for the interrupt */
 	priorities = ctrl_inl(priorityReg);
 	priorities &= ~MICRODEV_FPGA_INTPRI_MASK(fpgaIrq);
 	priorities |= MICRODEV_FPGA_INTPRI_LEVEL(fpgaIrq, pri);
 	ctrl_outl(priorities, priorityReg);
 
-		/* enable interupts on the FPGA INTC register */
+	/* enable interupts on the FPGA INTC register */
 	ctrl_outl(MICRODEV_FPGA_INTC_MASK(fpgaIrq), MICRODEV_FPGA_INTENB_REG);
-
-		/* restore interrupts */
-	local_irq_restore(flags);
 }
 
 	/* This functions sets the desired irq handler to be a MicroDev type */
@@ -158,9 +145,7 @@ static void mask_and_ack_microdev(unsigned int irq)
 static void end_microdev_irq(unsigned int irq)
 {
 	if (!(irq_desc[irq].status & (IRQ_DISABLED|IRQ_INPROGRESS)))
-	{
 		enable_microdev_irq(irq);
-	}
 }
 
 extern void __init init_microdev_irq(void)
@@ -171,9 +156,7 @@ extern void __init init_microdev_irq(void)
 	ctrl_outl(~0ul, MICRODEV_FPGA_INTDSB_REG);
 
 	for (i = 0; i < NUM_EXTERNAL_IRQS; i++)
-	{
 		make_microdev_irq(i);
-	}
 }
 
 extern void microdev_print_fpga_intc_status(void)
diff --git a/arch/sh/cchips/hd6446x/hd64461/setup.c b/arch/sh/cchips/hd6446x/hd64461/setup.c
index 6c961273a23a..38f1e8171a3a 100644
--- a/arch/sh/cchips/hd6446x/hd64461/setup.c
+++ b/arch/sh/cchips/hd6446x/hd64461/setup.c
@@ -11,35 +11,28 @@
 #include <linux/interrupt.h>
 #include <linux/init.h>
 #include <linux/irq.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/hd64461.h>
 
 static void disable_hd64461_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short nimr;
 	unsigned short mask = 1 << (irq - HD64461_IRQBASE);
 
-	local_irq_save(flags);
 	nimr = inw(HD64461_NIMR);
 	nimr |= mask;
 	outw(nimr, HD64461_NIMR);
-	local_irq_restore(flags);
 }
 
 static void enable_hd64461_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short nimr;
 	unsigned short mask = 1 << (irq - HD64461_IRQBASE);
 
-	local_irq_save(flags);
 	nimr = inw(HD64461_NIMR);
 	nimr &= ~mask;
 	outw(nimr, HD64461_NIMR);
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_hd64461(unsigned int irq)
diff --git a/arch/sh/cchips/hd6446x/hd64465/setup.c b/arch/sh/cchips/hd6446x/hd64465/setup.c
index d2b2851bc44b..30573d3e1966 100644
--- a/arch/sh/cchips/hd6446x/hd64465/setup.c
+++ b/arch/sh/cchips/hd6446x/hd64465/setup.c
@@ -25,31 +25,25 @@
 
 static void disable_hd64465_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short nimr;
 	unsigned short mask = 1 << (irq - HD64465_IRQ_BASE);
 
     	pr_debug("disable_hd64465_irq(%d): mask=%x\n", irq, mask);
-	local_irq_save(flags);
 	nimr = inw(HD64465_REG_NIMR);
 	nimr |= mask;
 	outw(nimr, HD64465_REG_NIMR);
-	local_irq_restore(flags);
 }
 
 
 static void enable_hd64465_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short nimr;
 	unsigned short mask = 1 << (irq - HD64465_IRQ_BASE);
 
     	pr_debug("enable_hd64465_irq(%d): mask=%x\n", irq, mask);
-	local_irq_save(flags);
 	nimr = inw(HD64465_REG_NIMR);
 	nimr &= ~mask;
 	outw(nimr, HD64465_REG_NIMR);
-	local_irq_restore(flags);
 }
 
 
diff --git a/arch/sh/cchips/voyagergx/irq.c b/arch/sh/cchips/voyagergx/irq.c
index 4cfa7c990553..392c8b12ce36 100644
--- a/arch/sh/cchips/voyagergx/irq.c
+++ b/arch/sh/cchips/voyagergx/irq.c
@@ -36,32 +36,26 @@
 
 static void disable_voyagergx_irq(unsigned int irq)
 {
-	unsigned long flags, val;
-	unsigned long  mask = 1 << (irq - VOYAGER_IRQ_BASE);
+	unsigned long val;
+	unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE);
 
     	pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask);
-	local_irq_save(flags);
         val = inl(VOYAGER_INT_MASK);
         val &= ~mask;
         outl(val, VOYAGER_INT_MASK);
-	local_irq_restore(flags);
 }
 
-
 static void enable_voyagergx_irq(unsigned int irq)
 {
-        unsigned long flags, val;
-        unsigned long  mask = 1 << (irq - VOYAGER_IRQ_BASE);
+        unsigned long val;
+        unsigned long mask = 1 << (irq - VOYAGER_IRQ_BASE);
 
         pr_debug("disable_voyagergx_irq(%d): mask=%x\n", irq, mask);
-        local_irq_save(flags);
         val = inl(VOYAGER_INT_MASK);
         val |= mask;
         outl(val, VOYAGER_INT_MASK);
-        local_irq_restore(flags);
 }
 
-
 static void mask_and_ack_voyagergx(unsigned int irq)
 {
 	disable_voyagergx_irq(irq);
@@ -94,7 +88,8 @@ static struct hw_interrupt_type voyagergx_irq_type = {
 	.end = end_voyagergx_irq,
 };
 
-static irqreturn_t voyagergx_interrupt(int irq, void *dev_id, struct pt_regs *regs)
+static irqreturn_t voyagergx_interrupt(int irq, void *dev_id,
+				      struct pt_regs *regs)
 {
 	printk(KERN_INFO
 	       "VoyagerGX: spurious interrupt, status: 0x%x\n",
@@ -102,9 +97,6 @@ static irqreturn_t voyagergx_interrupt(int irq, void *dev_id, struct pt_regs *re
 	return IRQ_HANDLED;
 }
 
-
-/*====================================================*/
-
 static struct {
 	int (*func)(int, void *);
 	void *dev;
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 00edee971bcb..4c4fd4118d1a 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -57,31 +57,27 @@ static struct hw_interrupt_type ipr_irq_type = {
 
 static void disable_ipr_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 	unsigned int addr = ipr_data[irq].addr;
 	unsigned short mask = 0xffff ^ (0x0f << ipr_data[irq].shift);
 
 	/* Set the priority in IPR to 0 */
-	local_irq_save(flags);
 	val = ctrl_inw(addr);
 	val &= mask;
 	ctrl_outw(val, addr);
-	local_irq_restore(flags);
 }
 
 static void enable_ipr_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 	unsigned int addr = ipr_data[irq].addr;
 	int priority = ipr_data[irq].priority;
 	unsigned short value = (priority << ipr_data[irq].shift);
 
 	/* Set priority in IPR back to original value */
-	local_irq_save(flags);
 	val = ctrl_inw(addr);
 	val |= value;
 	ctrl_outw(val, addr);
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_ipr(unsigned int irq)
diff --git a/arch/sh/kernel/cpu/irq/maskreg.c b/arch/sh/kernel/cpu/irq/maskreg.c
index 1cc0de15e4a6..492db31b3cab 100644
--- a/arch/sh/kernel/cpu/irq/maskreg.c
+++ b/arch/sh/kernel/cpu/irq/maskreg.c
@@ -52,32 +52,26 @@ static void shutdown_maskreg_irq(unsigned int irq)
 
 static void disable_maskreg_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val, mask = 0x01 << irq;
 
 	BUG_ON(!irq_mask_register);
 
 	/* Set "irq"th bit */
-	local_irq_save(flags);
 	val = ctrl_inw(irq_mask_register);
 	val |= mask;
 	ctrl_outw(val, irq_mask_register);
-	local_irq_restore(flags);
 }
 
 static void enable_maskreg_irq(unsigned int irq)
 {
-	unsigned long flags;
 	unsigned short val, mask = ~(0x01 << irq);
 
 	BUG_ON(!irq_mask_register);
 
 	/* Clear "irq"th bit */
-	local_irq_save(flags);
 	val = ctrl_inw(irq_mask_register);
 	val &= mask;
 	ctrl_outw(val, irq_mask_register);
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_maskreg(unsigned int irq)
diff --git a/arch/sh/kernel/cpu/irq/pint.c b/arch/sh/kernel/cpu/irq/pint.c
index 80cd8108d36a..17f47b373d6e 100644
--- a/arch/sh/kernel/cpu/irq/pint.c
+++ b/arch/sh/kernel/cpu/irq/pint.c
@@ -48,26 +48,22 @@ static struct hw_interrupt_type pint_irq_type = {
 
 static void disable_pint_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 
-	local_irq_save(flags);
 	val = ctrl_inw(INTC_INTER);
 	val &= ~(1 << (irq - PINT_IRQ_BASE));
 	ctrl_outw(val, INTC_INTER);	/* disable PINTn */
 	portcr_mask &= ~(3 << (irq - PINT_IRQ_BASE)*2);
-	local_irq_restore(flags);
 }
 
 static void enable_pint_irq(unsigned int irq)
 {
-	unsigned long val, flags;
+	unsigned long val;
 
-	local_irq_save(flags);
 	val = ctrl_inw(INTC_INTER);
 	val |= 1 << (irq - PINT_IRQ_BASE);
 	ctrl_outw(val, INTC_INTER);	/* enable PINTn */
 	portcr_mask |= 3 << (irq - PINT_IRQ_BASE)*2;
-	local_irq_restore(flags);
 }
 
 static void mask_and_ack_pint(unsigned int irq)
-- 
cgit v1.2.1


From 49c3f807f5734605c10cdfb462a8fd5c3075eb76 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:04:36 +0900
Subject: sh: Enable verbose BUG() support.

Add SH to the list of platforms interested in Verbose BUG().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 lib/Kconfig.debug | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index e7303fe54b01..b0f5ca72599f 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -284,7 +284,7 @@ config DEBUG_HIGHMEM
 config DEBUG_BUGVERBOSE
 	bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EMBEDDED
 	depends on BUG
-	depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV
+	depends on ARM || ARM26 || AVR32 || M32R || M68K || SPARC32 || SPARC64 || X86_32 || FRV || SUPERH
 	default !EMBEDDED
 	help
 	  Say Y here to make BUG() panics output the file name and line number
-- 
cgit v1.2.1


From 91b91d01416afba8d3f230a62b5d2784bd7af94a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:08:33 +0900
Subject: sh: SH7710VoIPGW board support.

This adds support for the SH7710 VoIP Gateway board.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                              |  11 +
 arch/sh/Makefile                             |   1 +
 arch/sh/boards/renesas/sh7710voipgw/Makefile |   1 +
 arch/sh/boards/renesas/sh7710voipgw/setup.c  | 109 ++++
 arch/sh/configs/sh7710voipgw_defconfig       | 913 +++++++++++++++++++++++++++
 arch/sh/tools/mach-types                     |   1 +
 6 files changed, 1036 insertions(+)
 create mode 100644 arch/sh/boards/renesas/sh7710voipgw/Makefile
 create mode 100644 arch/sh/boards/renesas/sh7710voipgw/setup.c
 create mode 100644 arch/sh/configs/sh7710voipgw_defconfig

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index d90fb893bea9..2549bd82c57f 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -149,6 +149,13 @@ config SH_HS7751RVOIP
 	  Select HS7751RVOIP if configuring for a Renesas Technology
 	  Sales VoIP board.
 
+config SH_7710VOIPGW
+	bool "SH7710-VOIP-GW"
+	select CPU_SUBTYPE_SH7710
+	help
+	  Select this option to build a kernel for the SH7710 based
+	  VOIP GW.
+
 config SH_RTS7751R2D
 	bool "RTS7751R2D"
 	select CPU_SUBTYPE_SH7751R
@@ -405,6 +412,8 @@ config ISA_DMA_API
 
 menu "Kernel features"
 
+source kernel/Kconfig.hz
+
 config KEXEC
 	bool "kexec system call (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
@@ -461,6 +470,8 @@ config NR_CPUS
 	  This is purely to save memory - each supported CPU adds
 	  approximately eight kilobytes to the kernel image.
 
+source "kernel/Kconfig.preempt"
+
 config CPU_HAS_SR_RB
 	bool "CPU has SR.RB"
 	depends on CPU_SH3 || CPU_SH4
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index d0168844c3c7..f0fc88e5f1ad 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -103,6 +103,7 @@ machdir-$(CONFIG_SH_RTS7751R2D)			:= renesas/rts7751r2d
 machdir-$(CONFIG_SH_7751_SYSTEMH)		:= renesas/systemh
 machdir-$(CONFIG_SH_EDOSK7705)			:= renesas/edosk7705
 machdir-$(CONFIG_SH_R7780RP)			:= renesas/r7780rp
+machdir-$(CONFIG_SH_7710VOIPGW)			:= renesas/sh7710voipgw
 machdir-$(CONFIG_SH_SH4202_MICRODEV)		:= superh/microdev
 machdir-$(CONFIG_SH_LANDISK)			:= landisk
 machdir-$(CONFIG_SH_TITAN)			:= titan
diff --git a/arch/sh/boards/renesas/sh7710voipgw/Makefile b/arch/sh/boards/renesas/sh7710voipgw/Makefile
new file mode 100644
index 000000000000..77037567633b
--- /dev/null
+++ b/arch/sh/boards/renesas/sh7710voipgw/Makefile
@@ -0,0 +1 @@
+obj-y	 := setup.o
diff --git a/arch/sh/boards/renesas/sh7710voipgw/setup.c b/arch/sh/boards/renesas/sh7710voipgw/setup.c
new file mode 100644
index 000000000000..e57e7afab8c6
--- /dev/null
+++ b/arch/sh/boards/renesas/sh7710voipgw/setup.c
@@ -0,0 +1,109 @@
+/*
+ * Renesas Technology SH7710 VoIP Gateway
+ *
+ * Copyright (C) 2006  Ranjit Deshpande
+ * Kenati Technologies Inc.
+ *
+ * May be copied or modified under the terms of the GNU General Public
+ * License.  See linux/COPYING for more information.
+ */
+#include <linux/init.h>
+#include <asm/machvec.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+
+/*
+ * Initialize IRQ setting
+ */
+static void __init sh7710voipgw_init_irq(void)
+{
+	/* Disable all interrupts in IPR registers */
+	ctrl_outw(0x0, INTC_IPRA);
+	ctrl_outw(0x0, INTC_IPRB);
+	ctrl_outw(0x0, INTC_IPRC);
+	ctrl_outw(0x0, INTC_IPRD);
+	ctrl_outw(0x0, INTC_IPRE);
+	ctrl_outw(0x0, INTC_IPRF);
+	ctrl_outw(0x0, INTC_IPRG);
+	ctrl_outw(0x0, INTC_IPRH);
+	ctrl_outw(0x0, INTC_IPRI);
+
+	/* Ack all interrupt sources in the IRR0 register */
+	ctrl_outb(0x3f, INTC_IRR0);
+
+	/* Use IRQ0 - IRQ3 as active low interrupt lines i.e. disable
+	 * IRL mode.
+	 */
+	ctrl_outw(0x2aa, INTC_ICR1);
+
+	/* Now make IPR interrupts */
+	make_ipr_irq(TIMER2_IRQ, TIMER2_IPR_ADDR,
+			TIMER2_IPR_POS, TIMER2_PRIORITY);
+	make_ipr_irq(WDT_IRQ, WDT_IPR_ADDR, WDT_IPR_POS, WDT_PRIORITY);
+
+	/* SCIF0 */
+	make_ipr_irq(SCIF0_ERI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS,
+			SCIF0_PRIORITY);
+	make_ipr_irq(SCIF0_RXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS,
+			SCIF0_PRIORITY);
+	make_ipr_irq(SCIF0_BRI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS,
+			SCIF0_PRIORITY);
+	make_ipr_irq(SCIF0_TXI_IRQ, SCIF0_IPR_ADDR, SCIF0_IPR_POS,
+			SCIF0_PRIORITY);
+
+	/* DMAC-1 */
+	make_ipr_irq(DMTE0_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+	make_ipr_irq(DMTE1_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+	make_ipr_irq(DMTE2_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+	make_ipr_irq(DMTE3_IRQ, DMA_IPR_ADDR, DMA_IPR_POS, DMA_PRIORITY);
+
+	/* DMAC-2 */
+	make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+	make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+
+	/* IPSEC */
+	make_ipr_irq(IPSEC_IRQ, IPSEC_IPR_ADDR, IPSEC_IPR_POS, IPSEC_PRIORITY);
+
+	/* EDMAC */
+	make_ipr_irq(EDMAC0_IRQ, EDMAC0_IPR_ADDR, EDMAC0_IPR_POS,
+			EDMAC0_PRIORITY);
+	make_ipr_irq(EDMAC1_IRQ, EDMAC1_IPR_ADDR, EDMAC1_IPR_POS,
+			EDMAC1_PRIORITY);
+	make_ipr_irq(EDMAC2_IRQ, EDMAC2_IPR_ADDR, EDMAC2_IPR_POS,
+			EDMAC2_PRIORITY);
+
+	/* SIOF0 */
+	make_ipr_irq(SIOF0_ERI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS,
+			SIOF0_PRIORITY);
+	make_ipr_irq(SIOF0_TXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS,
+			SIOF0_PRIORITY);
+	make_ipr_irq(SIOF0_RXI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS,
+			SIOF0_PRIORITY);
+	make_ipr_irq(SIOF0_CCI_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS,
+			SIOF0_PRIORITY);
+
+	/* SIOF1 */
+	make_ipr_irq(SIOF1_ERI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS,
+			SIOF1_PRIORITY);
+	make_ipr_irq(SIOF1_TXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS,
+			SIOF1_PRIORITY);
+	make_ipr_irq(SIOF1_RXI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS,
+			SIOF1_PRIORITY);
+	make_ipr_irq(SIOF1_CCI_IRQ, SIOF1_IPR_ADDR, SIOF1_IPR_POS,
+			SIOF1_PRIORITY);
+
+	/* SLIC IRQ's */
+	make_ipr_irq(IRQ1_IRQ, IRQ1_IPR_ADDR, IRQ1_IPR_POS, IRQ1_PRIORITY);
+	make_ipr_irq(IRQ2_IRQ, IRQ2_IPR_ADDR, IRQ2_IPR_POS, IRQ2_PRIORITY);
+}
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_sh7710voipgw __initmv = {
+	.mv_name		= "SH7710 VoIP Gateway",
+	.mv_nr_irqs		= 104,
+	.mv_init_irq		= sh7710voipgw_init_irq,
+};
+ALIAS_MV(sh7710voipgw)
diff --git a/arch/sh/configs/sh7710voipgw_defconfig b/arch/sh/configs/sh7710voipgw_defconfig
new file mode 100644
index 000000000000..ec9a3034daa5
--- /dev/null
+++ b/arch/sh/configs/sh7710voipgw_defconfig
@@ -0,0 +1,913 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.17
+# Mon Aug  7 17:07:06 2006
+#
+CONFIG_SUPERH=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SWAP is not set
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_UID16=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SHMEM is not set
+CONFIG_SLAB=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
+CONFIG_OBSOLETE_INTERMODULE=y
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+CONFIG_IOSCHED_DEADLINE=y
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+CONFIG_DEFAULT_DEADLINE=y
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="deadline"
+
+#
+# System type
+#
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+CONFIG_SH_7710VOIPGW=y
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_SHMIN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH3=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7706 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+CONFIG_CPU_SUBTYPE_SH7710=y
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x00800000
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_SH_FPU_EMU is not set
+CONFIG_SH_DSP=y
+# CONFIG_SH_ADC is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ=32768000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+# CONFIG_SH_DMA is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+
+#
+# Kernel features
+#
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+# CONFIG_KEXEC is not set
+# CONFIG_PREEMPT is not set
+# CONFIG_SMP is not set
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+# CONFIG_CMDLINE_BOOL is not set
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options (EXPERIMENTAL)
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_PNP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+
+#
+# IP: Virtual Server Configuration
+#
+# CONFIG_IP_VS is not set
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+CONFIG_NETFILTER=y
+# CONFIG_NETFILTER_DEBUG is not set
+
+#
+# Core Netfilter Configuration
+#
+# CONFIG_NETFILTER_NETLINK is not set
+# CONFIG_NETFILTER_XTABLES is not set
+
+#
+# IP: Netfilter Configuration
+#
+CONFIG_IP_NF_CONNTRACK=y
+# CONFIG_IP_NF_CT_ACCT is not set
+# CONFIG_IP_NF_CONNTRACK_MARK is not set
+# CONFIG_IP_NF_CONNTRACK_EVENTS is not set
+# CONFIG_IP_NF_CT_PROTO_SCTP is not set
+CONFIG_IP_NF_FTP=m
+# CONFIG_IP_NF_IRC is not set
+# CONFIG_IP_NF_NETBIOS_NS is not set
+# CONFIG_IP_NF_TFTP is not set
+# CONFIG_IP_NF_AMANDA is not set
+CONFIG_IP_NF_PPTP=m
+# CONFIG_IP_NF_H323 is not set
+# CONFIG_IP_NF_QUEUE is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_CLK_JIFFIES=y
+# CONFIG_NET_SCH_CLK_GETTIMEOFDAY is not set
+# CONFIG_NET_SCH_CLK_CPU is not set
+
+#
+# Queueing/Scheduling
+#
+CONFIG_NET_SCH_CBQ=y
+# CONFIG_NET_SCH_HTB is not set
+# CONFIG_NET_SCH_HFSC is not set
+# CONFIG_NET_SCH_PRIO is not set
+# CONFIG_NET_SCH_RED is not set
+# CONFIG_NET_SCH_SFQ is not set
+# CONFIG_NET_SCH_TEQL is not set
+# CONFIG_NET_SCH_TBF is not set
+# CONFIG_NET_SCH_GRED is not set
+# CONFIG_NET_SCH_DSMARK is not set
+# CONFIG_NET_SCH_NETEM is not set
+CONFIG_NET_SCH_INGRESS=y
+
+#
+# Classification
+#
+CONFIG_NET_CLS=y
+CONFIG_NET_CLS_BASIC=y
+CONFIG_NET_CLS_TCINDEX=y
+CONFIG_NET_CLS_ROUTE4=y
+CONFIG_NET_CLS_ROUTE=y
+# CONFIG_NET_CLS_FW is not set
+CONFIG_NET_CLS_U32=y
+# CONFIG_CLS_U32_PERF is not set
+# CONFIG_CLS_U32_MARK is not set
+# CONFIG_NET_CLS_RSVP is not set
+# CONFIG_NET_CLS_RSVP6 is not set
+# CONFIG_NET_EMATCH is not set
+# CONFIG_NET_CLS_ACT is not set
+CONFIG_NET_CLS_POLICE=y
+# CONFIG_NET_CLS_IND is not set
+CONFIG_NET_ESTIMATOR=y
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+# CONFIG_MTD_PHYSMAP is not set
+# CONFIG_MTD_SOLUTIONENGINE is not set
+CONFIG_MTD_SH7710VOIPGW=y
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# OneNAND Flash Device Drivers
+#
+# CONFIG_MTD_ONENAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_STNIC is not set
+# CONFIG_SMC91X is not set
+# CONFIG_NE2000 is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+CONFIG_PHONE=y
+# CONFIG_PHONE_IXJ is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+# CONFIG_I2C is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+CONFIG_VIDEO_V4L2=y
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+# CONFIG_FB is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS_FS is not set
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=0
+CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_SUMMARY is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+# CONFIG_NFS_FS is not set
+# CONFIG_NFSD is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index b30e784b3a31..06df309f1a6d 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -28,3 +28,4 @@ R7780RP			SH_R7780RP
 R7780MP			SH_R7780MP
 TITAN			SH_TITAN
 SHMIN			SH_SHMIN
+7710VOIPGW		SH_7710VOIPGW
-- 
cgit v1.2.1


From bc8fb5d0471473f775378d09db712dcb8eeece75 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:09:34 +0900
Subject: sh: Solution Engine SH7343 board support.

This adds support for the SE7343 board.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig                  |  35 +-
 arch/sh/Makefile                 |   1 +
 arch/sh/boards/se/7343/Makefile  |   7 +
 arch/sh/boards/se/7343/io.c      | 275 +++++++++++
 arch/sh/boards/se/7343/irq.c     | 193 ++++++++
 arch/sh/boards/se/7343/led.c     |  46 ++
 arch/sh/boards/se/7343/setup.c   |  84 ++++
 arch/sh/configs/se7343_defconfig | 997 +++++++++++++++++++++++++++++++++++++++
 arch/sh/tools/mach-types         |   1 +
 include/asm-sh/irq.h             |   2 +
 include/asm-sh/se7343.h          |  82 ++++
 11 files changed, 1710 insertions(+), 13 deletions(-)
 create mode 100644 arch/sh/boards/se/7343/Makefile
 create mode 100644 arch/sh/boards/se/7343/io.c
 create mode 100644 arch/sh/boards/se/7343/irq.c
 create mode 100644 arch/sh/boards/se/7343/led.c
 create mode 100644 arch/sh/boards/se/7343/setup.c
 create mode 100644 arch/sh/configs/se7343_defconfig
 create mode 100644 include/asm-sh/se7343.h

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 2549bd82c57f..9e5272e84fd8 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -51,18 +51,23 @@ source "init/Kconfig"
 
 menu "System type"
 
+config SOLUTION_ENGINE
+	bool
+
 choice
 	prompt "SuperH system type"
 	default SH_UNKNOWN
 
 config SH_SOLUTION_ENGINE
 	bool "SolutionEngine"
+	select SOLUTION_ENGINE
 	help
 	  Select SolutionEngine if configuring for a Hitachi SH7709
 	  or SH7750 evaluation board.
 
 config SH_7751_SOLUTION_ENGINE
 	bool "SolutionEngine7751"
+	select SOLUTION_ENGINE
 	select CPU_SUBTYPE_SH7751
 	help
 	  Select 7751 SolutionEngine if configuring for a Hitachi SH7751
@@ -70,17 +75,27 @@ config SH_7751_SOLUTION_ENGINE
 
 config SH_7300_SOLUTION_ENGINE
 	bool "SolutionEngine7300"
+	select SOLUTION_ENGINE
 	select CPU_SUBTYPE_SH7300
 	help
-	  Select 7300 SolutionEngine if configuring for a Hitachi SH7300(SH-Mobile V)
-	  evaluation board.
+	  Select 7300 SolutionEngine if configuring for a Hitachi
+	  SH7300(SH-Mobile V) evaluation board.
+
+config SH_7343_SOLUTION_ENGINE
+	bool "SolutionEngine7343"
+	select SOLUTION_ENGINE
+	select CPU_SUBTYPE_SH7343
+	help
+	  Select 7343 SolutionEngine if configuring for a Hitachi
+	  SH7343 (SH-Mobile 3AS) evaluation board.
 
 config SH_73180_SOLUTION_ENGINE
        bool "SolutionEngine73180"
-       select CPU_SUBTYPE_SH73180
-       help
-         Select 73180 SolutionEngine if configuring for a Hitachi SH73180(SH-Mobile 3)
-         evaluation board.
+	select SOLUTION_ENGINE
+	select CPU_SUBTYPE_SH73180
+	help
+	  Select 73180 SolutionEngine if configuring for a Hitachi
+	  SH73180(SH-Mobile 3) evaluation board.
 
 config SH_7751_SYSTEMH
 	bool "SystemH7751R"
@@ -394,9 +409,7 @@ source "arch/sh/cchips/Kconfig"
 config HEARTBEAT
 	bool "Heartbeat LED"
 	depends on SH_MPC1211 || SH_SH03 || \
-		   SH_BIGSUR || \
-		   SH_7751_SOLUTION_ENGINE || SH_7300_SOLUTION_ENGINE || \
-		   SH_73180_SOLUTION_ENGINE || SH_SOLUTION_ENGINE || \
+		   SH_BIGSUR || SOLUTION_ENGINE || \
 		   SH_RTS7751R2D || SH_SH4202_MICRODEV || SH_LANDISK
 	help
 	  Use the power-on LED on your machine as a load meter.  The exact
@@ -431,10 +444,6 @@ config KEXEC
 	  support.  As of this writing the exact hardware interface is
 	  strongly in flux, so no good recommendation can be made.
 
-config PREEMPT
-	bool "Preemptible Kernel (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
-
 config SMP
 	bool "Symmetric multi-processing support"
 	---help---
diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index f0fc88e5f1ad..26d62ff51a64 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -89,6 +89,7 @@ core-$(CONFIG_SH_FPU_EMU)	+= arch/sh/math-emu/
 machdir-$(CONFIG_SH_SOLUTION_ENGINE)		:= se/770x
 machdir-$(CONFIG_SH_7751_SOLUTION_ENGINE)	:= se/7751
 machdir-$(CONFIG_SH_7300_SOLUTION_ENGINE)	:= se/7300
+machdir-$(CONFIG_SH_7343_SOLUTION_ENGINE)	:= se/7343
 machdir-$(CONFIG_SH_73180_SOLUTION_ENGINE)	:= se/73180
 machdir-$(CONFIG_SH_HP6XX)			:= hp6xx
 machdir-$(CONFIG_SH_EC3104)			:= ec3104
diff --git a/arch/sh/boards/se/7343/Makefile b/arch/sh/boards/se/7343/Makefile
new file mode 100644
index 000000000000..4291069c0b4f
--- /dev/null
+++ b/arch/sh/boards/se/7343/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for the 7343 SolutionEngine specific parts of the kernel
+#
+
+obj-y	 := setup.o io.o irq.o
+
+obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/se/7343/io.c b/arch/sh/boards/se/7343/io.c
new file mode 100644
index 000000000000..646661a146ad
--- /dev/null
+++ b/arch/sh/boards/se/7343/io.c
@@ -0,0 +1,275 @@
+/*
+ * arch/sh/boards/se/7343/io.c
+ *
+ * I/O routine for SH-Mobile3AS 7343 SolutionEngine.
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <asm/io.h>
+#include <asm/mach/se7343.h>
+
+#define badio(fn, a) panic("bad i/o operation %s for %08lx.", #fn, a)
+
+struct iop {
+	unsigned long start, end;
+	unsigned long base;
+	struct iop *(*check) (struct iop * p, unsigned long port);
+	unsigned char (*inb) (struct iop * p, unsigned long port);
+	unsigned short (*inw) (struct iop * p, unsigned long port);
+	void (*outb) (struct iop * p, unsigned char value, unsigned long port);
+	void (*outw) (struct iop * p, unsigned short value, unsigned long port);
+};
+
+struct iop *
+simple_check(struct iop *p, unsigned long port)
+{
+	static int count;
+
+	if (count < 100)
+		count++;
+
+	port &= 0xFFFF;
+
+	if ((p->start <= port) && (port <= p->end))
+		return p;
+	else
+		badio(check, port);
+}
+
+struct iop *
+ide_check(struct iop *p, unsigned long port)
+{
+	if (((0x1f0 <= port) && (port <= 0x1f7)) || (port == 0x3f7))
+		return p;
+	return NULL;
+}
+
+unsigned char
+simple_inb(struct iop *p, unsigned long port)
+{
+	return *(unsigned char *) (p->base + port);
+}
+
+unsigned short
+simple_inw(struct iop *p, unsigned long port)
+{
+	return *(unsigned short *) (p->base + port);
+}
+
+void
+simple_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	*(unsigned char *) (p->base + port) = value;
+}
+
+void
+simple_outw(struct iop *p, unsigned short value, unsigned long port)
+{
+	*(unsigned short *) (p->base + port) = value;
+}
+
+unsigned char
+pcc_inb(struct iop *p, unsigned long port)
+{
+	unsigned long addr = p->base + port + 0x40000;
+	unsigned long v;
+
+	if (port & 1)
+		addr += 0x00400000;
+	v = *(volatile unsigned char *) addr;
+	return v;
+}
+
+void
+pcc_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	unsigned long addr = p->base + port + 0x40000;
+
+	if (port & 1)
+		addr += 0x00400000;
+	*(volatile unsigned char *) addr = value;
+}
+
+unsigned char
+bad_inb(struct iop *p, unsigned long port)
+{
+	badio(inb, port);
+}
+
+void
+bad_outb(struct iop *p, unsigned char value, unsigned long port)
+{
+	badio(inw, port);
+}
+
+#ifdef CONFIG_SMC91X
+/* MSTLANEX01 LAN at 0xb400:0000 */
+static struct iop laniop = {
+	.start = 0x00,
+	.end = 0x0F,
+	.base = 0x04000000,
+	.check = simple_check,
+	.inb = simple_inb,
+	.inw = simple_inw,
+	.outb = simple_outb,
+	.outw = simple_outw,
+};
+#endif
+
+#ifdef CONFIG_NE2000
+/* NE2000 pc card NIC */
+static struct iop neiop = {
+	.start = 0x280,
+	.end = 0x29f,
+	.base = 0xb0600000 + 0x80,	/* soft 0x280 -> hard 0x300 */
+	.check = simple_check,
+	.inb = pcc_inb,
+	.inw = simple_inw,
+	.outb = pcc_outb,
+	.outw = simple_outw,
+};
+#endif
+
+#ifdef CONFIG_IDE
+/* CF in CF slot */
+static struct iop cfiop = {
+	.base = 0xb0600000,
+	.check = ide_check,
+	.inb = pcc_inb,
+	.inw = simple_inw,
+	.outb = pcc_outb,
+	.outw = simple_outw,
+};
+#endif
+
+static __inline__ struct iop *
+port2iop(unsigned long port)
+{
+	if (0) ;
+#if defined(CONFIG_SMC91X)
+	else if (laniop.check(&laniop, port))
+		return &laniop;
+#endif
+#if defined(CONFIG_NE2000)
+	else if (neiop.check(&neiop, port))
+		return &neiop;
+#endif
+#if defined(CONFIG_IDE)
+	else if (cfiop.check(&cfiop, port))
+		return &cfiop;
+#endif
+	else
+		return NULL;
+}
+
+static inline void
+delay(void)
+{
+	ctrl_inw(0xac000000);
+	ctrl_inw(0xac000000);
+}
+
+unsigned char
+sh7343se_inb(unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	return (p->inb) (p, port);
+}
+
+unsigned char
+sh7343se_inb_p(unsigned long port)
+{
+	unsigned char v = sh7343se_inb(port);
+	delay();
+	return v;
+}
+
+unsigned short
+sh7343se_inw(unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	return (p->inw) (p, port);
+}
+
+unsigned int
+sh7343se_inl(unsigned long port)
+{
+	badio(inl, port);
+}
+
+void
+sh7343se_outb(unsigned char value, unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	(p->outb) (p, value, port);
+}
+
+void
+sh7343se_outb_p(unsigned char value, unsigned long port)
+{
+	sh7343se_outb(value, port);
+	delay();
+}
+
+void
+sh7343se_outw(unsigned short value, unsigned long port)
+{
+	struct iop *p = port2iop(port);
+	(p->outw) (p, value, port);
+}
+
+void
+sh7343se_outl(unsigned int value, unsigned long port)
+{
+	badio(outl, port);
+}
+
+void
+sh7343se_insb(unsigned long port, void *addr, unsigned long count)
+{
+	unsigned char *a = addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		*a++ = (p->inb) (p, port);
+}
+
+void
+sh7343se_insw(unsigned long port, void *addr, unsigned long count)
+{
+	unsigned short *a = addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		*a++ = (p->inw) (p, port);
+}
+
+void
+sh7343se_insl(unsigned long port, void *addr, unsigned long count)
+{
+	badio(insl, port);
+}
+
+void
+sh7343se_outsb(unsigned long port, const void *addr, unsigned long count)
+{
+	unsigned char *a = (unsigned char *) addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		(p->outb) (p, *a++, port);
+}
+
+void
+sh7343se_outsw(unsigned long port, const void *addr, unsigned long count)
+{
+	unsigned short *a = (unsigned short *) addr;
+	struct iop *p = port2iop(port);
+	while (count--)
+		(p->outw) (p, *a++, port);
+}
+
+void
+sh7343se_outsl(unsigned long port, const void *addr, unsigned long count)
+{
+	badio(outsw, port);
+}
diff --git a/arch/sh/boards/se/7343/irq.c b/arch/sh/boards/se/7343/irq.c
new file mode 100644
index 000000000000..b41e3d4ea37c
--- /dev/null
+++ b/arch/sh/boards/se/7343/irq.c
@@ -0,0 +1,193 @@
+/*
+ * arch/sh/boards/se/7343/irq.c
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <asm/irq.h>
+#include <asm/io.h>
+#include <asm/mach/se7343.h>
+
+static void
+disable_intreq_irq(unsigned int irq)
+{
+	int bit = irq - OFFCHIP_IRQ_BASE;
+	u16 val;
+
+	val = ctrl_inw(PA_CPLD_IMSK);
+	val |= 1 << bit;
+	ctrl_outw(val, PA_CPLD_IMSK);
+}
+
+static void
+enable_intreq_irq(unsigned int irq)
+{
+	int bit = irq - OFFCHIP_IRQ_BASE;
+	u16 val;
+
+	val = ctrl_inw(PA_CPLD_IMSK);
+	val &= ~(1 << bit);
+	ctrl_outw(val, PA_CPLD_IMSK);
+}
+
+static void
+mask_and_ack_intreq_irq(unsigned int irq)
+{
+	disable_intreq_irq(irq);
+}
+
+static unsigned int
+startup_intreq_irq(unsigned int irq)
+{
+	enable_intreq_irq(irq);
+	return 0;
+}
+
+static void
+shutdown_intreq_irq(unsigned int irq)
+{
+	disable_intreq_irq(irq);
+}
+
+static void
+end_intreq_irq(unsigned int irq)
+{
+	if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+		enable_intreq_irq(irq);
+}
+
+static struct hw_interrupt_type intreq_irq_type = {
+	.typename = "FPGA-IRQ",
+	.startup = startup_intreq_irq,
+	.shutdown = shutdown_intreq_irq,
+	.enable = enable_intreq_irq,
+	.disable = disable_intreq_irq,
+	.ack = mask_and_ack_intreq_irq,
+	.end = end_intreq_irq
+};
+
+static void
+make_intreq_irq(unsigned int irq)
+{
+	disable_irq_nosync(irq);
+	irq_desc[irq].handler = &intreq_irq_type;
+	disable_intreq_irq(irq);
+}
+
+int
+shmse_irq_demux(int irq)
+{
+	int bit;
+	volatile u16 val;
+
+	if (irq == IRQ5_IRQ) {
+		/* Read status Register */
+		val = ctrl_inw(PA_CPLD_ST);
+		bit = ffs(val);
+		if (bit != 0)
+			return OFFCHIP_IRQ_BASE + bit - 1;
+	}
+	return irq;
+}
+
+/* IRQ5 is multiplexed between the following sources:
+ * 1. PC Card socket
+ * 2. Extension slot
+ * 3. USB Controller
+ * 4. Serial Controller
+ *
+ * We configure IRQ5 as a cascade IRQ.
+ */
+static struct irqaction irq5 = { no_action, 0, CPU_MASK_NONE, "IRQ5-cascade",
+				NULL, NULL};
+
+/*
+ * Initialize IRQ setting
+ */
+void __init
+init_7343se_IRQ(void)
+{
+	/* Setup Multiplexed interrupts */
+	ctrl_outw(8, PA_CPLD_MODESET);	/* Set all CPLD interrupts to active
+					 * low.
+					 */
+	/* Mask all CPLD controller interrupts */
+	ctrl_outw(0x0fff, PA_CPLD_IMSK);
+
+	/* PC Card interrupts */
+	make_intreq_irq(PC_IRQ0);
+	make_intreq_irq(PC_IRQ1);
+	make_intreq_irq(PC_IRQ2);
+	make_intreq_irq(PC_IRQ3);
+
+	/* Extension Slot Interrupts */
+	make_intreq_irq(EXT_IRQ0);
+	make_intreq_irq(EXT_IRQ1);
+	make_intreq_irq(EXT_IRQ2);
+	make_intreq_irq(EXT_IRQ3);
+
+	/* USB Controller interrupts */
+	make_intreq_irq(USB_IRQ0);
+	make_intreq_irq(USB_IRQ1);
+
+	/* Serial Controller interrupts */
+	make_intreq_irq(UART_IRQ0);
+	make_intreq_irq(UART_IRQ1);
+
+	/* Setup all external interrupts to be active low */
+	ctrl_outw(0xaaaa, INTC_ICR1);
+
+	make_ipr_irq(IRQ5_IRQ, IRQ5_IPR_ADDR+2, IRQ5_IPR_POS, IRQ5_PRIORITY);
+	setup_irq(IRQ5_IRQ, &irq5);
+	/* Set port control to use IRQ5 */
+	*(u16 *)0xA4050108 &= ~0xc;
+
+	make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY);
+	make_ipr_irq(VPU_IRQ, VPU_IPR_ADDR, VPU_IPR_POS, 8);
+
+	ctrl_outb(0x0f, INTC_IMCR5);	/* enable SCIF IRQ */
+
+	make_ipr_irq(DMTE0_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE1_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE2_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE3_IRQ, DMA1_IPR_ADDR, DMA1_IPR_POS, DMA1_PRIORITY);
+	make_ipr_irq(DMTE4_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+	make_ipr_irq(DMTE5_IRQ, DMA2_IPR_ADDR, DMA2_IPR_POS, DMA2_PRIORITY);
+
+	/* I2C block */
+	make_ipr_irq(IIC0_ALI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY);
+	make_ipr_irq(IIC0_TACKI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS,
+		     IIC0_PRIORITY);
+	make_ipr_irq(IIC0_WAITI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS,
+		     IIC0_PRIORITY);
+	make_ipr_irq(IIC0_DTEI_IRQ, IIC0_IPR_ADDR, IIC0_IPR_POS, IIC0_PRIORITY);
+
+	make_ipr_irq(IIC1_ALI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY);
+	make_ipr_irq(IIC1_TACKI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS,
+		     IIC1_PRIORITY);
+	make_ipr_irq(IIC1_WAITI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS,
+		     IIC1_PRIORITY);
+	make_ipr_irq(IIC1_DTEI_IRQ, IIC1_IPR_ADDR, IIC1_IPR_POS, IIC1_PRIORITY);
+
+	/* SIOF */
+	make_ipr_irq(SIOF0_IRQ, SIOF0_IPR_ADDR, SIOF0_IPR_POS, SIOF0_PRIORITY);
+
+	/* SIU */
+	make_ipr_irq(SIU_IRQ, SIU_IPR_ADDR, SIU_IPR_POS, SIU_PRIORITY);
+
+	/* VIO interrupt */
+	make_ipr_irq(CEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+	make_ipr_irq(BEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+	make_ipr_irq(VEU_IRQ, VIO_IPR_ADDR, VIO_IPR_POS, VIO_PRIORITY);
+
+	/*MFI interrupt*/
+
+	make_ipr_irq(MFI_IRQ, MFI_IPR_ADDR, MFI_IPR_POS, MFI_PRIORITY);
+
+	/* LCD controller */
+	make_ipr_irq(LCDC_IRQ, LCDC_IPR_ADDR, LCDC_IPR_POS, LCDC_PRIORITY);
+	ctrl_outw(0x2000, PA_MRSHPC + 0x0c);	/* mrshpc irq enable */
+}
diff --git a/arch/sh/boards/se/7343/led.c b/arch/sh/boards/se/7343/led.c
new file mode 100644
index 000000000000..6a439cf83e46
--- /dev/null
+++ b/arch/sh/boards/se/7343/led.c
@@ -0,0 +1,46 @@
+/*
+ * arch/sh/boards/se/7343/led.c
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/sched.h>
+#include <asm/mach/se7343.h>
+
+/* Cycle the LED's in the clasic Knightrider/Sun pattern */
+void heartbeat_7343se(void)
+{
+	static unsigned int cnt = 0, period = 0;
+	volatile unsigned short *p = (volatile unsigned short *) PA_LED;
+	static unsigned bit = 0, up = 1;
+
+	cnt += 1;
+	if (cnt < period) {
+		return;
+	}
+
+	cnt = 0;
+
+	/* Go through the points (roughly!):
+	 * f(0)=10, f(1)=16, f(2)=20, f(5)=35,f(inf)->110
+	 */
+	period = 110 - ((300 << FSHIFT) / ((avenrun[0] / 5) + (3 << FSHIFT)));
+
+	if (up) {
+		if (bit == 7) {
+			bit--;
+			up = 0;
+		} else {
+			bit++;
+		}
+	} else {
+		if (bit == 0) {
+			bit++;
+			up = 1;
+		} else {
+			bit--;
+		}
+	}
+	*p = 1 << (bit + LED_SHIFT);
+
+}
diff --git a/arch/sh/boards/se/7343/setup.c b/arch/sh/boards/se/7343/setup.c
new file mode 100644
index 000000000000..787322291fb3
--- /dev/null
+++ b/arch/sh/boards/se/7343/setup.c
@@ -0,0 +1,84 @@
+#include <linux/config.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <asm/machvec.h>
+#include <asm/mach/se7343.h>
+#include <asm/irq.h>
+
+void heartbeat_7343se(void);
+void init_7343se_IRQ(void);
+
+static struct resource smc91x_resources[] = {
+	[0] = {
+		.start	= 0x10000000,
+		.end	= 0x1000000F,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		/*
+		 * shared with other devices via externel
+		 * interrupt controller in FPGA...
+		 */
+		.start	= EXT_IRQ2,
+		.end	= EXT_IRQ2,
+		.flags	= IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device smc91x_device = {
+	.name		= "smc91x",
+	.id		= 0,
+	.num_resources	= ARRAY_SIZE(smc91x_resources),
+	.resource	= smc91x_resources,
+};
+
+static struct platform_device *smc91x_platform_devices[] __initdata = {
+	&smc91x_device,
+};
+
+static int __init sh7343se_devices_setup(void)
+{
+	return platform_add_devices(smc91x_platform_devices,
+				    ARRAY_SIZE(smc91x_platform_devices));
+}
+
+static void __init sh7343se_setup(char **cmdline_p)
+{
+	device_initcall(sh7343se_devices_setup);
+}
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_7343se __initmv = {
+	.mv_name = "SolutionEngine 7343",
+	.mv_setup = sh7343se_setup,
+	.mv_nr_irqs = 108,
+	.mv_inb = sh7343se_inb,
+	.mv_inw = sh7343se_inw,
+	.mv_inl = sh7343se_inl,
+	.mv_outb = sh7343se_outb,
+	.mv_outw = sh7343se_outw,
+	.mv_outl = sh7343se_outl,
+
+	.mv_inb_p = sh7343se_inb_p,
+	.mv_inw_p = sh7343se_inw,
+	.mv_inl_p = sh7343se_inl,
+	.mv_outb_p = sh7343se_outb_p,
+	.mv_outw_p = sh7343se_outw,
+	.mv_outl_p = sh7343se_outl,
+
+	.mv_insb = sh7343se_insb,
+	.mv_insw = sh7343se_insw,
+	.mv_insl = sh7343se_insl,
+	.mv_outsb = sh7343se_outsb,
+	.mv_outsw = sh7343se_outsw,
+	.mv_outsl = sh7343se_outsl,
+
+	.mv_init_irq = init_7343se_IRQ,
+	.mv_irq_demux = shmse_irq_demux,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_7343se,
+#endif
+};
+ALIAS_MV(7343se)
diff --git a/arch/sh/configs/se7343_defconfig b/arch/sh/configs/se7343_defconfig
new file mode 100644
index 000000000000..948e507b52be
--- /dev/null
+++ b/arch/sh/configs/se7343_defconfig
@@ -0,0 +1,997 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.17
+# Mon Aug  7 20:14:44 2006
+#
+CONFIG_SUPERH=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+
+#
+# Code maturity level options
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+
+#
+# General setup
+#
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SWAP is not set
+CONFIG_SYSVIPC=y
+CONFIG_POSIX_MQUEUE=y
+# CONFIG_BSD_PROCESS_ACCT is not set
+CONFIG_SYSCTL=y
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+# CONFIG_RELAY is not set
+CONFIG_INITRAMFS_SOURCE=""
+CONFIG_UID16=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_EMBEDDED=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SHMEM is not set
+CONFIG_SLAB=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
+CONFIG_OBSOLETE_INTERMODULE=y
+
+#
+# Loadable module support
+#
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+CONFIG_MODULE_FORCE_UNLOAD=y
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+
+#
+# Block layer
+#
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+CONFIG_IOSCHED_DEADLINE=y
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+CONFIG_DEFAULT_DEADLINE=y
+# CONFIG_DEFAULT_CFQ is not set
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="deadline"
+
+#
+# System type
+#
+CONFIG_SOLUTION_ENGINE=y
+# CONFIG_SH_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SOLUTION_ENGINE is not set
+# CONFIG_SH_7300_SOLUTION_ENGINE is not set
+CONFIG_SH_7343_SOLUTION_ENGINE=y
+# CONFIG_SH_73180_SOLUTION_ENGINE is not set
+# CONFIG_SH_7751_SYSTEMH is not set
+# CONFIG_SH_HP6XX is not set
+# CONFIG_SH_EC3104 is not set
+# CONFIG_SH_SATURN is not set
+# CONFIG_SH_DREAMCAST is not set
+# CONFIG_SH_BIGSUR is not set
+# CONFIG_SH_MPC1211 is not set
+# CONFIG_SH_SH03 is not set
+# CONFIG_SH_SECUREEDGE5410 is not set
+# CONFIG_SH_HS7751RVOIP is not set
+# CONFIG_SH_7710VOIPGW is not set
+# CONFIG_SH_RTS7751R2D is not set
+# CONFIG_SH_R7780RP is not set
+# CONFIG_SH_EDOSK7705 is not set
+# CONFIG_SH_SH4202_MICRODEV is not set
+# CONFIG_SH_LANDISK is not set
+# CONFIG_SH_TITAN is not set
+# CONFIG_SH_SHMIN is not set
+# CONFIG_SH_UNKNOWN is not set
+
+#
+# Processor selection
+#
+CONFIG_CPU_SH4=y
+CONFIG_CPU_SH4A=y
+
+#
+# SH-2 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7604 is not set
+
+#
+# SH-3 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7300 is not set
+# CONFIG_CPU_SUBTYPE_SH7705 is not set
+# CONFIG_CPU_SUBTYPE_SH7706 is not set
+# CONFIG_CPU_SUBTYPE_SH7707 is not set
+# CONFIG_CPU_SUBTYPE_SH7708 is not set
+# CONFIG_CPU_SUBTYPE_SH7709 is not set
+# CONFIG_CPU_SUBTYPE_SH7710 is not set
+
+#
+# SH-4 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH7750 is not set
+# CONFIG_CPU_SUBTYPE_SH7091 is not set
+# CONFIG_CPU_SUBTYPE_SH7750R is not set
+# CONFIG_CPU_SUBTYPE_SH7750S is not set
+# CONFIG_CPU_SUBTYPE_SH7751 is not set
+# CONFIG_CPU_SUBTYPE_SH7751R is not set
+# CONFIG_CPU_SUBTYPE_SH7760 is not set
+# CONFIG_CPU_SUBTYPE_SH4_202 is not set
+
+#
+# ST40 Processor Support
+#
+# CONFIG_CPU_SUBTYPE_ST40STB1 is not set
+# CONFIG_CPU_SUBTYPE_ST40GX1 is not set
+
+#
+# SH-4A Processor Support
+#
+# CONFIG_CPU_SUBTYPE_SH73180 is not set
+CONFIG_CPU_SUBTYPE_SH7343=y
+# CONFIG_CPU_SUBTYPE_SH7770 is not set
+# CONFIG_CPU_SUBTYPE_SH7780 is not set
+
+#
+# Memory management options
+#
+CONFIG_MMU=y
+CONFIG_PAGE_OFFSET=0x80000000
+CONFIG_MEMORY_START=0x0c000000
+CONFIG_MEMORY_SIZE=0x01000000
+CONFIG_32BIT=y
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+
+#
+# Cache configuration
+#
+# CONFIG_SH_DIRECT_MAPPED is not set
+# CONFIG_SH_WRITETHROUGH is not set
+# CONFIG_SH_OCRAM is not set
+
+#
+# Processor features
+#
+CONFIG_CPU_LITTLE_ENDIAN=y
+# CONFIG_SH_FPU is not set
+# CONFIG_SH_FPU_EMU is not set
+CONFIG_SH_DSP=y
+# CONFIG_SH_STORE_QUEUES is not set
+CONFIG_CPU_HAS_INTEVT=y
+CONFIG_CPU_HAS_SR_RB=y
+
+#
+# Timer support
+#
+CONFIG_SH_TMU=y
+CONFIG_SH_PCLK_FREQ=27000000
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# DMA support
+#
+# CONFIG_SH_DMA is not set
+
+#
+# Companion Chips
+#
+# CONFIG_HD6446X_SERIES is not set
+CONFIG_HEARTBEAT=y
+
+#
+# Kernel features
+#
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+# CONFIG_KEXEC is not set
+# CONFIG_SMP is not set
+CONFIG_PREEMPT_NONE=y
+# CONFIG_PREEMPT_VOLUNTARY is not set
+# CONFIG_PREEMPT is not set
+
+#
+# Boot options
+#
+CONFIG_ZERO_PAGE_OFFSET=0x00001000
+CONFIG_BOOT_LINK_OFFSET=0x00800000
+# CONFIG_UBC_WAKEUP is not set
+# CONFIG_CMDLINE_BOOL is not set
+
+#
+# Bus options
+#
+# CONFIG_PCI is not set
+
+#
+# PCCARD (PCMCIA/CardBus) support
+#
+# CONFIG_PCCARD is not set
+
+#
+# PCI Hotplug Support
+#
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF=y
+# CONFIG_BINFMT_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options (EXPERIMENTAL)
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+# CONFIG_NETDEBUG is not set
+CONFIG_PACKET=y
+CONFIG_PACKET_MMAP=y
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_BIC=y
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETFILTER is not set
+
+#
+# DCCP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_DCCP is not set
+
+#
+# SCTP Configuration (EXPERIMENTAL)
+#
+# CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_NET_DIVERT is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+
+#
+# QoS and/or fair queueing
+#
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_IEEE80211 is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+CONFIG_FW_LOADER=y
+
+#
+# Connector - unified userspace <-> kernelspace linker
+#
+# CONFIG_CONNECTOR is not set
+
+#
+# Memory Technology Devices (MTD)
+#
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+CONFIG_MTD_CONCAT=y
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+# CONFIG_MTD_OBSOLETE_CHIPS is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+# CONFIG_MTD_PHYSMAP is not set
+# CONFIG_MTD_SOLUTIONENGINE is not set
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+
+#
+# NAND Flash Device Drivers
+#
+# CONFIG_MTD_NAND is not set
+
+#
+# OneNAND Flash Device Drivers
+#
+# CONFIG_MTD_ONENAND is not set
+
+#
+# Parallel port support
+#
+# CONFIG_PARPORT is not set
+
+#
+# Plug and Play support
+#
+
+#
+# Block devices
+#
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_RAM is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+
+#
+# ATA/ATAPI/MFM/RLL support
+#
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+
+#
+# Multi-device support (RAID and LVM)
+#
+# CONFIG_MD is not set
+
+#
+# Fusion MPT device support
+#
+# CONFIG_FUSION is not set
+
+#
+# IEEE 1394 (FireWire) support
+#
+
+#
+# I2O device support
+#
+
+#
+# Network device support
+#
+CONFIG_NETDEVICES=y
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+
+#
+# PHY device support
+#
+# CONFIG_PHYLIB is not set
+
+#
+# Ethernet (10 or 100Mbit)
+#
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+# CONFIG_STNIC is not set
+CONFIG_SMC91X=y
+# CONFIG_NE2000 is not set
+
+#
+# Ethernet (1000 Mbit)
+#
+
+#
+# Ethernet (10000 Mbit)
+#
+
+#
+# Token Ring devices
+#
+
+#
+# Wireless LAN (non-hamradio)
+#
+# CONFIG_NET_RADIO is not set
+
+#
+# Wan interfaces
+#
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+
+#
+# ISDN subsystem
+#
+# CONFIG_ISDN is not set
+
+#
+# Telephony Support
+#
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_TSDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_SH_SCI=y
+CONFIG_SERIAL_SH_SCI_NR_UARTS=2
+CONFIG_SERIAL_SH_SCI_CONSOLE=y
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+
+#
+# IPMI
+#
+# CONFIG_IPMI_HANDLER is not set
+
+#
+# Watchdog Cards
+#
+# CONFIG_WATCHDOG is not set
+# CONFIG_RTC is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_DTLK is not set
+# CONFIG_R3964 is not set
+
+#
+# Ftape, the floppy tape device driver
+#
+# CONFIG_RAW_DRIVER is not set
+
+#
+# TPM devices
+#
+# CONFIG_TCG_TPM is not set
+# CONFIG_TELCLOCK is not set
+
+#
+# I2C support
+#
+CONFIG_I2C=y
+CONFIG_I2C_CHARDEV=y
+
+#
+# I2C Algorithms
+#
+# CONFIG_I2C_ALGOBIT is not set
+# CONFIG_I2C_ALGOPCF is not set
+# CONFIG_I2C_ALGOPCA is not set
+
+#
+# I2C Hardware Bus support
+#
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_STUB is not set
+# CONFIG_I2C_PCA_ISA is not set
+CONFIG_I2C_SH7343=y
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
+# CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCA9539 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
+#
+# Dallas's 1-wire bus
+#
+# CONFIG_W1 is not set
+
+#
+# Hardware Monitoring support
+#
+# CONFIG_HWMON is not set
+# CONFIG_HWMON_VID is not set
+
+#
+# Misc devices
+#
+
+#
+# Multimedia devices
+#
+CONFIG_VIDEO_DEV=y
+CONFIG_VIDEO_V4L1=y
+CONFIG_VIDEO_V4L1_COMPAT=y
+CONFIG_VIDEO_V4L2=y
+
+#
+# Video Capture Adapters
+#
+
+#
+# Video Capture Adapters
+#
+# CONFIG_VIDEO_ADV_DEBUG is not set
+# CONFIG_VIDEO_VIVI is not set
+# CONFIG_VIDEO_CPIA is not set
+# CONFIG_VIDEO_SAA5246A is not set
+# CONFIG_VIDEO_SAA5249 is not set
+# CONFIG_TUNER_3036 is not set
+# CONFIG_VIDEO_OVCAMCHIP is not set
+
+#
+# Encoders and Decoders
+#
+# CONFIG_VIDEO_MSP3400 is not set
+# CONFIG_VIDEO_CS53L32A is not set
+# CONFIG_VIDEO_WM8775 is not set
+# CONFIG_VIDEO_WM8739 is not set
+# CONFIG_VIDEO_CX25840 is not set
+# CONFIG_VIDEO_SAA711X is not set
+# CONFIG_VIDEO_SAA7127 is not set
+# CONFIG_VIDEO_UPD64031A is not set
+# CONFIG_VIDEO_UPD64083 is not set
+
+#
+# Radio Adapters
+#
+# CONFIG_RADIO_MAESTRO is not set
+
+#
+# Digital Video Broadcasting Devices
+#
+# CONFIG_DVB is not set
+
+#
+# Graphics support
+#
+CONFIG_FB=y
+# CONFIG_FB_CFB_FILLRECT is not set
+# CONFIG_FB_CFB_COPYAREA is not set
+# CONFIG_FB_CFB_IMAGEBLIT is not set
+# CONFIG_FB_MACMODES is not set
+CONFIG_FB_FIRMWARE_EDID=y
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+# CONFIG_FB_EPSON1355 is not set
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_VIRTUAL is not set
+
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+# CONFIG_FRAMEBUFFER_CONSOLE is not set
+
+#
+# Logo configuration
+#
+# CONFIG_LOGO is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Sound
+#
+CONFIG_SOUND=y
+
+#
+# Advanced Linux Sound Architecture
+#
+CONFIG_SND=y
+CONFIG_SND_TIMER=y
+CONFIG_SND_PCM=y
+CONFIG_SND_SEQUENCER=y
+# CONFIG_SND_SEQ_DUMMY is not set
+CONFIG_SND_OSSEMUL=y
+# CONFIG_SND_MIXER_OSS is not set
+CONFIG_SND_PCM_OSS=y
+CONFIG_SND_PCM_OSS_PLUGINS=y
+# CONFIG_SND_SEQUENCER_OSS is not set
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
+CONFIG_SND_VERBOSE_PROCFS=y
+# CONFIG_SND_VERBOSE_PRINTK is not set
+# CONFIG_SND_DEBUG is not set
+
+#
+# Generic devices
+#
+# CONFIG_SND_DUMMY is not set
+# CONFIG_SND_VIRMIDI is not set
+# CONFIG_SND_MTPAV is not set
+# CONFIG_SND_SERIAL_U16550 is not set
+# CONFIG_SND_MPU401 is not set
+
+#
+# SuperH devices
+#
+CONFIG_SH7343_SIU=m
+CONFIG_AK4537_CODEC=y
+
+#
+# Open Sound System
+#
+# CONFIG_SOUND_PRIME is not set
+
+#
+# USB support
+#
+# CONFIG_USB_ARCH_HAS_HCD is not set
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+
+#
+# MMC/SD Card support
+#
+# CONFIG_MMC is not set
+
+#
+# LED devices
+#
+# CONFIG_NEW_LEDS is not set
+
+#
+# LED drivers
+#
+
+#
+# LED Triggers
+#
+
+#
+# InfiniBand support
+#
+
+#
+# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
+#
+
+#
+# Real Time Clock
+#
+# CONFIG_RTC_CLASS is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+# CONFIG_PROC_KCORE is not set
+CONFIG_SYSFS=y
+CONFIG_TMPFS=y
+# CONFIG_HUGETLBFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+CONFIG_RAMFS=y
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS_FS is not set
+CONFIG_JFFS2_FS=y
+CONFIG_JFFS2_FS_DEBUG=0
+CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_SUMMARY is not set
+# CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
+CONFIG_JFFS2_ZLIB=y
+CONFIG_JFFS2_RTIME=y
+# CONFIG_JFFS2_RUBIN is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Network File Systems
+#
+CONFIG_NFS_FS=y
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+CONFIG_NFSD=y
+# CONFIG_NFSD_V3 is not set
+# CONFIG_NFSD_TCP is not set
+CONFIG_ROOT_NFS=y
+CONFIG_LOCKD=y
+CONFIG_LOCKD_V4=y
+CONFIG_EXPORTFS=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=y
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+# CONFIG_9P_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+
+#
+# Native Language Support
+#
+# CONFIG_NLS is not set
+
+#
+# Profiling support
+#
+# CONFIG_PROFILING is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_DEBUG_KERNEL is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_EARLY_SCIF_CONSOLE is not set
+# CONFIG_KGDB is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+
+#
+# Cryptographic options
+#
+# CONFIG_CRYPTO is not set
+
+#
+# Hardware crypto devices
+#
+
+#
+# Library routines
+#
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+CONFIG_CRC32=y
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
+CONFIG_ZLIB_DEFLATE=y
diff --git a/arch/sh/tools/mach-types b/arch/sh/tools/mach-types
index 06df309f1a6d..ac57638977ee 100644
--- a/arch/sh/tools/mach-types
+++ b/arch/sh/tools/mach-types
@@ -8,6 +8,7 @@
 SE			SH_SOLUTION_ENGINE
 7751SE			SH_7751_SOLUTION_ENGINE		
 7300SE			SH_7300_SOLUTION_ENGINE
+7343SE			SH_7343_SOLUTION_ENGINE
 73180SE			SH_73180_SOLUTION_ENGINE
 7751SYSTEMH		SH_7751_SYSTEMH
 HP6XX			SH_HP6XX
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index 00886f9adb4d..dd05e102fc0e 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -316,6 +316,8 @@
 # define OFFCHIP_NR_IRQS 4
 #elif defined(CONFIG_SH_R7780RP)
 # define OFFCHIP_NR_IRQS 16
+#elif defined(CONFIG_SH_7343_SOLUTION_ENGINE)
+# define OFFCHIP_NR_IRQS 12
 #elif defined(CONFIG_SH_UNKNOWN)
 # define OFFCHIP_NR_IRQS 16	/* Must also be last */
 #else
diff --git a/include/asm-sh/se7343.h b/include/asm-sh/se7343.h
new file mode 100644
index 000000000000..e7914a54aa96
--- /dev/null
+++ b/include/asm-sh/se7343.h
@@ -0,0 +1,82 @@
+#ifndef __ASM_SH_HITACHI_SE7343_H
+#define __ASM_SH_HITACHI_SE7343_H
+
+/*
+ * include/asm-sh/se/se7343.h
+ *
+ * Copyright (C) 2003 Takashi Kusuda <kusuda-takashi@hitachi-ul.co.jp>
+ *
+ * SH-Mobile SolutionEngine 7343 support
+ */
+
+/* Box specific addresses.  */
+
+/* Area 0 */
+#define PA_ROM		0x00000000	/* EPROM */
+#define PA_ROM_SIZE	0x00400000	/* EPROM size 4M byte(Actually 2MB) */
+#define PA_FROM		0x00400000	/* Flash ROM */
+#define PA_FROM_SIZE	0x00400000	/* Flash size 4M byte */
+#define PA_SRAM		0x00800000	/* SRAM */
+#define PA_FROM_SIZE	0x00400000	/* SRAM size 4M byte */
+/* Area 1 */
+#define PA_EXT1		0x04000000
+#define PA_EXT1_SIZE	0x04000000
+/* Area 2 */
+#define PA_EXT2		0x08000000
+#define PA_EXT2_SIZE	0x04000000
+/* Area 3 */
+#define PA_SDRAM	0x0c000000
+#define PA_SDRAM_SIZE	0x04000000
+/* Area 4 */
+#define PA_PCIC		0x10000000	/* MR-SHPC-01 PCMCIA */
+#define PA_MRSHPC       0xb03fffe0      /* MR-SHPC-01 PCMCIA controller */
+#define PA_MRSHPC_MW1   0xb0400000      /* MR-SHPC-01 memory window base */
+#define PA_MRSHPC_MW2   0xb0500000      /* MR-SHPC-01 attribute window base */
+#define PA_MRSHPC_IO    0xb0600000      /* MR-SHPC-01 I/O window base */
+#define MRSHPC_OPTION   (PA_MRSHPC + 6)
+#define MRSHPC_CSR      (PA_MRSHPC + 8)
+#define MRSHPC_ISR      (PA_MRSHPC + 10)
+#define MRSHPC_ICR      (PA_MRSHPC + 12)
+#define MRSHPC_CPWCR    (PA_MRSHPC + 14)
+#define MRSHPC_MW0CR1   (PA_MRSHPC + 16)
+#define MRSHPC_MW1CR1   (PA_MRSHPC + 18)
+#define MRSHPC_IOWCR1   (PA_MRSHPC + 20)
+#define MRSHPC_MW0CR2   (PA_MRSHPC + 22)
+#define MRSHPC_MW1CR2   (PA_MRSHPC + 24)
+#define MRSHPC_IOWCR2   (PA_MRSHPC + 26)
+#define MRSHPC_CDCR     (PA_MRSHPC + 28)
+#define MRSHPC_PCIC_INFO (PA_MRSHPC + 30)
+#define PA_LED		0xb0C00000	/* LED */
+#define LED_SHIFT       0
+#define PA_DIPSW	0xb0900000	/* Dip switch 31 */
+#define PA_CPLD_MODESET	0xb1400004	/* CPLD Mode set register */
+#define PA_CPLD_ST	0xb1400008	/* CPLD Interrupt status register */
+#define PA_CPLD_IMSK	0xb140000a	/* CPLD Interrupt mask register */
+/* Area 5 */
+#define PA_EXT5		0x14000000
+#define PA_EXT5_SIZE	0x04000000
+/* Area 6 */
+#define PA_LCD1		0xb8000000
+#define PA_LCD2		0xb8800000
+
+#define __IO_PREFIX	sh7343se
+#include <asm/io_generic.h>
+
+/* External Multiplexed interrupts */
+#define PC_IRQ0		OFFCHIP_IRQ_BASE
+#define PC_IRQ1		(PC_IRQ0 + 1)
+#define PC_IRQ2		(PC_IRQ1 + 1)
+#define PC_IRQ3		(PC_IRQ2 + 1)
+
+#define EXT_IRQ0	(PC_IRQ3 + 1)
+#define EXT_IRQ1	(EXT_IRQ0 + 1)
+#define EXT_IRQ2	(EXT_IRQ1 + 1)
+#define EXT_IRQ3	(EXT_IRQ2 + 1)
+
+#define USB_IRQ0	(EXT_IRQ3 + 1)
+#define USB_IRQ1	(USB_IRQ0 + 1)
+
+#define UART_IRQ0	(USB_IRQ1 + 1)
+#define UART_IRQ1	(UART_IRQ0 + 1)
+
+#endif  /* __ASM_SH_HITACHI_SE7343_H */
-- 
cgit v1.2.1


From 2c7834a6f15fe6c50ed4766f1bb6f9183b9e2740 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:17:31 +0900
Subject: sh: machvec rework.

Some more machvec overhauling and setup code cleanup. Kill off
get_system_type() and platform_setup(), we can do these both
through the machvec. While we're add it, kill off more useless
mach.c's and drop some legacy cruft from setup.c.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/boards/bigsur/setup.c              |  47 +++++-------
 arch/sh/boards/dreamcast/setup.c           |  34 ++++-----
 arch/sh/boards/ec3104/setup.c              |  49 +++++--------
 arch/sh/boards/hp6xx/Makefile              |   4 +-
 arch/sh/boards/hp6xx/setup.c               |  47 ++++++++++--
 arch/sh/boards/landisk/setup.c             |  92 ++++++++++++-----------
 arch/sh/boards/mpc1211/setup.c             |  42 +++++------
 arch/sh/boards/renesas/edosk7705/setup.c   |  29 ++------
 arch/sh/boards/renesas/hs7751rvoip/setup.c |  96 +++++++++++-------------
 arch/sh/boards/renesas/r7780rp/setup.c     |  83 +++++++++++----------
 arch/sh/boards/renesas/rts7751r2d/Makefile |   4 +-
 arch/sh/boards/renesas/rts7751r2d/mach.c   |  66 -----------------
 arch/sh/boards/renesas/rts7751r2d/setup.c  |  62 ++++++++++++++--
 arch/sh/boards/renesas/systemh/setup.c     |  16 +---
 arch/sh/boards/saturn/setup.c              |  14 +---
 arch/sh/boards/se/7300/setup.c             |  17 +----
 arch/sh/boards/se/73180/setup.c            |  17 +----
 arch/sh/boards/se/770x/Makefile            |   3 +-
 arch/sh/boards/se/770x/mach.c              |  62 ----------------
 arch/sh/boards/se/770x/setup.c             |  61 ++++++++++++----
 arch/sh/boards/se/7751/Makefile            |   3 +-
 arch/sh/boards/se/7751/mach.c              |  48 ------------
 arch/sh/boards/se/7751/setup.c             | 103 ++++++++++----------------
 arch/sh/boards/sh03/setup.c                |  26 +++----
 arch/sh/boards/shmin/setup.c               |  10 +--
 arch/sh/boards/snapgear/setup.c            |  22 ++----
 arch/sh/boards/superh/microdev/setup.c     | 113 +++++++++++++----------------
 arch/sh/boards/titan/setup.c               |  13 +---
 arch/sh/boards/unknown/setup.c             |  13 +---
 arch/sh/kernel/setup.c                     |  54 +++-----------
 include/asm-sh/machvec.h                   |   7 +-
 31 files changed, 481 insertions(+), 776 deletions(-)
 delete mode 100644 arch/sh/boards/renesas/rts7751r2d/mach.c
 delete mode 100644 arch/sh/boards/se/770x/mach.c
 delete mode 100644 arch/sh/boards/se/7751/mach.c

diff --git a/arch/sh/boards/bigsur/setup.c b/arch/sh/boards/bigsur/setup.c
index dfeede9da50f..9711c20fc9e4 100644
--- a/arch/sh/boards/bigsur/setup.c
+++ b/arch/sh/boards/bigsur/setup.c
@@ -41,31 +41,7 @@
 //		Big Sur Init Routines	
 /*===========================================================*/
 
-const char *get_system_type(void)
-{
-	return "Big Sur";
-}
-
-/*
- * The Machine Vector
- */
-extern void heartbeat_bigsur(void);
-extern void init_bigsur_IRQ(void);
-
-struct sh_machine_vector mv_bigsur __initmv = {
-	.mv_nr_irqs		= NR_IRQS,     // Defined in <asm/irq.h>
-
-	.mv_isa_port2addr	= bigsur_isa_port2addr,
-	.mv_irq_demux       	= bigsur_irq_demux,
-
-	.mv_init_irq		= init_bigsur_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_bigsur,
-#endif
-};
-ALIAS_MV(bigsur)
-
-int __init platform_setup(void)
+static void __init bigsur_setup(char **cmdline_p)
 {
 	/* Mask all 2nd level IRQ's */
 	outb(-1,BIGSUR_IMR0);
@@ -89,7 +65,24 @@ int __init platform_setup(void)
 	outw(1, BIGSUR_ETHR+0xe);
 	/* set the IO port to BIGSUR_ETHER_IOPORT */
 	outw(BIGSUR_ETHER_IOPORT<<3, BIGSUR_ETHR+0x2);
-
-	return 0;
 }
 
+/*
+ * The Machine Vector
+ */
+extern void heartbeat_bigsur(void);
+extern void init_bigsur_IRQ(void);
+
+struct sh_machine_vector mv_bigsur __initmv = {
+	.mv_name		= "Big Sur",
+	.mv_setup		= bigsur_setup,
+
+	.mv_isa_port2addr	= bigsur_isa_port2addr,
+	.mv_irq_demux       	= bigsur_irq_demux,
+
+	.mv_init_irq		= init_bigsur_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_bigsur,
+#endif
+};
+ALIAS_MV(bigsur)
diff --git a/arch/sh/boards/dreamcast/setup.c b/arch/sh/boards/dreamcast/setup.c
index a00ac94beb05..f13017eeeb27 100644
--- a/arch/sh/boards/dreamcast/setup.c
+++ b/arch/sh/boards/dreamcast/setup.c
@@ -22,7 +22,6 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/device.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/rtc.h>
@@ -37,24 +36,7 @@ extern int systemasic_irq_demux(int);
 void *dreamcast_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
 int dreamcast_consistent_free(struct device *, size_t, void *, dma_addr_t);
 
-const char *get_system_type(void)
-{
-	return "Sega Dreamcast";
-}
-
-struct sh_machine_vector mv_dreamcast __initmv = {
-	.mv_nr_irqs		= NR_IRQS,
-
-	.mv_irq_demux		= systemasic_irq_demux,
-
-#ifdef CONFIG_PCI
-	.mv_consistent_alloc	= dreamcast_consistent_alloc,
-	.mv_consistent_free	= dreamcast_consistent_free,
-#endif
-};
-ALIAS_MV(dreamcast)
-
-int __init platform_setup(void)
+static void __init dreamcast_setup(char **cmdline_p)
 {
 	int i;
 
@@ -76,6 +58,16 @@ int __init platform_setup(void)
 	if (gapspci_init() < 0)
 		printk(KERN_WARNING "GAPSPCI was not detected.\n");
 #endif
-
-	return 0;
 }
+
+struct sh_machine_vector mv_dreamcast __initmv = {
+	.mv_name		= "Sega Dreamcast",
+	.mv_setup		= dreamcast_setup,
+	.mv_irq_demux		= systemasic_irq_demux,
+
+#ifdef CONFIG_PCI
+	.mv_consistent_alloc	= dreamcast_consistent_alloc,
+	.mv_consistent_free	= dreamcast_consistent_free,
+#endif
+};
+ALIAS_MV(dreamcast)
diff --git a/arch/sh/boards/ec3104/setup.c b/arch/sh/boards/ec3104/setup.c
index 4b3ef16a0e96..902bc975a13e 100644
--- a/arch/sh/boards/ec3104/setup.c
+++ b/arch/sh/boards/ec3104/setup.c
@@ -21,22 +21,36 @@
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/types.h>
-
 #include <asm/io.h>
 #include <asm/irq.h>
 #include <asm/machvec.h>
 #include <asm/mach/ec3104.h>
 
-const char *get_system_type(void)
+static void __init ec3104_setup(char **cmdline_p)
 {
-	return "EC3104";
+	char str[8];
+	int i;
+
+	for (i=0; i<8; i++)
+		str[i] = ctrl_readb(EC3104_BASE + i);
+
+	for (i = EC3104_IRQBASE; i < EC3104_IRQBASE + 32; i++)
+		irq_desc[i].handler = &ec3104_int;
+
+	printk("initializing EC3104 \"%.8s\" at %08x, IRQ %d, IRQ base %d\n",
+	       str, EC3104_BASE, EC3104_IRQ, EC3104_IRQBASE);
+
+	/* mask all interrupts.  this should have been done by the boot
+	 * loader for us but we want to be sure ... */
+	ctrl_writel(0xffffffff, EC3104_IMR);
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_ec3104 __initmv = {
+	.mv_name	= "EC3104",
+	.mv_setup	= ec3104_setup,
 	.mv_nr_irqs	= 96,
 
 	.mv_inb		= ec3104_inb,
@@ -48,31 +62,4 @@ struct sh_machine_vector mv_ec3104 __initmv = {
 
 	.mv_irq_demux	= ec3104_irq_demux,
 };
-
 ALIAS_MV(ec3104)
-
-int __init platform_setup(void)
-{
-	char str[8];
-	int i;
-	
-	if (0)
-		return 0;
-
-	for (i=0; i<8; i++)
-		str[i] = ctrl_readb(EC3104_BASE + i);
-
-	for (i = EC3104_IRQBASE; i < EC3104_IRQBASE + 32; i++)
-		irq_desc[i].chip = &ec3104_int;
-
-	printk("initializing EC3104 \"%.8s\" at %08x, IRQ %d, IRQ base %d\n",
-	       str, EC3104_BASE, EC3104_IRQ, EC3104_IRQBASE);
-
-
-	/* mask all interrupts.  this should have been done by the boot
-	 * loader for us but we want to be sure ... */
-	ctrl_writel(0xffffffff, EC3104_IMR);
-	
-	return 0;
-}
-
diff --git a/arch/sh/boards/hp6xx/Makefile b/arch/sh/boards/hp6xx/Makefile
index fdf98ffcf437..ff1b7f5b4e91 100644
--- a/arch/sh/boards/hp6xx/Makefile
+++ b/arch/sh/boards/hp6xx/Makefile
@@ -2,8 +2,6 @@
 # Makefile for the HP6xx specific parts of the kernel
 #
 
-obj-y	 		:= mach.o setup.o
+obj-y	 		:= setup.o
 obj-$(CONFIG_PM)	+= pm.o pm_wakeup.o
 obj-$(CONFIG_APM)	+= hp6xx_apm.o
-
-
diff --git a/arch/sh/boards/hp6xx/setup.c b/arch/sh/boards/hp6xx/setup.c
index 629016bec809..60ab17ad6054 100644
--- a/arch/sh/boards/hp6xx/setup.c
+++ b/arch/sh/boards/hp6xx/setup.c
@@ -19,12 +19,7 @@
 #define	SCPCR	0xa4000116
 #define SCPDR	0xa4000136
 
-const char *get_system_type(void)
-{
-	return "HP6xx";
-}
-
-int __init platform_setup(void)
+static void __init hp6xx_setup(char **cmdline_p)
 {
 	u8 v8;
 	u16 v;
@@ -64,6 +59,42 @@ int __init platform_setup(void)
 	v &= ~SCPCR_TS_MASK;
 	v |= SCPCR_TS_ENABLE;
 	ctrl_outw(v, SCPCR);
-
-	return 0;
 }
+
+/*
+ * XXX: This is stupid, we should have a generic machine vector for the cchips
+ * and just wrap the platform setup code in to this, as it's the only thing
+ * that ends up being different.
+ */
+struct sh_machine_vector mv_hp6xx __initmv = {
+	.mv_name = "hp6xx",
+	.mv_setup = hp6xx_setup,
+	.mv_nr_irqs = HD64461_IRQBASE + HD64461_IRQ_NUM,
+
+	.mv_inb = hd64461_inb,
+	.mv_inw = hd64461_inw,
+	.mv_inl = hd64461_inl,
+	.mv_outb = hd64461_outb,
+	.mv_outw = hd64461_outw,
+	.mv_outl = hd64461_outl,
+
+	.mv_inb_p = hd64461_inb_p,
+	.mv_inw_p = hd64461_inw,
+	.mv_inl_p = hd64461_inl,
+	.mv_outb_p = hd64461_outb_p,
+	.mv_outw_p = hd64461_outw,
+	.mv_outl_p = hd64461_outl,
+
+	.mv_insb = hd64461_insb,
+	.mv_insw = hd64461_insw,
+	.mv_insl = hd64461_insl,
+	.mv_outsb = hd64461_outsb,
+	.mv_outsw = hd64461_outsw,
+	.mv_outsl = hd64461_outsl,
+
+	.mv_readw = hd64461_readw,
+	.mv_writew = hd64461_writew,
+
+	.mv_irq_demux = hd64461_irq_demux,
+};
+ALIAS_MV(hp6xx)
diff --git a/arch/sh/boards/landisk/setup.c b/arch/sh/boards/landisk/setup.c
index 3a795cfb1eda..127b9e020e00 100644
--- a/arch/sh/boards/landisk/setup.c
+++ b/arch/sh/boards/landisk/setup.c
@@ -69,42 +69,6 @@ static void heartbeat_landisk(void)
 	landisk_buzzerparam >>= 1;
 }
 
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_landisk __initmv = {
-	.mv_nr_irqs = 72,
-	.mv_inb = landisk_inb,
-	.mv_inw = landisk_inw,
-	.mv_inl = landisk_inl,
-	.mv_outb = landisk_outb,
-	.mv_outw = landisk_outw,
-	.mv_outl = landisk_outl,
-	.mv_inb_p = landisk_inb_p,
-	.mv_inw_p = landisk_inw,
-	.mv_inl_p = landisk_inl,
-	.mv_outb_p = landisk_outb_p,
-	.mv_outw_p = landisk_outw,
-	.mv_outl_p = landisk_outl,
-	.mv_insb = landisk_insb,
-	.mv_insw = landisk_insw,
-	.mv_insl = landisk_insl,
-	.mv_outsb = landisk_outsb,
-	.mv_outsw = landisk_outsw,
-	.mv_outsl = landisk_outsl,
-        .mv_ioport_map = landisk_ioport_map,
-	.mv_init_irq = init_landisk_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat = heartbeat_landisk,
-#endif
-};
-ALIAS_MV(landisk)
-
-const char *get_system_type(void)
-{
-        return "LANDISK";
-}
-
 static void landisk_power_off(void)
 {
         ctrl_outb(0x01, PA_SHUTDOWN);
@@ -132,16 +96,6 @@ static void check_usl5p(void)
         }
 }
 
-void __init platform_setup(void)
-{
-        landisk_buzzerparam = 0;
-        check_usl5p();
-
-        printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
-        board_time_init = landisk_time_init;
-        pm_power_off = landisk_power_off;
-}
-
 void *area5_io_base;
 void *area6_io_base;
 
@@ -176,4 +130,48 @@ static int __init landisk_cf_init(void)
 	return 0;
 }
 
-__initcall(landisk_cf_init);
+static void __init landisk_setup(char **cmdline_p)
+{
+	device_initcall(landisk_cf_init);
+
+	landisk_buzzerparam = 0;
+	check_usl5p();
+
+	printk(KERN_INFO "I-O DATA DEVICE, INC. \"LANDISK Series\" support.\n");
+
+	board_time_init = landisk_time_init;
+	pm_power_off = landisk_power_off;
+}
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_landisk __initmv = {
+	.mv_name = "LANDISK",
+	.mv_setup = landisk_setup,
+	.mv_nr_irqs = 72,
+	.mv_inb = landisk_inb,
+	.mv_inw = landisk_inw,
+	.mv_inl = landisk_inl,
+	.mv_outb = landisk_outb,
+	.mv_outw = landisk_outw,
+	.mv_outl = landisk_outl,
+	.mv_inb_p = landisk_inb_p,
+	.mv_inw_p = landisk_inw,
+	.mv_inl_p = landisk_inl,
+	.mv_outb_p = landisk_outb_p,
+	.mv_outw_p = landisk_outw,
+	.mv_outl_p = landisk_outl,
+	.mv_insb = landisk_insb,
+	.mv_insw = landisk_insw,
+	.mv_insl = landisk_insl,
+	.mv_outsb = landisk_outsb,
+	.mv_outsw = landisk_outsw,
+	.mv_outsl = landisk_outsl,
+	.mv_ioport_map = landisk_ioport_map,
+	.mv_init_irq = init_landisk_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat = heartbeat_landisk,
+#endif
+};
+ALIAS_MV(landisk)
diff --git a/arch/sh/boards/mpc1211/setup.c b/arch/sh/boards/mpc1211/setup.c
index e9c8ff221dda..8eb5d4303972 100644
--- a/arch/sh/boards/mpc1211/setup.c
+++ b/arch/sh/boards/mpc1211/setup.c
@@ -10,14 +10,12 @@
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <linux/interrupt.h>
-
 #include <asm/io.h>
 #include <asm/machvec.h>
 #include <asm/mpc1211/mpc1211.h>
 #include <asm/mpc1211/pci.h>
 #include <asm/mpc1211/m1543c.h>
 
-
 /* ALI15X3 SMBus address offsets */
 #define SMBHSTSTS   (0 + 0x3100)
 #define SMBHSTCNT   (1 + 0x3100)
@@ -50,11 +48,6 @@
 #define ALI15X3_STS_TERM	0x80	/* terminated by abort */
 #define ALI15X3_STS_ERR		0xE0	/* all the bad error bits */
 
-const char *get_system_type(void)
-{
-	return "Interface MPC-1211(CTP/PCI/MPC-SH02)";
-}
-
 static void __init pci_write_config(unsigned long busNo,
 				    unsigned long devNo,
 				    unsigned long fncNo,
@@ -205,7 +198,7 @@ int mpc1211_irq_demux(int irq)
 	return irq;
 }
 
-void __init init_mpc1211_IRQ(void)
+static void __init init_mpc1211_IRQ(void)
 {
 	int i;
 	/*
@@ -289,26 +282,10 @@ static int put_smb_blk(unsigned char *p, int address, int command, int no)
 	return 0;
 }
 
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_mpc1211 __initmv = {
-	.mv_nr_irqs		= 48,
-	.mv_irq_demux		= mpc1211_irq_demux,
-	.mv_init_irq		= init_mpc1211_IRQ,
-
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_mpc1211,
-#endif
-};
-
-ALIAS_MV(mpc1211)
-
 /* arch/sh/boards/mpc1211/rtc.c */
 void mpc1211_time_init(void);
 
-int __init platform_setup(void)
+static void __init mpc1211_setup(char **cmdline_p)
 {
 	unsigned char spd_buf[128];
 
@@ -332,3 +309,18 @@ int __init platform_setup(void)
 	return 0;
 }
 
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_mpc1211 __initmv = {
+	.mv_name		= "Interface MPC-1211(CTP/PCI/MPC-SH02)",
+	.mv_setup		= mpc1211_setup,
+	.mv_nr_irqs		= 48,
+	.mv_irq_demux		= mpc1211_irq_demux,
+	.mv_init_irq		= init_mpc1211_IRQ,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_mpc1211,
+#endif
+};
+ALIAS_MV(mpc1211)
diff --git a/arch/sh/boards/renesas/edosk7705/setup.c b/arch/sh/boards/renesas/edosk7705/setup.c
index ba143fa4afaa..ec5be0107719 100644
--- a/arch/sh/boards/renesas/edosk7705/setup.c
+++ b/arch/sh/boards/renesas/edosk7705/setup.c
@@ -8,19 +8,21 @@
  * Modified for edosk7705 development
  * board by S. Dunn, 2003.
  */
-
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/machvec_init.h>
 #include <asm/edosk7705/io.h>
 
-static void init_edosk7705(void);
+static void __init sh_edosk7705_init_irq(void)
+{
+	/* This is the Ethernet interrupt */
+	make_imask_irq(0x09);
+}
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_edosk7705 __initmv = {
+	.mv_name		= "EDOSK7705",
 	.mv_nr_irqs		= 80,
 
 	.mv_inb			= sh_edosk7705_inb,
@@ -37,23 +39,6 @@ struct sh_machine_vector mv_edosk7705 __initmv = {
 	.mv_outsl		= sh_edosk7705_outsl,
 
 	.mv_isa_port2addr	= sh_edosk7705_isa_port2addr,
-	.mv_init_irq		= init_edosk7705,
+	.mv_init_irq		= sh_edosk7705_init_irq,
 };
 ALIAS_MV(edosk7705)
-
-static void __init init_edosk7705(void)
-{
-	/* This is the Ethernet interrupt */
-	make_imask_irq(0x09);
-}
-
-const char *get_system_type(void)
-{
-	return "EDOSK7705";
-}
-
-void __init platform_setup(void)
-{
-	/* Nothing .. */
-}
-
diff --git a/arch/sh/boards/renesas/hs7751rvoip/setup.c b/arch/sh/boards/renesas/hs7751rvoip/setup.c
index a2cbcc5d530f..0414c15c3458 100644
--- a/arch/sh/boards/renesas/hs7751rvoip/setup.c
+++ b/arch/sh/boards/renesas/hs7751rvoip/setup.c
@@ -1,15 +1,12 @@
 /*
- * linux/arch/sh/kernel/setup_hs7751rvoip.c
+ * Renesas Technology Sales HS7751RVoIP Support.
  *
  * Copyright (C) 2000  Kazumoto Kojima
  *
- * Renesas Technology Sales HS7751RVoIP Support.
- *
  * Modified for HS7751RVoIP by
  * Atom Create Engineering Co., Ltd. 2002.
  * Lineo uSolutions, Inc. 2003.
  */
-
 #include <linux/init.h>
 #include <linux/irq.h>
 #include <linux/mm.h>
@@ -23,8 +20,6 @@
 #include <asm/rtc.h>
 #include <asm/irq.h>
 
-unsigned int debug_counter;
-
 static void __init hs7751rvoip_init_irq(void)
 {
 #if defined(CONFIG_HS7751RVOIP_CODEC)
@@ -35,56 +30,11 @@ static void __init hs7751rvoip_init_irq(void)
 	init_hs7751rvoip_IRQ();
 }
 
-struct sh_machine_vector mv_hs7751rvoip __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= hs7751rvoip_inb,
-	.mv_inw			= hs7751rvoip_inw,
-	.mv_inl			= hs7751rvoip_inl,
-	.mv_outb		= hs7751rvoip_outb,
-	.mv_outw		= hs7751rvoip_outw,
-	.mv_outl		= hs7751rvoip_outl,
-
-	.mv_inb_p		= hs7751rvoip_inb_p,
-	.mv_inw_p		= hs7751rvoip_inw,
-	.mv_inl_p		= hs7751rvoip_inl,
-	.mv_outb_p		= hs7751rvoip_outb_p,
-	.mv_outw_p		= hs7751rvoip_outw,
-	.mv_outl_p		= hs7751rvoip_outl,
-
-	.mv_insb		= hs7751rvoip_insb,
-	.mv_insw		= hs7751rvoip_insw,
-	.mv_insl		= hs7751rvoip_insl,
-	.mv_outsb		= hs7751rvoip_outsb,
-	.mv_outsw		= hs7751rvoip_outsw,
-	.mv_outsl		= hs7751rvoip_outsl,
-
-	.mv_init_irq		= hs7751rvoip_init_irq,
-	.mv_ioport_map		= hs7751rvoip_ioport_map,
-};
-ALIAS_MV(hs7751rvoip)
-
-const char *get_system_type(void)
-{
-	return "HS7751RVoIP";
-}
-
 static void hs7751rvoip_power_off(void)
 {
 	ctrl_outw(ctrl_inw(PA_OUTPORTR) & 0xffdf, PA_OUTPORTR);
 }
 
-/*
- * Initialize the board
- */
-void __init platform_setup(void)
-{
-	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
-	ctrl_outb(0xf0, PA_OUTPORTR);
-	pm_power_off = hs7751rvoip_power_off;
-	debug_counter = 0;
-}
-
 void *area5_io8_base;
 void *area6_io8_base;
 void *area5_io16_base;
@@ -127,4 +77,46 @@ static int __init hs7751rvoip_cf_init(void)
 	return 0;
 }
 
-__initcall(hs7751rvoip_cf_init);
+/*
+ * Initialize the board
+ */
+static void __init hs7751rvoip_setup(char **cmdline_p)
+{
+	device_initcall(hs7751rvoip_cf_init);
+
+	ctrl_outb(0xf0, PA_OUTPORTR);
+	pm_power_off = hs7751rvoip_power_off;
+
+	printk(KERN_INFO "Renesas Technology Sales HS7751RVoIP-2 support.\n");
+}
+
+struct sh_machine_vector mv_hs7751rvoip __initmv = {
+	.mv_name		= "HS7751RVoIP",
+	.mv_setup		= hs7751rvoip_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= hs7751rvoip_inb,
+	.mv_inw			= hs7751rvoip_inw,
+	.mv_inl			= hs7751rvoip_inl,
+	.mv_outb		= hs7751rvoip_outb,
+	.mv_outw		= hs7751rvoip_outw,
+	.mv_outl		= hs7751rvoip_outl,
+
+	.mv_inb_p		= hs7751rvoip_inb_p,
+	.mv_inw_p		= hs7751rvoip_inw,
+	.mv_inl_p		= hs7751rvoip_inl,
+	.mv_outb_p		= hs7751rvoip_outb_p,
+	.mv_outw_p		= hs7751rvoip_outw,
+	.mv_outl_p		= hs7751rvoip_outl,
+
+	.mv_insb		= hs7751rvoip_insb,
+	.mv_insw		= hs7751rvoip_insw,
+	.mv_insl		= hs7751rvoip_insl,
+	.mv_outsb		= hs7751rvoip_outsb,
+	.mv_outsw		= hs7751rvoip_outsw,
+	.mv_outsl		= hs7751rvoip_outsl,
+
+	.mv_init_irq		= hs7751rvoip_init_irq,
+	.mv_ioport_map		= hs7751rvoip_ioport_map,
+};
+ALIAS_MV(hs7751rvoip)
diff --git a/arch/sh/boards/renesas/r7780rp/setup.c b/arch/sh/boards/renesas/r7780rp/setup.c
index 0e4d5e1b541c..b941aa0aa34e 100644
--- a/arch/sh/boards/renesas/r7780rp/setup.c
+++ b/arch/sh/boards/renesas/r7780rp/setup.c
@@ -20,41 +20,6 @@
 extern void heartbeat_r7780rp(void);
 extern void init_r7780rp_IRQ(void);
 
-/*
- * The Machine Vector
- */
-struct sh_machine_vector mv_r7780rp __initmv = {
-	.mv_nr_irqs		= 109,
-
-	.mv_inb			= r7780rp_inb,
-	.mv_inw			= r7780rp_inw,
-	.mv_inl			= r7780rp_inl,
-	.mv_outb		= r7780rp_outb,
-	.mv_outw		= r7780rp_outw,
-	.mv_outl		= r7780rp_outl,
-
-	.mv_inb_p		= r7780rp_inb_p,
-	.mv_inw_p		= r7780rp_inw,
-	.mv_inl_p		= r7780rp_inl,
-	.mv_outb_p		= r7780rp_outb_p,
-	.mv_outw_p		= r7780rp_outw,
-	.mv_outl_p		= r7780rp_outl,
-
-	.mv_insb		= r7780rp_insb,
-	.mv_insw		= r7780rp_insw,
-	.mv_insl		= r7780rp_insl,
-	.mv_outsb		= r7780rp_outsb,
-	.mv_outsw		= r7780rp_outsw,
-	.mv_outsl		= r7780rp_outsl,
-
-	.mv_ioport_map		= r7780rp_ioport_map,
-	.mv_init_irq		= init_r7780rp_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_r7780rp,
-#endif
-};
-ALIAS_MV(r7780rp)
-
 static struct resource m66596_usb_host_resources[] = {
 	[0] = {
 		.start	= 0xa4800000,
@@ -88,7 +53,6 @@ static int __init r7780rp_devices_setup(void)
 	return platform_add_devices(r7780rp_devices,
 				    ARRAY_SIZE(r7780rp_devices));
 }
-__initcall(r7780rp_devices_setup);
 
 /*
  * Platform specific clocks
@@ -117,11 +81,6 @@ static struct clk *r7780rp_clocks[] = {
 	&ivdr_clk,
 };
 
-const char *get_system_type(void)
-{
-	return "Highlander R7780RP-1";
-}
-
 static void r7780rp_power_off(void)
 {
 #ifdef CONFIG_SH_R7780MP
@@ -132,11 +91,13 @@ static void r7780rp_power_off(void)
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init r7780rp_setup(char **cmdline_p)
 {
 	u16 ver = ctrl_inw(PA_VERREG);
 	int i;
 
+	device_initcall(r7780rp_devices_setup);
+
 	printk(KERN_INFO "Renesas Solutions Highlander R7780RP-1 support.\n");
 
 	printk(KERN_INFO "Board version: %d (revision %d), "
@@ -162,3 +123,41 @@ void __init platform_setup(void)
 
 	pm_power_off = r7780rp_power_off;
 }
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_r7780rp __initmv = {
+	.mv_name		= "Highlander R7780RP-1",
+	.mv_setup		= r7780rp_setup,
+
+	.mv_nr_irqs		= 109,
+
+	.mv_inb			= r7780rp_inb,
+	.mv_inw			= r7780rp_inw,
+	.mv_inl			= r7780rp_inl,
+	.mv_outb		= r7780rp_outb,
+	.mv_outw		= r7780rp_outw,
+	.mv_outl		= r7780rp_outl,
+
+	.mv_inb_p		= r7780rp_inb_p,
+	.mv_inw_p		= r7780rp_inw,
+	.mv_inl_p		= r7780rp_inl,
+	.mv_outb_p		= r7780rp_outb_p,
+	.mv_outw_p		= r7780rp_outw,
+	.mv_outl_p		= r7780rp_outl,
+
+	.mv_insb		= r7780rp_insb,
+	.mv_insw		= r7780rp_insw,
+	.mv_insl		= r7780rp_insl,
+	.mv_outsb		= r7780rp_outsb,
+	.mv_outsw		= r7780rp_outsw,
+	.mv_outsl		= r7780rp_outsl,
+
+	.mv_ioport_map		= r7780rp_ioport_map,
+	.mv_init_irq		= init_r7780rp_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_r7780rp,
+#endif
+};
+ALIAS_MV(r7780rp)
diff --git a/arch/sh/boards/renesas/rts7751r2d/Makefile b/arch/sh/boards/renesas/rts7751r2d/Makefile
index eee1ed6f5727..686fc9ea5989 100644
--- a/arch/sh/boards/renesas/rts7751r2d/Makefile
+++ b/arch/sh/boards/renesas/rts7751r2d/Makefile
@@ -2,5 +2,5 @@
 # Makefile for the RTS7751R2D specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o led.o
-
+obj-y	 := setup.o io.o irq.o
+obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/renesas/rts7751r2d/mach.c b/arch/sh/boards/renesas/rts7751r2d/mach.c
deleted file mode 100644
index fe3e8735e9f8..000000000000
--- a/arch/sh/boards/renesas/rts7751r2d/mach.c
+++ /dev/null
@@ -1,66 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_rts7751r2d.c
- *
- * Minor tweak of mach_se.c file to reference rts7751r2d-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Renesas Technology sales RTS7751R2D
- */
-
-#include <linux/init.h>
-#include <linux/types.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/irq.h>
-#include <asm/mach/rts7751r2d.h>
-
-extern void heartbeat_rts7751r2d(void);
-extern void init_rts7751r2d_IRQ(void);
-extern int rts7751r2d_irq_demux(int irq);
-
-extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
-extern int voyagergx_consistent_free(struct device *, size_t, void *, dma_addr_t);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_rts7751r2d __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= rts7751r2d_inb,
-	.mv_inw			= rts7751r2d_inw,
-	.mv_inl			= rts7751r2d_inl,
-	.mv_outb		= rts7751r2d_outb,
-	.mv_outw		= rts7751r2d_outw,
-	.mv_outl		= rts7751r2d_outl,
-
-	.mv_inb_p		= rts7751r2d_inb_p,
-	.mv_inw_p		= rts7751r2d_inw,
-	.mv_inl_p		= rts7751r2d_inl,
-	.mv_outb_p		= rts7751r2d_outb_p,
-	.mv_outw_p		= rts7751r2d_outw,
-	.mv_outl_p		= rts7751r2d_outl,
-
-	.mv_insb		= rts7751r2d_insb,
-	.mv_insw		= rts7751r2d_insw,
-	.mv_insl		= rts7751r2d_insl,
-	.mv_outsb		= rts7751r2d_outsb,
-	.mv_outsw		= rts7751r2d_outsw,
-	.mv_outsl		= rts7751r2d_outsl,
-
-	.mv_init_irq		= init_rts7751r2d_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_rts7751r2d,
-#endif
-	.mv_irq_demux		= rts7751r2d_irq_demux,
-
-#ifdef CONFIG_USB_OHCI_HCD
-	.mv_consistent_alloc	= voyagergx_consistent_alloc,
-	.mv_consistent_free	= voyagergx_consistent_free,
-#endif
-};
-ALIAS_MV(rts7751r2d)
diff --git a/arch/sh/boards/renesas/rts7751r2d/setup.c b/arch/sh/boards/renesas/rts7751r2d/setup.c
index bb56c018f2ee..20597a6e6702 100644
--- a/arch/sh/boards/renesas/rts7751r2d/setup.c
+++ b/arch/sh/boards/renesas/rts7751r2d/setup.c
@@ -13,9 +13,17 @@
 #include <linux/serial_8250.h>
 #include <linux/pm.h>
 #include <asm/io.h>
+#include <asm/machvec.h>
 #include <asm/mach/rts7751r2d.h>
 #include <asm/voyagergx.h>
 
+extern void heartbeat_rts7751r2d(void);
+extern void init_rts7751r2d_IRQ(void);
+extern int rts7751r2d_irq_demux(int irq);
+
+extern void *voyagergx_consistent_alloc(struct device *, size_t, dma_addr_t *, gfp_t);
+extern int voyagergx_consistent_free(struct device *, size_t, void *, dma_addr_t);
+
 static struct plat_serial8250_port uart_platform_data[] = {
 	{
 		.membase	= (void *)VOYAGER_UART_BASE,
@@ -70,12 +78,6 @@ static int __init rts7751r2d_devices_setup(void)
 	return platform_add_devices(rts7751r2d_devices,
 				    ARRAY_SIZE(rts7751r2d_devices));
 }
-__initcall(rts7751r2d_devices_setup);
-
-const char *get_system_type(void)
-{
-	return "RTS7751R2D";
-}
 
 static void rts7751r2d_power_off(void)
 {
@@ -85,12 +87,56 @@ static void rts7751r2d_power_off(void)
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init rts7751r2d_setup(char **cmdline_p)
 {
-	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
+	device_initcall(rts7751r2d_devices_setup);
+
 	ctrl_outw(0x0000, PA_OUTPORT);
 	pm_power_off = rts7751r2d_power_off;
 
 	voyagergx_serial_init();
 
+	printk(KERN_INFO "Renesas Technology Sales RTS7751R2D support.\n");
 }
+
+/*
+ * The Machine Vector
+ */
+struct sh_machine_vector mv_rts7751r2d __initmv = {
+	.mv_name		= "RTS7751R2D",
+	.mv_setup		= rts7751r2d_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= rts7751r2d_inb,
+	.mv_inw			= rts7751r2d_inw,
+	.mv_inl			= rts7751r2d_inl,
+	.mv_outb		= rts7751r2d_outb,
+	.mv_outw		= rts7751r2d_outw,
+	.mv_outl		= rts7751r2d_outl,
+
+	.mv_inb_p		= rts7751r2d_inb_p,
+	.mv_inw_p		= rts7751r2d_inw,
+	.mv_inl_p		= rts7751r2d_inl,
+	.mv_outb_p		= rts7751r2d_outb_p,
+	.mv_outw_p		= rts7751r2d_outw,
+	.mv_outl_p		= rts7751r2d_outl,
+
+	.mv_insb		= rts7751r2d_insb,
+	.mv_insw		= rts7751r2d_insw,
+	.mv_insl		= rts7751r2d_insl,
+	.mv_outsb		= rts7751r2d_outsb,
+	.mv_outsw		= rts7751r2d_outsw,
+	.mv_outsl		= rts7751r2d_outsl,
+
+	.mv_init_irq		= init_rts7751r2d_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_rts7751r2d,
+#endif
+	.mv_irq_demux		= rts7751r2d_irq_demux,
+
+#ifdef CONFIG_USB_SM501
+	.mv_consistent_alloc	= voyagergx_consistent_alloc,
+	.mv_consistent_free	= voyagergx_consistent_free,
+#endif
+};
+ALIAS_MV(rts7751r2d)
diff --git a/arch/sh/boards/renesas/systemh/setup.c b/arch/sh/boards/renesas/systemh/setup.c
index 433c7c3f35e2..a8467bf90c25 100644
--- a/arch/sh/boards/renesas/systemh/setup.c
+++ b/arch/sh/boards/renesas/systemh/setup.c
@@ -20,20 +20,16 @@
 
 extern void make_systemh_irq(unsigned int irq);
 
-const char *get_system_type(void)
-{
-	return "7751 SystemH";
-}
-
 /*
  * Initialize IRQ setting
  */
-void __init init_7751systemh_IRQ(void)
+static void __init sh7751systemh_init_irq(void)
 {
 	make_systemh_irq(0xb);	/* Ethernet interrupt */
 }
 
 struct sh_machine_vector mv_7751systemh __initmv = {
+	.mv_name		= "7751 SystemH",
 	.mv_nr_irqs		= 72,
 
 	.mv_inb			= sh7751systemh_inb,
@@ -57,12 +53,6 @@ struct sh_machine_vector mv_7751systemh __initmv = {
 	.mv_outsw		= sh7751systemh_outsw,
 	.mv_outsl		= sh7751systemh_outsl,
 
-	.mv_init_irq		= init_7751systemh_IRQ,
+	.mv_init_irq		= sh7751system_init_irq,
 };
 ALIAS_MV(7751systemh)
-
-int __init platform_setup(void)
-{
-	return 0;
-}
-
diff --git a/arch/sh/boards/saturn/setup.c b/arch/sh/boards/saturn/setup.c
index bea6c572ad82..a3a37c9aad2e 100644
--- a/arch/sh/boards/saturn/setup.c
+++ b/arch/sh/boards/saturn/setup.c
@@ -9,22 +9,17 @@
  */
 #include <linux/kernel.h>
 #include <linux/init.h>
-
 #include <asm/io.h>
 #include <asm/machvec.h>
 #include <asm/mach/io.h>
 
 extern int saturn_irq_demux(int irq_nr);
 
-const char *get_system_type(void)
-{
-	return "Sega Saturn";
-}
-
 /*
  * The Machine Vector
  */
 struct sh_machine_vector mv_saturn __initmv = {
+	.mv_name		= "Sega Saturn",
 	.mv_nr_irqs		= 80,	/* Fix this later */
 
 	.mv_isa_port2addr	= saturn_isa_port2addr,
@@ -33,11 +28,4 @@ struct sh_machine_vector mv_saturn __initmv = {
 	.mv_ioremap		= saturn_ioremap,
 	.mv_iounmap		= saturn_iounmap,
 };
-
 ALIAS_MV(saturn)
-
-int __init platform_setup(void)
-{
-	return 0;
-}
-
diff --git a/arch/sh/boards/se/7300/setup.c b/arch/sh/boards/se/7300/setup.c
index bb7e1a189be8..6f082a722d42 100644
--- a/arch/sh/boards/se/7300/setup.c
+++ b/arch/sh/boards/se/7300/setup.c
@@ -14,17 +14,11 @@
 void heartbeat_7300se(void);
 void init_7300se_IRQ(void);
 
-const char *
-get_system_type(void)
-{
-	return "SolutionEngine 7300";
-}
-
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_7300se __initmv = {
+	.mv_name = "SolutionEngine 7300",
 	.mv_nr_irqs = 109,
 	.mv_inb = sh7300se_inb,
 	.mv_inw = sh7300se_inw,
@@ -52,13 +46,4 @@ struct sh_machine_vector mv_7300se __initmv = {
 	.mv_heartbeat = heartbeat_7300se,
 #endif
 };
-
 ALIAS_MV(7300se)
-/*
- * Initialize the board
- */
-void __init
-platform_setup(void)
-{
-
-}
diff --git a/arch/sh/boards/se/73180/setup.c b/arch/sh/boards/se/73180/setup.c
index 4daf53b1457f..b38ef50a160a 100644
--- a/arch/sh/boards/se/73180/setup.c
+++ b/arch/sh/boards/se/73180/setup.c
@@ -17,17 +17,11 @@
 void heartbeat_73180se(void);
 void init_73180se_IRQ(void);
 
-const char *
-get_system_type(void)
-{
-	return "SolutionEngine 73180";
-}
-
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_73180se __initmv = {
+	.mv_name = "SolutionEngine 73180",
 	.mv_nr_irqs = 108,
 	.mv_inb = sh73180se_inb,
 	.mv_inw = sh73180se_inw,
@@ -56,13 +50,4 @@ struct sh_machine_vector mv_73180se __initmv = {
 	.mv_heartbeat = heartbeat_73180se,
 #endif
 };
-
 ALIAS_MV(73180se)
-/*
- * Initialize the board
- */
-void __init
-platform_setup(void)
-{
-
-}
diff --git a/arch/sh/boards/se/770x/Makefile b/arch/sh/boards/se/770x/Makefile
index c96ed6933fe1..9a5035f80ec0 100644
--- a/arch/sh/boards/se/770x/Makefile
+++ b/arch/sh/boards/se/770x/Makefile
@@ -2,6 +2,5 @@
 # Makefile for the 770x SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o
-
+obj-y	 := setup.o io.o irq.o
 obj-$(CONFIG_HEARTBEAT) += led.o
diff --git a/arch/sh/boards/se/770x/mach.c b/arch/sh/boards/se/770x/mach.c
deleted file mode 100644
index e8968b71c353..000000000000
--- a/arch/sh/boards/se/770x/mach.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_se.c
- *
- * Copyright (C) 2000 Stuart Menefy (stuart.menefy@st.com)
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Hitachi SolutionEngine
- */
-
-#include <linux/init.h>
-
-#include <asm/machvec.h>
-#include <asm/rtc.h>
-#include <asm/se.h>
-
-void heartbeat_se(void);
-void init_se_IRQ(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_se __initmv = {
-#if defined(CONFIG_CPU_SH4)
-	.mv_nr_irqs		= 48,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
-	.mv_nr_irqs		= 32,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
-	.mv_nr_irqs		= 61,
-#elif defined(CONFIG_CPU_SUBTYPE_SH7705)
-	.mv_nr_irqs		= 86,
-#endif
-
-	.mv_inb			= se_inb,
-	.mv_inw			= se_inw,
-	.mv_inl			= se_inl,
-	.mv_outb		= se_outb,
-	.mv_outw		= se_outw,
-	.mv_outl		= se_outl,
-
-	.mv_inb_p		= se_inb_p,
-	.mv_inw_p		= se_inw,
-	.mv_inl_p		= se_inl,
-	.mv_outb_p		= se_outb_p,
-	.mv_outw_p		= se_outw,
-	.mv_outl_p		= se_outl,
-
-	.mv_insb		= se_insb,
-	.mv_insw		= se_insw,
-	.mv_insl		= se_insl,
-	.mv_outsb		= se_outsb,
-	.mv_outsw		= se_outsw,
-	.mv_outsl		= se_outsl,
-
-	.mv_init_irq		= init_se_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_se,
-#endif
-};
-ALIAS_MV(se)
diff --git a/arch/sh/boards/se/770x/setup.c b/arch/sh/boards/se/770x/setup.c
index c9f75272e751..f3f82b7c8217 100644
--- a/arch/sh/boards/se/770x/setup.c
+++ b/arch/sh/boards/se/770x/setup.c
@@ -7,15 +7,17 @@
  * Hitachi SolutionEngine Support.
  *
  */
-
 #include <linux/init.h>
 #include <linux/irq.h>
-
 #include <linux/hdreg.h>
 #include <linux/ide.h>
 #include <asm/io.h>
 #include <asm/se.h>
 #include <asm/smc37c93x.h>
+#include <asm/machvec.h>
+
+void heartbeat_se(void);
+void init_se_IRQ(void);
 
 /*
  * Configure the Super I/O chip
@@ -26,7 +28,8 @@ static void __init smsc_config(int index, int data)
 	outb_p(data, DATA_PORT);
 }
 
-static void __init init_smsc(void)
+/* XXX: Another candidate for a more generic cchip machine vector */
+static void __init smsc_setup(char **cmdline_p)
 {
 	outb_p(CONFIG_ENTER, CONFIG_PORT);
 	outb_p(CONFIG_ENTER, CONFIG_PORT);
@@ -69,16 +72,46 @@ static void __init init_smsc(void)
 	outb_p(CONFIG_EXIT, CONFIG_PORT);
 }
 
-const char *get_system_type(void)
-{
-	return "SolutionEngine";
-}
-
 /*
- * Initialize the board
+ * The Machine Vector
  */
-void __init platform_setup(void)
-{
-	init_smsc();
-	/* XXX: RTC setting comes here */
-}
+struct sh_machine_vector mv_se __initmv = {
+	.mv_name		= "SolutionEngine",
+	.mv_setup		= smsc_setup,
+#if defined(CONFIG_CPU_SH4)
+	.mv_nr_irqs		= 48,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7708)
+	.mv_nr_irqs		= 32,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7709)
+	.mv_nr_irqs		= 61,
+#elif defined(CONFIG_CPU_SUBTYPE_SH7705)
+	.mv_nr_irqs		= 86,
+#endif
+
+	.mv_inb			= se_inb,
+	.mv_inw			= se_inw,
+	.mv_inl			= se_inl,
+	.mv_outb		= se_outb,
+	.mv_outw		= se_outw,
+	.mv_outl		= se_outl,
+
+	.mv_inb_p		= se_inb_p,
+	.mv_inw_p		= se_inw,
+	.mv_inl_p		= se_inl,
+	.mv_outb_p		= se_outb_p,
+	.mv_outw_p		= se_outw,
+	.mv_outl_p		= se_outl,
+
+	.mv_insb		= se_insb,
+	.mv_insw		= se_insw,
+	.mv_insl		= se_insl,
+	.mv_outsb		= se_outsb,
+	.mv_outsw		= se_outsw,
+	.mv_outsl		= se_outsl,
+
+	.mv_init_irq		= init_se_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_se,
+#endif
+};
+ALIAS_MV(se)
diff --git a/arch/sh/boards/se/7751/Makefile b/arch/sh/boards/se/7751/Makefile
index 07bf17a4a19b..188900c48321 100644
--- a/arch/sh/boards/se/7751/Makefile
+++ b/arch/sh/boards/se/7751/Makefile
@@ -2,8 +2,7 @@
 # Makefile for the 7751 SolutionEngine specific parts of the kernel
 #
 
-obj-y	 := mach.o setup.o io.o irq.o
+obj-y	 := setup.o io.o irq.o
 
 obj-$(CONFIG_PCI) += pci.o
 obj-$(CONFIG_HEARTBEAT) += led.o
-
diff --git a/arch/sh/boards/se/7751/mach.c b/arch/sh/boards/se/7751/mach.c
deleted file mode 100644
index 1bb9047d863b..000000000000
--- a/arch/sh/boards/se/7751/mach.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * linux/arch/sh/kernel/mach_7751se.c
- *
- * Minor tweak of mach_se.c file to reference 7751se-specific items.
- *
- * May be copied or modified under the terms of the GNU General Public
- * License.  See linux/COPYING for more information.
- *
- * Machine vector for the Hitachi 7751 SolutionEngine
- */
-
-#include <linux/init.h>
-#include <asm/machvec.h>
-#include <asm/se7751.h>
-
-void heartbeat_7751se(void);
-void init_7751se_IRQ(void);
-
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_7751se __initmv = {
-	.mv_nr_irqs		= 72,
-
-	.mv_inb			= sh7751se_inb,
-	.mv_inw			= sh7751se_inw,
-	.mv_inl			= sh7751se_inl,
-	.mv_outb		= sh7751se_outb,
-	.mv_outw		= sh7751se_outw,
-	.mv_outl		= sh7751se_outl,
-
-	.mv_inb_p		= sh7751se_inb_p,
-	.mv_inw_p		= sh7751se_inw,
-	.mv_inl_p		= sh7751se_inl,
-	.mv_outb_p		= sh7751se_outb_p,
-	.mv_outw_p		= sh7751se_outw,
-	.mv_outl_p		= sh7751se_outl,
-
-	.mv_insl		= sh7751se_insl,
-	.mv_outsl		= sh7751se_outsl,
-
-	.mv_init_irq		= init_7751se_IRQ,
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= heartbeat_7751se,
-#endif
-};
-ALIAS_MV(7751se)
diff --git a/arch/sh/boards/se/7751/setup.c b/arch/sh/boards/se/7751/setup.c
index 8b693105893c..73e826310ba8 100644
--- a/arch/sh/boards/se/7751/setup.c
+++ b/arch/sh/boards/se/7751/setup.c
@@ -15,72 +15,11 @@
 #include <asm/io.h>
 #include <asm/se7751.h>
 
-#ifdef CONFIG_SH_KGDB
-#include <asm/kgdb.h>
-#endif
-
-/*
- * Configure the Super I/O chip
- */
-#if 0
-/* Leftover code from regular Solution Engine, for reference. */
-/* The SH7751 Solution Engine has a different SuperIO. */
-static void __init smsc_config(int index, int data)
-{
-	outb_p(index, INDEX_PORT);
-	outb_p(data, DATA_PORT);
-}
-
-static void __init init_smsc(void)
-{
-	outb_p(CONFIG_ENTER, CONFIG_PORT);
-	outb_p(CONFIG_ENTER, CONFIG_PORT);
-
-	/* FDC */
-	smsc_config(CURRENT_LDN_INDEX, LDN_FDC);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 6); /* IRQ6 */
-
-	/* IDE1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_IDE1);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 14); /* IRQ14 */
-
-	/* AUXIO (GPIO): to use IDE1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_AUXIO);
-	smsc_config(GPIO46_INDEX, 0x00); /* nIOROP */
-	smsc_config(GPIO47_INDEX, 0x00); /* nIOWOP */
-
-	/* COM1 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_COM1);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IO_BASE_HI_INDEX, 0x03);
-	smsc_config(IO_BASE_LO_INDEX, 0xf8);
-	smsc_config(IRQ_SELECT_INDEX, 4); /* IRQ4 */
-
-	/* COM2 */
-	smsc_config(CURRENT_LDN_INDEX, LDN_COM2);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IO_BASE_HI_INDEX, 0x02);
-	smsc_config(IO_BASE_LO_INDEX, 0xf8);
-	smsc_config(IRQ_SELECT_INDEX, 3); /* IRQ3 */
-
-	/* RTC */
-	smsc_config(CURRENT_LDN_INDEX, LDN_RTC);
-	smsc_config(ACTIVATE_INDEX, 0x01);
-	smsc_config(IRQ_SELECT_INDEX, 8); /* IRQ8 */
-
-	/* XXX: PARPORT, KBD, and MOUSE will come here... */
-	outb_p(CONFIG_EXIT, CONFIG_PORT);
-}
-#endif
-
-const char *get_system_type(void)
-{
-	return "7751 SolutionEngine";
-}
+void heartbeat_7751se(void);
+void init_7751se_IRQ(void);
 
 #ifdef CONFIG_SH_KGDB
+#include <asm/kgdb.h>
 static int kgdb_uart_setup(void);
 static struct kgdb_sermap kgdb_uart_sermap = 
 { "ttyS", 0, kgdb_uart_setup, NULL };
@@ -89,7 +28,7 @@ static struct kgdb_sermap kgdb_uart_sermap =
 /*
  * Initialize the board
  */
-void __init platform_setup(void)
+static void __init sh7751se_setup(char **cmdline_p)
 {
 	/* Call init_smsc() replacement to set up SuperIO. */
 	/* XXX: RTC setting comes here */
@@ -223,3 +162,37 @@ static int kgdb_uart_setup(void)
 	return 0;
 }
 #endif /* CONFIG_SH_KGDB */
+
+
+/*
+ * The Machine Vector
+ */
+
+struct sh_machine_vector mv_7751se __initmv = {
+	.mv_name		= "7751 SolutionEngine",
+	.mv_setup		= sh7751se_setup,
+	.mv_nr_irqs		= 72,
+
+	.mv_inb			= sh7751se_inb,
+	.mv_inw			= sh7751se_inw,
+	.mv_inl			= sh7751se_inl,
+	.mv_outb		= sh7751se_outb,
+	.mv_outw		= sh7751se_outw,
+	.mv_outl		= sh7751se_outl,
+
+	.mv_inb_p		= sh7751se_inb_p,
+	.mv_inw_p		= sh7751se_inw,
+	.mv_inl_p		= sh7751se_inl,
+	.mv_outb_p		= sh7751se_outb_p,
+	.mv_outw_p		= sh7751se_outw,
+	.mv_outl_p		= sh7751se_outl,
+
+	.mv_insl		= sh7751se_insl,
+	.mv_outsl		= sh7751se_outsl,
+
+	.mv_init_irq		= init_7751se_IRQ,
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= heartbeat_7751se,
+#endif
+};
+ALIAS_MV(7751se)
diff --git a/arch/sh/boards/sh03/setup.c b/arch/sh/boards/sh03/setup.c
index 7d31d6aed8a5..6c310587ddfe 100644
--- a/arch/sh/boards/sh03/setup.c
+++ b/arch/sh/boards/sh03/setup.c
@@ -13,12 +13,7 @@
 #include <asm/sh03/sh03.h>
 #include <asm/addrspace.h>
 
-const char *get_system_type(void)
-{
-	return "Interface (CTP/PCI-SH03)";
-}
-
-static void init_sh03_IRQ(void)
+static void __init init_sh03_IRQ(void)
 {
 	ctrl_outw(ctrl_inw(INTC_ICR) | INTC_ICR_IRLM, INTC_ICR);
 
@@ -41,7 +36,17 @@ static void __iomem *sh03_ioport_map(unsigned long port, unsigned int size)
         return (void __iomem *)(port + PCI_IO_BASE);
 }
 
+/* arch/sh/boards/sh03/rtc.c */
+void sh03_time_init(void);
+
+static void __init sh03_setup(char **cmdline_p)
+{
+	board_time_init = sh03_time_init;
+}
+
 struct sh_machine_vector mv_sh03 __initmv = {
+	.mv_name		= "Interface (CTP/PCI-SH03)",
+	.mv_setup		= sh03_setup,
 	.mv_nr_irqs		= 48,
 	.mv_ioport_map		= sh03_ioport_map,
 	.mv_init_irq		= init_sh03_IRQ,
@@ -51,12 +56,3 @@ struct sh_machine_vector mv_sh03 __initmv = {
 #endif
 };
 ALIAS_MV(sh03)
-
-/* arch/sh/boards/sh03/rtc.c */
-void sh03_time_init(void);
-
-int __init platform_setup(void)
-{
-	board_time_init = sh03_time_init;
-	return 0;
-}
diff --git a/arch/sh/boards/shmin/setup.c b/arch/sh/boards/shmin/setup.c
index a950ec4d765d..2f0c19706cf9 100644
--- a/arch/sh/boards/shmin/setup.c
+++ b/arch/sh/boards/shmin/setup.c
@@ -14,21 +14,12 @@
 
 #define PFC_PHCR	0xa400010e
 
-const char *get_system_type(void)
-{
-	return "SHMIN";
-}
-
 static void __init init_shmin_irq(void)
 {
 	ctrl_outw(0x2a00, PFC_PHCR);	// IRQ0-3=IRQ
 	ctrl_outw(0x0aaa, INTC_ICR1);	// IRQ0-3=IRQ-mode,Low-active.
 }
 
-void __init platform_setup(void)
-{
-}
-
 static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
 {
 	static int dummy;
@@ -43,6 +34,7 @@ static void __iomem *shmin_ioport_map(unsigned long port, unsigned int size)
 }
 
 struct sh_machine_vector mv_shmin __initmv = {
+	.mv_name	= "SHMIN",
 	.mv_init_irq	= init_shmin_irq,
 	.mv_ioport_map	= shmin_ioport_map,
 };
diff --git a/arch/sh/boards/snapgear/setup.c b/arch/sh/boards/snapgear/setup.c
index 66ce32f8b13c..f5e98c56b530 100644
--- a/arch/sh/boards/snapgear/setup.c
+++ b/arch/sh/boards/snapgear/setup.c
@@ -81,16 +81,20 @@ static void __init init_snapgear_IRQ(void)
 	make_ipr_irq(IRL3_IRQ, IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY);
 }
 
-const char *get_system_type(void)
+/*
+ * Initialize the board
+ */
+static void __init snapgear_setup(char **cmdline_p)
 {
-	return "SnapGear SecureEdge5410";
+	board_time_init = secureedge5410_rtc_init;
 }
 
 /*
  * The Machine Vector
  */
-
 struct sh_machine_vector mv_snapgear __initmv = {
+	.mv_name		= "SnapGear SecureEdge5410",
+	.mv_setup		= snapgear_setup,
 	.mv_nr_irqs		= 72,
 
 	.mv_inb			= snapgear_inb,
@@ -110,15 +114,3 @@ struct sh_machine_vector mv_snapgear __initmv = {
 	.mv_init_irq		= init_snapgear_IRQ,
 };
 ALIAS_MV(snapgear)
-
-/*
- * Initialize the board
- */
-
-int __init platform_setup(void)
-{
-	board_time_init = secureedge5410_rtc_init;
-
-	return 0;
-}
-
diff --git a/arch/sh/boards/superh/microdev/setup.c b/arch/sh/boards/superh/microdev/setup.c
index 61b402a3f5d7..031c814e6e76 100644
--- a/arch/sh/boards/superh/microdev/setup.c
+++ b/arch/sh/boards/superh/microdev/setup.c
@@ -10,7 +10,6 @@
  * May be copied or modified under the terms of the GNU General Public
  * License.  See linux/COPYING for more information.
  */
-
 #include <linux/init.h>
 #include <linux/platform_device.h>
 #include <linux/ioport.h>
@@ -21,41 +20,6 @@
 
 extern void microdev_heartbeat(void);
 
-/*
- * The Machine Vector
- */
-
-struct sh_machine_vector mv_sh4202_microdev __initmv = {
-	.mv_nr_irqs		= 72,		/* QQQ need to check this - use the MACRO */
-
-	.mv_inb			= microdev_inb,
-	.mv_inw			= microdev_inw,
-	.mv_inl			= microdev_inl,
-	.mv_outb		= microdev_outb,
-	.mv_outw		= microdev_outw,
-	.mv_outl		= microdev_outl,
-
-	.mv_inb_p		= microdev_inb_p,
-	.mv_inw_p		= microdev_inw_p,
-	.mv_inl_p		= microdev_inl_p,
-	.mv_outb_p		= microdev_outb_p,
-	.mv_outw_p		= microdev_outw_p,
-	.mv_outl_p		= microdev_outl_p,
-
-	.mv_insb		= microdev_insb,
-	.mv_insw		= microdev_insw,
-	.mv_insl		= microdev_insl,
-	.mv_outsb		= microdev_outsb,
-	.mv_outsw		= microdev_outsw,
-	.mv_outsl		= microdev_outsl,
-
-	.mv_init_irq		= init_microdev_irq,
-
-#ifdef CONFIG_HEARTBEAT
-	.mv_heartbeat		= microdev_heartbeat,
-#endif
-};
-ALIAS_MV(sh4202_microdev)
 
 /****************************************************************************/
 
@@ -113,11 +77,6 @@ ALIAS_MV(sh4202_microdev)
 	/* assume a Keyboard Controller is present */
 int microdev_kbd_controller_present = 1;
 
-const char *get_system_type(void)
-{
-	return "SH4-202 MicroDev";
-}
-
 static struct resource smc91x_resources[] = {
 	[0] = {
 		.start		= 0x300,
@@ -291,25 +250,9 @@ static int __init microdev_devices_setup(void)
 	return platform_add_devices(microdev_devices, ARRAY_SIZE(microdev_devices));
 }
 
-__initcall(microdev_devices_setup);
-
-void __init platform_setup(void)
-{
-	int * const fpgaRevisionRegister = (int*)(MICRODEV_FPGA_GP_BASE + 0x8ul);
-	const int fpgaRevision = *fpgaRevisionRegister;
-	int * const CacheControlRegister = (int*)CCR;
-
-	printk("SuperH %s board (FPGA rev: 0x%0x, CCR: 0x%0x)\n",
-		get_system_type(), fpgaRevision, *CacheControlRegister);
-}
-
-
-/****************************************************************************/
-
-
-	/*
-	 * Setup for the SMSC FDC37C93xAPM
-	 */
+/*
+ * Setup for the SMSC FDC37C93xAPM
+ */
 static int __init smsc_superio_setup(void)
 {
 
@@ -412,8 +355,52 @@ static int __init smsc_superio_setup(void)
 	return 0;
 }
 
+static void __init microdev_setup(char **cmdline_p)
+{
+	int * const fpgaRevisionRegister = (int*)(MICRODEV_FPGA_GP_BASE + 0x8ul);
+	const int fpgaRevision = *fpgaRevisionRegister;
+	int * const CacheControlRegister = (int*)CCR;
+
+	device_initcall(microdev_devices_setup);
+	device_initcall(smsc_superio_setup);
 
-/* This is grotty, but, because kernel is always referenced on the link line
- * before any devices, this is safe.
+	printk("SuperH %s board (FPGA rev: 0x%0x, CCR: 0x%0x)\n",
+		get_system_type(), fpgaRevision, *CacheControlRegister);
+}
+
+/*
+ * The Machine Vector
  */
-__initcall(smsc_superio_setup);
+struct sh_machine_vector mv_sh4202_microdev __initmv = {
+	.mv_name		= "SH4-202 MicroDev",
+	.mv_setup		= microdev_setup,
+	.mv_nr_irqs		= 72,		/* QQQ need to check this - use the MACRO */
+
+	.mv_inb			= microdev_inb,
+	.mv_inw			= microdev_inw,
+	.mv_inl			= microdev_inl,
+	.mv_outb		= microdev_outb,
+	.mv_outw		= microdev_outw,
+	.mv_outl		= microdev_outl,
+
+	.mv_inb_p		= microdev_inb_p,
+	.mv_inw_p		= microdev_inw_p,
+	.mv_inl_p		= microdev_inl_p,
+	.mv_outb_p		= microdev_outb_p,
+	.mv_outw_p		= microdev_outw_p,
+	.mv_outl_p		= microdev_outl_p,
+
+	.mv_insb		= microdev_insb,
+	.mv_insw		= microdev_insw,
+	.mv_insl		= microdev_insl,
+	.mv_outsb		= microdev_outsb,
+	.mv_outsw		= microdev_outsw,
+	.mv_outsl		= microdev_outsl,
+
+	.mv_init_irq		= init_microdev_irq,
+
+#ifdef CONFIG_HEARTBEAT
+	.mv_heartbeat		= microdev_heartbeat,
+#endif
+};
+ALIAS_MV(sh4202_microdev)
diff --git a/arch/sh/boards/titan/setup.c b/arch/sh/boards/titan/setup.c
index c8b431c1d0fd..52b66d8b8d2a 100644
--- a/arch/sh/boards/titan/setup.c
+++ b/arch/sh/boards/titan/setup.c
@@ -20,19 +20,8 @@ static void __init init_titan_irq(void)
 	make_ipr_irq( TITAN_IRQ_USB,   IRL3_IPR_ADDR, IRL3_IPR_POS, IRL3_PRIORITY); /* PCIRQ3 */
 }
 
-const char *get_system_type(void)
-{
-	return "Titan";
-}
-
-int __init platform_setup(void)
-{
-	printk("%s Platform Setup\n", get_system_type());
-	return 0;
-}
-
 struct sh_machine_vector mv_titan __initmv = {
-	.mv_nr_irqs =	NR_IRQS,
+	.mv_name =	"Titan",
 
 	.mv_inb =	titan_inb,
 	.mv_inw =	titan_inw,
diff --git a/arch/sh/boards/unknown/setup.c b/arch/sh/boards/unknown/setup.c
index c5e4ed10876b..1c941370a2e3 100644
--- a/arch/sh/boards/unknown/setup.c
+++ b/arch/sh/boards/unknown/setup.c
@@ -14,19 +14,8 @@
  */
 #include <linux/init.h>
 #include <asm/machvec.h>
-#include <asm/irq.h>
 
 struct sh_machine_vector mv_unknown __initmv = {
-	.mv_nr_irqs		= NR_IRQS,
+	.mv_name		= "Unknown",
 };
 ALIAS_MV(unknown)
-
-const char *get_system_type(void)
-{
-	return "Unknown";
-}
-
-void __init platform_setup(void)
-{
-}
-
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 86ef17fe48b5..de8df969d6af 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -43,27 +43,14 @@ extern void * __rd_start, * __rd_end;
  * The bigger value means no problem.
  */
 struct sh_cpuinfo boot_cpu_data = { CPU_SH_NONE, 10000000, };
+#ifdef CONFIG_VT
 struct screen_info screen_info;
+#endif
 
 #if defined(CONFIG_SH_UNKNOWN)
 struct sh_machine_vector sh_mv;
 #endif
 
-/* We need this to satisfy some external references. */
-struct screen_info screen_info = {
-        0, 25,                  /* orig-x, orig-y */
-        0,                      /* unused */
-        0,                      /* orig-video-page */
-        0,                      /* orig-video-mode */
-        80,                     /* orig-video-cols */
-        0,0,0,                  /* ega_ax, ega_bx, ega_cx */
-        25,                     /* orig-video-lines */
-        0,                      /* orig-video-isVGA */
-        16                      /* orig-video-points */
-};
-
-extern void platform_setup(void);
-extern char *get_system_type(void);
 extern int root_mountflags;
 
 #define MV_NAME_SIZE 32
@@ -90,29 +77,8 @@ static struct sh_machine_vector* __init get_mv_byname(const char* name);
 
 static char command_line[COMMAND_LINE_SIZE] = { 0, };
 
-struct resource standard_io_resources[] = {
-	{ "dma1", 0x00, 0x1f },
-	{ "pic1", 0x20, 0x3f },
-	{ "timer", 0x40, 0x5f },
-	{ "keyboard", 0x60, 0x6f },
-	{ "dma page reg", 0x80, 0x8f },
-	{ "pic2", 0xa0, 0xbf },
-	{ "dma2", 0xc0, 0xdf },
-	{ "fpu", 0xf0, 0xff }
-};
-
-#define STANDARD_IO_RESOURCES (sizeof(standard_io_resources)/sizeof(struct resource))
-
-/* System RAM - interrupted by the 640kB-1M hole */
-#define code_resource (ram_resources[3])
-#define data_resource (ram_resources[4])
-static struct resource ram_resources[] = {
-	{ "System RAM", 0x000000, 0x09ffff, IORESOURCE_BUSY },
-	{ "System RAM", 0x100000, 0x100000, IORESOURCE_BUSY },
-	{ "Video RAM area", 0x0a0000, 0x0bffff },
-	{ "Kernel code", 0x100000, 0 },
-	{ "Kernel data", 0, 0 }
-};
+static struct resource code_resource = { .name = "Kernel code", };
+static struct resource data_resource = { .name = "Kernel data", };
 
 unsigned long memory_start, memory_end;
 
@@ -255,6 +221,9 @@ static int __init sh_mv_setup(char **cmdline_p)
 	__set_io_port_base(mv_io_base);
 #endif
 
+	if (!sh_mv.mv_nr_irqs)
+		sh_mv.mv_nr_irqs = NR_IRQS;
+
 	return 0;
 }
 
@@ -263,7 +232,6 @@ void __init setup_arch(char **cmdline_p)
 	unsigned long bootmap_size;
 	unsigned long start_pfn, max_pfn, max_low_pfn;
 
-
 #ifdef CONFIG_CMDLINE_BOOL
         strcpy(COMMAND_LINE, CONFIG_CMDLINE);
 #endif
@@ -382,14 +350,14 @@ void __init setup_arch(char **cmdline_p)
 #endif
 
 	/* Perform the machine specific initialisation */
-	platform_setup();
+	if (likely(sh_mv.mv_setup))
+		sh_mv.mv_setup(cmdline_p);
 
 	paging_init();
 }
 
 struct sh_machine_vector* __init get_mv_byname(const char* name)
 {
-	extern int strcasecmp(const char *, const char *);
 	extern long __machvec_start, __machvec_end;
 	struct sh_machine_vector *all_vecs =
 		(struct sh_machine_vector *)&__machvec_start;
@@ -467,7 +435,8 @@ static void show_cpuflags(struct seq_file *m)
 	seq_printf(m, "\n");
 }
 
-static void show_cacheinfo(struct seq_file *m, const char *type, struct cache_info info)
+static void show_cacheinfo(struct seq_file *m, const char *type,
+			   struct cache_info info)
 {
 	unsigned int cache_size;
 
@@ -624,4 +593,3 @@ static int __init kgdb_parse_options(char *options)
 }
 __setup("kgdb=", kgdb_parse_options);
 #endif /* CONFIG_SH_KGDB */
-
diff --git a/include/asm-sh/machvec.h b/include/asm-sh/machvec.h
index 550501fa4fed..70389b72ffef 100644
--- a/include/asm-sh/machvec.h
+++ b/include/asm-sh/machvec.h
@@ -8,17 +8,18 @@
  */
 
 #ifndef _ASM_SH_MACHVEC_H
-#define _ASM_SH_MACHVEC_H 1
+#define _ASM_SH_MACHVEC_H
 
 #include <linux/types.h>
 #include <linux/time.h>
-
 #include <asm/machtypes.h>
 #include <asm/machvec_init.h>
 
 struct device;
 
 struct sh_machine_vector {
+	void (*mv_setup)(char **cmdline_p);
+	const char *mv_name;
 	int mv_nr_irqs;
 
 	u8 (*mv_inb)(unsigned long);
@@ -65,4 +66,6 @@ struct sh_machine_vector {
 
 extern struct sh_machine_vector sh_mv;
 
+#define get_system_type()	sh_mv.mv_name
+
 #endif /* _ASM_SH_MACHVEC_H */
-- 
cgit v1.2.1


From 7a440c950efb5cdc8a05cc6c3ec8fc864b60ef77 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:18:31 +0900
Subject: sh: select CONFIG_EMBEDDED.

This solution isn't very optimal, but it's generaly the behaviour that
we want..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 9e5272e84fd8..1cc5c9b27bfd 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -8,6 +8,7 @@ mainmenu "Linux/SuperH Kernel Configuration"
 config SUPERH
 	bool
 	default y
+	select EMBEDDED
 	help
 	  The SuperH is a RISC processor targeted for use in embedded systems
 	  and consumer electronics; it was also used in the Sega Dreamcast
-- 
cgit v1.2.1


From d153ea88dccf003173315b5d21acabebb897fb4a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:20:16 +0900
Subject: sh: stack debugging support.

This adds a DEBUG_STACK_USAGE and DEBUG_STACKOVERFLOW for SH.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug        | 16 ++++++++++++++++
 arch/sh/kernel/head.S        |  4 +++-
 arch/sh/kernel/irq.c         | 17 ++++++++++++++++-
 arch/sh/kernel/vmlinux.lds.S | 14 ++++++++------
 include/asm-sh/page.h        |  2 +-
 include/asm-sh/thread_info.h | 12 ++++++++----
 6 files changed, 52 insertions(+), 13 deletions(-)

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index 75ed1be8a49e..f0188e626be0 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -30,6 +30,22 @@ config EARLY_PRINTK
 	  when the kernel may crash or hang before the serial console is
 	  initialised. If unsure, say N.
 
+config DEBUG_STACKOVERFLOW
+	bool "Check for stack overflows"
+	depends on DEBUG_KERNEL
+	help
+	  This option will cause messages to be printed if free stack space
+	  drops below a certain limit.
+
+config DEBUG_STACK_USAGE
+	bool "Stack utilization instrumentation"
+	depends on DEBUG_KERNEL
+	help
+	  Enables the display of the minimum amount of free stack which each
+	  task has ever had available in the sysrq-T and sysrq-P debug output.
+
+	  This option will slow down process creation somewhat.
+
 config KGDB
 	bool "Include KGDB kernel debugger"
 	select FRAME_POINTER
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index c5e363872b91..3e7d00b7985a 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -11,6 +11,8 @@
  * Head.S contains the SH exception handlers and startup code.
  */
 #include <linux/linkage.h>
+#include <asm/thread_info.h>
+#include <asm/page.h>
 
 #ifdef CONFIG_CPU_SH4A
 #define SYNCO()		synco
@@ -95,7 +97,7 @@ ENTRY(_stext)
 
 	.balign 4
 1:	.long	0x400080F0		! MD=1, RB=0, BL=0, FD=1, IMASK=0xF
-2:	.long	init_thread_union+8192
+2:	.long	init_thread_union+THREAD_SIZE
 3:	.long	__bss_start
 4:	.long	_end
 5:	.long	start_kernel
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index c2e07f7f3496..7066611aeb72 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -60,7 +60,6 @@ unlock:
 }
 #endif
 
-
 asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 		      unsigned long r6, unsigned long r7,
 		      struct pt_regs regs)
@@ -69,6 +68,22 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 
 	irq_enter();
 
+#ifdef CONFIG_DEBUG_STACKOVERFLOW
+	/* Debugging check for stack overflow: is there less than 1KB free? */
+	{
+		long sp;
+
+		__asm__ __volatile__ ("and r15, %0" :
+					"=r" (sp) : "0" (THREAD_SIZE - 1));
+
+		if (unlikely(sp < (sizeof(struct thread_info) + STACK_WARN))) {
+			printk("do_IRQ: stack overflow: %ld\n",
+			       sp - sizeof(struct thread_info));
+			dump_stack();
+		}
+	}
+#endif
+
 #ifdef CONFIG_CPU_HAS_INTEVT
 	__asm__ __volatile__ (
 #ifdef CONFIG_CPU_HAS_SR_RB
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index eb860c51c697..0220d8a838a7 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -2,6 +2,8 @@
  * ld script to make SuperH Linux kernel
  * Written by Niibe Yutaka
  */
+#include <asm/thread_info.h>
+#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
@@ -40,16 +42,16 @@ SECTIONS
 	*(.data)
 
  	 /* Align the initial ramdisk image (INITRD) on page boundaries. */
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_start = .;
  	 *(.initrd)
- 	 . = ALIGN(4096);
+ 	 . = ALIGN(PAGE_SIZE);
  	 __rd_end = .;
 
 	CONSTRUCTORS
 	}
 
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   .data.page_aligned : { *(.data.idt) }
 
   . = ALIGN(32);
@@ -60,10 +62,10 @@ SECTIONS
 
   _edata = .;			/* End of data section */
 
-  . = ALIGN(8192);		/* init_task */
+  . = ALIGN(THREAD_SIZE);		/* init_task */
   .data.init_task : { *(.data.init_task) }
 
-  . = ALIGN(4096);		/* Init code and data */
+  . = ALIGN(PAGE_SIZE);		/* Init code and data */
   __init_begin = .;
   _sinittext = .;
   .init.text : { *(.init.text) }
@@ -94,7 +96,7 @@ SECTIONS
   __machvec_start = .;
   .init.machvec : { *(.init.machvec) }
   __machvec_end = .;
-  . = ALIGN(4096);
+  . = ALIGN(PAGE_SIZE);
   __init_end = .;
 
   . = ALIGN(4);
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index e9135532d00c..888d6fe0030e 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -16,7 +16,7 @@
 
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT	12
-#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#define PAGE_SIZE	(1 << PAGE_SHIFT)
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 5eb874b065a6..605259f88113 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -29,6 +29,8 @@ struct thread_info {
 #endif
 
 #define PREEMPT_ACTIVE		0x10000000
+#define THREAD_SIZE		(PAGE_SIZE * 2)
+#define STACK_WARN		(THREAD_SIZE / 8)
 
 /*
  * macros/functions for gaining access to the thread information structure
@@ -50,8 +52,6 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
-#define THREAD_SIZE (2*PAGE_SIZE)
-
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
@@ -73,8 +73,12 @@ static inline struct thread_info *current_thread_info(void)
 }
 
 /* thread information allocation */
-#define alloc_thread_info(ti) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
+#ifdef CONFIG_DEBUG_STACK_USAGE
+#define alloc_thread_info(ti)	kzalloc(THREAD_SIZE, GFP_KERNEL)
+#else
+#define alloc_thread_info(ti)	kmalloc(THREAD_SIZE, GFP_KERNEL)
+#endif
+#define free_thread_info(ti)	kfree(ti)
 
 #else /* !__ASSEMBLY__ */
 
-- 
cgit v1.2.1


From 2cb7ce3bb384f30a377f66336c78546b834604df Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:20:58 +0900
Subject: sh: Enable /proc/kcore support.

This was previously unimplemented..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/init.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 2c6dc3d8df2c..ad182b31d846 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -24,7 +24,7 @@
 #include <linux/highmem.h>
 #include <linux/bootmem.h>
 #include <linux/pagemap.h>
-
+#include <linux/proc_fs.h>
 #include <asm/processor.h>
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -223,6 +223,8 @@ void __init paging_init(void)
 	free_area_init_node(0, NODE_DATA(0), zones_size, __MEMORY_START >> PAGE_SHIFT, 0);
 }
 
+static struct kcore_list kcore_mem, kcore_vmalloc;
+
 void __init mem_init(void)
 {
 	extern unsigned long empty_zero_page[1024];
@@ -270,7 +272,12 @@ void __init mem_init(void)
 	datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
 	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;
 
-	printk("Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init)\n",
+	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
+	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
+		   VMALLOC_END - VMALLOC_START);
+
+	printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, "
+	       "%dk reserved, %dk data, %dk init)\n",
 		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
 		max_mapnr << (PAGE_SHIFT-10),
 		codesize >> 10,
-- 
cgit v1.2.1


From a6a31139897a5e539efe7ad3b7bd351fa9673ce8 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:22:14 +0900
Subject: sh: Add support for 4K stacks.

This enables support for 4K stacks on SH.

Currently this depends on DEBUG_KERNEL, but likely all boards
will switch to this as the default in the future.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/Kconfig.debug        |  10 +++
 arch/sh/kernel/cpu/irq/ipr.c |   2 +
 arch/sh/kernel/entry.S       |   4 +-
 arch/sh/kernel/head.S        |   5 +-
 arch/sh/kernel/irq.c         | 153 +++++++++++++++++++++++++++++++++++++++++--
 arch/sh/kernel/traps.c       |  18 ++---
 arch/sh/kernel/vmlinux.lds.S |   1 -
 include/asm-sh/irq.h         |   9 +++
 include/asm-sh/thread_info.h |  12 +++-
 9 files changed, 190 insertions(+), 24 deletions(-)

diff --git a/arch/sh/Kconfig.debug b/arch/sh/Kconfig.debug
index f0188e626be0..48479e014dac 100644
--- a/arch/sh/Kconfig.debug
+++ b/arch/sh/Kconfig.debug
@@ -46,6 +46,16 @@ config DEBUG_STACK_USAGE
 
 	  This option will slow down process creation somewhat.
 
+config 4KSTACKS
+	bool "Use 4Kb for kernel stacks instead of 8Kb"
+	depends on DEBUG_KERNEL
+	help
+	  If you say Y here the kernel will use a 4Kb stacksize for the
+	  kernel stack attached to each process/thread. This facilitates
+	  running more threads on a system and also reduces the pressure
+	  on the VM subsystem for higher order allocations. This option
+	  will also use IRQ stacks to compensate for the reduced stackspace.
+
 config KGDB
 	bool "Include KGDB kernel debugger"
 	select FRAME_POINTER
diff --git a/arch/sh/kernel/cpu/irq/ipr.c b/arch/sh/kernel/cpu/irq/ipr.c
index 4c4fd4118d1a..f785822cd5de 100644
--- a/arch/sh/kernel/cpu/irq/ipr.c
+++ b/arch/sh/kernel/cpu/irq/ipr.c
@@ -190,6 +190,8 @@ void __init init_IRQ(void)
 	/* Perform the machine specific initialisation */
 	if (sh_mv.mv_init_irq != NULL)
 		sh_mv.mv_init_irq();
+
+	irq_ctx_init(smp_processor_id());
 }
 
 #if !defined(CONFIG_CPU_HAS_PINT_IRQ)
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index fd5fe2349f20..fe8221855b28 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -716,8 +716,8 @@ ENTRY(handle_exception)
 	bt/s	1f		! It's a kernel to kernel transition.
 	 mov	r15, k0		! save original stack to k0
 	/* User space to kernel */
-	mov	#0x20, k1
-	shll8	k1		! k1 := 8192 (== THREAD_SIZE)
+	mov	#(THREAD_SIZE >> 8), k1
+	shll8	k1		! k1 := THREAD_SIZE
 	add	current, k1
 	mov	k1, r15		! change to kernel stack
 	!
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 3e7d00b7985a..f5f53d14f245 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -12,7 +12,6 @@
  */
 #include <linux/linkage.h>
 #include <asm/thread_info.h>
-#include <asm/page.h>
 
 #ifdef CONFIG_CPU_SH4A
 #define SYNCO()		synco
@@ -69,8 +68,8 @@ ENTRY(_stext)
 	!
 	mov.l	2f, r0
 	mov	r0, r15		! Set initial r15 (stack pointer)
-	mov	#0x20, r1	!
-	shll8	r1		! r1 = 8192
+	mov	#(THREAD_SIZE >> 8), r1
+	shll8	r1		! r1 = THREAD_SIZE
 	sub	r1, r0		!
 	ldc	r0, r7_bank	! ... and initial thread_info
 
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 7066611aeb72..c7ebd6aec951 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -1,5 +1,4 @@
-/* $Id: irq.c,v 1.20 2004/01/13 05:52:11 kkojima Exp $
- *
+/*
  * linux/arch/sh/kernel/irq.c
  *
  *	Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
@@ -7,13 +6,15 @@
  *
  * SuperH version:  Copyright (C) 1999  Niibe Yutaka
  */
-
 #include <linux/irq.h>
 #include <linux/interrupt.h>
+#include <linux/module.h>
 #include <linux/kernel_stat.h>
 #include <linux/seq_file.h>
 #include <asm/irq.h>
 #include <asm/processor.h>
+#include <asm/uaccess.h>
+#include <asm/thread_info.h>
 #include <asm/cpu/mmu_context.h>
 
 /*
@@ -60,11 +61,27 @@ unlock:
 }
 #endif
 
+#ifdef CONFIG_4KSTACKS
+/*
+ * per-CPU IRQ handling contexts (thread information and stack)
+ */
+union irq_ctx {
+	struct thread_info	tinfo;
+	u32			stack[THREAD_SIZE/sizeof(u32)];
+};
+
+static union irq_ctx *hardirq_ctx[NR_CPUS];
+static union irq_ctx *softirq_ctx[NR_CPUS];
+#endif
+
 asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 		      unsigned long r6, unsigned long r7,
 		      struct pt_regs regs)
 {
 	int irq = r4;
+#ifdef CONFIG_4KSTACKS
+	union irq_ctx *curctx, *irqctx;
+#endif
 
 	irq_enter();
 
@@ -102,7 +119,135 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 #endif
 
 	irq = irq_demux(irq);
-	__do_IRQ(irq, &regs);
+
+#ifdef CONFIG_4KSTACKS
+	curctx = (union irq_ctx *)current_thread_info();
+	irqctx = hardirq_ctx[smp_processor_id()];
+
+	/*
+	 * this is where we switch to the IRQ stack. However, if we are
+	 * already using the IRQ stack (because we interrupted a hardirq
+	 * handler) we can't do that and just have to keep using the
+	 * current stack (which is the irq stack already after all)
+	 */
+	if (curctx != irqctx) {
+		u32 *isp;
+
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+		irqctx->tinfo.task = curctx->tinfo.task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		__asm__ __volatile__ (
+			"mov	%0, r4		\n"
+			"mov	%1, r5		\n"
+			"mov	r15, r9		\n"
+			"jsr	@%2		\n"
+			/* swith to the irq stack */
+			" mov	%3, r15		\n"
+			/* restore the stack (ring zero) */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (irq), "r" (&regs), "r" (__do_IRQ), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "t", "pr"
+		);
+	} else
+#endif
+		__do_IRQ(irq, &regs);
+
 	irq_exit();
+
 	return 1;
 }
+
+#ifdef CONFIG_4KSTACKS
+/*
+ * These should really be __section__(".bss.page_aligned") as well, but
+ * gcc's 3.0 and earlier don't handle that correctly.
+ */
+static char softirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+static char hardirq_stack[NR_CPUS * THREAD_SIZE]
+		__attribute__((__aligned__(THREAD_SIZE)));
+
+/*
+ * allocate per-cpu stacks for hardirq and for softirq processing
+ */
+void irq_ctx_init(int cpu)
+{
+	union irq_ctx *irqctx;
+
+	if (hardirq_ctx[cpu])
+		return;
+
+	irqctx = (union irq_ctx *)&hardirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= HARDIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	hardirq_ctx[cpu] = irqctx;
+
+	irqctx = (union irq_ctx *)&softirq_stack[cpu * THREAD_SIZE];
+	irqctx->tinfo.task		= NULL;
+	irqctx->tinfo.exec_domain	= NULL;
+	irqctx->tinfo.cpu		= cpu;
+	irqctx->tinfo.preempt_count	= SOFTIRQ_OFFSET;
+	irqctx->tinfo.addr_limit	= MAKE_MM_SEG(0);
+
+	softirq_ctx[cpu] = irqctx;
+
+	printk("CPU %u irqstacks, hard=%p soft=%p\n",
+		cpu, hardirq_ctx[cpu], softirq_ctx[cpu]);
+}
+
+void irq_ctx_exit(int cpu)
+{
+	hardirq_ctx[cpu] = NULL;
+}
+
+extern asmlinkage void __do_softirq(void);
+
+asmlinkage void do_softirq(void)
+{
+	unsigned long flags;
+	struct thread_info *curctx;
+	union irq_ctx *irqctx;
+	u32 *isp;
+
+	if (in_interrupt())
+		return;
+
+	local_irq_save(flags);
+
+	if (local_softirq_pending()) {
+		curctx = current_thread_info();
+		irqctx = softirq_ctx[smp_processor_id()];
+		irqctx->tinfo.task = curctx->task;
+		irqctx->tinfo.previous_sp = current_stack_pointer;
+
+		/* build the stack frame on the softirq stack */
+		isp = (u32 *)((char *)irqctx + sizeof(*irqctx));
+
+		__asm__ __volatile__ (
+			"mov	r15, r9		\n"
+			"jsr	@%0		\n"
+			/* switch to the softirq stack */
+			" mov	%1, r15		\n"
+			/* restore the thread stack */
+			"mov	r9, r15		\n"
+			: /* no outputs */
+			: "r" (__do_softirq), "r" (isp)
+			/* XXX: A somewhat excessive clobber list? -PFM */
+			: "memory", "r0", "r1", "r2", "r3", "r4",
+			  "r5", "r6", "r7", "r8", "r9", "r15", "t", "pr"
+		);
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL(do_softirq);
+#endif
diff --git a/arch/sh/kernel/traps.c b/arch/sh/kernel/traps.c
index 95c810b3c97e..c2c597e09482 100644
--- a/arch/sh/kernel/traps.c
+++ b/arch/sh/kernel/traps.c
@@ -741,20 +741,12 @@ void show_stack(struct task_struct *tsk, unsigned long *sp)
 	unsigned long module_end = VMALLOC_END;
 	int i = 1;
 
-	if (tsk && !sp) {
+	if (!tsk)
+		tsk = current;
+	if (tsk == current)
+		sp = (unsigned long *)current_stack_pointer;
+	else
 		sp = (unsigned long *)tsk->thread.sp;
-	}
-
-	if (!sp) {
-		__asm__ __volatile__ (
-			"mov r15, %0\n\t"
-			"stc r7_bank, %1\n\t"
-			: "=r" (module_start),
-			  "=r" (module_end)
-		);
-		
-		sp = (unsigned long *)module_start;
-	}
 
 	stack = sp;
 
diff --git a/arch/sh/kernel/vmlinux.lds.S b/arch/sh/kernel/vmlinux.lds.S
index 0220d8a838a7..5eb930918186 100644
--- a/arch/sh/kernel/vmlinux.lds.S
+++ b/arch/sh/kernel/vmlinux.lds.S
@@ -3,7 +3,6 @@
  * Written by Niibe Yutaka
  */
 #include <asm/thread_info.h>
-#include <asm/page.h>
 #include <asm-generic/vmlinux.lds.h>
 
 #ifdef CONFIG_CPU_LITTLE_ENDIAN
diff --git a/include/asm-sh/irq.h b/include/asm-sh/irq.h
index dd05e102fc0e..0e5f365aff70 100644
--- a/include/asm-sh/irq.h
+++ b/include/asm-sh/irq.h
@@ -719,6 +719,15 @@ static inline int generic_irq_demux(int irq)
 #define irq_canonicalize(irq)	(irq)
 #define irq_demux(irq)		__irq_demux(sh_mv.mv_irq_demux(irq))
 
+#ifdef CONFIG_4KSTACKS
+extern void irq_ctx_init(int cpu);
+extern void irq_ctx_exit(int cpu);
+# define __ARCH_HAS_DO_SOFTIRQ
+#else
+# define irq_ctx_init(cpu) do { } while (0)
+# define irq_ctx_exit(cpu) do { } while (0)
+#endif
+
 #if defined(CONFIG_CPU_SUBTYPE_SH73180)
 #include <asm/irq-sh73180.h>
 #endif
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index 605259f88113..3ebc3f9039eb 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -9,8 +9,8 @@
  *  Copyright (C) 2002  David Howells (dhowells@redhat.com)
  *  - Incorporating suggestions made by Linus Torvalds and Dave Miller
  */
-
 #ifdef __KERNEL__
+#include <asm/page.h>
 
 #ifndef __ASSEMBLY__
 #include <asm/processor.h>
@@ -23,13 +23,20 @@ struct thread_info {
 	int			preempt_count; /* 0 => preemptable, <0 => BUG */
 	mm_segment_t		addr_limit;	/* thread address space */
 	struct restart_block	restart_block;
+	unsigned long		previous_sp;	/* sp of previous stack in case
+						   of nested IRQ stacks */
 	__u8			supervisor_stack[0];
 };
 
 #endif
 
 #define PREEMPT_ACTIVE		0x10000000
+
+#ifdef CONFIG_4KSTACKS
+#define THREAD_SIZE		(PAGE_SIZE)
+#else
 #define THREAD_SIZE		(PAGE_SIZE * 2)
+#endif
 #define STACK_WARN		(THREAD_SIZE / 8)
 
 /*
@@ -52,6 +59,9 @@ struct thread_info {
 #define init_thread_info	(init_thread_union.thread_info)
 #define init_stack		(init_thread_union.stack)
 
+/* how to get the current stack pointer from C */
+register unsigned long current_stack_pointer asm("r15") __attribute_used__;
+
 /* how to get the thread information struct from C */
 static inline struct thread_info *current_thread_info(void)
 {
-- 
cgit v1.2.1


From 315bb96824149614efe4844ded077a13fc908880 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:22:53 +0900
Subject: sh: CPU flags in AT_HWCAP in ELF auxvt.

Encode processor flags in AT_HWCAP in the ELF auxiliary vector.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c |  8 +++++---
 include/asm-sh/cpu-features.h  | 15 +++++++++++++++
 include/asm-sh/elf.h           | 10 ++++++----
 include/asm-sh/processor.h     | 12 +-----------
 4 files changed, 27 insertions(+), 18 deletions(-)
 create mode 100644 include/asm-sh/cpu-features.h

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 6e8a2b5268e8..0e65aa6ddcaa 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -76,6 +76,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->type = CPU_SH73180;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+		cpu_data->flags |= CPU_HAS_LLSC;
 		break;
 	case 0x2001:
 	case 0x2004:
@@ -83,7 +84,7 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		cpu_data->flags |= CPU_HAS_FPU;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_LLSC;
 		break;
 	case 0x2006:
 	case 0x200A:
@@ -95,13 +96,15 @@ int __init detect_cpu_and_cache_system(void)
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
 
-		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER;
+		cpu_data->flags |= CPU_HAS_FPU | CPU_HAS_PERF_COUNTER |
+				   CPU_HAS_LLSC;
 		break;
 	case 0x3000:
 	case 0x3003:
 		cpu_data->type = CPU_SH7343;
 		cpu_data->icache.ways = 4;
 		cpu_data->dcache.ways = 4;
+		cpu_data->flags |= CPU_HAS_LLSC;
 		break;
 	case 0x8000:
 		cpu_data->type = CPU_ST40RA;
@@ -180,4 +183,3 @@ int __init detect_cpu_and_cache_system(void)
 
 	return 0;
 }
-
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
new file mode 100644
index 000000000000..e398947ec01d
--- /dev/null
+++ b/include/asm-sh/cpu-features.h
@@ -0,0 +1,15 @@
+#ifndef __ASM_SH_CPU_FEATURES_H
+#define __ASM_SH_CPU_FEATURES_H
+
+/*
+ * Processor flags
+ */
+#define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
+#define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
+#define CPU_HAS_MMU_PAGE_ASSOC	0x0004	/* SH3: TLB way selection bit support */
+#define CPU_HAS_DSP		0x0008	/* SH-DSP: DSP support */
+#define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
+#define CPU_HAS_PTEA		0x0020	/* PTEA register */
+#define CPU_HAS_LLSC		0x0040	/* movli.l/movco.l */
+
+#endif /* __ASM_SH_CPU_FEATURES_H */
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index 1b63dfeea4f2..cc8e5e767345 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -1,6 +1,11 @@
 #ifndef __ASM_SH_ELF_H
 #define __ASM_SH_ELF_H
 
+#include <asm/processor.h>
+#include <asm/auxvec.h>
+#include <asm/ptrace.h>
+#include <asm/user.h>
+
 /* SH relocation types  */
 #define	R_SH_NONE		0
 #define	R_SH_DIR32		1
@@ -46,9 +51,6 @@
  * ELF register definitions..
  */
 
-#include <asm/ptrace.h>
-#include <asm/user.h>
-
 typedef unsigned long elf_greg_t;
 
 #define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t))
@@ -91,7 +93,7 @@ typedef struct user_fpu_struct elf_fpregset_t;
    instruction set this CPU supports.  This could be done in user space,
    but it's not easy, and we've already done it here.  */
 
-#define ELF_HWCAP	(0)
+#define ELF_HWCAP	(boot_cpu_data.flags)
 
 /* This yields a string that ld.so will use to load implementation
    specific libraries for optimization.  This is more specific in
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index 3b3ef4f2bf31..bdd472705546 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -14,6 +14,7 @@
 #include <asm/types.h>
 #include <asm/cache.h>
 #include <asm/ptrace.h>
+#include <asm/cpu-features.h>
 
 /*
  * Default implementation of macro that returns current
@@ -127,17 +128,6 @@ union sh_fpu_union {
 	struct sh_fpu_soft_struct soft;
 };
 
-/*
- * Processor flags
- */
-
-#define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
-#define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
-#define CPU_HAS_MMU_PAGE_ASSOC	0x0004	/* SH3: TLB way selection bit support */
-#define CPU_HAS_DSP		0x0008	/* SH-DSP: DSP support */
-#define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
-#define CPU_HAS_PTEA		0x0020	/* PTEA register */
-
 struct thread_struct {
 	unsigned long sp;
 	unsigned long pc;
-- 
cgit v1.2.1


From 2220d164933a8776d1336c814e3c2e5573256d34 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:24:28 +0900
Subject: sh: Report movli.l/movco.l capabilities.

Add llsc to cpu_flags[] and comment cpu-features.h.

Signed-off-by: Jamie Lenehan <nynaeve@twibble.org>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/setup.c        | 4 +++-
 include/asm-sh/cpu-features.h | 8 ++++++++
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index de8df969d6af..6810de3f8ed2 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -413,8 +413,10 @@ const char *get_cpu_subtype(void)
 }
 
 #ifdef CONFIG_PROC_FS
+/* Symbolic CPU flags, keep in sync with asm/cpu-features.h */
 static const char *cpu_flags[] = {
-	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr", "ptea", NULL
+	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr",
+	"ptea", "llsc", NULL
 };
 
 static void show_cpuflags(struct seq_file *m)
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
index e398947ec01d..e1260aae3ee3 100644
--- a/include/asm-sh/cpu-features.h
+++ b/include/asm-sh/cpu-features.h
@@ -3,6 +3,14 @@
 
 /*
  * Processor flags
+ *
+ * Note: When adding a new flag, keep cpu_flags[] in
+ * arch/sh/kernel/setup.c in sync so symbolic name
+ * mapping of the processor flags has a chance of being
+ * reasonably accurate.
+ *
+ * These flags are also available through the ELF
+ * auxiliary vector as AT_HWCAP.
  */
 #define CPU_HAS_FPU		0x0001	/* Hardware FPU support */
 #define CPU_HAS_P2_FLUSH_BUG	0x0002	/* Need to flush the cache in P2 area */
-- 
cgit v1.2.1


From 05ae91585167410dadd1bc8f2e207a062e638a16 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:25:24 +0900
Subject: sh: Optimized readsl()/writesl() support.

Implement optimized copies of readsl()/writesl().

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/io.c | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 include/asm-sh/io.h |  6 +++++
 2 files changed, 73 insertions(+)

diff --git a/arch/sh/kernel/io.c b/arch/sh/kernel/io.c
index 71c9fde2fd90..501fe03e3715 100644
--- a/arch/sh/kernel/io.c
+++ b/arch/sh/kernel/io.c
@@ -61,6 +61,73 @@ void memset_io(volatile void __iomem *dst, int c, unsigned long count)
 }
 EXPORT_SYMBOL(memset_io);
 
+void __raw_readsl(unsigned long addr, void *datap, int len)
+{
+	u32 *data;
+
+	for (data = datap; (len != 0) && (((u32)data & 0x1f) != 0); len--)
+		*data++ = ctrl_inl(addr);
+
+	if (likely(len >= (0x20 >> 2))) {
+		int tmp2, tmp3, tmp4, tmp5, tmp6;
+
+		__asm__ __volatile__(
+			"1:			\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	@%7, %2		\n\t"
+#ifdef CONFIG_CPU_SH4
+			"movca.l r0, @%0	\n\t"
+#else
+			"mov.l	r0, @%0		\n\t"
+#endif
+			"mov.l	@%7, %3		\n\t"
+			"mov.l	@%7, %4		\n\t"
+			"mov.l	@%7, %5		\n\t"
+			"mov.l	@%7, %6		\n\t"
+			"mov.l	@%7, r7		\n\t"
+			"mov.l	@%7, r0		\n\t"
+			"mov.l	%2, @(0x04,%0)	\n\t"
+			"mov	#0x20>>2, %2	\n\t"
+			"mov.l	%3, @(0x08,%0)	\n\t"
+			"sub	%2, %1		\n\t"
+			"mov.l	%4, @(0x0c,%0)	\n\t"
+			"cmp/hi	%1, %2		! T if 32 > len	\n\t"
+			"mov.l	%5, @(0x10,%0)	\n\t"
+			"mov.l	%6, @(0x14,%0)	\n\t"
+			"mov.l	r7, @(0x18,%0)	\n\t"
+			"mov.l	r0, @(0x1c,%0)	\n\t"
+			"bf.s	1b		\n\t"
+			" add	#0x20, %0	\n\t"
+			: "=&r" (data), "=&r" (len),
+			  "=&r" (tmp2), "=&r" (tmp3), "=&r" (tmp4),
+			  "=&r" (tmp5), "=&r" (tmp6)
+			: "r"(addr), "0" (data), "1" (len)
+			: "r0", "r7", "t", "memory");
+	}
+
+	for (; len != 0; len--)
+		*data++ = ctrl_inl(addr);
+}
+EXPORT_SYMBOL(__raw_readsl);
+
+void __raw_writesl(unsigned long addr, const void *data, int len)
+{
+	if (likely(len != 0)) {
+		int tmp1;
+
+		__asm__ __volatile__ (
+			"1:				\n\t"
+			"mov.l	@%0+, %1	\n\t"
+			"dt		%3		\n\t"
+			"bf.s		1b		\n\t"
+			" mov.l	%1, @%4		\n\t"
+			: "=&r" (data), "=&r" (tmp1)
+			: "0" (data), "r" (len), "r"(addr)
+			: "t", "memory");
+	}
+}
+EXPORT_SYMBOL(__raw_writesl);
+
 void __iomem *ioport_map(unsigned long port, unsigned int nr)
 {
 	return sh_mv.mv_ioport_map(port, nr);
diff --git a/include/asm-sh/io.h b/include/asm-sh/io.h
index 377160b86295..ed12d38e8c00 100644
--- a/include/asm-sh/io.h
+++ b/include/asm-sh/io.h
@@ -107,6 +107,9 @@
 #define __raw_writew(v, a)	__writew(v, (void __iomem *)(a))
 #define __raw_writel(v, a)	__writel(v, (void __iomem *)(a))
 
+void __raw_writesl(unsigned long addr, const void *data, int longlen);
+void __raw_readsl(unsigned long addr, void *data, int longlen);
+
 /*
  * The platform header files may define some of these macros to use
  * the inlined versions where appropriate.  These macros may also be
@@ -132,6 +135,9 @@
 # define writel(v,a)	({ __raw_writel((v),(a)); mb(); })
 #endif
 
+#define writesl __raw_writesl
+#define readsl  __raw_readsl
+
 #define readb_relaxed(a) readb(a)
 #define readw_relaxed(a) readw(a)
 #define readl_relaxed(a) readl(a)
-- 
cgit v1.2.1


From 9d549a7d8ef71f684a35cf1e438543957cf81d12 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:26:05 +0900
Subject: sh: Update kexec support for API changes.

This was falling a bit behind..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/machine_kexec.c |  6 ------
 arch/sh/kernel/process.c       | 10 ----------
 include/asm-sh/kexec.h         |  9 +++------
 3 files changed, 3 insertions(+), 22 deletions(-)

diff --git a/arch/sh/kernel/machine_kexec.c b/arch/sh/kernel/machine_kexec.c
index 6bcd8d92399f..08587cdb64d6 100644
--- a/arch/sh/kernel/machine_kexec.c
+++ b/arch/sh/kernel/machine_kexec.c
@@ -29,12 +29,6 @@ extern const unsigned char relocate_new_kernel[];
 extern const unsigned int relocate_new_kernel_size;
 extern void *gdb_vbr_vector;
 
-/*
- * Provide a dummy crash_notes definition while crash dump arrives to ppc.
- * This prevents breakage of crash_notes attribute in kernel/ksysfs.c.
- */
-void *crash_notes = NULL;
-
 void machine_shutdown(void)
 {
 }
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 2167746e88f1..630ec1af2483 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -81,16 +81,6 @@ void cpu_idle(void)
 
 void machine_restart(char * __unused)
 {
-
-#ifdef CONFIG_KEXEC
-	struct kimage *image;
-	image = xchg(&kexec_image, 0);
-	if (image) {
-		machine_shutdown();
-		machine_kexec(image);
-	}
-#endif
-
 	/* SR.BL=1 and invoke address error to let CPU reset (manual reset) */
 	asm volatile("ldc %0, sr\n\t"
 		     "mov.l @%1, %0" : : "r" (0x10000000), "r" (0x80000001));
diff --git a/include/asm-sh/kexec.h b/include/asm-sh/kexec.h
index a5f85e9e428d..9d235af20cdd 100644
--- a/include/asm-sh/kexec.h
+++ b/include/asm-sh/kexec.h
@@ -25,11 +25,8 @@
 
 #define MAX_NOTE_BYTES 1024
 
-#ifndef __ASSEMBLY__
-
-extern void machine_shutdown(void);
-extern void *crash_notes;
-
-#endif /* __ASSEMBLY__ */
+/* Provide a dummy definition to avoid build failures. */
+static inline void crash_setup_regs(struct pt_regs *newregs,
+					struct pt_regs *oldregs) { }
 
 #endif /* _SH_KEXEC_H */
-- 
cgit v1.2.1


From 72c35543f8cf1316773ffbd9619575bb84ac44fb Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:27:43 +0900
Subject: sh: Support for L2 cache on newer SH-4A CPUs.

This implements preliminary support for the L2 caches found
on newer SH-4A CPUs.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c | 47 +++++++++++++++++++++++++++++++++++++++++-
 arch/sh/kernel/setup.c         |  6 +++++-
 include/asm-sh/cpu-features.h  |  1 +
 include/asm-sh/processor.h     |  9 ++++----
 4 files changed, 57 insertions(+), 6 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index 0e65aa6ddcaa..bee00cac0b16 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -29,7 +29,7 @@ int __init detect_cpu_and_cache_system(void)
 		[9] = (1 << 16)
 	};
 
-	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffff;
+	pvr = (ctrl_inl(CCN_PVR) >> 8) & 0xffffff;
 	prr = (ctrl_inl(CCN_PRR) >> 4) & 0xff;
 	cvr = (ctrl_inl(CCN_CVR));
 
@@ -53,6 +53,26 @@ int __init detect_cpu_and_cache_system(void)
 	cpu_data->dcache.ways		= 1;
 	cpu_data->dcache.linesz		= L1_CACHE_BYTES;
 
+	/*
+	 * Setup some generic flags we can probe
+	 * (L2 and DSP detection only work on SH-4A)
+	 */
+	if (((pvr >> 16) & 0xff) == 0x10) {
+		if ((cvr & 0x02000000) == 0)
+			cpu_data->flags |= CPU_HAS_L2_CACHE;
+		if ((cvr & 0x10000000) == 0)
+			cpu_data->flags |= CPU_HAS_DSP;
+
+		cpu_data->flags |= CPU_HAS_LLSC;
+	}
+
+	/* FPU detection works for everyone */
+	if ((cvr & 0x20000000) == 1)
+		cpu_data->flags |= CPU_HAS_FPU;
+
+	/* Mask off the upper chip ID */
+	pvr &= 0xffff;
+
 	/*
 	 * Probe the underlying processor version/revision and
 	 * adjust cpu_data setup accordingly.
@@ -181,5 +201,30 @@ int __init detect_cpu_and_cache_system(void)
 	cpu_data->dcache.way_size = cpu_data->dcache.sets *
 				    cpu_data->dcache.linesz;
 
+	/*
+	 * Setup the L2 cache desc
+	 *
+	 * SH-4A's have an optional PIPT L2.
+	 */
+	if (cpu_data->flags & CPU_HAS_L2_CACHE) {
+		/*
+		 * Size calculation is much more sensible
+		 * than it is for the L1.
+		 *
+		 * Sizes are 128KB, 258KB, 512KB, and 1MB.
+		 */
+		size = (cvr & 0xf) << 17;
+
+		BUG_ON(!size);
+
+		cpu_data->scache.way_incr	= (1 << 16);
+		cpu_data->scache.entry_shift	= 5;
+		cpu_data->scache.entry_mask	= 0xffe0;
+		cpu_data->scache.ways		= 4;
+		cpu_data->scache.linesz		= L1_CACHE_BYTES;
+		cpu_data->scache.sets		= size /
+			(cpu_data->scache.linesz * cpu_data->scache.ways);
+	}
+
 	return 0;
 }
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index 6810de3f8ed2..5f587332234a 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -416,7 +416,7 @@ const char *get_cpu_subtype(void)
 /* Symbolic CPU flags, keep in sync with asm/cpu-features.h */
 static const char *cpu_flags[] = {
 	"none", "fpu", "p2flush", "mmuassoc", "dsp", "perfctr",
-	"ptea", "llsc", NULL
+	"ptea", "llsc", "l2", NULL
 };
 
 static void show_cpuflags(struct seq_file *m)
@@ -480,6 +480,10 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		show_cacheinfo(m, "dcache", boot_cpu_data.dcache);
 	}
 
+	/* Optional secondary cache */
+	if (boot_cpu_data.flags & CPU_HAS_L2_CACHE)
+		show_cacheinfo(m, "scache", boot_cpu_data.scache);
+
 	seq_printf(m, "bogomips\t: %lu.%02lu\n",
 		     boot_cpu_data.loops_per_jiffy/(500000/HZ),
 		     (boot_cpu_data.loops_per_jiffy/(5000/HZ)) % 100);
diff --git a/include/asm-sh/cpu-features.h b/include/asm-sh/cpu-features.h
index e1260aae3ee3..4bccd7c032f9 100644
--- a/include/asm-sh/cpu-features.h
+++ b/include/asm-sh/cpu-features.h
@@ -19,5 +19,6 @@
 #define CPU_HAS_PERF_COUNTER	0x0010	/* Hardware performance counters */
 #define CPU_HAS_PTEA		0x0020	/* PTEA register */
 #define CPU_HAS_LLSC		0x0040	/* movli.l/movco.l */
+#define CPU_HAS_L2_CACHE	0x0080	/* Secondary cache / URAM */
 
 #endif /* __ASM_SH_CPU_FEATURES_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index bdd472705546..b7cba4e91a72 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -54,14 +54,15 @@ enum cpu_type {
 };
 
 struct sh_cpuinfo {
-	enum cpu_type type;
+	unsigned int type;
 	unsigned long loops_per_jiffy;
 
-	struct cache_info icache;
-	struct cache_info dcache;
+	struct cache_info icache;	/* Primary I-cache */
+	struct cache_info dcache;	/* Primary D-cache */
+	struct cache_info scache;	/* Secondary cache */
 
 	unsigned long flags;
-};
+} __attribute__ ((aligned(SMP_CACHE_BYTES)));
 
 extern struct sh_cpuinfo boot_cpu_data;
 
-- 
cgit v1.2.1


From d15f456043175bdf3464514b92a825b88d0546ae Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:28:34 +0900
Subject: sh: More intelligent entry_mask/way_size calculation.

Figure out the cache desc entry_mask at runtime, and remove
hard-coded assumption about the cacheline size.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/sh4/probe.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/arch/sh/kernel/cpu/sh4/probe.c b/arch/sh/kernel/cpu/sh4/probe.c
index bee00cac0b16..c294de1e14a3 100644
--- a/arch/sh/kernel/cpu/sh4/probe.c
+++ b/arch/sh/kernel/cpu/sh4/probe.c
@@ -38,7 +38,6 @@ int __init detect_cpu_and_cache_system(void)
 	 */
 	cpu_data->icache.way_incr	= (1 << 13);
 	cpu_data->icache.entry_shift	= 5;
-	cpu_data->icache.entry_mask	= 0x1fe0;
 	cpu_data->icache.sets		= 256;
 	cpu_data->icache.ways		= 1;
 	cpu_data->icache.linesz		= L1_CACHE_BYTES;
@@ -48,7 +47,6 @@ int __init detect_cpu_and_cache_system(void)
 	 */
 	cpu_data->dcache.way_incr	= (1 << 14);
 	cpu_data->dcache.entry_shift	= 5;
-	cpu_data->dcache.entry_mask	= 0x3fe0;
 	cpu_data->dcache.sets		= 512;
 	cpu_data->dcache.ways		= 1;
 	cpu_data->dcache.linesz		= L1_CACHE_BYTES;
@@ -183,21 +181,26 @@ int __init detect_cpu_and_cache_system(void)
 		size = sizes[(cvr >> 20) & 0xf];
 		cpu_data->icache.way_incr	= (size >> 1);
 		cpu_data->icache.sets		= (size >> 6);
-		cpu_data->icache.entry_mask	=
-			(cpu_data->icache.way_incr - (1 << 5));
+
 	}
 
+	/* Setup the rest of the I-cache info */
+	cpu_data->icache.entry_mask = cpu_data->icache.way_incr -
+				      cpu_data->icache.linesz;
+
 	cpu_data->icache.way_size = cpu_data->icache.sets *
 				    cpu_data->icache.linesz;
 
+	/* And the rest of the D-cache */
 	if (cpu_data->dcache.ways > 1) {
 		size = sizes[(cvr >> 16) & 0xf];
 		cpu_data->dcache.way_incr	= (size >> 1);
 		cpu_data->dcache.sets		= (size >> 6);
-		cpu_data->dcache.entry_mask	=
-			(cpu_data->dcache.way_incr - (1 << 5));
 	}
 
+	cpu_data->dcache.entry_mask = cpu_data->dcache.way_incr -
+				      cpu_data->dcache.linesz;
+
 	cpu_data->dcache.way_size = cpu_data->dcache.sets *
 				    cpu_data->dcache.linesz;
 
@@ -219,11 +222,14 @@ int __init detect_cpu_and_cache_system(void)
 
 		cpu_data->scache.way_incr	= (1 << 16);
 		cpu_data->scache.entry_shift	= 5;
-		cpu_data->scache.entry_mask	= 0xffe0;
 		cpu_data->scache.ways		= 4;
 		cpu_data->scache.linesz		= L1_CACHE_BYTES;
+		cpu_data->scache.entry_mask	=
+			(cpu_data->scache.way_incr - cpu_data->scache.linesz);
 		cpu_data->scache.sets		= size /
 			(cpu_data->scache.linesz * cpu_data->scache.ways);
+		cpu_data->scache.way_size	=
+			(cpu_data->scache.sets * cpu_data->scache.linesz);
 	}
 
 	return 0;
-- 
cgit v1.2.1


From 28ccf7f91b1ac42ee1f18480a69d2a7486b625ce Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:30:07 +0900
Subject: sh: Selective flush_cache_mm() flushing.

flush_cache_mm() wraps in to flush_cache_all(), which is rather
excessive given that the number of PTEs within the specified context
are generally quite low.  Optimize for walking the mm's VMA list and
selectively flushing the VMA ranges from the dcache. Invalidate the
icache only if a VMA sets VM_EXEC.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c | 220 +++++++++++++++++++++++++++++--------------------
 1 file changed, 130 insertions(+), 90 deletions(-)

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index 2203bd6aadb3..aa4f62f0e374 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -2,29 +2,31 @@
  * arch/sh/mm/cache-sh4.c
  *
  * Copyright (C) 1999, 2000, 2002  Niibe Yutaka
- * Copyright (C) 2001, 2002, 2003, 2004, 2005  Paul Mundt
+ * Copyright (C) 2001 - 2006  Paul Mundt
  * Copyright (C) 2003  Richard Curnow
  *
  * This file is subject to the terms and conditions of the GNU General Public
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  */
-
 #include <linux/init.h>
-#include <linux/mman.h>
 #include <linux/mm.h>
-#include <linux/threads.h>
 #include <asm/addrspace.h>
-#include <asm/page.h>
 #include <asm/pgtable.h>
 #include <asm/processor.h>
 #include <asm/cache.h>
 #include <asm/io.h>
-#include <asm/uaccess.h>
 #include <asm/pgalloc.h>
 #include <asm/mmu_context.h>
 #include <asm/cacheflush.h>
 
+/*
+ * The maximum number of pages we support up to when doing ranged dcache
+ * flushing. Anything exceeding this will simply flush the dcache in its
+ * entirety.
+ */
+#define MAX_DCACHE_PAGES	64	/* XXX: Tune for ways */
+
 static void __flush_dcache_segment_1way(unsigned long start,
 					unsigned long extent);
 static void __flush_dcache_segment_2way(unsigned long start,
@@ -219,14 +221,14 @@ void flush_cache_sigtramp(unsigned long addr)
 static inline void flush_cache_4096(unsigned long start,
 				    unsigned long phys)
 {
-	unsigned long flags;
-
 	/*
 	 * All types of SH-4 require PC to be in P2 to operate on the I-cache.
 	 * Some types of SH-4 require PC to be in P2 to operate on the D-cache.
 	 */
-	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG)
-	   || start < CACHE_OC_ADDRESS_ARRAY) {
+	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG) ||
+	    (start < CACHE_OC_ADDRESS_ARRAY)) {
+		unsigned long flags;
+
 		local_irq_save(flags);
 		__flush_cache_4096(start | SH_CACHE_ASSOC,
 				   P1SEGADDR(phys), 0x20000000);
@@ -257,6 +259,7 @@ void flush_dcache_page(struct page *page)
 	wmb();
 }
 
+/* TODO: Selective icache invalidation through IC address array.. */
 static inline void flush_icache_all(void)
 {
 	unsigned long flags, ccr;
@@ -290,19 +293,121 @@ void flush_cache_all(void)
 	flush_icache_all();
 }
 
+static void __flush_cache_mm(struct mm_struct *mm, unsigned long start,
+			     unsigned long end)
+{
+	unsigned long d = 0, p = start & PAGE_MASK;
+	unsigned long alias_mask = cpu_data->dcache.alias_mask;
+	unsigned long n_aliases = cpu_data->dcache.n_aliases;
+	unsigned long select_bit;
+	unsigned long all_aliases_mask;
+	unsigned long addr_offset;
+	pgd_t *dir;
+	pmd_t *pmd;
+	pud_t *pud;
+	pte_t *pte;
+	int i;
+
+	dir = pgd_offset(mm, p);
+	pud = pud_offset(dir, p);
+	pmd = pmd_offset(pud, p);
+	end = PAGE_ALIGN(end);
+
+	all_aliases_mask = (1 << n_aliases) - 1;
+
+	do {
+		if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd))) {
+			p &= PMD_MASK;
+			p += PMD_SIZE;
+			pmd++;
+
+			continue;
+		}
+
+		pte = pte_offset_kernel(pmd, p);
+
+		do {
+			unsigned long phys;
+			pte_t entry = *pte;
+
+			if (!(pte_val(entry) & _PAGE_PRESENT)) {
+				pte++;
+				p += PAGE_SIZE;
+				continue;
+			}
+
+			phys = pte_val(entry) & PTE_PHYS_MASK;
+
+			if ((p ^ phys) & alias_mask) {
+				d |= 1 << ((p & alias_mask) >> PAGE_SHIFT);
+				d |= 1 << ((phys & alias_mask) >> PAGE_SHIFT);
+
+				if (d == all_aliases_mask)
+					goto loop_exit;
+			}
+
+			pte++;
+			p += PAGE_SIZE;
+		} while (p < end && ((unsigned long)pte & ~PAGE_MASK));
+		pmd++;
+	} while (p < end);
+
+loop_exit:
+	addr_offset = 0;
+	select_bit = 1;
+
+	for (i = 0; i < n_aliases; i++) {
+		if (d & select_bit) {
+			(*__flush_dcache_segment_fn)(addr_offset, PAGE_SIZE);
+			wmb();
+		}
+
+		select_bit <<= 1;
+		addr_offset += PAGE_SIZE;
+	}
+}
+
+/*
+ * Note : (RPC) since the caches are physically tagged, the only point
+ * of flush_cache_mm for SH-4 is to get rid of aliases from the
+ * D-cache.  The assumption elsewhere, e.g. flush_cache_range, is that
+ * lines can stay resident so long as the virtual address they were
+ * accessed with (hence cache set) is in accord with the physical
+ * address (i.e. tag).  It's no different here.  So I reckon we don't
+ * need to flush the I-cache, since aliases don't matter for that.  We
+ * should try that.
+ *
+ * Caller takes mm->mmap_sem.
+ */
 void flush_cache_mm(struct mm_struct *mm)
 {
 	/*
-	 * Note : (RPC) since the caches are physically tagged, the only point
-	 * of flush_cache_mm for SH-4 is to get rid of aliases from the
-	 * D-cache.  The assumption elsewhere, e.g. flush_cache_range, is that
-	 * lines can stay resident so long as the virtual address they were
-	 * accessed with (hence cache set) is in accord with the physical
-	 * address (i.e. tag).  It's no different here.  So I reckon we don't
-	 * need to flush the I-cache, since aliases don't matter for that.  We
-	 * should try that.
+	 * If cache is only 4k-per-way, there are never any 'aliases'.  Since
+	 * the cache is physically tagged, the data can just be left in there.
+	 */
+	if (cpu_data->dcache.n_aliases == 0)
+		return;
+
+	/*
+	 * Don't bother groveling around the dcache for the VMA ranges
+	 * if there are too many PTEs to make it worthwhile.
 	 */
-	flush_cache_all();
+	if (mm->nr_ptes >= MAX_DCACHE_PAGES)
+		flush_dcache_all();
+	else {
+		struct vm_area_struct *vma;
+
+		/*
+		 * In this case there are reasonably sized ranges to flush,
+		 * iterate through the VMA list and take care of any aliases.
+		 */
+		for (vma = mm->mmap; vma; vma = vma->vm_next)
+			__flush_cache_mm(mm, vma->vm_start, vma->vm_end);
+	}
+
+	/* Only touch the icache if one of the VMAs has VM_EXEC set. */
+	if (mm->exec_vm)
+		flush_icache_all();
 }
 
 /*
@@ -311,7 +416,8 @@ void flush_cache_mm(struct mm_struct *mm)
  * ADDR: Virtual Address (U0 address)
  * PFN: Physical page number
  */
-void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigned long pfn)
+void flush_cache_page(struct vm_area_struct *vma, unsigned long address,
+		      unsigned long pfn)
 {
 	unsigned long phys = pfn << PAGE_SHIFT;
 	unsigned int alias_mask;
@@ -358,87 +464,22 @@ void flush_cache_page(struct vm_area_struct *vma, unsigned long address, unsigne
 void flush_cache_range(struct vm_area_struct *vma, unsigned long start,
 		       unsigned long end)
 {
-	unsigned long d = 0, p = start & PAGE_MASK;
-	unsigned long alias_mask = cpu_data->dcache.alias_mask;
-	unsigned long n_aliases = cpu_data->dcache.n_aliases;
-	unsigned long select_bit;
-	unsigned long all_aliases_mask;
-	unsigned long addr_offset;
-	unsigned long phys;
-	pgd_t *dir;
-	pmd_t *pmd;
-	pud_t *pud;
-	pte_t *pte;
-	pte_t entry;
-	int i;
-
 	/*
 	 * If cache is only 4k-per-way, there are never any 'aliases'.  Since
 	 * the cache is physically tagged, the data can just be left in there.
 	 */
-	if (n_aliases == 0)
+	if (cpu_data->dcache.n_aliases == 0)
 		return;
 
-	all_aliases_mask = (1 << n_aliases) - 1;
-
 	/*
 	 * Don't bother with the lookup and alias check if we have a
 	 * wide range to cover, just blow away the dcache in its
 	 * entirety instead. -- PFM.
 	 */
-	if (((end - start) >> PAGE_SHIFT) >= 64) {
+	if (((end - start) >> PAGE_SHIFT) >= MAX_DCACHE_PAGES)
 		flush_dcache_all();
-
-		if (vma->vm_flags & VM_EXEC)
-			flush_icache_all();
-
-		return;
-	}
-
-	dir = pgd_offset(vma->vm_mm, p);
-	pud = pud_offset(dir, p);
-	pmd = pmd_offset(pud, p);
-	end = PAGE_ALIGN(end);
-
-	do {
-		if (pmd_none(*pmd) || pmd_bad(*pmd)) {
-			p &= ~((1 << PMD_SHIFT) - 1);
-			p += (1 << PMD_SHIFT);
-			pmd++;
-
-			continue;
-		}
-
-		pte = pte_offset_kernel(pmd, p);
-
-		do {
-			entry = *pte;
-
-			if ((pte_val(entry) & _PAGE_PRESENT)) {
-				phys = pte_val(entry) & PTE_PHYS_MASK;
-
-				if ((p ^ phys) & alias_mask) {
-					d |= 1 << ((p & alias_mask) >> PAGE_SHIFT);
-					d |= 1 << ((phys & alias_mask) >> PAGE_SHIFT);
-
-					if (d == all_aliases_mask)
-						goto loop_exit;
-				}
-			}
-
-			pte++;
-			p += PAGE_SIZE;
-		} while (p < end && ((unsigned long)pte & ~PAGE_MASK));
-		pmd++;
-	} while (p < end);
-
-loop_exit:
-	for (i = 0, select_bit = 0x1, addr_offset = 0x0; i < n_aliases;
-	     i++, select_bit <<= 1, addr_offset += PAGE_SIZE)
-		if (d & select_bit) {
-			(*__flush_dcache_segment_fn)(addr_offset, PAGE_SIZE);
-			wmb();
-		}
+	else
+		__flush_cache_mm(vma->vm_mm, start, end);
 
 	if (vma->vm_flags & VM_EXEC) {
 		/*
@@ -731,4 +772,3 @@ static void __flush_dcache_segment_4way(unsigned long start,
 		a3 += linesz;
 	} while (a0 < a0e);
 }
-
-- 
cgit v1.2.1


From 8c12b5dc13bf8516303a8224ab4e9708b33d5b00 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:31:06 +0900
Subject: sh: Clean up PAGE_SIZE definition for assembly use.

We want to be able to use PAGE_SIZE all over the place,
this is the same approach adopted by other architectures..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/page.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 888d6fe0030e..acf6977b4042 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -16,7 +16,13 @@
 
 /* PAGE_SHIFT determines the page size */
 #define PAGE_SHIFT	12
+
+#ifdef __ASSEMBLY__
 #define PAGE_SIZE	(1 << PAGE_SHIFT)
+#else
+#define PAGE_SIZE	(1UL << PAGE_SHIFT)
+#endif
+
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
-- 
cgit v1.2.1


From 19f9a34f87c48bbd270d617d1c986d0c23866a1a Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:33:49 +0900
Subject: sh: Initial vsyscall page support.

This implements initial support for the vsyscall page on SH.
At the moment we leave it configurable due to having nommu
to support from the same code base. We hook it up for the
signal trampoline return at present, with more to be added
later, once uClibc catches up.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/Makefile                      |   1 +
 arch/sh/kernel/process.c                     |   2 +-
 arch/sh/kernel/signal.c                      |  17 ++-
 arch/sh/kernel/vsyscall/Makefile             |  36 +++++++
 arch/sh/kernel/vsyscall/vsyscall-note.S      |  25 +++++
 arch/sh/kernel/vsyscall/vsyscall-sigreturn.S |  39 +++++++
 arch/sh/kernel/vsyscall/vsyscall-syscall.S   |  10 ++
 arch/sh/kernel/vsyscall/vsyscall-trapa.S     |  42 ++++++++
 arch/sh/kernel/vsyscall/vsyscall.c           | 150 +++++++++++++++++++++++++++
 arch/sh/kernel/vsyscall/vsyscall.lds.S       |  74 +++++++++++++
 arch/sh/mm/Kconfig                           |  13 +++
 arch/sh/mm/init.c                            |   3 +
 arch/sh/mm/tlb-flush.c                       |  18 ++--
 include/asm-sh/auxvec.h                      |  14 +++
 include/asm-sh/elf.h                         |  20 ++++
 include/asm-sh/mmu.h                         |   7 +-
 include/asm-sh/mmu_context.h                 |   8 +-
 include/asm-sh/page.h                        |   5 +
 include/asm-sh/processor.h                   |   6 ++
 19 files changed, 473 insertions(+), 17 deletions(-)
 create mode 100644 arch/sh/kernel/vsyscall/Makefile
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-note.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-syscall.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall-trapa.S
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall.c
 create mode 100644 arch/sh/kernel/vsyscall/vsyscall.lds.S

diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index 0e2148f63e7e..5da88a43d350 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -9,6 +9,7 @@ obj-y	:= process.o signal.o entry.o traps.o irq.o \
 	io.o io_generic.o sh_ksyms.o syscalls.o
 
 obj-y				+= cpu/ timers/
+obj-$(CONFIG_VSYSCALL)		+= vsyscall/
 
 obj-$(CONFIG_SMP)		+= smp.o
 obj-$(CONFIG_CF_ENABLER)	+= cf-enabler.o
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 630ec1af2483..0b1d5dd7a93b 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -355,7 +355,7 @@ struct task_struct *__switch_to(struct task_struct *prev, struct task_struct *ne
 	else if (next->thread.ubc_pc && next->mm) {
 		int asid = 0;
 #ifdef CONFIG_MMU
-		asid |= next->mm->context & MMU_CONTEXT_ASID_MASK;
+		asid |= next->mm->context.id & MMU_CONTEXT_ASID_MASK;
 #endif
 		ubc_set_tracing(asid, next->thread.ubc_pc);
 	} else {
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index 2f1c9545b49f..5213f5bc6ce0 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -8,7 +8,6 @@
  *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
  *
  */
-
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/smp.h>
@@ -21,6 +20,7 @@
 #include <linux/unistd.h>
 #include <linux/stddef.h>
 #include <linux/tty.h>
+#include <linux/elf.h>
 #include <linux/personality.h>
 #include <linux/binfmts.h>
 
@@ -29,8 +29,6 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#undef DEBUG
-
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
 /*
@@ -312,6 +310,11 @@ get_sigframe(struct k_sigaction *ka, unsigned long sp, size_t frame_size)
 	return (void __user *)((sp - frame_size) & -8ul);
 }
 
+/* These symbols are defined with the addresses in the vsyscall page.
+   See vsyscall-trapa.S.  */
+extern void __user __kernel_sigreturn;
+extern void __user __kernel_rt_sigreturn;
+
 static int setup_frame(int sig, struct k_sigaction *ka,
 			sigset_t *set, struct pt_regs *regs)
 {
@@ -340,6 +343,10 @@ static int setup_frame(int sig, struct k_sigaction *ka,
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
@@ -416,6 +423,10 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	   already in userspace.  */
 	if (ka->sa.sa_flags & SA_RESTORER) {
 		regs->pr = (unsigned long) ka->sa.sa_restorer;
+#ifdef CONFIG_VSYSCALL
+	} else if (likely(current->mm->context.vdso)) {
+		regs->pr = VDSO_SYM(&__kernel_rt_sigreturn);
+#endif
 	} else {
 		/* Generate return code (system call to rt_sigreturn) */
 		err |= __put_user(MOVW(7), &frame->retcode[0]);
diff --git a/arch/sh/kernel/vsyscall/Makefile b/arch/sh/kernel/vsyscall/Makefile
new file mode 100644
index 000000000000..4bbce1cfa359
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/Makefile
@@ -0,0 +1,36 @@
+obj-y += vsyscall.o vsyscall-syscall.o
+
+$(obj)/vsyscall-syscall.o: \
+	$(foreach F,trapa,$(obj)/vsyscall-$F.so)
+
+# Teach kbuild about targets
+targets += $(foreach F,trapa,vsyscall-$F.o vsyscall-$F.so)
+targets += vsyscall-note.o vsyscall.lds
+
+# The DSO images are built using a special linker script
+quiet_cmd_syscall = SYSCALL $@
+      cmd_syscall = $(CC) -nostdlib $(SYSCFLAGS_$(@F)) \
+			   -Wl,-T,$(filter-out FORCE,$^) -o $@
+
+export CPPFLAGS_vsyscall.lds += -P -C -Ush
+
+vsyscall-flags = -shared -s -Wl,-soname=linux-gate.so.1 \
+		$(call ld-option, -Wl$(comma)--hash-style=sysv)
+
+SYSCFLAGS_vsyscall-trapa.so	= $(vsyscall-flags)
+
+$(obj)/vsyscall-trapa.so: \
+$(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
+	$(call if_changed,syscall)
+
+# We also create a special relocatable object that should mirror the symbol
+# table and layout of the linked DSO.  With ld -R we can then refer to
+# these symbols in the kernel code rather than hand-coded addresses.
+extra-y += vsyscall-syms.o
+$(obj)/built-in.o: $(obj)/vsyscall-syms.o
+$(obj)/built-in.o: ld_flags += -R $(obj)/vsyscall-syms.o
+
+SYSCFLAGS_vsyscall-syms.o = -r
+$(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
+			$(obj)/vsyscall-trapa.o $(obj)/vsyscall-note.o FORCE
+	$(call if_changed,syscall)
diff --git a/arch/sh/kernel/vsyscall/vsyscall-note.S b/arch/sh/kernel/vsyscall/vsyscall-note.S
new file mode 100644
index 000000000000..d4b5be4f3d5f
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-note.S
@@ -0,0 +1,25 @@
+/*
+ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
+ * Here we can supply some information useful to userland.
+ */
+
+#include <linux/uts.h>
+#include <linux/version.h>
+
+#define ASM_ELF_NOTE_BEGIN(name, flags, vendor, type)			      \
+	.section name, flags;						      \
+	.balign 4;							      \
+	.long 1f - 0f;		/* name length */			      \
+	.long 3f - 2f;		/* data length */			      \
+	.long type;		/* note type */				      \
+0:	.asciz vendor;		/* vendor name */			      \
+1:	.balign 4;							      \
+2:
+
+#define ASM_ELF_NOTE_END						      \
+3:	.balign 4;		/* pad out section */			      \
+	.previous
+
+	ASM_ELF_NOTE_BEGIN(".note.kernel-version", "a", UTS_SYSNAME, 0)
+	.long LINUX_VERSION_CODE
+	ASM_ELF_NOTE_END
diff --git a/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
new file mode 100644
index 000000000000..555a64f124ca
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-sigreturn.S
@@ -0,0 +1,39 @@
+#include <asm/unistd.h>
+
+	.text
+	.balign 32
+	.globl __kernel_sigreturn
+	.type __kernel_sigreturn,@function
+__kernel_sigreturn:
+.LSTART_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_sigreturn
+.LEND_sigreturn:
+	.size __kernel_sigreturn,.-.LSTART_sigreturn
+
+	.balign 32
+	.globl __kernel_rt_sigreturn
+	.type __kernel_rt_sigreturn,@function
+__kernel_rt_sigreturn:
+.LSTART_rt_sigreturn:
+	mov.w	1f, r3
+	trapa	#0x10
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+	or	r0, r0
+
+1:	.short	__NR_rt_sigreturn
+.LEND_rt_sigreturn:
+	.size __kernel_rt_sigreturn,.-.LSTART_rt_sigreturn
+
+	.section .eh_frame,"a",@progbits
+	.previous
diff --git a/arch/sh/kernel/vsyscall/vsyscall-syscall.S b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
new file mode 100644
index 000000000000..c2ac7f0282b3
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-syscall.S
@@ -0,0 +1,10 @@
+#include <linux/init.h>
+
+__INITDATA
+
+	.globl vsyscall_trapa_start, vsyscall_trapa_end
+vsyscall_trapa_start:
+	.incbin "arch/sh/kernel/vsyscall/vsyscall-trapa.so"
+vsyscall_trapa_end:
+
+__FINIT
diff --git a/arch/sh/kernel/vsyscall/vsyscall-trapa.S b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
new file mode 100644
index 000000000000..3b6eb34c43fa
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall-trapa.S
@@ -0,0 +1,42 @@
+	.text
+	.globl __kernel_vsyscall
+	.type __kernel_vsyscall,@function
+__kernel_vsyscall:
+.LSTART_vsyscall:
+	/* XXX: We'll have to do something here once we opt to use the vDSO
+	 * page for something other than the signal trampoline.. as well as
+	 * fill out .eh_frame -- PFM. */
+.LEND_vsyscall:
+	.size __kernel_vsyscall,.-.LSTART_vsyscall
+	.previous
+
+	.section .eh_frame,"a",@progbits
+.LCIE:
+	.ualong	.LCIE_end - .LCIE_start
+.LCIE_start:
+	.ualong	0		/* CIE ID */
+	.byte	0x1		/* Version number */
+	.string	"zRS"		/* NUL-terminated augmentation string */
+	.uleb128 0x1		/* Code alignment factor */
+	.sleb128 -4		/* Data alignment factor */
+	.byte	0x11		/* Return address register column */
+				/* Augmentation length and data (none) */
+	.byte	0xc		/* DW_CFA_def_cfa */
+	.uleb128 0xf		/* r15 */
+	.uleb128 0x0		/* offset 0 */
+
+	.align 2
+.LCIE_end:
+
+	.ualong	.LFDE_end-.LFDE_start	/* Length FDE */
+.LFDE_start:
+	.ualong	.LCIE			/* CIE pointer */
+	.ualong	.LSTART_vsyscall-.	/* start address */
+	.ualong	.LEND_vsyscall-.LSTART_vsyscall
+	.uleb128 0
+	.align 2
+.LFDE_end:
+	.previous
+
+/* Get the common code for the sigreturn entry points */
+#include "vsyscall-sigreturn.S"
diff --git a/arch/sh/kernel/vsyscall/vsyscall.c b/arch/sh/kernel/vsyscall/vsyscall.c
new file mode 100644
index 000000000000..075d6cc1a2d7
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.c
@@ -0,0 +1,150 @@
+/*
+ * arch/sh/kernel/vsyscall.c
+ *
+ *  Copyright (C) 2006 Paul Mundt
+ *
+ * vDSO randomization
+ * Copyright(C) 2005-2006, Red Hat, Inc., Ingo Molnar
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/gfp.h>
+#include <linux/module.h>
+#include <linux/elf.h>
+
+/*
+ * Should the kernel map a VDSO page into processes and pass its
+ * address down to glibc upon exec()?
+ */
+unsigned int __read_mostly vdso_enabled = 1;
+EXPORT_SYMBOL_GPL(vdso_enabled);
+
+static int __init vdso_setup(char *s)
+{
+	vdso_enabled = simple_strtoul(s, NULL, 0);
+	return 1;
+}
+__setup("vdso=", vdso_setup);
+
+/*
+ * These symbols are defined by vsyscall.o to mark the bounds
+ * of the ELF DSO images included therein.
+ */
+extern const char vsyscall_trapa_start, vsyscall_trapa_end;
+static void *syscall_page;
+
+int __init vsyscall_init(void)
+{
+	syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
+
+	/*
+	 * XXX: Map this page to a fixmap entry if we get around
+	 * to adding the page to ELF core dumps
+	 */
+
+	memcpy(syscall_page,
+	       &vsyscall_trapa_start,
+	       &vsyscall_trapa_end - &vsyscall_trapa_start);
+
+	return 0;
+}
+
+static struct page *syscall_vma_nopage(struct vm_area_struct *vma,
+				       unsigned long address, int *type)
+{
+	unsigned long offset = address - vma->vm_start;
+	struct page *page;
+
+	if (address < vma->vm_start || address > vma->vm_end)
+		return NOPAGE_SIGBUS;
+
+	page = virt_to_page(syscall_page + offset);
+
+	get_page(page);
+
+	return page;
+}
+
+/* Prevent VMA merging */
+static void syscall_vma_close(struct vm_area_struct *vma)
+{
+}
+
+static struct vm_operations_struct syscall_vm_ops = {
+	.nopage	= syscall_vma_nopage,
+	.close	= syscall_vma_close,
+};
+
+/* Setup a VMA at program startup for the vsyscall page */
+int arch_setup_additional_pages(struct linux_binprm *bprm,
+				int executable_stack)
+{
+	struct vm_area_struct *vma;
+	struct mm_struct *mm = current->mm;
+	unsigned long addr;
+	int ret;
+
+	down_write(&mm->mmap_sem);
+	addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0);
+	if (IS_ERR_VALUE(addr)) {
+		ret = addr;
+		goto up_fail;
+	}
+
+	vma = kmem_cache_zalloc(vm_area_cachep, SLAB_KERNEL);
+	if (!vma) {
+		ret = -ENOMEM;
+		goto up_fail;
+	}
+
+	vma->vm_start = addr;
+	vma->vm_end = addr + PAGE_SIZE;
+	/* MAYWRITE to allow gdb to COW and set breakpoints */
+	vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE;
+	vma->vm_flags |= mm->def_flags;
+	vma->vm_page_prot = protection_map[vma->vm_flags & 7];
+	vma->vm_ops = &syscall_vm_ops;
+	vma->vm_mm = mm;
+
+	ret = insert_vm_struct(mm, vma);
+	if (unlikely(ret)) {
+		kmem_cache_free(vm_area_cachep, vma);
+		goto up_fail;
+	}
+
+	current->mm->context.vdso = (void *)addr;
+
+	mm->total_vm++;
+up_fail:
+	up_write(&mm->mmap_sem);
+	return ret;
+}
+
+const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso)
+		return "[vdso]";
+
+	return NULL;
+}
+
+struct vm_area_struct *get_gate_vma(struct task_struct *task)
+{
+	return NULL;
+}
+
+int in_gate_area(struct task_struct *task, unsigned long address)
+{
+	return 0;
+}
+
+int in_gate_area_no_task(unsigned long address)
+{
+	return 0;
+}
diff --git a/arch/sh/kernel/vsyscall/vsyscall.lds.S b/arch/sh/kernel/vsyscall/vsyscall.lds.S
new file mode 100644
index 000000000000..b13c3d439fee
--- /dev/null
+++ b/arch/sh/kernel/vsyscall/vsyscall.lds.S
@@ -0,0 +1,74 @@
+/*
+ * Linker script for vsyscall DSO.  The vsyscall page is an ELF shared
+ * object prelinked to its virtual address, and with only one read-only
+ * segment (that fits in one page).  This script controls its layout.
+ */
+#include <asm/asm-offsets.h>
+
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+OUTPUT_FORMAT("elf32-sh-linux", "elf32-sh-linux", "elf32-sh-linux")
+#else
+OUTPUT_FORMAT("elf32-shbig-linux", "elf32-shbig-linux", "elf32-shbig-linux")
+#endif
+OUTPUT_ARCH(sh)
+
+/* The ELF entry point can be used to set the AT_SYSINFO value.  */
+ENTRY(__kernel_vsyscall);
+
+SECTIONS
+{
+  . = SIZEOF_HEADERS;
+
+  .hash           : { *(.hash) }		:text
+  .gnu.hash       : { *(.gnu.hash) }
+  .dynsym         : { *(.dynsym) }
+  .dynstr         : { *(.dynstr) }
+  .gnu.version    : { *(.gnu.version) }
+  .gnu.version_d  : { *(.gnu.version_d) }
+  .gnu.version_r  : { *(.gnu.version_r) }
+
+  /* This linker script is used both with -r and with -shared.
+     For the layouts to match, we need to skip more than enough
+     space for the dynamic symbol table et al.  If this amount
+     is insufficient, ld -shared will barf.  Just increase it here.  */
+  . = 0x400;
+
+  .text           : { *(.text) }		:text =0x90909090
+  .note		  : { *(.note.*) }		:text :note
+  .eh_frame_hdr   : { *(.eh_frame_hdr) }	:text :eh_frame_hdr
+  .eh_frame       : { KEEP (*(.eh_frame)) }	:text
+  .dynamic        : { *(.dynamic) }		:text :dynamic
+  .useless        : {
+  	*(.got.plt) *(.got)
+	*(.data .data.* .gnu.linkonce.d.*)
+	*(.dynbss)
+	*(.bss .bss.* .gnu.linkonce.b.*)
+  }						:text
+}
+
+/*
+ * We must supply the ELF program headers explicitly to get just one
+ * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+ */
+PHDRS
+{
+  text PT_LOAD FILEHDR PHDRS FLAGS(5); /* PF_R|PF_X */
+  dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
+  note PT_NOTE FLAGS(4); /* PF_R */
+  eh_frame_hdr 0x6474e550; /* PT_GNU_EH_FRAME, but ld doesn't match the name */
+}
+
+/*
+ * This controls what symbols we export from the DSO.
+ */
+VERSION
+{
+  LINUX_2.6 {
+    global:
+    	__kernel_vsyscall;
+    	__kernel_sigreturn;
+    	__kernel_rt_sigreturn;
+
+    local: *;
+  };
+}
diff --git a/arch/sh/mm/Kconfig b/arch/sh/mm/Kconfig
index b445d02075e8..9dd606464d23 100644
--- a/arch/sh/mm/Kconfig
+++ b/arch/sh/mm/Kconfig
@@ -223,6 +223,19 @@ config 32BIT
 	  32-bits through the SH-4A PMB. If this is not set, legacy
 	  29-bit physical addressing will be used.
 
+config VSYSCALL
+	bool "Support vsyscall page"
+	depends on MMU
+	default y
+	help
+	  This will enable support for the kernel mapping a vDSO page
+	  in process space, and subsequently handing down the entry point
+	  to the libc through the ELF auxiliary vector.
+
+	  From the kernel side this is used for the signal trampoline.
+	  For systems with an MMU that can afford to give up a page,
+	  (the default value) say Y.
+
 choice
 	prompt "HugeTLB page size"
 	depends on HUGETLB_PAGE && CPU_SH4 && MMU
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index ad182b31d846..7154d1ce9785 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -286,6 +286,9 @@ void __init mem_init(void)
 		initsize >> 10);
 
 	p3_cache_init();
+
+	/* Initialize the vDSO */
+	vsyscall_init();
 }
 
 void free_initmem(void)
diff --git a/arch/sh/mm/tlb-flush.c b/arch/sh/mm/tlb-flush.c
index fd7e42bcaa40..73ec7f6084fa 100644
--- a/arch/sh/mm/tlb-flush.c
+++ b/arch/sh/mm/tlb-flush.c
@@ -14,12 +14,12 @@
 
 void flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
 {
-	if (vma->vm_mm && vma->vm_mm->context != NO_CONTEXT) {
+	if (vma->vm_mm && vma->vm_mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 		unsigned long asid;
 		unsigned long saved_asid = MMU_NO_ASID;
 
-		asid = vma->vm_mm->context & MMU_CONTEXT_ASID_MASK;
+		asid = vma->vm_mm->context.id & MMU_CONTEXT_ASID_MASK;
 		page &= PAGE_MASK;
 
 		local_irq_save(flags);
@@ -39,20 +39,21 @@ void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 {
 	struct mm_struct *mm = vma->vm_mm;
 
-	if (mm->context != NO_CONTEXT) {
+	if (mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 		int size;
 
 		local_irq_save(flags);
 		size = (end - start + (PAGE_SIZE - 1)) >> PAGE_SHIFT;
 		if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
-			mm->context = NO_CONTEXT;
+			mm->context.id = NO_CONTEXT;
 			if (mm == current->mm)
 				activate_context(mm);
 		} else {
-			unsigned long asid = mm->context&MMU_CONTEXT_ASID_MASK;
+			unsigned long asid;
 			unsigned long saved_asid = MMU_NO_ASID;
 
+			asid = mm->context.id & MMU_CONTEXT_ASID_MASK;
 			start &= PAGE_MASK;
 			end += (PAGE_SIZE - 1);
 			end &= PAGE_MASK;
@@ -81,9 +82,10 @@ void flush_tlb_kernel_range(unsigned long start, unsigned long end)
 	if (size > (MMU_NTLB_ENTRIES/4)) { /* Too many TLB to flush */
 		flush_tlb_all();
 	} else {
-		unsigned long asid = init_mm.context&MMU_CONTEXT_ASID_MASK;
+		unsigned long asid;
 		unsigned long saved_asid = get_asid();
 
+		asid = init_mm.context.id & MMU_CONTEXT_ASID_MASK;
 		start &= PAGE_MASK;
 		end += (PAGE_SIZE - 1);
 		end &= PAGE_MASK;
@@ -101,11 +103,11 @@ void flush_tlb_mm(struct mm_struct *mm)
 {
 	/* Invalidate all TLB of this process. */
 	/* Instead of invalidating each TLB, we get new MMU context. */
-	if (mm->context != NO_CONTEXT) {
+	if (mm->context.id != NO_CONTEXT) {
 		unsigned long flags;
 
 		local_irq_save(flags);
-		mm->context = NO_CONTEXT;
+		mm->context.id = NO_CONTEXT;
 		if (mm == current->mm)
 			activate_context(mm);
 		local_irq_restore(flags);
diff --git a/include/asm-sh/auxvec.h b/include/asm-sh/auxvec.h
index fc21e4db5881..1b6916e63e90 100644
--- a/include/asm-sh/auxvec.h
+++ b/include/asm-sh/auxvec.h
@@ -1,4 +1,18 @@
 #ifndef __ASM_SH_AUXVEC_H
 #define __ASM_SH_AUXVEC_H
 
+/*
+ * Architecture-neutral AT_ values in 0-17, leave some room
+ * for more of them.
+ */
+
+#ifdef CONFIG_VSYSCALL
+/*
+ * Only define this in the vsyscall case, the entry point to
+ * the vsyscall page gets placed here. The kernel will attempt
+ * to build a gate VMA we don't care about otherwise..
+ */
+#define AT_SYSINFO_EHDR		33
+#endif
+
 #endif /* __ASM_SH_AUXVEC_H */
diff --git a/include/asm-sh/elf.h b/include/asm-sh/elf.h
index cc8e5e767345..3a07ab40ac4d 100644
--- a/include/asm-sh/elf.h
+++ b/include/asm-sh/elf.h
@@ -121,4 +121,24 @@ extern int dump_task_fpu (struct task_struct *, elf_fpregset_t *);
 #define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
 #endif
 
+#ifdef CONFIG_VSYSCALL
+/* vDSO has arch_setup_additional_pages */
+#define ARCH_HAS_SETUP_ADDITIONAL_PAGES
+struct linux_binprm;
+extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+				       int executable_stack);
+
+extern unsigned int vdso_enabled;
+extern void __kernel_vsyscall;
+
+#define VDSO_BASE		((unsigned long)current->mm->context.vdso)
+#define VDSO_SYM(x)		(VDSO_BASE + (unsigned long)(x))
+
+#define ARCH_DLINFO						\
+do {								\
+	if (vdso_enabled)					\
+		NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_BASE);	\
+} while (0)
+#endif /* CONFIG_VSYSCALL */
+
 #endif /* __ASM_SH_ELF_H */
diff --git a/include/asm-sh/mmu.h b/include/asm-sh/mmu.h
index 6383dc84501e..cf47df79bb94 100644
--- a/include/asm-sh/mmu.h
+++ b/include/asm-sh/mmu.h
@@ -11,7 +11,12 @@ typedef struct {
 #else
 
 /* Default "unsigned long" context */
-typedef unsigned long mm_context_t;
+typedef unsigned long mm_context_id_t;
+
+typedef struct {
+	mm_context_id_t id;
+	void *vdso;
+} mm_context_t;
 
 #endif /* CONFIG_MMU */
 
diff --git a/include/asm-sh/mmu_context.h b/include/asm-sh/mmu_context.h
index 87678ba8d6b6..c7088efe579a 100644
--- a/include/asm-sh/mmu_context.h
+++ b/include/asm-sh/mmu_context.h
@@ -49,7 +49,7 @@ get_mmu_context(struct mm_struct *mm)
 	unsigned long mc = mmu_context_cache;
 
 	/* Check if we have old version of context. */
-	if (((mm->context ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
+	if (((mm->context.id ^ mc) & MMU_CONTEXT_VERSION_MASK) == 0)
 		/* It's up to date, do nothing */
 		return;
 
@@ -68,7 +68,7 @@ get_mmu_context(struct mm_struct *mm)
 		if (!mc)
 			mmu_context_cache = mc = MMU_CONTEXT_FIRST_VERSION;
 	}
-	mm->context = mc;
+	mm->context.id = mc;
 }
 
 /*
@@ -78,7 +78,7 @@ get_mmu_context(struct mm_struct *mm)
 static __inline__ int init_new_context(struct task_struct *tsk,
 				       struct mm_struct *mm)
 {
-	mm->context = NO_CONTEXT;
+	mm->context.id = NO_CONTEXT;
 
 	return 0;
 }
@@ -123,7 +123,7 @@ static __inline__ unsigned long get_asid(void)
 static __inline__ void activate_context(struct mm_struct *mm)
 {
 	get_mmu_context(mm);
-	set_asid(mm->context & MMU_CONTEXT_ASID_MASK);
+	set_asid(mm->context.id & MMU_CONTEXT_ASID_MASK);
 }
 
 /* MMU_TTB can be used for optimizing the fault handling.
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index acf6977b4042..3d8dae31a6f6 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -117,5 +117,10 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 
+/* vDSO support */
+#ifdef CONFIG_VSYSCALL
+#define __HAVE_ARCH_GATE_AREA
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PAGE_H */
diff --git a/include/asm-sh/processor.h b/include/asm-sh/processor.h
index b7cba4e91a72..474773853cd1 100644
--- a/include/asm-sh/processor.h
+++ b/include/asm-sh/processor.h
@@ -276,5 +276,11 @@ static inline void prefetch(void *x)
 #define prefetchw(x)	prefetch(x)
 #endif
 
+#ifdef CONFIG_VSYSCALL
+extern int vsyscall_init(void);
+#else
+#define vsyscall_init() do { } while (0)
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_PROCESSOR_H */
-- 
cgit v1.2.1


From 87b0ef91b6f27c07bf7dcce8584437481f473092 Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:34:41 +0900
Subject: sh: dma-mapping compile fixes.

Silly bug, make it build again..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 include/asm-sh/dma-mapping.h | 42 +++++++++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 17 deletions(-)

diff --git a/include/asm-sh/dma-mapping.h b/include/asm-sh/dma-mapping.h
index 124968f9866e..56cd4b977232 100644
--- a/include/asm-sh/dma-mapping.h
+++ b/include/asm-sh/dma-mapping.h
@@ -141,25 +141,35 @@ static inline void dma_sync_sg(struct device *dev, struct scatterlist *sg,
 	}
 }
 
-static void dma_sync_single_for_cpu(struct device *dev,
-				    dma_addr_t dma_handle, size_t size,
-				    enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_single")));
+static inline void dma_sync_single_for_cpu(struct device *dev,
+					   dma_addr_t dma_handle, size_t size,
+					   enum dma_data_direction dir)
+{
+	dma_sync_single(dev, dma_handle, size, dir);
+}
+
+static inline void dma_sync_single_for_device(struct device *dev,
+					      dma_addr_t dma_handle,
+					      size_t size,
+					      enum dma_data_direction dir)
+{
+	dma_sync_single(dev, dma_handle, size, dir);
+}
 
-static void dma_sync_single_for_device(struct device *dev,
-				       dma_addr_t dma_handle, size_t size,
+static inline void dma_sync_sg_for_cpu(struct device *dev,
+				       struct scatterlist *sg, int nelems,
 				       enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_single")));
+{
+	dma_sync_sg(dev, sg, nelems, dir);
+}
 
-static void dma_sync_sg_for_cpu(struct device *dev,
-				struct scatterlist *sg, int nelems,
-				enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_sg")));
+static inline void dma_sync_sg_for_device(struct device *dev,
+					  struct scatterlist *sg, int nelems,
+					  enum dma_data_direction dir)
+{
+	dma_sync_sg(dev, sg, nelems, dir);
+}
 
-static void dma_sync_sg_for_device(struct device *dev,
-				   struct scatterlist *sg, int nelems,
-				   enum dma_data_direction dir)
-	__attribute__ ((alias("dma_sync_sg")));
 
 static inline int dma_get_cache_alignment(void)
 {
@@ -174,6 +184,4 @@ static inline int dma_mapping_error(dma_addr_t dma_addr)
 {
 	return dma_addr == 0;
 }
-
 #endif /* __ASM_SH_DMA_MAPPING_H */
-
-- 
cgit v1.2.1


From f3c2575818fab45f8609e4aef2e43ab02b3a142e Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:36:17 +0900
Subject: sh: Calculate shm alignment at runtime.

Set the SHM alignment at runtime, based off of probed cache desc.
Optimize get_unmapped_area() to only colour align shared mappings.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/kernel/cpu/init.c           |  5 ++++
 arch/sh/kernel/sys_sh.c             | 54 +++++++++++++++++++++++--------------
 include/asm-sh/cacheflush.h         |  2 ++
 include/asm-sh/cpu-sh3/cacheflush.h |  8 ------
 include/asm-sh/cpu-sh4/cacheflush.h |  3 ---
 include/asm-sh/page.h               |  2 ++
 include/asm-sh/pgtable.h            |  2 --
 7 files changed, 43 insertions(+), 33 deletions(-)

diff --git a/arch/sh/kernel/cpu/init.c b/arch/sh/kernel/cpu/init.c
index 731dd61419dd..bfb90eb0b7a6 100644
--- a/arch/sh/kernel/cpu/init.c
+++ b/arch/sh/kernel/cpu/init.c
@@ -14,6 +14,7 @@
 #include <linux/kernel.h>
 #include <asm/processor.h>
 #include <asm/uaccess.h>
+#include <asm/page.h>
 #include <asm/system.h>
 #include <asm/cacheflush.h>
 #include <asm/cache.h>
@@ -198,6 +199,10 @@ asmlinkage void __init sh_cpu_init(void)
 	/* Init the cache */
 	cache_init();
 
+	shm_align_mask = max_t(unsigned long,
+			       cpu_data->dcache.way_size - 1,
+			       PAGE_SIZE - 1);
+
 	/* Disable the FPU */
 	if (fpu_disabled) {
 		printk("FPU Disabled\n");
diff --git a/arch/sh/kernel/sys_sh.c b/arch/sh/kernel/sys_sh.c
index 0ee7bf4cb238..b68ff705f067 100644
--- a/arch/sh/kernel/sys_sh.c
+++ b/arch/sh/kernel/sys_sh.c
@@ -21,6 +21,7 @@
 #include <linux/mman.h>
 #include <linux/file.h>
 #include <linux/utsname.h>
+#include <linux/module.h>
 #include <asm/cacheflush.h>
 #include <asm/uaccess.h>
 #include <asm/ipc.h>
@@ -44,11 +45,16 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	return error;
 }
 
-#if defined(HAVE_ARCH_UNMAPPED_AREA) && defined(CONFIG_MMU)
+unsigned long shm_align_mask = PAGE_SIZE - 1;	/* Sane caches */
+
+EXPORT_SYMBOL(shm_align_mask);
+
 /*
- * To avoid cache alias, we map the shard page with same color.
+ * To avoid cache aliases, we map the shared page with same color.
  */
-#define COLOUR_ALIGN(addr)	(((addr)+SHMLBA-1)&~(SHMLBA-1))
+#define COLOUR_ALIGN(addr, pgoff)				\
+	((((addr) + shm_align_mask) & ~shm_align_mask) +	\
+	 (((pgoff) << PAGE_SHIFT) & shm_align_mask))
 
 unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	unsigned long len, unsigned long pgoff, unsigned long flags)
@@ -56,43 +62,52 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr,
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
+	int do_colour_align;
 
 	if (flags & MAP_FIXED) {
 		/* We do not accept a shared mapping if it would violate
 		 * cache aliasing constraints.
 		 */
-		if ((flags & MAP_SHARED) && (addr & (SHMLBA - 1)))
+		if ((flags & MAP_SHARED) && (addr & shm_align_mask))
 			return -EINVAL;
 		return addr;
 	}
 
-	if (len > TASK_SIZE)
+	if (unlikely(len > TASK_SIZE))
 		return -ENOMEM;
 
+	do_colour_align = 0;
+	if (filp || (flags & MAP_SHARED))
+		do_colour_align = 1;
+
 	if (addr) {
-		if (flags & MAP_PRIVATE)
-			addr = PAGE_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 		else
-			addr = COLOUR_ALIGN(addr);
+			addr = PAGE_ALIGN(addr);
+
 		vma = find_vma(mm, addr);
 		if (TASK_SIZE - len >= addr &&
 		    (!vma || addr + len <= vma->vm_start))
 			return addr;
 	}
-	if (len <= mm->cached_hole_size) {
+
+	if (len > mm->cached_hole_size) {
+		start_addr = addr = mm->free_area_cache;
+	} else {
 	        mm->cached_hole_size = 0;
-		mm->free_area_cache = TASK_UNMAPPED_BASE;
+		start_addr = addr = TASK_UNMAPPED_BASE;
 	}
-	if (flags & MAP_PRIVATE)
-		addr = PAGE_ALIGN(mm->free_area_cache);
-	else
-		addr = COLOUR_ALIGN(mm->free_area_cache);
-	start_addr = addr;
 
 full_search:
+	if (do_colour_align)
+		addr = COLOUR_ALIGN(addr, pgoff);
+	else
+		addr = PAGE_ALIGN(mm->free_area_cache);
+
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
 		/* At this point:  (!vma || addr < vma->vm_end). */
-		if (TASK_SIZE - len < addr) {
+		if (unlikely(TASK_SIZE - len < addr)) {
 			/*
 			 * Start a new search - just in case we missed
 			 * some holes.
@@ -104,7 +119,7 @@ full_search:
 			}
 			return -ENOMEM;
 		}
-		if (!vma || addr + len <= vma->vm_start) {
+		if (likely(!vma || addr + len <= vma->vm_start)) {
 			/*
 			 * Remember the place where we stopped the search:
 			 */
@@ -115,11 +130,10 @@ full_search:
 		        mm->cached_hole_size = vma->vm_start - addr;
 
 		addr = vma->vm_end;
-		if (!(flags & MAP_PRIVATE))
-			addr = COLOUR_ALIGN(addr);
+		if (do_colour_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
 	}
 }
-#endif
 
 static inline long
 do_mmap2(unsigned long addr, unsigned long len, unsigned long prot, 
diff --git a/include/asm-sh/cacheflush.h b/include/asm-sh/cacheflush.h
index 92930b4a40d4..07f62ec9ff0c 100644
--- a/include/asm-sh/cacheflush.h
+++ b/include/asm-sh/cacheflush.h
@@ -28,5 +28,7 @@ extern void __flush_invalidate_region(void *start, int size);
 		memcpy(dst, src, len);				\
 	} while (0)
 
+#define HAVE_ARCH_UNMAPPED_AREA
+
 #endif /* __KERNEL__ */
 #endif /* __ASM_SH_CACHEFLUSH_H */
diff --git a/include/asm-sh/cpu-sh3/cacheflush.h b/include/asm-sh/cpu-sh3/cacheflush.h
index 97f5a64c2ab8..03fde97a7fd0 100644
--- a/include/asm-sh/cpu-sh3/cacheflush.h
+++ b/include/asm-sh/cpu-sh3/cacheflush.h
@@ -64,12 +64,4 @@ void flush_icache_page(struct vm_area_struct *vma, struct page *page);
 
 #define p3_cache_init()				do { } while (0)
 
-/*
- * We provide our own get_unmapped_area to avoid cache aliasing issues
- * on SH7705 with a 32KB cache, and to page align addresses in the
- * non-aliasing case.
- */
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #endif /* __ASM_CPU_SH3_CACHEFLUSH_H */
-
diff --git a/include/asm-sh/cpu-sh4/cacheflush.h b/include/asm-sh/cpu-sh4/cacheflush.h
index a95fc951aff6..515fd574267c 100644
--- a/include/asm-sh/cpu-sh4/cacheflush.h
+++ b/include/asm-sh/cpu-sh4/cacheflush.h
@@ -39,9 +39,6 @@ void p3_cache_init(void);
 
 #define PG_mapped	PG_arch_1
 
-/* We provide our own get_unmapped_area to avoid cache alias issue */
-#define HAVE_ARCH_UNMAPPED_AREA
-
 #ifdef CONFIG_MMU
 extern int remap_area_pages(unsigned long addr, unsigned long phys_addr,
 			    unsigned long size, unsigned long flags);
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 3d8dae31a6f6..ca8b26d90475 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -44,6 +44,8 @@
 extern void (*clear_page)(void *to);
 extern void (*copy_page)(void *to, void *from);
 
+extern unsigned long shm_align_mask;
+
 #ifdef CONFIG_MMU
 extern void clear_page_slow(void *to);
 extern void copy_page_slow(void *to, void *from);
diff --git a/include/asm-sh/pgtable.h b/include/asm-sh/pgtable.h
index 41c559d8ba87..2c8682ad1012 100644
--- a/include/asm-sh/pgtable.h
+++ b/include/asm-sh/pgtable.h
@@ -340,6 +340,4 @@ extern pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t
 #include <asm-generic/pgtable.h>
 
 #endif /* !__ASSEMBLY__ */
-
 #endif /* __ASM_SH_PAGE_H */
-
-- 
cgit v1.2.1


From 33573c0e3243aaa38b6ad96942de85a1b713c2ff Mon Sep 17 00:00:00 2001
From: Paul Mundt <lethal@linux-sh.org>
Date: Wed, 27 Sep 2006 18:37:30 +0900
Subject: sh: Fix occasional flush_cache_4096() stack corruption.

IRQs disabling in flush_cache_4096 for cache purge. Under certain
workloads we would get an IRQ in the middle of a purge operation,
and the cachelines would remain in an inconsistent state, leading
to occasional stack corruption.

Signed-off-by: Takeo Takahashi <takahashi.takeo@renesas.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
---
 arch/sh/mm/cache-sh4.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/arch/sh/mm/cache-sh4.c b/arch/sh/mm/cache-sh4.c
index aa4f62f0e374..e48cc22724d9 100644
--- a/arch/sh/mm/cache-sh4.c
+++ b/arch/sh/mm/cache-sh4.c
@@ -221,22 +221,20 @@ void flush_cache_sigtramp(unsigned long addr)
 static inline void flush_cache_4096(unsigned long start,
 				    unsigned long phys)
 {
+	unsigned long flags, exec_offset = 0;
+
 	/*
 	 * All types of SH-4 require PC to be in P2 to operate on the I-cache.
 	 * Some types of SH-4 require PC to be in P2 to operate on the D-cache.
 	 */
 	if ((cpu_data->flags & CPU_HAS_P2_FLUSH_BUG) ||
-	    (start < CACHE_OC_ADDRESS_ARRAY)) {
-		unsigned long flags;
-
-		local_irq_save(flags);
-		__flush_cache_4096(start | SH_CACHE_ASSOC,
-				   P1SEGADDR(phys), 0x20000000);
-		local_irq_restore(flags);
-	} else {
-		__flush_cache_4096(start | SH_CACHE_ASSOC,
-				   P1SEGADDR(phys), 0);
-	}
+	    (start < CACHE_OC_ADDRESS_ARRAY))
+	    	exec_offset = 0x20000000;
+
+	local_irq_save(flags);
+	__flush_cache_4096(start | SH_CACHE_ASSOC,
+			   P1SEGADDR(phys), exec_offset);
+	local_irq_restore(flags);
 }
 
 /*
-- 
cgit v1.2.1


From bda3028813bd07f34f30288a492fbf6f7b8712dd Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Wed, 27 Sep 2006 05:41:13 -0400
Subject: [libata] Don't use old-EH ->eng_timeout() hook when not needed

The PATA driver set got converted to the new error handling setup, but
the old hooks were accidentally left in place.  Now, removed.

Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/ata/ata_generic.c      |  2 +-
 drivers/ata/pata_ali.c         |  8 ++++----
 drivers/ata/pata_amd.c         | 12 ++++++------
 drivers/ata/pata_artop.c       |  4 +---
 drivers/ata/pata_atiixp.c      |  2 +-
 drivers/ata/pata_cmd64x.c      |  6 +++---
 drivers/ata/pata_cs5520.c      |  2 --
 drivers/ata/pata_cs5530.c      |  2 +-
 drivers/ata/pata_cs5535.c      |  2 +-
 drivers/ata/pata_cypress.c     |  2 +-
 drivers/ata/pata_efar.c        |  2 --
 drivers/ata/pata_hpt366.c      |  2 +-
 drivers/ata/pata_hpt37x.c      |  8 ++++----
 drivers/ata/pata_hpt3x2n.c     |  2 +-
 drivers/ata/pata_hpt3x3.c      |  2 +-
 drivers/ata/pata_isapnp.c      |  2 +-
 drivers/ata/pata_it821x.c      |  4 ++--
 drivers/ata/pata_jmicron.c     |  3 +--
 drivers/ata/pata_legacy.c      | 14 +++++++-------
 drivers/ata/pata_netcell.c     |  3 +--
 drivers/ata/pata_ns87410.c     |  2 +-
 drivers/ata/pata_opti.c        |  2 +-
 drivers/ata/pata_optidma.c     |  4 ++--
 drivers/ata/pata_pcmcia.c      |  2 +-
 drivers/ata/pata_qdi.c         |  4 ++--
 drivers/ata/pata_radisys.c     |  2 --
 drivers/ata/pata_rz1000.c      |  2 +-
 drivers/ata/pata_sc1200.c      |  2 +-
 drivers/ata/pata_serverworks.c |  4 ++--
 drivers/ata/pata_sil680.c      |  2 +-
 drivers/ata/pata_sis.c         | 10 ----------
 drivers/ata/pata_sl82c105.c    |  2 +-
 drivers/ata/pata_triflex.c     |  2 +-
 drivers/ata/pata_via.c         |  4 ++--
 34 files changed, 54 insertions(+), 74 deletions(-)

diff --git a/drivers/ata/ata_generic.c b/drivers/ata/ata_generic.c
index 1d1c30a2fcd0..377425e71391 100644
--- a/drivers/ata/ata_generic.c
+++ b/drivers/ata/ata_generic.c
@@ -143,7 +143,7 @@ static struct ata_port_operations generic_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.irq_handler	= ata_interrupt,
 	.irq_clear	= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_ali.c b/drivers/ata/pata_ali.c
index 8448ee6e0eed..87af3b5861ab 100644
--- a/drivers/ata/pata_ali.c
+++ b/drivers/ata/pata_ali.c
@@ -369,7 +369,7 @@ static struct ata_port_operations ali_early_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -410,7 +410,7 @@ static struct ata_port_operations ali_20_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -448,7 +448,7 @@ static struct ata_port_operations ali_c2_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -485,7 +485,7 @@ static struct ata_port_operations ali_c5_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_amd.c b/drivers/ata/pata_amd.c
index 793a5e1cb4cb..599ee266722c 100644
--- a/drivers/ata/pata_amd.c
+++ b/drivers/ata/pata_amd.c
@@ -359,7 +359,7 @@ static struct ata_port_operations amd33_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -393,7 +393,7 @@ static struct ata_port_operations amd66_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -427,7 +427,7 @@ static struct ata_port_operations amd100_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -461,7 +461,7 @@ static struct ata_port_operations amd133_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -495,7 +495,7 @@ static struct ata_port_operations nv100_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -529,7 +529,7 @@ static struct ata_port_operations nv133_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c
index 0f3b49176dc1..c4ccb75a4f1d 100644
--- a/drivers/ata/pata_artop.c
+++ b/drivers/ata/pata_artop.c
@@ -340,7 +340,7 @@ static const struct ata_port_operations artop6210_ops = {
 	.bmdma_status		= ata_bmdma_status,
 	.qc_prep		= ata_qc_prep,
 	.qc_issue		= ata_qc_issue_prot,
-	.eng_timeout		= ata_eng_timeout,
+
 	.data_xfer		= ata_pio_data_xfer,
 
 	.irq_handler		= ata_interrupt,
@@ -375,8 +375,6 @@ static const struct ata_port_operations artop6260_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_atiixp.c b/drivers/ata/pata_atiixp.c
index 060d019fd30b..6c2269b6bd3c 100644
--- a/drivers/ata/pata_atiixp.c
+++ b/drivers/ata/pata_atiixp.c
@@ -242,7 +242,7 @@ static struct ata_port_operations atiixp_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_cmd64x.c b/drivers/ata/pata_cmd64x.c
index abf1bb7bd322..e92b0ef43ec5 100644
--- a/drivers/ata/pata_cmd64x.c
+++ b/drivers/ata/pata_cmd64x.c
@@ -301,7 +301,7 @@ static struct ata_port_operations cmd64x_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -335,7 +335,7 @@ static struct ata_port_operations cmd646r1_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -369,7 +369,7 @@ static struct ata_port_operations cmd648_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_cs5520.c b/drivers/ata/pata_cs5520.c
index 792ce4828510..a6c6cebd0dae 100644
--- a/drivers/ata/pata_cs5520.c
+++ b/drivers/ata/pata_cs5520.c
@@ -193,8 +193,6 @@ static struct ata_port_operations cs5520_port_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_cs5530.c b/drivers/ata/pata_cs5530.c
index f3d8a3bc1e78..7bba4d954e9c 100644
--- a/drivers/ata/pata_cs5530.c
+++ b/drivers/ata/pata_cs5530.c
@@ -207,7 +207,7 @@ static struct ata_port_operations cs5530_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= cs5530_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_cs5535.c b/drivers/ata/pata_cs5535.c
index 69d6b4258724..d64fcdceaf01 100644
--- a/drivers/ata/pata_cs5535.c
+++ b/drivers/ata/pata_cs5535.c
@@ -211,7 +211,7 @@ static struct ata_port_operations cs5535_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_cypress.c b/drivers/ata/pata_cypress.c
index fd55474e0d15..dfa5ac539048 100644
--- a/drivers/ata/pata_cypress.c
+++ b/drivers/ata/pata_cypress.c
@@ -162,7 +162,7 @@ static struct ata_port_operations cy82c693_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_efar.c b/drivers/ata/pata_efar.c
index 92813827faa7..95cd1ca181f5 100644
--- a/drivers/ata/pata_efar.c
+++ b/drivers/ata/pata_efar.c
@@ -261,8 +261,6 @@ static const struct ata_port_operations efar_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_hpt366.c b/drivers/ata/pata_hpt366.c
index 94bb1dfc3f19..cf656ecbe507 100644
--- a/drivers/ata/pata_hpt366.c
+++ b/drivers/ata/pata_hpt366.c
@@ -360,7 +360,7 @@ static struct ata_port_operations hpt366_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_hpt37x.c b/drivers/ata/pata_hpt37x.c
index 532a7928f803..10318c0012ef 100644
--- a/drivers/ata/pata_hpt37x.c
+++ b/drivers/ata/pata_hpt37x.c
@@ -793,7 +793,7 @@ static struct ata_port_operations hpt370_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -832,7 +832,7 @@ static struct ata_port_operations hpt370a_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -872,7 +872,7 @@ static struct ata_port_operations hpt372_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -912,7 +912,7 @@ static struct ata_port_operations hpt374_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_hpt3x2n.c b/drivers/ata/pata_hpt3x2n.c
index 06c8db079b91..5c5d4f6ab901 100644
--- a/drivers/ata/pata_hpt3x2n.c
+++ b/drivers/ata/pata_hpt3x2n.c
@@ -372,7 +372,7 @@ static struct ata_port_operations hpt3x2n_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= hpt3x2n_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_hpt3x3.c b/drivers/ata/pata_hpt3x3.c
index 152770133ab1..1f084ab1ccc6 100644
--- a/drivers/ata/pata_hpt3x3.c
+++ b/drivers/ata/pata_hpt3x3.c
@@ -145,7 +145,7 @@ static struct ata_port_operations hpt3x3_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_isapnp.c b/drivers/ata/pata_isapnp.c
index 73948c8b7270..640b8b0954f5 100644
--- a/drivers/ata/pata_isapnp.c
+++ b/drivers/ata/pata_isapnp.c
@@ -52,7 +52,7 @@ static struct ata_port_operations isapnp_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_it821x.c b/drivers/ata/pata_it821x.c
index af39097d8081..82a46ff40000 100644
--- a/drivers/ata/pata_it821x.c
+++ b/drivers/ata/pata_it821x.c
@@ -703,7 +703,7 @@ static struct ata_port_operations it821x_smart_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= it821x_smart_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -739,7 +739,7 @@ static struct ata_port_operations it821x_passthru_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= it821x_passthru_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_clear	= ata_bmdma_irq_clear,
diff --git a/drivers/ata/pata_jmicron.c b/drivers/ata/pata_jmicron.c
index 3bdb6f2fa5ba..be3a866b111f 100644
--- a/drivers/ata/pata_jmicron.c
+++ b/drivers/ata/pata_jmicron.c
@@ -164,8 +164,7 @@ static const struct ata_port_operations jmicron_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	/* Timeout handling. Special recovery hooks here */
-	.eng_timeout		= ata_eng_timeout,
+	/* IRQ-related hooks */
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_legacy.c b/drivers/ata/pata_legacy.c
index ad37c220bb2c..10231ef731d1 100644
--- a/drivers/ata/pata_legacy.c
+++ b/drivers/ata/pata_legacy.c
@@ -161,7 +161,7 @@ static struct ata_port_operations simple_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
@@ -186,7 +186,7 @@ static struct ata_port_operations legacy_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
@@ -296,7 +296,7 @@ static struct ata_port_operations pdc20230_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= pdc_data_xfer_vlb,
 
 	.irq_handler	= ata_interrupt,
@@ -348,7 +348,7 @@ static struct ata_port_operations ht6560a_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,	/* Check vlb/noirq */
 
 	.irq_handler	= ata_interrupt,
@@ -411,7 +411,7 @@ static struct ata_port_operations ht6560b_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,	/* FIXME: Check 32bit and noirq */
 
 	.irq_handler	= ata_interrupt,
@@ -529,7 +529,7 @@ static struct ata_port_operations opti82c611a_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -659,7 +659,7 @@ static struct ata_port_operations opti82c46x_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= opti82c46x_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_netcell.c b/drivers/ata/pata_netcell.c
index 16cb254cb973..76eb9c90bee1 100644
--- a/drivers/ata/pata_netcell.c
+++ b/drivers/ata/pata_netcell.c
@@ -90,8 +90,7 @@ static const struct ata_port_operations netcell_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	/* Timeout handling. Special recovery hooks here */
-	.eng_timeout		= ata_eng_timeout,
+	/* IRQ-related hooks */
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_ns87410.c b/drivers/ata/pata_ns87410.c
index f529fe4c29b7..2005a95f48f6 100644
--- a/drivers/ata/pata_ns87410.c
+++ b/drivers/ata/pata_ns87410.c
@@ -176,7 +176,7 @@ static struct ata_port_operations ns87410_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ns87410_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_opti.c b/drivers/ata/pata_opti.c
index 136d7a65feb3..57fe21f3a975 100644
--- a/drivers/ata/pata_opti.c
+++ b/drivers/ata/pata_opti.c
@@ -227,7 +227,7 @@ static struct ata_port_operations opti_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_optidma.c b/drivers/ata/pata_optidma.c
index 18ec8e72bbb6..7296a20cd107 100644
--- a/drivers/ata/pata_optidma.c
+++ b/drivers/ata/pata_optidma.c
@@ -386,7 +386,7 @@ static struct ata_port_operations optidma_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -421,7 +421,7 @@ static struct ata_port_operations optiplus_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_pcmcia.c b/drivers/ata/pata_pcmcia.c
index 62b25cda409b..cb501e145a42 100644
--- a/drivers/ata/pata_pcmcia.c
+++ b/drivers/ata/pata_pcmcia.c
@@ -87,7 +87,7 @@ static struct ata_port_operations pcmcia_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_qdi.c b/drivers/ata/pata_qdi.c
index 35cfdf0ac3f0..7977f471d5e9 100644
--- a/drivers/ata/pata_qdi.c
+++ b/drivers/ata/pata_qdi.c
@@ -184,7 +184,7 @@ static struct ata_port_operations qdi6500_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= qdi_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= qdi_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -212,7 +212,7 @@ static struct ata_port_operations qdi6580_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= qdi_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= qdi_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_radisys.c b/drivers/ata/pata_radisys.c
index 277f8411b521..c20bcf43ed6d 100644
--- a/drivers/ata/pata_radisys.c
+++ b/drivers/ata/pata_radisys.c
@@ -255,8 +255,6 @@ static const struct ata_port_operations radisys_pata_ops = {
 	.qc_issue		= radisys_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_rz1000.c b/drivers/ata/pata_rz1000.c
index 3c6d84fd4312..eccc6fd45032 100644
--- a/drivers/ata/pata_rz1000.c
+++ b/drivers/ata/pata_rz1000.c
@@ -112,7 +112,7 @@ static struct ata_port_operations rz1000_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.freeze		= ata_bmdma_freeze,
diff --git a/drivers/ata/pata_sc1200.c b/drivers/ata/pata_sc1200.c
index 4166c1a8a9e8..107e6cd3dc0d 100644
--- a/drivers/ata/pata_sc1200.c
+++ b/drivers/ata/pata_sc1200.c
@@ -217,7 +217,7 @@ static struct ata_port_operations sc1200_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= sc1200_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_serverworks.c b/drivers/ata/pata_serverworks.c
index d62051b8da05..a5c8d7e121d1 100644
--- a/drivers/ata/pata_serverworks.c
+++ b/drivers/ata/pata_serverworks.c
@@ -352,7 +352,7 @@ static struct ata_port_operations serverworks_osb4_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -387,7 +387,7 @@ static struct ata_port_operations serverworks_csb_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_sil680.c b/drivers/ata/pata_sil680.c
index 8f7db9638d0a..c8b2e26db70d 100644
--- a/drivers/ata/pata_sil680.c
+++ b/drivers/ata/pata_sil680.c
@@ -251,7 +251,7 @@ static struct ata_port_operations sil680_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_sis.c b/drivers/ata/pata_sis.c
index fce74d9b2e67..17791e2785f9 100644
--- a/drivers/ata/pata_sis.c
+++ b/drivers/ata/pata_sis.c
@@ -573,8 +573,6 @@ static const struct ata_port_operations sis_133_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
@@ -608,8 +606,6 @@ static const struct ata_port_operations sis_133_early_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
@@ -644,8 +640,6 @@ static const struct ata_port_operations sis_100_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
@@ -679,8 +673,6 @@ static const struct ata_port_operations sis_66_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
@@ -714,8 +706,6 @@ static const struct ata_port_operations sis_old_ops = {
 	.qc_issue		= ata_qc_issue_prot,
 	.data_xfer		= ata_pio_data_xfer,
 
-	.eng_timeout		= ata_eng_timeout,
-
 	.irq_handler		= ata_interrupt,
 	.irq_clear		= ata_bmdma_irq_clear,
 
diff --git a/drivers/ata/pata_sl82c105.c b/drivers/ata/pata_sl82c105.c
index b3bc607f0293..5b762acc5687 100644
--- a/drivers/ata/pata_sl82c105.c
+++ b/drivers/ata/pata_sl82c105.c
@@ -261,7 +261,7 @@ static struct ata_port_operations sl82c105_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_triflex.c b/drivers/ata/pata_triflex.c
index 95adf92901fc..a954ed93a40c 100644
--- a/drivers/ata/pata_triflex.c
+++ b/drivers/ata/pata_triflex.c
@@ -218,7 +218,7 @@ static struct ata_port_operations triflex_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c
index 0e8b1303acd6..7b5dd2343b9a 100644
--- a/drivers/ata/pata_via.c
+++ b/drivers/ata/pata_via.c
@@ -322,7 +322,7 @@ static struct ata_port_operations via_port_ops = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer,
 
 	.irq_handler	= ata_interrupt,
@@ -357,7 +357,7 @@ static struct ata_port_operations via_port_ops_noirq = {
 
 	.qc_prep 	= ata_qc_prep,
 	.qc_issue	= ata_qc_issue_prot,
-	.eng_timeout	= ata_eng_timeout,
+
 	.data_xfer	= ata_pio_data_xfer_noirq,
 
 	.irq_handler	= ata_interrupt,
-- 
cgit v1.2.1


From d4fd1989ea142be620978dcac2e1b86929f3237a Mon Sep 17 00:00:00 2001
From: Maxime Bizon <mbizon@freebox.fr>
Date: Thu, 20 Jul 2006 18:52:02 +0200
Subject: [MIPS] Honour "panic_on_oops" sysctl.

Signed-off-by: Maxime Bizon <mbizon@freebox.fr>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 954a198494ef..368fdb779175 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -20,6 +20,7 @@
 #include <linux/spinlock.h>
 #include <linux/kallsyms.h>
 #include <linux/bootmem.h>
+#include <linux/interrupt.h>
 
 #include <asm/bootinfo.h>
 #include <asm/branch.h>
@@ -292,6 +293,16 @@ NORET_TYPE void ATTRIB_NORET die(const char * str, struct pt_regs * regs)
 	printk("%s[#%d]:\n", str, ++die_counter);
 	show_registers(regs);
 	spin_unlock_irq(&die_lock);
+
+	if (in_interrupt())
+		panic("Fatal exception in interrupt");
+
+	if (panic_on_oops) {
+		printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
+		ssleep(5);
+		panic("Fatal exception");
+	}
+
 	do_exit(SIGSEGV);
 }
 
-- 
cgit v1.2.1


From 53a73347faa38d3e80b46fc3c1999ee6bfbadbef Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Tue, 25 Jul 2006 23:24:47 +0900
Subject: [MIPS] Update e55 defconfig

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/e55_defconfig | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/mips/configs/e55_defconfig b/arch/mips/configs/e55_defconfig
index fa2996bb4b7c..d9a0d0eb0683 100644
--- a/arch/mips/configs/e55_defconfig
+++ b/arch/mips/configs/e55_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 10:04:02 2006
+# Linux kernel version: 2.6.18-rc2
+# Tue Jul 25 23:15:03 2006
 #
 CONFIG_MIPS=y
 
@@ -227,7 +227,6 @@ CONFIG_MMU=y
 #
 # PCCARD (PCMCIA/CardBus) support
 #
-# CONFIG_PCCARD is not set
 
 #
 # PCI Hotplug Support
@@ -254,7 +253,6 @@ CONFIG_TRAD_SIGNALS=y
 #
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
 # CONFIG_SYS_HYPERVISOR is not set
 
 #
@@ -284,6 +282,7 @@ CONFIG_PREVENT_FIRMWARE_BUILD=y
 CONFIG_BLK_DEV_RAM=m
 CONFIG_BLK_DEV_RAM_COUNT=16
 CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_BLK_DEV_INITRD is not set
 # CONFIG_CDROM_PKTCDVD is not set
 
@@ -643,6 +642,7 @@ CONFIG_MSDOS_PARTITION=y
 #
 # Kernel hacking
 #
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -650,7 +650,7 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_FS is not set
 CONFIG_CROSSCOMPILE=y
-CONFIG_CMDLINE="console=ttyVR0,19200 mem=8M"
+CONFIG_CMDLINE="console=ttyVR0,19200 ide0=0x1f0,0x3f6,40 mem=8M"
 
 #
 # Security options
-- 
cgit v1.2.1


From bbbdb8567e972f39a377a0bbd10b78b5b1ba4657 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Tue, 25 Jul 2006 23:24:54 +0900
Subject: [MIPS] Updat workpad defconfig

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/workpad_defconfig | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/arch/mips/configs/workpad_defconfig b/arch/mips/configs/workpad_defconfig
index 4891d02ef8ca..8f66ad71cb7e 100644
--- a/arch/mips/configs/workpad_defconfig
+++ b/arch/mips/configs/workpad_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 10:04:21 2006
+# Linux kernel version: 2.6.18-rc2
+# Tue Jul 25 23:13:04 2006
 #
 CONFIG_MIPS=y
 
@@ -166,6 +166,7 @@ CONFIG_SWAP=y
 CONFIG_SYSVIPC=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
 # CONFIG_IKCONFIG is not set
@@ -379,6 +380,7 @@ CONFIG_CONNECTOR=m
 CONFIG_BLK_DEV_RAM=m
 CONFIG_BLK_DEV_RAM_COUNT=16
 CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_BLK_DEV_INITRD is not set
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
@@ -855,7 +857,6 @@ CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
 # CONFIG_SMB_FS is not set
 # CONFIG_CIFS is not set
-# CONFIG_CIFS_DEBUG2 is not set
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
@@ -880,6 +881,7 @@ CONFIG_MSDOS_PARTITION=y
 #
 # Kernel hacking
 #
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -887,7 +889,7 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_FS is not set
 CONFIG_CROSSCOMPILE=y
-CONFIG_CMDLINE="console=ttyVR0,19200 mem=16M"
+CONFIG_CMDLINE="console=ttyVR0,19200 ide0=0x170,0x376,49 mem=16M"
 
 #
 # Security options
-- 
cgit v1.2.1


From e7dee3cea4dcd54744cf68864784c1336a42144f Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Tue, 25 Jul 2006 23:26:09 +0900
Subject: [MIPS] Updat mpc30x defconfig

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/mpc30x_defconfig | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/mips/configs/mpc30x_defconfig b/arch/mips/configs/mpc30x_defconfig
index 938b38ab5239..4405d127a941 100644
--- a/arch/mips/configs/mpc30x_defconfig
+++ b/arch/mips/configs/mpc30x_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 10:04:15 2006
+# Linux kernel version: 2.6.18-rc2
+# Tue Jul 25 23:16:46 2006
 #
 CONFIG_MIPS=y
 
@@ -71,7 +71,6 @@ CONFIG_MACH_VR41XX=y
 CONFIG_VICTOR_MPC30X=y
 # CONFIG_ZAO_CAPCELLA is not set
 CONFIG_PCI_VR41XX=y
-CONFIG_VRC4173=y
 CONFIG_RWSEM_GENERIC_SPINLOCK=y
 CONFIG_GENERIC_FIND_NEXT_BIT=y
 CONFIG_GENERIC_HWEIGHT=y
@@ -168,6 +167,7 @@ CONFIG_SWAP=y
 CONFIG_SYSVIPC=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
 # CONFIG_IKCONFIG is not set
@@ -841,7 +841,7 @@ CONFIG_USB_PEGASUS=m
 # CONFIG_USB_LEGOTOWER is not set
 # CONFIG_USB_LCD is not set
 # CONFIG_USB_LED is not set
-# CONFIG_USB_CY7C63 is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
 # CONFIG_USB_CYTHERM is not set
 # CONFIG_USB_PHIDGETKIT is not set
 # CONFIG_USB_PHIDGETSERVO is not set
@@ -982,7 +982,6 @@ CONFIG_SUNRPC=y
 # CONFIG_RPCSEC_GSS_SPKM3 is not set
 # CONFIG_SMB_FS is not set
 # CONFIG_CIFS is not set
-# CONFIG_CIFS_DEBUG2 is not set
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
@@ -1007,6 +1006,7 @@ CONFIG_MSDOS_PARTITION=y
 #
 # Kernel hacking
 #
+CONFIG_TRACE_IRQFLAGS_SUPPORT=y
 # CONFIG_PRINTK_TIME is not set
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -1014,7 +1014,7 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_DEBUG_FS is not set
 CONFIG_CROSSCOMPILE=y
-CONFIG_CMDLINE="mem=32M console=ttyVR0,19200"
+CONFIG_CMDLINE="mem=32M console=ttyVR0,19200 ide0=0x170,0x376,73"
 
 #
 # Security options
-- 
cgit v1.2.1


From e889d78fd5775d0b466f570928f5b0f0c9f05355 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 25 Jul 2006 23:51:36 +0900
Subject: [MIPS] Rearrange show_stack, show_trace

Print call-trace in show_stack() (like on other archs).  Also make
show_trace() static and simplify its argument list.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 51 ++++++++++++++++++++++--------------------------
 1 file changed, 23 insertions(+), 28 deletions(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 368fdb779175..c6f70467e750 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -79,6 +79,25 @@ void (*board_bind_eic_interrupt)(int irq, int regset);
  */
 #define MODULE_RANGE (8*1024*1024)
 
+static void show_trace(unsigned long *stack)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned long addr;
+
+	printk("Call Trace:");
+#ifdef CONFIG_KALLSYMS
+	printk("\n");
+#endif
+	while (!kstack_end(stack)) {
+		addr = *stack++;
+		if (__kernel_text_address(addr)) {
+			printk(" [<%0*lx>] ", field, addr);
+			print_symbol("%s\n", addr);
+		}
+	}
+	printk("\n");
+}
+
 /*
  * This routine abuses get_user()/put_user() to reference pointers
  * with at least a bit of error checking ...
@@ -88,6 +107,7 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 	const int field = 2 * sizeof(unsigned long);
 	long stackdata;
 	int i;
+	unsigned long *stack;
 
 	if (!sp) {
 		if (task && task != current)
@@ -95,6 +115,7 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 		else
 			sp = (unsigned long *) &sp;
 	}
+	stack = sp;
 
 	printk("Stack :");
 	i = 0;
@@ -115,32 +136,7 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 		i++;
 	}
 	printk("\n");
-}
-
-void show_trace(struct task_struct *task, unsigned long *stack)
-{
-	const int field = 2 * sizeof(unsigned long);
-	unsigned long addr;
-
-	if (!stack) {
-		if (task && task != current)
-			stack = (unsigned long *) task->thread.reg29;
-		else
-			stack = (unsigned long *) &stack;
-	}
-
-	printk("Call Trace:");
-#ifdef CONFIG_KALLSYMS
-	printk("\n");
-#endif
-	while (!kstack_end(stack)) {
-		addr = *stack++;
-		if (__kernel_text_address(addr)) {
-			printk(" [<%0*lx>] ", field, addr);
-			print_symbol("%s\n", addr);
-		}
-	}
-	printk("\n");
+	show_trace(stack);
 }
 
 /*
@@ -150,7 +146,7 @@ void dump_stack(void)
 {
 	unsigned long stack;
 
-	show_trace(current, &stack);
+	show_trace(&stack);
 }
 
 EXPORT_SYMBOL(dump_stack);
@@ -270,7 +266,6 @@ void show_registers(struct pt_regs *regs)
 	printk("Process %s (pid: %d, threadinfo=%p, task=%p)\n",
 	        current->comm, current->pid, current_thread_info(), current);
 	show_stack(current, (long *) regs->regs[29]);
-	show_trace(current, (long *) regs->regs[29]);
 	show_code((unsigned int *) regs->cp0_epc);
 	printk("\n");
 }
-- 
cgit v1.2.1


From 79495d876c2074af5552a0c4b7aea600c2320e83 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 26 Jul 2006 23:34:19 +0900
Subject: [MIPS] db1x00: Remove unused mirage_ts.c

CONFIG_WM97XX_COMODULE doesn't exist.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/au1000/db1x00/Makefile    |   1 -
 arch/mips/au1000/db1x00/mirage_ts.c | 260 ------------------------------------
 2 files changed, 261 deletions(-)
 delete mode 100644 arch/mips/au1000/db1x00/mirage_ts.c

diff --git a/arch/mips/au1000/db1x00/Makefile b/arch/mips/au1000/db1x00/Makefile
index 4c7d763f2113..51d62bd5d900 100644
--- a/arch/mips/au1000/db1x00/Makefile
+++ b/arch/mips/au1000/db1x00/Makefile
@@ -6,4 +6,3 @@
 # Makefile for the Alchemy Semiconductor Db1x00 board.
 
 lib-y := init.o board_setup.o irqmap.o
-obj-$(CONFIG_WM97XX_COMODULE) += mirage_ts.o
diff --git a/arch/mips/au1000/db1x00/mirage_ts.c b/arch/mips/au1000/db1x00/mirage_ts.c
deleted file mode 100644
index 0942dcf69518..000000000000
--- a/arch/mips/au1000/db1x00/mirage_ts.c
+++ /dev/null
@@ -1,260 +0,0 @@
-/*
- * linux/arch/mips/au1000/db1x00/mirage_ts.c
- *
- * BRIEF MODULE DESCRIPTION
- *	Glue between Mirage board-specific touchscreen pieces
- *	and generic Wolfson Codec touchscreen support.
- *
- *	Based on pb1100_ts.c used in Hydrogen II.
- *
- * Copyright (c) 2003 Embedded Edge, LLC
- *		dan@embeddededge.com
- *
- *  This program is free software; you can redistribute	 it and/or modify it
- *  under  the terms of	 the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the	License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED	  ``AS	IS'' AND   ANY	EXPRESS OR IMPLIED
- *  WARRANTIES,	  INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO	EVENT  SHALL   THE AUTHOR  BE	 LIABLE FOR ANY	  DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED	  TO, PROCUREMENT OF  SUBSTITUTE GOODS	OR SERVICES; LOSS OF
- *  USE, DATA,	OR PROFITS; OR	BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN	 CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-
-#include <linux/types.h>
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/fs.h>
-#include <linux/poll.h>
-#include <linux/proc_fs.h>
-#include <linux/smp.h>
-#include <linux/smp_lock.h>
-#include <linux/wait.h>
-
-#include <asm/segment.h>
-#include <asm/irq.h>
-#include <asm/uaccess.h>
-#include <asm/delay.h>
-#include <asm/au1000.h>
-
-/*
- *  Imported interface to Wolfson Codec driver.
- */
-extern void *wm97xx_ts_get_handle(int which);
-extern int wm97xx_ts_ready(void* ts_handle);
-extern void wm97xx_ts_set_cal(void* ts_handle, int xscale, int xtrans, int yscale, int ytrans);
-extern u16 wm97xx_ts_get_ac97(void* ts_handle, u8 reg);
-extern void wm97xx_ts_set_ac97(void* ts_handle, u8 reg, u16 val);
-extern int wm97xx_ts_read_data(void* ts_handle, long* x, long* y, long* pressure);
-extern void wm97xx_ts_send_data(void* ts_handle, long x, long y, long z);
-
-int wm97xx_comodule_present = 1;
-
-
-#define TS_NAME "mirage_ts"
-
-#define err(format, arg...) printk(KERN_ERR TS_NAME ": " format "\n" , ## arg)
-#define info(format, arg...) printk(KERN_INFO TS_NAME ": " format "\n" , ## arg)
-#define warn(format, arg...) printk(KERN_WARNING TS_NAME ": " format "\n" , ## arg)
-#define DPRINTK(format, arg...) printk("%s: " format "\n", __FUNCTION__ , ## arg)
-
-
-#define PEN_DOWN_IRQ	AU1000_GPIO_7
-
-static struct task_struct *ts_task = 0;
-static DECLARE_COMPLETION(ts_complete);
-static DECLARE_WAIT_QUEUE_HEAD(pendown_wait);
-
-#ifdef CONFIG_WM97XX_FIVEWIRETS
-static int release_pressure = 1;
-#else
-static int release_pressure = 50;
-#endif
-
-typedef struct {
-   long x;
-   long y;
-} DOWN_EVENT;
-
-#define SAMPLE_RATE	50	/* samples per second */
-#define PEN_DEBOUNCE	5	/* samples for settling - fn of SAMPLE_RATE */
-#define PEN_UP_TIMEOUT	10	/* in seconds */
-#define PEN_UP_SETTLE	5	/* samples per second */
-
-static struct {
-	int xscale;
-	int xtrans;
-	int yscale;
-	int ytrans;
-} mirage_ts_cal =
-{
-#if 0
-	.xscale   = 84,
-	.xtrans = -157,
-	.yscale   = 66,
-	.ytrans = -150,
-#else
-	.xscale   = 84,
-	.xtrans = -150,
-	.yscale   = 66,
-	.ytrans = -146,
-#endif
-};
-
-
-static void pendown_irq(int irqnr, void *devid, struct pt_regs *regs)
-{
-//DPRINTK("got one 0x%x", au_readl(SYS_PINSTATERD));
-	wake_up(&pendown_wait);
-}
-
-static int ts_thread(void *id)
-{
-	static int pen_was_down = 0;
-	static DOWN_EVENT pen_xy;
-	long x, y, z;
-	void *ts;	/* handle */
-	struct task_struct *tsk = current;
-	int timeout = HZ / SAMPLE_RATE;
-
-	ts_task = tsk;
-
-	daemonize();
-	tsk->tty = NULL;
-	tsk->policy = SCHED_FIFO;
-	tsk->rt_priority = 1;
-	strcpy(tsk->comm, "touchscreen");
-
-	/* only want to receive SIGKILL */
-	spin_lock_irq(&tsk->sigmask_lock);
-	siginitsetinv(&tsk->blocked, sigmask(SIGKILL));
-	recalc_sigpending(tsk);
-	spin_unlock_irq(&tsk->sigmask_lock);
-
-	/* get handle for codec */
-	ts = wm97xx_ts_get_handle(0);
-
-	/* proceed only after everybody is ready */
-	wait_event_timeout(pendown_wait, wm97xx_ts_ready(ts), HZ/4);
-
-	/* board-specific calibration */
-	wm97xx_ts_set_cal(ts,
-			mirage_ts_cal.xscale,
-			mirage_ts_cal.xtrans,
-			mirage_ts_cal.yscale,
-			mirage_ts_cal.ytrans);
-
-	/* route Wolfson pendown interrupts to our GPIO */
-	au_sync();
-	wm97xx_ts_set_ac97(ts, 0x4c, wm97xx_ts_get_ac97(ts, 0x4c) & ~0x0008);
-	au_sync();
-	wm97xx_ts_set_ac97(ts, 0x56, wm97xx_ts_get_ac97(ts, 0x56) & ~0x0008);
-	au_sync();
-	wm97xx_ts_set_ac97(ts, 0x52, wm97xx_ts_get_ac97(ts, 0x52) | 0x2008);
-	au_sync();
-
-	for (;;) {
-		interruptible_sleep_on_timeout(&pendown_wait, timeout);
-		disable_irq(PEN_DOWN_IRQ);
-		if (signal_pending(tsk)) {
-			break;
-		}
-
-		/* read codec */
-		if (!wm97xx_ts_read_data(ts, &x, &y, &z))
-			z = 0;	/* treat no-data and pen-up the same */
-
-		if (signal_pending(tsk)) {
-			break;
-		}
-
-		if (z >= release_pressure) {
-			y = ~y;	/* top to bottom */
-			if (pen_was_down > 1 /*&& pen_was_down < PEN_DEBOUNCE*/) {//THXXX
-				/* bounce ? */
-				x = pen_xy.x;
-				y = pen_xy.y;
-				--pen_was_down;
-			} else if (pen_was_down <= 1) {
-				pen_xy.x = x;
-				pen_xy.y = y;
-				if (pen_was_down)
-					wm97xx_ts_send_data(ts, x, y, z);
-				pen_was_down = PEN_DEBOUNCE;
-			}
-			//wm97xx_ts_send_data(ts, x, y, z);
-			timeout = HZ / SAMPLE_RATE;
-		} else {
-			if (pen_was_down) {
-				if (--pen_was_down)
-					z = release_pressure;
-				else //THXXX
-				wm97xx_ts_send_data(ts, pen_xy.x, pen_xy.y, z);
-			}
-			/* The pendown signal takes some time to settle after
-			 * reading the pen pressure so wait a little
-			 * before enabling the pen.
-			 */
-			if (! pen_was_down) {
-//				interruptible_sleep_on_timeout(&pendown_wait, HZ / PEN_UP_SETTLE);
-				timeout = HZ * PEN_UP_TIMEOUT;
-			}
-		}
-		enable_irq(PEN_DOWN_IRQ);
-	}
-	enable_irq(PEN_DOWN_IRQ);
-	ts_task = NULL;
-	complete(&ts_complete);
-	return 0;
-}
-
-static int __init ts_mirage_init(void)
-{
-	int ret;
-
-	/* pen down signal is connected to GPIO 7 */
-
-	ret = request_irq(PEN_DOWN_IRQ, pendown_irq, 0, "ts-pendown", NULL);
-	if (ret) {
-		err("unable to get pendown irq%d: [%d]", PEN_DOWN_IRQ, ret);
-		return ret;
-	}
-
-	lock_kernel();
-	ret = kernel_thread(ts_thread, NULL, CLONE_FS | CLONE_FILES);
-	if (ret < 0) {
-		unlock_kernel();
-		return ret;
-	}
-	unlock_kernel();
-
-	info("Mirage touchscreen IRQ initialized.");
-
-	return 0;
-}
-
-static void __exit ts_mirage_exit(void)
-{
-	if (ts_task) {
-		send_sig(SIGKILL, ts_task, 1);
-		wait_for_completion(&ts_complete);
-	}
-
-	free_irq(PEN_DOWN_IRQ, NULL);
-}
-
-module_init(ts_mirage_init);
-module_exit(ts_mirage_exit);
-
-- 
cgit v1.2.1


From f66686f70a2a61e53ee8c2284f75ca342e4c0dc8 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 29 Jul 2006 23:27:20 +0900
Subject: [MIPS] dump_stack() based on prologue code analysis

Instead of dump all possible address in the stack, unwind the stack frame
based on prologue code analysis, as like as get_wchan() does.  While the
code analysis might fail for some reason, there is a new kernel option
"raw_show_trace" to disable this feature.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 66 +++++++++++++++++++++++++------
 arch/mips/kernel/traps.c   | 98 ++++++++++++++++++++++++++++++++++++++++------
 2 files changed, 141 insertions(+), 23 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 7ab67f786bfe..8709a46a45c1 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -281,7 +281,7 @@ static struct mips_frame_info {
 } *schedule_frame, mfinfo[64];
 static int mfinfo_num;
 
-static int __init get_frame_info(struct mips_frame_info *info)
+static int get_frame_info(struct mips_frame_info *info)
 {
 	int i;
 	void *func = info->func;
@@ -329,14 +329,12 @@ static int __init get_frame_info(struct mips_frame_info *info)
 				ip->i_format.simmediate / sizeof(long);
 		}
 	}
-	if (info->pc_offset == -1 || info->frame_size == 0) {
-		if (func == schedule)
-			printk("Can't analyze prologue code at %p\n", func);
-		info->pc_offset = -1;
-		info->frame_size = 0;
-	}
-
-	return 0;
+	if (info->frame_size && info->pc_offset >= 0) /* nested */
+		return 0;
+	if (info->pc_offset < 0) /* leaf */
+		return 1;
+	/* prologue seems boggus... */
+	return -1;
 }
 
 static int __init frame_info_init(void)
@@ -367,8 +365,15 @@ static int __init frame_info_init(void)
 	mfinfo[0].func = schedule;
 	schedule_frame = &mfinfo[0];
 #endif
-	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++)
-		get_frame_info(&mfinfo[i]);
+	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++) {
+		struct mips_frame_info *info = &mfinfo[i];
+		if (get_frame_info(info)) {
+			/* leaf or unknown */
+			if (info->func == schedule)
+				printk("Can't analyze prologue code at %p\n",
+				       info->func);
+		}
+	}
 
 	mfinfo_num = i;
 	return 0;
@@ -427,6 +432,8 @@ unsigned long get_wchan(struct task_struct *p)
 		if (i < 0)
 			break;
 
+		if (mfinfo[i].pc_offset < 0)
+			break;
 		pc = ((unsigned long *)frame)[mfinfo[i].pc_offset];
 		if (!mfinfo[i].frame_size)
 			break;
@@ -437,3 +444,40 @@ unsigned long get_wchan(struct task_struct *p)
 	return pc;
 }
 
+#ifdef CONFIG_KALLSYMS
+/* used by show_frametrace() */
+unsigned long unwind_stack(struct task_struct *task,
+			   unsigned long **sp, unsigned long pc)
+{
+	unsigned long stack_page;
+	struct mips_frame_info info;
+	char *modname;
+	char namebuf[KSYM_NAME_LEN + 1];
+	unsigned long size, ofs;
+
+	stack_page = (unsigned long)task_stack_page(task);
+	if (!stack_page)
+		return 0;
+
+	if (!kallsyms_lookup(pc, &size, &ofs, &modname, namebuf))
+		return 0;
+	if (ofs == 0)
+		return 0;
+
+	info.func = (void *)(pc - ofs);
+	info.func_size = ofs;	/* analyze from start to ofs */
+	if (get_frame_info(&info)) {
+		/* leaf or unknown */
+		*sp += info.frame_size / sizeof(long);
+		return 0;
+	}
+	if ((unsigned long)*sp < stack_page ||
+	    (unsigned long)*sp + info.frame_size / sizeof(long) >
+	    stack_page + THREAD_SIZE - 32)
+		return 0;
+
+	pc = (*sp)[info.pc_offset];
+	*sp += info.frame_size / sizeof(long);
+	return pc;
+}
+#endif
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index c6f70467e750..7aa9dfc57b81 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -98,24 +98,53 @@ static void show_trace(unsigned long *stack)
 	printk("\n");
 }
 
+#ifdef CONFIG_KALLSYMS
+static int raw_show_trace;
+static int __init set_raw_show_trace(char *str)
+{
+	raw_show_trace = 1;
+	return 1;
+}
+__setup("raw_show_trace", set_raw_show_trace);
+
+extern unsigned long unwind_stack(struct task_struct *task,
+				  unsigned long **sp, unsigned long pc);
+static void show_frametrace(struct task_struct *task, struct pt_regs *regs)
+{
+	const int field = 2 * sizeof(unsigned long);
+	unsigned long *stack = (long *)regs->regs[29];
+	unsigned long pc = regs->cp0_epc;
+	int top = 1;
+
+	if (raw_show_trace || !__kernel_text_address(pc)) {
+		show_trace(stack);
+		return;
+	}
+	printk("Call Trace:\n");
+	while (__kernel_text_address(pc)) {
+		printk(" [<%0*lx>] ", field, pc);
+		print_symbol("%s\n", pc);
+		pc = unwind_stack(task, &stack, pc);
+		if (top && pc == 0)
+			pc = regs->regs[31];	/* leaf? */
+		top = 0;
+	}
+	printk("\n");
+}
+#else
+#define show_frametrace(task, r) show_trace((long *)(r)->regs[29]);
+#endif
+
 /*
  * This routine abuses get_user()/put_user() to reference pointers
  * with at least a bit of error checking ...
  */
-void show_stack(struct task_struct *task, unsigned long *sp)
+static void show_stacktrace(struct task_struct *task, struct pt_regs *regs)
 {
 	const int field = 2 * sizeof(unsigned long);
 	long stackdata;
 	int i;
-	unsigned long *stack;
-
-	if (!sp) {
-		if (task && task != current)
-			sp = (unsigned long *) task->thread.reg29;
-		else
-			sp = (unsigned long *) &sp;
-	}
-	stack = sp;
+	unsigned long *sp = (unsigned long *)regs->regs[29];
 
 	printk("Stack :");
 	i = 0;
@@ -136,7 +165,44 @@ void show_stack(struct task_struct *task, unsigned long *sp)
 		i++;
 	}
 	printk("\n");
-	show_trace(stack);
+	show_frametrace(task, regs);
+}
+
+static noinline void prepare_frametrace(struct pt_regs *regs)
+{
+	__asm__ __volatile__(
+		"1: la $2, 1b\n\t"
+#ifdef CONFIG_64BIT
+		"sd $2, %0\n\t"
+		"sd $29, %1\n\t"
+		"sd $31, %2\n\t"
+#else
+		"sw $2, %0\n\t"
+		"sw $29, %1\n\t"
+		"sw $31, %2\n\t"
+#endif
+		: "=m" (regs->cp0_epc),
+		"=m" (regs->regs[29]), "=m" (regs->regs[31])
+		: : "memory");
+}
+
+void show_stack(struct task_struct *task, unsigned long *sp)
+{
+	struct pt_regs regs;
+	if (sp) {
+		regs.regs[29] = (unsigned long)sp;
+		regs.regs[31] = 0;
+		regs.cp0_epc = 0;
+	} else {
+		if (task && task != current) {
+			regs.regs[29] = task->thread.reg29;
+			regs.regs[31] = 0;
+			regs.cp0_epc = task->thread.reg31;
+		} else {
+			prepare_frametrace(&regs);
+		}
+	}
+	show_stacktrace(task, &regs);
 }
 
 /*
@@ -146,6 +212,14 @@ void dump_stack(void)
 {
 	unsigned long stack;
 
+#ifdef CONFIG_KALLSYMS
+	if (!raw_show_trace) {
+		struct pt_regs regs;
+		prepare_frametrace(&regs);
+		show_frametrace(current, &regs);
+		return;
+	}
+#endif
 	show_trace(&stack);
 }
 
@@ -265,7 +339,7 @@ void show_registers(struct pt_regs *regs)
 	print_modules();
 	printk("Process %s (pid: %d, threadinfo=%p, task=%p)\n",
 	        current->comm, current->pid, current_thread_info(), current);
-	show_stack(current, (long *) regs->regs[29]);
+	show_stacktrace(current, regs);
 	show_code((unsigned int *) regs->cp0_epc);
 	printk("\n");
 }
-- 
cgit v1.2.1


From 1b4ee40a1330cad05e1cf65b60d8c82732bcafba Mon Sep 17 00:00:00 2001
From: Peter Watkins <treestem@gmail.com>
Date: Thu, 27 Jul 2006 16:05:49 -0400
Subject: [MIPS] N32 rt_sigqueueinfo uses O32 padding, not N64

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/scall64-n32.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 98abbc5a9f13..549b4bcb53a5 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -247,7 +247,7 @@ EXPORT(sysn32_call_table)
 	PTR	sys_capset
 	PTR	sys32_rt_sigpending		/* 6125 */
 	PTR	compat_sys_rt_sigtimedwait
-	PTR	sys_rt_sigqueueinfo
+	PTR	sys32_rt_sigqueueinfo
 	PTR	sysn32_rt_sigsuspend
 	PTR	sys32_sigaltstack
 	PTR	compat_sys_utime		/* 6130 */
-- 
cgit v1.2.1


From df586d59a4f069bb60a94ff4e5b64fe5c876f72c Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 1 Aug 2006 23:42:30 +0100
Subject: [MIPS] c-r4k: Typo fix.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/c-r4k.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 069803f58f3b..b3ccb8738d29 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -1242,7 +1242,7 @@ static inline void coherency_setup(void)
 		clear_c0_config(CONF_CU);
 		break;
 	/*
-	 * We need to catch the ealry Alchemy SOCs with
+	 * We need to catch the early Alchemy SOCs with
 	 * the write-only co_config.od bit and set it back to one...
 	 */
 	case CPU_AU1000: /* rev. DA, HA, HB */
-- 
cgit v1.2.1


From c0efbb6dc2726785482d85a284c883d541a6e0be Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:15 +0200
Subject: [MIPS] Make get_frame_info() more readable.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 67 +++++++++++++++++++++++++---------------------
 1 file changed, 36 insertions(+), 31 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 8709a46a45c1..93d5432759db 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -281,48 +281,53 @@ static struct mips_frame_info {
 } *schedule_frame, mfinfo[64];
 static int mfinfo_num;
 
+static inline int is_ra_save_ins(union mips_instruction *ip)
+{
+	/* sw / sd $ra, offset($sp) */
+	return (ip->i_format.opcode == sw_op || ip->i_format.opcode == sd_op) &&
+		ip->i_format.rs == 29 &&
+		ip->i_format.rt == 31;
+}
+
+static inline int is_jal_jalr_jr_ins(union mips_instruction *ip)
+{
+	if (ip->j_format.opcode == jal_op)
+		return 1;
+	if (ip->r_format.opcode != spec_op)
+		return 0;
+	return ip->r_format.func == jalr_op || ip->r_format.func == jr_op;
+}
+
+static inline int is_sp_move_ins(union mips_instruction *ip)
+{
+	/* addiu/daddiu sp,sp,-imm */
+	if (ip->i_format.rs != 29 || ip->i_format.rt != 29)
+		return 0;
+	if (ip->i_format.opcode == addiu_op || ip->i_format.opcode == daddiu_op)
+		return 1;
+	return 0;
+}
+
 static int get_frame_info(struct mips_frame_info *info)
 {
-	int i;
-	void *func = info->func;
-	union mips_instruction *ip = (union mips_instruction *)func;
+	union mips_instruction *ip = info->func;
+	int i, max_insns =
+		min(128UL, info->func_size / sizeof(union mips_instruction));
+
 	info->pc_offset = -1;
 	info->frame_size = 0;
-	for (i = 0; i < 128; i++, ip++) {
-		/* if jal, jalr, jr, stop. */
-		if (ip->j_format.opcode == jal_op ||
-		    (ip->r_format.opcode == spec_op &&
-		     (ip->r_format.func == jalr_op ||
-		      ip->r_format.func == jr_op)))
-			break;
 
-		if (info->func_size && i >= info->func_size / 4)
+	for (i = 0; i < max_insns; i++, ip++) {
+
+		if (is_jal_jalr_jr_ins(ip))
 			break;
-		if (
-#ifdef CONFIG_32BIT
-		    ip->i_format.opcode == addiu_op &&
-#endif
-#ifdef CONFIG_64BIT
-		    ip->i_format.opcode == daddiu_op &&
-#endif
-		    ip->i_format.rs == 29 &&
-		    ip->i_format.rt == 29) {
-			/* addiu/daddiu sp,sp,-imm */
+		if (is_sp_move_ins(ip)) {
 			if (info->frame_size)
 				continue;
 			info->frame_size = - ip->i_format.simmediate;
 		}
 
-		if (
-#ifdef CONFIG_32BIT
-		    ip->i_format.opcode == sw_op &&
-#endif
-#ifdef CONFIG_64BIT
-		    ip->i_format.opcode == sd_op &&
-#endif
-		    ip->i_format.rs == 29 &&
-		    ip->i_format.rt == 31) {
-			/* sw / sd $ra, offset($sp) */
+		if (is_ra_save_ins(ip)) {
 			if (info->pc_offset != -1)
 				continue;
 			info->pc_offset =
-- 
cgit v1.2.1


From cf495a333093a1c0eca90c7d8d98313a3b7f01a1 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:16 +0200
Subject: [MIPS] Remove unused MODULE_RANGE macro.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 7aa9dfc57b81..4a11a3d8447d 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -73,11 +73,6 @@ void (*board_nmi_handler_setup)(void);
 void (*board_ejtag_handler_setup)(void);
 void (*board_bind_eic_interrupt)(int irq, int regset);
 
-/*
- * These constant is for searching for possible module text segments.
- * MODULE_RANGE is a guess of how much space is likely to be vmalloced.
- */
-#define MODULE_RANGE (8*1024*1024)
 
 static void show_trace(unsigned long *stack)
 {
-- 
cgit v1.2.1


From 87151ae39bf5556abe83d69af0be9580c32c501b Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:17 +0200
Subject: [MIPS] Miscellaneous cleanup in prologue analysis code

We usually use backtrace term for dumping a call tree during
debug. Therefore this patch renames show_frametrace() into
show_backtrace() and show_trace() into show_raw_backtrace().

It also uses the new function print_ip_sym().

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 33 ++++++++++++++-------------------
 1 file changed, 14 insertions(+), 19 deletions(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 4a11a3d8447d..549cbb8209a3 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -74,21 +74,18 @@ void (*board_ejtag_handler_setup)(void);
 void (*board_bind_eic_interrupt)(int irq, int regset);
 
 
-static void show_trace(unsigned long *stack)
+static void show_raw_backtrace(unsigned long *sp)
 {
-	const int field = 2 * sizeof(unsigned long);
 	unsigned long addr;
 
 	printk("Call Trace:");
 #ifdef CONFIG_KALLSYMS
 	printk("\n");
 #endif
-	while (!kstack_end(stack)) {
-		addr = *stack++;
-		if (__kernel_text_address(addr)) {
-			printk(" [<%0*lx>] ", field, addr);
-			print_symbol("%s\n", addr);
-		}
+	while (!kstack_end(sp)) {
+		addr = *sp++;
+		if (__kernel_text_address(addr))
+			print_ip_sym(addr);
 	}
 	printk("\n");
 }
@@ -104,22 +101,20 @@ __setup("raw_show_trace", set_raw_show_trace);
 
 extern unsigned long unwind_stack(struct task_struct *task,
 				  unsigned long **sp, unsigned long pc);
-static void show_frametrace(struct task_struct *task, struct pt_regs *regs)
+static void show_backtrace(struct task_struct *task, struct pt_regs *regs)
 {
-	const int field = 2 * sizeof(unsigned long);
-	unsigned long *stack = (long *)regs->regs[29];
+	unsigned long *sp = (long *)regs->regs[29];
 	unsigned long pc = regs->cp0_epc;
 	int top = 1;
 
 	if (raw_show_trace || !__kernel_text_address(pc)) {
-		show_trace(stack);
+		show_raw_backtrace(sp);
 		return;
 	}
 	printk("Call Trace:\n");
 	while (__kernel_text_address(pc)) {
-		printk(" [<%0*lx>] ", field, pc);
-		print_symbol("%s\n", pc);
-		pc = unwind_stack(task, &stack, pc);
+		print_ip_sym(pc);
+		pc = unwind_stack(task, &sp, pc);
 		if (top && pc == 0)
 			pc = regs->regs[31];	/* leaf? */
 		top = 0;
@@ -127,7 +122,7 @@ static void show_frametrace(struct task_struct *task, struct pt_regs *regs)
 	printk("\n");
 }
 #else
-#define show_frametrace(task, r) show_trace((long *)(r)->regs[29]);
+#define show_backtrace(task, r) show_raw_backtrace((long *)(r)->regs[29]);
 #endif
 
 /*
@@ -160,7 +155,7 @@ static void show_stacktrace(struct task_struct *task, struct pt_regs *regs)
 		i++;
 	}
 	printk("\n");
-	show_frametrace(task, regs);
+	show_backtrace(task, regs);
 }
 
 static noinline void prepare_frametrace(struct pt_regs *regs)
@@ -211,11 +206,11 @@ void dump_stack(void)
 	if (!raw_show_trace) {
 		struct pt_regs regs;
 		prepare_frametrace(&regs);
-		show_frametrace(current, &regs);
+		show_backtrace(current, &regs);
 		return;
 	}
 #endif
-	show_trace(&stack);
+	show_raw_backtrace(&stack);
 }
 
 EXPORT_SYMBOL(dump_stack);
-- 
cgit v1.2.1


From 6057a7987608941a203f40f8b53513af433d8d2f Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:18 +0200
Subject: [MIPS] Make frame_info_init() more readable.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 93d5432759db..da332d707ce5 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -370,15 +370,15 @@ static int __init frame_info_init(void)
 	mfinfo[0].func = schedule;
 	schedule_frame = &mfinfo[0];
 #endif
-	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++) {
-		struct mips_frame_info *info = &mfinfo[i];
-		if (get_frame_info(info)) {
-			/* leaf or unknown */
-			if (info->func == schedule)
-				printk("Can't analyze prologue code at %p\n",
-				       info->func);
-		}
-	}
+	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++)
+		get_frame_info(mfinfo + i);
+
+	/*
+	 * Without schedule() frame info, result given by
+	 * thread_saved_pc() and get_wchan() are not reliable.
+	 */
+	if (schedule_frame->pc_offset < 0)
+		printk("Can't analyze schedule() prologue at %p\n", schedule);
 
 	mfinfo_num = i;
 	return 0;
-- 
cgit v1.2.1


From 1666a6fc73f724cdc6bd7d699f9ada4b04953efe Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:19 +0200
Subject: [MIPS] Simplify dump_stack()

Make dump_stack() code not depend on CONFIG_KALLSYMS.

It also make prepare_frametrace() always inlined to get
less false entries reported by show_raw_backtrace().

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 549cbb8209a3..303f00843021 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -158,7 +158,7 @@ static void show_stacktrace(struct task_struct *task, struct pt_regs *regs)
 	show_backtrace(task, regs);
 }
 
-static noinline void prepare_frametrace(struct pt_regs *regs)
+static __always_inline void prepare_frametrace(struct pt_regs *regs)
 {
 	__asm__ __volatile__(
 		"1: la $2, 1b\n\t"
@@ -200,17 +200,15 @@ void show_stack(struct task_struct *task, unsigned long *sp)
  */
 void dump_stack(void)
 {
-	unsigned long stack;
+	struct pt_regs regs;
 
-#ifdef CONFIG_KALLSYMS
-	if (!raw_show_trace) {
-		struct pt_regs regs;
-		prepare_frametrace(&regs);
-		show_backtrace(current, &regs);
-		return;
-	}
-#endif
-	show_raw_backtrace(&stack);
+	/*
+	 * Remove any garbage that may be in regs (specially func
+	 * addresses) to avoid show_raw_backtrace() to report them
+	 */
+	memset(&regs, 0, sizeof(regs));
+	prepare_frametrace(&regs);
+	show_backtrace(current, &regs);
 }
 
 EXPORT_SYMBOL(dump_stack);
-- 
cgit v1.2.1


From 0cceb4aa9acf6192a5f02134e764b1feeea8b9de Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:20 +0200
Subject: [MIPS] Make get_frame_info() more robust

Now get_frame_info() wants to detect move sp instruction first. It
assumes that the save ra in the stack instruction can't happen
before allocating frame size space into the stack.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index da332d707ce5..309bfa4a1520 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -321,17 +321,15 @@ static int get_frame_info(struct mips_frame_info *info)
 
 		if (is_jal_jalr_jr_ins(ip))
 			break;
-		if (is_sp_move_ins(ip)) {
-			if (info->frame_size)
-				continue;
-			info->frame_size = - ip->i_format.simmediate;
+		if (!info->frame_size) {
+			if (is_sp_move_ins(ip))
+				info->frame_size = - ip->i_format.simmediate;
+			continue;
 		}
-
-		if (is_ra_save_ins(ip)) {
-			if (info->pc_offset != -1)
-				continue;
+		if (info->pc_offset == -1 && is_ra_save_ins(ip)) {
 			info->pc_offset =
 				ip->i_format.simmediate / sizeof(long);
+			break;
 		}
 	}
 	if (info->frame_size && info->pc_offset >= 0) /* nested */
-- 
cgit v1.2.1


From 4d157d5eac29d7d5559fdcabf20f3961bc5cb3e7 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Thu, 3 Aug 2006 09:29:21 +0200
Subject: [MIPS] Improve unwind_stack()

This patch allows unwind_stack() to return ra for leaf function.
But it tries to detects cases where get_frame_info() wrongly
consider nested function as a leaf one.

It also pass 'unsinged long *sp' instead of 'unsigned long **sp'
as second parameter. The code looks cleaner.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 35 ++++++++++++++++++++++-------------
 arch/mips/kernel/traps.c   | 24 ++++++++++++------------
 2 files changed, 34 insertions(+), 25 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 309bfa4a1520..951bf9ca3ce9 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -448,15 +448,16 @@ unsigned long get_wchan(struct task_struct *p)
 }
 
 #ifdef CONFIG_KALLSYMS
-/* used by show_frametrace() */
-unsigned long unwind_stack(struct task_struct *task,
-			   unsigned long **sp, unsigned long pc)
+/* used by show_backtrace() */
+unsigned long unwind_stack(struct task_struct *task, unsigned long *sp,
+			   unsigned long pc, unsigned long ra)
 {
 	unsigned long stack_page;
 	struct mips_frame_info info;
 	char *modname;
 	char namebuf[KSYM_NAME_LEN + 1];
 	unsigned long size, ofs;
+	int leaf;
 
 	stack_page = (unsigned long)task_stack_page(task);
 	if (!stack_page)
@@ -469,18 +470,26 @@ unsigned long unwind_stack(struct task_struct *task,
 
 	info.func = (void *)(pc - ofs);
 	info.func_size = ofs;	/* analyze from start to ofs */
-	if (get_frame_info(&info)) {
-		/* leaf or unknown */
-		*sp += info.frame_size / sizeof(long);
+	leaf = get_frame_info(&info);
+	if (leaf < 0)
 		return 0;
-	}
-	if ((unsigned long)*sp < stack_page ||
-	    (unsigned long)*sp + info.frame_size / sizeof(long) >
-	    stack_page + THREAD_SIZE - 32)
+
+	if (*sp < stack_page ||
+	    *sp + info.frame_size > stack_page + THREAD_SIZE - 32)
 		return 0;
 
-	pc = (*sp)[info.pc_offset];
-	*sp += info.frame_size / sizeof(long);
-	return pc;
+	if (leaf)
+		/*
+		 * For some extreme cases, get_frame_info() can
+		 * consider wrongly a nested function as a leaf
+		 * one. In that cases avoid to return always the
+		 * same value.
+		 */
+		pc = pc != ra ? ra : 0;
+	else
+		pc = ((unsigned long *)(*sp))[info.pc_offset];
+
+	*sp += info.frame_size;
+	return __kernel_text_address(pc) ? pc : 0;
 }
 #endif
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 303f00843021..ab77034921c4 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -74,8 +74,9 @@ void (*board_ejtag_handler_setup)(void);
 void (*board_bind_eic_interrupt)(int irq, int regset);
 
 
-static void show_raw_backtrace(unsigned long *sp)
+static void show_raw_backtrace(unsigned long reg29)
 {
+	unsigned long *sp = (unsigned long *)reg29;
 	unsigned long addr;
 
 	printk("Call Trace:");
@@ -99,30 +100,29 @@ static int __init set_raw_show_trace(char *str)
 }
 __setup("raw_show_trace", set_raw_show_trace);
 
-extern unsigned long unwind_stack(struct task_struct *task,
-				  unsigned long **sp, unsigned long pc);
+extern unsigned long unwind_stack(struct task_struct *task, unsigned long *sp,
+				  unsigned long pc, unsigned long ra);
+
 static void show_backtrace(struct task_struct *task, struct pt_regs *regs)
 {
-	unsigned long *sp = (long *)regs->regs[29];
+	unsigned long sp = regs->regs[29];
+	unsigned long ra = regs->regs[31];
 	unsigned long pc = regs->cp0_epc;
-	int top = 1;
 
 	if (raw_show_trace || !__kernel_text_address(pc)) {
 		show_raw_backtrace(sp);
 		return;
 	}
 	printk("Call Trace:\n");
-	while (__kernel_text_address(pc)) {
+	do {
 		print_ip_sym(pc);
-		pc = unwind_stack(task, &sp, pc);
-		if (top && pc == 0)
-			pc = regs->regs[31];	/* leaf? */
-		top = 0;
-	}
+		pc = unwind_stack(task, &sp, pc, ra);
+		ra = 0;
+	} while (pc);
 	printk("\n");
 }
 #else
-#define show_backtrace(task, r) show_raw_backtrace((long *)(r)->regs[29]);
+#define show_backtrace(task, r) show_raw_backtrace((r)->regs[29]);
 #endif
 
 /*
-- 
cgit v1.2.1


From b4b30a5a0a270e6b3fef88373ad35d235a047fc1 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 3 Aug 2006 17:34:24 +0100
Subject: [MIPS] Cleanup leftovers of ARCH_HAS_IRQ_PER_CPU

CONFIG_IRQ_PER_CPU now controls the IRQ_PER_CPU stuff.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig      | 3 +--
 include/asm-mips/irq.h | 4 ----
 2 files changed, 1 insertion(+), 6 deletions(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 330f6abc7703..18e15a651d51 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1649,9 +1649,7 @@ config GENERIC_IRQ_PROBE
 	default y
 
 config IRQ_PER_CPU
-	depends on SMP
 	bool
-	default y
 
 #
 # - Highmem only makes sense for the 32-bit kernel.
@@ -1719,6 +1717,7 @@ source "mm/Kconfig"
 config SMP
 	bool "Multi-Processing support"
 	depends on SYS_SUPPORTS_SMP
+	select IRQ_PER_CPU
 	help
 	  This enables support for systems with more than one CPU. If you have
 	  a system with only one CPU, like most personal computers, say N. If
diff --git a/include/asm-mips/irq.h b/include/asm-mips/irq.h
index 896550bad322..d35c61776a02 100644
--- a/include/asm-mips/irq.h
+++ b/include/asm-mips/irq.h
@@ -76,8 +76,4 @@ extern int setup_irq_smtc(unsigned int irq, struct irqaction * new,
                           unsigned long hwmask);
 #endif /* CONFIG_MIPS_MT_SMTC */
 
-#ifdef CONFIG_SMP
-#define ARCH_HAS_IRQ_PER_CPU
-#endif
-
 #endif /* _ASM_IRQ_H */
-- 
cgit v1.2.1


From 048c6140c0f5d0bb1af8308c91dfa560ac65a296 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 3 Aug 2006 19:36:37 +0100
Subject: [MIPS] Don't call try_to_freeze in do_signal & co.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/irixsig.c  | 4 ----
 arch/mips/kernel/signal.c   | 4 ----
 arch/mips/kernel/signal32.c | 4 ----
 3 files changed, 12 deletions(-)

diff --git a/arch/mips/kernel/irixsig.c b/arch/mips/kernel/irixsig.c
index 676e868d26fb..e6f40609ba6a 100644
--- a/arch/mips/kernel/irixsig.c
+++ b/arch/mips/kernel/irixsig.c
@@ -186,9 +186,6 @@ asmlinkage int do_irix_signal(sigset_t *oldset, struct pt_regs *regs)
 	if (!user_mode(regs))
 		return 1;
 
-	if (try_to_freeze())
-		goto no_signal;
-
 	if (!oldset)
 		oldset = &current->blocked;
 
@@ -196,7 +193,6 @@ asmlinkage int do_irix_signal(sigset_t *oldset, struct pt_regs *regs)
 	if (signr > 0)
 		return handle_signal(signr, &info, &ka, oldset, regs);
 
-no_signal:
 	/*
 	 * Who's code doesn't conform to the restartable syscall convention
 	 * dies here!!!  The li instruction, a single machine instruction,
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index 6b4d9be31615..dc8956e3a396 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -424,9 +424,6 @@ void do_signal(struct pt_regs *regs)
 	if (!user_mode(regs))
 		return;
 
-	if (try_to_freeze())
-		goto no_signal;
-
 	if (test_thread_flag(TIF_RESTORE_SIGMASK))
 		oldset = &current->saved_sigmask;
 	else
@@ -448,7 +445,6 @@ void do_signal(struct pt_regs *regs)
 		}
 	}
 
-no_signal:
 	/*
 	 * Who's code doesn't conform to the restartable syscall convention
 	 * dies here!!!  The li instruction, a single machine instruction,
diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index f32a22997c3d..30980df56cd1 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -815,9 +815,6 @@ void do_signal32(struct pt_regs *regs)
 	if (!user_mode(regs))
 		return;
 
-	if (try_to_freeze())
-		goto no_signal;
-
 	if (test_thread_flag(TIF_RESTORE_SIGMASK))
 		oldset = &current->saved_sigmask;
 	else
@@ -838,7 +835,6 @@ void do_signal32(struct pt_regs *regs)
 		}
 	}
 
-no_signal:
 	/*
 	 * Who's code doesn't conform to the restartable syscall convention
 	 * dies here!!!  The li instruction, a single machine instruction,
-- 
cgit v1.2.1


From 1b223c861fd98080c2483f196b2496ac1f06f308 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 3 Aug 2006 21:53:10 +0100
Subject: [MIPS] IRIX: Crapectopy.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/irixsig.c | 55 +++++++++++++++++++++++++++++++++-------------
 1 file changed, 40 insertions(+), 15 deletions(-)

diff --git a/arch/mips/kernel/irixsig.c b/arch/mips/kernel/irixsig.c
index e6f40609ba6a..052ea15f1e80 100644
--- a/arch/mips/kernel/irixsig.c
+++ b/arch/mips/kernel/irixsig.c
@@ -17,6 +17,7 @@
 
 #include <asm/ptrace.h>
 #include <asm/uaccess.h>
+#include <asm/unistd.h>
 
 #undef DEBUG_SIG
 
@@ -172,11 +173,12 @@ static inline int handle_signal(unsigned long sig, siginfo_t *info,
 	return ret;
 }
 
-asmlinkage int do_irix_signal(sigset_t *oldset, struct pt_regs *regs)
+void do_irix_signal(struct pt_regs *regs)
 {
 	struct k_sigaction ka;
 	siginfo_t info;
 	int signr;
+	sigset_t *oldset;
 
 	/*
 	 * We want the common case to go fast, which is why we may in certain
@@ -184,14 +186,27 @@ asmlinkage int do_irix_signal(sigset_t *oldset, struct pt_regs *regs)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
-	if (signr > 0)
-		return handle_signal(signr, &info, &ka, oldset, regs);
+	if (signr > 0) {
+		/* Whee!  Actually deliver the signal.  */
+		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
+			/* a signal was successfully delivered; the saved
+			 * sigmask will have been stored in the signal frame,
+			 * and will be restored by sigreturn, so we can simply
+			 * clear the TIF_RESTORE_SIGMASK flag */
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
+
+		return;
+	}
 
 	/*
 	 * Who's code doesn't conform to the restartable syscall convention
@@ -204,8 +219,21 @@ asmlinkage int do_irix_signal(sigset_t *oldset, struct pt_regs *regs)
 		    regs->regs[2] == ERESTARTNOINTR) {
 			regs->cp0_epc -= 8;
 		}
+		if (regs->regs[2] == ERESTART_RESTARTBLOCK) {
+			regs->regs[2] = __NR_restart_syscall;
+			regs->regs[7] = regs->regs[26];
+			regs->cp0_epc -= 4;
+		}
+	}
+
+	/*
+	* If there's no signal to deliver, we just put the saved sigmask
+	* back
+	*/
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
 	}
-	return 0;
 }
 
 asmlinkage void
@@ -409,7 +437,7 @@ asmlinkage int irix_sigprocmask(int how, irix_sigset_t __user *new,
 
 asmlinkage int irix_sigsuspend(struct pt_regs *regs)
 {
-	sigset_t saveset, newset;
+	sigset_t newset;
 	sigset_t __user *uset;
 
 	uset = (sigset_t __user *) regs->regs[4];
@@ -418,18 +446,15 @@ asmlinkage int irix_sigsuspend(struct pt_regs *regs)
 	sigdelsetmask(&newset, ~_BLOCKABLE);
 
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	current->blocked = newset;
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs->regs[2] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_irix_signal(&saveset, regs))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 /* hate hate hate... */
-- 
cgit v1.2.1


From 45887e12f21780f00b861fdaff8e9d02a42eeb0f Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 3 Aug 2006 21:54:13 +0100
Subject: [MIPS] Add missing returns in signal code.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal.c   | 2 ++
 arch/mips/kernel/signal32.c | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index dc8956e3a396..56bf42911336 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -443,6 +443,8 @@ void do_signal(struct pt_regs *regs)
 			if (test_thread_flag(TIF_RESTORE_SIGMASK))
 				clear_thread_flag(TIF_RESTORE_SIGMASK);
 		}
+
+		return;
 	}
 
 	/*
diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index 30980df56cd1..282e85a7bae1 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -833,6 +833,8 @@ void do_signal32(struct pt_regs *regs)
 			if (test_thread_flag(TIF_RESTORE_SIGMASK))
 				clear_thread_flag(TIF_RESTORE_SIGMASK);
 		}
+
+		return;
 	}
 
 	/*
-- 
cgit v1.2.1


From bca70d24c09b740d6fd96b972011644cba8383d6 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Mon, 31 Jul 2006 23:05:04 +0900
Subject: [MIPS] Fix EV64120 PCI fixup in Makefile

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/pci/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/pci/Makefile b/arch/mips/pci/Makefile
index 35d5927706ea..368d3d896165 100644
--- a/arch/mips/pci/Makefile
+++ b/arch/mips/pci/Makefile
@@ -28,7 +28,7 @@ obj-$(CONFIG_DDB5477)		+= fixup-ddb5477.o pci-ddb5477.o ops-ddb5477.o
 obj-$(CONFIG_LASAT)		+= pci-lasat.o
 obj-$(CONFIG_MIPS_ATLAS)	+= fixup-atlas.o
 obj-$(CONFIG_MIPS_COBALT)	+= fixup-cobalt.o
-obj-$(CONFIG_MIPS_EV96100)	+= fixup-ev64120.o
+obj-$(CONFIG_MIPS_EV64120)	+= fixup-ev64120.o
 obj-$(CONFIG_MIPS_EV96100)	+= fixup-ev96100.o pci-ev96100.o
 obj-$(CONFIG_MIPS_ITE8172)	+= fixup-ite8172g.o
 obj-$(CONFIG_MIPS_IVR)		+= fixup-ivr.o
-- 
cgit v1.2.1


From 847b9dfccad7dd34b2e44b1c6ceeb1c4cb88084a Mon Sep 17 00:00:00 2001
From: Chris Dearman <chris@mips.com>
Date: Mon, 7 Aug 2006 15:08:01 +0100
Subject: [MIPS] MT: Initialise all writable bits in Cause register to zero.

Recent 34Ks come out of reset with WP enabled on VPE 1 so we take an
immediate exception when starting the second VPE.

Signed-off-by: Chris Dearman <chris@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smp-mt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/smp-mt.c b/arch/mips/kernel/smp-mt.c
index 93429a4d3012..766253c44f3f 100644
--- a/arch/mips/kernel/smp-mt.c
+++ b/arch/mips/kernel/smp-mt.c
@@ -203,7 +203,7 @@ void plat_smp_setup(void)
 				write_vpe_c0_config( read_c0_config());
 
 				/* make sure there are no software interrupts pending */
-				write_vpe_c0_cause(read_vpe_c0_cause() & ~(C_SW1|C_SW0));
+				write_vpe_c0_cause(0);
 
 				/* Propagate Config7 */
 				write_vpe_c0_config7(read_c0_config7());
-- 
cgit v1.2.1


From 13fdd31abec5f48cf97693bd14d2e11e0779b4ca Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 8 Aug 2006 03:47:01 +0100
Subject: [MIPS] Avoid double signal restarting.

In entry.S resume_userspace ... jal do_notify_resume form a loop through
which the kernel will iterate as long as work is pending.  If we
iterate through this loop more than once with no signal pending for at
least one but the last iteration we will take do the syscall restarting
multiple times resulting in a syscall return prior to the the syscall
instruction in userspace.  This may happen when debugging a multithreaded
program.

Debugging and original fix by Maciej; extended to other ABIs by me.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/irixsig.c  | 1 +
 arch/mips/kernel/signal.c   | 2 +-
 arch/mips/kernel/signal32.c | 1 +
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/mips/kernel/irixsig.c b/arch/mips/kernel/irixsig.c
index 052ea15f1e80..719364752e77 100644
--- a/arch/mips/kernel/irixsig.c
+++ b/arch/mips/kernel/irixsig.c
@@ -224,6 +224,7 @@ void do_irix_signal(struct pt_regs *regs)
 			regs->regs[7] = regs->regs[26];
 			regs->cp0_epc -= 4;
 		}
+		regs->regs[0] = 0;	/* Don't deal with this again.  */
 	}
 
 	/*
diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c
index 56bf42911336..b9d358e05214 100644
--- a/arch/mips/kernel/signal.c
+++ b/arch/mips/kernel/signal.c
@@ -429,7 +429,6 @@ void do_signal(struct pt_regs *regs)
 	else
 		oldset = &current->blocked;
 
-
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
 	if (signr > 0) {
 		/* Whee!  Actually deliver the signal.  */
@@ -464,6 +463,7 @@ void do_signal(struct pt_regs *regs)
 			regs->regs[7] = regs->regs[26];
 			regs->cp0_epc -= 4;
 		}
+		regs->regs[0] = 0;	/* Don't deal with this again.  */
 	}
 
 	/*
diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index 282e85a7bae1..c86a5ddff050 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -854,6 +854,7 @@ void do_signal32(struct pt_regs *regs)
 			regs->regs[7] = regs->regs[26];
 			regs->cp0_epc -= 4;
 		}
+		regs->regs[0] = 0;	/* Don't deal with this again.  */
 	}
 
 	/*
-- 
cgit v1.2.1


From 585fa72493edd7d5acb308806e7bb609412c6228 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 12 Aug 2006 16:40:08 +0100
Subject: [MIPS] Retire flush_icache_page from mm use.

On the 34K the redundant cache operations were causing excessive stalls
resulting in realtime code running on the second VPE missing its deadline.
For all other platforms this patch is just a significant performance
improvment as illustrated by below benchmark numbers.

Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host                 OS  Mhz null null      open slct sig  sig  fork exec sh
                             call  I/O stat clos TCP  inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
25Kf      2.6.18-rc4     533 0.49 1.16 7.57 33.4 30.5 1.34 12.4 5497 17.K 54.K
25Kf      2.6.18-rc4-p   533 0.49 1.16 6.68 23.0 30.7 1.36 8.55 5030 16.K 48.K
4Kc       2.6.18-rc4      80 4.21 15.0 131. 289. 261. 16.5 258. 18.K 70.K 227K
4Kc       2.6.18-rc4-p    80 4.34 13.1 128. 285. 262. 18.2 258. 12.K 52.K 176K
34Kc      2.6.18-rc4      40 5.01 14.0 61.6 90.0 477. 17.9 94.7 29.K 108K 342K
34Kc      2.6.18-rc4-p    40 4.98 13.9 61.2 89.7 475. 17.6 93.7 8758 44.K 158K
BCM1480   2.6.18-rc4     700 0.28 0.60 3.68 5.92 16.0 0.78 5.08 931. 3163 15.K
BCM1480   2.6.18-rc4-p   700 0.28 0.61 3.65 5.85 16.0 0.79 5.20 395. 1464 8385
TX49-16K  2.6.18-rc3     197 0.73 2.41 19.0 37.8 82.9 2.94 17.5 4438 14.K 56.K
TX49-16K  2.6.18-rc3-p   197 0.73 2.40 19.9 36.3 82.9 2.94 23.4 2577 9103 38.K
TX49-32K  2.6.18-rc3     396 0.36 1.19 6.80 11.8 41.0 1.46 8.17 2738 8465 32.K
TX49-32K  2.6.18-rc3-p   396 0.36 1.19 6.82 10.2 41.0 1.46 8.18 1330 4638 18.K

Original patch by me with enhancements by Atsushi Nemoto.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
---
 arch/mips/mm/c-r3k.c          |  2 +-
 arch/mips/mm/c-r4k.c          |  2 +-
 arch/mips/mm/c-sb1.c          |  2 +-
 arch/mips/mm/c-tx39.c         |  4 ++--
 arch/mips/mm/cache.c          | 18 ++++++++++--------
 include/asm-mips/cacheflush.h | 12 ++++++++----
 include/asm-mips/page.h       |  7 +++++--
 7 files changed, 28 insertions(+), 19 deletions(-)

diff --git a/arch/mips/mm/c-r3k.c b/arch/mips/mm/c-r3k.c
index bb041a22f20a..e1f35ef81145 100644
--- a/arch/mips/mm/c-r3k.c
+++ b/arch/mips/mm/c-r3k.c
@@ -335,7 +335,7 @@ void __init r3k_cache_init(void)
 	flush_cache_mm = r3k_flush_cache_mm;
 	flush_cache_range = r3k_flush_cache_range;
 	flush_cache_page = r3k_flush_cache_page;
-	flush_icache_page = r3k_flush_icache_page;
+	__flush_icache_page = r3k_flush_icache_page;
 	flush_icache_range = r3k_flush_icache_range;
 
 	flush_cache_sigtramp = r3k_flush_cache_sigtramp;
diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index b3ccb8738d29..4e1498246343 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -1291,7 +1291,7 @@ void __init r4k_cache_init(void)
 	__flush_cache_all	= r4k___flush_cache_all;
 	flush_cache_mm		= r4k_flush_cache_mm;
 	flush_cache_page	= r4k_flush_cache_page;
-	flush_icache_page	= r4k_flush_icache_page;
+	__flush_icache_page	= r4k_flush_icache_page;
 	flush_cache_range	= r4k_flush_cache_range;
 
 	flush_cache_sigtramp	= r4k_flush_cache_sigtramp;
diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c
index 2d71efb82ac5..4bd9ad8a5e07 100644
--- a/arch/mips/mm/c-sb1.c
+++ b/arch/mips/mm/c-sb1.c
@@ -520,7 +520,7 @@ void sb1_cache_init(void)
 
 	/* These routines are for Icache coherence with the Dcache */
 	flush_icache_range = sb1_flush_icache_range;
-	flush_icache_page = sb1_flush_icache_page;
+	__flush_icache_page = sb1_flush_icache_page;
 	flush_icache_all = __sb1_flush_icache_all; /* local only */
 
 	/* This implies an Icache flush too, so can't be nop'ed */
diff --git a/arch/mips/mm/c-tx39.c b/arch/mips/mm/c-tx39.c
index 5dfc9b1901f6..932a09d7ef84 100644
--- a/arch/mips/mm/c-tx39.c
+++ b/arch/mips/mm/c-tx39.c
@@ -382,7 +382,7 @@ void __init tx39_cache_init(void)
 		flush_cache_mm		= (void *) tx39h_flush_icache_all;
 		flush_cache_range	= (void *) tx39h_flush_icache_all;
 		flush_cache_page	= (void *) tx39h_flush_icache_all;
-		flush_icache_page	= (void *) tx39h_flush_icache_all;
+		__flush_icache_page	= (void *) tx39h_flush_icache_all;
 		flush_icache_range	= (void *) tx39h_flush_icache_all;
 
 		flush_cache_sigtramp	= (void *) tx39h_flush_icache_all;
@@ -408,7 +408,7 @@ void __init tx39_cache_init(void)
 		flush_cache_mm = tx39_flush_cache_mm;
 		flush_cache_range = tx39_flush_cache_range;
 		flush_cache_page = tx39_flush_cache_page;
-		flush_icache_page = tx39_flush_icache_page;
+		__flush_icache_page = tx39_flush_icache_page;
 		flush_icache_range = tx39_flush_icache_range;
 
 		flush_cache_sigtramp = tx39_flush_cache_sigtramp;
diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c
index ddd3a2de1d73..40c8b0235183 100644
--- a/arch/mips/mm/cache.c
+++ b/arch/mips/mm/cache.c
@@ -25,7 +25,7 @@ void (*flush_cache_range)(struct vm_area_struct *vma, unsigned long start,
 void (*flush_cache_page)(struct vm_area_struct *vma, unsigned long page,
 	unsigned long pfn);
 void (*flush_icache_range)(unsigned long start, unsigned long end);
-void (*flush_icache_page)(struct vm_area_struct *vma, struct page *page);
+void (*__flush_icache_page)(struct vm_area_struct *vma, struct page *page);
 
 /* MIPS specific cache operations */
 void (*flush_cache_sigtramp)(unsigned long addr);
@@ -70,6 +70,8 @@ void __flush_dcache_page(struct page *page)
 	struct address_space *mapping = page_mapping(page);
 	unsigned long addr;
 
+	if (PageHighMem(page))
+		return;
 	if (mapping && !mapping_mapped(mapping)) {
 		SetPageDcacheDirty(page);
 		return;
@@ -91,16 +93,16 @@ void __update_cache(struct vm_area_struct *vma, unsigned long address,
 {
 	struct page *page;
 	unsigned long pfn, addr;
+	int exec = (vma->vm_flags & VM_EXEC) && !cpu_has_ic_fills_f_dc;
 
 	pfn = pte_pfn(pte);
-	if (pfn_valid(pfn) && (page = pfn_to_page(pfn), page_mapping(page)) &&
-	    Page_dcache_dirty(page)) {
-		if (pages_do_alias((unsigned long)page_address(page),
-		                   address & PAGE_MASK)) {
-			addr = (unsigned long) page_address(page);
+	if (unlikely(!pfn_valid(pfn)))
+		return;
+	page = pfn_to_page(pfn);
+	if (page_mapping(page) && Page_dcache_dirty(page)) {
+		addr = (unsigned long) page_address(page);
+		if (exec || pages_do_alias(addr, address & PAGE_MASK))
 			flush_data_cache_page(addr);
-		}
-
 		ClearPageDcacheDirty(page);
 	}
 }
diff --git a/include/asm-mips/cacheflush.h b/include/asm-mips/cacheflush.h
index 47bc8f6c20d2..36416fdfcf68 100644
--- a/include/asm-mips/cacheflush.h
+++ b/include/asm-mips/cacheflush.h
@@ -21,7 +21,6 @@
  *  - flush_cache_range(vma, start, end) flushes a range of pages
  *  - flush_icache_range(start, end) flush a range of instructions
  *  - flush_dcache_page(pg) flushes(wback&invalidates) a page for dcache
- *  - flush_icache_page(vma, pg) flushes(invalidates) a page for icache
  *
  * MIPS specific flush operations:
  *
@@ -39,7 +38,7 @@ extern void __flush_dcache_page(struct page *page);
 
 static inline void flush_dcache_page(struct page *page)
 {
-	if (cpu_has_dc_aliases)
+	if (cpu_has_dc_aliases || !cpu_has_ic_fills_f_dc)
 		__flush_dcache_page(page);
 
 }
@@ -47,8 +46,13 @@ static inline void flush_dcache_page(struct page *page)
 #define flush_dcache_mmap_lock(mapping)		do { } while (0)
 #define flush_dcache_mmap_unlock(mapping)	do { } while (0)
 
-extern void (*flush_icache_page)(struct vm_area_struct *vma,
+extern void (*__flush_icache_page)(struct vm_area_struct *vma,
 	struct page *page);
+static inline void flush_icache_page(struct vm_area_struct *vma,
+	struct page *page)
+{
+}
+
 extern void (*flush_icache_range)(unsigned long start, unsigned long end);
 #define flush_cache_vmap(start, end)		flush_cache_all()
 #define flush_cache_vunmap(start, end)		flush_cache_all()
@@ -60,7 +64,7 @@ static inline void copy_to_user_page(struct vm_area_struct *vma,
 	if (cpu_has_dc_aliases)
 		flush_cache_page(vma, vaddr, page_to_pfn(page));
 	memcpy(dst, src, len);
-	flush_icache_page(vma, page);
+	__flush_icache_page(vma, page);
 }
 
 static inline void copy_from_user_page(struct vm_area_struct *vma,
diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index 219d359861f3..c3b872b047d7 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -34,6 +34,8 @@
 
 #ifndef __ASSEMBLY__
 
+#include <asm/cpu-features.h>
+
 extern void clear_page(void * page);
 extern void copy_page(void * to, void * from);
 
@@ -53,7 +55,7 @@ static inline void clear_user_page(void *addr, unsigned long vaddr,
 	extern void (*flush_data_cache_page)(unsigned long addr);
 
 	clear_page(addr);
-	if (pages_do_alias((unsigned long) addr, vaddr))
+	if (pages_do_alias((unsigned long) addr, vaddr & PAGE_MASK))
 		flush_data_cache_page((unsigned long)addr);
 }
 
@@ -63,7 +65,8 @@ static inline void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
 	extern void (*flush_data_cache_page)(unsigned long addr);
 
 	copy_page(vto, vfrom);
-	if (pages_do_alias((unsigned long)vto, vaddr))
+	if (!cpu_has_ic_fills_f_dc ||
+	    pages_do_alias((unsigned long)vto, vaddr & PAGE_MASK))
 		flush_data_cache_page((unsigned long)vto);
 }
 
-- 
cgit v1.2.1


From c487d2a5a03163a1e0bd35ea72f7e5a6bd8b0d56 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 12 Aug 2006 22:56:03 +0100
Subject: [MIPS] eXcite: Don't set SERIAL_RM9000.

The driver has not been merged yet so selecting it results in a warning
message.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 18e15a651d51..4bb813389e9e 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -126,7 +126,6 @@ config BASLER_EXCITE
 	select IRQ_CPU
 	select IRQ_CPU_RM7K
 	select IRQ_CPU_RM9K
-	select SERIAL_RM9000
 	select SYS_HAS_CPU_RM9000
 	select SYS_SUPPORTS_32BIT_KERNEL
 	select SYS_SUPPORTS_64BIT_KERNEL
-- 
cgit v1.2.1


From 73b76c78fdf3a25faecbffb52c925950d46152b6 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 15 Aug 2006 19:36:40 +0100
Subject: [MIPS] MT: When doing "select SMP" also select SMP's prerequesites or
 ...

... kconfig will do weird stuff.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 4bb813389e9e..991b329224e7 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -1526,6 +1526,7 @@ config MIPS_MT_SMTC
 	select CPU_MIPSR2_SRS
 	select MIPS_MT
 	select SMP
+	select SYS_SUPPORTS_SMP
 	help
 	  This is a kernel model which is known a SMTC or lately has been
 	  marketesed into SMVP.
@@ -1537,6 +1538,7 @@ config MIPS_MT_SMP
 	select CPU_MIPSR2_SRS
 	select MIPS_MT
 	select SMP
+	select SYS_SUPPORTS_SMP
 	help
 	  This is a kernel model which is also known a VSMP or lately
 	  has been marketesed into SMVP.
-- 
cgit v1.2.1


From dc41fb43966ae1318fe34a8e5f959924a7c05d81 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 16 Aug 2006 00:59:40 +0100
Subject: [MIPS] Fix 32-bit kernel by replacing 64-bit-only code.

dclz() expects its 64-bit argument being passed as a single register
but on 32-bit kernels it'll actually be in a register pair.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/sibyte/bcm1480/irq.c | 19 ++-----------------
 arch/mips/sibyte/sb1250/irq.c  | 17 +----------------
 2 files changed, 3 insertions(+), 33 deletions(-)

diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index ed325f0ab28a..a0222fa4416c 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -469,21 +469,6 @@ void bcm1480_kgdb_interrupt(struct pt_regs *regs)
 
 #endif 	/* CONFIG_KGDB */
 
-static inline int dclz(unsigned long long x)
-{
-	int lz;
-
-	__asm__ (
-	"	.set	push						\n"
-	"	.set	mips64						\n"
-	"	dclz	%0, %1						\n"
-	"	.set	pop						\n"
-	: "=r" (lz)
-	: "r" (x));
-
-	return lz;
-}
-
 extern void bcm1480_timer_interrupt(struct pt_regs *regs);
 extern void bcm1480_mailbox_interrupt(struct pt_regs *regs);
 extern void bcm1480_kgdb_interrupt(struct pt_regs *regs);
@@ -536,9 +521,9 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
 
 		if (mask_h) {
 			if (mask_h ^ 1)
-				do_IRQ(63 - dclz(mask_h), regs);
+				do_IRQ(fls64(mask_h) - 1, regs);
 			else
-				do_IRQ(127 - dclz(mask_l), regs);
+				do_IRQ(63 + fls64(mask_l), regs);
 		}
 	}
 }
diff --git a/arch/mips/sibyte/sb1250/irq.c b/arch/mips/sibyte/sb1250/irq.c
index 1de71adec6c6..a451b4c7732d 100644
--- a/arch/mips/sibyte/sb1250/irq.c
+++ b/arch/mips/sibyte/sb1250/irq.c
@@ -419,21 +419,6 @@ static void sb1250_kgdb_interrupt(struct pt_regs *regs)
 
 #endif 	/* CONFIG_KGDB */
 
-static inline int dclz(unsigned long long x)
-{
-	int lz;
-
-	__asm__ (
-	"	.set	push						\n"
-	"	.set	mips64						\n"
-	"	dclz	%0, %1						\n"
-	"	.set	pop						\n"
-	: "=r" (lz)
-	: "r" (x));
-
-	return lz;
-}
-
 extern void sb1250_timer_interrupt(struct pt_regs *regs);
 extern void sb1250_mailbox_interrupt(struct pt_regs *regs);
 extern void sb1250_kgdb_interrupt(struct pt_regs *regs);
@@ -490,6 +475,6 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
 		mask = __raw_readq(IOADDR(A_IMR_REGISTER(smp_processor_id(),
 		                              R_IMR_INTERRUPT_STATUS_BASE)));
 		if (mask)
-			do_IRQ(63 - dclz(mask), regs);
+			do_IRQ(fls64(mask) - 1, regs);
 	}
 }
-- 
cgit v1.2.1


From 6e74bae9a0c2fc59ffb0e25fec074b4ac0ac7048 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 16 Aug 2006 14:05:11 +0100
Subject: [MIPS] SMTC Build fix.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smtc-asm.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/smtc-asm.S b/arch/mips/kernel/smtc-asm.S
index 4cc3dea36612..76cb31d57482 100644
--- a/arch/mips/kernel/smtc-asm.S
+++ b/arch/mips/kernel/smtc-asm.S
@@ -8,7 +8,7 @@
 #include <asm/regdef.h>
 #include <asm/asmmacro.h>
 #include <asm/stackframe.h>
-#include <asm/stackframe.h>
+#include <asm/irqflags.h>
 
 /*
  * "Software Interrupt" linkage.
-- 
cgit v1.2.1


From a94d702049569401c65b579d0751ce282f962b41 Mon Sep 17 00:00:00 2001
From: Elizabeth Oldham <beth@mips.com>
Date: Thu, 17 Aug 2006 12:39:21 +0100
Subject: [MIPS] MT: Fix setting of XTC.

XTC can only be set if VPA is clear, which it may not be. There is
also the possibility of a back to back c0 register access hazard to
take care of.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/vpe.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c
index 9ee0ec2cd067..51ddd2166898 100644
--- a/arch/mips/kernel/vpe.c
+++ b/arch/mips/kernel/vpe.c
@@ -768,10 +768,16 @@ int vpe_run(struct vpe * v)
 	 */
  	write_tc_c0_tcbind((read_tc_c0_tcbind() & ~TCBIND_CURVPE) | v->minor);
 
+	write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() & ~(VPECONF0_VPA));
+
+	back_to_back_c0_hazard();
+
         /* Set up the XTC bit in vpeconf0 to point at our tc */
         write_vpe_c0_vpeconf0( (read_vpe_c0_vpeconf0() & ~(VPECONF0_XTC))
                                | (t->index << VPECONF0_XTC_SHIFT));
 
+	back_to_back_c0_hazard();
+
         /* enable this VPE */
         write_vpe_c0_vpeconf0(read_vpe_c0_vpeconf0() | VPECONF0_VPA);
 
-- 
cgit v1.2.1


From f6502791d780b22fc147150137704a07a05ba361 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Fri, 25 Aug 2006 17:55:31 +0900
Subject: [MIPS] Do not use drop_mmu_context to flusing other task's VIPT
 I-cache.

c-r4k.c and c-sb1.c use drop_mmu_context() to flush virtually tagged
I-caches, but this does not work for flushing other task's icache.  This
is for example triggered by copy_to_user_page() called from ptrace(2).
Use indexed flush for such cases.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/c-r4k.c |  4 ++--
 arch/mips/mm/c-sb1.c | 56 +++++++++++++++++++++++++++++-----------------------
 2 files changed, 33 insertions(+), 27 deletions(-)

diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 4e1498246343..2d729f6f6348 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -475,7 +475,7 @@ static inline void local_r4k_flush_cache_page(void *args)
 		}
 	}
 	if (exec) {
-		if (cpu_has_vtag_icache) {
+		if (cpu_has_vtag_icache && mm == current->active_mm) {
 			int cpu = smp_processor_id();
 
 			if (cpu_context(cpu, mm) != 0)
@@ -599,7 +599,7 @@ static inline void local_r4k_flush_icache_page(void *args)
 	 * We're not sure of the virtual address(es) involved here, so
 	 * we have to flush the entire I-cache.
 	 */
-	if (cpu_has_vtag_icache) {
+	if (cpu_has_vtag_icache && vma->vm_mm == current->active_mm) {
 		int cpu = smp_processor_id();
 
 		if (cpu_context(cpu, vma->vm_mm) != 0)
diff --git a/arch/mips/mm/c-sb1.c b/arch/mips/mm/c-sb1.c
index 4bd9ad8a5e07..16bad7c0a63f 100644
--- a/arch/mips/mm/c-sb1.c
+++ b/arch/mips/mm/c-sb1.c
@@ -154,6 +154,26 @@ static inline void __sb1_flush_icache_all(void)
 	}
 }
 
+/*
+ * Invalidate a range of the icache.  The addresses are virtual, and
+ * the cache is virtually indexed and tagged.  However, we don't
+ * necessarily have the right ASID context, so use index ops instead
+ * of hit ops.
+ */
+static inline void __sb1_flush_icache_range(unsigned long start,
+	unsigned long end)
+{
+	start &= ~(icache_line_size - 1);
+	end = (end + icache_line_size - 1) & ~(icache_line_size - 1);
+
+	while (start != end) {
+		cache_set_op(Index_Invalidate_I, start & icache_index_mask);
+		start += icache_line_size;
+	}
+	mispredict();
+	sync();
+}
+
 /*
  * Flush the icache for a given physical page.  Need to writeback the
  * dcache first, then invalidate the icache.  If the page isn't
@@ -173,8 +193,11 @@ static void local_sb1_flush_cache_page(struct vm_area_struct *vma, unsigned long
 	/*
 	 * Bumping the ASID is probably cheaper than the flush ...
 	 */
-	if (cpu_context(cpu, vma->vm_mm) != 0)
-		drop_mmu_context(vma->vm_mm, cpu);
+	if (vma->vm_mm == current->active_mm) {
+		if (cpu_context(cpu, vma->vm_mm) != 0)
+			drop_mmu_context(vma->vm_mm, cpu);
+	} else
+		__sb1_flush_icache_range(addr, addr + PAGE_SIZE);
 }
 
 #ifdef CONFIG_SMP
@@ -210,26 +233,6 @@ void sb1_flush_cache_page(struct vm_area_struct *vma, unsigned long addr, unsign
 	__attribute__((alias("local_sb1_flush_cache_page")));
 #endif
 
-/*
- * Invalidate a range of the icache.  The addresses are virtual, and
- * the cache is virtually indexed and tagged.  However, we don't
- * necessarily have the right ASID context, so use index ops instead
- * of hit ops.
- */
-static inline void __sb1_flush_icache_range(unsigned long start,
-	unsigned long end)
-{
-	start &= ~(icache_line_size - 1);
-	end = (end + icache_line_size - 1) & ~(icache_line_size - 1);
-
-	while (start != end) {
-		cache_set_op(Index_Invalidate_I, start & icache_index_mask);
-		start += icache_line_size;
-	}
-	mispredict();
-	sync();
-}
-
 
 /*
  * Invalidate all caches on this CPU
@@ -326,9 +329,12 @@ static void local_sb1_flush_icache_page(struct vm_area_struct *vma,
 	 * If there's a context, bump the ASID (cheaper than a flush,
 	 * since we don't know VAs!)
 	 */
-	if (cpu_context(cpu, vma->vm_mm) != 0) {
-		drop_mmu_context(vma->vm_mm, cpu);
-	}
+	if (vma->vm_mm == current->active_mm) {
+		if (cpu_context(cpu, vma->vm_mm) != 0)
+			drop_mmu_context(vma->vm_mm, cpu);
+	} else
+		__sb1_flush_icache_range(start, start + PAGE_SIZE);
+
 }
 
 #ifdef CONFIG_SMP
-- 
cgit v1.2.1


From 898d229107f7aa8ea45685c11e2930783755f9ba Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 29 Aug 2006 12:10:22 +0900
Subject: [MIPS] Make prepare_frametrace() not clobber v0

Since lmo commit 323a380bf9e1a1679a774a2b053e3c1f2aa3f179 ("Simplify
dump_stack()") made prepare_frametrace() always inlined, using $2 (v0)
in __asm__ is not safe anymore.  We can use $1 (at) instead.  Also we
should use "dla" instead of "la" for 64-bit kernel.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index ab77034921c4..e51d8fd9a152 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -161,16 +161,20 @@ static void show_stacktrace(struct task_struct *task, struct pt_regs *regs)
 static __always_inline void prepare_frametrace(struct pt_regs *regs)
 {
 	__asm__ __volatile__(
-		"1: la $2, 1b\n\t"
+		".set push\n\t"
+		".set noat\n\t"
 #ifdef CONFIG_64BIT
-		"sd $2, %0\n\t"
+		"1: dla $1, 1b\n\t"
+		"sd $1, %0\n\t"
 		"sd $29, %1\n\t"
 		"sd $31, %2\n\t"
 #else
-		"sw $2, %0\n\t"
+		"1: la $1, 1b\n\t"
+		"sw $1, %0\n\t"
 		"sw $29, %1\n\t"
 		"sw $31, %2\n\t"
 #endif
+		".set pop\n\t"
 		: "=m" (regs->cp0_epc),
 		"=m" (regs->regs[29]), "=m" (regs->regs[31])
 		: : "memory");
-- 
cgit v1.2.1


From 32136568a9828c27c07eedb8d2b3e3d9fe8d87ba Mon Sep 17 00:00:00 2001
From: Alexander Bigga <ab@mycable.de>
Date: Tue, 29 Aug 2006 16:48:34 +0200
Subject: [MIPS] Fix for pci config_access on alchemy au1x000

I've encountered a serious problem with PCI config space access on Au1x000
platforms with recent 2.6.x-kernel. With 2.4.31 the same hardware works fine.
So I was looking for the differences:

Symptoms:
- no PCI-device is seen on bootup though two or three cards are present
- lspci output is empty
- OR: lspci shows 20 times the same device
(- OR: in some slot-configurations it worked anyhow)

System(s):
1. platform with Au1500 and three PCI-devices (actually a mycable XXS1500
    with backplane for three PCI-devices)
2. platform with Au1550 and two PCI-devices (custom board)

Debugging:
I digged down to the config_access() of the au1xxx-processors in
arch/mips/pci/ops-au1000.c and switched on DEBUG.

The code of config_access() seems to be almost the same as of the
2.4.x-kernel. But the "pci_cfg_vm->addr" returned by get_vm_area(0x2000, 0)
once on booting is different. That's of course not forbidden. But the
alignment seems to be wrong. In my case, I received:

2.4.31: pci_cfg_vm->addr = c0000000
2.6.18-rc5: pci_cfg_vm->addr = c0101000

To make it short: With 2.6.x it fails on the first config-access with:
"PCI ERR detected: status 83a00356".

Fixup:
My fix is now, to use the VM_IOREMAP-flag in the get_vm_area call. This flag
seems to be introduced in mm/vmalloc.c a long time ago (in 2.6.7-bk13, I
found in gitweb).
Now, the returned address is pci_cfg_vm->addr = c0104000 and everything works
fine.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/pci/ops-au1000.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/pci/ops-au1000.c b/arch/mips/pci/ops-au1000.c
index 0c0c1e6519f9..8ae46481fcb7 100644
--- a/arch/mips/pci/ops-au1000.c
+++ b/arch/mips/pci/ops-au1000.c
@@ -110,7 +110,7 @@ static int config_access(unsigned char access_type, struct pci_bus *bus,
 	if (first_cfg) {
 		/* reserve a wired entry for pci config accesses */
 		first_cfg = 0;
-		pci_cfg_vm = get_vm_area(0x2000, 0);
+		pci_cfg_vm = get_vm_area(0x2000, VM_IOREMAP);
 		if (!pci_cfg_vm)
 			panic (KERN_ERR "PCI unable to get vm area\n");
 		pci_cfg_wired_entry = read_c0_wired();
-- 
cgit v1.2.1


From 09f451bfb9fe5ceea763a9ebd6ec73dd05e4faf8 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@mips.com>
Date: Wed, 30 Aug 2006 14:29:57 +0100
Subject: [MIPS] SEAD defconfig build fix

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Makefile | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index d333ce4ba26b..f4227d24227d 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -330,6 +330,7 @@ load-$(CONFIG_MIPS_MALTA)	+= 0xffffffff80100000
 # MIPS SEAD board
 #
 core-$(CONFIG_MIPS_SEAD)	+= arch/mips/mips-boards/sead/
+cflags-$(CONFIG_MIPS_SEAD)	+= -Iinclude/asm-mips/mach-mips
 load-$(CONFIG_MIPS_SEAD)	+= 0xffffffff80100000
 
 #
-- 
cgit v1.2.1


From 65316fd13ad9d82560edbad0a940d684380f7461 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 31 Aug 2006 14:16:06 +0100
Subject: [MIPS] Replace generic__raw_read_trylock usage

generic__raw_read_trylock() is a defect generic function actually doing
a __raw_read_lock ...

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/spinlock.h | 47 ++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 46 insertions(+), 1 deletion(-)

diff --git a/include/asm-mips/spinlock.h b/include/asm-mips/spinlock.h
index 669b8e349ff2..4c1a1b53aeaf 100644
--- a/include/asm-mips/spinlock.h
+++ b/include/asm-mips/spinlock.h
@@ -239,7 +239,51 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 	: "memory");
 }
 
-#define __raw_read_trylock(lock) generic__raw_read_trylock(lock)
+static inline int __raw_read_trylock(raw_rwlock_t *rw)
+{
+	unsigned int tmp;
+	int ret;
+
+	if (R10000_LLSC_WAR) {
+		__asm__ __volatile__(
+		"	.set	noreorder	# __raw_read_trylock	\n"
+		"	li	%2, 0					\n"
+		"1:	ll	%1, %3					\n"
+		"	bnez	%1, 2f					\n"
+		"	 addu	%1, 1					\n"
+		"	sc	%1, %0					\n"
+		"	beqzl	%1, 1b					\n"
+		"	.set	reorder					\n"
+#ifdef CONFIG_SMP
+		"	 sync						\n"
+#endif
+		"	li	%2, 1					\n"
+		"2:							\n"
+		: "=m" (rw->lock), "=&r" (tmp), "=&r" (ret)
+		: "m" (rw->lock)
+		: "memory");
+	} else {
+		__asm__ __volatile__(
+		"	.set	noreorder	# __raw_read_trylock	\n"
+		"	li	%2, 0					\n"
+		"1:	ll	%1, %3					\n"
+		"	bnez	%1, 2f					\n"
+		"	 addu	%1, 1					\n"
+		"	sc	%1, %0					\n"
+		"	beqz	%1, 1b					\n"
+		"	.set	reorder					\n"
+#ifdef CONFIG_SMP
+		"	 sync						\n"
+#endif
+		"	li	%2, 1					\n"
+		"2:							\n"
+		: "=m" (rw->lock), "=&r" (tmp), "=&r" (ret)
+		: "m" (rw->lock)
+		: "memory");
+	}
+
+	return ret;
+}
 
 static inline int __raw_write_trylock(raw_rwlock_t *rw)
 {
@@ -283,4 +327,5 @@ static inline int __raw_write_trylock(raw_rwlock_t *rw)
 	return ret;
 }
 
+
 #endif /* _ASM_SPINLOCK_H */
-- 
cgit v1.2.1


From d34555fb20e7abf33f86d7aa3ec0826343f38256 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 31 Aug 2006 19:39:09 +0100
Subject: [MIPS] Do not lose upper 32-bit on MIPS32 with 64-bit addresses in
 __pte().

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/page.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index c3b872b047d7..85b258ee7090 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -77,15 +77,17 @@ static inline void copy_user_page(void *vto, void *vfrom, unsigned long vaddr,
   #ifdef CONFIG_CPU_MIPS32
     typedef struct { unsigned long pte_low, pte_high; } pte_t;
     #define pte_val(x)    ((x).pte_low | ((unsigned long long)(x).pte_high << 32))
+    #define __pte(x)      ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; })
   #else
      typedef struct { unsigned long long pte; } pte_t;
      #define pte_val(x)	((x).pte)
+     #define __pte(x)	((pte_t) { (x) } )
   #endif
 #else
 typedef struct { unsigned long pte; } pte_t;
 #define pte_val(x)	((x).pte)
-#endif
 #define __pte(x)	((pte_t) { (x) } )
+#endif
 
 /*
  * For 3-level pagetables we defines these ourselves, for 2-level the
-- 
cgit v1.2.1


From 8f9a2b324644d3f8c233287f0a7764d61655cda4 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Thu, 7 Sep 2006 01:00:22 +0900
Subject: [MIPS] Fix errors detected by "make headers_check"

* export asm/sgidefs.h
* include asm/isadep.h only if in kernel
* do not export contents of asm/timex.h and asm/user.h

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/Kbuild   | 2 ++
 include/asm-mips/ptrace.h | 3 +--
 include/asm-mips/timex.h  | 4 ++++
 include/asm-mips/user.h   | 4 ++++
 4 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/include/asm-mips/Kbuild b/include/asm-mips/Kbuild
index c68e1680da01..01ea3e045652 100644
--- a/include/asm-mips/Kbuild
+++ b/include/asm-mips/Kbuild
@@ -1 +1,3 @@
 include include/asm-generic/Kbuild.asm
+
+header-y += sgidefs.h
diff --git a/include/asm-mips/ptrace.h b/include/asm-mips/ptrace.h
index 4113316ee0da..4fb0fc43ffd7 100644
--- a/include/asm-mips/ptrace.h
+++ b/include/asm-mips/ptrace.h
@@ -10,8 +10,6 @@
 #define _ASM_PTRACE_H
 
 
-#include <asm/isadep.h>
-
 /* 0 - 31 are integer registers, 32 - 63 are fp registers.  */
 #define FPR_BASE	32
 #define PC		64
@@ -73,6 +71,7 @@ struct pt_regs {
 #ifdef __KERNEL__
 
 #include <linux/linkage.h>
+#include <asm/isadep.h>
 
 /*
  * Does the process account for user or for system time?
diff --git a/include/asm-mips/timex.h b/include/asm-mips/timex.h
index 98aa737b34aa..b80de8e0fbbd 100644
--- a/include/asm-mips/timex.h
+++ b/include/asm-mips/timex.h
@@ -8,6 +8,8 @@
 #ifndef _ASM_TIMEX_H
 #define _ASM_TIMEX_H
 
+#ifdef __KERNEL__
+
 #include <asm/mipsregs.h>
 
 /*
@@ -51,4 +53,6 @@ static inline cycles_t get_cycles (void)
 	return read_c0_count();
 }
 
+#endif /* __KERNEL__ */
+
 #endif /*  _ASM_TIMEX_H */
diff --git a/include/asm-mips/user.h b/include/asm-mips/user.h
index 89bf8b4cab3c..61f2a093b91b 100644
--- a/include/asm-mips/user.h
+++ b/include/asm-mips/user.h
@@ -8,6 +8,8 @@
 #ifndef _ASM_USER_H
 #define _ASM_USER_H
 
+#ifdef __KERNEL__
+
 #include <asm/page.h>
 #include <asm/reg.h>
 
@@ -55,4 +57,6 @@ struct user {
 #define HOST_DATA_START_ADDR	(u.start_data)
 #define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
 
+#endif /* __KERNEL__ */
+
 #endif /* _ASM_USER_H */
-- 
cgit v1.2.1


From 7fdeb048141b363a23b8cf6f6a226d74aca4d724 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 6 Sep 2006 22:42:02 +0900
Subject: [MIPS] Wire up set_robust_list(2) and get_robust_list(2)

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/scall32-o32.S |  2 ++
 arch/mips/kernel/scall64-64.S  |  2 ++
 arch/mips/kernel/scall64-n32.S |  4 +++-
 arch/mips/kernel/scall64-o32.S |  2 ++
 include/asm-mips/unistd.h      | 18 ++++++++++++------
 5 files changed, 21 insertions(+), 7 deletions(-)

diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index ba1bcd83c7d3..e71785102206 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -662,6 +662,8 @@ einval:	li	v0, -EINVAL
 	sys	sys_tee			4
 	sys	sys_vmsplice		4
 	sys	sys_move_pages		6
+	sys	sys_set_robust_list	2
+	sys	sys_get_robust_list	3
 	.endm
 
 	/* We pre-compute the number of _instruction_ bytes needed to
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index 939e172db953..4c22d0b4825d 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -466,3 +466,5 @@ sys_call_table:
 	PTR	sys_tee				/* 5265 */
 	PTR	sys_vmsplice
 	PTR	sys_move_pages
+	PTR	sys_set_robust_list
+	PTR	sys_get_robust_list
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 549b4bcb53a5..f25c2a2f1038 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -390,5 +390,7 @@ EXPORT(sysn32_call_table)
 	PTR	sys_splice
 	PTR	sys_sync_file_range
 	PTR	sys_tee
-	PTR	sys_vmsplice			/* 6271 */
+	PTR	sys_vmsplice			/* 6270 */
 	PTR	sys_move_pages
+	PTR	compat_sys_set_robust_list
+	PTR	compat_sys_get_robust_list
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 505c9ee54009..2ac01412f747 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -514,4 +514,6 @@ sys_call_table:
 	PTR	sys_tee
 	PTR	sys_vmsplice
 	PTR	compat_sys_move_pages
+	PTR	compat_sys_set_robust_list
+	PTR	compat_sys_get_robust_list	/* 4310 */
 	.size	sys_call_table,.-sys_call_table
diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index 610ccb8a50b3..558e3cba4f92 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -329,16 +329,18 @@
 #define __NR_tee			(__NR_Linux + 306)
 #define __NR_vmsplice			(__NR_Linux + 307)
 #define __NR_move_pages			(__NR_Linux + 308)
+#define __NR_set_robust_list		(__NR_Linux + 309)
+#define __NR_get_robust_list		(__NR_Linux + 310)
 
 /*
  * Offset of the last Linux o32 flavoured syscall
  */
-#define __NR_Linux_syscalls		308
+#define __NR_Linux_syscalls		310
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */
 
 #define __NR_O32_Linux			4000
-#define __NR_O32_Linux_syscalls		308
+#define __NR_O32_Linux_syscalls		310
 
 #if _MIPS_SIM == _MIPS_SIM_ABI64
 
@@ -614,16 +616,18 @@
 #define __NR_tee			(__NR_Linux + 265)
 #define __NR_vmsplice			(__NR_Linux + 266)
 #define __NR_move_pages			(__NR_Linux + 267)
+#define __NR_set_robust_list		(__NR_Linux + 268)
+#define __NR_get_robust_list		(__NR_Linux + 269)
 
 /*
  * Offset of the last Linux 64-bit flavoured syscall
  */
-#define __NR_Linux_syscalls		267
+#define __NR_Linux_syscalls		269
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */
 
 #define __NR_64_Linux			5000
-#define __NR_64_Linux_syscalls		267
+#define __NR_64_Linux_syscalls		269
 
 #if _MIPS_SIM == _MIPS_SIM_NABI32
 
@@ -903,16 +907,18 @@
 #define __NR_tee			(__NR_Linux + 269)
 #define __NR_vmsplice			(__NR_Linux + 270)
 #define __NR_move_pages			(__NR_Linux + 271)
+#define __NR_set_robust_list		(__NR_Linux + 272)
+#define __NR_get_robust_list		(__NR_Linux + 273)
 
 /*
  * Offset of the last N32 flavoured syscall
  */
-#define __NR_Linux_syscalls		271
+#define __NR_Linux_syscalls		273
 
 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */
 
 #define __NR_N32_Linux			6000
-#define __NR_N32_Linux_syscalls		271
+#define __NR_N32_Linux_syscalls		273
 
 #ifdef __KERNEL__
 
-- 
cgit v1.2.1


From 60a6c3777ec607c5b19df9eac35088db4e142a6b Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Thu, 8 Jun 2006 01:09:01 +0900
Subject: [MIPS] Reduce race between cpu_wait() and need_resched() checking

If a thread became runnable between need_resched() and the WAIT
instruction, switching to the thread will delay until a next interrupt.
Some CPUs can execute the WAIT instruction with interrupt disabled, so
we can get rid of this race on them (at least UP case).

Original Patch by Atsushi with fixing up for MIPS Technology's cores by
Ralf based on feedback from the RTL designers.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/cpu-probe.c | 62 ++++++++++++++++++++++++++++++++------------
 1 file changed, 45 insertions(+), 17 deletions(-)

diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index aa2caa67299a..9fbf8430c849 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -38,15 +38,40 @@ static void r3081_wait(void)
 
 static void r39xx_wait(void)
 {
-	unsigned long cfg = read_c0_conf();
-	write_c0_conf(cfg | TX39_CONF_HALT);
+	local_irq_disable();
+	if (!need_resched())
+		write_c0_conf(read_c0_conf() | TX39_CONF_HALT);
+	local_irq_enable();
 }
 
+/*
+ * There is a race when WAIT instruction executed with interrupt
+ * enabled.
+ * But it is implementation-dependent wheter the pipelie restarts when
+ * a non-enabled interrupt is requested.
+ */
 static void r4k_wait(void)
 {
-	__asm__(".set\tmips3\n\t"
-		"wait\n\t"
-		".set\tmips0");
+	__asm__("	.set	mips3			\n"
+		"	wait				\n"
+		"	.set	mips0			\n");
+}
+
+/*
+ * This variant is preferable as it allows testing need_resched and going to
+ * sleep depending on the outcome atomically.  Unfortunately the "It is
+ * implementation-dependent whether the pipeline restarts when a non-enabled
+ * interrupt is requested" restriction in the MIPS32/MIPS64 architecture makes
+ * using this version a gamble.
+ */
+static void r4k_wait_irqoff(void)
+{
+	local_irq_disable();
+	if (!need_resched())
+		__asm__("	.set	mips3		\n"
+			"	wait			\n"
+			"	.set	mips0		\n");
+	local_irq_enable();
 }
 
 /* The Au1xxx wait is available only if using 32khz counter or
@@ -56,17 +81,17 @@ int allow_au1k_wait;
 static void au1k_wait(void)
 {
 	/* using the wait instruction makes CP0 counter unusable */
-	__asm__(".set mips3\n\t"
-		"cache 0x14, 0(%0)\n\t"
-		"cache 0x14, 32(%0)\n\t"
-		"sync\n\t"
-		"nop\n\t"
-		"wait\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		"nop\n\t"
-		".set mips0\n\t"
+	__asm__("	.set	mips3			\n"
+		"	cache	0x14, 0(%0)		\n"
+		"	cache	0x14, 32(%0)		\n"
+		"	sync				\n"
+		"	nop				\n"
+		"	wait				\n"
+		"	nop				\n"
+		"	nop				\n"
+		"	nop				\n"
+		"	nop				\n"
+		"	.set	mips0			\n"
 		: : "r" (au1k_wait));
 }
 
@@ -111,7 +136,6 @@ static inline void check_wait(void)
 	case CPU_NEVADA:
 	case CPU_RM7000:
 	case CPU_RM9000:
-	case CPU_TX49XX:
 	case CPU_4KC:
 	case CPU_4KEC:
 	case CPU_4KSC:
@@ -125,6 +149,10 @@ static inline void check_wait(void)
 		cpu_wait = r4k_wait;
 		printk(" available.\n");
 		break;
+	case CPU_TX49XX:
+		cpu_wait = r4k_wait_irqoff;
+		printk(" available.\n");
+		break;
 	case CPU_AU1000:
 	case CPU_AU1100:
 	case CPU_AU1500:
-- 
cgit v1.2.1


From 846acaa2b4974ae2e28038d024dedcfc184efbb7 Mon Sep 17 00:00:00 2001
From: "Kevin D. Kissell" <kevink@mips.com>
Date: Tue, 12 Sep 2006 12:08:08 +0200
Subject: [MIPS] Patch to arch/mips/mips-boards/generic/time.c

In hooking up the perf counter overflow interrupt to the experimental
deprecated-real-soon-now /proc/perf interface last night, I had to
revisit arch/mips/mips-boards/generic/time.c, and discovered that
when the 2.6.9-based SMTC prototype was merged with the more
recent tree, it was missed that arch/mips/kernel/time.c had changed
so that even in SMP kernels, timer_interrupt() calls
local_timer_interrupt(), so there is no longer a need to invoke it
directly from mips_timer_interrupt() in those cases where
timer_interrupt() has been called.  So I got rid of that, and added the
invocation of perf_irq() if Cause.PCI is set, more-or-less following the
same logic as in the non-SMTC case, with the modifications that (a) a
runtime check for Release 2 isn't done, because it's redundant in SMTC),
and (b) we check for a clock interrupt regardless of the value returned
by the perf counter service - I don't understand why we'd want to control
that with perf_irq(), but maybe one of you knows the story.  I also got
rid of the stupid warning about the unused variable when compiled for
SMTC (another artifact of the merge). The result hasn't been beaten to
death, but boots, seems stable, and supports extended precision event
counting.

Signed-off-by: Kevin D. Kissell <kevink@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mips-boards/generic/time.c | 57 +++++++++++++++++++++++-------------
 1 file changed, 36 insertions(+), 21 deletions(-)

diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c
index 557bf961f36a..de5798e1e015 100644
--- a/arch/mips/mips-boards/generic/time.c
+++ b/arch/mips/mips-boards/generic/time.c
@@ -92,10 +92,9 @@ extern int (*perf_irq)(struct pt_regs *regs);
 irqreturn_t mips_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
 	int cpu = smp_processor_id();
-	int r2 = cpu_has_mips_r2;
 
 #ifdef CONFIG_MIPS_MT_SMTC
-        /*
+	/*
 	 *  In an SMTC system, one Count/Compare set exists per VPE.
 	 *  Which TC within a VPE gets the interrupt is essentially
 	 *  random - we only know that it shouldn't be one with
@@ -108,29 +107,46 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 	 *  the general MIPS timer_interrupt routine.
 	 */
 
+	int vpflags;
+
 	/*
-	 * DVPE is necessary so long as cross-VPE interrupts
-	 * are done via read-modify-write of Cause register.
+	 * We could be here due to timer interrupt,
+	 * perf counter overflow, or both.
 	 */
-	int vpflags = dvpe();
-	write_c0_compare (read_c0_count() - 1);
-	clear_c0_cause(CPUCTR_IMASKBIT);
-	evpe(vpflags);
-
-	if (cpu_data[cpu].vpe_id == 0) {
-		timer_interrupt(irq, dev_id, regs);
-		scroll_display_message();
-	} else
-		write_c0_compare (read_c0_count() + ( mips_hpt_frequency/HZ));
-	smtc_timer_broadcast(cpu_data[cpu].vpe_id);
+	if (read_c0_cause() & (1 << 26))
+		perf_irq(regs);
 
-	if (cpu != 0)
+	if (read_c0_cause() & (1 << 30)) {
+		/* If timer interrupt, make it de-assert */
+		write_c0_compare (read_c0_count() - 1);
 		/*
-		 * Other CPUs should do profiling and process accounting
+		 * DVPE is necessary so long as cross-VPE interrupts
+		 * are done via read-modify-write of Cause register.
 		 */
-		local_timer_interrupt(irq, dev_id, regs);
-
+		vpflags = dvpe();
+		clear_c0_cause(CPUCTR_IMASKBIT);
+		evpe(vpflags);
+		/*
+		 * There are things we only want to do once per tick
+		 * in an "MP" system.   One TC of each VPE will take
+		 * the actual timer interrupt.  The others will get
+		 * timer broadcast IPIs. We use whoever it is that takes
+		 * the tick on VPE 0 to run the full timer_interrupt().
+		 */
+		if (cpu_data[cpu].vpe_id == 0) {
+				timer_interrupt(irq, NULL, regs);
+				smtc_timer_broadcast(cpu_data[cpu].vpe_id);
+				scroll_display_message();
+		} else {
+			write_c0_compare(read_c0_count() +
+			                 (mips_hpt_frequency/HZ));
+			local_timer_interrupt(irq, dev_id, regs);
+			smtc_timer_broadcast(cpu_data[cpu].vpe_id);
+		}
+	}
 #else /* CONFIG_MIPS_MT_SMTC */
+	int r2 = cpu_has_mips_r2;
+
 	if (cpu == 0) {
 		/*
 		 * CPU 0 handles the global timer interrupt job and process
@@ -161,9 +177,8 @@ irqreturn_t mips_timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 		 */
 		local_timer_interrupt(irq, dev_id, regs);
 	}
-#endif /* CONFIG_MIPS_MT_SMTC */
-
 out:
+#endif /* CONFIG_MIPS_MT_SMTC */
 	return IRQ_HANDLED;
 }
 
-- 
cgit v1.2.1


From 3ee24e1b1e0b5ae413a85ba63677a7110915e3af Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@mips.com>
Date: Tue, 12 Sep 2006 19:02:44 +0100
Subject: [MIPS] Atlas: Fix building the RTC driver

Atlas maps its RTC chip in the host mmio space rather than using the
"traditional" location in the PCI/ISA port space.  A change that has
happened to the generic RTC header requires to define ARCH_RTC_LOCATION
now.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mach-atlas/mc146818rtc.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/asm-mips/mach-atlas/mc146818rtc.h b/include/asm-mips/mach-atlas/mc146818rtc.h
index 397522ea5565..fc09928571d6 100644
--- a/include/asm-mips/mach-atlas/mc146818rtc.h
+++ b/include/asm-mips/mach-atlas/mc146818rtc.h
@@ -28,6 +28,8 @@
 #include <asm/mips-boards/atlas.h>
 #include <asm/mips-boards/atlasint.h>
 
+#define ARCH_RTC_LOCATION
+
 #define RTC_PORT(x)	(ATLAS_RTC_ADR_REG + (x) * 8)
 #define RTC_IO_EXTENT	0x100
 #define RTC_IOMAPPED	0
-- 
cgit v1.2.1


From fc095a902181b72ce77a10feb7b36ba1cbacd736 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@mips.com>
Date: Tue, 12 Sep 2006 19:12:18 +0100
Subject: [MIPS] Atlas: update interrupt handling

The following change updates the Atlas interrupt handling to match that
of Malta.  Tested with a 5Kc and a 34Kf successfully.

Signed-off-by: Maciej W. Rozycki <macro@mips.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mips-boards/atlas/atlas_int.c   | 105 ++++++++++++++++++++-----
 arch/mips/mips-boards/atlas/atlas_setup.c |   2 +-
 arch/mips/mips-boards/generic/time.c      |   7 +-
 arch/mips/pci/fixup-atlas.c               |  20 ++---
 include/asm-mips/mach-atlas/mc146818rtc.h |   2 +-
 include/asm-mips/mips-boards/atlasint.h   | 124 +++++++++++++++++++++---------
 6 files changed, 189 insertions(+), 71 deletions(-)

diff --git a/arch/mips/mips-boards/atlas/atlas_int.c b/arch/mips/mips-boards/atlas/atlas_int.c
index fb25e0377f11..a020a3cb4f4b 100644
--- a/arch/mips/mips-boards/atlas/atlas_int.c
+++ b/arch/mips/mips-boards/atlas/atlas_int.c
@@ -1,6 +1,8 @@
 /*
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 1999, 2000, 2006  MIPS Technologies, Inc.
+ *	All rights reserved.
+ *	Authors: Carsten Langgaard <carstenl@mips.com>
+ *		 Maciej W. Rozycki <macro@mips.com>
  *
  * ########################################################################
  *
@@ -25,17 +27,20 @@
  */
 #include <linux/compiler.h>
 #include <linux/init.h>
+#include <linux/irq.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 #include <linux/kernel_stat.h>
 
-#include <asm/irq.h>
+#include <asm/gdb-stub.h>
 #include <asm/io.h>
+#include <asm/irq_cpu.h>
+#include <asm/msc01_ic.h>
+
 #include <asm/mips-boards/atlas.h>
 #include <asm/mips-boards/atlasint.h>
-#include <asm/gdb-stub.h>
-
+#include <asm/mips-boards/generic.h>
 
 static struct atlas_ictrl_regs *atlas_hw0_icregs;
 
@@ -47,13 +52,13 @@ static struct atlas_ictrl_regs *atlas_hw0_icregs;
 
 void disable_atlas_irq(unsigned int irq_nr)
 {
-	atlas_hw0_icregs->intrsten = (1 << (irq_nr-ATLASINT_BASE));
+	atlas_hw0_icregs->intrsten = 1 << (irq_nr - ATLAS_INT_BASE);
 	iob();
 }
 
 void enable_atlas_irq(unsigned int irq_nr)
 {
-	atlas_hw0_icregs->intseten = (1 << (irq_nr-ATLASINT_BASE));
+	atlas_hw0_icregs->intseten = 1 << (irq_nr - ATLAS_INT_BASE);
 	iob();
 }
 
@@ -107,7 +112,7 @@ static inline void atlas_hw0_irqdispatch(struct pt_regs *regs)
 	if (unlikely(int_status == 0))
 		return;
 
-	irq = ATLASINT_BASE + ls1bit32(int_status);
+	irq = ATLAS_INT_BASE + ls1bit32(int_status);
 
 	DEBUG_INT("atlas_hw0_irqdispatch: irq=%d\n", irq);
 
@@ -161,15 +166,14 @@ static inline unsigned int irq_ffs(unsigned int pending)
 }
 
 /*
- * IRQs on the Atlas board look basically (barring software IRQs which we
- * don't use at all and all external interrupt sources are combined together
- * on hardware interrupt 0 (MIPS IRQ 2)) like:
+ * IRQs on the Atlas board look basically like (all external interrupt
+ * sources are combined together on hardware interrupt 0 (MIPS IRQ 2)):
  *
- *	MIPS IRQ	Source
+ *      MIPS IRQ        Source
  *      --------        ------
- *             0	Software (ignored)
- *             1        Software (ignored)
- *             2        Combined hardware interrupt (hw0)
+ *             0        Software 0 (reschedule IPI on MT)
+ *             1        Software 1 (remote call IPI on MT)
+ *             2        Combined Atlas hardware interrupt (hw0)
  *             3        Hardware (ignored)
  *             4        Hardware (ignored)
  *             5        Hardware (ignored)
@@ -179,7 +183,7 @@ static inline unsigned int irq_ffs(unsigned int pending)
  * We handle the IRQ according to _our_ priority which is:
  *
  * Highest ----     R4k Timer
- * Lowest  ----     Combined hardware interrupt
+ * Lowest  ----     Software 0
  *
  * then we just return, if multiple IRQs are pending then we will just take
  * another exception, big deal.
@@ -193,17 +197,19 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
 
 	if (irq == MIPSCPU_INT_ATLAS)
 		atlas_hw0_irqdispatch(regs);
-	else if (irq > 0)
+	else if (irq >= 0)
 		do_IRQ(MIPSCPU_INT_BASE + irq, regs);
 	else
 		spurious_interrupt(regs);
 }
 
-void __init arch_init_irq(void)
+static inline void init_atlas_irqs (int base)
 {
 	int i;
 
-	atlas_hw0_icregs = (struct atlas_ictrl_regs *)ioremap (ATLAS_ICTRL_REGS_BASE, sizeof(struct atlas_ictrl_regs *));
+	atlas_hw0_icregs = (struct atlas_ictrl_regs *)
+			   ioremap(ATLAS_ICTRL_REGS_BASE,
+				   sizeof(struct atlas_ictrl_regs *));
 
 	/*
 	 * Mask out all interrupt by writing "1" to all bit position in
@@ -211,7 +217,7 @@ void __init arch_init_irq(void)
 	 */
 	atlas_hw0_icregs->intrsten = 0xffffffff;
 
-	for (i = ATLASINT_BASE; i <= ATLASINT_END; i++) {
+	for (i = ATLAS_INT_BASE; i <= ATLAS_INT_END; i++) {
 		irq_desc[i].status	= IRQ_DISABLED;
 		irq_desc[i].action	= 0;
 		irq_desc[i].depth	= 1;
@@ -219,3 +225,62 @@ void __init arch_init_irq(void)
 		spin_lock_init(&irq_desc[i].lock);
 	}
 }
+
+static struct irqaction atlasirq = {
+	.handler = no_action,
+	.name = "Atlas cascade"
+};
+
+msc_irqmap_t __initdata msc_irqmap[] = {
+	{MSC01C_INT_TMR,		MSC01_IRQ_EDGE, 0},
+	{MSC01C_INT_PCI,		MSC01_IRQ_LEVEL, 0},
+};
+int __initdata msc_nr_irqs = sizeof(msc_irqmap) / sizeof(*msc_irqmap);
+
+msc_irqmap_t __initdata msc_eicirqmap[] = {
+	{MSC01E_INT_SW0,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_SW1,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_ATLAS,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_TMR,		MSC01_IRQ_EDGE, 0},
+	{MSC01E_INT_PCI,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_PERFCTR,		MSC01_IRQ_LEVEL, 0},
+	{MSC01E_INT_CPUCTR,		MSC01_IRQ_LEVEL, 0}
+};
+int __initdata msc_nr_eicirqs = sizeof(msc_eicirqmap) / sizeof(*msc_eicirqmap);
+
+void __init arch_init_irq(void)
+{
+	init_atlas_irqs(ATLAS_INT_BASE);
+
+	if (!cpu_has_veic)
+		mips_cpu_irq_init(MIPSCPU_INT_BASE);
+
+	switch(mips_revision_corid) {
+	case MIPS_REVISION_CORID_CORE_MSC:
+	case MIPS_REVISION_CORID_CORE_FPGA2:
+	case MIPS_REVISION_CORID_CORE_FPGA3:
+	case MIPS_REVISION_CORID_CORE_24K:
+	case MIPS_REVISION_CORID_CORE_EMUL_MSC:
+		if (cpu_has_veic)
+			init_msc_irqs (MSC01E_INT_BASE,
+				       msc_eicirqmap, msc_nr_eicirqs);
+		else
+			init_msc_irqs (MSC01C_INT_BASE,
+				       msc_irqmap, msc_nr_irqs);
+	}
+
+
+	if (cpu_has_veic) {
+		set_vi_handler (MSC01E_INT_ATLAS, atlas_hw0_irqdispatch);
+		setup_irq (MSC01E_INT_BASE + MSC01E_INT_ATLAS, &atlasirq);
+	} else if (cpu_has_vint) {
+		set_vi_handler (MIPSCPU_INT_ATLAS, atlas_hw0_irqdispatch);
+#ifdef CONFIG_MIPS_MT_SMTC
+		setup_irq_smtc (MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS,
+				&atlasirq, (0x100 << MIPSCPU_INT_ATLAS));
+#else /* Not SMTC */
+		setup_irq(MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS, &atlasirq);
+#endif /* CONFIG_MIPS_MT_SMTC */
+	} else
+		setup_irq(MIPSCPU_INT_BASE + MIPSCPU_INT_ATLAS, &atlasirq);
+}
diff --git a/arch/mips/mips-boards/atlas/atlas_setup.c b/arch/mips/mips-boards/atlas/atlas_setup.c
index 9871a91fdb07..0c6b0ce15028 100644
--- a/arch/mips/mips-boards/atlas/atlas_setup.c
+++ b/arch/mips/mips-boards/atlas/atlas_setup.c
@@ -77,7 +77,7 @@ static void __init serial_init(void)
 #else
 	s.iobase = ATLAS_UART_REGS_BASE+3;
 #endif
-	s.irq = ATLASINT_UART;
+	s.irq = ATLAS_INT_UART;
 	s.uartclk = ATLAS_BASE_BAUD * 16;
 	s.flags = UPF_BOOT_AUTOCONF | UPF_SKIP_TEST | UPF_AUTO_IRQ;
 	s.iotype = UPIO_PORT;
diff --git a/arch/mips/mips-boards/generic/time.c b/arch/mips/mips-boards/generic/time.c
index de5798e1e015..8d15861fce61 100644
--- a/arch/mips/mips-boards/generic/time.c
+++ b/arch/mips/mips-boards/generic/time.c
@@ -41,8 +41,13 @@
 
 #include <asm/mips-boards/generic.h>
 #include <asm/mips-boards/prom.h>
+
+#ifdef CONFIG_MIPS_ATLAS
+#include <asm/mips-boards/atlasint.h>
+#endif
+#ifdef CONFIG_MIPS_MALTA
 #include <asm/mips-boards/maltaint.h>
-#include <asm/mc146818-time.h>
+#endif
 
 unsigned long cpu_khz;
 
diff --git a/arch/mips/pci/fixup-atlas.c b/arch/mips/pci/fixup-atlas.c
index 439510af3037..c6cd6e9cdfbc 100644
--- a/arch/mips/pci/fixup-atlas.c
+++ b/arch/mips/pci/fixup-atlas.c
@@ -21,16 +21,16 @@
 
 #include <asm/mips-boards/atlasint.h>
 
-#define PCIA		ATLASINT_PCIA
-#define PCIB		ATLASINT_PCIB
-#define PCIC		ATLASINT_PCIC
-#define PCID		ATLASINT_PCID
-#define INTA		ATLASINT_INTA
-#define INTB		ATLASINT_INTB
-#define ETH		ATLASINT_ETH
-#define INTC		ATLASINT_INTC
-#define SCSI		ATLASINT_SCSI
-#define INTD		ATLASINT_INTD
+#define PCIA		ATLAS_INT_PCIA
+#define PCIB		ATLAS_INT_PCIB
+#define PCIC		ATLAS_INT_PCIC
+#define PCID		ATLAS_INT_PCID
+#define INTA		ATLAS_INT_INTA
+#define INTB		ATLAS_INT_INTB
+#define ETH		ATLAS_INT_ETH
+#define INTC		ATLAS_INT_INTC
+#define SCSI		ATLAS_INT_SCSI
+#define INTD		ATLAS_INT_INTD
 
 static char irq_tab[][5] __initdata = {
 	/*      INTA    INTB    INTC    INTD */
diff --git a/include/asm-mips/mach-atlas/mc146818rtc.h b/include/asm-mips/mach-atlas/mc146818rtc.h
index fc09928571d6..a73a5698420c 100644
--- a/include/asm-mips/mach-atlas/mc146818rtc.h
+++ b/include/asm-mips/mach-atlas/mc146818rtc.h
@@ -33,7 +33,7 @@
 #define RTC_PORT(x)	(ATLAS_RTC_ADR_REG + (x) * 8)
 #define RTC_IO_EXTENT	0x100
 #define RTC_IOMAPPED	0
-#define RTC_IRQ		ATLASINT_RTC
+#define RTC_IRQ		ATLAS_INT_RTC
 
 static inline unsigned char CMOS_READ(unsigned long addr)
 {
diff --git a/include/asm-mips/mips-boards/atlasint.h b/include/asm-mips/mips-boards/atlasint.h
index fd7ebc54fa90..b15e4ea0b091 100644
--- a/include/asm-mips/mips-boards/atlasint.h
+++ b/include/asm-mips/mips-boards/atlasint.h
@@ -1,6 +1,7 @@
 /*
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999 MIPS Technologies, Inc.  All rights reserved.
+ * Copyright (C) 1999, 2006  MIPS Technologies, Inc.  All rights reserved.
+ *	Authors: Carsten Langgaard <carstenl@mips.com>
+ *		 Maciej W. Rozycki <macro@mips.com>
  *
  * ########################################################################
  *
@@ -25,41 +26,88 @@
 #ifndef _MIPS_ATLASINT_H
 #define _MIPS_ATLASINT_H
 
-#define ATLASINT_BASE		1
-#define ATLASINT_UART		(ATLASINT_BASE+0)
-#define ATLASINT_TIM0		(ATLASINT_BASE+1)
-#define ATLASINT_RES2		(ATLASINT_BASE+2)
-#define ATLASINT_RES3		(ATLASINT_BASE+3)
-#define ATLASINT_RTC		(ATLASINT_BASE+4)
-#define ATLASINT_COREHI		(ATLASINT_BASE+5)
-#define ATLASINT_CORELO		(ATLASINT_BASE+6)
-#define ATLASINT_RES7		(ATLASINT_BASE+7)
-#define ATLASINT_PCIA		(ATLASINT_BASE+8)
-#define ATLASINT_PCIB		(ATLASINT_BASE+9)
-#define ATLASINT_PCIC		(ATLASINT_BASE+10)
-#define ATLASINT_PCID		(ATLASINT_BASE+11)
-#define ATLASINT_ENUM		(ATLASINT_BASE+12)
-#define ATLASINT_DEG		(ATLASINT_BASE+13)
-#define ATLASINT_ATXFAIL	(ATLASINT_BASE+14)
-#define ATLASINT_INTA		(ATLASINT_BASE+15)
-#define ATLASINT_INTB		(ATLASINT_BASE+16)
-#define ATLASINT_ETH		ATLASINT_INTB
-#define ATLASINT_INTC		(ATLASINT_BASE+17)
-#define ATLASINT_SCSI		ATLASINT_INTC
-#define ATLASINT_INTD		(ATLASINT_BASE+18)
-#define ATLASINT_SERR		(ATLASINT_BASE+19)
-#define ATLASINT_RES20		(ATLASINT_BASE+20)
-#define ATLASINT_RES21		(ATLASINT_BASE+21)
-#define ATLASINT_RES22		(ATLASINT_BASE+22)
-#define ATLASINT_RES23		(ATLASINT_BASE+23)
-#define ATLASINT_RES24		(ATLASINT_BASE+24)
-#define ATLASINT_RES25		(ATLASINT_BASE+25)
-#define ATLASINT_RES26		(ATLASINT_BASE+26)
-#define ATLASINT_RES27		(ATLASINT_BASE+27)
-#define ATLASINT_RES28		(ATLASINT_BASE+28)
-#define ATLASINT_RES29		(ATLASINT_BASE+29)
-#define ATLASINT_RES30		(ATLASINT_BASE+30)
-#define ATLASINT_RES31		(ATLASINT_BASE+31)
-#define ATLASINT_END		(ATLASINT_BASE+31)
+/*
+ * Interrupts 0..7 are used for Atlas CPU interrupts (nonEIC mode)
+ */
+#define MIPSCPU_INT_BASE	0
+
+/* CPU interrupt offsets */
+#define MIPSCPU_INT_SW0		0
+#define MIPSCPU_INT_SW1		1
+#define MIPSCPU_INT_MB0		2
+#define MIPSCPU_INT_ATLAS	MIPSCPU_INT_MB0
+#define MIPSCPU_INT_MB1		3
+#define MIPSCPU_INT_MB2		4
+#define MIPSCPU_INT_MB3		5
+#define MIPSCPU_INT_MB4		6
+#define MIPSCPU_INT_CPUCTR	7
+
+/*
+ * Interrupts 8..39 are used for Atlas interrupt controller interrupts
+ */
+#define ATLAS_INT_BASE		8
+#define ATLAS_INT_UART		(ATLAS_INT_BASE + 0)
+#define ATLAS_INT_TIM0		(ATLAS_INT_BASE + 1)
+#define ATLAS_INT_RES2		(ATLAS_INT_BASE + 2)
+#define ATLAS_INT_RES3		(ATLAS_INT_BASE + 3)
+#define ATLAS_INT_RTC		(ATLAS_INT_BASE + 4)
+#define ATLAS_INT_COREHI	(ATLAS_INT_BASE + 5)
+#define ATLAS_INT_CORELO	(ATLAS_INT_BASE + 6)
+#define ATLAS_INT_RES7		(ATLAS_INT_BASE + 7)
+#define ATLAS_INT_PCIA		(ATLAS_INT_BASE + 8)
+#define ATLAS_INT_PCIB		(ATLAS_INT_BASE + 9)
+#define ATLAS_INT_PCIC		(ATLAS_INT_BASE + 10)
+#define ATLAS_INT_PCID		(ATLAS_INT_BASE + 11)
+#define ATLAS_INT_ENUM		(ATLAS_INT_BASE + 12)
+#define ATLAS_INT_DEG		(ATLAS_INT_BASE + 13)
+#define ATLAS_INT_ATXFAIL	(ATLAS_INT_BASE + 14)
+#define ATLAS_INT_INTA		(ATLAS_INT_BASE + 15)
+#define ATLAS_INT_INTB		(ATLAS_INT_BASE + 16)
+#define ATLAS_INT_ETH		ATLAS_INT_INTB
+#define ATLAS_INT_INTC		(ATLAS_INT_BASE + 17)
+#define ATLAS_INT_SCSI		ATLAS_INT_INTC
+#define ATLAS_INT_INTD		(ATLAS_INT_BASE + 18)
+#define ATLAS_INT_SERR		(ATLAS_INT_BASE + 19)
+#define ATLAS_INT_RES20		(ATLAS_INT_BASE + 20)
+#define ATLAS_INT_RES21		(ATLAS_INT_BASE + 21)
+#define ATLAS_INT_RES22		(ATLAS_INT_BASE + 22)
+#define ATLAS_INT_RES23		(ATLAS_INT_BASE + 23)
+#define ATLAS_INT_RES24		(ATLAS_INT_BASE + 24)
+#define ATLAS_INT_RES25		(ATLAS_INT_BASE + 25)
+#define ATLAS_INT_RES26		(ATLAS_INT_BASE + 26)
+#define ATLAS_INT_RES27		(ATLAS_INT_BASE + 27)
+#define ATLAS_INT_RES28		(ATLAS_INT_BASE + 28)
+#define ATLAS_INT_RES29		(ATLAS_INT_BASE + 29)
+#define ATLAS_INT_RES30		(ATLAS_INT_BASE + 30)
+#define ATLAS_INT_RES31		(ATLAS_INT_BASE + 31)
+#define ATLAS_INT_END		(ATLAS_INT_BASE + 31)
+
+/*
+ * Interrupts 64..127 are used for Soc-it Classic interrupts
+ */
+#define MSC01C_INT_BASE		64
+
+/* SOC-it Classic interrupt offsets */
+#define MSC01C_INT_TMR		0
+#define MSC01C_INT_PCI		1
+
+/*
+ * Interrupts 64..127 are used for Soc-it EIC interrupts
+ */
+#define MSC01E_INT_BASE		64
+
+/* SOC-it EIC interrupt offsets */
+#define	MSC01E_INT_SW0		1
+#define	MSC01E_INT_SW1		2
+#define	MSC01E_INT_MB0		3
+#define	MSC01E_INT_ATLAS	MSC01E_INT_MB0
+#define	MSC01E_INT_MB1		4
+#define	MSC01E_INT_MB2		5
+#define	MSC01E_INT_MB3		6
+#define	MSC01E_INT_MB4		7
+#define	MSC01E_INT_TMR		8
+#define	MSC01E_INT_PCI		9
+#define	MSC01E_INT_PERFCTR	10
+#define	MSC01E_INT_CPUCTR	11
 
 #endif /* !(_MIPS_ATLASINT_H) */
-- 
cgit v1.2.1


From 00932ba3052a84ed73fab34ff1ec5a413b77fbab Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 16 Sep 2006 01:29:37 +0100
Subject: [MIPS] Make PROT_WRITE imply PROT_READ.

---
 arch/mips/mm/fault.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
index e3a617224868..a4f8c45c4e8e 100644
--- a/arch/mips/mm/fault.c
+++ b/arch/mips/mm/fault.c
@@ -89,7 +89,7 @@ good_area:
 		if (!(vma->vm_flags & VM_WRITE))
 			goto bad_area;
 	} else {
-		if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
+		if (!(vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)))
 			goto bad_area;
 	}
 
-- 
cgit v1.2.1


From 717736d4d73f8966fcc9802732f52f9e85830247 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 19 Sep 2006 17:16:56 +0100
Subject: [MIPS] TLS: Delete unused sys32_set_thread_area

There is no need for a compat version.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/linux32.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 450ac592da57..dc500e20cf14 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -1296,9 +1296,3 @@ _sys32_clone(nabi_no_regargs struct pt_regs regs)
 	return do_fork(clone_flags, newsp, &regs, 0,
 	               parent_tidptr, child_tidptr);
 }
-
-extern asmlinkage void sys_set_thread_area(u32 addr);
-asmlinkage void sys32_set_thread_area(u32 addr)
-{
-	sys_set_thread_area(AA(addr));
-}
-- 
cgit v1.2.1


From 06be375b9ae237b295648980e8a52048ab640f92 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 19 Sep 2006 17:18:53 +0100
Subject: [MIPS] TLS: set_thread_area returns  asmlinkage int not void.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/syscall.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
index 0721314db657..9951240cc3fd 100644
--- a/arch/mips/kernel/syscall.c
+++ b/arch/mips/kernel/syscall.c
@@ -263,7 +263,7 @@ asmlinkage int sys_olduname(struct oldold_utsname __user * name)
 	return error;
 }
 
-void sys_set_thread_area(unsigned long addr)
+asmlinkage int sys_set_thread_area(unsigned long addr)
 {
 	struct thread_info *ti = task_thread_info(current);
 
@@ -271,6 +271,8 @@ void sys_set_thread_area(unsigned long addr)
 
 	/* If some future MIPS implementation has this register in hardware,
 	 * we will need to update it here (and in context switches).  */
+
+	return 0;
 }
 
 asmlinkage int _sys_sysmips(int cmd, long arg1, int arg2, int arg3)
-- 
cgit v1.2.1


From a00f6310183e164ad2cfa5f4ec5d8c724ff0f799 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 1 Aug 2006 23:39:42 +0100
Subject: [MIPS] c-r4k: Convert init functions from inline to __init.

With more recent compilers inline doesn't necessarily means a function
will always be inlined.  So leave that decission to the compiler and
make the function as __init.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/c-r4k.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index 2d729f6f6348..0b2da53750bd 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -89,7 +89,7 @@ static inline void r4k_blast_dcache_page_dc32(unsigned long addr)
 	blast_dcache32_page(addr);
 }
 
-static inline void r4k_blast_dcache_page_setup(void)
+static void __init r4k_blast_dcache_page_setup(void)
 {
 	unsigned long  dc_lsize = cpu_dcache_line_size();
 
@@ -103,7 +103,7 @@ static inline void r4k_blast_dcache_page_setup(void)
 
 static void (* r4k_blast_dcache_page_indexed)(unsigned long addr);
 
-static inline void r4k_blast_dcache_page_indexed_setup(void)
+static void __init r4k_blast_dcache_page_indexed_setup(void)
 {
 	unsigned long dc_lsize = cpu_dcache_line_size();
 
@@ -117,7 +117,7 @@ static inline void r4k_blast_dcache_page_indexed_setup(void)
 
 static void (* r4k_blast_dcache)(void);
 
-static inline void r4k_blast_dcache_setup(void)
+static void __init r4k_blast_dcache_setup(void)
 {
 	unsigned long dc_lsize = cpu_dcache_line_size();
 
@@ -202,7 +202,7 @@ static inline void tx49_blast_icache32_page_indexed(unsigned long page)
 
 static void (* r4k_blast_icache_page)(unsigned long addr);
 
-static inline void r4k_blast_icache_page_setup(void)
+static void __init r4k_blast_icache_page_setup(void)
 {
 	unsigned long ic_lsize = cpu_icache_line_size();
 
@@ -219,7 +219,7 @@ static inline void r4k_blast_icache_page_setup(void)
 
 static void (* r4k_blast_icache_page_indexed)(unsigned long addr);
 
-static inline void r4k_blast_icache_page_indexed_setup(void)
+static void __init r4k_blast_icache_page_indexed_setup(void)
 {
 	unsigned long ic_lsize = cpu_icache_line_size();
 
@@ -243,7 +243,7 @@ static inline void r4k_blast_icache_page_indexed_setup(void)
 
 static void (* r4k_blast_icache)(void);
 
-static inline void r4k_blast_icache_setup(void)
+static void __init r4k_blast_icache_setup(void)
 {
 	unsigned long ic_lsize = cpu_icache_line_size();
 
@@ -264,7 +264,7 @@ static inline void r4k_blast_icache_setup(void)
 
 static void (* r4k_blast_scache_page)(unsigned long addr);
 
-static inline void r4k_blast_scache_page_setup(void)
+static void __init r4k_blast_scache_page_setup(void)
 {
 	unsigned long sc_lsize = cpu_scache_line_size();
 
@@ -282,7 +282,7 @@ static inline void r4k_blast_scache_page_setup(void)
 
 static void (* r4k_blast_scache_page_indexed)(unsigned long addr);
 
-static inline void r4k_blast_scache_page_indexed_setup(void)
+static void __init r4k_blast_scache_page_indexed_setup(void)
 {
 	unsigned long sc_lsize = cpu_scache_line_size();
 
@@ -300,7 +300,7 @@ static inline void r4k_blast_scache_page_indexed_setup(void)
 
 static void (* r4k_blast_scache)(void);
 
-static inline void r4k_blast_scache_setup(void)
+static void __init r4k_blast_scache_setup(void)
 {
 	unsigned long sc_lsize = cpu_scache_line_size();
 
@@ -1221,7 +1221,7 @@ void au1x00_fixup_config_od(void)
 	}
 }
 
-static inline void coherency_setup(void)
+static void __init coherency_setup(void)
 {
 	change_c0_config(CONF_CM_CMASK, CONF_CM_DEFAULT);
 
-- 
cgit v1.2.1


From 6b3e5f44b56745daad8cd913ccc7bcd9a9ece5ea Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Mon, 31 Jul 2006 23:01:37 +0900
Subject: [MIPS] Use common definitions from asm-generic/signal.h

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/signal.h | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h
index 87a1dff95199..5fc3a599d014 100644
--- a/include/asm-mips/signal.h
+++ b/include/asm-mips/signal.h
@@ -111,14 +111,7 @@ typedef unsigned long old_sigset_t;		/* at least 32 bits */
 #define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
 				   set only the low 32 bit of the sigset.  */
 
-/* Type of a signal handler.  */
-typedef void __signalfn_t(int);
-typedef __signalfn_t __user *__sighandler_t;
-
-/* Fake signal functions */
-#define SIG_DFL	((__sighandler_t)0)	/* default signal handling */
-#define SIG_IGN	((__sighandler_t)1)	/* ignore signal */
-#define SIG_ERR	((__sighandler_t)-1)	/* error return from signal */
+#include <asm-generic/signal.h>
 
 struct sigaction {
 	unsigned int	sa_flags;
-- 
cgit v1.2.1


From 633fd568c1db825b9eb563681af09260ecee87b6 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 4 Aug 2006 01:49:31 +0100
Subject: [MIPS] Move definition of IRIX compat constant into IRIX compat code.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/irixsig.c | 3 +++
 include/asm-mips/signal.h  | 2 --
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/irixsig.c b/arch/mips/kernel/irixsig.c
index 719364752e77..2132485caa74 100644
--- a/arch/mips/kernel/irixsig.c
+++ b/arch/mips/kernel/irixsig.c
@@ -323,6 +323,9 @@ struct sigact_irix5 {
 	int _unused0[2];
 };
 
+#define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
+				   set only the low 32 bit of the sigset.  */
+
 #ifdef DEBUG_SIG
 static inline void dump_sigact_irix5(struct sigact_irix5 *p)
 {
diff --git a/include/asm-mips/signal.h b/include/asm-mips/signal.h
index 5fc3a599d014..8b391a2f0814 100644
--- a/include/asm-mips/signal.h
+++ b/include/asm-mips/signal.h
@@ -108,8 +108,6 @@ typedef unsigned long old_sigset_t;		/* at least 32 bits */
 #define SIG_BLOCK	1	/* for blocking signals */
 #define SIG_UNBLOCK	2	/* for unblocking signals */
 #define SIG_SETMASK	3	/* for setting the signal mask */
-#define SIG_SETMASK32	256	/* Goodie from SGI for BSD compatibility:
-				   set only the low 32 bit of the sigset.  */
 
 #include <asm-generic/signal.h>
 
-- 
cgit v1.2.1


From bdb37c8d63f9455cd2ea505d62d0d6de221f3c76 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 16 Aug 2006 23:10:00 +0900
Subject: [MIPS] Remove F_SETSIG and F_GETSIG in favor of the asm-generic
 definitions.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/fcntl.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/asm-mips/fcntl.h b/include/asm-mips/fcntl.h
index 787220e6c1fc..00a50ec1c19f 100644
--- a/include/asm-mips/fcntl.h
+++ b/include/asm-mips/fcntl.h
@@ -25,8 +25,6 @@
 
 #define F_SETOWN	24	/*  for sockets. */
 #define F_GETOWN	23	/*  for sockets. */
-#define F_SETSIG	10	/*  for sockets. */
-#define F_GETSIG	11	/*  for sockets. */
 
 #ifndef __mips64
 #define F_GETLK64	33	/*  using 'struct flock64' */
-- 
cgit v1.2.1


From 3c70f12bfaec6e1a1c4bfb94aec0c17675bc9310 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sun, 20 Aug 2006 00:33:38 +0900
Subject: [MIPS] Qemu does not have D-cache aliases

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mach-qemu/cpu-feature-overrides.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-mips/mach-qemu/cpu-feature-overrides.h b/include/asm-mips/mach-qemu/cpu-feature-overrides.h
index f4e370e27168..529445dacedb 100644
--- a/include/asm-mips/mach-qemu/cpu-feature-overrides.h
+++ b/include/asm-mips/mach-qemu/cpu-feature-overrides.h
@@ -20,7 +20,7 @@
 
 #define cpu_has_llsc		1
 #define cpu_has_vtag_icache	0
-#define cpu_has_dc_aliases	(PAGE_SIZE < 0x4000)
+#define cpu_has_dc_aliases	0
 #define cpu_has_ic_fills_f_dc	0
 
 #define cpu_has_dsp		0
-- 
cgit v1.2.1


From 6b8aab09309c577318f8484fce401e5c3546bb3b Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 25 Aug 2006 12:34:33 +0100
Subject: [MIPS] Reformat missformated SMTC bits.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mmu_context.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/asm-mips/mmu_context.h b/include/asm-mips/mmu_context.h
index 18b69de87daa..fe065d6070ca 100644
--- a/include/asm-mips/mmu_context.h
+++ b/include/asm-mips/mmu_context.h
@@ -262,10 +262,10 @@ drop_mmu_context(struct mm_struct *mm, unsigned cpu)
 		/* See comments for similar code above */
 		prevvpe = dvpe();
 		oldasid = (read_c0_entryhi() & ASID_MASK);
-		if(smtc_live_asid[mytlb][oldasid]) {
-		  smtc_live_asid[mytlb][oldasid] &= ~(0x1 << cpu);
-		  if(smtc_live_asid[mytlb][oldasid] == 0)
-			smtc_flush_tlb_asid(oldasid);
+		if (smtc_live_asid[mytlb][oldasid]) {
+			smtc_live_asid[mytlb][oldasid] &= ~(0x1 << cpu);
+			if(smtc_live_asid[mytlb][oldasid] == 0)
+				smtc_flush_tlb_asid(oldasid);
 		}
 		/* See comments for similar code above */
 		write_c0_entryhi((read_c0_entryhi() & ~HW_ASID_MASK)
-- 
cgit v1.2.1


From 48712a96e31d11dce6dcb2a886894bd4d8b890f1 Mon Sep 17 00:00:00 2001
From: "thomas@koeller.dyndns.org" <thomas@koeller.dyndns.org>
Date: Sun, 27 Aug 2006 13:53:16 +0200
Subject: [MIPS] Suppress compiler warnings

The excite platform exports hardware resources for device drivers to use.
Any driver wanting to use these resources will look up them by their names.
Since these resources are declared to have static linkage, but are not
used in the source file defining them, the compiler used to emit an
'unused' warning, which this patch suppresses.

Signed-off-by: Thomas Koeller <thomas.koeller@baslerweb.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/basler/excite/excite_device.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/arch/mips/basler/excite/excite_device.c b/arch/mips/basler/excite/excite_device.c
index bbb4ea43da88..cc1ce77eab4a 100644
--- a/arch/mips/basler/excite/excite_device.c
+++ b/arch/mips/basler/excite/excite_device.c
@@ -68,7 +68,7 @@ enum {
 
 
 static struct resource
-	excite_ctr_resource = {
+	excite_ctr_resource __attribute__((unused)) = {
 		.name		= "GPI counters",
 		.start		= 0,
 		.end		= 5,
@@ -77,7 +77,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_gpislice_resource = {
+	excite_gpislice_resource __attribute__((unused)) = {
 		.name		= "GPI slices",
 		.start		= 0,
 		.end		= 1,
@@ -86,7 +86,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_mdio_channel_resource = {
+	excite_mdio_channel_resource __attribute__((unused)) = {
 		.name		= "MDIO channels",
 		.start		= 0,
 		.end		= 1,
@@ -95,7 +95,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_fifomem_resource = {
+	excite_fifomem_resource __attribute__((unused)) = {
 		.name		= "FIFO memory",
 		.start		= 0,
 		.end		= 767,
@@ -104,7 +104,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_scram_resource = {
+	excite_scram_resource __attribute__((unused)) = {
 		.name		= "Scratch RAM",
 		.start		= EXCITE_PHYS_SCRAM,
 		.end		= EXCITE_PHYS_SCRAM + EXCITE_SIZE_SCRAM - 1,
@@ -113,7 +113,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_fpga_resource = {
+	excite_fpga_resource __attribute__((unused)) = {
 		.name		= "System FPGA",
 		.start		= EXCITE_PHYS_FPGA,
 		.end		= EXCITE_PHYS_FPGA + EXCITE_SIZE_FPGA - 1,
@@ -122,7 +122,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_nand_resource = {
+	excite_nand_resource __attribute__((unused)) = {
 		.name		= "NAND flash control",
 		.start		= EXCITE_PHYS_NAND,
 		.end		= EXCITE_PHYS_NAND + EXCITE_SIZE_NAND - 1,
@@ -131,7 +131,7 @@ static struct resource
 		.sibling	= NULL,
 		.child		= NULL
 	},
-	excite_titan_resource = {
+	excite_titan_resource __attribute__((unused)) = {
 		.name		= "TITAN registers",
 		.start		= EXCITE_PHYS_TITAN,
 		.end		= EXCITE_PHYS_TITAN + EXCITE_SIZE_TITAN - 1,
-- 
cgit v1.2.1


From 0c68a9b6a7da0cc9095c117bea573f9058b00fff Mon Sep 17 00:00:00 2001
From: "thomas@koeller.dyndns.org" <thomas@koeller.dyndns.org>
Date: Sun, 27 Aug 2006 13:54:31 +0200
Subject: [MIPS] Move excite_fpga.h to include/asm-mips/mach-excite

excite_fpga.h, like all platform headers, really belongs in the
platform header directory.

Signed-off-by: Thomas Koeller <thomas.koeller@baslerweb.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/basler/excite/excite_fpga.h      | 80 ------------------------------
 include/asm-mips/mach-excite/excite_fpga.h | 80 ++++++++++++++++++++++++++++++
 2 files changed, 80 insertions(+), 80 deletions(-)
 delete mode 100644 arch/mips/basler/excite/excite_fpga.h
 create mode 100644 include/asm-mips/mach-excite/excite_fpga.h

diff --git a/arch/mips/basler/excite/excite_fpga.h b/arch/mips/basler/excite/excite_fpga.h
deleted file mode 100644
index 38fcda703a0b..000000000000
--- a/arch/mips/basler/excite/excite_fpga.h
+++ /dev/null
@@ -1,80 +0,0 @@
-#ifndef EXCITE_FPGA_H_INCLUDED
-#define EXCITE_FPGA_H_INCLUDED
-
-
-/**
- * Adress alignment of the individual FPGA bytes.
- * The address arrangement of the individual bytes of the FPGA is two
- * byte aligned at the embedded MK2 platform.
- */
-#ifdef EXCITE_CCI_FPGA_MK2
-typedef unsigned char excite_cci_fpga_align_t __attribute__ ((aligned(2)));
-#else
-typedef unsigned char excite_cci_fpga_align_t;
-#endif
-
-
-/**
- * Size of Dual Ported RAM.
- */
-#define EXCITE_DPR_SIZE 263
-
-
-/**
- * Size of Reserved Status Fields in Dual Ported RAM.
- */
-#define EXCITE_DPR_STATUS_SIZE 7
-
-
-
-/**
- * FPGA.
- * Hardware register layout of the FPGA interface. The FPGA must accessed
- * byte wise solely.
- * @see EXCITE_CCI_DPR_MK2
- */
-typedef struct excite_fpga {
-
-	/**
-	 * Dual Ported RAM.
-	 */
-	excite_cci_fpga_align_t dpr[EXCITE_DPR_SIZE];
-
-	/**
-	 * Status.
-	 */
-	excite_cci_fpga_align_t status[EXCITE_DPR_STATUS_SIZE];
-
-#ifdef EXCITE_CCI_FPGA_MK2
-	/**
-	 * RM9000 Interrupt.
-	 * Write access initiates interrupt at the RM9000 (MIPS) processor of the eXcite.
-	 */
-	excite_cci_fpga_align_t rm9k_int;
-#else
-	/**
-	 * MK2 Interrupt.
-	 * Write access initiates interrupt at the ARM processor of the MK2.
-	 */
-	excite_cci_fpga_align_t mk2_int;
-
-	excite_cci_fpga_align_t gap[0x1000-0x10f];
-
-	/**
-	 * IRQ Source/Acknowledge.
-	 */
-	excite_cci_fpga_align_t rm9k_irq_src;
-
-	/**
-	 * IRQ Mask.
-	 * Set bits enable the related interrupt.
-	 */
-	excite_cci_fpga_align_t rm9k_irq_mask;
-#endif
-
-
-} excite_fpga;
-
-
-
-#endif	/* ndef EXCITE_FPGA_H_INCLUDED */
diff --git a/include/asm-mips/mach-excite/excite_fpga.h b/include/asm-mips/mach-excite/excite_fpga.h
new file mode 100644
index 000000000000..38fcda703a0b
--- /dev/null
+++ b/include/asm-mips/mach-excite/excite_fpga.h
@@ -0,0 +1,80 @@
+#ifndef EXCITE_FPGA_H_INCLUDED
+#define EXCITE_FPGA_H_INCLUDED
+
+
+/**
+ * Adress alignment of the individual FPGA bytes.
+ * The address arrangement of the individual bytes of the FPGA is two
+ * byte aligned at the embedded MK2 platform.
+ */
+#ifdef EXCITE_CCI_FPGA_MK2
+typedef unsigned char excite_cci_fpga_align_t __attribute__ ((aligned(2)));
+#else
+typedef unsigned char excite_cci_fpga_align_t;
+#endif
+
+
+/**
+ * Size of Dual Ported RAM.
+ */
+#define EXCITE_DPR_SIZE 263
+
+
+/**
+ * Size of Reserved Status Fields in Dual Ported RAM.
+ */
+#define EXCITE_DPR_STATUS_SIZE 7
+
+
+
+/**
+ * FPGA.
+ * Hardware register layout of the FPGA interface. The FPGA must accessed
+ * byte wise solely.
+ * @see EXCITE_CCI_DPR_MK2
+ */
+typedef struct excite_fpga {
+
+	/**
+	 * Dual Ported RAM.
+	 */
+	excite_cci_fpga_align_t dpr[EXCITE_DPR_SIZE];
+
+	/**
+	 * Status.
+	 */
+	excite_cci_fpga_align_t status[EXCITE_DPR_STATUS_SIZE];
+
+#ifdef EXCITE_CCI_FPGA_MK2
+	/**
+	 * RM9000 Interrupt.
+	 * Write access initiates interrupt at the RM9000 (MIPS) processor of the eXcite.
+	 */
+	excite_cci_fpga_align_t rm9k_int;
+#else
+	/**
+	 * MK2 Interrupt.
+	 * Write access initiates interrupt at the ARM processor of the MK2.
+	 */
+	excite_cci_fpga_align_t mk2_int;
+
+	excite_cci_fpga_align_t gap[0x1000-0x10f];
+
+	/**
+	 * IRQ Source/Acknowledge.
+	 */
+	excite_cci_fpga_align_t rm9k_irq_src;
+
+	/**
+	 * IRQ Mask.
+	 * Set bits enable the related interrupt.
+	 */
+	excite_cci_fpga_align_t rm9k_irq_mask;
+#endif
+
+
+} excite_fpga;
+
+
+
+#endif	/* ndef EXCITE_FPGA_H_INCLUDED */
-- 
cgit v1.2.1


From 663c3d9009c1a41194b8390abdcea61e3ade1c8f Mon Sep 17 00:00:00 2001
From: "thomas@koeller.dyndns.org" <thomas@koeller.dyndns.org>
Date: Sun, 27 Aug 2006 13:51:48 +0200
Subject: [MIPS] Add configuration variables for RM9xxx processor

This patch introduces a number of configuration variables. These allow to
specify presence/absence of integrated peripherals found on the MIPS
RM9xxx processor family, based on the particular processor model used.

Signed-off-by: Thomas Koeller <thomas.koeller@baslerweb.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Kconfig | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 991b329224e7..235208d658ff 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -126,6 +126,7 @@ config BASLER_EXCITE
 	select IRQ_CPU
 	select IRQ_CPU_RM7K
 	select IRQ_CPU_RM9K
+	select MIPS_RM9122
 	select SYS_HAS_CPU_RM9000
 	select SYS_SUPPORTS_32BIT_KERNEL
 	select SYS_SUPPORTS_64BIT_KERNEL
@@ -973,6 +974,12 @@ config MIPS_TX3927
 	bool
 	select HAS_TXX9_SERIAL
 
+config MIPS_RM9122
+	bool
+	select SERIAL_RM9000
+	select GPI_RM9000
+	select WDT_RM9000
+
 config PCI_MARVELL
 	bool
 
@@ -1023,6 +1030,15 @@ config EMMA2RH
 	depends on MARKEINS
 	default y
 
+config SERIAL_RM9000
+	bool
+
+config GPI_RM9000
+	bool
+
+config WDT_RM9000
+	bool
+
 #
 # Unfortunately not all GT64120 systems run the chip at the same clock.
 # As the user for the clock rate and try to minimize the available options.
-- 
cgit v1.2.1


From 9dbd7b9142e95867ee8a56da5d45c72884c107d3 Mon Sep 17 00:00:00 2001
From: Peter Watkins <treestem@gmail.com>
Date: Wed, 23 Aug 2006 11:15:49 -0400
Subject: [MIPS] Fix USER_PTRS_PER_PGD for 64K page size.

The code in pgtable-64.h assumes TASK_SIZE is always bigger than a first
level PGDIR_SIZE. This is not the case for 64K pages, where task size is
40 bits (1TB) and a pgd entry can map 42 bits. This leads to
USER_PTRS_PER_PGD being zero for 64K pages.

Signed-off-by: Peter Watkins <treestem@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/pgtable-64.h | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/include/asm-mips/pgtable-64.h b/include/asm-mips/pgtable-64.h
index c59a1e21f5b0..d05fb6f38aa7 100644
--- a/include/asm-mips/pgtable-64.h
+++ b/include/asm-mips/pgtable-64.h
@@ -93,8 +93,12 @@
 #define PTRS_PER_PMD	((PAGE_SIZE << PMD_ORDER) / sizeof(pmd_t))
 #define PTRS_PER_PTE	((PAGE_SIZE << PTE_ORDER) / sizeof(pte_t))
 
+#if PGDIR_SIZE >= TASK_SIZE
+#define USER_PTRS_PER_PGD       (1)
+#else
 #define USER_PTRS_PER_PGD	(TASK_SIZE / PGDIR_SIZE)
-#define FIRST_USER_ADDRESS	0
+#endif
+#define FIRST_USER_ADDRESS	0UL
 
 #define VMALLOC_START		MAP_BASE
 #define VMALLOC_END	\
-- 
cgit v1.2.1


From da79e827d4164ba1b961b62774800a8a52a3b6e4 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 7 Sep 2006 02:29:57 +0200
Subject: [MIPS] IP27: Delete useless declaration of allocate_irqno().

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/pci/pci-ip27.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/arch/mips/pci/pci-ip27.c b/arch/mips/pci/pci-ip27.c
index 80eb9af9ecdf..405ce0152739 100644
--- a/arch/mips/pci/pci-ip27.c
+++ b/arch/mips/pci/pci-ip27.c
@@ -16,8 +16,6 @@
 #include <asm/sn/intr.h>
 #include <asm/sn/sn0/hub.h>
 
-extern unsigned int allocate_irqno(void);
-
 /*
  * Max #PCI busses we can handle; ie, max #PCI bridges.
  */
-- 
cgit v1.2.1


From d7d86aa88a1f3922b85e39edd8a6d6c01e939842 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 8 Sep 2006 04:13:49 +0200
Subject: [MIPS] Cleanup hazard handling.

Mostly based on patch by Chris Dearman and cleanups from Yoichi.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/hazards.h | 362 ++++++++++++++++-----------------------------
 1 file changed, 126 insertions(+), 236 deletions(-)

diff --git a/include/asm-mips/hazards.h b/include/asm-mips/hazards.h
index 25f5e8a4177d..bf0b80404946 100644
--- a/include/asm-mips/hazards.h
+++ b/include/asm-mips/hazards.h
@@ -11,103 +11,96 @@
 #define _ASM_HAZARDS_H
 
 
-#ifdef __ASSEMBLY__
-
-	.macro	_ssnop
-	sll	$0, $0, 1
-	.endm
-
-	.macro	_ehb
-	sll	$0, $0, 3
-	.endm
-
-/*
- * RM9000 hazards.  When the JTLB is updated by tlbwi or tlbwr, a subsequent
- * use of the JTLB for instructions should not occur for 4 cpu cycles and use
- * for data translations should not occur for 3 cpu cycles.
- */
-#ifdef CONFIG_CPU_RM9000
-
-	.macro	mtc0_tlbw_hazard
-	.set	push
-	.set	mips32
-	_ssnop; _ssnop; _ssnop; _ssnop
-	.set	pop
-	.endm
-
-	.macro	tlbw_eret_hazard
-	.set	push
-	.set	mips32
-	_ssnop; _ssnop; _ssnop; _ssnop
-	.set	pop
-	.endm
-
+#ifdef __ASSEMBLER__
+#define ASMMACRO(name, code...) .macro name; code; .endm
 #else
 
-/*
- * The taken branch will result in a two cycle penalty for the two killed
- * instructions on R4000 / R4400.  Other processors only have a single cycle
- * hazard so this is nice trick to have an optimal code for a range of
- * processors.
- */
-	.macro	mtc0_tlbw_hazard
-	b	. + 8
-	.endm
+#define ASMMACRO(name, code...)						\
+__asm__(".macro " #name "; " #code "; .endm");				\
+									\
+static inline void name(void)						\
+{									\
+	__asm__ __volatile__ (#name);					\
+}
 
-	.macro	tlbw_eret_hazard
-	.endm
 #endif
 
+ASMMACRO(_ssnop,
+	 sll	$0, $0, 1
+	)
+
+ASMMACRO(_ehb,
+	 sll	$0, $0, 3
+	)
+
 /*
- * mtc0->mfc0 hazard
- * The 24K has a 2 cycle mtc0/mfc0 execution hazard.
- * It is a MIPS32R2 processor so ehb will clear the hazard.
+ * TLB hazards
  */
+#if defined(CONFIG_CPU_MIPSR2)
 
-#ifdef CONFIG_CPU_MIPSR2
 /*
- * Use a macro for ehb unless explicit support for MIPSR2 is enabled
+ * MIPSR2 defines ehb for hazard avoidance
  */
 
-#define irq_enable_hazard						\
+ASMMACRO(mtc0_tlbw_hazard,
+	 _ehb
+	)
+ASMMACRO(tlbw_use_hazard,
+	 _ehb
+	)
+ASMMACRO(tlb_probe_hazard,
+	 _ehb
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
 	_ehb
-
-#define irq_disable_hazard						\
-	_ehb
-
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000)
-
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	 _ehb
+	)
 /*
- * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
+ * gcc has a tradition of misscompiling the previous construct using the
+ * address of a label as argument to inline assembler.  Gas otoh has the
+ * annoying difference between la and dla which are only usable for 32-bit
+ * rsp. 64-bit code, so can't be used without conditional compilation.
+ * The alterantive is switching the assembler to 64-bit code which happens
+ * to work right even for 32-bit code ...
  */
+#define instruction_hazard()						\
+do {									\
+	unsigned long tmp;						\
+									\
+	__asm__ __volatile__(						\
+	"	.set	mips64r2				\n"	\
+	"	dla	%0, 1f					\n"	\
+	"	jr.hb	%0					\n"	\
+	"	.set	mips0					\n"	\
+	"1:							\n"	\
+	: "=r" (tmp));							\
+} while (0)
 
-#define irq_enable_hazard
-
-#define irq_disable_hazard
-
-#else
+#elif defined(CONFIG_CPU_R10000)
 
 /*
- * Classic MIPS needs 1 - 3 nops or ssnops
+ * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
  */
-#define irq_enable_hazard
-#define irq_disable_hazard						\
-	_ssnop; _ssnop; _ssnop
 
-#endif
-
-#else /* __ASSEMBLY__ */
-
-__asm__(
-	"	.macro	_ssnop					\n"
-	"	sll	$0, $0, 1				\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	_ehb					\n"
-	"	sll	$0, $0, 3				\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	)
+ASMMACRO(tlbw_use_hazard,
+	)
+ASMMACRO(tlb_probe_hazard,
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
-#ifdef CONFIG_CPU_RM9000
+#elif defined(CONFIG_CPU_RM9000)
 
 /*
  * RM9000 hazards.  When the JTLB is updated by tlbwi or tlbwr, a subsequent
@@ -115,176 +108,73 @@ __asm__(
  * for data translations should not occur for 3 cpu cycles.
  */
 
-#define mtc0_tlbw_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	mips32					\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	.set	mips0					\n")
-
-#define tlbw_use_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	mips32					\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	_ssnop						\n"	\
-	"	.set	mips0					\n")
-
-#else
-
-/*
- * Overkill warning ...
- */
-#define mtc0_tlbw_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	noreorder				\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	.set	reorder					\n")
-
-#define tlbw_use_hazard()						\
-	__asm__ __volatile__(						\
-	"	.set	noreorder				\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	nop						\n"	\
-	"	.set	reorder					\n")
-
-#endif
-
-/*
- * Interrupt enable/disable hazards
- * Some processors have hazards when modifying
- * the status register to change the interrupt state
- */
-
-#ifdef CONFIG_CPU_MIPSR2
-
-__asm__("	.macro	irq_enable_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	irq_disable_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(tlbw_use_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(tlb_probe_hazard,
+	 _ssnop; _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000)
+#elif defined(CONFIG_CPU_SB1)
 
 /*
- * R10000 rocks - all hazards handled in hardware, so this becomes a nobrainer.
+ * Mostly like R4000 for historic reasons
  */
-
-__asm__(
-	"	.macro	irq_enable_hazard			\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro	irq_disable_hazard			\n"
-	"	.endm						\n");
+ASMMACRO(mtc0_tlbw_hazard,
+	)
+ASMMACRO(tlbw_use_hazard,
+	)
+ASMMACRO(tlb_probe_hazard,
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	 _ssnop; _ssnop; _ssnop
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	)
+#define instruction_hazard() do { } while (0)
 
 #else
 
 /*
- * Default for classic MIPS processors.  Assume worst case hazards but don't
- * care about the irq_enable_hazard - sooner or later the hardware will
- * enable it and we don't care when exactly.
- */
-
-__asm__(
-	"	#						\n"
-	"	# There is a hazard but we do not care		\n"
-	"	#						\n"
-	"	.macro\tirq_enable_hazard			\n"
-	"	.endm						\n"
-	"							\n"
-	"	.macro\tirq_disable_hazard			\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	.endm						\n");
-
-#endif
-
-#define irq_enable_hazard()						\
-	__asm__ __volatile__("irq_enable_hazard")
-#define irq_disable_hazard()						\
-	__asm__ __volatile__("irq_disable_hazard")
-
-
-/*
- * Back-to-back hazards -
+ * Finally the catchall case for all other processors including R4000, R4400,
+ * R4600, R4700, R5000, RM7000, NEC VR41xx etc.
  *
- * What is needed to separate a move to cp0 from a subsequent read from the
- * same cp0 register?
- */
-#ifdef CONFIG_CPU_MIPSR2
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	_ehb						\n"
-	"	.endm						\n");
-
-#elif defined(CONFIG_CPU_R10000) || defined(CONFIG_CPU_RM9000) || \
-      defined(CONFIG_CPU_SB1)
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	.endm						\n");
-
-#else
-
-__asm__("	.macro	back_to_back_c0_hazard			\n"
-	"	.set	noreorder				\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	_ssnop						\n"
-	"	.set	reorder					\n"
-	"	.endm");
-
-#endif
-
-#define back_to_back_c0_hazard()					\
-	__asm__ __volatile__("back_to_back_c0_hazard")
-
-
-/*
- * Instruction execution hazard
- */
-#ifdef CONFIG_CPU_MIPSR2
-/*
- * gcc has a tradition of misscompiling the previous construct using the
- * address of a label as argument to inline assembler.  Gas otoh has the
- * annoying difference between la and dla which are only usable for 32-bit
- * rsp. 64-bit code, so can't be used without conditional compilation.
- * The alterantive is switching the assembler to 64-bit code which happens
- * to work right even for 32-bit code ...
+ * The taken branch will result in a two cycle penalty for the two killed
+ * instructions on R4000 / R4400.  Other processors only have a single cycle
+ * hazard so this is nice trick to have an optimal code for a range of
+ * processors.
  */
-#define instruction_hazard()						\
-do {									\
-	unsigned long tmp;						\
-									\
-	__asm__ __volatile__(						\
-	"	.set	mips64r2				\n"	\
-	"	dla	%0, 1f					\n"	\
-	"	jr.hb	%0					\n"	\
-	"	.set	mips0					\n"	\
-	"1:							\n"	\
-	: "=r" (tmp));							\
-} while (0)
-
-#else
+ASMMACRO(mtc0_tlbw_hazard,
+	nop
+	)
+ASMMACRO(tlbw_use_hazard,
+	nop; nop; nop
+	)
+ASMMACRO(tlb_probe_hazard,
+	 nop; nop; nop
+	)
+ASMMACRO(irq_enable_hazard,
+	)
+ASMMACRO(irq_disable_hazard,
+	nop; nop; nop
+	)
+ASMMACRO(back_to_back_c0_hazard,
+	 _ssnop; _ssnop; _ssnop;
+	)
 #define instruction_hazard() do { } while (0)
-#endif
-
-extern void mips_ihb(void);
 
-#endif /* __ASSEMBLY__ */
+#endif
 
 #endif /* _ASM_HAZARDS_H */
-- 
cgit v1.2.1


From 432bef2a31668a0562e5738eaa59a43854f26567 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 8 Sep 2006 04:16:21 +0200
Subject: [MIPS] Replace BARRIER with more appropriate hazard barrier.

This is the unchanged part 2 of Chris' hazard cleanup.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/tlb-r4k.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)

diff --git a/arch/mips/mm/tlb-r4k.c b/arch/mips/mm/tlb-r4k.c
index 2cde1b772443..2e0e21ef433e 100644
--- a/arch/mips/mm/tlb-r4k.c
+++ b/arch/mips/mm/tlb-r4k.c
@@ -26,11 +26,6 @@ extern void build_tlb_refill_handler(void);
  */
 #define UNIQUE_ENTRYHI(idx) (CKSEG0 + ((idx) << (PAGE_SHIFT + 1)))
 
-/* CP0 hazard avoidance. */
-#define BARRIER __asm__ __volatile__(".set noreorder\n\t" \
-				     "nop; nop; nop; nop; nop; nop;\n\t" \
-				     ".set reorder\n\t")
-
 /* Atomicity and interruptability */
 #ifdef CONFIG_MIPS_MT_SMTC
 
@@ -126,7 +121,7 @@ void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 				start += (PAGE_SIZE << 1);
 				mtc0_tlbw_hazard();
 				tlb_probe();
-				BARRIER;
+				tlb_probe_hazard();
 				idx = read_c0_index();
 				write_c0_entrylo0(0);
 				write_c0_entrylo1(0);
@@ -168,7 +163,7 @@ void local_flush_tlb_kernel_range(unsigned long start, unsigned long end)
 			start += (PAGE_SIZE << 1);
 			mtc0_tlbw_hazard();
 			tlb_probe();
-			BARRIER;
+			tlb_probe_hazard();
 			idx = read_c0_index();
 			write_c0_entrylo0(0);
 			write_c0_entrylo1(0);
@@ -202,7 +197,7 @@ void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long page)
 		write_c0_entryhi(page | newpid);
 		mtc0_tlbw_hazard();
 		tlb_probe();
-		BARRIER;
+		tlb_probe_hazard();
 		idx = read_c0_index();
 		write_c0_entrylo0(0);
 		write_c0_entrylo1(0);
@@ -235,7 +230,7 @@ void local_flush_tlb_one(unsigned long page)
 	write_c0_entryhi(page);
 	mtc0_tlbw_hazard();
 	tlb_probe();
-	BARRIER;
+	tlb_probe_hazard();
 	idx = read_c0_index();
 	write_c0_entrylo0(0);
 	write_c0_entrylo1(0);
@@ -279,7 +274,7 @@ void __update_tlb(struct vm_area_struct * vma, unsigned long address, pte_t pte)
 	pgdp = pgd_offset(vma->vm_mm, address);
 	mtc0_tlbw_hazard();
 	tlb_probe();
-	BARRIER;
+	tlb_probe_hazard();
 	pudp = pud_offset(pgdp, address);
 	pmdp = pmd_offset(pudp, address);
 	idx = read_c0_index();
@@ -320,7 +315,7 @@ static void r4k_update_mmu_cache_hwbug(struct vm_area_struct * vma,
 	pgdp = pgd_offset(vma->vm_mm, address);
 	mtc0_tlbw_hazard();
 	tlb_probe();
-	BARRIER;
+	tlb_probe_hazard();
 	pmdp = pmd_offset(pgdp, address);
 	idx = read_c0_index();
 	ptep = pte_offset_map(pmdp, address);
@@ -351,7 +346,7 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 	wired = read_c0_wired();
 	write_c0_wired(wired + 1);
 	write_c0_index(wired);
-	BARRIER;
+	tlbw_use_hazard();	/* What is the hazard here? */
 	write_c0_pagemask(pagemask);
 	write_c0_entryhi(entryhi);
 	write_c0_entrylo0(entrylo0);
@@ -361,7 +356,7 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 	tlbw_use_hazard();
 
 	write_c0_entryhi(old_ctx);
-	BARRIER;
+	tlbw_use_hazard();	/* What is the hazard here? */
 	write_c0_pagemask(old_pagemask);
 	local_flush_tlb_all();
 	EXIT_CRITICAL(flags);
-- 
cgit v1.2.1


From d48f1de2d8170814fb64effa320848410c466f95 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 20 Sep 2006 20:56:02 +0100
Subject: [MIPS] Remove EV96100 as previously announced.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 Documentation/feature-removal-schedule.txt   |   8 -
 Documentation/kernel-parameters.txt          |   2 -
 arch/mips/Kconfig                            |  24 -
 arch/mips/Makefile                           |   7 -
 arch/mips/configs/atlas_defconfig            |   1 -
 arch/mips/configs/bigsur_defconfig           |   1 -
 arch/mips/configs/capcella_defconfig         |   1 -
 arch/mips/configs/cobalt_defconfig           |   1 -
 arch/mips/configs/db1000_defconfig           |   1 -
 arch/mips/configs/db1100_defconfig           |   1 -
 arch/mips/configs/db1200_defconfig           |   1 -
 arch/mips/configs/db1500_defconfig           |   1 -
 arch/mips/configs/db1550_defconfig           |   1 -
 arch/mips/configs/ddb5477_defconfig          |   1 -
 arch/mips/configs/decstation_defconfig       |   1 -
 arch/mips/configs/e55_defconfig              |   1 -
 arch/mips/configs/emma2rh_defconfig          |   1 -
 arch/mips/configs/ev64120_defconfig          |   1 -
 arch/mips/configs/ev96100_defconfig          | 850 ---------------------------
 arch/mips/configs/excite_defconfig           |   1 -
 arch/mips/configs/ip22_defconfig             |   1 -
 arch/mips/configs/ip27_defconfig             |   1 -
 arch/mips/configs/ip32_defconfig             |   1 -
 arch/mips/configs/it8172_defconfig           |   1 -
 arch/mips/configs/ivr_defconfig              |   1 -
 arch/mips/configs/jaguar-atx_defconfig       |   1 -
 arch/mips/configs/jmr3927_defconfig          |   1 -
 arch/mips/configs/lasat200_defconfig         |   1 -
 arch/mips/configs/malta_defconfig            |   1 -
 arch/mips/configs/mipssim_defconfig          |   1 -
 arch/mips/configs/mpc30x_defconfig           |   1 -
 arch/mips/configs/ocelot_3_defconfig         |   1 -
 arch/mips/configs/ocelot_c_defconfig         |   1 -
 arch/mips/configs/ocelot_defconfig           |   1 -
 arch/mips/configs/ocelot_g_defconfig         |   1 -
 arch/mips/configs/pb1100_defconfig           |   1 -
 arch/mips/configs/pb1500_defconfig           |   1 -
 arch/mips/configs/pb1550_defconfig           |   1 -
 arch/mips/configs/pnx8550-jbs_defconfig      |   1 -
 arch/mips/configs/pnx8550-v2pci_defconfig    |   1 -
 arch/mips/configs/qemu_defconfig             |   1 -
 arch/mips/configs/rbhma4500_defconfig        |   1 -
 arch/mips/configs/rm200_defconfig            |   1 -
 arch/mips/configs/sb1250-swarm_defconfig     |   1 -
 arch/mips/configs/sead_defconfig             |   1 -
 arch/mips/configs/tb0226_defconfig           |   1 -
 arch/mips/configs/tb0229_defconfig           |   1 -
 arch/mips/configs/tb0287_defconfig           |   1 -
 arch/mips/configs/workpad_defconfig          |   1 -
 arch/mips/configs/wrppmc_defconfig           |   1 -
 arch/mips/configs/yosemite_defconfig         |   1 -
 arch/mips/defconfig                          |   1 -
 arch/mips/galileo-boards/ev96100/Makefile    |   9 -
 arch/mips/galileo-boards/ev96100/init.c      | 173 ------
 arch/mips/galileo-boards/ev96100/irq.c       |  77 ---
 arch/mips/galileo-boards/ev96100/puts.c      | 138 -----
 arch/mips/galileo-boards/ev96100/reset.c     |  70 ---
 arch/mips/galileo-boards/ev96100/setup.c     | 159 -----
 arch/mips/galileo-boards/ev96100/time.c      |  88 ---
 arch/mips/pci/Makefile                       |   2 -
 arch/mips/pci/fixup-ev96100.c                |  48 --
 arch/mips/pci/ops-gt96100.c                  | 169 ------
 arch/mips/pci/pci-ev96100.c                  |  63 --
 include/asm-mips/bootinfo.h                  |   3 +-
 include/asm-mips/galileo-boards/gt96100.h    | 427 --------------
 include/asm-mips/mach-ev96100/mach-gt64120.h |  46 --
 include/asm-mips/serial.h                    |   4 +-
 include/linux/pci_ids.h                      |   3 -
 68 files changed, 3 insertions(+), 2414 deletions(-)
 delete mode 100644 arch/mips/configs/ev96100_defconfig
 delete mode 100644 arch/mips/galileo-boards/ev96100/Makefile
 delete mode 100644 arch/mips/galileo-boards/ev96100/init.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/irq.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/puts.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/reset.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/setup.c
 delete mode 100644 arch/mips/galileo-boards/ev96100/time.c
 delete mode 100644 arch/mips/pci/fixup-ev96100.c
 delete mode 100644 arch/mips/pci/ops-gt96100.c
 delete mode 100644 arch/mips/pci/pci-ev96100.c
 delete mode 100644 include/asm-mips/galileo-boards/gt96100.h
 delete mode 100644 include/asm-mips/mach-ev96100/mach-gt64120.h

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 611acc32fdf5..bf56b20652b0 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -217,14 +217,6 @@ Who:	Nick Piggin <npiggin@suse.de>
 
 ---------------------------
 
-What:	Support for the MIPS EV96100 evaluation board
-When:	September 2006
-Why:	Does no longer build since at least November 15, 2003, apparently
-	no userbase left.
-Who:	Ralf Baechle <ralf@linux-mips.org>
-
----------------------------
-
 What:	Support for the Momentum / PMC-Sierra Jaguar ATX evaluation board
 When:	September 2006
 Why:	Does no longer build since quite some time, and was never popular,
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index c918cc3f65fb..255ec535bba8 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -573,8 +573,6 @@ running once the system is up.
 	gscd=		[HW,CD]
 			Format: <io>
 
-	gt96100eth=	[NET] MIPS GT96100 Advanced Communication Controller
-
 	gus=		[HW,OSS]
 			Format: <io>,<irq>,<dma>,<dma16>
 
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 235208d658ff..30750c54bdf5 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -203,26 +203,6 @@ config MIPS_EV64120
 	  <http://www.marvell.com/>.  Say Y here if you wish to build a
 	  kernel for this platform.
 
-config MIPS_EV96100
-	bool "Galileo EV96100 Evaluation board (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
-	select DMA_NONCOHERENT
-	select HW_HAS_PCI
-	select IRQ_CPU
-	select MIPS_GT96100
-	select RM7000_CPU_SCACHE
-	select SWAP_IO_SPACE
-	select SYS_HAS_CPU_R5000
-	select SYS_HAS_CPU_RM7000
-	select SYS_SUPPORTS_32BIT_KERNEL
-	select SYS_SUPPORTS_64BIT_KERNEL if EXPERIMENTAL
-	select SYS_SUPPORTS_BIG_ENDIAN
-	help
-	  This is an evaluation board based on the Galileo GT-96100 LAN/WAN
-	  communications controllers containing a MIPS R5000 compatible core
-	  running at 83MHz. Their website is <http://www.marvell.com/>. Say Y
-	  here if you wish to build a kernel for this platform.
-
 config MIPS_IVR
 	bool "Globespan IVR board"
 	select DMA_NONCOHERENT
@@ -1069,10 +1049,6 @@ config AU1X00_USB_DEVICE
 	depends on MIPS_PB1500 || MIPS_PB1100 || MIPS_PB1000
 	default n
 
-config MIPS_GT96100
-	bool
-	select MIPS_GT64120
-
 config IT8172_CIR
 	bool
 	depends on MIPS_ITE8172 || MIPS_IVR
diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index f4227d24227d..e521826b4234 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -279,13 +279,6 @@ core-$(CONFIG_MIPS_EV64120)	+= arch/mips/gt64120/common/
 cflags-$(CONFIG_MIPS_EV64120)	+= -Iinclude/asm-mips/mach-ev64120
 load-$(CONFIG_MIPS_EV64120)	+= 0xffffffff80100000
 
-#
-# Galileo EV96100 Board
-#
-core-$(CONFIG_MIPS_EV96100)	+= arch/mips/galileo-boards/ev96100/
-cflags-$(CONFIG_MIPS_EV96100)	+= -Iinclude/asm-mips/mach-ev96100
-load-$(CONFIG_MIPS_EV96100)	+= 0xffffffff80100000
-
 #
 # Wind River PPMC Board (4KC + GT64120)
 #
diff --git a/arch/mips/configs/atlas_defconfig b/arch/mips/configs/atlas_defconfig
index 54274065e9a5..2774ba9606b7 100644
--- a/arch/mips/configs/atlas_defconfig
+++ b/arch/mips/configs/atlas_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/bigsur_defconfig b/arch/mips/configs/bigsur_defconfig
index 887fd959482a..e12a475dcbf4 100644
--- a/arch/mips/configs/bigsur_defconfig
+++ b/arch/mips/configs/bigsur_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/capcella_defconfig b/arch/mips/configs/capcella_defconfig
index a01344f3a4c2..bfade9abb767 100644
--- a/arch/mips/configs/capcella_defconfig
+++ b/arch/mips/configs/capcella_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/cobalt_defconfig b/arch/mips/configs/cobalt_defconfig
index c95682445a28..78db2a8a711b 100644
--- a/arch/mips/configs/cobalt_defconfig
+++ b/arch/mips/configs/cobalt_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 CONFIG_MIPS_COBALT=y
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1000_defconfig b/arch/mips/configs/db1000_defconfig
index c2f33d3af62c..93cca1585bc3 100644
--- a/arch/mips/configs/db1000_defconfig
+++ b/arch/mips/configs/db1000_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1000=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1100_defconfig b/arch/mips/configs/db1100_defconfig
index 8c44d16ae9a2..ffd99252a837 100644
--- a/arch/mips/configs/db1100_defconfig
+++ b/arch/mips/configs/db1100_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1100=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1200_defconfig b/arch/mips/configs/db1200_defconfig
index c13768e75ac5..63eac5e89b9c 100644
--- a/arch/mips/configs/db1200_defconfig
+++ b/arch/mips/configs/db1200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1200=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1500_defconfig b/arch/mips/configs/db1500_defconfig
index 8aea73fae7fb..25a095f7dc4e 100644
--- a/arch/mips/configs/db1500_defconfig
+++ b/arch/mips/configs/db1500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1500=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/db1550_defconfig b/arch/mips/configs/db1550_defconfig
index 90ccb7359630..dda469c842b3 100644
--- a/arch/mips/configs/db1550_defconfig
+++ b/arch/mips/configs/db1550_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_DB1550=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ddb5477_defconfig b/arch/mips/configs/ddb5477_defconfig
index b598cf08f156..fcd3dd19bc74 100644
--- a/arch/mips/configs/ddb5477_defconfig
+++ b/arch/mips/configs/ddb5477_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/decstation_defconfig b/arch/mips/configs/decstation_defconfig
index 597150b14077..8683e0df12e0 100644
--- a/arch/mips/configs/decstation_defconfig
+++ b/arch/mips/configs/decstation_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 CONFIG_MACH_DECSTATION=y
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/e55_defconfig b/arch/mips/configs/e55_defconfig
index d9a0d0eb0683..4ace61c95778 100644
--- a/arch/mips/configs/e55_defconfig
+++ b/arch/mips/configs/e55_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/emma2rh_defconfig b/arch/mips/configs/emma2rh_defconfig
index 375b2ac24a49..5847c916c130 100644
--- a/arch/mips/configs/emma2rh_defconfig
+++ b/arch/mips/configs/emma2rh_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ev64120_defconfig b/arch/mips/configs/ev64120_defconfig
index b0afc118bd5c..bc4c4f125c48 100644
--- a/arch/mips/configs/ev64120_defconfig
+++ b/arch/mips/configs/ev64120_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 CONFIG_MIPS_EV64120=y
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ev96100_defconfig b/arch/mips/configs/ev96100_defconfig
deleted file mode 100644
index 0bdc10f11610..000000000000
--- a/arch/mips/configs/ev96100_defconfig
+++ /dev/null
@@ -1,850 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 10:04:05 2006
-#
-CONFIG_MIPS=y
-
-#
-# Machine selection
-#
-# CONFIG_MIPS_MTX1 is not set
-# CONFIG_MIPS_BOSPORUS is not set
-# CONFIG_MIPS_PB1000 is not set
-# CONFIG_MIPS_PB1100 is not set
-# CONFIG_MIPS_PB1500 is not set
-# CONFIG_MIPS_PB1550 is not set
-# CONFIG_MIPS_PB1200 is not set
-# CONFIG_MIPS_DB1000 is not set
-# CONFIG_MIPS_DB1100 is not set
-# CONFIG_MIPS_DB1500 is not set
-# CONFIG_MIPS_DB1550 is not set
-# CONFIG_MIPS_DB1200 is not set
-# CONFIG_MIPS_MIRAGE is not set
-# CONFIG_BASLER_EXCITE is not set
-# CONFIG_MIPS_COBALT is not set
-# CONFIG_MACH_DECSTATION is not set
-# CONFIG_MIPS_EV64120 is not set
-CONFIG_MIPS_EV96100=y
-# CONFIG_MIPS_IVR is not set
-# CONFIG_MIPS_ITE8172 is not set
-# CONFIG_MACH_JAZZ is not set
-# CONFIG_LASAT is not set
-# CONFIG_MIPS_ATLAS is not set
-# CONFIG_MIPS_MALTA is not set
-# CONFIG_MIPS_SEAD is not set
-# CONFIG_WR_PPMC is not set
-# CONFIG_MIPS_SIM is not set
-# CONFIG_MOMENCO_JAGUAR_ATX is not set
-# CONFIG_MOMENCO_OCELOT is not set
-# CONFIG_MOMENCO_OCELOT_3 is not set
-# CONFIG_MOMENCO_OCELOT_C is not set
-# CONFIG_MOMENCO_OCELOT_G is not set
-# CONFIG_MIPS_XXS1500 is not set
-# CONFIG_PNX8550_V2PCI is not set
-# CONFIG_PNX8550_JBS is not set
-# CONFIG_DDB5477 is not set
-# CONFIG_MACH_VR41XX is not set
-# CONFIG_PMC_YOSEMITE is not set
-# CONFIG_QEMU is not set
-# CONFIG_MARKEINS is not set
-# CONFIG_SGI_IP22 is not set
-# CONFIG_SGI_IP27 is not set
-# CONFIG_SGI_IP32 is not set
-# CONFIG_SIBYTE_BIGSUR is not set
-# CONFIG_SIBYTE_SWARM is not set
-# CONFIG_SIBYTE_SENTOSA is not set
-# CONFIG_SIBYTE_RHONE is not set
-# CONFIG_SIBYTE_CARMEL is not set
-# CONFIG_SIBYTE_PTSWARM is not set
-# CONFIG_SIBYTE_LITTLESUR is not set
-# CONFIG_SIBYTE_CRHINE is not set
-# CONFIG_SIBYTE_CRHONE is not set
-# CONFIG_SNI_RM200_PCI is not set
-# CONFIG_TOSHIBA_JMR3927 is not set
-# CONFIG_TOSHIBA_RBTX4927 is not set
-# CONFIG_TOSHIBA_RBTX4938 is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-CONFIG_GENERIC_FIND_NEXT_BIT=y
-CONFIG_GENERIC_HWEIGHT=y
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
-CONFIG_DMA_NONCOHERENT=y
-CONFIG_DMA_NEED_PCI_MAP_STATE=y
-CONFIG_CPU_BIG_ENDIAN=y
-# CONFIG_CPU_LITTLE_ENDIAN is not set
-CONFIG_SYS_SUPPORTS_BIG_ENDIAN=y
-CONFIG_IRQ_CPU=y
-CONFIG_MIPS_GT64120=y
-CONFIG_SWAP_IO_SPACE=y
-CONFIG_MIPS_GT96100=y
-CONFIG_MIPS_L1_CACHE_SHIFT=5
-
-#
-# CPU selection
-#
-# CONFIG_CPU_MIPS32_R1 is not set
-# CONFIG_CPU_MIPS32_R2 is not set
-# CONFIG_CPU_MIPS64_R1 is not set
-# CONFIG_CPU_MIPS64_R2 is not set
-# CONFIG_CPU_R3000 is not set
-# CONFIG_CPU_TX39XX is not set
-# CONFIG_CPU_VR41XX is not set
-# CONFIG_CPU_R4300 is not set
-# CONFIG_CPU_R4X00 is not set
-# CONFIG_CPU_TX49XX is not set
-# CONFIG_CPU_R5000 is not set
-# CONFIG_CPU_R5432 is not set
-# CONFIG_CPU_R6000 is not set
-# CONFIG_CPU_NEVADA is not set
-# CONFIG_CPU_R8000 is not set
-# CONFIG_CPU_R10000 is not set
-CONFIG_CPU_RM7000=y
-# CONFIG_CPU_RM9000 is not set
-# CONFIG_CPU_SB1 is not set
-CONFIG_SYS_HAS_CPU_R5000=y
-CONFIG_SYS_HAS_CPU_RM7000=y
-CONFIG_SYS_SUPPORTS_32BIT_KERNEL=y
-CONFIG_SYS_SUPPORTS_64BIT_KERNEL=y
-CONFIG_CPU_SUPPORTS_32BIT_KERNEL=y
-CONFIG_CPU_SUPPORTS_64BIT_KERNEL=y
-
-#
-# Kernel type
-#
-CONFIG_32BIT=y
-# CONFIG_64BIT is not set
-CONFIG_PAGE_SIZE_4KB=y
-# CONFIG_PAGE_SIZE_8KB is not set
-# CONFIG_PAGE_SIZE_16KB is not set
-# CONFIG_PAGE_SIZE_64KB is not set
-CONFIG_BOARD_SCACHE=y
-CONFIG_RM7000_CPU_SCACHE=y
-CONFIG_CPU_HAS_PREFETCH=y
-CONFIG_MIPS_MT_DISABLED=y
-# CONFIG_MIPS_MT_SMTC is not set
-# CONFIG_MIPS_MT_SMP is not set
-# CONFIG_MIPS_VPE_LOADER is not set
-# CONFIG_64BIT_PHYS_ADDR is not set
-CONFIG_CPU_HAS_LLSC=y
-CONFIG_CPU_HAS_SYNC=y
-CONFIG_GENERIC_HARDIRQS=y
-CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_CPU_SUPPORTS_HIGHMEM=y
-CONFIG_ARCH_FLATMEM_ENABLE=y
-CONFIG_SELECT_MEMORY_MODEL=y
-CONFIG_FLATMEM_MANUAL=y
-# CONFIG_DISCONTIGMEM_MANUAL is not set
-# CONFIG_SPARSEMEM_MANUAL is not set
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-# CONFIG_SPARSEMEM_STATIC is not set
-CONFIG_SPLIT_PTLOCK_CPUS=4
-# CONFIG_RESOURCES_64BIT is not set
-# CONFIG_HZ_48 is not set
-# CONFIG_HZ_100 is not set
-# CONFIG_HZ_128 is not set
-# CONFIG_HZ_250 is not set
-# CONFIG_HZ_256 is not set
-CONFIG_HZ_1000=y
-# CONFIG_HZ_1024 is not set
-CONFIG_SYS_SUPPORTS_ARBIT_HZ=y
-CONFIG_HZ=1000
-CONFIG_PREEMPT_NONE=y
-# CONFIG_PREEMPT_VOLUNTARY is not set
-# CONFIG_PREEMPT is not set
-CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
-
-#
-# Code maturity level options
-#
-CONFIG_EXPERIMENTAL=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-CONFIG_LOCALVERSION_AUTO=y
-CONFIG_SWAP=y
-CONFIG_SYSVIPC=y
-# CONFIG_POSIX_MQUEUE is not set
-# CONFIG_BSD_PROCESS_ACCT is not set
-CONFIG_SYSCTL=y
-# CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
-CONFIG_RELAY=y
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
-CONFIG_EMBEDDED=y
-CONFIG_KALLSYMS=y
-# CONFIG_KALLSYMS_EXTRA_PASS is not set
-# CONFIG_HOTPLUG is not set
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-CONFIG_ELF_CORE=y
-CONFIG_BASE_FULL=y
-CONFIG_RT_MUTEXES=y
-CONFIG_FUTEX=y
-CONFIG_EPOLL=y
-CONFIG_SHMEM=y
-CONFIG_SLAB=y
-CONFIG_VM_EVENT_COUNTERS=y
-# CONFIG_TINY_SHMEM is not set
-CONFIG_BASE_SMALL=0
-# CONFIG_SLOB is not set
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-# CONFIG_MODULE_FORCE_UNLOAD is not set
-CONFIG_MODVERSIONS=y
-CONFIG_MODULE_SRCVERSION_ALL=y
-# CONFIG_KMOD is not set
-
-#
-# Block layer
-#
-# CONFIG_LBD is not set
-# CONFIG_BLK_DEV_IO_TRACE is not set
-# CONFIG_LSF is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-CONFIG_IOSCHED_AS=y
-CONFIG_IOSCHED_DEADLINE=y
-CONFIG_IOSCHED_CFQ=y
-CONFIG_DEFAULT_AS=y
-# CONFIG_DEFAULT_DEADLINE is not set
-# CONFIG_DEFAULT_CFQ is not set
-# CONFIG_DEFAULT_NOOP is not set
-CONFIG_DEFAULT_IOSCHED="anticipatory"
-
-#
-# Bus options (PCI, PCMCIA, EISA, ISA, TC)
-#
-CONFIG_HW_HAS_PCI=y
-# CONFIG_PCI is not set
-CONFIG_MMU=y
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_ELF=y
-# CONFIG_BINFMT_MISC is not set
-CONFIG_TRAD_SIGNALS=y
-
-#
-# Networking
-#
-CONFIG_NET=y
-
-#
-# Networking options
-#
-# CONFIG_NETDEBUG is not set
-# CONFIG_PACKET is not set
-CONFIG_UNIX=y
-CONFIG_XFRM=y
-CONFIG_XFRM_USER=m
-CONFIG_NET_KEY=y
-CONFIG_INET=y
-# CONFIG_IP_MULTICAST is not set
-# CONFIG_IP_ADVANCED_ROUTER is not set
-CONFIG_IP_FIB_HASH=y
-CONFIG_IP_PNP=y
-# CONFIG_IP_PNP_DHCP is not set
-CONFIG_IP_PNP_BOOTP=y
-# CONFIG_IP_PNP_RARP is not set
-# CONFIG_NET_IPIP is not set
-# CONFIG_NET_IPGRE is not set
-# CONFIG_ARPD is not set
-# CONFIG_SYN_COOKIES is not set
-# CONFIG_INET_AH is not set
-# CONFIG_INET_ESP is not set
-# CONFIG_INET_IPCOMP is not set
-# CONFIG_INET_XFRM_TUNNEL is not set
-# CONFIG_INET_TUNNEL is not set
-CONFIG_INET_XFRM_MODE_TRANSPORT=m
-CONFIG_INET_XFRM_MODE_TUNNEL=m
-CONFIG_INET_DIAG=y
-CONFIG_INET_TCP_DIAG=y
-# CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_BIC=y
-# CONFIG_IPV6 is not set
-# CONFIG_INET6_XFRM_TUNNEL is not set
-# CONFIG_INET6_TUNNEL is not set
-CONFIG_NETWORK_SECMARK=y
-# CONFIG_NETFILTER is not set
-
-#
-# DCCP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_DCCP is not set
-
-#
-# SCTP Configuration (EXPERIMENTAL)
-#
-# CONFIG_IP_SCTP is not set
-
-#
-# TIPC Configuration (EXPERIMENTAL)
-#
-# CONFIG_TIPC is not set
-# CONFIG_ATM is not set
-# CONFIG_BRIDGE is not set
-# CONFIG_VLAN_8021Q is not set
-# CONFIG_DECNET is not set
-# CONFIG_LLC2 is not set
-# CONFIG_IPX is not set
-# CONFIG_ATALK is not set
-# CONFIG_X25 is not set
-# CONFIG_LAPB is not set
-# CONFIG_NET_DIVERT is not set
-# CONFIG_ECONET is not set
-# CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
-# CONFIG_NET_SCHED is not set
-
-#
-# Network testing
-#
-# CONFIG_NET_PKTGEN is not set
-# CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
-# CONFIG_BT is not set
-CONFIG_IEEE80211=m
-# CONFIG_IEEE80211_DEBUG is not set
-CONFIG_IEEE80211_CRYPT_WEP=m
-CONFIG_IEEE80211_CRYPT_CCMP=m
-CONFIG_IEEE80211_SOFTMAC=m
-# CONFIG_IEEE80211_SOFTMAC_DEBUG is not set
-CONFIG_WIRELESS_EXT=y
-
-#
-# Device Drivers
-#
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
-# CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
-CONFIG_CONNECTOR=m
-
-#
-# Memory Technology Devices (MTD)
-#
-# CONFIG_MTD is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_NBD is not set
-# CONFIG_BLK_DEV_RAM is not set
-# CONFIG_BLK_DEV_INITRD is not set
-CONFIG_CDROM_PKTCDVD=m
-CONFIG_CDROM_PKTCDVD_BUFFERS=8
-# CONFIG_CDROM_PKTCDVD_WCACHE is not set
-CONFIG_ATA_OVER_ETH=m
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-# CONFIG_IDE is not set
-
-#
-# SCSI device support
-#
-CONFIG_RAID_ATTRS=m
-# CONFIG_SCSI is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-
-#
-# I2O device support
-#
-
-#
-# Network device support
-#
-CONFIG_NETDEVICES=y
-# CONFIG_DUMMY is not set
-# CONFIG_BONDING is not set
-# CONFIG_EQUALIZER is not set
-# CONFIG_TUN is not set
-
-#
-# PHY device support
-#
-CONFIG_PHYLIB=m
-
-#
-# MII PHY device drivers
-#
-CONFIG_MARVELL_PHY=m
-CONFIG_DAVICOM_PHY=m
-CONFIG_QSEMI_PHY=m
-CONFIG_LXT_PHY=m
-CONFIG_CICADA_PHY=m
-CONFIG_VITESSE_PHY=m
-CONFIG_SMSC_PHY=m
-
-#
-# Ethernet (10 or 100Mbit)
-#
-CONFIG_NET_ETHERNET=y
-# CONFIG_MII is not set
-CONFIG_MIPS_GT96100ETH=y
-# CONFIG_DM9000 is not set
-
-#
-# Ethernet (1000 Mbit)
-#
-
-#
-# Ethernet (10000 Mbit)
-#
-
-#
-# Token Ring devices
-#
-
-#
-# Wireless LAN (non-hamradio)
-#
-# CONFIG_NET_RADIO is not set
-
-#
-# Wan interfaces
-#
-# CONFIG_WAN is not set
-# CONFIG_PPP is not set
-# CONFIG_SLIP is not set
-# CONFIG_SHAPER is not set
-# CONFIG_NETCONSOLE is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-# CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
-# CONFIG_PHONE is not set
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-
-#
-# Userland interfaces
-#
-CONFIG_INPUT_MOUSEDEV=y
-CONFIG_INPUT_MOUSEDEV_PSAUX=y
-CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
-CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-CONFIG_SERIO_SERPORT=y
-# CONFIG_SERIO_LIBPS2 is not set
-CONFIG_SERIO_RAW=m
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-CONFIG_VT=y
-CONFIG_VT_CONSOLE=y
-CONFIG_HW_CONSOLE=y
-CONFIG_VT_HW_CONSOLE_BINDING=y
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-CONFIG_SERIAL_8250=y
-CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=4
-CONFIG_SERIAL_8250_RUNTIME_UARTS=4
-# CONFIG_SERIAL_8250_EXTENDED is not set
-
-#
-# Non-8250 serial port support
-#
-CONFIG_SERIAL_CORE=y
-CONFIG_SERIAL_CORE_CONSOLE=y
-CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-# CONFIG_HW_RANDOM is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-# CONFIG_TCG_TPM is not set
-# CONFIG_TELCLOCK is not set
-
-#
-# I2C support
-#
-# CONFIG_I2C is not set
-
-#
-# SPI support
-#
-# CONFIG_SPI is not set
-# CONFIG_SPI_MASTER is not set
-
-#
-# Dallas's 1-wire bus
-#
-# CONFIG_W1 is not set
-
-#
-# Hardware Monitoring support
-#
-# CONFIG_HWMON is not set
-# CONFIG_HWMON_VID is not set
-
-#
-# Misc devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-CONFIG_VIDEO_V4L2=y
-
-#
-# Digital Video Broadcasting Devices
-#
-# CONFIG_DVB is not set
-
-#
-# Graphics support
-#
-# CONFIG_FIRMWARE_EDID is not set
-# CONFIG_FB is not set
-
-#
-# Console display driver support
-#
-# CONFIG_VGA_CONSOLE is not set
-CONFIG_DUMMY_CONSOLE=y
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-# CONFIG_USB_ARCH_HAS_HCD is not set
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# MMC/SD Card support
-#
-# CONFIG_MMC is not set
-
-#
-# LED devices
-#
-# CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
-# CONFIG_RTC_CLASS is not set
-
-#
-# DMA Engine support
-#
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# File systems
-#
-CONFIG_EXT2_FS=y
-# CONFIG_EXT2_FS_XATTR is not set
-# CONFIG_EXT2_FS_XIP is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-# CONFIG_XFS_FS is not set
-# CONFIG_OCFS2_FS is not set
-# CONFIG_MINIX_FS is not set
-# CONFIG_ROMFS_FS is not set
-CONFIG_INOTIFY=y
-CONFIG_INOTIFY_USER=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-CONFIG_FUSE_FS=m
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_PROC_KCORE=y
-CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-# CONFIG_CONFIGFS_FS is not set
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_ADFS_FS is not set
-# CONFIG_AFFS_FS is not set
-# CONFIG_HFS_FS is not set
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_BEFS_FS is not set
-# CONFIG_BFS_FS is not set
-# CONFIG_EFS_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
-CONFIG_NFS_FS=y
-# CONFIG_NFS_V3 is not set
-# CONFIG_NFS_V4 is not set
-# CONFIG_NFS_DIRECTIO is not set
-# CONFIG_NFSD is not set
-CONFIG_ROOT_NFS=y
-CONFIG_LOCKD=y
-CONFIG_NFS_COMMON=y
-CONFIG_SUNRPC=y
-# CONFIG_RPCSEC_GSS_KRB5 is not set
-# CONFIG_RPCSEC_GSS_SPKM3 is not set
-# CONFIG_SMB_FS is not set
-# CONFIG_CIFS is not set
-# CONFIG_CIFS_DEBUG2 is not set
-# CONFIG_NCP_FS is not set
-# CONFIG_CODA_FS is not set
-# CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Profiling support
-#
-# CONFIG_PROFILING is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-# CONFIG_MAGIC_SYSRQ is not set
-# CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_KERNEL is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_DEBUG_FS is not set
-CONFIG_CROSSCOMPILE=y
-CONFIG_CMDLINE=""
-
-#
-# Security options
-#
-CONFIG_KEYS=y
-CONFIG_KEYS_DEBUG_PROC_KEYS=y
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-CONFIG_CRYPTO=y
-CONFIG_CRYPTO_HMAC=y
-CONFIG_CRYPTO_NULL=m
-CONFIG_CRYPTO_MD4=m
-CONFIG_CRYPTO_MD5=m
-CONFIG_CRYPTO_SHA1=m
-CONFIG_CRYPTO_SHA256=m
-CONFIG_CRYPTO_SHA512=m
-CONFIG_CRYPTO_WP512=m
-CONFIG_CRYPTO_TGR192=m
-CONFIG_CRYPTO_DES=m
-CONFIG_CRYPTO_BLOWFISH=m
-CONFIG_CRYPTO_TWOFISH=m
-CONFIG_CRYPTO_SERPENT=m
-CONFIG_CRYPTO_AES=m
-CONFIG_CRYPTO_CAST5=m
-CONFIG_CRYPTO_CAST6=m
-CONFIG_CRYPTO_TEA=m
-CONFIG_CRYPTO_ARC4=m
-CONFIG_CRYPTO_KHAZAD=m
-CONFIG_CRYPTO_ANUBIS=m
-CONFIG_CRYPTO_DEFLATE=m
-CONFIG_CRYPTO_MICHAEL_MIC=m
-CONFIG_CRYPTO_CRC32C=m
-# CONFIG_CRYPTO_TEST is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-CONFIG_CRC16=m
-CONFIG_CRC32=m
-CONFIG_LIBCRC32C=m
-CONFIG_ZLIB_INFLATE=m
-CONFIG_ZLIB_DEFLATE=m
-CONFIG_PLIST=y
diff --git a/arch/mips/configs/excite_defconfig b/arch/mips/configs/excite_defconfig
index 045ebd089893..eb87cbbfd037 100644
--- a/arch/mips/configs/excite_defconfig
+++ b/arch/mips/configs/excite_defconfig
@@ -26,7 +26,6 @@ CONFIG_BASLER_EXCITE=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index ef16d1fb5071..9f22d13e729d 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 4bf1ee7f5f00..414a649dff8a 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ip32_defconfig b/arch/mips/configs/ip32_defconfig
index f83dc09c3ca9..dec2ba6ba03f 100644
--- a/arch/mips/configs/ip32_defconfig
+++ b/arch/mips/configs/ip32_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/it8172_defconfig b/arch/mips/configs/it8172_defconfig
index a91d72a9ca86..37f9dd7187b1 100644
--- a/arch/mips/configs/it8172_defconfig
+++ b/arch/mips/configs/it8172_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 CONFIG_MIPS_ITE8172=y
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ivr_defconfig b/arch/mips/configs/ivr_defconfig
index cebc67212d06..18874a4c24fe 100644
--- a/arch/mips/configs/ivr_defconfig
+++ b/arch/mips/configs/ivr_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 CONFIG_MIPS_IVR=y
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/jaguar-atx_defconfig b/arch/mips/configs/jaguar-atx_defconfig
index 5d9eb11aba3d..9b529857f802 100644
--- a/arch/mips/configs/jaguar-atx_defconfig
+++ b/arch/mips/configs/jaguar-atx_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/jmr3927_defconfig b/arch/mips/configs/jmr3927_defconfig
index be45a9044d06..fded3f73815f 100644
--- a/arch/mips/configs/jmr3927_defconfig
+++ b/arch/mips/configs/jmr3927_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/lasat200_defconfig b/arch/mips/configs/lasat200_defconfig
index 64dc9f45a19c..0354bfa18b39 100644
--- a/arch/mips/configs/lasat200_defconfig
+++ b/arch/mips/configs/lasat200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/malta_defconfig b/arch/mips/configs/malta_defconfig
index 2690baf15a85..b21ca470273a 100644
--- a/arch/mips/configs/malta_defconfig
+++ b/arch/mips/configs/malta_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/mipssim_defconfig b/arch/mips/configs/mipssim_defconfig
index c298979c18ae..adbeeadddb8f 100644
--- a/arch/mips/configs/mipssim_defconfig
+++ b/arch/mips/configs/mipssim_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/mpc30x_defconfig b/arch/mips/configs/mpc30x_defconfig
index 4405d127a941..79fd544fcb2a 100644
--- a/arch/mips/configs/mpc30x_defconfig
+++ b/arch/mips/configs/mpc30x_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_3_defconfig b/arch/mips/configs/ocelot_3_defconfig
index ec5758f22676..4d87da2b99fd 100644
--- a/arch/mips/configs/ocelot_3_defconfig
+++ b/arch/mips/configs/ocelot_3_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_c_defconfig b/arch/mips/configs/ocelot_c_defconfig
index 0d33d87de1a1..ce7c26475328 100644
--- a/arch/mips/configs/ocelot_c_defconfig
+++ b/arch/mips/configs/ocelot_c_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_defconfig b/arch/mips/configs/ocelot_defconfig
index 4b999102715e..3da2a43d1608 100644
--- a/arch/mips/configs/ocelot_defconfig
+++ b/arch/mips/configs/ocelot_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/ocelot_g_defconfig b/arch/mips/configs/ocelot_g_defconfig
index 827b344f6010..3a3e5ff4a3ef 100644
--- a/arch/mips/configs/ocelot_g_defconfig
+++ b/arch/mips/configs/ocelot_g_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1100_defconfig b/arch/mips/configs/pb1100_defconfig
index 9ed60fef69e0..1a16e92900cb 100644
--- a/arch/mips/configs/pb1100_defconfig
+++ b/arch/mips/configs/pb1100_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1100=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1500_defconfig b/arch/mips/configs/pb1500_defconfig
index 6774254b1be6..9ea8edea6f29 100644
--- a/arch/mips/configs/pb1500_defconfig
+++ b/arch/mips/configs/pb1500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1500=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pb1550_defconfig b/arch/mips/configs/pb1550_defconfig
index 1afe5bf6e765..c4a158976f8f 100644
--- a/arch/mips/configs/pb1550_defconfig
+++ b/arch/mips/configs/pb1550_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS_PB1550=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pnx8550-jbs_defconfig b/arch/mips/configs/pnx8550-jbs_defconfig
index ac616c82d348..1cbf270c301c 100644
--- a/arch/mips/configs/pnx8550-jbs_defconfig
+++ b/arch/mips/configs/pnx8550-jbs_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/pnx8550-v2pci_defconfig b/arch/mips/configs/pnx8550-v2pci_defconfig
index a8eb51bae3f3..bec30b15b9bd 100644
--- a/arch/mips/configs/pnx8550-v2pci_defconfig
+++ b/arch/mips/configs/pnx8550-v2pci_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/qemu_defconfig b/arch/mips/configs/qemu_defconfig
index 6a63a113b7ea..daa40c8ff694 100644
--- a/arch/mips/configs/qemu_defconfig
+++ b/arch/mips/configs/qemu_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/rbhma4500_defconfig b/arch/mips/configs/rbhma4500_defconfig
index 6779f449bd2d..2f5650227ba3 100644
--- a/arch/mips/configs/rbhma4500_defconfig
+++ b/arch/mips/configs/rbhma4500_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/rm200_defconfig b/arch/mips/configs/rm200_defconfig
index b7826d3a2b77..f26b338333ac 100644
--- a/arch/mips/configs/rm200_defconfig
+++ b/arch/mips/configs/rm200_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/sb1250-swarm_defconfig b/arch/mips/configs/sb1250-swarm_defconfig
index 625c1c619b6b..9041f095f96f 100644
--- a/arch/mips/configs/sb1250-swarm_defconfig
+++ b/arch/mips/configs/sb1250-swarm_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/sead_defconfig b/arch/mips/configs/sead_defconfig
index 4401b602118f..02abb2f1bfaf 100644
--- a/arch/mips/configs/sead_defconfig
+++ b/arch/mips/configs/sead_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0226_defconfig b/arch/mips/configs/tb0226_defconfig
index 2ba4e25e8c34..ca3d0c4ba15b 100644
--- a/arch/mips/configs/tb0226_defconfig
+++ b/arch/mips/configs/tb0226_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0229_defconfig b/arch/mips/configs/tb0229_defconfig
index fc8a407c1add..4e2009ace278 100644
--- a/arch/mips/configs/tb0229_defconfig
+++ b/arch/mips/configs/tb0229_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/tb0287_defconfig b/arch/mips/configs/tb0287_defconfig
index effcb63b81a3..535a813d01a9 100644
--- a/arch/mips/configs/tb0287_defconfig
+++ b/arch/mips/configs/tb0287_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/workpad_defconfig b/arch/mips/configs/workpad_defconfig
index 8f66ad71cb7e..3a3ef20b21cc 100644
--- a/arch/mips/configs/workpad_defconfig
+++ b/arch/mips/configs/workpad_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/wrppmc_defconfig b/arch/mips/configs/wrppmc_defconfig
index 3e4b16b39827..e6b1dea55842 100644
--- a/arch/mips/configs/wrppmc_defconfig
+++ b/arch/mips/configs/wrppmc_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/configs/yosemite_defconfig b/arch/mips/configs/yosemite_defconfig
index 3a68d8a25b66..06a072b77b1c 100644
--- a/arch/mips/configs/yosemite_defconfig
+++ b/arch/mips/configs/yosemite_defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/defconfig b/arch/mips/defconfig
index fff6fcc96212..d620c463a78b 100644
--- a/arch/mips/defconfig
+++ b/arch/mips/defconfig
@@ -25,7 +25,6 @@ CONFIG_MIPS=y
 # CONFIG_MIPS_COBALT is not set
 # CONFIG_MACH_DECSTATION is not set
 # CONFIG_MIPS_EV64120 is not set
-# CONFIG_MIPS_EV96100 is not set
 # CONFIG_MIPS_IVR is not set
 # CONFIG_MIPS_ITE8172 is not set
 # CONFIG_MACH_JAZZ is not set
diff --git a/arch/mips/galileo-boards/ev96100/Makefile b/arch/mips/galileo-boards/ev96100/Makefile
deleted file mode 100644
index cd868ec78cbc..000000000000
--- a/arch/mips/galileo-boards/ev96100/Makefile
+++ /dev/null
@@ -1,9 +0,0 @@
-#
-#  Copyright 2000 MontaVista Software Inc.
-#  Author: MontaVista Software, Inc.
-#     	ppopov@mvista.com or source@mvista.com
-#
-# Makefile for the Galileo EV96100 board.
-#
-
-obj-y		+= init.o irq.o puts.o reset.o time.o setup.o
diff --git a/arch/mips/galileo-boards/ev96100/init.c b/arch/mips/galileo-boards/ev96100/init.c
deleted file mode 100644
index a01fe9b36f2c..000000000000
--- a/arch/mips/galileo-boards/ev96100/init.c
+++ /dev/null
@@ -1,173 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/generic.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/bootmem.h>
-#include <linux/string.h>
-#include <linux/kernel.h>
-
-#include <asm/addrspace.h>
-#include <asm/bootinfo.h>
-#include <asm/gt64120.h>
-
-
-/* Environment variable */
-
-typedef struct {
-	char *name;
-	char *val;
-} t_env_var;
-
-int prom_argc;
-char **prom_argv, **prom_envp;
-
-int init_debug = 0;
-
-char * __init prom_getcmdline(void)
-{
-	return &(arcs_cmdline[0]);
-}
-
-unsigned long __init prom_free_prom_memory(void)
-{
-	return 0;
-}
-
-void  __init prom_init_cmdline(void)
-{
-	char *cp;
-	int actr;
-
-	actr = 1; /* Always ignore argv[0] */
-
-	cp = &(arcs_cmdline[0]);
-	while(actr < prom_argc) {
-	        strcpy(cp, prom_argv[actr]);
-		cp += strlen(prom_argv[actr]);
-		*cp++ = ' ';
-		actr++;
-	}
-	if (cp != &(arcs_cmdline[0])) /* get rid of trailing space */
-		--cp;
-	*cp = '\0';
-}
-
-char *prom_getenv(char *envname)
-{
-	/*
-	 * Return a pointer to the given environment variable.
-	 */
-
-	t_env_var *env = (t_env_var *) prom_envp;
-	int i;
-
-	i = strlen(envname);
-
-	while (env->name) {
-		if (strncmp(envname, env->name, i) == 0) {
-			return (env->val);
-		}
-		env++;
-	}
-	return (NULL);
-}
-
-static inline unsigned char str2hexnum(unsigned char c)
-{
-	if (c >= '0' && c <= '9')
-		return c - '0';
-	if (c >= 'a' && c <= 'f')
-		return c - 'a' + 10;
-	return 0;		/* foo */
-}
-
-static inline void str2eaddr(unsigned char *ea, unsigned char *str)
-{
-	int i;
-
-	for (i = 0; i < 6; i++) {
-		unsigned char num;
-
-		if ((*str == '.') || (*str == ':'))
-			str++;
-		num = str2hexnum(*str++) << 4;
-		num |= (str2hexnum(*str++));
-		ea[i] = num;
-	}
-}
-
-int get_ethernet_addr(char *ethernet_addr)
-{
-	char *ethaddr_str;
-
-	ethaddr_str = prom_getenv("ethaddr");
-	if (!ethaddr_str) {
-		printk("ethaddr not set in boot prom\n");
-		return -1;
-	}
-	str2eaddr(ethernet_addr, ethaddr_str);
-
-	if (init_debug > 1) {
-		int i;
-		printk("get_ethernet_addr: ");
-		for (i = 0; i < 5; i++)
-			printk("%02x:",
-			       (unsigned char) *(ethernet_addr + i));
-		printk("%02x\n", *(ethernet_addr + i));
-	}
-
-	return 0;
-}
-
-const char *get_system_type(void)
-{
-	return "Galileo EV96100";
-}
-
-void __init prom_init(void)
-{
-	volatile unsigned char *uart;
-	char ppbuf[8];
-
-	prom_argc = fw_arg0;
-	prom_argv = (char **) fw_arg1;
-	prom_envp = (char **) fw_arg2;
-
-	mips_machgroup = MACH_GROUP_GALILEO;
-	mips_machtype = MACH_EV96100;
-
-	prom_init_cmdline();
-
-	/* 32 MB upgradable */
-	add_memory_region(0, 32 << 20, BOOT_MEM_RAM);
-}
diff --git a/arch/mips/galileo-boards/ev96100/irq.c b/arch/mips/galileo-boards/ev96100/irq.c
deleted file mode 100644
index ee5d6720f23b..000000000000
--- a/arch/mips/galileo-boards/ev96100/irq.c
+++ /dev/null
@@ -1,77 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_int.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/errno.h>
-#include <linux/init.h>
-#include <linux/kernel_stat.h>
-#include <linux/irq.h>
-#include <linux/module.h>
-#include <linux/signal.h>
-#include <linux/sched.h>
-#include <linux/types.h>
-#include <linux/interrupt.h>
-#include <asm/irq_cpu.h>
-
-static inline unsigned int ffz8(unsigned int word)
-{
-	unsigned long k;
-
-	k = 7;
-	if (word & 0x0fUL) { k -= 4;  word <<= 4;  }
-	if (word & 0x30UL) { k -= 2;  word <<= 2;  }
-	if (word & 0x40UL) { k -= 1; }
-
-	return k;
-}
-
-extern void mips_timer_interrupt(struct pt_regs *regs);
-
-asmlinkage void ev96100_cpu_irq(unsigned int pending, struct pt_regs *regs)
-{
-	do_IRQ(ffz8(pending >> 8), regs);
-}
-
-asmlinkage void plat_irq_dispatch(struct pt_regs *regs)
-{
-	unsigned int pending = read_c0_cause() & read_c0_status() & ST0_IM;
-
-	if (pending & CAUSEF_IP7)
-		mips_timer_interrupt(regs);
-	else if (pending)
-		ev96100_cpu_irq(pending, regs);
-	else
-		spurious_interrupt(regs);
-}
-
-void __init arch_init_irq(void)
-{
-	mips_cpu_irq_init(0);
-}
diff --git a/arch/mips/galileo-boards/ev96100/puts.c b/arch/mips/galileo-boards/ev96100/puts.c
deleted file mode 100644
index 49dc6d137b9c..000000000000
--- a/arch/mips/galileo-boards/ev96100/puts.c
+++ /dev/null
@@ -1,138 +0,0 @@
-
-/*
- * Debug routines which directly access the uart.
- */
-
-#include <linux/types.h>
-#include <asm/gt64120.h>
-
-
-//#define SERIAL_BASE    EV96100_UART0_REGS_BASE
-#define SERIAL_BASE    0xBD000020
-#define NS16550_BASE   SERIAL_BASE
-
-#define SERA_CMD       0x0D
-#define SERA_DATA      0x08
-//#define SERB_CMD       0x05
-#define SERB_CMD       20
-#define SERB_DATA      0x00
-#define TX_BUSY        0x20
-
-#define TIMEOUT    0xffff
-#undef SLOW_DOWN
-
-static const char digits[16] = "0123456789abcdef";
-static volatile unsigned char *const com1 = (unsigned char *) SERIAL_BASE;
-
-
-#ifdef SLOW_DOWN
-static inline void slow_down()
-{
-	int k;
-	for (k = 0; k < 10000; k++);
-}
-#else
-#define slow_down()
-#endif
-
-void putch(const unsigned char c)
-{
-	unsigned char ch;
-	int i = 0;
-
-	do {
-		ch = com1[SERB_CMD];
-		slow_down();
-		i++;
-		if (i > TIMEOUT) {
-			break;
-		}
-	} while (0 == (ch & TX_BUSY));
-	com1[SERB_DATA] = c;
-}
-
-void putchar(const unsigned char c)
-{
-	unsigned char ch;
-	int i = 0;
-
-	do {
-		ch = com1[SERB_CMD];
-		slow_down();
-		i++;
-		if (i > TIMEOUT) {
-			break;
-		}
-	} while (0 == (ch & TX_BUSY));
-	com1[SERB_DATA] = c;
-}
-
-void puts(unsigned char *cp)
-{
-	unsigned char ch;
-	int i = 0;
-
-	while (*cp) {
-		do {
-			ch = com1[SERB_CMD];
-			slow_down();
-			i++;
-			if (i > TIMEOUT) {
-				break;
-			}
-		} while (0 == (ch & TX_BUSY));
-		com1[SERB_DATA] = *cp++;
-	}
-	putch('\r');
-	putch('\n');
-}
-
-void fputs(unsigned char *cp)
-{
-	unsigned char ch;
-	int i = 0;
-
-	while (*cp) {
-
-		do {
-			ch = com1[SERB_CMD];
-			slow_down();
-			i++;
-			if (i > TIMEOUT) {
-				break;
-			}
-		} while (0 == (ch & TX_BUSY));
-		com1[SERB_DATA] = *cp++;
-	}
-}
-
-
-void put64(uint64_t ul)
-{
-	int cnt;
-	unsigned ch;
-
-	cnt = 16;		/* 16 nibbles in a 64 bit long */
-	putch('0');
-	putch('x');
-	do {
-		cnt--;
-		ch = (unsigned char) (ul >> cnt * 4) & 0x0F;
-		putch(digits[ch]);
-	} while (cnt > 0);
-}
-
-void put32(unsigned u)
-{
-	int cnt;
-	unsigned ch;
-
-	cnt = 8;		/* 8 nibbles in a 32 bit long */
-	putch('0');
-	putch('x');
-	do {
-		cnt--;
-		ch = (unsigned char) (u >> cnt * 4) & 0x0F;
-		putch(digits[ch]);
-	} while (cnt > 0);
-}
diff --git a/arch/mips/galileo-boards/ev96100/reset.c b/arch/mips/galileo-boards/ev96100/reset.c
deleted file mode 100644
index 5ef9b7f896e6..000000000000
--- a/arch/mips/galileo-boards/ev96100/reset.c
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 reset routines.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/reset.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/sched.h>
-#include <linux/mm.h>
-#include <asm/io.h>
-#include <asm/pgtable.h>
-#include <asm/processor.h>
-#include <asm/reboot.h>
-#include <asm/system.h>
-#include <asm/gt64120.h>
-
-static void mips_machine_restart(char *command);
-static void mips_machine_halt(void);
-
-static void mips_machine_restart(char *command)
-{
-	set_c0_status(ST0_BEV | ST0_ERL);
-	change_c0_config(CONF_CM_CMASK, CONF_CM_UNCACHED);
-	flush_cache_all();
-	write_c0_wired(0);
-	__asm__ __volatile__("jr\t%0"::"r"(0xbfc00000));
-	while (1);
-}
-
-static void mips_machine_halt(void)
-{
-	printk(KERN_NOTICE "You can safely turn off the power\n");
-	while (1)
-		__asm__(".set\tmips3\n\t"
-	                "wait\n\t"
-			".set\tmips0");
-}
-
-void mips_reboot_setup(void)
-{
-	_machine_restart = mips_machine_restart;
-	_machine_halt = mips_machine_halt;
-}
diff --git a/arch/mips/galileo-boards/ev96100/setup.c b/arch/mips/galileo-boards/ev96100/setup.c
deleted file mode 100644
index 639ad5562c63..000000000000
--- a/arch/mips/galileo-boards/ev96100/setup.c
+++ /dev/null
@@ -1,159 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 setup.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_setup.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/sched.h>
-#include <linux/ioport.h>
-#include <linux/string.h>
-#include <linux/ctype.h>
-#include <linux/pci.h>
-
-#include <asm/cpu.h>
-#include <asm/bootinfo.h>
-#include <asm/mipsregs.h>
-#include <asm/irq.h>
-#include <asm/delay.h>
-#include <asm/gt64120.h>
-#include <asm/galileo-boards/ev96100int.h>
-
-
-extern char *__init prom_getcmdline(void);
-
-extern void mips_reboot_setup(void);
-
-unsigned char mac_0_1[12];
-
-void __init plat_mem_setup(void)
-{
-	unsigned int config = read_c0_config();
-	unsigned int status = read_c0_status();
-	unsigned int info = read_c0_info();
-	u32 tmp;
-
-	char *argptr;
-
-	clear_c0_status(ST0_FR);
-
-	if (config & 0x8)
-		printk("Secondary cache is enabled\n");
-	else
-		printk("Secondary cache is disabled\n");
-
-	if (status & (1 << 27))
-		printk("User-mode cache ops enabled\n");
-	else
-		printk("User-mode cache ops disabled\n");
-
-	printk("CP0 info reg: %x\n", (unsigned) info);
-	if (info & (1 << 28))
-		printk("burst mode Scache RAMS\n");
-	else
-		printk("pipelined Scache RAMS\n");
-
-	if (info & 0x1)
-		printk("Atomic Enable is set\n");
-
-	argptr = prom_getcmdline();
-#ifdef CONFIG_SERIAL_CONSOLE
-	if (strstr(argptr, "console=") == NULL) {
-		argptr = prom_getcmdline();
-		strcat(argptr, " console=ttyS0,115200");
-	}
-#endif
-
-	mips_reboot_setup();
-
-	set_io_port_base(KSEG1);
-	ioport_resource.start = GT_PCI_IO_BASE;
-	ioport_resource.end = GT_PCI_IO_BASE + 0x01ffffff;
-
-#ifdef CONFIG_BLK_DEV_INITRD
-	ROOT_DEV = MKDEV(RAMDISK_MAJOR, 0);
-#endif
-
-
-	/*
-	 * Setup GT controller master bit so we can do config cycles
-	 */
-
-	/* Clear cause register bits */
-	GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-				     GT_INTRCAUSE_TARABORT0_BIT));
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(2);
-	tmp = GT_READ(GT_PCI0_CFGDATA_OFS);
-
-	tmp |= (PCI_COMMAND_IO | PCI_COMMAND_MEMORY |
-		PCI_COMMAND_MASTER | PCI_COMMAND_SERR);
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-	udelay(2);
-	GT_WRITE(GT_PCI0_CFGDATA_OFS, tmp);
-
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_COMMAND / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(2);
-	tmp = GT_READ(GT_PCI0_CFGDATA_OFS);
-}
-
-unsigned short get_gt_devid(void)
-{
-	u32 gt_devid;
-
-	/* Figure out if this is a gt96100 or gt96100A */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (0 << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (0 << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((PCI_VENDOR_ID / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-
-	udelay(4);
-	gt_devid = GT_READ(GT_PCI0_CFGDATA_OFS);
-
-	return gt_devid >> 16;
-}
diff --git a/arch/mips/galileo-boards/ev96100/time.c b/arch/mips/galileo-boards/ev96100/time.c
deleted file mode 100644
index 8cbe8426491a..000000000000
--- a/arch/mips/galileo-boards/ev96100/time.c
+++ /dev/null
@@ -1,88 +0,0 @@
-/*
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 rtc routines.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/atlas/atlas_rtc.c.
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/kernel_stat.h>
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/spinlock.h>
-#include <linux/timex.h>
-
-#include <asm/mipsregs.h>
-#include <asm/ptrace.h>
-#include <asm/time.h>
-
-
-#define ALLINTS (IE_IRQ0 | IE_IRQ1 | IE_IRQ2 | IE_IRQ3 | IE_IRQ4 | IE_IRQ5)
-
-extern volatile unsigned long wall_jiffies;
-unsigned long missed_heart_beats = 0;
-
-static unsigned long r4k_offset; /* Amount to increment compare reg each time */
-static unsigned long r4k_cur;    /* What counter should be at next timer irq */
-
-static inline void ack_r4ktimer(unsigned long newval)
-{
-	write_c0_compare(newval);
-}
-
-/*
- * There are a lot of conceptually broken versions of the MIPS timer interrupt
- * handler floating around.  This one is rather different, but the algorithm
- * is probably more robust.
- */
-void mips_timer_interrupt(struct pt_regs *regs)
-{
-        int irq = 7; /* FIX ME */
-
-	if (r4k_offset == 0) {
-            goto null;
-        }
-
-	do {
-		kstat_this_cpu.irqs[irq]++;
-		do_timer(regs);
-#ifndef CONFIG_SMP
-		update_process_times(user_mode(regs));
-#endif
-		r4k_cur += r4k_offset;
-		ack_r4ktimer(r4k_cur);
-
-	} while (((unsigned long)read_c0_count()
-                    - r4k_cur) < 0x7fffffff);
-	return;
-
-null:
-	ack_r4ktimer(0);
-}
diff --git a/arch/mips/pci/Makefile b/arch/mips/pci/Makefile
index 368d3d896165..edefa97b2330 100644
--- a/arch/mips/pci/Makefile
+++ b/arch/mips/pci/Makefile
@@ -11,7 +11,6 @@ obj-$(CONFIG_ITE_BOARD_GEN)	+= ops-it8172.o
 obj-$(CONFIG_MIPS_BONITO64)	+= ops-bonito64.o
 obj-$(CONFIG_MIPS_GT64111)	+= ops-gt64111.o
 obj-$(CONFIG_MIPS_GT64120)	+= ops-gt64120.o
-obj-$(CONFIG_MIPS_GT96100)	+= ops-gt96100.o
 obj-$(CONFIG_PCI_MARVELL)	+= ops-marvell.o
 obj-$(CONFIG_MIPS_MSC)		+= ops-msc.o
 obj-$(CONFIG_MIPS_NILE4)	+= ops-nile4.o
@@ -29,7 +28,6 @@ obj-$(CONFIG_LASAT)		+= pci-lasat.o
 obj-$(CONFIG_MIPS_ATLAS)	+= fixup-atlas.o
 obj-$(CONFIG_MIPS_COBALT)	+= fixup-cobalt.o
 obj-$(CONFIG_MIPS_EV64120)	+= fixup-ev64120.o
-obj-$(CONFIG_MIPS_EV96100)	+= fixup-ev96100.o pci-ev96100.o
 obj-$(CONFIG_MIPS_ITE8172)	+= fixup-ite8172g.o
 obj-$(CONFIG_MIPS_IVR)		+= fixup-ivr.o
 obj-$(CONFIG_SOC_AU1500)	+= fixup-au1000.o ops-au1000.o
diff --git a/arch/mips/pci/fixup-ev96100.c b/arch/mips/pci/fixup-ev96100.c
deleted file mode 100644
index e2bc977b6d58..000000000000
--- a/arch/mips/pci/fixup-ev96100.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- *
- * BRIEF MODULE DESCRIPTION
- *	EV96100 Board specific pci fixups.
- *
- * Copyright 2001 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/init.h>
-#include <linux/types.h>
-#include <linux/pci.h>
-
-static char irq_tab_ev96100[][5] __initdata = {
- [8] = { 0, 5, 5, 5, 5 },
- [9] = { 0, 2, 2, 2, 2 }
-};
-
-int __init pcibios_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
-{
-	return irq_tab_ev96100[slot][pin];
-}
-
-/* Do platform specific device initialization at pci_enable_device() time */
-int pcibios_plat_dev_init(struct pci_dev *dev)
-{
-	return 0;
-}
diff --git a/arch/mips/pci/ops-gt96100.c b/arch/mips/pci/ops-gt96100.c
deleted file mode 100644
index 9e4ea6627e21..000000000000
--- a/arch/mips/pci/ops-gt96100.c
+++ /dev/null
@@ -1,169 +0,0 @@
-/*
- *
- * BRIEF MODULE DESCRIPTION
- *	Galileo EV96100 board specific pci support.
- *
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * This file was derived from Carsten Langgaard's
- * arch/mips/mips-boards/generic/pci.c
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-
-#include <asm/delay.h>
-#include <asm/gt64120.h>
-#include <asm/galileo-boards/ev96100.h>
-
-#define PCI_ACCESS_READ  0
-#define PCI_ACCESS_WRITE 1
-
-static int static gt96100_config_access(unsigned char access_type,
-	struct pci_bus *bus, unsigned int devfn, int where, u32 * data)
-{
-	unsigned char bus = bus->number;
-	u32 intr;
-
-	/*
-	 * Because of a bug in the galileo (for slot 31).
-	 */
-	if (bus == 0 && devfn >= PCI_DEVFN(31, 0))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	/* Clear cause register bits */
-	GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-				     GT_INTRCAUSE_TARABORT0_BIT));
-
-	/* Setup address */
-	GT_WRITE(GT_PCI0_CFGADDR_OFS,
-		 (bus << GT_PCI0_CFGADDR_BUSNUM_SHF) |
-		 (devfn << GT_PCI0_CFGADDR_FUNCTNUM_SHF) |
-		 ((where / 4) << GT_PCI0_CFGADDR_REGNUM_SHF) |
-		 GT_PCI0_CFGADDR_CONFIGEN_BIT);
-	udelay(2);
-
-
-	if (access_type == PCI_ACCESS_WRITE) {
-		if (devfn != 0)
-			*data = le32_to_cpu(*data);
-		GT_WRITE(GT_PCI0_CFGDATA_OFS, *data);
-	} else {
-		*data = GT_READ(GT_PCI0_CFGDATA_OFS);
-		if (devfn != 0)
-			*data = le32_to_cpu(*data);
-	}
-
-	udelay(2);
-
-	/* Check for master or target abort */
-	intr = GT_READ(GT_INTRCAUSE_OFS);
-
-	if (intr & (GT_INTRCAUSE_MASABORT0_BIT | GT_INTRCAUSE_TARABORT0_BIT)) {
-		/* Error occured */
-
-		/* Clear bits */
-		GT_WRITE(GT_INTRCAUSE_OFS, ~(GT_INTRCAUSE_MASABORT0_BIT |
-					     GT_INTRCAUSE_TARABORT0_BIT));
-		return -1;
-	}
-	return 0;
-}
-
-/*
- * We can't address 8 and 16 bit words directly.  Instead we have to
- * read/write a 32bit word and mask/modify the data we actually want.
- */
-static int gt96100_pcibios_read(struct pci_bus *bus, unsigned int devfn,
-	int where, int size, u32 * val)
-{
-	u32 data = 0;
-
-	if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-		return PCIBIOS_DEVICE_NOT_FOUND;
-
-	switch (size) {
-	case 1:
-		*val = (data >> ((where & 3) << 3)) & 0xff;
-		break;
-
-	case 2:
-		*val = (data >> ((where & 3) << 3)) & 0xffff;
-		break;
-
-	case 4:
-		*val = data;
-		break;
-	}
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int gt96100_pcibios_write(struct pci_bus *bus, unsigned int devfn,
-	int where, int size, u32 val)
-{
-	u32 data = 0;
-
-	switch (size) {
-	case 1:
-		if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-			return -1;
-
-		data = (data & ~(0xff << ((where & 3) << 3))) |
-		       (val << ((where & 3) << 3));
-
-		if (gt96100_config_access(PCI_ACCESS_WRITE, bus, devfn, where, &data))
-			return -1;
-
-		return PCIBIOS_SUCCESSFUL;
-
-	case 2:
-		if (gt96100_config_access(PCI_ACCESS_READ, bus, devfn, where, &data))
-			return -1;
-
-		data = (data & ~(0xffff << ((where & 3) << 3))) |
-		       (val << ((where & 3) << 3));
-
-		if (gt96100_config_access(PCI_ACCESS_WRITE, dev, where, &data))
-			return -1;
-
-
-		return PCIBIOS_SUCCESSFUL;
-
-	case 4:
-		if (gt96100_config_access(PCI_ACCESS_WRITE, dev, where, &val))
-			return -1;
-
-		return PCIBIOS_SUCCESSFUL;
-	}
-}
-
-struct pci_ops gt96100_pci_ops = {
-	.read	= gt96100_pcibios_read,
-	.write	= gt96100_pcibios_write
-};
diff --git a/arch/mips/pci/pci-ev96100.c b/arch/mips/pci/pci-ev96100.c
deleted file mode 100644
index f9457ea00def..000000000000
--- a/arch/mips/pci/pci-ev96100.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	ppopov@mvista.com or source@mvista.com
- *
- * Carsten Langgaard, carstenl@mips.com
- * Copyright (C) 1999,2000 MIPS Technologies, Inc.  All rights reserved.
- *
- * Copyright (C) 2004 by Ralf Baechle (ralf@linux-mips.org)
- *
- *  This program is free software; you can redistribute  it and/or modify it
- *  under  the terms of  the GNU General  Public License as published by the
- *  Free Software Foundation;  either version 2 of the  License, or (at your
- *  option) any later version.
- *
- *  THIS  SOFTWARE  IS PROVIDED   ``AS  IS'' AND   ANY  EXPRESS OR IMPLIED
- *  WARRANTIES,   INCLUDING, BUT NOT  LIMITED  TO, THE IMPLIED WARRANTIES OF
- *  MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN
- *  NO  EVENT  SHALL   THE AUTHOR  BE    LIABLE FOR ANY   DIRECT, INDIRECT,
- *  INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
- *  NOT LIMITED   TO, PROCUREMENT OF  SUBSTITUTE GOODS  OR SERVICES; LOSS OF
- *  USE, DATA,  OR PROFITS; OR  BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
- *  ANY THEORY OF LIABILITY, WHETHER IN  CONTRACT, STRICT LIABILITY, OR TORT
- *  (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
- *  THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
- *
- *  You should have received a copy of the  GNU General Public License along
- *  with this program; if not, write  to the Free Software Foundation, Inc.,
- *  675 Mass Ave, Cambridge, MA 02139, USA.
- */
-#include <linux/types.h>
-#include <linux/pci.h>
-#include <linux/kernel.h>
-#include <linux/init.h>
-
-static struct resource pci_io_resource = {
-	.name	= "io pci IO space",
-	.start	= 0x10000000,
-	.end	= 0x11ffffff,
-	.flags	= IORESOURCE_IO
-};
-
-static struct resource pci_mem_resource = {
-	.name	= "ext pci memory space",
-	.start	= 0x12000000,
-	.end	= 0x13ffffff,
-	.flags	= IORESOURCE_MEM
-};
-
-extern struct pci_ops gt96100_pci_ops;
-
-struct pci_controller ev96100_controller = {
-	.pci_ops	= &gt96100_pci_ops,
-	.io_resource	= &pci_io_resource,
-	.mem_resource	= &pci_mem_resource,
-};
-
-static void ev96100_pci_init(void)
-{
-	register_pci_controller(&ev96100_controller);
-}
-
-arch_initcall(ev96100_pci_init);
diff --git a/include/asm-mips/bootinfo.h b/include/asm-mips/bootinfo.h
index 3b745e76f429..78c35ec46362 100644
--- a/include/asm-mips/bootinfo.h
+++ b/include/asm-mips/bootinfo.h
@@ -112,8 +112,7 @@
  * Valid machtype for group GALILEO
  */
 #define MACH_GROUP_GALILEO     11	/* Galileo Eval Boards		*/
-#define  MACH_EV96100		0	/* EV96100 */
-#define  MACH_EV64120A		1	/* EV64120A */
+#define  MACH_EV64120A		0	/* EV64120A */
 
 /*
  * Valid machtype for group MOMENCO
diff --git a/include/asm-mips/galileo-boards/gt96100.h b/include/asm-mips/galileo-boards/gt96100.h
deleted file mode 100644
index aabd1b629c19..000000000000
--- a/include/asm-mips/galileo-boards/gt96100.h
+++ /dev/null
@@ -1,427 +0,0 @@
-/*
- * Copyright 2000 MontaVista Software Inc.
- * Author: MontaVista Software, Inc.
- *         	stevel@mvista.com or source@mvista.com
- *
- *  This program is free software; you can distribute it and/or modify it
- *  under the terms of the GNU General Public License (Version 2) as
- *  published by the Free Software Foundation.
- *
- *  This program is distributed in the hope it will be useful, but WITHOUT
- *  ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- *  FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
- *  for more details.
- *
- *  You should have received a copy of the GNU General Public License along
- *  with this program; if not, write to the Free Software Foundation, Inc.,
- *  59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
- *
- * Register offsets of the MIPS GT96100 Advanced Communication Controller.
- */
-#ifndef _GT96100_H
-#define _GT96100_H
-
-/*
- * Galileo GT96100 internal register base.
- */
-#define MIPS_GT96100_BASE (KSEG1ADDR(0x14000000))
-
-#define GT96100_WRITE(ofs, data) \
-    *(volatile u32 *)(MIPS_GT96100_BASE+ofs) = cpu_to_le32(data)
-#define GT96100_READ(ofs) \
-    le32_to_cpu(*(volatile u32 *)(MIPS_GT96100_BASE+ofs))
-
-#define GT96100_ETH_IO_SIZE 0x4000
-
-/************************************************************************
- *  Register offset addresses follow
- ************************************************************************/
-
-/* CPU Interface Control Registers */
-#define GT96100_CPU_INTERF_CONFIG 0x000000
-
-/* Ethernet Ports */
-#define GT96100_ETH_PHY_ADDR_REG             0x080800
-#define GT96100_ETH_SMI_REG                  0x080810
-/*
-  These are offsets to port 0 registers. Add GT96100_ETH_IO_SIZE to
-  get offsets to port 1 registers.
-*/
-#define GT96100_ETH_PORT_CONFIG          0x084800
-#define GT96100_ETH_PORT_CONFIG_EXT      0x084808
-#define GT96100_ETH_PORT_COMM            0x084810
-#define GT96100_ETH_PORT_STATUS          0x084818
-#define GT96100_ETH_SER_PARAM            0x084820
-#define GT96100_ETH_HASH_TBL_PTR         0x084828
-#define GT96100_ETH_FLOW_CNTRL_SRC_ADDR_L    0x084830
-#define GT96100_ETH_FLOW_CNTRL_SRC_ADDR_H    0x084838
-#define GT96100_ETH_SDMA_CONFIG          0x084840
-#define GT96100_ETH_SDMA_COMM            0x084848
-#define GT96100_ETH_INT_CAUSE            0x084850
-#define GT96100_ETH_INT_MASK             0x084858
-#define GT96100_ETH_1ST_RX_DESC_PTR0         0x084880
-#define GT96100_ETH_1ST_RX_DESC_PTR1         0x084884
-#define GT96100_ETH_1ST_RX_DESC_PTR2         0x084888
-#define GT96100_ETH_1ST_RX_DESC_PTR3         0x08488C
-#define GT96100_ETH_CURR_RX_DESC_PTR0        0x0848A0
-#define GT96100_ETH_CURR_RX_DESC_PTR1        0x0848A4
-#define GT96100_ETH_CURR_RX_DESC_PTR2        0x0848A8
-#define GT96100_ETH_CURR_RX_DESC_PTR3        0x0848AC
-#define GT96100_ETH_CURR_TX_DESC_PTR0        0x0848E0
-#define GT96100_ETH_CURR_TX_DESC_PTR1        0x0848E4
-#define GT96100_ETH_MIB_COUNT_BASE           0x085800
-
-/* SDMAs */
-#define GT96100_SDMA_GROUP_CONFIG           0x101AF0
-/* SDMA Group 0 */
-#define GT96100_SDMA_G0_CHAN0_CONFIG        0x000900
-#define GT96100_SDMA_G0_CHAN0_COMM          0x000908
-#define GT96100_SDMA_G0_CHAN0_RX_DESC_BASE      0x008900
-#define GT96100_SDMA_G0_CHAN0_CURR_RX_DESC_PTR  0x008910
-#define GT96100_SDMA_G0_CHAN0_TX_DESC_BASE      0x00C900
-#define GT96100_SDMA_G0_CHAN0_CURR_TX_DESC_PTR  0x00C910
-#define GT96100_SDMA_G0_CHAN0_1ST_TX_DESC_PTR   0x00C914
-#define GT96100_SDMA_G0_CHAN1_CONFIG        0x010900
-#define GT96100_SDMA_G0_CHAN1_COMM          0x010908
-#define GT96100_SDMA_G0_CHAN1_RX_DESC_BASE      0x018900
-#define GT96100_SDMA_G0_CHAN1_CURR_RX_DESC_PTR  0x018910
-#define GT96100_SDMA_G0_CHAN1_TX_DESC_BASE      0x01C900
-#define GT96100_SDMA_G0_CHAN1_CURR_TX_DESC_PTR  0x01C910
-#define GT96100_SDMA_G0_CHAN1_1ST_TX_DESC_PTR   0x01C914
-#define GT96100_SDMA_G0_CHAN2_CONFIG        0x020900
-#define GT96100_SDMA_G0_CHAN2_COMM          0x020908
-#define GT96100_SDMA_G0_CHAN2_RX_DESC_BASE      0x028900
-#define GT96100_SDMA_G0_CHAN2_CURR_RX_DESC_PTR  0x028910
-#define GT96100_SDMA_G0_CHAN2_TX_DESC_BASE      0x02C900
-#define GT96100_SDMA_G0_CHAN2_CURR_TX_DESC_PTR  0x02C910
-#define GT96100_SDMA_G0_CHAN2_1ST_TX_DESC_PTR   0x02C914
-#define GT96100_SDMA_G0_CHAN3_CONFIG        0x030900
-#define GT96100_SDMA_G0_CHAN3_COMM          0x030908
-#define GT96100_SDMA_G0_CHAN3_RX_DESC_BASE      0x038900
-#define GT96100_SDMA_G0_CHAN3_CURR_RX_DESC_PTR  0x038910
-#define GT96100_SDMA_G0_CHAN3_TX_DESC_BASE      0x03C900
-#define GT96100_SDMA_G0_CHAN3_CURR_TX_DESC_PTR  0x03C910
-#define GT96100_SDMA_G0_CHAN3_1ST_TX_DESC_PTR   0x03C914
-#define GT96100_SDMA_G0_CHAN4_CONFIG        0x040900
-#define GT96100_SDMA_G0_CHAN4_COMM          0x040908
-#define GT96100_SDMA_G0_CHAN4_RX_DESC_BASE      0x048900
-#define GT96100_SDMA_G0_CHAN4_CURR_RX_DESC_PTR  0x048910
-#define GT96100_SDMA_G0_CHAN4_TX_DESC_BASE      0x04C900
-#define GT96100_SDMA_G0_CHAN4_CURR_TX_DESC_PTR  0x04C910
-#define GT96100_SDMA_G0_CHAN4_1ST_TX_DESC_PTR   0x04C914
-#define GT96100_SDMA_G0_CHAN5_CONFIG        0x050900
-#define GT96100_SDMA_G0_CHAN5_COMM          0x050908
-#define GT96100_SDMA_G0_CHAN5_RX_DESC_BASE      0x058900
-#define GT96100_SDMA_G0_CHAN5_CURR_RX_DESC_PTR  0x058910
-#define GT96100_SDMA_G0_CHAN5_TX_DESC_BASE      0x05C900
-#define GT96100_SDMA_G0_CHAN5_CURR_TX_DESC_PTR  0x05C910
-#define GT96100_SDMA_G0_CHAN5_1ST_TX_DESC_PTR   0x05C914
-#define GT96100_SDMA_G0_CHAN6_CONFIG        0x060900
-#define GT96100_SDMA_G0_CHAN6_COMM          0x060908
-#define GT96100_SDMA_G0_CHAN6_RX_DESC_BASE      0x068900
-#define GT96100_SDMA_G0_CHAN6_CURR_RX_DESC_PTR  0x068910
-#define GT96100_SDMA_G0_CHAN6_TX_DESC_BASE      0x06C900
-#define GT96100_SDMA_G0_CHAN6_CURR_TX_DESC_PTR  0x06C910
-#define GT96100_SDMA_G0_CHAN6_1ST_TX_DESC_PTR   0x06C914
-#define GT96100_SDMA_G0_CHAN7_CONFIG        0x070900
-#define GT96100_SDMA_G0_CHAN7_COMM          0x070908
-#define GT96100_SDMA_G0_CHAN7_RX_DESC_BASE      0x078900
-#define GT96100_SDMA_G0_CHAN7_CURR_RX_DESC_PTR  0x078910
-#define GT96100_SDMA_G0_CHAN7_TX_DESC_BASE      0x07C900
-#define GT96100_SDMA_G0_CHAN7_CURR_TX_DESC_PTR  0x07C910
-#define GT96100_SDMA_G0_CHAN7_1ST_TX_DESC_PTR   0x07C914
-/* SDMA Group 1 */
-#define GT96100_SDMA_G1_CHAN0_CONFIG        0x100900
-#define GT96100_SDMA_G1_CHAN0_COMM          0x100908
-#define GT96100_SDMA_G1_CHAN0_RX_DESC_BASE      0x108900
-#define GT96100_SDMA_G1_CHAN0_CURR_RX_DESC_PTR  0x108910
-#define GT96100_SDMA_G1_CHAN0_TX_DESC_BASE      0x10C900
-#define GT96100_SDMA_G1_CHAN0_CURR_TX_DESC_PTR  0x10C910
-#define GT96100_SDMA_G1_CHAN0_1ST_TX_DESC_PTR   0x10C914
-#define GT96100_SDMA_G1_CHAN1_CONFIG        0x110900
-#define GT96100_SDMA_G1_CHAN1_COMM          0x110908
-#define GT96100_SDMA_G1_CHAN1_RX_DESC_BASE      0x118900
-#define GT96100_SDMA_G1_CHAN1_CURR_RX_DESC_PTR  0x118910
-#define GT96100_SDMA_G1_CHAN1_TX_DESC_BASE      0x11C900
-#define GT96100_SDMA_G1_CHAN1_CURR_TX_DESC_PTR  0x11C910
-#define GT96100_SDMA_G1_CHAN1_1ST_TX_DESC_PTR   0x11C914
-#define GT96100_SDMA_G1_CHAN2_CONFIG        0x120900
-#define GT96100_SDMA_G1_CHAN2_COMM          0x120908
-#define GT96100_SDMA_G1_CHAN2_RX_DESC_BASE      0x128900
-#define GT96100_SDMA_G1_CHAN2_CURR_RX_DESC_PTR  0x128910
-#define GT96100_SDMA_G1_CHAN2_TX_DESC_BASE      0x12C900
-#define GT96100_SDMA_G1_CHAN2_CURR_TX_DESC_PTR  0x12C910
-#define GT96100_SDMA_G1_CHAN2_1ST_TX_DESC_PTR   0x12C914
-#define GT96100_SDMA_G1_CHAN3_CONFIG        0x130900
-#define GT96100_SDMA_G1_CHAN3_COMM          0x130908
-#define GT96100_SDMA_G1_CHAN3_RX_DESC_BASE      0x138900
-#define GT96100_SDMA_G1_CHAN3_CURR_RX_DESC_PTR  0x138910
-#define GT96100_SDMA_G1_CHAN3_TX_DESC_BASE      0x13C900
-#define GT96100_SDMA_G1_CHAN3_CURR_TX_DESC_PTR  0x13C910
-#define GT96100_SDMA_G1_CHAN3_1ST_TX_DESC_PTR   0x13C914
-#define GT96100_SDMA_G1_CHAN4_CONFIG        0x140900
-#define GT96100_SDMA_G1_CHAN4_COMM          0x140908
-#define GT96100_SDMA_G1_CHAN4_RX_DESC_BASE      0x148900
-#define GT96100_SDMA_G1_CHAN4_CURR_RX_DESC_PTR  0x148910
-#define GT96100_SDMA_G1_CHAN4_TX_DESC_BASE      0x14C900
-#define GT96100_SDMA_G1_CHAN4_CURR_TX_DESC_PTR  0x14C910
-#define GT96100_SDMA_G1_CHAN4_1ST_TX_DESC_PTR   0x14C914
-#define GT96100_SDMA_G1_CHAN5_CONFIG        0x150900
-#define GT96100_SDMA_G1_CHAN5_COMM          0x150908
-#define GT96100_SDMA_G1_CHAN5_RX_DESC_BASE      0x158900
-#define GT96100_SDMA_G1_CHAN5_CURR_RX_DESC_PTR  0x158910
-#define GT96100_SDMA_G1_CHAN5_TX_DESC_BASE      0x15C900
-#define GT96100_SDMA_G1_CHAN5_CURR_TX_DESC_PTR  0x15C910
-#define GT96100_SDMA_G1_CHAN5_1ST_TX_DESC_PTR   0x15C914
-#define GT96100_SDMA_G1_CHAN6_CONFIG        0x160900
-#define GT96100_SDMA_G1_CHAN6_COMM          0x160908
-#define GT96100_SDMA_G1_CHAN6_RX_DESC_BASE      0x168900
-#define GT96100_SDMA_G1_CHAN6_CURR_RX_DESC_PTR  0x168910
-#define GT96100_SDMA_G1_CHAN6_TX_DESC_BASE      0x16C900
-#define GT96100_SDMA_G1_CHAN6_CURR_TX_DESC_PTR  0x16C910
-#define GT96100_SDMA_G1_CHAN6_1ST_TX_DESC_PTR   0x16C914
-#define GT96100_SDMA_G1_CHAN7_CONFIG        0x170900
-#define GT96100_SDMA_G1_CHAN7_COMM          0x170908
-#define GT96100_SDMA_G1_CHAN7_RX_DESC_BASE      0x178900
-#define GT96100_SDMA_G1_CHAN7_CURR_RX_DESC_PTR  0x178910
-#define GT96100_SDMA_G1_CHAN7_TX_DESC_BASE      0x17C900
-#define GT96100_SDMA_G1_CHAN7_CURR_TX_DESC_PTR  0x17C910
-#define GT96100_SDMA_G1_CHAN7_1ST_TX_DESC_PTR   0x17C914
-/*  MPSCs  */
-#define GT96100_MPSC0_MAIN_CONFIG_LOW   0x000A00
-#define GT96100_MPSC0_MAIN_CONFIG_HIGH  0x000A04
-#define GT96100_MPSC0_PROTOCOL_CONFIG   0x000A08
-#define GT96100_MPSC_CHAN0_REG1         0x000A0C
-#define GT96100_MPSC_CHAN0_REG2         0x000A10
-#define GT96100_MPSC_CHAN0_REG3         0x000A14
-#define GT96100_MPSC_CHAN0_REG4         0x000A18
-#define GT96100_MPSC_CHAN0_REG5         0x000A1C
-#define GT96100_MPSC_CHAN0_REG6         0x000A20
-#define GT96100_MPSC_CHAN0_REG7         0x000A24
-#define GT96100_MPSC_CHAN0_REG8         0x000A28
-#define GT96100_MPSC_CHAN0_REG9         0x000A2C
-#define GT96100_MPSC_CHAN0_REG10        0x000A30
-#define GT96100_MPSC_CHAN0_REG11        0x000A34
-#define GT96100_MPSC1_MAIN_CONFIG_LOW   0x008A00
-#define GT96100_MPSC1_MAIN_CONFIG_HIGH  0x008A04
-#define GT96100_MPSC1_PROTOCOL_CONFIG   0x008A08
-#define GT96100_MPSC_CHAN1_REG1         0x008A0C
-#define GT96100_MPSC_CHAN1_REG2         0x008A10
-#define GT96100_MPSC_CHAN1_REG3         0x008A14
-#define GT96100_MPSC_CHAN1_REG4         0x008A18
-#define GT96100_MPSC_CHAN1_REG5         0x008A1C
-#define GT96100_MPSC_CHAN1_REG6         0x008A20
-#define GT96100_MPSC_CHAN1_REG7         0x008A24
-#define GT96100_MPSC_CHAN1_REG8         0x008A28
-#define GT96100_MPSC_CHAN1_REG9         0x008A2C
-#define GT96100_MPSC_CHAN1_REG10        0x008A30
-#define GT96100_MPSC_CHAN1_REG11        0x008A34
-#define GT96100_MPSC2_MAIN_CONFIG_LOW   0x010A00
-#define GT96100_MPSC2_MAIN_CONFIG_HIGH  0x010A04
-#define GT96100_MPSC2_PROTOCOL_CONFIG   0x010A08
-#define GT96100_MPSC_CHAN2_REG1         0x010A0C
-#define GT96100_MPSC_CHAN2_REG2         0x010A10
-#define GT96100_MPSC_CHAN2_REG3         0x010A14
-#define GT96100_MPSC_CHAN2_REG4         0x010A18
-#define GT96100_MPSC_CHAN2_REG5         0x010A1C
-#define GT96100_MPSC_CHAN2_REG6         0x010A20
-#define GT96100_MPSC_CHAN2_REG7         0x010A24
-#define GT96100_MPSC_CHAN2_REG8         0x010A28
-#define GT96100_MPSC_CHAN2_REG9         0x010A2C
-#define GT96100_MPSC_CHAN2_REG10        0x010A30
-#define GT96100_MPSC_CHAN2_REG11        0x010A34
-#define GT96100_MPSC3_MAIN_CONFIG_LOW   0x018A00
-#define GT96100_MPSC3_MAIN_CONFIG_HIGH  0x018A04
-#define GT96100_MPSC3_PROTOCOL_CONFIG   0x018A08
-#define GT96100_MPSC_CHAN3_REG1         0x018A0C
-#define GT96100_MPSC_CHAN3_REG2         0x018A10
-#define GT96100_MPSC_CHAN3_REG3         0x018A14
-#define GT96100_MPSC_CHAN3_REG4         0x018A18
-#define GT96100_MPSC_CHAN3_REG5         0x018A1C
-#define GT96100_MPSC_CHAN3_REG6         0x018A20
-#define GT96100_MPSC_CHAN3_REG7         0x018A24
-#define GT96100_MPSC_CHAN3_REG8         0x018A28
-#define GT96100_MPSC_CHAN3_REG9         0x018A2C
-#define GT96100_MPSC_CHAN3_REG10        0x018A30
-#define GT96100_MPSC_CHAN3_REG11        0x018A34
-#define GT96100_MPSC4_MAIN_CONFIG_LOW   0x020A00
-#define GT96100_MPSC4_MAIN_CONFIG_HIGH  0x020A04
-#define GT96100_MPSC4_PROTOCOL_CONFIG   0x020A08
-#define GT96100_MPSC_CHAN4_REG1         0x020A0C
-#define GT96100_MPSC_CHAN4_REG2         0x020A10
-#define GT96100_MPSC_CHAN4_REG3         0x020A14
-#define GT96100_MPSC_CHAN4_REG4         0x020A18
-#define GT96100_MPSC_CHAN4_REG5         0x020A1C
-#define GT96100_MPSC_CHAN4_REG6         0x020A20
-#define GT96100_MPSC_CHAN4_REG7         0x020A24
-#define GT96100_MPSC_CHAN4_REG8         0x020A28
-#define GT96100_MPSC_CHAN4_REG9         0x020A2C
-#define GT96100_MPSC_CHAN4_REG10        0x020A30
-#define GT96100_MPSC_CHAN4_REG11        0x020A34
-#define GT96100_MPSC5_MAIN_CONFIG_LOW   0x028A00
-#define GT96100_MPSC5_MAIN_CONFIG_HIGH  0x028A04
-#define GT96100_MPSC5_PROTOCOL_CONFIG   0x028A08
-#define GT96100_MPSC_CHAN5_REG1         0x028A0C
-#define GT96100_MPSC_CHAN5_REG2         0x028A10
-#define GT96100_MPSC_CHAN5_REG3         0x028A14
-#define GT96100_MPSC_CHAN5_REG4         0x028A18
-#define GT96100_MPSC_CHAN5_REG5         0x028A1C
-#define GT96100_MPSC_CHAN5_REG6         0x028A20
-#define GT96100_MPSC_CHAN5_REG7         0x028A24
-#define GT96100_MPSC_CHAN5_REG8         0x028A28
-#define GT96100_MPSC_CHAN5_REG9         0x028A2C
-#define GT96100_MPSC_CHAN5_REG10        0x028A30
-#define GT96100_MPSC_CHAN5_REG11        0x028A34
-#define GT96100_MPSC6_MAIN_CONFIG_LOW   0x030A00
-#define GT96100_MPSC6_MAIN_CONFIG_HIGH  0x030A04
-#define GT96100_MPSC6_PROTOCOL_CONFIG   0x030A08
-#define GT96100_MPSC_CHAN6_REG1         0x030A0C
-#define GT96100_MPSC_CHAN6_REG2         0x030A10
-#define GT96100_MPSC_CHAN6_REG3         0x030A14
-#define GT96100_MPSC_CHAN6_REG4         0x030A18
-#define GT96100_MPSC_CHAN6_REG5         0x030A1C
-#define GT96100_MPSC_CHAN6_REG6         0x030A20
-#define GT96100_MPSC_CHAN6_REG7         0x030A24
-#define GT96100_MPSC_CHAN6_REG8         0x030A28
-#define GT96100_MPSC_CHAN6_REG9         0x030A2C
-#define GT96100_MPSC_CHAN6_REG10        0x030A30
-#define GT96100_MPSC_CHAN6_REG11        0x030A34
-#define GT96100_MPSC7_MAIN_CONFIG_LOW   0x038A00
-#define GT96100_MPSC7_MAIN_CONFIG_HIGH  0x038A04
-#define GT96100_MPSC7_PROTOCOL_CONFIG   0x038A08
-#define GT96100_MPSC_CHAN7_REG1         0x038A0C
-#define GT96100_MPSC_CHAN7_REG2         0x038A10
-#define GT96100_MPSC_CHAN7_REG3         0x038A14
-#define GT96100_MPSC_CHAN7_REG4         0x038A18
-#define GT96100_MPSC_CHAN7_REG5         0x038A1C
-#define GT96100_MPSC_CHAN7_REG6         0x038A20
-#define GT96100_MPSC_CHAN7_REG7         0x038A24
-#define GT96100_MPSC_CHAN7_REG8         0x038A28
-#define GT96100_MPSC_CHAN7_REG9         0x038A2C
-#define GT96100_MPSC_CHAN7_REG10        0x038A30
-#define GT96100_MPSC_CHAN7_REG11        0x038A34
-/*  FlexTDMs  */
-/* TDPR0 - Transmit Dual Port RAM. block size 0xff */
-#define GT96100_FXTDM0_TDPR0_BLK0_BASE  0x000B00
-#define GT96100_FXTDM0_TDPR0_BLK1_BASE  0x001B00
-#define GT96100_FXTDM0_TDPR0_BLK2_BASE  0x002B00
-#define GT96100_FXTDM0_TDPR0_BLK3_BASE  0x003B00
-/* RDPR0 - Receive Dual Port RAM. block size 0xff */
-#define GT96100_FXTDM0_RDPR0_BLK0_BASE  0x004B00
-#define GT96100_FXTDM0_RDPR0_BLK1_BASE  0x005B00
-#define GT96100_FXTDM0_RDPR0_BLK2_BASE  0x006B00
-#define GT96100_FXTDM0_RDPR0_BLK3_BASE  0x007B00
-#define GT96100_FXTDM0_TX_READ_PTR      0x008B00
-#define GT96100_FXTDM0_RX_READ_PTR      0x008B04
-#define GT96100_FXTDM0_CONFIG       0x008B08
-#define GT96100_FXTDM0_AUX_CHANA_TX 0x008B0C
-#define GT96100_FXTDM0_AUX_CHANA_RX 0x008B10
-#define GT96100_FXTDM0_AUX_CHANB_TX 0x008B14
-#define GT96100_FXTDM0_AUX_CHANB_RX 0x008B18
-#define GT96100_FXTDM1_TDPR1_BLK0_BASE  0x010B00
-#define GT96100_FXTDM1_TDPR1_BLK1_BASE  0x011B00
-#define GT96100_FXTDM1_TDPR1_BLK2_BASE  0x012B00
-#define GT96100_FXTDM1_TDPR1_BLK3_BASE  0x013B00
-#define GT96100_FXTDM1_RDPR1_BLK0_BASE  0x014B00
-#define GT96100_FXTDM1_RDPR1_BLK1_BASE  0x015B00
-#define GT96100_FXTDM1_RDPR1_BLK2_BASE  0x016B00
-#define GT96100_FXTDM1_RDPR1_BLK3_BASE  0x017B00
-#define GT96100_FXTDM1_TX_READ_PTR      0x018B00
-#define GT96100_FXTDM1_RX_READ_PTR      0x018B04
-#define GT96100_FXTDM1_CONFIG       0x018B08
-#define GT96100_FXTDM1_AUX_CHANA_TX 0x018B0C
-#define GT96100_FXTDM1_AUX_CHANA_RX 0x018B10
-#define GT96100_FLTDM1_AUX_CHANB_TX 0x018B14
-#define GT96100_FLTDM1_AUX_CHANB_RX 0x018B18
-#define GT96100_FLTDM2_TDPR2_BLK0_BASE  0x020B00
-#define GT96100_FLTDM2_TDPR2_BLK1_BASE  0x021B00
-#define GT96100_FLTDM2_TDPR2_BLK2_BASE  0x022B00
-#define GT96100_FLTDM2_TDPR2_BLK3_BASE  0x023B00
-#define GT96100_FLTDM2_RDPR2_BLK0_BASE  0x024B00
-#define GT96100_FLTDM2_RDPR2_BLK1_BASE  0x025B00
-#define GT96100_FLTDM2_RDPR2_BLK2_BASE  0x026B00
-#define GT96100_FLTDM2_RDPR2_BLK3_BASE  0x027B00
-#define GT96100_FLTDM2_TX_READ_PTR      0x028B00
-#define GT96100_FLTDM2_RX_READ_PTR      0x028B04
-#define GT96100_FLTDM2_CONFIG       0x028B08
-#define GT96100_FLTDM2_AUX_CHANA_TX 0x028B0C
-#define GT96100_FLTDM2_AUX_CHANA_RX 0x028B10
-#define GT96100_FLTDM2_AUX_CHANB_TX 0x028B14
-#define GT96100_FLTDM2_AUX_CHANB_RX 0x028B18
-#define GT96100_FLTDM3_TDPR3_BLK0_BASE  0x030B00
-#define GT96100_FLTDM3_TDPR3_BLK1_BASE  0x031B00
-#define GT96100_FLTDM3_TDPR3_BLK2_BASE  0x032B00
-#define GT96100_FLTDM3_TDPR3_BLK3_BASE  0x033B00
-#define GT96100_FXTDM3_RDPR3_BLK0_BASE  0x034B00
-#define GT96100_FXTDM3_RDPR3_BLK1_BASE  0x035B00
-#define GT96100_FXTDM3_RDPR3_BLK2_BASE  0x036B00
-#define GT96100_FXTDM3_RDPR3_BLK3_BASE  0x037B00
-#define GT96100_FXTDM3_TX_READ_PTR      0x038B00
-#define GT96100_FXTDM3_RX_READ_PTR      0x038B04
-#define GT96100_FXTDM3_CONFIG       0x038B08
-#define GT96100_FXTDM3_AUX_CHANA_TX 0x038B0C
-#define GT96100_FXTDM3_AUX_CHANA_RX 0x038B10
-#define GT96100_FXTDM3_AUX_CHANB_TX 0x038B14
-#define GT96100_FXTDM3_AUX_CHANB_RX 0x038B18
-/*  Baud Rate Generators  */
-#define GT96100_BRG0_CONFIG     0x102A00
-#define GT96100_BRG0_BAUD_TUNE  0x102A04
-#define GT96100_BRG1_CONFIG     0x102A08
-#define GT96100_BRG1_BAUD_TUNE  0x102A0C
-#define GT96100_BRG2_CONFIG     0x102A10
-#define GT96100_BRG2_BAUD_TUNE  0x102A14
-#define GT96100_BRG3_CONFIG     0x102A18
-#define GT96100_BRG3_BAUD_TUNE  0x102A1C
-#define GT96100_BRG4_CONFIG     0x102A20
-#define GT96100_BRG4_BAUD_TUNE  0x102A24
-#define GT96100_BRG5_CONFIG     0x102A28
-#define GT96100_BRG5_BAUD_TUNE  0x102A2C
-#define GT96100_BRG6_CONFIG     0x102A30
-#define GT96100_BRG6_BAUD_TUNE  0x102A34
-#define GT96100_BRG7_CONFIG     0x102A38
-#define GT96100_BRG7_BAUD_TUNE  0x102A3C
-/*  Routing Registers  */
-#define GT96100_ROUTE_MAIN      0x101A00
-#define GT96100_ROUTE_RX_CLOCK  0x101A10
-#define GT96100_ROUTE_TX_CLOCK  0x101A20
-/*  General Purpose Ports  */
-#define GT96100_GPP_CONFIG0     0x100A00
-#define GT96100_GPP_CONFIG1     0x100A04
-#define GT96100_GPP_CONFIG2     0x100A08
-#define GT96100_GPP_CONFIG3     0x100A0C
-#define GT96100_GPP_IO0         0x100A20
-#define GT96100_GPP_IO1         0x100A24
-#define GT96100_GPP_IO2         0x100A28
-#define GT96100_GPP_IO3         0x100A2C
-#define GT96100_GPP_DATA0       0x100A40
-#define GT96100_GPP_DATA1       0x100A44
-#define GT96100_GPP_DATA2       0x100A48
-#define GT96100_GPP_DATA3       0x100A4C
-#define GT96100_GPP_LEVEL0      0x100A60
-#define GT96100_GPP_LEVEL1      0x100A64
-#define GT96100_GPP_LEVEL2      0x100A68
-#define GT96100_GPP_LEVEL3      0x100A6C
-/*  Watchdog  */
-#define GT96100_WD_CONFIG   0x101A80
-#define GT96100_WD_VALUE    0x101A84
-/* Communication Unit Arbiter  */
-#define GT96100_COMM_UNIT_ARBTR_CONFIG 0x101AC0
-/*  PCI Arbiters  */
-#define GT96100_PCI0_ARBTR_CONFIG 0x101AE0
-#define GT96100_PCI1_ARBTR_CONFIG 0x101AE4
-/* CIU Arbiter */
-#define GT96100_CIU_ARBITER_CONFIG 0x101AC0
-/* Interrupt Controller */
-#define GT96100_MAIN_CAUSE     0x000C18
-#define GT96100_INT0_MAIN_MASK 0x000C1C
-#define GT96100_INT1_MAIN_MASK 0x000C24
-#define GT96100_HIGH_CAUSE     0x000C98
-#define GT96100_INT0_HIGH_MASK 0x000C9C
-#define GT96100_INT1_HIGH_MASK 0x000CA4
-#define GT96100_INT0_SELECT    0x000C70
-#define GT96100_INT1_SELECT    0x000C74
-#define GT96100_SERIAL_CAUSE   0x103A00
-#define GT96100_SERINT0_MASK   0x103A80
-#define GT96100_SERINT1_MASK   0x103A88
-
-#endif /*  _GT96100_H */
diff --git a/include/asm-mips/mach-ev96100/mach-gt64120.h b/include/asm-mips/mach-ev96100/mach-gt64120.h
deleted file mode 100644
index 0ef1e6c25acf..000000000000
--- a/include/asm-mips/mach-ev96100/mach-gt64120.h
+++ /dev/null
@@ -1,46 +0,0 @@
-/*
- *  This is a direct copy of the ev96100.h file, with a global
- * search and replace.  The numbers are the same.
- *
- *  The reason I'm duplicating this is so that the 64120/96100
- * defines won't be confusing in the source code.
- */
-#ifndef _ASM_GT64120_EV96100_GT64120_DEP_H
-#define _ASM_GT64120_EV96100_GT64120_DEP_H
-
-/*
- *   GT96100 config space base address
- */
-#define GT64120_BASE	(KSEG1ADDR(0x14000000))
-
-/*
- *   PCI Bus allocation
- *
- *   (Guessing ...)
- */
-#define GT_PCI_MEM_BASE	0x12000000UL
-#define GT_PCI_MEM_SIZE	0x02000000UL
-#define GT_PCI_IO_BASE	0x10000000UL
-#define GT_PCI_IO_SIZE	0x02000000UL
-#define GT_ISA_IO_BASE	PCI_IO_BASE
-
-/*
- *   Duart I/O ports.
- */
-#define EV96100_COM1_BASE_ADDR	(0xBD000000 + 0x20)
-#define EV96100_COM2_BASE_ADDR	(0xBD000000 + 0x00)
-
-
-/*
- *   EV96100 interrupt controller register base.
- */
-#define EV96100_ICTRL_REGS_BASE	(KSEG1ADDR(0x1f000000))
-
-/*
- *   EV96100 UART register base.
- */
-#define EV96100_UART0_REGS_BASE	EV96100_COM1_BASE_ADDR
-#define EV96100_UART1_REGS_BASE	EV96100_COM2_BASE_ADDR
-#define EV96100_BASE_BAUD	( 3686400 / 16 )
-
-#endif /* _ASM_GT64120_EV96100_GT64120_DEP_H */
diff --git a/include/asm-mips/serial.h b/include/asm-mips/serial.h
index 584bd9c0ab2e..035637c67e7c 100644
--- a/include/asm-mips/serial.h
+++ b/include/asm-mips/serial.h
@@ -52,9 +52,9 @@
 #endif
 
 /*
- * Both Galileo boards have the same UART mappings.
+ * Galileo EV64120 evaluation board
  */
-#if defined (CONFIG_MIPS_EV96100) || defined (CONFIG_MIPS_EV64120)
+#ifdef CONFIG_MIPS_EV64120
 #include <asm/galileo-boards/ev96100.h>
 #include <asm/galileo-boards/ev96100int.h>
 #define EV96100_SERIAL_PORT_DEFNS                                  \
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 6a1e09834559..5c1c698a92ac 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1482,9 +1482,6 @@
 #define PCI_DEVICE_ID_MARVELL_GT64260	0x6430
 #define PCI_DEVICE_ID_MARVELL_MV64360	0x6460
 #define PCI_DEVICE_ID_MARVELL_MV64460	0x6480
-#define PCI_DEVICE_ID_MARVELL_GT96100	0x9652
-#define PCI_DEVICE_ID_MARVELL_GT96100A	0x9653
-
 
 #define PCI_VENDOR_ID_V3		0x11b0
 #define PCI_DEVICE_ID_V3_V960		0x0001
-- 
cgit v1.2.1


From 63415dbb54fb929a6ea597f2f6b885d570225d90 Mon Sep 17 00:00:00 2001
From: Richard Sandiford <richard@codesourcery.com>
Date: Sun, 17 Sep 2006 20:30:46 +0100
Subject: [MIPS] The o32 fstatat syscall behaves differently on 32 and 64 bit
 kernels

While working on a glibc patch to support the fstatat() functions[1],
I noticed that the o32 implementation behaves differently on 32-bit and
64-bit kernels; the former provides a stat64 while the latter provides
a plain (o32) stat.  I think the former is what's intended, as there is
no separate fstatat64.  It's also what x86 does.

I think this is just a case of a compat too far.

[1] I've seen Khem's patch, but I don't think it's right.

Signed-off-by: Richard Sandiford <richard@codesourcery.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/scall64-o32.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 2ac01412f747..288ee4ac4dbb 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -498,7 +498,7 @@ sys_call_table:
 	PTR	sys_mknodat			/* 4290 */
 	PTR	sys_fchownat
 	PTR	compat_sys_futimesat
-	PTR	compat_sys_newfstatat
+	PTR	sys_newfstatat
 	PTR	sys_unlinkat
 	PTR	sys_renameat			/* 4295 */
 	PTR	sys_linkat
-- 
cgit v1.2.1


From ddb1199c4cd74a02d70f4c297373400893d70aa9 Mon Sep 17 00:00:00 2001
From: Richard Sandiford <richard@codesourcery.com>
Date: Sun, 17 Sep 2006 20:38:39 +0100
Subject: [MIPS] fstatat syscall names

MIPS is the only port to call its fstatat()-related syscalls
"__NR_fstatat".  Now I can see why that might be seen as every
other port being wrong, but I think for o32, it is at best confusing.
__NR_fstat provides a plain (32-bit) stat while __NR_fstatat provides a
64-bit stat.  Changing the name to __NR_fstatat64 would make things more
explicit, match x86, and make the glibc port slightly easier.

The current name is more appropriate for n32 and n64, but it would be
appropriate for other 64-bit targets too, and those targets have chosen
to call it __NR_newfstatat instead.  Using the same name for MIPS would
again be more consistent and make the glibc port slightly easier.

I'm not wedded to this idea if the current names are preferred,
but FWIW...

Signed-off-by: Richard Sandiford <richard@codesourcery.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/unistd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index 558e3cba4f92..c39142920fe6 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -313,7 +313,7 @@
 #define __NR_mknodat			(__NR_Linux + 290)
 #define __NR_fchownat			(__NR_Linux + 291)
 #define __NR_futimesat			(__NR_Linux + 292)
-#define __NR_fstatat			(__NR_Linux + 293)
+#define __NR_fstatat64			(__NR_Linux + 293)
 #define __NR_unlinkat			(__NR_Linux + 294)
 #define __NR_renameat			(__NR_Linux + 295)
 #define __NR_linkat			(__NR_Linux + 296)
@@ -600,7 +600,7 @@
 #define __NR_mknodat			(__NR_Linux + 249)
 #define __NR_fchownat			(__NR_Linux + 250)
 #define __NR_futimesat			(__NR_Linux + 251)
-#define __NR_fstatat			(__NR_Linux + 252)
+#define __NR_newfstatat			(__NR_Linux + 252)
 #define __NR_unlinkat			(__NR_Linux + 253)
 #define __NR_renameat			(__NR_Linux + 254)
 #define __NR_linkat			(__NR_Linux + 255)
@@ -891,7 +891,7 @@
 #define __NR_mknodat			(__NR_Linux + 253)
 #define __NR_fchownat			(__NR_Linux + 254)
 #define __NR_futimesat			(__NR_Linux + 255)
-#define __NR_fstatat			(__NR_Linux + 256)
+#define __NR_newfstatat			(__NR_Linux + 256)
 #define __NR_unlinkat			(__NR_Linux + 257)
 #define __NR_renameat			(__NR_Linux + 258)
 #define __NR_linkat			(__NR_Linux + 259)
-- 
cgit v1.2.1


From e584ade1a6679bfb417620f15951796bed9805d9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 23 Sep 2006 18:08:36 +0100
Subject: [MIPS] Have headers_install install <asm/cachectl.h> and
 <asm/sysmips.h>.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/Kbuild | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-mips/Kbuild b/include/asm-mips/Kbuild
index 01ea3e045652..7897f05e3165 100644
--- a/include/asm-mips/Kbuild
+++ b/include/asm-mips/Kbuild
@@ -1,3 +1,3 @@
 include include/asm-generic/Kbuild.asm
 
-header-y += sgidefs.h
+header-y += cachectl.h sgidefs.h sysmips.h
-- 
cgit v1.2.1


From 36396f3c36b04f79438f87a0fccfa76aa3de6af9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 25 Sep 2006 15:49:49 +0100
Subject: [MIPS] s/__ASSEMBLER__/__ASSEMBLY__/ for clarity sake.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/hazards.h            | 2 +-
 include/asm-mips/mach-excite/excite.h | 2 +-
 include/asm-mips/sibyte/sb1250_defs.h | 6 +++---
 include/asm-mips/sibyte/sb1250_scd.h  | 2 +-
 4 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/include/asm-mips/hazards.h b/include/asm-mips/hazards.h
index bf0b80404946..0fe02945feba 100644
--- a/include/asm-mips/hazards.h
+++ b/include/asm-mips/hazards.h
@@ -11,7 +11,7 @@
 #define _ASM_HAZARDS_H
 
 
-#ifdef __ASSEMBLER__
+#ifdef __ASSEMBLY__
 #define ASMMACRO(name, code...) .macro name; code; .endm
 #else
 
diff --git a/include/asm-mips/mach-excite/excite.h b/include/asm-mips/mach-excite/excite.h
index 130bd4b8edce..4c29ba44992c 100644
--- a/include/asm-mips/mach-excite/excite.h
+++ b/include/asm-mips/mach-excite/excite.h
@@ -7,7 +7,7 @@
 
 #define EXCITE_CPU_EXT_CLOCK 100000000
 
-#if !defined(__ASSEMBLER__)
+#if !defined(__ASSEMBLY__)
 void __init excite_kgdb_init(void);
 void excite_procfs_init(void);
 extern unsigned long memsize;
diff --git a/include/asm-mips/sibyte/sb1250_defs.h b/include/asm-mips/sibyte/sb1250_defs.h
index 335dbaf1d831..a885491217c1 100644
--- a/include/asm-mips/sibyte/sb1250_defs.h
+++ b/include/asm-mips/sibyte/sb1250_defs.h
@@ -212,7 +212,7 @@
  * Note: you'll need to define uint32_t and uint64_t in your headers.
  */
 
-#if !defined(__ASSEMBLER__)
+#if !defined(__ASSEMBLY__)
 #define _SB_MAKE64(x) ((uint64_t)(x))
 #define _SB_MAKE32(x) ((uint32_t)(x))
 #else
@@ -251,9 +251,9 @@
  */
 
 
-#if defined(__mips64) && !defined(__ASSEMBLER__)
+#if defined(__mips64) && !defined(__ASSEMBLY__)
 #define SBWRITECSR(csr,val) *((volatile uint64_t *) PHYS_TO_K1(csr)) = (val)
 #define SBREADCSR(csr) (*((volatile uint64_t *) PHYS_TO_K1(csr)))
-#endif /* __ASSEMBLER__ */
+#endif /* __ASSEMBLY__ */
 
 #endif
diff --git a/include/asm-mips/sibyte/sb1250_scd.h b/include/asm-mips/sibyte/sb1250_scd.h
index f4178bdcfcb0..7ed0bb611e56 100644
--- a/include/asm-mips/sibyte/sb1250_scd.h
+++ b/include/asm-mips/sibyte/sb1250_scd.h
@@ -149,7 +149,7 @@
  * (For the assembler version, sysrev and dest may be the same register.
  * Also, it clobbers AT.)
  */
-#ifdef __ASSEMBLER__
+#ifdef __ASSEMBLY__
 #define SYS_SOC_TYPE(dest, sysrev)					\
 	.set push ;							\
 	.set reorder ;							\
-- 
cgit v1.2.1


From f83b854a1d2d28bc1ed86fec8a80940b59f8d932 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 26 Sep 2006 03:07:22 +0100
Subject: [MIPS] Enable tmpfs for anything that possibly runs a full
 distribution.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/atlas_defconfig      | 2 +-
 arch/mips/configs/cobalt_defconfig     | 2 +-
 arch/mips/configs/ip22_defconfig       | 2 +-
 arch/mips/configs/ip27_defconfig       | 2 +-
 arch/mips/configs/jaguar-atx_defconfig | 2 +-
 arch/mips/configs/lasat200_defconfig   | 2 +-
 arch/mips/configs/malta_defconfig      | 2 +-
 arch/mips/configs/ocelot_c_defconfig   | 2 +-
 arch/mips/configs/ocelot_defconfig     | 2 +-
 arch/mips/configs/ocelot_g_defconfig   | 2 +-
 arch/mips/configs/qemu_defconfig       | 2 +-
 arch/mips/configs/rm200_defconfig      | 2 +-
 arch/mips/defconfig                    | 4 ++--
 13 files changed, 14 insertions(+), 14 deletions(-)

diff --git a/arch/mips/configs/atlas_defconfig b/arch/mips/configs/atlas_defconfig
index 2774ba9606b7..d3705284de39 100644
--- a/arch/mips/configs/atlas_defconfig
+++ b/arch/mips/configs/atlas_defconfig
@@ -1192,7 +1192,7 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/cobalt_defconfig b/arch/mips/configs/cobalt_defconfig
index 78db2a8a711b..4baf2ff1128a 100644
--- a/arch/mips/configs/cobalt_defconfig
+++ b/arch/mips/configs/cobalt_defconfig
@@ -827,7 +827,7 @@ CONFIG_FUSE_FS=y
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/ip22_defconfig b/arch/mips/configs/ip22_defconfig
index 9f22d13e729d..cc9b24eda9e8 100644
--- a/arch/mips/configs/ip22_defconfig
+++ b/arch/mips/configs/ip22_defconfig
@@ -1012,7 +1012,7 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index 414a649dff8a..50092ba8aa71 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -899,7 +899,7 @@ CONFIG_FUSE_FS=m
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/jaguar-atx_defconfig b/arch/mips/configs/jaguar-atx_defconfig
index 9b529857f802..9f1e3048d623 100644
--- a/arch/mips/configs/jaguar-atx_defconfig
+++ b/arch/mips/configs/jaguar-atx_defconfig
@@ -730,7 +730,7 @@ CONFIG_FUSE_FS=m
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 
diff --git a/arch/mips/configs/lasat200_defconfig b/arch/mips/configs/lasat200_defconfig
index 0354bfa18b39..320b8cdd6e58 100644
--- a/arch/mips/configs/lasat200_defconfig
+++ b/arch/mips/configs/lasat200_defconfig
@@ -904,7 +904,7 @@ CONFIG_FUSE_FS=m
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/malta_defconfig b/arch/mips/configs/malta_defconfig
index b21ca470273a..0ba1ef5048fb 100644
--- a/arch/mips/configs/malta_defconfig
+++ b/arch/mips/configs/malta_defconfig
@@ -1229,7 +1229,7 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/ocelot_c_defconfig b/arch/mips/configs/ocelot_c_defconfig
index ce7c26475328..a7ac2b0a8273 100644
--- a/arch/mips/configs/ocelot_c_defconfig
+++ b/arch/mips/configs/ocelot_c_defconfig
@@ -773,7 +773,7 @@ CONFIG_FUSE_FS=y
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/ocelot_defconfig b/arch/mips/configs/ocelot_defconfig
index 3da2a43d1608..853e7bba5122 100644
--- a/arch/mips/configs/ocelot_defconfig
+++ b/arch/mips/configs/ocelot_defconfig
@@ -722,7 +722,7 @@ CONFIG_FUSE_FS=y
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/ocelot_g_defconfig b/arch/mips/configs/ocelot_g_defconfig
index 3a3e5ff4a3ef..8524efa23a49 100644
--- a/arch/mips/configs/ocelot_g_defconfig
+++ b/arch/mips/configs/ocelot_g_defconfig
@@ -776,7 +776,7 @@ CONFIG_FUSE_FS=y
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/configs/qemu_defconfig b/arch/mips/configs/qemu_defconfig
index daa40c8ff694..f5f799e93707 100644
--- a/arch/mips/configs/qemu_defconfig
+++ b/arch/mips/configs/qemu_defconfig
@@ -686,7 +686,7 @@ CONFIG_FUSE_FS=y
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 
diff --git a/arch/mips/configs/rm200_defconfig b/arch/mips/configs/rm200_defconfig
index f26b338333ac..4fee90b2b100 100644
--- a/arch/mips/configs/rm200_defconfig
+++ b/arch/mips/configs/rm200_defconfig
@@ -1441,7 +1441,7 @@ CONFIG_NTFS_FS=m
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
diff --git a/arch/mips/defconfig b/arch/mips/defconfig
index d620c463a78b..cc9b24eda9e8 100644
--- a/arch/mips/defconfig
+++ b/arch/mips/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
 # Linux kernel version: 2.6.18-rc1
-# Thu Jul  6 09:49:33 2006
+# Thu Jul  6 10:04:10 2006
 #
 CONFIG_MIPS=y
 
@@ -1012,7 +1012,7 @@ CONFIG_FAT_DEFAULT_IOCHARSET="iso8859-1"
 CONFIG_PROC_FS=y
 CONFIG_PROC_KCORE=y
 CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
+CONFIG_TMPFS=y
 # CONFIG_HUGETLB_PAGE is not set
 CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
-- 
cgit v1.2.1


From 1fd6909802b837ed5510603846c0ce5938d296a1 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 18 Aug 2006 16:18:07 +0200
Subject: [MIPS] unwind_stack(): return ra if an exception occured at the first
 instruction

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 951bf9ca3ce9..e7b0b385fb2b 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -465,8 +465,11 @@ unsigned long unwind_stack(struct task_struct *task, unsigned long *sp,
 
 	if (!kallsyms_lookup(pc, &size, &ofs, &modname, namebuf))
 		return 0;
-	if (ofs == 0)
-		return 0;
+	/*
+	 * Return ra if an exception occured at the first instruction
+	 */
+	if (unlikely(ofs == 0))
+		return ra;
 
 	info.func = (void *)(pc - ofs);
 	info.func_size = ofs;	/* analyze from start to ofs */
-- 
cgit v1.2.1


From 29b376ff10aaea69ee4d93b70d0fbb2ebfd80f4e Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 18 Aug 2006 16:18:08 +0200
Subject: [MIPS] get_frame_info(): null function size means size is unknown

This patch adds 2 sanity checks.

The first one test that the start address of the function to analyze has been
set by the caller. If not return an error since nothing usefull can be done
without.

The second one checks that the function's size has been set. A null size can
happen if CONFIG_KALLSYMS is not set and it means that we don't know the size
of the function to analyze. In this case, we make it equal to 128 instructions
by default.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index e7b0b385fb2b..b160ea30de0f 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -311,12 +311,19 @@ static inline int is_sp_move_ins(union mips_instruction *ip)
 static int get_frame_info(struct mips_frame_info *info)
 {
 	union mips_instruction *ip = info->func;
-	int i, max_insns =
-		min(128UL, info->func_size / sizeof(union mips_instruction));
+	unsigned max_insns = info->func_size / sizeof(union mips_instruction);
+	unsigned i;
 
 	info->pc_offset = -1;
 	info->frame_size = 0;
 
+	if (!ip)
+		goto err;
+
+	if (max_insns == 0)
+		max_insns = 128U;	/* unknown function size */
+	max_insns = min(128U, max_insns);
+
 	for (i = 0; i < max_insns; i++, ip++) {
 
 		if (is_jal_jalr_jr_ins(ip))
@@ -337,6 +344,7 @@ static int get_frame_info(struct mips_frame_info *info)
 	if (info->pc_offset < 0) /* leaf */
 		return 1;
 	/* prologue seems boggus... */
+err:
 	return -1;
 }
 
-- 
cgit v1.2.1


From b5943182592ba256639a569c7d5305cf60360733 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 18 Aug 2006 16:18:09 +0200
Subject: [MIPS] get_wchan(): remove uses of mfinfo[64]

This array was used to 'cache' some frame info about scheduler
functions to speed up get_wchan(). This array was 1Ko size and
was only used when CONFIG_KALLSYMS was set but declared for all
configs.

Rather than make the array statement conditional, this patches
removes this array and its uses. Indeed the common case doesn't
seem to use this array and get_wchan() is not a critical path
anyways.

It results in a smaller bss and a smaller/cleaner code:

   text    data     bss     dec     hex filename
2543808  254148  139296 2937252  2cd1a4 vmlinux-new-get-wchan
2544080  254148  143392 2941620  2ce2b4 vmlinux~old

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 132 +++++++++++++++++----------------------------
 1 file changed, 50 insertions(+), 82 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index b160ea30de0f..2613a0dd4b82 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -273,13 +273,15 @@ long kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
 	return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, &regs, 0, NULL, NULL);
 }
 
-static struct mips_frame_info {
-	void *func;
-	unsigned long func_size;
-	int frame_size;
-	int pc_offset;
-} *schedule_frame, mfinfo[64];
-static int mfinfo_num;
+/*
+ *
+ */
+struct mips_frame_info {
+	void		*func;
+	unsigned long	func_size;
+	int		frame_size;
+	int		pc_offset;
+};
 
 static inline int is_ra_save_ins(union mips_instruction *ip)
 {
@@ -348,45 +350,30 @@ err:
 	return -1;
 }
 
+static struct mips_frame_info schedule_mfi __read_mostly;
+
 static int __init frame_info_init(void)
 {
-	int i;
+	unsigned long size = 0;
 #ifdef CONFIG_KALLSYMS
+	unsigned long ofs;
 	char *modname;
 	char namebuf[KSYM_NAME_LEN + 1];
-	unsigned long start, size, ofs;
-	extern char __sched_text_start[], __sched_text_end[];
-	extern char __lock_text_start[], __lock_text_end[];
-
-	start = (unsigned long)__sched_text_start;
-	for (i = 0; i < ARRAY_SIZE(mfinfo); i++) {
-		if (start == (unsigned long)schedule)
-			schedule_frame = &mfinfo[i];
-		if (!kallsyms_lookup(start, &size, &ofs, &modname, namebuf))
-			break;
-		mfinfo[i].func = (void *)(start + ofs);
-		mfinfo[i].func_size = size;
-		start += size - ofs;
-		if (start >= (unsigned long)__lock_text_end)
-			break;
-		if (start == (unsigned long)__sched_text_end)
-			start = (unsigned long)__lock_text_start;
-	}
-#else
-	mfinfo[0].func = schedule;
-	schedule_frame = &mfinfo[0];
+
+	kallsyms_lookup((unsigned long)schedule, &size, &ofs, &modname, namebuf);
 #endif
-	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++)
-		get_frame_info(mfinfo + i);
+	schedule_mfi.func = schedule;
+	schedule_mfi.func_size = size;
+
+	get_frame_info(&schedule_mfi);
 
 	/*
 	 * Without schedule() frame info, result given by
 	 * thread_saved_pc() and get_wchan() are not reliable.
 	 */
-	if (schedule_frame->pc_offset < 0)
+	if (schedule_mfi.pc_offset < 0)
 		printk("Can't analyze schedule() prologue at %p\n", schedule);
 
-	mfinfo_num = i;
 	return 0;
 }
 
@@ -402,58 +389,11 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	/* New born processes are a special case */
 	if (t->reg31 == (unsigned long) ret_from_fork)
 		return t->reg31;
-
-	if (!schedule_frame || schedule_frame->pc_offset < 0)
+	if (schedule_mfi.pc_offset < 0)
 		return 0;
-	return ((unsigned long *)t->reg29)[schedule_frame->pc_offset];
+	return ((unsigned long *)t->reg29)[schedule_mfi.pc_offset];
 }
 
-/* get_wchan - a maintenance nightmare^W^Wpain in the ass ...  */
-unsigned long get_wchan(struct task_struct *p)
-{
-	unsigned long stack_page;
-	unsigned long pc;
-#ifdef CONFIG_KALLSYMS
-	unsigned long frame;
-#endif
-
-	if (!p || p == current || p->state == TASK_RUNNING)
-		return 0;
-
-	stack_page = (unsigned long)task_stack_page(p);
-	if (!stack_page || !mfinfo_num)
-		return 0;
-
-	pc = thread_saved_pc(p);
-#ifdef CONFIG_KALLSYMS
-	if (!in_sched_functions(pc))
-		return pc;
-
-	frame = p->thread.reg29 + schedule_frame->frame_size;
-	do {
-		int i;
-
-		if (frame < stack_page || frame > stack_page + THREAD_SIZE - 32)
-			return 0;
-
-		for (i = mfinfo_num - 1; i >= 0; i--) {
-			if (pc >= (unsigned long) mfinfo[i].func)
-				break;
-		}
-		if (i < 0)
-			break;
-
-		if (mfinfo[i].pc_offset < 0)
-			break;
-		pc = ((unsigned long *)frame)[mfinfo[i].pc_offset];
-		if (!mfinfo[i].frame_size)
-			break;
-		frame += mfinfo[i].frame_size;
-	} while (in_sched_functions(pc));
-#endif
-
-	return pc;
-}
 
 #ifdef CONFIG_KALLSYMS
 /* used by show_backtrace() */
@@ -504,3 +444,31 @@ unsigned long unwind_stack(struct task_struct *task, unsigned long *sp,
 	return __kernel_text_address(pc) ? pc : 0;
 }
 #endif
+
+/*
+ * get_wchan - a maintenance nightmare^W^Wpain in the ass ...
+ */
+unsigned long get_wchan(struct task_struct *task)
+{
+	unsigned long pc = 0;
+#ifdef CONFIG_KALLSYMS
+	unsigned long sp;
+#endif
+
+	if (!task || task == current || task->state == TASK_RUNNING)
+		goto out;
+	if (!task_stack_page(task))
+		goto out;
+
+	pc = thread_saved_pc(task);
+
+#ifdef CONFIG_KALLSYMS
+	sp = task->thread.reg29 + schedule_mfi.frame_size;
+
+	while (in_sched_functions(pc))
+		pc = unwind_stack(task, &sp, pc, 0);
+#endif
+
+out:
+	return pc;
+}
-- 
cgit v1.2.1


From b6f1f0dea1469e0c956eb89399916d60dd2a3808 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:48 +0200
Subject: [MIPS] setup.c: cleanup bootmem_init()

This function although doing simple thing is hard to follow. It's
mainly due to:

    - a lot of #ifdef
    - bad local names
    - redundant tests

So this patch try to address these issues. It also do not use
max_pfn global which is marked as an unused exported symbol.

As a bonus side, it's now really easy to see what part of the
code is for no-numa system.

There's also no point to make this function inline.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 135 ++++++++++++++++++++---------------------------
 1 file changed, 57 insertions(+), 78 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 8c2b596a136f..bcbd926e58f7 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -96,6 +96,12 @@ void __init add_memory_region(phys_t start, phys_t size, long type)
 	int x = boot_mem_map.nr_map;
 	struct boot_mem_map_entry *prev = boot_mem_map.map + x - 1;
 
+	/* Sanity check */
+	if (start + size < start) {
+		printk("Trying to add an invalid memory region, skipped\n");
+		return;
+	}
+
 	/*
 	 * Try to merge with previous entry if any.  This is far less than
 	 * perfect but is sufficient for most real world cases.
@@ -257,15 +263,16 @@ static inline int parse_rd_cmdline(unsigned long* rd_start, unsigned long* rd_en
 	return 0;
 }
 
-#define MAXMEM		HIGHMEM_START
-#define MAXMEM_PFN	PFN_DOWN(MAXMEM)
-
-static inline void bootmem_init(void)
+/*
+ * Initialize the bootmem allocator. It also setup initrd related data
+ * if needed.
+ */
+static void __init bootmem_init(void)
 {
-	unsigned long start_pfn;
 	unsigned long reserved_end = (unsigned long)&_end;
 #ifndef CONFIG_SGI_IP27
-	unsigned long first_usable_pfn;
+	unsigned long highest = 0;
+	unsigned long mapstart = -1UL;
 	unsigned long bootmap_size;
 	int i;
 #endif
@@ -281,7 +288,7 @@ static inline void bootmem_init(void)
 		unsigned long tmp;
 		u32 *initrd_header;
 
-		tmp = ((reserved_end + PAGE_SIZE-1) & PAGE_MASK) - sizeof(u32) * 2;
+		tmp = PAGE_ALIGN(reserved_end) - sizeof(u32) * 2;
 		if (tmp < reserved_end)
 			tmp += PAGE_SIZE;
 		initrd_header = (u32 *)tmp;
@@ -294,16 +301,15 @@ static inline void bootmem_init(void)
 	}
 #endif	/* CONFIG_BLK_DEV_INITRD */
 
+#ifndef CONFIG_SGI_IP27
 	/*
-	 * Partially used pages are not usable - thus
-	 * we are rounding upwards.
+	 * reserved_end is now a pfn
 	 */
-	start_pfn = PFN_UP(CPHYSADDR(reserved_end));
+	reserved_end = PFN_UP(CPHYSADDR(reserved_end));
 
-#ifndef CONFIG_SGI_IP27
-	/* Find the highest page frame number we have available.  */
-	max_pfn = 0;
-	first_usable_pfn = -1UL;
+	/*
+	 * Find the highest page frame number we have available.
+	 */
 	for (i = 0; i < boot_mem_map.nr_map; i++) {
 		unsigned long start, end;
 
@@ -312,56 +318,38 @@ static inline void bootmem_init(void)
 
 		start = PFN_UP(boot_mem_map.map[i].addr);
 		end = PFN_DOWN(boot_mem_map.map[i].addr
-		      + boot_mem_map.map[i].size);
+				+ boot_mem_map.map[i].size);
 
-		if (start >= end)
+		if (end > highest)
+			highest = end;
+		if (end <= reserved_end)
 			continue;
-		if (end > max_pfn)
-			max_pfn = end;
-		if (start < first_usable_pfn) {
-			if (start > start_pfn) {
-				first_usable_pfn = start;
-			} else if (end > start_pfn) {
-				first_usable_pfn = start_pfn;
-			}
-		}
+		if (start >= mapstart)
+			continue;
+		mapstart = max(reserved_end, start);
 	}
 
 	/*
 	 * Determine low and high memory ranges
 	 */
-	max_low_pfn = max_pfn;
-	if (max_low_pfn > MAXMEM_PFN) {
-		max_low_pfn = MAXMEM_PFN;
-#ifndef CONFIG_HIGHMEM
-		/* Maximum memory usable is what is directly addressable */
-		printk(KERN_WARNING "Warning only %ldMB will be used.\n",
-		       MAXMEM >> 20);
-		printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
+	if (highest > PFN_DOWN(HIGHMEM_START)) {
+#ifdef CONFIG_HIGHMEM
+		highstart_pfn = PFN_DOWN(HIGHMEM_START);
+		highend_pfn = highest;
 #endif
+		highest = PFN_DOWN(HIGHMEM_START);
 	}
 
-#ifdef CONFIG_HIGHMEM
 	/*
-	 * Crude, we really should make a better attempt at detecting
-	 * highstart_pfn
+	 * Initialize the boot-time allocator with low memory only.
 	 */
-	highstart_pfn = highend_pfn = max_pfn;
-	if (max_pfn > MAXMEM_PFN) {
-		highstart_pfn = MAXMEM_PFN;
-		printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
-		       (highend_pfn - highstart_pfn) >> (20 - PAGE_SHIFT));
-	}
-#endif
-
-	/* Initialize the boot-time allocator with low memory only.  */
-	bootmap_size = init_bootmem(first_usable_pfn, max_low_pfn);
+	bootmap_size = init_bootmem(mapstart, highest);
 
 	/*
 	 * Register fully available low RAM pages with the bootmem allocator.
 	 */
 	for (i = 0; i < boot_mem_map.nr_map; i++) {
-		unsigned long curr_pfn, last_pfn, size;
+		unsigned long start, end, size;
 
 		/*
 		 * Reserve usable memory.
@@ -369,49 +357,37 @@ static inline void bootmem_init(void)
 		if (boot_mem_map.map[i].type != BOOT_MEM_RAM)
 			continue;
 
-		/*
-		 * We are rounding up the start address of usable memory:
-		 */
-		curr_pfn = PFN_UP(boot_mem_map.map[i].addr);
-		if (curr_pfn >= max_low_pfn)
-			continue;
-		if (curr_pfn < start_pfn)
-			curr_pfn = start_pfn;
-
-		/*
-		 * ... and at the end of the usable range downwards:
-		 */
-		last_pfn = PFN_DOWN(boot_mem_map.map[i].addr
+		start = PFN_UP(boot_mem_map.map[i].addr);
+		end   = PFN_DOWN(boot_mem_map.map[i].addr
 				    + boot_mem_map.map[i].size);
-
-		if (last_pfn > max_low_pfn)
-			last_pfn = max_low_pfn;
-
 		/*
-		 * Only register lowmem part of lowmem segment with bootmem.
+		 * We are rounding up the start address of usable memory
+		 * and at the end of the usable range downwards.
 		 */
-		size = last_pfn - curr_pfn;
-		if (curr_pfn > PFN_DOWN(HIGHMEM_START))
-			continue;
-		if (curr_pfn + size - 1 > PFN_DOWN(HIGHMEM_START))
-			size = PFN_DOWN(HIGHMEM_START) - curr_pfn;
-		if (!size)
+		if (start >= max_low_pfn)
 			continue;
+		if (start < reserved_end)
+			start = reserved_end;
+		if (end > max_low_pfn)
+			end = max_low_pfn;
 
 		/*
-		 * ... finally, did all the rounding and playing
-		 * around just make the area go away?
+		 * ... finally, is the area going away?
 		 */
-		if (last_pfn <= curr_pfn)
+		if (end <= start)
 			continue;
+		size = end - start;
 
 		/* Register lowmem ranges */
-		free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size));
-		memory_present(0, curr_pfn, curr_pfn + size - 1);
+		free_bootmem(PFN_PHYS(start), size << PAGE_SHIFT);
+		memory_present(0, start, end);
 	}
 
-	/* Reserve the bootmap memory.  */
-	reserve_bootmem(PFN_PHYS(first_usable_pfn), bootmap_size);
+	/*
+	 * Reserve the bootmap memory.
+	 */
+	reserve_bootmem(PFN_PHYS(mapstart), bootmap_size);
+
 #endif /* CONFIG_SGI_IP27 */
 
 #ifdef CONFIG_BLK_DEV_INITRD
@@ -483,6 +459,9 @@ static void __init arch_mem_init(char **cmdline_p)
 	paging_init();
 }
 
+#define MAXMEM		HIGHMEM_START
+#define MAXMEM_PFN	PFN_DOWN(MAXMEM)
+
 static inline void resource_init(void)
 {
 	int i;
-- 
cgit v1.2.1


From d2043ca8484d434201816267c0e23c5999df6cfd Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:49 +0200
Subject: [MIPS] setup.c: move initrd code inside dedicated functions

NUMA specific code could rely on them too.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 144 ++++++++++++++++++++++++++++-------------------
 1 file changed, 86 insertions(+), 58 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index bcbd926e58f7..489b27320dfa 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -200,7 +200,12 @@ static inline void parse_cmdline_early(void)
 	}
 }
 
-static inline int parse_rd_cmdline(unsigned long* rd_start, unsigned long* rd_end)
+/*
+ * Manage initrd
+ */
+#ifdef CONFIG_BLK_DEV_INITRD
+
+static int __init parse_rd_cmdline(unsigned long *rd_start, unsigned long *rd_end)
 {
 	/*
 	 * "rd_start=0xNNNNNNNN" defines the memory address of an initrd
@@ -263,49 +268,93 @@ static inline int parse_rd_cmdline(unsigned long* rd_start, unsigned long* rd_en
 	return 0;
 }
 
+static unsigned long __init init_initrd(void)
+{
+	unsigned long tmp, end;
+	u32 *initrd_header;
+
+	ROOT_DEV = Root_RAM0;
+
+	if (parse_rd_cmdline(&initrd_start, &initrd_end))
+		return initrd_end;
+	/*
+	 * Board specific code should have set up initrd_start
+	 * and initrd_end...
+	 */
+	end = (unsigned long)&_end;
+	tmp = PAGE_ALIGN(end) - sizeof(u32) * 2;
+	if (tmp < end)
+		tmp += PAGE_SIZE;
+
+	initrd_header = (u32 *)tmp;
+	if (initrd_header[0] == 0x494E5244) {
+		initrd_start = (unsigned long)&initrd_header[2];
+		initrd_end = initrd_start + initrd_header[1];
+	}
+	return initrd_end;
+}
+
+static void __init finalize_initrd(void)
+{
+	unsigned long size = initrd_end - initrd_start;
+
+	if (size == 0) {
+		printk(KERN_INFO "Initrd not found or empty");
+		goto disable;
+	}
+	if (CPHYSADDR(initrd_end) > PFN_PHYS(max_low_pfn)) {
+		printk("Initrd extends beyond end of memory");
+		goto disable;
+	}
+
+	reserve_bootmem(CPHYSADDR(initrd_start), size);
+	initrd_below_start_ok = 1;
+
+	printk(KERN_INFO "Initial ramdisk at: 0x%lx (%lu bytes)\n",
+	       initrd_start, size);
+	return;
+disable:
+	printk(" - disabling initrd\n");
+	initrd_start = 0;
+	initrd_end = 0;
+}
+
+#else  /* !CONFIG_BLK_DEV_INITRD */
+
+#define init_initrd()		0
+#define finalize_initrd()	do {} while (0)
+
+#endif
+
 /*
  * Initialize the bootmem allocator. It also setup initrd related data
  * if needed.
  */
+#ifdef CONFIG_SGI_IP27
+
 static void __init bootmem_init(void)
 {
-	unsigned long reserved_end = (unsigned long)&_end;
-#ifndef CONFIG_SGI_IP27
+	init_initrd();
+	finalize_initrd();
+}
+
+#else  /* !CONFIG_SGI_IP27 */
+
+static void __init bootmem_init(void)
+{
+	unsigned long reserved_end;
 	unsigned long highest = 0;
 	unsigned long mapstart = -1UL;
 	unsigned long bootmap_size;
 	int i;
-#endif
-#ifdef CONFIG_BLK_DEV_INITRD
-	int initrd_reserve_bootmem = 0;
-
-	/* Board specific code should have set up initrd_start and initrd_end */
- 	ROOT_DEV = Root_RAM0;
-	if (parse_rd_cmdline(&initrd_start, &initrd_end)) {
-		reserved_end = max(reserved_end, initrd_end);
-		initrd_reserve_bootmem = 1;
-	} else {
-		unsigned long tmp;
-		u32 *initrd_header;
-
-		tmp = PAGE_ALIGN(reserved_end) - sizeof(u32) * 2;
-		if (tmp < reserved_end)
-			tmp += PAGE_SIZE;
-		initrd_header = (u32 *)tmp;
-		if (initrd_header[0] == 0x494E5244) {
-			initrd_start = (unsigned long)&initrd_header[2];
-			initrd_end = initrd_start + initrd_header[1];
-			reserved_end = max(reserved_end, initrd_end);
-			initrd_reserve_bootmem = 1;
-		}
-	}
-#endif	/* CONFIG_BLK_DEV_INITRD */
 
-#ifndef CONFIG_SGI_IP27
 	/*
-	 * reserved_end is now a pfn
+	 * Init any data related to initrd. It's a nop if INITRD is
+	 * not selected. Once that done we can determine the low bound
+	 * of usable memory.
 	 */
-	reserved_end = PFN_UP(CPHYSADDR(reserved_end));
+	reserved_end = init_initrd();
+	reserved_end = PFN_UP(CPHYSADDR(max(reserved_end, (unsigned long)&_end)));
 
 	/*
 	 * Find the highest page frame number we have available.
@@ -388,35 +437,14 @@ static void __init bootmem_init(void)
 	 */
 	reserve_bootmem(PFN_PHYS(mapstart), bootmap_size);
 
-#endif /* CONFIG_SGI_IP27 */
-
-#ifdef CONFIG_BLK_DEV_INITRD
-	initrd_below_start_ok = 1;
-	if (initrd_start) {
-		unsigned long initrd_size = ((unsigned char *)initrd_end) -
-			((unsigned char *)initrd_start);
-		const int width = sizeof(long) * 2;
-
-		printk("Initial ramdisk at: 0x%p (%lu bytes)\n",
-		       (void *)initrd_start, initrd_size);
-
-		if (CPHYSADDR(initrd_end) > PFN_PHYS(max_low_pfn)) {
-			printk("initrd extends beyond end of memory "
-			       "(0x%0*Lx > 0x%0*Lx)\ndisabling initrd\n",
-			       width,
-			       (unsigned long long) CPHYSADDR(initrd_end),
-			       width,
-			       (unsigned long long) PFN_PHYS(max_low_pfn));
-			initrd_start = initrd_end = 0;
-			initrd_reserve_bootmem = 0;
-		}
-
-		if (initrd_reserve_bootmem)
-			reserve_bootmem(CPHYSADDR(initrd_start), initrd_size);
-	}
-#endif /* CONFIG_BLK_DEV_INITRD  */
+	/*
+	 * Reserve initrd memory if needed.
+	 */
+	finalize_initrd();
 }
 
+#endif	/* CONFIG_SGI_IP27 */
+
 /*
  * arch_mem_init - initialize memory managment subsystem
  *
-- 
cgit v1.2.1


From 8ff7bc4808f25e4dd75c3775dba313ddf35277e1 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:50 +0200
Subject: [MIPS] setup.c: remove useless includes.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 489b27320dfa..0bdc67116deb 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -10,29 +10,15 @@
  * Copyright (C) 1999 Silicon Graphics, Inc.
  * Copyright (C) 2000 2001, 2002  Maciej W. Rozycki
  */
-#include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/ioport.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
 #include <linux/module.h>
-#include <linux/stddef.h>
-#include <linux/string.h>
-#include <linux/unistd.h>
-#include <linux/slab.h>
-#include <linux/user.h>
-#include <linux/utsname.h>
-#include <linux/a.out.h>
 #include <linux/screen_info.h>
 #include <linux/bootmem.h>
 #include <linux/initrd.h>
-#include <linux/major.h>
-#include <linux/kdev_t.h>
 #include <linux/root_dev.h>
 #include <linux/highmem.h>
 #include <linux/console.h>
-#include <linux/mmzone.h>
 #include <linux/pfn.h>
 
 #include <asm/addrspace.h>
-- 
cgit v1.2.1


From 8df32c636e785069ba7d223ceb9b72c182902295 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:51 +0200
Subject: [MIPS] setup.c: do not inline functions

There's no point to inline any functions in setup.c. Let's GCC
doing its job, it's good enough for that now.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 0bdc67116deb..f66af9cc7220 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -135,7 +135,7 @@ static void __init print_memory_map(void)
 	}
 }
 
-static inline void parse_cmdline_early(void)
+static void __init parse_cmdline_early(void)
 {
 	char c = ' ', *to = command_line, *from = saved_command_line;
 	unsigned long start_at, mem_size;
@@ -476,7 +476,7 @@ static void __init arch_mem_init(char **cmdline_p)
 #define MAXMEM		HIGHMEM_START
 #define MAXMEM_PFN	PFN_DOWN(MAXMEM)
 
-static inline void resource_init(void)
+static void __init resource_init(void)
 {
 	int i;
 
-- 
cgit v1.2.1


From 1c6fd44d7ed3d105b2eaa29d72b415ca51e40d32 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:52 +0200
Subject: [MIPS] setup.c: remove MAXMEM macro

It doesn't improve readability.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index f66af9cc7220..804b93dc3c5a 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -473,9 +473,6 @@ static void __init arch_mem_init(char **cmdline_p)
 	paging_init();
 }
 
-#define MAXMEM		HIGHMEM_START
-#define MAXMEM_PFN	PFN_DOWN(MAXMEM)
-
 static void __init resource_init(void)
 {
 	int i;
@@ -497,10 +494,10 @@ static void __init resource_init(void)
 
 		start = boot_mem_map.map[i].addr;
 		end = boot_mem_map.map[i].addr + boot_mem_map.map[i].size - 1;
-		if (start >= MAXMEM)
+		if (start >= HIGHMEM_START)
 			continue;
-		if (end >= MAXMEM)
-			end = MAXMEM - 1;
+		if (end >= HIGHMEM_START)
+			end = HIGHMEM_START - 1;
 
 		res = alloc_bootmem(sizeof(struct resource));
 		switch (boot_mem_map.map[i].type) {
@@ -529,9 +526,6 @@ static void __init resource_init(void)
 	}
 }
 
-#undef MAXMEM
-#undef MAXMEM_PFN
-
 void __init setup_arch(char **cmdline_p)
 {
 	cpu_probe();
-- 
cgit v1.2.1


From a09fc446fb6d541281d9559fe7215d7c0d3cc9ce Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Fri, 11 Aug 2006 17:51:53 +0200
Subject: [MIPS] setup.c: use early_param() for early command line parsing

There's no point to rewrite some logic to parse command line
to pass initrd parameters or to declare a user memory area.
We could use instead parse_early_param() that does the same
thing.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c | 172 +++++++++++++++++------------------------------
 1 file changed, 60 insertions(+), 112 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 804b93dc3c5a..fdbb508661c5 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -135,138 +135,54 @@ static void __init print_memory_map(void)
 	}
 }
 
-static void __init parse_cmdline_early(void)
-{
-	char c = ' ', *to = command_line, *from = saved_command_line;
-	unsigned long start_at, mem_size;
-	int len = 0;
-	int usermem = 0;
-
-	printk("Determined physical RAM map:\n");
-	print_memory_map();
-
-	for (;;) {
-		/*
-		 * "mem=XXX[kKmM]" defines a memory region from
-		 * 0 to <XXX>, overriding the determined size.
-		 * "mem=XXX[KkmM]@YYY[KkmM]" defines a memory region from
-		 * <YYY> to <YYY>+<XXX>, overriding the determined size.
-		 */
-		if (c == ' ' && !memcmp(from, "mem=", 4)) {
-			if (to != command_line)
-				to--;
-			/*
-			 * If a user specifies memory size, we
-			 * blow away any automatically generated
-			 * size.
-			 */
-			if (usermem == 0) {
-				boot_mem_map.nr_map = 0;
-				usermem = 1;
-			}
-			mem_size = memparse(from + 4, &from);
-			if (*from == '@')
-				start_at = memparse(from + 1, &from);
-			else
-				start_at = 0;
-			add_memory_region(start_at, mem_size, BOOT_MEM_RAM);
-		}
-		c = *(from++);
-		if (!c)
-			break;
-		if (CL_SIZE <= ++len)
-			break;
-		*(to++) = c;
-	}
-	*to = '\0';
-
-	if (usermem) {
-		printk("User-defined physical RAM map:\n");
-		print_memory_map();
-	}
-}
-
 /*
  * Manage initrd
  */
 #ifdef CONFIG_BLK_DEV_INITRD
 
-static int __init parse_rd_cmdline(unsigned long *rd_start, unsigned long *rd_end)
+static int __init rd_start_early(char *p)
 {
-	/*
-	 * "rd_start=0xNNNNNNNN" defines the memory address of an initrd
-	 * "rd_size=0xNN" it's size
-	 */
-	unsigned long start = 0;
-	unsigned long size = 0;
-	unsigned long end;
-	char cmd_line[CL_SIZE];
-	char *start_str;
-	char *size_str;
-	char *tmp;
-
-	strcpy(cmd_line, command_line);
-	*command_line = 0;
-	tmp = cmd_line;
-	/* Ignore "rd_start=" strings in other parameters. */
-	start_str = strstr(cmd_line, "rd_start=");
-	if (start_str && start_str != cmd_line && *(start_str - 1) != ' ')
-		start_str = strstr(start_str, " rd_start=");
-	while (start_str) {
-		if (start_str != cmd_line)
-			strncat(command_line, tmp, start_str - tmp);
-		start = memparse(start_str + 9, &start_str);
-		tmp = start_str + 1;
-		start_str = strstr(start_str, " rd_start=");
-	}
-	if (*tmp)
-		strcat(command_line, tmp);
-
-	strcpy(cmd_line, command_line);
-	*command_line = 0;
-	tmp = cmd_line;
-	/* Ignore "rd_size" strings in other parameters. */
-	size_str = strstr(cmd_line, "rd_size=");
-	if (size_str && size_str != cmd_line && *(size_str - 1) != ' ')
-		size_str = strstr(size_str, " rd_size=");
-	while (size_str) {
-		if (size_str != cmd_line)
-			strncat(command_line, tmp, size_str - tmp);
-		size = memparse(size_str + 8, &size_str);
-		tmp = size_str + 1;
-		size_str = strstr(size_str, " rd_size=");
-	}
-	if (*tmp)
-		strcat(command_line, tmp);
+	unsigned long start = memparse(p, &p);
 
 #ifdef CONFIG_64BIT
 	/* HACK: Guess if the sign extension was forgotten */
 	if (start > 0x0000000080000000 && start < 0x00000000ffffffff)
 		start |= 0xffffffff00000000UL;
 #endif
+	initrd_start = start;
+	initrd_end += start;
+
+	return 0;
+}
+early_param("rd_start", rd_start_early);
+
+static int __init rd_size_early(char *p)
+{
+	initrd_end += memparse(p, &p);
 
-	end = start + size;
-	if (start && end) {
-		*rd_start = start;
-		*rd_end = end;
-		return 1;
-	}
 	return 0;
 }
+early_param("rd_size", rd_size_early);
 
 static unsigned long __init init_initrd(void)
 {
-	unsigned long tmp, end;
+	unsigned long tmp, end, size;
 	u32 *initrd_header;
 
 	ROOT_DEV = Root_RAM0;
 
-	if (parse_rd_cmdline(&initrd_start, &initrd_end))
-		return initrd_end;
 	/*
-	 * Board specific code should have set up initrd_start
-	 * and initrd_end...
+	 * Board specific code or command line parser should have
+	 * already set up initrd_start and initrd_end. In these cases
+	 * perfom sanity checks and use them if all looks good.
 	 */
+	size = initrd_end - initrd_start;
+	if (initrd_end == 0 || size == 0) {
+		initrd_start = 0;
+		initrd_end = 0;
+	} else
+		return initrd_end;
+
 	end = (unsigned long)&_end;
 	tmp = PAGE_ALIGN(end) - sizeof(u32) * 2;
 	if (tmp < end)
@@ -436,8 +352,6 @@ static void __init bootmem_init(void)
  *
  *  o plat_mem_setup() detects the memory configuration and will record detected
  *    memory areas using add_memory_region.
- *  o parse_cmdline_early() parses the command line for mem= options which,
- *    iff detected, will override the results of the automatic detection.
  *
  * At this stage the memory configuration of the system is known to the
  * kernel but generic memory managment system is still entirely uninitialized.
@@ -455,19 +369,53 @@ static void __init bootmem_init(void)
  * initialization hook for anything else was introduced.
  */
 
-extern void plat_mem_setup(void);
+static int usermem __initdata = 0;
+
+static int __init early_parse_mem(char *p)
+{
+	unsigned long start, size;
+
+	/*
+	 * If a user specifies memory size, we
+	 * blow away any automatically generated
+	 * size.
+	 */
+	if (usermem == 0) {
+		boot_mem_map.nr_map = 0;
+		usermem = 1;
+ 	}
+	start = 0;
+	size = memparse(p, &p);
+	if (*p == '@')
+		start = memparse(p + 1, &p);
+
+	add_memory_region(start, size, BOOT_MEM_RAM);
+	return 0;
+}
+early_param("mem", early_parse_mem);
 
 static void __init arch_mem_init(char **cmdline_p)
 {
+	extern void plat_mem_setup(void);
+
 	/* call board setup routine */
 	plat_mem_setup();
 
+	printk("Determined physical RAM map:\n");
+	print_memory_map();
+
 	strlcpy(command_line, arcs_cmdline, sizeof(command_line));
 	strlcpy(saved_command_line, command_line, COMMAND_LINE_SIZE);
 
 	*cmdline_p = command_line;
 
-	parse_cmdline_early();
+	parse_early_param();
+
+	if (usermem) {
+		printk("User-defined physical RAM map:\n");
+		print_memory_map();
+	}
+
 	bootmem_init();
 	sparse_init();
 	paging_init();
-- 
cgit v1.2.1


From bbf2bef9f50eb119ffadd735eb0966ac8a04f91f Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 27 Sep 2006 01:49:25 -0700
Subject: [PATCH] fix "cpu to node relationship fixup: map cpu to node"

Fix build error introduced by 3212fe1594e577463bc8601d28aa008f520c3377

Non-NUMA case should be handled.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/kernel/topology.c | 2 --
 include/asm-ia64/numa.h     | 2 ++
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c
index 05bdf7affb43..5629b45e89c6 100644
--- a/arch/ia64/kernel/topology.c
+++ b/arch/ia64/kernel/topology.c
@@ -36,9 +36,7 @@ int arch_register_cpu(int num)
 	 */
 	if (!can_cpei_retarget() && is_cpu_cpei_target(num))
 		sysfs_cpus[num].cpu.no_control = 1;
-#ifdef CONFIG_NUMA
 	map_cpu_to_node(num, node_cpuid[num].nid);
-#endif
 #endif
 
 	return register_cpu(&sysfs_cpus[num].cpu, num);
diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h
index e0a1d173e42d..7d5e2ccc37a0 100644
--- a/include/asm-ia64/numa.h
+++ b/include/asm-ia64/numa.h
@@ -69,6 +69,8 @@ extern void unmap_cpu_from_node(int cpu, int nid);
 
 
 #else /* !CONFIG_NUMA */
+#define map_cpu_to_node(cpu, nid)	do{}while(0)
+#define unmap_cpu_from_node(cpu, nid)	do{}while(0)
 
 #define paddr_to_nid(addr)	0
 
-- 
cgit v1.2.1


From e7ab8d65055e9b9dfc131d0467cfc5a8368d7ee4 Mon Sep 17 00:00:00 2001
From: Josh Triplett <josht@us.ibm.com>
Date: Wed, 27 Sep 2006 01:49:26 -0700
Subject: [PATCH] jbd: add lock annotation to jbd_sync_bh

jbd_sync_bh releases journal->j_list_lock.  Add a lock annotation to this
function so that sparse can check callers for lock pairing, and so that
sparse will not complain about this function since it intentionally uses
the lock in this manner.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/checkpoint.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 47678a26c13b..d0685596e5a6 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -145,6 +145,7 @@ void __log_wait_for_space(journal_t *journal)
  * jbd_unlock_bh_state().
  */
 static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh)
+	__releases(journal->j_list_lock)
 {
 	get_bh(bh);
 	spin_unlock(&journal->j_list_lock);
-- 
cgit v1.2.1


From ae6ddcc5f24d6b06ae9231dc128904750a4155e0 Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Wed, 27 Sep 2006 01:49:27 -0700
Subject: [PATCH] ext3 and jbd cleanup: remove whitespace

Remove whitespace from ext3 and jbd, before we clone ext4.

Signed-off-by: Mingming Cao<cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/balloc.c         |  16 +++---
 fs/ext3/bitmap.c         |   2 +-
 fs/ext3/dir.c            |  14 +++---
 fs/ext3/file.c           |   2 +-
 fs/ext3/fsync.c          |   6 +--
 fs/ext3/hash.c           |   6 +--
 fs/ext3/ialloc.c         |  48 +++++++++---------
 fs/ext3/inode.c          |  64 ++++++++++++------------
 fs/ext3/namei.c          |  28 +++++------
 fs/ext3/super.c          |  24 ++++-----
 fs/jbd/checkpoint.c      |  30 +++++------
 fs/jbd/journal.c         |  56 ++++++++++-----------
 fs/jbd/recovery.c        |  54 ++++++++++----------
 fs/jbd/revoke.c          |  70 +++++++++++++-------------
 fs/jbd/transaction.c     | 128 +++++++++++++++++++++++------------------------
 include/linux/ext3_jbd.h |  10 ++--
 include/linux/jbd.h      |  56 ++++++++++-----------
 17 files changed, 307 insertions(+), 307 deletions(-)

diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index 063d994bda0b..e6b983707008 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -74,7 +74,7 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb,
 }
 
 /*
- * Read the bitmap for a given block_group, reading into the specified 
+ * Read the bitmap for a given block_group, reading into the specified
  * slot in the superblock's bitmap cache.
  *
  * Return buffer_head on success or NULL in case of failure.
@@ -419,8 +419,8 @@ do_more:
 		}
 		/* @@@ This prevents newly-allocated data from being
 		 * freed and then reallocated within the same
-		 * transaction. 
-		 * 
+		 * transaction.
+		 *
 		 * Ideally we would want to allow that to happen, but to
 		 * do so requires making journal_forget() capable of
 		 * revoking the queued write of a data block, which
@@ -433,7 +433,7 @@ do_more:
 		 * safe not to set the allocation bit in the committed
 		 * bitmap, because we know that there is no outstanding
 		 * activity on the buffer any more and so it is safe to
-		 * reallocate it.  
+		 * reallocate it.
 		 */
 		BUFFER_TRACE(bitmap_bh, "set in b_committed_data");
 		J_ASSERT_BH(bitmap_bh,
@@ -518,7 +518,7 @@ void ext3_free_blocks(handle_t *handle, struct inode *inode,
  * data would allow the old block to be overwritten before the
  * transaction committed (because we force data to disk before commit).
  * This would lead to corruption if we crashed between overwriting the
- * data and committing the delete. 
+ * data and committing the delete.
  *
  * @@@ We may want to make this allocation behaviour conditional on
  * data-writes at some point, and disable it for metadata allocations or
@@ -584,7 +584,7 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 
 	if (start > 0) {
 		/*
-		 * The goal was occupied; search forward for a free 
+		 * The goal was occupied; search forward for a free
 		 * block within the next XX blocks.
 		 *
 		 * end_goal is more or less random, but it has to be
@@ -1194,7 +1194,7 @@ int ext3_should_retry_alloc(struct super_block *sb, int *retries)
 /*
  * ext3_new_block uses a goal block to assist allocation.  If the goal is
  * free, or there is a free block within 32 blocks of the goal, that block
- * is allocated.  Otherwise a forward search is made for a free block; within 
+ * is allocated.  Otherwise a forward search is made for a free block; within
  * each block group the search first looks for an entire free byte in the block
  * bitmap, and then for any free bit if that fails.
  * This function also updates quota and i_blocks field.
@@ -1303,7 +1303,7 @@ retry_alloc:
 	smp_rmb();
 
 	/*
-	 * Now search the rest of the groups.  We assume that 
+	 * Now search the rest of the groups.  We assume that
 	 * i and gdp correctly point to the last group visited.
 	 */
 	for (bgi = 0; bgi < ngroups; bgi++) {
diff --git a/fs/ext3/bitmap.c b/fs/ext3/bitmap.c
index ce4f82b9e528..b9176eed98d1 100644
--- a/fs/ext3/bitmap.c
+++ b/fs/ext3/bitmap.c
@@ -20,7 +20,7 @@ unsigned long ext3_count_free (struct buffer_head * map, unsigned int numchars)
 	unsigned int i;
 	unsigned long sum = 0;
 
-	if (!map) 
+	if (!map)
 		return (0);
 	for (i = 0; i < numchars; i++)
 		sum += nibblemap[map->b_data[i] & 0xf] +
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index fbb0d4ed07d4..6f9e5a523c87 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -59,7 +59,7 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 
 	return (ext3_filetype_table[filetype]);
 }
-			       
+
 
 int ext3_check_dir_entry (const char * function, struct inode * dir,
 			  struct ext3_dir_entry_2 * de,
@@ -162,7 +162,7 @@ revalidate:
 		 * to make sure. */
 		if (filp->f_version != inode->i_version) {
 			for (i = 0; i < sb->s_blocksize && i < offset; ) {
-				de = (struct ext3_dir_entry_2 *) 
+				de = (struct ext3_dir_entry_2 *)
 					(bh->b_data + i);
 				/* It's too expensive to do a full
 				 * dirent test each time round this
@@ -181,7 +181,7 @@ revalidate:
 			filp->f_version = inode->i_version;
 		}
 
-		while (!error && filp->f_pos < inode->i_size 
+		while (!error && filp->f_pos < inode->i_size
 		       && offset < sb->s_blocksize) {
 			de = (struct ext3_dir_entry_2 *) (bh->b_data + offset);
 			if (!ext3_check_dir_entry ("ext3_readdir", inode, de,
@@ -229,7 +229,7 @@ out:
 /*
  * These functions convert from the major/minor hash to an f_pos
  * value.
- * 
+ *
  * Currently we only use major hash numer.  This is unfortunate, but
  * on 32-bit machines, the same VFS interface is used for lseek and
  * llseek, so if we use the 64 bit offset, then the 32-bit versions of
@@ -250,7 +250,7 @@ out:
 struct fname {
 	__u32		hash;
 	__u32		minor_hash;
-	struct rb_node	rb_hash; 
+	struct rb_node	rb_hash;
 	struct fname	*next;
 	__u32		inode;
 	__u8		name_len;
@@ -410,7 +410,7 @@ static int call_filldir(struct file * filp, void * dirent,
 	curr_pos = hash2pos(fname->hash, fname->minor_hash);
 	while (fname) {
 		error = filldir(dirent, fname->name,
-				fname->name_len, curr_pos, 
+				fname->name_len, curr_pos,
 				fname->inode,
 				get_dtype(sb, fname->file_type));
 		if (error) {
@@ -465,7 +465,7 @@ static int ext3_dx_readdir(struct file * filp,
 		/*
 		 * Fill the rbtree if we have no more entries,
 		 * or the inode has changed since we last read in the
-		 * cached entries. 
+		 * cached entries.
 		 */
 		if ((!info->curr_node) ||
 		    (filp->f_version != inode->i_version)) {
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 1efefb630ea9..994efd189f4e 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -100,7 +100,7 @@ ext3_file_write(struct kiocb *iocb, const char __user *buf, size_t count, loff_t
 
 force_commit:
 	err = ext3_force_commit(inode->i_sb);
-	if (err) 
+	if (err)
 		return err;
 	return ret;
 }
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index 49382a208e05..dd1fd3c0fc05 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -8,14 +8,14 @@
  *                      Universite Pierre et Marie Curie (Paris VI)
  *  from
  *  linux/fs/minix/truncate.c   Copyright (C) 1991, 1992  Linus Torvalds
- * 
+ *
  *  ext3fs fsync primitive
  *
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
- * 
+ *
  *  Removed unnecessary code duplication for little endian machines
- *  and excessive __inline__s. 
+ *  and excessive __inline__s.
  *        Andi Kleen, 1997
  *
  * Major simplications and cleanup - we only need to do the metadata, because
diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c
index 5a2d1235ead0..7fa637cd322a 100644
--- a/fs/ext3/hash.c
+++ b/fs/ext3/hash.c
@@ -4,7 +4,7 @@
  * Copyright (C) 2002 by Theodore Ts'o
  *
  * This file is released under the GPL v2.
- * 
+ *
  * This file may be redistributed under the terms of the GNU Public
  * License.
  */
@@ -80,11 +80,11 @@ static void str2hashbuf(const char *msg, int len, __u32 *buf, int num)
  * Returns the hash of a filename.  If len is 0 and name is NULL, then
  * this function can be used to test whether or not a hash version is
  * supported.
- * 
+ *
  * The seed is an 4 longword (32 bits) "secret" which can be used to
  * uniquify a hash.  If the seed is all zero's, then some default seed
  * may be used.
- * 
+ *
  * A particular hash version specifies whether or not the seed is
  * represented, and whether or not the returned hash is 32 bits or 64
  * bits.  32 bit hashes will return 0 for the minor hash.
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 36546ed36a14..5e288368499b 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -216,7 +216,7 @@ static int find_group_dir(struct super_block *sb, struct inode *parent)
 			continue;
 		if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei)
 			continue;
-		if (!best_desc || 
+		if (!best_desc ||
 		    (le16_to_cpu(desc->bg_free_blocks_count) >
 		     le16_to_cpu(best_desc->bg_free_blocks_count))) {
 			best_group = group;
@@ -226,30 +226,30 @@ static int find_group_dir(struct super_block *sb, struct inode *parent)
 	return best_group;
 }
 
-/* 
- * Orlov's allocator for directories. 
- * 
+/*
+ * Orlov's allocator for directories.
+ *
  * We always try to spread first-level directories.
  *
- * If there are blockgroups with both free inodes and free blocks counts 
- * not worse than average we return one with smallest directory count. 
- * Otherwise we simply return a random group. 
- * 
- * For the rest rules look so: 
- * 
- * It's OK to put directory into a group unless 
- * it has too many directories already (max_dirs) or 
- * it has too few free inodes left (min_inodes) or 
- * it has too few free blocks left (min_blocks) or 
- * it's already running too large debt (max_debt). 
- * Parent's group is prefered, if it doesn't satisfy these 
- * conditions we search cyclically through the rest. If none 
- * of the groups look good we just look for a group with more 
- * free inodes than average (starting at parent's group). 
- * 
- * Debt is incremented each time we allocate a directory and decremented 
- * when we allocate an inode, within 0--255. 
- */ 
+ * If there are blockgroups with both free inodes and free blocks counts
+ * not worse than average we return one with smallest directory count.
+ * Otherwise we simply return a random group.
+ *
+ * For the rest rules look so:
+ *
+ * It's OK to put directory into a group unless
+ * it has too many directories already (max_dirs) or
+ * it has too few free inodes left (min_inodes) or
+ * it has too few free blocks left (min_blocks) or
+ * it's already running too large debt (max_debt).
+ * Parent's group is prefered, if it doesn't satisfy these
+ * conditions we search cyclically through the rest. If none
+ * of the groups look good we just look for a group with more
+ * free inodes than average (starting at parent's group).
+ *
+ * Debt is incremented each time we allocate a directory and decremented
+ * when we allocate an inode, within 0--255.
+ */
 
 #define INODE_COST 64
 #define BLOCK_COST 256
@@ -454,7 +454,7 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
 			group = find_group_dir(sb, dir);
 		else
 			group = find_group_orlov(sb, dir);
-	} else 
+	} else
 		group = find_group_other(sb, dir);
 
 	err = -ENOSPC;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 84be02e93652..473d206b1d7e 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -55,7 +55,7 @@ static int ext3_inode_is_fast_symlink(struct inode *inode)
 /*
  * The ext3 forget function must perform a revoke if we are freeing data
  * which has been journaled.  Metadata (eg. indirect blocks) must be
- * revoked in all cases. 
+ * revoked in all cases.
  *
  * "bh" may be NULL: a metadata block may have been freed from memory
  * but there may still be a record of it in the journal, and that record
@@ -105,7 +105,7 @@ int ext3_forget(handle_t *handle, int is_metadata, struct inode *inode,
  * Work out how many blocks we need to proceed with the next chunk of a
  * truncate transaction.
  */
-static unsigned long blocks_for_truncate(struct inode *inode) 
+static unsigned long blocks_for_truncate(struct inode *inode)
 {
 	unsigned long needed;
 
@@ -122,13 +122,13 @@ static unsigned long blocks_for_truncate(struct inode *inode)
 
 	/* But we need to bound the transaction so we don't overflow the
 	 * journal. */
-	if (needed > EXT3_MAX_TRANS_DATA) 
+	if (needed > EXT3_MAX_TRANS_DATA)
 		needed = EXT3_MAX_TRANS_DATA;
 
 	return EXT3_DATA_TRANS_BLOCKS(inode->i_sb) + needed;
 }
 
-/* 
+/*
  * Truncate transactions can be complex and absolutely huge.  So we need to
  * be able to restart the transaction at a conventient checkpoint to make
  * sure we don't overflow the journal.
@@ -136,9 +136,9 @@ static unsigned long blocks_for_truncate(struct inode *inode)
  * start_transaction gets us a new handle for a truncate transaction,
  * and extend_transaction tries to extend the existing one a bit.  If
  * extend fails, we need to propagate the failure up and restart the
- * transaction in the top-level truncate loop. --sct 
+ * transaction in the top-level truncate loop. --sct
  */
-static handle_t *start_transaction(struct inode *inode) 
+static handle_t *start_transaction(struct inode *inode)
 {
 	handle_t *result;
 
@@ -215,12 +215,12 @@ void ext3_delete_inode (struct inode * inode)
 	ext3_orphan_del(handle, inode);
 	EXT3_I(inode)->i_dtime	= get_seconds();
 
-	/* 
+	/*
 	 * One subtle ordering requirement: if anything has gone wrong
 	 * (transaction abort, IO errors, whatever), then we can still
 	 * do these next steps (the fs will already have been marked as
 	 * having errors), but we can't free the inode if the mark_dirty
-	 * fails.  
+	 * fails.
 	 */
 	if (ext3_mark_inode_dirty(handle, inode))
 		/* If that failed, just do the required in-core inode clear. */
@@ -398,7 +398,7 @@ no_block:
  *	  + if there is a block to the left of our position - allocate near it.
  *	  + if pointer will live in indirect block - allocate near that block.
  *	  + if pointer will live in inode - allocate in the same
- *	    cylinder group. 
+ *	    cylinder group.
  *
  * In the latter case we colour the starting block by the callers PID to
  * prevent it from clashing with concurrent allocations for a different inode
@@ -744,7 +744,7 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode,
 		jbd_debug(5, "splicing indirect only\n");
 		BUFFER_TRACE(where->bh, "call ext3_journal_dirty_metadata");
 		err = ext3_journal_dirty_metadata(handle, where->bh);
-		if (err) 
+		if (err)
 			goto err_out;
 	} else {
 		/*
@@ -1137,7 +1137,7 @@ static int walk_page_buffers(	handle_t *handle,
  * So what we do is to rely on the fact that journal_stop/journal_start
  * will _not_ run commit under these circumstances because handle->h_ref
  * is elevated.  We'll still have enough credits for the tiny quotafile
- * write.  
+ * write.
  */
 static int do_journal_get_write_access(handle_t *handle,
 					struct buffer_head *bh)
@@ -1282,7 +1282,7 @@ static int ext3_journalled_commit_write(struct file *file,
 	if (inode->i_size > EXT3_I(inode)->i_disksize) {
 		EXT3_I(inode)->i_disksize = inode->i_size;
 		ret2 = ext3_mark_inode_dirty(handle, inode);
-		if (!ret) 
+		if (!ret)
 			ret = ret2;
 	}
 	ret2 = ext3_journal_stop(handle);
@@ -1291,7 +1291,7 @@ static int ext3_journalled_commit_write(struct file *file,
 	return ret;
 }
 
-/* 
+/*
  * bmap() is special.  It gets used by applications such as lilo and by
  * the swapper to find the on-disk block of a specific piece of data.
  *
@@ -1300,10 +1300,10 @@ static int ext3_journalled_commit_write(struct file *file,
  * filesystem and enables swap, then they may get a nasty shock when the
  * data getting swapped to that swapfile suddenly gets overwritten by
  * the original zero's written out previously to the journal and
- * awaiting writeback in the kernel's buffer cache. 
+ * awaiting writeback in the kernel's buffer cache.
  *
  * So, if we see any bmap calls here on a modified, data-journaled file,
- * take extra steps to flush any blocks which might be in the cache. 
+ * take extra steps to flush any blocks which might be in the cache.
  */
 static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
 {
@@ -1312,16 +1312,16 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
 	int err;
 
 	if (EXT3_I(inode)->i_state & EXT3_STATE_JDATA) {
-		/* 
+		/*
 		 * This is a REALLY heavyweight approach, but the use of
 		 * bmap on dirty files is expected to be extremely rare:
 		 * only if we run lilo or swapon on a freshly made file
-		 * do we expect this to happen. 
+		 * do we expect this to happen.
 		 *
 		 * (bmap requires CAP_SYS_RAWIO so this does not
 		 * represent an unprivileged user DOS attack --- we'd be
 		 * in trouble if mortal users could trigger this path at
-		 * will.) 
+		 * will.)
 		 *
 		 * NB. EXT3_STATE_JDATA is not set on files other than
 		 * regular files.  If somebody wants to bmap a directory
@@ -1457,7 +1457,7 @@ static int ext3_ordered_writepage(struct page *page,
 	 */
 
 	/*
-	 * And attach them to the current transaction.  But only if 
+	 * And attach them to the current transaction.  But only if
 	 * block_write_full_page() succeeded.  Otherwise they are unmapped,
 	 * and generally junk.
 	 */
@@ -1644,7 +1644,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb,
 		}
 	}
 
-	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, 
+	ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
 				 offset, nr_segs,
 				 ext3_get_block, NULL);
 
@@ -2025,7 +2025,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode,
 			   __le32 *first, __le32 *last)
 {
 	ext3_fsblk_t block_to_free = 0;    /* Starting block # of a run */
-	unsigned long count = 0;	    /* Number of blocks in the run */ 
+	unsigned long count = 0;	    /* Number of blocks in the run */
 	__le32 *block_to_free_p = NULL;	    /* Pointer into inode/ind
 					       corresponding to
 					       block_to_free */
@@ -2054,7 +2054,7 @@ static void ext3_free_data(handle_t *handle, struct inode *inode,
 			} else if (nr == block_to_free + count) {
 				count++;
 			} else {
-				ext3_clear_blocks(handle, inode, this_bh, 
+				ext3_clear_blocks(handle, inode, this_bh,
 						  block_to_free,
 						  count, block_to_free_p, p);
 				block_to_free = nr;
@@ -2184,7 +2184,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode,
 					*p = 0;
 					BUFFER_TRACE(parent_bh,
 					"call ext3_journal_dirty_metadata");
-					ext3_journal_dirty_metadata(handle, 
+					ext3_journal_dirty_metadata(handle,
 								    parent_bh);
 				}
 			}
@@ -2704,7 +2704,7 @@ void ext3_read_inode(struct inode * inode)
 		if (raw_inode->i_block[0])
 			init_special_inode(inode, inode->i_mode,
 			   old_decode_dev(le32_to_cpu(raw_inode->i_block[0])));
-		else 
+		else
 			init_special_inode(inode, inode->i_mode,
 			   new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
 	}
@@ -2724,8 +2724,8 @@ bad_inode:
  *
  * The caller must have write access to iloc->bh.
  */
-static int ext3_do_update_inode(handle_t *handle, 
-				struct inode *inode, 
+static int ext3_do_update_inode(handle_t *handle,
+				struct inode *inode,
 				struct ext3_iloc *iloc)
 {
 	struct ext3_inode *raw_inode = ext3_raw_inode(iloc);
@@ -2900,7 +2900,7 @@ int ext3_write_inode(struct inode *inode, int wait)
  * commit will leave the blocks being flushed in an unused state on
  * disk.  (On recovery, the inode will get truncated and the blocks will
  * be freed, so we have a strong guarantee that no future commit will
- * leave these blocks visible to the user.)  
+ * leave these blocks visible to the user.)
  *
  * Called with inode->sem down.
  */
@@ -3043,13 +3043,13 @@ int ext3_mark_iloc_dirty(handle_t *handle,
 	return err;
 }
 
-/* 
+/*
  * On success, We end up with an outstanding reference count against
- * iloc->bh.  This _must_ be cleaned up later. 
+ * iloc->bh.  This _must_ be cleaned up later.
  */
 
 int
-ext3_reserve_inode_write(handle_t *handle, struct inode *inode, 
+ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
 			 struct ext3_iloc *iloc)
 {
 	int err = 0;
@@ -3139,7 +3139,7 @@ out:
 }
 
 #if 0
-/* 
+/*
  * Bind an inode's backing buffer_head into this transaction, to prevent
  * it from being flushed to disk early.  Unlike
  * ext3_reserve_inode_write, this leaves behind no bh reference and
@@ -3157,7 +3157,7 @@ static int ext3_pin_inode(handle_t *handle, struct inode *inode)
 			BUFFER_TRACE(iloc.bh, "get_write_access");
 			err = journal_get_write_access(handle, iloc.bh);
 			if (!err)
-				err = ext3_journal_dirty_metadata(handle, 
+				err = ext3_journal_dirty_metadata(handle,
 								  iloc.bh);
 			brelse(iloc.bh);
 		}
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 2aa7101b27cd..4123f5261bcd 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -76,7 +76,7 @@ static struct buffer_head *ext3_append(handle_t *handle,
 #ifdef DX_DEBUG
 #define dxtrace(command) command
 #else
-#define dxtrace(command) 
+#define dxtrace(command)
 #endif
 
 struct fake_dirent
@@ -169,7 +169,7 @@ static struct ext3_dir_entry_2* dx_pack_dirents (char *base, int size);
 static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block);
 static int ext3_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
-				 struct dx_frame *frames, 
+				 struct dx_frame *frames,
 				 __u32 *start_hash);
 static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry,
 		       struct ext3_dir_entry_2 **res_dir, int *err);
@@ -250,7 +250,7 @@ static void dx_show_index (char * label, struct dx_entry *entries)
 }
 
 struct stats
-{ 
+{
 	unsigned names;
 	unsigned space;
 	unsigned bcount;
@@ -464,7 +464,7 @@ static void dx_release (struct dx_frame *frames)
  */
 static int ext3_htree_next_block(struct inode *dir, __u32 hash,
 				 struct dx_frame *frame,
-				 struct dx_frame *frames, 
+				 struct dx_frame *frames,
 				 __u32 *start_hash)
 {
 	struct dx_frame *p;
@@ -632,7 +632,7 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 		}
 		count += ret;
 		hashval = ~0;
-		ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS, 
+		ret = ext3_htree_next_block(dir, HASH_NB_ALWAYS,
 					    frame, frames, &hashval);
 		*next_hash = hashval;
 		if (ret < 0) {
@@ -649,7 +649,7 @@ int ext3_htree_fill_tree(struct file *dir_file, __u32 start_hash,
 			break;
 	}
 	dx_release(frames);
-	dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", 
+	dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n",
 		       count, *next_hash));
 	return count;
 errout:
@@ -1050,7 +1050,7 @@ struct dentry *ext3_get_parent(struct dentry *child)
 		parent = ERR_PTR(-ENOMEM);
 	}
 	return parent;
-} 
+}
 
 #define S_SHIFT 12
 static unsigned char ext3_type_by_mode[S_IFMT >> S_SHIFT] = {
@@ -1198,7 +1198,7 @@ errout:
  * add_dirent_to_buf will attempt search the directory block for
  * space.  It will return -ENOSPC if no space is available, and -EIO
  * and -EEXIST if directory entry already exists.
- * 
+ *
  * NOTE!  bh is NOT released in the case where ENOSPC is returned.  In
  * all other cases bh is released.
  */
@@ -1572,7 +1572,7 @@ cleanup:
  * ext3_delete_entry deletes a directory entry by merging it with the
  * previous entry
  */
-static int ext3_delete_entry (handle_t *handle, 
+static int ext3_delete_entry (handle_t *handle,
 			      struct inode * dir,
 			      struct ext3_dir_entry_2 * de_del,
 			      struct buffer_head * bh)
@@ -1643,12 +1643,12 @@ static int ext3_add_nondir(handle_t *handle,
  * is so far negative - it has no inode.
  *
  * If the create succeeds, we fill in the inode information
- * with d_instantiate(). 
+ * with d_instantiate().
  */
 static int ext3_create (struct inode * dir, struct dentry * dentry, int mode,
 		struct nameidata *nd)
 {
-	handle_t *handle; 
+	handle_t *handle;
 	struct inode * inode;
 	int err, retries = 0;
 
@@ -1813,7 +1813,7 @@ static int empty_dir (struct inode * inode)
 	de1 = (struct ext3_dir_entry_2 *)
 			((char *) de + le16_to_cpu(de->rec_len));
 	if (le32_to_cpu(de->inode) != inode->i_ino ||
-			!le32_to_cpu(de1->inode) || 
+			!le32_to_cpu(de1->inode) ||
 			strcmp (".", de->name) ||
 			strcmp ("..", de1->name)) {
 	    	ext3_warning (inode->i_sb, "empty_dir",
@@ -1883,7 +1883,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 	 * being truncated, or files being unlinked. */
 
 	/* @@@ FIXME: Observation from aviro:
-	 * I think I can trigger J_ASSERT in ext3_orphan_add().  We block 
+	 * I think I can trigger J_ASSERT in ext3_orphan_add().  We block
 	 * here (on lock_super()), so race with ext3_link() which might bump
 	 * ->i_nlink. For, say it, character device. Not a regular file,
 	 * not a directory, not a symlink and ->i_nlink > 0.
@@ -2393,4 +2393,4 @@ struct inode_operations ext3_special_inode_operations = {
 	.removexattr	= generic_removexattr,
 #endif
 	.permission	= ext3_permission,
-}; 
+};
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 3559086eee5f..4b95bfe4c8f7 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -62,13 +62,13 @@ static void ext3_unlockfs(struct super_block *sb);
 static void ext3_write_super (struct super_block * sb);
 static void ext3_write_super_lockfs(struct super_block *sb);
 
-/* 
+/*
  * Wrappers for journal_start/end.
  *
  * The only special thing we need to do here is to make sure that all
  * journal_end calls result in the superblock being marked dirty, so
  * that sync() will call the filesystem's write_super callback if
- * appropriate. 
+ * appropriate.
  */
 handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks)
 {
@@ -90,11 +90,11 @@ handle_t *ext3_journal_start_sb(struct super_block *sb, int nblocks)
 	return journal_start(journal, nblocks);
 }
 
-/* 
+/*
  * The only special thing we need to do here is to make sure that all
  * journal_stop calls result in the superblock being marked dirty, so
  * that sync() will call the filesystem's write_super callback if
- * appropriate. 
+ * appropriate.
  */
 int __ext3_journal_stop(const char *where, handle_t *handle)
 {
@@ -369,7 +369,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 {
 	struct list_head *l;
 
-	printk(KERN_ERR "sb orphan head is %d\n", 
+	printk(KERN_ERR "sb orphan head is %d\n",
 	       le32_to_cpu(sbi->s_es->s_last_orphan));
 
 	printk(KERN_ERR "sb_info orphan list:\n");
@@ -378,7 +378,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 		printk(KERN_ERR "  "
 		       "inode %s:%ld at %p: mode %o, nlink %d, next %d\n",
 		       inode->i_sb->s_id, inode->i_ino, inode,
-		       inode->i_mode, inode->i_nlink, 
+		       inode->i_mode, inode->i_nlink,
 		       NEXT_ORPHAN(inode));
 	}
 }
@@ -475,7 +475,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
 		inode_init_once(&ei->vfs_inode);
 	}
 }
- 
+
 static int init_inodecache(void)
 {
 	ext3_inode_cachep = kmem_cache_create("ext3_inode_cache",
@@ -1483,7 +1483,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	    (EXT3_HAS_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_RO_COMPAT_FEATURE(sb, ~0U) ||
 	     EXT3_HAS_INCOMPAT_FEATURE(sb, ~0U)))
-		printk(KERN_WARNING 
+		printk(KERN_WARNING
 		       "EXT3-fs warning: feature flags set on rev 0 fs, "
 		       "running e2fsck is recommended\n");
 	/*
@@ -1509,7 +1509,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	if (blocksize < EXT3_MIN_BLOCK_SIZE ||
 	    blocksize > EXT3_MAX_BLOCK_SIZE) {
-		printk(KERN_ERR 
+		printk(KERN_ERR
 		       "EXT3-fs: Unsupported filesystem blocksize %d on %s.\n",
 		       blocksize, sb->s_id);
 		goto failed_mount;
@@ -1533,14 +1533,14 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 		offset = (sb_block * EXT3_MIN_BLOCK_SIZE) % blocksize;
 		bh = sb_bread(sb, logic_sb_block);
 		if (!bh) {
-			printk(KERN_ERR 
+			printk(KERN_ERR
 			       "EXT3-fs: Can't read superblock on 2nd try.\n");
 			goto failed_mount;
 		}
 		es = (struct ext3_super_block *)(((char *)bh->b_data) + offset);
 		sbi->s_es = es;
 		if (es->s_magic != cpu_to_le16(EXT3_SUPER_MAGIC)) {
-			printk (KERN_ERR 
+			printk (KERN_ERR
 				"EXT3-fs: Magic mismatch, very weird !\n");
 			goto failed_mount;
 		}
@@ -1820,7 +1820,7 @@ out_fail:
 /*
  * Setup any per-fs journal parameters now.  We'll do this both on
  * initial mount, once the journal has been initialised but before we've
- * done any recovery; and again on any subsequent remount. 
+ * done any recovery; and again on any subsequent remount.
  */
 static void ext3_init_journal_params(struct super_block *sb, journal_t *journal)
 {
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index d0685596e5a6..961ada28db5e 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/checkpoint.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
  *
  * Copyright 1999 Red Hat Software --- All Rights Reserved
@@ -9,8 +9,8 @@
  * the terms of the GNU General Public License, version 2, or at your
  * option, any later version, incorporated herein by reference.
  *
- * Checkpoint routines for the generic filesystem journaling code.  
- * Part of the ext2fs journaling system.  
+ * Checkpoint routines for the generic filesystem journaling code.
+ * Part of the ext2fs journaling system.
  *
  * Checkpointing is the process of ensuring that a section of the log is
  * committed fully to disk, so that that portion of the log can be
@@ -226,7 +226,7 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Try to flush one buffer from the checkpoint list to disk.
  *
  * Return 1 if something happened which requires us to abort the current
- * scan of the checkpoint list.  
+ * scan of the checkpoint list.
  *
  * Called with j_list_lock held and drops it if 1 is returned
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
@@ -270,7 +270,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		 * possibly block, while still holding the journal lock.
 		 * We cannot afford to let the transaction logic start
 		 * messing around with this buffer before we write it to
-		 * disk, as that would break recoverability.  
+		 * disk, as that would break recoverability.
 		 */
 		BUFFER_TRACE(bh, "queue");
 		get_bh(bh);
@@ -293,7 +293,7 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
  * Perform an actual checkpoint. We take the first transaction on the
  * list of transactions to be checkpointed and send all its buffers
  * to disk. We submit larger chunks of data at once.
- * 
+ *
  * The journal should be locked before calling this function.
  */
 int log_do_checkpoint(journal_t *journal)
@@ -304,10 +304,10 @@ int log_do_checkpoint(journal_t *journal)
 
 	jbd_debug(1, "Start checkpoint\n");
 
-	/* 
+	/*
 	 * First thing: if there are any transactions in the log which
 	 * don't need checkpointing, just eliminate them from the
-	 * journal straight away.  
+	 * journal straight away.
 	 */
 	result = cleanup_journal_tail(journal);
 	jbd_debug(1, "cleanup_journal_tail returned %d\n", result);
@@ -385,9 +385,9 @@ out:
  * we have already got rid of any since the last update of the log tail
  * in the journal superblock.  If so, we can instantly roll the
  * superblock forward to remove those transactions from the log.
- * 
+ *
  * Return <0 on error, 0 on success, 1 if there was nothing to clean up.
- * 
+ *
  * Called with the journal lock held.
  *
  * This is the only part of the journaling code which really needs to be
@@ -404,8 +404,8 @@ int cleanup_journal_tail(journal_t *journal)
 	unsigned long	blocknr, freed;
 
 	/* OK, work out the oldest transaction remaining in the log, and
-	 * the log block it starts at. 
-	 * 
+	 * the log block it starts at.
+	 *
 	 * If the log is now empty, we need to work out which is the
 	 * next transaction ID we will write, and where it will
 	 * start. */
@@ -558,7 +558,7 @@ out:
 	return ret;
 }
 
-/* 
+/*
  * journal_remove_checkpoint: called after a buffer has been committed
  * to disk (either by being write-back flushed to disk, or being
  * committed to the log).
@@ -636,7 +636,7 @@ out:
  * Called with the journal locked.
  * Called with j_list_lock held.
  */
-void __journal_insert_checkpoint(struct journal_head *jh, 
+void __journal_insert_checkpoint(struct journal_head *jh,
 			       transaction_t *transaction)
 {
 	JBUFFER_TRACE(jh, "entry");
@@ -658,7 +658,7 @@ void __journal_insert_checkpoint(struct journal_head *jh,
 
 /*
  * We've finished with this transaction structure: adios...
- * 
+ *
  * The transaction must have no links except for the checkpoint by this
  * point.
  *
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index f66724ce443a..87c5a6d00805 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -578,7 +578,7 @@ int journal_next_log_block(journal_t *journal, unsigned long *retp)
  * this is a no-op.  If needed, we can use j_blk_offset - everything is
  * ready.
  */
-int journal_bmap(journal_t *journal, unsigned long blocknr, 
+int journal_bmap(journal_t *journal, unsigned long blocknr,
 		 unsigned long *retp)
 {
 	int err = 0;
@@ -699,10 +699,10 @@ fail:
  *  @len:  Lenght of the journal in blocks.
  *  @blocksize: blocksize of journalling device
  *  @returns: a newly created journal_t *
- *  
+ *
  *  journal_init_dev creates a journal which maps a fixed contiguous
  *  range of blocks on an arbitrary block device.
- * 
+ *
  */
 journal_t * journal_init_dev(struct block_device *bdev,
 			struct block_device *fs_dev,
@@ -739,11 +739,11 @@ journal_t * journal_init_dev(struct block_device *bdev,
 
 	return journal;
 }
- 
-/** 
+
+/**
  *  journal_t * journal_init_inode () - creates a journal which maps to a inode.
  *  @inode: An inode to create the journal in
- *  
+ *
  * journal_init_inode creates a journal which maps an on-disk inode as
  * the journal.  The inode must exist already, must support bmap() and
  * must have all data blocks preallocated.
@@ -763,7 +763,7 @@ journal_t * journal_init_inode (struct inode *inode)
 	journal->j_inode = inode;
 	jbd_debug(1,
 		  "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n",
-		  journal, inode->i_sb->s_id, inode->i_ino, 
+		  journal, inode->i_sb->s_id, inode->i_ino,
 		  (long long) inode->i_size,
 		  inode->i_sb->s_blocksize_bits, inode->i_sb->s_blocksize);
 
@@ -798,10 +798,10 @@ journal_t * journal_init_inode (struct inode *inode)
 	return journal;
 }
 
-/* 
+/*
  * If the journal init or create aborts, we need to mark the journal
  * superblock as being NULL to prevent the journal destroy from writing
- * back a bogus superblock. 
+ * back a bogus superblock.
  */
 static void journal_fail_superblock (journal_t *journal)
 {
@@ -844,13 +844,13 @@ static int journal_reset(journal_t *journal)
 	return 0;
 }
 
-/** 
+/**
  * int journal_create() - Initialise the new journal file
  * @journal: Journal to create. This structure must have been initialised
- * 
+ *
  * Given a journal_t structure which tells us which disk blocks we can
  * use, create a new journal superblock and initialise all of the
- * journal fields from scratch.  
+ * journal fields from scratch.
  **/
 int journal_create(journal_t *journal)
 {
@@ -915,7 +915,7 @@ int journal_create(journal_t *journal)
 	return journal_reset(journal);
 }
 
-/** 
+/**
  * void journal_update_superblock() - Update journal sb on disk.
  * @journal: The journal to update.
  * @wait: Set to '0' if you don't want to wait for IO completion.
@@ -939,7 +939,7 @@ void journal_update_superblock(journal_t *journal, int wait)
 				journal->j_transaction_sequence) {
 		jbd_debug(1,"JBD: Skipping superblock update on recovered sb "
 			"(start %ld, seq %d, errno %d)\n",
-			journal->j_tail, journal->j_tail_sequence, 
+			journal->j_tail, journal->j_tail_sequence,
 			journal->j_errno);
 		goto out;
 	}
@@ -1062,7 +1062,7 @@ static int load_superblock(journal_t *journal)
 /**
  * int journal_load() - Read journal from disk.
  * @journal: Journal to act on.
- * 
+ *
  * Given a journal_t structure which tells us which disk blocks contain
  * a journal, read the journal from disk to initialise the in-memory
  * structures.
@@ -1172,9 +1172,9 @@ void journal_destroy(journal_t *journal)
  * @compat: bitmask of compatible features
  * @ro: bitmask of features that force read-only mount
  * @incompat: bitmask of incompatible features
- * 
+ *
  * Check whether the journal uses all of a given set of
- * features.  Return true (non-zero) if it does. 
+ * features.  Return true (non-zero) if it does.
  **/
 
 int journal_check_used_features (journal_t *journal, unsigned long compat,
@@ -1203,7 +1203,7 @@ int journal_check_used_features (journal_t *journal, unsigned long compat,
  * @compat: bitmask of compatible features
  * @ro: bitmask of features that force read-only mount
  * @incompat: bitmask of incompatible features
- * 
+ *
  * Check whether the journaling code supports the use of
  * all of a given set of features on this journal.  Return true
  * (non-zero) if it can. */
@@ -1241,7 +1241,7 @@ int journal_check_available_features (journal_t *journal, unsigned long compat,
  * @incompat: bitmask of incompatible features
  *
  * Mark a given journal feature as present on the
- * superblock.  Returns true if the requested features could be set. 
+ * superblock.  Returns true if the requested features could be set.
  *
  */
 
@@ -1327,7 +1327,7 @@ static int journal_convert_superblock_v1(journal_t *journal,
 /**
  * int journal_flush () - Flush journal
  * @journal: Journal to act on.
- * 
+ *
  * Flush all data for a given journal to disk and empty the journal.
  * Filesystems can use this when remounting readonly to ensure that
  * recovery does not need to happen on remount.
@@ -1394,7 +1394,7 @@ int journal_flush(journal_t *journal)
  * int journal_wipe() - Wipe journal contents
  * @journal: Journal to act on.
  * @write: flag (see below)
- * 
+ *
  * Wipe out all of the contents of a journal, safely.  This will produce
  * a warning if the journal contains any valid recovery information.
  * Must be called between journal_init_*() and journal_load().
@@ -1449,7 +1449,7 @@ static const char *journal_dev_name(journal_t *journal, char *buffer)
 
 /*
  * Journal abort has very specific semantics, which we describe
- * for journal abort. 
+ * for journal abort.
  *
  * Two internal function, which provide abort to te jbd layer
  * itself are here.
@@ -1504,7 +1504,7 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * Perform a complete, immediate shutdown of the ENTIRE
  * journal (not of a single transaction).  This operation cannot be
  * undone without closing and reopening the journal.
- *           
+ *
  * The journal_abort function is intended to support higher level error
  * recovery mechanisms such as the ext2/ext3 remount-readonly error
  * mode.
@@ -1538,7 +1538,7 @@ static void __journal_abort_soft (journal_t *journal, int errno)
  * supply an errno; a null errno implies that absolutely no further
  * writes are done to the journal (unless there are any already in
  * progress).
- * 
+ *
  */
 
 void journal_abort(journal_t *journal, int errno)
@@ -1546,7 +1546,7 @@ void journal_abort(journal_t *journal, int errno)
 	__journal_abort_soft(journal, errno);
 }
 
-/** 
+/**
  * int journal_errno () - returns the journal's error state.
  * @journal: journal to examine.
  *
@@ -1570,7 +1570,7 @@ int journal_errno(journal_t *journal)
 	return err;
 }
 
-/** 
+/**
  * int journal_clear_err () - clears the journal's error state
  * @journal: journal to act on.
  *
@@ -1590,7 +1590,7 @@ int journal_clear_err(journal_t *journal)
 	return err;
 }
 
-/** 
+/**
  * void journal_ack_err() - Ack journal err.
  * @journal: journal to act on.
  *
@@ -1612,7 +1612,7 @@ int journal_blocks_per_page(struct inode *inode)
 
 /*
  * Simple support for retrying memory allocations.  Introduced to help to
- * debug different VM deadlock avoidance strategies. 
+ * debug different VM deadlock avoidance strategies.
  */
 void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry)
 {
diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index de5bafb4e853..73bb64806ed3 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/recovery.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1999
  *
  * Copyright 1999-2000 Red Hat Software --- All Rights Reserved
@@ -10,7 +10,7 @@
  * option, any later version, incorporated herein by reference.
  *
  * Journal recovery routines for the generic filesystem journaling code;
- * part of the ext2fs journaling system.  
+ * part of the ext2fs journaling system.
  */
 
 #ifndef __KERNEL__
@@ -25,9 +25,9 @@
 
 /*
  * Maintain information about the progress of the recovery job, so that
- * the different passes can carry information between them. 
+ * the different passes can carry information between them.
  */
-struct recovery_info 
+struct recovery_info
 {
 	tid_t		start_transaction;
 	tid_t		end_transaction;
@@ -116,7 +116,7 @@ static int do_readahead(journal_t *journal, unsigned int start)
 	err = 0;
 
 failed:
-	if (nbufs) 
+	if (nbufs)
 		journal_brelse_array(bufs, nbufs);
 	return err;
 }
@@ -128,7 +128,7 @@ failed:
  * Read a block from the journal
  */
 
-static int jread(struct buffer_head **bhp, journal_t *journal, 
+static int jread(struct buffer_head **bhp, journal_t *journal,
 		 unsigned int offset)
 {
 	int err;
@@ -212,14 +212,14 @@ do {									\
 /**
  * journal_recover - recovers a on-disk journal
  * @journal: the journal to recover
- * 
+ *
  * The primary function for recovering the log contents when mounting a
- * journaled device.  
+ * journaled device.
  *
  * Recovery is done in three passes.  In the first pass, we look for the
  * end of the log.  In the second, we assemble the list of revoke
  * blocks.  In the third and final pass, we replay any un-revoked blocks
- * in the log.  
+ * in the log.
  */
 int journal_recover(journal_t *journal)
 {
@@ -231,10 +231,10 @@ int journal_recover(journal_t *journal)
 	memset(&info, 0, sizeof(info));
 	sb = journal->j_superblock;
 
-	/* 
+	/*
 	 * The journal superblock's s_start field (the current log head)
 	 * is always zero if, and only if, the journal was cleanly
-	 * unmounted.  
+	 * unmounted.
 	 */
 
 	if (!sb->s_start) {
@@ -253,7 +253,7 @@ int journal_recover(journal_t *journal)
 	jbd_debug(0, "JBD: recovery, exit status %d, "
 		  "recovered transactions %u to %u\n",
 		  err, info.start_transaction, info.end_transaction);
-	jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", 
+	jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n",
 		  info.nr_replays, info.nr_revoke_hits, info.nr_revokes);
 
 	/* Restart the log at the next transaction ID, thus invalidating
@@ -268,15 +268,15 @@ int journal_recover(journal_t *journal)
 /**
  * journal_skip_recovery - Start journal and wipe exiting records
  * @journal: journal to startup
- * 
+ *
  * Locate any valid recovery information from the journal and set up the
  * journal structures in memory to ignore it (presumably because the
- * caller has evidence that it is out of date).  
+ * caller has evidence that it is out of date).
  * This function does'nt appear to be exorted..
  *
  * We perform one pass over the journal to allow us to tell the user how
  * much recovery information is being erased, and to let us initialise
- * the journal transaction sequence numbers to the next unused ID. 
+ * the journal transaction sequence numbers to the next unused ID.
  */
 int journal_skip_recovery(journal_t *journal)
 {
@@ -297,7 +297,7 @@ int journal_skip_recovery(journal_t *journal)
 #ifdef CONFIG_JBD_DEBUG
 		int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence);
 #endif
-		jbd_debug(0, 
+		jbd_debug(0,
 			  "JBD: ignoring %d transaction%s from the journal.\n",
 			  dropped, (dropped == 1) ? "" : "s");
 		journal->j_transaction_sequence = ++info.end_transaction;
@@ -324,10 +324,10 @@ static int do_one_pass(journal_t *journal,
 	MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t))
 			       / sizeof(journal_block_tag_t));
 
-	/* 
+	/*
 	 * First thing is to establish what we expect to find in the log
 	 * (in terms of transaction IDs), and where (in terms of log
-	 * block offsets): query the superblock.  
+	 * block offsets): query the superblock.
 	 */
 
 	sb = journal->j_superblock;
@@ -344,7 +344,7 @@ static int do_one_pass(journal_t *journal,
 	 * Now we walk through the log, transaction by transaction,
 	 * making sure that each transaction has a commit block in the
 	 * expected place.  Each complete transaction gets replayed back
-	 * into the main filesystem. 
+	 * into the main filesystem.
 	 */
 
 	while (1) {
@@ -379,8 +379,8 @@ static int do_one_pass(journal_t *journal,
 		next_log_block++;
 		wrap(journal, next_log_block);
 
-		/* What kind of buffer is it? 
-		 * 
+		/* What kind of buffer is it?
+		 *
 		 * If it is a descriptor block, check that it has the
 		 * expected sequence number.  Otherwise, we're all done
 		 * here. */
@@ -394,7 +394,7 @@ static int do_one_pass(journal_t *journal,
 
 		blocktype = be32_to_cpu(tmp->h_blocktype);
 		sequence = be32_to_cpu(tmp->h_sequence);
-		jbd_debug(3, "Found magic %d, sequence %d\n", 
+		jbd_debug(3, "Found magic %d, sequence %d\n",
 			  blocktype, sequence);
 
 		if (sequence != next_commit_ID) {
@@ -438,7 +438,7 @@ static int do_one_pass(journal_t *journal,
 					/* Recover what we can, but
 					 * report failure at the end. */
 					success = err;
-					printk (KERN_ERR 
+					printk (KERN_ERR
 						"JBD: IO error %d recovering "
 						"block %ld in log\n",
 						err, io_block);
@@ -452,7 +452,7 @@ static int do_one_pass(journal_t *journal,
 					 * revoked, then we're all done
 					 * here. */
 					if (journal_test_revoke
-					    (journal, blocknr, 
+					    (journal, blocknr,
 					     next_commit_ID)) {
 						brelse(obh);
 						++info->nr_revoke_hits;
@@ -465,7 +465,7 @@ static int do_one_pass(journal_t *journal,
 							blocknr,
 							journal->j_blocksize);
 					if (nbh == NULL) {
-						printk(KERN_ERR 
+						printk(KERN_ERR
 						       "JBD: Out of memory "
 						       "during recovery.\n");
 						err = -ENOMEM;
@@ -537,7 +537,7 @@ static int do_one_pass(journal_t *journal,
 	}
 
  done:
-	/* 
+	/*
 	 * We broke out of the log scan loop: either we came to the
 	 * known end of the log or we found an unexpected block in the
 	 * log.  If the latter happened, then we know that the "current"
@@ -567,7 +567,7 @@ static int do_one_pass(journal_t *journal,
 
 /* Scan a revoke record, marking all blocks mentioned as revoked. */
 
-static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, 
+static int scan_revoke_records(journal_t *journal, struct buffer_head *bh,
 			       tid_t sequence, struct recovery_info *info)
 {
 	journal_revoke_header_t *header;
diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
index a56144183462..c532429d8d9b 100644
--- a/fs/jbd/revoke.c
+++ b/fs/jbd/revoke.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/revoke.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 2000
  *
  * Copyright 2000 Red Hat corp --- All Rights Reserved
@@ -15,10 +15,10 @@
  * Revoke is the mechanism used to prevent old log records for deleted
  * metadata from being replayed on top of newer data using the same
  * blocks.  The revoke mechanism is used in two separate places:
- * 
+ *
  * + Commit: during commit we write the entire list of the current
  *   transaction's revoked blocks to the journal
- * 
+ *
  * + Recovery: during recovery we record the transaction ID of all
  *   revoked blocks.  If there are multiple revoke records in the log
  *   for a single block, only the last one counts, and if there is a log
@@ -29,7 +29,7 @@
  * single transaction:
  *
  * Block is revoked and then journaled:
- *   The desired end result is the journaling of the new block, so we 
+ *   The desired end result is the journaling of the new block, so we
  *   cancel the revoke before the transaction commits.
  *
  * Block is journaled and then revoked:
@@ -41,7 +41,7 @@
  *   transaction must have happened after the block was journaled and so
  *   the revoke must take precedence.
  *
- * Block is revoked and then written as data: 
+ * Block is revoked and then written as data:
  *   The data write is allowed to succeed, but the revoke is _not_
  *   cancelled.  We still need to prevent old log records from
  *   overwriting the new data.  We don't even need to clear the revoke
@@ -54,7 +54,7 @@
  *			buffer has not been revoked, and cancel_revoke
  *			need do nothing.
  * RevokeValid set, Revoked set:
- *			buffer has been revoked.  
+ *			buffer has been revoked.
  */
 
 #ifndef __KERNEL__
@@ -77,7 +77,7 @@ static kmem_cache_t *revoke_table_cache;
    journal replay, this involves recording the transaction ID of the
    last transaction to revoke this block. */
 
-struct jbd_revoke_record_s 
+struct jbd_revoke_record_s
 {
 	struct list_head  hash;
 	tid_t		  sequence;	/* Used for recovery only */
@@ -90,8 +90,8 @@ struct jbd_revoke_table_s
 {
 	/* It is conceivable that we might want a larger hash table
 	 * for recovery.  Must be a power of two. */
-	int		  hash_size; 
-	int		  hash_shift; 
+	int		  hash_size;
+	int		  hash_shift;
 	struct list_head *hash_table;
 };
 
@@ -301,22 +301,22 @@ void journal_destroy_revoke(journal_t *journal)
 
 #ifdef __KERNEL__
 
-/* 
+/*
  * journal_revoke: revoke a given buffer_head from the journal.  This
  * prevents the block from being replayed during recovery if we take a
  * crash after this current transaction commits.  Any subsequent
  * metadata writes of the buffer in this transaction cancel the
- * revoke.  
+ * revoke.
  *
  * Note that this call may block --- it is up to the caller to make
  * sure that there are no further calls to journal_write_metadata
  * before the revoke is complete.  In ext3, this implies calling the
  * revoke before clearing the block bitmap when we are deleting
- * metadata. 
+ * metadata.
  *
  * Revoke performs a journal_forget on any buffer_head passed in as a
  * parameter, but does _not_ forget the buffer_head if the bh was only
- * found implicitly. 
+ * found implicitly.
  *
  * bh_in may not be a journalled buffer - it may have come off
  * the hash tables without an attached journal_head.
@@ -325,7 +325,7 @@ void journal_destroy_revoke(journal_t *journal)
  * by one.
  */
 
-int journal_revoke(handle_t *handle, unsigned long blocknr, 
+int journal_revoke(handle_t *handle, unsigned long blocknr,
 		   struct buffer_head *bh_in)
 {
 	struct buffer_head *bh = NULL;
@@ -487,7 +487,7 @@ void journal_switch_revoke_table(journal_t *journal)
 	else
 		journal->j_revoke = journal->j_revoke_table[0];
 
-	for (i = 0; i < journal->j_revoke->hash_size; i++) 
+	for (i = 0; i < journal->j_revoke->hash_size; i++)
 		INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]);
 }
 
@@ -498,7 +498,7 @@ void journal_switch_revoke_table(journal_t *journal)
  * Called with the journal lock held.
  */
 
-void journal_write_revoke_records(journal_t *journal, 
+void journal_write_revoke_records(journal_t *journal,
 				  transaction_t *transaction)
 {
 	struct journal_head *descriptor;
@@ -507,7 +507,7 @@ void journal_write_revoke_records(journal_t *journal,
 	struct list_head *hash_list;
 	int i, offset, count;
 
-	descriptor = NULL; 
+	descriptor = NULL;
 	offset = 0;
 	count = 0;
 
@@ -519,10 +519,10 @@ void journal_write_revoke_records(journal_t *journal,
 		hash_list = &revoke->hash_table[i];
 
 		while (!list_empty(hash_list)) {
-			record = (struct jbd_revoke_record_s *) 
+			record = (struct jbd_revoke_record_s *)
 				hash_list->next;
 			write_one_revoke_record(journal, transaction,
-						&descriptor, &offset, 
+						&descriptor, &offset,
 						record);
 			count++;
 			list_del(&record->hash);
@@ -534,14 +534,14 @@ void journal_write_revoke_records(journal_t *journal,
 	jbd_debug(1, "Wrote %d revoke records\n", count);
 }
 
-/* 
+/*
  * Write out one revoke record.  We need to create a new descriptor
- * block if the old one is full or if we have not already created one.  
+ * block if the old one is full or if we have not already created one.
  */
 
-static void write_one_revoke_record(journal_t *journal, 
+static void write_one_revoke_record(journal_t *journal,
 				    transaction_t *transaction,
-				    struct journal_head **descriptorp, 
+				    struct journal_head **descriptorp,
 				    int *offsetp,
 				    struct jbd_revoke_record_s *record)
 {
@@ -584,21 +584,21 @@ static void write_one_revoke_record(journal_t *journal,
 		*descriptorp = descriptor;
 	}
 
-	* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = 
+	* ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) =
 		cpu_to_be32(record->blocknr);
 	offset += 4;
 	*offsetp = offset;
 }
 
-/* 
+/*
  * Flush a revoke descriptor out to the journal.  If we are aborting,
  * this is a noop; otherwise we are generating a buffer which needs to
  * be waited for during commit, so it has to go onto the appropriate
  * journal buffer list.
  */
 
-static void flush_descriptor(journal_t *journal, 
-			     struct journal_head *descriptor, 
+static void flush_descriptor(journal_t *journal,
+			     struct journal_head *descriptor,
 			     int offset)
 {
 	journal_revoke_header_t *header;
@@ -618,7 +618,7 @@ static void flush_descriptor(journal_t *journal,
 }
 #endif
 
-/* 
+/*
  * Revoke support for recovery.
  *
  * Recovery needs to be able to:
@@ -629,7 +629,7 @@ static void flush_descriptor(journal_t *journal,
  *  check whether a given block in a given transaction should be replayed
  *  (ie. has not been revoked by a revoke record in that or a subsequent
  *  transaction)
- * 
+ *
  *  empty the revoke table after recovery.
  */
 
@@ -637,11 +637,11 @@ static void flush_descriptor(journal_t *journal,
  * First, setting revoke records.  We create a new revoke record for
  * every block ever revoked in the log as we scan it for recovery, and
  * we update the existing records if we find multiple revokes for a
- * single block. 
+ * single block.
  */
 
-int journal_set_revoke(journal_t *journal, 
-		       unsigned long blocknr, 
+int journal_set_revoke(journal_t *journal,
+		       unsigned long blocknr,
 		       tid_t sequence)
 {
 	struct jbd_revoke_record_s *record;
@@ -653,18 +653,18 @@ int journal_set_revoke(journal_t *journal,
 		if (tid_gt(sequence, record->sequence))
 			record->sequence = sequence;
 		return 0;
-	} 
+	}
 	return insert_revoke_hash(journal, blocknr, sequence);
 }
 
-/* 
+/*
  * Test revoke records.  For a given block referenced in the log, has
  * that block been revoked?  A revoke record with a given transaction
  * sequence number revokes all blocks in that transaction and earlier
  * ones, but later transactions still need replayed.
  */
 
-int journal_test_revoke(journal_t *journal, 
+int journal_test_revoke(journal_t *journal,
 			unsigned long blocknr,
 			tid_t sequence)
 {
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index f5169a96260e..bf7fd7117817 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -1,6 +1,6 @@
 /*
  * linux/fs/transaction.c
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>, 1998
  *
  * Copyright 1998 Red Hat corp --- All Rights Reserved
@@ -10,7 +10,7 @@
  * option, any later version, incorporated herein by reference.
  *
  * Generic filesystem transaction handling code; part of the ext2fs
- * journaling system.  
+ * journaling system.
  *
  * This file manages transactions (compound commits managed by the
  * journaling code) and handles (individual atomic operations by the
@@ -74,7 +74,7 @@ get_transaction(journal_t *journal, transaction_t *transaction)
  * start_this_handle: Given a handle, deal with any locking or stalling
  * needed to make sure that there is enough journal space for the handle
  * to begin.  Attach the handle to a transaction and set up the
- * transaction's buffer credits.  
+ * transaction's buffer credits.
  */
 
 static int start_this_handle(journal_t *journal, handle_t *handle)
@@ -117,7 +117,7 @@ repeat_locked:
 	if (is_journal_aborted(journal) ||
 	    (journal->j_errno != 0 && !(journal->j_flags & JFS_ACK_ERR))) {
 		spin_unlock(&journal->j_state_lock);
-		ret = -EROFS; 
+		ret = -EROFS;
 		goto out;
 	}
 
@@ -182,7 +182,7 @@ repeat_locked:
 		goto repeat;
 	}
 
-	/* 
+	/*
 	 * The commit code assumes that it can get enough log space
 	 * without forcing a checkpoint.  This is *critical* for
 	 * correctness: a checkpoint of a buffer which is also
@@ -191,7 +191,7 @@ repeat_locked:
 	 *
 	 * We must therefore ensure the necessary space in the journal
 	 * *before* starting to dirty potentially checkpointed buffers
-	 * in the new transaction. 
+	 * in the new transaction.
 	 *
 	 * The worst part is, any transaction currently committing can
 	 * reduce the free space arbitrarily.  Be careful to account for
@@ -246,13 +246,13 @@ static handle_t *new_handle(int nblocks)
 }
 
 /**
- * handle_t *journal_start() - Obtain a new handle.  
+ * handle_t *journal_start() - Obtain a new handle.
  * @journal: Journal to start transaction on.
  * @nblocks: number of block buffer we might modify
  *
  * We make sure that the transaction can guarantee at least nblocks of
  * modified buffers in the log.  We block until the log can guarantee
- * that much space.  
+ * that much space.
  *
  * This function is visible to journal users (like ext3fs), so is not
  * called with the journal already locked.
@@ -292,11 +292,11 @@ handle_t *journal_start(journal_t *journal, int nblocks)
  * int journal_extend() - extend buffer credits.
  * @handle:  handle to 'extend'
  * @nblocks: nr blocks to try to extend by.
- * 
+ *
  * Some transactions, such as large extends and truncates, can be done
  * atomically all at once or in several stages.  The operation requests
  * a credit for a number of buffer modications in advance, but can
- * extend its credit if it needs more.  
+ * extend its credit if it needs more.
  *
  * journal_extend tries to give the running handle more buffer credits.
  * It does not guarantee that allocation - this is a best-effort only.
@@ -363,7 +363,7 @@ out:
  * int journal_restart() - restart a handle .
  * @handle:  handle to restart
  * @nblocks: nr credits requested
- * 
+ *
  * Restart a handle for a multi-transaction filesystem
  * operation.
  *
@@ -462,7 +462,7 @@ void journal_lock_updates(journal_t *journal)
 /**
  * void journal_unlock_updates (journal_t* journal) - release barrier
  * @journal:  Journal to release the barrier on.
- * 
+ *
  * Release a transaction barrier obtained with journal_lock_updates().
  *
  * Should be called without the journal lock held.
@@ -547,8 +547,8 @@ repeat:
 	jbd_lock_bh_state(bh);
 
 	/* We now hold the buffer lock so it is safe to query the buffer
-	 * state.  Is the buffer dirty? 
-	 * 
+	 * state.  Is the buffer dirty?
+	 *
 	 * If so, there are two possibilities.  The buffer may be
 	 * non-journaled, and undergoing a quite legitimate writeback.
 	 * Otherwise, it is journaled, and we don't expect dirty buffers
@@ -566,7 +566,7 @@ repeat:
 		 */
 		if (jh->b_transaction) {
 			J_ASSERT_JH(jh,
-				jh->b_transaction == transaction || 
+				jh->b_transaction == transaction ||
 				jh->b_transaction ==
 					journal->j_committing_transaction);
 			if (jh->b_next_transaction)
@@ -653,7 +653,7 @@ repeat:
 		 * buffer had better remain locked during the kmalloc,
 		 * but that should be true --- we hold the journal lock
 		 * still and the buffer is already on the BUF_JOURNAL
-		 * list so won't be flushed. 
+		 * list so won't be flushed.
 		 *
 		 * Subtle point, though: if this is a get_undo_access,
 		 * then we will be relying on the frozen_data to contain
@@ -765,8 +765,8 @@ int journal_get_write_access(handle_t *handle, struct buffer_head *bh)
  * manually rather than reading off disk), then we need to keep the
  * buffer_head locked until it has been completely filled with new
  * data.  In this case, we should be able to make the assertion that
- * the bh is not already part of an existing transaction.  
- * 
+ * the bh is not already part of an existing transaction.
+ *
  * The buffer should already be locked by the caller by this point.
  * There is no lock ranking violation: it was a newly created,
  * unlocked buffer beforehand. */
@@ -778,7 +778,7 @@ int journal_get_write_access(handle_t *handle, struct buffer_head *bh)
  *
  * Call this if you create a new bh.
  */
-int journal_get_create_access(handle_t *handle, struct buffer_head *bh) 
+int journal_get_create_access(handle_t *handle, struct buffer_head *bh)
 {
 	transaction_t *transaction = handle->h_transaction;
 	journal_t *journal = transaction->t_journal;
@@ -847,13 +847,13 @@ out:
  * do not reuse freed space until the deallocation has been committed,
  * since if we overwrote that space we would make the delete
  * un-rewindable in case of a crash.
- * 
+ *
  * To deal with that, journal_get_undo_access requests write access to a
  * buffer for parts of non-rewindable operations such as delete
  * operations on the bitmaps.  The journaling code must keep a copy of
  * the buffer's contents prior to the undo_access call until such time
  * as we know that the buffer has definitely been committed to disk.
- * 
+ *
  * We never need to know which transaction the committed data is part
  * of, buffers touched here are guaranteed to be dirtied later and so
  * will be committed to a new transaction in due course, at which point
@@ -911,13 +911,13 @@ out:
 	return err;
 }
 
-/** 
+/**
  * int journal_dirty_data() -  mark a buffer as containing dirty data which
  *                             needs to be flushed before we can commit the
- *                             current transaction.  
+ *                             current transaction.
  * @handle: transaction
  * @bh: bufferhead to mark
- * 
+ *
  * The buffer is placed on the transaction's data list and is marked as
  * belonging to the transaction.
  *
@@ -946,15 +946,15 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh)
 
 	/*
 	 * What if the buffer is already part of a running transaction?
-	 * 
+	 *
 	 * There are two cases:
 	 * 1) It is part of the current running transaction.  Refile it,
 	 *    just in case we have allocated it as metadata, deallocated
-	 *    it, then reallocated it as data. 
+	 *    it, then reallocated it as data.
 	 * 2) It is part of the previous, still-committing transaction.
 	 *    If all we want to do is to guarantee that the buffer will be
 	 *    written to disk before this new transaction commits, then
-	 *    being sure that the *previous* transaction has this same 
+	 *    being sure that the *previous* transaction has this same
 	 *    property is sufficient for us!  Just leave it on its old
 	 *    transaction.
 	 *
@@ -1076,18 +1076,18 @@ no_journal:
 	return 0;
 }
 
-/** 
+/**
  * int journal_dirty_metadata() -  mark a buffer as containing dirty metadata
  * @handle: transaction to add buffer to.
- * @bh: buffer to mark 
- * 
+ * @bh: buffer to mark
+ *
  * mark dirty metadata which needs to be journaled as part of the current
  * transaction.
  *
  * The buffer is placed on the transaction's metadata list and is marked
- * as belonging to the transaction.  
+ * as belonging to the transaction.
  *
- * Returns error number or 0 on success.  
+ * Returns error number or 0 on success.
  *
  * Special care needs to be taken if the buffer already belongs to the
  * current committing transaction (in which case we should have frozen
@@ -1135,11 +1135,11 @@ int journal_dirty_metadata(handle_t *handle, struct buffer_head *bh)
 
 	set_buffer_jbddirty(bh);
 
-	/* 
+	/*
 	 * Metadata already on the current transaction list doesn't
 	 * need to be filed.  Metadata on another transaction's list must
 	 * be committing, and will be refiled once the commit completes:
-	 * leave it alone for now. 
+	 * leave it alone for now.
 	 */
 	if (jh->b_transaction != transaction) {
 		JBUFFER_TRACE(jh, "already on other transaction");
@@ -1165,7 +1165,7 @@ out:
 	return 0;
 }
 
-/* 
+/*
  * journal_release_buffer: undo a get_write_access without any buffer
  * updates, if the update decided in the end that it didn't need access.
  *
@@ -1176,20 +1176,20 @@ journal_release_buffer(handle_t *handle, struct buffer_head *bh)
 	BUFFER_TRACE(bh, "entry");
 }
 
-/** 
+/**
  * void journal_forget() - bforget() for potentially-journaled buffers.
  * @handle: transaction handle
  * @bh:     bh to 'forget'
  *
  * We can only do the bforget if there are no commits pending against the
  * buffer.  If the buffer is dirty in the current running transaction we
- * can safely unlink it. 
+ * can safely unlink it.
  *
  * bh may not be a journalled buffer at all - it may be a non-JBD
  * buffer which came off the hashtable.  Check for this.
  *
  * Decrements bh->b_count by one.
- * 
+ *
  * Allow this call even if the handle has aborted --- it may be part of
  * the caller's cleanup after an abort.
  */
@@ -1237,7 +1237,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 
 		drop_reserve = 1;
 
-		/* 
+		/*
 		 * We are no longer going to journal this buffer.
 		 * However, the commit of this transaction is still
 		 * important to the buffer: the delete that we are now
@@ -1246,7 +1246,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 		 *
 		 * So, if we have a checkpoint on the buffer, we should
 		 * now refile the buffer on our BJ_Forget list so that
-		 * we know to remove the checkpoint after we commit. 
+		 * we know to remove the checkpoint after we commit.
 		 */
 
 		if (jh->b_cp_transaction) {
@@ -1264,7 +1264,7 @@ int journal_forget (handle_t *handle, struct buffer_head *bh)
 			}
 		}
 	} else if (jh->b_transaction) {
-		J_ASSERT_JH(jh, (jh->b_transaction == 
+		J_ASSERT_JH(jh, (jh->b_transaction ==
 				 journal->j_committing_transaction));
 		/* However, if the buffer is still owned by a prior
 		 * (committing) transaction, we can't drop it yet... */
@@ -1294,7 +1294,7 @@ drop:
 /**
  * int journal_stop() - complete a transaction
  * @handle: tranaction to complete.
- * 
+ *
  * All done for a particular handle.
  *
  * There is not much action needed here.  We just return any remaining
@@ -1303,7 +1303,7 @@ drop:
  * filesystem is marked for synchronous update.
  *
  * journal_stop itself will not usually return an error, but it may
- * do so in unusual circumstances.  In particular, expect it to 
+ * do so in unusual circumstances.  In particular, expect it to
  * return -EIO if a journal_abort has been executed since the
  * transaction began.
  */
@@ -1388,7 +1388,7 @@ int journal_stop(handle_t *handle)
 
 		/*
 		 * Special case: JFS_SYNC synchronous updates require us
-		 * to wait for the commit to complete.  
+		 * to wait for the commit to complete.
 		 */
 		if (handle->h_sync && !(current->flags & PF_MEMALLOC))
 			err = log_wait_commit(journal, tid);
@@ -1439,7 +1439,7 @@ int journal_force_commit(journal_t *journal)
  * jbd_lock_bh_state(jh2bh(jh)) is held.
  */
 
-static inline void 
+static inline void
 __blist_add_buffer(struct journal_head **list, struct journal_head *jh)
 {
 	if (!*list) {
@@ -1454,7 +1454,7 @@ __blist_add_buffer(struct journal_head **list, struct journal_head *jh)
 	}
 }
 
-/* 
+/*
  * Remove a buffer from a transaction list, given the transaction's list
  * head pointer.
  *
@@ -1475,7 +1475,7 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh)
 	jh->b_tnext->b_tprev = jh->b_tprev;
 }
 
-/* 
+/*
  * Remove a buffer from the appropriate transaction list.
  *
  * Note that this function can *change* the value of
@@ -1595,17 +1595,17 @@ out:
 }
 
 
-/** 
+/**
  * int journal_try_to_free_buffers() - try to free page buffers.
  * @journal: journal for operation
  * @page: to try and free
  * @unused_gfp_mask: unused
  *
- * 
+ *
  * For all the buffers on this page,
  * if they are fully written out ordered data, move them onto BUF_CLEAN
  * so try_to_free_buffers() can reap them.
- * 
+ *
  * This function returns non-zero if we wish try_to_free_buffers()
  * to be called. We do this if the page is releasable by try_to_free_buffers().
  * We also do it if the page has locked or dirty buffers and the caller wants
@@ -1629,7 +1629,7 @@ out:
  * cannot happen because we never reallocate freed data as metadata
  * while the data is part of a transaction.  Yes?
  */
-int journal_try_to_free_buffers(journal_t *journal, 
+int journal_try_to_free_buffers(journal_t *journal,
 				struct page *page, gfp_t unused_gfp_mask)
 {
 	struct buffer_head *head;
@@ -1697,7 +1697,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
 }
 
 /*
- * journal_invalidatepage 
+ * journal_invalidatepage
  *
  * This code is tricky.  It has a number of cases to deal with.
  *
@@ -1705,15 +1705,15 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
  *
  * i_size must be updated on disk before we start calling invalidatepage on the
  * data.
- * 
+ *
  *  This is done in ext3 by defining an ext3_setattr method which
  *  updates i_size before truncate gets going.  By maintaining this
  *  invariant, we can be sure that it is safe to throw away any buffers
  *  attached to the current transaction: once the transaction commits,
  *  we know that the data will not be needed.
- * 
+ *
  *  Note however that we can *not* throw away data belonging to the
- *  previous, committing transaction!  
+ *  previous, committing transaction!
  *
  * Any disk blocks which *are* part of the previous, committing
  * transaction (and which therefore cannot be discarded immediately) are
@@ -1732,7 +1732,7 @@ static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction)
  * don't make guarantees about the order in which data hits disk --- in
  * particular we don't guarantee that new dirty data is flushed before
  * transaction commit --- so it is always safe just to discard data
- * immediately in that mode.  --sct 
+ * immediately in that mode.  --sct
  */
 
 /*
@@ -1876,9 +1876,9 @@ zap_buffer_unlocked:
 	return may_free;
 }
 
-/** 
+/**
  * void journal_invalidatepage()
- * @journal: journal to use for flush... 
+ * @journal: journal to use for flush...
  * @page:    page to flush
  * @offset:  length of page to invalidate.
  *
@@ -1886,7 +1886,7 @@ zap_buffer_unlocked:
  *
  */
 void journal_invalidatepage(journal_t *journal,
-		      struct page *page, 
+		      struct page *page,
 		      unsigned long offset)
 {
 	struct buffer_head *head, *bh, *next;
@@ -1924,8 +1924,8 @@ void journal_invalidatepage(journal_t *journal,
 	}
 }
 
-/* 
- * File a buffer on the given transaction list. 
+/*
+ * File a buffer on the given transaction list.
  */
 void __journal_file_buffer(struct journal_head *jh,
 			transaction_t *transaction, int jlist)
@@ -1948,7 +1948,7 @@ void __journal_file_buffer(struct journal_head *jh,
 	 * with __jbd_unexpected_dirty_buffer()'s handling of dirty
 	 * state. */
 
-	if (jlist == BJ_Metadata || jlist == BJ_Reserved || 
+	if (jlist == BJ_Metadata || jlist == BJ_Reserved ||
 	    jlist == BJ_Shadow || jlist == BJ_Forget) {
 		if (test_clear_buffer_dirty(bh) ||
 		    test_clear_buffer_jbddirty(bh))
@@ -2008,7 +2008,7 @@ void journal_file_buffer(struct journal_head *jh,
 	jbd_unlock_bh_state(jh2bh(jh));
 }
 
-/* 
+/*
  * Remove a buffer from its current buffer list in preparation for
  * dropping it from its current transaction entirely.  If the buffer has
  * already started to be used by a subsequent transaction, refile the
@@ -2060,7 +2060,7 @@ void __journal_refile_buffer(struct journal_head *jh)
  * to the caller to remove the journal_head if necessary.  For the
  * unlocked journal_refile_buffer call, the caller isn't going to be
  * doing anything else to the buffer so we need to do the cleanup
- * ourselves to avoid a jh leak. 
+ * ourselves to avoid a jh leak.
  *
  * *** The journal_head may be freed by this call! ***
  */
diff --git a/include/linux/ext3_jbd.h b/include/linux/ext3_jbd.h
index c8307c02dd07..ce0e6109aff0 100644
--- a/include/linux/ext3_jbd.h
+++ b/include/linux/ext3_jbd.h
@@ -23,7 +23,7 @@
 
 /* Define the number of blocks we need to account to a transaction to
  * modify one block of data.
- * 
+ *
  * We may have to touch one inode, one bitmap buffer, up to three
  * indirection blocks, the group and superblock summaries, and the data
  * block to complete the transaction.  */
@@ -88,16 +88,16 @@
 #endif
 
 int
-ext3_mark_iloc_dirty(handle_t *handle, 
+ext3_mark_iloc_dirty(handle_t *handle,
 		     struct inode *inode,
 		     struct ext3_iloc *iloc);
 
-/* 
+/*
  * On success, We end up with an outstanding reference count against
- * iloc->bh.  This _must_ be cleaned up later. 
+ * iloc->bh.  This _must_ be cleaned up later.
  */
 
-int ext3_reserve_inode_write(handle_t *handle, struct inode *inode, 
+int ext3_reserve_inode_write(handle_t *handle, struct inode *inode,
 			struct ext3_iloc *iloc);
 
 int ext3_mark_inode_dirty(handle_t *handle, struct inode *inode);
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index a04c154c5207..7d847931ee5a 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -1,6 +1,6 @@
 /*
  * linux/include/linux/jbd.h
- * 
+ *
  * Written by Stephen C. Tweedie <sct@redhat.com>
  *
  * Copyright 1998-2000 Red Hat, Inc --- All Rights Reserved
@@ -97,8 +97,8 @@ extern void jbd_slab_free(void *ptr, size_t size);
  * number of outstanding buffers possible at any time.  When the
  * operation completes, any buffer credits not used are credited back to
  * the transaction, so that at all times we know how many buffers the
- * outstanding updates on a transaction might possibly touch. 
- * 
+ * outstanding updates on a transaction might possibly touch.
+ *
  * This is an opaque datatype.
  **/
 typedef struct handle_s		handle_t;	/* Atomic operation type */
@@ -108,7 +108,7 @@ typedef struct handle_s		handle_t;	/* Atomic operation type */
  * typedef journal_t - The journal_t maintains all of the journaling state information for a single filesystem.
  *
  * journal_t is linked to from the fs superblock structure.
- * 
+ *
  * We use the journal_t to keep track of all outstanding transaction
  * activity on the filesystem, and to manage the state of the log
  * writing process.
@@ -128,7 +128,7 @@ typedef struct journal_s	journal_t;	/* Journal control structure */
  * On-disk structures
  */
 
-/* 
+/*
  * Descriptor block types:
  */
 
@@ -149,8 +149,8 @@ typedef struct journal_header_s
 } journal_header_t;
 
 
-/* 
- * The block tag: used to describe a single buffer in the journal 
+/*
+ * The block tag: used to describe a single buffer in the journal
  */
 typedef struct journal_block_tag_s
 {
@@ -158,9 +158,9 @@ typedef struct journal_block_tag_s
 	__be32		t_flags;	/* See below */
 } journal_block_tag_t;
 
-/* 
+/*
  * The revoke descriptor: used on disk to describe a series of blocks to
- * be revoked from the log 
+ * be revoked from the log
  */
 typedef struct journal_revoke_header_s
 {
@@ -374,10 +374,10 @@ struct jbd_revoke_table_s;
  **/
 
 /* Docbook can't yet cope with the bit fields, but will leave the documentation
- * in so it can be fixed later. 
+ * in so it can be fixed later.
  */
 
-struct handle_s 
+struct handle_s
 {
 	/* Which compound transaction is this update a part of? */
 	transaction_t		*h_transaction;
@@ -435,7 +435,7 @@ struct handle_s
  *
  */
 
-struct transaction_s 
+struct transaction_s
 {
 	/* Pointer to the journal for this transaction. [no locking] */
 	journal_t		*t_journal;
@@ -455,7 +455,7 @@ struct transaction_s
 		T_RUNDOWN,
 		T_FLUSH,
 		T_COMMIT,
-		T_FINISHED 
+		T_FINISHED
 	}			t_state;
 
 	/*
@@ -569,7 +569,7 @@ struct transaction_s
  *     journal_t.
  * @j_flags:  General journaling state flags
  * @j_errno:  Is there an outstanding uncleared error on the journal (from a
- *     prior abort)? 
+ *     prior abort)?
  * @j_sb_buffer: First part of superblock buffer
  * @j_superblock: Second part of superblock buffer
  * @j_format_version: Version of the superblock format
@@ -583,7 +583,7 @@ struct transaction_s
  * @j_wait_transaction_locked: Wait queue for waiting for a locked transaction
  *  to start committing, or for a barrier lock to be released
  * @j_wait_logspace: Wait queue for waiting for checkpointing to complete
- * @j_wait_done_commit: Wait queue for waiting for commit to complete 
+ * @j_wait_done_commit: Wait queue for waiting for commit to complete
  * @j_wait_checkpoint:  Wait queue to trigger checkpointing
  * @j_wait_commit: Wait queue to trigger commit
  * @j_wait_updates: Wait queue to wait for updates to complete
@@ -592,7 +592,7 @@ struct transaction_s
  * @j_tail: Journal tail - identifies the oldest still-used block in the
  *  journal.
  * @j_free: Journal free - how many free blocks are there in the journal?
- * @j_first: The block number of the first usable block 
+ * @j_first: The block number of the first usable block
  * @j_last: The block number one beyond the last usable block
  * @j_dev: Device where we store the journal
  * @j_blocksize: blocksize for the location where we store the journal.
@@ -604,12 +604,12 @@ struct transaction_s
  * @j_list_lock: Protects the buffer lists and internal buffer state.
  * @j_inode: Optional inode where we store the journal.  If present, all journal
  *     block numbers are mapped into this inode via bmap().
- * @j_tail_sequence:  Sequence number of the oldest transaction in the log 
+ * @j_tail_sequence:  Sequence number of the oldest transaction in the log
  * @j_transaction_sequence: Sequence number of the next transaction to grant
  * @j_commit_sequence: Sequence number of the most recently committed
  *  transaction
  * @j_commit_request: Sequence number of the most recent transaction wanting
- *     commit 
+ *     commit
  * @j_uuid: Uuid of client object.
  * @j_task: Pointer to the current commit thread for this journal
  * @j_max_transaction_buffers:  Maximum number of metadata buffers to allow in a
@@ -823,8 +823,8 @@ struct journal_s
 	void *j_private;
 };
 
-/* 
- * Journal flag definitions 
+/*
+ * Journal flag definitions
  */
 #define JFS_UNMOUNT	0x001	/* Journal thread is being destroyed */
 #define JFS_ABORT	0x002	/* Journaling has been aborted for errors. */
@@ -833,7 +833,7 @@ struct journal_s
 #define JFS_LOADED	0x010	/* The journal superblock has been loaded */
 #define JFS_BARRIER	0x020	/* Use IDE barriers */
 
-/* 
+/*
  * Function declarations for the journaling transaction and buffer
  * management
  */
@@ -862,7 +862,7 @@ int __journal_remove_checkpoint(struct journal_head *);
 void __journal_insert_checkpoint(struct journal_head *, transaction_t *);
 
 /* Buffer IO */
-extern int 
+extern int
 journal_write_metadata_buffer(transaction_t	  *transaction,
 			      struct journal_head  *jh_in,
 			      struct journal_head **jh_out,
@@ -890,7 +890,7 @@ static inline handle_t *journal_current_handle(void)
 /* The journaling code user interface:
  *
  * Create and destroy handles
- * Register buffer modifications against the current transaction. 
+ * Register buffer modifications against the current transaction.
  */
 
 extern handle_t *journal_start(journal_t *, int nblocks);
@@ -917,11 +917,11 @@ extern journal_t * journal_init_dev(struct block_device *bdev,
 				int start, int len, int bsize);
 extern journal_t * journal_init_inode (struct inode *);
 extern int	   journal_update_format (journal_t *);
-extern int	   journal_check_used_features 
+extern int	   journal_check_used_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
-extern int	   journal_check_available_features 
+extern int	   journal_check_available_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
-extern int	   journal_set_features 
+extern int	   journal_set_features
 		   (journal_t *, unsigned long, unsigned long, unsigned long);
 extern int	   journal_create     (journal_t *);
 extern int	   journal_load       (journal_t *journal);
@@ -1015,7 +1015,7 @@ do {								           \
  * bit, when set, indicates that we have had a fatal error somewhere,
  * either inside the journaling layer or indicated to us by the client
  * (eg. ext3), and that we and should not commit any further
- * transactions.  
+ * transactions.
  */
 
 static inline int is_journal_aborted(journal_t *journal)
@@ -1082,7 +1082,7 @@ static inline int jbd_space_needed(journal_t *journal)
 #define BJ_Reserved	7	/* Buffer is reserved for access by journal */
 #define BJ_Locked	8	/* Locked for I/O during commit */
 #define BJ_Types	9
- 
+
 extern int jbd_blocks_per_page(struct inode *inode);
 
 #ifdef __KERNEL__
-- 
cgit v1.2.1


From 2aed3484695ecb03f1395bb62f1099e8b0826124 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 01:49:28 -0700
Subject: [PATCH] jbd: use BUILD_BUG_ON in journal init

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Stephen Tweedie <sct@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/journal.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 87c5a6d00805..2613fca92740 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -2047,13 +2047,7 @@ static int __init journal_init(void)
 {
 	int ret;
 
-/* Static check for data structure consistency.  There's no code
- * invoked --- we'll just get a linker failure if things aren't right.
- */
-	extern void journal_bad_superblock_size(void);
-	if (sizeof(struct journal_superblock_s) != 1024)
-		journal_bad_superblock_size();
-
+	BUILD_BUG_ON(sizeof(struct journal_superblock_s) != 1024);
 
 	ret = journal_init_caches();
 	if (ret != 0)
-- 
cgit v1.2.1


From 855565e81ad8940cc645b5110ec2c7f124a76d23 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:29 -0700
Subject: [PATCH] fix ext3 mounts at 16T

I need to do some actual IO testing now, but this gets things mounting for
a 16T ext3 filesystem.  (patched up e2fsprogs is needed too, I'll send that
off the kernel list)

This patch fixes these issues in the kernel:

o sbi->s_groups_count overflows in ext3_fill_super()

	sbi->s_groups_count = (le32_to_cpu(es->s_blocks_count) -
			       le32_to_cpu(es->s_first_data_block) +
			       EXT3_BLOCKS_PER_GROUP(sb) - 1) /
			      EXT3_BLOCKS_PER_GROUP(sb);

  at 16T, s_blocks_count is already maxed out; adding
  EXT3_BLOCKS_PER_GROUP(sb) overflows it and groups_count comes out to 0.
  Not really what we want, and causes a failed mount.

  Feel free to check my math (actually, please do!), but changing it this
  way should work & avoid the overflow:

  (A + B - 1)/B changed to: ((A - 1)/B) + 1

o ext3_check_descriptors() overflows range checks

  ext3_check_descriptors() iterates over all block groups making sure
  that various bits are within the right block ranges...  on the last pass
  through, it is checking the error case

   [item] >= block + EXT3_BLOCKS_PER_GROUP(sb)

  where "block" is the first block in the last block group.  The last
  block in this group (and the last one that will fit in 32 bits) is block
  + EXT3_BLOCKS_PER_GROUP(sb)- 1.  block + EXT3_BLOCKS_PER_GROUP(sb) wraps
  back around to 0.

  so, make things clearer with "first_block" and "last_block" where those
  are first and last, inclusive, and use <, > rather than <, >=.

  Finally, the last block group may be smaller than the rest, so account
  for this on the last pass through: last_block = sb->s_blocks_count - 1;

(a similar patch could be done for ext2; does anyone in their right mind
use ext2 at 16T?  I'll send an ext2 patch doing the same thing if that's
warranted)

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Cc: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/super.c | 34 +++++++++++++++++++---------------
 1 file changed, 19 insertions(+), 15 deletions(-)

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 4b95bfe4c8f7..762dff7b56fb 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1174,7 +1174,8 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es,
 static int ext3_check_descriptors (struct super_block * sb)
 {
 	struct ext3_sb_info *sbi = EXT3_SB(sb);
-	ext3_fsblk_t block = le32_to_cpu(sbi->s_es->s_first_data_block);
+	ext3_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block);
+	ext3_fsblk_t last_block;
 	struct ext3_group_desc * gdp = NULL;
 	int desc_block = 0;
 	int i;
@@ -1183,12 +1184,17 @@ static int ext3_check_descriptors (struct super_block * sb)
 
 	for (i = 0; i < sbi->s_groups_count; i++)
 	{
+		if (i == sbi->s_groups_count - 1)
+			last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1;
+		else
+			last_block = first_block +
+				(EXT3_BLOCKS_PER_GROUP(sb) - 1);
+
 		if ((i % EXT3_DESC_PER_BLOCK(sb)) == 0)
 			gdp = (struct ext3_group_desc *)
 					sbi->s_group_desc[desc_block++]->b_data;
-		if (le32_to_cpu(gdp->bg_block_bitmap) < block ||
-		    le32_to_cpu(gdp->bg_block_bitmap) >=
-				block + EXT3_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_block_bitmap) < first_block ||
+		    le32_to_cpu(gdp->bg_block_bitmap) > last_block)
 		{
 			ext3_error (sb, "ext3_check_descriptors",
 				    "Block bitmap for group %d"
@@ -1197,9 +1203,8 @@ static int ext3_check_descriptors (struct super_block * sb)
 					le32_to_cpu(gdp->bg_block_bitmap));
 			return 0;
 		}
-		if (le32_to_cpu(gdp->bg_inode_bitmap) < block ||
-		    le32_to_cpu(gdp->bg_inode_bitmap) >=
-				block + EXT3_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_inode_bitmap) < first_block ||
+		    le32_to_cpu(gdp->bg_inode_bitmap) > last_block)
 		{
 			ext3_error (sb, "ext3_check_descriptors",
 				    "Inode bitmap for group %d"
@@ -1208,9 +1213,9 @@ static int ext3_check_descriptors (struct super_block * sb)
 					le32_to_cpu(gdp->bg_inode_bitmap));
 			return 0;
 		}
-		if (le32_to_cpu(gdp->bg_inode_table) < block ||
-		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >=
-		    block + EXT3_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_inode_table) < first_block ||
+		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >
+		    last_block)
 		{
 			ext3_error (sb, "ext3_check_descriptors",
 				    "Inode table for group %d"
@@ -1219,7 +1224,7 @@ static int ext3_check_descriptors (struct super_block * sb)
 					le32_to_cpu(gdp->bg_inode_table));
 			return 0;
 		}
-		block += EXT3_BLOCKS_PER_GROUP(sb);
+		first_block += EXT3_BLOCKS_PER_GROUP(sb);
 		gdp++;
 	}
 
@@ -1622,10 +1627,9 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 
 	if (EXT3_BLOCKS_PER_GROUP(sb) == 0)
 		goto cantfind_ext3;
-	sbi->s_groups_count = (le32_to_cpu(es->s_blocks_count) -
-			       le32_to_cpu(es->s_first_data_block) +
-			       EXT3_BLOCKS_PER_GROUP(sb) - 1) /
-			      EXT3_BLOCKS_PER_GROUP(sb);
+	sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) -
+			       le32_to_cpu(es->s_first_data_block) - 1)
+				       / EXT3_BLOCKS_PER_GROUP(sb)) + 1;
 	db_count = (sbi->s_groups_count + EXT3_DESC_PER_BLOCK(sb) - 1) /
 		   EXT3_DESC_PER_BLOCK(sb);
 	sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *),
-- 
cgit v1.2.1


From 41f04d852e359582518f950d12b2287766613022 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:30 -0700
Subject: [PATCH] ext2: fix mounts at 16T

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext2/super.c | 32 +++++++++++++++++++-------------
 1 file changed, 19 insertions(+), 13 deletions(-)

diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 4286ff6330b6..d978d3febb8c 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -544,17 +544,24 @@ static int ext2_check_descriptors (struct super_block * sb)
 	int i;
 	int desc_block = 0;
 	struct ext2_sb_info *sbi = EXT2_SB(sb);
-	unsigned long block = le32_to_cpu(sbi->s_es->s_first_data_block);
+	unsigned long first_block = le32_to_cpu(sbi->s_es->s_first_data_block);
+	unsigned long last_block;
 	struct ext2_group_desc * gdp = NULL;
 
 	ext2_debug ("Checking group descriptors");
 
 	for (i = 0; i < sbi->s_groups_count; i++)
 	{
+		if (i == sbi->s_groups_count - 1)
+			last_block = le32_to_cpu(sbi->s_es->s_blocks_count) - 1;
+		else
+			last_block = first_block +
+				(EXT2_BLOCKS_PER_GROUP(sb) - 1);
+
 		if ((i % EXT2_DESC_PER_BLOCK(sb)) == 0)
 			gdp = (struct ext2_group_desc *) sbi->s_group_desc[desc_block++]->b_data;
-		if (le32_to_cpu(gdp->bg_block_bitmap) < block ||
-		    le32_to_cpu(gdp->bg_block_bitmap) >= block + EXT2_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_block_bitmap) < first_block ||
+		    le32_to_cpu(gdp->bg_block_bitmap) > last_block)
 		{
 			ext2_error (sb, "ext2_check_descriptors",
 				    "Block bitmap for group %d"
@@ -562,8 +569,8 @@ static int ext2_check_descriptors (struct super_block * sb)
 				    i, (unsigned long) le32_to_cpu(gdp->bg_block_bitmap));
 			return 0;
 		}
-		if (le32_to_cpu(gdp->bg_inode_bitmap) < block ||
-		    le32_to_cpu(gdp->bg_inode_bitmap) >= block + EXT2_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_inode_bitmap) < first_block ||
+		    le32_to_cpu(gdp->bg_inode_bitmap) > last_block)
 		{
 			ext2_error (sb, "ext2_check_descriptors",
 				    "Inode bitmap for group %d"
@@ -571,9 +578,9 @@ static int ext2_check_descriptors (struct super_block * sb)
 				    i, (unsigned long) le32_to_cpu(gdp->bg_inode_bitmap));
 			return 0;
 		}
-		if (le32_to_cpu(gdp->bg_inode_table) < block ||
-		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >=
-		    block + EXT2_BLOCKS_PER_GROUP(sb))
+		if (le32_to_cpu(gdp->bg_inode_table) < first_block ||
+		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >
+		    last_block)
 		{
 			ext2_error (sb, "ext2_check_descriptors",
 				    "Inode table for group %d"
@@ -581,7 +588,7 @@ static int ext2_check_descriptors (struct super_block * sb)
 				    i, (unsigned long) le32_to_cpu(gdp->bg_inode_table));
 			return 0;
 		}
-		block += EXT2_BLOCKS_PER_GROUP(sb);
+		first_block += EXT2_BLOCKS_PER_GROUP(sb);
 		gdp++;
 	}
 	return 1;
@@ -861,10 +868,9 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 
 	if (EXT2_BLOCKS_PER_GROUP(sb) == 0)
 		goto cantfind_ext2;
-	sbi->s_groups_count = (le32_to_cpu(es->s_blocks_count) -
-				        le32_to_cpu(es->s_first_data_block) +
-				       EXT2_BLOCKS_PER_GROUP(sb) - 1) /
-				       EXT2_BLOCKS_PER_GROUP(sb);
+ 	sbi->s_groups_count = ((le32_to_cpu(es->s_blocks_count) -
+ 				le32_to_cpu(es->s_first_data_block) - 1)
+ 					/ EXT2_BLOCKS_PER_GROUP(sb)) + 1;
 	db_count = (sbi->s_groups_count + EXT2_DESC_PER_BLOCK(sb) - 1) /
 		   EXT2_DESC_PER_BLOCK(sb);
 	sbi->s_group_desc = kmalloc (db_count * sizeof (struct buffer_head *), GFP_KERNEL);
-- 
cgit v1.2.1


From eee194e76c681dbdbf5024b889fda1181b66ef57 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:30 -0700
Subject: [PATCH] ext3: inode numbers are unsigned long

This is primarily format string fixes, with changes to ialloc.c where large
inode counts could overflow, and also pass around journal_inum as an
unsigned long, just to be pedantic about it....

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Cc: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/ialloc.c |  6 +++---
 fs/ext3/inode.c  |  2 +-
 fs/ext3/namei.c  |  4 ++--
 fs/ext3/super.c  | 25 +++++++++++++------------
 fs/ext3/xattr.c  | 16 ++++++++--------
 5 files changed, 27 insertions(+), 26 deletions(-)

diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 5e288368499b..1e4ded8aa3cd 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -202,7 +202,7 @@ error_return:
 static int find_group_dir(struct super_block *sb, struct inode *parent)
 {
 	int ngroups = EXT3_SB(sb)->s_groups_count;
-	int freei, avefreei;
+	unsigned int freei, avefreei;
 	struct ext3_group_desc *desc, *best_desc = NULL;
 	struct buffer_head *bh;
 	int group, best_group = -1;
@@ -261,10 +261,10 @@ static int find_group_orlov(struct super_block *sb, struct inode *parent)
 	struct ext3_super_block *es = sbi->s_es;
 	int ngroups = sbi->s_groups_count;
 	int inodes_per_group = EXT3_INODES_PER_GROUP(sb);
-	int freei, avefreei;
+	unsigned int freei, avefreei;
 	ext3_fsblk_t freeb, avefreeb;
 	ext3_fsblk_t blocks_per_dir;
-	int ndirs;
+	unsigned int ndirs;
 	int max_debt, max_dirs, min_inodes;
 	ext3_grpblk_t min_blocks;
 	int group = -1, i;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 473d206b1d7e..56665fab027d 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2115,7 +2115,7 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode,
 			 */
 			if (!bh) {
 				ext3_error(inode->i_sb, "ext3_free_branches",
-					   "Read failure, inode=%ld, block="E3FSBLK,
+					   "Read failure, inode=%lu, block="E3FSBLK,
 					   inode->i_ino, nr);
 				continue;
 			}
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 4123f5261bcd..70dc5206ebfa 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -1919,8 +1919,8 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode)
 	if (!err)
 		list_add(&EXT3_I(inode)->i_orphan, &EXT3_SB(sb)->s_orphan);
 
-	jbd_debug(4, "superblock will point to %ld\n", inode->i_ino);
-	jbd_debug(4, "orphan inode %ld will point to %d\n",
+	jbd_debug(4, "superblock will point to %lu\n", inode->i_ino);
+	jbd_debug(4, "orphan inode %lu will point to %d\n",
 			inode->i_ino, NEXT_ORPHAN(inode));
 out_unlock:
 	unlock_super(sb);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 762dff7b56fb..bc7a768f9bb4 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -45,7 +45,7 @@
 static int ext3_load_journal(struct super_block *, struct ext3_super_block *,
 			     unsigned long journal_devnum);
 static int ext3_create_journal(struct super_block *, struct ext3_super_block *,
-			       int);
+			       unsigned int);
 static void ext3_commit_super (struct super_block * sb,
 			       struct ext3_super_block * es,
 			       int sync);
@@ -376,7 +376,7 @@ static void dump_orphan_list(struct super_block *sb, struct ext3_sb_info *sbi)
 	list_for_each(l, &sbi->s_orphan) {
 		struct inode *inode = orphan_list_entry(l);
 		printk(KERN_ERR "  "
-		       "inode %s:%ld at %p: mode %o, nlink %d, next %d\n",
+		       "inode %s:%lu at %p: mode %o, nlink %d, next %d\n",
 		       inode->i_sb->s_id, inode->i_ino, inode,
 		       inode->i_mode, inode->i_nlink,
 		       NEXT_ORPHAN(inode));
@@ -753,7 +753,7 @@ static ext3_fsblk_t get_sb_block(void **data)
 }
 
 static int parse_options (char *options, struct super_block *sb,
-			  unsigned long *inum, unsigned long *journal_devnum,
+			  unsigned int *inum, unsigned long *journal_devnum,
 			  ext3_fsblk_t *n_blocks_count, int is_remount)
 {
 	struct ext3_sb_info *sbi = EXT3_SB(sb);
@@ -1306,17 +1306,17 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 		DQUOT_INIT(inode);
 		if (inode->i_nlink) {
 			printk(KERN_DEBUG
-				"%s: truncating inode %ld to %Ld bytes\n",
+				"%s: truncating inode %lu to %Ld bytes\n",
 				__FUNCTION__, inode->i_ino, inode->i_size);
-			jbd_debug(2, "truncating inode %ld to %Ld bytes\n",
+			jbd_debug(2, "truncating inode %lu to %Ld bytes\n",
 				  inode->i_ino, inode->i_size);
 			ext3_truncate(inode);
 			nr_truncates++;
 		} else {
 			printk(KERN_DEBUG
-				"%s: deleting unreferenced inode %ld\n",
+				"%s: deleting unreferenced inode %lu\n",
 				__FUNCTION__, inode->i_ino);
-			jbd_debug(2, "deleting unreferenced inode %ld\n",
+			jbd_debug(2, "deleting unreferenced inode %lu\n",
 				  inode->i_ino);
 			nr_orphans++;
 		}
@@ -1395,7 +1395,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	ext3_fsblk_t sb_block = get_sb_block(&data);
 	ext3_fsblk_t logic_sb_block;
 	unsigned long offset = 0;
-	unsigned long journal_inum = 0;
+	unsigned int journal_inum = 0;
 	unsigned long journal_devnum = 0;
 	unsigned long def_mount_opts;
 	struct inode *root;
@@ -1844,7 +1844,8 @@ static void ext3_init_journal_params(struct super_block *sb, journal_t *journal)
 	spin_unlock(&journal->j_state_lock);
 }
 
-static journal_t *ext3_get_journal(struct super_block *sb, int journal_inum)
+static journal_t *ext3_get_journal(struct super_block *sb,
+				   unsigned int journal_inum)
 {
 	struct inode *journal_inode;
 	journal_t *journal;
@@ -1979,7 +1980,7 @@ static int ext3_load_journal(struct super_block *sb,
 			     unsigned long journal_devnum)
 {
 	journal_t *journal;
-	int journal_inum = le32_to_cpu(es->s_journal_inum);
+	unsigned int journal_inum = le32_to_cpu(es->s_journal_inum);
 	dev_t journal_dev;
 	int err = 0;
 	int really_read_only;
@@ -2065,7 +2066,7 @@ static int ext3_load_journal(struct super_block *sb,
 
 static int ext3_create_journal(struct super_block * sb,
 			       struct ext3_super_block * es,
-			       int journal_inum)
+			       unsigned int journal_inum)
 {
 	journal_t *journal;
 
@@ -2078,7 +2079,7 @@ static int ext3_create_journal(struct super_block * sb,
 	if (!(journal = ext3_get_journal(sb, journal_inum)))
 		return -EINVAL;
 
-	printk(KERN_INFO "EXT3-fs: creating new journal on inode %d\n",
+	printk(KERN_INFO "EXT3-fs: creating new journal on inode %u\n",
 	       journal_inum);
 
 	if (journal_create(journal)) {
diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c
index a44a0562203a..f86f2482f01d 100644
--- a/fs/ext3/xattr.c
+++ b/fs/ext3/xattr.c
@@ -75,7 +75,7 @@
 
 #ifdef EXT3_XATTR_DEBUG
 # define ea_idebug(inode, f...) do { \
-		printk(KERN_DEBUG "inode %s:%ld: ", \
+		printk(KERN_DEBUG "inode %s:%lu: ", \
 			inode->i_sb->s_id, inode->i_ino); \
 		printk(f); \
 		printk("\n"); \
@@ -233,7 +233,7 @@ ext3_xattr_block_get(struct inode *inode, int name_index, const char *name,
 		atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount));
 	if (ext3_xattr_check_block(bh)) {
 bad_block:	ext3_error(inode->i_sb, __FUNCTION__,
-			   "inode %ld: bad block "E3FSBLK, inode->i_ino,
+			   "inode %lu: bad block "E3FSBLK, inode->i_ino,
 			   EXT3_I(inode)->i_file_acl);
 		error = -EIO;
 		goto cleanup;
@@ -375,7 +375,7 @@ ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size)
 		atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount));
 	if (ext3_xattr_check_block(bh)) {
 		ext3_error(inode->i_sb, __FUNCTION__,
-			   "inode %ld: bad block "E3FSBLK, inode->i_ino,
+			   "inode %lu: bad block "E3FSBLK, inode->i_ino,
 			   EXT3_I(inode)->i_file_acl);
 		error = -EIO;
 		goto cleanup;
@@ -647,7 +647,7 @@ ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i,
 			le32_to_cpu(BHDR(bs->bh)->h_refcount));
 		if (ext3_xattr_check_block(bs->bh)) {
 			ext3_error(sb, __FUNCTION__,
-				"inode %ld: bad block "E3FSBLK, inode->i_ino,
+				"inode %lu: bad block "E3FSBLK, inode->i_ino,
 				EXT3_I(inode)->i_file_acl);
 			error = -EIO;
 			goto cleanup;
@@ -848,7 +848,7 @@ cleanup_dquot:
 
 bad_block:
 	ext3_error(inode->i_sb, __FUNCTION__,
-		   "inode %ld: bad block "E3FSBLK, inode->i_ino,
+		   "inode %lu: bad block "E3FSBLK, inode->i_ino,
 		   EXT3_I(inode)->i_file_acl);
 	goto cleanup;
 
@@ -1077,14 +1077,14 @@ ext3_xattr_delete_inode(handle_t *handle, struct inode *inode)
 	bh = sb_bread(inode->i_sb, EXT3_I(inode)->i_file_acl);
 	if (!bh) {
 		ext3_error(inode->i_sb, __FUNCTION__,
-			"inode %ld: block "E3FSBLK" read error", inode->i_ino,
+			"inode %lu: block "E3FSBLK" read error", inode->i_ino,
 			EXT3_I(inode)->i_file_acl);
 		goto cleanup;
 	}
 	if (BHDR(bh)->h_magic != cpu_to_le32(EXT3_XATTR_MAGIC) ||
 	    BHDR(bh)->h_blocks != cpu_to_le32(1)) {
 		ext3_error(inode->i_sb, __FUNCTION__,
-			"inode %ld: bad block "E3FSBLK, inode->i_ino,
+			"inode %lu: bad block "E3FSBLK, inode->i_ino,
 			EXT3_I(inode)->i_file_acl);
 		goto cleanup;
 	}
@@ -1211,7 +1211,7 @@ again:
 		bh = sb_bread(inode->i_sb, ce->e_block);
 		if (!bh) {
 			ext3_error(inode->i_sb, __FUNCTION__,
-				"inode %ld: block %lu read error",
+				"inode %lu: block %lu read error",
 				inode->i_ino, (unsigned long) ce->e_block);
 		} else if (le32_to_cpu(BHDR(bh)->h_refcount) >=
 				EXT3_XATTR_REFCOUNT_MAX) {
-- 
cgit v1.2.1


From 37ed322290eb6d5cf2ab33915793ed4219eae1d6 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:31 -0700
Subject: [PATCH] JBD: 16T fixes

These are a few places I've found in jbd that look like they may not be
16T-safe, or consistent with the use of unsigned longs for block
containers.  Problems here would be somewhat hard to hit, would require
journal blocks past the 8T boundary, which would not be terribly common.
Still, should fix.

(some of these have come from the ext4 work on jbd as well).

I think there's one more possibility that the wrap() function may not be
safe IF your last block in the journal butts right up against the 232 block
boundary, but that seems like a VERY remote possibility, and I'm not
worrying about it at this point.

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/journal.c    | 6 +++---
 include/linux/jbd.h | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 2613fca92740..88d970e09ce6 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -271,7 +271,7 @@ static void journal_kill_thread(journal_t *journal)
 int journal_write_metadata_buffer(transaction_t *transaction,
 				  struct journal_head  *jh_in,
 				  struct journal_head **jh_out,
-				  int blocknr)
+				  unsigned long blocknr)
 {
 	int need_copy_out = 0;
 	int done_copy_out = 0;
@@ -696,7 +696,7 @@ fail:
  *  @bdev: Block device on which to create the journal
  *  @fs_dev: Device which hold journalled filesystem for this journal.
  *  @start: Block nr Start of journal.
- *  @len:  Lenght of the journal in blocks.
+ *  @len:  Length of the journal in blocks.
  *  @blocksize: blocksize of journalling device
  *  @returns: a newly created journal_t *
  *
@@ -820,7 +820,7 @@ static void journal_fail_superblock (journal_t *journal)
 static int journal_reset(journal_t *journal)
 {
 	journal_superblock_t *sb = journal->j_superblock;
-	unsigned int first, last;
+	unsigned long first, last;
 
 	first = be32_to_cpu(sb->s_first);
 	last = be32_to_cpu(sb->s_maxlen);
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 7d847931ee5a..dc262624efa5 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -732,7 +732,7 @@ struct journal_s
 	 */
 	struct block_device	*j_dev;
 	int			j_blocksize;
-	unsigned int		j_blk_offset;
+	unsigned long		j_blk_offset;
 
 	/*
 	 * Device which holds the client fs.  For internal journal this will be
@@ -866,7 +866,7 @@ extern int
 journal_write_metadata_buffer(transaction_t	  *transaction,
 			      struct journal_head  *jh_in,
 			      struct journal_head **jh_out,
-			      int		   blocknr);
+			      unsigned long	   blocknr);
 
 /* Transaction locking */
 extern void		__wait_on_journal (journal_t *);
-- 
cgit v1.2.1


From 321fb9e81821a81b2cda1328698b0c19d57c717f Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Wed, 27 Sep 2006 01:49:32 -0700
Subject: [PATCH] ext3: turn on reservation dump on block allocation errors

In the past there were a few kernel panics related to block reservation
tree operations failure (insert/remove etc).  It would be very useful to
get the block allocation reservation map info when such error happens.

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/balloc.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index e6b983707008..2bd836f30ca6 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -111,7 +111,7 @@ error_out:
  * we could easily switch to that without changing too much
  * code.
  */
-#if 0
+#if 1
 static void __rsv_window_dump(struct rb_root *root, int verbose,
 			      const char *fn)
 {
@@ -129,7 +129,7 @@ restart:
 		rsv = list_entry(n, struct ext3_reserve_window_node, rsv_node);
 		if (verbose)
 			printk("reservation window 0x%p "
-			       "start:  %d, end:  %d\n",
+			       "start:  %lu, end:  %lu\n",
 			       rsv, rsv->rsv_start, rsv->rsv_end);
 		if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) {
 			printk("Bad reservation %p (start >= end)\n",
@@ -236,8 +236,10 @@ void ext3_rsv_window_add(struct super_block *sb,
 			p = &(*p)->rb_left;
 		else if (start > this->rsv_end)
 			p = &(*p)->rb_right;
-		else
+		else {
+			rsv_window_dump(root, 1);
 			BUG();
+		}
 	}
 
 	rb_link_node(node, parent, p);
@@ -1133,8 +1135,10 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 					*count-my_rsv->rsv_end + grp_goal - 1);
 
 		if ((my_rsv->rsv_start >= group_first_block + EXT3_BLOCKS_PER_GROUP(sb))
-		    || (my_rsv->rsv_end < group_first_block))
+		    || (my_rsv->rsv_end < group_first_block)) {
+			rsv_window_dump(&EXT3_SB(sb)->s_rsv_window_root, 1);
 			BUG();
+		}
 		ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, grp_goal,
 					   &num, &my_rsv->rsv_window);
 		if (ret >= 0) {
-- 
cgit v1.2.1


From 36faadc144477b4929c8fe60b8053f4472eeb3d2 Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Wed, 27 Sep 2006 01:49:32 -0700
Subject: [PATCH] ext3: more comments about block allocation/reservation code

Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Acked-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/balloc.c | 292 ++++++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 247 insertions(+), 45 deletions(-)

diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index 2bd836f30ca6..f1897feb18f4 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -38,6 +38,13 @@
 
 #define in_range(b, first, len)	((b) >= (first) && (b) <= (first) + (len) - 1)
 
+/**
+ * ext3_get_group_desc() -- load group descriptor from disk
+ * @sb: 		super block
+ * @block_group:	given block group
+ * @bh:			pointer to the buffer head to store the block
+ *			group descriptor
+ */
 struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb,
 					     unsigned int block_group,
 					     struct buffer_head ** bh)
@@ -73,7 +80,11 @@ struct ext3_group_desc * ext3_get_group_desc(struct super_block * sb,
 	return desc + offset;
 }
 
-/*
+/**
+ * read_block_bitmap()
+ * @sb:			super block
+ * @block_group:	given block group
+ *
  * Read the bitmap for a given block_group, reading into the specified
  * slot in the superblock's bitmap cache.
  *
@@ -103,13 +114,20 @@ error_out:
  * Operations include:
  * dump, find, add, remove, is_empty, find_next_reservable_window, etc.
  *
- * We use sorted double linked list for the per-filesystem reservation
- * window list. (like in vm_region).
+ * We use a red-black tree to represent per-filesystem reservation
+ * windows.
  *
- * Initially, we keep those small operations in the abstract functions,
- * so later if we need a better searching tree than double linked-list,
- * we could easily switch to that without changing too much
- * code.
+ */
+
+/**
+ * __rsv_window_dump() -- Dump the filesystem block allocation reservation map
+ * @rb_root:		root of per-filesystem reservation rb tree
+ * @verbose:		verbose mode
+ * @fn:			function which wishes to dump the reservation map
+ *
+ * If verbose is turned on, it will print the whole block reservation
+ * windows(start, end).	Otherwise, it will only print out the "bad" windows,
+ * those windows that overlap with their immediate neighbors.
  */
 #if 1
 static void __rsv_window_dump(struct rb_root *root, int verbose,
@@ -161,6 +179,22 @@ restart:
 #define rsv_window_dump(root, verbose) do {} while (0)
 #endif
 
+/**
+ * goal_in_my_reservation()
+ * @rsv:		inode's reservation window
+ * @grp_goal:		given goal block relative to the allocation block group
+ * @group:		the current allocation block group
+ * @sb:			filesystem super block
+ *
+ * Test if the given goal block (group relative) is within the file's
+ * own block reservation window range.
+ *
+ * If the reservation window is outside the goal allocation group, return 0;
+ * grp_goal (given goal block) could be -1, which means no specific
+ * goal block. In this case, always return 1.
+ * If the goal block is within the reservation window, return 1;
+ * otherwise, return 0;
+ */
 static int
 goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal,
 			unsigned int group, struct super_block * sb)
@@ -179,7 +213,11 @@ goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal,
 	return 1;
 }
 
-/*
+/**
+ * search_reserve_window()
+ * @rb_root:		root of reservation tree
+ * @goal:		target allocation block
+ *
  * Find the reserved window which includes the goal, or the previous one
  * if the goal is not in any window.
  * Returns NULL if there are no windows or if all windows start after the goal.
@@ -216,6 +254,13 @@ search_reserve_window(struct rb_root *root, ext3_fsblk_t goal)
 	return rsv;
 }
 
+/**
+ * ext3_rsv_window_add() -- Insert a window to the block reservation rb tree.
+ * @sb:			super block
+ * @rsv:		reservation window to add
+ *
+ * Must be called with rsv_lock hold.
+ */
 void ext3_rsv_window_add(struct super_block *sb,
 		    struct ext3_reserve_window_node *rsv)
 {
@@ -246,6 +291,15 @@ void ext3_rsv_window_add(struct super_block *sb,
 	rb_insert_color(node, root);
 }
 
+/**
+ * ext3_rsv_window_remove() -- unlink a window from the reservation rb tree
+ * @sb:			super block
+ * @rsv:		reservation window to remove
+ *
+ * Mark the block reservation window as not allocated, and unlink it
+ * from the filesystem reservation window rb tree. Must be called with
+ * rsv_lock hold.
+ */
 static void rsv_window_remove(struct super_block *sb,
 			      struct ext3_reserve_window_node *rsv)
 {
@@ -255,11 +309,39 @@ static void rsv_window_remove(struct super_block *sb,
 	rb_erase(&rsv->rsv_node, &EXT3_SB(sb)->s_rsv_window_root);
 }
 
+/*
+ * rsv_is_empty() -- Check if the reservation window is allocated.
+ * @rsv:		given reservation window to check
+ *
+ * returns 1 if the end block is EXT3_RESERVE_WINDOW_NOT_ALLOCATED.
+ */
 static inline int rsv_is_empty(struct ext3_reserve_window *rsv)
 {
 	/* a valid reservation end block could not be 0 */
-	return (rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED);
+	return rsv->_rsv_end == EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
 }
+
+/**
+ * ext3_init_block_alloc_info()
+ * @inode:		file inode structure
+ *
+ * Allocate and initialize the	reservation window structure, and
+ * link the window to the ext3 inode structure at last
+ *
+ * The reservation window structure is only dynamically allocated
+ * and linked to ext3 inode the first time the open file
+ * needs a new block. So, before every ext3_new_block(s) call, for
+ * regular files, we should check whether the reservation window
+ * structure exists or not. In the latter case, this function is called.
+ * Fail to do so will result in block reservation being turned off for that
+ * open file.
+ *
+ * This function is called from ext3_get_blocks_handle(), also called
+ * when setting the reservation window size through ioctl before the file
+ * is open for write (needs block allocation).
+ *
+ * Needs truncate_mutex protection prior to call this function.
+ */
 void ext3_init_block_alloc_info(struct inode *inode)
 {
 	struct ext3_inode_info *ei = EXT3_I(inode);
@@ -289,6 +371,19 @@ void ext3_init_block_alloc_info(struct inode *inode)
 	ei->i_block_alloc_info = block_i;
 }
 
+/**
+ * ext3_discard_reservation()
+ * @inode:		inode
+ *
+ * Discard(free) block reservation window on last file close, or truncate
+ * or at last iput().
+ *
+ * It is being called in three cases:
+ *	ext3_release_file(): last writer close the file
+ *	ext3_clear_inode(): last iput(), when nobody link to this file.
+ *	ext3_truncate(): when the block indirect map is about to change.
+ *
+ */
 void ext3_discard_reservation(struct inode *inode)
 {
 	struct ext3_inode_info *ei = EXT3_I(inode);
@@ -308,7 +403,14 @@ void ext3_discard_reservation(struct inode *inode)
 	}
 }
 
-/* Free given blocks, update quota and i_blocks field */
+/**
+ * ext3_free_blocks_sb() -- Free given blocks and update quota
+ * @handle:			handle to this transaction
+ * @sb:				super block
+ * @block:			start physcial block to free
+ * @count:			number of blocks to free
+ * @pdquot_freed_blocks:	pointer to quota
+ */
 void ext3_free_blocks_sb(handle_t *handle, struct super_block *sb,
 			 ext3_fsblk_t block, unsigned long count,
 			 unsigned long *pdquot_freed_blocks)
@@ -492,7 +594,13 @@ error_return:
 	return;
 }
 
-/* Free given blocks, update quota and i_blocks field */
+/**
+ * ext3_free_blocks() -- Free given blocks and update quota
+ * @handle:		handle for this transaction
+ * @inode:		inode
+ * @block:		start physical block to free
+ * @count:		number of blocks to count
+ */
 void ext3_free_blocks(handle_t *handle, struct inode *inode,
 			ext3_fsblk_t block, unsigned long count)
 {
@@ -510,7 +618,11 @@ void ext3_free_blocks(handle_t *handle, struct inode *inode,
 	return;
 }
 
-/*
+/**
+ * ext3_test_allocatable()
+ * @nr:			given allocation block group
+ * @bh:			bufferhead contains the bitmap of the given block group
+ *
  * For ext3 allocations, we must not reuse any blocks which are
  * allocated in the bitmap buffer's "last committed data" copy.  This
  * prevents deletes from freeing up the page for reuse until we have
@@ -543,6 +655,16 @@ static int ext3_test_allocatable(ext3_grpblk_t nr, struct buffer_head *bh)
 	return ret;
 }
 
+/**
+ * bitmap_search_next_usable_block()
+ * @start:		the starting block (group relative) of the search
+ * @bh:			bufferhead contains the block group bitmap
+ * @maxblocks:		the ending block (group relative) of the reservation
+ *
+ * The bitmap search --- search forward alternately through the actual
+ * bitmap on disk and the last-committed copy in journal, until we find a
+ * bit free in both bitmaps.
+ */
 static ext3_grpblk_t
 bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 					ext3_grpblk_t maxblocks)
@@ -550,11 +672,6 @@ bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 	ext3_grpblk_t next;
 	struct journal_head *jh = bh2jh(bh);
 
-	/*
-	 * The bitmap search --- search forward alternately through the actual
-	 * bitmap and the last-committed copy until we find a bit free in
-	 * both
-	 */
 	while (start < maxblocks) {
 		next = ext3_find_next_zero_bit(bh->b_data, maxblocks, start);
 		if (next >= maxblocks)
@@ -570,8 +687,14 @@ bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 	return -1;
 }
 
-/*
- * Find an allocatable block in a bitmap.  We honour both the bitmap and
+/**
+ * find_next_usable_block()
+ * @start:		the starting block (group relative) to find next
+ *			allocatable block in bitmap.
+ * @bh:			bufferhead contains the block group bitmap
+ * @maxblocks:		the ending block (group relative) for the search
+ *
+ * Find an allocatable block in a bitmap.  We honor both the bitmap and
  * its last-committed copy (if that exists), and perform the "most
  * appropriate allocation" algorithm of looking for a free block near
  * the initial goal; then for a free byte somewhere in the bitmap; then
@@ -622,7 +745,11 @@ find_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 	return here;
 }
 
-/*
+/**
+ * claim_block()
+ * @block:		the free block (group relative) to allocate
+ * @bh:			the bufferhead containts the block group bitmap
+ *
  * We think we can allocate this block in this bitmap.  Try to set the bit.
  * If that succeeds then check that nobody has allocated and then freed the
  * block since we saw that is was not marked in b_committed_data.  If it _was_
@@ -648,7 +775,26 @@ claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh)
 	return ret;
 }
 
-/*
+/**
+ * ext3_try_to_allocate()
+ * @sb:			superblock
+ * @handle:		handle to this transaction
+ * @group:		given allocation block group
+ * @bitmap_bh:		bufferhead holds the block bitmap
+ * @grp_goal:		given target block within the group
+ * @count:		target number of blocks to allocate
+ * @my_rsv:		reservation window
+ *
+ * Attempt to allocate blocks within a give range. Set the range of allocation
+ * first, then find the first free bit(s) from the bitmap (within the range),
+ * and at last, allocate the blocks by claiming the found free bit as allocated.
+ *
+ * To set the range of this allocation:
+ * 	if there is a reservation window, only try to allocate block(s) from the
+ *	file's own reservation window;
+ *	Otherwise, the allocation range starts from the give goal block, ends at
+ * 	the block group's last block.
+ *
  * If we failed to allocate the desired block then we may end up crossing to a
  * new bitmap.  In that case we must release write access to the old one via
  * ext3_journal_release_buffer(), else we'll run out of credits.
@@ -705,7 +851,8 @@ repeat:
 	}
 	start = grp_goal;
 
-	if (!claim_block(sb_bgl_lock(EXT3_SB(sb), group), grp_goal, bitmap_bh)) {
+	if (!claim_block(sb_bgl_lock(EXT3_SB(sb), group),
+		grp_goal, bitmap_bh)) {
 		/*
 		 * The block was allocated by another thread, or it was
 		 * allocated and then freed by another thread
@@ -720,7 +867,8 @@ repeat:
 	grp_goal++;
 	while (num < *count && grp_goal < end
 		&& ext3_test_allocatable(grp_goal, bitmap_bh)
-		&& claim_block(sb_bgl_lock(EXT3_SB(sb), group), grp_goal, bitmap_bh)) {
+		&& claim_block(sb_bgl_lock(EXT3_SB(sb), group),
+				grp_goal, bitmap_bh)) {
 		num++;
 		grp_goal++;
 	}
@@ -931,9 +1079,10 @@ static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv,
 		if ((my_rsv->rsv_alloc_hit >
 		     (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) {
 			/*
-			 * if we previously allocation hit ration is greater than half
-			 * we double the size of reservation window next time
-			 * otherwise keep the same
+			 * if the previously allocation hit ratio is
+			 * greater than 1/2, then we double the size of
+			 * the reservation window the next time,
+			 * otherwise we keep the same size window
 			 */
 			size = size * 2;
 			if (size > EXT3_MAX_RESERVE_BLOCKS)
@@ -1012,6 +1161,23 @@ retry:
 	goto retry;
 }
 
+/**
+ * try_to_extend_reservation()
+ * @my_rsv:		given reservation window
+ * @sb:			super block
+ * @size:		the delta to extend
+ *
+ * Attempt to expand the reservation window large enough to have
+ * required number of free blocks
+ *
+ * Since ext3_try_to_allocate() will always allocate blocks within
+ * the reservation window range, if the window size is too small,
+ * multiple blocks allocation has to stop at the end of the reservation
+ * window. To make this more efficient, given the total number of
+ * blocks needed and the current size of the window, we try to
+ * expand the reservation window size if necessary on a best-effort
+ * basis before ext3_new_blocks() tries to allocate blocks,
+ */
 static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv,
 			struct super_block *sb, int size)
 {
@@ -1037,7 +1203,17 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv,
 	spin_unlock(rsv_lock);
 }
 
-/*
+/**
+ * ext3_try_to_allocate_with_rsv()
+ * @sb:			superblock
+ * @handle:		handle to this transaction
+ * @group:		given allocation block group
+ * @bitmap_bh:		bufferhead holds the block bitmap
+ * @grp_goal:		given target block within the group
+ * @count:		target number of blocks to allocate
+ * @my_rsv:		reservation window
+ * @errp:		pointer to store the error code
+ *
  * This is the main function used to allocate a new block and its reservation
  * window.
  *
@@ -1053,9 +1229,7 @@ static void try_to_extend_reservation(struct ext3_reserve_window_node *my_rsv,
  * reservation), and there are lots of free blocks, but they are all
  * being reserved.
  *
- * We use a sorted double linked list for the per-filesystem reservation list.
- * The insert, remove and find a free space(non-reserved) operations for the
- * sorted double linked list should be fast.
+ * We use a red-black tree for the per-filesystem reservation list.
  *
  */
 static ext3_grpblk_t
@@ -1120,7 +1294,8 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 	 */
 	while (1) {
 		if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) ||
-			!goal_in_my_reservation(&my_rsv->rsv_window, grp_goal, group, sb)) {
+			!goal_in_my_reservation(&my_rsv->rsv_window,
+						grp_goal, group, sb)) {
 			if (my_rsv->rsv_goal_size < *count)
 				my_rsv->rsv_goal_size = *count;
 			ret = alloc_new_reservation(my_rsv, grp_goal, sb,
@@ -1128,19 +1303,22 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 			if (ret < 0)
 				break;			/* failed */
 
-			if (!goal_in_my_reservation(&my_rsv->rsv_window, grp_goal, group, sb))
+			if (!goal_in_my_reservation(&my_rsv->rsv_window,
+							grp_goal, group, sb))
 				grp_goal = -1;
-		} else if (grp_goal > 0 && (my_rsv->rsv_end-grp_goal+1) < *count)
+		} else if (grp_goal > 0 &&
+			  (my_rsv->rsv_end-grp_goal+1) < *count)
 			try_to_extend_reservation(my_rsv, sb,
 					*count-my_rsv->rsv_end + grp_goal - 1);
 
-		if ((my_rsv->rsv_start >= group_first_block + EXT3_BLOCKS_PER_GROUP(sb))
+		if ((my_rsv->rsv_start >= group_first_block +
+					EXT3_BLOCKS_PER_GROUP(sb))
 		    || (my_rsv->rsv_end < group_first_block)) {
 			rsv_window_dump(&EXT3_SB(sb)->s_rsv_window_root, 1);
 			BUG();
 		}
-		ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh, grp_goal,
-					   &num, &my_rsv->rsv_window);
+		ret = ext3_try_to_allocate(sb, handle, group, bitmap_bh,
+					   grp_goal, &num, &my_rsv->rsv_window);
 		if (ret >= 0) {
 			my_rsv->rsv_alloc_hit += num;
 			*count = num;
@@ -1165,6 +1343,12 @@ out:
 	return ret;
 }
 
+/**
+ * ext3_has_free_blocks()
+ * @sbi:		in-core super block structure.
+ *
+ * Check if filesystem has at least 1 free block available for allocation.
+ */
 static int ext3_has_free_blocks(struct ext3_sb_info *sbi)
 {
 	ext3_fsblk_t free_blocks, root_blocks;
@@ -1179,11 +1363,17 @@ static int ext3_has_free_blocks(struct ext3_sb_info *sbi)
 	return 1;
 }
 
-/*
+/**
+ * ext3_should_retry_alloc()
+ * @sb:			super block
+ * @retries		number of attemps has been made
+ *
  * ext3_should_retry_alloc() is called when ENOSPC is returned, and if
  * it is profitable to retry the operation, this function will wait
  * for the current or commiting transaction to complete, and then
  * return TRUE.
+ *
+ * if the total number of retries exceed three times, return FALSE.
  */
 int ext3_should_retry_alloc(struct super_block *sb, int *retries)
 {
@@ -1195,13 +1385,19 @@ int ext3_should_retry_alloc(struct super_block *sb, int *retries)
 	return journal_force_commit_nested(EXT3_SB(sb)->s_journal);
 }
 
-/*
- * ext3_new_block uses a goal block to assist allocation.  If the goal is
- * free, or there is a free block within 32 blocks of the goal, that block
- * is allocated.  Otherwise a forward search is made for a free block; within
- * each block group the search first looks for an entire free byte in the block
- * bitmap, and then for any free bit if that fails.
- * This function also updates quota and i_blocks field.
+/**
+ * ext3_new_blocks() -- core block(s) allocation function
+ * @handle:		handle to this transaction
+ * @inode:		file inode
+ * @goal:		given target block(filesystem wide)
+ * @count:		target number of blocks to allocate
+ * @errp:		error code
+ *
+ * ext3_new_blocks uses a goal block to assist allocation.  It tries to
+ * allocate block(s) from the block group contains the goal block first. If that
+ * fails, it will try to allocate block(s) from other block groups without
+ * any specific goal block.
+ *
  */
 ext3_fsblk_t ext3_new_blocks(handle_t *handle, struct inode *inode,
 			ext3_fsblk_t goal, unsigned long *count, int *errp)
@@ -1432,7 +1628,7 @@ allocated:
 
 	spin_lock(sb_bgl_lock(sbi, group_no));
 	gdp->bg_free_blocks_count =
-			cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count) - num);
+			cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num);
 	spin_unlock(sb_bgl_lock(sbi, group_no));
 	percpu_counter_mod(&sbi->s_freeblocks_counter, -num);
 
@@ -1475,6 +1671,12 @@ ext3_fsblk_t ext3_new_block(handle_t *handle, struct inode *inode,
 	return ext3_new_blocks(handle, inode, goal, &count, errp);
 }
 
+/**
+ * ext3_count_free_blocks() -- count filesystem free blocks
+ * @sb:		superblock
+ *
+ * Adds up the number of free blocks from each block group.
+ */
 ext3_fsblk_t ext3_count_free_blocks(struct super_block *sb)
 {
 	ext3_fsblk_t desc_count;
-- 
cgit v1.2.1


From 7543fc7b3abfee8c6cd6349ebd5e5fde02fac984 Mon Sep 17 00:00:00 2001
From: Vasily Averin <vvs@sw.ru>
Date: Wed, 27 Sep 2006 01:49:33 -0700
Subject: [PATCH] ext3: wrong error behavior

SWsoft Virtuozzo/OpenVZ Linux kernel team has discovered that ext3 error
behavior was broken in linux kernels since 2.5.x versions by the following
patch:

2002/10/31 02:15:26-05:00 tytso@snap.thunk.org
Default mount options from superblock for ext2/3 filesystems
http://linux.bkbits.net:8080/linux-2.6/gnupatch@3dc0d88eKbV9ivV4ptRNM8fBuA3JBQ

In case ext3 file system is mounted with errors=continue
(EXT3_ERRORS_CONTINUE) errors should be ignored when possible.  However at
present in case of any error kernel aborts journal and remounts filesystem
to read-only.  Such behavior was hit number of times and noted to differ
from that of 2.4.x kernels.

This patch fixes this:
- do nothing in case of EXT3_ERRORS_CONTINUE,
- set EXT3_MOUNT_ABORT and call journal_abort() in all other cases
- panic() should be called after ext3_commit_super() to save
 sb marked as EXT3_ERROR_FS

Signed-off-by: Vasily Averin <vvs@sw.ru>
Acked-by: Kirill Korotaev <dev@sw.ru>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: "Stephen C. Tweedie" <sct@redhat.com>
Cc: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/super.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index bc7a768f9bb4..4b526b496102 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -159,20 +159,21 @@ static void ext3_handle_error(struct super_block *sb)
 	if (sb->s_flags & MS_RDONLY)
 		return;
 
-	if (test_opt (sb, ERRORS_RO)) {
-		printk (KERN_CRIT "Remounting filesystem read-only\n");
-		sb->s_flags |= MS_RDONLY;
-	} else {
+	if (!test_opt (sb, ERRORS_CONT)) {
 		journal_t *journal = EXT3_SB(sb)->s_journal;
 
 		EXT3_SB(sb)->s_mount_opt |= EXT3_MOUNT_ABORT;
 		if (journal)
 			journal_abort(journal, -EIO);
 	}
+	if (test_opt (sb, ERRORS_RO)) {
+		printk (KERN_CRIT "Remounting filesystem read-only\n");
+		sb->s_flags |= MS_RDONLY;
+	}
+	ext3_commit_super(sb, es, 1);
 	if (test_opt(sb, ERRORS_PANIC))
 		panic("EXT3-fs (device %s): panic forced after error\n",
 			sb->s_id);
-	ext3_commit_super(sb, es, 1);
 }
 
 void ext3_error (struct super_block * sb, const char * function,
-- 
cgit v1.2.1


From e9ad5620bfb901df8a7a2603c88689ededeecaf1 Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@austin.ibm.com>
Date: Wed, 27 Sep 2006 01:49:35 -0700
Subject: [PATCH] ext3: More whitespace cleanups

More white space cleanups in preparation of cloning ext4 from ext3.
Removing spaces that precede a tab.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/acl.c             |  2 +-
 fs/ext3/balloc.c          | 32 ++++++++++++++++----------------
 fs/ext3/dir.c             |  2 +-
 fs/ext3/hash.c            |  2 +-
 fs/ext3/inode.c           |  8 ++++----
 fs/ext3/namei.c           | 18 +++++++++---------
 fs/ext3/super.c           |  6 +++---
 fs/jbd/checkpoint.c       |  2 +-
 fs/jbd/journal.c          |  2 +-
 fs/jbd/recovery.c         |  2 +-
 fs/jbd/transaction.c      |  6 +++---
 include/linux/ext3_fs.h   |  2 +-
 include/linux/ext3_fs_i.h |  2 +-
 include/linux/jbd.h       | 10 +++++-----
 14 files changed, 48 insertions(+), 48 deletions(-)

diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 0d21d558b87a..92bf78221429 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -258,7 +258,7 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type,
 		default:
 			return -EINVAL;
 	}
- 	if (acl) {
+	if (acl) {
 		value = ext3_acl_to_disk(acl, &size);
 		if (IS_ERR(value))
 			return (int)PTR_ERR(value);
diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index f1897feb18f4..c19be8fc3e51 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -40,7 +40,7 @@
 
 /**
  * ext3_get_group_desc() -- load group descriptor from disk
- * @sb: 		super block
+ * @sb:			super block
  * @block_group:	given block group
  * @bh:			pointer to the buffer head to store the block
  *			group descriptor
@@ -355,7 +355,7 @@ void ext3_init_block_alloc_info(struct inode *inode)
 		rsv->rsv_start = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
 		rsv->rsv_end = EXT3_RESERVE_WINDOW_NOT_ALLOCATED;
 
-	 	/*
+		/*
 		 * if filesystem is mounted with NORESERVATION, the goal
 		 * reservation window size is set to zero to indicate
 		 * block reservation is off
@@ -681,7 +681,7 @@ bitmap_search_next_usable_block(ext3_grpblk_t start, struct buffer_head *bh,
 		jbd_lock_bh_state(bh);
 		if (jh->b_committed_data)
 			start = ext3_find_next_zero_bit(jh->b_committed_data,
-						 	maxblocks, next);
+							maxblocks, next);
 		jbd_unlock_bh_state(bh);
 	}
 	return -1;
@@ -790,10 +790,10 @@ claim_block(spinlock_t *lock, ext3_grpblk_t block, struct buffer_head *bh)
  * and at last, allocate the blocks by claiming the found free bit as allocated.
  *
  * To set the range of this allocation:
- * 	if there is a reservation window, only try to allocate block(s) from the
+ *	if there is a reservation window, only try to allocate block(s) from the
  *	file's own reservation window;
  *	Otherwise, the allocation range starts from the give goal block, ends at
- * 	the block group's last block.
+ *	the block group's last block.
  *
  * If we failed to allocate the desired block then we may end up crossing to a
  * new bitmap.  In that case we must release write access to the old one via
@@ -880,12 +880,12 @@ fail_access:
 }
 
 /**
- * 	find_next_reservable_window():
+ *	find_next_reservable_window():
  *		find a reservable space within the given range.
  *		It does not allocate the reservation window for now:
  *		alloc_new_reservation() will do the work later.
  *
- * 	@search_head: the head of the searching list;
+ *	@search_head: the head of the searching list;
  *		This is not necessarily the list head of the whole filesystem
  *
  *		We have both head and start_block to assist the search
@@ -893,12 +893,12 @@ fail_access:
  *		but we will shift to the place where start_block is,
  *		then start from there, when looking for a reservable space.
  *
- * 	@size: the target new reservation window size
+ *	@size: the target new reservation window size
  *
- * 	@group_first_block: the first block we consider to start
+ *	@group_first_block: the first block we consider to start
  *			the real search from
  *
- * 	@last_block:
+ *	@last_block:
  *		the maximum block number that our goal reservable space
  *		could start from. This is normally the last block in this
  *		group. The search will end when we found the start of next
@@ -906,10 +906,10 @@ fail_access:
  *		This could handle the cross boundary reservation window
  *		request.
  *
- * 	basically we search from the given range, rather than the whole
- * 	reservation double linked list, (start_block, last_block)
- * 	to find a free region that is of my size and has not
- * 	been reserved.
+ *	basically we search from the given range, rather than the whole
+ *	reservation double linked list, (start_block, last_block)
+ *	to find a free region that is of my size and has not
+ *	been reserved.
  *
  */
 static int find_next_reservable_window(
@@ -962,7 +962,7 @@ static int find_next_reservable_window(
 			/*
 			 * Found a reserveable space big enough.  We could
 			 * have a reservation across the group boundary here
-		 	 */
+			 */
 			break;
 		}
 	}
@@ -998,7 +998,7 @@ static int find_next_reservable_window(
 }
 
 /**
- * 	alloc_new_reservation()--allocate a new reservation window
+ *	alloc_new_reservation()--allocate a new reservation window
  *
  *		To make a new reservation, we search part of the filesystem
  *		reservation list (the list that inside the group). We try to
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 6f9e5a523c87..4d97f437cd48 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -67,7 +67,7 @@ int ext3_check_dir_entry (const char * function, struct inode * dir,
 			  unsigned long offset)
 {
 	const char * error_msg = NULL;
- 	const int rlen = le16_to_cpu(de->rec_len);
+	const int rlen = le16_to_cpu(de->rec_len);
 
 	if (rlen < EXT3_DIR_REC_LEN(1))
 		error_msg = "rec_len is smaller than minimal";
diff --git a/fs/ext3/hash.c b/fs/ext3/hash.c
index 7fa637cd322a..deeb27b5ba83 100644
--- a/fs/ext3/hash.c
+++ b/fs/ext3/hash.c
@@ -95,7 +95,7 @@ int ext3fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo)
 	__u32	minor_hash = 0;
 	const char	*p;
 	int		i;
-	__u32 		in[8], buf[4];
+	__u32		in[8], buf[4];
 
 	/* Initialize the default seed for the hash checksum functions */
 	buf[0] = 0x67452301;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 56665fab027d..c9fc15f8b609 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -13,11 +13,11 @@
  *  Copyright (C) 1991, 1992  Linus Torvalds
  *
  *  Goal-directed block allocation by Stephen Tweedie
- * 	(sct@redhat.com), 1993, 1998
+ *	(sct@redhat.com), 1993, 1998
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
  *  64-bit file support on 64-bit platforms by Jakub Jelinek
- * 	(jj@sunsite.ms.mff.cuni.cz)
+ *	(jj@sunsite.ms.mff.cuni.cz)
  *
  *  Assorted race fixes, rewrite of ext3_get_block() by Al Viro, 2000
  */
@@ -470,7 +470,7 @@ static ext3_fsblk_t ext3_find_goal(struct inode *inode, long block,
  *	ext3_blks_to_allocate: Look up the block map and count the number
  *	of direct blocks need to be allocated for the given branch.
  *
- * 	@branch: chain of indirect blocks
+ *	@branch: chain of indirect blocks
  *	@k: number of blocks need for indirect blocks
  *	@blks: number of data blocks to be mapped.
  *	@blocks_to_boundary:  the offset in the indirect block
@@ -1098,7 +1098,7 @@ static int walk_page_buffers(	handle_t *handle,
 
 	for (	bh = head, block_start = 0;
 		ret == 0 && (bh != head || !block_start);
-	    	block_start = block_end, bh = next)
+		block_start = block_end, bh = next)
 	{
 		next = bh->b_this_page;
 		block_end = block_start + blocksize;
diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 70dc5206ebfa..85d132c37ee0 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -15,13 +15,13 @@
  *  Big-endian to little-endian byte-swapping/bitmaps by
  *        David S. Miller (davem@caip.rutgers.edu), 1995
  *  Directory entry file type support and forward compatibility hooks
- *  	for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998
+ *	for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998
  *  Hash Tree Directory indexing (c)
- *  	Daniel Phillips, 2001
+ *	Daniel Phillips, 2001
  *  Hash Tree Directory indexing porting
- *  	Christopher Li, 2002
+ *	Christopher Li, 2002
  *  Hash Tree Directory indexing cleanup
- * 	Theodore Ts'o, 2002
+ *	Theodore Ts'o, 2002
  */
 
 #include <linux/fs.h>
@@ -278,7 +278,7 @@ static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext3_dir_ent
 				       ((char *) de - base));
 			}
 			space += EXT3_DIR_REC_LEN(de->name_len);
-	 		names++;
+			names++;
 		}
 		de = (struct ext3_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len));
 	}
@@ -1688,7 +1688,7 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry,
 
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 +
 					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
@@ -1816,7 +1816,7 @@ static int empty_dir (struct inode * inode)
 			!le32_to_cpu(de1->inode) ||
 			strcmp (".", de->name) ||
 			strcmp ("..", de1->name)) {
-	    	ext3_warning (inode->i_sb, "empty_dir",
+		ext3_warning (inode->i_sb, "empty_dir",
 			      "bad directory (dir #%lu) - no `.' or `..'",
 			      inode->i_ino);
 		brelse (bh);
@@ -2129,7 +2129,7 @@ static int ext3_symlink (struct inode * dir,
 
 retry:
 	handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 +
 					2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb));
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
@@ -2227,7 +2227,7 @@ static int ext3_rename (struct inode * old_dir, struct dentry *old_dentry,
 		DQUOT_INIT(new_dentry->d_inode);
 	handle = ext3_journal_start(old_dir, 2 *
 					EXT3_DATA_TRANS_BLOCKS(old_dir->i_sb) +
-			 		EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
+					EXT3_INDEX_EXTRA_TRANS_BLOCKS + 2);
 	if (IS_ERR(handle))
 		return PTR_ERR(handle);
 
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 4b526b496102..0544325d9d7e 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -734,8 +734,8 @@ static match_table_t tokens = {
 
 static ext3_fsblk_t get_sb_block(void **data)
 {
-	ext3_fsblk_t 	sb_block;
-	char 		*options = (char *) *data;
+	ext3_fsblk_t	sb_block;
+	char		*options = (char *) *data;
 
 	if (!options || strncmp(options, "sb=", 3) != 0)
 		return 1;	/* Default location */
@@ -2740,7 +2740,7 @@ static int __init init_ext3_fs(void)
 out:
 	destroy_inodecache();
 out1:
- 	exit_ext3_xattr();
+	exit_ext3_xattr();
 	return err;
 }
 
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 961ada28db5e..0208cc7ac5d0 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -480,7 +480,7 @@ static int journal_clean_one_cp_list(struct journal_head *jh, int *released)
 	if (!jh)
 		return 0;
 
- 	last_jh = jh->b_cpprev;
+	last_jh = jh->b_cpprev;
 	do {
 		jh = next_jh;
 		next_jh = jh->b_cpnext;
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 88d970e09ce6..b571209d6a1e 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -181,7 +181,7 @@ loop:
 						transaction->t_expires))
 			should_sleep = 0;
 		if (journal->j_flags & JFS_UNMOUNT)
- 			should_sleep = 0;
+			should_sleep = 0;
 		if (should_sleep) {
 			spin_unlock(&journal->j_state_lock);
 			schedule();
diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c
index 73bb64806ed3..445eed6ce5dc 100644
--- a/fs/jbd/recovery.c
+++ b/fs/jbd/recovery.c
@@ -314,7 +314,7 @@ static int do_one_pass(journal_t *journal,
 	unsigned long		next_log_block;
 	int			err, success = 0;
 	journal_superblock_t *	sb;
-	journal_header_t * 	tmp;
+	journal_header_t *	tmp;
 	struct buffer_head *	bh;
 	unsigned int		sequence;
 	int			blocktype;
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index bf7fd7117817..e1b3c8af4d17 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -580,7 +580,7 @@ repeat:
 		 */
 		JBUFFER_TRACE(jh, "Unexpected dirty buffer");
 		jbd_unexpected_dirty_buffer(jh);
- 	}
+	}
 
 	unlock_buffer(bh);
 
@@ -1373,7 +1373,7 @@ int journal_stop(handle_t *handle)
 	if (handle->h_sync ||
 			transaction->t_outstanding_credits >
 				journal->j_max_transaction_buffers ||
-	    		time_after_eq(jiffies, transaction->t_expires)) {
+			time_after_eq(jiffies, transaction->t_expires)) {
 		/* Do this even for aborted journals: an abort still
 		 * completes the commit thread, it just doesn't write
 		 * anything to disk. */
@@ -1908,7 +1908,7 @@ void journal_invalidatepage(journal_t *journal,
 		next = bh->b_this_page;
 
 		if (offset <= curr_off) {
-		 	/* This block is wholly outside the truncation point */
+			/* This block is wholly outside the truncation point */
 			lock_buffer(bh);
 			may_free &= journal_unmap_buffer(journal, bh);
 			unlock_buffer(bh);
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 0eed918b3816..369c67502346 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -473,7 +473,7 @@ struct ext3_super_block {
 	__u8	s_reserved_char_pad;
 	__u16	s_reserved_word_pad;
 	__le32	s_default_mount_opts;
-	__le32	s_first_meta_bg; 	/* First metablock block group */
+	__le32	s_first_meta_bg;	/* First metablock block group */
 	__u32	s_reserved[190];	/* Padding to the end of the block */
 };
 
diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h
index 2f18b9511f21..4395e5206746 100644
--- a/include/linux/ext3_fs_i.h
+++ b/include/linux/ext3_fs_i.h
@@ -35,7 +35,7 @@ struct ext3_reserve_window {
 };
 
 struct ext3_reserve_window_node {
-	struct rb_node	 	rsv_node;
+	struct rb_node		rsv_node;
 	__u32			rsv_goal_size;
 	__u32			rsv_alloc_hit;
 	struct ext3_reserve_window	rsv_window;
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index dc262624efa5..a6d9daa38c6d 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -64,7 +64,7 @@ extern int journal_enable_debug;
 		if ((n) <= journal_enable_debug) {			\
 			printk (KERN_DEBUG "(%s, %d): %s: ",		\
 				__FILE__, __LINE__, __FUNCTION__);	\
-		  	printk (f, ## a);				\
+			printk (f, ## a);				\
 		}							\
 	} while (0)
 #else
@@ -201,9 +201,9 @@ typedef struct journal_superblock_s
 
 /* 0x0024 */
 	/* Remaining fields are only valid in a version-2 superblock */
-	__be32	s_feature_compat; 	/* compatible feature set */
-	__be32	s_feature_incompat; 	/* incompatible feature set */
-	__be32	s_feature_ro_compat; 	/* readonly-compatible feature set */
+	__be32	s_feature_compat;	/* compatible feature set */
+	__be32	s_feature_incompat;	/* incompatible feature set */
+	__be32	s_feature_ro_compat;	/* readonly-compatible feature set */
 /* 0x0030 */
 	__u8	s_uuid[16];		/* 128-bit uuid for journal */
 
@@ -699,7 +699,7 @@ struct journal_s
 	wait_queue_head_t	j_wait_updates;
 
 	/* Semaphore for locking against concurrent checkpoints */
-	struct mutex	 	j_checkpoint_mutex;
+	struct mutex		j_checkpoint_mutex;
 
 	/*
 	 * Journal head: identifies the first unused block in the journal.
-- 
cgit v1.2.1


From a4e4de36dc446b2193bdc8ebb96a96e44b69dd94 Mon Sep 17 00:00:00 2001
From: Dave Kleikamp <shaggy@austin.ibm.com>
Date: Wed, 27 Sep 2006 01:49:36 -0700
Subject: [PATCH] ext3: Fix sparse warnings

Fixing up some endian-ness warnings in preparation to clone ext4 from ext3.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/resize.c        | 21 +++++++++++----------
 fs/ext3/super.c         |  4 +---
 fs/jbd/journal.c        |  2 +-
 include/linux/ext3_fs.h |  2 +-
 4 files changed, 14 insertions(+), 15 deletions(-)

diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 5e1337fd878a..e186f7fb698b 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -336,7 +336,7 @@ static int verify_reserved_gdb(struct super_block *sb,
 	unsigned five = 5;
 	unsigned seven = 7;
 	unsigned grp;
-	__u32 *p = (__u32 *)primary->b_data;
+	__le32 *p = (__le32 *)primary->b_data;
 	int gdbackups = 0;
 
 	while ((grp = ext3_list_backups(sb, &three, &five, &seven)) < end) {
@@ -380,7 +380,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	struct buffer_head *dind;
 	int gdbackups;
 	struct ext3_iloc iloc;
-	__u32 *data;
+	__le32 *data;
 	int err;
 
 	if (test_opt(sb, DEBUG))
@@ -417,7 +417,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 		goto exit_bh;
 	}
 
-	data = (__u32 *)dind->b_data;
+	data = (__le32 *)dind->b_data;
 	if (le32_to_cpu(data[gdb_num % EXT3_ADDR_PER_BLOCK(sb)]) != gdblock) {
 		ext3_warning(sb, __FUNCTION__,
 			     "new group %u GDT block "E3FSBLK" not reserved",
@@ -519,7 +519,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	struct buffer_head *dind;
 	struct ext3_iloc iloc;
 	ext3_fsblk_t blk;
-	__u32 *data, *end;
+	__le32 *data, *end;
 	int gdbackups = 0;
 	int res, i;
 	int err;
@@ -536,8 +536,8 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	}
 
 	blk = EXT3_SB(sb)->s_sbh->b_blocknr + 1 + EXT3_SB(sb)->s_gdb_count;
-	data = (__u32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count;
-	end = (__u32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb);
+	data = (__le32 *)dind->b_data + EXT3_SB(sb)->s_gdb_count;
+	end = (__le32 *)dind->b_data + EXT3_ADDR_PER_BLOCK(sb);
 
 	/* Get each reserved primary GDT block and verify it holds backups */
 	for (res = 0; res < reserved_gdb; res++, blk++) {
@@ -545,7 +545,8 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 			ext3_warning(sb, __FUNCTION__,
 				     "reserved block "E3FSBLK
 				     " not at offset %ld",
-				     blk, (long)(data - (__u32 *)dind->b_data));
+				     blk,
+				     (long)(data - (__le32 *)dind->b_data));
 			err = -EINVAL;
 			goto exit_bh;
 		}
@@ -560,7 +561,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 			goto exit_bh;
 		}
 		if (++data >= end)
-			data = (__u32 *)dind->b_data;
+			data = (__le32 *)dind->b_data;
 	}
 
 	for (i = 0; i < reserved_gdb; i++) {
@@ -584,7 +585,7 @@ static int reserve_backup_gdb(handle_t *handle, struct inode *inode,
 	blk = input->group * EXT3_BLOCKS_PER_GROUP(sb);
 	for (i = 0; i < reserved_gdb; i++) {
 		int err2;
-		data = (__u32 *)primary[i]->b_data;
+		data = (__le32 *)primary[i]->b_data;
 		/* printk("reserving backup %lu[%u] = %lu\n",
 		       primary[i]->b_blocknr, gdbackups,
 		       blk + primary[i]->b_blocknr); */
@@ -689,7 +690,7 @@ exit_err:
 			     "can't update backup for group %d (err %d), "
 			     "forcing fsck on next reboot", group, err);
 		sbi->s_mount_state &= ~EXT3_VALID_FS;
-		sbi->s_es->s_state &= ~cpu_to_le16(EXT3_VALID_FS);
+		sbi->s_es->s_state &= cpu_to_le16(~EXT3_VALID_FS);
 		mark_buffer_dirty(sbi->s_sbh);
 	}
 }
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 0544325d9d7e..10985017765b 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2348,10 +2348,8 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data)
 			 */
 			ext3_clear_journal_err(sb, es);
 			sbi->s_mount_state = le16_to_cpu(es->s_state);
-			if ((ret = ext3_group_extend(sb, es, n_blocks_count))) {
-				err = ret;
+			if ((err = ext3_group_extend(sb, es, n_blocks_count)))
 				goto restore_opts;
-			}
 			if (!ext3_setup_super (sb, es, 0))
 				sb->s_flags &= ~MS_RDONLY;
 		}
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index b571209d6a1e..2fc66c3e6681 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -1094,7 +1094,7 @@ int journal_load(journal_t *journal)
 	/*
 	 * Create a slab for this blocksize
 	 */
-	err = journal_create_jbd_slab(cpu_to_be32(sb->s_blocksize));
+	err = journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize));
 	if (err)
 		return err;
 
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 369c67502346..cc08f56750da 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -460,7 +460,7 @@ struct ext3_super_block {
 	 */
 	__u8	s_prealloc_blocks;	/* Nr of blocks to try to preallocate*/
 	__u8	s_prealloc_dir_blocks;	/* Nr to preallocate for dirs */
-	__u16	s_reserved_gdt_blocks;	/* Per group desc for online growth */
+	__le16	s_reserved_gdt_blocks;	/* Per group desc for online growth */
 	/*
 	 * Journaling support valid if EXT3_FEATURE_COMPAT_HAS_JOURNAL set.
 	 */
-- 
cgit v1.2.1


From 32c2d2bc4bed61323f14f2a7d69ccbd567253d8a Mon Sep 17 00:00:00 2001
From: Eric Sandeen <esandeen@redhat.com>
Date: Wed, 27 Sep 2006 01:49:36 -0700
Subject: [PATCH] more ext3 16T overflow fixes

Some of the changes in balloc.c are just cosmetic, as Andreas pointed out -
if they overflow they'll then underflow and things are fine.

5th hunk actually fixes an overflow problem.

Also check for potential overflows in inode & block counts when resizing.

Signed-off-by: Eric Sandeen <esandeen@redhat.com>
Cc: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext3/balloc.c | 12 ++++++------
 fs/ext3/resize.c | 17 +++++++++++++++++
 2 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c
index c19be8fc3e51..b41a7d7e20f0 100644
--- a/fs/ext3/balloc.c
+++ b/fs/ext3/balloc.c
@@ -202,7 +202,7 @@ goal_in_my_reservation(struct ext3_reserve_window *rsv, ext3_grpblk_t grp_goal,
 	ext3_fsblk_t group_first_block, group_last_block;
 
 	group_first_block = ext3_group_first_block_no(sb, group);
-	group_last_block = group_first_block + EXT3_BLOCKS_PER_GROUP(sb) - 1;
+	group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1);
 
 	if ((rsv->_rsv_start > group_last_block) ||
 	    (rsv->_rsv_end < group_first_block))
@@ -1047,7 +1047,7 @@ static int alloc_new_reservation(struct ext3_reserve_window_node *my_rsv,
 	spinlock_t *rsv_lock = &EXT3_SB(sb)->s_rsv_window_lock;
 
 	group_first_block = ext3_group_first_block_no(sb, group);
-	group_end_block = group_first_block + EXT3_BLOCKS_PER_GROUP(sb) - 1;
+	group_end_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1);
 
 	if (grp_goal < 0)
 		start_block = group_first_block;
@@ -1239,7 +1239,7 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 			struct ext3_reserve_window_node * my_rsv,
 			unsigned long *count, int *errp)
 {
-	ext3_fsblk_t group_first_block;
+	ext3_fsblk_t group_first_block, group_last_block;
 	ext3_grpblk_t ret = 0;
 	int fatal;
 	unsigned long num = *count;
@@ -1276,6 +1276,7 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 	 * first block is the block number of the first block in this group
 	 */
 	group_first_block = ext3_group_first_block_no(sb, group);
+	group_last_block = group_first_block + (EXT3_BLOCKS_PER_GROUP(sb) - 1);
 
 	/*
 	 * Basically we will allocate a new block from inode's reservation
@@ -1311,9 +1312,8 @@ ext3_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle,
 			try_to_extend_reservation(my_rsv, sb,
 					*count-my_rsv->rsv_end + grp_goal - 1);
 
-		if ((my_rsv->rsv_start >= group_first_block +
-					EXT3_BLOCKS_PER_GROUP(sb))
-		    || (my_rsv->rsv_end < group_first_block)) {
+		if ((my_rsv->rsv_start > group_last_block) ||
+				(my_rsv->rsv_end < group_first_block)) {
 			rsv_window_dump(&EXT3_SB(sb)->s_rsv_window_root, 1);
 			BUG();
 		}
diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index e186f7fb698b..9e318a5cfc87 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -731,6 +731,18 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input)
 		return -EPERM;
 	}
 
+	if (le32_to_cpu(es->s_blocks_count) + input->blocks_count <
+	    le32_to_cpu(es->s_blocks_count)) {
+		ext3_warning(sb, __FUNCTION__, "blocks_count overflow\n");
+		return -EINVAL;
+	}
+
+	if (le32_to_cpu(es->s_inodes_count) + EXT3_INODES_PER_GROUP(sb) <
+	    le32_to_cpu(es->s_inodes_count)) {
+		ext3_warning(sb, __FUNCTION__, "inodes_count overflow\n");
+		return -EINVAL;
+	}
+
 	if (reserved_gdb || gdb_off == 0) {
 		if (!EXT3_HAS_COMPAT_FEATURE(sb,
 					     EXT3_FEATURE_COMPAT_RESIZE_INODE)){
@@ -959,6 +971,11 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es,
 
 	add = EXT3_BLOCKS_PER_GROUP(sb) - last;
 
+	if (o_blocks_count + add < o_blocks_count) {
+		ext3_warning(sb, __FUNCTION__, "blocks_count overflow");
+		return -EINVAL;
+	}
+
 	if (o_blocks_count + add > n_blocks_count)
 		add = n_blocks_count - o_blocks_count;
 
-- 
cgit v1.2.1


From f8314dc60ccba7e41f425048c4160dc7f63377d5 Mon Sep 17 00:00:00 2001
From: Panagiotis Issaris <takis@issaris.org>
Date: Wed, 27 Sep 2006 01:49:37 -0700
Subject: [PATCH] fs: Conversions from kmalloc+memset to k(z|c)alloc

Conversions from kmalloc+memset to kzalloc.

Signed-off-by: Panagiotis Issaris <takis@issaris.org>
Jffs2-bit-acked-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/adfs/super.c     | 3 +--
 fs/affs/super.c     | 3 +--
 fs/afs/vlocation.c  | 3 +--
 fs/afs/volume.c     | 3 +--
 fs/autofs/inode.c   | 3 +--
 fs/bfs/inode.c      | 6 ++----
 fs/configfs/file.c  | 3 +--
 fs/configfs/inode.c | 3 +--
 fs/cramfs/inode.c   | 3 +--
 fs/efs/super.c      | 3 +--
 fs/ext2/super.c     | 3 +--
 fs/ext2/xattr.c     | 3 +--
 fs/ext3/dir.c       | 3 +--
 fs/ext3/super.c     | 3 +--
 fs/fat/inode.c      | 3 +--
 fs/hfs/bnode.c      | 3 +--
 fs/hfs/btree.c      | 3 +--
 fs/hfs/super.c      | 3 +--
 fs/hfsplus/bnode.c  | 3 +--
 fs/hfsplus/btree.c  | 3 +--
 fs/hpfs/super.c     | 3 +--
 fs/isofs/inode.c    | 3 +--
 fs/jffs2/super.c    | 3 +--
 fs/lockd/host.c     | 4 ++--
 fs/lockd/svcsubs.c  | 3 +--
 fs/minix/inode.c    | 6 ++----
 fs/ncpfs/inode.c    | 3 +--
 fs/nfsd/nfs4idmap.c | 3 +--
 fs/nfsd/nfs4state.c | 3 +--
 fs/partitions/efi.c | 9 +++------
 fs/proc/kcore.c     | 6 ++----
 fs/qnx4/inode.c     | 3 +--
 fs/sysv/super.c     | 6 ++----
 fs/udf/ialloc.c     | 6 ++----
 fs/ufs/super.c      | 3 +--
 35 files changed, 43 insertions(+), 84 deletions(-)

diff --git a/fs/adfs/super.c b/fs/adfs/super.c
index 82011019494c..959dbf1f8e63 100644
--- a/fs/adfs/super.c
+++ b/fs/adfs/super.c
@@ -339,11 +339,10 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent)
 
 	sb->s_flags |= MS_NODIRATIME;
 
-	asb = kmalloc(sizeof(*asb), GFP_KERNEL);
+	asb = kzalloc(sizeof(*asb), GFP_KERNEL);
 	if (!asb)
 		return -ENOMEM;
 	sb->s_fs_info = asb;
-	memset(asb, 0, sizeof(*asb));
 
 	/* set default options */
 	asb->s_uid = 0;
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 17352011ab67..b35f514479bf 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -280,11 +280,10 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_op                = &affs_sops;
 	sb->s_flags |= MS_NODIRATIME;
 
-	sbi = kmalloc(sizeof(struct affs_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct affs_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(*sbi));
 	init_MUTEX(&sbi->s_bmlock);
 
 	if (!parse_options(data,&uid,&gid,&i,&reserved,&root_block,
diff --git a/fs/afs/vlocation.c b/fs/afs/vlocation.c
index 331f730a1fb3..782ee7c600ca 100644
--- a/fs/afs/vlocation.c
+++ b/fs/afs/vlocation.c
@@ -281,11 +281,10 @@ int afs_vlocation_lookup(struct afs_cell *cell,
 	spin_unlock(&cell->vl_gylock);
 
 	/* not in the cell's in-memory lists - create a new record */
-	vlocation = kmalloc(sizeof(struct afs_vlocation), GFP_KERNEL);
+	vlocation = kzalloc(sizeof(struct afs_vlocation), GFP_KERNEL);
 	if (!vlocation)
 		return -ENOMEM;
 
-	memset(vlocation, 0, sizeof(struct afs_vlocation));
 	atomic_set(&vlocation->usage, 1);
 	INIT_LIST_HEAD(&vlocation->link);
 	rwlock_init(&vlocation->lock);
diff --git a/fs/afs/volume.c b/fs/afs/volume.c
index 0ff4b86476e3..768c6dbd323a 100644
--- a/fs/afs/volume.c
+++ b/fs/afs/volume.c
@@ -186,11 +186,10 @@ int afs_volume_lookup(const char *name, struct afs_cell *cell, int rwpath,
 	_debug("creating new volume record");
 
 	ret = -ENOMEM;
-	volume = kmalloc(sizeof(struct afs_volume), GFP_KERNEL);
+	volume = kzalloc(sizeof(struct afs_volume), GFP_KERNEL);
 	if (!volume)
 		goto error_up;
 
-	memset(volume, 0, sizeof(struct afs_volume));
 	atomic_set(&volume->usage, 1);
 	volume->type		= type;
 	volume->type_force	= force;
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index af2efbbb5d76..c81d6b8c2828 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -129,10 +129,9 @@ int autofs_fill_super(struct super_block *s, void *data, int silent)
 	struct autofs_sb_info *sbi;
 	int minproto, maxproto;
 
-	sbi = kmalloc(sizeof(*sbi), GFP_KERNEL);
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if ( !sbi )
 		goto fail_unlock;
-	memset(sbi, 0, sizeof(*sbi));
 	DPRINTK(("autofs: starting up, sbi = %p\n",sbi));
 
 	s->s_fs_info = sbi;
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index cf74f3d4d966..3e4d6c767b9b 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -311,11 +311,10 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 	unsigned i, imap_len;
 	struct bfs_sb_info * info;
 
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
 	if (!info)
 		return -ENOMEM;
 	s->s_fs_info = info;
-	memset(info, 0, sizeof(*info));
 
 	sb_set_blocksize(s, BFS_BSIZE);
 
@@ -338,10 +337,9 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 			+ BFS_ROOT_INO - 1;
 
 	imap_len = info->si_lasti/8 + 1;
-	info->si_imap = kmalloc(imap_len, GFP_KERNEL);
+	info->si_imap = kzalloc(imap_len, GFP_KERNEL);
 	if (!info->si_imap)
 		goto out;
-	memset(info->si_imap, 0, imap_len);
 	for (i=0; i<BFS_ROOT_INO; i++) 
 		set_bit(i, info->si_imap);
 
diff --git a/fs/configfs/file.c b/fs/configfs/file.c
index f499803743e0..85105e50f7db 100644
--- a/fs/configfs/file.c
+++ b/fs/configfs/file.c
@@ -274,9 +274,8 @@ static int check_perm(struct inode * inode, struct file * file)
 	/* No error? Great, allocate a buffer for the file, and store it
 	 * it in file->private_data for easy access.
 	 */
-	buffer = kmalloc(sizeof(struct configfs_buffer),GFP_KERNEL);
+	buffer = kzalloc(sizeof(struct configfs_buffer),GFP_KERNEL);
 	if (buffer) {
-		memset(buffer,0,sizeof(struct configfs_buffer));
 		init_MUTEX(&buffer->sem);
 		buffer->needs_read_fill = 1;
 		buffer->ops = ops;
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index e14488ca6411..5047e6a7581e 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -76,11 +76,10 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr)
 
 	if (!sd_iattr) {
 		/* setting attributes for the first time, allocate now */
-		sd_iattr = kmalloc(sizeof(struct iattr), GFP_KERNEL);
+		sd_iattr = kzalloc(sizeof(struct iattr), GFP_KERNEL);
 		if (!sd_iattr)
 			return -ENOMEM;
 		/* assign default attributes */
-		memset(sd_iattr, 0, sizeof(struct iattr));
 		sd_iattr->ia_mode = sd->s_mode;
 		sd_iattr->ia_uid = 0;
 		sd_iattr->ia_gid = 0;
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 223c0431042d..d09b6777c41a 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -242,11 +242,10 @@ static int cramfs_fill_super(struct super_block *sb, void *data, int silent)
 
 	sb->s_flags |= MS_RDONLY;
 
-	sbi = kmalloc(sizeof(struct cramfs_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct cramfs_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(struct cramfs_sb_info));
 
 	/* Invalidate the read buffers on mount: think disk change.. */
 	mutex_lock(&read_mutex);
diff --git a/fs/efs/super.c b/fs/efs/super.c
index 8ac2462ae5dd..7089269ee9ae 100644
--- a/fs/efs/super.c
+++ b/fs/efs/super.c
@@ -248,11 +248,10 @@ static int efs_fill_super(struct super_block *s, void *d, int silent)
 	struct buffer_head *bh;
 	struct inode *root;
 
- 	sb = kmalloc(sizeof(struct efs_sb_info), GFP_KERNEL);
+ 	sb = kzalloc(sizeof(struct efs_sb_info), GFP_KERNEL);
 	if (!sb)
 		return -ENOMEM;
 	s->s_fs_info = sb;
-	memset(sb, 0, sizeof(struct efs_sb_info));
  
 	s->s_magic		= EFS_SUPER_MAGIC;
 	if (!sb_set_blocksize(s, EFS_BLOCKSIZE)) {
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index d978d3febb8c..69435fbe53f2 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -655,11 +655,10 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 	int i, j;
 	__le32 features;
 
-	sbi = kmalloc(sizeof(*sbi), GFP_KERNEL);
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(*sbi));
 
 	/*
 	 * See what the current blocksize for the device is, and
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index 86ae8e93adb9..af52a7f8b291 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -521,11 +521,10 @@ bad_block:		ext2_error(sb, "ext2_xattr_set",
 		}
 	} else {
 		/* Allocate a buffer where we construct the new block. */
-		header = kmalloc(sb->s_blocksize, GFP_KERNEL);
+		header = kzalloc(sb->s_blocksize, GFP_KERNEL);
 		error = -ENOMEM;
 		if (header == NULL)
 			goto cleanup;
-		memset(header, 0, sb->s_blocksize);
 		end = (char *)header + sb->s_blocksize;
 		header->h_magic = cpu_to_le32(EXT2_XATTR_MAGIC);
 		header->h_blocks = header->h_refcount = cpu_to_le32(1);
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 4d97f437cd48..429acbb4e064 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -343,10 +343,9 @@ int ext3_htree_store_dirent(struct file *dir_file, __u32 hash,
 
 	/* Create and allocate the fname structure */
 	len = sizeof(struct fname) + dirent->name_len + 1;
-	new_fn = kmalloc(len, GFP_KERNEL);
+	new_fn = kzalloc(len, GFP_KERNEL);
 	if (!new_fn)
 		return -ENOMEM;
-	memset(new_fn, 0, len);
 	new_fn->hash = hash;
 	new_fn->minor_hash = minor_hash;
 	new_fn->inode = le32_to_cpu(dirent->inode);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 10985017765b..09b4b313ca07 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1407,11 +1407,10 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
 	int needs_recovery;
 	__le32 features;
 
-	sbi = kmalloc(sizeof(*sbi), GFP_KERNEL);
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(*sbi));
 	sbi->s_mount_opt = 0;
 	sbi->s_resuid = EXT3_DEF_RESUID;
 	sbi->s_resgid = EXT3_DEF_RESGID;
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 31b7174176ba..bc4da3a48c8c 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -1168,11 +1168,10 @@ int fat_fill_super(struct super_block *sb, void *data, int silent,
 	long error;
 	char buf[50];
 
-	sbi = kmalloc(sizeof(struct msdos_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct msdos_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(struct msdos_sb_info));
 
 	sb->s_flags |= MS_NODIRATIME;
 	sb->s_magic = MSDOS_SUPER_MAGIC;
diff --git a/fs/hfs/bnode.c b/fs/hfs/bnode.c
index 13231dd5ce66..0d200068d0af 100644
--- a/fs/hfs/bnode.c
+++ b/fs/hfs/bnode.c
@@ -249,10 +249,9 @@ static struct hfs_bnode *__hfs_bnode_create(struct hfs_btree *tree, u32 cnid)
 	sb = tree->inode->i_sb;
 	size = sizeof(struct hfs_bnode) + tree->pages_per_bnode *
 		sizeof(struct page *);
-	node = kmalloc(size, GFP_KERNEL);
+	node = kzalloc(size, GFP_KERNEL);
 	if (!node)
 		return NULL;
-	memset(node, 0, size);
 	node->tree = tree;
 	node->this = cnid;
 	set_bit(HFS_BNODE_NEW, &node->flags);
diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c
index 400357994319..5fd0ed71f923 100644
--- a/fs/hfs/btree.c
+++ b/fs/hfs/btree.c
@@ -21,10 +21,9 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id, btree_keycmp ke
 	struct page *page;
 	unsigned int size;
 
-	tree = kmalloc(sizeof(*tree), GFP_KERNEL);
+	tree = kzalloc(sizeof(*tree), GFP_KERNEL);
 	if (!tree)
 		return NULL;
-	memset(tree, 0, sizeof(*tree));
 
 	init_MUTEX(&tree->tree_lock);
 	spin_lock_init(&tree->hash_lock);
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index 34937ee83ab1..05dffff1a372 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -356,11 +356,10 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent)
 	struct inode *root_inode;
 	int res;
 
-	sbi = kmalloc(sizeof(struct hfs_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct hfs_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(struct hfs_sb_info));
 	INIT_HLIST_HEAD(&sbi->rsrc_inodes);
 
 	res = -EINVAL;
diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c
index 77bf434da679..29da6574ba77 100644
--- a/fs/hfsplus/bnode.c
+++ b/fs/hfsplus/bnode.c
@@ -409,10 +409,9 @@ static struct hfs_bnode *__hfs_bnode_create(struct hfs_btree *tree, u32 cnid)
 	sb = tree->inode->i_sb;
 	size = sizeof(struct hfs_bnode) + tree->pages_per_bnode *
 		sizeof(struct page *);
-	node = kmalloc(size, GFP_KERNEL);
+	node = kzalloc(size, GFP_KERNEL);
 	if (!node)
 		return NULL;
-	memset(node, 0, size);
 	node->tree = tree;
 	node->this = cnid;
 	set_bit(HFS_BNODE_NEW, &node->flags);
diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c
index cfc852fdd1b5..a9b9e872e29a 100644
--- a/fs/hfsplus/btree.c
+++ b/fs/hfsplus/btree.c
@@ -24,10 +24,9 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id)
 	struct page *page;
 	unsigned int size;
 
-	tree = kmalloc(sizeof(*tree), GFP_KERNEL);
+	tree = kzalloc(sizeof(*tree), GFP_KERNEL);
 	if (!tree)
 		return NULL;
-	memset(tree, 0, sizeof(*tree));
 
 	init_MUTEX(&tree->tree_lock);
 	spin_lock_init(&tree->hash_lock);
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index 8fe51c343786..b76d60832375 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -462,11 +462,10 @@ static int hpfs_fill_super(struct super_block *s, void *options, int silent)
 
 	int o;
 
-	sbi = kmalloc(sizeof(*sbi), GFP_KERNEL);
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	s->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(*sbi));
 
 	sbi->sb_bmp_dir = NULL;
 	sbi->sb_cp_table = NULL;
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 14391361c886..ab15e9072151 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -557,11 +557,10 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent)
 	struct iso9660_options		opt;
 	struct isofs_sb_info	      * sbi;
 
-	sbi = kmalloc(sizeof(*sbi), GFP_KERNEL);
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	s->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(*sbi));
 
 	if (!parse_options((char *)data, &opt))
 		goto out_freesbi;
diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c
index 68e3953419b4..6de374513c01 100644
--- a/fs/jffs2/super.c
+++ b/fs/jffs2/super.c
@@ -119,10 +119,9 @@ static int jffs2_get_sb_mtd(struct file_system_type *fs_type,
 	struct jffs2_sb_info *c;
 	int ret;
 
-	c = kmalloc(sizeof(*c), GFP_KERNEL);
+	c = kzalloc(sizeof(*c), GFP_KERNEL);
 	if (!c)
 		return -ENOMEM;
-	memset(c, 0, sizeof(*c));
 	c->mtd = mtd;
 
 	sb = sget(fs_type, jffs2_sb_compare, jffs2_sb_set, c);
diff --git a/fs/lockd/host.c b/fs/lockd/host.c
index 703fb038c813..a0d0b58ce7a4 100644
--- a/fs/lockd/host.c
+++ b/fs/lockd/host.c
@@ -99,9 +99,9 @@ nlm_lookup_host(int server, struct sockaddr_in *sin,
 	/* Ooops, no host found, create it */
 	dprintk("lockd: creating host entry\n");
 
-	if (!(host = (struct nlm_host *) kmalloc(sizeof(*host), GFP_KERNEL)))
+	host = kzalloc(sizeof(*host), GFP_KERNEL);
+	if (!host)
 		goto nohost;
-	memset(host, 0, sizeof(*host));
 
 	addr = sin->sin_addr.s_addr;
 	sprintf(host->h_name, "%u.%u.%u.%u", NIPQUAD(addr));
diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c
index 01b4db9e5466..a92dd98f8401 100644
--- a/fs/lockd/svcsubs.c
+++ b/fs/lockd/svcsubs.c
@@ -100,11 +100,10 @@ nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result,
 	nlm_debug_print_fh("creating file for", f);
 
 	nfserr = nlm_lck_denied_nolocks;
-	file = (struct nlm_file *) kmalloc(sizeof(*file), GFP_KERNEL);
+	file = kzalloc(sizeof(*file), GFP_KERNEL);
 	if (!file)
 		goto out_unlock;
 
-	memset(file, 0, sizeof(*file));
 	memcpy(&file->f_handle, f, sizeof(struct nfs_fh));
 	file->f_hash = hash;
 	init_MUTEX(&file->f_sema);
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 330ff9fc7cf0..e24be7fc0651 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -145,11 +145,10 @@ static int minix_fill_super(struct super_block *s, void *data, int silent)
 	struct inode *root_inode;
 	struct minix_sb_info *sbi;
 
-	sbi = kmalloc(sizeof(struct minix_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct minix_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
 	s->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(struct minix_sb_info));
 
 	/* N.B. These should be compile-time tests.
 	   Unfortunately that is impossible. */
@@ -207,10 +206,9 @@ static int minix_fill_super(struct super_block *s, void *data, int silent)
 	if (sbi->s_imap_blocks == 0 || sbi->s_zmap_blocks == 0)
 		goto out_illegal_sb;
 	i = (sbi->s_imap_blocks + sbi->s_zmap_blocks) * sizeof(bh);
-	map = kmalloc(i, GFP_KERNEL);
+	map = kzalloc(i, GFP_KERNEL);
 	if (!map)
 		goto out_no_map;
-	memset(map, 0, i);
 	sbi->s_imap = &map[0];
 	sbi->s_zmap = &map[sbi->s_imap_blocks];
 
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 1ddf77b0b825..752f02e43ab0 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -411,11 +411,10 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
 #endif
 	struct ncp_entry_info finfo;
 
-	server = kmalloc(sizeof(struct ncp_server), GFP_KERNEL);
+	server = kzalloc(sizeof(struct ncp_server), GFP_KERNEL);
 	if (!server)
 		return -ENOMEM;
 	sb->s_fs_info = server;
-	memset(server, 0, sizeof(struct ncp_server));
 
 	error = -EFAULT;
 	if (raw_data == NULL)
diff --git a/fs/nfsd/nfs4idmap.c b/fs/nfsd/nfs4idmap.c
index bea6b9478114..b1902ebaab41 100644
--- a/fs/nfsd/nfs4idmap.c
+++ b/fs/nfsd/nfs4idmap.c
@@ -573,10 +573,9 @@ idmap_lookup(struct svc_rqst *rqstp,
 	struct idmap_defer_req *mdr;
 	int ret;
 
-	mdr = kmalloc(sizeof(*mdr), GFP_KERNEL);
+	mdr = kzalloc(sizeof(*mdr), GFP_KERNEL);
 	if (!mdr)
 		return -ENOMEM;
-	memset(mdr, 0, sizeof(*mdr));
 	atomic_set(&mdr->count, 1);
 	init_waitqueue_head(&mdr->waitq);
 	mdr->req.defer = idmap_defer;
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 9daa0b9feb8d..1a25c01ecdf1 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -339,8 +339,7 @@ alloc_client(struct xdr_netobj name)
 {
 	struct nfs4_client *clp;
 
-	if ((clp = kmalloc(sizeof(struct nfs4_client), GFP_KERNEL))!= NULL) {
-		memset(clp, 0, sizeof(*clp));
+	if ((clp = kzalloc(sizeof(struct nfs4_client), GFP_KERNEL))!= NULL) {
 		if ((clp->cl_name.data = kmalloc(name.len, GFP_KERNEL)) != NULL) {
 			memcpy(clp->cl_name.data, name.data, name.len);
 			clp->cl_name.len = name.len;
diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c
index 63730282ad81..1bea610078b3 100644
--- a/fs/partitions/efi.c
+++ b/fs/partitions/efi.c
@@ -238,10 +238,9 @@ alloc_read_gpt_entries(struct block_device *bdev, gpt_header *gpt)
                 le32_to_cpu(gpt->sizeof_partition_entry);
 	if (!count)
 		return NULL;
-	pte = kmalloc(count, GFP_KERNEL);
+	pte = kzalloc(count, GFP_KERNEL);
 	if (!pte)
 		return NULL;
-	memset(pte, 0, count);
 
 	if (read_lba(bdev, le64_to_cpu(gpt->partition_entry_lba),
                      (u8 *) pte,
@@ -269,10 +268,9 @@ alloc_read_gpt_header(struct block_device *bdev, u64 lba)
 	if (!bdev)
 		return NULL;
 
-	gpt = kmalloc(sizeof (gpt_header), GFP_KERNEL);
+	gpt = kzalloc(sizeof (gpt_header), GFP_KERNEL);
 	if (!gpt)
 		return NULL;
-	memset(gpt, 0, sizeof (gpt_header));
 
 	if (read_lba(bdev, lba, (u8 *) gpt,
 		     sizeof (gpt_header)) < sizeof (gpt_header)) {
@@ -526,9 +524,8 @@ find_valid_gpt(struct block_device *bdev, gpt_header **gpt, gpt_entry **ptes)
 	lastlba = last_lba(bdev);
         if (!force_gpt) {
                 /* This will be added to the EFI Spec. per Intel after v1.02. */
-                legacymbr = kmalloc(sizeof (*legacymbr), GFP_KERNEL);
+                legacymbr = kzalloc(sizeof (*legacymbr), GFP_KERNEL);
                 if (legacymbr) {
-                        memset(legacymbr, 0, sizeof (*legacymbr));
                         read_lba(bdev, 0, (u8 *) legacymbr,
                                  sizeof (*legacymbr));
                         good_pmbr = is_pmbr_valid(legacymbr, lastlba);
diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
index 6a984f64edd7..3ceff3857272 100644
--- a/fs/proc/kcore.c
+++ b/fs/proc/kcore.c
@@ -279,12 +279,11 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 		tsz = elf_buflen - *fpos;
 		if (buflen < tsz)
 			tsz = buflen;
-		elf_buf = kmalloc(elf_buflen, GFP_ATOMIC);
+		elf_buf = kzalloc(elf_buflen, GFP_ATOMIC);
 		if (!elf_buf) {
 			read_unlock(&kclist_lock);
 			return -ENOMEM;
 		}
-		memset(elf_buf, 0, elf_buflen);
 		elf_kcore_store_hdr(elf_buf, nphdr, elf_buflen);
 		read_unlock(&kclist_lock);
 		if (copy_to_user(buffer, elf_buf + *fpos, tsz)) {
@@ -330,10 +329,9 @@ read_kcore(struct file *file, char __user *buffer, size_t buflen, loff_t *fpos)
 			unsigned long curstart = start;
 			unsigned long cursize = tsz;
 
-			elf_buf = kmalloc(tsz, GFP_KERNEL);
+			elf_buf = kzalloc(tsz, GFP_KERNEL);
 			if (!elf_buf)
 				return -ENOMEM;
-			memset(elf_buf, 0, tsz);
 
 			read_lock(&vmlist_lock);
 			for (m=vmlist; m && cursize; m=m->next) {
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 5a903491e697..8497609f5022 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -358,11 +358,10 @@ static int qnx4_fill_super(struct super_block *s, void *data, int silent)
 	const char *errmsg;
 	struct qnx4_sb_info *qs;
 
-	qs = kmalloc(sizeof(struct qnx4_sb_info), GFP_KERNEL);
+	qs = kzalloc(sizeof(struct qnx4_sb_info), GFP_KERNEL);
 	if (!qs)
 		return -ENOMEM;
 	s->s_fs_info = qs;
-	memset(qs, 0, sizeof(struct qnx4_sb_info));
 
 	sb_set_blocksize(s, QNX4_BLOCK_SIZE);
 
diff --git a/fs/sysv/super.c b/fs/sysv/super.c
index 876639b93321..350cba5d6803 100644
--- a/fs/sysv/super.c
+++ b/fs/sysv/super.c
@@ -369,10 +369,9 @@ static int sysv_fill_super(struct super_block *sb, void *data, int silent)
 	if (64 != sizeof (struct sysv_inode))
 		panic("sysv fs: bad inode size");
 
-	sbi = kmalloc(sizeof(struct sysv_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct sysv_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
-	memset(sbi, 0, sizeof(struct sysv_sb_info));
 
 	sbi->s_sb = sb;
 	sbi->s_block_base = 0;
@@ -453,10 +452,9 @@ static int v7_fill_super(struct super_block *sb, void *data, int silent)
 	if (64 != sizeof (struct sysv_inode))
 		panic("sysv fs: bad i-node size");
 
-	sbi = kmalloc(sizeof(struct sysv_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct sysv_sb_info), GFP_KERNEL);
 	if (!sbi)
 		return -ENOMEM;
-	memset(sbi, 0, sizeof(struct sysv_sb_info));
 
 	sbi->s_sb = sb;
 	sbi->s_block_base = 0;
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index 33323473e3c4..d1d38238ce65 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -130,14 +130,12 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
 	{
 		UDF_I_EFE(inode) = 1;
 		UDF_UPDATE_UDFREV(inode->i_sb, UDF_VERS_USE_EXTENDED_FE);
-		UDF_I_DATA(inode) = kmalloc(inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry), GFP_KERNEL);
-		memset(UDF_I_DATA(inode), 0x00, inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry));
+		UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct extendedFileEntry), GFP_KERNEL);
 	}
 	else
 	{
 		UDF_I_EFE(inode) = 0;
-		UDF_I_DATA(inode) = kmalloc(inode->i_sb->s_blocksize - sizeof(struct fileEntry), GFP_KERNEL);
-		memset(UDF_I_DATA(inode), 0x00, inode->i_sb->s_blocksize - sizeof(struct fileEntry));
+		UDF_I_DATA(inode) = kzalloc(inode->i_sb->s_blocksize - sizeof(struct fileEntry), GFP_KERNEL);
 	}
 	if (UDF_QUERY_FLAG(inode->i_sb, UDF_FLAG_USE_AD_IN_ICB))
 		UDF_I_ALLOCTYPE(inode) = ICBTAG_FLAG_AD_IN_ICB;
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 992ee0b87cc3..ef910e784034 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -611,11 +611,10 @@ static int ufs_fill_super(struct super_block *sb, void *data, int silent)
 	
 	UFSD("ENTER\n");
 		
-	sbi = kmalloc(sizeof(struct ufs_sb_info), GFP_KERNEL);
+	sbi = kzalloc(sizeof(struct ufs_sb_info), GFP_KERNEL);
 	if (!sbi)
 		goto failed_nomem;
 	sb->s_fs_info = sbi;
-	memset(sbi, 0, sizeof(struct ufs_sb_info));
 
 	UFSD("flag %u\n", (int)(sb->s_flags & MS_RDONLY));
 	
-- 
cgit v1.2.1


From f52720ca5f48574e347dff35ffe6b389ace61537 Mon Sep 17 00:00:00 2001
From: Panagiotis Issaris <takis@issaris.org>
Date: Wed, 27 Sep 2006 01:49:39 -0700
Subject: [PATCH] fs: Removing useless casts

* Removing useless casts
* Removing useless wrapper
* Conversion from kmalloc+memset to kzalloc

Signed-off-by: Panagiotis Issaris <takis@issaris.org>
Acked-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/coda/dir.c               |  2 +-
 fs/ext2/acl.c               |  4 ++--
 fs/ext3/acl.c               |  4 ++--
 fs/ext3/resize.c            |  4 ++--
 fs/hpfs/buffer.c            |  2 +-
 fs/jffs/inode-v23.c         |  8 ++++----
 fs/jffs/intrep.c            | 11 ++++-------
 fs/jffs/jffs_fm.c           |  6 ++----
 fs/jfs/jfs_txnmgr.c         |  4 ++--
 fs/lockd/clntproc.c         |  2 +-
 fs/ncpfs/symlink.c          |  4 ++--
 fs/nfs/delegation.c         |  7 +------
 fs/nfs/inode.c              |  2 +-
 fs/nfs/nfs3proc.c           |  2 +-
 fs/nfs/proc.c               |  2 +-
 fs/ntfs/dir.c               |  5 ++---
 fs/ntfs/inode.c             |  2 +-
 fs/ntfs/mft.c               |  4 ++--
 fs/ntfs/unistr.c            |  4 ++--
 fs/xfs/linux-2.6/xfs_file.c |  2 +-
 fs/xfs/linux-2.6/xfs_iops.c |  4 ++--
 21 files changed, 37 insertions(+), 48 deletions(-)

diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 71f2ea632e53..8651ea6a23b7 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -513,7 +513,7 @@ static int coda_venus_readdir(struct file *filp, filldir_t filldir,
 	ino_t ino;
 	int ret, i;
 
-	vdir = (struct venus_dirent *)kmalloc(sizeof(*vdir), GFP_KERNEL);
+	vdir = kmalloc(sizeof(*vdir), GFP_KERNEL);
 	if (!vdir) return -ENOMEM;
 
 	i = filp->f_pos;
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index da52b4a5db64..7c420b800c34 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -89,8 +89,8 @@ ext2_acl_to_disk(const struct posix_acl *acl, size_t *size)
 	size_t n;
 
 	*size = ext2_acl_size(acl->a_count);
-	ext_acl = (ext2_acl_header *)kmalloc(sizeof(ext2_acl_header) +
-		acl->a_count * sizeof(ext2_acl_entry), GFP_KERNEL);
+	ext_acl = kmalloc(sizeof(ext2_acl_header) + acl->a_count *
+			sizeof(ext2_acl_entry), GFP_KERNEL);
 	if (!ext_acl)
 		return ERR_PTR(-ENOMEM);
 	ext_acl->a_version = cpu_to_le32(EXT2_ACL_VERSION);
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index 92bf78221429..1e5038d9a01b 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -90,8 +90,8 @@ ext3_acl_to_disk(const struct posix_acl *acl, size_t *size)
 	size_t n;
 
 	*size = ext3_acl_size(acl->a_count);
-	ext_acl = (ext3_acl_header *)kmalloc(sizeof(ext3_acl_header) +
-		acl->a_count * sizeof(ext3_acl_entry), GFP_KERNEL);
+	ext_acl = kmalloc(sizeof(ext3_acl_header) + acl->a_count *
+			sizeof(ext3_acl_entry), GFP_KERNEL);
 	if (!ext_acl)
 		return ERR_PTR(-ENOMEM);
 	ext_acl->a_version = cpu_to_le32(EXT3_ACL_VERSION);
diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c
index 9e318a5cfc87..b73cba12f79c 100644
--- a/fs/ext3/resize.c
+++ b/fs/ext3/resize.c
@@ -439,8 +439,8 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
 	if ((err = ext3_reserve_inode_write(handle, inode, &iloc)))
 		goto exit_dindj;
 
-	n_group_desc = (struct buffer_head **)kmalloc((gdb_num + 1) *
-				sizeof(struct buffer_head *), GFP_KERNEL);
+	n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *),
+			GFP_KERNEL);
 	if (!n_group_desc) {
 		err = -ENOMEM;
 		ext3_warning (sb, __FUNCTION__,
diff --git a/fs/hpfs/buffer.c b/fs/hpfs/buffer.c
index 2807aa833e62..b52b7381d10f 100644
--- a/fs/hpfs/buffer.c
+++ b/fs/hpfs/buffer.c
@@ -76,7 +76,7 @@ void *hpfs_map_4sectors(struct super_block *s, unsigned secno, struct quad_buffe
 		return NULL;
 	}
 
-	qbh->data = data = (char *)kmalloc(2048, GFP_NOFS);
+	qbh->data = data = kmalloc(2048, GFP_NOFS);
 	if (!data) {
 		printk("HPFS: hpfs_map_4sectors: out of memory\n");
 		goto bail;
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index 93068697a9bf..b59553d28d13 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -652,7 +652,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	lock_kernel();
 
 	D3({
-		char *s = (char *)kmalloc(len + 1, GFP_KERNEL);
+		char *s = kmalloc(len + 1, GFP_KERNEL);
 		memcpy(s, name, len);
 		s[len] = '\0';
 		printk("jffs_lookup(): dir: 0x%p, name: \"%s\"\n", dir, s);
@@ -1173,8 +1173,8 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 	lock_kernel();
 	D1({
 		int len = dentry->d_name.len; 
-		char *_name = (char *)kmalloc(len + 1, GFP_KERNEL);
-		char *_symname = (char *)kmalloc(symname_len + 1, GFP_KERNEL);
+		char *_name = kmalloc(len + 1, GFP_KERNEL);
+		char *_symname = kmalloc(symname_len + 1, GFP_KERNEL);
 		memcpy(_name, dentry->d_name.name, len);
 		_name[len] = '\0';
 		memcpy(_symname, symname, symname_len);
@@ -1282,7 +1282,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
 	lock_kernel();
 	D1({
 		int len = dentry->d_name.len;
-		char *s = (char *)kmalloc(len + 1, GFP_KERNEL);
+		char *s = kmalloc(len + 1, GFP_KERNEL);
 		memcpy(s, dentry->d_name.name, len);
 		s[len] = '\0';
 		printk("jffs_create(): dir: 0x%p, name: \"%s\"\n", dir, s);
diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c
index 9000f1effedf..4a543e114970 100644
--- a/fs/jffs/intrep.c
+++ b/fs/jffs/intrep.c
@@ -488,13 +488,11 @@ jffs_create_file(struct jffs_control *c,
 {
 	struct jffs_file *f;
 
-	if (!(f = (struct jffs_file *)kmalloc(sizeof(struct jffs_file),
-					      GFP_KERNEL))) {
+	if (!(f = kzalloc(sizeof(*f), GFP_KERNEL))) {
 		D(printk("jffs_create_file(): Failed!\n"));
 		return NULL;
 	}
 	no_jffs_file++;
-	memset(f, 0, sizeof(struct jffs_file));
 	f->ino = raw_inode->ino;
 	f->pino = raw_inode->pino;
 	f->nlink = raw_inode->nlink;
@@ -516,7 +514,7 @@ jffs_create_control(struct super_block *sb)
 
 	D2(printk("jffs_create_control()\n"));
 
-	if (!(c = (struct jffs_control *)kmalloc(s, GFP_KERNEL))) {
+	if (!(c = kmalloc(s, GFP_KERNEL))) {
 		goto fail_control;
 	}
 	DJM(no_jffs_control++);
@@ -524,7 +522,7 @@ jffs_create_control(struct super_block *sb)
 	c->gc_task = NULL;
 	c->hash_len = JFFS_HASH_SIZE;
 	s = sizeof(struct list_head) * c->hash_len;
-	if (!(c->hash = (struct list_head *)kmalloc(s, GFP_KERNEL))) {
+	if (!(c->hash = kmalloc(s, GFP_KERNEL))) {
 		goto fail_hash;
 	}
 	DJM(no_hash++);
@@ -593,8 +591,7 @@ jffs_add_virtual_root(struct jffs_control *c)
 	D2(printk("jffs_add_virtual_root(): "
 		  "Creating a virtual root directory.\n"));
 
-	if (!(root = (struct jffs_file *)kmalloc(sizeof(struct jffs_file),
-						 GFP_KERNEL))) {
+	if (!(root = kmalloc(sizeof(struct jffs_file), GFP_KERNEL))) {
 		return -ENOMEM;
 	}
 	no_jffs_file++;
diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c
index 7d8ca1aeace2..29b68d939bd9 100644
--- a/fs/jffs/jffs_fm.c
+++ b/fs/jffs/jffs_fm.c
@@ -94,8 +94,7 @@ jffs_build_begin(struct jffs_control *c, int unit)
 	struct mtd_info *mtd;
 	
 	D3(printk("jffs_build_begin()\n"));
-	fmc = (struct jffs_fmcontrol *)kmalloc(sizeof(struct jffs_fmcontrol),
-					       GFP_KERNEL);
+	fmc = kmalloc(sizeof(*fmc), GFP_KERNEL);
 	if (!fmc) {
 		D(printk("jffs_build_begin(): Allocation of "
 			 "struct jffs_fmcontrol failed!\n"));
@@ -486,8 +485,7 @@ jffs_add_node(struct jffs_node *node)
 
 	D3(printk("jffs_add_node(): ino = %u\n", node->ino));
 
-	ref = (struct jffs_node_ref *)kmalloc(sizeof(struct jffs_node_ref),
-					      GFP_KERNEL);
+	ref = kmalloc(sizeof(*ref), GFP_KERNEL);
 	if (!ref)
 		return -ENOMEM;
 
diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c
index efbb586bed4b..3856efc399c1 100644
--- a/fs/jfs/jfs_txnmgr.c
+++ b/fs/jfs/jfs_txnmgr.c
@@ -282,7 +282,7 @@ int txInit(void)
 	TxLockVHWM = (nTxLock * 8) / 10;
 
 	size = sizeof(struct tblock) * nTxBlock;
-	TxBlock = (struct tblock *) vmalloc(size);
+	TxBlock = vmalloc(size);
 	if (TxBlock == NULL)
 		return -ENOMEM;
 
@@ -307,7 +307,7 @@ int txInit(void)
 	 * tlock id = 0 is reserved.
 	 */
 	size = sizeof(struct tlock) * nTxLock;
-	TxLock = (struct tlock *) vmalloc(size);
+	TxLock = vmalloc(size);
 	if (TxLock == NULL) {
 		vfree(TxBlock);
 		return -ENOMEM;
diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index 50dbb67ae0c4..271e2165fff6 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -100,7 +100,7 @@ static struct nlm_lockowner *nlm_find_lockowner(struct nlm_host *host, fl_owner_
 	res = __nlm_find_lockowner(host, owner);
 	if (res == NULL) {
 		spin_unlock(&host->h_lock);
-		new = (struct nlm_lockowner *)kmalloc(sizeof(*new), GFP_KERNEL);
+		new = kmalloc(sizeof(*new), GFP_KERNEL);
 		spin_lock(&host->h_lock);
 		res = __nlm_find_lockowner(host, owner);
 		if (res == NULL && new != NULL) {
diff --git a/fs/ncpfs/symlink.c b/fs/ncpfs/symlink.c
index ca92c2406635..e3d26c1bd105 100644
--- a/fs/ncpfs/symlink.c
+++ b/fs/ncpfs/symlink.c
@@ -48,7 +48,7 @@ static int ncp_symlink_readpage(struct file *file, struct page *page)
 	char *buf = kmap(page);
 
 	error = -ENOMEM;
-	rawlink=(char *)kmalloc(NCP_MAX_SYMLINK_SIZE, GFP_KERNEL);
+	rawlink = kmalloc(NCP_MAX_SYMLINK_SIZE, GFP_KERNEL);
 	if (!rawlink)
 		goto fail;
 
@@ -126,7 +126,7 @@ int ncp_symlink(struct inode *dir, struct dentry *dentry, const char *symname) {
 	/* EPERM is returned by VFS if symlink procedure does not exist */
 		return -EPERM;
   
-	rawlink=(char *)kmalloc(NCP_MAX_SYMLINK_SIZE, GFP_KERNEL);
+	rawlink = kmalloc(NCP_MAX_SYMLINK_SIZE, GFP_KERNEL);
 	if (!rawlink)
 		return -ENOMEM;
 
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 57133678db16..841c99a9b11c 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -20,11 +20,6 @@
 #include "delegation.h"
 #include "internal.h"
 
-static struct nfs_delegation *nfs_alloc_delegation(void)
-{
-	return (struct nfs_delegation *)kmalloc(sizeof(struct nfs_delegation), GFP_KERNEL);
-}
-
 static void nfs_free_delegation(struct nfs_delegation *delegation)
 {
 	if (delegation->cred)
@@ -124,7 +119,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct
 	if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR)))
 		__nfs_revalidate_inode(NFS_SERVER(inode), inode);
 
-	delegation = nfs_alloc_delegation();
+	delegation = kmalloc(sizeof(*delegation), GFP_KERNEL);
 	if (delegation == NULL)
 		return -ENOMEM;
 	memcpy(delegation->stateid.data, res->delegation.data,
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index e8c143d182c4..a9d0f71eb5f7 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -443,7 +443,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, str
 {
 	struct nfs_open_context *ctx;
 
-	ctx = (struct nfs_open_context *)kmalloc(sizeof(*ctx), GFP_KERNEL);
+	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
 	if (ctx != NULL) {
 		atomic_set(&ctx->count, 1);
 		ctx->dentry = dget(dentry);
diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c
index f8688eaa0001..3b234d4601e7 100644
--- a/fs/nfs/nfs3proc.c
+++ b/fs/nfs/nfs3proc.c
@@ -449,7 +449,7 @@ nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr
 		struct nfs_fattr res;
 	} *ptr;
 
-	ptr = (struct unlinkxdr *)kmalloc(sizeof(*ptr), GFP_KERNEL);
+	ptr = kmalloc(sizeof(*ptr), GFP_KERNEL);
 	if (!ptr)
 		return -ENOMEM;
 	ptr->arg.fh = NFS_FH(dir->d_inode);
diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c
index 630e50647bbb..4529cc4f3f8f 100644
--- a/fs/nfs/proc.c
+++ b/fs/nfs/proc.c
@@ -352,7 +352,7 @@ nfs_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr *
 {
 	struct nfs_diropargs	*arg;
 
-	arg = (struct nfs_diropargs *)kmalloc(sizeof(*arg), GFP_KERNEL);
+	arg = kmalloc(sizeof(*arg), GFP_KERNEL);
 	if (!arg)
 		return -ENOMEM;
 	arg->fh = NFS_FH(dir->d_inode);
diff --git a/fs/ntfs/dir.c b/fs/ntfs/dir.c
index d1e2c6f9f05e..85c36b8ca452 100644
--- a/fs/ntfs/dir.c
+++ b/fs/ntfs/dir.c
@@ -1149,8 +1149,7 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	 * Allocate a buffer to store the current name being processed
 	 * converted to format determined by current NLS.
 	 */
-	name = (u8*)kmalloc(NTFS_MAX_NAME_LEN * NLS_MAX_CHARSET_SIZE + 1,
-			GFP_NOFS);
+	name = kmalloc(NTFS_MAX_NAME_LEN * NLS_MAX_CHARSET_SIZE + 1, GFP_NOFS);
 	if (unlikely(!name)) {
 		err = -ENOMEM;
 		goto err_out;
@@ -1191,7 +1190,7 @@ static int ntfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	 * map the mft record without deadlocking.
 	 */
 	rc = le32_to_cpu(ctx->attr->data.resident.value_length);
-	ir = (INDEX_ROOT*)kmalloc(rc, GFP_NOFS);
+	ir = kmalloc(rc, GFP_NOFS);
 	if (unlikely(!ir)) {
 		err = -ENOMEM;
 		goto err_out;
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index d313f356e66a..31852121b3f5 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -137,7 +137,7 @@ static int ntfs_init_locked_inode(struct inode *vi, ntfs_attr *na)
 
 		BUG_ON(!na->name);
 		i = na->name_len * sizeof(ntfschar);
-		ni->name = (ntfschar*)kmalloc(i + sizeof(ntfschar), GFP_ATOMIC);
+		ni->name = kmalloc(i + sizeof(ntfschar), GFP_ATOMIC);
 		if (!ni->name)
 			return -ENOMEM;
 		memcpy(ni->name, na->name, i);
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 2438c00ec0ce..578fb3d5e803 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -331,7 +331,7 @@ map_err_out:
 		ntfs_inode **tmp;
 		int new_size = (base_ni->nr_extents + 4) * sizeof(ntfs_inode *);
 
-		tmp = (ntfs_inode **)kmalloc(new_size, GFP_NOFS);
+		tmp = kmalloc(new_size, GFP_NOFS);
 		if (unlikely(!tmp)) {
 			ntfs_error(base_ni->vol->sb, "Failed to allocate "
 					"internal buffer.");
@@ -2893,7 +2893,7 @@ rollback:
 	if (!(base_ni->nr_extents & 3)) {
 		int new_size = (base_ni->nr_extents + 4) * sizeof(ntfs_inode*);
 
-		extent_nis = (ntfs_inode**)kmalloc(new_size, GFP_NOFS);
+		extent_nis = kmalloc(new_size, GFP_NOFS);
 		if (unlikely(!extent_nis)) {
 			ntfs_error(vol->sb, "Failed to allocate internal "
 					"buffer during rollback.%s", es);
diff --git a/fs/ntfs/unistr.c b/fs/ntfs/unistr.c
index b123c0fa6bf6..a1b572196fe4 100644
--- a/fs/ntfs/unistr.c
+++ b/fs/ntfs/unistr.c
@@ -350,7 +350,7 @@ int ntfs_ucstonls(const ntfs_volume *vol, const ntfschar *ins,
 		}
 		if (!ns) {
 			ns_len = ins_len * NLS_MAX_CHARSET_SIZE;
-			ns = (unsigned char*)kmalloc(ns_len + 1, GFP_NOFS);
+			ns = kmalloc(ns_len + 1, GFP_NOFS);
 			if (!ns)
 				goto mem_err_out;
 		}
@@ -365,7 +365,7 @@ retry:			wc = nls->uni2char(le16_to_cpu(ins[i]), ns + o,
 			else if (wc == -ENAMETOOLONG && ns != *outs) {
 				unsigned char *tc;
 				/* Grow in multiples of 64 bytes. */
-				tc = (unsigned char*)kmalloc((ns_len + 64) &
+				tc = kmalloc((ns_len + 64) &
 						~63, GFP_NOFS);
 				if (tc) {
 					memcpy(tc, ns, ns_len);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 3d4f6dff2113..41cfcba7ce49 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -370,7 +370,7 @@ xfs_file_readdir(
 
 	/* Try fairly hard to get memory */
 	do {
-		if ((read_buf = (caddr_t)kmalloc(rlen, GFP_KERNEL)))
+		if ((read_buf = kmalloc(rlen, GFP_KERNEL)))
 			break;
 		rlen >>= 1;
 	} while (rlen >= 1024);
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index d9180020de63..22e3b714f629 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -553,13 +553,13 @@ xfs_vn_follow_link(
 	ASSERT(dentry);
 	ASSERT(nd);
 
-	link = (char *)kmalloc(MAXPATHLEN+1, GFP_KERNEL);
+	link = kmalloc(MAXPATHLEN+1, GFP_KERNEL);
 	if (!link) {
 		nd_set_link(nd, ERR_PTR(-ENOMEM));
 		return NULL;
 	}
 
-	uio = (uio_t *)kmalloc(sizeof(uio_t), GFP_KERNEL);
+	uio = kmalloc(sizeof(uio_t), GFP_KERNEL);
 	if (!uio) {
 		kfree(link);
 		nd_set_link(nd, ERR_PTR(-ENOMEM));
-- 
cgit v1.2.1


From 1a1d92c10dd24bbdc28b3d6e2d03ec199dd3a65b Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 01:49:40 -0700
Subject: [PATCH] Really ignore kmem_cache_destroy return value

* Rougly half of callers already do it by not checking return value
* Code in drivers/acpi/osl.c does the following to be sure:

	(void)kmem_cache_destroy(cache);

* Those who check it printk something, however, slab_error already printed
  the name of failed cache.
* XFS BUGs on failed kmem_cache_destroy which is not the decision
  low-level filesystem driver should make. Converted to ignore.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/acpi/osl.c            |  2 +-
 drivers/infiniband/core/mad.c |  5 +----
 drivers/usb/host/uhci-hcd.c   |  8 ++------
 fs/adfs/super.c               |  3 +--
 fs/affs/super.c               |  3 +--
 fs/befs/linuxvfs.c            |  4 +---
 fs/bfs/inode.c                |  3 +--
 fs/cifs/cifsfs.c              | 20 +++++---------------
 fs/coda/inode.c               |  3 +--
 fs/efs/super.c                |  3 +--
 fs/ext2/super.c               |  3 +--
 fs/ext3/super.c               |  3 +--
 fs/fat/cache.c                |  3 +--
 fs/fat/inode.c                |  3 +--
 fs/hfs/super.c                |  3 +--
 fs/hfsplus/super.c            |  3 +--
 fs/hpfs/super.c               |  3 +--
 fs/isofs/inode.c              |  4 +---
 fs/minix/inode.c              |  3 +--
 fs/ncpfs/inode.c              |  3 +--
 fs/nfs/direct.c               |  3 +--
 fs/nfs/inode.c                |  3 +--
 fs/nfs/pagelist.c             |  3 +--
 fs/nfs/read.c                 |  3 +--
 fs/nfs/write.c                |  3 +--
 fs/nfsd/nfs4state.c           |  5 +----
 fs/ntfs/super.c               | 28 +++++-----------------------
 fs/ocfs2/dlm/dlmfs.c          |  4 +---
 fs/qnx4/inode.c               |  4 +---
 fs/reiserfs/super.c           |  4 +---
 fs/romfs/inode.c              |  3 +--
 fs/smbfs/inode.c              |  3 +--
 fs/smbfs/request.c            |  3 +--
 fs/udf/super.c                |  3 +--
 fs/ufs/super.c                |  3 +--
 fs/xfs/linux-2.6/kmem.h       |  4 ++--
 ipc/mqueue.c                  |  5 +----
 mm/shmem.c                    |  3 +--
 net/sunrpc/rpc_pipe.c         |  3 +--
 net/sunrpc/sched.c            |  8 ++++----
 40 files changed, 53 insertions(+), 130 deletions(-)

diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c
index 507f051d1cef..20beea778ea2 100644
--- a/drivers/acpi/osl.c
+++ b/drivers/acpi/osl.c
@@ -1079,7 +1079,7 @@ acpi_status acpi_os_purge_cache(acpi_cache_t * cache)
 
 acpi_status acpi_os_delete_cache(acpi_cache_t * cache)
 {
-	(void)kmem_cache_destroy(cache);
+	kmem_cache_destroy(cache);
 	return (AE_OK);
 }
 
diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index 082f03c158f0..493f4c65c7a2 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -2987,10 +2987,7 @@ error1:
 static void __exit ib_mad_cleanup_module(void)
 {
 	ib_unregister_client(&mad_client);
-
-	if (kmem_cache_destroy(ib_mad_cache)) {
-		printk(KERN_DEBUG PFX "Failed to destroy ib_mad cache\n");
-	}
+	kmem_cache_destroy(ib_mad_cache);
 }
 
 module_init(ib_mad_init_module);
diff --git a/drivers/usb/host/uhci-hcd.c b/drivers/usb/host/uhci-hcd.c
index b7402ceb3e93..eb4eab98e8bf 100644
--- a/drivers/usb/host/uhci-hcd.c
+++ b/drivers/usb/host/uhci-hcd.c
@@ -913,8 +913,7 @@ static int __init uhci_hcd_init(void)
 	return 0;
 
 init_failed:
-	if (kmem_cache_destroy(uhci_up_cachep))
-		warn("not all urb_privs were freed!");
+	kmem_cache_destroy(uhci_up_cachep);
 
 up_failed:
 	debugfs_remove(uhci_debugfs_root);
@@ -930,10 +929,7 @@ errbuf_failed:
 static void __exit uhci_hcd_cleanup(void) 
 {
 	pci_unregister_driver(&uhci_pci_driver);
-	
-	if (kmem_cache_destroy(uhci_up_cachep))
-		warn("not all urb_privs were freed!");
-
+	kmem_cache_destroy(uhci_up_cachep);
 	debugfs_remove(uhci_debugfs_root);
 	kfree(errbuf);
 }
diff --git a/fs/adfs/super.c b/fs/adfs/super.c
index 959dbf1f8e63..9ade139086fc 100644
--- a/fs/adfs/super.c
+++ b/fs/adfs/super.c
@@ -251,8 +251,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(adfs_inode_cachep))
-		printk(KERN_INFO "adfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(adfs_inode_cachep);
 }
 
 static struct super_operations adfs_sops = {
diff --git a/fs/affs/super.c b/fs/affs/super.c
index b35f514479bf..5ea72c3a16c3 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -109,8 +109,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(affs_inode_cachep))
-		printk(KERN_INFO "affs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(affs_inode_cachep);
 }
 
 static struct super_operations affs_sops = {
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index 50cfca5c7efd..f6676fbe9484 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -446,9 +446,7 @@ befs_init_inodecache(void)
 static void
 befs_destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(befs_inode_cachep))
-		printk(KERN_ERR "befs_destroy_inodecache: "
-		       "not all structures were freed\n");
+	kmem_cache_destroy(befs_inode_cachep);
 }
 
 /*
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 3e4d6c767b9b..8fc2e8e49dbe 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -268,8 +268,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(bfs_inode_cachep))
-		printk(KERN_INFO "bfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(bfs_inode_cachep);
 }
 
 static struct super_operations bfs_sops = {
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index c3ef1c0d0e68..4197a5043f13 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -699,8 +699,7 @@ cifs_init_inodecache(void)
 static void
 cifs_destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(cifs_inode_cachep))
-		printk(KERN_WARNING "cifs_inode_cache: error freeing\n");
+	kmem_cache_destroy(cifs_inode_cachep);
 }
 
 static int
@@ -778,13 +777,9 @@ static void
 cifs_destroy_request_bufs(void)
 {
 	mempool_destroy(cifs_req_poolp);
-	if (kmem_cache_destroy(cifs_req_cachep))
-		printk(KERN_WARNING
-		       "cifs_destroy_request_cache: error not all structures were freed\n");
+	kmem_cache_destroy(cifs_req_cachep);
 	mempool_destroy(cifs_sm_req_poolp);
-	if (kmem_cache_destroy(cifs_sm_req_cachep))
-		printk(KERN_WARNING
-		      "cifs_destroy_request_cache: cifs_small_rq free error\n");
+	kmem_cache_destroy(cifs_sm_req_cachep);
 }
 
 static int
@@ -819,13 +814,8 @@ static void
 cifs_destroy_mids(void)
 {
 	mempool_destroy(cifs_mid_poolp);
-	if (kmem_cache_destroy(cifs_mid_cachep))
-		printk(KERN_WARNING
-		       "cifs_destroy_mids: error not all structures were freed\n");
-
-	if (kmem_cache_destroy(cifs_oplock_cachep))
-		printk(KERN_WARNING
-		       "error not all oplock structures were freed\n");
+	kmem_cache_destroy(cifs_mid_cachep);
+	kmem_cache_destroy(cifs_oplock_cachep);
 }
 
 static int cifs_oplock_thread(void * dummyarg)
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index 87f1dc8aa24b..88d123321164 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -80,8 +80,7 @@ int coda_init_inodecache(void)
 
 void coda_destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(coda_inode_cachep))
-		printk(KERN_INFO "coda_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(coda_inode_cachep);
 }
 
 static int coda_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/efs/super.c b/fs/efs/super.c
index 7089269ee9ae..b3f50651eb6b 100644
--- a/fs/efs/super.c
+++ b/fs/efs/super.c
@@ -90,8 +90,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(efs_inode_cachep))
-		printk(KERN_INFO "efs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(efs_inode_cachep);
 }
 
 static void efs_put_super(struct super_block *s)
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 69435fbe53f2..513cd421ac0b 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -184,8 +184,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(ext2_inode_cachep))
-		printk(KERN_INFO "ext2_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(ext2_inode_cachep);
 }
 
 static void ext2_clear_inode(struct inode *inode)
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 09b4b313ca07..8bfd56ef18ca 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -491,8 +491,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(ext3_inode_cachep))
-		printk(KERN_INFO "ext3_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(ext3_inode_cachep);
 }
 
 static void ext3_clear_inode(struct inode *inode)
diff --git a/fs/fat/cache.c b/fs/fat/cache.c
index 97b967b84fc6..82cc4f59e3ba 100644
--- a/fs/fat/cache.c
+++ b/fs/fat/cache.c
@@ -58,8 +58,7 @@ int __init fat_cache_init(void)
 
 void fat_cache_destroy(void)
 {
-	if (kmem_cache_destroy(fat_cache_cachep))
-		printk(KERN_INFO "fat_cache: not all structures were freed\n");
+	kmem_cache_destroy(fat_cache_cachep);
 }
 
 static inline struct fat_cache *fat_cache_alloc(struct inode *inode)
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index bc4da3a48c8c..d30151190434 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -528,8 +528,7 @@ static int __init fat_init_inodecache(void)
 
 static void __exit fat_destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(fat_inode_cachep))
-		printk(KERN_INFO "fat_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(fat_inode_cachep);
 }
 
 static int fat_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index 05dffff1a372..d43b4fcc8ad3 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -454,8 +454,7 @@ static int __init init_hfs_fs(void)
 static void __exit exit_hfs_fs(void)
 {
 	unregister_filesystem(&hfs_fs_type);
-	if (kmem_cache_destroy(hfs_inode_cachep))
-		printk(KERN_ERR "hfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(hfs_inode_cachep);
 }
 
 module_init(init_hfs_fs)
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index d279d5924f28..194eede52fa4 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -493,8 +493,7 @@ static int __init init_hfsplus_fs(void)
 static void __exit exit_hfsplus_fs(void)
 {
 	unregister_filesystem(&hfsplus_fs_type);
-	if (kmem_cache_destroy(hfsplus_inode_cachep))
-		printk(KERN_ERR "hfsplus_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(hfsplus_inode_cachep);
 }
 
 module_init(init_hfsplus_fs)
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index b76d60832375..450b5e0b4785 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -203,8 +203,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(hpfs_inode_cachep))
-		printk(KERN_INFO "hpfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(hpfs_inode_cachep);
 }
 
 /*
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index ab15e9072151..10e47897bac7 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -96,9 +96,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(isofs_inode_cachep))
-		printk(KERN_INFO "iso_inode_cache: not all structures were "
-					"freed\n");
+	kmem_cache_destroy(isofs_inode_cachep);
 }
 
 static int isofs_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index e24be7fc0651..826b9d830650 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -90,8 +90,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(minix_inode_cachep))
-		printk(KERN_INFO "minix_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(minix_inode_cachep);
 }
 
 static struct super_operations minix_sops = {
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 752f02e43ab0..8244710e97dd 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -81,8 +81,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(ncp_inode_cachep))
-		printk(KERN_INFO "ncp_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(ncp_inode_cachep);
 }
 
 static int ncp_remount(struct super_block *sb, int *flags, char* data)
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 76ca1cbc38f9..377839bed172 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -855,6 +855,5 @@ int __init nfs_init_directcache(void)
  */
 void nfs_destroy_directcache(void)
 {
-	if (kmem_cache_destroy(nfs_direct_cachep))
-		printk(KERN_INFO "nfs_direct_cache: not all structures were freed\n");
+	kmem_cache_destroy(nfs_direct_cachep);
 }
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index a9d0f71eb5f7..931f52a19579 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1134,8 +1134,7 @@ static int __init nfs_init_inodecache(void)
 
 static void nfs_destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(nfs_inode_cachep))
-		printk(KERN_INFO "nfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(nfs_inode_cachep);
 }
 
 /*
diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c
index 36e902a88ca1..829af323f288 100644
--- a/fs/nfs/pagelist.c
+++ b/fs/nfs/pagelist.c
@@ -392,7 +392,6 @@ int __init nfs_init_nfspagecache(void)
 
 void nfs_destroy_nfspagecache(void)
 {
-	if (kmem_cache_destroy(nfs_page_cachep))
-		printk(KERN_INFO "nfs_page: not all structures were freed\n");
+	kmem_cache_destroy(nfs_page_cachep);
 }
 
diff --git a/fs/nfs/read.c b/fs/nfs/read.c
index 69f1549da2b9..c2e49c397a27 100644
--- a/fs/nfs/read.c
+++ b/fs/nfs/read.c
@@ -737,6 +737,5 @@ int __init nfs_init_readpagecache(void)
 void nfs_destroy_readpagecache(void)
 {
 	mempool_destroy(nfs_rdata_mempool);
-	if (kmem_cache_destroy(nfs_rdata_cachep))
-		printk(KERN_INFO "nfs_read_data: not all structures were freed\n");
+	kmem_cache_destroy(nfs_rdata_cachep);
 }
diff --git a/fs/nfs/write.c b/fs/nfs/write.c
index c12effb46fe5..b674462793d3 100644
--- a/fs/nfs/write.c
+++ b/fs/nfs/write.c
@@ -1565,7 +1565,6 @@ void nfs_destroy_writepagecache(void)
 {
 	mempool_destroy(nfs_commit_mempool);
 	mempool_destroy(nfs_wdata_mempool);
-	if (kmem_cache_destroy(nfs_wdata_cachep))
-		printk(KERN_INFO "nfs_write_data: not all structures were freed\n");
+	kmem_cache_destroy(nfs_wdata_cachep);
 }
 
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 1a25c01ecdf1..ebcf226a9e4a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -1005,13 +1005,10 @@ alloc_init_file(struct inode *ino)
 static void
 nfsd4_free_slab(kmem_cache_t **slab)
 {
-	int status;
-
 	if (*slab == NULL)
 		return;
-	status = kmem_cache_destroy(*slab);
+	kmem_cache_destroy(*slab);
 	*slab = NULL;
-	WARN_ON(status);
 }
 
 static void
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index 74e0ee8fce72..6b2712f10dd2 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -3248,32 +3248,14 @@ ictx_err_out:
 
 static void __exit exit_ntfs_fs(void)
 {
-	int err = 0;
-
 	ntfs_debug("Unregistering NTFS driver.");
 
 	unregister_filesystem(&ntfs_fs_type);
-
-	if (kmem_cache_destroy(ntfs_big_inode_cache) && (err = 1))
-		printk(KERN_CRIT "NTFS: Failed to destory %s.\n",
-				ntfs_big_inode_cache_name);
-	if (kmem_cache_destroy(ntfs_inode_cache) && (err = 1))
-		printk(KERN_CRIT "NTFS: Failed to destory %s.\n",
-				ntfs_inode_cache_name);
-	if (kmem_cache_destroy(ntfs_name_cache) && (err = 1))
-		printk(KERN_CRIT "NTFS: Failed to destory %s.\n",
-				ntfs_name_cache_name);
-	if (kmem_cache_destroy(ntfs_attr_ctx_cache) && (err = 1))
-		printk(KERN_CRIT "NTFS: Failed to destory %s.\n",
-				ntfs_attr_ctx_cache_name);
-	if (kmem_cache_destroy(ntfs_index_ctx_cache) && (err = 1))
-		printk(KERN_CRIT "NTFS: Failed to destory %s.\n",
-				ntfs_index_ctx_cache_name);
-	if (err)
-		printk(KERN_CRIT "NTFS: This causes memory to leak! There is "
-				"probably a BUG in the driver! Please report "
-				"you saw this message to "
-				"linux-ntfs-dev@lists.sourceforge.net\n");
+	kmem_cache_destroy(ntfs_big_inode_cache);
+	kmem_cache_destroy(ntfs_inode_cache);
+	kmem_cache_destroy(ntfs_name_cache);
+	kmem_cache_destroy(ntfs_attr_ctx_cache);
+	kmem_cache_destroy(ntfs_index_ctx_cache);
 	/* Unregister the ntfs sysctls. */
 	ntfs_sysctl(0);
 }
diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c
index 033ad1701232..0ff0898a0b9c 100644
--- a/fs/ocfs2/dlm/dlmfs.c
+++ b/fs/ocfs2/dlm/dlmfs.c
@@ -629,9 +629,7 @@ static void __exit exit_dlmfs_fs(void)
 	flush_workqueue(user_dlm_worker);
 	destroy_workqueue(user_dlm_worker);
 
-	if (kmem_cache_destroy(dlmfs_inode_cache))
-		printk(KERN_INFO "dlmfs_inode_cache: not all structures "
-		       "were freed\n");
+	kmem_cache_destroy(dlmfs_inode_cache);
 }
 
 MODULE_AUTHOR("Oracle");
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index 8497609f5022..fddbd61c68d0 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -556,9 +556,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(qnx4_inode_cachep))
-		printk(KERN_INFO
-		       "qnx4_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(qnx4_inode_cachep);
 }
 
 static int qnx4_get_sb(struct file_system_type *fs_type,
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 5567328f1041..1cd4d387f690 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -530,9 +530,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(reiserfs_inode_cachep))
-		reiserfs_warning(NULL,
-				 "reiserfs_inode_cache: not all structures were freed");
+	kmem_cache_destroy(reiserfs_inode_cachep);
 }
 
 /* we don't mark inodes dirty, we just log them */
diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c
index 22eed61ebf69..ddcd9e1ef282 100644
--- a/fs/romfs/inode.c
+++ b/fs/romfs/inode.c
@@ -589,8 +589,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(romfs_inode_cachep))
-		printk(KERN_INFO "romfs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(romfs_inode_cachep);
 }
 
 static int romfs_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c
index a1ed657c3c84..92cf60aa6121 100644
--- a/fs/smbfs/inode.c
+++ b/fs/smbfs/inode.c
@@ -89,8 +89,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(smb_inode_cachep))
-		printk(KERN_INFO "smb_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(smb_inode_cachep);
 }
 
 static int smb_remount(struct super_block *sb, int *flags, char *data)
diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c
index c8e96195b96e..0fb74697abc4 100644
--- a/fs/smbfs/request.c
+++ b/fs/smbfs/request.c
@@ -49,8 +49,7 @@ int smb_init_request_cache(void)
 
 void smb_destroy_request_cache(void)
 {
-	if (kmem_cache_destroy(req_cachep))
-		printk(KERN_INFO "smb_destroy_request_cache: not all structures were freed\n");
+	kmem_cache_destroy(req_cachep);
 }
 
 /*
diff --git a/fs/udf/super.c b/fs/udf/super.c
index fcce1a21a51b..5dd356cbbda6 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -156,8 +156,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(udf_inode_cachep))
-		printk(KERN_INFO "udf_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(udf_inode_cachep);
 }
 
 /* Superblock operations */
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index ef910e784034..ec79e3091d1b 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -1244,8 +1244,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(ufs_inode_cachep))
-		printk(KERN_INFO "ufs_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(ufs_inode_cachep);
 }
 
 #ifdef CONFIG_QUOTA
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index 939bd84bc7ee..0e8293c5a32f 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -91,8 +91,8 @@ kmem_zone_free(kmem_zone_t *zone, void *ptr)
 static inline void
 kmem_zone_destroy(kmem_zone_t *zone)
 {
-	if (zone && kmem_cache_destroy(zone))
-		BUG();
+	if (zone)
+		kmem_cache_destroy(zone);
 }
 
 extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast);
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 02e6f6798972..f8381f0660b2 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1275,10 +1275,7 @@ out_filesystem:
 out_sysctl:
 	if (mq_sysctl_table)
 		unregister_sysctl_table(mq_sysctl_table);
-	if (kmem_cache_destroy(mqueue_inode_cachep)) {
-		printk(KERN_INFO
-			"mqueue_inode_cache: not all structures were freed\n");
-	}
+	kmem_cache_destroy(mqueue_inode_cachep);
 	return error;
 }
 
diff --git a/mm/shmem.c b/mm/shmem.c
index 8631be45b40d..0a8e29cf87e0 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2157,8 +2157,7 @@ static int init_inodecache(void)
 
 static void destroy_inodecache(void)
 {
-	if (kmem_cache_destroy(shmem_inode_cachep))
-		printk(KERN_INFO "shmem_inode_cache: not all structures were freed\n");
+	kmem_cache_destroy(shmem_inode_cachep);
 }
 
 static const struct address_space_operations shmem_aops = {
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index dfa504fe383f..f65bea74734d 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -858,7 +858,6 @@ int register_rpc_pipefs(void)
 
 void unregister_rpc_pipefs(void)
 {
-	if (kmem_cache_destroy(rpc_inode_cachep))
-		printk(KERN_WARNING "RPC: unable to free inode cache\n");
+	kmem_cache_destroy(rpc_inode_cachep);
 	unregister_filesystem(&rpc_pipe_fs_type);
 }
diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c
index 6390461a9756..a1ab4eed41f4 100644
--- a/net/sunrpc/sched.c
+++ b/net/sunrpc/sched.c
@@ -1059,10 +1059,10 @@ rpc_destroy_mempool(void)
 		mempool_destroy(rpc_buffer_mempool);
 	if (rpc_task_mempool)
 		mempool_destroy(rpc_task_mempool);
-	if (rpc_task_slabp && kmem_cache_destroy(rpc_task_slabp))
-		printk(KERN_INFO "rpc_task: not all structures were freed\n");
-	if (rpc_buffer_slabp && kmem_cache_destroy(rpc_buffer_slabp))
-		printk(KERN_INFO "rpc_buffers: not all structures were freed\n");
+	if (rpc_task_slabp)
+		kmem_cache_destroy(rpc_task_slabp);
+	if (rpc_buffer_slabp)
+		kmem_cache_destroy(rpc_buffer_slabp);
 }
 
 int
-- 
cgit v1.2.1


From 133d205a18b7a4d8cb52959c5310f6664277cf61 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Wed, 27 Sep 2006 01:49:41 -0700
Subject: [PATCH] Make kmem_cache_destroy() return void

un-, de-, -free, -destroy, -exit, etc functions should in general return
void.  Also,

There is very little, say, filesystem driver code can do upon failed
kmem_cache_destroy().  If it will be decided to BUG in this case, BUG
should be put in generic code, instead.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/slab.h | 4 ++--
 mm/slab.c            | 6 ++----
 mm/slob.c            | 3 +--
 3 files changed, 5 insertions(+), 8 deletions(-)

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 66d6eb78d1c6..a96fd9310d55 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -60,7 +60,7 @@ extern void __init kmem_cache_init(void);
 extern kmem_cache_t *kmem_cache_create(const char *, size_t, size_t, unsigned long,
 				       void (*)(void *, kmem_cache_t *, unsigned long),
 				       void (*)(void *, kmem_cache_t *, unsigned long));
-extern int kmem_cache_destroy(kmem_cache_t *);
+extern void kmem_cache_destroy(kmem_cache_t *);
 extern int kmem_cache_shrink(kmem_cache_t *);
 extern void *kmem_cache_alloc(kmem_cache_t *, gfp_t);
 extern void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
@@ -249,7 +249,7 @@ struct kmem_cache *kmem_cache_create(const char *c, size_t, size_t,
 	unsigned long,
 	void (*)(void *, struct kmem_cache *, unsigned long),
 	void (*)(void *, struct kmem_cache *, unsigned long));
-int kmem_cache_destroy(struct kmem_cache *c);
+void kmem_cache_destroy(struct kmem_cache *c);
 void *kmem_cache_alloc(struct kmem_cache *c, gfp_t flags);
 void *kmem_cache_zalloc(struct kmem_cache *, gfp_t);
 void kmem_cache_free(struct kmem_cache *c, void *b);
diff --git a/mm/slab.c b/mm/slab.c
index 7a48eb1a60c8..c52ebf9c4462 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2442,7 +2442,6 @@ EXPORT_SYMBOL(kmem_cache_shrink);
  * @cachep: the cache to destroy
  *
  * Remove a struct kmem_cache object from the slab cache.
- * Returns 0 on success.
  *
  * It is expected this function will be called by a module when it is
  * unloaded.  This will remove the cache completely, and avoid a duplicate
@@ -2454,7 +2453,7 @@ EXPORT_SYMBOL(kmem_cache_shrink);
  * The caller must guarantee that noone will allocate memory from the cache
  * during the kmem_cache_destroy().
  */
-int kmem_cache_destroy(struct kmem_cache *cachep)
+void kmem_cache_destroy(struct kmem_cache *cachep)
 {
 	BUG_ON(!cachep || in_interrupt());
 
@@ -2475,7 +2474,7 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 		list_add(&cachep->next, &cache_chain);
 		mutex_unlock(&cache_chain_mutex);
 		unlock_cpu_hotplug();
-		return 1;
+		return;
 	}
 
 	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU))
@@ -2483,7 +2482,6 @@ int kmem_cache_destroy(struct kmem_cache *cachep)
 
 	__kmem_cache_destroy(cachep);
 	unlock_cpu_hotplug();
-	return 0;
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
diff --git a/mm/slob.c b/mm/slob.c
index 20188627347c..542394184a58 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -270,10 +270,9 @@ struct kmem_cache *kmem_cache_create(const char *name, size_t size,
 }
 EXPORT_SYMBOL(kmem_cache_create);
 
-int kmem_cache_destroy(struct kmem_cache *c)
+void kmem_cache_destroy(struct kmem_cache *c)
 {
 	slob_free(c, sizeof(struct kmem_cache));
-	return 0;
 }
 EXPORT_SYMBOL(kmem_cache_destroy);
 
-- 
cgit v1.2.1


From 2bd0cfbde2c0a74e209acbf045f1298cc2f61e01 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:49:42 -0700
Subject: [PATCH] fix x86_64-mm-spinlock-cleanup

We need processor.h for cpu_relax().

Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-x86_64/spinlock.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/asm-x86_64/spinlock.h b/include/asm-x86_64/spinlock.h
index be7a9e629fb2..3daf5b005905 100644
--- a/include/asm-x86_64/spinlock.h
+++ b/include/asm-x86_64/spinlock.h
@@ -4,6 +4,7 @@
 #include <asm/atomic.h>
 #include <asm/rwlock.h>
 #include <asm/page.h>
+#include <asm/processor.h>
 
 /*
  * Your basic SMP spinlocks, allowing only a single CPU anywhere
-- 
cgit v1.2.1


From c713216deebd95d2b0ab38fef8bb2361c0180c2d Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:43 -0700
Subject: [PATCH] Introduce mechanism for registering active regions of memory

At a basic level, architectures define structures to record where active
ranges of page frames are located.  Once located, the code to calculate zone
sizes and holes in each architecture is very similar.  Some of this zone and
hole sizing code is difficult to read for no good reason.  This set of patches
eliminates the similar-looking architecture-specific code.

The patches introduce a mechanism where architectures register where the
active ranges of page frames are with add_active_range().  When all areas have
been discovered, free_area_init_nodes() is called to initialise the pgdat and
zones.  The zone sizes and holes are then calculated in an architecture
independent manner.

Patch 1 introduces the mechanism for registering and initialising PFN ranges
Patch 2 changes ppc to use the mechanism - 139 arch-specific LOC removed
Patch 3 changes x86 to use the mechanism - 136 arch-specific LOC removed
Patch 4 changes x86_64 to use the mechanism - 74 arch-specific LOC removed
Patch 5 changes ia64 to use the mechanism - 52 arch-specific LOC removed
Patch 6 accounts for mem_map as a memory hole as the pages are not reclaimable.
	It adjusts the watermarks slightly

Tony Luck has successfully tested for ia64 on Itanium with tiger_defconfig,
gensparse_defconfig and defconfig.  Bob Picco has also tested and debugged on
IA64.  Jack Steiner successfully boot tested on a mammoth SGI IA64-based
machine.  These were on patches against 2.6.17-rc1 and release 3 of these
patches but there have been no ia64-changes since release 3.

There are differences in the zone sizes for x86_64 as the arch-specific code
for x86_64 accounts the kernel image and the starting mem_maps as memory holes
but the architecture-independent code accounts the memory as present.

The big benefit of this set of patches is a sizable reduction of
architecture-specific code, some of which is very hairy.  There should be a
greater reduction when other architectures use the same mechanisms for zone
and hole sizing but I lack the hardware to test on.

Additional credit;
	Dave Hansen for the initial suggestion and comments on early patches
	Andy Whitcroft for reviewing early versions and catching numerous
		errors
	Tony Luck for testing and debugging on IA64
	Bob Picco for fixing bugs related to pfn registration, reviewing a
		number of patch revisions, providing a number of suggestions
		on future direction and testing heavily
	Jack Steiner and Robin Holt for testing on IA64 and clarifying
		issues related to memory holes
	Yasunori for testing on IA64
	Andi Kleen for reviewing and feeding back about x86_64
	Christian Kujau for providing valuable information related to ACPI
		problems on x86_64 and testing potential fixes

This patch:

Define the structure to represent an active range of page frames within a node
in an architecture independent manner.  Architectures are expected to register
active ranges of PFNs using add_active_range(nid, start_pfn, end_pfn) and call
free_area_init_nodes() passing the PFNs of the end of each zone.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     |  47 +++++
 include/linux/mmzone.h |  10 +-
 mm/page_alloc.c        | 552 ++++++++++++++++++++++++++++++++++++++++++++++---
 3 files changed, 584 insertions(+), 25 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 856f0ee7e84a..c0402da7cce0 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -937,6 +937,53 @@ extern void free_area_init(unsigned long * zones_size);
 extern void free_area_init_node(int nid, pg_data_t *pgdat,
 	unsigned long * zones_size, unsigned long zone_start_pfn, 
 	unsigned long *zholes_size);
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/*
+ * With CONFIG_ARCH_POPULATES_NODE_MAP set, an architecture may initialise its
+ * zones, allocate the backing mem_map and account for memory holes in a more
+ * architecture independent manner. This is a substitute for creating the
+ * zone_sizes[] and zholes_size[] arrays and passing them to
+ * free_area_init_node()
+ *
+ * An architecture is expected to register range of page frames backed by
+ * physical memory with add_active_range() before calling
+ * free_area_init_nodes() passing in the PFN each zone ends at. At a basic
+ * usage, an architecture is expected to do something like
+ *
+ * unsigned long max_zone_pfns[MAX_NR_ZONES] = {max_dma, max_normal_pfn,
+ * 							 max_highmem_pfn};
+ * for_each_valid_physical_page_range()
+ * 	add_active_range(node_id, start_pfn, end_pfn)
+ * free_area_init_nodes(max_zone_pfns);
+ *
+ * If the architecture guarantees that there are no holes in the ranges
+ * registered with add_active_range(), free_bootmem_active_regions()
+ * will call free_bootmem_node() for each registered physical page range.
+ * Similarly sparse_memory_present_with_active_regions() calls
+ * memory_present() for each range when SPARSEMEM is enabled.
+ *
+ * See mm/page_alloc.c for more information on each function exposed by
+ * CONFIG_ARCH_POPULATES_NODE_MAP
+ */
+extern void free_area_init_nodes(unsigned long *max_zone_pfn);
+extern void add_active_range(unsigned int nid, unsigned long start_pfn,
+					unsigned long end_pfn);
+extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
+						unsigned long new_end_pfn);
+extern void remove_all_active_ranges(void);
+extern unsigned long absent_pages_in_range(unsigned long start_pfn,
+						unsigned long end_pfn);
+extern void get_pfn_range_for_nid(unsigned int nid,
+			unsigned long *start_pfn, unsigned long *end_pfn);
+extern unsigned long find_min_pfn_with_active_regions(void);
+extern unsigned long find_max_pfn_with_active_regions(void);
+extern void free_bootmem_with_active_regions(int nid,
+						unsigned long max_low_pfn);
+extern void sparse_memory_present_with_active_regions(int nid);
+#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+extern int early_pfn_to_nid(unsigned long pfn);
+#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long);
 extern void setup_per_zone_pages_min(void);
 extern void mem_init(void);
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 3693f1a52788..7fa1cbe9fa7a 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -305,6 +305,13 @@ struct zonelist {
 	struct zone *zones[MAX_NUMNODES * MAX_NR_ZONES + 1]; // NULL delimited
 };
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+struct node_active_region {
+	unsigned long start_pfn;
+	unsigned long end_pfn;
+	int nid;
+};
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
 /*
  * The pg_data_t structure is used in machines with CONFIG_DISCONTIGMEM
@@ -518,7 +525,8 @@ extern struct zone *next_zone(struct zone *zone);
 
 #endif
 
-#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+#if !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID) && \
+	!defined(CONFIG_ARCH_POPULATES_NODE_MAP)
 #define early_pfn_to_nid(nid)  (0UL)
 #endif
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9810f0a60db7..26c9939857fa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -37,6 +37,8 @@
 #include <linux/vmalloc.h>
 #include <linux/mempolicy.h>
 #include <linux/stop_machine.h>
+#include <linux/sort.h>
+#include <linux/pfn.h>
 
 #include <asm/tlbflush.h>
 #include <asm/div64.h>
@@ -103,6 +105,33 @@ int min_free_kbytes = 1024;
 unsigned long __meminitdata nr_kernel_pages;
 unsigned long __meminitdata nr_all_pages;
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+  /*
+   * MAX_ACTIVE_REGIONS determines the maxmimum number of distinct
+   * ranges of memory (RAM) that may be registered with add_active_range().
+   * Ranges passed to add_active_range() will be merged if possible
+   * so the number of times add_active_range() can be called is
+   * related to the number of nodes and the number of holes
+   */
+  #ifdef CONFIG_MAX_ACTIVE_REGIONS
+    /* Allow an architecture to set MAX_ACTIVE_REGIONS to save memory */
+    #define MAX_ACTIVE_REGIONS CONFIG_MAX_ACTIVE_REGIONS
+  #else
+    #if MAX_NUMNODES >= 32
+      /* If there can be many nodes, allow up to 50 holes per node */
+      #define MAX_ACTIVE_REGIONS (MAX_NUMNODES*50)
+    #else
+      /* By default, allow up to 256 distinct regions */
+      #define MAX_ACTIVE_REGIONS 256
+    #endif
+  #endif
+
+  struct node_active_region __initdata early_node_map[MAX_ACTIVE_REGIONS];
+  int __initdata nr_nodemap_entries;
+  unsigned long __initdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES];
+  unsigned long __initdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+
 #ifdef CONFIG_DEBUG_VM
 static int page_outside_zone_boundaries(struct zone *zone, struct page *page)
 {
@@ -1642,25 +1671,6 @@ static inline unsigned long wait_table_bits(unsigned long size)
 
 #define LONG_ALIGN(x) (((x)+(sizeof(long))-1)&~((sizeof(long))-1))
 
-static void __init calculate_zone_totalpages(struct pglist_data *pgdat,
-		unsigned long *zones_size, unsigned long *zholes_size)
-{
-	unsigned long realtotalpages, totalpages = 0;
-	enum zone_type i;
-
-	for (i = 0; i < MAX_NR_ZONES; i++)
-		totalpages += zones_size[i];
-	pgdat->node_spanned_pages = totalpages;
-
-	realtotalpages = totalpages;
-	if (zholes_size)
-		for (i = 0; i < MAX_NR_ZONES; i++)
-			realtotalpages -= zholes_size[i];
-	pgdat->node_present_pages = realtotalpages;
-	printk(KERN_DEBUG "On node %d totalpages: %lu\n", pgdat->node_id, realtotalpages);
-}
-
-
 /*
  * Initially all pages are reserved - free ones are freed
  * up by free_all_bootmem() once the early boot process is
@@ -1977,6 +1987,272 @@ __meminit int init_currently_empty_zone(struct zone *zone,
 	return 0;
 }
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/*
+ * Basic iterator support. Return the first range of PFNs for a node
+ * Note: nid == MAX_NUMNODES returns first region regardless of node
+ */
+static int __init first_active_region_index_in_nid(int nid)
+{
+	int i;
+
+	for (i = 0; i < nr_nodemap_entries; i++)
+		if (nid == MAX_NUMNODES || early_node_map[i].nid == nid)
+			return i;
+
+	return -1;
+}
+
+/*
+ * Basic iterator support. Return the next active range of PFNs for a node
+ * Note: nid == MAX_NUMNODES returns next region regardles of node
+ */
+static int __init next_active_region_index_in_nid(int index, int nid)
+{
+	for (index = index + 1; index < nr_nodemap_entries; index++)
+		if (nid == MAX_NUMNODES || early_node_map[index].nid == nid)
+			return index;
+
+	return -1;
+}
+
+#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID
+/*
+ * Required by SPARSEMEM. Given a PFN, return what node the PFN is on.
+ * Architectures may implement their own version but if add_active_range()
+ * was used and there are no special requirements, this is a convenient
+ * alternative
+ */
+int __init early_pfn_to_nid(unsigned long pfn)
+{
+	int i;
+
+	for (i = 0; i < nr_nodemap_entries; i++) {
+		unsigned long start_pfn = early_node_map[i].start_pfn;
+		unsigned long end_pfn = early_node_map[i].end_pfn;
+
+		if (start_pfn <= pfn && pfn < end_pfn)
+			return early_node_map[i].nid;
+	}
+
+	return 0;
+}
+#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
+
+/* Basic iterator support to walk early_node_map[] */
+#define for_each_active_range_index_in_nid(i, nid) \
+	for (i = first_active_region_index_in_nid(nid); i != -1; \
+				i = next_active_region_index_in_nid(i, nid))
+
+/**
+ * free_bootmem_with_active_regions - Call free_bootmem_node for each active range
+ * @nid: The node to free memory on. If MAX_NUMNODES, all nodes are freed
+ * @max_low_pfn: The highest PFN that till be passed to free_bootmem_node
+ *
+ * If an architecture guarantees that all ranges registered with
+ * add_active_ranges() contain no holes and may be freed, this
+ * this function may be used instead of calling free_bootmem() manually.
+ */
+void __init free_bootmem_with_active_regions(int nid,
+						unsigned long max_low_pfn)
+{
+	int i;
+
+	for_each_active_range_index_in_nid(i, nid) {
+		unsigned long size_pages = 0;
+		unsigned long end_pfn = early_node_map[i].end_pfn;
+
+		if (early_node_map[i].start_pfn >= max_low_pfn)
+			continue;
+
+		if (end_pfn > max_low_pfn)
+			end_pfn = max_low_pfn;
+
+		size_pages = end_pfn - early_node_map[i].start_pfn;
+		free_bootmem_node(NODE_DATA(early_node_map[i].nid),
+				PFN_PHYS(early_node_map[i].start_pfn),
+				size_pages << PAGE_SHIFT);
+	}
+}
+
+/**
+ * sparse_memory_present_with_active_regions - Call memory_present for each active range
+ * @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used
+ *
+ * If an architecture guarantees that all ranges registered with
+ * add_active_ranges() contain no holes and may be freed, this
+ * this function may be used instead of calling memory_present() manually.
+ */
+void __init sparse_memory_present_with_active_regions(int nid)
+{
+	int i;
+
+	for_each_active_range_index_in_nid(i, nid)
+		memory_present(early_node_map[i].nid,
+				early_node_map[i].start_pfn,
+				early_node_map[i].end_pfn);
+}
+
+/**
+ * get_pfn_range_for_nid - Return the start and end page frames for a node
+ * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned
+ * @start_pfn: Passed by reference. On return, it will have the node start_pfn
+ * @end_pfn: Passed by reference. On return, it will have the node end_pfn
+ *
+ * It returns the start and end page frame of a node based on information
+ * provided by an arch calling add_active_range(). If called for a node
+ * with no available memory, a warning is printed and the start and end
+ * PFNs will be 0
+ */
+void __init get_pfn_range_for_nid(unsigned int nid,
+			unsigned long *start_pfn, unsigned long *end_pfn)
+{
+	int i;
+	*start_pfn = -1UL;
+	*end_pfn = 0;
+
+	for_each_active_range_index_in_nid(i, nid) {
+		*start_pfn = min(*start_pfn, early_node_map[i].start_pfn);
+		*end_pfn = max(*end_pfn, early_node_map[i].end_pfn);
+	}
+
+	if (*start_pfn == -1UL) {
+		printk(KERN_WARNING "Node %u active with no memory\n", nid);
+		*start_pfn = 0;
+	}
+}
+
+/*
+ * Return the number of pages a zone spans in a node, including holes
+ * present_pages = zone_spanned_pages_in_node() - zone_absent_pages_in_node()
+ */
+unsigned long __init zone_spanned_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *ignored)
+{
+	unsigned long node_start_pfn, node_end_pfn;
+	unsigned long zone_start_pfn, zone_end_pfn;
+
+	/* Get the start and end of the node and zone */
+	get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn);
+	zone_start_pfn = arch_zone_lowest_possible_pfn[zone_type];
+	zone_end_pfn = arch_zone_highest_possible_pfn[zone_type];
+
+	/* Check that this node has pages within the zone's required range */
+	if (zone_end_pfn < node_start_pfn || zone_start_pfn > node_end_pfn)
+		return 0;
+
+	/* Move the zone boundaries inside the node if necessary */
+	zone_end_pfn = min(zone_end_pfn, node_end_pfn);
+	zone_start_pfn = max(zone_start_pfn, node_start_pfn);
+
+	/* Return the spanned pages */
+	return zone_end_pfn - zone_start_pfn;
+}
+
+/*
+ * Return the number of holes in a range on a node. If nid is MAX_NUMNODES,
+ * then all holes in the requested range will be accounted for
+ */
+unsigned long __init __absent_pages_in_range(int nid,
+				unsigned long range_start_pfn,
+				unsigned long range_end_pfn)
+{
+	int i = 0;
+	unsigned long prev_end_pfn = 0, hole_pages = 0;
+	unsigned long start_pfn;
+
+	/* Find the end_pfn of the first active range of pfns in the node */
+	i = first_active_region_index_in_nid(nid);
+	if (i == -1)
+		return 0;
+
+	prev_end_pfn = early_node_map[i].start_pfn;
+
+	/* Find all holes for the zone within the node */
+	for (; i != -1; i = next_active_region_index_in_nid(i, nid)) {
+
+		/* No need to continue if prev_end_pfn is outside the zone */
+		if (prev_end_pfn >= range_end_pfn)
+			break;
+
+		/* Make sure the end of the zone is not within the hole */
+		start_pfn = min(early_node_map[i].start_pfn, range_end_pfn);
+		prev_end_pfn = max(prev_end_pfn, range_start_pfn);
+
+		/* Update the hole size cound and move on */
+		if (start_pfn > range_start_pfn) {
+			BUG_ON(prev_end_pfn > start_pfn);
+			hole_pages += start_pfn - prev_end_pfn;
+		}
+		prev_end_pfn = early_node_map[i].end_pfn;
+	}
+
+	return hole_pages;
+}
+
+/**
+ * absent_pages_in_range - Return number of page frames in holes within a range
+ * @start_pfn: The start PFN to start searching for holes
+ * @end_pfn: The end PFN to stop searching for holes
+ *
+ * It returns the number of pages frames in memory holes within a range
+ */
+unsigned long __init absent_pages_in_range(unsigned long start_pfn,
+							unsigned long end_pfn)
+{
+	return __absent_pages_in_range(MAX_NUMNODES, start_pfn, end_pfn);
+}
+
+/* Return the number of page frames in holes in a zone on a node */
+unsigned long __init zone_absent_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *ignored)
+{
+	return __absent_pages_in_range(nid,
+				arch_zone_lowest_possible_pfn[zone_type],
+				arch_zone_highest_possible_pfn[zone_type]);
+}
+#else
+static inline unsigned long zone_spanned_pages_in_node(int nid,
+					unsigned long zone_type,
+					unsigned long *zones_size)
+{
+	return zones_size[zone_type];
+}
+
+static inline unsigned long zone_absent_pages_in_node(int nid,
+						unsigned long zone_type,
+						unsigned long *zholes_size)
+{
+	if (!zholes_size)
+		return 0;
+
+	return zholes_size[zone_type];
+}
+#endif
+
+static void __init calculate_node_totalpages(struct pglist_data *pgdat,
+		unsigned long *zones_size, unsigned long *zholes_size)
+{
+	unsigned long realtotalpages, totalpages = 0;
+	enum zone_type i;
+
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		totalpages += zone_spanned_pages_in_node(pgdat->node_id, i,
+								zones_size);
+	pgdat->node_spanned_pages = totalpages;
+
+	realtotalpages = totalpages;
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		realtotalpages -=
+			zone_absent_pages_in_node(pgdat->node_id, i,
+								zholes_size);
+	pgdat->node_present_pages = realtotalpages;
+	printk(KERN_DEBUG "On node %d totalpages: %lu\n", pgdat->node_id,
+							realtotalpages);
+}
+
 /*
  * Set up the zone data structures:
  *   - mark all pages reserved
@@ -2000,9 +2276,9 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		struct zone *zone = pgdat->node_zones + j;
 		unsigned long size, realsize;
 
-		realsize = size = zones_size[j];
-		if (zholes_size)
-			realsize -= zholes_size[j];
+		size = zone_spanned_pages_in_node(nid, j, zones_size);
+		realsize = size - zone_absent_pages_in_node(nid, j,
+								zholes_size);
 
 		if (!is_highmem_idx(j))
 			nr_kernel_pages += realsize;
@@ -2073,8 +2349,13 @@ static void __init alloc_node_mem_map(struct pglist_data *pgdat)
 	/*
 	 * With no DISCONTIG, the global mem_map is just set as node 0's
 	 */
-	if (pgdat == NODE_DATA(0))
+	if (pgdat == NODE_DATA(0)) {
 		mem_map = NODE_DATA(0)->node_mem_map;
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+		if (page_to_pfn(mem_map) != pgdat->node_start_pfn)
+			mem_map -= pgdat->node_start_pfn;
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+	}
 #endif
 #endif /* CONFIG_FLAT_NODE_MEM_MAP */
 }
@@ -2085,13 +2366,236 @@ void __meminit free_area_init_node(int nid, struct pglist_data *pgdat,
 {
 	pgdat->node_id = nid;
 	pgdat->node_start_pfn = node_start_pfn;
-	calculate_zone_totalpages(pgdat, zones_size, zholes_size);
+	calculate_node_totalpages(pgdat, zones_size, zholes_size);
 
 	alloc_node_mem_map(pgdat);
 
 	free_area_init_core(pgdat, zones_size, zholes_size);
 }
 
+#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
+/**
+ * add_active_range - Register a range of PFNs backed by physical memory
+ * @nid: The node ID the range resides on
+ * @start_pfn: The start PFN of the available physical memory
+ * @end_pfn: The end PFN of the available physical memory
+ *
+ * These ranges are stored in an early_node_map[] and later used by
+ * free_area_init_nodes() to calculate zone sizes and holes. If the
+ * range spans a memory hole, it is up to the architecture to ensure
+ * the memory is not freed by the bootmem allocator. If possible
+ * the range being registered will be merged with existing ranges.
+ */
+void __init add_active_range(unsigned int nid, unsigned long start_pfn,
+						unsigned long end_pfn)
+{
+	int i;
+
+	printk(KERN_DEBUG "Entering add_active_range(%d, %lu, %lu) "
+			  "%d entries of %d used\n",
+			  nid, start_pfn, end_pfn,
+			  nr_nodemap_entries, MAX_ACTIVE_REGIONS);
+
+	/* Merge with existing active regions if possible */
+	for (i = 0; i < nr_nodemap_entries; i++) {
+		if (early_node_map[i].nid != nid)
+			continue;
+
+		/* Skip if an existing region covers this new one */
+		if (start_pfn >= early_node_map[i].start_pfn &&
+				end_pfn <= early_node_map[i].end_pfn)
+			return;
+
+		/* Merge forward if suitable */
+		if (start_pfn <= early_node_map[i].end_pfn &&
+				end_pfn > early_node_map[i].end_pfn) {
+			early_node_map[i].end_pfn = end_pfn;
+			return;
+		}
+
+		/* Merge backward if suitable */
+		if (start_pfn < early_node_map[i].end_pfn &&
+				end_pfn >= early_node_map[i].start_pfn) {
+			early_node_map[i].start_pfn = start_pfn;
+			return;
+		}
+	}
+
+	/* Check that early_node_map is large enough */
+	if (i >= MAX_ACTIVE_REGIONS) {
+		printk(KERN_CRIT "More than %d memory regions, truncating\n",
+							MAX_ACTIVE_REGIONS);
+		return;
+	}
+
+	early_node_map[i].nid = nid;
+	early_node_map[i].start_pfn = start_pfn;
+	early_node_map[i].end_pfn = end_pfn;
+	nr_nodemap_entries = i + 1;
+}
+
+/**
+ * shrink_active_range - Shrink an existing registered range of PFNs
+ * @nid: The node id the range is on that should be shrunk
+ * @old_end_pfn: The old end PFN of the range
+ * @new_end_pfn: The new PFN of the range
+ *
+ * i386 with NUMA use alloc_remap() to store a node_mem_map on a local node.
+ * The map is kept at the end physical page range that has already been
+ * registered with add_active_range(). This function allows an arch to shrink
+ * an existing registered range.
+ */
+void __init shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
+						unsigned long new_end_pfn)
+{
+	int i;
+
+	/* Find the old active region end and shrink */
+	for_each_active_range_index_in_nid(i, nid)
+		if (early_node_map[i].end_pfn == old_end_pfn) {
+			early_node_map[i].end_pfn = new_end_pfn;
+			break;
+		}
+}
+
+/**
+ * remove_all_active_ranges - Remove all currently registered regions
+ * During discovery, it may be found that a table like SRAT is invalid
+ * and an alternative discovery method must be used. This function removes
+ * all currently registered regions.
+ */
+void __init remove_all_active_ranges()
+{
+	memset(early_node_map, 0, sizeof(early_node_map));
+	nr_nodemap_entries = 0;
+}
+
+/* Compare two active node_active_regions */
+static int __init cmp_node_active_region(const void *a, const void *b)
+{
+	struct node_active_region *arange = (struct node_active_region *)a;
+	struct node_active_region *brange = (struct node_active_region *)b;
+
+	/* Done this way to avoid overflows */
+	if (arange->start_pfn > brange->start_pfn)
+		return 1;
+	if (arange->start_pfn < brange->start_pfn)
+		return -1;
+
+	return 0;
+}
+
+/* sort the node_map by start_pfn */
+static void __init sort_node_map(void)
+{
+	sort(early_node_map, (size_t)nr_nodemap_entries,
+			sizeof(struct node_active_region),
+			cmp_node_active_region, NULL);
+}
+
+/* Find the lowest pfn for a node. This depends on a sorted early_node_map */
+unsigned long __init find_min_pfn_for_node(unsigned long nid)
+{
+	int i;
+
+	/* Assuming a sorted map, the first range found has the starting pfn */
+	for_each_active_range_index_in_nid(i, nid)
+		return early_node_map[i].start_pfn;
+
+	printk(KERN_WARNING "Could not find start_pfn for node %lu\n", nid);
+	return 0;
+}
+
+/**
+ * find_min_pfn_with_active_regions - Find the minimum PFN registered
+ *
+ * It returns the minimum PFN based on information provided via
+ * add_active_range()
+ */
+unsigned long __init find_min_pfn_with_active_regions(void)
+{
+	return find_min_pfn_for_node(MAX_NUMNODES);
+}
+
+/**
+ * find_max_pfn_with_active_regions - Find the maximum PFN registered
+ *
+ * It returns the maximum PFN based on information provided via
+ * add_active_range()
+ */
+unsigned long __init find_max_pfn_with_active_regions(void)
+{
+	int i;
+	unsigned long max_pfn = 0;
+
+	for (i = 0; i < nr_nodemap_entries; i++)
+		max_pfn = max(max_pfn, early_node_map[i].end_pfn);
+
+	return max_pfn;
+}
+
+/**
+ * free_area_init_nodes - Initialise all pg_data_t and zone data
+ * @arch_max_dma_pfn: The maximum PFN usable for ZONE_DMA
+ * @arch_max_dma32_pfn: The maximum PFN usable for ZONE_DMA32
+ * @arch_max_low_pfn: The maximum PFN usable for ZONE_NORMAL
+ * @arch_max_high_pfn: The maximum PFN usable for ZONE_HIGHMEM
+ *
+ * This will call free_area_init_node() for each active node in the system.
+ * Using the page ranges provided by add_active_range(), the size of each
+ * zone in each node and their holes is calculated. If the maximum PFN
+ * between two adjacent zones match, it is assumed that the zone is empty.
+ * For example, if arch_max_dma_pfn == arch_max_dma32_pfn, it is assumed
+ * that arch_max_dma32_pfn has no pages. It is also assumed that a zone
+ * starts where the previous one ended. For example, ZONE_DMA32 starts
+ * at arch_max_dma_pfn.
+ */
+void __init free_area_init_nodes(unsigned long *max_zone_pfn)
+{
+	unsigned long nid;
+	enum zone_type i;
+
+	/* Record where the zone boundaries are */
+	memset(arch_zone_lowest_possible_pfn, 0,
+				sizeof(arch_zone_lowest_possible_pfn));
+	memset(arch_zone_highest_possible_pfn, 0,
+				sizeof(arch_zone_highest_possible_pfn));
+	arch_zone_lowest_possible_pfn[0] = find_min_pfn_with_active_regions();
+	arch_zone_highest_possible_pfn[0] = max_zone_pfn[0];
+	for (i = 1; i < MAX_NR_ZONES; i++) {
+		arch_zone_lowest_possible_pfn[i] =
+			arch_zone_highest_possible_pfn[i-1];
+		arch_zone_highest_possible_pfn[i] =
+			max(max_zone_pfn[i], arch_zone_lowest_possible_pfn[i]);
+	}
+
+	/* Regions in the early_node_map can be in any order */
+	sort_node_map();
+
+	/* Print out the zone ranges */
+	printk("Zone PFN ranges:\n");
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		printk("  %-8s %8lu -> %8lu\n",
+				zone_names[i],
+				arch_zone_lowest_possible_pfn[i],
+				arch_zone_highest_possible_pfn[i]);
+
+	/* Print out the early_node_map[] */
+	printk("early_node_map[%d] active PFN ranges\n", nr_nodemap_entries);
+	for (i = 0; i < nr_nodemap_entries; i++)
+		printk("  %3d: %8lu -> %8lu\n", early_node_map[i].nid,
+						early_node_map[i].start_pfn,
+						early_node_map[i].end_pfn);
+
+	/* Initialise every node */
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		free_area_init_node(nid, pgdat, NULL,
+				find_min_pfn_for_node(nid), NULL);
+	}
+}
+#endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 static bootmem_data_t contig_bootmem_data;
 struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
-- 
cgit v1.2.1


From c67c3cb4c99fb2ee63c8733943c353d745f45b84 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:49 -0700
Subject: [PATCH] Have Power use add_active_range() and free_area_init_nodes()

Size zones and holes in an architecture independent manner for Power.

[judith@osdl.org: build fix]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/Kconfig   |   7 +--
 arch/powerpc/mm/mem.c  |  51 ++++++----------
 arch/powerpc/mm/numa.c | 159 +++++--------------------------------------------
 arch/ppc/Kconfig       |   3 +
 arch/ppc/mm/init.c     |  23 ++++---
 5 files changed, 52 insertions(+), 191 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index de1ef2fa1a20..a0dd1b0ee483 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -731,11 +731,10 @@ config ARCH_SPARSEMEM_DEFAULT
 	def_bool y
 	depends on SMP && PPC_PSERIES
 
-source "mm/Kconfig"
-
-config HAVE_ARCH_EARLY_PFN_TO_NID
+config ARCH_POPULATES_NODE_MAP
 	def_bool y
-	depends on NEED_MULTIPLE_NODES
+
+source "mm/Kconfig"
 
 config ARCH_MEMORY_PROBE
 	def_bool y
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index eebd8b83a6b0..16fe027bbc12 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -256,20 +256,22 @@ void __init do_init_bootmem(void)
 
 	boot_mapsize = init_bootmem(start >> PAGE_SHIFT, total_pages);
 
+	/* Add active regions with valid PFNs */
+	for (i = 0; i < lmb.memory.cnt; i++) {
+		unsigned long start_pfn, end_pfn;
+		start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
+		end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
+		add_active_range(0, start_pfn, end_pfn);
+	}
+
 	/* Add all physical memory to the bootmem map, mark each area
 	 * present.
 	 */
-	for (i = 0; i < lmb.memory.cnt; i++) {
-		unsigned long base = lmb.memory.region[i].base;
-		unsigned long size = lmb_size_bytes(&lmb.memory, i);
 #ifdef CONFIG_HIGHMEM
-		if (base >= total_lowmem)
-			continue;
-		if (base + size > total_lowmem)
-			size = total_lowmem - base;
+	free_bootmem_with_active_regions(0, total_lowmem >> PAGE_SHIFT);
+#else
+	free_bootmem_with_active_regions(0, max_pfn);
 #endif
-		free_bootmem(base, size);
-	}
 
 	/* reserve the sections we're already using */
 	for (i = 0; i < lmb.reserved.cnt; i++)
@@ -277,9 +279,8 @@ void __init do_init_bootmem(void)
 				lmb_size_bytes(&lmb.reserved, i));
 
 	/* XXX need to clip this if using highmem? */
-	for (i = 0; i < lmb.memory.cnt; i++)
-		memory_present(0, lmb_start_pfn(&lmb.memory, i),
-			       lmb_end_pfn(&lmb.memory, i));
+	sparse_memory_present_with_active_regions(0);
+
 	init_bootmem_done = 1;
 }
 
@@ -288,10 +289,9 @@ void __init do_init_bootmem(void)
  */
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES];
-	unsigned long zholes_size[MAX_NR_ZONES];
 	unsigned long total_ram = lmb_phys_mem_size();
 	unsigned long top_of_ram = lmb_end_of_DRAM();
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 #ifdef CONFIG_HIGHMEM
 	map_page(PKMAP_BASE, 0, 0);	/* XXX gross */
@@ -307,26 +307,13 @@ void __init paging_init(void)
 	       top_of_ram, total_ram);
 	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
 	       (top_of_ram - total_ram) >> 20);
-	/*
-	 * All pages are DMA-able so we put them all in the DMA zone.
-	 */
-	memset(zones_size, 0, sizeof(zones_size));
-	memset(zholes_size, 0, sizeof(zholes_size));
-
-	zones_size[ZONE_DMA] = top_of_ram >> PAGE_SHIFT;
-	zholes_size[ZONE_DMA] = (top_of_ram - total_ram) >> PAGE_SHIFT;
-
 #ifdef CONFIG_HIGHMEM
-	zones_size[ZONE_DMA] = total_lowmem >> PAGE_SHIFT;
-	zones_size[ZONE_HIGHMEM] = (total_memory - total_lowmem) >> PAGE_SHIFT;
-	zholes_size[ZONE_HIGHMEM] = (top_of_ram - total_ram) >> PAGE_SHIFT;
+	max_zone_pfns[0] = total_lowmem >> PAGE_SHIFT;
+	max_zone_pfns[1] = top_of_ram >> PAGE_SHIFT;
 #else
-	zones_size[ZONE_DMA] = top_of_ram >> PAGE_SHIFT;
-	zholes_size[ZONE_DMA] = (top_of_ram - total_ram) >> PAGE_SHIFT;
-#endif /* CONFIG_HIGHMEM */
-
-	free_area_init_node(0, NODE_DATA(0), zones_size,
-			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, zholes_size);
+	max_zone_pfns[0] = top_of_ram >> PAGE_SHIFT;
+#endif
+	free_area_init_nodes(max_zone_pfns);
 }
 #endif /* ! CONFIG_NEED_MULTIPLE_NODES */
 
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 6c0f1c7d83e5..43c272075e1a 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -39,96 +39,6 @@ static bootmem_data_t __initdata plat_node_bdata[MAX_NUMNODES];
 static int min_common_depth;
 static int n_mem_addr_cells, n_mem_size_cells;
 
-/*
- * We need somewhere to store start/end/node for each region until we have
- * allocated the real node_data structures.
- */
-#define MAX_REGIONS	(MAX_LMB_REGIONS*2)
-static struct {
-	unsigned long start_pfn;
-	unsigned long end_pfn;
-	int nid;
-} init_node_data[MAX_REGIONS] __initdata;
-
-int __init early_pfn_to_nid(unsigned long pfn)
-{
-	unsigned int i;
-
-	for (i = 0; init_node_data[i].end_pfn; i++) {
-		unsigned long start_pfn = init_node_data[i].start_pfn;
-		unsigned long end_pfn = init_node_data[i].end_pfn;
-
-		if ((start_pfn <= pfn) && (pfn < end_pfn))
-			return init_node_data[i].nid;
-	}
-
-	return -1;
-}
-
-void __init add_region(unsigned int nid, unsigned long start_pfn,
-		       unsigned long pages)
-{
-	unsigned int i;
-
-	dbg("add_region nid %d start_pfn 0x%lx pages 0x%lx\n",
-		nid, start_pfn, pages);
-
-	for (i = 0; init_node_data[i].end_pfn; i++) {
-		if (init_node_data[i].nid != nid)
-			continue;
-		if (init_node_data[i].end_pfn == start_pfn) {
-			init_node_data[i].end_pfn += pages;
-			return;
-		}
-		if (init_node_data[i].start_pfn == (start_pfn + pages)) {
-			init_node_data[i].start_pfn -= pages;
-			return;
-		}
-	}
-
-	/*
-	 * Leave last entry NULL so we dont iterate off the end (we use
-	 * entry.end_pfn to terminate the walk).
-	 */
-	if (i >= (MAX_REGIONS - 1)) {
-		printk(KERN_ERR "WARNING: too many memory regions in "
-				"numa code, truncating\n");
-		return;
-	}
-
-	init_node_data[i].start_pfn = start_pfn;
-	init_node_data[i].end_pfn = start_pfn + pages;
-	init_node_data[i].nid = nid;
-}
-
-/* We assume init_node_data has no overlapping regions */
-void __init get_region(unsigned int nid, unsigned long *start_pfn,
-		       unsigned long *end_pfn, unsigned long *pages_present)
-{
-	unsigned int i;
-
-	*start_pfn = -1UL;
-	*end_pfn = *pages_present = 0;
-
-	for (i = 0; init_node_data[i].end_pfn; i++) {
-		if (init_node_data[i].nid != nid)
-			continue;
-
-		*pages_present += init_node_data[i].end_pfn -
-			init_node_data[i].start_pfn;
-
-		if (init_node_data[i].start_pfn < *start_pfn)
-			*start_pfn = init_node_data[i].start_pfn;
-
-		if (init_node_data[i].end_pfn > *end_pfn)
-			*end_pfn = init_node_data[i].end_pfn;
-	}
-
-	/* We didnt find a matching region, return start/end as 0 */
-	if (*start_pfn == -1UL)
-		*start_pfn = 0;
-}
-
 static void __cpuinit map_cpu_to_node(int cpu, int node)
 {
 	numa_cpu_lookup_table[cpu] = node;
@@ -468,8 +378,8 @@ new_range:
 				continue;
 		}
 
-		add_region(nid, start >> PAGE_SHIFT,
-			   size >> PAGE_SHIFT);
+		add_active_range(nid, start >> PAGE_SHIFT,
+				(start >> PAGE_SHIFT) + (size >> PAGE_SHIFT));
 
 		if (--ranges)
 			goto new_range;
@@ -482,6 +392,7 @@ static void __init setup_nonnuma(void)
 {
 	unsigned long top_of_ram = lmb_end_of_DRAM();
 	unsigned long total_ram = lmb_phys_mem_size();
+	unsigned long start_pfn, end_pfn;
 	unsigned int i;
 
 	printk(KERN_DEBUG "Top of RAM: 0x%lx, Total RAM: 0x%lx\n",
@@ -489,9 +400,11 @@ static void __init setup_nonnuma(void)
 	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
 	       (top_of_ram - total_ram) >> 20);
 
-	for (i = 0; i < lmb.memory.cnt; ++i)
-		add_region(0, lmb.memory.region[i].base >> PAGE_SHIFT,
-			   lmb_size_pages(&lmb.memory, i));
+	for (i = 0; i < lmb.memory.cnt; ++i) {
+		start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
+		end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
+		add_active_range(0, start_pfn, end_pfn);
+	}
 	node_set_online(0);
 }
 
@@ -630,11 +543,11 @@ void __init do_init_bootmem(void)
 			  (void *)(unsigned long)boot_cpuid);
 
 	for_each_online_node(nid) {
-		unsigned long start_pfn, end_pfn, pages_present;
+		unsigned long start_pfn, end_pfn;
 		unsigned long bootmem_paddr;
 		unsigned long bootmap_pages;
 
-		get_region(nid, &start_pfn, &end_pfn, &pages_present);
+		get_pfn_range_for_nid(nid, &start_pfn, &end_pfn);
 
 		/* Allocate the node structure node local if possible */
 		NODE_DATA(nid) = careful_allocation(nid,
@@ -667,19 +580,7 @@ void __init do_init_bootmem(void)
 		init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT,
 				  start_pfn, end_pfn);
 
-		/* Add free regions on this node */
-		for (i = 0; init_node_data[i].end_pfn; i++) {
-			unsigned long start, end;
-
-			if (init_node_data[i].nid != nid)
-				continue;
-
-			start = init_node_data[i].start_pfn << PAGE_SHIFT;
-			end = init_node_data[i].end_pfn << PAGE_SHIFT;
-
-			dbg("free_bootmem %lx %lx\n", start, end - start);
-  			free_bootmem_node(NODE_DATA(nid), start, end - start);
-		}
+		free_bootmem_with_active_regions(nid, end_pfn);
 
 		/* Mark reserved regions on this node */
 		for (i = 0; i < lmb.reserved.cnt; i++) {
@@ -710,44 +611,16 @@ void __init do_init_bootmem(void)
 			}
 		}
 
-		/* Add regions into sparsemem */
-		for (i = 0; init_node_data[i].end_pfn; i++) {
-			unsigned long start, end;
-
-			if (init_node_data[i].nid != nid)
-				continue;
-
-			start = init_node_data[i].start_pfn;
-			end = init_node_data[i].end_pfn;
-
-			memory_present(nid, start, end);
-		}
+		sparse_memory_present_with_active_regions(nid);
 	}
 }
 
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES];
-	unsigned long zholes_size[MAX_NR_ZONES];
-	int nid;
-
-	memset(zones_size, 0, sizeof(zones_size));
-	memset(zholes_size, 0, sizeof(zholes_size));
-
-	for_each_online_node(nid) {
-		unsigned long start_pfn, end_pfn, pages_present;
-
-		get_region(nid, &start_pfn, &end_pfn, &pages_present);
-
-		zones_size[ZONE_DMA] = end_pfn - start_pfn;
-		zholes_size[ZONE_DMA] = zones_size[ZONE_DMA] - pages_present;
-
-		dbg("free_area_init node %d %lx %lx (hole: %lx)\n", nid,
-		    zones_size[ZONE_DMA], start_pfn, zholes_size[ZONE_DMA]);
-
-		free_area_init_node(nid, NODE_DATA(nid), zones_size, start_pfn,
-				    zholes_size);
-	}
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {
+				lmb_end_of_DRAM() >> PAGE_SHIFT
+	};
+	free_area_init_nodes(max_zone_pfns);
 }
 
 static int __init early_numa(char *p)
diff --git a/arch/ppc/Kconfig b/arch/ppc/Kconfig
index 8fa10cf661a8..fdd9e7b66244 100644
--- a/arch/ppc/Kconfig
+++ b/arch/ppc/Kconfig
@@ -953,6 +953,9 @@ config NR_CPUS
 config HIGHMEM
 	bool "High memory support"
 
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
+
 source kernel/Kconfig.hz
 source kernel/Kconfig.preempt
 source "mm/Kconfig"
diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c
index 523392d460fa..410200046af1 100644
--- a/arch/ppc/mm/init.c
+++ b/arch/ppc/mm/init.c
@@ -358,8 +358,8 @@ void __init do_init_bootmem(void)
  */
 void __init paging_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES], i;
-
+	unsigned long start_pfn, end_pfn;
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 #ifdef CONFIG_HIGHMEM
 	map_page(PKMAP_BASE, 0, 0);	/* XXX gross */
 	pkmap_page_table = pte_offset_kernel(pmd_offset(pgd_offset_k
@@ -369,19 +369,18 @@ void __init paging_init(void)
 			(KMAP_FIX_BEGIN), KMAP_FIX_BEGIN), KMAP_FIX_BEGIN);
 	kmap_prot = PAGE_KERNEL;
 #endif /* CONFIG_HIGHMEM */
-
-	/*
-	 * All pages are DMA-able so we put them all in the DMA zone.
-	 */
-	zones_size[ZONE_DMA] = total_lowmem >> PAGE_SHIFT;
-	for (i = 1; i < MAX_NR_ZONES; i++)
-		zones_size[i] = 0;
+	/* All pages are DMA-able so we put them all in the DMA zone. */
+	start_pfn = __pa(PAGE_OFFSET) >> PAGE_SHIFT;
+	end_pfn = start_pfn + (total_memory >> PAGE_SHIFT);
+	add_active_range(0, start_pfn, end_pfn);
 
 #ifdef CONFIG_HIGHMEM
-	zones_size[ZONE_HIGHMEM] = (total_memory - total_lowmem) >> PAGE_SHIFT;
+	max_zone_pfns[0] = total_lowmem >> PAGE_SHIFT;
+	max_zone_pfns[1] = total_memory >> PAGE_SHIFT;
+#else
+	max_zone_pfns[0] = total_memory >> PAGE_SHIFT;
 #endif /* CONFIG_HIGHMEM */
-
-	free_area_init(zones_size);
+	free_area_init_nodes(max_zone_pfns);
 }
 
 void __init mem_init(void)
-- 
cgit v1.2.1


From 4cfee88ad30acc47f02b8b7ba3db8556262dce1e Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:51 -0700
Subject: [PATCH] Have x86 use add_active_range() and free_area_init_nodes

Size zones and holes in an architecture independent manner for x86.

[akpm@osdl.org: build fix]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig        |  8 ++--
 arch/i386/kernel/setup.c | 26 ++++++-------
 arch/i386/kernel/srat.c  | 97 +-----------------------------------------------
 arch/i386/mm/discontig.c | 69 +++++++++-------------------------
 4 files changed, 34 insertions(+), 166 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 758044f5e718..3fd9f1e8b093 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -598,12 +598,10 @@ config ARCH_SELECT_MEMORY_MODEL
 	def_bool y
 	depends on ARCH_SPARSEMEM_ENABLE
 
-source "mm/Kconfig"
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
 
-config HAVE_ARCH_EARLY_PFN_TO_NID
-	bool
-	default y
-	depends on NUMA
+source "mm/Kconfig"
 
 config HIGHPTE
 	bool "Allocate 3rd-level pagetables from highmem"
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index 76a524b4c90f..814cdebf7377 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1089,22 +1089,20 @@ static unsigned long __init setup_memory(void)
 
 void __init zone_sizes_init(void)
 {
-	unsigned long zones_size[MAX_NR_ZONES] = { 0, };
-	unsigned int max_dma, low;
-
-	max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-	low = max_low_pfn;
-
-	if (low < max_dma)
-		zones_size[ZONE_DMA] = low;
-	else {
-		zones_size[ZONE_DMA] = max_dma;
-		zones_size[ZONE_NORMAL] = low - max_dma;
 #ifdef CONFIG_HIGHMEM
-		zones_size[ZONE_HIGHMEM] = highend_pfn - low;
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {
+			virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT,
+			max_low_pfn,
+			highend_pfn};
+	add_active_range(0, 0, highend_pfn);
+#else
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {
+			virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT,
+			max_low_pfn};
+	add_active_range(0, 0, max_low_pfn);
 #endif
-	}
-	free_area_init(zones_size);
+
+	free_area_init_nodes(max_zone_pfns);
 }
 #else
 extern unsigned long __init setup_memory(void);
diff --git a/arch/i386/kernel/srat.c b/arch/i386/kernel/srat.c
index 83db411b3aa7..32413122c4c2 100644
--- a/arch/i386/kernel/srat.c
+++ b/arch/i386/kernel/srat.c
@@ -54,8 +54,6 @@ struct node_memory_chunk_s {
 static struct node_memory_chunk_s node_memory_chunk[MAXCHUNKS];
 
 static int num_memory_chunks;		/* total number of memory chunks */
-static int zholes_size_init;
-static unsigned long zholes_size[MAX_NUMNODES * MAX_NR_ZONES];
 
 extern void * boot_ioremap(unsigned long, unsigned long);
 
@@ -135,47 +133,6 @@ static void __init parse_memory_affinity_structure (char *sratp)
 		 "enabled and removable" : "enabled" ) );
 }
 
-/* Take a chunk of pages from page frame cstart to cend and count the number
- * of pages in each zone, returned via zones[].
- */
-static __init void chunk_to_zones(unsigned long cstart, unsigned long cend, 
-		unsigned long *zones)
-{
-	unsigned long max_dma;
-	extern unsigned long max_low_pfn;
-
-	int z;
-	unsigned long rend;
-
-	/* FIXME: MAX_DMA_ADDRESS and max_low_pfn are trying to provide
-	 * similarly scoped information and should be handled in a consistant
-	 * manner.
-	 */
-	max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-
-	/* Split the hole into the zones in which it falls.  Repeatedly
-	 * take the segment in which the remaining hole starts, round it
-	 * to the end of that zone.
-	 */
-	memset(zones, 0, MAX_NR_ZONES * sizeof(long));
-	while (cstart < cend) {
-		if (cstart < max_dma) {
-			z = ZONE_DMA;
-			rend = (cend < max_dma)? cend : max_dma;
-
-		} else if (cstart < max_low_pfn) {
-			z = ZONE_NORMAL;
-			rend = (cend < max_low_pfn)? cend : max_low_pfn;
-
-		} else {
-			z = ZONE_HIGHMEM;
-			rend = cend;
-		}
-		zones[z] += rend - cstart;
-		cstart = rend;
-	}
-}
-
 /*
  * The SRAT table always lists ascending addresses, so can always
  * assume that the first "start" address that you see is the real
@@ -220,7 +177,6 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp)
 
 	memset(pxm_bitmap, 0, sizeof(pxm_bitmap));	/* init proximity domain bitmap */
 	memset(node_memory_chunk, 0, sizeof(node_memory_chunk));
-	memset(zholes_size, 0, sizeof(zholes_size));
 
 	num_memory_chunks = 0;
 	while (p < end) {
@@ -284,6 +240,7 @@ static int __init acpi20_parse_srat(struct acpi_table_srat *sratp)
 		printk("chunk %d nid %d start_pfn %08lx end_pfn %08lx\n",
 		       j, chunk->nid, chunk->start_pfn, chunk->end_pfn);
 		node_read_chunk(chunk->nid, chunk);
+		add_active_range(chunk->nid, chunk->start_pfn, chunk->end_pfn);
 	}
  
 	for_each_online_node(nid) {
@@ -392,57 +349,7 @@ int __init get_memcfg_from_srat(void)
 		return acpi20_parse_srat((struct acpi_table_srat *)header);
 	}
 out_err:
+	remove_all_active_ranges();
 	printk("failed to get NUMA memory information from SRAT table\n");
 	return 0;
 }
-
-/* For each node run the memory list to determine whether there are
- * any memory holes.  For each hole determine which ZONE they fall
- * into.
- *
- * NOTE#1: this requires knowledge of the zone boundries and so
- * _cannot_ be performed before those are calculated in setup_memory.
- * 
- * NOTE#2: we rely on the fact that the memory chunks are ordered by
- * start pfn number during setup.
- */
-static void __init get_zholes_init(void)
-{
-	int nid;
-	int c;
-	int first;
-	unsigned long end = 0;
-
-	for_each_online_node(nid) {
-		first = 1;
-		for (c = 0; c < num_memory_chunks; c++){
-			if (node_memory_chunk[c].nid == nid) {
-				if (first) {
-					end = node_memory_chunk[c].end_pfn;
-					first = 0;
-
-				} else {
-					/* Record any gap between this chunk
-					 * and the previous chunk on this node
-					 * against the zones it spans.
-					 */
-					chunk_to_zones(end,
-						node_memory_chunk[c].start_pfn,
-						&zholes_size[nid * MAX_NR_ZONES]);
-				}
-			}
-		}
-	}
-}
-
-unsigned long * __init get_zholes_size(int nid)
-{
-	if (!zholes_size_init) {
-		zholes_size_init++;
-		get_zholes_init();
-	}
-	if (nid >= MAX_NUMNODES || !node_online(nid))
-		printk("%s: nid = %d is invalid/offline. num_online_nodes = %d",
-		       __FUNCTION__, nid, num_online_nodes());
-	return &zholes_size[nid * MAX_NR_ZONES];
-}
diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c
index 941d1a5ebabb..51e3739dd227 100644
--- a/arch/i386/mm/discontig.c
+++ b/arch/i386/mm/discontig.c
@@ -157,21 +157,6 @@ static void __init find_max_pfn_node(int nid)
 		BUG();
 }
 
-/* Find the owning node for a pfn. */
-int early_pfn_to_nid(unsigned long pfn)
-{
-	int nid;
-
-	for_each_node(nid) {
-		if (node_end_pfn[nid] == 0)
-			break;
-		if (node_start_pfn[nid] <= pfn && node_end_pfn[nid] >= pfn)
-			return nid;
-	}
-
-	return 0;
-}
-
 /* 
  * Allocate memory for the pg_data_t for this node via a crude pre-bootmem
  * method.  For node zero take this from the bottom of memory, for
@@ -227,6 +212,8 @@ static unsigned long calculate_numa_remap_pages(void)
 	unsigned long pfn;
 
 	for_each_online_node(nid) {
+		unsigned old_end_pfn = node_end_pfn[nid];
+
 		/*
 		 * The acpi/srat node info can show hot-add memroy zones
 		 * where memory could be added but not currently present.
@@ -276,6 +263,7 @@ static unsigned long calculate_numa_remap_pages(void)
 
 		node_end_pfn[nid] -= size;
 		node_remap_start_pfn[nid] = node_end_pfn[nid];
+		shrink_active_range(nid, old_end_pfn, node_end_pfn[nid]);
 	}
 	printk("Reserving total of %ld pages for numa KVA remap\n",
 			reserve_pages);
@@ -369,45 +357,22 @@ void __init numa_kva_reserve(void)
 void __init zone_sizes_init(void)
 {
 	int nid;
-
-
-	for_each_online_node(nid) {
-		unsigned long zones_size[MAX_NR_ZONES] = {0, };
-		unsigned long *zholes_size;
-		unsigned int max_dma;
-
-		unsigned long low = max_low_pfn;
-		unsigned long start = node_start_pfn[nid];
-		unsigned long high = node_end_pfn[nid];
-
-		max_dma = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
-
-		if (node_has_online_mem(nid)){
-			if (start > low) {
-#ifdef CONFIG_HIGHMEM
-				BUG_ON(start > high);
-				zones_size[ZONE_HIGHMEM] = high - start;
-#endif
-			} else {
-				if (low < max_dma)
-					zones_size[ZONE_DMA] = low;
-				else {
-					BUG_ON(max_dma > low);
-					BUG_ON(low > high);
-					zones_size[ZONE_DMA] = max_dma;
-					zones_size[ZONE_NORMAL] = low - max_dma;
-#ifdef CONFIG_HIGHMEM
-					zones_size[ZONE_HIGHMEM] = high - low;
-#endif
-				}
-			}
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {
+		virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT,
+		max_low_pfn,
+		highend_pfn
+	};
+
+	/* If SRAT has not registered memory, register it now */
+	if (find_max_pfn_with_active_regions() == 0) {
+		for_each_online_node(nid) {
+			if (node_has_online_mem(nid))
+				add_active_range(nid, node_start_pfn[nid],
+							node_end_pfn[nid]);
 		}
-
-		zholes_size = get_zholes_size(nid);
-
-		free_area_init_node(nid, NODE_DATA(nid), zones_size, start,
-				zholes_size);
 	}
+
+	free_area_init_nodes(max_zone_pfns);
 	return;
 }
 
-- 
cgit v1.2.1


From 5cb248abf5ab65ab543b2d5fc16c738b28031fc0 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:52 -0700
Subject: [PATCH] Have x86_64 use add_active_range() and free_area_init_nodes

Size zones and holes in an architecture independent manner for x86_64.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/Kconfig         |   3 ++
 arch/x86_64/kernel/e820.c   | 125 ++++++++++++++++----------------------------
 arch/x86_64/kernel/setup.c  |   7 ++-
 arch/x86_64/mm/init.c       |  65 +++--------------------
 arch/x86_64/mm/k8topology.c |   3 ++
 arch/x86_64/mm/numa.c       |  21 ++++----
 arch/x86_64/mm/srat.c       |  11 ++--
 include/asm-x86_64/e820.h   |   5 +-
 include/asm-x86_64/proto.h  |   2 -
 9 files changed, 84 insertions(+), 158 deletions(-)

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index efe249e7d6b3..326aff7a87ea 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -85,6 +85,9 @@ config ARCH_MAY_HAVE_PC_FDC
 	bool
 	default y
 
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
+
 config DMI
 	bool
 	default y
diff --git a/arch/x86_64/kernel/e820.c b/arch/x86_64/kernel/e820.c
index c0af3828df45..b3f0908668ec 100644
--- a/arch/x86_64/kernel/e820.c
+++ b/arch/x86_64/kernel/e820.c
@@ -162,59 +162,14 @@ unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsi
 	return -1UL;		
 } 
 
-/* 
- * Free bootmem based on the e820 table for a node.
- */
-void __init e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end)
-{
-	int i;
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i]; 
-		unsigned long last, addr;
-
-		if (ei->type != E820_RAM || 
-		    ei->addr+ei->size <= start || 
-		    ei->addr >= end)
-			continue;
-
-		addr = round_up(ei->addr, PAGE_SIZE);
-		if (addr < start) 
-			addr = start;
-
-		last = round_down(ei->addr + ei->size, PAGE_SIZE); 
-		if (last >= end)
-			last = end; 
-
-		if (last > addr && last-addr >= PAGE_SIZE)
-			free_bootmem_node(pgdat, addr, last-addr);
-	}
-}
-
 /*
  * Find the highest page frame number we have available
  */
 unsigned long __init e820_end_of_ram(void)
 {
-	int i;
 	unsigned long end_pfn = 0;
+	end_pfn = find_max_pfn_with_active_regions();
 	
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i]; 
-		unsigned long start, end;
-
-		start = round_up(ei->addr, PAGE_SIZE); 
-		end = round_down(ei->addr + ei->size, PAGE_SIZE); 
-		if (start >= end)
-			continue;
-		if (ei->type == E820_RAM) { 
-		if (end > end_pfn<<PAGE_SHIFT)
-			end_pfn = end>>PAGE_SHIFT;
-		} else { 
-			if (end > end_pfn_map<<PAGE_SHIFT) 
-				end_pfn_map = end>>PAGE_SHIFT;
-		} 
-	}
-
 	if (end_pfn > end_pfn_map) 
 		end_pfn_map = end_pfn;
 	if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
@@ -224,43 +179,10 @@ unsigned long __init e820_end_of_ram(void)
 	if (end_pfn > end_pfn_map) 
 		end_pfn = end_pfn_map; 
 
+	printk("end_pfn_map = %lu\n", end_pfn_map);
 	return end_pfn;	
 }
 
-/* 
- * Compute how much memory is missing in a range.
- * Unlike the other functions in this file the arguments are in page numbers.
- */
-unsigned long __init
-e820_hole_size(unsigned long start_pfn, unsigned long end_pfn)
-{
-	unsigned long ram = 0;
-	unsigned long start = start_pfn << PAGE_SHIFT;
-	unsigned long end = end_pfn << PAGE_SHIFT;
-	int i;
-	for (i = 0; i < e820.nr_map; i++) {
-		struct e820entry *ei = &e820.map[i];
-		unsigned long last, addr;
-
-		if (ei->type != E820_RAM ||
-		    ei->addr+ei->size <= start ||
-		    ei->addr >= end)
-			continue;
-
-		addr = round_up(ei->addr, PAGE_SIZE);
-		if (addr < start)
-			addr = start;
-
-		last = round_down(ei->addr + ei->size, PAGE_SIZE);
-		if (last >= end)
-			last = end;
-
-		if (last > addr)
-			ram += last - addr;
-	}
-	return ((end - start) - ram) >> PAGE_SHIFT;
-}
-
 /*
  * Mark e820 reserved areas as busy for the resource manager.
  */
@@ -342,6 +264,49 @@ void __init e820_mark_nosave_regions(void)
 	}
 }
 
+/* Walk the e820 map and register active regions within a node */
+void __init
+e820_register_active_regions(int nid, unsigned long start_pfn,
+							unsigned long end_pfn)
+{
+	int i;
+	unsigned long ei_startpfn, ei_endpfn;
+	for (i = 0; i < e820.nr_map; i++) {
+		struct e820entry *ei = &e820.map[i];
+		ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
+		ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE)
+								>> PAGE_SHIFT;
+
+		/* Skip map entries smaller than a page */
+		if (ei_startpfn > ei_endpfn)
+			continue;
+
+		/* Check if end_pfn_map should be updated */
+		if (ei->type != E820_RAM && ei_endpfn > end_pfn_map)
+			end_pfn_map = ei_endpfn;
+
+		/* Skip if map is outside the node */
+		if (ei->type != E820_RAM ||
+				ei_endpfn <= start_pfn ||
+				ei_startpfn >= end_pfn)
+			continue;
+
+		/* Check for overlaps */
+		if (ei_startpfn < start_pfn)
+			ei_startpfn = start_pfn;
+		if (ei_endpfn > end_pfn)
+			ei_endpfn = end_pfn;
+
+		/* Obey end_user_pfn to save on memmap */
+		if (ei_startpfn >= end_user_pfn)
+			continue;
+		if (ei_endpfn > end_user_pfn)
+			ei_endpfn = end_user_pfn;
+
+		add_active_range(nid, ei_startpfn, ei_endpfn);
+	}
+}
+
 /* 
  * Add a memory region to the kernel e820 map.
  */ 
diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index f98e48cae6da..0b00bb2ea576 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -292,7 +292,8 @@ contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	if (bootmap == -1L)
 		panic("Cannot find bootmem map of size %ld\n",bootmap_size);
 	bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
-	e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
+	e820_register_active_regions(0, start_pfn, end_pfn);
+	free_bootmem_with_active_regions(0, end_pfn);
 	reserve_bootmem(bootmap, bootmap_size);
 } 
 #endif
@@ -384,6 +385,7 @@ void __init setup_arch(char **cmdline_p)
 
 	finish_e820_parsing();
 
+	e820_register_active_regions(0, 0, -1UL);
 	/*
 	 * partially used pages are not usable - thus
 	 * we are rounding upwards:
@@ -414,6 +416,9 @@ void __init setup_arch(char **cmdline_p)
 	max_pfn = end_pfn;
 	high_memory = (void *)__va(end_pfn * PAGE_SIZE - 1) + 1;
 
+	/* Remove active ranges so rediscovery with NUMA-awareness happens */
+	remove_all_active_ranges();
+
 #ifdef CONFIG_ACPI_NUMA
 	/*
 	 * Parse SRAT to discover nodes.
diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 1e4669fa5734..47928399e38a 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -403,69 +403,15 @@ void __cpuinit zap_low_mappings(int cpu)
 	__flush_tlb_all();
 }
 
-/* Compute zone sizes for the DMA and DMA32 zones in a node. */
-__init void
-size_zones(unsigned long *z, unsigned long *h,
-	   unsigned long start_pfn, unsigned long end_pfn)
-{
- 	int i;
- 	unsigned long w;
-
- 	for (i = 0; i < MAX_NR_ZONES; i++)
- 		z[i] = 0;
-
- 	if (start_pfn < MAX_DMA_PFN)
- 		z[ZONE_DMA] = MAX_DMA_PFN - start_pfn;
- 	if (start_pfn < MAX_DMA32_PFN) {
- 		unsigned long dma32_pfn = MAX_DMA32_PFN;
- 		if (dma32_pfn > end_pfn)
- 			dma32_pfn = end_pfn;
- 		z[ZONE_DMA32] = dma32_pfn - start_pfn;
- 	}
- 	z[ZONE_NORMAL] = end_pfn - start_pfn;
-
- 	/* Remove lower zones from higher ones. */
- 	w = 0;
- 	for (i = 0; i < MAX_NR_ZONES; i++) {
- 		if (z[i])
- 			z[i] -= w;
- 	        w += z[i];
-	}
-
-	/* Compute holes */
-	w = start_pfn;
-	for (i = 0; i < MAX_NR_ZONES; i++) {
-		unsigned long s = w;
-		w += z[i];
-		h[i] = e820_hole_size(s, w);
-	}
-
-	/* Add the space pace needed for mem_map to the holes too. */
-	for (i = 0; i < MAX_NR_ZONES; i++)
-		h[i] += (z[i] * sizeof(struct page)) / PAGE_SIZE;
-
-	/* The 16MB DMA zone has the kernel and other misc mappings.
- 	   Account them too */
-	if (h[ZONE_DMA]) {
-		h[ZONE_DMA] += dma_reserve;
-		if (h[ZONE_DMA] >= z[ZONE_DMA]) {
-			printk(KERN_WARNING
-				"Kernel too large and filling up ZONE_DMA?\n");
-			h[ZONE_DMA] = z[ZONE_DMA];
-		}
-	}
-}
-
 #ifndef CONFIG_NUMA
 void __init paging_init(void)
 {
-	unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
-
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = {MAX_DMA_PFN,
+							MAX_DMA32_PFN,
+							end_pfn};
 	memory_present(0, 0, end_pfn);
 	sparse_init();
-	size_zones(zones, holes, 0, end_pfn);
-	free_area_init_node(0, NODE_DATA(0), zones,
-			    __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
+	free_area_init_nodes(max_zone_pfns);
 }
 #endif
 
@@ -608,7 +554,8 @@ void __init mem_init(void)
 #else
 	totalram_pages = free_all_bootmem();
 #endif
-	reservedpages = end_pfn - totalram_pages - e820_hole_size(0, end_pfn);
+	reservedpages = end_pfn - totalram_pages -
+					absent_pages_in_range(0, end_pfn);
 
 	after_bootmem = 1;
 
diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index 5cf594f9230d..b5b8dba28b4e 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -149,6 +149,9 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 		
 		nodes[nodeid].start = base; 
 		nodes[nodeid].end = limit;
+		e820_register_active_regions(nodeid,
+				nodes[nodeid].start >> PAGE_SHIFT,
+				nodes[nodeid].end >> PAGE_SHIFT);
 
 		prevbase = base;
 
diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 322bf45fc36a..829a008bd39b 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -161,7 +161,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 					 bootmap_start >> PAGE_SHIFT, 
 					 start_pfn, end_pfn); 
 
-	e820_bootmem_free(NODE_DATA(nodeid), start, end);
+	free_bootmem_with_active_regions(nodeid, end);
 
 	reserve_bootmem_node(NODE_DATA(nodeid), nodedata_phys, pgdat_size); 
 	reserve_bootmem_node(NODE_DATA(nodeid), bootmap_start, bootmap_pages<<PAGE_SHIFT);
@@ -175,13 +175,11 @@ void __init setup_node_bootmem(int nodeid, unsigned long start, unsigned long en
 void __init setup_node_zones(int nodeid)
 { 
 	unsigned long start_pfn, end_pfn, memmapsize, limit;
-	unsigned long zones[MAX_NR_ZONES];
-	unsigned long holes[MAX_NR_ZONES];
 
  	start_pfn = node_start_pfn(nodeid);
  	end_pfn = node_end_pfn(nodeid);
 
-	Dprintk(KERN_INFO "Setting up node %d %lx-%lx\n",
+	Dprintk(KERN_INFO "Setting up memmap for node %d %lx-%lx\n",
 		nodeid, start_pfn, end_pfn);
 
 	/* Try to allocate mem_map at end to not fill up precious <4GB
@@ -195,10 +193,6 @@ void __init setup_node_zones(int nodeid)
 				round_down(limit - memmapsize, PAGE_SIZE), 
 				limit);
 #endif
-
-	size_zones(zones, holes, start_pfn, end_pfn);
-	free_area_init_node(nodeid, NODE_DATA(nodeid), zones,
-			    start_pfn, holes);
 } 
 
 void __init numa_init_array(void)
@@ -259,8 +253,11 @@ static int __init numa_emulation(unsigned long start_pfn, unsigned long end_pfn)
  		printk(KERN_ERR "No NUMA hash function found. Emulation disabled.\n");
  		return -1;
  	}
- 	for_each_online_node(i)
+ 	for_each_online_node(i) {
+		e820_register_active_regions(i, nodes[i].start >> PAGE_SHIFT,
+						nodes[i].end >> PAGE_SHIFT);
  		setup_node_bootmem(i, nodes[i].start, nodes[i].end);
+	}
  	numa_init_array();
  	return 0;
 }
@@ -299,6 +296,7 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
 	for (i = 0; i < NR_CPUS; i++)
 		numa_set_node(i, 0);
 	node_to_cpumask[0] = cpumask_of_cpu(0);
+	e820_register_active_regions(0, start_pfn, end_pfn);
 	setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT);
 }
 
@@ -340,12 +338,17 @@ static void __init arch_sparse_init(void)
 void __init paging_init(void)
 { 
 	int i;
+	unsigned long max_zone_pfns[MAX_NR_ZONES] = { MAX_DMA_PFN,
+		MAX_DMA32_PFN,
+		end_pfn};
 
 	arch_sparse_init();
 
 	for_each_online_node(i) {
 		setup_node_zones(i); 
 	}
+
+	free_area_init_nodes(max_zone_pfns);
 } 
 
 static __init int numa_setup(char *opt)
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index ca10701e7a90..7b50bb1caabe 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -93,6 +93,7 @@ static __init void bad_srat(void)
 		apicid_to_node[i] = NUMA_NO_NODE;
 	for (i = 0; i < MAX_NUMNODES; i++)
 		nodes_add[i].start = nodes[i].end = 0;
+	remove_all_active_ranges();
 }
 
 static __init inline int srat_disabled(void)
@@ -175,7 +176,7 @@ static int hotadd_enough_memory(struct bootnode *nd)
 
 	if (mem < 0)
 		return 0;
-	allowed = (end_pfn - e820_hole_size(0, end_pfn)) * PAGE_SIZE;
+	allowed = (end_pfn - absent_pages_in_range(0, end_pfn)) * PAGE_SIZE;
 	allowed = (allowed / 100) * hotadd_percent;
 	if (allocated + mem > allowed) {
 		unsigned long range;
@@ -225,7 +226,7 @@ static int reserve_hotadd(int node, unsigned long start, unsigned long end)
 	}
 
 	/* This check might be a bit too strict, but I'm keeping it for now. */
-	if (e820_hole_size(s_pfn, e_pfn) != e_pfn - s_pfn) {
+	if (absent_pages_in_range(s_pfn, e_pfn) != e_pfn - s_pfn) {
 		printk(KERN_ERR "SRAT: Hotplug area has existing memory\n");
 		return -1;
 	}
@@ -319,6 +320,8 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma)
 
 	printk(KERN_INFO "SRAT: Node %u PXM %u %Lx-%Lx\n", node, pxm,
 	       nd->start, nd->end);
+	e820_register_active_regions(node, nd->start >> PAGE_SHIFT,
+						nd->end >> PAGE_SHIFT);
 
 #ifdef RESERVE_HOTADD
  	if (ma->flags.hot_pluggable && reserve_hotadd(node, start, end) < 0) {
@@ -343,13 +346,13 @@ static int nodes_cover_memory(void)
 		unsigned long s = nodes[i].start >> PAGE_SHIFT;
 		unsigned long e = nodes[i].end >> PAGE_SHIFT;
 		pxmram += e - s;
-		pxmram -= e820_hole_size(s, e);
+		pxmram -= absent_pages_in_range(s, e);
 		pxmram -= nodes_add[i].end - nodes_add[i].start;
 		if ((long)pxmram < 0)
 			pxmram = 0;
 	}
 
-	e820ram = end_pfn - e820_hole_size(0, end_pfn);
+	e820ram = end_pfn - absent_pages_in_range(0, end_pfn);
 	/* We seem to lose 3 pages somewhere. Allow a bit of slack. */
 	if ((long)(e820ram - pxmram) >= 1*1024*1024) {
 		printk(KERN_ERR
diff --git a/include/asm-x86_64/e820.h b/include/asm-x86_64/e820.h
index e15d3c8628f3..fa2086774105 100644
--- a/include/asm-x86_64/e820.h
+++ b/include/asm-x86_64/e820.h
@@ -47,10 +47,9 @@ extern void e820_print_map(char *who);
 extern int e820_any_mapped(unsigned long start, unsigned long end, unsigned type);
 extern int e820_all_mapped(unsigned long start, unsigned long end, unsigned type);
 
-extern void e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end);
 extern void e820_setup_gap(void);
-extern unsigned long e820_hole_size(unsigned long start_pfn,
-				    unsigned long end_pfn);
+extern void e820_register_active_regions(int nid,
+				unsigned long start_pfn, unsigned long end_pfn);
 
 extern void finish_e820_parsing(void);
 
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index b73d0c76613c..c28fc2db2171 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -24,8 +24,6 @@ extern void mtrr_bp_init(void);
 #define mtrr_bp_init() do {} while (0)
 #endif
 extern void init_memory_mapping(unsigned long start, unsigned long end);
-extern void size_zones(unsigned long *z, unsigned long *h,
-			unsigned long start_pfn, unsigned long end_pfn);
 
 extern void system_call(void); 
 extern int kernel_syscall(void);
-- 
cgit v1.2.1


From 05e0caad3b7bd0d0fbeff980bca22f186241a501 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:54 -0700
Subject: [PATCH] Have ia64 use add_active_range() and free_area_init_nodes

Size zones and holes in an architecture independent manner for ia64.

[bob.picco@hp.com: fix ia64 FLATMEM+VIRTUAL_MEM_MAP]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/Kconfig          |  3 +++
 arch/ia64/mm/contig.c      | 67 +++++++++++-----------------------------------
 arch/ia64/mm/discontig.c   | 44 +++++++-----------------------
 arch/ia64/mm/init.c        | 12 +++++++++
 include/asm-ia64/meminit.h |  1 +
 5 files changed, 40 insertions(+), 87 deletions(-)

diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index f521f2f60a78..d5ee4fc8fe66 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -356,6 +356,9 @@ config NODES_SHIFT
 	  MAX_NUMNODES will be 2^(This value).
 	  If in doubt, use the default.
 
+config ARCH_POPULATES_NODE_MAP
+	def_bool y
+
 # VIRTUAL_MEM_MAP and FLAT_NODE_MEM_MAP are functionally equivalent.
 # VIRTUAL_MEM_MAP has been retained for historical reasons.
 config VIRTUAL_MEM_MAP
diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c
index e004143ba86b..719d476e71ba 100644
--- a/arch/ia64/mm/contig.c
+++ b/arch/ia64/mm/contig.c
@@ -26,7 +26,6 @@
 #include <asm/mca.h>
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
-static unsigned long num_dma_physpages;
 static unsigned long max_gap;
 #endif
 
@@ -218,18 +217,6 @@ count_pages (u64 start, u64 end, void *arg)
 	return 0;
 }
 
-#ifdef CONFIG_VIRTUAL_MEM_MAP
-static int
-count_dma_pages (u64 start, u64 end, void *arg)
-{
-	unsigned long *count = arg;
-
-	if (start < MAX_DMA_ADDRESS)
-		*count += (min(end, MAX_DMA_ADDRESS) - start) >> PAGE_SHIFT;
-	return 0;
-}
-#endif
-
 /*
  * Set up the page tables.
  */
@@ -238,45 +225,22 @@ void __init
 paging_init (void)
 {
 	unsigned long max_dma;
-	unsigned long zones_size[MAX_NR_ZONES];
-#ifdef CONFIG_VIRTUAL_MEM_MAP
-	unsigned long zholes_size[MAX_NR_ZONES];
-#endif
-
-	/* initialize mem_map[] */
-
-	memset(zones_size, 0, sizeof(zones_size));
+	unsigned long nid = 0;
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 	num_physpages = 0;
 	efi_memmap_walk(count_pages, &num_physpages);
 
 	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+	max_zone_pfns[ZONE_DMA] = max_dma;
+	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
-	memset(zholes_size, 0, sizeof(zholes_size));
-
-	num_dma_physpages = 0;
-	efi_memmap_walk(count_dma_pages, &num_dma_physpages);
-
-	if (max_low_pfn < max_dma) {
-		zones_size[ZONE_DMA] = max_low_pfn;
-		zholes_size[ZONE_DMA] = max_low_pfn - num_dma_physpages;
-	} else {
-		zones_size[ZONE_DMA] = max_dma;
-		zholes_size[ZONE_DMA] = max_dma - num_dma_physpages;
-		if (num_physpages > num_dma_physpages) {
-			zones_size[ZONE_NORMAL] = max_low_pfn - max_dma;
-			zholes_size[ZONE_NORMAL] =
-				((max_low_pfn - max_dma) -
-				 (num_physpages - num_dma_physpages));
-		}
-	}
-
+	efi_memmap_walk(register_active_ranges, &nid);
 	efi_memmap_walk(find_largest_hole, (u64 *)&max_gap);
 	if (max_gap < LARGE_GAP) {
 		vmem_map = (struct page *) 0;
-		free_area_init_node(0, NODE_DATA(0), zones_size, 0,
-				    zholes_size);
+		free_area_init_nodes(max_zone_pfns);
 	} else {
 		unsigned long map_size;
 
@@ -288,20 +252,19 @@ paging_init (void)
 		vmem_map = (struct page *) vmalloc_end;
 		efi_memmap_walk(create_mem_map_page_table, NULL);
 
-		NODE_DATA(0)->node_mem_map = vmem_map;
-		free_area_init_node(0, NODE_DATA(0), zones_size,
-				    0, zholes_size);
+		/*
+		 * alloc_node_mem_map makes an adjustment for mem_map
+		 * which isn't compatible with vmem_map.
+		 */
+		NODE_DATA(0)->node_mem_map = vmem_map +
+			find_min_pfn_with_active_regions();
+		free_area_init_nodes(max_zone_pfns);
 
 		printk("Virtual mem_map starts at 0x%p\n", mem_map);
 	}
 #else /* !CONFIG_VIRTUAL_MEM_MAP */
-	if (max_low_pfn < max_dma)
-		zones_size[ZONE_DMA] = max_low_pfn;
-	else {
-		zones_size[ZONE_DMA] = max_dma;
-		zones_size[ZONE_NORMAL] = max_low_pfn - max_dma;
-	}
-	free_area_init(zones_size);
+	add_active_range(0, 0, max_low_pfn);
+	free_area_init_nodes(max_zone_pfns);
 #endif /* !CONFIG_VIRTUAL_MEM_MAP */
 	zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page));
 }
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index d260bffa01ab..7bd28079dcc4 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -654,6 +654,7 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
 {
 	unsigned long end = start + len;
 
+	add_active_range(node, start >> PAGE_SHIFT, end >> PAGE_SHIFT);
 	mem_data[node].num_physpages += len >> PAGE_SHIFT;
 	if (start <= __pa(MAX_DMA_ADDRESS))
 		mem_data[node].num_dma_physpages +=
@@ -678,10 +679,10 @@ static __init int count_node_pages(unsigned long start, unsigned long len, int n
 void __init paging_init(void)
 {
 	unsigned long max_dma;
-	unsigned long zones_size[MAX_NR_ZONES];
-	unsigned long zholes_size[MAX_NR_ZONES];
 	unsigned long pfn_offset = 0;
+	unsigned long max_pfn = 0;
 	int node;
+	unsigned long max_zone_pfns[MAX_NR_ZONES];
 
 	max_dma = virt_to_phys((void *) MAX_DMA_ADDRESS) >> PAGE_SHIFT;
 
@@ -698,47 +699,20 @@ void __init paging_init(void)
 #endif
 
 	for_each_online_node(node) {
-		memset(zones_size, 0, sizeof(zones_size));
-		memset(zholes_size, 0, sizeof(zholes_size));
-
 		num_physpages += mem_data[node].num_physpages;
-
-		if (mem_data[node].min_pfn >= max_dma) {
-			/* All of this node's memory is above ZONE_DMA */
-			zones_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn -
-				mem_data[node].num_physpages;
-		} else if (mem_data[node].max_pfn < max_dma) {
-			/* All of this node's memory is in ZONE_DMA */
-			zones_size[ZONE_DMA] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_DMA] = mem_data[node].max_pfn -
-				mem_data[node].min_pfn -
-				mem_data[node].num_dma_physpages;
-		} else {
-			/* This node has memory in both zones */
-			zones_size[ZONE_DMA] = max_dma -
-				mem_data[node].min_pfn;
-			zholes_size[ZONE_DMA] = zones_size[ZONE_DMA] -
-				mem_data[node].num_dma_physpages;
-			zones_size[ZONE_NORMAL] = mem_data[node].max_pfn -
-				max_dma;
-			zholes_size[ZONE_NORMAL] = zones_size[ZONE_NORMAL] -
-				(mem_data[node].num_physpages -
-				 mem_data[node].num_dma_physpages);
-		}
-
 		pfn_offset = mem_data[node].min_pfn;
 
 #ifdef CONFIG_VIRTUAL_MEM_MAP
 		NODE_DATA(node)->node_mem_map = vmem_map + pfn_offset;
 #endif
-		free_area_init_node(node, NODE_DATA(node), zones_size,
-				    pfn_offset, zholes_size);
+		if (mem_data[node].max_pfn > max_pfn)
+			max_pfn = mem_data[node].max_pfn;
 	}
 
+	max_zone_pfns[ZONE_DMA] = max_dma;
+	max_zone_pfns[ZONE_NORMAL] = max_pfn;
+	free_area_init_nodes(max_zone_pfns);
+
 	zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page));
 }
 
diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c
index 30617ccb4f7e..ff87a5cba399 100644
--- a/arch/ia64/mm/init.c
+++ b/arch/ia64/mm/init.c
@@ -593,6 +593,18 @@ find_largest_hole (u64 start, u64 end, void *arg)
 	last_end = end;
 	return 0;
 }
+
+int __init
+register_active_ranges(u64 start, u64 end, void *nid)
+{
+	BUG_ON(nid == NULL);
+	BUG_ON(*(unsigned long *)nid >= MAX_NUMNODES);
+
+	add_active_range(*(unsigned long *)nid,
+				__pa(start) >> PAGE_SHIFT,
+				__pa(end) >> PAGE_SHIFT);
+	return 0;
+}
 #endif /* CONFIG_VIRTUAL_MEM_MAP */
 
 static int __init
diff --git a/include/asm-ia64/meminit.h b/include/asm-ia64/meminit.h
index 6a33a07b3f1d..c3b1f862e6e7 100644
--- a/include/asm-ia64/meminit.h
+++ b/include/asm-ia64/meminit.h
@@ -55,6 +55,7 @@ extern void efi_memmap_init(unsigned long *, unsigned long *);
   extern unsigned long vmalloc_end;
   extern struct page *vmem_map;
   extern int find_largest_hole (u64 start, u64 end, void *arg);
+  extern int register_active_ranges (u64 start, u64 end, void *arg);
   extern int create_mem_map_page_table (u64 start, u64 end, void *arg);
   extern int vmemmap_find_next_valid_pfn(int, int);
 #else
-- 
cgit v1.2.1


From 0e0b864e069c52a7b3e4a7da56e29b03a012fd75 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 27 Sep 2006 01:49:56 -0700
Subject: [PATCH] Account for memmap and optionally the kernel image as holes

The x86_64 code accounted for memmap and some portions of the the DMA zone as
holes.  This was because those areas would never be reclaimed and accounting
for them as memory affects min watermarks.  This patch will account for the
memmap as a memory hole.  Architectures may optionally use set_dma_reserve()
if they wish to account for a portion of memory in ZONE_DMA as a hole.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/init.c |  4 +++-
 include/linux/mm.h    |  1 +
 mm/page_alloc.c       | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 3 files changed, 63 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c
index 47928399e38a..3e16fe08150e 100644
--- a/arch/x86_64/mm/init.c
+++ b/arch/x86_64/mm/init.c
@@ -655,8 +655,10 @@ void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
 #else       		
 	reserve_bootmem(phys, len);    
 #endif
-	if (phys+len <= MAX_DMA_PFN*PAGE_SIZE)
+	if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) {
 		dma_reserve += len / PAGE_SIZE;
+		set_dma_reserve(dma_reserve);
+	}
 }
 
 int kern_addr_valid(unsigned long addr) 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index c0402da7cce0..22936e1fcdf2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -984,6 +984,7 @@ extern void sparse_memory_present_with_active_regions(int nid);
 extern int early_pfn_to_nid(unsigned long pfn);
 #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
+extern void set_dma_reserve(unsigned long new_dma_reserve);
 extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long);
 extern void setup_per_zone_pages_min(void);
 extern void mem_init(void);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 26c9939857fa..8d9a1eb9fbba 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -104,6 +104,7 @@ int min_free_kbytes = 1024;
 
 unsigned long __meminitdata nr_kernel_pages;
 unsigned long __meminitdata nr_all_pages;
+static unsigned long __initdata dma_reserve;
 
 #ifdef CONFIG_ARCH_POPULATES_NODE_MAP
   /*
@@ -2213,6 +2214,20 @@ unsigned long __init zone_absent_pages_in_node(int nid,
 				arch_zone_lowest_possible_pfn[zone_type],
 				arch_zone_highest_possible_pfn[zone_type]);
 }
+
+/* Return the zone index a PFN is in */
+int memmap_zone_idx(struct page *lmem_map)
+{
+	int i;
+	unsigned long phys_addr = virt_to_phys(lmem_map);
+	unsigned long pfn = phys_addr >> PAGE_SHIFT;
+
+	for (i = 0; i < MAX_NR_ZONES; i++)
+		if (pfn < arch_zone_highest_possible_pfn[i])
+			break;
+
+	return i;
+}
 #else
 static inline unsigned long zone_spanned_pages_in_node(int nid,
 					unsigned long zone_type,
@@ -2230,6 +2245,11 @@ static inline unsigned long zone_absent_pages_in_node(int nid,
 
 	return zholes_size[zone_type];
 }
+
+static inline int memmap_zone_idx(struct page *lmem_map)
+{
+	return MAX_NR_ZONES;
+}
 #endif
 
 static void __init calculate_node_totalpages(struct pglist_data *pgdat,
@@ -2274,12 +2294,35 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 	
 	for (j = 0; j < MAX_NR_ZONES; j++) {
 		struct zone *zone = pgdat->node_zones + j;
-		unsigned long size, realsize;
+		unsigned long size, realsize, memmap_pages;
 
 		size = zone_spanned_pages_in_node(nid, j, zones_size);
 		realsize = size - zone_absent_pages_in_node(nid, j,
 								zholes_size);
 
+		/*
+		 * Adjust realsize so that it accounts for how much memory
+		 * is used by this zone for memmap. This affects the watermark
+		 * and per-cpu initialisations
+		 */
+		memmap_pages = (size * sizeof(struct page)) >> PAGE_SHIFT;
+		if (realsize >= memmap_pages) {
+			realsize -= memmap_pages;
+			printk(KERN_DEBUG
+				"  %s zone: %lu pages used for memmap\n",
+				zone_names[j], memmap_pages);
+		} else
+			printk(KERN_WARNING
+				"  %s zone: %lu pages exceeds realsize %lu\n",
+				zone_names[j], memmap_pages, realsize);
+
+		/* Account for reserved DMA pages */
+		if (j == ZONE_DMA && realsize > dma_reserve) {
+			realsize -= dma_reserve;
+			printk(KERN_DEBUG "  DMA zone: %lu pages reserved\n",
+								dma_reserve);
+		}
+
 		if (!is_highmem_idx(j))
 			nr_kernel_pages += realsize;
 		nr_all_pages += realsize;
@@ -2596,6 +2639,21 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
 }
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
+/**
+ * set_dma_reserve - Account the specified number of pages reserved in ZONE_DMA
+ * @new_dma_reserve - The number of pages to mark reserved
+ *
+ * The per-cpu batchsize and zone watermarks are determined by present_pages.
+ * In the DMA zone, a significant percentage may be consumed by kernel image
+ * and other unfreeable allocations which can skew the watermarks badly. This
+ * function may optionally be used to account for unfreeable pages in
+ * ZONE_DMA. The effect will be lower watermarks and smaller per-cpu batchsize
+ */
+void __init set_dma_reserve(unsigned long new_dma_reserve)
+{
+	dma_reserve = new_dma_reserve;
+}
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 static bootmem_data_t contig_bootmem_data;
 struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
-- 
cgit v1.2.1


From 9c7cd6877cf8db15269163deda69392263124c1e Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@skynet.ie>
Date: Wed, 27 Sep 2006 01:49:58 -0700
Subject: [PATCH] Account for holes that are outside the range of physical
 memory

absent_pages_in_range() made the assumption that users of the API would not
care about holes beyound the end of physical memory.  This was not the
case.  This patch will account for ranges outside of physical memory as
holes correctly.

Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/srat.c |  4 +++-
 mm/page_alloc.c       | 22 +++++++++++++++++++---
 2 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 7b50bb1caabe..db1b2e11cf8f 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -227,7 +227,9 @@ static int reserve_hotadd(int node, unsigned long start, unsigned long end)
 
 	/* This check might be a bit too strict, but I'm keeping it for now. */
 	if (absent_pages_in_range(s_pfn, e_pfn) != e_pfn - s_pfn) {
-		printk(KERN_ERR "SRAT: Hotplug area has existing memory\n");
+		printk(KERN_ERR
+			"SRAT: Hotplug area %lu -> %lu has existing memory\n",
+			s_pfn, e_pfn);
 		return -1;
 	}
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8d9a1eb9fbba..75133e1dc4b1 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2168,6 +2168,10 @@ unsigned long __init __absent_pages_in_range(int nid,
 	if (i == -1)
 		return 0;
 
+	/* Account for ranges before physical memory on this node */
+	if (early_node_map[i].start_pfn > range_start_pfn)
+		hole_pages = early_node_map[i].start_pfn - range_start_pfn;
+
 	prev_end_pfn = early_node_map[i].start_pfn;
 
 	/* Find all holes for the zone within the node */
@@ -2189,6 +2193,11 @@ unsigned long __init __absent_pages_in_range(int nid,
 		prev_end_pfn = early_node_map[i].end_pfn;
 	}
 
+	/* Account for ranges past physical memory on this node */
+	if (range_end_pfn > prev_end_pfn)
+		hole_pages = range_end_pfn -
+				max(range_start_pfn, prev_end_pfn);
+
 	return hole_pages;
 }
 
@@ -2210,9 +2219,16 @@ unsigned long __init zone_absent_pages_in_node(int nid,
 					unsigned long zone_type,
 					unsigned long *ignored)
 {
-	return __absent_pages_in_range(nid,
-				arch_zone_lowest_possible_pfn[zone_type],
-				arch_zone_highest_possible_pfn[zone_type]);
+	unsigned long node_start_pfn, node_end_pfn;
+	unsigned long zone_start_pfn, zone_end_pfn;
+
+	get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn);
+	zone_start_pfn = max(arch_zone_lowest_possible_pfn[zone_type],
+							node_start_pfn);
+	zone_end_pfn = min(arch_zone_highest_possible_pfn[zone_type],
+							node_end_pfn);
+
+	return __absent_pages_in_range(nid, zone_start_pfn, zone_end_pfn);
 }
 
 /* Return the zone index a PFN is in */
-- 
cgit v1.2.1


From fb01439c5b778d5974a488c5d4fe85e6d0e18a68 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@skynet.ie>
Date: Wed, 27 Sep 2006 01:49:59 -0700
Subject: [PATCH] Allow an arch to expand node boundaries

Arch-independent zone-sizing determines the size of a node
(pgdat->node_spanned_pages) based on the physical memory that was
registered by the architecture.  However, when
CONFIG_MEMORY_HOTPLUG_RESERVE is set, the architecture expects that the
spanned_pages will be much larger and that mem_map will be allocated that
is used lated on memory hot-add.

This patch allows an architecture that sets CONFIG_MEMORY_HOTPLUG_RESERVE
to call push_node_boundaries() which will set the node beginning and end to
at *least* the requested boundary.

Cc: Dave Hansen <haveblue@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Andi Kleen <ak@muc.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Keith Mannthey" <kmannth@gmail.com>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/srat.c |  2 ++
 include/linux/mm.h    |  2 ++
 mm/page_alloc.c       | 67 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 71 insertions(+)

diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index db1b2e11cf8f..f8c04d6935c9 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -324,6 +324,8 @@ acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma)
 	       nd->start, nd->end);
 	e820_register_active_regions(node, nd->start >> PAGE_SHIFT,
 						nd->end >> PAGE_SHIFT);
+	push_node_boundaries(node, nd->start >> PAGE_SHIFT,
+						nd->end >> PAGE_SHIFT);
 
 #ifdef RESERVE_HOTADD
  	if (ma->flags.hot_pluggable && reserve_hotadd(node, start, end) < 0) {
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 22936e1fcdf2..9d046db31e76 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -970,6 +970,8 @@ extern void add_active_range(unsigned int nid, unsigned long start_pfn,
 					unsigned long end_pfn);
 extern void shrink_active_range(unsigned int nid, unsigned long old_end_pfn,
 						unsigned long new_end_pfn);
+extern void push_node_boundaries(unsigned int nid, unsigned long start_pfn,
+					unsigned long end_pfn);
 extern void remove_all_active_ranges(void);
 extern unsigned long absent_pages_in_range(unsigned long start_pfn,
 						unsigned long end_pfn);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 75133e1dc4b1..acbf58f8a1b7 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -131,6 +131,10 @@ static unsigned long __initdata dma_reserve;
   int __initdata nr_nodemap_entries;
   unsigned long __initdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES];
   unsigned long __initdata arch_zone_highest_possible_pfn[MAX_NR_ZONES];
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+  unsigned long __initdata node_boundary_start_pfn[MAX_NUMNODES];
+  unsigned long __initdata node_boundary_end_pfn[MAX_NUMNODES];
+#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
 #ifdef CONFIG_DEBUG_VM
@@ -2094,6 +2098,62 @@ void __init sparse_memory_present_with_active_regions(int nid)
 				early_node_map[i].end_pfn);
 }
 
+/**
+ * push_node_boundaries - Push node boundaries to at least the requested boundary
+ * @nid: The nid of the node to push the boundary for
+ * @start_pfn: The start pfn of the node
+ * @end_pfn: The end pfn of the node
+ *
+ * In reserve-based hot-add, mem_map is allocated that is unused until hotadd
+ * time. Specifically, on x86_64, SRAT will report ranges that can potentially
+ * be hotplugged even though no physical memory exists. This function allows
+ * an arch to push out the node boundaries so mem_map is allocated that can
+ * be used later.
+ */
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+void __init push_node_boundaries(unsigned int nid,
+		unsigned long start_pfn, unsigned long end_pfn)
+{
+	printk(KERN_DEBUG "Entering push_node_boundaries(%u, %lu, %lu)\n",
+			nid, start_pfn, end_pfn);
+
+	/* Initialise the boundary for this node if necessary */
+	if (node_boundary_end_pfn[nid] == 0)
+		node_boundary_start_pfn[nid] = -1UL;
+
+	/* Update the boundaries */
+	if (node_boundary_start_pfn[nid] > start_pfn)
+		node_boundary_start_pfn[nid] = start_pfn;
+	if (node_boundary_end_pfn[nid] < end_pfn)
+		node_boundary_end_pfn[nid] = end_pfn;
+}
+
+/* If necessary, push the node boundary out for reserve hotadd */
+static void __init account_node_boundary(unsigned int nid,
+		unsigned long *start_pfn, unsigned long *end_pfn)
+{
+	printk(KERN_DEBUG "Entering account_node_boundary(%u, %lu, %lu)\n",
+			nid, *start_pfn, *end_pfn);
+
+	/* Return if boundary information has not been provided */
+	if (node_boundary_end_pfn[nid] == 0)
+		return;
+
+	/* Check the boundaries and update if necessary */
+	if (node_boundary_start_pfn[nid] < *start_pfn)
+		*start_pfn = node_boundary_start_pfn[nid];
+	if (node_boundary_end_pfn[nid] > *end_pfn)
+		*end_pfn = node_boundary_end_pfn[nid];
+}
+#else
+void __init push_node_boundaries(unsigned int nid,
+		unsigned long start_pfn, unsigned long end_pfn) {}
+
+static void __init account_node_boundary(unsigned int nid,
+		unsigned long *start_pfn, unsigned long *end_pfn) {}
+#endif
+
+
 /**
  * get_pfn_range_for_nid - Return the start and end page frames for a node
  * @nid: The nid to return the range for. If MAX_NUMNODES, the min and max PFN are returned
@@ -2121,6 +2181,9 @@ void __init get_pfn_range_for_nid(unsigned int nid,
 		printk(KERN_WARNING "Node %u active with no memory\n", nid);
 		*start_pfn = 0;
 	}
+
+	/* Push the node boundaries out if requested */
+	account_node_boundary(nid, start_pfn, end_pfn);
 }
 
 /*
@@ -2527,6 +2590,10 @@ void __init remove_all_active_ranges()
 {
 	memset(early_node_map, 0, sizeof(early_node_map));
 	nr_nodemap_entries = 0;
+#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
+	memset(node_boundary_start_pfn, 0, sizeof(node_boundary_start_pfn));
+	memset(node_boundary_end_pfn, 0, sizeof(node_boundary_end_pfn));
+#endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
 }
 
 /* Compare two active node_active_regions */
-- 
cgit v1.2.1


From e129b5c23c2b471d47f1c5d2b8b193fc2034af43 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:50:00 -0700
Subject: [PATCH] vm: add per-zone writeout counter

The VM is supposed to minimise the number of pages which get written off the
LRU (for IO scheduling efficiency, and for high reclaim-success rates).  But
we don't actually have a clear way of showing how true this is.

So add `nr_vmscan_write' to /proc/vmstat and /proc/zoneinfo - the number of
pages which have been written by the vm scanner in this zone and globally.

Cc: Christoph Lameter <clameter@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mmzone.h | 1 +
 mm/vmscan.c            | 3 ++-
 mm/vmstat.c            | 1 +
 3 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 7fa1cbe9fa7a..1b0680cd84d2 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -58,6 +58,7 @@ enum zone_stat_item {
 	NR_WRITEBACK,
 	NR_UNSTABLE_NFS,	/* NFS unstable pages */
 	NR_BOUNCE,
+	NR_VMSCAN_WRITE,
 #ifdef CONFIG_NUMA
 	NUMA_HIT,		/* allocated in intended node */
 	NUMA_MISS,		/* allocated in non intended node */
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 87779dda4ec6..87d340999ce8 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -19,6 +19,7 @@
 #include <linux/pagemap.h>
 #include <linux/init.h>
 #include <linux/highmem.h>
+#include <linux/vmstat.h>
 #include <linux/file.h>
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
@@ -370,7 +371,7 @@ static pageout_t pageout(struct page *page, struct address_space *mapping)
 			/* synchronous write or broken a_ops? */
 			ClearPageReclaim(page);
 		}
-
+		inc_zone_page_state(page, NR_VMSCAN_WRITE);
 		return PAGE_SUCCESS;
 	}
 
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 490d8c1a0ded..69c657132e1f 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -465,6 +465,7 @@ static char *vmstat_text[] = {
 	"nr_writeback",
 	"nr_unstable",
 	"nr_bounce",
+	"nr_vmscan_write",
 
 #ifdef CONFIG_NUMA
 	"numa_hit",
-- 
cgit v1.2.1


From 5b99cd0effaf846240a15441aec459a592577eaf Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Wed, 27 Sep 2006 01:50:01 -0700
Subject: [PATCH] own header file for struct page

This moves the definition of struct page from mm.h to its own header file
page-struct.h.  This is a prereq to fix SetPageUptodate which is broken on
s390:

#define SetPageUptodate(_page)
       do {
               struct page *__page = (_page);
               if (!test_and_set_bit(PG_uptodate, &__page->flags))
                       page_test_and_clear_dirty(_page);
       } while (0)

_page gets used twice in this macro which can cause subtle bugs.  Using
__page for the page_test_and_clear_dirty call doesn't work since it causes
yet another problem with the page_test_and_clear_dirty macro as well.

In order to avoid all these problems caused by macros it seems to be a good
idea to get rid of them and convert them to static inline functions.
Because of header file include order it's necessary to have a seperate
header file for the struct page definition.

Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h       | 62 +-------------------------------------------
 include/linux/mm_types.h | 67 ++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/mmzone.h   |  5 ++++
 3 files changed, 73 insertions(+), 61 deletions(-)
 create mode 100644 include/linux/mm_types.h

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 9d046db31e76..7477fb59c4f2 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -16,6 +16,7 @@
 #include <linux/mutex.h>
 #include <linux/debug_locks.h>
 #include <linux/backing-dev.h>
+#include <linux/mm_types.h>
 
 struct mempolicy;
 struct anon_vma;
@@ -215,62 +216,6 @@ struct vm_operations_struct {
 struct mmu_gather;
 struct inode;
 
-/*
- * Each physical page in the system has a struct page associated with
- * it to keep track of whatever it is we are using the page for at the
- * moment. Note that we have no way to track which tasks are using
- * a page, though if it is a pagecache page, rmap structures can tell us
- * who is mapping it.
- */
-struct page {
-	unsigned long flags;		/* Atomic flags, some possibly
-					 * updated asynchronously */
-	atomic_t _count;		/* Usage count, see below. */
-	atomic_t _mapcount;		/* Count of ptes mapped in mms,
-					 * to show when page is mapped
-					 * & limit reverse map searches.
-					 */
-	union {
-	    struct {
-		unsigned long private;		/* Mapping-private opaque data:
-					 	 * usually used for buffer_heads
-						 * if PagePrivate set; used for
-						 * swp_entry_t if PageSwapCache;
-						 * indicates order in the buddy
-						 * system if PG_buddy is set.
-						 */
-		struct address_space *mapping;	/* If low bit clear, points to
-						 * inode address_space, or NULL.
-						 * If page mapped as anonymous
-						 * memory, low bit is set, and
-						 * it points to anon_vma object:
-						 * see PAGE_MAPPING_ANON below.
-						 */
-	    };
-#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
-	    spinlock_t ptl;
-#endif
-	};
-	pgoff_t index;			/* Our offset within mapping. */
-	struct list_head lru;		/* Pageout list, eg. active_list
-					 * protected by zone->lru_lock !
-					 */
-	/*
-	 * On machines where all RAM is mapped into kernel address space,
-	 * we can simply calculate the virtual address. On machines with
-	 * highmem some memory is mapped into kernel virtual memory
-	 * dynamically, so we need a place to store that address.
-	 * Note that this field could be 16 bits on x86 ... ;)
-	 *
-	 * Architectures with slow multiplication can define
-	 * WANT_PAGE_VIRTUAL in asm/page.h
-	 */
-#if defined(WANT_PAGE_VIRTUAL)
-	void *virtual;			/* Kernel virtual address (NULL if
-					   not kmapped, ie. highmem) */
-#endif /* WANT_PAGE_VIRTUAL */
-};
-
 #define page_private(page)		((page)->private)
 #define set_page_private(page, v)	((page)->private = (v))
 
@@ -546,11 +491,6 @@ static inline void set_page_links(struct page *page, enum zone_type zone,
  */
 #include <linux/vmstat.h>
 
-#ifndef CONFIG_DISCONTIGMEM
-/* The array of struct pages - for discontigmem use pgdat->lmem_map */
-extern struct page *mem_map;
-#endif
-
 static __always_inline void *lowmem_page_address(struct page *page)
 {
 	return __va(page_to_pfn(page) << PAGE_SHIFT);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
new file mode 100644
index 000000000000..c3852fd4a1cc
--- /dev/null
+++ b/include/linux/mm_types.h
@@ -0,0 +1,67 @@
+#ifndef _LINUX_MM_TYPES_H
+#define _LINUX_MM_TYPES_H
+
+#include <linux/types.h>
+#include <linux/threads.h>
+#include <linux/list.h>
+#include <linux/spinlock.h>
+
+struct address_space;
+
+/*
+ * Each physical page in the system has a struct page associated with
+ * it to keep track of whatever it is we are using the page for at the
+ * moment. Note that we have no way to track which tasks are using
+ * a page, though if it is a pagecache page, rmap structures can tell us
+ * who is mapping it.
+ */
+struct page {
+	unsigned long flags;		/* Atomic flags, some possibly
+					 * updated asynchronously */
+	atomic_t _count;		/* Usage count, see below. */
+	atomic_t _mapcount;		/* Count of ptes mapped in mms,
+					 * to show when page is mapped
+					 * & limit reverse map searches.
+					 */
+	union {
+	    struct {
+		unsigned long private;		/* Mapping-private opaque data:
+					 	 * usually used for buffer_heads
+						 * if PagePrivate set; used for
+						 * swp_entry_t if PageSwapCache;
+						 * indicates order in the buddy
+						 * system if PG_buddy is set.
+						 */
+		struct address_space *mapping;	/* If low bit clear, points to
+						 * inode address_space, or NULL.
+						 * If page mapped as anonymous
+						 * memory, low bit is set, and
+						 * it points to anon_vma object:
+						 * see PAGE_MAPPING_ANON below.
+						 */
+	    };
+#if NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS
+	    spinlock_t ptl;
+#endif
+	};
+	pgoff_t index;			/* Our offset within mapping. */
+	struct list_head lru;		/* Pageout list, eg. active_list
+					 * protected by zone->lru_lock !
+					 */
+	/*
+	 * On machines where all RAM is mapped into kernel address space,
+	 * we can simply calculate the virtual address. On machines with
+	 * highmem some memory is mapped into kernel virtual memory
+	 * dynamically, so we need a place to store that address.
+	 * Note that this field could be 16 bits on x86 ... ;)
+	 *
+	 * Architectures with slow multiplication can define
+	 * WANT_PAGE_VIRTUAL in asm/page.h
+	 */
+#if defined(WANT_PAGE_VIRTUAL)
+	void *virtual;			/* Kernel virtual address (NULL if
+					   not kmapped, ie. highmem) */
+#endif /* WANT_PAGE_VIRTUAL */
+};
+
+#endif /* _LINUX_MM_TYPES_H */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 1b0680cd84d2..562cf7a8f3ee 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -314,6 +314,11 @@ struct node_active_region {
 };
 #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */
 
+#ifndef CONFIG_DISCONTIGMEM
+/* The array of struct pages - for discontigmem use pgdat->lmem_map */
+extern struct page *mem_map;
+#endif
+
 /*
  * The pg_data_t structure is used in machines with CONFIG_DISCONTIGMEM
  * (mostly NUMA machines?) to denote a higher-level memory zone than the
-- 
cgit v1.2.1


From 0fd0e6b05aa096622f151cac2f81f2e6844fb1bb Mon Sep 17 00:00:00 2001
From: Nick Piggin <nickpiggin@yahoo.com.au>
Date: Wed, 27 Sep 2006 01:50:02 -0700
Subject: [PATCH] page invalidation cleanup

Clean up the invalidate code, and use a common function to safely remove
the page from pagecache.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/truncate.c | 25 ++++++++-----------------
 mm/vmscan.c   | 27 +++++++++++++++++++++++----
 2 files changed, 31 insertions(+), 21 deletions(-)

diff --git a/mm/truncate.c b/mm/truncate.c
index c6ab55ec6883..a654928323dc 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -9,6 +9,7 @@
 
 #include <linux/kernel.h>
 #include <linux/mm.h>
+#include <linux/swap.h>
 #include <linux/module.h>
 #include <linux/pagemap.h>
 #include <linux/pagevec.h>
@@ -52,36 +53,26 @@ truncate_complete_page(struct address_space *mapping, struct page *page)
 /*
  * This is for invalidate_inode_pages().  That function can be called at
  * any time, and is not supposed to throw away dirty pages.  But pages can
- * be marked dirty at any time too.  So we re-check the dirtiness inside
- * ->tree_lock.  That provides exclusion against the __set_page_dirty
- * functions.
+ * be marked dirty at any time too, so use remove_mapping which safely
+ * discards clean, unused pages.
  *
  * Returns non-zero if the page was successfully invalidated.
  */
 static int
 invalidate_complete_page(struct address_space *mapping, struct page *page)
 {
+	int ret;
+
 	if (page->mapping != mapping)
 		return 0;
 
 	if (PagePrivate(page) && !try_to_release_page(page, 0))
 		return 0;
 
-	write_lock_irq(&mapping->tree_lock);
-	if (PageDirty(page))
-		goto failed;
-	if (page_count(page) != 2)	/* caller's ref + pagecache ref */
-		goto failed;
-
-	BUG_ON(PagePrivate(page));
-	__remove_from_page_cache(page);
-	write_unlock_irq(&mapping->tree_lock);
+	ret = remove_mapping(mapping, page);
 	ClearPageUptodate(page);
-	page_cache_release(page);	/* pagecache ref */
-	return 1;
-failed:
-	write_unlock_irq(&mapping->tree_lock);
-	return 0;
+
+	return ret;
 }
 
 /**
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 87d340999ce8..eca70310adb2 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -384,11 +384,30 @@ int remove_mapping(struct address_space *mapping, struct page *page)
 	BUG_ON(mapping != page_mapping(page));
 
 	write_lock_irq(&mapping->tree_lock);
-
 	/*
-	 * The non-racy check for busy page.  It is critical to check
-	 * PageDirty _after_ making sure that the page is freeable and
-	 * not in use by anybody. 	(pagecache + us == 2)
+	 * The non racy check for a busy page.
+	 *
+	 * Must be careful with the order of the tests. When someone has
+	 * a ref to the page, it may be possible that they dirty it then
+	 * drop the reference. So if PageDirty is tested before page_count
+	 * here, then the following race may occur:
+	 *
+	 * get_user_pages(&page);
+	 * [user mapping goes away]
+	 * write_to(page);
+	 *				!PageDirty(page)    [good]
+	 * SetPageDirty(page);
+	 * put_page(page);
+	 *				!page_count(page)   [good, discard it]
+	 *
+	 * [oops, our write_to data is lost]
+	 *
+	 * Reversing the order of the tests ensures such a situation cannot
+	 * escape unnoticed. The smp_rmb is needed to ensure the page->flags
+	 * load is not satisfied before that of page->_count.
+	 *
+	 * Note that if SetPageDirty is always performed via set_page_dirty,
+	 * and thus under tree_lock, then this ordering is not required.
 	 */
 	if (unlikely(page_count(page) != 2))
 		goto cannot_free;
-- 
cgit v1.2.1


From de3083ec3e6bfb1ab60bc8a410f37702529f953c Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:03 -0700
Subject: [PATCH] slab: fix kmalloc_node applying memory policies if nodeid ==
 numa_node_id()

kmalloc_node() falls back to ___cache_alloc() under certain conditions and
at that point memory policies may be applied redirecting the allocation
away from the current node.  Therefore kmalloc_node(...,numa_node_id()) or
kmalloc_node(...,-1) may not return memory from the local node.

Fix this by doing the policy check in __cache_alloc() instead of
____cache_alloc().

This version here is a cleanup of Kiran's patch.

- Tested on ia64.
- Extra material removed.
- Consolidate the exit path if alternate_node_alloc() returned an object.

[akpm@osdl.org: warning fix]
Signed-off-by: Alok N Kataria <alok.kataria@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index c52ebf9c4462..69e11c45002f 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -3028,14 +3028,6 @@ static inline void *____cache_alloc(struct kmem_cache *cachep, gfp_t flags)
 	void *objp;
 	struct array_cache *ac;
 
-#ifdef CONFIG_NUMA
-	if (unlikely(current->flags & (PF_SPREAD_SLAB | PF_MEMPOLICY))) {
-		objp = alternate_node_alloc(cachep, flags);
-		if (objp != NULL)
-			return objp;
-	}
-#endif
-
 	check_irq_off();
 	ac = cpu_cache_get(cachep);
 	if (likely(ac->avail)) {
@@ -3053,12 +3045,19 @@ static __always_inline void *__cache_alloc(struct kmem_cache *cachep,
 						gfp_t flags, void *caller)
 {
 	unsigned long save_flags;
-	void *objp;
+	void *objp = NULL;
 
 	cache_alloc_debugcheck_before(cachep, flags);
 
 	local_irq_save(save_flags);
-	objp = ____cache_alloc(cachep, flags);
+
+#ifdef CONFIG_NUMA
+	if (unlikely(current->flags & (PF_SPREAD_SLAB | PF_MEMPOLICY)))
+		objp = alternate_node_alloc(cachep, flags);
+#endif
+
+	if (!objp)
+		objp = ____cache_alloc(cachep, flags);
 	local_irq_restore(save_flags);
 	objp = cache_alloc_debugcheck_after(cachep, flags, objp,
 					    caller);
-- 
cgit v1.2.1


From c72419138fa34e1bc1f1c6fa54ee77df55a05ed0 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 27 Sep 2006 01:50:05 -0700
Subject: [PATCH] Condense output of show_free_areas()

On larger systems, the amount of output dumped on the console when you do
SysRq-M is beyond insane.  This patch is trying to reduce it somewhat as
even with the smaller NUMA systems that have hit the desktop this seems to
be a fair thing to do.

The philosophy I have taken is as follows:
 1) If a zone is empty, don't tell, we don't need yet another line
    telling us so. The information is available since one can look up
    the fact how many zones were initialized in the first place.
 2) Put as much information on a line is possible, if it can be done
    in one line, rahter than two, then do it in one. I tried to format
    the temperature stuff for easy reading.

Change show_free_areas() to not print lines for empty zones.  If no zone
output is printed, the zone is empty.  This reduces the number of lines
dumped to the console in sysrq on a large system by several thousand lines.

Change the zone temperature printouts to use one line per CPU instead of
two lines (one hot, one cold).  On a 1024 CPU, 1024 node system, this
reduces the console output by over a million lines of output.

While this is a bigger problem on large NUMA systems, it is also applicable
to smaller desktop sized and mid range NUMA systems.

Old format:

Mem-info:
Node 0 DMA per-cpu:
cpu 0 hot: high 42, batch 7 used:24
cpu 0 cold: high 14, batch 3 used:1
cpu 1 hot: high 42, batch 7 used:34
cpu 1 cold: high 14, batch 3 used:0
cpu 2 hot: high 42, batch 7 used:0
cpu 2 cold: high 14, batch 3 used:0
cpu 3 hot: high 42, batch 7 used:0
cpu 3 cold: high 14, batch 3 used:0
cpu 4 hot: high 42, batch 7 used:0
cpu 4 cold: high 14, batch 3 used:0
cpu 5 hot: high 42, batch 7 used:0
cpu 5 cold: high 14, batch 3 used:0
cpu 6 hot: high 42, batch 7 used:0
cpu 6 cold: high 14, batch 3 used:0
cpu 7 hot: high 42, batch 7 used:0
cpu 7 cold: high 14, batch 3 used:0
Node 0 DMA32 per-cpu: empty
Node 0 Normal per-cpu: empty
Node 0 HighMem per-cpu: empty
Node 1 DMA per-cpu:
[snip]
Free pages:     5410688kB (0kB HighMem)
Active:9536 inactive:4261 dirty:6 writeback:0 unstable:0 free:338168 slab:1931 mapped:1900 pagetables:208
Node 0 DMA free:1676304kB min:3264kB low:4080kB high:4896kB active:128048kB inactive:61568kB present:1970880kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 0 DMA32 free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 0 Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 0 HighMem free:0kB min:512kB low:512kB high:512kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 1 DMA free:1951728kB min:3280kB low:4096kB high:4912kB active:5632kB inactive:1504kB present:1982464kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
....

New format:

Mem-info:
Node 0 DMA per-cpu:
CPU    0: Hot: hi:   42, btch:   7 usd:  41   Cold: hi:   14, btch:   3 usd:   2
CPU    1: Hot: hi:   42, btch:   7 usd:  40   Cold: hi:   14, btch:   3 usd:   1
CPU    2: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
CPU    3: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
CPU    4: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
CPU    5: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
CPU    6: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
CPU    7: Hot: hi:   42, btch:   7 usd:   0   Cold: hi:   14, btch:   3 usd:   0
Node 1 DMA per-cpu:
[snip]
Free pages:     5411088kB (0kB HighMem)
Active:9558 inactive:4233 dirty:6 writeback:0 unstable:0 free:338193 slab:1942 mapped:1918 pagetables:208
Node 0 DMA free:1677648kB min:3264kB low:4080kB high:4896kB active:129296kB inactive:58864kB present:1970880kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
Node 1 DMA free:1948448kB min:3280kB low:4096kB high:4912kB active:6864kB inactive:3536kB present:1982464kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0

Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 36 +++++++++++++++++-------------------
 1 file changed, 17 insertions(+), 19 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index acbf58f8a1b7..8a6418b0d2c5 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1305,34 +1305,30 @@ void si_meminfo_node(struct sysinfo *val, int nid)
  */
 void show_free_areas(void)
 {
-	int cpu, temperature;
+	int cpu;
 	unsigned long active;
 	unsigned long inactive;
 	unsigned long free;
 	struct zone *zone;
 
 	for_each_zone(zone) {
-		show_node(zone);
-		printk("%s per-cpu:", zone->name);
-
-		if (!populated_zone(zone)) {
-			printk(" empty\n");
+		if (!populated_zone(zone))
 			continue;
-		} else
-			printk("\n");
+
+		show_node(zone);
+		printk("%s per-cpu:\n", zone->name);
 
 		for_each_online_cpu(cpu) {
 			struct per_cpu_pageset *pageset;
 
 			pageset = zone_pcp(zone, cpu);
 
-			for (temperature = 0; temperature < 2; temperature++)
-				printk("cpu %d %s: high %d, batch %d used:%d\n",
-					cpu,
-					temperature ? "cold" : "hot",
-					pageset->pcp[temperature].high,
-					pageset->pcp[temperature].batch,
-					pageset->pcp[temperature].count);
+			printk("CPU %4d: Hot: hi:%5d, btch:%4d usd:%4d   "
+			       "Cold: hi:%5d, btch:%4d usd:%4d\n",
+			       cpu, pageset->pcp[0].high,
+			       pageset->pcp[0].batch, pageset->pcp[0].count,
+			       pageset->pcp[1].high, pageset->pcp[1].batch,
+			       pageset->pcp[1].count);
 		}
 	}
 
@@ -1354,6 +1350,9 @@ void show_free_areas(void)
 	for_each_zone(zone) {
 		int i;
 
+		if (!populated_zone(zone))
+			continue;
+
 		show_node(zone);
 		printk("%s"
 			" free:%lukB"
@@ -1386,12 +1385,11 @@ void show_free_areas(void)
 	for_each_zone(zone) {
  		unsigned long nr[MAX_ORDER], flags, order, total = 0;
 
+		if (!populated_zone(zone))
+			continue;
+
 		show_node(zone);
 		printk("%s: ", zone->name);
-		if (!populated_zone(zone)) {
-			printk("empty\n");
-			continue;
-		}
 
 		spin_lock_irqsave(&zone->lock, flags);
 		for (order = 0; order < MAX_ORDER; order++) {
-- 
cgit v1.2.1


From 08e0f6a9705376732fd3bc9bf8ba97a6b5211eb1 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:06 -0700
Subject: [PATCH] Add NUMA_BUILD definition in kernel.h to avoid #ifdef
 CONFIG_NUMA

The NUMA_BUILD constant is always available and will be set to 1 on
NUMA_BUILDs.  That way checks valid only under CONFIG_NUMA can easily be done
without #ifdef CONFIG_NUMA

F.e.

if (NUMA_BUILD && <numa_condition>) {
...
}

[akpm: not a thing we'd normally do, but CONFIG_NUMA is special: it is
 causing ifdef explosion in core kernel, so let's see if this is a comfortable
 way in whcih to control that]

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kernel.h |  7 +++++++
 mm/page_alloc.c        | 12 +++++-------
 2 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 4fa373bb18ac..4d00988dad03 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -350,4 +350,11 @@ struct sysinfo {
 /* Trap pasters of __FUNCTION__ at compile-time */
 #define __FUNCTION__ (__func__)
 
+/* This helps us to avoid #ifdef CONFIG_NUMA */
+#ifdef CONFIG_NUMA
+#define NUMA_BUILD 1
+#else
+#define NUMA_BUILD 0
+#endif
+
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8a6418b0d2c5..4c76188b1681 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -942,7 +942,7 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order,
 	 */
 	do {
 		zone = *z;
-		if (unlikely((gfp_mask & __GFP_THISNODE) &&
+		if (unlikely(NUMA_BUILD && (gfp_mask & __GFP_THISNODE) &&
 			zone->zone_pgdat != zonelist->zones[0]->zone_pgdat))
 				break;
 		if ((alloc_flags & ALLOC_CPUSET) &&
@@ -1256,14 +1256,12 @@ unsigned int nr_free_pagecache_pages(void)
 {
 	return nr_free_zone_pages(gfp_zone(GFP_HIGHUSER));
 }
-#ifdef CONFIG_NUMA
-static void show_node(struct zone *zone)
+
+static inline void show_node(struct zone *zone)
 {
-	printk("Node %ld ", zone_to_nid(zone));
+	if (NUMA_BUILD)
+		printk("Node %ld ", zone_to_nid(zone));
 }
-#else
-#define show_node(zone)	do { } while (0)
-#endif
 
 void si_meminfo(struct sysinfo *val)
 {
-- 
cgit v1.2.1


From 77f700dab4c05f8ee17584ec869672796d7bcb87 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:07 -0700
Subject: [PATCH] Disable GFP_THISNODE in the non-NUMA case

GFP_THISNODE must be set to 0 in the non numa case otherwise we disable retry
and warnings for failing allocations in the SMP and UP case.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/gfp.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 8b34aabfe4c6..bf2b6bc3f6fd 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -67,7 +67,12 @@ struct vm_area_struct;
 #define GFP_HIGHUSER	(__GFP_WAIT | __GFP_IO | __GFP_FS | __GFP_HARDWALL | \
 			 __GFP_HIGHMEM)
 
+#ifdef CONFIG_NUMA
 #define GFP_THISNODE	(__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
+#else
+#define GFP_THISNODE	0
+#endif
+
 
 /* Flag - indicates that the buffer will be suitable for DMA.  Ignored on some
    platforms, used as appropriate on others */
-- 
cgit v1.2.1


From 765c4507af71c39aba21006bbd3ec809fe9714ff Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:08 -0700
Subject: [PATCH] GFP_THISNODE for the slab allocator

This patch insures that the slab node lists in the NUMA case only contain
slabs that belong to that specific node.  All slab allocations use
GFP_THISNODE when calling into the page allocator.  If an allocation fails
then we fall back in the slab allocator according to the zonelists appropriate
for a certain context.

This allows a replication of the behavior of alloc_pages and alloc_pages node
in the slab layer.

Currently allocations requested from the page allocator may be redirected via
cpusets to other nodes.  This results in remote pages on nodelists and that in
turn results in interrupt latency issues during cache draining.  Plus the slab
is handing out memory as local when it is really remote.

Fallback for slab memory allocations will occur within the slab allocator and
not in the page allocator.  This is necessary in order to be able to use the
existing pools of objects on the nodes that we fall back to before adding more
pages to a slab.

The fallback function insures that the nodes we fall back to obey cpuset
restrictions of the current context.  We do not allocate objects from outside
of the current cpuset context like before.

Note that the implementation of locality constraints within the slab allocator
requires importing logic from the page allocator.  This is a mischmash that is
not that great.  Other allocators (uncached allocator, vmalloc, huge pages)
face similar problems and have similar minimal reimplementations of the basic
fallback logic of the page allocator.  There is another way of implementing a
slab by avoiding per node lists (see modular slab) but this wont work within
the existing slab.

V1->V2:
- Use NUMA_BUILD to avoid #ifdef CONFIG_NUMA
- Exploit GFP_THISNODE being 0 in the NON_NUMA case to avoid another
  #ifdef

[akpm@osdl.org: build fix]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c |   4 ++-
 mm/slab.c      | 107 +++++++++++++++++++++++++++++++++++++++++----------------
 2 files changed, 81 insertions(+), 30 deletions(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 38f89650bc84..cf18f0942553 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1136,7 +1136,9 @@ static unsigned interleave_nodes(struct mempolicy *policy)
  */
 unsigned slab_node(struct mempolicy *policy)
 {
-	switch (policy->policy) {
+	int pol = policy ? policy->policy : MPOL_DEFAULT;
+
+	switch (pol) {
 	case MPOL_INTERLEAVE:
 		return interleave_nodes(policy);
 
diff --git a/mm/slab.c b/mm/slab.c
index 69e11c45002f..792bfe320a8b 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -972,7 +972,39 @@ static int transfer_objects(struct array_cache *to,
 	return nr;
 }
 
-#ifdef CONFIG_NUMA
+#ifndef CONFIG_NUMA
+
+#define drain_alien_cache(cachep, alien) do { } while (0)
+#define reap_alien(cachep, l3) do { } while (0)
+
+static inline struct array_cache **alloc_alien_cache(int node, int limit)
+{
+	return (struct array_cache **)BAD_ALIEN_MAGIC;
+}
+
+static inline void free_alien_cache(struct array_cache **ac_ptr)
+{
+}
+
+static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
+{
+	return 0;
+}
+
+static inline void *alternate_node_alloc(struct kmem_cache *cachep,
+		gfp_t flags)
+{
+	return NULL;
+}
+
+static inline void *__cache_alloc_node(struct kmem_cache *cachep,
+		 gfp_t flags, int nodeid)
+{
+	return NULL;
+}
+
+#else	/* CONFIG_NUMA */
+
 static void *__cache_alloc_node(struct kmem_cache *, gfp_t, int);
 static void *alternate_node_alloc(struct kmem_cache *, gfp_t);
 
@@ -1101,26 +1133,6 @@ static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
 	}
 	return 1;
 }
-
-#else
-
-#define drain_alien_cache(cachep, alien) do { } while (0)
-#define reap_alien(cachep, l3) do { } while (0)
-
-static inline struct array_cache **alloc_alien_cache(int node, int limit)
-{
-	return (struct array_cache **)BAD_ALIEN_MAGIC;
-}
-
-static inline void free_alien_cache(struct array_cache **ac_ptr)
-{
-}
-
-static inline int cache_free_alien(struct kmem_cache *cachep, void *objp)
-{
-	return 0;
-}
-
 #endif
 
 static int __cpuinit cpuup_callback(struct notifier_block *nfb,
@@ -1564,7 +1576,13 @@ static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 	 */
 	flags |= __GFP_COMP;
 #endif
-	flags |= cachep->gfpflags;
+
+	/*
+	 * Under NUMA we want memory on the indicated node. We will handle
+	 * the needed fallback ourselves since we want to serve from our
+	 * per node object lists first for other nodes.
+	 */
+	flags |= cachep->gfpflags | GFP_THISNODE;
 
 	page = alloc_pages_node(nodeid, flags, cachep->gfporder);
 	if (!page)
@@ -3051,13 +3069,18 @@ static __always_inline void *__cache_alloc(struct kmem_cache *cachep,
 
 	local_irq_save(save_flags);
 
-#ifdef CONFIG_NUMA
-	if (unlikely(current->flags & (PF_SPREAD_SLAB | PF_MEMPOLICY)))
+	if (unlikely(NUMA_BUILD &&
+			current->flags & (PF_SPREAD_SLAB | PF_MEMPOLICY)))
 		objp = alternate_node_alloc(cachep, flags);
-#endif
 
 	if (!objp)
 		objp = ____cache_alloc(cachep, flags);
+	/*
+	 * We may just have run out of memory on the local node.
+	 * __cache_alloc_node() knows how to locate memory on other nodes
+	 */
+ 	if (NUMA_BUILD && !objp)
+ 		objp = __cache_alloc_node(cachep, flags, numa_node_id());
 	local_irq_restore(save_flags);
 	objp = cache_alloc_debugcheck_after(cachep, flags, objp,
 					    caller);
@@ -3076,7 +3099,7 @@ static void *alternate_node_alloc(struct kmem_cache *cachep, gfp_t flags)
 {
 	int nid_alloc, nid_here;
 
-	if (in_interrupt())
+	if (in_interrupt() || (flags & __GFP_THISNODE))
 		return NULL;
 	nid_alloc = nid_here = numa_node_id();
 	if (cpuset_do_slab_mem_spread() && (cachep->flags & SLAB_MEM_SPREAD))
@@ -3088,6 +3111,28 @@ static void *alternate_node_alloc(struct kmem_cache *cachep, gfp_t flags)
 	return NULL;
 }
 
+/*
+ * Fallback function if there was no memory available and no objects on a
+ * certain node and we are allowed to fall back. We mimick the behavior of
+ * the page allocator. We fall back according to a zonelist determined by
+ * the policy layer while obeying cpuset constraints.
+ */
+void *fallback_alloc(struct kmem_cache *cache, gfp_t flags)
+{
+	struct zonelist *zonelist = &NODE_DATA(slab_node(current->mempolicy))
+					->node_zonelists[gfp_zone(flags)];
+	struct zone **z;
+	void *obj = NULL;
+
+	for (z = zonelist->zones; *z && !obj; z++)
+		if (zone_idx(*z) <= ZONE_NORMAL &&
+				cpuset_zone_allowed(*z, flags))
+			obj = __cache_alloc_node(cache,
+					flags | __GFP_THISNODE,
+					zone_to_nid(*z));
+	return obj;
+}
+
 /*
  * A interface to enable slab creation on nodeid
  */
@@ -3141,11 +3186,15 @@ retry:
 must_grow:
 	spin_unlock(&l3->list_lock);
 	x = cache_grow(cachep, flags, nodeid);
+	if (x)
+		goto retry;
 
-	if (!x)
-		return NULL;
+	if (!(flags & __GFP_THISNODE))
+		/* Unable to grow the cache. Fall back to other nodes. */
+		return fallback_alloc(cachep, flags);
+
+	return NULL;
 
-	goto retry;
 done:
 	return obj;
 }
-- 
cgit v1.2.1


From d5f541ed6e31518508c688912e7464facf253c87 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:08 -0700
Subject: [PATCH] Add node to zone for the NUMA case

Add the node in order to optimize zone_to_nid.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Acked-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h     | 6 +++++-
 include/linux/mmzone.h | 1 +
 mm/page_alloc.c        | 1 +
 3 files changed, 7 insertions(+), 1 deletion(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7477fb59c4f2..8e433bbc6e7e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -446,7 +446,11 @@ static inline struct zone *page_zone(struct page *page)
 
 static inline unsigned long zone_to_nid(struct zone *zone)
 {
-	return zone->zone_pgdat->node_id;
+#ifdef CONFIG_NUMA
+	return zone->node;
+#else
+	return 0;
+#endif
 }
 
 static inline unsigned long page_to_nid(struct page *page)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 562cf7a8f3ee..59855b8718a0 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -168,6 +168,7 @@ struct zone {
 	unsigned long		lowmem_reserve[MAX_NR_ZONES];
 
 #ifdef CONFIG_NUMA
+	int node;
 	/*
 	 * zone reclaim becomes active if more unmapped pages exist.
 	 */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 4c76188b1681..d0432e44f77d 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2405,6 +2405,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat,
 		zone->spanned_pages = size;
 		zone->present_pages = realsize;
 #ifdef CONFIG_NUMA
+		zone->node = nid;
 		zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio)
 						/ 100;
 		zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100;
-- 
cgit v1.2.1


From 66a550308b8e4cbaba185d0326cb05d1bd758101 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:09 -0700
Subject: [PATCH] Do not allocate pagesets for unpopulated zones.

We do not need to allocate pagesets for unpopulated zones.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index d0432e44f77d..490aee1f2ce8 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1829,6 +1829,9 @@ static int __cpuinit process_zones(int cpu)
 
 	for_each_zone(zone) {
 
+		if (!populated_zone(zone))
+			continue;
+
 		zone_pcp(zone, cpu) = kmalloc_node(sizeof(struct per_cpu_pageset),
 					 GFP_KERNEL, cpu_to_node(cpu));
 		if (!zone_pcp(zone, cpu))
-- 
cgit v1.2.1


From 5d2923436217ba8bd05c5ee157712a391891c382 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@sgi.com>
Date: Wed, 27 Sep 2006 01:50:10 -0700
Subject: [PATCH] zone_statistics: Use hot node instead of cold zone_pgdat

Now that we have the node in the hot zone of struct zone we can avoid
accessing zone_pgdat in zone_statistics.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmstat.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/vmstat.c b/mm/vmstat.c
index 69c657132e1f..a2b6a9f96e5c 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -371,7 +371,7 @@ void zone_statistics(struct zonelist *zonelist, struct zone *z)
 		__inc_zone_state(z, NUMA_MISS);
 		__inc_zone_state(zonelist->zones[0], NUMA_FOREIGN);
 	}
-	if (z->zone_pgdat == NODE_DATA(numa_node_id()))
+	if (z->node == numa_node_id())
 		__inc_zone_state(z, NUMA_LOCAL);
 	else
 		__inc_zone_state(z, NUMA_OTHER);
-- 
cgit v1.2.1


From f4b81804a2d1ab341a4613089dc31ecce0800ed8 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 27 Sep 2006 01:50:10 -0700
Subject: [PATCH] do_no_pfn()

Implement do_no_pfn() for handling mapping of memory without a struct page
backing it.  This avoids creating fake page table entries for regions which
are not backed by real memory.

This feature is used by the MSPEC driver and other users, where it is
highly undesirable to have a struct page sitting behind the page (for
instance if the page is accessed in cached mode via the struct page in
parallel to the the driver accessing it uncached, which can result in data
corruption on some architectures, such as ia64).

This version uses specific NOPFN_{SIGBUS,OOM} return values, rather than
expect all negative pfn values would be an error.  It also bugs on cow
mappings as this would not work with the VM.

[akpm@osdl.org: micro-optimise]
Signed-off-by: Jes Sorensen <jes@sgi.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h |  7 ++++++
 mm/memory.c        | 64 +++++++++++++++++++++++++++++++++++++++++++++++++-----
 2 files changed, 66 insertions(+), 5 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 8e433bbc6e7e..22165cb18906 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -199,6 +199,7 @@ struct vm_operations_struct {
 	void (*open)(struct vm_area_struct * area);
 	void (*close)(struct vm_area_struct * area);
 	struct page * (*nopage)(struct vm_area_struct * area, unsigned long address, int *type);
+	unsigned long (*nopfn)(struct vm_area_struct * area, unsigned long address);
 	int (*populate)(struct vm_area_struct * area, unsigned long address, unsigned long len, pgprot_t prot, unsigned long pgoff, int nonblock);
 
 	/* notification that a previously read-only page is about to become
@@ -593,6 +594,12 @@ static inline int page_mapped(struct page *page)
 #define NOPAGE_SIGBUS	(NULL)
 #define NOPAGE_OOM	((struct page *) (-1))
 
+/*
+ * Error return values for the *_nopfn functions
+ */
+#define NOPFN_SIGBUS	((unsigned long) -1)
+#define NOPFN_OOM	((unsigned long) -2)
+
 /*
  * Different kinds of faults, as returned by handle_mm_fault().
  * Used to decide whether a process gets delivered SIGBUS or
diff --git a/mm/memory.c b/mm/memory.c
index 92a3ebd8d795..f2ef1dcfff77 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2255,6 +2255,54 @@ oom:
 	return VM_FAULT_OOM;
 }
 
+/*
+ * do_no_pfn() tries to create a new page mapping for a page without
+ * a struct_page backing it
+ *
+ * As this is called only for pages that do not currently exist, we
+ * do not need to flush old virtual caches or the TLB.
+ *
+ * We enter with non-exclusive mmap_sem (to exclude vma changes,
+ * but allow concurrent faults), and pte mapped but not yet locked.
+ * We return with mmap_sem still held, but pte unmapped and unlocked.
+ *
+ * It is expected that the ->nopfn handler always returns the same pfn
+ * for a given virtual mapping.
+ *
+ * Mark this `noinline' to prevent it from bloating the main pagefault code.
+ */
+static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma,
+		     unsigned long address, pte_t *page_table, pmd_t *pmd,
+		     int write_access)
+{
+	spinlock_t *ptl;
+	pte_t entry;
+	unsigned long pfn;
+	int ret = VM_FAULT_MINOR;
+
+	pte_unmap(page_table);
+	BUG_ON(!(vma->vm_flags & VM_PFNMAP));
+	BUG_ON(is_cow_mapping(vma->vm_flags));
+
+	pfn = vma->vm_ops->nopfn(vma, address & PAGE_MASK);
+	if (pfn == NOPFN_OOM)
+		return VM_FAULT_OOM;
+	if (pfn == NOPFN_SIGBUS)
+		return VM_FAULT_SIGBUS;
+
+	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
+
+	/* Only go through if we didn't race with anybody else... */
+	if (pte_none(*page_table)) {
+		entry = pfn_pte(pfn, vma->vm_page_prot);
+		if (write_access)
+			entry = maybe_mkwrite(pte_mkdirty(entry), vma);
+		set_pte_at(mm, address, page_table, entry);
+	}
+	pte_unmap_unlock(page_table, ptl);
+	return ret;
+}
+
 /*
  * Fault of a previously existing named mapping. Repopulate the pte
  * from the encoded file_pte if possible. This enables swappable
@@ -2317,11 +2365,17 @@ static inline int handle_pte_fault(struct mm_struct *mm,
 	old_entry = entry = *pte;
 	if (!pte_present(entry)) {
 		if (pte_none(entry)) {
-			if (!vma->vm_ops || !vma->vm_ops->nopage)
-				return do_anonymous_page(mm, vma, address,
-					pte, pmd, write_access);
-			return do_no_page(mm, vma, address,
-					pte, pmd, write_access);
+			if (vma->vm_ops) {
+				if (vma->vm_ops->nopage)
+					return do_no_page(mm, vma, address,
+							  pte, pmd,
+							  write_access);
+				if (unlikely(vma->vm_ops->nopfn))
+					return do_no_pfn(mm, vma, address, pte,
+							 pmd, write_access);
+			}
+			return do_anonymous_page(mm, vma, address,
+						 pte, pmd, write_access);
 		}
 		if (pte_file(entry))
 			return do_file_page(mm, vma, address,
-- 
cgit v1.2.1


From 17a3b05047119b7fc72ef03962e202becc659579 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Wed, 27 Sep 2006 01:50:11 -0700
Subject: [PATCH] mspec driver

Implement the special memory driver (mspec) based on the do_no_pfn
approach.  The driver is currently used only on SN2 hardware with special
fetchop support but could be beneficial on other architectures using the
uncached mode.

Signed-off-by: Jes Sorensen <jes@sgi.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/Kconfig  |   8 +
 drivers/char/Makefile |   1 +
 drivers/char/mspec.c  | 421 ++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 430 insertions(+)
 create mode 100644 drivers/char/mspec.c

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 52ea94b891f5..6a3d3af45c59 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -439,6 +439,14 @@ config SGI_MBCS
          If you have an SGI Altix with an attached SABrick
          say Y or M here, otherwise say N.
 
+config MSPEC
+	tristate "Memory special operations driver"
+	depends on IA64
+	help
+	  If you have an ia64 and you want to enable memory special
+	  operations support (formerly known as fetchop), say Y here,
+	  otherwise say N.
+
 source "drivers/serial/Kconfig"
 
 config UNIX98_PTYS
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 8c6dfc621520..b583d0cd9fbe 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -47,6 +47,7 @@ obj-$(CONFIG_HVC_RTAS)		+= hvc_rtas.o
 obj-$(CONFIG_HVC_DRIVER)	+= hvc_console.o
 obj-$(CONFIG_RAW_DRIVER)	+= raw.o
 obj-$(CONFIG_SGI_SNSC)		+= snsc.o snsc_event.o
+obj-$(CONFIG_MSPEC)		+= mspec.o
 obj-$(CONFIG_MMTIMER)		+= mmtimer.o
 obj-$(CONFIG_VIOCONS)		+= viocons.o
 obj-$(CONFIG_VIOTAPE)		+= viotape.o
diff --git a/drivers/char/mspec.c b/drivers/char/mspec.c
new file mode 100644
index 000000000000..5426b1e5595f
--- /dev/null
+++ b/drivers/char/mspec.c
@@ -0,0 +1,421 @@
+/*
+ * Copyright (C) 2001-2006 Silicon Graphics, Inc.  All rights
+ * reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ */
+
+/*
+ * SN Platform Special Memory (mspec) Support
+ *
+ * This driver exports the SN special memory (mspec) facility to user
+ * processes.
+ * There are three types of memory made available thru this driver:
+ * fetchops, uncached and cached.
+ *
+ * Fetchops are atomic memory operations that are implemented in the
+ * memory controller on SGI SN hardware.
+ *
+ * Uncached are used for memory write combining feature of the ia64
+ * cpu.
+ *
+ * Cached are used for areas of memory that are used as cached addresses
+ * on our partition and used as uncached addresses from other partitions.
+ * Due to a design constraint of the SN2 Shub, you can not have processors
+ * on the same FSB perform both a cached and uncached reference to the
+ * same cache line.  These special memory cached regions prevent the
+ * kernel from ever dropping in a TLB entry and therefore prevent the
+ * processor from ever speculating a cache line from this page.
+ */
+
+#include <linux/config.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/miscdevice.h>
+#include <linux/spinlock.h>
+#include <linux/mm.h>
+#include <linux/vmalloc.h>
+#include <linux/string.h>
+#include <linux/slab.h>
+#include <linux/numa.h>
+#include <asm/page.h>
+#include <asm/system.h>
+#include <asm/pgtable.h>
+#include <asm/atomic.h>
+#include <asm/tlbflush.h>
+#include <asm/uncached.h>
+#include <asm/sn/addrs.h>
+#include <asm/sn/arch.h>
+#include <asm/sn/mspec.h>
+#include <asm/sn/sn_cpuid.h>
+#include <asm/sn/io.h>
+#include <asm/sn/bte.h>
+#include <asm/sn/shubio.h>
+
+
+#define FETCHOP_ID	"SGI Fetchop,"
+#define CACHED_ID	"Cached,"
+#define UNCACHED_ID	"Uncached"
+#define REVISION	"4.0"
+#define MSPEC_BASENAME	"mspec"
+
+/*
+ * Page types allocated by the device.
+ */
+enum {
+	MSPEC_FETCHOP = 1,
+	MSPEC_CACHED,
+	MSPEC_UNCACHED
+};
+
+static int is_sn2;
+
+/*
+ * One of these structures is allocated when an mspec region is mmaped. The
+ * structure is pointed to by the vma->vm_private_data field in the vma struct.
+ * This structure is used to record the addresses of the mspec pages.
+ */
+struct vma_data {
+	atomic_t refcnt;	/* Number of vmas sharing the data. */
+	spinlock_t lock;	/* Serialize access to the vma. */
+	int count;		/* Number of pages allocated. */
+	int type;		/* Type of pages allocated. */
+	unsigned long maddr[0];	/* Array of MSPEC addresses. */
+};
+
+/* used on shub2 to clear FOP cache in the HUB */
+static unsigned long scratch_page[MAX_NUMNODES];
+#define SH2_AMO_CACHE_ENTRIES	4
+
+static inline int
+mspec_zero_block(unsigned long addr, int len)
+{
+	int status;
+
+	if (is_sn2) {
+		if (is_shub2()) {
+			int nid;
+			void *p;
+			int i;
+
+			nid = nasid_to_cnodeid(get_node_number(__pa(addr)));
+			p = (void *)TO_AMO(scratch_page[nid]);
+
+			for (i=0; i < SH2_AMO_CACHE_ENTRIES; i++) {
+				FETCHOP_LOAD_OP(p, FETCHOP_LOAD);
+				p += FETCHOP_VAR_SIZE;
+			}
+		}
+
+		status = bte_copy(0, addr & ~__IA64_UNCACHED_OFFSET, len,
+				  BTE_WACQUIRE | BTE_ZERO_FILL, NULL);
+	} else {
+		memset((char *) addr, 0, len);
+		status = 0;
+	}
+	return status;
+}
+
+/*
+ * mspec_open
+ *
+ * Called when a device mapping is created by a means other than mmap
+ * (via fork, etc.).  Increments the reference count on the underlying
+ * mspec data so it is not freed prematurely.
+ */
+static void
+mspec_open(struct vm_area_struct *vma)
+{
+	struct vma_data *vdata;
+
+	vdata = vma->vm_private_data;
+	atomic_inc(&vdata->refcnt);
+}
+
+/*
+ * mspec_close
+ *
+ * Called when unmapping a device mapping. Frees all mspec pages
+ * belonging to the vma.
+ */
+static void
+mspec_close(struct vm_area_struct *vma)
+{
+	struct vma_data *vdata;
+	int i, pages, result, vdata_size;
+
+	vdata = vma->vm_private_data;
+	if (!atomic_dec_and_test(&vdata->refcnt))
+		return;
+
+	pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+	vdata_size = sizeof(struct vma_data) + pages * sizeof(long);
+	for (i = 0; i < pages; i++) {
+		if (vdata->maddr[i] == 0)
+			continue;
+		/*
+		 * Clear the page before sticking it back
+		 * into the pool.
+		 */
+		result = mspec_zero_block(vdata->maddr[i], PAGE_SIZE);
+		if (!result)
+			uncached_free_page(vdata->maddr[i]);
+		else
+			printk(KERN_WARNING "mspec_close(): "
+			       "failed to zero page %i\n",
+			       result);
+	}
+
+	if (vdata_size <= PAGE_SIZE)
+		kfree(vdata);
+	else
+		vfree(vdata);
+}
+
+
+/*
+ * mspec_nopfn
+ *
+ * Creates a mspec page and maps it to user space.
+ */
+static unsigned long
+mspec_nopfn(struct vm_area_struct *vma, unsigned long address)
+{
+	unsigned long paddr, maddr;
+	unsigned long pfn;
+	int index;
+	struct vma_data *vdata = vma->vm_private_data;
+
+	index = (address - vma->vm_start) >> PAGE_SHIFT;
+	maddr = (volatile unsigned long) vdata->maddr[index];
+	if (maddr == 0) {
+		maddr = uncached_alloc_page(numa_node_id());
+		if (maddr == 0)
+			return NOPFN_OOM;
+
+		spin_lock(&vdata->lock);
+		if (vdata->maddr[index] == 0) {
+			vdata->count++;
+			vdata->maddr[index] = maddr;
+		} else {
+			uncached_free_page(maddr);
+			maddr = vdata->maddr[index];
+		}
+		spin_unlock(&vdata->lock);
+	}
+
+	if (vdata->type == MSPEC_FETCHOP)
+		paddr = TO_AMO(maddr);
+	else
+		paddr = __pa(TO_CAC(maddr));
+
+	pfn = paddr >> PAGE_SHIFT;
+
+	return pfn;
+}
+
+static struct vm_operations_struct mspec_vm_ops = {
+	.open = mspec_open,
+	.close = mspec_close,
+	.nopfn = mspec_nopfn
+};
+
+/*
+ * mspec_mmap
+ *
+ * Called when mmaping the device.  Initializes the vma with a fault handler
+ * and private data structure necessary to allocate, track, and free the
+ * underlying pages.
+ */
+static int
+mspec_mmap(struct file *file, struct vm_area_struct *vma, int type)
+{
+	struct vma_data *vdata;
+	int pages, vdata_size;
+
+	if (vma->vm_pgoff != 0)
+		return -EINVAL;
+
+	if ((vma->vm_flags & VM_SHARED) == 0)
+		return -EINVAL;
+
+	if ((vma->vm_flags & VM_WRITE) == 0)
+		return -EPERM;
+
+	pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
+	vdata_size = sizeof(struct vma_data) + pages * sizeof(long);
+	if (vdata_size <= PAGE_SIZE)
+		vdata = kmalloc(vdata_size, GFP_KERNEL);
+	else
+		vdata = vmalloc(vdata_size);
+	if (!vdata)
+		return -ENOMEM;
+	memset(vdata, 0, vdata_size);
+
+	vdata->type = type;
+	spin_lock_init(&vdata->lock);
+	vdata->refcnt = ATOMIC_INIT(1);
+	vma->vm_private_data = vdata;
+
+	vma->vm_flags |= (VM_IO | VM_LOCKED | VM_RESERVED | VM_PFNMAP);
+	if (vdata->type == MSPEC_FETCHOP || vdata->type == MSPEC_UNCACHED)
+		vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	vma->vm_ops = &mspec_vm_ops;
+
+	return 0;
+}
+
+static int
+fetchop_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	return mspec_mmap(file, vma, MSPEC_FETCHOP);
+}
+
+static int
+cached_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	return mspec_mmap(file, vma, MSPEC_CACHED);
+}
+
+static int
+uncached_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	return mspec_mmap(file, vma, MSPEC_UNCACHED);
+}
+
+static struct file_operations fetchop_fops = {
+	.owner = THIS_MODULE,
+	.mmap = fetchop_mmap
+};
+
+static struct miscdevice fetchop_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "sgi_fetchop",
+	.fops = &fetchop_fops
+};
+
+static struct file_operations cached_fops = {
+	.owner = THIS_MODULE,
+	.mmap = cached_mmap
+};
+
+static struct miscdevice cached_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "mspec_cached",
+	.fops = &cached_fops
+};
+
+static struct file_operations uncached_fops = {
+	.owner = THIS_MODULE,
+	.mmap = uncached_mmap
+};
+
+static struct miscdevice uncached_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "mspec_uncached",
+	.fops = &uncached_fops
+};
+
+/*
+ * mspec_init
+ *
+ * Called at boot time to initialize the mspec facility.
+ */
+static int __init
+mspec_init(void)
+{
+	int ret;
+	int nid;
+
+	/*
+	 * The fetchop device only works on SN2 hardware, uncached and cached
+	 * memory drivers should both be valid on all ia64 hardware
+	 */
+	if (ia64_platform_is("sn2")) {
+		is_sn2 = 1;
+		if (is_shub2()) {
+			ret = -ENOMEM;
+			for_each_online_node(nid) {
+				int actual_nid;
+				int nasid;
+				unsigned long phys;
+
+				scratch_page[nid] = uncached_alloc_page(nid);
+				if (scratch_page[nid] == 0)
+					goto free_scratch_pages;
+				phys = __pa(scratch_page[nid]);
+				nasid = get_node_number(phys);
+				actual_nid = nasid_to_cnodeid(nasid);
+				if (actual_nid != nid)
+					goto free_scratch_pages;
+			}
+		}
+
+		ret = misc_register(&fetchop_miscdev);
+		if (ret) {
+			printk(KERN_ERR
+			       "%s: failed to register device %i\n",
+			       FETCHOP_ID, ret);
+			goto free_scratch_pages;
+		}
+	}
+	ret = misc_register(&cached_miscdev);
+	if (ret) {
+		printk(KERN_ERR "%s: failed to register device %i\n",
+		       CACHED_ID, ret);
+		if (is_sn2)
+			misc_deregister(&fetchop_miscdev);
+		goto free_scratch_pages;
+	}
+	ret = misc_register(&uncached_miscdev);
+	if (ret) {
+		printk(KERN_ERR "%s: failed to register device %i\n",
+		       UNCACHED_ID, ret);
+		misc_deregister(&cached_miscdev);
+		if (is_sn2)
+			misc_deregister(&fetchop_miscdev);
+		goto free_scratch_pages;
+	}
+
+	printk(KERN_INFO "%s %s initialized devices: %s %s %s\n",
+	       MSPEC_BASENAME, REVISION, is_sn2 ? FETCHOP_ID : "",
+	       CACHED_ID, UNCACHED_ID);
+
+	return 0;
+
+ free_scratch_pages:
+	for_each_node(nid) {
+		if (scratch_page[nid] != 0)
+			uncached_free_page(scratch_page[nid]);
+	}
+	return ret;
+}
+
+static void __exit
+mspec_exit(void)
+{
+	int nid;
+
+	misc_deregister(&uncached_miscdev);
+	misc_deregister(&cached_miscdev);
+	if (is_sn2) {
+		misc_deregister(&fetchop_miscdev);
+
+		for_each_node(nid) {
+			if (scratch_page[nid] != 0)
+				uncached_free_page(scratch_page[nid]);
+		}
+	}
+}
+
+module_init(mspec_init);
+module_exit(mspec_exit);
+
+MODULE_AUTHOR("Silicon Graphics, Inc. <linux-altix@sgi.com>");
+MODULE_DESCRIPTION("Driver for SGI SN special memory operations");
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 423b41d773abe443bb546ce91361192073b96f88 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Wed, 27 Sep 2006 01:50:12 -0700
Subject: [PATCH] mm/page_alloc: use NULL instead of 0 for ptr

Use NULL instead of 0 for pointer value, eliminate sparse warnings.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 490aee1f2ce8..4f59d90b81e6 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1591,7 +1591,7 @@ static int __meminit __build_all_zonelists(void *dummy)
 void __meminit build_all_zonelists(void)
 {
 	if (system_state == SYSTEM_BOOTING) {
-		__build_all_zonelists(0);
+		__build_all_zonelists(NULL);
 		cpuset_init_current_mems_allowed();
 	} else {
 		/* we have to stop all cpus to guaranntee there is no user
-- 
cgit v1.2.1


From ead04089b138ed669658f80fafbe11fc7d97740b Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Wed, 27 Sep 2006 01:50:13 -0700
Subject: [PATCH] Fix kerneldoc comments in mm/vmalloc.c

The empty line between the short description and the first argument
description causes a section to appear twice in the generated manpage.
Also the short description should really be short: the script can't handle
multiple lines.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Acked-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmalloc.c | 28 ++++++++--------------------
 1 file changed, 8 insertions(+), 20 deletions(-)

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 9aad8b0cc6ee..659ec634856a 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -241,7 +241,6 @@ struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 
 /**
  *	get_vm_area  -  reserve a contingous kernel virtual area
- *
  *	@size:		size of the area
  *	@flags:		%VM_IOREMAP for I/O mappings or VM_ALLOC
  *
@@ -296,7 +295,6 @@ found:
 
 /**
  *	remove_vm_area  -  find and remove a contingous kernel virtual area
- *
  *	@addr:		base address
  *
  *	Search for the kernel VM area starting at @addr, and remove it.
@@ -355,7 +353,6 @@ void __vunmap(void *addr, int deallocate_pages)
 
 /**
  *	vfree  -  release memory allocated by vmalloc()
- *
  *	@addr:		memory base address
  *
  *	Free the virtually contiguous memory area starting at @addr, as
@@ -373,7 +370,6 @@ EXPORT_SYMBOL(vfree);
 
 /**
  *	vunmap  -  release virtual mapping obtained by vmap()
- *
  *	@addr:		memory base address
  *
  *	Free the virtually contiguous memory area starting at @addr,
@@ -390,7 +386,6 @@ EXPORT_SYMBOL(vunmap);
 
 /**
  *	vmap  -  map an array of pages into virtually contiguous space
- *
  *	@pages:		array of page pointers
  *	@count:		number of pages to map
  *	@flags:		vm_area->flags
@@ -471,7 +466,6 @@ void *__vmalloc_area(struct vm_struct *area, gfp_t gfp_mask, pgprot_t prot)
 
 /**
  *	__vmalloc_node  -  allocate virtually contiguous memory
- *
  *	@size:		allocation size
  *	@gfp_mask:	flags for the page level allocator
  *	@prot:		protection mask for the allocated pages
@@ -505,9 +499,7 @@ EXPORT_SYMBOL(__vmalloc);
 
 /**
  *	vmalloc  -  allocate virtually contiguous memory
- *
  *	@size:		allocation size
- *
  *	Allocate enough pages to cover @size from the page level
  *	allocator and map them into contiguous kernel virtual space.
  *
@@ -521,11 +513,11 @@ void *vmalloc(unsigned long size)
 EXPORT_SYMBOL(vmalloc);
 
 /**
- *	vmalloc_user  -  allocate virtually contiguous memory which has
- *			   been zeroed so it can be mapped to userspace without
- *			   leaking data.
+ * vmalloc_user - allocate zeroed virtually contiguous memory for userspace
+ * @size: allocation size
  *
- *	@size:		allocation size
+ * The resulting memory area is zeroed so it can be mapped to userspace
+ * without leaking data.
  */
 void *vmalloc_user(unsigned long size)
 {
@@ -544,7 +536,6 @@ EXPORT_SYMBOL(vmalloc_user);
 
 /**
  *	vmalloc_node  -  allocate memory on a specific node
- *
  *	@size:		allocation size
  *	@node:		numa node
  *
@@ -566,7 +557,6 @@ EXPORT_SYMBOL(vmalloc_node);
 
 /**
  *	vmalloc_exec  -  allocate virtually contiguous, executable memory
- *
  *	@size:		allocation size
  *
  *	Kernel-internal function to allocate enough pages to cover @size
@@ -584,7 +574,6 @@ void *vmalloc_exec(unsigned long size)
 
 /**
  *	vmalloc_32  -  allocate virtually contiguous memory (32bit addressable)
- *
  *	@size:		allocation size
  *
  *	Allocate enough 32bit PA addressable pages to cover @size from the
@@ -597,11 +586,11 @@ void *vmalloc_32(unsigned long size)
 EXPORT_SYMBOL(vmalloc_32);
 
 /**
- *	vmalloc_32_user  -  allocate virtually contiguous memory (32bit
- *			      addressable) which is zeroed so it can be
- *			      mapped to userspace without leaking data.
- *
+ * vmalloc_32_user - allocate zeroed virtually contiguous 32bit memory
  *	@size:		allocation size
+ *
+ * The resulting memory area is 32bit addressable and zeroed so it can be
+ * mapped to userspace without leaking data.
  */
 void *vmalloc_32_user(unsigned long size)
 {
@@ -695,7 +684,6 @@ finished:
 
 /**
  *	remap_vmalloc_range  -  map vmalloc pages to userspace
- *
  *	@vma:		vma to cover (map full range of vma)
  *	@addr:		vmalloc memory
  *	@pgoff:		number of pages into addr before first page to map
-- 
cgit v1.2.1


From d24afc57d51b1be41f95521e81399061fa5875a6 Mon Sep 17 00:00:00 2001
From: Rolf Eike Beer <eike-kernel@sf-tec.de>
Date: Wed, 27 Sep 2006 01:50:13 -0700
Subject: [PATCH] Mark __remove_vm_area() static

The function is exported but not used from anywhere else.  It's also marked as
"not for driver use" so noone out there should really care.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/vmalloc.h | 1 -
 mm/vmalloc.c            | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/include/linux/vmalloc.h b/include/linux/vmalloc.h
index dee88c6b6fa7..ce5f1482e6be 100644
--- a/include/linux/vmalloc.h
+++ b/include/linux/vmalloc.h
@@ -62,7 +62,6 @@ extern struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 extern struct vm_struct *get_vm_area_node(unsigned long size,
 					unsigned long flags, int node);
 extern struct vm_struct *remove_vm_area(void *addr);
-extern struct vm_struct *__remove_vm_area(void *addr);
 extern int map_vm_area(struct vm_struct *area, pgprot_t prot,
 			struct page ***pages);
 extern void unmap_vm_area(struct vm_struct *area);
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 659ec634856a..1ac191ce5641 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -272,7 +272,7 @@ static struct vm_struct *__find_vm_area(void *addr)
 }
 
 /* Caller must hold vmlist_lock */
-struct vm_struct *__remove_vm_area(void *addr)
+static struct vm_struct *__remove_vm_area(void *addr)
 {
 	struct vm_struct **p, *tmp;
 
-- 
cgit v1.2.1


From 361f6ed1d00f666a1a7c33f3e9aaccb713f9b9e4 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <hskinnemoen@atmel.com>
Date: Wed, 27 Sep 2006 01:50:14 -0700
Subject: [PATCH] AVR32: Use unsigned long flags for saving interrupt state

Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/avr32/mm/tlb.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/avr32/mm/tlb.c b/arch/avr32/mm/tlb.c
index 5d0523bbe298..7b073052203d 100644
--- a/arch/avr32/mm/tlb.c
+++ b/arch/avr32/mm/tlb.c
@@ -15,7 +15,8 @@
 
 void show_dtlb_entry(unsigned int index)
 {
-	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save;
+	unsigned long flags;
 
 	local_irq_save(flags);
 	mmucr_save = sysreg_read(MMUCR);
@@ -305,7 +306,8 @@ static void tlb_stop(struct seq_file *tlb, void *v)
 
 static int tlb_show(struct seq_file *tlb, void *v)
 {
-	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save, flags;
+	unsigned int tlbehi, tlbehi_save, tlbelo, mmucr, mmucr_save;
+	unsigned long flags;
 	unsigned long *index = v;
 
 	if (*index == 0)
-- 
cgit v1.2.1


From 0ec76a110f432e98277e464b82ace8dd66571689 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:15 -0700
Subject: [PATCH] NOMMU: Check that access_process_vm() has a valid target

Check that access_process_vm() is accessing a valid mapping in the target
process.

This limits ptrace() accesses and accesses through /proc/<pid>/maps to only
those regions actually mapped by a program.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/ptrace.c | 54 ------------------------------------------------------
 mm/memory.c     | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++
 mm/nommu.c      | 47 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 100 insertions(+), 54 deletions(-)

diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 9a111f70145c..8aad0331d82e 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -241,60 +241,6 @@ int ptrace_detach(struct task_struct *child, unsigned int data)
 	return 0;
 }
 
-/*
- * Access another process' address space.
- * Source/target buffer must be kernel space, 
- * Do not walk the page table directly, use get_user_pages
- */
-
-int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
-{
-	struct mm_struct *mm;
-	struct vm_area_struct *vma;
-	struct page *page;
-	void *old_buf = buf;
-
-	mm = get_task_mm(tsk);
-	if (!mm)
-		return 0;
-
-	down_read(&mm->mmap_sem);
-	/* ignore errors, just check how much was sucessfully transfered */
-	while (len) {
-		int bytes, ret, offset;
-		void *maddr;
-
-		ret = get_user_pages(tsk, mm, addr, 1,
-				write, 1, &page, &vma);
-		if (ret <= 0)
-			break;
-
-		bytes = len;
-		offset = addr & (PAGE_SIZE-1);
-		if (bytes > PAGE_SIZE-offset)
-			bytes = PAGE_SIZE-offset;
-
-		maddr = kmap(page);
-		if (write) {
-			copy_to_user_page(vma, page, addr,
-					  maddr + offset, buf, bytes);
-			set_page_dirty_lock(page);
-		} else {
-			copy_from_user_page(vma, page, addr,
-					    buf, maddr + offset, bytes);
-		}
-		kunmap(page);
-		page_cache_release(page);
-		len -= bytes;
-		buf += bytes;
-		addr += bytes;
-	}
-	up_read(&mm->mmap_sem);
-	mmput(mm);
-	
-	return buf - old_buf;
-}
-
 int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len)
 {
 	int copied = 0;
diff --git a/mm/memory.c b/mm/memory.c
index f2ef1dcfff77..601159a46ab6 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2604,3 +2604,56 @@ int in_gate_area_no_task(unsigned long addr)
 }
 
 #endif	/* __HAVE_ARCH_GATE_AREA */
+
+/*
+ * Access another process' address space.
+ * Source/target buffer must be kernel space,
+ * Do not walk the page table directly, use get_user_pages
+ */
+int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
+{
+	struct mm_struct *mm;
+	struct vm_area_struct *vma;
+	struct page *page;
+	void *old_buf = buf;
+
+	mm = get_task_mm(tsk);
+	if (!mm)
+		return 0;
+
+	down_read(&mm->mmap_sem);
+	/* ignore errors, just check how much was sucessfully transfered */
+	while (len) {
+		int bytes, ret, offset;
+		void *maddr;
+
+		ret = get_user_pages(tsk, mm, addr, 1,
+				write, 1, &page, &vma);
+		if (ret <= 0)
+			break;
+
+		bytes = len;
+		offset = addr & (PAGE_SIZE-1);
+		if (bytes > PAGE_SIZE-offset)
+			bytes = PAGE_SIZE-offset;
+
+		maddr = kmap(page);
+		if (write) {
+			copy_to_user_page(vma, page, addr,
+					  maddr + offset, buf, bytes);
+			set_page_dirty_lock(page);
+		} else {
+			copy_from_user_page(vma, page, addr,
+					    buf, maddr + offset, bytes);
+		}
+		kunmap(page);
+		page_cache_release(page);
+		len -= bytes;
+		buf += bytes;
+		addr += bytes;
+	}
+	up_read(&mm->mmap_sem);
+	mmput(mm);
+
+	return buf - old_buf;
+}
diff --git a/mm/nommu.c b/mm/nommu.c
index d99dea31e443..d08acdae0036 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1206,3 +1206,50 @@ struct page *filemap_nopage(struct vm_area_struct *area,
 	BUG();
 	return NULL;
 }
+
+/*
+ * Access another process' address space.
+ * - source/target buffer must be kernel space
+ */
+int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
+{
+	struct vm_list_struct *vml;
+	struct vm_area_struct *vma;
+	struct mm_struct *mm;
+
+	if (addr + len < addr)
+		return 0;
+
+	mm = get_task_mm(tsk);
+	if (!mm)
+		return 0;
+
+	down_read(&mm->mmap_sem);
+
+	/* the access must start within one of the target process's mappings */
+	for (vml = mm->context.vmlist; vml; vml = vml->next)
+		if (addr >= vml->vma->vm_start && addr < vml->vma->vm_end)
+			break;
+
+	if (vml) {
+		vma = vml->vma;
+
+		/* don't overrun this mapping */
+		if (addr + len >= vma->vm_end)
+			len = vma->vm_end - addr;
+
+		/* only read or write mappings where it is permitted */
+		if (write && vma->vm_flags & VM_WRITE)
+			len -= copy_to_user((void *) addr, buf, len);
+		else if (!write && vma->vm_flags & VM_READ)
+			len -= copy_from_user(buf, (void *) addr, len);
+		else
+			len = 0;
+	} else {
+		len = 0;
+	}
+
+	up_read(&mm->mmap_sem);
+	mmput(mm);
+	return len;
+}
-- 
cgit v1.2.1


From 5da6185bca064e35aa73a7c1f27488d2b96434f4 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:16 -0700
Subject: [PATCH] NOMMU: Set BDI capabilities for /dev/mem and /dev/kmem

Set the backing device info capabilities for /dev/mem and /dev/kmem to
permit direct sharing under no-MMU conditions and full mapping capabilities
under MMU conditions.  Make the BDI used by these available to all directly
mappable character devices.

Also comment the capabilities for /dev/zero.

[akpm@osdl.org: ifdef reductions]
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/mem.c   | 39 +++++++++++++++++++++++++++++++++++++++
 fs/char_dev.c        | 20 ++++++++++++++++++++
 include/linux/cdev.h |  2 ++
 3 files changed, 61 insertions(+)

diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index 917b20402664..4ac70ec697f0 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -238,6 +238,32 @@ static pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
 }
 #endif
 
+#ifndef CONFIG_MMU
+static unsigned long get_unmapped_area_mem(struct file *file,
+					   unsigned long addr,
+					   unsigned long len,
+					   unsigned long pgoff,
+					   unsigned long flags)
+{
+	if (!valid_mmap_phys_addr_range(pgoff, len))
+		return (unsigned long) -EINVAL;
+	return pgoff;
+}
+
+/* can't do an in-place private mapping if there's no MMU */
+static inline int private_mapping_ok(struct vm_area_struct *vma)
+{
+	return vma->vm_flags & VM_MAYSHARE;
+}
+#else
+#define get_unmapped_area_mem	NULL
+
+static inline int private_mapping_ok(struct vm_area_struct *vma)
+{
+	return 1;
+}
+#endif
+
 static int mmap_mem(struct file * file, struct vm_area_struct * vma)
 {
 	size_t size = vma->vm_end - vma->vm_start;
@@ -245,6 +271,9 @@ static int mmap_mem(struct file * file, struct vm_area_struct * vma)
 	if (!valid_mmap_phys_addr_range(vma->vm_pgoff, size))
 		return -EINVAL;
 
+	if (!private_mapping_ok(vma))
+		return -ENOSYS;
+
 	vma->vm_page_prot = phys_mem_access_prot(file, vma->vm_pgoff,
 						 size,
 						 vma->vm_page_prot);
@@ -782,6 +811,7 @@ static const struct file_operations mem_fops = {
 	.write		= write_mem,
 	.mmap		= mmap_mem,
 	.open		= open_mem,
+	.get_unmapped_area = get_unmapped_area_mem,
 };
 
 static const struct file_operations kmem_fops = {
@@ -790,6 +820,7 @@ static const struct file_operations kmem_fops = {
 	.write		= write_kmem,
 	.mmap		= mmap_kmem,
 	.open		= open_kmem,
+	.get_unmapped_area = get_unmapped_area_mem,
 };
 
 static const struct file_operations null_fops = {
@@ -815,6 +846,10 @@ static const struct file_operations zero_fops = {
 	.mmap		= mmap_zero,
 };
 
+/*
+ * capabilities for /dev/zero
+ * - permits private mappings, "copies" are taken of the source of zeros
+ */
 static struct backing_dev_info zero_bdi = {
 	.capabilities	= BDI_CAP_MAP_COPY,
 };
@@ -862,9 +897,13 @@ static int memory_open(struct inode * inode, struct file * filp)
 	switch (iminor(inode)) {
 		case 1:
 			filp->f_op = &mem_fops;
+			filp->f_mapping->backing_dev_info =
+				&directly_mappable_cdev_bdi;
 			break;
 		case 2:
 			filp->f_op = &kmem_fops;
+			filp->f_mapping->backing_dev_info =
+				&directly_mappable_cdev_bdi;
 			break;
 		case 3:
 			filp->f_op = &null_fops;
diff --git a/fs/char_dev.c b/fs/char_dev.c
index 3483d3cf8087..0009346d827f 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -19,11 +19,30 @@
 #include <linux/kobj_map.h>
 #include <linux/cdev.h>
 #include <linux/mutex.h>
+#include <linux/backing-dev.h>
 
 #ifdef CONFIG_KMOD
 #include <linux/kmod.h>
 #endif
 
+/*
+ * capabilities for /dev/mem, /dev/kmem and similar directly mappable character
+ * devices
+ * - permits shared-mmap for read, write and/or exec
+ * - does not permit private mmap in NOMMU mode (can't do COW)
+ * - no readahead or I/O queue unplugging required
+ */
+struct backing_dev_info directly_mappable_cdev_bdi = {
+	.capabilities	= (
+#ifdef CONFIG_MMU
+		/* permit private copies of the data to be taken */
+		BDI_CAP_MAP_COPY |
+#endif
+		/* permit direct mmap, for read, write or exec */
+		BDI_CAP_MAP_DIRECT |
+		BDI_CAP_READ_MAP | BDI_CAP_WRITE_MAP | BDI_CAP_EXEC_MAP),
+};
+
 static struct kobj_map *cdev_map;
 
 static DEFINE_MUTEX(chrdevs_lock);
@@ -461,3 +480,4 @@ EXPORT_SYMBOL(cdev_del);
 EXPORT_SYMBOL(cdev_add);
 EXPORT_SYMBOL(register_chrdev);
 EXPORT_SYMBOL(unregister_chrdev);
+EXPORT_SYMBOL(directly_mappable_cdev_bdi);
diff --git a/include/linux/cdev.h b/include/linux/cdev.h
index 2216638962d2..ee5f53f2ca15 100644
--- a/include/linux/cdev.h
+++ b/include/linux/cdev.h
@@ -23,5 +23,7 @@ void cdev_del(struct cdev *);
 
 void cd_forget(struct inode *);
 
+extern struct backing_dev_info directly_mappable_cdev_bdi;
+
 #endif
 #endif
-- 
cgit v1.2.1


From 0159b141d8b1f9b9f9cffacae47bec1e05c63b8b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:16 -0700
Subject: [PATCH] NOMMU: Use find_vma() rather than reimplementing a VMA search

Use find_vma() in the NOMMU version of access_process_vm() rather than
reimplementing it.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/nommu.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/mm/nommu.c b/mm/nommu.c
index d08acdae0036..00ffa974c90c 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1039,6 +1039,7 @@ unsigned long do_mremap(unsigned long addr,
 
 /*
  * Look up the first VMA which satisfies  addr < vm_end,  NULL if none
+ * - should be called with mm->mmap_sem at least readlocked
  */
 struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
 {
@@ -1213,7 +1214,6 @@ struct page *filemap_nopage(struct vm_area_struct *area,
  */
 int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
 {
-	struct vm_list_struct *vml;
 	struct vm_area_struct *vma;
 	struct mm_struct *mm;
 
@@ -1227,13 +1227,8 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 	down_read(&mm->mmap_sem);
 
 	/* the access must start within one of the target process's mappings */
-	for (vml = mm->context.vmlist; vml; vml = vml->next)
-		if (addr >= vml->vma->vm_start && addr < vml->vma->vm_end)
-			break;
-
-	if (vml) {
-		vma = vml->vma;
-
+	vma = find_vma(mm, addr);
+	if (vma) {
 		/* don't overrun this mapping */
 		if (addr + len >= vma->vm_end)
 			len = vma->vm_end - addr;
-- 
cgit v1.2.1


From 910e46da4b4e93d56ffea318c64afa41868d5e6d Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.adi@gmail.com>
Date: Wed, 27 Sep 2006 01:50:17 -0700
Subject: [PATCH] Check if start address is in vma region in NOMMU function
 get_user_pages()

In NOMMU arch, if run "cat /proc/self/mem", data from physical address 0
are read.  This behavior is different from MMU arch.  In IA32, message
"cat: /proc/self/mem: Input/output error" is reported.

This issue is rootcaused by not validate the start address in NOMMU
function get_user_pages().  Following patch solves this issue.

Signed-off-by: Sonic Zhang <sonic.adi@gmail.com>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/nommu.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/mm/nommu.c b/mm/nommu.c
index 00ffa974c90c..2af50831183f 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -129,16 +129,20 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	struct page **pages, struct vm_area_struct **vmas)
 {
 	int i;
-	static struct vm_area_struct dummy_vma;
+	struct vm_area_struct *vma;
 
 	for (i = 0; i < len; i++) {
+		vma = find_vma(mm, start);
+		if(!vma)
+			return i ? : -EFAULT;
+
 		if (pages) {
 			pages[i] = virt_to_page(start);
 			if (pages[i])
 				page_cache_get(pages[i]);
 		}
 		if (vmas)
-			vmas[i] = &dummy_vma;
+			vmas[i] = vma;
 		start += PAGE_SIZE;
 	}
 	return(i);
-- 
cgit v1.2.1


From 7b4d5b8b39fd3701ed3693a89f2bd8f6ef49bce2 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:18 -0700
Subject: [PATCH] NOMMU: Check VMA protections

Check the VMA protections in get_user_pages() against what's being asked.

This checks to see that we don't accidentally write on a non-writable VMA or
permit an I/O mapping VMA to be accessed (which may lack page structs).

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/nommu.c | 30 +++++++++++++++++++++++++-----
 1 file changed, 25 insertions(+), 5 deletions(-)

diff --git a/mm/nommu.c b/mm/nommu.c
index 2af50831183f..2e140a6ae22e 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -122,19 +122,35 @@ unsigned int kobjsize(const void *objp)
 }
 
 /*
- * The nommu dodgy version :-)
+ * get a list of pages in an address range belonging to the specified process
+ * and indicate the VMA that covers each page
+ * - this is potentially dodgy as we may end incrementing the page count of a
+ *   slab page or a secondary page from a compound page
+ * - don't permit access to VMAs that don't support it, such as I/O mappings
  */
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 	unsigned long start, int len, int write, int force,
 	struct page **pages, struct vm_area_struct **vmas)
 {
-	int i;
 	struct vm_area_struct *vma;
+	unsigned long vm_flags;
+	int i;
+
+	/* calculate required read or write permissions.
+	 * - if 'force' is set, we only require the "MAY" flags.
+	 */
+	vm_flags  = write ? (VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD);
+	vm_flags &= force ? (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);
 
 	for (i = 0; i < len; i++) {
 		vma = find_vma(mm, start);
-		if(!vma)
-			return i ? : -EFAULT;
+		if (!vma)
+			goto finish_or_fault;
+
+		/* protect what we can, including chardevs */
+		if (vma->vm_flags & (VM_IO | VM_PFNMAP) ||
+		    !(vm_flags & vma->vm_flags))
+			goto finish_or_fault;
 
 		if (pages) {
 			pages[i] = virt_to_page(start);
@@ -145,7 +161,11 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
 			vmas[i] = vma;
 		start += PAGE_SIZE;
 	}
-	return(i);
+
+	return i;
+
+finish_or_fault:
+	return i ? : -EFAULT;
 }
 
 EXPORT_SYMBOL(get_user_pages);
-- 
cgit v1.2.1


From d00c7b993712e4bb16d0012b35b654e40159b327 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:19 -0700
Subject: [PATCH] NOMMU: Permit ptrace to ignore non-PROT_WRITE VMAs in NOMMU
 mode

Permit ptrace to modify a section that's non-shared but is marked
unwritable, such as is obtained by mapping the text segment of an ELF-FDPIC
executable binary with into a binary that's being ptraced[*].

[*] Under NOMMU conditions ptrace causes read-only MAP_PRIVATE mmaps to become
    totally private copies because if a private mapping was actually shared
    then the debugging setting breakpoints in it would potentially crash
    other processes.

This is done by using the VM_MAYWRITE flag rather than the VM_WRITE flag
when deciding whether to permit a write.

Without this patch a debugger can't set breakpoints in the mapped text
sections of executables that are mapped read-only private, even if the
mmap() syscall has taken a private copy because PT_PTRACED is set.

In addition, VM_MAYREAD is used instead of VM_READ for similar reasons.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/nommu.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/nommu.c b/mm/nommu.c
index 2e140a6ae22e..829fc904de11 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1258,9 +1258,9 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 			len = vma->vm_end - addr;
 
 		/* only read or write mappings where it is permitted */
-		if (write && vma->vm_flags & VM_WRITE)
+		if (write && vma->vm_flags & VM_MAYWRITE)
 			len -= copy_to_user((void *) addr, buf, len);
-		else if (!write && vma->vm_flags & VM_READ)
+		else if (!write && vma->vm_flags & VM_MAYREAD)
 			len -= copy_from_user(buf, (void *) addr, len);
 		else
 			len = 0;
-- 
cgit v1.2.1


From dbf8685c8e21404e3a8ed244bd0219d3c4b89101 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:19 -0700
Subject: [PATCH] NOMMU: Implement /proc/pid/maps for NOMMU

Implement /proc/pid/maps for NOMMU by reading the vm_area_list attached to
current->mm->context.vmlist.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/nommu-mmap.txt |  3 ++
 fs/proc/internal.h           |  1 +
 fs/proc/nommu.c              | 20 ++++++++----
 fs/proc/task_nommu.c         | 74 +++++++++++++++++++++++++++++++++++---------
 include/linux/proc_fs.h      |  2 ++
 5 files changed, 80 insertions(+), 20 deletions(-)

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index b88ebe4d808c..83f4b083d57e 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -116,6 +116,9 @@ FURTHER NOTES ON NO-MMU MMAP
  (*) A list of all the mappings on the system is visible through /proc/maps in
      no-MMU mode.
 
+ (*) A list of all the mappings in use by a process is visible through
+     /proc/<pid>/maps in no-MMU mode.
+
  (*) Supplying MAP_FIXED or a requesting a particular mapping address will
      result in an error.
 
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 146a434ba944..987c773dbb20 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -28,6 +28,7 @@ do {						\
 	(vmi)->largest_chunk = 0;		\
 } while(0)
 
+extern int nommu_vma_show(struct seq_file *, struct vm_area_struct *);
 #endif
 
 extern void create_seq_entry(char *name, mode_t mode, const struct file_operations *f);
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
index cff10ab1af63..d7dbdf9e0f49 100644
--- a/fs/proc/nommu.c
+++ b/fs/proc/nommu.c
@@ -33,19 +33,15 @@
 #include "internal.h"
 
 /*
- * display a list of all the VMAs the kernel knows about
- * - nommu kernals have a single flat list
+ * display a single VMA to a sequenced file
  */
-static int nommu_vma_list_show(struct seq_file *m, void *v)
+int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
 {
-	struct vm_area_struct *vma;
 	unsigned long ino = 0;
 	struct file *file;
 	dev_t dev = 0;
 	int flags, len;
 
-	vma = rb_entry((struct rb_node *) v, struct vm_area_struct, vm_rb);
-
 	flags = vma->vm_flags;
 	file = vma->vm_file;
 
@@ -78,6 +74,18 @@ static int nommu_vma_list_show(struct seq_file *m, void *v)
 	return 0;
 }
 
+/*
+ * display a list of all the VMAs the kernel knows about
+ * - nommu kernals have a single flat list
+ */
+static int nommu_vma_list_show(struct seq_file *m, void *v)
+{
+	struct vm_area_struct *vma;
+
+	vma = rb_entry((struct rb_node *) v, struct vm_area_struct, vm_rb);
+	return nommu_vma_show(m, vma);
+}
+
 static void *nommu_vma_list_start(struct seq_file *m, loff_t *_pos)
 {
 	struct rb_node *_rb;
diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
index 4616ed50ffcd..091aa8e48e02 100644
--- a/fs/proc/task_nommu.c
+++ b/fs/proc/task_nommu.c
@@ -138,25 +138,63 @@ out:
 }
 
 /*
- * Albert D. Cahalan suggested to fake entries for the traditional
- * sections here.  This might be worth investigating.
+ * display mapping lines for a particular process's /proc/pid/maps
  */
-static int show_map(struct seq_file *m, void *v)
+static int show_map(struct seq_file *m, void *_vml)
 {
-	return 0;
+	struct vm_list_struct *vml = _vml;
+	return nommu_vma_show(m, vml->vma);
 }
+
 static void *m_start(struct seq_file *m, loff_t *pos)
 {
+	struct proc_maps_private *priv = m->private;
+	struct vm_list_struct *vml;
+	struct mm_struct *mm;
+	loff_t n = *pos;
+
+	/* pin the task and mm whilst we play with them */
+	priv->task = get_pid_task(priv->pid, PIDTYPE_PID);
+	if (!priv->task)
+		return NULL;
+
+	mm = get_task_mm(priv->task);
+	if (!mm) {
+		put_task_struct(priv->task);
+		priv->task = NULL;
+		return NULL;
+	}
+
+	down_read(&mm->mmap_sem);
+
+	/* start from the Nth VMA */
+	for (vml = mm->context.vmlist; vml; vml = vml->next)
+		if (n-- == 0)
+			return vml;
 	return NULL;
 }
-static void m_stop(struct seq_file *m, void *v)
+
+static void m_stop(struct seq_file *m, void *_vml)
 {
+	struct proc_maps_private *priv = m->private;
+
+	if (priv->task) {
+		struct mm_struct *mm = priv->task->mm;
+		up_read(&mm->mmap_sem);
+		mmput(mm);
+		put_task_struct(priv->task);
+	}
 }
-static void *m_next(struct seq_file *m, void *v, loff_t *pos)
+
+static void *m_next(struct seq_file *m, void *_vml, loff_t *pos)
 {
-	return NULL;
+	struct vm_list_struct *vml = _vml;
+
+	(*pos)++;
+	return vml ? vml->next : NULL;
 }
-static struct seq_operations proc_pid_maps_op = {
+
+static struct seq_operations proc_pid_maps_ops = {
 	.start	= m_start,
 	.next	= m_next,
 	.stop	= m_stop,
@@ -165,11 +203,19 @@ static struct seq_operations proc_pid_maps_op = {
 
 static int maps_open(struct inode *inode, struct file *file)
 {
-	int ret;
-	ret = seq_open(file, &proc_pid_maps_op);
-	if (!ret) {
-		struct seq_file *m = file->private_data;
-		m->private = NULL;
+	struct proc_maps_private *priv;
+	int ret = -ENOMEM;
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (priv) {
+		priv->pid = proc_pid(inode);
+		ret = seq_open(file, &proc_pid_maps_ops);
+		if (!ret) {
+			struct seq_file *m = file->private_data;
+			m->private = priv;
+		} else {
+			kfree(priv);
+		}
 	}
 	return ret;
 }
@@ -178,6 +224,6 @@ struct file_operations proc_maps_operations = {
 	.open		= maps_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release,
+	.release	= seq_release_private,
 };
 
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index 3435ca38dd14..57f70bc8b24b 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -268,7 +268,9 @@ static inline struct proc_dir_entry *PDE(const struct inode *inode)
 struct proc_maps_private {
 	struct pid *pid;
 	struct task_struct *task;
+#ifdef CONFIG_MMU
 	struct vm_area_struct *tail_vma;
+#endif
 };
 
 #endif /* _LINUX_PROC_FS_H */
-- 
cgit v1.2.1


From 3034097a5017dd9281b1f795e80af9859627850e Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:20 -0700
Subject: [PATCH] NOMMU: Order the per-mm_struct VMA list

Order the per-mm_struct VMA list by address so that searching it can be cut
short when the appropriate address has been exceeded.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/nommu.c | 104 ++++++++++++++++++++++++++++++++++++++++++-------------------
 1 file changed, 72 insertions(+), 32 deletions(-)

diff --git a/mm/nommu.c b/mm/nommu.c
index 829fc904de11..4f87b2f43a2b 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -310,6 +310,48 @@ static void show_process_blocks(void)
 }
 #endif /* DEBUG */
 
+/*
+ * add a VMA into a process's mm_struct in the appropriate place in the list
+ * - should be called with mm->mmap_sem held writelocked
+ */
+static void add_vma_to_mm(struct mm_struct *mm, struct vm_list_struct *vml)
+{
+	struct vm_list_struct **ppv;
+
+	for (ppv = &current->mm->context.vmlist; *ppv; ppv = &(*ppv)->next)
+		if ((*ppv)->vma->vm_start > vml->vma->vm_start)
+			break;
+
+	vml->next = *ppv;
+	*ppv = vml;
+}
+
+/*
+ * look up the first VMA in which addr resides, NULL if none
+ * - should be called with mm->mmap_sem at least held readlocked
+ */
+struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
+{
+	struct vm_list_struct *loop, *vml;
+
+	/* search the vm_start ordered list */
+	vml = NULL;
+	for (loop = mm->context.vmlist; loop; loop = loop->next) {
+		if (loop->vma->vm_start > addr)
+			break;
+		vml = loop;
+	}
+
+	if (vml && vml->vma->vm_end > addr)
+		return vml->vma;
+
+	return NULL;
+}
+EXPORT_SYMBOL(find_vma);
+
+/*
+ * find a VMA in the global tree
+ */
 static inline struct vm_area_struct *find_nommu_vma(unsigned long start)
 {
 	struct vm_area_struct *vma;
@@ -329,6 +371,9 @@ static inline struct vm_area_struct *find_nommu_vma(unsigned long start)
 	return NULL;
 }
 
+/*
+ * add a VMA in the global tree
+ */
 static void add_nommu_vma(struct vm_area_struct *vma)
 {
 	struct vm_area_struct *pvma;
@@ -375,6 +420,9 @@ static void add_nommu_vma(struct vm_area_struct *vma)
 	rb_insert_color(&vma->vm_rb, &nommu_vma_tree);
 }
 
+/*
+ * delete a VMA from the global list
+ */
 static void delete_nommu_vma(struct vm_area_struct *vma)
 {
 	struct address_space *mapping;
@@ -852,8 +900,7 @@ unsigned long do_mmap_pgoff(struct file *file,
 	realalloc += kobjsize(vml);
 	askedalloc += sizeof(*vml);
 
-	vml->next = current->mm->context.vmlist;
-	current->mm->context.vmlist = vml;
+	add_vma_to_mm(current->mm, vml);
 
 	up_write(&nommu_vma_sem);
 
@@ -932,6 +979,11 @@ static void put_vma(struct vm_area_struct *vma)
 	}
 }
 
+/*
+ * release a mapping
+ * - under NOMMU conditions the parameters must match exactly to the mapping to
+ *   be removed
+ */
 int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len)
 {
 	struct vm_list_struct *vml, **parent;
@@ -941,10 +993,13 @@ int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len)
 	printk("do_munmap:\n");
 #endif
 
-	for (parent = &mm->context.vmlist; *parent; parent = &(*parent)->next)
+	for (parent = &mm->context.vmlist; *parent; parent = &(*parent)->next) {
+		if ((*parent)->vma->vm_start > addr)
+			break;
 		if ((*parent)->vma->vm_start == addr &&
 		    ((len == 0) || ((*parent)->vma->vm_end == end)))
 			goto found;
+	}
 
 	printk("munmap of non-mmaped memory by process %d (%s): %p\n",
 	       current->pid, current->comm, (void *) addr);
@@ -970,7 +1025,20 @@ int do_munmap(struct mm_struct *mm, unsigned long addr, size_t len)
 	return 0;
 }
 
-/* Release all mmaps. */
+asmlinkage long sys_munmap(unsigned long addr, size_t len)
+{
+	int ret;
+	struct mm_struct *mm = current->mm;
+
+	down_write(&mm->mmap_sem);
+	ret = do_munmap(mm, addr, len);
+	up_write(&mm->mmap_sem);
+	return ret;
+}
+
+/*
+ * Release all mappings
+ */
 void exit_mmap(struct mm_struct * mm)
 {
 	struct vm_list_struct *tmp;
@@ -997,17 +1065,6 @@ void exit_mmap(struct mm_struct * mm)
 	}
 }
 
-asmlinkage long sys_munmap(unsigned long addr, size_t len)
-{
-	int ret;
-	struct mm_struct *mm = current->mm;
-
-	down_write(&mm->mmap_sem);
-	ret = do_munmap(mm, addr, len);
-	up_write(&mm->mmap_sem);
-	return ret;
-}
-
 unsigned long do_brk(unsigned long addr, unsigned long len)
 {
 	return -ENOMEM;
@@ -1061,23 +1118,6 @@ unsigned long do_mremap(unsigned long addr,
 	return vml->vma->vm_start;
 }
 
-/*
- * Look up the first VMA which satisfies  addr < vm_end,  NULL if none
- * - should be called with mm->mmap_sem at least readlocked
- */
-struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
-{
-	struct vm_list_struct *vml;
-
-	for (vml = mm->context.vmlist; vml; vml = vml->next)
-		if (addr >= vml->vma->vm_start && addr < vml->vma->vm_end)
-			return vml->vma;
-
-	return NULL;
-}
-
-EXPORT_SYMBOL(find_vma);
-
 struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
 			unsigned int foll_flags)
 {
-- 
cgit v1.2.1


From 6fa5f80bc34da1a49b42117602b44441402cac2f Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:21 -0700
Subject: [PATCH] NOMMU: Make mremap() partially work for NOMMU kernels

Make mremap() partially work for NOMMU kernels.  It may resize a VMA provided
that it doesn't exceed the size of the slab object in which the storage is
allocated that the VMA refers to.  Shareable VMAs may not be resized.

Moving VMAs (as permitted by MREMAP_MAYMOVE) is not currently supported.

This patch also makes use of the fact that the VMA list is now ordered to cut
it short when possible.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/nommu-mmap.txt | 24 +++++++++++++++++
 mm/nommu.c                   | 63 ++++++++++++++++++++++++++++++++------------
 2 files changed, 70 insertions(+), 17 deletions(-)

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 83f4b083d57e..3ce890664936 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -128,6 +128,30 @@ FURTHER NOTES ON NO-MMU MMAP
      error will result if they don't. This is most likely to be encountered
      with character device files, pipes, fifos and sockets.
 
+
+=============
+NO-MMU MREMAP
+=============
+
+The mremap() function is partially supported.  It may change the size of a
+mapping, and may move it[*] if MREMAP_MAYMOVE is specified and if the new size
+of the mapping exceeds the size of the slab object currently occupied by the
+memory to which the mapping refers, or if a smaller slab object could be used.
+
+MREMAP_FIXED is not supported, though it is ignored if there's no change of
+address and the object does not need to be moved.
+
+Shared mappings may not be moved.  Shareable mappings may not be moved either,
+even if they are not currently shared.
+
+The mremap() function must be given an exact match for base address and size of
+a previously mapped object.  It may not be used to create holes in existing
+mappings, move parts of existing mappings or resize parts of mappings.  It must
+act on a complete mapping.
+
+[*] Not currently supported.
+
+
 ============================================
 PROVIDING SHAREABLE CHARACTER DEVICE SUPPORT
 ============================================
diff --git a/mm/nommu.c b/mm/nommu.c
index 4f87b2f43a2b..23cfa8ec914a 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -349,6 +349,26 @@ struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
 }
 EXPORT_SYMBOL(find_vma);
 
+/*
+ * look up the first VMA exactly that exactly matches addr
+ * - should be called with mm->mmap_sem at least held readlocked
+ */
+static inline struct vm_area_struct *find_vma_exact(struct mm_struct *mm,
+						    unsigned long addr)
+{
+	struct vm_list_struct *vml;
+
+	/* search the vm_start ordered list */
+	for (vml = mm->context.vmlist; vml; vml = vml->next) {
+		if (vml->vma->vm_start == addr)
+			return vml->vma;
+		if (vml->vma->vm_start > addr)
+			break;
+	}
+
+	return NULL;
+}
+
 /*
  * find a VMA in the global tree
  */
@@ -1071,20 +1091,20 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 }
 
 /*
- * Expand (or shrink) an existing mapping, potentially moving it at the
- * same time (controlled by the MREMAP_MAYMOVE flag and available VM space)
+ * expand (or shrink) an existing mapping, potentially moving it at the same
+ * time (controlled by the MREMAP_MAYMOVE flag and available VM space)
  *
- * MREMAP_FIXED option added 5-Dec-1999 by Benjamin LaHaise
- * This option implies MREMAP_MAYMOVE.
+ * under NOMMU conditions, we only permit changing a mapping's size, and only
+ * as long as it stays within the hole allocated by the kmalloc() call in
+ * do_mmap_pgoff() and the block is not shareable
  *
- * on uClinux, we only permit changing a mapping's size, and only as long as it stays within the
- * hole allocated by the kmalloc() call in do_mmap_pgoff() and the block is not shareable
+ * MREMAP_FIXED is not supported under NOMMU conditions
  */
 unsigned long do_mremap(unsigned long addr,
 			unsigned long old_len, unsigned long new_len,
 			unsigned long flags, unsigned long new_addr)
 {
-	struct vm_list_struct *vml = NULL;
+	struct vm_area_struct *vma;
 
 	/* insanity checks first */
 	if (new_len == 0)
@@ -1093,29 +1113,38 @@ unsigned long do_mremap(unsigned long addr,
 	if (flags & MREMAP_FIXED && new_addr != addr)
 		return (unsigned long) -EINVAL;
 
-	for (vml = current->mm->context.vmlist; vml; vml = vml->next)
-		if (vml->vma->vm_start == addr)
-			goto found;
-
-	return (unsigned long) -EINVAL;
+	vma = find_vma_exact(current->mm, addr);
+	if (!vma)
+		return (unsigned long) -EINVAL;
 
- found:
-	if (vml->vma->vm_end != vml->vma->vm_start + old_len)
+	if (vma->vm_end != vma->vm_start + old_len)
 		return (unsigned long) -EFAULT;
 
-	if (vml->vma->vm_flags & VM_MAYSHARE)
+	if (vma->vm_flags & VM_MAYSHARE)
 		return (unsigned long) -EPERM;
 
 	if (new_len > kobjsize((void *) addr))
 		return (unsigned long) -ENOMEM;
 
 	/* all checks complete - do it */
-	vml->vma->vm_end = vml->vma->vm_start + new_len;
+	vma->vm_end = vma->vm_start + new_len;
 
 	askedalloc -= old_len;
 	askedalloc += new_len;
 
-	return vml->vma->vm_start;
+	return vma->vm_start;
+}
+
+asmlinkage unsigned long sys_mremap(unsigned long addr,
+	unsigned long old_len, unsigned long new_len,
+	unsigned long flags, unsigned long new_addr)
+{
+	unsigned long ret;
+
+	down_write(&current->mm->mmap_sem);
+	ret = do_mremap(addr, old_len, new_len, flags, new_addr);
+	up_write(&current->mm->mmap_sem);
+	return ret;
 }
 
 struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
-- 
cgit v1.2.1


From 0112c4c6461fed12ed9bcb249f967bc17a23f6c4 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:21 -0700
Subject: [PATCH] NOMMU: Add docs about shared memory

Add documentation about using shared memory in NOMMU mode.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/nommu-mmap.txt | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 3ce890664936..4db7c18bf68c 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -129,6 +129,15 @@ FURTHER NOTES ON NO-MMU MMAP
      with character device files, pipes, fifos and sockets.
 
 
+==========================
+INTERPROCESS SHARED MEMORY
+==========================
+
+Both SYSV IPC SHM shared memory and POSIX shared memory is supported in NOMMU
+mode.  The former through the usual mechanism, the latter through files created
+on ramfs or tmpfs mounts.
+
+
 =============
 NO-MMU MREMAP
 =============
-- 
cgit v1.2.1


From 930e652a21a08986b03d1f370f933057dc0db2dc Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:22 -0700
Subject: [PATCH] NOMMU: Make futexes work under NOMMU conditions

Make futexes work under NOMMU conditions.

This can be tested by running this in one shell:

	#define SYSERROR(X, Y) \
		do { if ((long)(X) == -1L) { perror(Y); exit(1); }} while(0)

	int main()
	{
		int shmid, tmp, *f, n;

		shmid = shmget(23, 4, IPC_CREAT|0666);
		SYSERROR(shmid, "shmget");

		f = shmat(shmid, NULL, 0);
		SYSERROR(f, "shmat");

		n = *f;
		printf("WAIT: %p{%x}\n", f, n);
		tmp = futex(f, FUTEX_WAIT, n, NULL, NULL, 0);
		SYSERROR(tmp, "futex");
		printf("WAITED: %d\n", tmp);

		tmp = shmdt(f);
		SYSERROR(tmp, "shmdt");

		exit(0);
	}

And then this in the other shell:

	#define SYSERROR(X, Y) \
		do { if ((long)(X) == -1L) { perror(Y); exit(1); }} while(0)

	int main()
	{
		int shmid, tmp, *f;

		shmid = shmget(23, 4, IPC_CREAT|0666);
		SYSERROR(shmid, "shmget");

		f = shmat(shmid, NULL, 0);
		SYSERROR(f, "shmat");

		(*f)++;
		printf("WAKE: %p{%x}\n", f, *f);
		tmp = futex(f, FUTEX_WAKE, 1, NULL, NULL, 0);
		SYSERROR(tmp, "futex");
		printf("WOKE: %d\n", tmp);

		tmp = shmdt(f);
		SYSERROR(tmp, "shmdt");

		exit(0);
	}

The first program will set up a SYSV IPC SHM segment and wait on a futex in it
for the number at the start to change.  The program will increment that number
and wake the first program up.  This leads to output of the form:

	SHELL 1			SHELL 2
	=======================	=======================
	# /dowait
	WAIT: 0xc32ac000{0}
				# /dowake
				WAKE: 0xc32ac000{1}
	WAITED: 0		WOKE: 1

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/nommu-mmap.txt | 10 ++++++++++
 mm/nommu.c                   | 14 +++++++++-----
 2 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/Documentation/nommu-mmap.txt b/Documentation/nommu-mmap.txt
index 4db7c18bf68c..7714f57caad5 100644
--- a/Documentation/nommu-mmap.txt
+++ b/Documentation/nommu-mmap.txt
@@ -138,6 +138,16 @@ mode.  The former through the usual mechanism, the latter through files created
 on ramfs or tmpfs mounts.
 
 
+=======
+FUTEXES
+=======
+
+Futexes are supported in NOMMU mode if the arch supports them.  An error will
+be given if an address passed to the futex system call lies outside the
+mappings made by a process or if the mapping in which the address lies does not
+support futexes (such as an I/O chardev mapping).
+
+
 =============
 NO-MMU MREMAP
 =============
diff --git a/mm/nommu.c b/mm/nommu.c
index 23cfa8ec914a..564540662192 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -349,6 +349,15 @@ struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
 }
 EXPORT_SYMBOL(find_vma);
 
+/*
+ * find a VMA
+ * - we don't extend stack VMAs under NOMMU conditions
+ */
+struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
+{
+	return find_vma(mm, addr);
+}
+
 /*
  * look up the first VMA exactly that exactly matches addr
  * - should be called with mm->mmap_sem at least held readlocked
@@ -1153,11 +1162,6 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
 	return NULL;
 }
 
-struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
-{
-	return NULL;
-}
-
 int remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
 		unsigned long to, unsigned long size, pgprot_t prot)
 {
-- 
cgit v1.2.1


From f269fdd1829acc5e53bf57b145003e5733133f2b Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 27 Sep 2006 01:50:23 -0700
Subject: [PATCH] NOMMU: move the fallback arch_vma_name() to a sensible place

Move the fallback arch_vma_name() to a sensible place (kernel/signal.c).

Currently it's in fs/proc/task_mmu.c, a file that is dependent on both
CONFIG_PROC_FS and CONFIG_MMU being enabled, but it's used from
kernel/signal.c from where it is called unconditionally.

[akpm@osdl.org: build fix]
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/proc/task_mmu.c | 5 -----
 include/linux/mm.h | 2 +-
 kernel/signal.c    | 5 +++++
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 0a163a4f7764..6b769afac55a 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -122,11 +122,6 @@ struct mem_size_stats
 	unsigned long private_dirty;
 };
 
-__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma)
-{
-	return NULL;
-}
-
 static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats *mss)
 {
 	struct proc_maps_private *priv = m->private;
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 22165cb18906..7b703b6d4358 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1131,7 +1131,7 @@ void drop_slab(void);
 extern int randomize_va_space;
 #endif
 
-const char *arch_vma_name(struct vm_area_struct *vma);
+__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma);
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */
diff --git a/kernel/signal.c b/kernel/signal.c
index bfdb5686fa3e..05853a7337e3 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2577,6 +2577,11 @@ asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize)
 }
 #endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */
 
+__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma)
+{
+	return NULL;
+}
+
 void __init signals_init(void)
 {
 	sigqueue_cachep =
-- 
cgit v1.2.1


From a27f3113322edff36743014cc9e752a21ffc0324 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Wed, 27 Sep 2006 01:50:24 -0700
Subject: [PATCH] m32r: Fix "value computed not used" warnings

Fix to remove annoying gcc-4.1 warnings "value computed not used" for m32r;
Modify set_mb to cast to void for SMP.

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/system.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-m32r/system.h b/include/asm-m32r/system.h
index 9e618afec6ed..4ce0619f6989 100644
--- a/include/asm-m32r/system.h
+++ b/include/asm-m32r/system.h
@@ -328,15 +328,15 @@ __cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
 #define smp_rmb()	rmb()
 #define smp_wmb()	wmb()
 #define smp_read_barrier_depends()	read_barrier_depends()
+#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
 #else
 #define smp_mb()	barrier()
 #define smp_rmb()	barrier()
 #define smp_wmb()	barrier()
 #define smp_read_barrier_depends()	do { } while (0)
+#define set_mb(var, value) do { var = value; barrier(); } while (0)
 #endif
 
-#define set_mb(var, value) do { xchg(&var, value); } while (0)
-
 #define arch_align_stack(x) (x)
 
 #endif  /* _ASM_M32R_SYSTEM_H */
-- 
cgit v1.2.1


From 85f651794c46e8e3faf204a767d1caa7f9f278f0 Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata.hirokazu@renesas.com>
Date: Wed, 27 Sep 2006 01:50:24 -0700
Subject: [PATCH] m32r: revise __raw_read_trylock()

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/spinlock.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/asm-m32r/spinlock.h b/include/asm-m32r/spinlock.h
index f94c1a673569..f9f90727a4a1 100644
--- a/include/asm-m32r/spinlock.h
+++ b/include/asm-m32r/spinlock.h
@@ -298,7 +298,14 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw)
 	);
 }
 
-#define __raw_read_trylock(lock) generic__raw_read_trylock(lock)
+static inline int __raw_read_trylock(raw_rwlock_t *lock)
+{
+	atomic_t *count = (atomic_t*)lock;
+	if (atomic_dec_return(count) >= 0)
+		return 1;
+	atomic_inc(count);
+	return 0;
+}
 
 static inline int __raw_write_trylock(raw_rwlock_t *lock)
 {
-- 
cgit v1.2.1


From 06837504de7b4883e92af207dbbab4310d0db0ed Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Wed, 27 Sep 2006 01:50:27 -0700
Subject: [PATCH] uml: use -mcmodel=kernel for x86_64

We have never used this flag and recently one user experienced a complaining
warning about this (there was a symbol in the positive half of the address space
IIRC). So fix it.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/Makefile-x86_64 | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/um/Makefile-x86_64 b/arch/um/Makefile-x86_64
index 9558a7cf34d5..11154b6773ec 100644
--- a/arch/um/Makefile-x86_64
+++ b/arch/um/Makefile-x86_64
@@ -4,10 +4,13 @@
 core-y += arch/um/sys-x86_64/
 START := 0x60000000
 
+_extra_flags_ = -fno-builtin -m64 -mcmodel=kernel
+
 #We #undef __x86_64__ for kernelspace, not for userspace where
 #it's needed for headers to work!
-CFLAGS += -U__$(SUBARCH)__ -fno-builtin -m64
-USER_CFLAGS += -fno-builtin -m64
+CFLAGS += -U__$(SUBARCH)__ $(_extra_flags_)
+USER_CFLAGS += $(_extra_flags_)
+
 CHECKFLAGS  += -m64
 AFLAGS += -m64
 LDFLAGS += -m elf_x86_64
-- 
cgit v1.2.1


From 48af05ed54ddf8dc6eceea4f009e063d7e784b37 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Wed, 27 Sep 2006 01:50:31 -0700
Subject: [PATCH] uml: fix proc-vs-interrupt context spinlock deadlock

This spinlock can be taken on interrupt too, so spin_lock_irq[save] must be
used.

However, Documentation/networking/netdevices.txt explains we are called with
rtnl_lock() held - so we don't need to care about other concurrent opens.
Verified also in LDD3 and by direct checking.  Also verified that the network
layer (through a state machine) guarantees us that nobody will close the
interface while it's being used.  Please correct me if I'm wrong.

Also, we must check we don't sleep with irqs disabled!!!  But anyway, this is
not news - we already can't sleep while holding a spinlock.  Who says this is
guaranted really by the present code?

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/net_kern.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 4a7966b21931..657dfacd5ba8 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -114,8 +114,6 @@ static int uml_net_open(struct net_device *dev)
 	struct uml_net_private *lp = dev->priv;
 	int err;
 
-	spin_lock(&lp->lock);
-
 	if(lp->fd >= 0){
 		err = -ENXIO;
 		goto out;
@@ -149,8 +147,6 @@ static int uml_net_open(struct net_device *dev)
 	 */
 	while((err = uml_net_rx(dev)) > 0) ;
 
-	spin_unlock(&lp->lock);
-
 	spin_lock(&opened_lock);
 	list_add(&lp->list, &opened);
 	spin_unlock(&opened_lock);
@@ -160,7 +156,6 @@ out_close:
 	if(lp->close != NULL) (*lp->close)(lp->fd, &lp->user);
 	lp->fd = -1;
 out:
-	spin_unlock(&lp->lock);
 	return err;
 }
 
@@ -169,15 +164,12 @@ static int uml_net_close(struct net_device *dev)
 	struct uml_net_private *lp = dev->priv;
 	
 	netif_stop_queue(dev);
-	spin_lock(&lp->lock);
 
 	free_irq(dev->irq, dev);
 	if(lp->close != NULL)
 		(*lp->close)(lp->fd, &lp->user);
 	lp->fd = -1;
 
-	spin_unlock(&lp->lock);
-
 	spin_lock(&opened_lock);
 	list_del(&lp->list);
 	spin_unlock(&opened_lock);
@@ -246,9 +238,9 @@ static int uml_net_set_mac(struct net_device *dev, void *addr)
 	struct uml_net_private *lp = dev->priv;
 	struct sockaddr *hwaddr = addr;
 
-	spin_lock(&lp->lock);
+	spin_lock_irq(&lp->lock);
 	set_ether_mac(dev, hwaddr->sa_data);
-	spin_unlock(&lp->lock);
+	spin_unlock_irq(&lp->lock);
 
 	return(0);
 }
@@ -258,7 +250,7 @@ static int uml_net_change_mtu(struct net_device *dev, int new_mtu)
 	struct uml_net_private *lp = dev->priv;
 	int err = 0;
 
-	spin_lock(&lp->lock);
+	spin_lock_irq(&lp->lock);
 
 	new_mtu = (*lp->set_mtu)(new_mtu, &lp->user);
 	if(new_mtu < 0){
@@ -269,7 +261,7 @@ static int uml_net_change_mtu(struct net_device *dev, int new_mtu)
 	dev->mtu = new_mtu;
 
  out:
-	spin_unlock(&lp->lock);
+	spin_unlock_irq(&lp->lock);
 	return err;
 }
 
-- 
cgit v1.2.1


From 5e7672ec3f059f764fcc5c78216e24bb16c44dba Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:33 -0700
Subject: [PATCH] uml: const more data

Make lots of structures const in order to make it obvious that they need no
locking.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/chan_kern.c              | 16 ++++++++--------
 arch/um/drivers/daemon.h                 |  2 +-
 arch/um/drivers/daemon_kern.c            |  2 +-
 arch/um/drivers/daemon_user.c            |  2 +-
 arch/um/drivers/fd.c                     |  4 ++--
 arch/um/drivers/hostaudio_kern.c         |  4 ++--
 arch/um/drivers/line.c                   |  6 +++---
 arch/um/drivers/mcast.h                  |  2 +-
 arch/um/drivers/mcast_kern.c             |  2 +-
 arch/um/drivers/mcast_user.c             |  2 +-
 arch/um/drivers/mmapper_kern.c           |  4 ++--
 arch/um/drivers/null.c                   |  4 ++--
 arch/um/drivers/pcap_kern.c              |  2 +-
 arch/um/drivers/pcap_user.c              |  2 +-
 arch/um/drivers/port_user.c              |  4 ++--
 arch/um/drivers/pty.c                    |  6 +++---
 arch/um/drivers/random.c                 |  2 +-
 arch/um/drivers/slip.h                   |  2 +-
 arch/um/drivers/slip_kern.c              |  2 +-
 arch/um/drivers/slip_user.c              |  2 +-
 arch/um/drivers/slirp.h                  |  2 +-
 arch/um/drivers/slirp_kern.c             |  2 +-
 arch/um/drivers/slirp_user.c             |  2 +-
 arch/um/drivers/ssl.c                    |  4 ++--
 arch/um/drivers/stdio_console.c          |  2 +-
 arch/um/drivers/tty.c                    |  4 ++--
 arch/um/drivers/xterm.c                  |  4 ++--
 arch/um/include/chan_kern.h              |  6 +++---
 arch/um/include/chan_user.h              |  6 +++---
 arch/um/include/kern_util.h              |  2 +-
 arch/um/include/line.h                   |  4 ++--
 arch/um/include/net_kern.h               |  4 ++--
 arch/um/kernel/trap.c                    |  2 +-
 arch/um/kernel/um_arch.c                 |  2 +-
 arch/um/os-Linux/drivers/etap.h          |  2 +-
 arch/um/os-Linux/drivers/ethertap_kern.c |  2 +-
 arch/um/os-Linux/drivers/ethertap_user.c |  2 +-
 arch/um/os-Linux/drivers/tuntap.h        |  2 +-
 arch/um/os-Linux/drivers/tuntap_kern.c   |  2 +-
 arch/um/os-Linux/drivers/tuntap_user.c   |  2 +-
 40 files changed, 65 insertions(+), 65 deletions(-)

diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c
index e82764f75e7f..3576b3cc505e 100644
--- a/arch/um/drivers/chan_kern.c
+++ b/arch/um/drivers/chan_kern.c
@@ -110,7 +110,7 @@ static void not_configged_free(void *data)
 	       "UML\n");
 }
 
-static struct chan_ops not_configged_ops = {
+static const struct chan_ops not_configged_ops = {
 	.init		= not_configged_init,
 	.open		= not_configged_open,
 	.close		= not_configged_close,
@@ -373,7 +373,7 @@ int console_write_chan(struct list_head *chans, const char *buf, int len)
 }
 
 int console_open_chan(struct line *line, struct console *co,
-		      struct chan_opts *opts)
+		      const struct chan_opts *opts)
 {
 	int err;
 
@@ -494,10 +494,10 @@ int chan_config_string(struct list_head *chans, char *str, int size,
 
 struct chan_type {
 	char *key;
-	struct chan_ops *ops;
+	const struct chan_ops *ops;
 };
 
-static struct chan_type chan_table[] = {
+static const struct chan_type chan_table[] = {
 	{ "fd", &fd_ops },
 
 #ifdef CONFIG_NULL_CHAN
@@ -534,10 +534,10 @@ static struct chan_type chan_table[] = {
 };
 
 static struct chan *parse_chan(struct line *line, char *str, int device,
-			       struct chan_opts *opts)
+			       const struct chan_opts *opts)
 {
-	struct chan_type *entry;
-	struct chan_ops *ops;
+	const struct chan_type *entry;
+	const struct chan_ops *ops;
 	struct chan *chan;
 	void *data;
 	int i;
@@ -582,7 +582,7 @@ static struct chan *parse_chan(struct line *line, char *str, int device,
 }
 
 int parse_chan_pair(char *str, struct line *line, int device,
-		    struct chan_opts *opts)
+		    const struct chan_opts *opts)
 {
 	struct list_head *chans = &line->chan_list;
 	struct chan *new, *chan;
diff --git a/arch/um/drivers/daemon.h b/arch/um/drivers/daemon.h
index 7326c42f7ef9..3bc3cf6b94aa 100644
--- a/arch/um/drivers/daemon.h
+++ b/arch/um/drivers/daemon.h
@@ -18,7 +18,7 @@ struct daemon_data {
 	void *dev;
 };
 
-extern struct net_user_info daemon_user_info;
+extern const struct net_user_info daemon_user_info;
 
 extern int daemon_user_write(int fd, void *buf, int len, 
 			     struct daemon_data *pri);
diff --git a/arch/um/drivers/daemon_kern.c b/arch/um/drivers/daemon_kern.c
index 53d09ed78b42..824386974f88 100644
--- a/arch/um/drivers/daemon_kern.c
+++ b/arch/um/drivers/daemon_kern.c
@@ -57,7 +57,7 @@ static int daemon_write(int fd, struct sk_buff **skb,
 				 (struct daemon_data *) &lp->user));
 }
 
-static struct net_kern_info daemon_kern_info = {
+static const struct net_kern_info daemon_kern_info = {
 	.init			= daemon_init,
 	.protocol		= eth_protocol,
 	.read			= daemon_read,
diff --git a/arch/um/drivers/daemon_user.c b/arch/um/drivers/daemon_user.c
index c944265955e2..77954ea77043 100644
--- a/arch/um/drivers/daemon_user.c
+++ b/arch/um/drivers/daemon_user.c
@@ -182,7 +182,7 @@ static int daemon_set_mtu(int mtu, void *data)
 	return(mtu);
 }
 
-struct net_user_info daemon_user_info = {
+const struct net_user_info daemon_user_info = {
 	.init		= daemon_user_init,
 	.open		= daemon_open,
 	.close	 	= NULL,
diff --git a/arch/um/drivers/fd.c b/arch/um/drivers/fd.c
index c41f75e4acb5..108b7dafbd0e 100644
--- a/arch/um/drivers/fd.c
+++ b/arch/um/drivers/fd.c
@@ -20,7 +20,7 @@ struct fd_chan {
 	char str[sizeof("1234567890\0")];
 };
 
-static void *fd_init(char *str, int device, struct chan_opts *opts)
+static void *fd_init(char *str, int device, const struct chan_opts *opts)
 {
 	struct fd_chan *data;
 	char *end;
@@ -77,7 +77,7 @@ static void fd_close(int fd, void *d)
 	}
 }
 
-struct chan_ops fd_ops = {
+const struct chan_ops fd_ops = {
 	.type		= "fd",
 	.init		= fd_init,
 	.open		= fd_open,
diff --git a/arch/um/drivers/hostaudio_kern.c b/arch/um/drivers/hostaudio_kern.c
index 37232f908cd7..d247ef45c374 100644
--- a/arch/um/drivers/hostaudio_kern.c
+++ b/arch/um/drivers/hostaudio_kern.c
@@ -280,7 +280,7 @@ static int hostmixer_release(struct inode *inode, struct file *file)
 
 /* kernel module operations */
 
-static struct file_operations hostaudio_fops = {
+static const struct file_operations hostaudio_fops = {
         .owner          = THIS_MODULE,
         .llseek         = no_llseek,
         .read           = hostaudio_read,
@@ -292,7 +292,7 @@ static struct file_operations hostaudio_fops = {
         .release        = hostaudio_release,
 };
 
-static struct file_operations hostmixer_fops = {
+static const struct file_operations hostmixer_fops = {
         .owner          = THIS_MODULE,
         .llseek         = no_llseek,
         .ioctl          = hostmixer_ioctl_mixdev,
diff --git a/arch/um/drivers/line.c b/arch/um/drivers/line.c
index ebebaabb78ad..563ce7690a1e 100644
--- a/arch/um/drivers/line.c
+++ b/arch/um/drivers/line.c
@@ -251,7 +251,7 @@ void line_set_termios(struct tty_struct *tty, struct termios * old)
 	/* nothing */
 }
 
-static struct {
+static const struct {
 	int  cmd;
 	char *level;
 	char *name;
@@ -405,7 +405,7 @@ static irqreturn_t line_write_interrupt(int irq, void *data,
 
 int line_setup_irq(int fd, int input, int output, struct line *line, void *data)
 {
-	struct line_driver *driver = line->driver;
+	const struct line_driver *driver = line->driver;
 	int err = 0, flags = IRQF_DISABLED | IRQF_SHARED | IRQF_SAMPLE_RANDOM;
 
 	if (input)
@@ -558,7 +558,7 @@ int line_setup(struct line *lines, unsigned int num, char *init)
 }
 
 int line_config(struct line *lines, unsigned int num, char *str,
-		struct chan_opts *opts)
+		const struct chan_opts *opts)
 {
 	struct line *line;
 	char *new;
diff --git a/arch/um/drivers/mcast.h b/arch/um/drivers/mcast.h
index a2c6db243458..bc56af9d3e53 100644
--- a/arch/um/drivers/mcast.h
+++ b/arch/um/drivers/mcast.h
@@ -13,7 +13,7 @@ struct mcast_data {
 	void *dev;
 };
 
-extern struct net_user_info mcast_user_info;
+extern const struct net_user_info mcast_user_info;
 
 extern int mcast_user_write(int fd, void *buf, int len, 
 			    struct mcast_data *pri);
diff --git a/arch/um/drivers/mcast_kern.c b/arch/um/drivers/mcast_kern.c
index 3a7af18cf944..c090fbd464e7 100644
--- a/arch/um/drivers/mcast_kern.c
+++ b/arch/um/drivers/mcast_kern.c
@@ -61,7 +61,7 @@ static int mcast_write(int fd, struct sk_buff **skb,
 				 (struct mcast_data *) &lp->user);
 }
 
-static struct net_kern_info mcast_kern_info = {
+static const struct net_kern_info mcast_kern_info = {
 	.init			= mcast_init,
 	.protocol		= eth_protocol,
 	.read			= mcast_read,
diff --git a/arch/um/drivers/mcast_user.c b/arch/um/drivers/mcast_user.c
index afe85bfa66e0..4d2bd39a85bc 100644
--- a/arch/um/drivers/mcast_user.c
+++ b/arch/um/drivers/mcast_user.c
@@ -152,7 +152,7 @@ static int mcast_set_mtu(int mtu, void *data)
 	return(mtu);
 }
 
-struct net_user_info mcast_user_info = {
+const struct net_user_info mcast_user_info = {
 	.init		= mcast_user_init,
 	.open		= mcast_open,
 	.close	 	= mcast_close,
diff --git a/arch/um/drivers/mmapper_kern.c b/arch/um/drivers/mmapper_kern.c
index 022f67bb6873..9a3b5daf6250 100644
--- a/arch/um/drivers/mmapper_kern.c
+++ b/arch/um/drivers/mmapper_kern.c
@@ -85,7 +85,7 @@ mmapper_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static struct file_operations mmapper_fops = {
+static const struct file_operations mmapper_fops = {
 	.owner		= THIS_MODULE,
 	.read		= mmapper_read,
 	.write		= mmapper_write,
@@ -95,7 +95,7 @@ static struct file_operations mmapper_fops = {
 	.release	= mmapper_release,
 };
 
-static struct miscdevice mmapper_dev = {
+static const struct miscdevice mmapper_dev = {
 	.minor		= MISC_DYNAMIC_MINOR,
 	.name		= "mmapper",
 	.fops		= &mmapper_fops
diff --git a/arch/um/drivers/null.c b/arch/um/drivers/null.c
index 14cc5f78398a..3683ed44315d 100644
--- a/arch/um/drivers/null.c
+++ b/arch/um/drivers/null.c
@@ -10,7 +10,7 @@
 
 static int null_chan;
 
-static void *null_init(char *str, int device, struct chan_opts *opts)
+static void *null_init(char *str, int device, const struct chan_opts *opts)
 {
 	return(&null_chan);
 }
@@ -31,7 +31,7 @@ static void null_free(void *data)
 {
 }
 
-struct chan_ops null_ops = {
+const struct chan_ops null_ops = {
 	.type		= "null",
 	.init		= null_init,
 	.open		= null_open,
diff --git a/arch/um/drivers/pcap_kern.c b/arch/um/drivers/pcap_kern.c
index 4c767c7adb96..6e1ef8558283 100644
--- a/arch/um/drivers/pcap_kern.c
+++ b/arch/um/drivers/pcap_kern.c
@@ -46,7 +46,7 @@ static int pcap_write(int fd, struct sk_buff **skb, struct uml_net_private *lp)
 	return(-EPERM);
 }
 
-static struct net_kern_info pcap_kern_info = {
+static const struct net_kern_info pcap_kern_info = {
 	.init			= pcap_init,
 	.protocol		= eth_protocol,
 	.read			= pcap_read,
diff --git a/arch/um/drivers/pcap_user.c b/arch/um/drivers/pcap_user.c
index edfcb29273e1..2ef641ded960 100644
--- a/arch/um/drivers/pcap_user.c
+++ b/arch/um/drivers/pcap_user.c
@@ -120,7 +120,7 @@ int pcap_user_read(int fd, void *buffer, int len, struct pcap_data *pri)
 	return(hdata.len);
 }
 
-struct net_user_info pcap_user_info = {
+const struct net_user_info pcap_user_info = {
 	.init		= pcap_user_init,
 	.open		= pcap_open,
 	.close	 	= NULL,
diff --git a/arch/um/drivers/port_user.c b/arch/um/drivers/port_user.c
index c43e8bb32502..f2e8fc42ecc2 100644
--- a/arch/um/drivers/port_user.c
+++ b/arch/um/drivers/port_user.c
@@ -27,7 +27,7 @@ struct port_chan {
 	char dev[sizeof("32768\0")];
 };
 
-static void *port_init(char *str, int device, struct chan_opts *opts)
+static void *port_init(char *str, int device, const struct chan_opts *opts)
 {
 	struct port_chan *data;
 	void *kern_data;
@@ -100,7 +100,7 @@ static void port_close(int fd, void *d)
 	os_close_file(fd);
 }
 
-struct chan_ops port_ops = {
+const struct chan_ops port_ops = {
 	.type		= "port",
 	.init		= port_init,
 	.open		= port_open,
diff --git a/arch/um/drivers/pty.c b/arch/um/drivers/pty.c
index 1c555c38de4d..abec620e8380 100644
--- a/arch/um/drivers/pty.c
+++ b/arch/um/drivers/pty.c
@@ -22,7 +22,7 @@ struct pty_chan {
 	char dev_name[sizeof("/dev/pts/0123456\0")];
 };
 
-static void *pty_chan_init(char *str, int device, struct chan_opts *opts)
+static void *pty_chan_init(char *str, int device, const struct chan_opts *opts)
 {
 	struct pty_chan *data;
 
@@ -118,7 +118,7 @@ static int pty_open(int input, int output, int primary, void *d,
 	return(fd);
 }
 
-struct chan_ops pty_ops = {
+const struct chan_ops pty_ops = {
 	.type		= "pty",
 	.init		= pty_chan_init,
 	.open		= pty_open,
@@ -131,7 +131,7 @@ struct chan_ops pty_ops = {
 	.winch		= 0,
 };
 
-struct chan_ops pts_ops = {
+const struct chan_ops pts_ops = {
 	.type		= "pts",
 	.init		= pty_chan_init,
 	.open		= pts_open,
diff --git a/arch/um/drivers/random.c b/arch/um/drivers/random.c
index ba471f5864a6..ae9909415b9c 100644
--- a/arch/um/drivers/random.c
+++ b/arch/um/drivers/random.c
@@ -68,7 +68,7 @@ static ssize_t rng_dev_read (struct file *filp, char __user *buf, size_t size,
 	return ret;
 }
 
-static struct file_operations rng_chrdev_ops = {
+static const struct file_operations rng_chrdev_ops = {
 	.owner		= THIS_MODULE,
 	.open		= rng_dev_open,
 	.read		= rng_dev_read,
diff --git a/arch/um/drivers/slip.h b/arch/um/drivers/slip.h
index bb0dab41c2e4..c64f8c61d274 100644
--- a/arch/um/drivers/slip.h
+++ b/arch/um/drivers/slip.h
@@ -12,7 +12,7 @@ struct slip_data {
 	struct slip_proto slip;
 };
 
-extern struct net_user_info slip_user_info;
+extern const struct net_user_info slip_user_info;
 
 extern int slip_user_read(int fd, void *buf, int len, struct slip_data *pri);
 extern int slip_user_write(int fd, void *buf, int len, struct slip_data *pri);
diff --git a/arch/um/drivers/slip_kern.c b/arch/um/drivers/slip_kern.c
index 163ee0d5f75e..ccea2d7885e5 100644
--- a/arch/um/drivers/slip_kern.c
+++ b/arch/um/drivers/slip_kern.c
@@ -61,7 +61,7 @@ static int slip_write(int fd, struct sk_buff **skb,
 			       (struct slip_data *) &lp->user));
 }
 
-struct net_kern_info slip_kern_info = {
+const struct net_kern_info slip_kern_info = {
 	.init			= slip_init,
 	.protocol		= slip_protocol,
 	.read			= slip_read,
diff --git a/arch/um/drivers/slip_user.c b/arch/um/drivers/slip_user.c
index 89fbec185cc1..8460285c69a5 100644
--- a/arch/um/drivers/slip_user.c
+++ b/arch/um/drivers/slip_user.c
@@ -241,7 +241,7 @@ static void slip_del_addr(unsigned char *addr, unsigned char *netmask,
 	close_addr(addr, netmask, pri->name);
 }
 
-struct net_user_info slip_user_info = {
+const struct net_user_info slip_user_info = {
 	.init		= slip_user_init,
 	.open		= slip_open,
 	.close	 	= slip_close,
diff --git a/arch/um/drivers/slirp.h b/arch/um/drivers/slirp.h
index 6cf88ab580c9..89ccf83b7577 100644
--- a/arch/um/drivers/slirp.h
+++ b/arch/um/drivers/slirp.h
@@ -24,7 +24,7 @@ struct slirp_data {
 	struct slip_proto slip;
 };
 
-extern struct net_user_info slirp_user_info;
+extern const struct net_user_info slirp_user_info;
 
 extern int slirp_user_read(int fd, void *buf, int len, struct slirp_data *pri);
 extern int slirp_user_write(int fd, void *buf, int len,
diff --git a/arch/um/drivers/slirp_kern.c b/arch/um/drivers/slirp_kern.c
index 95e50c943e14..ae322e1c8a87 100644
--- a/arch/um/drivers/slirp_kern.c
+++ b/arch/um/drivers/slirp_kern.c
@@ -64,7 +64,7 @@ static int slirp_write(int fd, struct sk_buff **skb,
 			       (struct slirp_data *) &lp->user));
 }
 
-struct net_kern_info slirp_kern_info = {
+const struct net_kern_info slirp_kern_info = {
 	.init			= slirp_init,
 	.protocol		= slirp_protocol,
 	.read			= slirp_read,
diff --git a/arch/um/drivers/slirp_user.c b/arch/um/drivers/slirp_user.c
index 33c5f6e625e8..ce5e85d1de3d 100644
--- a/arch/um/drivers/slirp_user.c
+++ b/arch/um/drivers/slirp_user.c
@@ -126,7 +126,7 @@ static int slirp_set_mtu(int mtu, void *data)
 	return(mtu);
 }
 
-struct net_user_info slirp_user_info = {
+const struct net_user_info slirp_user_info = {
 	.init		= slirp_user_init,
 	.open		= slirp_open,
 	.close	 	= slirp_close,
diff --git a/arch/um/drivers/ssl.c b/arch/um/drivers/ssl.c
index 6dafd6fbfdae..6f13e7c71a82 100644
--- a/arch/um/drivers/ssl.c
+++ b/arch/um/drivers/ssl.c
@@ -23,7 +23,7 @@
 #include "irq_user.h"
 #include "mconsole_kern.h"
 
-static int ssl_version = 1;
+static const int ssl_version = 1;
 
 /* Referenced only by tty_driver below - presumably it's locked correctly
  * by the tty driver.
@@ -123,7 +123,7 @@ void ssl_hangup(struct tty_struct *tty)
 }
 #endif
 
-static struct tty_operations ssl_ops = {
+static const struct tty_operations ssl_ops = {
 	.open 	 		= ssl_open,
 	.close 	 		= line_close,
 	.write 	 		= line_write,
diff --git a/arch/um/drivers/stdio_console.c b/arch/um/drivers/stdio_console.c
index 856f568c2687..5e44adb07051 100644
--- a/arch/um/drivers/stdio_console.c
+++ b/arch/um/drivers/stdio_console.c
@@ -110,7 +110,7 @@ static int con_open(struct tty_struct *tty, struct file *filp)
 
 static int con_init_done = 0;
 
-static struct tty_operations console_ops = {
+static const struct tty_operations console_ops = {
 	.open 	 		= con_open,
 	.close 	 		= line_close,
 	.write 	 		= line_write,
diff --git a/arch/um/drivers/tty.c b/arch/um/drivers/tty.c
index 9f70edf5d8ef..11de3ac1eb5c 100644
--- a/arch/um/drivers/tty.c
+++ b/arch/um/drivers/tty.c
@@ -18,7 +18,7 @@ struct tty_chan {
 	struct termios tt;
 };
 
-static void *tty_chan_init(char *str, int device, struct chan_opts *opts)
+static void *tty_chan_init(char *str, int device, const struct chan_opts *opts)
 {
 	struct tty_chan *data;
 
@@ -62,7 +62,7 @@ static int tty_open(int input, int output, int primary, void *d,
 	return fd;
 }
 
-struct chan_ops tty_ops = {
+const struct chan_ops tty_ops = {
 	.type		= "tty",
 	.init		= tty_chan_init,
 	.open		= tty_open,
diff --git a/arch/um/drivers/xterm.c b/arch/um/drivers/xterm.c
index aaa636661043..386f8b952982 100644
--- a/arch/um/drivers/xterm.c
+++ b/arch/um/drivers/xterm.c
@@ -31,7 +31,7 @@ struct xterm_chan {
 };
 
 /* Not static because it's called directly by the tt mode gdb code */
-void *xterm_init(char *str, int device, struct chan_opts *opts)
+void *xterm_init(char *str, int device, const struct chan_opts *opts)
 {
 	struct xterm_chan *data;
 
@@ -194,7 +194,7 @@ static void xterm_free(void *d)
 	free(d);
 }
 
-struct chan_ops xterm_ops = {
+const struct chan_ops xterm_ops = {
 	.type		= "xterm",
 	.init		= xterm_init,
 	.open		= xterm_open,
diff --git a/arch/um/include/chan_kern.h b/arch/um/include/chan_kern.h
index 1bb5e9d94270..572d286ed2c6 100644
--- a/arch/um/include/chan_kern.h
+++ b/arch/um/include/chan_kern.h
@@ -23,21 +23,21 @@ struct chan {
 	unsigned int opened:1;
 	unsigned int enabled:1;
 	int fd;
-	struct chan_ops *ops;
+	const struct chan_ops *ops;
 	void *data;
 };
 
 extern void chan_interrupt(struct list_head *chans, struct work_struct *task,
 			   struct tty_struct *tty, int irq);
 extern int parse_chan_pair(char *str, struct line *line, int device,
-			   struct chan_opts *opts);
+			   const struct chan_opts *opts);
 extern int open_chan(struct list_head *chans);
 extern int write_chan(struct list_head *chans, const char *buf, int len,
 			     int write_irq);
 extern int console_write_chan(struct list_head *chans, const char *buf, 
 			      int len);
 extern int console_open_chan(struct line *line, struct console *co,
-			     struct chan_opts *opts);
+			     const struct chan_opts *opts);
 extern void deactivate_chan(struct list_head *chans, int irq);
 extern void reactivate_chan(struct list_head *chans, int irq);
 extern void chan_enable_winch(struct list_head *chans, struct tty_struct *tty);
diff --git a/arch/um/include/chan_user.h b/arch/um/include/chan_user.h
index 659bb3cac32f..a795547a1dbd 100644
--- a/arch/um/include/chan_user.h
+++ b/arch/um/include/chan_user.h
@@ -20,7 +20,7 @@ enum chan_init_pri { INIT_STATIC, INIT_ALL, INIT_ONE };
 
 struct chan_ops {
 	char *type;
-	void *(*init)(char *, int, struct chan_opts *);
+	void *(*init)(char *, int, const struct chan_opts *);
 	int (*open)(int, int, int, void *, char **);
 	void (*close)(int, void *);
 	int (*read)(int, char *, void *);
@@ -31,8 +31,8 @@ struct chan_ops {
 	int winch;
 };
 
-extern struct chan_ops fd_ops, null_ops, port_ops, pts_ops, pty_ops, tty_ops,
-	xterm_ops;
+extern const struct chan_ops fd_ops, null_ops, port_ops, pts_ops, pty_ops,
+	tty_ops, xterm_ops;
 
 extern void generic_close(int fd, void *unused);
 extern int generic_read(int fd, char *c_out, void *unused);
diff --git a/arch/um/include/kern_util.h b/arch/um/include/kern_util.h
index 89e1dc835a5b..59cfa9e0cad0 100644
--- a/arch/um/include/kern_util.h
+++ b/arch/um/include/kern_util.h
@@ -21,7 +21,7 @@ struct kern_handlers {
 	kern_hndl timer_handler;
 };
 
-extern struct kern_handlers handlinfo_kern;
+extern const struct kern_handlers handlinfo_kern;
 
 extern int ncpus;
 extern char *linux_prog;
diff --git a/arch/um/include/line.h b/arch/um/include/line.h
index 27bf2f6fbc05..642c9a0320f9 100644
--- a/arch/um/include/line.h
+++ b/arch/um/include/line.h
@@ -52,7 +52,7 @@ struct line {
 
 	int sigio;
 	struct work_struct task;
-	struct line_driver *driver;
+	const struct line_driver *driver;
 	int have_irq;
 };
 
@@ -99,7 +99,7 @@ extern void lines_init(struct line *lines, int nlines, struct chan_opts *opts);
 extern void close_lines(struct line *lines, int nlines);
 
 extern int line_config(struct line *lines, unsigned int sizeof_lines,
-		       char *str, struct chan_opts *opts);
+		       char *str, const struct chan_opts *opts);
 extern int line_id(char **str, int *start_out, int *end_out);
 extern int line_remove(struct line *lines, unsigned int sizeof_lines, int n);
 extern int line_get_config(char *dev, struct line *lines,
diff --git a/arch/um/include/net_kern.h b/arch/um/include/net_kern.h
index f7de6df60dd7..769fba43ee03 100644
--- a/arch/um/include/net_kern.h
+++ b/arch/um/include/net_kern.h
@@ -54,8 +54,8 @@ struct transport {
 	struct list_head list;
 	char *name;
 	int (*setup)(char *, char **, void *);
-	struct net_user_info *user;
-	struct net_kern_info *kern;
+	const struct net_user_info *user;
+	const struct net_kern_info *kern;
 	int private_size;
 	int setup_size;
 };
diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index e5eeaf2b6af1..e44ff6eca496 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -140,7 +140,7 @@ void segv_handler(int sig, union uml_pt_regs *regs)
 	segv(*fi, UPT_IP(regs), UPT_IS_USER(regs), regs);
 }
 
-struct kern_handlers handlinfo_kern = {
+const struct kern_handlers handlinfo_kern = {
 	.relay_signal = relay_signal,
 	.winch = winch,
 	.bus_handler = relay_signal,
diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c
index 7896cf98232d..55005710dcbb 100644
--- a/arch/um/kernel/um_arch.c
+++ b/arch/um/kernel/um_arch.c
@@ -106,7 +106,7 @@ static void c_stop(struct seq_file *m, void *v)
 {
 }
 
-struct seq_operations cpuinfo_op = {
+const struct seq_operations cpuinfo_op = {
 	.start	= c_start,
 	.next	= c_next,
 	.stop	= c_stop,
diff --git a/arch/um/os-Linux/drivers/etap.h b/arch/um/os-Linux/drivers/etap.h
index b84f6c4740f7..57ecdaf2f67e 100644
--- a/arch/um/os-Linux/drivers/etap.h
+++ b/arch/um/os-Linux/drivers/etap.h
@@ -13,7 +13,7 @@ struct ethertap_data {
 	void *dev;
 };
 
-extern struct net_user_info ethertap_user_info;
+extern const struct net_user_info ethertap_user_info;
 
 /*
  * Overrides for Emacs so that we follow Linus's tabbing style.
diff --git a/arch/um/os-Linux/drivers/ethertap_kern.c b/arch/um/os-Linux/drivers/ethertap_kern.c
index 768606bec233..16385e2ada85 100644
--- a/arch/um/os-Linux/drivers/ethertap_kern.c
+++ b/arch/um/os-Linux/drivers/ethertap_kern.c
@@ -65,7 +65,7 @@ static int etap_write(int fd, struct sk_buff **skb, struct uml_net_private *lp)
 	return(net_send(fd, (*skb)->data, (*skb)->len));
 }
 
-struct net_kern_info ethertap_kern_info = {
+const struct net_kern_info ethertap_kern_info = {
 	.init			= etap_init,
 	.protocol		= eth_protocol,
 	.read			= etap_read,
diff --git a/arch/um/os-Linux/drivers/ethertap_user.c b/arch/um/os-Linux/drivers/ethertap_user.c
index 8f49507e64ef..f559bdf746e6 100644
--- a/arch/um/os-Linux/drivers/ethertap_user.c
+++ b/arch/um/os-Linux/drivers/ethertap_user.c
@@ -216,7 +216,7 @@ static void etap_del_addr(unsigned char *addr, unsigned char *netmask,
 	etap_close_addr(addr, netmask, &pri->control_fd);
 }
 
-struct net_user_info ethertap_user_info = {
+const struct net_user_info ethertap_user_info = {
 	.init		= etap_user_init,
 	.open		= etap_open,
 	.close	 	= etap_close,
diff --git a/arch/um/os-Linux/drivers/tuntap.h b/arch/um/os-Linux/drivers/tuntap.h
index 25d4a2868814..d3e8d3af6245 100644
--- a/arch/um/os-Linux/drivers/tuntap.h
+++ b/arch/um/os-Linux/drivers/tuntap.h
@@ -16,7 +16,7 @@ struct tuntap_data {
 	void *dev;
 };
 
-extern struct net_user_info tuntap_user_info;
+extern const struct net_user_info tuntap_user_info;
 
 #endif
 
diff --git a/arch/um/os-Linux/drivers/tuntap_kern.c b/arch/um/os-Linux/drivers/tuntap_kern.c
index 190009a6f89c..0edbac63c527 100644
--- a/arch/um/os-Linux/drivers/tuntap_kern.c
+++ b/arch/um/os-Linux/drivers/tuntap_kern.c
@@ -53,7 +53,7 @@ static int tuntap_write(int fd, struct sk_buff **skb,
 	return(net_write(fd, (*skb)->data, (*skb)->len));
 }
 
-struct net_kern_info tuntap_kern_info = {
+const struct net_kern_info tuntap_kern_info = {
 	.init			= tuntap_init,
 	.protocol		= eth_protocol,
 	.read			= tuntap_read,
diff --git a/arch/um/os-Linux/drivers/tuntap_user.c b/arch/um/os-Linux/drivers/tuntap_user.c
index 87c3aa0252db..e846b23f7558 100644
--- a/arch/um/os-Linux/drivers/tuntap_user.c
+++ b/arch/um/os-Linux/drivers/tuntap_user.c
@@ -205,7 +205,7 @@ static int tuntap_set_mtu(int mtu, void *data)
 	return(mtu);
 }
 
-struct net_user_info tuntap_user_info = {
+const struct net_user_info tuntap_user_info = {
 	.init		= tuntap_user_init,
 	.open		= tuntap_open,
 	.close	 	= tuntap_close,
-- 
cgit v1.2.1


From 0715501bf1d915002d92e34e8a78ea889e5a0049 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:34 -0700
Subject: [PATCH] uml: get rid of ZONE_DMA use

ZONE_DMA might become dependent on CONFIG_ZONE_DMA, which UML doesn't define
(we're still arguing about this) So, let's change ZONE_DMA to ZONE_NORMAL.

This is prompted by optional-zone_dma-in-the-vm.patch, but should be harmless
on its own.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/mem.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index 93121c6d26e5..c95855ba6ab5 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -226,7 +226,8 @@ void paging_init(void)
 	for(i = 0; i < ARRAY_SIZE(zones_size); i++)
 		zones_size[i] = 0;
 
-	zones_size[ZONE_DMA] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT);
+	zones_size[ZONE_NORMAL] = (end_iomem >> PAGE_SHIFT) -
+		(uml_physmem >> PAGE_SHIFT);
 #ifdef CONFIG_HIGHMEM
 	zones_size[ZONE_HIGHMEM] = highmem >> PAGE_SHIFT;
 #endif
-- 
cgit v1.2.1


From a8b4fc4d7c3ccf80d4fa1805cee85c06c2aa653e Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:35 -0700
Subject: [PATCH] uml: fix missing x86_64 register definitions

The UML/x86_64 headers were missing ptrace support for some segment registers.
 The underlying problem was that the x86_64 kernel uses user_regs_struct
rather than the ptrace register definitions in ptrace.  This patch switches
UML/x86_64 to using user_regs_struct for its definitions of the host's
registers.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/sysdep-x86_64/ptrace.h | 43 +++++++++++++++++++++++++++++++---
 arch/um/include/sysdep-x86_64/sc.h     |  2 +-
 include/asm-um/ptrace-x86_64.h         |  3 +++
 3 files changed, 44 insertions(+), 4 deletions(-)

diff --git a/arch/um/include/sysdep-x86_64/ptrace.h b/arch/um/include/sysdep-x86_64/ptrace.h
index 8d353f0feec1..617bb9efc934 100644
--- a/arch/um/include/sysdep-x86_64/ptrace.h
+++ b/arch/um/include/sysdep-x86_64/ptrace.h
@@ -50,6 +50,21 @@
 #define HOST_FS 25
 #define HOST_GS 26
 
+/* Also defined in asm/ptrace-x86_64.h, but not in libc headers.  So, these
+ * are already defined for kernel code, but not for userspace code.
+ */
+#ifndef FS_BASE
+/* These aren't defined in ptrace.h, but exist in struct user_regs_struct,
+ * which is what x86_64 ptrace actually uses.
+ */
+#define FS_BASE (HOST_FS_BASE * sizeof(long))
+#define GS_BASE (HOST_GS_BASE * sizeof(long))
+#define DS (HOST_DS * sizeof(long))
+#define ES (HOST_ES * sizeof(long))
+#define FS (HOST_FS * sizeof(long))
+#define GS (HOST_GS * sizeof(long))
+#endif
+
 #define REGS_FS_BASE(r) ((r)[HOST_FS_BASE])
 #define REGS_GS_BASE(r) ((r)[HOST_GS_BASE])
 #define REGS_DS(r) ((r)[HOST_DS])
@@ -89,9 +104,12 @@ union uml_pt_regs {
 #endif
 #ifdef UML_CONFIG_MODE_SKAS
 	struct skas_regs {
-		/* XXX */
-		unsigned long regs[27];
-		unsigned long fp[65];
+		/* x86_64 ptrace uses sizeof(user_regs_struct) as its register
+		 * file size, while i386 uses FRAME_SIZE.  Therefore, we need
+		 * to use UM_FRAME_SIZE here instead of HOST_FRAME_SIZE.
+		 */
+		unsigned long regs[UM_FRAME_SIZE];
+		unsigned long fp[HOST_FP_SIZE];
                 struct faultinfo faultinfo;
 		long syscall;
 		int is_user;
@@ -120,11 +138,16 @@ extern int mode_tt;
 #define UPT_R14(r) __CHOOSE_MODE(SC_R14(UPT_SC(r)), REGS_R14((r)->skas.regs))
 #define UPT_R15(r) __CHOOSE_MODE(SC_R15(UPT_SC(r)), REGS_R15((r)->skas.regs))
 #define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs))
+#define UPT_FS_BASE(r) \
+	__CHOOSE_MODE(SC_FS_BASE(UPT_SC(r)), REGS_FS_BASE((r)->skas.regs))
 #define UPT_FS(r) __CHOOSE_MODE(SC_FS(UPT_SC(r)), REGS_FS((r)->skas.regs))
+#define UPT_GS_BASE(r) \
+	__CHOOSE_MODE(SC_GS_BASE(UPT_SC(r)), REGS_GS_BASE((r)->skas.regs))
 #define UPT_GS(r) __CHOOSE_MODE(SC_GS(UPT_SC(r)), REGS_GS((r)->skas.regs))
 #define UPT_DS(r) __CHOOSE_MODE(SC_DS(UPT_SC(r)), REGS_DS((r)->skas.regs))
 #define UPT_ES(r) __CHOOSE_MODE(SC_ES(UPT_SC(r)), REGS_ES((r)->skas.regs))
 #define UPT_CS(r) __CHOOSE_MODE(SC_CS(UPT_SC(r)), REGS_CS((r)->skas.regs))
+#define UPT_SS(r) __CHOOSE_MODE(SC_SS(UPT_SC(r)), REGS_SS((r)->skas.regs))
 #define UPT_ORIG_RAX(r) \
 	__CHOOSE_MODE((r)->tt.orig_rax, REGS_ORIG_RAX((r)->skas.regs))
 
@@ -183,6 +206,13 @@ struct syscall_args {
                 case RBP: val = UPT_RBP(regs); break; \
                 case ORIG_RAX: val = UPT_ORIG_RAX(regs); break; \
                 case CS: val = UPT_CS(regs); break; \
+                case SS: val = UPT_SS(regs); break; \
+		case FS_BASE: val = UPT_FS_BASE(regs); break; \
+                case GS_BASE: val = UPT_GS_BASE(regs); break; \
+                case DS: val = UPT_DS(regs); break; \
+                case ES: val = UPT_ES(regs); break; \
+                case FS : val = UPT_FS (regs); break; \
+		case GS: val = UPT_GS(regs); break;	    \
                 case EFLAGS: val = UPT_EFLAGS(regs); break; \
                 default :  \
                         panic("Bad register in UPT_REG : %d\n", reg);  \
@@ -214,6 +244,13 @@ struct syscall_args {
                 case RBP: UPT_RBP(regs) = __upt_val; break; \
                 case ORIG_RAX: UPT_ORIG_RAX(regs) = __upt_val; break; \
                 case CS: UPT_CS(regs) = __upt_val; break; \
+                case SS: UPT_SS(regs) = __upt_val; break; \
+                case FS_BASE: UPT_FS_BASE(regs) = __upt_val; break; \
+                case GS_BASE: UPT_GS_BASE(regs) = __upt_val; break; \
+                case DS: UPT_DS(regs) = __upt_val; break; \
+                case ES: UPT_ES(regs) = __upt_val; break; \
+                case FS: UPT_FS(regs) = __upt_val; break; \
+                case GS: UPT_GS(regs) = __upt_val; break; \
                 case EFLAGS: UPT_EFLAGS(regs) = __upt_val; break; \
                 default :  \
                         panic("Bad register in UPT_SET : %d\n", reg);  \
diff --git a/arch/um/include/sysdep-x86_64/sc.h b/arch/um/include/sysdep-x86_64/sc.h
index a160d9fcc596..8aee45b07434 100644
--- a/arch/um/include/sysdep-x86_64/sc.h
+++ b/arch/um/include/sysdep-x86_64/sc.h
@@ -35,11 +35,11 @@
 #define SC_GS(sc) SC_OFFSET(sc, SC_GS)
 #define SC_EFLAGS(sc) SC_OFFSET(sc, SC_EFLAGS)
 #define SC_SIGMASK(sc) SC_OFFSET(sc, SC_SIGMASK)
+#define SC_SS(sc) SC_OFFSET(sc, SC_SS)
 #if 0
 #define SC_ORIG_RAX(sc) SC_OFFSET(sc, SC_ORIG_RAX)
 #define SC_DS(sc) SC_OFFSET(sc, SC_DS)
 #define SC_ES(sc) SC_OFFSET(sc, SC_ES)
-#define SC_SS(sc) SC_OFFSET(sc, SC_SS)
 #endif
 
 #endif
diff --git a/include/asm-um/ptrace-x86_64.h b/include/asm-um/ptrace-x86_64.h
index 2074483e6ca4..03b4af4ac09a 100644
--- a/include/asm-um/ptrace-x86_64.h
+++ b/include/asm-um/ptrace-x86_64.h
@@ -16,12 +16,15 @@
 
 #define HOST_AUDIT_ARCH AUDIT_ARCH_X86_64
 
+/* Also defined in sysdep/ptrace.h, so may already be defined. */
+#ifndef FS_BASE
 #define FS_BASE (21 * sizeof(unsigned long))
 #define GS_BASE (22 * sizeof(unsigned long))
 #define DS (23 * sizeof(unsigned long))
 #define ES (24 * sizeof(unsigned long))
 #define FS (25 * sizeof(unsigned long))
 #define GS (26 * sizeof(unsigned long))
+#endif
 
 #define PT_REGS_RBX(r) UPT_RBX(&(r)->regs)
 #define PT_REGS_RCX(r) UPT_RCX(&(r)->regs)
-- 
cgit v1.2.1


From 09b185a3169c5a8f9b7baf97f16d32add7e10ca4 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:36 -0700
Subject: [PATCH] uml: fix gcov support

Make __bb_init_func weak in order to avoid a link failure with some libcs
and/or gccs.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/gmon_syms.c | 13 +------------
 1 file changed, 1 insertion(+), 12 deletions(-)

diff --git a/arch/um/kernel/gmon_syms.c b/arch/um/kernel/gmon_syms.c
index 2c86e7fdb014..13aa115cd1b4 100644
--- a/arch/um/kernel/gmon_syms.c
+++ b/arch/um/kernel/gmon_syms.c
@@ -5,7 +5,7 @@
 
 #include "linux/module.h"
 
-extern void __bb_init_func(void *);
+extern void __bb_init_func(void *)  __attribute__((weak));
 EXPORT_SYMBOL(__bb_init_func);
 
 /* This is defined (and referred to in profiling stub code) only by some GCC
@@ -21,14 +21,3 @@ EXPORT_SYMBOL(__gcov_init);
 
 extern void __gcov_merge_add(void *) __attribute__((weak));
 EXPORT_SYMBOL(__gcov_merge_add);
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
-- 
cgit v1.2.1


From 53dd2b55c50c810073615c79816bc06aea02dba9 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:37 -0700
Subject: [PATCH] uml: use correct SIGBUS handler

BB noticed that we had the wrong bus error handler.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/trap.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index e44ff6eca496..61a23fff4395 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -140,14 +140,6 @@ void segv_handler(int sig, union uml_pt_regs *regs)
 	segv(*fi, UPT_IP(regs), UPT_IS_USER(regs), regs);
 }
 
-const struct kern_handlers handlinfo_kern = {
-	.relay_signal = relay_signal,
-	.winch = winch,
-	.bus_handler = relay_signal,
-	.page_fault = segv_handler,
-	.sigio_handler = sigio_handler,
-	.timer_handler = timer_handler
-};
 /*
  * We give a *copy* of the faultinfo in the regs to segv.
  * This must be done, since nesting SEGVs could overwrite
@@ -253,6 +245,15 @@ void winch(int sig, union uml_pt_regs *regs)
 	do_IRQ(WINCH_IRQ, regs);
 }
 
+const struct kern_handlers handlinfo_kern = {
+	.relay_signal = relay_signal,
+	.winch = winch,
+	.bus_handler = bus_handler,
+	.page_fault = segv_handler,
+	.sigio_handler = sigio_handler,
+	.timer_handler = timer_handler
+};
+
 void trap_init(void)
 {
 }
-- 
cgit v1.2.1


From e3ccf6e3699c879973b616fe681ecaa363457d41 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:37 -0700
Subject: [PATCH] uml: add checkstack support

Make checkstack work for UML.  We need to pass the underlying architecture
name, rather than "um" to checkstack.pl.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Acked-by: Matt Mackall <mpm@selenic.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Makefile | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/Makefile b/Makefile
index 80dac0245d66..4c6c5e32ef96 100644
--- a/Makefile
+++ b/Makefile
@@ -1385,9 +1385,13 @@ endif #ifeq ($(config-targets),1)
 endif #ifeq ($(mixed-targets),1)
 
 PHONY += checkstack kernelrelease kernelversion
+
+# Use $(SUBARCH) here instead of $(ARCH) so that this works for UML.
+# In the UML case, $(SUBARCH) is the name of the underlying
+# architecture, while for all other arches, it is the same as $(ARCH).
 checkstack:
 	$(OBJDUMP) -d vmlinux $$(find . -name '*.ko') | \
-	$(PERL) $(src)/scripts/checkstack.pl $(ARCH)
+	$(PERL) $(src)/scripts/checkstack.pl $(SUBARCH)
 
 kernelrelease:
 	$(if $(wildcard include/config/kernel.release), $(Q)echo $(KERNELRELEASE), \
-- 
cgit v1.2.1


From 0915ee38c7db57ff41a7a68bcc02d0dd3b7b849b Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:38 -0700
Subject: [PATCH] uml: mark some tt-mode code

Mark a symbol and file as being tt-mode only.  This shrinks the binary
slightly when tt mode support is compiled out and makes it easier to identity
stuff when tt mode is removed.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h      | 3 ++-
 arch/um/kernel/ksyms.c    | 2 +-
 arch/um/os-Linux/Makefile | 8 ++++++--
 3 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index 24fb6d8680e1..be71b3d4c961 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -198,7 +198,9 @@ extern long os_ptrace_ldt(long pid, long addr, long data);
 extern int os_getpid(void);
 extern int os_getpgrp(void);
 
+#ifdef UML_CONFIG_MODE_TT
 extern void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int));
+#endif
 extern void init_new_thread_signals(void);
 extern int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr);
 
@@ -216,7 +218,6 @@ extern void os_flush_stdout(void);
  */
 extern void forward_ipi(int fd, int pid);
 extern void kill_child_dead(int pid);
-extern void stop(void);
 extern int wait_for_stop(int pid, int sig, int cont_type, void *relay);
 extern int protect_memory(unsigned long addr, unsigned long len,
 			  int r, int w, int x, int must_succeed);
diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c
index c97045d6d89f..9cbd9bb7e6d0 100644
--- a/arch/um/kernel/ksyms.c
+++ b/arch/um/kernel/ksyms.c
@@ -21,7 +21,6 @@
 #include "mem_user.h"
 #include "os.h"
 
-EXPORT_SYMBOL(stop);
 EXPORT_SYMBOL(uml_physmem);
 EXPORT_SYMBOL(set_signals);
 EXPORT_SYMBOL(get_signals);
@@ -41,6 +40,7 @@ EXPORT_SYMBOL(handle_page_fault);
 EXPORT_SYMBOL(find_iomem);
 
 #ifdef CONFIG_MODE_TT
+EXPORT_SYMBOL(stop);
 EXPORT_SYMBOL(strncpy_from_user_tt);
 EXPORT_SYMBOL(copy_from_user_tt);
 EXPORT_SYMBOL(copy_to_user_tt);
diff --git a/arch/um/os-Linux/Makefile b/arch/um/os-Linux/Makefile
index f4bfc4c7ccac..b4183929b32c 100644
--- a/arch/um/os-Linux/Makefile
+++ b/arch/um/os-Linux/Makefile
@@ -4,15 +4,19 @@
 #
 
 obj-y = aio.o elf_aux.o file.o helper.o irq.o main.o mem.o process.o sigio.o \
-	signal.o start_up.o time.o trap.o tt.o tty.o uaccess.o umid.o tls.o \
+	signal.o start_up.o time.o trap.o tty.o uaccess.o umid.o tls.o \
 	user_syms.o util.o drivers/ sys-$(SUBARCH)/
 
 obj-$(CONFIG_MODE_SKAS) += skas/
+
+obj-$(CONFIG_MODE_TT) += tt.o
+user-objs-$(CONFIG_MODE_TT) += tt.o
+
 obj-$(CONFIG_TTY_LOG) += tty_log.o
 user-objs-$(CONFIG_TTY_LOG) += tty_log.o
 
 USER_OBJS := $(user-objs-y) aio.o elf_aux.o file.o helper.o irq.o main.o mem.o \
-	process.o sigio.o signal.o start_up.o time.o trap.o tt.o tty.o tls.o \
+	process.o sigio.o signal.o start_up.o time.o trap.o tty.o tls.o \
 	uaccess.o umid.o util.o
 
 CFLAGS_user_syms.o += -DSUBARCH_$(SUBARCH)
-- 
cgit v1.2.1


From 3c9173509985b957bea692ea887a8a0e5055cfe8 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:40 -0700
Subject: [PATCH] uml: thread creation tidying

fork on UML has always somewhat subtle.  The underlying cause has been the
need to initialize a stack for the new process.  The only portable way to
initialize a new stack is to set it as the alternate signal stack and take a
signal.  The signal handler does whatever initialization is needed and jumps
back to the original stack, where the fork processing is finished.  The basic
context switching mechanism is a jmp_buf for each process.  You switch to a
new process by longjmping to its jmp_buf.

Now that UML has its own implementation of setjmp and longjmp, and I can poke
around inside a jmp_buf without fear that libc will change the structure, a
much simpler mechanism is possible.  The jmpbuf can simply be initialized by
hand.

This eliminates -
	the need to set up and remove the alternate signal stack
	sending and handling a signal
	the signal blocking needed around the stack switching, since
there is no stack switching
	setting up the jmp_buf needed to jump back to the original
stack after the new one is set up

In addition, since jmp_buf is now defined by UML, and not by libc, it can be
embedded in the thread struct.  This makes it unnecessary to have it exist on
the stack, where it used to be.  It also simplifies interfaces, since the
switch jmp_buf used to be a void * inside the thread struct, and functions
which took it as an argument needed to define a jmp_buf variable and assign it
from the void *.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h                       | 10 ++---
 arch/um/include/skas/skas.h                |  3 +-
 arch/um/include/sysdep-i386/archsetjmp.h   |  3 ++
 arch/um/include/sysdep-x86_64/archsetjmp.h |  3 ++
 arch/um/kernel/skas/process_kern.c         | 36 ++++++----------
 arch/um/os-Linux/skas/process.c            | 68 +++++++-----------------------
 include/asm-um/processor-generic.h         |  6 +--
 7 files changed, 43 insertions(+), 86 deletions(-)

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index be71b3d4c961..120ca21a513a 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -14,6 +14,7 @@
 #include "skas/mm_id.h"
 #include "irq_user.h"
 #include "sysdep/tls.h"
+#include "sysdep/archsetjmp.h"
 
 #define OS_TYPE_FILE 1
 #define OS_TYPE_DIR 2
@@ -308,12 +309,9 @@ extern int copy_context_skas0(unsigned long stack, int pid);
 extern void userspace(union uml_pt_regs *regs);
 extern void map_stub_pages(int fd, unsigned long code,
 			   unsigned long data, unsigned long stack);
-extern void new_thread(void *stack, void **switch_buf_ptr,
-			 void **fork_buf_ptr, void (*handler)(int));
-extern void thread_wait(void *sw, void *fb);
-extern void switch_threads(void *me, void *next);
-extern int start_idle_thread(void *stack, void *switch_buf_ptr,
-			     void **fork_buf_ptr);
+extern void new_thread(void *stack, jmp_buf *buf, void (*handler)(void));
+extern void switch_threads(jmp_buf *me, jmp_buf *you);
+extern int start_idle_thread(void *stack, jmp_buf *switch_buf);
 extern void initial_thread_cb_skas(void (*proc)(void *),
 				 void *arg);
 extern void halt_skas(void);
diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h
index 853b26f148c5..e88926b16072 100644
--- a/arch/um/include/skas/skas.h
+++ b/arch/um/include/skas/skas.h
@@ -14,8 +14,7 @@ extern int proc_mm, ptrace_faultinfo, ptrace_ldt;
 extern int skas_needs_stub;
 
 extern int user_thread(unsigned long stack, int flags);
-extern void new_thread_proc(void *stack, void (*handler)(int sig));
-extern void new_thread_handler(int sig);
+extern void new_thread_handler(void);
 extern void handle_syscall(union uml_pt_regs *regs);
 extern int new_mm(unsigned long stack);
 extern void get_skas_faultinfo(int pid, struct faultinfo * fi);
diff --git a/arch/um/include/sysdep-i386/archsetjmp.h b/arch/um/include/sysdep-i386/archsetjmp.h
index ea1ba3d42aee..11bafab669e9 100644
--- a/arch/um/include/sysdep-i386/archsetjmp.h
+++ b/arch/um/include/sysdep-i386/archsetjmp.h
@@ -16,4 +16,7 @@ struct __jmp_buf {
 
 typedef struct __jmp_buf jmp_buf[1];
 
+#define JB_IP __eip
+#define JB_SP __esp
+
 #endif				/* _SETJMP_H */
diff --git a/arch/um/include/sysdep-x86_64/archsetjmp.h b/arch/um/include/sysdep-x86_64/archsetjmp.h
index 454fc60aff6d..9a5e1a6ec800 100644
--- a/arch/um/include/sysdep-x86_64/archsetjmp.h
+++ b/arch/um/include/sysdep-x86_64/archsetjmp.h
@@ -18,4 +18,7 @@ struct __jmp_buf {
 
 typedef struct __jmp_buf jmp_buf[1];
 
+#define JB_IP __rip
+#define JB_SP __rsp
+
 #endif				/* _SETJMP_H */
diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c
index 55caeec8b257..c73d47c57abe 100644
--- a/arch/um/kernel/skas/process_kern.c
+++ b/arch/um/kernel/skas/process_kern.c
@@ -33,7 +33,7 @@ void switch_to_skas(void *prev, void *next)
 		switch_timers(0);
 
 	switch_threads(&from->thread.mode.skas.switch_buf,
-		       to->thread.mode.skas.switch_buf);
+		       &to->thread.mode.skas.switch_buf);
 
 	arch_switch_to_skas(current->thread.prev_sched, current);
 
@@ -43,21 +43,21 @@ void switch_to_skas(void *prev, void *next)
 
 extern void schedule_tail(struct task_struct *prev);
 
-void new_thread_handler(int sig)
+/* This is called magically, by its address being stuffed in a jmp_buf
+ * and being longjmp-d to.
+ */
+void new_thread_handler(void)
 {
 	int (*fn)(void *), n;
 	void *arg;
 
-	fn = current->thread.request.u.thread.proc;
-	arg = current->thread.request.u.thread.arg;
-	os_usr1_signal(1);
-	thread_wait(&current->thread.mode.skas.switch_buf,
-		    current->thread.mode.skas.fork_buf);
-
 	if(current->thread.prev_sched != NULL)
 		schedule_tail(current->thread.prev_sched);
 	current->thread.prev_sched = NULL;
 
+	fn = current->thread.request.u.thread.proc;
+	arg = current->thread.request.u.thread.arg;
+
 	/* The return value is 1 if the kernel thread execs a process,
 	 * 0 if it just exits
 	 */
@@ -70,22 +70,13 @@ void new_thread_handler(int sig)
 	else do_exit(0);
 }
 
-void new_thread_proc(void *stack, void (*handler)(int sig))
-{
-	init_new_thread_stack(stack, handler);
-	os_usr1_process(os_getpid());
-}
-
 void release_thread_skas(struct task_struct *task)
 {
 }
 
-void fork_handler(int sig)
+/* Called magically, see new_thread_handler above */
+void fork_handler(void)
 {
-	os_usr1_signal(1);
-	thread_wait(&current->thread.mode.skas.switch_buf,
-		    current->thread.mode.skas.fork_buf);
-  	
 	force_flush_all();
 	if(current->thread.prev_sched == NULL)
 		panic("blech");
@@ -109,7 +100,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
 		     unsigned long stack_top, struct task_struct * p,
 		     struct pt_regs *regs)
 {
-  	void (*handler)(int);
+  	void (*handler)(void);
 
 	if(current->thread.forking){
 	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
@@ -128,7 +119,7 @@ int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
 	}
 
 	new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf,
-		   &p->thread.mode.skas.fork_buf, handler);
+		   handler);
 	return(0);
 }
 
@@ -182,8 +173,7 @@ int start_uml_skas(void)
 	init_task.thread.request.u.thread.proc = start_kernel_proc;
 	init_task.thread.request.u.thread.arg = NULL;
 	return(start_idle_thread(task_stack_page(&init_task),
-				 &init_task.thread.mode.skas.switch_buf,
-				 &init_task.thread.mode.skas.fork_buf));
+				 &init_task.thread.mode.skas.switch_buf));
 }
 
 int external_pid_skas(struct task_struct *task)
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 42e3d1ed802c..cb9ab54146cc 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -444,56 +444,22 @@ void map_stub_pages(int fd, unsigned long code,
 	}
 }
 
-void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr,
-		void (*handler)(int))
+void new_thread(void *stack, jmp_buf *buf, void (*handler)(void))
 {
-	unsigned long flags;
-	jmp_buf switch_buf, fork_buf;
-
-	*switch_buf_ptr = &switch_buf;
-	*fork_buf_ptr = &fork_buf;
-
-	/* Somewhat subtle - siglongjmp restores the signal mask before doing
-	 * the longjmp.  This means that when jumping from one stack to another
-	 * when the target stack has interrupts enabled, an interrupt may occur
-	 * on the source stack.  This is bad when starting up a process because
-	 * it's not supposed to get timer ticks until it has been scheduled.
-	 * So, we disable interrupts around the sigsetjmp to ensure that
-	 * they can't happen until we get back here where they are safe.
-	 */
-	flags = get_signals();
-	block_signals();
-	if(UML_SETJMP(&fork_buf) == 0)
-		new_thread_proc(stack, handler);
-
-	remove_sigstack();
-
-	set_signals(flags);
+	(*buf)[0].JB_IP = (unsigned long) handler;
+	(*buf)[0].JB_SP = (unsigned long) stack +
+		(PAGE_SIZE << UML_CONFIG_KERNEL_STACK_ORDER) - sizeof(void *);
 }
 
 #define INIT_JMP_NEW_THREAD 0
-#define INIT_JMP_REMOVE_SIGSTACK 1
-#define INIT_JMP_CALLBACK 2
-#define INIT_JMP_HALT 3
-#define INIT_JMP_REBOOT 4
-
-void thread_wait(void *sw, void *fb)
-{
-	jmp_buf buf, **switch_buf = sw, *fork_buf;
-
-	*switch_buf = &buf;
-	fork_buf = fb;
-	if(UML_SETJMP(&buf) == 0)
-		UML_LONGJMP(fork_buf, INIT_JMP_REMOVE_SIGSTACK);
-}
+#define INIT_JMP_CALLBACK 1
+#define INIT_JMP_HALT 2
+#define INIT_JMP_REBOOT 3
 
-void switch_threads(void *me, void *next)
+void switch_threads(jmp_buf *me, jmp_buf *you)
 {
-	jmp_buf my_buf, **me_ptr = me, *next_buf = next;
-
-	*me_ptr = &my_buf;
-	if(UML_SETJMP(&my_buf) == 0)
-		UML_LONGJMP(next_buf, 1);
+	if(UML_SETJMP(me) == 0)
+		UML_LONGJMP(you, 1);
 }
 
 static jmp_buf initial_jmpbuf;
@@ -503,23 +469,21 @@ static void (*cb_proc)(void *arg);
 static void *cb_arg;
 static jmp_buf *cb_back;
 
-int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr)
+int start_idle_thread(void *stack, jmp_buf *switch_buf)
 {
-	jmp_buf **switch_buf = switch_buf_ptr;
 	int n;
 
 	set_handler(SIGWINCH, (__sighandler_t) sig_handler,
 		    SA_ONSTACK | SA_RESTART, SIGUSR1, SIGIO, SIGALRM,
 		    SIGVTALRM, -1);
 
-	*fork_buf_ptr = &initial_jmpbuf;
 	n = UML_SETJMP(&initial_jmpbuf);
 	switch(n){
 	case INIT_JMP_NEW_THREAD:
-		new_thread_proc((void *) stack, new_thread_handler);
-		break;
-	case INIT_JMP_REMOVE_SIGSTACK:
-		remove_sigstack();
+		(*switch_buf)[0].JB_IP = (unsigned long) new_thread_handler;
+		(*switch_buf)[0].JB_SP = (unsigned long) stack +
+			(PAGE_SIZE << UML_CONFIG_KERNEL_STACK_ORDER) -
+			sizeof(void *);
 		break;
 	case INIT_JMP_CALLBACK:
 		(*cb_proc)(cb_arg);
@@ -534,7 +498,7 @@ int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr)
 	default:
 		panic("Bad sigsetjmp return in start_idle_thread - %d\n", n);
 	}
-	UML_LONGJMP(*switch_buf, 1);
+	UML_LONGJMP(switch_buf, 1);
 }
 
 void initial_thread_cb_skas(void (*proc)(void *), void *arg)
diff --git a/include/asm-um/processor-generic.h b/include/asm-um/processor-generic.h
index afa4fe1ca9f1..d99bbddffdb9 100644
--- a/include/asm-um/processor-generic.h
+++ b/include/asm-um/processor-generic.h
@@ -13,6 +13,7 @@ struct task_struct;
 #include "asm/ptrace.h"
 #include "choose-mode.h"
 #include "registers.h"
+#include "sysdep/archsetjmp.h"
 
 struct mm_struct;
 
@@ -43,8 +44,7 @@ struct thread_struct {
 #endif
 #ifdef CONFIG_MODE_SKAS
 		struct {
-			void *switch_buf;
-			void *fork_buf;
+			jmp_buf switch_buf;
 			int mm_count;
 		} skas;
 #endif
@@ -138,7 +138,7 @@ extern struct cpuinfo_um cpu_data[];
 
 #ifdef CONFIG_MODE_SKAS
 #define KSTK_REG(tsk, reg) \
-	get_thread_reg(reg, tsk->thread.mode.skas.switch_buf)
+	get_thread_reg(reg, &tsk->thread.mode.skas.switch_buf)
 #else
 #define KSTK_REG(tsk, reg) (0xbadbabe)
 #endif
-- 
cgit v1.2.1


From 995473aec0be1d9993205accc03e19d32d4e4a2a Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:40 -0700
Subject: [PATCH] uml: file renaming

Move some foo_kern.c files to foo.c now that the old foo.c files are out
of the way.

Also cleaned up some whitespace and an emacs formatting comment.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/Makefile            |   2 +-
 arch/um/kernel/process.c           | 483 +++++++++++++++++++++++++++++++++
 arch/um/kernel/process_kern.c      | 484 ---------------------------------
 arch/um/kernel/skas/Makefile       |   3 +-
 arch/um/kernel/skas/exec.c         |  30 +++
 arch/um/kernel/skas/exec_kern.c    |  41 ---
 arch/um/kernel/skas/process.c      | 217 +++++++++++++++
 arch/um/kernel/skas/process_kern.c | 529 ++++++++++++++++++++++++++++---------
 8 files changed, 1130 insertions(+), 659 deletions(-)
 create mode 100644 arch/um/kernel/process.c
 delete mode 100644 arch/um/kernel/process_kern.c
 create mode 100644 arch/um/kernel/skas/exec.c
 delete mode 100644 arch/um/kernel/skas/exec_kern.c
 create mode 100644 arch/um/kernel/skas/process.c

diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile
index a2d93065b2d0..6fa63a2a89e3 100644
--- a/arch/um/kernel/Makefile
+++ b/arch/um/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := vmlinux.lds
 clean-files :=
 
 obj-y = config.o exec.o exitcode.o init_task.o irq.o ksyms.o mem.o \
-	physmem.o process_kern.o ptrace.o reboot.o resource.o sigio.o \
+	physmem.o process.o ptrace.o reboot.o resource.o sigio.o \
 	signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o uaccess.o \
 	um_arch.o umid.o
 
diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
new file mode 100644
index 000000000000..fe6c64abda5b
--- /dev/null
+++ b/arch/um/kernel/process.c
@@ -0,0 +1,483 @@
+/*
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright 2003 PathScale, Inc.
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "linux/sched.h"
+#include "linux/interrupt.h"
+#include "linux/string.h"
+#include "linux/mm.h"
+#include "linux/slab.h"
+#include "linux/utsname.h"
+#include "linux/fs.h"
+#include "linux/utime.h"
+#include "linux/smp_lock.h"
+#include "linux/module.h"
+#include "linux/init.h"
+#include "linux/capability.h"
+#include "linux/vmalloc.h"
+#include "linux/spinlock.h"
+#include "linux/proc_fs.h"
+#include "linux/ptrace.h"
+#include "linux/random.h"
+#include "linux/personality.h"
+#include "asm/unistd.h"
+#include "asm/mman.h"
+#include "asm/segment.h"
+#include "asm/stat.h"
+#include "asm/pgtable.h"
+#include "asm/processor.h"
+#include "asm/tlbflush.h"
+#include "asm/uaccess.h"
+#include "asm/user.h"
+#include "user_util.h"
+#include "kern_util.h"
+#include "kern.h"
+#include "signal_kern.h"
+#include "init.h"
+#include "irq_user.h"
+#include "mem_user.h"
+#include "tlb.h"
+#include "frame_kern.h"
+#include "sigcontext.h"
+#include "os.h"
+#include "mode.h"
+#include "mode_kern.h"
+#include "choose-mode.h"
+
+/* This is a per-cpu array.  A processor only modifies its entry and it only
+ * cares about its entry, so it's OK if another processor is modifying its
+ * entry.
+ */
+struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } };
+
+int external_pid(void *t)
+{
+	struct task_struct *task = t ? t : current;
+
+	return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task));
+}
+
+int pid_to_processor_id(int pid)
+{
+	int i;
+
+	for(i = 0; i < ncpus; i++){
+		if(cpu_tasks[i].pid == pid) return(i);
+	}
+	return(-1);
+}
+
+void free_stack(unsigned long stack, int order)
+{
+	free_pages(stack, order);
+}
+
+unsigned long alloc_stack(int order, int atomic)
+{
+	unsigned long page;
+	gfp_t flags = GFP_KERNEL;
+
+	if (atomic)
+		flags = GFP_ATOMIC;
+	page = __get_free_pages(flags, order);
+	if(page == 0)
+		return(0);
+	stack_protections(page);
+	return(page);
+}
+
+int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+{
+	int pid;
+
+	current->thread.request.u.thread.proc = fn;
+	current->thread.request.u.thread.arg = arg;
+	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0,
+		      &current->thread.regs, 0, NULL, NULL);
+	if(pid < 0)
+		panic("do_fork failed in kernel_thread, errno = %d", pid);
+	return(pid);
+}
+
+void set_current(void *t)
+{
+	struct task_struct *task = t;
+
+	cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task)
+		{ external_pid(task), task });
+}
+
+void *_switch_to(void *prev, void *next, void *last)
+{
+	struct task_struct *from = prev;
+	struct task_struct *to= next;
+
+	to->thread.prev_sched = from;
+	set_current(to);
+
+	do {
+		current->thread.saved_task = NULL ;
+		CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next);
+		if(current->thread.saved_task)
+			show_regs(&(current->thread.regs));
+		next= current->thread.saved_task;
+		prev= current;
+	} while(current->thread.saved_task);
+
+	return(current->thread.prev_sched);
+
+}
+
+void interrupt_end(void)
+{
+	if(need_resched()) schedule();
+	if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal();
+}
+
+void release_thread(struct task_struct *task)
+{
+	CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task));
+}
+
+void exit_thread(void)
+{
+	unprotect_stack((unsigned long) current_thread);
+}
+
+void *get_current(void)
+{
+	return(current);
+}
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
+		unsigned long stack_top, struct task_struct * p,
+		struct pt_regs *regs)
+{
+	int ret;
+
+	p->thread = (struct thread_struct) INIT_THREAD;
+	ret = CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr,
+				clone_flags, sp, stack_top, p, regs);
+
+	if (ret || !current->thread.forking)
+		goto out;
+
+	clear_flushed_tls(p);
+
+	/*
+	 * Set a new TLS for the child thread?
+	 */
+	if (clone_flags & CLONE_SETTLS)
+		ret = arch_copy_tls(p);
+
+out:
+	return ret;
+}
+
+void initial_thread_cb(void (*proc)(void *), void *arg)
+{
+	int save_kmalloc_ok = kmalloc_ok;
+
+	kmalloc_ok = 0;
+	CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc,
+			 arg);
+	kmalloc_ok = save_kmalloc_ok;
+}
+
+unsigned long stack_sp(unsigned long page)
+{
+	return(page + PAGE_SIZE - sizeof(void *));
+}
+
+int current_pid(void)
+{
+	return(current->pid);
+}
+
+void default_idle(void)
+{
+	CHOOSE_MODE(uml_idle_timer(), (void) 0);
+
+	while(1){
+		/* endless idle loop with no priority at all */
+
+		/*
+		 * although we are an idle CPU, we do not want to
+		 * get into the scheduler unnecessarily.
+		 */
+		if(need_resched())
+			schedule();
+
+		idle_sleep(10);
+	}
+}
+
+void cpu_idle(void)
+{
+	CHOOSE_MODE(init_idle_tt(), init_idle_skas());
+}
+
+int page_size(void)
+{
+	return(PAGE_SIZE);
+}
+
+void *um_virt_to_phys(struct task_struct *task, unsigned long addr,
+		      pte_t *pte_out)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	pte_t ptent;
+
+	if(task->mm == NULL)
+		return(ERR_PTR(-EINVAL));
+	pgd = pgd_offset(task->mm, addr);
+	if(!pgd_present(*pgd))
+		return(ERR_PTR(-EINVAL));
+
+	pud = pud_offset(pgd, addr);
+	if(!pud_present(*pud))
+		return(ERR_PTR(-EINVAL));
+
+	pmd = pmd_offset(pud, addr);
+	if(!pmd_present(*pmd))
+		return(ERR_PTR(-EINVAL));
+
+	pte = pte_offset_kernel(pmd, addr);
+	ptent = *pte;
+	if(!pte_present(ptent))
+		return(ERR_PTR(-EINVAL));
+
+	if(pte_out != NULL)
+		*pte_out = ptent;
+	return((void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK));
+}
+
+char *current_cmd(void)
+{
+#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM)
+	return("(Unknown)");
+#else
+	void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL);
+	return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr);
+#endif
+}
+
+void force_sigbus(void)
+{
+	printk(KERN_ERR "Killing pid %d because of a lack of memory\n",
+	       current->pid);
+	lock_kernel();
+	sigaddset(&current->pending.signal, SIGBUS);
+	recalc_sigpending();
+	current->flags |= PF_SIGNALED;
+	do_exit(SIGBUS | 0x80);
+}
+
+void dump_thread(struct pt_regs *regs, struct user *u)
+{
+}
+
+void enable_hlt(void)
+{
+	panic("enable_hlt");
+}
+
+EXPORT_SYMBOL(enable_hlt);
+
+void disable_hlt(void)
+{
+	panic("disable_hlt");
+}
+
+EXPORT_SYMBOL(disable_hlt);
+
+void *um_kmalloc(int size)
+{
+	return kmalloc(size, GFP_KERNEL);
+}
+
+void *um_kmalloc_atomic(int size)
+{
+	return kmalloc(size, GFP_ATOMIC);
+}
+
+void *um_vmalloc(int size)
+{
+	return vmalloc(size);
+}
+
+void *um_vmalloc_atomic(int size)
+{
+	return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL);
+}
+
+int __cant_sleep(void) {
+	return in_atomic() || irqs_disabled() || in_interrupt();
+	/* Is in_interrupt() really needed? */
+}
+
+unsigned long get_fault_addr(void)
+{
+	return((unsigned long) current->thread.fault_addr);
+}
+
+EXPORT_SYMBOL(get_fault_addr);
+
+void not_implemented(void)
+{
+	printk(KERN_DEBUG "Something isn't implemented in here\n");
+}
+
+EXPORT_SYMBOL(not_implemented);
+
+int user_context(unsigned long sp)
+{
+	unsigned long stack;
+
+	stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER);
+	return(stack != (unsigned long) current_thread);
+}
+
+extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end;
+
+void do_uml_exitcalls(void)
+{
+	exitcall_t *call;
+
+	call = &__uml_exitcall_end;
+	while (--call >= &__uml_exitcall_begin)
+		(*call)();
+}
+
+char *uml_strdup(char *string)
+{
+	return kstrdup(string, GFP_KERNEL);
+}
+
+int copy_to_user_proc(void __user *to, void *from, int size)
+{
+	return(copy_to_user(to, from, size));
+}
+
+int copy_from_user_proc(void *to, void __user *from, int size)
+{
+	return(copy_from_user(to, from, size));
+}
+
+int clear_user_proc(void __user *buf, int size)
+{
+	return(clear_user(buf, size));
+}
+
+int strlen_user_proc(char __user *str)
+{
+	return(strlen_user(str));
+}
+
+int smp_sigio_handler(void)
+{
+#ifdef CONFIG_SMP
+	int cpu = current_thread->cpu;
+	IPI_handler(cpu);
+	if(cpu != 0)
+		return(1);
+#endif
+	return(0);
+}
+
+int cpu(void)
+{
+	return(current_thread->cpu);
+}
+
+static atomic_t using_sysemu = ATOMIC_INIT(0);
+int sysemu_supported;
+
+void set_using_sysemu(int value)
+{
+	if (value > sysemu_supported)
+		return;
+	atomic_set(&using_sysemu, value);
+}
+
+int get_using_sysemu(void)
+{
+	return atomic_read(&using_sysemu);
+}
+
+static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
+{
+	if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
+		*eof = 1;
+
+	return strlen(buf);
+}
+
+static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data)
+{
+	char tmp[2];
+
+	if (copy_from_user(tmp, buf, 1))
+		return -EFAULT;
+
+	if (tmp[0] >= '0' && tmp[0] <= '2')
+		set_using_sysemu(tmp[0] - '0');
+	return count; /*We use the first char, but pretend to write everything*/
+}
+
+int __init make_proc_sysemu(void)
+{
+	struct proc_dir_entry *ent;
+	if (!sysemu_supported)
+		return 0;
+
+	ent = create_proc_entry("sysemu", 0600, &proc_root);
+
+	if (ent == NULL)
+	{
+		printk(KERN_WARNING "Failed to register /proc/sysemu\n");
+		return(0);
+	}
+
+	ent->read_proc  = proc_read_sysemu;
+	ent->write_proc = proc_write_sysemu;
+
+	return 0;
+}
+
+late_initcall(make_proc_sysemu);
+
+int singlestepping(void * t)
+{
+	struct task_struct *task = t ? t : current;
+
+	if ( ! (task->ptrace & PT_DTRACE) )
+		return(0);
+
+	if (task->thread.singlestep_syscall)
+		return(1);
+
+	return 2;
+}
+
+/*
+ * Only x86 and x86_64 have an arch_align_stack().
+ * All other arches have "#define arch_align_stack(x) (x)"
+ * in their asm/system.h
+ * As this is included in UML from asm-um/system-generic.h,
+ * we can use it to behave as the subarch does.
+ */
+#ifndef arch_align_stack
+unsigned long arch_align_stack(unsigned long sp)
+{
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+		sp -= get_random_int() % 8192;
+	return sp & ~0xf;
+}
+#endif
diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c
deleted file mode 100644
index 537895d68ad1..000000000000
--- a/arch/um/kernel/process_kern.c
+++ /dev/null
@@ -1,484 +0,0 @@
-/* 
- * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
- * Copyright 2003 PathScale, Inc.
- * Licensed under the GPL
- */
-
-#include "linux/config.h"
-#include "linux/kernel.h"
-#include "linux/sched.h"
-#include "linux/interrupt.h"
-#include "linux/string.h"
-#include "linux/mm.h"
-#include "linux/slab.h"
-#include "linux/utsname.h"
-#include "linux/fs.h"
-#include "linux/utime.h"
-#include "linux/smp_lock.h"
-#include "linux/module.h"
-#include "linux/init.h"
-#include "linux/capability.h"
-#include "linux/vmalloc.h"
-#include "linux/spinlock.h"
-#include "linux/proc_fs.h"
-#include "linux/ptrace.h"
-#include "linux/random.h"
-#include "linux/personality.h"
-#include "asm/unistd.h"
-#include "asm/mman.h"
-#include "asm/segment.h"
-#include "asm/stat.h"
-#include "asm/pgtable.h"
-#include "asm/processor.h"
-#include "asm/tlbflush.h"
-#include "asm/uaccess.h"
-#include "asm/user.h"
-#include "user_util.h"
-#include "kern_util.h"
-#include "kern.h"
-#include "signal_kern.h"
-#include "init.h"
-#include "irq_user.h"
-#include "mem_user.h"
-#include "tlb.h"
-#include "frame_kern.h"
-#include "sigcontext.h"
-#include "os.h"
-#include "mode.h"
-#include "mode_kern.h"
-#include "choose-mode.h"
-
-/* This is a per-cpu array.  A processor only modifies its entry and it only
- * cares about its entry, so it's OK if another processor is modifying its
- * entry.
- */
-struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } };
-
-int external_pid(void *t)
-{
-	struct task_struct *task = t ? t : current;
-
-	return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task));
-}
-
-int pid_to_processor_id(int pid)
-{
-	int i;
-
-	for(i = 0; i < ncpus; i++){
-		if(cpu_tasks[i].pid == pid) return(i);
-	}
-	return(-1);
-}
-
-void free_stack(unsigned long stack, int order)
-{
-	free_pages(stack, order);
-}
-
-unsigned long alloc_stack(int order, int atomic)
-{
-	unsigned long page;
-	gfp_t flags = GFP_KERNEL;
-
-	if (atomic)
-		flags = GFP_ATOMIC;
-	page = __get_free_pages(flags, order);
-	if(page == 0)
-		return(0);
-	stack_protections(page);
-	return(page);
-}
-
-int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
-{
-	int pid;
-
-	current->thread.request.u.thread.proc = fn;
-	current->thread.request.u.thread.arg = arg;
-	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0,
-		      &current->thread.regs, 0, NULL, NULL);
-	if(pid < 0)
-		panic("do_fork failed in kernel_thread, errno = %d", pid);
-	return(pid);
-}
-
-void set_current(void *t)
-{
-	struct task_struct *task = t;
-
-	cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task)
-		{ external_pid(task), task });
-}
-
-void *_switch_to(void *prev, void *next, void *last)
-{
-        struct task_struct *from = prev;
-        struct task_struct *to= next;
-
-        to->thread.prev_sched = from;
-        set_current(to);
-
-	do {
-		current->thread.saved_task = NULL ;
-		CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next);
-		if(current->thread.saved_task)
-			show_regs(&(current->thread.regs));
-		next= current->thread.saved_task;
-		prev= current;
-	} while(current->thread.saved_task);
-
-        return(current->thread.prev_sched);
-
-}
-
-void interrupt_end(void)
-{
-	if(need_resched()) schedule();
-	if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal();
-}
-
-void release_thread(struct task_struct *task)
-{
-	CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task));
-}
- 
-void exit_thread(void)
-{
-	unprotect_stack((unsigned long) current_thread);
-}
- 
-void *get_current(void)
-{
-	return(current);
-}
-
-int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
-		unsigned long stack_top, struct task_struct * p, 
-		struct pt_regs *regs)
-{
-	int ret;
-
-	p->thread = (struct thread_struct) INIT_THREAD;
-	ret = CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr,
-				clone_flags, sp, stack_top, p, regs);
-
-	if (ret || !current->thread.forking)
-		goto out;
-
-	clear_flushed_tls(p);
-
-	/*
-	 * Set a new TLS for the child thread?
-	 */
-	if (clone_flags & CLONE_SETTLS)
-		ret = arch_copy_tls(p);
-
-out:
-	return ret;
-}
-
-void initial_thread_cb(void (*proc)(void *), void *arg)
-{
-	int save_kmalloc_ok = kmalloc_ok;
-
-	kmalloc_ok = 0;
-	CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc, 
-			 arg);
-	kmalloc_ok = save_kmalloc_ok;
-}
- 
-unsigned long stack_sp(unsigned long page)
-{
-	return(page + PAGE_SIZE - sizeof(void *));
-}
-
-int current_pid(void)
-{
-	return(current->pid);
-}
-
-void default_idle(void)
-{
-	CHOOSE_MODE(uml_idle_timer(), (void) 0);
-
-	while(1){
-		/* endless idle loop with no priority at all */
-
-		/*
-		 * although we are an idle CPU, we do not want to
-		 * get into the scheduler unnecessarily.
-		 */
-		if(need_resched())
-			schedule();
-		
-		idle_sleep(10);
-	}
-}
-
-void cpu_idle(void)
-{
-	CHOOSE_MODE(init_idle_tt(), init_idle_skas());
-}
-
-int page_size(void)
-{
-	return(PAGE_SIZE);
-}
-
-void *um_virt_to_phys(struct task_struct *task, unsigned long addr, 
-		      pte_t *pte_out)
-{
-	pgd_t *pgd;
-	pud_t *pud;
-	pmd_t *pmd;
-	pte_t *pte;
-	pte_t ptent;
-
-	if(task->mm == NULL) 
-		return(ERR_PTR(-EINVAL));
-	pgd = pgd_offset(task->mm, addr);
-	if(!pgd_present(*pgd))
-		return(ERR_PTR(-EINVAL));
-
-	pud = pud_offset(pgd, addr);
-	if(!pud_present(*pud))
-		return(ERR_PTR(-EINVAL));
-
-	pmd = pmd_offset(pud, addr);
-	if(!pmd_present(*pmd)) 
-		return(ERR_PTR(-EINVAL));
-
-	pte = pte_offset_kernel(pmd, addr);
-	ptent = *pte;
-	if(!pte_present(ptent))
-		return(ERR_PTR(-EINVAL));
-
-	if(pte_out != NULL)
-		*pte_out = ptent;
-	return((void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK));
-}
-
-char *current_cmd(void)
-{
-#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM)
-	return("(Unknown)");
-#else
-	void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL);
-	return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr);
-#endif
-}
-
-void force_sigbus(void)
-{
-	printk(KERN_ERR "Killing pid %d because of a lack of memory\n", 
-	       current->pid);
-	lock_kernel();
-	sigaddset(&current->pending.signal, SIGBUS);
-	recalc_sigpending();
-	current->flags |= PF_SIGNALED;
-	do_exit(SIGBUS | 0x80);
-}
-
-void dump_thread(struct pt_regs *regs, struct user *u)
-{
-}
-
-void enable_hlt(void)
-{
-	panic("enable_hlt");
-}
-
-EXPORT_SYMBOL(enable_hlt);
-
-void disable_hlt(void)
-{
-	panic("disable_hlt");
-}
-
-EXPORT_SYMBOL(disable_hlt);
-
-void *um_kmalloc(int size)
-{
-	return kmalloc(size, GFP_KERNEL);
-}
-
-void *um_kmalloc_atomic(int size)
-{
-	return kmalloc(size, GFP_ATOMIC);
-}
-
-void *um_vmalloc(int size)
-{
-	return vmalloc(size);
-}
-
-void *um_vmalloc_atomic(int size)
-{
-	return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL);
-}
-
-int __cant_sleep(void) {
-	return in_atomic() || irqs_disabled() || in_interrupt();
-	/* Is in_interrupt() really needed? */
-}
-
-unsigned long get_fault_addr(void)
-{
-	return((unsigned long) current->thread.fault_addr);
-}
-
-EXPORT_SYMBOL(get_fault_addr);
-
-void not_implemented(void)
-{
-	printk(KERN_DEBUG "Something isn't implemented in here\n");
-}
-
-EXPORT_SYMBOL(not_implemented);
-
-int user_context(unsigned long sp)
-{
-	unsigned long stack;
-
-	stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER);
-	return(stack != (unsigned long) current_thread);
-}
-
-extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end;
-
-void do_uml_exitcalls(void)
-{
-	exitcall_t *call;
-
-	call = &__uml_exitcall_end;
-	while (--call >= &__uml_exitcall_begin)
-		(*call)();
-}
-
-char *uml_strdup(char *string)
-{
-	return kstrdup(string, GFP_KERNEL);
-}
-
-int copy_to_user_proc(void __user *to, void *from, int size)
-{
-	return(copy_to_user(to, from, size));
-}
-
-int copy_from_user_proc(void *to, void __user *from, int size)
-{
-	return(copy_from_user(to, from, size));
-}
-
-int clear_user_proc(void __user *buf, int size)
-{
-	return(clear_user(buf, size));
-}
-
-int strlen_user_proc(char __user *str)
-{
-	return(strlen_user(str));
-}
-
-int smp_sigio_handler(void)
-{
-#ifdef CONFIG_SMP
-	int cpu = current_thread->cpu;
-	IPI_handler(cpu);
-	if(cpu != 0)
-		return(1);
-#endif
-	return(0);
-}
-
-int cpu(void)
-{
-	return(current_thread->cpu);
-}
-
-static atomic_t using_sysemu = ATOMIC_INIT(0);
-int sysemu_supported;
-
-void set_using_sysemu(int value)
-{
-	if (value > sysemu_supported)
-		return;
-	atomic_set(&using_sysemu, value);
-}
-
-int get_using_sysemu(void)
-{
-	return atomic_read(&using_sysemu);
-}
-
-static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
-{
-	if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
-		*eof = 1;
-
-	return strlen(buf);
-}
-
-static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data)
-{
-	char tmp[2];
-
-	if (copy_from_user(tmp, buf, 1))
-		return -EFAULT;
-
-	if (tmp[0] >= '0' && tmp[0] <= '2')
-		set_using_sysemu(tmp[0] - '0');
-	return count; /*We use the first char, but pretend to write everything*/
-}
-
-int __init make_proc_sysemu(void)
-{
-	struct proc_dir_entry *ent;
-	if (!sysemu_supported)
-		return 0;
-
-	ent = create_proc_entry("sysemu", 0600, &proc_root);
-
-	if (ent == NULL)
-	{
-		printk(KERN_WARNING "Failed to register /proc/sysemu\n");
-		return(0);
-	}
-
-	ent->read_proc  = proc_read_sysemu;
-	ent->write_proc = proc_write_sysemu;
-
-	return 0;
-}
-
-late_initcall(make_proc_sysemu);
-
-int singlestepping(void * t)
-{
-	struct task_struct *task = t ? t : current;
-
-	if ( ! (task->ptrace & PT_DTRACE) )
-		return(0);
-
-	if (task->thread.singlestep_syscall)
-		return(1);
-
-	return 2;
-}
-
-/*
- * Only x86 and x86_64 have an arch_align_stack().
- * All other arches have "#define arch_align_stack(x) (x)"
- * in their asm/system.h
- * As this is included in UML from asm-um/system-generic.h,
- * we can use it to behave as the subarch does.
- */
-#ifndef arch_align_stack
-unsigned long arch_align_stack(unsigned long sp)
-{
-	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
-		sp -= get_random_int() % 8192;
-	return sp & ~0xf;
-}
-#endif
diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile
index ea3a8e409a6e..3e3fa7e7e3cf 100644
--- a/arch/um/kernel/skas/Makefile
+++ b/arch/um/kernel/skas/Makefile
@@ -3,8 +3,7 @@
 # Licensed under the GPL
 #
 
-obj-y := clone.o exec_kern.o mem.o mmu.o process_kern.o \
-	syscall.o tlb.o uaccess.o
+obj-y := clone.o exec.o mem.o mmu.o process.o syscall.o tlb.o uaccess.o
 
 # clone.o is in the stub, so it can't be built with profiling
 # GCC hardened also auto-enables -fpic, but we need %ebx so it can't work ->
diff --git a/arch/um/kernel/skas/exec.c b/arch/um/kernel/skas/exec.c
new file mode 100644
index 000000000000..54b795951372
--- /dev/null
+++ b/arch/um/kernel/skas/exec.c
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/kernel.h"
+#include "asm/current.h"
+#include "asm/page.h"
+#include "asm/signal.h"
+#include "asm/ptrace.h"
+#include "asm/uaccess.h"
+#include "asm/mmu_context.h"
+#include "tlb.h"
+#include "skas.h"
+#include "um_mmu.h"
+#include "os.h"
+
+void flush_thread_skas(void)
+{
+	force_flush_all();
+	switch_mm_skas(&current->mm->context.skas.id);
+}
+
+void start_thread_skas(struct pt_regs *regs, unsigned long eip,
+		       unsigned long esp)
+{
+	set_fs(USER_DS);
+	PT_REGS_IP(regs) = eip;
+	PT_REGS_SP(regs) = esp;
+}
diff --git a/arch/um/kernel/skas/exec_kern.c b/arch/um/kernel/skas/exec_kern.c
deleted file mode 100644
index 77ed7bbab219..000000000000
--- a/arch/um/kernel/skas/exec_kern.c
+++ /dev/null
@@ -1,41 +0,0 @@
-/* 
- * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
- * Licensed under the GPL
- */
-
-#include "linux/kernel.h"
-#include "asm/current.h"
-#include "asm/page.h"
-#include "asm/signal.h"
-#include "asm/ptrace.h"
-#include "asm/uaccess.h"
-#include "asm/mmu_context.h"
-#include "tlb.h"
-#include "skas.h"
-#include "um_mmu.h"
-#include "os.h"
-
-void flush_thread_skas(void)
-{
-	force_flush_all();
-        switch_mm_skas(&current->mm->context.skas.id);
-}
-
-void start_thread_skas(struct pt_regs *regs, unsigned long eip, 
-		       unsigned long esp)
-{
-	set_fs(USER_DS);
-        PT_REGS_IP(regs) = eip;
-	PT_REGS_SP(regs) = esp;
-}
-
-/*
- * Overrides for Emacs so that we follow Linus's tabbing style.
- * Emacs will notice this stuff at the end of the file and automatically
- * adjust the settings for this buffer only.  This must remain at the end
- * of the file.
- * ---------------------------------------------------------------------------
- * Local variables:
- * c-file-style: "linux"
- * End:
- */
diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c
new file mode 100644
index 000000000000..ae4fa71d3b8b
--- /dev/null
+++ b/arch/um/kernel/skas/process.c
@@ -0,0 +1,217 @@
+/*
+ * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Licensed under the GPL
+ */
+
+#include "linux/sched.h"
+#include "linux/slab.h"
+#include "linux/ptrace.h"
+#include "linux/proc_fs.h"
+#include "linux/file.h"
+#include "linux/errno.h"
+#include "linux/init.h"
+#include "asm/uaccess.h"
+#include "asm/atomic.h"
+#include "kern_util.h"
+#include "skas.h"
+#include "os.h"
+#include "user_util.h"
+#include "tlb.h"
+#include "kern.h"
+#include "mode.h"
+#include "registers.h"
+
+void switch_to_skas(void *prev, void *next)
+{
+	struct task_struct *from, *to;
+
+	from = prev;
+	to = next;
+
+	/* XXX need to check runqueues[cpu].idle */
+	if(current->pid == 0)
+		switch_timers(0);
+
+	switch_threads(&from->thread.mode.skas.switch_buf,
+		       &to->thread.mode.skas.switch_buf);
+
+	arch_switch_to_skas(current->thread.prev_sched, current);
+
+	if(current->pid == 0)
+		switch_timers(1);
+}
+
+extern void schedule_tail(struct task_struct *prev);
+
+/* This is called magically, by its address being stuffed in a jmp_buf
+ * and being longjmp-d to.
+ */
+void new_thread_handler(void)
+{
+	int (*fn)(void *), n;
+	void *arg;
+
+	if(current->thread.prev_sched != NULL)
+		schedule_tail(current->thread.prev_sched);
+	current->thread.prev_sched = NULL;
+
+	fn = current->thread.request.u.thread.proc;
+	arg = current->thread.request.u.thread.arg;
+
+	/* The return value is 1 if the kernel thread execs a process,
+	 * 0 if it just exits
+	 */
+	n = run_kernel_thread(fn, arg, &current->thread.exec_buf);
+	if(n == 1){
+		/* Handle any immediate reschedules or signals */
+		interrupt_end();
+		userspace(&current->thread.regs.regs);
+	}
+	else do_exit(0);
+}
+
+void release_thread_skas(struct task_struct *task)
+{
+}
+
+/* Called magically, see new_thread_handler above */
+void fork_handler(void)
+{
+	force_flush_all();
+	if(current->thread.prev_sched == NULL)
+		panic("blech");
+
+	schedule_tail(current->thread.prev_sched);
+
+	/* XXX: if interrupt_end() calls schedule, this call to
+	 * arch_switch_to_skas isn't needed. We could want to apply this to
+	 * improve performance. -bb */
+	arch_switch_to_skas(current->thread.prev_sched, current);
+
+	current->thread.prev_sched = NULL;
+
+/* Handle any immediate reschedules or signals */
+	interrupt_end();
+
+	userspace(&current->thread.regs.regs);
+}
+
+int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
+		     unsigned long stack_top, struct task_struct * p,
+		     struct pt_regs *regs)
+{
+	void (*handler)(void);
+
+	if(current->thread.forking){
+	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
+		       sizeof(p->thread.regs.regs.skas));
+		REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0);
+		if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp;
+
+		handler = fork_handler;
+
+		arch_copy_thread(&current->thread.arch, &p->thread.arch);
+	}
+	else {
+		init_thread_registers(&p->thread.regs.regs);
+		p->thread.request.u.thread = current->thread.request.u.thread;
+		handler = new_thread_handler;
+	}
+
+	new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf,
+		   handler);
+	return(0);
+}
+
+int new_mm(unsigned long stack)
+{
+	int fd;
+
+	fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0);
+	if(fd < 0)
+		return(fd);
+
+	if(skas_needs_stub)
+		map_stub_pages(fd, CONFIG_STUB_CODE, CONFIG_STUB_DATA, stack);
+
+	return(fd);
+}
+
+void init_idle_skas(void)
+{
+	cpu_tasks[current_thread->cpu].pid = os_getpid();
+	default_idle();
+}
+
+extern void start_kernel(void);
+
+static int start_kernel_proc(void *unused)
+{
+	int pid;
+
+	block_signals();
+	pid = os_getpid();
+
+	cpu_tasks[0].pid = pid;
+	cpu_tasks[0].task = current;
+#ifdef CONFIG_SMP
+	cpu_online_map = cpumask_of_cpu(0);
+#endif
+	start_kernel();
+	return(0);
+}
+
+extern int userspace_pid[];
+
+int start_uml_skas(void)
+{
+	if(proc_mm)
+		userspace_pid[0] = start_userspace(0);
+
+	init_new_thread_signals();
+
+	init_task.thread.request.u.thread.proc = start_kernel_proc;
+	init_task.thread.request.u.thread.arg = NULL;
+	return(start_idle_thread(task_stack_page(&init_task),
+				 &init_task.thread.mode.skas.switch_buf));
+}
+
+int external_pid_skas(struct task_struct *task)
+{
+#warning Need to look up userspace_pid by cpu
+	return(userspace_pid[0]);
+}
+
+int thread_pid_skas(struct task_struct *task)
+{
+#warning Need to look up userspace_pid by cpu
+	return(userspace_pid[0]);
+}
+
+void kill_off_processes_skas(void)
+{
+	if(proc_mm)
+#warning need to loop over userspace_pids in kill_off_processes_skas
+		os_kill_ptraced_process(userspace_pid[0], 1);
+	else {
+		struct task_struct *p;
+		int pid, me;
+
+		me = os_getpid();
+		for_each_process(p){
+			if(p->mm == NULL)
+				continue;
+
+			pid = p->mm->context.skas.id.u.pid;
+			os_kill_ptraced_process(pid, 1);
+		}
+	}
+}
+
+unsigned long current_stub_stack(void)
+{
+	if(current->mm == NULL)
+		return(0);
+
+	return(current->mm->context.skas.id.stack);
+}
diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c
index c73d47c57abe..0f3d5d084dc7 100644
--- a/arch/um/kernel/skas/process_kern.c
+++ b/arch/um/kernel/skas/process_kern.c
@@ -1,217 +1,484 @@
 /*
- * Copyright (C) 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com)
+ * Copyright 2003 PathScale, Inc.
  * Licensed under the GPL
  */
 
+#include "linux/config.h"
+#include "linux/kernel.h"
 #include "linux/sched.h"
+#include "linux/interrupt.h"
+#include "linux/string.h"
+#include "linux/mm.h"
 #include "linux/slab.h"
-#include "linux/ptrace.h"
-#include "linux/proc_fs.h"
-#include "linux/file.h"
-#include "linux/errno.h"
+#include "linux/utsname.h"
+#include "linux/fs.h"
+#include "linux/utime.h"
+#include "linux/smp_lock.h"
+#include "linux/module.h"
 #include "linux/init.h"
+#include "linux/capability.h"
+#include "linux/vmalloc.h"
+#include "linux/spinlock.h"
+#include "linux/proc_fs.h"
+#include "linux/ptrace.h"
+#include "linux/random.h"
+#include "linux/personality.h"
+#include "asm/unistd.h"
+#include "asm/mman.h"
+#include "asm/segment.h"
+#include "asm/stat.h"
+#include "asm/pgtable.h"
+#include "asm/processor.h"
+#include "asm/tlbflush.h"
 #include "asm/uaccess.h"
-#include "asm/atomic.h"
-#include "kern_util.h"
-#include "skas.h"
-#include "os.h"
+#include "asm/user.h"
 #include "user_util.h"
-#include "tlb.h"
+#include "kern_util.h"
 #include "kern.h"
+#include "signal_kern.h"
+#include "init.h"
+#include "irq_user.h"
+#include "mem_user.h"
+#include "tlb.h"
+#include "frame_kern.h"
+#include "sigcontext.h"
+#include "os.h"
 #include "mode.h"
-#include "registers.h"
+#include "mode_kern.h"
+#include "choose-mode.h"
+
+/* This is a per-cpu array.  A processor only modifies its entry and it only
+ * cares about its entry, so it's OK if another processor is modifying its
+ * entry.
+ */
+struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } };
+
+int external_pid(void *t)
+{
+	struct task_struct *task = t ? t : current;
+
+	return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task));
+}
+
+int pid_to_processor_id(int pid)
+{
+	int i;
 
-void switch_to_skas(void *prev, void *next)
+	for(i = 0; i < ncpus; i++){
+		if(cpu_tasks[i].pid == pid) return(i);
+	}
+	return(-1);
+}
+
+void free_stack(unsigned long stack, int order)
 {
-	struct task_struct *from, *to;
+	free_pages(stack, order);
+}
 
-	from = prev;
-	to = next;
+unsigned long alloc_stack(int order, int atomic)
+{
+	unsigned long page;
+	gfp_t flags = GFP_KERNEL;
 
-	/* XXX need to check runqueues[cpu].idle */
-	if(current->pid == 0)
-		switch_timers(0);
+	if (atomic)
+		flags = GFP_ATOMIC;
+	page = __get_free_pages(flags, order);
+	if(page == 0)
+		return(0);
+	stack_protections(page);
+	return(page);
+}
 
-	switch_threads(&from->thread.mode.skas.switch_buf,
-		       &to->thread.mode.skas.switch_buf);
+int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
+{
+	int pid;
+
+	current->thread.request.u.thread.proc = fn;
+	current->thread.request.u.thread.arg = arg;
+	pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0,
+		      &current->thread.regs, 0, NULL, NULL);
+	if(pid < 0)
+		panic("do_fork failed in kernel_thread, errno = %d", pid);
+	return(pid);
+}
 
-	arch_switch_to_skas(current->thread.prev_sched, current);
+void set_current(void *t)
+{
+	struct task_struct *task = t;
 
-	if(current->pid == 0)
-		switch_timers(1);
+	cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task)
+		{ external_pid(task), task });
 }
 
-extern void schedule_tail(struct task_struct *prev);
+void *_switch_to(void *prev, void *next, void *last)
+{
+        struct task_struct *from = prev;
+        struct task_struct *to= next;
 
-/* This is called magically, by its address being stuffed in a jmp_buf
- * and being longjmp-d to.
- */
-void new_thread_handler(void)
+        to->thread.prev_sched = from;
+        set_current(to);
+
+	do {
+		current->thread.saved_task = NULL ;
+		CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next);
+		if(current->thread.saved_task)
+			show_regs(&(current->thread.regs));
+		next= current->thread.saved_task;
+		prev= current;
+	} while(current->thread.saved_task);
+
+        return(current->thread.prev_sched);
+
+}
+
+void interrupt_end(void)
 {
-	int (*fn)(void *), n;
-	void *arg;
+	if(need_resched()) schedule();
+	if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal();
+}
 
-	if(current->thread.prev_sched != NULL)
-		schedule_tail(current->thread.prev_sched);
-	current->thread.prev_sched = NULL;
+void release_thread(struct task_struct *task)
+{
+	CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task));
+}
 
-	fn = current->thread.request.u.thread.proc;
-	arg = current->thread.request.u.thread.arg;
+void exit_thread(void)
+{
+	unprotect_stack((unsigned long) current_thread);
+}
 
-	/* The return value is 1 if the kernel thread execs a process,
-	 * 0 if it just exits
+void *get_current(void)
+{
+	return(current);
+}
+
+int copy_thread(int nr, unsigned long clone_flags, unsigned long sp,
+		unsigned long stack_top, struct task_struct * p,
+		struct pt_regs *regs)
+{
+	int ret;
+
+	p->thread = (struct thread_struct) INIT_THREAD;
+	ret = CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr,
+				clone_flags, sp, stack_top, p, regs);
+
+	if (ret || !current->thread.forking)
+		goto out;
+
+	clear_flushed_tls(p);
+
+	/*
+	 * Set a new TLS for the child thread?
 	 */
-	n = run_kernel_thread(fn, arg, &current->thread.exec_buf);
-	if(n == 1){
-		/* Handle any immediate reschedules or signals */
-		interrupt_end();
-		userspace(&current->thread.regs.regs);
+	if (clone_flags & CLONE_SETTLS)
+		ret = arch_copy_tls(p);
+
+out:
+	return ret;
+}
+
+void initial_thread_cb(void (*proc)(void *), void *arg)
+{
+	int save_kmalloc_ok = kmalloc_ok;
+
+	kmalloc_ok = 0;
+	CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc,
+			 arg);
+	kmalloc_ok = save_kmalloc_ok;
+}
+
+unsigned long stack_sp(unsigned long page)
+{
+	return(page + PAGE_SIZE - sizeof(void *));
+}
+
+int current_pid(void)
+{
+	return(current->pid);
+}
+
+void default_idle(void)
+{
+	CHOOSE_MODE(uml_idle_timer(), (void) 0);
+
+	while(1){
+		/* endless idle loop with no priority at all */
+
+		/*
+		 * although we are an idle CPU, we do not want to
+		 * get into the scheduler unnecessarily.
+		 */
+		if(need_resched())
+			schedule();
+
+		idle_sleep(10);
 	}
-	else do_exit(0);
 }
 
-void release_thread_skas(struct task_struct *task)
+void cpu_idle(void)
 {
+	CHOOSE_MODE(init_idle_tt(), init_idle_skas());
 }
 
-/* Called magically, see new_thread_handler above */
-void fork_handler(void)
+int page_size(void)
 {
-	force_flush_all();
-	if(current->thread.prev_sched == NULL)
-		panic("blech");
+	return(PAGE_SIZE);
+}
 
-	schedule_tail(current->thread.prev_sched);
+void *um_virt_to_phys(struct task_struct *task, unsigned long addr,
+		      pte_t *pte_out)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	pte_t ptent;
+
+	if(task->mm == NULL)
+		return(ERR_PTR(-EINVAL));
+	pgd = pgd_offset(task->mm, addr);
+	if(!pgd_present(*pgd))
+		return(ERR_PTR(-EINVAL));
+
+	pud = pud_offset(pgd, addr);
+	if(!pud_present(*pud))
+		return(ERR_PTR(-EINVAL));
+
+	pmd = pmd_offset(pud, addr);
+	if(!pmd_present(*pmd))
+		return(ERR_PTR(-EINVAL));
+
+	pte = pte_offset_kernel(pmd, addr);
+	ptent = *pte;
+	if(!pte_present(ptent))
+		return(ERR_PTR(-EINVAL));
+
+	if(pte_out != NULL)
+		*pte_out = ptent;
+	return((void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK));
+}
 
-	/* XXX: if interrupt_end() calls schedule, this call to
-	 * arch_switch_to_skas isn't needed. We could want to apply this to
-	 * improve performance. -bb */
-	arch_switch_to_skas(current->thread.prev_sched, current);
+char *current_cmd(void)
+{
+#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM)
+	return("(Unknown)");
+#else
+	void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL);
+	return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr);
+#endif
+}
 
-	current->thread.prev_sched = NULL;
+void force_sigbus(void)
+{
+	printk(KERN_ERR "Killing pid %d because of a lack of memory\n",
+	       current->pid);
+	lock_kernel();
+	sigaddset(&current->pending.signal, SIGBUS);
+	recalc_sigpending();
+	current->flags |= PF_SIGNALED;
+	do_exit(SIGBUS | 0x80);
+}
 
-/* Handle any immediate reschedules or signals */
-	interrupt_end();
+void dump_thread(struct pt_regs *regs, struct user *u)
+{
+}
 
-	userspace(&current->thread.regs.regs);
+void enable_hlt(void)
+{
+	panic("enable_hlt");
 }
 
-int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp,
-		     unsigned long stack_top, struct task_struct * p,
-		     struct pt_regs *regs)
+EXPORT_SYMBOL(enable_hlt);
+
+void disable_hlt(void)
 {
-  	void (*handler)(void);
+	panic("disable_hlt");
+}
 
-	if(current->thread.forking){
-	  	memcpy(&p->thread.regs.regs.skas, &regs->regs.skas,
-		       sizeof(p->thread.regs.regs.skas));
-		REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0);
-		if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp;
+EXPORT_SYMBOL(disable_hlt);
 
-		handler = fork_handler;
+void *um_kmalloc(int size)
+{
+	return kmalloc(size, GFP_KERNEL);
+}
 
-		arch_copy_thread(&current->thread.arch, &p->thread.arch);
-	}
-	else {
-		init_thread_registers(&p->thread.regs.regs);
-                p->thread.request.u.thread = current->thread.request.u.thread;
-		handler = new_thread_handler;
-	}
+void *um_kmalloc_atomic(int size)
+{
+	return kmalloc(size, GFP_ATOMIC);
+}
 
-	new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf,
-		   handler);
-	return(0);
+void *um_vmalloc(int size)
+{
+	return vmalloc(size);
 }
 
-int new_mm(unsigned long stack)
+void *um_vmalloc_atomic(int size)
 {
-	int fd;
+	return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL);
+}
 
-	fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0);
-	if(fd < 0)
-		return(fd);
+int __cant_sleep(void) {
+	return in_atomic() || irqs_disabled() || in_interrupt();
+	/* Is in_interrupt() really needed? */
+}
 
-	if(skas_needs_stub)
-		map_stub_pages(fd, CONFIG_STUB_CODE, CONFIG_STUB_DATA, stack);
+unsigned long get_fault_addr(void)
+{
+	return((unsigned long) current->thread.fault_addr);
+}
 
-	return(fd);
+EXPORT_SYMBOL(get_fault_addr);
+
+void not_implemented(void)
+{
+	printk(KERN_DEBUG "Something isn't implemented in here\n");
 }
 
-void init_idle_skas(void)
+EXPORT_SYMBOL(not_implemented);
+
+int user_context(unsigned long sp)
 {
-	cpu_tasks[current_thread->cpu].pid = os_getpid();
-	default_idle();
+	unsigned long stack;
+
+	stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER);
+	return(stack != (unsigned long) current_thread);
 }
 
-extern void start_kernel(void);
+extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end;
 
-static int start_kernel_proc(void *unused)
+void do_uml_exitcalls(void)
 {
-	int pid;
+	exitcall_t *call;
+
+	call = &__uml_exitcall_end;
+	while (--call >= &__uml_exitcall_begin)
+		(*call)();
+}
 
-	block_signals();
-	pid = os_getpid();
+char *uml_strdup(char *string)
+{
+	return kstrdup(string, GFP_KERNEL);
+}
 
-	cpu_tasks[0].pid = pid;
-	cpu_tasks[0].task = current;
+int copy_to_user_proc(void __user *to, void *from, int size)
+{
+	return(copy_to_user(to, from, size));
+}
+
+int copy_from_user_proc(void *to, void __user *from, int size)
+{
+	return(copy_from_user(to, from, size));
+}
+
+int clear_user_proc(void __user *buf, int size)
+{
+	return(clear_user(buf, size));
+}
+
+int strlen_user_proc(char __user *str)
+{
+	return(strlen_user(str));
+}
+
+int smp_sigio_handler(void)
+{
 #ifdef CONFIG_SMP
- 	cpu_online_map = cpumask_of_cpu(0);
+	int cpu = current_thread->cpu;
+	IPI_handler(cpu);
+	if(cpu != 0)
+		return(1);
 #endif
-	start_kernel();
 	return(0);
 }
 
-extern int userspace_pid[];
-
-int start_uml_skas(void)
+int cpu(void)
 {
-	if(proc_mm)
-		userspace_pid[0] = start_userspace(0);
+	return(current_thread->cpu);
+}
 
-	init_new_thread_signals();
+static atomic_t using_sysemu = ATOMIC_INIT(0);
+int sysemu_supported;
 
-	init_task.thread.request.u.thread.proc = start_kernel_proc;
-	init_task.thread.request.u.thread.arg = NULL;
-	return(start_idle_thread(task_stack_page(&init_task),
-				 &init_task.thread.mode.skas.switch_buf));
+void set_using_sysemu(int value)
+{
+	if (value > sysemu_supported)
+		return;
+	atomic_set(&using_sysemu, value);
 }
 
-int external_pid_skas(struct task_struct *task)
+int get_using_sysemu(void)
 {
-#warning Need to look up userspace_pid by cpu
-	return(userspace_pid[0]);
+	return atomic_read(&using_sysemu);
 }
 
-int thread_pid_skas(struct task_struct *task)
+static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data)
 {
-#warning Need to look up userspace_pid by cpu
-	return(userspace_pid[0]);
+	if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/
+		*eof = 1;
+
+	return strlen(buf);
 }
 
-void kill_off_processes_skas(void)
+static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data)
 {
-	if(proc_mm)
-#warning need to loop over userspace_pids in kill_off_processes_skas
-		os_kill_ptraced_process(userspace_pid[0], 1);
-	else {
-		struct task_struct *p;
-		int pid, me;
+	char tmp[2];
 
-		me = os_getpid();
-		for_each_process(p){
-			if(p->mm == NULL)
-				continue;
+	if (copy_from_user(tmp, buf, 1))
+		return -EFAULT;
 
-			pid = p->mm->context.skas.id.u.pid;
-			os_kill_ptraced_process(pid, 1);
-		}
+	if (tmp[0] >= '0' && tmp[0] <= '2')
+		set_using_sysemu(tmp[0] - '0');
+	return count; /*We use the first char, but pretend to write everything*/
+}
+
+int __init make_proc_sysemu(void)
+{
+	struct proc_dir_entry *ent;
+	if (!sysemu_supported)
+		return 0;
+
+	ent = create_proc_entry("sysemu", 0600, &proc_root);
+
+	if (ent == NULL)
+	{
+		printk(KERN_WARNING "Failed to register /proc/sysemu\n");
+		return(0);
 	}
+
+	ent->read_proc  = proc_read_sysemu;
+	ent->write_proc = proc_write_sysemu;
+
+	return 0;
 }
 
-unsigned long current_stub_stack(void)
+late_initcall(make_proc_sysemu);
+
+int singlestepping(void * t)
 {
-	if(current->mm == NULL)
+	struct task_struct *task = t ? t : current;
+
+	if ( ! (task->ptrace & PT_DTRACE) )
 		return(0);
 
-	return(current->mm->context.skas.id.stack);
+	if (task->thread.singlestep_syscall)
+		return(1);
+
+	return 2;
+}
+
+/*
+ * Only x86 and x86_64 have an arch_align_stack().
+ * All other arches have "#define arch_align_stack(x) (x)"
+ * in their asm/system.h
+ * As this is included in UML from asm-um/system-generic.h,
+ * we can use it to behave as the subarch does.
+ */
+#ifndef arch_align_stack
+unsigned long arch_align_stack(unsigned long sp)
+{
+	if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space)
+		sp -= get_random_int() % 8192;
+	return sp & ~0xf;
 }
+#endif
-- 
cgit v1.2.1


From db9d988eaa48fa8d64cf3234a301eb9406b91de1 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:41 -0700
Subject: [PATCH] uml: add an export

Some modules need strnlen_user_skas.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/ksyms.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c
index 9cbd9bb7e6d0..f030e44262ba 100644
--- a/arch/um/kernel/ksyms.c
+++ b/arch/um/kernel/ksyms.c
@@ -47,6 +47,7 @@ EXPORT_SYMBOL(copy_to_user_tt);
 #endif
 
 #ifdef CONFIG_MODE_SKAS
+EXPORT_SYMBOL(strnlen_user_skas);
 EXPORT_SYMBOL(strncpy_from_user_skas);
 EXPORT_SYMBOL(copy_to_user_skas);
 EXPORT_SYMBOL(copy_from_user_skas);
-- 
cgit v1.2.1


From c1b4098d4852f823ecf1abb2d369cb1beb902653 Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:42 -0700
Subject: [PATCH] uml: fix sleep length bug

um_timer shouldn't add local_offset to the host time since get_time already
did it.  This threw off sleep when a settimeofday or equivalent had happened.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/kernel/time.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index 2454bbd9555d..820affbf3e16 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -95,7 +95,7 @@ irqreturn_t um_timer(int irq, void *dev, struct pt_regs *regs)
 
 	do_timer(regs);
 
-	nsecs = get_time() + local_offset;
+	nsecs = get_time();
 	xtime.tv_sec = nsecs / NSEC_PER_SEC;
 	xtime.tv_nsec = nsecs - xtime.tv_sec * NSEC_PER_SEC;
 
-- 
cgit v1.2.1


From 1183dc943cae8b0fddca0b310c26052b2355e04b Mon Sep 17 00:00:00 2001
From: Jeff Dike <jdike@addtoit.com>
Date: Wed, 27 Sep 2006 01:50:43 -0700
Subject: [PATCH] uml: fix allocation size

Fix an instance of ptr=alloc(sizeof(ptr)).  Grepping showed no more instances
of this pattern.

Also fixed the formatting in the area.

Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/net_kern.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/arch/um/drivers/net_kern.c b/arch/um/drivers/net_kern.c
index 657dfacd5ba8..664c2e2fb820 100644
--- a/arch/um/drivers/net_kern.c
+++ b/arch/um/drivers/net_kern.c
@@ -561,12 +561,13 @@ static int eth_setup(char *str)
 	int n, err;
 
 	err = eth_parse(str, &n, &str);
-	if(err) return(1);
+	if(err)
+		return 1;
 
-	new = alloc_bootmem(sizeof(new));
+	new = alloc_bootmem(sizeof(*new));
 	if (new == NULL){
 		printk("eth_init : alloc_bootmem failed\n");
-		return(1);
+		return 1;
 	}
 
 	INIT_LIST_HEAD(&new->list);
@@ -574,7 +575,7 @@ static int eth_setup(char *str)
 	new->init = str;
 
 	list_add_tail(&new->list, &eth_cmd_line);
-	return(1);
+	return 1;
 }
 
 __setup("eth", eth_setup);
-- 
cgit v1.2.1


From bcdc5e019d9f525a9f181a7de642d3a9c27c7610 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 27 Sep 2006 01:50:44 -0700
Subject: [PATCH] autofs4 needs to force fail return revalidate

For a long time now I have had a problem with not being able to return a
lookup failure on an existsing directory.  In autofs this corresponds to a
mount failure on a autofs managed mount entry that is browsable (and so the
mount point directory exists).

While this problem has been present for a long time I've avoided resolving
it because it was not very visible.  But now that autofs v5 has "mount and
expire on demand" of nested multiple mounts, such as is found when mounting
an export list from a server, solving the problem cannot be avoided any
longer.

I've tried very hard to find a way to do this entirely within the autofs4
module but have not been able to find a satisfactory way to achieve it.

So, I need to propose a change to the VFS.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/autofs4/root.c | 38 ++++++++++++++++++++++++++++++--------
 fs/namei.c        | 50 +++++++++++++++++++++++++++++++++++---------------
 2 files changed, 65 insertions(+), 23 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 5100f984783f..27e17f96cada 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -137,7 +137,9 @@ static int autofs4_dir_open(struct inode *inode, struct file *file)
 		nd.flags = LOOKUP_DIRECTORY;
 		ret = (dentry->d_op->d_revalidate)(dentry, &nd);
 
-		if (!ret) {
+		if (ret <= 0) {
+			if (ret < 0)
+				status = ret;
 			dcache_dir_close(inode, file);
 			goto out;
 		}
@@ -400,13 +402,23 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 	struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb);
 	int oz_mode = autofs4_oz_mode(sbi);
 	int flags = nd ? nd->flags : 0;
-	int status = 0;
+	int status = 1;
 
 	/* Pending dentry */
 	if (autofs4_ispending(dentry)) {
-		if (!oz_mode)
-			status = try_to_fill_dentry(dentry, flags);
-		return !status;
+		/* The daemon never causes a mount to trigger */
+		if (oz_mode)
+			return 1;
+
+		/*
+		 * A zero status is success otherwise we have a
+		 * negative error code.
+		 */
+		status = try_to_fill_dentry(dentry, flags);
+		if (status == 0)
+				return 1;
+
+		return status;
 	}
 
 	/* Negative dentry.. invalidate if "old" */
@@ -421,9 +433,19 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		DPRINTK("dentry=%p %.*s, emptydir",
 			 dentry, dentry->d_name.len, dentry->d_name.name);
 		spin_unlock(&dcache_lock);
-		if (!oz_mode)
-			status = try_to_fill_dentry(dentry, flags);
-		return !status;
+		/* The daemon never causes a mount to trigger */
+		if (oz_mode)
+			return 1;
+
+		/*
+		 * A zero status is success otherwise we have a
+		 * negative error code.
+		 */
+		status = try_to_fill_dentry(dentry, flags);
+		if (status == 0)
+			return 1;
+
+		return status;
 	}
 	spin_unlock(&dcache_lock);
 
diff --git a/fs/namei.c b/fs/namei.c
index 6b591c01b09f..808e4ea2bb94 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -372,6 +372,30 @@ void release_open_intent(struct nameidata *nd)
 		fput(nd->intent.open.file);
 }
 
+static inline struct dentry *
+do_revalidate(struct dentry *dentry, struct nameidata *nd)
+{
+	int status = dentry->d_op->d_revalidate(dentry, nd);
+	if (unlikely(status <= 0)) {
+		/*
+		 * The dentry failed validation.
+		 * If d_revalidate returned 0 attempt to invalidate
+		 * the dentry otherwise d_revalidate is asking us
+		 * to return a fail status.
+		 */
+		if (!status) {
+			if (!d_invalidate(dentry)) {
+				dput(dentry);
+				dentry = NULL;
+			}
+		} else {
+			dput(dentry);
+			dentry = ERR_PTR(status);
+		}
+	}
+	return dentry;
+}
+
 /*
  * Internal lookup() using the new generic dcache.
  * SMP-safe
@@ -386,12 +410,9 @@ static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name,
 	if (!dentry)
 		dentry = d_lookup(parent, name);
 
-	if (dentry && dentry->d_op && dentry->d_op->d_revalidate) {
-		if (!dentry->d_op->d_revalidate(dentry, nd) && !d_invalidate(dentry)) {
-			dput(dentry);
-			dentry = NULL;
-		}
-	}
+	if (dentry && dentry->d_op && dentry->d_op->d_revalidate)
+		dentry = do_revalidate(dentry, nd);
+
 	return dentry;
 }
 
@@ -484,10 +505,9 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s
 	 */
 	mutex_unlock(&dir->i_mutex);
 	if (result->d_op && result->d_op->d_revalidate) {
-		if (!result->d_op->d_revalidate(result, nd) && !d_invalidate(result)) {
-			dput(result);
+		result = do_revalidate(result, nd);
+		if (!result)
 			result = ERR_PTR(-ENOENT);
-		}
 	}
 	return result;
 }
@@ -767,12 +787,12 @@ need_lookup:
 	goto done;
 
 need_revalidate:
-	if (dentry->d_op->d_revalidate(dentry, nd))
-		goto done;
-	if (d_invalidate(dentry))
-		goto done;
-	dput(dentry);
-	goto need_lookup;
+	dentry = do_revalidate(dentry, nd);
+	if (!dentry)
+		goto need_lookup;
+	if (IS_ERR(dentry))
+		goto fail;
+	goto done;
 
 fail:
 	return PTR_ERR(dentry);
-- 
cgit v1.2.1


From 7e96287ddc4f42081e18248b6167041c0908004c Mon Sep 17 00:00:00 2001
From: Vivek Goyal <vgoyal@in.ibm.com>
Date: Wed, 27 Sep 2006 01:50:44 -0700
Subject: [PATCH] kdump: introduce "reset_devices" command line option

Resetting the devices during driver initialization can be a costly
operation in terms of time (especially scsi devices).  This option can be
used by drivers to know that user forcibly wants the devices to be reset
during initialization.

This option can be useful while kernel is booting in unreliable
environment.  For ex.  during kdump boot where devices are in unknown
random state and BIOS execution has been skipped.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |  3 +++
 include/linux/init.h                |  1 +
 init/main.c                         | 20 ++++++++++++++++++++
 3 files changed, 24 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 255ec535bba8..54983246930d 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1370,6 +1370,9 @@ running once the system is up.
 			Reserves a hole at the top of the kernel virtual
 			address space.
 
+	reset_devices	[KNL] Force drivers to reset the underlying device
+			during initialization.
+
 	resume=		[SWSUSP]
 			Specify the partition device for software suspend
 
diff --git a/include/linux/init.h b/include/linux/init.h
index 6667785dd1ff..e92b1455d7af 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -68,6 +68,7 @@ extern initcall_t __security_initcall_start[], __security_initcall_end[];
 
 /* Defined in init/main.c */
 extern char saved_command_line[];
+extern unsigned int reset_devices;
 
 /* used by init/main.c */
 extern void setup_arch(char **);
diff --git a/init/main.c b/init/main.c
index 913e48d658ee..0766e69712b2 100644
--- a/init/main.c
+++ b/init/main.c
@@ -127,6 +127,18 @@ static char *ramdisk_execute_command;
 /* Setup configured maximum number of CPUs to activate */
 static unsigned int max_cpus = NR_CPUS;
 
+/*
+ * If set, this is an indication to the drivers that reset the underlying
+ * device before going ahead with the initialization otherwise driver might
+ * rely on the BIOS and skip the reset operation.
+ *
+ * This is useful if kernel is booting in an unreliable environment.
+ * For ex. kdump situaiton where previous kernel has crashed, BIOS has been
+ * skipped and devices will be in unknown state.
+ */
+unsigned int reset_devices;
+EXPORT_SYMBOL(reset_devices);
+
 /*
  * Setup routine for controlling SMP activation
  *
@@ -153,6 +165,14 @@ static int __init maxcpus(char *str)
 
 __setup("maxcpus=", maxcpus);
 
+static int __init set_reset_devices(char *str)
+{
+	reset_devices = 1;
+	return 1;
+}
+
+__setup("reset_devices", set_reset_devices);
+
 static char * argv_init[MAX_INIT_ARGS+2] = { "init", NULL, };
 char * envp_init[MAX_INIT_ENVS+2] = { "HOME=/", "TERM=linux", NULL, };
 static const char *panic_later, *panic_param;
-- 
cgit v1.2.1


From 6a1d9805ec506d8b9d04450997707da5f643d87c Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Wed, 27 Sep 2006 01:50:45 -0700
Subject: [PATCH] fat: cleanup fat_get_block(s)

get_blocks() was removed.  So, this removes it on fat, and will take
advantage of the multi block mapping.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/fat/inode.c | 29 ++++++++++++-----------------
 1 file changed, 12 insertions(+), 17 deletions(-)

diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index d30151190434..e1035a590664 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -50,14 +50,14 @@ static int fat_add_cluster(struct inode *inode)
 	return err;
 }
 
-static int __fat_get_blocks(struct inode *inode, sector_t iblock,
-			    unsigned long *max_blocks,
-			    struct buffer_head *bh_result, int create)
+static inline int __fat_get_block(struct inode *inode, sector_t iblock,
+				  unsigned long *max_blocks,
+				  struct buffer_head *bh_result, int create)
 {
 	struct super_block *sb = inode->i_sb;
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
-	sector_t phys;
 	unsigned long mapped_blocks;
+	sector_t phys;
 	int err, offset;
 
 	err = fat_bmap(inode, iblock, &phys, &mapped_blocks);
@@ -73,7 +73,7 @@ static int __fat_get_blocks(struct inode *inode, sector_t iblock,
 
 	if (iblock != MSDOS_I(inode)->mmu_private >> sb->s_blocksize_bits) {
 		fat_fs_panic(sb, "corrupted file size (i_pos %lld, %lld)",
-			     MSDOS_I(inode)->i_pos, MSDOS_I(inode)->mmu_private);
+			MSDOS_I(inode)->i_pos, MSDOS_I(inode)->mmu_private);
 		return -EIO;
 	}
 
@@ -93,34 +93,29 @@ static int __fat_get_blocks(struct inode *inode, sector_t iblock,
 	err = fat_bmap(inode, iblock, &phys, &mapped_blocks);
 	if (err)
 		return err;
+
 	BUG_ON(!phys);
 	BUG_ON(*max_blocks != mapped_blocks);
 	set_buffer_new(bh_result);
 	map_bh(bh_result, sb, phys);
+
 	return 0;
 }
 
-static int fat_get_blocks(struct inode *inode, sector_t iblock,
-			  struct buffer_head *bh_result, int create)
+static int fat_get_block(struct inode *inode, sector_t iblock,
+			 struct buffer_head *bh_result, int create)
 {
 	struct super_block *sb = inode->i_sb;
-	int err;
 	unsigned long max_blocks = bh_result->b_size >> inode->i_blkbits;
+	int err;
 
-	err = __fat_get_blocks(inode, iblock, &max_blocks, bh_result, create);
+	err = __fat_get_block(inode, iblock, &max_blocks, bh_result, create);
 	if (err)
 		return err;
 	bh_result->b_size = max_blocks << sb->s_blocksize_bits;
 	return 0;
 }
 
-static int fat_get_block(struct inode *inode, sector_t iblock,
-			 struct buffer_head *bh_result, int create)
-{
-	unsigned long max_blocks = 1;
-	return __fat_get_blocks(inode, iblock, &max_blocks, bh_result, create);
-}
-
 static int fat_writepage(struct page *page, struct writeback_control *wbc)
 {
 	return block_write_full_page(page, fat_get_block, wbc);
@@ -188,7 +183,7 @@ static ssize_t fat_direct_IO(int rw, struct kiocb *iocb,
 	 * condition of fat_get_block() and ->truncate().
 	 */
 	return blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
-				  offset, nr_segs, fat_get_blocks, NULL);
+				  offset, nr_segs, fat_get_block, NULL);
 }
 
 static sector_t _fat_bmap(struct address_space *mapping, sector_t block)
-- 
cgit v1.2.1


From 8e18e2941c53416aa219708e7dcad21fb4bd6794 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:46 -0700
Subject: [PATCH] inode_diet: Replace inode.u.generic_ip with inode.i_private

The following patches reduce the size of the VFS inode structure by 28 bytes
on a UP x86.  (It would be more on an x86_64 system).  This is a 10% reduction
in the inode size on a UP kernel that is configured in a production mode
(i.e., with no spinlock or other debugging functions enabled; if you want to
save memory taken up by in-core inodes, the first thing you should do is
disable the debugging options; they are responsible for a huge amount of bloat
in the VFS inode structure).

This patch:

The filesystem or device-specific pointer in the inode is inside a union,
which is pretty pointless given that all 30+ users of this field have been
using the void pointer.  Get rid of the union and rename it to i_private, with
a comment to explain who is allowed to use the void pointer.  This is just a
cleanup, but it allows us to reuse the union 'u' for something something where
the union will actually be used.

[judith@osdl.org: powerpc build fix]
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Judith Lebzelter <judith@osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/cell/spufs/inode.c      |  2 +-
 arch/powerpc/platforms/pseries/hvCall_inst.c   |  2 +-
 arch/s390/hypfs/inode.c                        |  6 ++---
 arch/s390/kernel/debug.c                       |  2 +-
 block/blktrace.c                               |  2 +-
 drivers/i2c/chips/tps65010.c                   |  2 +-
 drivers/infiniband/hw/ipath/ipath_fs.c         | 12 ++++-----
 drivers/infiniband/ulp/ipoib/ipoib_fs.c        |  4 +--
 drivers/misc/ibmasm/ibmasmfs.c                 | 16 ++++++------
 drivers/net/irda/vlsi_ir.h                     |  2 +-
 drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c |  2 +-
 drivers/oprofile/oprofilefs.c                  | 10 ++++----
 drivers/pci/hotplug/cpqphp_sysfs.c             |  2 +-
 drivers/usb/core/devio.c                       |  2 +-
 drivers/usb/core/inode.c                       |  6 ++---
 drivers/usb/gadget/inode.c                     |  6 ++---
 drivers/usb/host/isp116x-hcd.c                 |  2 +-
 drivers/usb/host/uhci-debug.c                  |  2 +-
 drivers/usb/mon/mon_stat.c                     |  2 +-
 drivers/usb/mon/mon_text.c                     |  4 +--
 fs/autofs/inode.c                              |  2 +-
 fs/autofs/symlink.c                            |  2 +-
 fs/binfmt_misc.c                               |  8 +++---
 fs/debugfs/file.c                              |  4 +--
 fs/debugfs/inode.c                             |  4 +--
 fs/devpts/inode.c                              |  4 +--
 fs/freevxfs/vxfs.h                             |  2 +-
 fs/freevxfs/vxfs_inode.c                       |  4 +--
 fs/fuse/control.c                              |  6 ++---
 fs/inode.c                                     |  2 +-
 fs/jffs/inode-v23.c                            | 34 +++++++++++++-------------
 fs/libfs.c                                     |  2 +-
 fs/ocfs2/dlmglue.c                             |  2 +-
 include/linux/fs.h                             |  4 +--
 kernel/relay.c                                 |  2 +-
 security/inode.c                               |  8 +++---
 36 files changed, 88 insertions(+), 90 deletions(-)

diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index 7b4572805db9..b837f12a84ae 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -120,7 +120,7 @@ spufs_new_file(struct super_block *sb, struct dentry *dentry,
 	ret = 0;
 	inode->i_op = &spufs_file_iops;
 	inode->i_fop = fops;
-	inode->u.generic_ip = SPUFS_I(inode)->i_ctx = get_spu_context(ctx);
+	inode->i_private = SPUFS_I(inode)->i_ctx = get_spu_context(ctx);
 	d_add(dentry, inode);
 out:
 	return ret;
diff --git a/arch/powerpc/platforms/pseries/hvCall_inst.c b/arch/powerpc/platforms/pseries/hvCall_inst.c
index 641e6511cf06..446e17d162a5 100644
--- a/arch/powerpc/platforms/pseries/hvCall_inst.c
+++ b/arch/powerpc/platforms/pseries/hvCall_inst.c
@@ -85,7 +85,7 @@ static int hcall_inst_seq_open(struct inode *inode, struct file *file)
 
 	rc = seq_open(file, &hcall_inst_seq_ops);
 	seq = file->private_data;
-	seq->private = file->f_dentry->d_inode->u.generic_ip;
+	seq->private = file->f_dentry->d_inode->i_private;
 
 	return rc;
 }
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index bdade5f2e325..8735024d235b 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -104,13 +104,13 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode)
 
 static void hypfs_drop_inode(struct inode *inode)
 {
-	kfree(inode->u.generic_ip);
+	kfree(inode->i_private);
 	generic_delete_inode(inode);
 }
 
 static int hypfs_open(struct inode *inode, struct file *filp)
 {
-	char *data = filp->f_dentry->d_inode->u.generic_ip;
+	char *data = filp->f_dentry->d_inode->i_private;
 	struct hypfs_sb_info *fs_info;
 
 	if (filp->f_mode & FMODE_WRITE) {
@@ -352,7 +352,7 @@ static struct dentry *hypfs_create_file(struct super_block *sb,
 		parent->d_inode->i_nlink++;
 	} else
 		BUG();
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 	d_instantiate(dentry, inode);
 	dget(dentry);
 	return dentry;
diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c
index 7ba20922a535..43f3d0c7e132 100644
--- a/arch/s390/kernel/debug.c
+++ b/arch/s390/kernel/debug.c
@@ -603,7 +603,7 @@ debug_open(struct inode *inode, struct file *file)
 	debug_info_t *debug_info, *debug_info_snapshot;
 
 	down(&debug_lock);
-	debug_info = (struct debug_info*)file->f_dentry->d_inode->u.generic_ip;
+	debug_info = file->f_dentry->d_inode->i_private;
 	/* find debug view */
 	for (i = 0; i < DEBUG_MAX_VIEWS; i++) {
 		if (!debug_info->views[i])
diff --git a/block/blktrace.c b/block/blktrace.c
index 265f7a830619..2b4ef2b89b8d 100644
--- a/block/blktrace.c
+++ b/block/blktrace.c
@@ -217,7 +217,7 @@ static int blk_trace_remove(request_queue_t *q)
 
 static int blk_dropped_open(struct inode *inode, struct file *filp)
 {
-	filp->private_data = inode->u.generic_ip;
+	filp->private_data = inode->i_private;
 
 	return 0;
 }
diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c
index 0be6fd6a267d..6a7578217177 100644
--- a/drivers/i2c/chips/tps65010.c
+++ b/drivers/i2c/chips/tps65010.c
@@ -305,7 +305,7 @@ static int dbg_show(struct seq_file *s, void *_)
 
 static int dbg_tps_open(struct inode *inode, struct file *file)
 {
-	return single_open(file, dbg_show, inode->u.generic_ip);
+	return single_open(file, dbg_show, inode->i_private);
 }
 
 static struct file_operations debug_fops = {
diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c
index a5eb30a06a5c..055cdd089b28 100644
--- a/drivers/infiniband/hw/ipath/ipath_fs.c
+++ b/drivers/infiniband/hw/ipath/ipath_fs.c
@@ -64,7 +64,7 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 	if ((mode & S_IFMT) == S_IFDIR) {
 		inode->i_op = &simple_dir_inode_operations;
 		inode->i_nlink++;
@@ -119,7 +119,7 @@ static ssize_t atomic_counters_read(struct file *file, char __user *buf,
 	u16 i;
 	struct ipath_devdata *dd;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	for (i = 0; i < NUM_COUNTERS; i++)
 		counters[i] = ipath_snap_cntr(dd, i);
@@ -139,7 +139,7 @@ static ssize_t atomic_node_info_read(struct file *file, char __user *buf,
 	struct ipath_devdata *dd;
 	u64 guid;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	guid = be64_to_cpu(dd->ipath_guid);
 
@@ -178,7 +178,7 @@ static ssize_t atomic_port_info_read(struct file *file, char __user *buf,
 	u32 tmp, tmp2;
 	struct ipath_devdata *dd;
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 
 	/* so we only initialize non-zero fields. */
 	memset(portinfo, 0, sizeof portinfo);
@@ -325,7 +325,7 @@ static ssize_t flash_read(struct file *file, char __user *buf,
 		goto bail;
 	}
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 	if (ipath_eeprom_read(dd, pos, tmp, count)) {
 		ipath_dev_err(dd, "failed to read from flash\n");
 		ret = -ENXIO;
@@ -381,7 +381,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
 		goto bail_tmp;
 	}
 
-	dd = file->f_dentry->d_inode->u.generic_ip;
+	dd = file->f_dentry->d_inode->i_private;
 	if (ipath_eeprom_write(dd, pos, tmp, count)) {
 		ret = -ENXIO;
 		ipath_dev_err(dd, "failed to write to flash\n");
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c
index 5dde380e8dbe..f1cb83688b31 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c
@@ -141,7 +141,7 @@ static int ipoib_mcg_open(struct inode *inode, struct file *file)
 		return ret;
 
 	seq = file->private_data;
-	seq->private = inode->u.generic_ip;
+	seq->private = inode->i_private;
 
 	return 0;
 }
@@ -247,7 +247,7 @@ static int ipoib_path_open(struct inode *inode, struct file *file)
 		return ret;
 
 	seq = file->private_data;
-	seq->private = inode->u.generic_ip;
+	seq->private = inode->i_private;
 
 	return 0;
 }
diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
index 4a35caff5d02..0e909b617226 100644
--- a/drivers/misc/ibmasm/ibmasmfs.c
+++ b/drivers/misc/ibmasm/ibmasmfs.c
@@ -175,7 +175,7 @@ static struct dentry *ibmasmfs_create_file (struct super_block *sb,
 	}
 
 	inode->i_fop = fops;
-	inode->u.generic_ip = data;
+	inode->i_private = data;
 
 	d_add(dentry, inode);
 	return dentry;
@@ -244,7 +244,7 @@ static int command_file_open(struct inode *inode, struct file *file)
 {
 	struct ibmasmfs_command_data *command_data;
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
 	command_data = kmalloc(sizeof(struct ibmasmfs_command_data), GFP_KERNEL);
@@ -252,7 +252,7 @@ static int command_file_open(struct inode *inode, struct file *file)
 		return -ENOMEM;
 
 	command_data->command = NULL;
-	command_data->sp = inode->u.generic_ip;
+	command_data->sp = inode->i_private;
 	file->private_data = command_data;
 	return 0;
 }
@@ -351,10 +351,10 @@ static int event_file_open(struct inode *inode, struct file *file)
 	struct ibmasmfs_event_data *event_data;
 	struct service_processor *sp; 
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
-	sp = inode->u.generic_ip;
+	sp = inode->i_private;
 
 	event_data = kmalloc(sizeof(struct ibmasmfs_event_data), GFP_KERNEL);
 	if (!event_data)
@@ -439,14 +439,14 @@ static int r_heartbeat_file_open(struct inode *inode, struct file *file)
 {
 	struct ibmasmfs_heartbeat_data *rhbeat;
 
-	if (!inode->u.generic_ip)
+	if (!inode->i_private)
 		return -ENODEV;
 
 	rhbeat = kmalloc(sizeof(struct ibmasmfs_heartbeat_data), GFP_KERNEL);
 	if (!rhbeat)
 		return -ENOMEM;
 
-	rhbeat->sp = (struct service_processor *)inode->u.generic_ip;
+	rhbeat->sp = inode->i_private;
 	rhbeat->active = 0;
 	ibmasm_init_reverse_heartbeat(rhbeat->sp, &rhbeat->heartbeat);
 	file->private_data = rhbeat;
@@ -508,7 +508,7 @@ static ssize_t r_heartbeat_file_write(struct file *file, const char __user *buf,
 
 static int remote_settings_file_open(struct inode *inode, struct file *file)
 {
-	file->private_data = inode->u.generic_ip;
+	file->private_data = inode->i_private;
 	return 0;
 }
 
diff --git a/drivers/net/irda/vlsi_ir.h b/drivers/net/irda/vlsi_ir.h
index a82a4ba8de4f..c37f0bc4c7f9 100644
--- a/drivers/net/irda/vlsi_ir.h
+++ b/drivers/net/irda/vlsi_ir.h
@@ -58,7 +58,7 @@ typedef void irqreturn_t;
 
 /* PDE() introduced in 2.5.4 */
 #ifdef CONFIG_PROC_FS
-#define PDE(inode) ((inode)->u.generic_ip)
+#define PDE(inode) ((inode)->i_private)
 #endif
 
 /* irda crc16 calculation exported in 2.5.42 */
diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
index 923275ea0789..b9df06a06ea9 100644
--- a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
+++ b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c
@@ -54,7 +54,7 @@ static ssize_t write_file_dummy(struct file *file, const char __user *buf,
 
 static int open_file_generic(struct inode *inode, struct file *file)
 {
-	file->private_data = inode->u.generic_ip;
+	file->private_data = inode->i_private;
 	return 0;
 }
 
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index 71c2da277d6e..deb37354785b 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -110,8 +110,8 @@ static ssize_t ulong_write_file(struct file * file, char const __user * buf, siz
 
 static int default_open(struct inode * inode, struct file * filp)
 {
-	if (inode->u.generic_ip)
-		filp->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		filp->private_data = inode->i_private;
 	return 0;
 }
 
@@ -158,7 +158,7 @@ int oprofilefs_create_ulong(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
@@ -171,7 +171,7 @@ int oprofilefs_create_ro_ulong(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
@@ -197,7 +197,7 @@ int oprofilefs_create_ro_atomic(struct super_block * sb, struct dentry * root,
 	if (!d)
 		return -EFAULT;
 
-	d->d_inode->u.generic_ip = val;
+	d->d_inode->i_private = val;
 	return 0;
 }
 
diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c
index 8b3da007e859..5bab666cd67e 100644
--- a/drivers/pci/hotplug/cpqphp_sysfs.c
+++ b/drivers/pci/hotplug/cpqphp_sysfs.c
@@ -140,7 +140,7 @@ struct ctrl_dbg {
 
 static int open(struct inode *inode, struct file *file)
 {
-	struct controller *ctrl = inode->u.generic_ip;
+	struct controller *ctrl = inode->i_private;
 	struct ctrl_dbg *dbg;
 	int retval = -ENOMEM;
 
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 218621b9958e..32e03000420c 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -555,7 +555,7 @@ static int usbdev_open(struct inode *inode, struct file *file)
 	if (imajor(inode) == USB_DEVICE_MAJOR)
 		dev = usbdev_lookup_minor(iminor(inode));
 	if (!dev)
-		dev = inode->u.generic_ip;
+		dev = inode->i_private;
 	if (!dev) {
 		kfree(ps);
 		goto out;
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 3182c2224ba2..482f253085e5 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -402,8 +402,8 @@ static loff_t default_file_lseek (struct file *file, loff_t offset, int orig)
 
 static int default_open (struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
@@ -509,7 +509,7 @@ static struct dentry *fs_create_file (const char *name, mode_t mode,
 	} else {
 		if (dentry->d_inode) {
 			if (data)
-				dentry->d_inode->u.generic_ip = data;
+				dentry->d_inode->i_private = data;
 			if (fops)
 				dentry->d_inode->i_fop = fops;
 			dentry->d_inode->i_uid = uid;
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 3bdc5e3ba234..ffaa8c1afad8 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -844,7 +844,7 @@ fail1:
 static int
 ep_open (struct inode *inode, struct file *fd)
 {
-	struct ep_data		*data = inode->u.generic_ip;
+	struct ep_data		*data = inode->i_private;
 	int			value = -EBUSY;
 
 	if (down_interruptible (&data->lock) != 0)
@@ -1909,7 +1909,7 @@ fail:
 static int
 dev_open (struct inode *inode, struct file *fd)
 {
-	struct dev_data		*dev = inode->u.generic_ip;
+	struct dev_data		*dev = inode->i_private;
 	int			value = -EBUSY;
 
 	if (dev->state == STATE_DEV_DISABLED) {
@@ -1970,7 +1970,7 @@ gadgetfs_make_inode (struct super_block *sb,
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime
 				= CURRENT_TIME;
-		inode->u.generic_ip = data;
+		inode->i_private = data;
 		inode->i_fop = fops;
 	}
 	return inode;
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
index 5147ed4a6662..8c6b38a0b5bb 100644
--- a/drivers/usb/host/isp116x-hcd.c
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -1204,7 +1204,7 @@ static int isp116x_show_dbg(struct seq_file *s, void *unused)
 
 static int isp116x_open_seq(struct inode *inode, struct file *file)
 {
-	return single_open(file, isp116x_show_dbg, inode->u.generic_ip);
+	return single_open(file, isp116x_show_dbg, inode->i_private);
 }
 
 static struct file_operations isp116x_debug_fops = {
diff --git a/drivers/usb/host/uhci-debug.c b/drivers/usb/host/uhci-debug.c
index dc286a48cafd..d1372cb27f33 100644
--- a/drivers/usb/host/uhci-debug.c
+++ b/drivers/usb/host/uhci-debug.c
@@ -428,7 +428,7 @@ struct uhci_debug {
 
 static int uhci_debug_open(struct inode *inode, struct file *file)
 {
-	struct uhci_hcd *uhci = inode->u.generic_ip;
+	struct uhci_hcd *uhci = inode->i_private;
 	struct uhci_debug *up;
 	int ret = -ENOMEM;
 	unsigned long flags;
diff --git a/drivers/usb/mon/mon_stat.c b/drivers/usb/mon/mon_stat.c
index 1fe01d994a79..86ad2b381c4b 100644
--- a/drivers/usb/mon/mon_stat.c
+++ b/drivers/usb/mon/mon_stat.c
@@ -28,7 +28,7 @@ static int mon_stat_open(struct inode *inode, struct file *file)
 	if ((sp = kmalloc(sizeof(struct snap), GFP_KERNEL)) == NULL)
 		return -ENOMEM;
 
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 
 	sp->slen = snprintf(sp->str, STAT_BUF_SIZE,
 	    "nreaders %d events %u text_lost %u\n",
diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c
index f961a770cee2..2fd39b4fa166 100644
--- a/drivers/usb/mon/mon_text.c
+++ b/drivers/usb/mon/mon_text.c
@@ -238,7 +238,7 @@ static int mon_text_open(struct inode *inode, struct file *file)
 	int rc;
 
 	mutex_lock(&mon_lock);
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 	ubus = mbus->u_bus;
 
 	rp = kzalloc(sizeof(struct mon_reader_text), GFP_KERNEL);
@@ -401,7 +401,7 @@ static int mon_text_release(struct inode *inode, struct file *file)
 	struct mon_event_text *ep;
 
 	mutex_lock(&mon_lock);
-	mbus = inode->u.generic_ip;
+	mbus = inode->i_private;
 
 	if (mbus->nreaders <= 0) {
 		printk(KERN_ERR TAG ": consistency error on close\n");
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index c81d6b8c2828..d39d2acd9b38 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -241,7 +241,7 @@ static void autofs_read_inode(struct inode *inode)
 		
 		inode->i_op = &autofs_symlink_inode_operations;
 		sl = &sbi->symlink[n];
-		inode->u.generic_ip = sl;
+		inode->i_private = sl;
 		inode->i_mode = S_IFLNK | S_IRWXUGO;
 		inode->i_mtime.tv_sec = inode->i_ctime.tv_sec = sl->mtime;
 		inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0;
diff --git a/fs/autofs/symlink.c b/fs/autofs/symlink.c
index 52e8772b066e..c74f2eb65775 100644
--- a/fs/autofs/symlink.c
+++ b/fs/autofs/symlink.c
@@ -15,7 +15,7 @@
 /* Nothing to release.. */
 static void *autofs_follow_link(struct dentry *dentry, struct nameidata *nd)
 {
-	char *s=((struct autofs_symlink *)dentry->d_inode->u.generic_ip)->data;
+	char *s=((struct autofs_symlink *)dentry->d_inode->i_private)->data;
 	nd_set_link(nd, s);
 	return NULL;
 }
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 34ebbc191e46..6759b9839ce8 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -517,7 +517,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
 
 static void bm_clear_inode(struct inode *inode)
 {
-	kfree(inode->u.generic_ip);
+	kfree(inode->i_private);
 }
 
 static void kill_node(Node *e)
@@ -545,7 +545,7 @@ static void kill_node(Node *e)
 static ssize_t
 bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos)
 {
-	Node *e = file->f_dentry->d_inode->u.generic_ip;
+	Node *e = file->f_dentry->d_inode->i_private;
 	loff_t pos = *ppos;
 	ssize_t res;
 	char *page;
@@ -579,7 +579,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer,
 				size_t count, loff_t *ppos)
 {
 	struct dentry *root;
-	Node *e = file->f_dentry->d_inode->u.generic_ip;
+	Node *e = file->f_dentry->d_inode->i_private;
 	int res = parse_command(buffer, count);
 
 	switch (res) {
@@ -646,7 +646,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer,
 	}
 
 	e->dentry = dget(dentry);
-	inode->u.generic_ip = e;
+	inode->i_private = e;
 	inode->i_fop = &bm_entry_operations;
 
 	d_instantiate(dentry, inode);
diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c
index e4b430552c88..bf3901ab1744 100644
--- a/fs/debugfs/file.c
+++ b/fs/debugfs/file.c
@@ -32,8 +32,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf,
 
 static int default_open(struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 3ca268d2e5a2..717f4821ed02 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -168,7 +168,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode,
  *          directory dentry if set.  If this paramater is NULL, then the
  *          file will be created in the root of the debugfs filesystem.
  * @data: a pointer to something that the caller will want to get to later
- *        on.  The inode.u.generic_ip pointer will point to this value on
+ *        on.  The inode.i_private pointer will point to this value on
  *        the open() call.
  * @fops: a pointer to a struct file_operations that should be used for
  *        this file.
@@ -209,7 +209,7 @@ struct dentry *debugfs_create_file(const char *name, mode_t mode,
 
 	if (dentry->d_inode) {
 		if (data)
-			dentry->d_inode->u.generic_ip = data;
+			dentry->d_inode->i_private = data;
 		if (fops)
 			dentry->d_inode->i_fop = fops;
 	}
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index f7aef5bb584a..5bf06a10dddf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -177,7 +177,7 @@ int devpts_pty_new(struct tty_struct *tty)
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	init_special_inode(inode, S_IFCHR|config.mode, device);
-	inode->u.generic_ip = tty;
+	inode->i_private = tty;
 
 	dentry = get_node(number);
 	if (!IS_ERR(dentry) && !dentry->d_inode)
@@ -196,7 +196,7 @@ struct tty_struct *devpts_get_tty(int number)
 	tty = NULL;
 	if (!IS_ERR(dentry)) {
 		if (dentry->d_inode)
-			tty = dentry->d_inode->u.generic_ip;
+			tty = dentry->d_inode->i_private;
 		dput(dentry);
 	}
 
diff --git a/fs/freevxfs/vxfs.h b/fs/freevxfs/vxfs.h
index d35979a58743..c8a92652612a 100644
--- a/fs/freevxfs/vxfs.h
+++ b/fs/freevxfs/vxfs.h
@@ -252,7 +252,7 @@ enum {
  * Get filesystem private data from VFS inode.
  */
 #define VXFS_INO(ip) \
-	((struct vxfs_inode_info *)(ip)->u.generic_ip)
+	((struct vxfs_inode_info *)(ip)->i_private)
 
 /*
  * Get filesystem private data from VFS superblock.
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index ca6a39714771..32a82ed108e4 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -243,7 +243,7 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip)
 	ip->i_blocks = vip->vii_blocks;
 	ip->i_generation = vip->vii_gen;
 
-	ip->u.generic_ip = (void *)vip;
+	ip->i_private = vip;
 	
 }
 
@@ -338,5 +338,5 @@ vxfs_read_inode(struct inode *ip)
 void
 vxfs_clear_inode(struct inode *ip)
 {
-	kmem_cache_free(vxfs_inode_cachep, ip->u.generic_ip);
+	kmem_cache_free(vxfs_inode_cachep, ip->i_private);
 }
diff --git a/fs/fuse/control.c b/fs/fuse/control.c
index 46fe60b2da23..79ec1f23d4d2 100644
--- a/fs/fuse/control.c
+++ b/fs/fuse/control.c
@@ -23,7 +23,7 @@ static struct fuse_conn *fuse_ctl_file_conn_get(struct file *file)
 {
 	struct fuse_conn *fc;
 	mutex_lock(&fuse_mutex);
-	fc = file->f_dentry->d_inode->u.generic_ip;
+	fc = file->f_dentry->d_inode->i_private;
 	if (fc)
 		fc = fuse_conn_get(fc);
 	mutex_unlock(&fuse_mutex);
@@ -98,7 +98,7 @@ static struct dentry *fuse_ctl_add_dentry(struct dentry *parent,
 		inode->i_op = iop;
 	inode->i_fop = fop;
 	inode->i_nlink = nlink;
-	inode->u.generic_ip = fc;
+	inode->i_private = fc;
 	d_add(dentry, inode);
 	return dentry;
 }
@@ -150,7 +150,7 @@ void fuse_ctl_remove_conn(struct fuse_conn *fc)
 
 	for (i = fc->ctl_ndents - 1; i >= 0; i--) {
 		struct dentry *dentry = fc->ctl_dentry[i];
-		dentry->d_inode->u.generic_ip = NULL;
+		dentry->d_inode->i_private = NULL;
 		d_drop(dentry);
 		dput(dentry);
 	}
diff --git a/fs/inode.c b/fs/inode.c
index 0bf9f0444a96..77e254792025 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -163,7 +163,7 @@ static struct inode *alloc_inode(struct super_block *sb)
 				bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info;
 			mapping->backing_dev_info = bdi;
 		}
-		memset(&inode->u, 0, sizeof(inode->u));
+		inode->i_private = 0;
 		inode->i_mapping = mapping;
 	}
 	return inode;
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index b59553d28d13..7358ef87f16b 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -369,7 +369,7 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode,
 
 	f = jffs_find_file(c, raw_inode->ino);
 
-	inode->u.generic_ip = (void *)f;
+	inode->i_private = (void *)f;
 	insert_inode_hash(inode);
 
 	return inode;
@@ -442,7 +442,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	});
 
 	result = -ENOTDIR;
-	if (!(old_dir_f = (struct jffs_file *)old_dir->u.generic_ip)) {
+	if (!(old_dir_f = old_dir->i_private)) {
 		D(printk("jffs_rename(): Old dir invalid.\n"));
 		goto jffs_rename_end;
 	}
@@ -456,7 +456,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
 
 	/* Find the new directory.  */
 	result = -ENOTDIR;
-	if (!(new_dir_f = (struct jffs_file *)new_dir->u.generic_ip)) {
+	if (!(new_dir_f = new_dir->i_private)) {
 		D(printk("jffs_rename(): New dir invalid.\n"));
 		goto jffs_rename_end;
 	}
@@ -593,7 +593,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		}
 		else {
 			ddino = ((struct jffs_file *)
-				 inode->u.generic_ip)->pino;
+				 inode->i_private)->pino;
 		}
 		D3(printk("jffs_readdir(): \"..\" %u\n", ddino));
 		if (filldir(dirent, "..", 2, filp->f_pos, ddino, DT_DIR) < 0) {
@@ -604,7 +604,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
 		}
 		filp->f_pos++;
 	}
-	f = ((struct jffs_file *)inode->u.generic_ip)->children;
+	f = ((struct jffs_file *)inode->i_private)->children;
 
 	j = 2;
 	while(f && (f->deleted || j++ < filp->f_pos )) {
@@ -668,7 +668,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 	}
 
 	r = -EACCES;
-	if (!(d = (struct jffs_file *)dir->u.generic_ip)) {
+	if (!(d = (struct jffs_file *)dir->i_private)) {
 		D(printk("jffs_lookup(): No such inode! (%lu)\n",
 			 dir->i_ino));
 		goto jffs_lookup_end;
@@ -739,7 +739,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
 	unsigned long read_len;
 	int result;
 	struct inode *inode = (struct inode*)page->mapping->host;
-	struct jffs_file *f = (struct jffs_file *)inode->u.generic_ip;
+	struct jffs_file *f = (struct jffs_file *)inode->i_private;
 	struct jffs_control *c = (struct jffs_control *)inode->i_sb->s_fs_info;
 	int r;
 	loff_t offset;
@@ -828,7 +828,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 	});
 
 	lock_kernel();
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_mkdir(): No reference to a "
@@ -972,7 +972,7 @@ jffs_remove(struct inode *dir, struct dentry *dentry, int type)
 		kfree(_name);
 	});
 
-	dir_f = (struct jffs_file *) dir->u.generic_ip;
+	dir_f = dir->i_private;
 	c = dir_f->c;
 
 	result = -ENOENT;
@@ -1082,7 +1082,7 @@ jffs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
 	if (!old_valid_dev(rdev))
 		return -EINVAL;
 	lock_kernel();
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	c = dir_f->c;
 
 	D3(printk (KERN_NOTICE "mknod(): down biglock\n"));
@@ -1186,7 +1186,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
 		kfree(_symname);
 	});
 
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_symlink(): No reference to a "
 		       "jffs_file struct in inode.\n");
@@ -1289,7 +1289,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
 		kfree(s);
 	});
 
-	dir_f = (struct jffs_file *)dir->u.generic_ip;
+	dir_f = dir->i_private;
 	ASSERT(if (!dir_f) {
 		printk(KERN_ERR "jffs_create(): No reference to a "
 		       "jffs_file struct in inode.\n");
@@ -1403,9 +1403,9 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
 		goto out_isem;
 	}
 
-	if (!(f = (struct jffs_file *)inode->u.generic_ip)) {
-		D(printk("jffs_file_write(): inode->u.generic_ip = 0x%p\n",
-				inode->u.generic_ip));
+	if (!(f = inode->i_private)) {
+		D(printk("jffs_file_write(): inode->i_private = 0x%p\n",
+				inode->i_private));
 		goto out_isem;
 	}
 
@@ -1693,7 +1693,7 @@ jffs_read_inode(struct inode *inode)
 		mutex_unlock(&c->fmc->biglock);
 		return;
 	}
-	inode->u.generic_ip = (void *)f;
+	inode->i_private = f;
 	inode->i_mode = f->mode;
 	inode->i_nlink = f->nlink;
 	inode->i_uid = f->uid;
@@ -1748,7 +1748,7 @@ jffs_delete_inode(struct inode *inode)
 	lock_kernel();
 	inode->i_size = 0;
 	inode->i_blocks = 0;
-	inode->u.generic_ip = NULL;
+	inode->i_private = NULL;
 	clear_inode(inode);
 	if (inode->i_nlink == 0) {
 		c = (struct jffs_control *) inode->i_sb->s_fs_info;
diff --git a/fs/libfs.c b/fs/libfs.c
index ac02ea602c3d..2751793beeaa 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -547,7 +547,7 @@ int simple_attr_open(struct inode *inode, struct file *file,
 
 	attr->get = get;
 	attr->set = set;
-	attr->data = inode->u.generic_ip;
+	attr->data = inode->i_private;
 	attr->fmt = fmt;
 	mutex_init(&attr->mutex);
 
diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c
index de887063dcfc..8801e41afe80 100644
--- a/fs/ocfs2/dlmglue.c
+++ b/fs/ocfs2/dlmglue.c
@@ -2052,7 +2052,7 @@ static int ocfs2_dlm_debug_open(struct inode *inode, struct file *file)
 		mlog_errno(ret);
 		goto out;
 	}
-	osb = (struct ocfs2_super *) inode->u.generic_ip;
+	osb = inode->i_private;
 	ocfs2_get_dlm_debug(osb->osb_dlm_debug);
 	priv->p_dlm_debug = osb->osb_dlm_debug;
 	INIT_LIST_HEAD(&priv->p_iter_res.l_debug_list);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1d3e601ece73..4f77ec9c3353 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -554,9 +554,7 @@ struct inode {
 
 	atomic_t		i_writecount;
 	void			*i_security;
-	union {
-		void		*generic_ip;
-	} u;
+	void			*i_private; /* fs or device private pointer */
 #ifdef __NEED_I_SIZE_ORDERED
 	seqcount_t		i_size_seqcount;
 #endif
diff --git a/kernel/relay.c b/kernel/relay.c
index 33345e73485c..85786ff2a4f9 100644
--- a/kernel/relay.c
+++ b/kernel/relay.c
@@ -669,7 +669,7 @@ EXPORT_SYMBOL_GPL(relay_flush);
  */
 static int relay_file_open(struct inode *inode, struct file *filp)
 {
-	struct rchan_buf *buf = inode->u.generic_ip;
+	struct rchan_buf *buf = inode->i_private;
 	kref_get(&buf->kref);
 	filp->private_data = buf;
 
diff --git a/security/inode.c b/security/inode.c
index 47eb63480dac..176aacea8ca4 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -44,8 +44,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf,
 
 static int default_open(struct inode *inode, struct file *file)
 {
-	if (inode->u.generic_ip)
-		file->private_data = inode->u.generic_ip;
+	if (inode->i_private)
+		file->private_data = inode->i_private;
 
 	return 0;
 }
@@ -194,7 +194,7 @@ static int create_by_name(const char *name, mode_t mode,
  *          directory dentry if set.  If this paramater is NULL, then the
  *          file will be created in the root of the securityfs filesystem.
  * @data: a pointer to something that the caller will want to get to later
- *        on.  The inode.u.generic_ip pointer will point to this value on
+ *        on.  The inode.i_private pointer will point to this value on
  *        the open() call.
  * @fops: a pointer to a struct file_operations that should be used for
  *        this file.
@@ -240,7 +240,7 @@ struct dentry *securityfs_create_file(const char *name, mode_t mode,
 		if (fops)
 			dentry->d_inode->i_fop = fops;
 		if (data)
-			dentry->d_inode->u.generic_ip = data;
+			dentry->d_inode->i_private = data;
 	}
 exit:
 	return dentry;
-- 
cgit v1.2.1


From 4c1541680f8d189d21dd07b053bc12996574646e Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:47 -0700
Subject: [PATCH] inode-diet: Move i_pipe into a union

Move the i_pipe pointer into a union that will be shared with i_bdev and
i_cdev.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/fs.h | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4f77ec9c3353..ca695fc8d69c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -528,9 +528,10 @@ struct inode {
 #ifdef CONFIG_QUOTA
 	struct dquot		*i_dquot[MAXQUOTAS];
 #endif
-	/* These three should probably be a union */
 	struct list_head	i_devices;
-	struct pipe_inode_info	*i_pipe;
+	union {
+		struct pipe_inode_info	*i_pipe;
+	};
 	struct block_device	*i_bdev;
 	struct cdev		*i_cdev;
 	int			i_cindex;
-- 
cgit v1.2.1


From eaf796e7ef6014f208c409b2b14fddcfaafe7e3a Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:48 -0700
Subject: [PATCH] inode-diet: Move i_bdev into a union

Move the i_bdev pointer in struct inode into a union.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/inode.c         | 2 +-
 include/linux/fs.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/inode.c b/fs/inode.c
index 77e254792025..41b462232f7f 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -254,7 +254,7 @@ void clear_inode(struct inode *inode)
 	DQUOT_DROP(inode);
 	if (inode->i_sb && inode->i_sb->s_op->clear_inode)
 		inode->i_sb->s_op->clear_inode(inode);
-	if (inode->i_bdev)
+	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
 		bd_forget(inode);
 	if (inode->i_cdev)
 		cd_forget(inode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index ca695fc8d69c..98ff684a5b1c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -531,8 +531,8 @@ struct inode {
 	struct list_head	i_devices;
 	union {
 		struct pipe_inode_info	*i_pipe;
+		struct block_device	*i_bdev;
 	};
-	struct block_device	*i_bdev;
 	struct cdev		*i_cdev;
 	int			i_cindex;
 
-- 
cgit v1.2.1


From 577c4eb09d1034d0739e3135fd2cff50588024be Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:49 -0700
Subject: [PATCH] inode-diet: Move i_cdev into a union

Move the i_cdev pointer in struct inode into a union.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/file_table.c    | 2 +-
 fs/inode.c         | 2 +-
 include/linux/fs.h | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/fs/file_table.c b/fs/file_table.c
index 0131ba06e1ee..bc35a40417d7 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -169,7 +169,7 @@ void fastcall __fput(struct file *file)
 	if (file->f_op && file->f_op->release)
 		file->f_op->release(inode, file);
 	security_file_free(file);
-	if (unlikely(inode->i_cdev != NULL))
+	if (unlikely(S_ISCHR(inode->i_mode) && inode->i_cdev != NULL))
 		cdev_put(inode->i_cdev);
 	fops_put(file->f_op);
 	if (file->f_mode & FMODE_WRITE)
diff --git a/fs/inode.c b/fs/inode.c
index 41b462232f7f..f5c04dd9ae8a 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -256,7 +256,7 @@ void clear_inode(struct inode *inode)
 		inode->i_sb->s_op->clear_inode(inode);
 	if (S_ISBLK(inode->i_mode) && inode->i_bdev)
 		bd_forget(inode);
-	if (inode->i_cdev)
+	if (S_ISCHR(inode->i_mode) && inode->i_cdev)
 		cd_forget(inode);
 	inode->i_state = I_CLEAR;
 }
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 98ff684a5b1c..192e69bb55b5 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -532,8 +532,8 @@ struct inode {
 	union {
 		struct pipe_inode_info	*i_pipe;
 		struct block_device	*i_bdev;
+		struct cdev		*i_cdev;
 	};
-	struct cdev		*i_cdev;
 	int			i_cindex;
 
 	__u32			i_generation;
-- 
cgit v1.2.1


From ba52de123d454b57369f291348266d86f4b35070 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Wed, 27 Sep 2006 01:50:49 -0700
Subject: [PATCH] inode-diet: Eliminate i_blksize from the inode structure

This eliminates the i_blksize field from struct inode.  Filesystems that want
to provide a per-inode st_blksize can do so by providing their own getattr
routine instead of using the generic_fillattr() function.

Note that some filesystems were providing pretty much random (and incorrect)
values for i_blksize.

[bunk@stusta.de: cleanup]
[akpm@osdl.org: generic_fillattr() fix]
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/cell/spufs/inode.c |  1 -
 arch/s390/hypfs/inode.c                   |  1 -
 drivers/block/loop.c                      |  7 +++++--
 drivers/infiniband/hw/ipath/ipath_fs.c    |  1 -
 drivers/isdn/capi/capifs.c                |  2 --
 drivers/misc/ibmasm/ibmasmfs.c            |  1 -
 drivers/oprofile/oprofilefs.c             |  1 -
 drivers/usb/core/inode.c                  |  1 -
 drivers/usb/gadget/inode.c                |  1 -
 fs/9p/vfs_inode.c                         |  4 +---
 fs/adfs/inode.c                           |  1 -
 fs/afs/inode.c                            |  1 -
 fs/autofs/inode.c                         |  1 -
 fs/autofs4/inode.c                        |  1 -
 fs/befs/linuxvfs.c                        |  1 -
 fs/bfs/dir.c                              |  2 +-
 fs/bfs/inode.c                            |  1 -
 fs/binfmt_misc.c                          |  1 -
 fs/cifs/cifsfs.c                          |  1 -
 fs/cifs/readdir.c                         |  5 ++---
 fs/coda/coda_linux.c                      |  2 --
 fs/configfs/inode.c                       |  1 -
 fs/cramfs/inode.c                         |  1 -
 fs/debugfs/inode.c                        |  1 -
 fs/devpts/inode.c                         |  2 --
 fs/eventpoll.c                            |  1 -
 fs/ext2/ialloc.c                          |  1 -
 fs/ext2/inode.c                           |  1 -
 fs/ext3/ialloc.c                          |  1 -
 fs/ext3/inode.c                           |  3 ---
 fs/fat/inode.c                            |  3 ---
 fs/freevxfs/vxfs_inode.c                  |  1 -
 fs/fuse/inode.c                           |  1 -
 fs/hfs/inode.c                            |  2 --
 fs/hfsplus/inode.c                        |  2 --
 fs/hostfs/hostfs_kern.c                   |  1 -
 fs/hpfs/inode.c                           |  1 -
 fs/hppfs/hppfs_kern.c                     |  1 -
 fs/hugetlbfs/inode.c                      |  1 -
 fs/isofs/inode.c                          |  3 +--
 fs/jffs/inode-v23.c                       |  2 --
 fs/jffs2/fs.c                             |  2 --
 fs/jfs/jfs_extent.c                       |  2 +-
 fs/jfs/jfs_imap.c                         |  1 -
 fs/jfs/jfs_inode.c                        |  1 -
 fs/jfs/jfs_metapage.c                     |  2 +-
 fs/libfs.c                                |  2 --
 fs/minix/bitmap.c                         |  2 +-
 fs/minix/inode.c                          |  4 ++--
 fs/ncpfs/inode.c                          |  1 -
 fs/nfs/inode.c                            |  4 ----
 fs/ntfs/inode.c                           |  4 ----
 fs/ntfs/mft.c                             |  5 -----
 fs/ocfs2/dlm/dlmfs.c                      |  2 --
 fs/ocfs2/inode.c                          |  4 ----
 fs/pipe.c                                 |  1 -
 fs/qnx4/inode.c                           |  1 -
 fs/ramfs/inode.c                          |  1 -
 fs/reiserfs/inode.c                       |  4 ----
 fs/smbfs/inode.c                          |  2 --
 fs/smbfs/proc.c                           |  1 -
 fs/stat.c                                 |  3 ++-
 fs/sysfs/inode.c                          |  1 -
 fs/sysv/ialloc.c                          |  2 +-
 fs/sysv/inode.c                           |  2 +-
 fs/udf/ialloc.c                           |  1 -
 fs/udf/inode.c                            |  2 --
 fs/ufs/ialloc.c                           |  1 -
 fs/ufs/inode.c                            |  1 -
 fs/xfs/linux-2.6/xfs_super.c              |  1 -
 fs/xfs/linux-2.6/xfs_vnode.c              |  1 -
 include/linux/fs.h                        |  1 -
 include/linux/nfsd/nfsfh.h                | 10 ++--------
 include/linux/smb.h                       |  1 -
 ipc/mqueue.c                              |  1 -
 kernel/cpuset.c                           |  1 -
 mm/shmem.c                                |  1 -
 net/sunrpc/rpc_pipe.c                     |  1 -
 security/inode.c                          |  1 -
 security/selinux/selinuxfs.c              |  1 -
 80 files changed, 21 insertions(+), 125 deletions(-)

diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index b837f12a84ae..3950ddccb2c8 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -82,7 +82,6 @@ spufs_new_inode(struct super_block *sb, int mode)
 	inode->i_mode = mode;
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 out:
diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c
index 8735024d235b..813fc21358f9 100644
--- a/arch/s390/hypfs/inode.c
+++ b/arch/s390/hypfs/inode.c
@@ -91,7 +91,6 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode)
 		ret->i_mode = mode;
 		ret->i_uid = hypfs_info->uid;
 		ret->i_gid = hypfs_info->gid;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 		if (mode & S_IFDIR)
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 7b3b94ddddcc..c774121684d7 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -662,7 +662,8 @@ static void do_loop_switch(struct loop_device *lo, struct switch_request *p)
 
 	mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask);
 	lo->lo_backing_file = file;
-	lo->lo_blocksize = mapping->host->i_blksize;
+	lo->lo_blocksize = S_ISBLK(mapping->host->i_mode) ?
+		mapping->host->i_bdev->bd_block_size : PAGE_SIZE;
 	lo->old_gfp_mask = mapping_gfp_mask(mapping);
 	mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
 	complete(&p->wait);
@@ -794,7 +795,9 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file,
 		if (!(lo_flags & LO_FLAGS_USE_AOPS) && !file->f_op->write)
 			lo_flags |= LO_FLAGS_READ_ONLY;
 
-		lo_blocksize = inode->i_blksize;
+		lo_blocksize = S_ISBLK(inode->i_mode) ?
+			inode->i_bdev->bd_block_size : PAGE_SIZE;
+
 		error = 0;
 	} else {
 		goto out_putf;
diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c
index 055cdd089b28..c8a8af0fe471 100644
--- a/drivers/infiniband/hw/ipath/ipath_fs.c
+++ b/drivers/infiniband/hw/ipath/ipath_fs.c
@@ -61,7 +61,6 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry,
 	inode->i_mode = mode;
 	inode->i_uid = 0;
 	inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	inode->i_private = data;
diff --git a/drivers/isdn/capi/capifs.c b/drivers/isdn/capi/capifs.c
index 9ea6bd0ddc35..2dd1b57b0ba4 100644
--- a/drivers/isdn/capi/capifs.c
+++ b/drivers/isdn/capi/capifs.c
@@ -104,7 +104,6 @@ capifs_fill_super(struct super_block *s, void *data, int silent)
 	inode->i_ino = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 	inode->i_uid = inode->i_gid = 0;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
 	inode->i_op = &simple_dir_inode_operations;
@@ -149,7 +148,6 @@ void capifs_new_ncci(unsigned int number, dev_t device)
 	if (!inode)
 		return;
 	inode->i_ino = number+2;
-	inode->i_blksize = 1024;
 	inode->i_uid = config.setuid ? config.uid : current->fsuid;
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c
index 0e909b617226..b99dc507de2e 100644
--- a/drivers/misc/ibmasm/ibmasmfs.c
+++ b/drivers/misc/ibmasm/ibmasmfs.c
@@ -147,7 +147,6 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode)
 	if (ret) {
 		ret->i_mode = mode;
 		ret->i_uid = ret->i_gid = 0;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 	}
diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c
index deb37354785b..5756401fb15b 100644
--- a/drivers/oprofile/oprofilefs.c
+++ b/drivers/oprofile/oprofilefs.c
@@ -31,7 +31,6 @@ static struct inode * oprofilefs_get_inode(struct super_block * sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	}
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 482f253085e5..58b4b1012120 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -249,7 +249,6 @@ static struct inode *usbfs_get_inode (struct super_block *sb, int mode, dev_t de
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index ffaa8c1afad8..2a7162d89799 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1966,7 +1966,6 @@ gadgetfs_make_inode (struct super_block *sb,
 		inode->i_mode = mode;
 		inode->i_uid = default_uid;
 		inode->i_gid = default_gid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime
 				= CURRENT_TIME;
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index eae50c9d6dc4..7a7ec2d1d2f4 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -204,7 +204,6 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = sb->s_blocksize;
 		inode->i_blocks = 0;
 		inode->i_rdev = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
@@ -950,9 +949,8 @@ v9fs_stat2inode(struct v9fs_stat *stat, struct inode *inode,
 
 	inode->i_size = stat->length;
 
-	inode->i_blksize = sb->s_blocksize;
 	inode->i_blocks =
-	    (inode->i_size + inode->i_blksize - 1) >> sb->s_blocksize_bits;
+	    (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits;
 }
 
 /**
diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c
index 534f3eecc985..7e7a04be1278 100644
--- a/fs/adfs/inode.c
+++ b/fs/adfs/inode.c
@@ -269,7 +269,6 @@ adfs_iget(struct super_block *sb, struct object_info *obj)
 	inode->i_ino	 = obj->file_id;
 	inode->i_size	 = obj->size;
 	inode->i_nlink	 = 2;
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks	 = (inode->i_size + sb->s_blocksize - 1) >>
 			    sb->s_blocksize_bits;
 
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index 4ebb30a50ed5..6f37754906c2 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -72,7 +72,6 @@ static int afs_inode_map_status(struct afs_vnode *vnode)
 	inode->i_ctime.tv_sec	= vnode->status.mtime_server;
 	inode->i_ctime.tv_nsec	= 0;
 	inode->i_atime		= inode->i_mtime = inode->i_ctime;
-	inode->i_blksize	= PAGE_CACHE_SIZE;
 	inode->i_blocks		= 0;
 	inode->i_version	= vnode->fid.unique;
 	inode->i_mapping->a_ops	= &afs_fs_aops;
diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c
index d39d2acd9b38..2c9759baad61 100644
--- a/fs/autofs/inode.c
+++ b/fs/autofs/inode.c
@@ -216,7 +216,6 @@ static void autofs_read_inode(struct inode *inode)
 	inode->i_nlink = 2;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 
 	if ( ino == AUTOFS_ROOT_INO ) {
 		inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 11a6a9ae51b7..800ce876caec 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -447,7 +447,6 @@ struct inode *autofs4_get_inode(struct super_block *sb,
 		inode->i_uid = 0;
 		inode->i_gid = 0;
 	}
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index f6676fbe9484..57020c7a7e65 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -365,7 +365,6 @@ befs_read_inode(struct inode *inode)
 	inode->i_mtime.tv_nsec = 0;   /* lower 16 bits are not a time */	
 	inode->i_ctime = inode->i_mtime;
 	inode->i_atime = inode->i_mtime;
-	inode->i_blksize = befs_sb->block_size;
 
 	befs_ino->i_inode_num = fsrun_to_cpu(sb, raw_inode->inode_num);
 	befs_ino->i_parent = fsrun_to_cpu(sb, raw_inode->parent);
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index 26fad9621738..dcf04cb13283 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -102,7 +102,7 @@ static int bfs_create(struct inode * dir, struct dentry * dentry, int mode,
 	inode->i_uid = current->fsuid;
 	inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	inode->i_op = &bfs_file_inops;
 	inode->i_fop = &bfs_file_operations;
 	inode->i_mapping->a_ops = &bfs_aops;
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 8fc2e8e49dbe..ed27ffb3459e 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -76,7 +76,6 @@ static void bfs_read_inode(struct inode * inode)
 	inode->i_size = BFS_FILESIZE(di);
 	inode->i_blocks = BFS_FILEBLOCKS(di);
         if (inode->i_size || inode->i_blocks) dprintf("Registered inode with %lld size, %ld blocks\n", inode->i_size, inode->i_blocks);
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_atime.tv_sec =  le32_to_cpu(di->i_atime);
 	inode->i_mtime.tv_sec =  le32_to_cpu(di->i_mtime);
 	inode->i_ctime.tv_sec =  le32_to_cpu(di->i_ctime);
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 6759b9839ce8..66ba137f8661 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -507,7 +507,6 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime =
 			current_fs_time(inode->i_sb);
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 4197a5043f13..22bcf4d7e7ae 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -253,7 +253,6 @@ cifs_alloc_inode(struct super_block *sb)
 	file data or metadata */
 	cifs_inode->clientCanCacheRead = FALSE;
 	cifs_inode->clientCanCacheAll = FALSE;
-	cifs_inode->vfs_inode.i_blksize = CIFS_MAX_MSGSIZE;
 	cifs_inode->vfs_inode.i_blkbits = 14;  /* 2**14 = CIFS_MAX_MSGSIZE */
 	cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME;
 	INIT_LIST_HEAD(&cifs_inode->openFileList);
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 9aeb58a7d369..b27b34537bf2 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -216,10 +216,9 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type,
 
 	if (allocation_size < end_of_file)
 		cFYI(1, ("May be sparse file, allocation less than file size"));
-	cFYI(1, ("File Size %ld and blocks %llu and blocksize %ld",
+	cFYI(1, ("File Size %ld and blocks %llu",
 		(unsigned long)tmp_inode->i_size,
-		(unsigned long long)tmp_inode->i_blocks,
-		tmp_inode->i_blksize));
+		(unsigned long long)tmp_inode->i_blocks));
 	if (S_ISREG(tmp_inode->i_mode)) {
 		cFYI(1, ("File inode"));
 		tmp_inode->i_op = &cifs_file_inode_ops;
diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c
index 5597080cb811..95a54253c047 100644
--- a/fs/coda/coda_linux.c
+++ b/fs/coda/coda_linux.c
@@ -110,8 +110,6 @@ void coda_vattr_to_iattr(struct inode *inode, struct coda_vattr *attr)
 	        inode->i_nlink = attr->va_nlink;
 	if (attr->va_size != -1)
 	        inode->i_size = attr->va_size;
-	if (attr->va_blocksize != -1)
-		inode->i_blksize = attr->va_blocksize;
 	if (attr->va_size != -1)
 		inode->i_blocks = (attr->va_size + 511) >> 9;
 	if (attr->va_atime.tv_sec != -1) 
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 5047e6a7581e..fb18917954a9 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -135,7 +135,6 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd)
 {
 	struct inode * inode = new_inode(configfs_sb);
 	if (inode) {
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &configfs_aops;
 		inode->i_mapping->backing_dev_info = &configfs_backing_dev_info;
diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index d09b6777c41a..ad96b6990715 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -73,7 +73,6 @@ static int cramfs_iget5_set(struct inode *inode, void *opaque)
 	inode->i_uid = cramfs_inode->uid;
 	inode->i_size = cramfs_inode->size;
 	inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_gid = cramfs_inode->gid;
 	/* Struct copy intentional */
 	inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime;
diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c
index 717f4821ed02..269e649e6dc6 100644
--- a/fs/debugfs/inode.c
+++ b/fs/debugfs/inode.c
@@ -40,7 +40,6 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 5bf06a10dddf..5f7b5a6025bf 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -113,7 +113,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent)
 	inode->i_ino = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	inode->i_blocks = 0;
-	inode->i_blksize = 1024;
 	inode->i_uid = inode->i_gid = 0;
 	inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR;
 	inode->i_op = &simple_dir_inode_operations;
@@ -172,7 +171,6 @@ int devpts_pty_new(struct tty_struct *tty)
 		return -ENOMEM;
 
 	inode->i_ino = number+2;
-	inode->i_blksize = 1024;
 	inode->i_uid = config.setuid ? config.uid : current->fsuid;
 	inode->i_gid = config.setgid ? config.gid : current->fsgid;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 3a3567433b92..8d544334bcd2 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1590,7 +1590,6 @@ static struct inode *ep_eventpoll_inode(void)
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->i_blksize = PAGE_SIZE;
 	return inode;
 
 eexit_1:
diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c
index 695f69ccf908..2cb545bf0f3c 100644
--- a/fs/ext2/ialloc.c
+++ b/fs/ext2/ialloc.c
@@ -574,7 +574,6 @@ got:
 	inode->i_mode = mode;
 
 	inode->i_ino = ino;
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	memset(ei->i_data, 0, sizeof(ei->i_data));
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index fb4d3220eb8d..dd4e14c221e0 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -1094,7 +1094,6 @@ void ext2_read_inode (struct inode * inode)
 		brelse (bh);
 		goto bad_inode;
 	}
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 	ei->i_faddr = le32_to_cpu(raw_inode->i_faddr);
diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 1e4ded8aa3cd..e45dbd651736 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -559,7 +559,6 @@ got:
 
 	inode->i_ino = ino;
 	/* This is the optimal IO size (for stat), not the fs block size */
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index c9fc15f8b609..dcf4f1dd108b 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2632,9 +2632,6 @@ void ext3_read_inode(struct inode * inode)
 		 * recovery code: that's fine, we're about to complete
 		 * the process of deleting those. */
 	}
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size
-					 * (for stat), not the fs block
-					 * size */  
 	inode->i_blocks = le32_to_cpu(raw_inode->i_blocks);
 	ei->i_flags = le32_to_cpu(raw_inode->i_flags);
 #ifdef EXT3_FRAGMENTS
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index e1035a590664..ab96ae823753 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -370,8 +370,6 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 			inode->i_flags |= S_IMMUTABLE;
 	}
 	MSDOS_I(inode)->i_attrs = de->attr & ATTR_UNUSED;
-	/* this is as close to the truth as we can get ... */
-	inode->i_blksize = sbi->cluster_size;
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
 	inode->i_mtime.tv_sec =
@@ -1131,7 +1129,6 @@ static int fat_read_root(struct inode *inode)
 		MSDOS_I(inode)->i_start = 0;
 		inode->i_size = sbi->dir_entries * sizeof(struct msdos_dir_entry);
 	}
-	inode->i_blksize = sbi->cluster_size;
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
 	MSDOS_I(inode)->i_logstart = 0;
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index 32a82ed108e4..4786d51ad3bd 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -239,7 +239,6 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip)
 	ip->i_ctime.tv_nsec = 0;
 	ip->i_mtime.tv_nsec = 0;
 
-	ip->i_blksize = PAGE_SIZE;
 	ip->i_blocks = vip->vii_blocks;
 	ip->i_generation = vip->vii_gen;
 
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 7d25092262ae..cb7cadb0b790 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -118,7 +118,6 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr)
 	inode->i_uid     = attr->uid;
 	inode->i_gid     = attr->gid;
 	i_size_write(inode, attr->size);
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks  = attr->blocks;
 	inode->i_atime.tv_sec   = attr->atime;
 	inode->i_atime.tv_nsec  = attr->atimensec;
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 315cf44a90b2..d05641c35fc9 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -154,7 +154,6 @@ struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, int mode)
 	inode->i_gid = current->fsgid;
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blksize = HFS_SB(sb)->alloc_blksz;
 	HFS_I(inode)->flags = 0;
 	HFS_I(inode)->rsrc_inode = NULL;
 	HFS_I(inode)->fs_blocks = 0;
@@ -284,7 +283,6 @@ static int hfs_read_inode(struct inode *inode, void *data)
 	inode->i_uid = hsb->s_uid;
 	inode->i_gid = hsb->s_gid;
 	inode->i_nlink = 1;
-	inode->i_blksize = HFS_SB(inode->i_sb)->alloc_blksz;
 
 	if (idata->key)
 		HFS_I(inode)->cat_key = *idata->key;
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 924ecdef8091..0eb1a6092668 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -304,7 +304,6 @@ struct inode *hfsplus_new_inode(struct super_block *sb, int mode)
 	inode->i_gid = current->fsgid;
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blksize = HFSPLUS_SB(sb).alloc_blksz;
 	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
 	init_MUTEX(&HFSPLUS_I(inode).extents_lock);
 	atomic_set(&HFSPLUS_I(inode).opencnt, 0);
@@ -407,7 +406,6 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd)
 	type = hfs_bnode_read_u16(fd->bnode, fd->entryoffset);
 
 	HFSPLUS_I(inode).dev = 0;
-	inode->i_blksize = HFSPLUS_SB(inode->i_sb).alloc_blksz;
 	if (type == HFSPLUS_FOLDER) {
 		struct hfsplus_cat_folder *folder = &entry.folder;
 
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index b82e3d9c8790..322e876c35ed 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -156,7 +156,6 @@ static int read_name(struct inode *ino, char *name)
 	ino->i_mode = i_mode;
 	ino->i_nlink = i_nlink;
 	ino->i_size = i_size;
-	ino->i_blksize = i_blksize;
 	ino->i_blocks = i_blocks;
 	return(0);
 }
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index 56f2c338c4d9..bcf6ee36e065 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -17,7 +17,6 @@ void hpfs_init_inode(struct inode *i)
 	i->i_gid = hpfs_sb(sb)->sb_gid;
 	i->i_mode = hpfs_sb(sb)->sb_mode;
 	hpfs_inode->i_conv = hpfs_sb(sb)->sb_conv;
-	i->i_blksize = 512;
 	i->i_size = -1;
 	i->i_blocks = -1;
 	
diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c
index 3a9bdf58166f..dcb6d2e988b8 100644
--- a/fs/hppfs/hppfs_kern.c
+++ b/fs/hppfs/hppfs_kern.c
@@ -152,7 +152,6 @@ static void hppfs_read_inode(struct inode *ino)
 	ino->i_mode = proc_ino->i_mode;
 	ino->i_nlink = proc_ino->i_nlink;
 	ino->i_size = proc_ino->i_size;
-	ino->i_blksize = proc_ino->i_blksize;
 	ino->i_blocks = proc_ino->i_blocks;
 }
 
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index c3920c96dadf..e025a31b4c64 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -357,7 +357,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid,
 		inode->i_mode = mode;
 		inode->i_uid = uid;
 		inode->i_gid = gid;
-		inode->i_blksize = HPAGE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &hugetlbfs_aops;
 		inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info;
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 10e47897bac7..4527692f432b 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -1235,7 +1235,7 @@ static void isofs_read_inode(struct inode *inode)
 	}
 	inode->i_uid = sbi->s_uid;
 	inode->i_gid = sbi->s_gid;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 
 	ei->i_format_parm[0] = 0;
 	ei->i_format_parm[1] = 0;
@@ -1291,7 +1291,6 @@ static void isofs_read_inode(struct inode *inode)
 			      isonum_711 (de->ext_attr_length));
 
 	/* Set the number of blocks for stat() - should be done before RR */
-	inode->i_blksize = PAGE_CACHE_SIZE; /* For stat() only */
 	inode->i_blocks  = (inode->i_size + 511) >> 9;
 
 	/*
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index 7358ef87f16b..f5cf9c93e243 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -364,7 +364,6 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode,
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 
 	f = jffs_find_file(c, raw_inode->ino);
@@ -1706,7 +1705,6 @@ jffs_read_inode(struct inode *inode)
 	inode->i_mtime.tv_nsec = 
 	inode->i_ctime.tv_nsec = 0;
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 	if (S_ISREG(inode->i_mode)) {
 		inode->i_op = &jffs_file_inode_operations;
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index 4780f82825d6..72d9909d95ff 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -263,7 +263,6 @@ void jffs2_read_inode (struct inode *inode)
 
 	inode->i_nlink = f->inocache->nlink;
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = (inode->i_size + 511) >> 9;
 
 	switch (inode->i_mode & S_IFMT) {
@@ -449,7 +448,6 @@ struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_i
 	inode->i_atime = inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC;
 	ri->atime = ri->mtime = ri->ctime = cpu_to_je32(I_SEC(inode->i_mtime));
 
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_size = 0;
 
diff --git a/fs/jfs/jfs_extent.c b/fs/jfs/jfs_extent.c
index 4d52593a5fc6..4c74f0944f7e 100644
--- a/fs/jfs/jfs_extent.c
+++ b/fs/jfs/jfs_extent.c
@@ -468,7 +468,7 @@ int extRecord(struct inode *ip, xad_t * xp)
 int extFill(struct inode *ip, xad_t * xp)
 {
 	int rc, nbperpage = JFS_SBI(ip->i_sb)->nbperpage;
-	s64 blkno = offsetXAD(xp) >> ip->i_blksize;
+	s64 blkno = offsetXAD(xp) >> ip->i_blkbits;
 
 //      assert(ISSPARSE(ip));
 
diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
index ccbe60aff83d..369d7f39c040 100644
--- a/fs/jfs/jfs_imap.c
+++ b/fs/jfs/jfs_imap.c
@@ -3115,7 +3115,6 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip)
 	ip->i_mtime.tv_nsec = le32_to_cpu(dip->di_mtime.tv_nsec);
 	ip->i_ctime.tv_sec = le32_to_cpu(dip->di_ctime.tv_sec);
 	ip->i_ctime.tv_nsec = le32_to_cpu(dip->di_ctime.tv_nsec);
-	ip->i_blksize = ip->i_sb->s_blocksize;
 	ip->i_blocks = LBLK2PBLK(ip->i_sb, le64_to_cpu(dip->di_nblocks));
 	ip->i_generation = le32_to_cpu(dip->di_gen);
 
diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c
index 495df402916d..bffaca9ae3a2 100644
--- a/fs/jfs/jfs_inode.c
+++ b/fs/jfs/jfs_inode.c
@@ -115,7 +115,6 @@ struct inode *ialloc(struct inode *parent, umode_t mode)
 	}
 	jfs_inode->mode2 |= mode;
 
-	inode->i_blksize = sb->s_blocksize;
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	jfs_inode->otime = inode->i_ctime.tv_sec;
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index e1e0a6e6ebdf..f5afc129d6b1 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -257,7 +257,7 @@ static sector_t metapage_get_blocks(struct inode *inode, sector_t lblock,
 	int rc = 0;
 	int xflag;
 	s64 xaddr;
-	sector_t file_blocks = (inode->i_size + inode->i_blksize - 1) >>
+	sector_t file_blocks = (inode->i_size + inode->i_sb->s_blocksize - 1) >>
 			       inode->i_blkbits;
 
 	if (lblock >= file_blocks)
diff --git a/fs/libfs.c b/fs/libfs.c
index 2751793beeaa..8db5afb7b0a7 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -383,7 +383,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files
 		return -ENOMEM;
 	inode->i_mode = S_IFDIR | 0755;
 	inode->i_uid = inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	inode->i_op = &simple_dir_inode_operations;
@@ -405,7 +404,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files
 			goto out;
 		inode->i_mode = S_IFREG | files->mode;
 		inode->i_uid = inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		inode->i_fop = files->ops;
diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c
index 4a6abc49418e..df6b1075b549 100644
--- a/fs/minix/bitmap.c
+++ b/fs/minix/bitmap.c
@@ -254,7 +254,7 @@ struct inode * minix_new_inode(const struct inode * dir, int * error)
 	inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid;
 	inode->i_ino = j;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	memset(&minix_i(inode)->u, 0, sizeof(minix_i(inode)->u));
 	insert_inode_hash(inode);
 	mark_inode_dirty(inode);
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 826b9d830650..c11a4b9fb863 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -396,7 +396,7 @@ static void V1_minix_read_inode(struct inode * inode)
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_ctime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	for (i = 0; i < 9; i++)
 		minix_inode->u.i1_data[i] = raw_inode->i_zone[i];
 	minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0]));
@@ -429,7 +429,7 @@ static void V2_minix_read_inode(struct inode * inode)
 	inode->i_mtime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_ctime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	for (i = 0; i < 10; i++)
 		minix_inode->u.i2_data[i] = raw_inode->i_zone[i];
 	minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0]));
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 8244710e97dd..42e3bef270c9 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -223,7 +223,6 @@ static void ncp_set_attr(struct inode *inode, struct ncp_entry_info *nwinfo)
 	inode->i_nlink = 1;
 	inode->i_uid = server->m.uid;
 	inode->i_gid = server->m.gid;
-	inode->i_blksize = NCP_BLOCK_SIZE;
 
 	ncp_update_dates(inode, &nwinfo->i);
 	ncp_update_inode(inode, nwinfo);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index 931f52a19579..bc9376ca86cd 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -277,10 +277,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr)
 			 * report the blocks in 512byte units
 			 */
 			inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
-			inode->i_blksize = inode->i_sb->s_blocksize;
 		} else {
 			inode->i_blocks = fattr->du.nfs2.blocks;
-			inode->i_blksize = fattr->du.nfs2.blocksize;
 		}
 		nfsi->attrtimeo = NFS_MINATTRTIMEO(inode);
 		nfsi->attrtimeo_timestamp = jiffies;
@@ -969,10 +967,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
 		 * report the blocks in 512byte units
 		 */
 		inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used);
-		inode->i_blksize = inode->i_sb->s_blocksize;
  	} else {
  		inode->i_blocks = fattr->du.nfs2.blocks;
- 		inode->i_blksize = fattr->du.nfs2.blocksize;
  	}
 
 	if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 &&
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 31852121b3f5..933dbd89c2a4 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -556,8 +556,6 @@ static int ntfs_read_locked_inode(struct inode *vi)
 
 	/* Setup the generic vfs inode parts now. */
 
-	/* This is the optimal IO size (for stat), not the fs block size. */
-	vi->i_blksize = PAGE_CACHE_SIZE;
 	/*
 	 * This is for checking whether an inode has changed w.r.t. a file so
 	 * that the file can be updated if necessary (compare with f_version).
@@ -1234,7 +1232,6 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi)
 	base_ni = NTFS_I(base_vi);
 
 	/* Just mirror the values from the base inode. */
-	vi->i_blksize	= base_vi->i_blksize;
 	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
@@ -1504,7 +1501,6 @@ static int ntfs_read_locked_index_inode(struct inode *base_vi, struct inode *vi)
 	ni	= NTFS_I(vi);
 	base_ni = NTFS_I(base_vi);
 	/* Just mirror the values from the base inode. */
-	vi->i_blksize	= base_vi->i_blksize;
 	vi->i_version	= base_vi->i_version;
 	vi->i_uid	= base_vi->i_uid;
 	vi->i_gid	= base_vi->i_gid;
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 578fb3d5e803..584260fd6848 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -2637,11 +2637,6 @@ mft_rec_already_initialized:
 			goto undo_mftbmp_alloc;
 		}
 		vi->i_ino = bit;
-		/*
-		 * This is the optimal IO size (for stat), not the fs block
-		 * size.
-		 */
-		vi->i_blksize = PAGE_CACHE_SIZE;
 		/*
 		 * This is for checking whether an inode has changed w.r.t. a
 		 * file so that the file can be updated if necessary (compare
diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c
index 0ff0898a0b9c..0368c6402182 100644
--- a/fs/ocfs2/dlm/dlmfs.c
+++ b/fs/ocfs2/dlm/dlmfs.c
@@ -335,7 +335,6 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
@@ -362,7 +361,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent,
 	inode->i_mode = mode;
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 69d3db569166..16e8e74dc966 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -269,7 +269,6 @@ int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe,
 	inode->i_mode = le16_to_cpu(fe->i_mode);
 	inode->i_uid = le32_to_cpu(fe->i_uid);
 	inode->i_gid = le32_to_cpu(fe->i_gid);
-	inode->i_blksize = (u32)osb->s_clustersize;
 
 	/* Fast symlinks will have i_size but no allocated clusters. */
 	if (S_ISLNK(inode->i_mode) && !fe->i_clusters)
@@ -1258,8 +1257,6 @@ leave:
 void ocfs2_refresh_inode(struct inode *inode,
 			 struct ocfs2_dinode *fe)
 {
-	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
-
 	spin_lock(&OCFS2_I(inode)->ip_lock);
 
 	OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters);
@@ -1270,7 +1267,6 @@ void ocfs2_refresh_inode(struct inode *inode,
 	inode->i_uid = le32_to_cpu(fe->i_uid);
 	inode->i_gid = le32_to_cpu(fe->i_gid);
 	inode->i_mode = le16_to_cpu(fe->i_mode);
-	inode->i_blksize = (u32) osb->s_clustersize;
 	if (S_ISLNK(inode->i_mode) && le32_to_cpu(fe->i_clusters) == 0)
 		inode->i_blocks = 0;
 	else
diff --git a/fs/pipe.c b/fs/pipe.c
index 20352573e025..f3b6f71e9d0b 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -879,7 +879,6 @@ static struct inode * get_pipe_inode(void)
 	inode->i_uid = current->fsuid;
 	inode->i_gid = current->fsgid;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	inode->i_blksize = PAGE_SIZE;
 
 	return inode;
 
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index fddbd61c68d0..5a41db2a218d 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -496,7 +496,6 @@ static void qnx4_read_inode(struct inode *inode)
 	inode->i_ctime.tv_sec   = le32_to_cpu(raw_inode->di_ctime);
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_blocks  = le32_to_cpu(raw_inode->di_first_xtnt.xtnt_size);
-	inode->i_blksize = QNX4_DIR_ENTRY_SIZE;
 
 	memcpy(qnx4_inode, raw_inode, QNX4_DIR_ENTRY_SIZE);
 	if (S_ISREG(inode->i_mode)) {
diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index b9677335cc8d..bc0e51662424 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -58,7 +58,6 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &ramfs_aops;
 		inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info;
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 52f1e2136546..8810fda0da46 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -17,8 +17,6 @@
 #include <linux/writeback.h>
 #include <linux/quotaops.h>
 
-extern int reiserfs_default_io_size;	/* default io size devuned in super.c */
-
 static int reiserfs_commit_write(struct file *f, struct page *page,
 				 unsigned from, unsigned to);
 static int reiserfs_prepare_write(struct file *f, struct page *page,
@@ -1122,7 +1120,6 @@ static void init_inode(struct inode *inode, struct path *path)
 	ih = PATH_PITEM_HEAD(path);
 
 	copy_key(INODE_PKEY(inode), &(ih->ih_key));
-	inode->i_blksize = reiserfs_default_io_size;
 
 	INIT_LIST_HEAD(&(REISERFS_I(inode)->i_prealloc_list));
 	REISERFS_I(inode)->i_flags = 0;
@@ -1877,7 +1874,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th,
 	}
 	// these do not go to on-disk stat data
 	inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid);
-	inode->i_blksize = reiserfs_default_io_size;
 
 	// store in in-core inode the key of stat data and version all
 	// object items will have (directory items will have old offset
diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c
index 92cf60aa6121..2c122ee83adb 100644
--- a/fs/smbfs/inode.c
+++ b/fs/smbfs/inode.c
@@ -166,7 +166,6 @@ smb_get_inode_attr(struct inode *inode, struct smb_fattr *fattr)
 	fattr->f_mtime	= inode->i_mtime;
 	fattr->f_ctime	= inode->i_ctime;
 	fattr->f_atime	= inode->i_atime;
-	fattr->f_blksize= inode->i_blksize;
 	fattr->f_blocks	= inode->i_blocks;
 
 	fattr->attr	= SMB_I(inode)->attr;
@@ -200,7 +199,6 @@ smb_set_inode_attr(struct inode *inode, struct smb_fattr *fattr)
 	inode->i_uid	= fattr->f_uid;
 	inode->i_gid	= fattr->f_gid;
 	inode->i_ctime	= fattr->f_ctime;
-	inode->i_blksize= fattr->f_blksize;
 	inode->i_blocks = fattr->f_blocks;
 	inode->i_size	= fattr->f_size;
 	inode->i_mtime	= fattr->f_mtime;
diff --git a/fs/smbfs/proc.c b/fs/smbfs/proc.c
index c3495059889d..40e174db9872 100644
--- a/fs/smbfs/proc.c
+++ b/fs/smbfs/proc.c
@@ -1826,7 +1826,6 @@ smb_init_dirent(struct smb_sb_info *server, struct smb_fattr *fattr)
 	fattr->f_nlink = 1;
 	fattr->f_uid = server->mnt->uid;
 	fattr->f_gid = server->mnt->gid;
-	fattr->f_blksize = SMB_ST_BLKSIZE;
 	fattr->f_unix = 0;
 }
 
diff --git a/fs/stat.c b/fs/stat.c
index 3a44dcf97da2..60a31d5e5966 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -14,6 +14,7 @@
 #include <linux/namei.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
+#include <linux/pagemap.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -32,7 +33,7 @@ void generic_fillattr(struct inode *inode, struct kstat *stat)
 	stat->ctime = inode->i_ctime;
 	stat->size = i_size_read(inode);
 	stat->blocks = inode->i_blocks;
-	stat->blksize = inode->i_blksize;
+	stat->blksize = (1 << inode->i_blkbits);
 }
 
 EXPORT_SYMBOL(generic_fillattr);
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index fd7cd5f843d2..e79e38d52c00 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -125,7 +125,6 @@ struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent * sd)
 {
 	struct inode * inode = new_inode(sysfs_sb);
 	if (inode) {
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &sysfs_aops;
 		inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info;
diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c
index 9b585d1081c0..115ab0d6f4bc 100644
--- a/fs/sysv/ialloc.c
+++ b/fs/sysv/ialloc.c
@@ -170,7 +170,7 @@ struct inode * sysv_new_inode(const struct inode * dir, mode_t mode)
 	inode->i_uid = current->fsuid;
 	inode->i_ino = fs16_to_cpu(sbi, ino);
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 	memset(SYSV_I(inode)->i_data, 0, sizeof(SYSV_I(inode)->i_data));
 	SYSV_I(inode)->i_dir_start_lookup = 0;
 	insert_inode_hash(inode);
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index 58b2d22142ba..d63c5e48b050 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -201,7 +201,7 @@ static void sysv_read_inode(struct inode *inode)
 	inode->i_ctime.tv_nsec = 0;
 	inode->i_atime.tv_nsec = 0;
 	inode->i_mtime.tv_nsec = 0;
-	inode->i_blocks = inode->i_blksize = 0;
+	inode->i_blocks = 0;
 
 	si = SYSV_I(inode);
 	for (block = 0; block < 10+1+1+1; block++)
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index d1d38238ce65..8206983f2ebf 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -121,7 +121,6 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
 	UDF_I_LOCATION(inode).logicalBlockNum = block;
 	UDF_I_LOCATION(inode).partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum;
 	inode->i_ino = udf_get_lb_pblock(sb, UDF_I_LOCATION(inode), 0);
-	inode->i_blksize = PAGE_SIZE;
 	inode->i_blocks = 0;
 	UDF_I_LENEATTR(inode) = 0;
 	UDF_I_LENALLOC(inode) = 0;
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 605f5111b6d8..b223b32db991 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -916,8 +916,6 @@ __udf_read_inode(struct inode *inode)
 	 *      i_nlink = 1
 	 *      i_op = NULL;
 	 */
-	inode->i_blksize = PAGE_SIZE;
-
 	bh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 0, &ident);
 
 	if (!bh)
diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c
index 9501dcd3b213..2ad1259c6eca 100644
--- a/fs/ufs/ialloc.c
+++ b/fs/ufs/ialloc.c
@@ -255,7 +255,6 @@ cg_found:
 		inode->i_gid = current->fsgid;
 
 	inode->i_ino = cg * uspi->s_ipg + bit;
-	inode->i_blksize = PAGE_SIZE;	/* This is the optimal IO size (for stat), not the fs block size */
 	inode->i_blocks = 0;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	ufsi->i_flags = UFS_I(dir)->i_flags;
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index 30c6e8a9446c..ee1eaa6f4ec2 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -741,7 +741,6 @@ void ufs_read_inode(struct inode * inode)
 		ufs1_read_inode(inode, ufs_inode + ufs_inotofsbo(inode->i_ino));
 	}
 
-	inode->i_blksize = PAGE_SIZE;/*This is the optimal IO size (for stat)*/
 	inode->i_version++;
 	ufsi->i_lastfrag =
 		(inode->i_size + uspi->s_fsize - 1) >> uspi->s_fshift;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 4754f342a5d3..9df9ed37d219 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -171,7 +171,6 @@ xfs_revalidate_inode(
 		break;
 	}
 
-	inode->i_blksize = xfs_preferred_iosize(mp);
 	inode->i_generation = ip->i_d.di_gen;
 	i_size_write(inode, ip->i_d.di_size);
 	inode->i_blocks =
diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c
index 6628d96b6fd6..553fa731ade5 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.c
+++ b/fs/xfs/linux-2.6/xfs_vnode.c
@@ -122,7 +122,6 @@ vn_revalidate_core(
 	inode->i_blocks	    = vap->va_nblocks;
 	inode->i_mtime	    = vap->va_mtime;
 	inode->i_ctime	    = vap->va_ctime;
-	inode->i_blksize    = vap->va_blocksize;
 	if (vap->va_xflags & XFS_XFLAG_IMMUTABLE)
 		inode->i_flags |= S_IMMUTABLE;
 	else
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 192e69bb55b5..8f74dfbb2edd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -512,7 +512,6 @@ struct inode {
 	struct timespec		i_mtime;
 	struct timespec		i_ctime;
 	unsigned int		i_blkbits;
-	unsigned long		i_blksize;
 	unsigned long		i_version;
 	blkcnt_t		i_blocks;
 	unsigned short          i_bytes;
diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h
index f9edcd2ff3c8..31a3cb617ce0 100644
--- a/include/linux/nfsd/nfsfh.h
+++ b/include/linux/nfsd/nfsfh.h
@@ -269,14 +269,8 @@ fill_post_wcc(struct svc_fh *fhp)
 	fhp->fh_post_uid	= inode->i_uid;
 	fhp->fh_post_gid	= inode->i_gid;
 	fhp->fh_post_size       = inode->i_size;
-	if (inode->i_blksize) {
-		fhp->fh_post_blksize    = inode->i_blksize;
-		fhp->fh_post_blocks     = inode->i_blocks;
-	} else {
-		fhp->fh_post_blksize    = BLOCK_SIZE;
-		/* how much do we care for accuracy with MinixFS? */
-		fhp->fh_post_blocks     = (inode->i_size+511) >> 9;
-	}
+	fhp->fh_post_blksize    = BLOCK_SIZE;
+	fhp->fh_post_blocks     = inode->i_blocks;
 	fhp->fh_post_rdev[0]    = htonl((u32)imajor(inode));
 	fhp->fh_post_rdev[1]    = htonl((u32)iminor(inode));
 	fhp->fh_post_atime      = inode->i_atime;
diff --git a/include/linux/smb.h b/include/linux/smb.h
index 6df3b1501559..f098dff93f6b 100644
--- a/include/linux/smb.h
+++ b/include/linux/smb.h
@@ -89,7 +89,6 @@ struct smb_fattr {
 	struct timespec	f_atime;
 	struct timespec f_mtime;
 	struct timespec f_ctime;
-	unsigned long	f_blksize;
 	unsigned long	f_blocks;
 	int		f_unix;
 };
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index f8381f0660b2..840f8a6fb85f 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -115,7 +115,6 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode,
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mtime = inode->i_ctime = inode->i_atime =
 				CURRENT_TIME;
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index cff41511269f..1b32c2c04c15 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -289,7 +289,6 @@ static struct inode *cpuset_new_inode(mode_t mode)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		inode->i_mapping->backing_dev_info = &cpuset_backing_dev_info;
diff --git a/mm/shmem.c b/mm/shmem.c
index 0a8e29cf87e0..eda907c3a86a 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1351,7 +1351,6 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = current->fsuid;
 		inode->i_gid = current->fsgid;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_mapping->a_ops = &shmem_aops;
 		inode->i_mapping->backing_dev_info = &shmem_backing_dev_info;
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index f65bea74734d..700c6e061a04 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -488,7 +488,6 @@ rpc_get_inode(struct super_block *sb, int mode)
 		return NULL;
 	inode->i_mode = mode;
 	inode->i_uid = inode->i_gid = 0;
-	inode->i_blksize = PAGE_CACHE_SIZE;
 	inode->i_blocks = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 	switch(mode & S_IFMT) {
diff --git a/security/inode.c b/security/inode.c
index 176aacea8ca4..49ee51529396 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -64,7 +64,6 @@ static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev)
 		inode->i_mode = mode;
 		inode->i_uid = 0;
 		inode->i_gid = 0;
-		inode->i_blksize = PAGE_CACHE_SIZE;
 		inode->i_blocks = 0;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
 		switch (mode & S_IFMT) {
diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c
index 00534c302ba2..bab7b386cb8d 100644
--- a/security/selinux/selinuxfs.c
+++ b/security/selinux/selinuxfs.c
@@ -771,7 +771,6 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode)
 	if (ret) {
 		ret->i_mode = mode;
 		ret->i_uid = ret->i_gid = 0;
-		ret->i_blksize = PAGE_CACHE_SIZE;
 		ret->i_blocks = 0;
 		ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
 	}
-- 
cgit v1.2.1


From 36b756f2b5762e6d5acba0c18e75cb5c11f11c1b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Wed, 27 Sep 2006 01:50:50 -0700
Subject: [PATCH] reiserfs: warn about the useless nolargeio option

Since the nolargeio option no longer has any effect, print a warning
instead of setting a write-only variable.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <mason@suse.com>
Cc: Hans Reiser <reiser@namesys.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/super.c | 21 ++-------------------
 1 file changed, 2 insertions(+), 19 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 1cd4d387f690..b40d4d64d598 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -723,12 +723,6 @@ static const arg_desc_t error_actions[] = {
 	{NULL, 0, 0},
 };
 
-int reiserfs_default_io_size = 128 * 1024;	/* Default recommended I/O size is 128k.
-						   There might be broken applications that are
-						   confused by this. Use nolargeio mount option
-						   to get usual i/o size = PAGE_SIZE.
-						 */
-
 /* proceed only one option from a list *cur - string containing of mount options
    opts - array of options which are accepted
    opt_arg - if option is found and requires an argument and if it is specifed
@@ -957,19 +951,8 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 		}
 
 		if (c == 'w') {
-			char *p = NULL;
-			int val = simple_strtoul(arg, &p, 0);
-
-			if (*p != '\0') {
-				reiserfs_warning(s,
-						 "reiserfs_parse_options: non-numeric value %s for nolargeio option",
-						 arg);
-				return 0;
-			}
-			if (val)
-				reiserfs_default_io_size = PAGE_SIZE;
-			else
-				reiserfs_default_io_size = 128 * 1024;
+			reiserfs_warning(s, "reiserfs: nolargeio option is no longer supported");
+			return 0;
 		}
 
 		if (c == 'j') {
-- 
cgit v1.2.1


From 9a3110bf4bb0466b43b898533bfd4952001bc38f Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 27 Sep 2006 01:50:51 -0700
Subject: [PATCH] x86 microcode: microcode driver cleanup.

Clean up microcode update driver and make it more readable.

[akpm@osdl.org: cleanups]
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Tigran Aivazian <tigran@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig            |   5 +
 arch/i386/kernel/microcode.c | 508 +++++++++++++++++++++----------------------
 arch/x86_64/Kconfig          |   5 +
 3 files changed, 260 insertions(+), 258 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 3fd9f1e8b093..0cbeb26ac3fb 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -416,6 +416,11 @@ config MICROCODE
 	  To compile this driver as a module, choose M here: the
 	  module will be called microcode.
 
+config MICROCODE_OLD_INTERFACE
+	bool
+	depends on MICROCODE
+	default y
+
 config X86_MSR
 	tristate "/dev/cpu/*/msr - Model-specific register support"
 	help
diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index 40b44cc0d14b..00bdc3dd693e 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -2,6 +2,7 @@
  *	Intel CPU Microcode Update Driver for Linux
  *
  *	Copyright (C) 2000-2004 Tigran Aivazian
+ *		      2006	Shaohua Li <shaohua.li@intel.com>
  *
  *	This driver allows to upgrade microcode on Intel processors
  *	belonging to IA-32 family - PentiumPro, Pentium II, 
@@ -91,9 +92,6 @@ MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver");
 MODULE_AUTHOR("Tigran Aivazian <tigran@veritas.com>");
 MODULE_LICENSE("GPL");
 
-static int verbose;
-module_param(verbose, int, 0644);
-
 #define MICROCODE_VERSION 	"1.14a"
 
 #define DEFAULT_UCODE_DATASIZE 	(2000) 	  /* 2000 bytes */
@@ -120,55 +118,40 @@ static DEFINE_SPINLOCK(microcode_update_lock);
 /* no concurrent ->write()s are allowed on /dev/cpu/microcode */
 static DEFINE_MUTEX(microcode_mutex);
 
-static void __user *user_buffer;	/* user area microcode data buffer */
-static unsigned int user_buffer_size;	/* it's size */
-
-typedef enum mc_error_code {
-	MC_SUCCESS 	= 0,
-	MC_IGNORED 	= 1,
-	MC_NOTFOUND 	= 2,
-	MC_MARKED 	= 3,
-	MC_ALLOCATED 	= 4,
-} mc_error_code_t;
-
 static struct ucode_cpu_info {
+	int valid;
 	unsigned int sig;
-	unsigned int pf, orig_pf;
+	unsigned int pf;
 	unsigned int rev;
-	unsigned int cksum;
-	mc_error_code_t err;
 	microcode_t *mc;
 } ucode_cpu_info[NR_CPUS];
-				
-static int microcode_open (struct inode *unused1, struct file *unused2)
-{
-	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
-}
 
-static void collect_cpu_info (void *unused)
+static void collect_cpu_info(int cpu_num)
 {
-	int cpu_num = smp_processor_id();
 	struct cpuinfo_x86 *c = cpu_data + cpu_num;
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
 	unsigned int val[2];
 
-	uci->sig = uci->pf = uci->rev = uci->cksum = 0;
-	uci->err = MC_NOTFOUND; 
+	/* We should bind the task to the CPU */
+	BUG_ON(raw_smp_processor_id() != cpu_num);
+	uci->pf = uci->rev = 0;
 	uci->mc = NULL;
+	uci->valid = 1;
 
 	if (c->x86_vendor != X86_VENDOR_INTEL || c->x86 < 6 ||
 	    	cpu_has(c, X86_FEATURE_IA64)) {
-		printk(KERN_ERR "microcode: CPU%d not a capable Intel processor\n", cpu_num);
+		printk(KERN_ERR "microcode: CPU%d not a capable Intel "
+			"processor\n", cpu_num);
+		uci->valid = 0;
 		return;
-	} else {
-		uci->sig = cpuid_eax(0x00000001);
+	}
 
-		if ((c->x86_model >= 5) || (c->x86 > 6)) {
-			/* get processor flags from MSR 0x17 */
-			rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
-			uci->pf = 1 << ((val[1] >> 18) & 7);
-		}
-		uci->orig_pf = uci->pf;
+	uci->sig = cpuid_eax(0x00000001);
+
+	if ((c->x86_model >= 5) || (c->x86 > 6)) {
+		/* get processor flags from MSR 0x17 */
+		rdmsr(MSR_IA32_PLATFORM_ID, val[0], val[1]);
+		uci->pf = 1 << ((val[1] >> 18) & 7);
 	}
 
 	wrmsr(MSR_IA32_UCODE_REV, 0, 0);
@@ -180,218 +163,160 @@ static void collect_cpu_info (void *unused)
 			uci->sig, uci->pf, uci->rev);
 }
 
-static inline void mark_microcode_update (int cpu_num, microcode_header_t *mc_header, int sig, int pf, int cksum)
+static inline int microcode_update_match(int cpu_num,
+	microcode_header_t *mc_header, int sig, int pf)
 {
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
 
-	pr_debug("Microcode Found.\n");
-	pr_debug("   Header Revision 0x%x\n", mc_header->hdrver);
-	pr_debug("   Loader Revision 0x%x\n", mc_header->ldrver);
-	pr_debug("   Revision 0x%x \n", mc_header->rev);
-	pr_debug("   Date %x/%x/%x\n",
-		((mc_header->date >> 24 ) & 0xff),
-		((mc_header->date >> 16 ) & 0xff),
-		(mc_header->date & 0xFFFF));
-	pr_debug("   Signature 0x%x\n", sig);
-	pr_debug("   Type 0x%x Family 0x%x Model 0x%x Stepping 0x%x\n",
-		((sig >> 12) & 0x3),
-		((sig >> 8) & 0xf),
-		((sig >> 4) & 0xf),
-		((sig & 0xf)));
-	pr_debug("   Processor Flags 0x%x\n", pf);
-	pr_debug("   Checksum 0x%x\n", cksum);
-
-	if (mc_header->rev < uci->rev) {
-		if (uci->err == MC_NOTFOUND) {
-			uci->err = MC_IGNORED;
-			uci->cksum = mc_header->rev;
-		} else if (uci->err == MC_IGNORED && uci->cksum < mc_header->rev)
-			uci->cksum = mc_header->rev;
-	} else if (mc_header->rev == uci->rev) {
-		if (uci->err < MC_MARKED) {
-			/* notify the caller of success on this cpu */
-			uci->err = MC_SUCCESS;
-		}
-	} else if (uci->err != MC_ALLOCATED || mc_header->rev > uci->mc->hdr.rev) {
-		pr_debug("microcode: CPU%d found a matching microcode update with "
-			" revision 0x%x (current=0x%x)\n", cpu_num, mc_header->rev, uci->rev);
-		uci->cksum = cksum;
-		uci->pf = pf; /* keep the original mc pf for cksum calculation */
-		uci->err = MC_MARKED; /* found the match */
-		for_each_online_cpu(cpu_num) {
-			if (ucode_cpu_info + cpu_num != uci
-			    && ucode_cpu_info[cpu_num].mc == uci->mc) {
-				uci->mc = NULL;
-				break;
-			}
-		}
-		if (uci->mc != NULL) {
-			vfree(uci->mc);
-			uci->mc = NULL;
-		}
-	}
-	return;
+	if (!sigmatch(sig, uci->sig, pf, uci->pf)
+		|| mc_header->rev <= uci->rev)
+		return 0;
+	return 1;
 }
 
-static int find_matching_ucodes (void) 
+static int microcode_sanity_check(void *mc)
 {
-	int cursor = 0;
-	int error = 0;
-
-	while (cursor + MC_HEADER_SIZE < user_buffer_size) {
-		microcode_header_t mc_header;
-		void *newmc = NULL;
-		int i, sum, cpu_num, allocated_flag, total_size, data_size, ext_table_size;
+	microcode_header_t *mc_header = mc;
+	struct extended_sigtable *ext_header = NULL;
+	struct extended_signature *ext_sig;
+	unsigned long total_size, data_size, ext_table_size;
+	int sum, orig_sum, ext_sigcount = 0, i;
+
+	total_size = get_totalsize(mc_header);
+	data_size = get_datasize(mc_header);
+	if ((data_size + MC_HEADER_SIZE > total_size)
+		|| (data_size < DEFAULT_UCODE_DATASIZE)) {
+		printk(KERN_ERR "microcode: error! "
+			"Bad data size in microcode data file\n");
+		return -EINVAL;
+	}
 
-		if (copy_from_user(&mc_header, user_buffer + cursor, MC_HEADER_SIZE)) {
-			printk(KERN_ERR "microcode: error! Can not read user data\n");
-			error = -EFAULT;
-			goto out;
+	if (mc_header->ldrver != 1 || mc_header->hdrver != 1) {
+		printk(KERN_ERR "microcode: error! "
+			"Unknown microcode update format\n");
+		return -EINVAL;
+	}
+	ext_table_size = total_size - (MC_HEADER_SIZE + data_size);
+	if (ext_table_size) {
+		if ((ext_table_size < EXT_HEADER_SIZE)
+		 || ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) {
+			printk(KERN_ERR "microcode: error! "
+				"Small exttable size in microcode data file\n");
+			return -EINVAL;
 		}
-
-		total_size = get_totalsize(&mc_header);
-		if ((cursor + total_size > user_buffer_size) || (total_size < DEFAULT_UCODE_TOTALSIZE)) {
-			printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
-			error = -EINVAL;
-			goto out;
+		ext_header = mc + MC_HEADER_SIZE + data_size;
+		if (ext_table_size != exttable_size(ext_header)) {
+			printk(KERN_ERR "microcode: error! "
+				"Bad exttable size in microcode data file\n");
+			return -EFAULT;
 		}
+		ext_sigcount = ext_header->count;
+	}
 
-		data_size = get_datasize(&mc_header);
-		if ((data_size + MC_HEADER_SIZE > total_size) || (data_size < DEFAULT_UCODE_DATASIZE)) {
-			printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
-			error = -EINVAL;
-			goto out;
+	/* check extended table checksum */
+	if (ext_table_size) {
+		int ext_table_sum = 0;
+		int * ext_tablep = (int *)ext_header;
+
+		i = ext_table_size / DWSIZE;
+		while (i--)
+			ext_table_sum += ext_tablep[i];
+		if (ext_table_sum) {
+			printk(KERN_WARNING "microcode: aborting, "
+				"bad extended signature table checksum\n");
+			return -EINVAL;
 		}
+	}
 
-		if (mc_header.ldrver != 1 || mc_header.hdrver != 1) {
-			printk(KERN_ERR "microcode: error! Unknown microcode update format\n");
-			error = -EINVAL;
-			goto out;
+	/* calculate the checksum */
+	orig_sum = 0;
+	i = (MC_HEADER_SIZE + data_size) / DWSIZE;
+	while (i--)
+		orig_sum += ((int *)mc)[i];
+	if (orig_sum) {
+		printk(KERN_ERR "microcode: aborting, bad checksum\n");
+		return -EINVAL;
+	}
+	if (!ext_table_size)
+		return 0;
+	/* check extended signature checksum */
+	for (i = 0; i < ext_sigcount; i++) {
+		ext_sig = (struct extended_signature *)((void *)ext_header
+			+ EXT_HEADER_SIZE + EXT_SIGNATURE_SIZE * i);
+		sum = orig_sum
+			- (mc_header->sig + mc_header->pf + mc_header->cksum)
+			+ (ext_sig->sig + ext_sig->pf + ext_sig->cksum);
+		if (sum) {
+			printk(KERN_ERR "microcode: aborting, bad checksum\n");
+			return -EINVAL;
 		}
+	}
+	return 0;
+}
 
-		for_each_online_cpu(cpu_num) {
-			struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-
-			if (sigmatch(mc_header.sig, uci->sig, mc_header.pf, uci->orig_pf))
-				mark_microcode_update(cpu_num, &mc_header, mc_header.sig, mc_header.pf, mc_header.cksum);
-		}
+/*
+ * return 0 - no update found
+ * return 1 - found update
+ * return < 0 - error
+ */
+static int get_maching_microcode(void *mc, int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+	microcode_header_t *mc_header = mc;
+	struct extended_sigtable *ext_header;
+	unsigned long total_size = get_totalsize(mc_header);
+	int ext_sigcount, i;
+	struct extended_signature *ext_sig;
+	void *new_mc;
+
+	if (microcode_update_match(cpu, mc_header,
+			mc_header->sig, mc_header->pf))
+		goto find;
+
+	if (total_size <= get_datasize(mc_header) + MC_HEADER_SIZE)
+		return 0;
+
+	ext_header = (struct extended_sigtable *)(mc +
+			get_datasize(mc_header) + MC_HEADER_SIZE);
+	ext_sigcount = ext_header->count;
+	ext_sig = (struct extended_signature *)((void *)ext_header
+			+ EXT_HEADER_SIZE);
+	for (i = 0; i < ext_sigcount; i++) {
+		if (microcode_update_match(cpu, mc_header,
+				ext_sig->sig, ext_sig->pf))
+			goto find;
+		ext_sig++;
+	}
+	return 0;
+find:
+	pr_debug("microcode: CPU %d found a matching microcode update with"
+		" version 0x%x (current=0x%x)\n", cpu, mc_header->rev,uci->rev);
+	new_mc = vmalloc(total_size);
+	if (!new_mc) {
+		printk(KERN_ERR "microcode: error! Can not allocate memory\n");
+		return -ENOMEM;
+	}
 
-		ext_table_size = total_size - (MC_HEADER_SIZE + data_size);
-		if (ext_table_size) {
-			struct extended_sigtable ext_header;
-			struct extended_signature ext_sig;
-			int ext_sigcount;
+	/* free previous update file */
+	vfree(uci->mc);
 
-			if ((ext_table_size < EXT_HEADER_SIZE) 
-					|| ((ext_table_size - EXT_HEADER_SIZE) % EXT_SIGNATURE_SIZE)) {
-				printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
-				error = -EINVAL;
-				goto out;
-			}
-			if (copy_from_user(&ext_header, user_buffer + cursor 
-					+ MC_HEADER_SIZE + data_size, EXT_HEADER_SIZE)) {
-				printk(KERN_ERR "microcode: error! Can not read user data\n");
-				error = -EFAULT;
-				goto out;
-			}
-			if (ext_table_size != exttable_size(&ext_header)) {
-				printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
-				error = -EFAULT;
-				goto out;
-			}
-
-			ext_sigcount = ext_header.count;
-			
-			for (i = 0; i < ext_sigcount; i++) {
-				if (copy_from_user(&ext_sig, user_buffer + cursor + MC_HEADER_SIZE + data_size + EXT_HEADER_SIZE 
-						+ EXT_SIGNATURE_SIZE * i, EXT_SIGNATURE_SIZE)) {
-					printk(KERN_ERR "microcode: error! Can not read user data\n");
-					error = -EFAULT;
-					goto out;
-				}
-				for_each_online_cpu(cpu_num) {
-					struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-
-					if (sigmatch(ext_sig.sig, uci->sig, ext_sig.pf, uci->orig_pf)) {
-						mark_microcode_update(cpu_num, &mc_header, ext_sig.sig, ext_sig.pf, ext_sig.cksum);
-					}
-				}
-			}
-		}
-		/* now check if any cpu has matched */
-		allocated_flag = 0;
-		sum = 0;
-		for_each_online_cpu(cpu_num) {
-			if (ucode_cpu_info[cpu_num].err == MC_MARKED) { 
-				struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
-				if (!allocated_flag) {
-					allocated_flag = 1;
-					newmc = vmalloc(total_size);
-					if (!newmc) {
-						printk(KERN_ERR "microcode: error! Can not allocate memory\n");
-						error = -ENOMEM;
-						goto out;
-					}
-					if (copy_from_user(newmc + MC_HEADER_SIZE, 
-								user_buffer + cursor + MC_HEADER_SIZE, 
-								total_size - MC_HEADER_SIZE)) {
-						printk(KERN_ERR "microcode: error! Can not read user data\n");
-						vfree(newmc);
-						error = -EFAULT;
-						goto out;
-					}
-					memcpy(newmc, &mc_header, MC_HEADER_SIZE);
-					/* check extended table checksum */
-					if (ext_table_size) {
-						int ext_table_sum = 0;
-						int * ext_tablep = (((void *) newmc) + MC_HEADER_SIZE + data_size);
-						i = ext_table_size / DWSIZE;
-						while (i--) ext_table_sum += ext_tablep[i];
-						if (ext_table_sum) {
-							printk(KERN_WARNING "microcode: aborting, bad extended signature table checksum\n");
-							vfree(newmc);
-							error = -EINVAL;
-							goto out;
-						}
-					}
-
-					/* calculate the checksum */
-					i = (MC_HEADER_SIZE + data_size) / DWSIZE;
-					while (i--) sum += ((int *)newmc)[i];
-					sum -= (mc_header.sig + mc_header.pf + mc_header.cksum);
-				}
-				ucode_cpu_info[cpu_num].mc = newmc;
-				ucode_cpu_info[cpu_num].err = MC_ALLOCATED; /* mc updated */
-				if (sum + uci->sig + uci->pf + uci->cksum != 0) {
-					printk(KERN_ERR "microcode: CPU%d aborting, bad checksum\n", cpu_num);
-					error = -EINVAL;
-					goto out;
-				}
-			}
-		}
-		cursor += total_size; /* goto the next update patch */
-	} /* end of while */
-out:
-	return error;
+	memcpy(new_mc, mc, total_size);
+	uci->mc = new_mc;
+	return 1;
 }
 
-static void do_update_one (void * unused)
+static void apply_microcode(int cpu)
 {
 	unsigned long flags;
 	unsigned int val[2];
-	int cpu_num = smp_processor_id();
+	int cpu_num = raw_smp_processor_id();
 	struct ucode_cpu_info *uci = ucode_cpu_info + cpu_num;
 
-	if (uci->mc == NULL) {
-		if (verbose) {
-			if (uci->err == MC_SUCCESS)
-				printk(KERN_INFO "microcode: CPU%d already at revision 0x%x\n",
-					cpu_num, uci->rev);
-			else
-				printk(KERN_INFO "microcode: No new microcode data for CPU%d\n", cpu_num);
-		}
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu_num != cpu);
+
+	if (uci->mc == NULL)
 		return;
-	}
 
 	/* serialize access to the physical write to MSR 0x79 */
 	spin_lock_irqsave(&microcode_update_lock, flags);          
@@ -408,54 +333,98 @@ static void do_update_one (void * unused)
 	/* get the current revision from MSR 0x8B */
 	rdmsr(MSR_IA32_UCODE_REV, val[0], val[1]);
 
-	/* notify the caller of success on this cpu */
-	uci->err = MC_SUCCESS;
 	spin_unlock_irqrestore(&microcode_update_lock, flags);
-	printk(KERN_INFO "microcode: CPU%d updated from revision "
+	if (val[1] != uci->mc->hdr.rev) {
+		printk(KERN_ERR "microcode: CPU%d updated from revision "
+			"0x%x to 0x%x failed\n", cpu_num, uci->rev, val[1]);
+		return;
+	}
+	pr_debug("microcode: CPU%d updated from revision "
 	       "0x%x to 0x%x, date = %08x \n", 
 	       cpu_num, uci->rev, val[1], uci->mc->hdr.date);
-	return;
+	uci->rev = val[1];
 }
 
-static int do_microcode_update (void)
-{
-	int i, error;
+#ifdef CONFIG_MICROCODE_OLD_INTERFACE
+static void __user *user_buffer;	/* user area microcode data buffer */
+static unsigned int user_buffer_size;	/* it's size */
 
-	if (on_each_cpu(collect_cpu_info, NULL, 1, 1) != 0) {
-		printk(KERN_ERR "microcode: Error! Could not run on all processors\n");
-		error = -EIO;
-		goto out;
+static long get_next_ucode(void **mc, long offset)
+{
+	microcode_header_t mc_header;
+	unsigned long total_size;
+
+	/* No more data */
+	if (offset >= user_buffer_size)
+		return 0;
+	if (copy_from_user(&mc_header, user_buffer + offset, MC_HEADER_SIZE)) {
+		printk(KERN_ERR "microcode: error! Can not read user data\n");
+		return -EFAULT;
 	}
-
-	if ((error = find_matching_ucodes())) {
-		printk(KERN_ERR "microcode: Error in the microcode data\n");
-		goto out_free;
+	total_size = get_totalsize(&mc_header);
+	if ((offset + total_size > user_buffer_size)
+		|| (total_size < DEFAULT_UCODE_TOTALSIZE)) {
+		printk(KERN_ERR "microcode: error! Bad total size in microcode "
+				"data file\n");
+		return -EINVAL;
 	}
-
-	if (on_each_cpu(do_update_one, NULL, 1, 1) != 0) {
-		printk(KERN_ERR "microcode: Error! Could not run on all processors\n");
-		error = -EIO;
+	*mc = vmalloc(total_size);
+	if (!*mc)
+		return -ENOMEM;
+	if (copy_from_user(*mc, user_buffer + offset, total_size)) {
+		printk(KERN_ERR "microcode: error! Can not read user data\n");
+		vfree(*mc);
+		return -EFAULT;
 	}
+	return offset + total_size;
+}
 
-out_free:
-	for_each_online_cpu(i) {
-		if (ucode_cpu_info[i].mc) {
-			int j;
-			void *tmp = ucode_cpu_info[i].mc;
-			vfree(tmp);
-			for_each_online_cpu(j) {
-				if (ucode_cpu_info[j].mc == tmp)
-					ucode_cpu_info[j].mc = NULL;
-			}
+static int do_microcode_update (void)
+{
+	long cursor = 0;
+	int error = 0;
+	void * new_mc;
+	int cpu;
+	cpumask_t old;
+
+	old = current->cpus_allowed;
+
+	while ((cursor = get_next_ucode(&new_mc, cursor)) > 0) {
+		error = microcode_sanity_check(new_mc);
+		if (error)
+			goto out;
+		/*
+		 * It's possible the data file has multiple matching ucode,
+		 * lets keep searching till the latest version
+		 */
+		for_each_online_cpu(cpu) {
+			struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+			if (!uci->valid)
+				continue;
+			set_cpus_allowed(current, cpumask_of_cpu(cpu));
+			error = get_maching_microcode(new_mc, cpu);
+			if (error < 0)
+				goto out;
+			if (error == 1)
+				apply_microcode(cpu);
 		}
-		if (ucode_cpu_info[i].err == MC_IGNORED && verbose)
-			printk(KERN_WARNING "microcode: CPU%d not 'upgrading' to earlier revision"
-			       " 0x%x (current=0x%x)\n", i, ucode_cpu_info[i].cksum, ucode_cpu_info[i].rev);
+		vfree(new_mc);
 	}
 out:
+	if (cursor > 0)
+		vfree(new_mc);
+	if (cursor < 0)
+		error = cursor;
+	set_cpus_allowed(current, old);
 	return error;
 }
 
+static int microcode_open (struct inode *unused1, struct file *unused2)
+{
+	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
+}
+
 static ssize_t microcode_write (struct file *file, const char __user *buf, size_t len, loff_t *ppos)
 {
 	ssize_t ret;
@@ -470,6 +439,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_
 		return -EINVAL;
 	}
 
+	lock_cpu_hotplug();
 	mutex_lock(&microcode_mutex);
 
 	user_buffer = (void __user *) buf;
@@ -480,6 +450,7 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_
 		ret = (ssize_t)len;
 
 	mutex_unlock(&microcode_mutex);
+	unlock_cpu_hotplug();
 
 	return ret;
 }
@@ -496,7 +467,7 @@ static struct miscdevice microcode_dev = {
 	.fops		= &microcode_fops,
 };
 
-static int __init microcode_init (void)
+static int __init microcode_dev_init (void)
 {
 	int error;
 
@@ -508,6 +479,28 @@ static int __init microcode_init (void)
 		return error;
 	}
 
+	return 0;
+}
+
+static void __exit microcode_dev_exit (void)
+{
+	misc_deregister(&microcode_dev);
+}
+
+MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
+#else
+#define microcode_dev_init() 0
+#define microcode_dev_exit() do { } while(0)
+#endif
+
+static int __init microcode_init (void)
+{
+	int error;
+
+	error = microcode_dev_init();
+	if (error)
+		return error;
+
 	printk(KERN_INFO 
 		"IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@veritas.com>\n");
 	return 0;
@@ -515,9 +508,8 @@ static int __init microcode_init (void)
 
 static void __exit microcode_exit (void)
 {
-	misc_deregister(&microcode_dev);
+	microcode_dev_exit();
 }
 
 module_init(microcode_init)
 module_exit(microcode_exit)
-MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 326aff7a87ea..ab4cbcb9e3b3 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -186,6 +186,11 @@ config MICROCODE
 	  If you use modprobe or kmod you may also want to add the line
 	  'alias char-major-10-184 microcode' to your /etc/modules.conf file.
 
+config MICROCODE_OLD_INTERFACE
+	bool
+	depends on MICROCODE
+	default y
+
 config X86_MSR
 	tristate "/dev/cpu/*/msr - Model-specific register support"
 	help
-- 
cgit v1.2.1


From a30a6a2cb0fdc2c9701d6ddfb21affeb8146c038 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 27 Sep 2006 01:50:52 -0700
Subject: [PATCH] x86 microcode: using request_firmware to pull microcode

Using request_firmware to pull ucode from userspace, so we don't need the
application 'microcode_ctl' to assist.  We name each ucode file according
to CPU's info as intel-ucode/family-model-stepping.  In this way we could
split ucode file as small one.  This has a lot of advantages such as
selectively update and validate microcode for specific models, better
manage microcode file, easily write tools for administerators and so on.
with the changes, we should put all intel-ucode/xx-xx-xx microcode files
into the firmware dir (I had a tool to split previous big data file into
small one and later we will release new style data file).  The init script
should be changed to just loading the driver without unloading

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Tigran Aivazian <tigran@veritas.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/microcode.c  | 116 ++++++++++++++++++++++++++++++++++++++++++
 drivers/base/firmware_class.c |   2 +-
 2 files changed, 117 insertions(+), 1 deletion(-)

diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index 00bdc3dd693e..f8f1ec5f26c1 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -83,6 +83,9 @@
 #include <linux/spinlock.h>
 #include <linux/mm.h>
 #include <linux/mutex.h>
+#include <linux/cpu.h>
+#include <linux/firmware.h>
+#include <linux/platform_device.h>
 
 #include <asm/msr.h>
 #include <asm/uaccess.h>
@@ -493,6 +496,112 @@ MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
 #define microcode_dev_exit() do { } while(0)
 #endif
 
+static long get_next_ucode_from_buffer(void **mc, void *buf,
+	unsigned long size, long offset)
+{
+	microcode_header_t *mc_header;
+	unsigned long total_size;
+
+	/* No more data */
+	if (offset >= size)
+		return 0;
+	mc_header = (microcode_header_t *)(buf + offset);
+	total_size = get_totalsize(mc_header);
+
+	if ((offset + total_size > size)
+		|| (total_size < DEFAULT_UCODE_TOTALSIZE)) {
+		printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
+		return -EINVAL;
+	}
+
+	*mc = vmalloc(total_size);
+	if (!*mc) {
+		printk(KERN_ERR "microcode: error! Can not allocate memory\n");
+		return -ENOMEM;
+	}
+	memcpy(*mc, buf + offset, total_size);
+	return offset + total_size;
+}
+
+/* fake device for request_firmware */
+static struct platform_device *microcode_pdev;
+
+static int cpu_request_microcode(int cpu)
+{
+	char name[30];
+	struct cpuinfo_x86 *c = cpu_data + cpu;
+	const struct firmware *firmware;
+	void * buf;
+	unsigned long size;
+	long offset = 0;
+	int error;
+	void *mc;
+
+	/* We should bind the task to the CPU */
+	BUG_ON(cpu != raw_smp_processor_id());
+	sprintf(name,"intel-ucode/%02x-%02x-%02x",
+		c->x86, c->x86_model, c->x86_mask);
+	error = request_firmware(&firmware, name, &microcode_pdev->dev);
+	if (error) {
+		pr_debug("ucode data file %s load failed\n", name);
+		return error;
+	}
+	buf = (void *)firmware->data;
+	size = firmware->size;
+	while ((offset = get_next_ucode_from_buffer(&mc, buf, size, offset))
+			> 0) {
+		error = microcode_sanity_check(mc);
+		if (error)
+			break;
+		error = get_maching_microcode(mc, cpu);
+		if (error < 0)
+			break;
+		/*
+		 * It's possible the data file has multiple matching ucode,
+		 * lets keep searching till the latest version
+		 */
+		if (error == 1) {
+			apply_microcode(cpu);
+			error = 0;
+		}
+		vfree(mc);
+	}
+	if (offset > 0)
+		vfree(mc);
+	if (offset < 0)
+		error = offset;
+	release_firmware(firmware);
+
+	return error;
+}
+
+static void microcode_init_cpu(int cpu)
+{
+	cpumask_t old;
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	old = current->cpus_allowed;
+
+	set_cpus_allowed(current, cpumask_of_cpu(cpu));
+	mutex_lock(&microcode_mutex);
+	collect_cpu_info(cpu);
+	if (uci->valid)
+		cpu_request_microcode(cpu);
+	mutex_unlock(&microcode_mutex);
+	set_cpus_allowed(current, old);
+}
+
+static void microcode_fini_cpu(int cpu)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	mutex_lock(&microcode_mutex);
+	uci->valid = 0;
+	vfree(uci->mc);
+	uci->mc = NULL;
+	mutex_unlock(&microcode_mutex);
+}
+
 static int __init microcode_init (void)
 {
 	int error;
@@ -500,6 +609,12 @@ static int __init microcode_init (void)
 	error = microcode_dev_init();
 	if (error)
 		return error;
+	microcode_pdev = platform_device_register_simple("microcode", -1,
+							 NULL, 0);
+	if (IS_ERR(microcode_pdev)) {
+		microcode_dev_exit();
+		return PTR_ERR(microcode_pdev);
+	}
 
 	printk(KERN_INFO 
 		"IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@veritas.com>\n");
@@ -509,6 +624,7 @@ static int __init microcode_init (void)
 static void __exit microcode_exit (void)
 {
 	microcode_dev_exit();
+	platform_device_unregister(microcode_pdev);
 }
 
 module_init(microcode_init)
diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c
index 5d6c011183f5..77bf8826e2f9 100644
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -602,7 +602,7 @@ firmware_class_exit(void)
 	class_unregister(&firmware_class);
 }
 
-module_init(firmware_class_init);
+fs_initcall(firmware_class_init);
 module_exit(firmware_class_exit);
 
 EXPORT_SYMBOL(release_firmware);
-- 
cgit v1.2.1


From 9a4b9efa1d39d7d31bed08fbe5a9b2a03b2759d4 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 27 Sep 2006 01:50:53 -0700
Subject: [PATCH] x86 microcode: add sysfs and hotplug support

Add sysfs support.  Currently each CPU has three microcode related
attributes.  One is 'version' which shows current ucode version of CPU.
Tools can use the attribute do validation or show CPU ucode status.  one is
'reload' which allows manually reloading ucode.  Another is
'processor_flags', which exports processor flags, so we can write tools to
check if CPU has latest ucode.  Also add suspend/resume and CPU hotplug
support.

[akpm@osdl.org: cleanups, build fix]
[bunk@stusta.de: Kconfig fixes]
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Acked-by: Tigran Aivazian <tigran@veritas.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig            |   1 +
 arch/i386/kernel/microcode.c | 154 ++++++++++++++++++++++++++++++++++++++++++-
 arch/x86_64/Kconfig          |   1 +
 3 files changed, 153 insertions(+), 3 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 0cbeb26ac3fb..3fd2f256f2be 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -401,6 +401,7 @@ config X86_REBOOTFIXUPS
 
 config MICROCODE
 	tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
+	select FW_LOADER
 	---help---
 	  If you say Y here and also to "/dev file system support" in the
 	  'File systems' section, you will be able to update the microcode on
diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index f8f1ec5f26c1..467901ebb992 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -219,7 +219,7 @@ static int microcode_sanity_check(void *mc)
 	/* check extended table checksum */
 	if (ext_table_size) {
 		int ext_table_sum = 0;
-		int * ext_tablep = (int *)ext_header;
+		int *ext_tablep = (int *)ext_header;
 
 		i = ext_table_size / DWSIZE;
 		while (i--)
@@ -386,7 +386,7 @@ static int do_microcode_update (void)
 {
 	long cursor = 0;
 	int error = 0;
-	void * new_mc;
+	void *new_mc;
 	int cpu;
 	cpumask_t old;
 
@@ -531,7 +531,7 @@ static int cpu_request_microcode(int cpu)
 	char name[30];
 	struct cpuinfo_x86 *c = cpu_data + cpu;
 	const struct firmware *firmware;
-	void * buf;
+	void *buf;
 	unsigned long size;
 	long offset = 0;
 	int error;
@@ -602,6 +602,136 @@ static void microcode_fini_cpu(int cpu)
 	mutex_unlock(&microcode_mutex);
 }
 
+static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+	char *end;
+	unsigned long val = simple_strtoul(buf, &end, 0);
+	int err = 0;
+	int cpu = dev->id;
+
+	if (end == buf)
+		return -EINVAL;
+	if (val == 1) {
+		cpumask_t old;
+
+		old = current->cpus_allowed;
+
+		lock_cpu_hotplug();
+		set_cpus_allowed(current, cpumask_of_cpu(cpu));
+
+		mutex_lock(&microcode_mutex);
+		if (uci->valid)
+			err = cpu_request_microcode(cpu);
+		mutex_unlock(&microcode_mutex);
+		unlock_cpu_hotplug();
+		set_cpus_allowed(current, old);
+	}
+	if (err)
+		return err;
+	return sz;
+}
+
+static ssize_t version_show(struct sys_device *dev, char *buf)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+	return sprintf(buf, "0x%x\n", uci->rev);
+}
+
+static ssize_t pf_show(struct sys_device *dev, char *buf)
+{
+	struct ucode_cpu_info *uci = ucode_cpu_info + dev->id;
+
+	return sprintf(buf, "0x%x\n", uci->pf);
+}
+
+static SYSDEV_ATTR(reload, 0200, NULL, reload_store);
+static SYSDEV_ATTR(version, 0400, version_show, NULL);
+static SYSDEV_ATTR(processor_flags, 0400, pf_show, NULL);
+
+static struct attribute *mc_default_attrs[] = {
+	&attr_reload.attr,
+	&attr_version.attr,
+	&attr_processor_flags.attr,
+	NULL
+};
+
+static struct attribute_group mc_attr_group = {
+	.attrs = mc_default_attrs,
+	.name = "microcode",
+};
+
+static int mc_sysdev_add(struct sys_device *sys_dev)
+{
+	int cpu = sys_dev->id;
+	struct ucode_cpu_info *uci = ucode_cpu_info + cpu;
+
+	if (!cpu_online(cpu))
+		return 0;
+	pr_debug("Microcode:CPU %d added\n", cpu);
+	memset(uci, 0, sizeof(*uci));
+	sysfs_create_group(&sys_dev->kobj, &mc_attr_group);
+
+	microcode_init_cpu(cpu);
+	return 0;
+}
+
+static int mc_sysdev_remove(struct sys_device *sys_dev)
+{
+	int cpu = sys_dev->id;
+
+	if (!cpu_online(cpu))
+		return 0;
+	pr_debug("Microcode:CPU %d removed\n", cpu);
+	microcode_fini_cpu(cpu);
+	sysfs_remove_group(&sys_dev->kobj, &mc_attr_group);
+	return 0;
+}
+
+static int mc_sysdev_resume(struct sys_device *dev)
+{
+	int cpu = dev->id;
+
+	if (!cpu_online(cpu))
+		return 0;
+	pr_debug("Microcode:CPU %d resumed\n", cpu);
+	/* only CPU 0 will apply ucode here */
+	apply_microcode(0);
+	return 0;
+}
+
+static struct sysdev_driver mc_sysdev_driver = {
+	.add = mc_sysdev_add,
+	.remove = mc_sysdev_remove,
+	.resume = mc_sysdev_resume,
+};
+
+#ifdef CONFIG_HOTPLUG_CPU
+static __cpuinit int
+mc_cpu_callback(struct notifier_block *nb, unsigned long action, void *hcpu)
+{
+	unsigned int cpu = (unsigned long)hcpu;
+	struct sys_device *sys_dev;
+
+	sys_dev = get_cpu_sysdev(cpu);
+	switch (action) {
+	case CPU_ONLINE:
+	case CPU_DOWN_FAILED:
+		mc_sysdev_add(sys_dev);
+		break;
+	case CPU_DOWN_PREPARE:
+		mc_sysdev_remove(sys_dev);
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block mc_cpu_notifier = {
+	.notifier_call = mc_cpu_callback,
+};
+#endif
+
 static int __init microcode_init (void)
 {
 	int error;
@@ -616,6 +746,17 @@ static int __init microcode_init (void)
 		return PTR_ERR(microcode_pdev);
 	}
 
+	lock_cpu_hotplug();
+	error = sysdev_driver_register(&cpu_sysdev_class, &mc_sysdev_driver);
+	unlock_cpu_hotplug();
+	if (error) {
+		microcode_dev_exit();
+		platform_device_unregister(microcode_pdev);
+		return error;
+	}
+
+	register_hotcpu_notifier(&mc_cpu_notifier);
+
 	printk(KERN_INFO 
 		"IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@veritas.com>\n");
 	return 0;
@@ -624,6 +765,13 @@ static int __init microcode_init (void)
 static void __exit microcode_exit (void)
 {
 	microcode_dev_exit();
+
+	unregister_hotcpu_notifier(&mc_cpu_notifier);
+
+	lock_cpu_hotplug();
+	sysdev_driver_unregister(&cpu_sysdev_class, &mc_sysdev_driver);
+	unlock_cpu_hotplug();
+
 	platform_device_unregister(microcode_pdev);
 }
 
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index ab4cbcb9e3b3..32ae1378f35c 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -171,6 +171,7 @@ config X86_GOOD_APIC
 
 config MICROCODE
 	tristate "/dev/cpu/microcode - Intel CPU microcode support"
+	select FW_LOADER
 	---help---
 	  If you say Y here the 'File systems' section, you will be
 	  able to update the microcode on Intel processors. You will
-- 
cgit v1.2.1


From bd8e39f9e4c0960541c8c69e1f7cb321574d7c90 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 27 Sep 2006 01:50:54 -0700
Subject: [PATCH] x86 microcode: don't check the size

IA32 manual says if micorcode update's size is 0, then the size is
default size (2048 bytes). But this doesn't suggest all microcode
update's size should be above 2048 bytes to me. We actually had a
microcode update whose size is 1024 bytes. The patch just removed the
check.

Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Cc: Tigran Aivazian <tigran@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/microcode.c | 14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

diff --git a/arch/i386/kernel/microcode.c b/arch/i386/kernel/microcode.c
index 467901ebb992..9b9479768d5e 100644
--- a/arch/i386/kernel/microcode.c
+++ b/arch/i386/kernel/microcode.c
@@ -187,8 +187,7 @@ static int microcode_sanity_check(void *mc)
 
 	total_size = get_totalsize(mc_header);
 	data_size = get_datasize(mc_header);
-	if ((data_size + MC_HEADER_SIZE > total_size)
-		|| (data_size < DEFAULT_UCODE_DATASIZE)) {
+	if (data_size + MC_HEADER_SIZE > total_size) {
 		printk(KERN_ERR "microcode: error! "
 			"Bad data size in microcode data file\n");
 		return -EINVAL;
@@ -365,8 +364,7 @@ static long get_next_ucode(void **mc, long offset)
 		return -EFAULT;
 	}
 	total_size = get_totalsize(&mc_header);
-	if ((offset + total_size > user_buffer_size)
-		|| (total_size < DEFAULT_UCODE_TOTALSIZE)) {
+	if (offset + total_size > user_buffer_size) {
 		printk(KERN_ERR "microcode: error! Bad total size in microcode "
 				"data file\n");
 		return -EINVAL;
@@ -432,11 +430,6 @@ static ssize_t microcode_write (struct file *file, const char __user *buf, size_
 {
 	ssize_t ret;
 
-	if (len < DEFAULT_UCODE_TOTALSIZE) {
-		printk(KERN_ERR "microcode: not enough data\n"); 
-		return -EINVAL;
-	}
-
 	if ((len >> PAGE_SHIFT) > num_physpages) {
 		printk(KERN_ERR "microcode: too much data (max %ld pages)\n", num_physpages);
 		return -EINVAL;
@@ -508,8 +501,7 @@ static long get_next_ucode_from_buffer(void **mc, void *buf,
 	mc_header = (microcode_header_t *)(buf + offset);
 	total_size = get_totalsize(mc_header);
 
-	if ((offset + total_size > size)
-		|| (total_size < DEFAULT_UCODE_TOTALSIZE)) {
+	if (offset + total_size > size) {
 		printk(KERN_ERR "microcode: error! Bad data in microcode data file\n");
 		return -EINVAL;
 	}
-- 
cgit v1.2.1


From ebba5f9fcb882306bef7175dee987342ec6fcf2f Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Wed, 27 Sep 2006 01:50:55 -0700
Subject: [PATCH] consistently use MAX_ERRNO in __syscall_return

Consistently use MAX_ERRNO when checking for errors in __syscall_return().

[ralf@linux-mips.org: build fix]
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-arm/unistd.h       | 3 ++-
 include/asm-arm26/unistd.h     | 3 ++-
 include/asm-frv/unistd.h       | 3 ++-
 include/asm-h8300/unistd.h     | 6 +++---
 include/asm-i386/unistd.h      | 5 +++--
 include/asm-m32r/unistd.h      | 5 +++--
 include/asm-m68k/unistd.h      | 5 +++--
 include/asm-m68knommu/unistd.h | 5 +++--
 include/asm-s390/unistd.h      | 4 +++-
 include/asm-sh/unistd.h        | 7 +++++--
 include/asm-sh64/unistd.h      | 6 ++++--
 include/asm-v850/unistd.h      | 5 +++--
 include/asm-x86_64/unistd.h    | 5 +++--
 include/linux/err.h            | 4 ++++
 14 files changed, 43 insertions(+), 23 deletions(-)

diff --git a/include/asm-arm/unistd.h b/include/asm-arm/unistd.h
index 1e891f860ef3..2ab4078334bf 100644
--- a/include/asm-arm/unistd.h
+++ b/include/asm-arm/unistd.h
@@ -377,6 +377,7 @@
 #endif
 
 #ifdef __KERNEL__
+#include <linux/err.h>
 #include <linux/linkage.h>
 
 #define __sys2(x) #x
@@ -396,7 +397,7 @@
 
 #define __syscall_return(type, res)					\
 do {									\
-	if ((unsigned long)(res) >= (unsigned long)(-129)) {		\
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) {	\
 		errno = -(res);						\
 		res = -1;						\
 	}								\
diff --git a/include/asm-arm26/unistd.h b/include/asm-arm26/unistd.h
index 70eb6d91cfd0..c6d2436c9d34 100644
--- a/include/asm-arm26/unistd.h
+++ b/include/asm-arm26/unistd.h
@@ -311,6 +311,7 @@
 #define __ARM_NR_usr26			(__ARM_NR_BASE+3)
 
 #ifdef __KERNEL__
+#include <linux/err.h>
 #include <linux/linkage.h>
 
 #define __sys2(x) #x
@@ -322,7 +323,7 @@
 
 #define __syscall_return(type, res)					\
 do {									\
-	if ((unsigned long)(res) >= (unsigned long)(-125)) {		\
+	if ((unsigned long)(res) >= (unsigned long)-MAX_ERRNO) {	\
 		errno = -(res);						\
 		res = -1;						\
 	}								\
diff --git a/include/asm-frv/unistd.h b/include/asm-frv/unistd.h
index b80dbd839475..d104d1b91d39 100644
--- a/include/asm-frv/unistd.h
+++ b/include/asm-frv/unistd.h
@@ -320,6 +320,7 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 310
+#include <linux/err.h>
 
 /*
  * process the return value of a syscall, consigning it to one of two possible fates
@@ -329,7 +330,7 @@
 #define __syscall_return(type, res)					\
 do {									\
         unsigned long __sr2 = (res);					\
-	if (__builtin_expect(__sr2 >= (unsigned long)(-4095), 0)) {	\
+	if (__builtin_expect(__sr2 >= (unsigned long)(-MAX_ERRNO), 0)) { \
 		errno = (-__sr2);					\
 		__sr2 = ~0UL;						\
 	}								\
diff --git a/include/asm-h8300/unistd.h b/include/asm-h8300/unistd.h
index 226dd596c2da..a2dd90462d80 100644
--- a/include/asm-h8300/unistd.h
+++ b/include/asm-h8300/unistd.h
@@ -295,14 +295,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 289
+#include <linux/err.h>
 
-
-/* user-visible error numbers are in the range -1 - -122: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index 565d0897b205..bd9987087adc 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -328,14 +328,15 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 319
+#include <linux/err.h>
 
 /*
- * user-visible error numbers are in the range -1 - -128: see
+ * user-visible error numbers are in the range -1 - -MAX_ERRNO: see
  * <asm-i386/errno.h>
  */
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-(128 + 1))) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res); \
 		res = -1; \
 	} \
diff --git a/include/asm-m32r/unistd.h b/include/asm-m32r/unistd.h
index 89f376e6229f..5c6a9ac6cf1a 100644
--- a/include/asm-m32r/unistd.h
+++ b/include/asm-m32r/unistd.h
@@ -296,8 +296,9 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls 285
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -124: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
  * <asm-m32r/errno.h>
  */
 
@@ -305,7 +306,7 @@
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-(124 + 1))) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* Avoid using "res" which is declared to be in register r0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-m68k/unistd.h b/include/asm-m68k/unistd.h
index 7c0b6296b45c..751632b904db 100644
--- a/include/asm-m68k/unistd.h
+++ b/include/asm-m68k/unistd.h
@@ -288,13 +288,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls		282
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -124: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-m68knommu/unistd.h b/include/asm-m68knommu/unistd.h
index 1b2abdf281e1..21fdc37c5c2c 100644
--- a/include/asm-m68knommu/unistd.h
+++ b/include/asm-m68knommu/unistd.h
@@ -289,13 +289,14 @@
 #ifdef __KERNEL__
 
 #define NR_syscalls		282
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -122: see
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO: see
    <asm-m68k/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* avoid using res which is declared to be in register d0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index 02b942d85c37..d49c54cb5505 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -342,9 +342,11 @@
 
 #ifdef __KERNEL__
 
+#include <linux/err.h>
+
 #define __syscall_return(type, res)			     \
 do {							     \
-	if ((unsigned long)(res) >= (unsigned long)(-4095)) {\
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res);				     \
 		res = -1;				     \
 	}						     \
diff --git a/include/asm-sh/unistd.h b/include/asm-sh/unistd.h
index 76b5430cb458..da127d7901af 100644
--- a/include/asm-sh/unistd.h
+++ b/include/asm-sh/unistd.h
@@ -306,11 +306,14 @@
 
 #ifdef __KERNEL__
 
-/* user-visible error numbers are in the range -1 - -124: see <asm-sh/errno.h> */
+#include <linux/err.h>
+
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO:
+ * see <asm-sh/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-124)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 	/* Avoid using "res" which is declared to be in register r0; \
 	   errno might expand to a function call and clobber it.  */ \
 		int __err = -(res); \
diff --git a/include/asm-sh64/unistd.h b/include/asm-sh64/unistd.h
index 9a1590fffc15..c113566bef33 100644
--- a/include/asm-sh64/unistd.h
+++ b/include/asm-sh64/unistd.h
@@ -347,8 +347,10 @@
 #ifdef __KERNEL__ 
 
 #define NR_syscalls 321
+#include <linux/err.h>
 
-/* user-visible error numbers are in the range -1 - -125: see <asm-sh64/errno.h> */
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO:
+ * see <asm-sh64/errno.h> */
 
 #define __syscall_return(type, res) \
 do { \
@@ -358,7 +360,7 @@ do { \
 	**       life easier in the system call epilogue (see entry.S)      \
 	*/								    \
         register unsigned long __sr2 __asm__ ("r2") = res;		    \
-	if ((unsigned long)(res) >= (unsigned long)(-125)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) {	    \
 		errno = -(res);						    \
 		__sr2 = -1; 						    \
 	} \
diff --git a/include/asm-v850/unistd.h b/include/asm-v850/unistd.h
index bcb44bfe577a..552b7c873a57 100644
--- a/include/asm-v850/unistd.h
+++ b/include/asm-v850/unistd.h
@@ -238,12 +238,13 @@
 #ifdef __KERNEL__
 
 #include <asm/clinkage.h>
+#include <linux/err.h>
 
 #define __syscall_return(type, res)					      \
   do {									      \
-	  /* user-visible error numbers are in the range -1 - -124:	      \
+	  /* user-visible error numbers are in the range -1 - -MAX_ERRNO:      \
 	     see <asm-v850/errno.h> */					      \
-	  if (__builtin_expect ((unsigned long)(res) >= (unsigned long)(-125), 0)) { \
+	  if (__builtin_expect ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO), 0)) { \
 		  errno = -(res);					      \
 		  res = -1;						      \
 	  }								      \
diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h
index eeb98c168e98..6137146516d3 100644
--- a/include/asm-x86_64/unistd.h
+++ b/include/asm-x86_64/unistd.h
@@ -623,16 +623,17 @@ __SYSCALL(__NR_move_pages, sys_move_pages)
 #ifdef __KERNEL__
 
 #define __NR_syscall_max __NR_move_pages
+#include <linux/err.h>
 
 #ifndef __NO_STUBS
 
-/* user-visible error numbers are in the range -1 - -4095 */
+/* user-visible error numbers are in the range -1 - -MAX_ERRNO */
 
 #define __syscall_clobber "r11","rcx","memory" 
 
 #define __syscall_return(type, res) \
 do { \
-	if ((unsigned long)(res) >= (unsigned long)(-127)) { \
+	if ((unsigned long)(res) >= (unsigned long)(-MAX_ERRNO)) { \
 		errno = -(res); \
 		res = -1; \
 	} \
diff --git a/include/linux/err.h b/include/linux/err.h
index cd3b367f7445..1ab1d44f8d3b 100644
--- a/include/linux/err.h
+++ b/include/linux/err.h
@@ -15,6 +15,8 @@
  */
 #define MAX_ERRNO	4095
 
+#ifndef __ASSEMBLY__
+
 #define IS_ERR_VALUE(x) unlikely((x) >= (unsigned long)-MAX_ERRNO)
 
 static inline void *ERR_PTR(long error)
@@ -32,4 +34,6 @@ static inline long IS_ERR(const void *ptr)
 	return IS_ERR_VALUE((unsigned long)ptr);
 }
 
+#endif
+
 #endif /* _LINUX_ERR_H */
-- 
cgit v1.2.1


From 07563c711fbc25389e58ab9c9f0b9de2fce56760 Mon Sep 17 00:00:00 2001
From: Michael Tokarev <mjt@tls.mks.ru>
Date: Wed, 27 Sep 2006 01:50:56 -0700
Subject: [PATCH] EISA bus MODALIAS attributes support

Add modalias attribute support for the almost forgotten now EISA bus and
(at least some) EISA-aware modules.

The modalias entry looks like (for an 3c509 NIC):

 eisa:sTCM5093

and the in-module alias like:

 eisa:sTCM5093*

The patch moves struct eisa_device_id declaration from include/linux/eisa.h
to include/linux/mod_devicetable.h (so that the former now #includes the
latter), adds proper MODULE_DEVICE_TABLE(eisa, ...) statements for all
drivers with EISA IDs I found (some drivers already have that DEVICE_TABLE
declared), and adds recognision of __mod_eisa_device_table to
scripts/mod/file2alias.c so that proper modules.alias will be generated.

There's no support for /lib/modules/$kver/modules.eisamap, as it's not used
by any existing tools, and because with in-kernel modalias mechanism those
maps are obsolete anyway.

The rationale for this patch is:

 a) to make EISA bus to act as other busses with modalias
    support, to unify driver loading

 b) to foget about EISA finally - with this patch, kernel
    (who still supports EISA) will be the only one who knows
    how to choose the necessary drivers for this bus ;)

[akpm@osdl.org: fix the kbuild bit]
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Randy Dunlap <rdunlap@xenotime.net>
Acked-the-net-bits-by: Jeff Garzik <jeff@garzik.org>
Acked-the-tulip-bit-by: Valerie Henson <val_henson@linux.intel.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/eisa/eisa-bus.c            | 23 +++++++++++++++++++++++
 drivers/net/3c509.c                |  1 +
 drivers/net/3c59x.c                |  1 +
 drivers/net/ne3210.c               |  1 +
 drivers/net/tulip/de4x5.c          |  1 +
 drivers/scsi/aha1740.c             |  1 +
 drivers/scsi/aic7xxx/aic7770_osm.c |  3 ++-
 drivers/scsi/sim710.c              |  1 +
 include/linux/eisa.h               |  8 +-------
 include/linux/mod_devicetable.h    | 12 ++++++++++++
 scripts/mod/file2alias.c           | 12 ++++++++++++
 11 files changed, 56 insertions(+), 8 deletions(-)

diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c
index 6078e2f58817..3a365e159d89 100644
--- a/drivers/eisa/eisa-bus.c
+++ b/drivers/eisa/eisa-bus.c
@@ -128,9 +128,23 @@ static int eisa_bus_match (struct device *dev, struct device_driver *drv)
 	return 0;
 }
 
+static int eisa_bus_uevent(struct device *dev, char **envp, int num_envp,
+			   char *buffer, int buffer_size)
+{
+	struct eisa_device *edev = to_eisa_device(dev);
+	int i = 0;
+	int length = 0;
+
+	add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
+		       "MODALIAS=" EISA_DEVICE_MODALIAS_FMT, edev->id.sig);
+	envp[i] = NULL;
+	return 0;
+}
+
 struct bus_type eisa_bus_type = {
 	.name  = "eisa",
 	.match = eisa_bus_match,
+	.uevent = eisa_bus_uevent,
 };
 
 int eisa_driver_register (struct eisa_driver *edrv)
@@ -160,6 +174,14 @@ static ssize_t eisa_show_state (struct device *dev, struct device_attribute *att
 
 static DEVICE_ATTR(enabled, S_IRUGO, eisa_show_state, NULL);
 
+static ssize_t eisa_show_modalias (struct device *dev, struct device_attribute *attr, char *buf)
+{
+        struct eisa_device *edev = to_eisa_device (dev);
+        return sprintf (buf, EISA_DEVICE_MODALIAS_FMT "\n", edev->id.sig);
+}
+
+static DEVICE_ATTR(modalias, S_IRUGO, eisa_show_modalias, NULL);
+
 static int __init eisa_init_device (struct eisa_root_device *root,
 				    struct eisa_device *edev,
 				    int slot)
@@ -209,6 +231,7 @@ static int __init eisa_register_device (struct eisa_device *edev)
 
 	device_create_file (&edev->dev, &dev_attr_signature);
 	device_create_file (&edev->dev, &dev_attr_enabled);
+	device_create_file (&edev->dev, &dev_attr_modalias);
 
 	return 0;
 }
diff --git a/drivers/net/3c509.c b/drivers/net/3c509.c
index 59c33925be62..b936373ab2a5 100644
--- a/drivers/net/3c509.c
+++ b/drivers/net/3c509.c
@@ -225,6 +225,7 @@ static struct eisa_device_id el3_eisa_ids[] = {
 		{ "TCM5095" },
 		{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, el3_eisa_ids);
 
 static int el3_eisa_probe (struct device *device);
 
diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c
index af301f09d674..df42e28cc80f 100644
--- a/drivers/net/3c59x.c
+++ b/drivers/net/3c59x.c
@@ -851,6 +851,7 @@ static struct eisa_device_id vortex_eisa_ids[] = {
 	{ "TCM5970", CH_3C597 },
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, vortex_eisa_ids);
 
 static int vortex_eisa_probe(struct device *device);
 static int vortex_eisa_remove(struct device *device);
diff --git a/drivers/net/ne3210.c b/drivers/net/ne3210.c
index 0fa8e4d22769..d66328975425 100644
--- a/drivers/net/ne3210.c
+++ b/drivers/net/ne3210.c
@@ -343,6 +343,7 @@ static struct eisa_device_id ne3210_ids[] = {
 	{ "NVL1801" },
 	{ "" },
 };
+MODULE_DEVICE_TABLE(eisa, ne3210_ids);
 
 static struct eisa_driver ne3210_eisa_driver = {
 	.id_table = ne3210_ids,
diff --git a/drivers/net/tulip/de4x5.c b/drivers/net/tulip/de4x5.c
index e661d0a9cc64..fb5fa7d68888 100644
--- a/drivers/net/tulip/de4x5.c
+++ b/drivers/net/tulip/de4x5.c
@@ -2114,6 +2114,7 @@ static struct eisa_device_id de4x5_eisa_ids[] = {
         { "DEC4250", 0 },	/* 0 is the board name index... */
         { "" }
 };
+MODULE_DEVICE_TABLE(eisa, de4x5_eisa_ids);
 
 static struct eisa_driver de4x5_eisa_driver = {
         .id_table = de4x5_eisa_ids,
diff --git a/drivers/scsi/aha1740.c b/drivers/scsi/aha1740.c
index 0e4a7ebe300a..6b35ed8301e0 100644
--- a/drivers/scsi/aha1740.c
+++ b/drivers/scsi/aha1740.c
@@ -681,6 +681,7 @@ static struct eisa_device_id aha1740_ids[] = {
 	{ "ADP0400" },		/* 1744  */
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, aha1740_ids);
 
 static struct eisa_driver aha1740_driver = {
 	.id_table = aha1740_ids,
diff --git a/drivers/scsi/aic7xxx/aic7770_osm.c b/drivers/scsi/aic7xxx/aic7770_osm.c
index 867cbe23579b..1ac119733bac 100644
--- a/drivers/scsi/aic7xxx/aic7770_osm.c
+++ b/drivers/scsi/aic7xxx/aic7770_osm.c
@@ -132,7 +132,8 @@ static struct eisa_device_id aic7770_ids[] = {
 	{ "ADP7770", 5 }, /* AIC7770 generic */
 	{ "" }
 };
-  
+MODULE_DEVICE_TABLE(eisa, aic7770_ids);
+
 static struct eisa_driver aic7770_driver = {
 	.id_table	= aic7770_ids,
 	.driver = {
diff --git a/drivers/scsi/sim710.c b/drivers/scsi/sim710.c
index b27e85428daa..551baccec523 100644
--- a/drivers/scsi/sim710.c
+++ b/drivers/scsi/sim710.c
@@ -282,6 +282,7 @@ static struct eisa_device_id sim710_eisa_ids[] = {
 	{ "HWP0C80" },
 	{ "" }
 };
+MODULE_DEVICE_TABLE(eisa, sim710_eisa_ids);
 
 static __init int
 sim710_eisa_probe(struct device *dev)
diff --git a/include/linux/eisa.h b/include/linux/eisa.h
index 4079242dced8..1ff7c1392525 100644
--- a/include/linux/eisa.h
+++ b/include/linux/eisa.h
@@ -3,8 +3,8 @@
 
 #include <linux/ioport.h>
 #include <linux/device.h>
+#include <linux/mod_devicetable.h>
 
-#define EISA_SIG_LEN   8
 #define EISA_MAX_SLOTS 8
 
 #define EISA_MAX_RESOURCES 4
@@ -27,12 +27,6 @@
 #define EISA_CONFIG_ENABLED         1
 #define EISA_CONFIG_FORCED          2
 
-/* The EISA signature, in ASCII form, null terminated */
-struct eisa_device_id {
-	char          sig[EISA_SIG_LEN];
-	unsigned long driver_data;
-};
-
 /* There is not much we can say about an EISA device, apart from
  * signature, slot number, and base address. dma_mask is set by
  * default to parent device mask..*/
diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h
index f7ca0b09075d..e0c393cc7240 100644
--- a/include/linux/mod_devicetable.h
+++ b/include/linux/mod_devicetable.h
@@ -308,4 +308,16 @@ struct input_device_id {
 	kernel_ulong_t driver_info;
 };
 
+/* EISA */
+
+#define EISA_SIG_LEN   8
+
+/* The EISA signature, in ASCII form, null terminated */
+struct eisa_device_id {
+	char          sig[EISA_SIG_LEN];
+	kernel_ulong_t driver_data;
+};
+
+#define EISA_DEVICE_MODALIAS_FMT "eisa:s%s"
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index de76da80443f..f61c9ccef6aa 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -444,6 +444,14 @@ static int do_input_entry(const char *filename, struct input_device_id *id,
 	return 1;
 }
 
+static int do_eisa_entry(const char *filename, struct eisa_device_id *eisa,
+		char *alias)
+{
+	if (eisa->sig[0])
+		sprintf(alias, EISA_DEVICE_MODALIAS_FMT "*", eisa->sig);
+	return 1;
+}
+
 /* Ignore any prefix, eg. v850 prepends _ */
 static inline int sym_is(const char *symbol, const char *name)
 {
@@ -547,6 +555,10 @@ void handle_moddevtable(struct module *mod, struct elf_info *info,
 		do_table(symval, sym->st_size,
 			 sizeof(struct input_device_id), "input",
 			 do_input_entry, mod);
+	else if (sym_is(symname, "__mod_eisa_device_table"))
+		do_table(symval, sym->st_size,
+			 sizeof(struct eisa_device_id), "eisa",
+			 do_eisa_entry, mod);
 }
 
 /* Now add out buffered information to the generated C source */
-- 
cgit v1.2.1


From 8b0e330b7720a206339887044fa275bf537a5264 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:51:02 -0700
Subject: [PATCH] alloc_fdtable() cleanup

free_fdset(NULL, ...) is legal.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/file.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/fs/file.c b/fs/file.c
index b3c6b82e6a9d..8d3bfca7714b 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -281,10 +281,8 @@ static struct fdtable *alloc_fdtable(int nr)
 out2:
 	nfds = fdt->max_fdset;
 out:
-  	if (new_openset)
-  		free_fdset(new_openset, nfds);
-  	if (new_execset)
-  		free_fdset(new_execset, nfds);
+	free_fdset(new_openset, nfds);
+	free_fdset(new_execset, nfds);
 	kfree(fdt);
 	return NULL;
 }
-- 
cgit v1.2.1


From 7583ddfd3aae1007bc4fc67ea4c07d573d376e9e Mon Sep 17 00:00:00 2001
From: Marcelo Tosatti <marcelo@kvack.org>
Date: Wed, 27 Sep 2006 01:51:02 -0700
Subject: [PATCH] Include __param section in read-only data range

The param section is an array of "kernel_param" structures, storing only
constant data: pointer to name, permission of the variable pointed to by
(void *)arg and pointers to set/get methods.

Move end_rodata down to include __param section in the read-only range used
by CONFIG_DEBUG_RODATA.

Signed-off-by: Marcelo Tosatti <marcelo@kvack.org>
Acked-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-generic/vmlinux.lds.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 253ae1328271..69240b52f8e1 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -118,15 +118,15 @@
         __ksymtab_strings : AT(ADDR(__ksymtab_strings) - LOAD_OFFSET) {	\
 		*(__ksymtab_strings)					\
 	}								\
-	__end_rodata = .;						\
-	. = ALIGN(4096);						\
 									\
 	/* Built-in module parameters. */				\
 	__param : AT(ADDR(__param) - LOAD_OFFSET) {			\
 		VMLINUX_SYMBOL(__start___param) = .;			\
 		*(__param)						\
 		VMLINUX_SYMBOL(__stop___param) = .;			\
-	}
+	}								\
+	__end_rodata = .;						\
+	. = ALIGN(4096);
 
 #define SECURITY_INIT							\
 	.security_initcall.init : AT(ADDR(.security_initcall.init) - LOAD_OFFSET) { \
-- 
cgit v1.2.1


From 571817849c76aabf34d534c905b5e604f2e824c5 Mon Sep 17 00:00:00 2001
From: Pekka J Enberg <penberg@cs.Helsinki.FI>
Date: Wed, 27 Sep 2006 01:51:03 -0700
Subject: [PATCH] msi: use kmem_cache_zalloc()

Simpler, cleaner.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/pci/msi.c | 13 +++----------
 1 file changed, 3 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/msi.c b/drivers/pci/msi.c
index 008235947aa4..27a057409eca 100644
--- a/drivers/pci/msi.c
+++ b/drivers/pci/msi.c
@@ -45,16 +45,10 @@ msi_register(struct msi_ops *ops)
 	return 0;
 }
 
-static void msi_cache_ctor(void *p, kmem_cache_t *cache, unsigned long flags)
-{
-	memset(p, 0, sizeof(struct msi_desc));
-}
-
 static int msi_cache_init(void)
 {
-	msi_cachep = kmem_cache_create("msi_cache",
-			sizeof(struct msi_desc),
-		       	0, SLAB_HWCACHE_ALIGN, msi_cache_ctor, NULL);
+	msi_cachep = kmem_cache_create("msi_cache", sizeof(struct msi_desc),
+					0, SLAB_HWCACHE_ALIGN, NULL, NULL);
 	if (!msi_cachep)
 		return -ENOMEM;
 
@@ -402,11 +396,10 @@ static struct msi_desc* alloc_msi_entry(void)
 {
 	struct msi_desc *entry;
 
-	entry = kmem_cache_alloc(msi_cachep, SLAB_KERNEL);
+	entry = kmem_cache_zalloc(msi_cachep, GFP_KERNEL);
 	if (!entry)
 		return NULL;
 
-	memset(entry, 0, sizeof(struct msi_desc));
 	entry->link.tail = entry->link.head = 0;	/* single message */
 	entry->dev = NULL;
 
-- 
cgit v1.2.1


From b89a81712f486e4f7a606987413e387605fdeaf4 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:04 -0700
Subject: [PATCH] sysctl: Allow /proc/sys without sys_sysctl

Since sys_sysctl is deprecated start allow it to be compiled out.  This
should catch any remaining user space code that cares, and paves the way
for further sysctl cleanups.

[akpm@osdl.org: If sys_sysctl() is not compiled-in, emit a warning]
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/ia64/ia32/sys_ia32.c         |   2 +-
 arch/mips/kernel/linux32.c        |   4 +-
 arch/powerpc/kernel/sys_ppc32.c   |   2 +-
 arch/s390/kernel/compat_linux.c   |   2 +-
 arch/sparc64/kernel/sys_sparc32.c |   2 +-
 arch/x86_64/ia32/sys_ia32.c       |   2 +-
 fs/Kconfig                        |  19 +++++++
 init/Kconfig                      |  31 ++++++-----
 kernel/sysctl.c                   | 113 +++++++++++++-------------------------
 9 files changed, 81 insertions(+), 96 deletions(-)

diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c
index 6aa3c51619ca..bddbd22706ed 100644
--- a/arch/ia64/ia32/sys_ia32.c
+++ b/arch/ia64/ia32/sys_ia32.c
@@ -1942,7 +1942,7 @@ struct sysctl32 {
 	unsigned int	__unused[4];
 };
 
-#ifdef CONFIG_SYSCTL
+#ifdef CONFIG_SYSCTL_SYSCALL
 asmlinkage long
 sys32_sysctl (struct sysctl32 __user *args)
 {
diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index dc500e20cf14..43b1162d714f 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -991,7 +991,7 @@ struct sysctl_args32
 	unsigned int __unused[4];
 };
 
-#ifdef CONFIG_SYSCTL
+#ifdef CONFIG_SYSCTL_SYSCALL
 
 asmlinkage long sys32_sysctl(struct sysctl_args32 __user *args)
 {
@@ -1032,7 +1032,7 @@ asmlinkage long sys32_sysctl(struct sysctl_args32 __user *args)
 	return error;
 }
 
-#endif /* CONFIG_SYSCTL */
+#endif /* CONFIG_SYSCTL_SYSCALL */
 
 asmlinkage long sys32_newuname(struct new_utsname __user * name)
 {
diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c
index 2e292863e982..5e391fc25340 100644
--- a/arch/powerpc/kernel/sys_ppc32.c
+++ b/arch/powerpc/kernel/sys_ppc32.c
@@ -740,7 +740,7 @@ asmlinkage long compat_sys_umask(u32 mask)
 	return sys_umask((int)mask);
 }
 
-#ifdef CONFIG_SYSCTL
+#ifdef CONFIG_SYSCTL_SYSCALL
 struct __sysctl_args32 {
 	u32 name;
 	int nlen;
diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c
index 785c9f70ac98..91b2884fa5c4 100644
--- a/arch/s390/kernel/compat_linux.c
+++ b/arch/s390/kernel/compat_linux.c
@@ -708,7 +708,7 @@ asmlinkage long sys32_sendfile64(int out_fd, int in_fd,
 	return ret;
 }
 
-#ifdef CONFIG_SYSCTL
+#ifdef CONFIG_SYSCTL_SYSCALL
 struct __sysctl_args32 {
 	u32 name;
 	int nlen;
diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c
index c88ae23ce812..69444f266e2d 100644
--- a/arch/sparc64/kernel/sys_sparc32.c
+++ b/arch/sparc64/kernel/sys_sparc32.c
@@ -1016,7 +1016,7 @@ struct __sysctl_args32 {
 
 asmlinkage long sys32_sysctl(struct __sysctl_args32 __user *args)
 {
-#ifndef CONFIG_SYSCTL
+#ifndef CONFIG_SYSCTL_SYSCALL
 	return -ENOSYS;
 #else
 	struct __sysctl_args32 tmp;
diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c
index b0e82c7947dc..f280d3665f4b 100644
--- a/arch/x86_64/ia32/sys_ia32.c
+++ b/arch/x86_64/ia32/sys_ia32.c
@@ -648,7 +648,7 @@ sys32_pause(void)
 }
 
 
-#ifdef CONFIG_SYSCTL
+#ifdef CONFIG_SYSCTL_SYSCALL
 struct sysctl_ia32 {
 	unsigned int	name;
 	int		nlen;
diff --git a/fs/Kconfig b/fs/Kconfig
index a27002668bd3..d311198bba43 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -826,6 +826,25 @@ config PROC_VMCORE
         help
         Exports the dump image of crashed kernel in ELF format.
 
+config PROC_SYSCTL
+	bool "Sysctl support (/proc/sys)" if EMBEDDED
+	depends on PROC_FS
+	select SYSCTL
+	default y
+	---help---
+	  The sysctl interface provides a means of dynamically changing
+	  certain kernel parameters and variables on the fly without requiring
+	  a recompile of the kernel or reboot of the system.  The primary
+	  interface is through /proc/sys.  If you say Y here a tree of
+	  modifiable sysctl entries will be generated beneath the
+          /proc/sys directory. They are explained in the files
+	  in <file:Documentation/sysctl/>.  Note that enabling this
+	  option will enlarge the kernel by at least 8 KB.
+
+	  As it is generally a good thing, you should say Y here unless
+	  building a kernel for install/rescue disks or your system is very
+	  limited in memory.
+
 config SYSFS
 	bool "sysfs file system support" if EMBEDDED
 	default y
diff --git a/init/Kconfig b/init/Kconfig
index 9a7656f0b5ec..4381006dd666 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -273,21 +273,24 @@ config UID16
 	  This enables the legacy 16-bit UID syscall wrappers.
 
 config SYSCTL
-	bool "Sysctl support" if EMBEDDED
-	default y
+	bool
+
+config SYSCTL_SYSCALL
+	bool "Sysctl syscall support"
+	default n
+	select SYSCTL
 	---help---
-	  The sysctl interface provides a means of dynamically changing
-	  certain kernel parameters and variables on the fly without requiring
-	  a recompile of the kernel or reboot of the system.  The primary
-	  interface consists of a system call, but if you say Y to "/proc
-	  file system support", a tree of modifiable sysctl entries will be
-	  generated beneath the /proc/sys directory. They are explained in the
-	  files in <file:Documentation/sysctl/>.  Note that enabling this
-	  option will enlarge the kernel by at least 8 KB.
-
-	  As it is generally a good thing, you should say Y here unless
-	  building a kernel for install/rescue disks or your system is very
-	  limited in memory.
+	  Enable the deprecated sysctl system call.  sys_sysctl uses
+	  binary paths that have been found to be a major pain to maintain
+	  and use.  The interface in /proc/sys is now the primary and what
+	  everyone uses.
+
+	  Nothing has been using the binary sysctl interface for some time
+	  time now so nothing should break if you disable sysctl syscall
+	  support, and you kernel will get marginally smaller.
+
+	  Unless you have an application that uses the sys_syscall interface
+ 	  you should probably say N here.
 
 config KALLSYMS
 	 bool "Load all symbols for debugging/kksymoops" if EMBEDDED
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index bcb3a181dbb2..8bfa7d117c54 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -137,8 +137,11 @@ extern int no_unaligned_warning;
 extern int max_lock_depth;
 #endif
 
-static int parse_table(int __user *, int, void __user *, size_t __user *, void __user *, size_t,
-		       ctl_table *, void **);
+#ifdef CONFIG_SYSCTL_SYSCALL
+static int parse_table(int __user *, int, void __user *, size_t __user *,
+		void __user *, size_t, ctl_table *, void **);
+#endif
+
 static int proc_doutsstring(ctl_table *table, int write, struct file *filp,
 		  void __user *buffer, size_t *lenp, loff_t *ppos);
 
@@ -165,7 +168,7 @@ int sysctl_legacy_va_layout;
 
 /* /proc declarations: */
 
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_PROC_SYSCTL
 
 static ssize_t proc_readsys(struct file *, char __user *, size_t, loff_t *);
 static ssize_t proc_writesys(struct file *, const char __user *, size_t, loff_t *);
@@ -1166,12 +1169,13 @@ static void start_unregistering(struct ctl_table_header *p)
 
 void __init sysctl_init(void)
 {
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_PROC_SYSCTL
 	register_proc_table(root_table, proc_sys_root, &root_table_header);
 	init_irq_proc();
 #endif
 }
 
+#ifdef CONFIG_SYSCTL_SYSCALL
 int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp,
 	       void __user *newval, size_t newlen)
 {
@@ -1225,6 +1229,7 @@ asmlinkage long sys_sysctl(struct __sysctl_args __user *args)
 	unlock_kernel();
 	return error;
 }
+#endif /* CONFIG_SYSCTL_SYSCALL */
 
 /*
  * ctl_perm does NOT grant the superuser all rights automatically, because
@@ -1251,6 +1256,7 @@ static inline int ctl_perm(ctl_table *table, int op)
 	return test_perm(table->mode, op);
 }
 
+#ifdef CONFIG_SYSCTL_SYSCALL
 static int parse_table(int __user *name, int nlen,
 		       void __user *oldval, size_t __user *oldlenp,
 		       void __user *newval, size_t newlen,
@@ -1340,6 +1346,7 @@ int do_sysctl_strategy (ctl_table *table,
 	}
 	return 0;
 }
+#endif /* CONFIG_SYSCTL_SYSCALL */
 
 /**
  * register_sysctl_table - register a sysctl hierarchy
@@ -1427,7 +1434,7 @@ struct ctl_table_header *register_sysctl_table(ctl_table * table,
 	else
 		list_add_tail(&tmp->ctl_entry, &root_table_header.ctl_entry);
 	spin_unlock(&sysctl_lock);
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_PROC_SYSCTL
 	register_proc_table(table, proc_sys_root, tmp);
 #endif
 	return tmp;
@@ -1445,18 +1452,31 @@ void unregister_sysctl_table(struct ctl_table_header * header)
 	might_sleep();
 	spin_lock(&sysctl_lock);
 	start_unregistering(header);
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_PROC_SYSCTL
 	unregister_proc_table(header->ctl_table, proc_sys_root);
 #endif
 	spin_unlock(&sysctl_lock);
 	kfree(header);
 }
 
+#else /* !CONFIG_SYSCTL */
+struct ctl_table_header * register_sysctl_table(ctl_table * table,
+						int insert_at_head)
+{
+	return NULL;
+}
+
+void unregister_sysctl_table(struct ctl_table_header * table)
+{
+}
+
+#endif /* CONFIG_SYSCTL */
+
 /*
  * /proc/sys support
  */
 
-#ifdef CONFIG_PROC_FS
+#ifdef CONFIG_PROC_SYSCTL
 
 /* Scan the sysctl entries in table and add them all into /proc */
 static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, void *set)
@@ -2318,6 +2338,7 @@ int proc_doulongvec_ms_jiffies_minmax(ctl_table *table, int write,
 #endif /* CONFIG_PROC_FS */
 
 
+#ifdef CONFIG_SYSCTL_SYSCALL
 /*
  * General sysctl support routines 
  */
@@ -2460,11 +2481,19 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen,
 	return 1;
 }
 
-#else /* CONFIG_SYSCTL */
+#else /* CONFIG_SYSCTL_SYSCALL */
 
 
 asmlinkage long sys_sysctl(struct __sysctl_args __user *args)
 {
+	static int msg_count;
+
+	if (msg_count < 5) {
+		msg_count++;
+		printk(KERN_INFO
+			"warning: process `%s' used the removed sysctl "
+			"system call\n", current->comm);
+	}
 	return -ENOSYS;
 }
 
@@ -2496,73 +2525,7 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen,
 	return -ENOSYS;
 }
 
-int proc_dostring(ctl_table *table, int write, struct file *filp,
-		  void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec(ctl_table *table, int write, struct file *filp,
-		  void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec_bset(ctl_table *table, int write, struct file *filp,
-			void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec_minmax(ctl_table *table, int write, struct file *filp,
-		    void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec_jiffies(ctl_table *table, int write, struct file *filp,
-			  void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec_userhz_jiffies(ctl_table *table, int write, struct file *filp,
-			  void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_dointvec_ms_jiffies(ctl_table *table, int write, struct file *filp,
-			     void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_doulongvec_minmax(ctl_table *table, int write, struct file *filp,
-		    void __user *buffer, size_t *lenp, loff_t *ppos)
-{
-	return -ENOSYS;
-}
-
-int proc_doulongvec_ms_jiffies_minmax(ctl_table *table, int write,
-				      struct file *filp,
-				      void __user *buffer,
-				      size_t *lenp, loff_t *ppos)
-{
-    return -ENOSYS;
-}
-
-struct ctl_table_header * register_sysctl_table(ctl_table * table, 
-						int insert_at_head)
-{
-	return NULL;
-}
-
-void unregister_sysctl_table(struct ctl_table_header * table)
-{
-}
-
-#endif /* CONFIG_SYSCTL */
+#endif /* CONFIG_SYSCTL_SYSCALL */
 
 /*
  * No sense putting this after each symbol definition, twice,
-- 
cgit v1.2.1


From 35fa2048ab13d1be846be612e395c15c200bd51c Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:05 -0700
Subject: [PATCH] sysctl: Document that sys_sysctl will be removed

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/feature-removal-schedule.txt | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 93b10bd784ed..436697cb9388 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -70,6 +70,18 @@ Who:	Mauro Carvalho Chehab <mchehab@brturbo.com.br>
 
 ---------------------------
 
+What:	sys_sysctl
+When:	January 2007
+Why:	The same information is available through /proc/sys and that is the
+	interface user space prefers to use. And there do not appear to be
+	any existing user in user space of sys_sysctl.  The additional
+	maintenance overhead of keeping a set of binary names gets
+	in the way of doing a good job of maintaining this interface.
+
+Who:	Eric Biederman <ebiederm@xmission.com>
+
+---------------------------
+
 What:	PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
 When:	November 2005
 Files:	drivers/pcmcia/: pcmcia_ioctl.c
-- 
cgit v1.2.1


From c18258c6f0848f97e85287f6271c511a092bb784 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:06 -0700
Subject: [PATCH] pid: Implement transfer_pid and use it to simplify de_thread

In de_thread we move pids from one process to another, a rather ugly case.
The function transfer_pid makes it clear what we are doing, and makes the
action atomic.  This is useful we ever want to atomically traverse the
process group and session lists, in a rcu safe manner.

Even if the atomic properties this change should be a win as transfer_pid
should be less code to execute than executing both attach_pid and
detach_pid, and this should make de_thread slightly smaller as only a
single function call needs to be emitted.  The only downside is that the
code might be slower to execute as the odds are against transfer_pid being
in cache.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/exec.c           | 11 ++++-------
 include/linux/pid.h |  2 ++
 kernel/pid.c        |  9 +++++++++
 3 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 54135df2a966..b7aa3d6422d6 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -696,23 +696,20 @@ static int de_thread(struct task_struct *tsk)
 		 */
 
 		/* Become a process group leader with the old leader's pid.
-		 * Note: The old leader also uses thispid until release_task
+		 * The old leader becomes a thread of the this thread group.
+		 * Note: The old leader also uses this pid until release_task
 		 *       is called.  Odd but simple and correct.
 		 */
 		detach_pid(current, PIDTYPE_PID);
 		current->pid = leader->pid;
 		attach_pid(current, PIDTYPE_PID,  current->pid);
-		attach_pid(current, PIDTYPE_PGID, current->signal->pgrp);
-		attach_pid(current, PIDTYPE_SID,  current->signal->session);
+		transfer_pid(leader, current, PIDTYPE_PGID);
+		transfer_pid(leader, current, PIDTYPE_SID);
 		list_replace_rcu(&leader->tasks, &current->tasks);
 
 		current->group_leader = current;
 		leader->group_leader = current;
 
-		/* Reduce leader to a thread */
-		detach_pid(leader, PIDTYPE_PGID);
-		detach_pid(leader, PIDTYPE_SID);
-
 		current->exit_signal = SIGCHLD;
 
 		BUG_ON(leader->exit_state != EXIT_ZOMBIE);
diff --git a/include/linux/pid.h b/include/linux/pid.h
index 29960b03bef7..93da7e2d9f30 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -76,6 +76,8 @@ extern int FASTCALL(attach_pid(struct task_struct *task,
 				enum pid_type type, int nr));
 
 extern void FASTCALL(detach_pid(struct task_struct *task, enum pid_type));
+extern void FASTCALL(transfer_pid(struct task_struct *old,
+				  struct task_struct *new, enum pid_type));
 
 /*
  * look up a PID in the hash table. Must be called with the tasklist_lock
diff --git a/kernel/pid.c b/kernel/pid.c
index 93e212f20671..6db82b68e2f8 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -252,6 +252,15 @@ void fastcall detach_pid(struct task_struct *task, enum pid_type type)
 	free_pid(pid);
 }
 
+/* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */
+void fastcall transfer_pid(struct task_struct *old, struct task_struct *new,
+			   enum pid_type type)
+{
+	new->pids[type].pid = old->pids[type].pid;
+	hlist_replace_rcu(&old->pids[type].node, &new->pids[type].node);
+	old->pids[type].pid = NULL;
+}
+
 struct task_struct * fastcall pid_task(struct pid *pid, enum pid_type type)
 {
 	struct task_struct *result = NULL;
-- 
cgit v1.2.1


From 65800ac77e080cf159d6c1207b6886e18f22bc08 Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:11 -0700
Subject: [PATCH] pid: remove temporary debug code in attach_pid

With the patches flying between Oleg and myself somehow this temporary
debug code got left in pid.c.  It was never intended to make it to the
stable kernel.

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/pid.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/kernel/pid.c b/kernel/pid.c
index 6db82b68e2f8..8387e8c68193 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -223,9 +223,6 @@ int fastcall attach_pid(struct task_struct *task, enum pid_type type, int nr)
 	struct pid_link *link;
 	struct pid *pid;
 
-	WARN_ON(!task->pid); /* to be removed soon */
-	WARN_ON(!nr); /* to be removed soon */
-
 	link = &task->pids[type];
 	link->pid = pid = find_pid(nr);
 	hlist_add_head_rcu(&link->node, &pid->tasks[type]);
-- 
cgit v1.2.1


From b7b52630deca785c4fc2bf4c53500cee066786af Mon Sep 17 00:00:00 2001
From: Martin Bligh <mbligh@mbligh.org>
Date: Wed, 27 Sep 2006 01:51:12 -0700
Subject: [PATCH] add newline to nfs dprintk

Add missing \n to dprintk

Signed-off-by: Martin Bligh <mbligh@google.com>
Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/lockd/clntlock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index 52774feab93f..f95cc3f3c42d 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -160,7 +160,7 @@ static void nlmclnt_prepare_reclaim(struct nlm_host *host)
 	 */
 	list_splice_init(&host->h_granted, &host->h_reclaim);
 
-	dprintk("NLM: reclaiming locks for host %s", host->h_name);
+	dprintk("NLM: reclaiming locks for host %s\n", host->h_name);
 }
 
 static void nlmclnt_finish_reclaim(struct nlm_host *host)
-- 
cgit v1.2.1


From 66f37509fc7191df468a8d183374f48b13bacb73 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Wed, 27 Sep 2006 01:51:13 -0700
Subject: [PATCH] fs/nfs/: make code static

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/nfs/namespace.c     | 12 +++++++++---
 include/linux/nfs_fs.h |  4 ----
 2 files changed, 9 insertions(+), 7 deletions(-)

diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c
index 77b00684894d..60408646176b 100644
--- a/fs/nfs/namespace.c
+++ b/fs/nfs/namespace.c
@@ -26,6 +26,11 @@ LIST_HEAD(nfs_automount_list);
 static DECLARE_WORK(nfs_automount_task, nfs_expire_automounts, &nfs_automount_list);
 int nfs_mountpoint_expiry_timeout = 500 * HZ;
 
+static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
+					const struct dentry *dentry,
+					struct nfs_fh *fh,
+					struct nfs_fattr *fattr);
+
 /*
  * nfs_path - reconstruct the path given an arbitrary dentry
  * @base - arbitrary string to prepend to the path
@@ -209,9 +214,10 @@ static struct vfsmount *nfs_do_clone_mount(struct nfs_server *server,
  * @fattr - attributes for new root inode
  *
  */
-struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-		const struct dentry *dentry, struct nfs_fh *fh,
-		struct nfs_fattr *fattr)
+static struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
+					const struct dentry *dentry,
+					struct nfs_fh *fh,
+					struct nfs_fattr *fattr)
 {
 	struct nfs_clone_mount mountdata = {
 		.sb = mnt_parent->mnt_sb,
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 36f5bcf513b0..98c9b9f667a5 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -315,10 +315,6 @@ extern void nfs_end_data_update(struct inode *);
 extern struct nfs_open_context *get_nfs_open_context(struct nfs_open_context *ctx);
 extern void put_nfs_open_context(struct nfs_open_context *ctx);
 extern struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode);
-extern struct vfsmount *nfs_do_submount(const struct vfsmount *mnt_parent,
-					const struct dentry *dentry,
-					struct nfs_fh *fh,
-					struct nfs_fattr *fattr);
 
 /* linux/net/ipv4/ipconfig.c: trims ip addr off front of name, too. */
 extern u32 root_nfs_parse_addr(char *name); /*__init*/
-- 
cgit v1.2.1


From aafe6c2a2b6bce5a3a4913ce5c07e85ea143144d Mon Sep 17 00:00:00 2001
From: "Eric W. Biederman" <ebiederm@xmission.com>
Date: Wed, 27 Sep 2006 01:51:13 -0700
Subject: [PATCH] de_thread: Use tsk not current

Ingo Oeser pointed out that because current expands to an inline function
it is more space efficient and somewhat faster to simply keep a cached copy
of current in another variable.  This patch implements that for the
de_thread function.

(akpm: saves nearly 100 bytes of text on x86)

Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/exec.c | 46 +++++++++++++++++++++++-----------------------
 1 file changed, 23 insertions(+), 23 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index b7aa3d6422d6..97df6e0aeaee 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -595,7 +595,7 @@ static int de_thread(struct task_struct *tsk)
 	if (!newsighand)
 		return -ENOMEM;
 
-	if (thread_group_empty(current))
+	if (thread_group_empty(tsk))
 		goto no_thread_group;
 
 	/*
@@ -620,17 +620,17 @@ static int de_thread(struct task_struct *tsk)
 	 * Reparenting needs write_lock on tasklist_lock,
 	 * so it is safe to do it under read_lock.
 	 */
-	if (unlikely(current->group_leader == child_reaper))
-		child_reaper = current;
+	if (unlikely(tsk->group_leader == child_reaper))
+		child_reaper = tsk;
 
-	zap_other_threads(current);
+	zap_other_threads(tsk);
 	read_unlock(&tasklist_lock);
 
 	/*
 	 * Account for the thread group leader hanging around:
 	 */
 	count = 1;
-	if (!thread_group_leader(current)) {
+	if (!thread_group_leader(tsk)) {
 		count = 2;
 		/*
 		 * The SIGALRM timer survives the exec, but needs to point
@@ -639,14 +639,14 @@ static int de_thread(struct task_struct *tsk)
 		 * synchronize with any firing (by calling del_timer_sync)
 		 * before we can safely let the old group leader die.
 		 */
-		sig->tsk = current;
+		sig->tsk = tsk;
 		spin_unlock_irq(lock);
 		if (hrtimer_cancel(&sig->real_timer))
 			hrtimer_restart(&sig->real_timer);
 		spin_lock_irq(lock);
 	}
 	while (atomic_read(&sig->count) > count) {
-		sig->group_exit_task = current;
+		sig->group_exit_task = tsk;
 		sig->notify_count = count;
 		__set_current_state(TASK_UNINTERRUPTIBLE);
 		spin_unlock_irq(lock);
@@ -662,13 +662,13 @@ static int de_thread(struct task_struct *tsk)
 	 * do is to wait for the thread group leader to become inactive,
 	 * and to assume its PID:
 	 */
-	if (!thread_group_leader(current)) {
+	if (!thread_group_leader(tsk)) {
 		/*
 		 * Wait for the thread group leader to be a zombie.
 		 * It should already be zombie at this point, most
 		 * of the time.
 		 */
-		leader = current->group_leader;
+		leader = tsk->group_leader;
 		while (leader->exit_state != EXIT_ZOMBIE)
 			yield();
 
@@ -682,12 +682,12 @@ static int de_thread(struct task_struct *tsk)
 		 * When we take on its identity by switching to its PID, we
 		 * also take its birthdate (always earlier than our own).
 		 */
-		current->start_time = leader->start_time;
+		tsk->start_time = leader->start_time;
 
 		write_lock_irq(&tasklist_lock);
 
-		BUG_ON(leader->tgid != current->tgid);
-		BUG_ON(current->pid == current->tgid);
+		BUG_ON(leader->tgid != tsk->tgid);
+		BUG_ON(tsk->pid == tsk->tgid);
 		/*
 		 * An exec() starts a new thread group with the
 		 * TGID of the previous thread group. Rehash the
@@ -700,17 +700,17 @@ static int de_thread(struct task_struct *tsk)
 		 * Note: The old leader also uses this pid until release_task
 		 *       is called.  Odd but simple and correct.
 		 */
-		detach_pid(current, PIDTYPE_PID);
-		current->pid = leader->pid;
-		attach_pid(current, PIDTYPE_PID,  current->pid);
-		transfer_pid(leader, current, PIDTYPE_PGID);
-		transfer_pid(leader, current, PIDTYPE_SID);
-		list_replace_rcu(&leader->tasks, &current->tasks);
+		detach_pid(tsk, PIDTYPE_PID);
+		tsk->pid = leader->pid;
+		attach_pid(tsk, PIDTYPE_PID,  tsk->pid);
+		transfer_pid(leader, tsk, PIDTYPE_PGID);
+		transfer_pid(leader, tsk, PIDTYPE_SID);
+		list_replace_rcu(&leader->tasks, &tsk->tasks);
 
-		current->group_leader = current;
-		leader->group_leader = current;
+		tsk->group_leader = tsk;
+		leader->group_leader = tsk;
 
-		current->exit_signal = SIGCHLD;
+		tsk->exit_signal = SIGCHLD;
 
 		BUG_ON(leader->exit_state != EXIT_ZOMBIE);
 		leader->exit_state = EXIT_DEAD;
@@ -750,7 +750,7 @@ no_thread_group:
 		spin_lock(&oldsighand->siglock);
 		spin_lock_nested(&newsighand->siglock, SINGLE_DEPTH_NESTING);
 
-		rcu_assign_pointer(current->sighand, newsighand);
+		rcu_assign_pointer(tsk->sighand, newsighand);
 		recalc_sigpending();
 
 		spin_unlock(&newsighand->siglock);
@@ -761,7 +761,7 @@ no_thread_group:
 			kmem_cache_free(sighand_cachep, oldsighand);
 	}
 
-	BUG_ON(!thread_group_leader(current));
+	BUG_ON(!thread_group_leader(tsk));
 	return 0;
 }
 	
-- 
cgit v1.2.1


From 1b79e5513d52e8533a08af35a3595dad80c74d1f Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:51:14 -0700
Subject: [PATCH] add probe_kernel_address()

Add a version of __get_user() which is safe to call inside mmap_sem.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/uaccess.h | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
index 391e7ed1eb3f..a48d7f11c7be 100644
--- a/include/linux/uaccess.h
+++ b/include/linux/uaccess.h
@@ -19,4 +19,26 @@ static inline unsigned long __copy_from_user_nocache(void *to,
 
 #endif		/* ARCH_HAS_NOCACHE_UACCESS */
 
+/**
+ * probe_kernel_address(): safely attempt to read from a location
+ * @addr: address to read from - its type is type typeof(retval)*
+ * @retval: read into this variable
+ *
+ * Safely read from address @addr into variable @revtal.  If a kernel fault
+ * happens, handle that and return -EFAULT.
+ * We ensure that the __get_user() is executed in atomic context so that
+ * do_page_fault() doesn't attempt to take mmap_sem.  This makes
+ * probe_kernel_address() suitable for use within regions where the caller
+ * already holds mmap_sem, or other locks which nest inside mmap_sem.
+ */
+#define probe_kernel_address(addr, retval)		\
+	({						\
+		long ret;				\
+							\
+		inc_preempt_count();			\
+		ret = __get_user(retval, addr);		\
+		dec_preempt_count();			\
+		ret;					\
+	})
+
 #endif		/* __LINUX_UACCESS_H__ */
-- 
cgit v1.2.1


From 1e2af92e089d4c845248844a3e7ee06ce889af46 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 01:51:15 -0700
Subject: [PATCH] x86: use probe_kernel_address in handle_BUG()

Avoid possible deadlock on a BUG() inside down_write(mmap_sem).  The deadlock
can only occur if something has gone horridly wrong, because a fault here
shouldn't happen.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/traps.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 21aa1cd57773..a13037fe0ee3 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -28,6 +28,7 @@
 #include <linux/kprobes.h>
 #include <linux/kexec.h>
 #include <linux/unwind.h>
+#include <linux/uaccess.h>
 
 #ifdef CONFIG_EISA
 #include <linux/ioport.h>
@@ -40,7 +41,6 @@
 
 #include <asm/processor.h>
 #include <asm/system.h>
-#include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/atomic.h>
 #include <asm/debugreg.h>
@@ -409,7 +409,7 @@ static void handle_BUG(struct pt_regs *regs)
 
 	if (eip < PAGE_OFFSET)
 		return;
-	if (__get_user(ud2, (unsigned short __user *)eip))
+	if (probe_kernel_address((unsigned short __user *)eip, ud2))
 		return;
 	if (ud2 != 0x0b0f)
 		return;
@@ -422,7 +422,8 @@ static void handle_BUG(struct pt_regs *regs)
 		char *file;
 		char c;
 
-		if (__get_user(line, (unsigned short __user *)(eip + 2)))
+		if (probe_kernel_address((unsigned short __user *)(eip + 2),
+					line))
 			break;
 		if (__get_user(file, (char * __user *)(eip + 4)) ||
 		    (unsigned long)file < PAGE_OFFSET || __get_user(c, file))
-- 
cgit v1.2.1


From b97b196c9d351a501ed89fc836e4e2fe71ff93c9 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Thu, 14 Sep 2006 15:18:54 -0400
Subject: USB: unusual_devs entry for Lacie DVD+-RW

This patch (as781) adds an entry to unusual_devs.h for the Lacie DVD+-RW
drive.  Apparently its USB interface has requirements similar to the
Genesys Logic interface; it doesn't like data to be sent too soon after
a command.

This fixes Bugzilla #6817.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index b130e170b4a8..4b06ac34430f 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -631,6 +631,13 @@ UNUSUAL_DEV(  0x0595, 0x4343, 0x0000, 0x2210,
 		"Digital Camera EX-20 DSC",
 		US_SC_8070, US_PR_DEVICE, NULL, 0 ),
 
+/* Reported by <Hendryk.Pfeiffer@gmx.de> */
+UNUSUAL_DEV(  0x059f, 0x0643, 0x0000, 0x0000,
+		"LaCie",
+		"DVD+-RW",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_GO_SLOW ),
+
 /* Submitted by Joel Bourquard <numlock@freesurf.ch>
  * Some versions of this device need the SubClass and Protocol overrides
  * while others don't.
-- 
cgit v1.2.1


From 39c2f3ac04663bb2546a7058701a621e526eabb2 Mon Sep 17 00:00:00 2001
From: Phil Dibowitz <phil@ipom.com>
Date: Mon, 11 Sep 2006 00:27:40 -0700
Subject: USB: unusual_dev entry for Sony P990i

This patch is a re-diffed version of one originally sent by Jan Mate
<mate@fiit.stuba.sk>.

Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 4b06ac34430f..fa49357289c4 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1261,6 +1261,13 @@ UNUSUAL_DEV(  0x0fce, 0xd008, 0x0000, 0x0000,
 		US_SC_DEVICE, US_PR_DEVICE, NULL,
 		US_FL_NO_WP_DETECT ),
 
+/* Reported by Jan Mate <mate@fiit.stuba.sk> */
+UNUSUAL_DEV(  0x0fce, 0xe030, 0x0000, 0x0000,
+		"Sony Ericsson",
+		"P990i",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_FIX_CAPACITY ),
+
 /* Reported by Emmanuel Vasilakis <evas@forthnet.gr> */
 UNUSUAL_DEV(  0x0fce, 0xe031, 0x0000, 0x0000,
 		"Sony Ericsson",
-- 
cgit v1.2.1


From 341323269258ab8484077a9c70db804fbae6ae93 Mon Sep 17 00:00:00 2001
From: Sam Bishop <sam@bishop.dhs.org>
Date: Mon, 28 Aug 2006 16:42:10 -0600
Subject: USB: doc patch 1

Grammar, spelling, and stylistic edits.

Signed-off-by: Sam Bishop <sam@bishop.dhs.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/DocBook/usb.tmpl | 98 +++++++++++++++++++-----------------------
 1 file changed, 44 insertions(+), 54 deletions(-)

diff --git a/Documentation/DocBook/usb.tmpl b/Documentation/DocBook/usb.tmpl
index 320af25de3a2..8a28f76b9359 100644
--- a/Documentation/DocBook/usb.tmpl
+++ b/Documentation/DocBook/usb.tmpl
@@ -43,59 +43,52 @@
 
     <para>A Universal Serial Bus (USB) is used to connect a host,
     such as a PC or workstation, to a number of peripheral
-    devices.  USB uses a tree structure, with the host at the
+    devices.  USB uses a tree structure, with the host as the
     root (the system's master), hubs as interior nodes, and
-    peripheral devices as leaves (and slaves).
+    peripherals as leaves (and slaves).
     Modern PCs support several such trees of USB devices, usually
     one USB 2.0 tree (480 Mbit/sec each) with
     a few USB 1.1 trees (12 Mbit/sec each) that are used when you
     connect a USB 1.1 device directly to the machine's "root hub".
     </para>
 
-    <para>That master/slave asymmetry was designed in part for
-    ease of use.  It is not physically possible to assemble
-    (legal) USB cables incorrectly:  all upstream "to-the-host"
-    connectors are the rectangular type, matching the sockets on
-    root hubs, and the downstream type are the squarish type
-    (or they are built in to the peripheral).
-    Software doesn't need to deal with distributed autoconfiguration
-    since the pre-designated master node manages all that.
-    At the electrical level, bus protocol overhead is reduced by
-    eliminating arbitration and moving scheduling into host software.
+    <para>That master/slave asymmetry was designed-in for a number of
+    reasons, one being ease of use.  It is not physically possible to
+    assemble (legal) USB cables incorrectly:  all upstream "to the host"
+    connectors are the rectangular type (matching the sockets on
+    root hubs), and all downstream connectors are the squarish type
+    (or they are built into the peripheral).
+    Also, the host software doesn't need to deal with distributed
+    auto-configuration since the pre-designated master node manages all that.
+    And finally, at the electrical level, bus protocol overhead is reduced by
+    eliminating arbitration and moving scheduling into the host software.
     </para>
 
-    <para>USB 1.0 was announced in January 1996, and was revised
+    <para>USB 1.0 was announced in January 1996 and was revised
     as USB 1.1 (with improvements in hub specification and
     support for interrupt-out transfers) in September 1998.
-    USB 2.0 was released in April 2000, including high speed
-    transfers and transaction translating hubs (used for USB 1.1
+    USB 2.0 was released in April 2000, adding high-speed
+    transfers and transaction-translating hubs (used for USB 1.1
     and 1.0 backward compatibility).
     </para>
 
-    <para>USB support was added to Linux early in the 2.2 kernel series
-    shortly before the 2.3 development forked off.  Updates
-    from 2.3 were regularly folded back into 2.2 releases, bringing
-    new features such as <filename>/sbin/hotplug</filename> support,
-    more drivers, and more robustness.
-    The 2.5 kernel series continued such improvements, and also
-    worked on USB 2.0 support,
-    higher performance,
-    better consistency between host controller drivers,
-    API simplification (to make bugs less likely),
-    and providing internal "kerneldoc" documentation.
+    <para>Kernel developers added USB support to Linux early in the 2.2 kernel
+    series, shortly before 2.3 development forked.  Updates from 2.3 were
+    regularly folded back into 2.2 releases, which improved reliability and
+    brought <filename>/sbin/hotplug</filename> support as well more drivers.
+    Such improvements were continued in the 2.5 kernel series, where they added
+    USB 2.0 support, improved performance, and made the host controller drivers
+    (HCDs) more consistent.  They also simplified the API (to make bugs less
+    likely) and added internal "kerneldoc" documentation.
     </para>
 
     <para>Linux can run inside USB devices as well as on
     the hosts that control the devices.
-    Because the Linux 2.x USB support evolved to support mass market
-    platforms such as Apple Macintosh or PC-compatible systems,
-    it didn't address design concerns for those types of USB systems.
-    So it can't be used inside mass-market PDAs, or other peripherals.
-    USB device drivers running inside those Linux peripherals
+    But USB device drivers running inside those peripherals
     don't do the same things as the ones running inside hosts,
-    and so they've been given a different name:
-    they're called <emphasis>gadget drivers</emphasis>.
-    This document does not present gadget drivers.
+    so they've been given a different name:
+    <emphasis>gadget drivers</emphasis>.
+    This document does not cover gadget drivers.
     </para>
 
     </chapter>
@@ -103,17 +96,14 @@
 <chapter id="host">
     <title>USB Host-Side API Model</title>
 
-    <para>Within the kernel,
-    host-side drivers for USB devices talk to the "usbcore" APIs.
-    There are two types of public "usbcore" APIs, targetted at two different
-    layers of USB driver.  Those are
-    <emphasis>general purpose</emphasis> drivers, exposed through
-    driver frameworks such as block, character, or network devices;
-    and drivers that are <emphasis>part of the core</emphasis>,
-    which are involved in managing a USB bus.
-    Such core drivers include the <emphasis>hub</emphasis> driver,
-    which manages trees of USB devices, and several different kinds
-    of <emphasis>host controller driver (HCD)</emphasis>,
+    <para>Host-side drivers for USB devices talk to the "usbcore" APIs.
+    There are two.  One is intended for
+    <emphasis>general-purpose</emphasis> drivers (exposed through
+    driver frameworks), and the other is for drivers that are
+    <emphasis>part of the core</emphasis>.
+    Such core drivers include the <emphasis>hub</emphasis> driver
+    (which manages trees of USB devices) and several different kinds
+    of <emphasis>host controller drivers</emphasis>,
     which control individual busses.
     </para>
 
@@ -122,21 +112,21 @@
      
     <itemizedlist>
 
-	<listitem><para>USB supports four kinds of data transfer
-	(control, bulk, interrupt, and isochronous).  Two transfer
-	types use bandwidth as it's available (control and bulk),
-	while the other two types of transfer (interrupt and isochronous)
+	<listitem><para>USB supports four kinds of data transfers
+	(control, bulk, interrupt, and isochronous).  Two of them (control
+	and bulk) use bandwidth as it's available,
+	while the other two (interrupt and isochronous)
 	are scheduled to provide guaranteed bandwidth.
 	</para></listitem>
 
 	<listitem><para>The device description model includes one or more
 	"configurations" per device, only one of which is active at a time.
-	Devices that are capable of high speed operation must also support
-	full speed configurations, along with a way to ask about the
-	"other speed" configurations that might be used.
+	Devices that are capable of high-speed operation must also support
+	full-speed configurations, along with a way to ask about the
+	"other speed" configurations which might be used.
 	</para></listitem>
 
-	<listitem><para>Configurations have one or more "interface", each
+	<listitem><para>Configurations have one or more "interfaces", each
 	of which may have "alternate settings".  Interfaces may be
 	standardized by USB "Class" specifications, or may be specific to
 	a vendor or device.</para>
@@ -162,7 +152,7 @@
 	</para></listitem>
 
 	<listitem><para>The Linux USB API supports synchronous calls for
-	control and bulk messaging.
+	control and bulk messages.
 	It also supports asynchnous calls for all kinds of data transfer,
 	using request structures called "URBs" (USB Request Blocks).
 	</para></listitem>
-- 
cgit v1.2.1


From 06afff00bcab0e384afbef70194fd3469532abdf Mon Sep 17 00:00:00 2001
From: Sam Bishop <sam@bishop.dhs.org>
Date: Mon, 28 Aug 2006 16:52:15 -0600
Subject: USB doc patch 2

A little more detail on how and when to poll() /proc/bus/usb/devices.

Signed-off-by: Sam Bishop <sam@bishop.dhs.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/DocBook/usb.tmpl | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/Documentation/DocBook/usb.tmpl b/Documentation/DocBook/usb.tmpl
index 8a28f76b9359..3608472d7b74 100644
--- a/Documentation/DocBook/usb.tmpl
+++ b/Documentation/DocBook/usb.tmpl
@@ -453,14 +453,25 @@
 	    file in your Linux kernel sources.
 	    </para>
 
-	    <para>Otherwise the main use for this file from programs
-	    is to poll() it to get notifications of usb devices
-	    as they're plugged or unplugged.
-	    To see what changed, you'd need to read the file and
-	    compare "before" and "after" contents, scan the filesystem,
-	    or see its hotplug event.
+	    <para>This file, in combination with the poll() system call, can
+	    also be used to detect when devices are added or removed:
+<programlisting>int fd;
+struct pollfd pfd;
+
+fd = open("/proc/bus/usb/devices", O_RDONLY);
+pfd = { fd, POLLIN, 0 };
+for (;;) {
+	/* The first time through, this call will return immediately. */
+	poll(&amp;pfd, 1, -1);
+
+	/* To see what's changed, compare the file's previous and current
+	   contents or scan the filesystem.  (Scanning is more precise.) */
+}</programlisting>
+	    Note that this behavior is intended to be used for informational
+	    and debug purposes.  It would be more appropriate to use programs
+	    such as udev or HAL to initialize a device or start a user-mode
+	    helper program, for instance.
 	    </para>
-
 	</sect1>
 
 	<sect1>
-- 
cgit v1.2.1


From d413984ae936fad46678403b38d79c595e5aaafe Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Fri, 4 Aug 2006 11:31:55 -0700
Subject: USB: OHCI avoids root hub timer polling

This teaches OHCI to use the root hub status change (RHSC) IRQ, bypassing
root hub timers most of the time and switching over to the "new" root hub
polling scheme.  It's complicated by the fact that implementations of OHCI
trigger and ack that IRQ differently (the spec is vague there).

Avoiding root hub timers helps mechanisms like "dynamic tick" leave the
CPU in lowpower modes for longer intervals.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ohci-at91.c    |  2 +-
 drivers/usb/host/ohci-au1xxx.c  |  5 +--
 drivers/usb/host/ohci-dbg.c     |  5 +++
 drivers/usb/host/ohci-ep93xx.c  |  1 +
 drivers/usb/host/ohci-hcd.c     | 41 +++++++++++++++++++------
 drivers/usb/host/ohci-hub.c     | 67 ++++++++++++++++++++++++++++-------------
 drivers/usb/host/ohci-lh7a404.c |  5 +--
 drivers/usb/host/ohci-omap.c    |  5 ++-
 drivers/usb/host/ohci-pci.c     |  5 ++-
 drivers/usb/host/ohci-ppc-soc.c |  1 +
 drivers/usb/host/ohci-pxa27x.c  |  1 +
 drivers/usb/host/ohci-s3c2410.c |  1 +
 drivers/usb/host/ohci-sa1111.c  |  5 +--
 13 files changed, 99 insertions(+), 45 deletions(-)

diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
index 85cc059705a6..33b75087bc0c 100644
--- a/drivers/usb/host/ohci-at91.c
+++ b/drivers/usb/host/ohci-at91.c
@@ -239,7 +239,7 @@ static const struct hc_driver ohci_at91_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
-
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-au1xxx.c b/drivers/usb/host/ohci-au1xxx.c
index f7a975d5db09..44ed3a4c01ef 100644
--- a/drivers/usb/host/ohci-au1xxx.c
+++ b/drivers/usb/host/ohci-au1xxx.c
@@ -268,10 +268,6 @@ static const struct hc_driver ohci_au1xxx_hc_driver = {
 	 * basic lifecycle operations
 	 */
 	.start =		ohci_au1xxx_start,
-#ifdef	CONFIG_PM
-	/* suspend:		ohci_au1xxx_suspend,  -- tbd */
-	/* resume:		ohci_au1xxx_resume,   -- tbd */
-#endif /*CONFIG_PM*/
 	.stop =			ohci_stop,
 
 	/*
@@ -291,6 +287,7 @@ static const struct hc_driver ohci_au1xxx_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c
index 7bfffcbbd226..da52609a9290 100644
--- a/drivers/usb/host/ohci-dbg.c
+++ b/drivers/usb/host/ohci-dbg.c
@@ -667,6 +667,11 @@ show_registers (struct class_device *class_dev, char *buf)
 	size -= temp;
 	next += temp;
 
+	temp = scnprintf (next, size, "hub poll timer %s\n",
+			ohci_to_hcd(ohci)->poll_rh ? "ON" : "off");
+	size -= temp;
+	next += temp;
+
 	/* roothub */
 	ohci_dump_roothub (ohci, 1, &next, &size);
 
diff --git a/drivers/usb/host/ohci-ep93xx.c b/drivers/usb/host/ohci-ep93xx.c
index 6531c4d26527..1a1d320b7995 100644
--- a/drivers/usb/host/ohci-ep93xx.c
+++ b/drivers/usb/host/ohci-ep93xx.c
@@ -134,6 +134,7 @@ static struct hc_driver ohci_ep93xx_hc_driver = {
 	.get_frame_number	= ohci_get_frame,
 	.hub_status_data	= ohci_hub_status_data,
 	.hub_control		= ohci_hub_control,
+	.hub_irq_enable		= ohci_rhsc_enable,
 #ifdef CONFIG_PM
 	.bus_suspend		= ohci_bus_suspend,
 	.bus_resume		= ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 94d8cf4b36c1..0684f57c14a0 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -101,7 +101,7 @@
 
 #include "../core/hcd.h"
 
-#define DRIVER_VERSION "2005 April 22"
+#define DRIVER_VERSION "2006 August 04"
 #define DRIVER_AUTHOR "Roman Weissgaerber, David Brownell"
 #define DRIVER_DESC "USB 1.1 'Open' Host Controller (OHCI) Driver"
 
@@ -110,9 +110,10 @@
 #undef OHCI_VERBOSE_DEBUG	/* not always helpful */
 
 /* For initializing controller (mask in an HCFS mode too) */
-#define	OHCI_CONTROL_INIT 	OHCI_CTRL_CBSR
+#define	OHCI_CONTROL_INIT	OHCI_CTRL_CBSR
 #define	OHCI_INTR_INIT \
-	(OHCI_INTR_MIE | OHCI_INTR_UE | OHCI_INTR_RD | OHCI_INTR_WDH)
+		(OHCI_INTR_MIE | OHCI_INTR_RHSC | OHCI_INTR_UE \
+		| OHCI_INTR_RD | OHCI_INTR_WDH)
 
 #ifdef __hppa__
 /* On PA-RISC, PDC can leave IR set incorrectly; ignore it there. */
@@ -128,6 +129,8 @@
 
 static const char	hcd_name [] = "ohci_hcd";
 
+#define	STATECHANGE_DELAY	msecs_to_jiffies(300)
+
 #include "ohci.h"
 
 static void ohci_dump (struct ohci_hcd *ohci, int verbose);
@@ -446,7 +449,6 @@ static int ohci_init (struct ohci_hcd *ohci)
 
 	disable (ohci);
 	ohci->regs = hcd->regs;
-	ohci->next_statechange = jiffies;
 
 	/* REVISIT this BIOS handshake is now moved into PCI "quirks", and
 	 * was never needed for most non-PCI systems ... remove the code?
@@ -637,10 +639,14 @@ retry:
 		return -EOVERFLOW;
 	}
 
- 	/* start controller operations */
+	/* use rhsc irqs after khubd is fully initialized */
+	hcd->poll_rh = 1;
+	hcd->uses_new_polling = 1;
+
+	/* start controller operations */
 	ohci->hc_control &= OHCI_CTRL_RWC;
- 	ohci->hc_control |= OHCI_CONTROL_INIT | OHCI_USB_OPER;
- 	ohci_writel (ohci, ohci->hc_control, &ohci->regs->control);
+	ohci->hc_control |= OHCI_CONTROL_INIT | OHCI_USB_OPER;
+	ohci_writel (ohci, ohci->hc_control, &ohci->regs->control);
 	hcd->state = HC_STATE_RUNNING;
 
 	/* wake on ConnectStatusChange, matching external hubs */
@@ -648,7 +654,7 @@ retry:
 
 	/* Choose the interrupts we care about now, others later on demand */
 	mask = OHCI_INTR_INIT;
-	ohci_writel (ohci, mask, &ohci->regs->intrstatus);
+	ohci_writel (ohci, ~0, &ohci->regs->intrstatus);
 	ohci_writel (ohci, mask, &ohci->regs->intrenable);
 
 	/* handle root hub init quirks ... */
@@ -672,6 +678,7 @@ retry:
 	// flush those writes
 	(void) ohci_readl (ohci, &ohci->regs->control);
 
+	ohci->next_statechange = jiffies + STATECHANGE_DELAY;
 	spin_unlock_irq (&ohci->lock);
 
 	// POTPGT delay is bits 24-31, in 2 ms units.
@@ -709,7 +716,23 @@ static irqreturn_t ohci_irq (struct usb_hcd *hcd, struct pt_regs *ptregs)
 	/* interrupt for some other device? */
 	} else if ((ints &= ohci_readl (ohci, &regs->intrenable)) == 0) {
 		return IRQ_NOTMINE;
-	} 
+	}
+
+	/* NOTE:  vendors didn't always make the same implementation
+	 * choices for RHSC.  Sometimes it triggers on an edge (like
+	 * setting and maybe clearing a port status change bit); and
+	 * it's level-triggered on other silicon, active until khubd
+	 * clears all active port status change bits.  Poll by timer
+	 * til it's fully debounced and the difference won't matter.
+	 */
+	if (ints & OHCI_INTR_RHSC) {
+		ohci_vdbg (ohci, "rhsc\n");
+		ohci_writel (ohci, OHCI_INTR_RHSC, &regs->intrdisable);
+		hcd->poll_rh = 1;
+		ohci->next_statechange = jiffies + STATECHANGE_DELAY;
+		ohci_writel (ohci, OHCI_INTR_RHSC, &regs->intrstatus);
+		usb_hcd_poll_rh_status(hcd);
+	}
 
 	if (ints & OHCI_INTR_UE) {
 		disable (ohci);
diff --git a/drivers/usb/host/ohci-hub.c b/drivers/usb/host/ohci-hub.c
index 5b0a23fd798b..f1b1ed086bde 100644
--- a/drivers/usb/host/ohci-hub.c
+++ b/drivers/usb/host/ohci-hub.c
@@ -36,6 +36,14 @@
 
 /*-------------------------------------------------------------------------*/
 
+/* hcd->hub_irq_enable() */
+static void ohci_rhsc_enable (struct usb_hcd *hcd)
+{
+	struct ohci_hcd		*ohci = hcd_to_ohci (hcd);
+
+	ohci_writel (ohci, OHCI_INTR_RHSC, &ohci->regs->intrenable);
+}
+
 #ifdef	CONFIG_PM
 
 #define OHCI_SCHED_ENABLES \
@@ -123,6 +131,9 @@ static int ohci_bus_suspend (struct usb_hcd *hcd)
 	/* no resumes until devices finish suspending */
 	ohci->next_statechange = jiffies + msecs_to_jiffies (5);
 
+	/* no timer polling */
+	hcd->poll_rh = 0;
+
 done:
 	/* external suspend vs self autosuspend ... same effect */
 	if (status == 0)
@@ -256,8 +267,8 @@ static int ohci_bus_resume (struct usb_hcd *hcd)
 	/* TRSMRCY */
 	msleep (10);
 
-	/* keep it alive for ~5x suspend + resume costs */
-	ohci->next_statechange = jiffies + msecs_to_jiffies (250);
+	/* keep it alive for more than ~5x suspend + resume costs */
+	ohci->next_statechange = jiffies + STATECHANGE_DELAY;
 
 	/* maybe turn schedules back on */
 	enables = 0;
@@ -302,9 +313,10 @@ ohci_hub_status_data (struct usb_hcd *hcd, char *buf)
 {
 	struct ohci_hcd	*ohci = hcd_to_ohci (hcd);
 	int		i, changed = 0, length = 1;
-	int		can_suspend = device_may_wakeup(&hcd->self.root_hub->dev);
+	int		can_suspend;
 	unsigned long	flags;
 
+	can_suspend = device_may_wakeup(&hcd->self.root_hub->dev);
 	spin_lock_irqsave (&ohci->lock, flags);
 
 	/* handle autosuspended root:  finish resuming before
@@ -339,6 +351,10 @@ ohci_hub_status_data (struct usb_hcd *hcd, char *buf)
 	for (i = 0; i < ohci->num_ports; i++) {
 		u32	status = roothub_portstatus (ohci, i);
 
+		/* can't autosuspend with active ports */
+		if ((status & RH_PS_PES) && !(status & RH_PS_PSS))
+			can_suspend = 0;
+
 		if (status & (RH_PS_CSC | RH_PS_PESC | RH_PS_PSSC
 				| RH_PS_OCIC | RH_PS_PRSC)) {
 			changed = 1;
@@ -348,32 +364,41 @@ ohci_hub_status_data (struct usb_hcd *hcd, char *buf)
 			    buf [1] |= 1 << (i - 7);
 			continue;
 		}
+	}
 
-		/* can suspend if no ports are enabled; or if all all
-		 * enabled ports are suspended AND remote wakeup is on.
-		 */
-		if (!(status & RH_PS_CCS))
-			continue;
-		if ((status & RH_PS_PSS) && can_suspend)
-			continue;
+	/* after root hub changes, stop polling after debouncing
+	 * for a while and maybe kicking in autosuspend
+	 */
+	if (changed) {
+		ohci->next_statechange = jiffies + STATECHANGE_DELAY;
 		can_suspend = 0;
+	} else if (time_before (jiffies, ohci->next_statechange)) {
+		can_suspend = 0;
+	} else {
+#ifdef	CONFIG_PM
+		can_suspend = can_suspend
+			&& !ohci->ed_rm_list
+			&& ((OHCI_CTRL_HCFS | OHCI_SCHED_ENABLES)
+					& ohci->hc_control)
+				== OHCI_USB_OPER;
+#endif
+		if (hcd->uses_new_polling) {
+			hcd->poll_rh = 0;
+			/* use INTR_RHSC iff INTR_RD won't apply */
+			if (!can_suspend)
+				ohci_writel (ohci, OHCI_INTR_RHSC,
+						&ohci->regs->intrenable);
+		}
 	}
+
 done:
 	spin_unlock_irqrestore (&ohci->lock, flags);
 
-#ifdef CONFIG_PM
-	/* save power by suspending idle root hubs;
+#ifdef	CONFIG_PM
+	/* save power by autosuspending idle root hubs;
 	 * INTR_RD wakes us when there's work
 	 */
-	if (can_suspend
-			&& !changed
-			&& !ohci->ed_rm_list
-			&& ((OHCI_CTRL_HCFS | OHCI_SCHED_ENABLES)
-					& ohci->hc_control)
-				== OHCI_USB_OPER
-			&& time_after (jiffies, ohci->next_statechange)
-			&& usb_trylock_device (hcd->self.root_hub) == 0
-			) {
+	if (can_suspend && usb_trylock_device (hcd->self.root_hub) == 0) {
 		ohci_vdbg (ohci, "autosuspend\n");
 		(void) ohci_bus_suspend (hcd);
 		usb_unlock_device (hcd->self.root_hub);
diff --git a/drivers/usb/host/ohci-lh7a404.c b/drivers/usb/host/ohci-lh7a404.c
index 5602da9bd52c..f2c9161d6d6a 100644
--- a/drivers/usb/host/ohci-lh7a404.c
+++ b/drivers/usb/host/ohci-lh7a404.c
@@ -173,10 +173,6 @@ static const struct hc_driver ohci_lh7a404_hc_driver = {
 	 * basic lifecycle operations
 	 */
 	.start =		ohci_lh7a404_start,
-#ifdef	CONFIG_PM
-	/* suspend:		ohci_lh7a404_suspend,  -- tbd */
-	/* resume:		ohci_lh7a404_resume,   -- tbd */
-#endif /*CONFIG_PM*/
 	.stop =			ohci_stop,
 
 	/*
@@ -196,6 +192,7 @@ static const struct hc_driver ohci_lh7a404_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-omap.c b/drivers/usb/host/ohci-omap.c
index c4c4babd4767..47f1c9bbef87 100644
--- a/drivers/usb/host/ohci-omap.c
+++ b/drivers/usb/host/ohci-omap.c
@@ -382,8 +382,10 @@ ohci_omap_start (struct usb_hcd *hcd)
 	int		ret;
 
 	config = hcd->self.controller->platform_data;
-	if (config->otg || config->rwc)
+	if (config->otg || config->rwc) {
+		ohci->hc_control = OHCI_CTRL_RWC;
 		writel(OHCI_CTRL_RWC, &ohci->regs->control);
+	}
 
 	if ((ret = ohci_run (ohci)) < 0) {
 		dev_err(hcd->self.controller, "can't start\n");
@@ -429,6 +431,7 @@ static const struct hc_driver ohci_omap_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index 37e122812b67..ef874443aa9f 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -176,11 +176,13 @@ static const struct hc_driver ohci_pci_hc_driver = {
 	 */
 	.reset =		ohci_pci_reset,
 	.start =		ohci_pci_start,
+	.stop =			ohci_stop,
+
 #ifdef	CONFIG_PM
+	/* these suspend/resume entries are for upstream PCI glue ONLY */
 	.suspend =		ohci_pci_suspend,
 	.resume =		ohci_pci_resume,
 #endif
-	.stop =			ohci_stop,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -199,6 +201,7 @@ static const struct hc_driver ohci_pci_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-ppc-soc.c b/drivers/usb/host/ohci-ppc-soc.c
index 9fe56ff1615d..270aaaad8c6d 100644
--- a/drivers/usb/host/ohci-ppc-soc.c
+++ b/drivers/usb/host/ohci-ppc-soc.c
@@ -166,6 +166,7 @@ static const struct hc_driver ohci_ppc_soc_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-pxa27x.c b/drivers/usb/host/ohci-pxa27x.c
index 6f559e102789..2752d36c2a78 100644
--- a/drivers/usb/host/ohci-pxa27x.c
+++ b/drivers/usb/host/ohci-pxa27x.c
@@ -288,6 +288,7 @@ static const struct hc_driver ohci_pxa27x_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef  CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-s3c2410.c b/drivers/usb/host/ohci-s3c2410.c
index d2fc6969a9f7..c4c77d26ca1f 100644
--- a/drivers/usb/host/ohci-s3c2410.c
+++ b/drivers/usb/host/ohci-s3c2410.c
@@ -465,6 +465,7 @@ static const struct hc_driver ohci_s3c2410_hc_driver = {
 	 */
 	.hub_status_data =	ohci_s3c2410_hub_status_data,
 	.hub_control =		ohci_s3c2410_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
diff --git a/drivers/usb/host/ohci-sa1111.c b/drivers/usb/host/ohci-sa1111.c
index ce3de106cadc..71371de32ada 100644
--- a/drivers/usb/host/ohci-sa1111.c
+++ b/drivers/usb/host/ohci-sa1111.c
@@ -212,10 +212,6 @@ static const struct hc_driver ohci_sa1111_hc_driver = {
 	 * basic lifecycle operations
 	 */
 	.start =		ohci_sa1111_start,
-#ifdef	CONFIG_PM
-	/* suspend:		ohci_sa1111_suspend,  -- tbd */
-	/* resume:		ohci_sa1111_resume,   -- tbd */
-#endif
 	.stop =			ohci_stop,
 
 	/*
@@ -235,6 +231,7 @@ static const struct hc_driver ohci_sa1111_hc_driver = {
 	 */
 	.hub_status_data =	ohci_hub_status_data,
 	.hub_control =		ohci_hub_control,
+	.hub_irq_enable =	ohci_rhsc_enable,
 #ifdef	CONFIG_PM
 	.bus_suspend =		ohci_bus_suspend,
 	.bus_resume =		ohci_bus_resume,
-- 
cgit v1.2.1


From b2a8e097d0f3bbb7ef550103022db92fc3832842 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 26 Jun 2006 23:36:07 +0100
Subject: USB: ohci-s3c2410.c: clock now usb-bus-host

With the newer Samsung S3C2412 and S3C2413 SoC devices,
the 48MHz USB clock has been given an individual gate
into the USB OHCI and gadget blocks.

This clock is called usb-bus-clock, and we need to
replace the old use of the USB PLL (upll) directly
with the new usb-bus-host.

The S3C2410 clock driver has been updated already to
provide a virtual clock which is a child of the UPLL
to maintain compatibility. The S3C2412 clock driver
correctly enables the PLL when either usb-bus clock
is active.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ohci-s3c2410.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/host/ohci-s3c2410.c b/drivers/usb/host/ohci-s3c2410.c
index c4c77d26ca1f..cd37eddf7d42 100644
--- a/drivers/usb/host/ohci-s3c2410.c
+++ b/drivers/usb/host/ohci-s3c2410.c
@@ -370,7 +370,7 @@ static int usb_hcd_s3c2410_probe (const struct hc_driver *driver,
 		goto err_mem;
 	}
 
-	usb_clk = clk_get(&dev->dev, "upll");
+	usb_clk = clk_get(&dev->dev, "usb-bus-host");
 	if (IS_ERR(usb_clk)) {
 		dev_err(&dev->dev, "cannot get usb-host clock\n");
 		retval = -ENOENT;
-- 
cgit v1.2.1


From 60bbfc84b6d916f5a10127762d0dcb3f4726450d Mon Sep 17 00:00:00 2001
From: Vitaly Wool <vwool@ru.mvista.com>
Date: Thu, 29 Jun 2006 18:28:18 +0400
Subject: USB OHCI controller support for PNX4008

inlined is the patch that adds basic support for USB OHCI controller
support for PNX4008 Philips PNX4008 ARM board. Due to HW design, it
depends on I2C driver for PNX4008 which I've recetnly posted to LKML and
i2c at lm-sensors.

Signed-off-by: Vitaly Wool <vitalywool@gmail.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/Kconfig             |   1 +
 drivers/usb/host/Kconfig        |   2 +
 drivers/usb/host/ohci-hcd.c     |   5 +
 drivers/usb/host/ohci-pnx4008.c | 476 ++++++++++++++++++++++++++++++++++++++++
 4 files changed, 484 insertions(+)
 create mode 100644 drivers/usb/host/ohci-pnx4008.c

diff --git a/drivers/usb/Kconfig b/drivers/usb/Kconfig
index 005043197527..f9b1719b9a37 100644
--- a/drivers/usb/Kconfig
+++ b/drivers/usb/Kconfig
@@ -25,6 +25,7 @@ config USB_ARCH_HAS_OHCI
 	default y if PXA27x
 	default y if ARCH_EP93XX
 	default y if (ARCH_AT91RM9200 || ARCH_AT91SAM9261)
+	default y if ARCH_PNX4008
 	# PPC:
 	default y if STB03xxx
 	default y if PPC_MPC52xx
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index b93d71d28db7..77ba2b0bf0b7 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -83,6 +83,8 @@ config USB_OHCI_HCD
 	tristate "OHCI HCD support"
 	depends on USB && USB_ARCH_HAS_OHCI
 	select ISP1301_OMAP if MACH_OMAP_H2 || MACH_OMAP_H3
+	select I2C if ARCH_PNX4008
+	select I2C_PNX if ARCH_PNX4008
 	---help---
 	  The Open Host Controller Interface (OHCI) is a standard for accessing
 	  USB 1.1 host controller hardware.  It does more in hardware than Intel's
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 0684f57c14a0..cbf38ae5ae76 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -940,6 +940,10 @@ MODULE_LICENSE ("GPL");
 #include "ohci-at91.c"
 #endif
 
+#ifdef CONFIG_ARCH_PNX4008
+#include "ohci-pnx4008.c"
+#endif
+
 #if !(defined(CONFIG_PCI) \
       || defined(CONFIG_SA1111) \
       || defined(CONFIG_ARCH_S3C2410) \
@@ -951,6 +955,7 @@ MODULE_LICENSE ("GPL");
       || defined (CONFIG_USB_OHCI_HCD_PPC_SOC) \
       || defined (CONFIG_ARCH_AT91RM9200) \
       || defined (CONFIG_ARCH_AT91SAM9261) \
+      || defined (CONFIG_ARCH_PNX4008) \
 	)
 #error "missing bus glue for ohci-hcd"
 #endif
diff --git a/drivers/usb/host/ohci-pnx4008.c b/drivers/usb/host/ohci-pnx4008.c
new file mode 100644
index 000000000000..82cb22f002e7
--- /dev/null
+++ b/drivers/usb/host/ohci-pnx4008.c
@@ -0,0 +1,476 @@
+/*
+ * drivers/usb/host/ohci-pnx4008.c
+ *
+ * driver for Philips PNX4008 USB Host
+ *
+ * Authors: Dmitry Chigirev <source@mvista.com>
+ * 	    Vitaly Wool <vitalywool@gmail.com>
+ *
+ * register initialization is based on code examples provided by Philips
+ * Copyright (c) 2005 Koninklijke Philips Electronics N.V.
+ *
+ * NOTE: This driver does not have suspend/resume functionality
+ * This driver is intended for engineering development purposes only
+ *
+ * 2005-2006 (c) MontaVista Software, Inc. This file is licensed under
+ * the terms of the GNU General Public License version 2. This program
+ * is licensed "as is" without any warranty of any kind, whether express
+ * or implied.
+ */
+#include <linux/clk.h>
+#include <linux/platform_device.h>
+#include <linux/i2c.h>
+
+#include <asm/hardware.h>
+#include <asm/io.h>
+#include <asm/mach-types.h>
+
+#include <asm/arch/platform.h>
+#include <asm/arch/irqs.h>
+#include <asm/arch/gpio.h>
+
+#define USB_CTRL 	IO_ADDRESS(PNX4008_PWRMAN_BASE + 0x64)
+
+/* USB_CTRL bit defines */
+#define USB_SLAVE_HCLK_EN	(1 << 24)
+#define USB_HOST_NEED_CLK_EN	(1 << 21)
+
+#define USB_OTG_CLK_CTRL	IO_ADDRESS(PNX4008_USB_CONFIG_BASE + 0xFF4)
+#define USB_OTG_CLK_STAT	IO_ADDRESS(PNX4008_USB_CONFIG_BASE + 0xFF8)
+
+/* USB_OTG_CLK_CTRL bit defines */
+#define AHB_M_CLOCK_ON		(1 << 4)
+#define OTG_CLOCK_ON		(1 << 3)
+#define I2C_CLOCK_ON		(1 << 2)
+#define DEV_CLOCK_ON		(1 << 1)
+#define HOST_CLOCK_ON		(1 << 0)
+
+#define USB_OTG_STAT_CONTROL	IO_ADDRESS(PNX4008_USB_CONFIG_BASE + 0x110)
+
+/* USB_OTG_STAT_CONTROL bit defines */
+#define TRANSPARENT_I2C_EN	(1 << 7)
+#define HOST_EN			(1 << 0)
+
+/* ISP1301 USB transceiver I2C registers */
+#define	ISP1301_MODE_CONTROL_1		0x04	/* u8 read, set, +1 clear */
+
+#define	MC1_SPEED_REG		(1 << 0)
+#define	MC1_SUSPEND_REG		(1 << 1)
+#define	MC1_DAT_SE0		(1 << 2)
+#define	MC1_TRANSPARENT		(1 << 3)
+#define	MC1_BDIS_ACON_EN	(1 << 4)
+#define	MC1_OE_INT_EN		(1 << 5)
+#define	MC1_UART_EN		(1 << 6)
+#define	MC1_MASK		0x7f
+
+#define	ISP1301_MODE_CONTROL_2		0x12	/* u8 read, set, +1 clear */
+
+#define	MC2_GLOBAL_PWR_DN	(1 << 0)
+#define	MC2_SPD_SUSP_CTRL	(1 << 1)
+#define	MC2_BI_DI		(1 << 2)
+#define	MC2_TRANSP_BDIR0	(1 << 3)
+#define	MC2_TRANSP_BDIR1	(1 << 4)
+#define	MC2_AUDIO_EN		(1 << 5)
+#define	MC2_PSW_EN		(1 << 6)
+#define	MC2_EN2V7		(1 << 7)
+
+#define	ISP1301_OTG_CONTROL_1		0x06	/* u8 read, set, +1 clear */
+#	define	OTG1_DP_PULLUP		(1 << 0)
+#	define	OTG1_DM_PULLUP		(1 << 1)
+#	define	OTG1_DP_PULLDOWN	(1 << 2)
+#	define	OTG1_DM_PULLDOWN	(1 << 3)
+#	define	OTG1_ID_PULLDOWN	(1 << 4)
+#	define	OTG1_VBUS_DRV		(1 << 5)
+#	define	OTG1_VBUS_DISCHRG	(1 << 6)
+#	define	OTG1_VBUS_CHRG		(1 << 7)
+#define	ISP1301_OTG_STATUS		0x10	/* u8 readonly */
+#	define	OTG_B_SESS_END		(1 << 6)
+#	define	OTG_B_SESS_VLD		(1 << 7)
+
+#define ISP1301_I2C_ADDR 0x2C
+
+#define ISP1301_I2C_MODE_CONTROL_1 0x4
+#define ISP1301_I2C_MODE_CONTROL_2 0x12
+#define ISP1301_I2C_OTG_CONTROL_1 0x6
+#define ISP1301_I2C_OTG_CONTROL_2 0x10
+#define ISP1301_I2C_INTERRUPT_SOURCE 0x8
+#define ISP1301_I2C_INTERRUPT_LATCH 0xA
+#define ISP1301_I2C_INTERRUPT_FALLING 0xC
+#define ISP1301_I2C_INTERRUPT_RISING 0xE
+#define ISP1301_I2C_REG_CLEAR_ADDR 1
+
+struct i2c_driver isp1301_driver;
+struct i2c_client *isp1301_i2c_client;
+
+extern int usb_disabled(void);
+extern int ocpi_enable(void);
+
+static struct clk *usb_clk;
+
+static int isp1301_probe(struct i2c_adapter *adap);
+static int isp1301_detach(struct i2c_client *client);
+static int isp1301_command(struct i2c_client *client, unsigned int cmd,
+			   void *arg);
+
+static unsigned short normal_i2c[] =
+    { ISP1301_I2C_ADDR, ISP1301_I2C_ADDR + 1, I2C_CLIENT_END };
+static unsigned short dummy_i2c_addrlist[] = { I2C_CLIENT_END };
+
+static struct i2c_client_address_data addr_data = {
+	.normal_i2c = normal_i2c,
+	.probe = dummy_i2c_addrlist,
+	.ignore = dummy_i2c_addrlist,
+};
+
+struct i2c_driver isp1301_driver = {
+	.id = I2C_DRIVERID_I2CDEV,	/* Fake Id */
+	.class = I2C_CLASS_HWMON,
+	.attach_adapter = isp1301_probe,
+	.detach_client = isp1301_detach,
+	.command = isp1301_command
+};
+
+static int isp1301_attach(struct i2c_adapter *adap, int addr, int kind)
+{
+	struct i2c_client *c;
+
+	c = (struct i2c_client *)kzalloc(sizeof(*c), SLAB_KERNEL);
+
+	if (!c)
+		return -ENOMEM;
+
+	strcpy(c->name, "isp1301");
+	c->flags = 0;
+	c->addr = addr;
+	c->adapter = adap;
+	c->driver = &isp1301_driver;
+
+	isp1301_i2c_client = c;
+
+	return i2c_attach_client(c);
+}
+
+static int isp1301_probe(struct i2c_adapter *adap)
+{
+	return i2c_probe(adap, &addr_data, isp1301_attach);
+}
+
+static int isp1301_detach(struct i2c_client *client)
+{
+	i2c_detach_client(client);
+	kfree(isp1301_i2c_client);
+	return 0;
+}
+
+/* No commands defined */
+static int isp1301_command(struct i2c_client *client, unsigned int cmd,
+			   void *arg)
+{
+	return 0;
+}
+
+static void i2c_write(u8 buf, u8 subaddr)
+{
+	char tmpbuf[2];
+
+	tmpbuf[0] = subaddr;	/*register number */
+	tmpbuf[1] = buf;	/*register data */
+	i2c_master_send(isp1301_i2c_client, &tmpbuf[0], 2);
+}
+
+static void isp1301_configure(void)
+{
+	/* PNX4008 only supports DAT_SE0 USB mode */
+	/* PNX4008 R2A requires setting the MAX603 to output 3.6V */
+	/* Power up externel charge-pump */
+
+	i2c_write(MC1_DAT_SE0 | MC1_SPEED_REG, ISP1301_I2C_MODE_CONTROL_1);
+	i2c_write(~(MC1_DAT_SE0 | MC1_SPEED_REG),
+		  ISP1301_I2C_MODE_CONTROL_1 | ISP1301_I2C_REG_CLEAR_ADDR);
+	i2c_write(MC2_BI_DI | MC2_PSW_EN | MC2_SPD_SUSP_CTRL,
+		  ISP1301_I2C_MODE_CONTROL_2);
+	i2c_write(~(MC2_BI_DI | MC2_PSW_EN | MC2_SPD_SUSP_CTRL),
+		  ISP1301_I2C_MODE_CONTROL_2 | ISP1301_I2C_REG_CLEAR_ADDR);
+	i2c_write(OTG1_DM_PULLDOWN | OTG1_DP_PULLDOWN,
+		  ISP1301_I2C_OTG_CONTROL_1);
+	i2c_write(~(OTG1_DM_PULLDOWN | OTG1_DP_PULLDOWN),
+		  ISP1301_I2C_OTG_CONTROL_1 | ISP1301_I2C_REG_CLEAR_ADDR);
+	i2c_write(0xFF,
+		  ISP1301_I2C_INTERRUPT_LATCH | ISP1301_I2C_REG_CLEAR_ADDR);
+	i2c_write(0xFF,
+		  ISP1301_I2C_INTERRUPT_FALLING | ISP1301_I2C_REG_CLEAR_ADDR);
+	i2c_write(0xFF,
+		  ISP1301_I2C_INTERRUPT_RISING | ISP1301_I2C_REG_CLEAR_ADDR);
+
+}
+
+static inline void isp1301_vbus_on(void)
+{
+	i2c_write(OTG1_VBUS_DRV, ISP1301_I2C_OTG_CONTROL_1);
+}
+
+static inline void isp1301_vbus_off(void)
+{
+	i2c_write(OTG1_VBUS_DRV,
+		  ISP1301_I2C_OTG_CONTROL_1 | ISP1301_I2C_REG_CLEAR_ADDR);
+}
+
+static void pnx4008_start_hc(void)
+{
+	unsigned long tmp = __raw_readl(USB_OTG_STAT_CONTROL) | HOST_EN;
+	__raw_writel(tmp, USB_OTG_STAT_CONTROL);
+	isp1301_vbus_on();
+}
+
+static void pnx4008_stop_hc(void)
+{
+	unsigned long tmp;
+	isp1301_vbus_off();
+	tmp = __raw_readl(USB_OTG_STAT_CONTROL) & ~HOST_EN;
+	__raw_writel(tmp, USB_OTG_STAT_CONTROL);
+}
+
+static int __devinit ohci_pnx4008_start(struct usb_hcd *hcd)
+{
+	struct ohci_hcd *ohci = hcd_to_ohci(hcd);
+	int ret;
+
+	if ((ret = ohci_init(ohci)) < 0)
+		return ret;
+
+	if ((ret = ohci_run(ohci)) < 0) {
+		dev_err(hcd->self.controller, "can't start\n");
+		ohci_stop(hcd);
+		return ret;
+	}
+	return 0;
+}
+
+static const struct hc_driver ohci_pnx4008_hc_driver = {
+	.description = hcd_name,
+	.product_desc =		"pnx4008 OHCI",
+
+	/*
+	 * generic hardware linkage
+	 */
+	.irq = ohci_irq,
+	.flags = HCD_USB11 | HCD_MEMORY,
+
+	.hcd_priv_size =	sizeof(struct ohci_hcd),
+	/*
+	 * basic lifecycle operations
+	 */
+	.start = ohci_pnx4008_start,
+	.stop = ohci_stop,
+
+	/*
+	 * managing i/o requests and associated device resources
+	 */
+	.urb_enqueue = ohci_urb_enqueue,
+	.urb_dequeue = ohci_urb_dequeue,
+	.endpoint_disable = ohci_endpoint_disable,
+
+	/*
+	 * scheduling support
+	 */
+	.get_frame_number = ohci_get_frame,
+
+	/*
+	 * root hub support
+	 */
+	.hub_status_data = ohci_hub_status_data,
+	.hub_control = ohci_hub_control,
+
+	.start_port_reset = ohci_start_port_reset,
+};
+
+#define USB_CLOCK_MASK (AHB_M_CLOCK_ON| OTG_CLOCK_ON | HOST_CLOCK_ON | I2C_CLOCK_ON)
+
+static void pnx4008_set_usb_bits(void)
+{
+	start_int_set_falling_edge(SE_USB_OTG_ATX_INT_N);
+	start_int_ack(SE_USB_OTG_ATX_INT_N);
+	start_int_umask(SE_USB_OTG_ATX_INT_N);
+
+	start_int_set_rising_edge(SE_USB_OTG_TIMER_INT);
+	start_int_ack(SE_USB_OTG_TIMER_INT);
+	start_int_umask(SE_USB_OTG_TIMER_INT);
+
+	start_int_set_rising_edge(SE_USB_I2C_INT);
+	start_int_ack(SE_USB_I2C_INT);
+	start_int_umask(SE_USB_I2C_INT);
+
+	start_int_set_rising_edge(SE_USB_INT);
+	start_int_ack(SE_USB_INT);
+	start_int_umask(SE_USB_INT);
+
+	start_int_set_rising_edge(SE_USB_NEED_CLK_INT);
+	start_int_ack(SE_USB_NEED_CLK_INT);
+	start_int_umask(SE_USB_NEED_CLK_INT);
+
+	start_int_set_rising_edge(SE_USB_AHB_NEED_CLK_INT);
+	start_int_ack(SE_USB_AHB_NEED_CLK_INT);
+	start_int_umask(SE_USB_AHB_NEED_CLK_INT);
+}
+
+static void pnx4008_unset_usb_bits(void)
+{
+	start_int_mask(SE_USB_OTG_ATX_INT_N);
+	start_int_mask(SE_USB_OTG_TIMER_INT);
+	start_int_mask(SE_USB_I2C_INT);
+	start_int_mask(SE_USB_INT);
+	start_int_mask(SE_USB_NEED_CLK_INT);
+	start_int_mask(SE_USB_AHB_NEED_CLK_INT);
+}
+
+static int __devinit usb_hcd_pnx4008_probe(struct platform_device *pdev)
+{
+	struct usb_hcd *hcd = 0;
+	struct ohci_hcd *ohci;
+	const struct hc_driver *driver = &ohci_pnx4008_hc_driver;
+
+	int ret = 0, irq;
+
+	dev_dbg(&pdev->dev, "%s: " DRIVER_INFO " (pnx4008)\n", hcd_name);
+	if (usb_disabled()) {
+		err("USB is disabled");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	if (pdev->num_resources != 2
+	    || pdev->resource[0].flags != IORESOURCE_MEM
+	    || pdev->resource[1].flags != IORESOURCE_IRQ) {
+		err("Invalid resource configuration");
+		ret = -ENODEV;
+		goto out;
+	}
+
+	/* Enable AHB slave USB clock, needed for further USB clock control */
+	__raw_writel(USB_SLAVE_HCLK_EN | (1 << 19), USB_CTRL);
+
+	ret = i2c_add_driver(&isp1301_driver);
+	if (ret < 0) {
+		err("failed to connect I2C to ISP1301 USB Transceiver");
+		goto out;
+	}
+
+	isp1301_configure();
+
+	/* Enable USB PLL */
+	usb_clk = clk_get(&pdev->dev, "ck_pll5");
+	if (IS_ERR(usb_clk)) {
+		err("failed to acquire USB PLL");
+		ret = PTR_ERR(usb_clk);
+		goto out1;
+	}
+
+	ret = clk_enable(usb_clk);
+	if (ret < 0) {
+		err("failed to start USB PLL");
+		goto out2;
+	}
+
+	ret = clk_set_rate(usb_clk, 48000);
+	if (ret < 0) {
+		err("failed to set USB clock rate");
+		goto out3;
+	}
+
+	__raw_writel(__raw_readl(USB_CTRL) | USB_HOST_NEED_CLK_EN, USB_CTRL);
+
+	/* Set to enable all needed USB clocks */
+	__raw_writel(USB_CLOCK_MASK, USB_OTG_CLK_CTRL);
+
+	while ((__raw_readl(USB_OTG_CLK_STAT) & USB_CLOCK_MASK) !=
+	       USB_CLOCK_MASK) ;
+
+	hcd = usb_create_hcd (driver, &pdev->dev, pdev->dev.bus_id);
+	if (!hcd) {
+		err("Failed to allocate HC buffer");
+		ret = -ENOMEM;
+		goto out3;
+	}
+
+	/* Set all USB bits in the Start Enable register */
+	pnx4008_set_usb_bits();
+
+	hcd->rsrc_start = pdev->resource[0].start;
+	hcd->rsrc_len = pdev->resource[0].end - pdev->resource[0].start + 1;
+	if (!request_mem_region(hcd->rsrc_start, hcd->rsrc_len, hcd_name)) {
+		dev_dbg(&pdev->dev, "request_mem_region failed\n");
+		ret =  -ENOMEM;
+		goto out4;
+	}
+	hcd->regs = (void __iomem *)pdev->resource[0].start;
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		ret = -ENXIO;
+		goto out4;
+	}
+
+	hcd->self.hcpriv = (void *)hcd;
+
+	pnx4008_start_hc();
+	platform_set_drvdata(pdev, hcd);
+	ohci = hcd_to_ohci(hcd);
+	ohci_hcd_init(ohci);
+
+	dev_info(&pdev->dev, "at 0x%p, irq %d\n", hcd->regs, hcd->irq);
+	ret = usb_add_hcd(hcd, irq, SA_INTERRUPT);
+	if (ret == 0)
+		return ret;
+
+	pnx4008_stop_hc();
+out4:
+	pnx4008_unset_usb_bits();
+	usb_put_hcd(hcd);
+out3:
+	clk_disable(usb_clk);
+out2:
+	clk_put(usb_clk);
+out1:
+	i2c_del_driver(&isp1301_driver);
+out:
+	return ret;
+}
+
+static int usb_hcd_pnx4008_remove(struct platform_device *pdev)
+{
+	struct usb_hcd *hcd = platform_get_drvdata(pdev);
+
+	usb_remove_hcd(hcd);
+	pnx4008_stop_hc();
+	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
+	usb_put_hcd(hcd);
+	pnx4008_unset_usb_bits();
+	clk_disable(usb_clk);
+	clk_put(usb_clk);
+	i2c_del_driver(&isp1301_driver);
+
+	platform_set_drvdata(pdev, NULL);
+
+	return 0;
+}
+
+static struct platform_driver usb_hcd_pnx4008_driver = {
+	.driver = {
+		.name = "usb-ohci",
+	},
+	.probe = usb_hcd_pnx4008_probe,
+	.remove = usb_hcd_pnx4008_remove,
+};
+
+static int __init usb_hcd_pnx4008_init(void)
+{
+	return platform_driver_register(&usb_hcd_pnx4008_driver);
+}
+
+static void __exit usb_hcd_pnx4008_cleanup(void)
+{
+	return platform_driver_unregister(&usb_hcd_pnx4008_driver);
+}
+
+module_init(usb_hcd_pnx4008_init);
+module_exit(usb_hcd_pnx4008_cleanup);
-- 
cgit v1.2.1


From 26facdff1d39c0028565137ad27e8bd5a5cafcf1 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 7 Jul 2006 00:11:45 -0700
Subject: USB: kill usb kconfig warning

drivers/usb/host/Kconfig:87:warning: 'select' used by config symbol 'USB_OHCI_HCD' refer to undefined symbol 'I2C_PNX'


Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index 77ba2b0bf0b7..4bd5cddae8a5 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -84,7 +84,6 @@ config USB_OHCI_HCD
 	depends on USB && USB_ARCH_HAS_OHCI
 	select ISP1301_OMAP if MACH_OMAP_H2 || MACH_OMAP_H3
 	select I2C if ARCH_PNX4008
-	select I2C_PNX if ARCH_PNX4008
 	---help---
 	  The Open Host Controller Interface (OHCI) is a standard for accessing
 	  USB 1.1 host controller hardware.  It does more in hardware than Intel's
-- 
cgit v1.2.1


From 3a16f7b4a75d68364c3278523f51ac141a12758a Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 29 Jun 2006 12:27:23 -0700
Subject: USB: move <linux/usb_otg.h> to <linux/usb/otg.h>

Move <linux/usb_otg.h> to <linux/usb/otg.h>.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/plat-omap/usb.c         |   2 +-
 drivers/i2c/chips/isp1301_omap.c |   2 +-
 drivers/usb/gadget/omap_udc.c    |   2 +-
 drivers/usb/host/ohci-hcd.c      |   2 +-
 include/linux/usb/otg.h          | 131 +++++++++++++++++++++++++++++++++++++++
 include/linux/usb_otg.h          | 131 ---------------------------------------
 6 files changed, 135 insertions(+), 135 deletions(-)
 create mode 100644 include/linux/usb/otg.h
 delete mode 100644 include/linux/usb_otg.h

diff --git a/arch/arm/plat-omap/usb.c b/arch/arm/plat-omap/usb.c
index 9b815327b6a5..7e8096809be2 100644
--- a/arch/arm/plat-omap/usb.c
+++ b/arch/arm/plat-omap/usb.c
@@ -26,7 +26,7 @@
 #include <linux/errno.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 
 #include <asm/io.h>
 #include <asm/irq.h>
diff --git a/drivers/i2c/chips/isp1301_omap.c b/drivers/i2c/chips/isp1301_omap.c
index f92505b94c61..182f04953466 100644
--- a/drivers/i2c/chips/isp1301_omap.c
+++ b/drivers/i2c/chips/isp1301_omap.c
@@ -30,7 +30,7 @@
 #include <linux/usb_ch9.h>
 #include <linux/usb_gadget.h>
 #include <linux/usb.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/i2c.h>
 #include <linux/workqueue.h>
 
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 2de9748ee673..81f0389fcc94 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -40,7 +40,7 @@
 #include <linux/platform_device.h>
 #include <linux/usb_ch9.h>
 #include <linux/usb_gadget.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/dma-mapping.h>
 
 #include <asm/byteorder.h>
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index cbf38ae5ae76..7c3d8c60a60f 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -88,7 +88,7 @@
 #include <linux/timer.h>
 #include <linux/list.h>
 #include <linux/usb.h>
-#include <linux/usb_otg.h>
+#include <linux/usb/otg.h>
 #include <linux/dma-mapping.h> 
 #include <linux/dmapool.h>
 #include <linux/reboot.h>
diff --git a/include/linux/usb/otg.h b/include/linux/usb/otg.h
new file mode 100644
index 000000000000..9897f7a818c5
--- /dev/null
+++ b/include/linux/usb/otg.h
@@ -0,0 +1,131 @@
+// include/linux/usb/otg.h
+
+/*
+ * These APIs may be used between USB controllers.  USB device drivers
+ * (for either host or peripheral roles) don't use these calls; they
+ * continue to use just usb_device and usb_gadget.
+ */
+
+
+/* OTG defines lots of enumeration states before device reset */
+enum usb_otg_state {
+	OTG_STATE_UNDEFINED = 0,
+
+	/* single-role peripheral, and dual-role default-b */
+	OTG_STATE_B_IDLE,
+	OTG_STATE_B_SRP_INIT,
+	OTG_STATE_B_PERIPHERAL,
+
+	/* extra dual-role default-b states */
+	OTG_STATE_B_WAIT_ACON,
+	OTG_STATE_B_HOST,
+
+	/* dual-role default-a */
+	OTG_STATE_A_IDLE,
+	OTG_STATE_A_WAIT_VRISE,
+	OTG_STATE_A_WAIT_BCON,
+	OTG_STATE_A_HOST,
+	OTG_STATE_A_SUSPEND,
+	OTG_STATE_A_PERIPHERAL,
+	OTG_STATE_A_WAIT_VFALL,
+	OTG_STATE_A_VBUS_ERR,
+};
+
+/*
+ * the otg driver needs to interact with both device side and host side
+ * usb controllers.  it decides which controller is active at a given
+ * moment, using the transceiver, ID signal, HNP and sometimes static
+ * configuration information (including "board isn't wired for otg").
+ */
+struct otg_transceiver {
+	struct device		*dev;
+	const char		*label;
+
+	u8			default_a;
+	enum usb_otg_state	state;
+
+	struct usb_bus		*host;
+	struct usb_gadget	*gadget;
+
+	/* to pass extra port status to the root hub */
+	u16			port_status;
+	u16			port_change;
+
+	/* bind/unbind the host controller */
+	int	(*set_host)(struct otg_transceiver *otg,
+				struct usb_bus *host);
+
+	/* bind/unbind the peripheral controller */
+	int	(*set_peripheral)(struct otg_transceiver *otg,
+				struct usb_gadget *gadget);
+
+	/* effective for B devices, ignored for A-peripheral */
+	int	(*set_power)(struct otg_transceiver *otg,
+				unsigned mA);
+
+	/* for non-OTG B devices: set transceiver into suspend mode */
+	int	(*set_suspend)(struct otg_transceiver *otg,
+				int suspend);
+
+	/* for B devices only:  start session with A-Host */
+	int	(*start_srp)(struct otg_transceiver *otg);
+
+	/* start or continue HNP role switch */
+	int	(*start_hnp)(struct otg_transceiver *otg);
+
+};
+
+
+/* for board-specific init logic */
+extern int otg_set_transceiver(struct otg_transceiver *);
+
+
+/* for usb host and peripheral controller drivers */
+extern struct otg_transceiver *otg_get_transceiver(void);
+
+static inline int
+otg_start_hnp(struct otg_transceiver *otg)
+{
+	return otg->start_hnp(otg);
+}
+
+
+/* for HCDs */
+static inline int
+otg_set_host(struct otg_transceiver *otg, struct usb_bus *host)
+{
+	return otg->set_host(otg, host);
+}
+
+
+/* for usb peripheral controller drivers */
+static inline int
+otg_set_peripheral(struct otg_transceiver *otg, struct usb_gadget *periph)
+{
+	return otg->set_peripheral(otg, periph);
+}
+
+static inline int
+otg_set_power(struct otg_transceiver *otg, unsigned mA)
+{
+	return otg->set_power(otg, mA);
+}
+
+static inline int
+otg_set_suspend(struct otg_transceiver *otg, int suspend)
+{
+	if (otg->set_suspend != NULL)
+		return otg->set_suspend(otg, suspend);
+	else
+		return 0;
+}
+
+static inline int
+otg_start_srp(struct otg_transceiver *otg)
+{
+	return otg->start_srp(otg);
+}
+
+
+/* for OTG controller drivers (and maybe other stuff) */
+extern int usb_bus_start_enum(struct usb_bus *bus, unsigned port_num);
diff --git a/include/linux/usb_otg.h b/include/linux/usb_otg.h
deleted file mode 100644
index f827f6e203c2..000000000000
--- a/include/linux/usb_otg.h
+++ /dev/null
@@ -1,131 +0,0 @@
-// include/linux/usb_otg.h 
-
-/*
- * These APIs may be used between USB controllers.  USB device drivers
- * (for either host or peripheral roles) don't use these calls; they
- * continue to use just usb_device and usb_gadget.
- */
-
-
-/* OTG defines lots of enumeration states before device reset */
-enum usb_otg_state {
-	OTG_STATE_UNDEFINED = 0,
-
-	/* single-role peripheral, and dual-role default-b */
-	OTG_STATE_B_IDLE,
-	OTG_STATE_B_SRP_INIT,
-	OTG_STATE_B_PERIPHERAL,
-
-	/* extra dual-role default-b states */
-	OTG_STATE_B_WAIT_ACON,
-	OTG_STATE_B_HOST,
-
-	/* dual-role default-a */
-	OTG_STATE_A_IDLE,
-	OTG_STATE_A_WAIT_VRISE,
-	OTG_STATE_A_WAIT_BCON,
-	OTG_STATE_A_HOST,
-	OTG_STATE_A_SUSPEND,
-	OTG_STATE_A_PERIPHERAL,
-	OTG_STATE_A_WAIT_VFALL,
-	OTG_STATE_A_VBUS_ERR,
-};
-
-/*
- * the otg driver needs to interact with both device side and host side
- * usb controllers.  it decides which controller is active at a given
- * moment, using the transceiver, ID signal, HNP and sometimes static
- * configuration information (including "board isn't wired for otg").
- */
-struct otg_transceiver {
-	struct device		*dev;
-	const char		*label;
-
-	u8			default_a;
-	enum usb_otg_state	state;
-
-	struct usb_bus		*host;
-	struct usb_gadget	*gadget;
-
-	/* to pass extra port status to the root hub */
-	u16			port_status;
-	u16			port_change;
-
-	/* bind/unbind the host controller */
-	int 	(*set_host)(struct otg_transceiver *otg,
-				struct usb_bus *host);
-
-	/* bind/unbind the peripheral controller */
-	int	(*set_peripheral)(struct otg_transceiver *otg,
-				struct usb_gadget *gadget);
-
-	/* effective for B devices, ignored for A-peripheral */
-	int	(*set_power)(struct otg_transceiver *otg,
-				unsigned mA);
-
-	/* for non-OTG B devices: set transceiver into suspend mode */
-	int	(*set_suspend)(struct otg_transceiver *otg,
-				int suspend);
-
-	/* for B devices only:  start session with A-Host */
-	int	(*start_srp)(struct otg_transceiver *otg);
-
-	/* start or continue HNP role switch */
-	int	(*start_hnp)(struct otg_transceiver *otg);
-
-};
-
-
-/* for board-specific init logic */
-extern int otg_set_transceiver(struct otg_transceiver *);
-
-
-/* for usb host and peripheral controller drivers */
-extern struct otg_transceiver *otg_get_transceiver(void);
-
-static inline int
-otg_start_hnp(struct otg_transceiver *otg)
-{
-	return otg->start_hnp(otg);
-}
-
-
-/* for HCDs */
-static inline int
-otg_set_host(struct otg_transceiver *otg, struct usb_bus *host)
-{
-	return otg->set_host(otg, host);
-}
-
-
-/* for usb peripheral controller drivers */
-static inline int
-otg_set_peripheral(struct otg_transceiver *otg, struct usb_gadget *periph)
-{
-	return otg->set_peripheral(otg, periph);
-}
-
-static inline int
-otg_set_power(struct otg_transceiver *otg, unsigned mA)
-{
-	return otg->set_power(otg, mA);
-}
-
-static inline int
-otg_set_suspend(struct otg_transceiver *otg, int suspend)
-{
-	if (otg->set_suspend != NULL)
-		return otg->set_suspend(otg, suspend);
-	else
-		return 0;
-}
-
-static inline int
-otg_start_srp(struct otg_transceiver *otg)
-{
-	return otg->start_srp(otg);
-}
-
-
-/* for OTG controller drivers (and maybe other stuff) */
-extern int usb_bus_start_enum(struct usb_bus *bus, unsigned port_num);
-- 
cgit v1.2.1


From b2bbb20b37d734443d1c279d0033a64f6095db54 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 29 Jun 2006 12:25:39 -0700
Subject: USB: pxa2xx_udc understands GPIO based VBUS sensing

This updates the PXA 25x UDC board-independent infrastructure for VBUS sensing
and the D+ pullup.  The original code evolved from rather bizarre support on
Intel's "Lubbock" reference hardware, so that on more sensible hardware it
doesn't work as well as it could/should.

The change is just to teach the UDC driver how to use built-in PXA GPIO pins
directly.  This reduces the amount of board-specfic object code needed, and
enables the use of a VBUS sensing IRQ on boards (like Gumstix) that have one.
With VBUS sensing, the UDC is unclocked until a host is actually connected.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 arch/arm/mach-pxa/corgi.c       | 15 +--------
 drivers/usb/gadget/pxa2xx_udc.c | 70 +++++++++++++++++++++++++++++++++++++++--
 drivers/usb/gadget/pxa2xx_udc.h | 24 +++++---------
 include/asm-arm/arch-pxa/udc.h  |  8 +++++
 4 files changed, 85 insertions(+), 32 deletions(-)

diff --git a/arch/arm/mach-pxa/corgi.c b/arch/arm/mach-pxa/corgi.c
index cce26576999e..337c01c4ac37 100644
--- a/arch/arm/mach-pxa/corgi.c
+++ b/arch/arm/mach-pxa/corgi.c
@@ -284,21 +284,9 @@ static struct pxaficp_platform_data corgi_ficp_platform_data = {
 /*
  * USB Device Controller
  */
-static void corgi_udc_command(int cmd)
-{
-	switch(cmd)	{
-	case PXA2XX_UDC_CMD_CONNECT:
-		GPSR(CORGI_GPIO_USB_PULLUP) = GPIO_bit(CORGI_GPIO_USB_PULLUP);
-		break;
-	case PXA2XX_UDC_CMD_DISCONNECT:
-		GPCR(CORGI_GPIO_USB_PULLUP) = GPIO_bit(CORGI_GPIO_USB_PULLUP);
-		break;
-	}
-}
-
 static struct pxa2xx_udc_mach_info udc_info __initdata = {
 	/* no connect GPIO; corgi can't tell connection status */
-	.udc_command		= corgi_udc_command,
+	.gpio_pullup		= CORGI_GPIO_USB_PULLUP,
 };
 
 
@@ -350,7 +338,6 @@ static void __init corgi_init(void)
 	corgi_ssp_set_machinfo(&corgi_ssp_machinfo);
 
 	pxa_gpio_mode(CORGI_GPIO_IR_ON | GPIO_OUT);
-	pxa_gpio_mode(CORGI_GPIO_USB_PULLUP | GPIO_OUT);
 	pxa_gpio_mode(CORGI_GPIO_HSYNC | GPIO_IN);
 
  	pxa_set_udc_info(&udc_info);
diff --git a/drivers/usb/gadget/pxa2xx_udc.c b/drivers/usb/gadget/pxa2xx_udc.c
index fff027d30a09..f1adcf8b2023 100644
--- a/drivers/usb/gadget/pxa2xx_udc.c
+++ b/drivers/usb/gadget/pxa2xx_udc.c
@@ -150,6 +150,39 @@ MODULE_PARM_DESC (fifo_mode, "pxa2xx udc fifo mode");
 static void pxa2xx_ep_fifo_flush (struct usb_ep *ep);
 static void nuke (struct pxa2xx_ep *, int status);
 
+/* one GPIO should be used to detect VBUS from the host */
+static int is_vbus_present(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_vbus)
+		return pxa_gpio_get(mach->gpio_vbus);
+	if (mach->udc_is_connected)
+		return mach->udc_is_connected();
+	return 1;
+}
+
+/* one GPIO should control a D+ pullup, so host sees this device (or not) */
+static void pullup_off(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_pullup)
+		pxa_gpio_set(mach->gpio_pullup, 0);
+	else if (mach->udc_command)
+		mach->udc_command(PXA2XX_UDC_CMD_DISCONNECT);
+}
+
+static void pullup_on(void)
+{
+	struct pxa2xx_udc_mach_info		*mach = the_controller->mach;
+
+	if (mach->gpio_pullup)
+		pxa_gpio_set(mach->gpio_pullup, 1);
+	else if (mach->udc_command)
+		mach->udc_command(PXA2XX_UDC_CMD_CONNECT);
+}
+
 static void pio_irq_enable(int bEndpointAddress)
 {
         bEndpointAddress &= 0xf;
@@ -1721,6 +1754,16 @@ lubbock_vbus_irq(int irq, void *_dev, struct pt_regs *r)
 
 #endif
 
+static irqreturn_t
+udc_vbus_irq(int irq, void *_dev, struct pt_regs *r)
+{
+	struct pxa2xx_udc	*dev = _dev;
+	int			vbus = pxa_gpio_get(dev->mach->gpio_vbus);
+
+	pxa2xx_udc_vbus_session(&dev->gadget, vbus);
+	return IRQ_HANDLED;
+}
+
 
 /*-------------------------------------------------------------------------*/
 
@@ -2438,7 +2481,7 @@ static struct pxa2xx_udc memory = {
 static int __init pxa2xx_udc_probe(struct platform_device *pdev)
 {
 	struct pxa2xx_udc *dev = &memory;
-	int retval, out_dma = 1;
+	int retval, out_dma = 1, vbus_irq;
 	u32 chiprev;
 
 	/* insist on Intel/ARM/XScale */
@@ -2502,6 +2545,16 @@ static int __init pxa2xx_udc_probe(struct platform_device *pdev)
 	/* other non-static parts of init */
 	dev->dev = &pdev->dev;
 	dev->mach = pdev->dev.platform_data;
+	if (dev->mach->gpio_vbus) {
+		vbus_irq = IRQ_GPIO(dev->mach->gpio_vbus & GPIO_MD_MASK_NR);
+		pxa_gpio_mode((dev->mach->gpio_vbus & GPIO_MD_MASK_NR)
+				| GPIO_IN);
+		set_irq_type(vbus_irq, IRQT_BOTHEDGE);
+	} else
+		vbus_irq = 0;
+	if (dev->mach->gpio_pullup)
+		pxa_gpio_mode((dev->mach->gpio_pullup & GPIO_MD_MASK_NR)
+				| GPIO_OUT | GPIO_DFLT_LOW);
 
 	init_timer(&dev->timer);
 	dev->timer.function = udc_watchdog;
@@ -2557,8 +2610,19 @@ lubbock_fail0:
 		HEX_DISPLAY(dev->stats.irqs);
 		LUB_DISC_BLNK_LED &= 0xff;
 #endif
-	}
+	} else
 #endif
+	if (vbus_irq) {
+		retval = request_irq(vbus_irq, udc_vbus_irq,
+				SA_INTERRUPT | SA_SAMPLE_RANDOM,
+				driver_name, dev);
+		if (retval != 0) {
+			printk(KERN_ERR "%s: can't get irq %i, err %d\n",
+				driver_name, vbus_irq, retval);
+			free_irq(IRQ_USB, dev);
+			return -EBUSY;
+		}
+	}
 	create_proc_files();
 
 	return 0;
@@ -2587,6 +2651,8 @@ static int __exit pxa2xx_udc_remove(struct platform_device *pdev)
 		free_irq(LUBBOCK_USB_IRQ, dev);
 	}
 #endif
+	if (dev->mach->gpio_vbus)
+		free_irq(IRQ_GPIO(dev->mach->gpio_vbus), dev);
 	platform_set_drvdata(pdev, NULL);
 	the_controller = NULL;
 	return 0;
diff --git a/drivers/usb/gadget/pxa2xx_udc.h b/drivers/usb/gadget/pxa2xx_udc.h
index 19a883f7d1b8..8e598c8bf4e3 100644
--- a/drivers/usb/gadget/pxa2xx_udc.h
+++ b/drivers/usb/gadget/pxa2xx_udc.h
@@ -177,27 +177,19 @@ struct pxa2xx_udc {
 
 static struct pxa2xx_udc *the_controller;
 
-/* one GPIO should be used to detect VBUS from the host */
-static inline int is_vbus_present(void)
+static inline int pxa_gpio_get(unsigned gpio)
 {
-	if (!the_controller->mach->udc_is_connected)
-		return 1;
-	return the_controller->mach->udc_is_connected();
+	return (GPLR(gpio) & GPIO_bit(gpio)) != 0;
 }
 
-/* one GPIO should control a D+ pullup, so host sees this device (or not) */
-static inline void pullup_off(void)
+static inline void pxa_gpio_set(unsigned gpio, int is_on)
 {
-	if (!the_controller->mach->udc_command)
-		return;
-	the_controller->mach->udc_command(PXA2XX_UDC_CMD_DISCONNECT);
-}
+	int mask = GPIO_bit(gpio);
 
-static inline void pullup_on(void)
-{
-	if (!the_controller->mach->udc_command)
-		return;
-	the_controller->mach->udc_command(PXA2XX_UDC_CMD_CONNECT);
+	if (is_on)
+		GPSR(gpio) = mask;
+	else
+		GPCR(gpio) = mask;
 }
 
 /*-------------------------------------------------------------------------*/
diff --git a/include/asm-arm/arch-pxa/udc.h b/include/asm-arm/arch-pxa/udc.h
index 30548a30c773..121cd241115d 100644
--- a/include/asm-arm/arch-pxa/udc.h
+++ b/include/asm-arm/arch-pxa/udc.h
@@ -12,6 +12,14 @@ struct pxa2xx_udc_mach_info {
         void (*udc_command)(int cmd);
 #define	PXA2XX_UDC_CMD_CONNECT		0	/* let host see us */
 #define	PXA2XX_UDC_CMD_DISCONNECT	1	/* so host won't see us */
+
+	/* Boards following the design guidelines in the developer's manual,
+	 * with on-chip GPIOs not Lubbock's wierd hardware, can have a sane
+	 * VBUS IRQ and omit the methods above.  Store the GPIO number
+	 * here; for GPIO 0, also mask in one of the pxa_gpio_mode() bits.
+	 */
+	u16	gpio_vbus;			/* high == vbus present */
+	u16	gpio_pullup;			/* high == pullup activated */
 };
 
 extern void pxa_set_udc_info(struct pxa2xx_udc_mach_info *info);
-- 
cgit v1.2.1


From bfb2c965d669045b7629fd577b7834c87c2dfd54 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Thu, 29 Jun 2006 22:27:36 -0700
Subject: USB: Allow compile in g_ether, fix typo

Allows compiling g_ether in and fixes a typo with MUSB_HDRC

Signed-off-by: Tony Lindgren <tony@atomide.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/ether.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index 30299c620d97..fed484da593c 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -262,7 +262,7 @@ MODULE_PARM_DESC(host_addr, "Host Ethernet Address");
 #define DEV_CONFIG_CDC
 #endif
 
-#ifdef CONFIG_USB_GADGET_MUSBHDRC
+#ifdef CONFIG_USB_GADGET_MUSB_HDRC
 #define DEV_CONFIG_CDC
 #endif
 
@@ -2564,7 +2564,7 @@ static struct usb_gadget_driver eth_driver = {
 
 	.function	= (char *) driver_desc,
 	.bind		= eth_bind,
-	.unbind		= __exit_p(eth_unbind),
+	.unbind		= eth_unbind,
 
 	.setup		= eth_setup,
 	.disconnect	= eth_disconnect,
-- 
cgit v1.2.1


From 2f430b4bbae7faa167730f954252eb7db4ac58dd Mon Sep 17 00:00:00 2001
From: Werner Lemberg <wl@gnu.org>
Date: Tue, 18 Jul 2006 17:00:52 +0200
Subject: USB: ark3116: Add TIOCGSERIAL and TIOCSSERIAL ioctl calls.

Add (dummy?) support for TIOCGSERIAL and TIOCSSERIAL ioctl calls to the USB
serial driver file `ark3116.c'.  This is sufficient for me to run wvdial
successfully, receive my email, and do webbrowsing with firefox.  On the
other hand, running the cvs program to update archives seems not to work,
and the traceroute command sometimes says

  send failed: No buffer space available

Looks like a buffering problem...  My knowledge of serial device drivers is
zero, so I can't fix this -- I just did a cut'n'paste from other USB serial
drivers...

Signed-off-by: Werner Lemberg <wl@gnu.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/ark3116.c | 34 +++++++++++++++++++++++++++++++---
 1 file changed, 31 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c
index 970d9ef0a7a5..d37300e1811a 100644
--- a/drivers/usb/serial/ark3116.c
+++ b/drivers/usb/serial/ark3116.c
@@ -1,4 +1,7 @@
 /*
+ * Copyright (C) 2006
+ *   Simon Schulz (ark3116_driver <at> auctionant.de)
+ *
  * ark3116
  * - implements a driver for the arkmicro ark3116 chipset (vendor=0x6547,
  *   productid=0x0232) (used in a datacable called KQ-U8A)
@@ -8,8 +11,6 @@
  *
  *  - based on logs created by usbsnoopy
  *
- *  Author   : Simon Schulz [ark3116_driver<AT>auctionant.de]
- *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -22,6 +23,8 @@
 #include <linux/module.h>
 #include <linux/usb.h>
 #include <linux/usb/serial.h>
+#include <linux/serial.h>
+#include <asm/uaccess.h>
 
 
 static int debug;
@@ -379,7 +382,32 @@ static int ark3116_open(struct usb_serial_port *port, struct file *filp)
 static int ark3116_ioctl(struct usb_serial_port *port, struct file *file,
 			 unsigned int cmd, unsigned long arg)
 {
-	dbg("ioctl not supported yet...");
+	struct serial_struct serstruct;
+	void __user *user_arg = (void __user *)arg;
+
+	switch (cmd) {
+	case TIOCGSERIAL:
+		/* XXX: Some of these values are probably wrong. */
+		memset(&serstruct, 0, sizeof (serstruct));
+		serstruct.type = PORT_16654;
+		serstruct.line = port->serial->minor;
+		serstruct.port = port->number;
+		serstruct.custom_divisor = 0;
+		serstruct.baud_base = 460800;
+
+		if (copy_to_user(user_arg, &serstruct, sizeof (serstruct)))
+			return -EFAULT;
+
+		return 0;
+	case TIOCSSERIAL:
+		if (copy_from_user(&serstruct, user_arg, sizeof (serstruct)))
+			return -EFAULT;
+		return 0;
+	default:
+		dbg("%s cmd 0x%04x not supported", __FUNCTION__, cmd);
+		break;
+	}
+
 	return -ENOIOCTLCMD;
 }
 
-- 
cgit v1.2.1


From 988440e7e51c6f8061c98d03d618ba090e7b84ef Mon Sep 17 00:00:00 2001
From: Werner Lemberg <wl@gnu.org>
Date: Tue, 18 Jul 2006 17:00:22 +0200
Subject: USB: ark3116: Formatting cleanups

Formatting only.

Signed-off-by: Werner Lemberg <wl@gnu.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/ark3116.c | 199 +++++++++++++++++++++----------------------
 1 file changed, 99 insertions(+), 100 deletions(-)

diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c
index d37300e1811a..ca52f12f0e24 100644
--- a/drivers/usb/serial/ark3116.c
+++ b/drivers/usb/serial/ark3116.c
@@ -46,10 +46,10 @@ static inline void ARK3116_SND(struct usb_serial *serial, int seq,
 {
 	int result;
 	result = usb_control_msg(serial->dev,
-				 usb_sndctrlpipe(serial->dev,0),
+				 usb_sndctrlpipe(serial->dev, 0),
 				 request, requesttype, value, index,
-				 NULL,0x00, 1000);
-	dbg("%03d > ok",seq);
+				 NULL, 0x00, 1000);
+	dbg("%03d > ok", seq);
 }
 
 static inline void ARK3116_RCV(struct usb_serial *serial, int seq,
@@ -59,27 +59,25 @@ static inline void ARK3116_RCV(struct usb_serial *serial, int seq,
 {
 	int result;
 	result = usb_control_msg(serial->dev,
-			      usb_rcvctrlpipe(serial->dev,0),
-			      request, requesttype, value, index,
-			      buf, 0x0000001, 1000);
+				 usb_rcvctrlpipe(serial->dev, 0),
+				 request, requesttype, value, index,
+				 buf, 0x0000001, 1000);
 	if (result)
-		dbg("%03d < %d bytes [0x%02X]",seq, result, buf[0]);
+		dbg("%03d < %d bytes [0x%02X]", seq, result, buf[0]);
 	else
 		dbg("%03d < 0 bytes", seq);
 }
 
-
 static inline void ARK3116_RCV_QUIET(struct usb_serial *serial,
 				     __u8 request, __u8 requesttype,
 				     __u16 value, __u16 index, char *buf)
 {
 	usb_control_msg(serial->dev,
-			usb_rcvctrlpipe(serial->dev,0),
+			usb_rcvctrlpipe(serial->dev, 0),
 			request, requesttype, value, index,
 			buf, 0x0000001, 1000);
 }
 
-
 static int ark3116_attach(struct usb_serial *serial)
 {
 	char *buf;
@@ -87,10 +85,10 @@ static int ark3116_attach(struct usb_serial *serial)
 	int i;
 
 	for (i = 0; i < serial->num_ports; ++i) {
-		priv = kmalloc (sizeof (struct ark3116_private), GFP_KERNEL);
+		priv = kmalloc(sizeof (struct ark3116_private), GFP_KERNEL);
 		if (!priv)
 			goto cleanup;
-		memset (priv, 0x00, sizeof (struct ark3116_private));
+		memset(priv, 0x00, sizeof (struct ark3116_private));
 		spin_lock_init(&priv->lock);
 
 		usb_set_serial_port_data(serial->port[i], priv);
@@ -98,63 +96,62 @@ static int ark3116_attach(struct usb_serial *serial)
 
 	buf = kmalloc(1, GFP_KERNEL);
 	if (!buf) {
-		dbg("error kmalloc -> out of mem ?");
+		dbg("error kmalloc -> out of mem?");
 		goto cleanup;
 	}
 
 	/* 3 */
-	ARK3116_SND(serial, 3,0xFE,0x40,0x0008,0x0002);
-	ARK3116_SND(serial, 4,0xFE,0x40,0x0008,0x0001);
-	ARK3116_SND(serial, 5,0xFE,0x40,0x0000,0x0008);
-	ARK3116_SND(serial, 6,0xFE,0x40,0x0000,0x000B);
+	ARK3116_SND(serial, 3, 0xFE, 0x40, 0x0008, 0x0002);
+	ARK3116_SND(serial, 4, 0xFE, 0x40, 0x0008, 0x0001);
+	ARK3116_SND(serial, 5, 0xFE, 0x40, 0x0000, 0x0008);
+	ARK3116_SND(serial, 6, 0xFE, 0x40, 0x0000, 0x000B);
 
 	/* <-- seq7 */
-	ARK3116_RCV(serial, 7,0xFE,0xC0,0x0000,0x0003, 0x00, buf);
-	ARK3116_SND(serial, 8,0xFE,0x40,0x0080,0x0003);
-	ARK3116_SND(serial, 9,0xFE,0x40,0x001A,0x0000);
-	ARK3116_SND(serial,10,0xFE,0x40,0x0000,0x0001);
-	ARK3116_SND(serial,11,0xFE,0x40,0x0000,0x0003);
+	ARK3116_RCV(serial,  7, 0xFE, 0xC0, 0x0000, 0x0003, 0x00, buf);
+	ARK3116_SND(serial,  8, 0xFE, 0x40, 0x0080, 0x0003);
+	ARK3116_SND(serial,  9, 0xFE, 0x40, 0x001A, 0x0000);
+	ARK3116_SND(serial, 10, 0xFE, 0x40, 0x0000, 0x0001);
+	ARK3116_SND(serial, 11, 0xFE, 0x40, 0x0000, 0x0003);
 
 	/* <-- seq12 */
-	ARK3116_RCV(serial,12,0xFE,0xC0,0x0000,0x0004, 0x00, buf);
-	ARK3116_SND(serial,13,0xFE,0x40,0x0000,0x0004);
+	ARK3116_RCV(serial, 12, 0xFE, 0xC0, 0x0000, 0x0004, 0x00, buf);
+	ARK3116_SND(serial, 13, 0xFE, 0x40, 0x0000, 0x0004);
 
 	/* 14 */
-	ARK3116_RCV(serial,14,0xFE,0xC0,0x0000,0x0004, 0x00, buf);
-	ARK3116_SND(serial,15,0xFE,0x40,0x0000,0x0004);
+	ARK3116_RCV(serial, 14, 0xFE, 0xC0, 0x0000, 0x0004, 0x00, buf);
+	ARK3116_SND(serial, 15, 0xFE, 0x40, 0x0000, 0x0004);
 
 	/* 16 */
-	ARK3116_RCV(serial,16,0xFE,0xC0,0x0000,0x0004, 0x00, buf);
+	ARK3116_RCV(serial, 16, 0xFE, 0xC0, 0x0000, 0x0004, 0x00, buf);
 	/* --> seq17 */
-	ARK3116_SND(serial,17,0xFE,0x40,0x0001,0x0004);
+	ARK3116_SND(serial, 17, 0xFE, 0x40, 0x0001, 0x0004);
 
 	/* <-- seq18 */
-	ARK3116_RCV(serial,18,0xFE,0xC0,0x0000,0x0004, 0x01, buf);
+	ARK3116_RCV(serial, 18, 0xFE, 0xC0, 0x0000, 0x0004, 0x01, buf);
 
 	/* --> seq19 */
-	ARK3116_SND(serial,19,0xFE,0x40,0x0003,0x0004);
-
+	ARK3116_SND(serial, 19, 0xFE, 0x40, 0x0003, 0x0004);
 
 	/* <-- seq20 */
-	/* seems like serial port status info (RTS, CTS,...) */
-	/* returns modem control line status ?! */
-	ARK3116_RCV(serial,20,0xFE,0xC0,0x0000,0x0006, 0xFF, buf);
-
-	/* set 9600 baud & do some init ?! */
-	ARK3116_SND(serial,147,0xFE,0x40,0x0083,0x0003);
-	ARK3116_SND(serial,148,0xFE,0x40,0x0038,0x0000);
-	ARK3116_SND(serial,149,0xFE,0x40,0x0001,0x0001);
-	ARK3116_SND(serial,150,0xFE,0x40,0x0003,0x0003);
-	ARK3116_RCV(serial,151,0xFE,0xC0,0x0000,0x0004,0x03, buf);
-	ARK3116_SND(serial,152,0xFE,0x40,0x0000,0x0003);
-	ARK3116_RCV(serial,153,0xFE,0xC0,0x0000,0x0003,0x00, buf);
-	ARK3116_SND(serial,154,0xFE,0x40,0x0003,0x0003);
+	/* seems like serial port status info (RTS, CTS, ...) */
+	/* returns modem control line status?! */
+	ARK3116_RCV(serial, 20, 0xFE, 0xC0, 0x0000, 0x0006, 0xFF, buf);
+
+	/* set 9600 baud & do some init?! */
+	ARK3116_SND(serial, 147, 0xFE, 0x40, 0x0083, 0x0003);
+	ARK3116_SND(serial, 148, 0xFE, 0x40, 0x0038, 0x0000);
+	ARK3116_SND(serial, 149, 0xFE, 0x40, 0x0001, 0x0001);
+	ARK3116_SND(serial, 150, 0xFE, 0x40, 0x0003, 0x0003);
+	ARK3116_RCV(serial, 151, 0xFE, 0xC0, 0x0000, 0x0004, 0x03, buf);
+	ARK3116_SND(serial, 152, 0xFE, 0x40, 0x0000, 0x0003);
+	ARK3116_RCV(serial, 153, 0xFE, 0xC0, 0x0000, 0x0003, 0x00, buf);
+	ARK3116_SND(serial, 154, 0xFE, 0x40, 0x0003, 0x0003);
 
 	kfree(buf);
-	return(0);
+	return 0;
 
 cleanup:
-	for (--i; i>=0; --i)
+	for (--i; i >= 0; --i)
 		usb_set_serial_port_data(serial->port[i], NULL);
 	return -ENOMEM;
 }
@@ -183,7 +180,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
 		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL;
+		port->tty->termios->c_cflag = B9600 | CS8
+					      | CREAD | HUPCL | CLOCAL;
 		priv->termios_initialized = 1;
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -207,8 +205,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	}
 
 	/* set data bit count (8/7/6/5) */
-	if (cflag & CSIZE){
-		switch (cflag & CSIZE){
+	if (cflag & CSIZE) {
+		switch (cflag & CSIZE) {
 		case CS5:
 			config |= 0x00;
 			dbg("setting CS5");
@@ -222,7 +220,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 			dbg("setting CS7");
 			break;
 		default:
-			err ("CSIZE was set but not CS5-CS8, using CS8!");
+			err("CSIZE was set but not CS5-CS8, using CS8!");
+			/* fall through */
 		case CS8:
 			config |= 0x03;
 			dbg("setting CS8");
@@ -230,8 +229,8 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 		}
 	}
 
-	/* set parity (NONE,EVEN,ODD) */
-	if (cflag & PARENB){
+	/* set parity (NONE/EVEN/ODD) */
+	if (cflag & PARENB) {
 		if (cflag & PARODD) {
 			config |= 0x08;
 			dbg("setting parity to ODD");
@@ -243,20 +242,19 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 		dbg("setting parity to NONE");
 	}
 
-	/* SET STOPBIT (1/2) */
+	/* set stop bit (1/2) */
 	if (cflag & CSTOPB) {
 		config |= 0x04;
-		dbg ("setting 2 stop bits");
+		dbg("setting 2 stop bits");
 	} else {
-		dbg ("setting 1 stop bit");
+		dbg("setting 1 stop bit");
 	}
 
-
-	/* set baudrate: */
+	/* set baudrate */
 	baud = 0;
-	switch (cflag & CBAUD){
+	switch (cflag & CBAUD) {
 		case B0:
-			err("can't set 0baud, using 9600 instead");
+			err("can't set 0 baud, using 9600 instead");
 			break;
 		case B75:	baud = 75;	break;
 		case B150:	baud = 150;	break;
@@ -288,38 +286,40 @@ static void ark3116_set_termios(struct usb_serial_port *port,
 	 */
 	if (baud == 460800)
 		/* strange, for 460800 the formula is wrong
-		 * (dont use round(), then 9600baud is wrong) */
+		 * if using round() then 9600baud is wrong) */
 		ark3116_baud = 7;
 	else
 		ark3116_baud = 3000000 / baud;
 
 	/* ? */
-	ARK3116_RCV(serial,0,0xFE,0xC0,0x0000,0x0003, 0x03, buf);
+	ARK3116_RCV(serial, 0, 0xFE, 0xC0, 0x0000, 0x0003, 0x03, buf);
+
 	/* offset = buf[0]; */
 	/* offset = 0x03; */
-	/* dbg("using 0x%04X as target for 0x0003:",0x0080+offset); */
-
+	/* dbg("using 0x%04X as target for 0x0003:", 0x0080 + offset); */
 
 	/* set baudrate */
-	dbg("setting baudrate to %d (->reg=%d)",baud,ark3116_baud);
-	ARK3116_SND(serial,147,0xFE,0x40,0x0083,0x0003);
-	ARK3116_SND(serial,148,0xFE,0x40,(ark3116_baud & 0x00FF)   ,0x0000);
-	ARK3116_SND(serial,149,0xFE,0x40,(ark3116_baud & 0xFF00)>>8,0x0001);
-	ARK3116_SND(serial,150,0xFE,0x40,0x0003,0x0003);
+	dbg("setting baudrate to %d (->reg=%d)", baud, ark3116_baud);
+	ARK3116_SND(serial, 147, 0xFE, 0x40, 0x0083, 0x0003);
+	ARK3116_SND(serial, 148, 0xFE, 0x40,
+			    (ark3116_baud & 0x00FF), 0x0000);
+	ARK3116_SND(serial, 149, 0xFE, 0x40,
+			    (ark3116_baud & 0xFF00) >> 8, 0x0001);
+	ARK3116_SND(serial, 150, 0xFE, 0x40, 0x0003, 0x0003);
 
 	/* ? */
-	ARK3116_RCV(serial,151,0xFE,0xC0,0x0000,0x0004,0x03, buf);
-	ARK3116_SND(serial,152,0xFE,0x40,0x0000,0x0003);
+	ARK3116_RCV(serial, 151, 0xFE, 0xC0, 0x0000, 0x0004, 0x03, buf);
+	ARK3116_SND(serial, 152, 0xFE, 0x40, 0x0000, 0x0003);
 
 	/* set data bit count, stop bit count & parity: */
 	dbg("updating bit count, stop bit or parity (cfg=0x%02X)", config);
-	ARK3116_RCV(serial,153,0xFE,0xC0,0x0000,0x0003,0x00, buf);
-	ARK3116_SND(serial,154,0xFE,0x40,config,0x0003);
+	ARK3116_RCV(serial, 153, 0xFE, 0xC0, 0x0000, 0x0003, 0x00, buf);
+	ARK3116_SND(serial, 154, 0xFE, 0x40, config, 0x0003);
 
 	if (cflag & CRTSCTS)
-		dbg("CRTSCTS not supported by chipset ?!");
+		dbg("CRTSCTS not supported by chipset?!");
 
-	/* TEST ARK3116_SND(154,0xFE,0x40,0xFFFF, 0x0006); */
+	/* TEST ARK3116_SND(154, 0xFE, 0x40, 0xFFFF, 0x0006); */
 
 	kfree(buf);
 	return;
@@ -332,11 +332,11 @@ static int ark3116_open(struct usb_serial_port *port, struct file *filp)
 	char *buf;
 	int result = 0;
 
-	dbg("%s -  port %d", __FUNCTION__, port->number);
+	dbg("%s - port %d", __FUNCTION__, port->number);
 
 	buf = kmalloc(1, GFP_KERNEL);
 	if (!buf) {
-		dbg("error kmalloc -> out of mem ?");
+		dbg("error kmalloc -> out of mem?");
 		return -ENOMEM;
 	}
 
@@ -345,38 +345,37 @@ static int ark3116_open(struct usb_serial_port *port, struct file *filp)
 		return result;
 
 	/* open */
-	ARK3116_RCV(serial,111,0xFE,0xC0,0x0000,0x0003, 0x02, buf);
+	ARK3116_RCV(serial, 111, 0xFE, 0xC0, 0x0000, 0x0003, 0x02, buf);
 
-	ARK3116_SND(serial,112,0xFE,0x40,0x0082,0x0003);
-	ARK3116_SND(serial,113,0xFE,0x40,0x001A,0x0000);
-	ARK3116_SND(serial,114,0xFE,0x40,0x0000,0x0001);
-	ARK3116_SND(serial,115,0xFE,0x40,0x0002,0x0003);
+	ARK3116_SND(serial, 112, 0xFE, 0x40, 0x0082, 0x0003);
+	ARK3116_SND(serial, 113, 0xFE, 0x40, 0x001A, 0x0000);
+	ARK3116_SND(serial, 114, 0xFE, 0x40, 0x0000, 0x0001);
+	ARK3116_SND(serial, 115, 0xFE, 0x40, 0x0002, 0x0003);
 
-	ARK3116_RCV(serial,116,0xFE,0xC0,0x0000,0x0004, 0x03, buf);
-	ARK3116_SND(serial,117,0xFE,0x40,0x0002,0x0004);
+	ARK3116_RCV(serial, 116, 0xFE, 0xC0, 0x0000, 0x0004, 0x03, buf);
+	ARK3116_SND(serial, 117, 0xFE, 0x40, 0x0002, 0x0004);
 
-	ARK3116_RCV(serial,118,0xFE,0xC0,0x0000,0x0004, 0x02, buf);
-	ARK3116_SND(serial,119,0xFE,0x40,0x0000,0x0004);
+	ARK3116_RCV(serial, 118, 0xFE, 0xC0, 0x0000, 0x0004, 0x02, buf);
+	ARK3116_SND(serial, 119, 0xFE, 0x40, 0x0000, 0x0004);
 
-	ARK3116_RCV(serial,120,0xFE,0xC0,0x0000,0x0004, 0x00, buf);
+	ARK3116_RCV(serial, 120, 0xFE, 0xC0, 0x0000, 0x0004, 0x00, buf);
 
-	ARK3116_SND(serial,121,0xFE,0x40,0x0001,0x0004);
+	ARK3116_SND(serial, 121, 0xFE, 0x40, 0x0001, 0x0004);
 
-	ARK3116_RCV(serial,122,0xFE,0xC0,0x0000,0x0004, 0x01, buf);
+	ARK3116_RCV(serial, 122, 0xFE, 0xC0, 0x0000, 0x0004, 0x01, buf);
 
-	ARK3116_SND(serial,123,0xFE,0x40,0x0003,0x0004);
+	ARK3116_SND(serial, 123, 0xFE, 0x40, 0x0003, 0x0004);
 
-	/* returns different values (control lines ?!) */
-	ARK3116_RCV(serial,124,0xFE,0xC0,0x0000,0x0006, 0xFF, buf);
+	/* returns different values (control lines?!) */
+	ARK3116_RCV(serial, 124, 0xFE, 0xC0, 0x0000, 0x0006, 0xFF, buf);
 
-	/* initialise termios: */
+	/* initialise termios */
 	if (port->tty)
 		ark3116_set_termios(port, &tmp_termios);
 
 	kfree(buf);
 
 	return result;
-
 }
 
 static int ark3116_ioctl(struct usb_serial_port *port, struct file *file,
@@ -417,7 +416,7 @@ static int ark3116_tiocmget(struct usb_serial_port *port, struct file *file)
 	char *buf;
 	char temp;
 
-	/* seems like serial port status info (RTS, CTS,...) is stored
+	/* seems like serial port status info (RTS, CTS, ...) is stored
 	 * in reg(?) 0x0006
 	 * pcb connection point 11 = GND -> sets bit4 of response
 	 * pcb connection point  7 = GND -> sets bit6 of response
@@ -429,16 +428,16 @@ static int ark3116_tiocmget(struct usb_serial_port *port, struct file *file)
 		return -ENOMEM;
 	}
 
-	/* read register: */
-	ARK3116_RCV_QUIET(serial,0xFE,0xC0,0x0000,0x0006,buf);
+	/* read register */
+	ARK3116_RCV_QUIET(serial, 0xFE, 0xC0, 0x0000, 0x0006, buf);
 	temp = buf[0];
 	kfree(buf);
 
-	/* i do not really know if bit4=CTS and bit6=DSR... was just a
-	 * quick guess !!
+	/* i do not really know if bit4=CTS and bit6=DSR... just a
+	 * quick guess!
 	 */
-	return  (temp & (1<<4) ? TIOCM_CTS : 0) |
-	        (temp & (1<<6) ? TIOCM_DSR : 0);
+	return (temp & (1<<4) ? TIOCM_CTS : 0)
+	       | (temp & (1<<6) ? TIOCM_DSR : 0);
 }
 
 static struct usb_driver ark3116_driver = {
-- 
cgit v1.2.1


From b94badbb47cb50f4fca8440efdaa8ebd32604fe4 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor@insightbb.com>
Date: Tue, 1 Aug 2006 22:33:34 -0400
Subject: USB: Make usb_buffer_free() NULL-safe

kfree() handles NULL arguments which is handy in error handling paths as one
does need to  insert bunch of ifs. How about making usb_buffer_free() do the
same?

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 184c24660a4c..ab766e0fe4ef 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -761,7 +761,9 @@ void usb_buffer_free (
 )
 {
 	if (!dev || !dev->bus || !dev->bus->op || !dev->bus->op->buffer_free)
-	    	return;
+		return;
+	if (!addr)
+		return;
 	dev->bus->op->buffer_free (dev->bus, size, addr, dma);
 }
 
-- 
cgit v1.2.1


From b6eb2d84d2bb01e9fcc46a032a3429b4747b1c47 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Thu, 6 Jul 2006 15:37:42 -0400
Subject: usbcore: add configuration_string to attribute group

This patch (as737b) does a very small cleanup of core/sysfs.c by adding
the configuration_string attribute file to the existing attribute group
instead of treating it separately.  It doesn't need this separate
treatment because unlike the other device string attributes, it changes
along with the active configuration.

The patch also fixes a simple typo (which, oddly enough, doesn't seem to
bother the compiler).

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/sysfs.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/sysfs.c b/drivers/usb/core/sysfs.c
index dec973affb0f..cd2286246f6a 100644
--- a/drivers/usb/core/sysfs.c
+++ b/drivers/usb/core/sysfs.c
@@ -60,7 +60,7 @@ static ssize_t
 set_bConfigurationValue (struct device *dev, struct device_attribute *attr,
 		const char *buf, size_t count)
 {
-	struct usb_device	*udev = udev = to_usb_device (dev);
+	struct usb_device	*udev = to_usb_device (dev);
 	int			config, value;
 
 	if (sscanf (buf, "%u", &config) != 1 || config > 255)
@@ -186,6 +186,7 @@ usb_descriptor_attr (bMaxPacketSize0, "%d\n")
 
 static struct attribute *dev_attrs[] = {
 	/* current configuration's attributes */
+	&dev_attr_configuration.attr,
 	&dev_attr_bNumInterfaces.attr,
 	&dev_attr_bConfigurationValue.attr,
 	&dev_attr_bmAttributes.attr,
@@ -221,7 +222,6 @@ void usb_create_sysfs_dev_files (struct usb_device *udev)
 		device_create_file (dev, &dev_attr_product);
 	if (udev->serial)
 		device_create_file (dev, &dev_attr_serial);
-	device_create_file (dev, &dev_attr_configuration);
 	usb_create_ep_files(dev, &udev->ep0, udev);
 }
 
@@ -238,7 +238,6 @@ void usb_remove_sysfs_dev_files (struct usb_device *udev)
 		device_remove_file(dev, &dev_attr_product);
 	if (udev->serial)
 		device_remove_file(dev, &dev_attr_serial);
-	device_remove_file (dev, &dev_attr_configuration);
 }
 
 /* Interface fields */
-- 
cgit v1.2.1


From d5176b413dcce85334e270021fc0d723d1714c84 Mon Sep 17 00:00:00 2001
From: Sean Young <sean@mess.org>
Date: Sun, 9 Jul 2006 13:01:02 +0000
Subject: USB: Add driver for PhidgetMotorControl

This driver add support for the Phidgets Inc., MotorControl via sysfs. Also
some minor fixes for the InterfaceKit.

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/hid-core.c           |  12 +-
 drivers/usb/misc/Kconfig               |  16 +-
 drivers/usb/misc/Makefile              |   1 +
 drivers/usb/misc/phidgetmotorcontrol.c | 427 +++++++++++++++++++++++++++++++++
 4 files changed, 451 insertions(+), 5 deletions(-)
 create mode 100644 drivers/usb/misc/phidgetmotorcontrol.c

diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index 3305fb6079eb..8f8d4af3f6ef 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1535,13 +1535,17 @@ void hid_init_reports(struct hid_device *hid)
 #define USB_VENDOR_ID_GLAB		0x06c2
 #define USB_DEVICE_ID_4_PHIDGETSERVO_30	0x0038
 #define USB_DEVICE_ID_1_PHIDGETSERVO_30	0x0039
-#define USB_DEVICE_ID_8_8_8_IF_KIT	0x0045
 #define USB_DEVICE_ID_0_0_4_IF_KIT	0x0040
+#define USB_DEVICE_ID_0_16_16_IF_KIT	0x0044
+#define USB_DEVICE_ID_8_8_8_IF_KIT	0x0045
+#define USB_DEVICE_ID_0_8_7_IF_KIT	0x0051
 #define USB_DEVICE_ID_0_8_8_IF_KIT	0x0053
+#define USB_DEVICE_ID_PHIDGET_MOTORCONTROL	0x0058
 
 #define USB_VENDOR_ID_WISEGROUP		0x0925
 #define USB_DEVICE_ID_1_PHIDGETSERVO_20	0x8101
 #define USB_DEVICE_ID_4_PHIDGETSERVO_20	0x8104
+#define USB_DEVICE_ID_8_8_4_IF_KIT	0x8201
 #define USB_DEVICE_ID_DUAL_USB_JOYPAD   0x8866
 
 #define USB_VENDOR_ID_WISEGROUP_LTD	0x6677
@@ -1620,9 +1624,12 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_ESSENTIAL_REALITY, USB_DEVICE_ID_ESSENTIAL_REALITY_P5, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_4_PHIDGETSERVO_30, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_1_PHIDGETSERVO_30, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_8_8_8_IF_KIT, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_0_0_4_IF_KIT, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_0_16_16_IF_KIT, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_8_8_8_IF_KIT, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_0_8_7_IF_KIT, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_0_8_8_IF_KIT, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_GLAB, USB_DEVICE_ID_PHIDGET_MOTORCONTROL, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_POWERMATE, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_SOUNDKNOB, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GTCO, USB_DEVICE_ID_GTCO_90, HID_QUIRK_IGNORE },
@@ -1701,6 +1708,7 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_CYCLOPS, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_WISEGROUP, USB_DEVICE_ID_4_PHIDGETSERVO_20, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_WISEGROUP, USB_DEVICE_ID_1_PHIDGETSERVO_20, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_WISEGROUP, USB_DEVICE_ID_8_8_4_IF_KIT, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_YEALINK, USB_DEVICE_ID_YEALINK_P1K_P4K_B2K, HID_QUIRK_IGNORE },
 
 	{ USB_VENDOR_ID_ACECAD, USB_DEVICE_ID_ACECAD_FLAIR, HID_QUIRK_IGNORE },
diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
index 88928a4be805..8affc137d618 100644
--- a/drivers/usb/misc/Kconfig
+++ b/drivers/usb/misc/Kconfig
@@ -116,15 +116,25 @@ config USB_CYTHERM
 	  module will be called cytherm.
 
 config USB_PHIDGETKIT
-	tristate "USB PhidgetKit support"
+	tristate "USB PhidgetInterfaceKit support"
 	depends on USB
 	help
-	  Say Y here if you want to connect a PhidgetKit USB device from
-	  Phidgets Inc.
+	  Say Y here if you want to connect a PhidgetInterfaceKit USB device
+	  from Phidgets Inc.
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called phidgetkit.
 
+config USB_PHIDGETMOTORCONTROL
+	tristate "USB PhidgetMotorControl support"
+	depends on USB
+	help
+	  Say Y here if you want to connect a PhidgetMotorControl USB device
+	  from Phidgets Inc.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called phidgetmotorcontrol.
+
 config USB_PHIDGETSERVO
 	tristate "USB PhidgetServo support"
 	depends on USB
diff --git a/drivers/usb/misc/Makefile b/drivers/usb/misc/Makefile
index 2927260c5812..5f329176451a 100644
--- a/drivers/usb/misc/Makefile
+++ b/drivers/usb/misc/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_USB_LD)		+= ldusb.o
 obj-$(CONFIG_USB_LED)		+= usbled.o
 obj-$(CONFIG_USB_LEGOTOWER)	+= legousbtower.o
 obj-$(CONFIG_USB_PHIDGETKIT)	+= phidgetkit.o
+obj-$(CONFIG_USB_PHIDGETMOTORCONTROL)	+= phidgetmotorcontrol.o
 obj-$(CONFIG_USB_PHIDGETSERVO)	+= phidgetservo.o
 obj-$(CONFIG_USB_RIO500)	+= rio500.o
 obj-$(CONFIG_USB_TEST)		+= usbtest.o
diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c
new file mode 100644
index 000000000000..2972dc2eb27f
--- /dev/null
+++ b/drivers/usb/misc/phidgetmotorcontrol.c
@@ -0,0 +1,427 @@
+/*
+ * USB Phidget MotorControl driver
+ *
+ * Copyright (C) 2006  Sean Young <sean@mess.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
+#define DRIVER_AUTHOR "Sean Young <sean@mess.org>"
+#define DRIVER_DESC "USB PhidgetMotorControl Driver"
+
+#define USB_VENDOR_ID_GLAB		0x06c2
+#define USB_DEVICE_ID_MOTORCONTROL	0x0058
+
+#define URB_INT_SIZE			8
+
+struct motorcontrol {
+	struct usb_device *udev;
+	struct usb_interface *intf;
+	u8 inputs[4];
+	s8 desired_speed[2];
+	s8 speed[2];
+	s16 _current[2];
+	s8 acceleration[2];
+	struct urb *irq;
+	unsigned char *data;
+	dma_addr_t data_dma;
+
+	struct work_struct do_notify;
+	unsigned long input_events;
+	unsigned long speed_events;
+	unsigned long exceed_events;
+};
+
+static struct usb_device_id id_table[] = {
+	{ USB_DEVICE(USB_VENDOR_ID_GLAB, USB_DEVICE_ID_MOTORCONTROL) },
+	{}
+};
+MODULE_DEVICE_TABLE(usb, id_table);
+
+static int set_motor(struct motorcontrol *mc, int motor)
+{
+	u8 *buffer;
+	int speed, speed2, acceleration;
+	int retval;
+
+	buffer = kzalloc(8, GFP_KERNEL);
+	if (!buffer) {
+		dev_err(&mc->intf->dev, "%s - out of memory\n", __FUNCTION__);
+		return -ENOMEM;
+	}
+
+	acceleration = mc->acceleration[motor] * 10;
+	/* -127 <= speed <= 127 */
+	speed = (mc->desired_speed[motor] * 127) / 100;
+	/* -0x7300 <= speed2 <= 0x7300 */
+	speed2 = (mc->desired_speed[motor] * 230 * 128) / 100;
+
+	buffer[0] = motor;
+	buffer[1] = speed;
+	buffer[2] = acceleration >> 8;
+	buffer[3] = acceleration;
+	buffer[4] = speed2 >> 8;
+	buffer[5] = speed2;
+
+	retval = usb_control_msg(mc->udev,
+			 usb_sndctrlpipe(mc->udev, 0),
+			 0x09, 0x21, 0x0200, 0x0000, buffer, 8, 2000);
+
+	if (retval != 8)
+		dev_err(&mc->intf->dev, "usb_control_msg returned %d\n",
+				retval);
+	kfree(buffer);
+
+	return retval < 0 ? retval : 0;
+}
+
+static void motorcontrol_irq(struct urb *urb, struct pt_regs *regs)
+{
+	struct motorcontrol *mc = urb->context;
+	unsigned char *buffer = mc->data;
+	int i, level;
+	int status;
+
+	switch (urb->status) {
+	case 0:			/* success */
+		break;
+	case -ECONNRESET:	/* unlink */
+	case -ENOENT:
+	case -ESHUTDOWN:
+		return;
+	/* -EPIPE:  should clear the halt */
+	default:		/* error */
+		goto resubmit;
+	}
+
+	/* digital inputs */
+	for (i=0; i<4; i++) {
+		level = (buffer[0] >> i) & 1;
+		if (mc->inputs[i] != level) {
+			mc->inputs[i] = level;
+			set_bit(i, &mc->input_events);
+		}
+	}
+
+	/* motor speed */
+	if (buffer[2] == 0) {
+		for (i=0; i<2; i++) {
+		level = ((s8)buffer[4+i]) * 100 / 127;
+			if (mc->speed[i] != level) {
+				mc->speed[i] = level;
+				set_bit(i, &mc->speed_events);
+			}
+		}
+	} else {
+		int index = buffer[3] & 1;
+
+		level = ((s8)buffer[4] << 8) | buffer[5];
+		level = level * 100 / 29440;
+		if (mc->speed[index] != level) {
+			mc->speed[index] = level;
+			set_bit(index, &mc->speed_events);
+		}
+
+		level = ((s8)buffer[6] << 8) | buffer[7];
+		mc->_current[index] = level * 100 / 1572;
+	}
+
+	if (buffer[1] & 1)
+		set_bit(0, &mc->exceed_events);
+
+	if (buffer[1] & 2)
+		set_bit(1, &mc->exceed_events);
+
+	if (mc->input_events || mc->exceed_events || mc->speed_events)
+		schedule_work(&mc->do_notify);
+
+resubmit:
+	status = usb_submit_urb(urb, SLAB_ATOMIC);
+	if (status)
+		dev_err(&mc->intf->dev,
+			"can't resubmit intr, %s-%s/motorcontrol0, status %d",
+			mc->udev->bus->bus_name,
+			mc->udev->devpath, status);
+}
+
+static void do_notify(void *data)
+{
+	struct motorcontrol *mc = data;
+	int i;
+	char sysfs_file[8];
+
+	for (i=0; i<4; i++) {
+		if (test_and_clear_bit(i, &mc->input_events)) {
+			sprintf(sysfs_file, "input%d", i);
+			sysfs_notify(&mc->intf->dev.kobj, NULL, sysfs_file);
+		}
+	}
+
+	for (i=0; i<2; i++) {
+		if (test_and_clear_bit(i, &mc->speed_events)) {
+			sprintf(sysfs_file, "speed%d", i);
+			sysfs_notify(&mc->intf->dev.kobj, NULL, sysfs_file);
+		}
+	}
+
+	for (i=0; i<2; i++) {
+		if (test_and_clear_bit(i, &mc->exceed_events))
+			dev_warn(&mc->intf->dev,
+				"motor #%d exceeds 1.5 Amp current limit\n", i);
+	}
+}
+
+#define show_set_speed(value)		\
+static ssize_t set_speed##value(struct device *dev, 			\
+	struct device_attribute *attr, const char *buf,	size_t count)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	int speed;							\
+	int retval;							\
+									\
+	if (sscanf(buf, "%d", &speed) < 1)				\
+		return -EINVAL;						\
+									\
+	if (speed < -100 || speed > 100)				\
+		return -EINVAL;						\
+									\
+	mc->desired_speed[value] = speed;				\
+									\
+	retval = set_motor(mc, value);					\
+									\
+	return retval ? retval : count;					\
+}									\
+									\
+static ssize_t show_speed##value(struct device *dev, 			\
+			struct device_attribute *attr, char *buf)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+									\
+	return sprintf(buf, "%d\n", mc->speed[value]);			\
+}									\
+static DEVICE_ATTR(speed##value, S_IWUGO | S_IRUGO,			\
+		show_speed##value, set_speed##value);
+show_set_speed(0);
+show_set_speed(1);
+
+#define show_set_acceleration(value)		\
+static ssize_t set_acceleration##value(struct device *dev, 		\
+	struct device_attribute *attr, const char *buf,	size_t count)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	int acceleration;						\
+	int retval;							\
+									\
+	if (sscanf(buf, "%d", &acceleration) < 1)			\
+		return -EINVAL;						\
+									\
+	if (acceleration < 0 || acceleration > 100)			\
+		return -EINVAL;						\
+									\
+	mc->acceleration[value] = acceleration;				\
+									\
+	retval = set_motor(mc, value);					\
+									\
+	return retval ? retval : count;					\
+}									\
+									\
+static ssize_t show_acceleration##value(struct device *dev, 		\
+			struct device_attribute *attr, char *buf)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+									\
+	return sprintf(buf, "%d\n", mc->acceleration[value]);		\
+}									\
+static DEVICE_ATTR(acceleration##value, S_IWUGO | S_IRUGO,		\
+		show_acceleration##value, set_acceleration##value);
+show_set_acceleration(0);
+show_set_acceleration(1);
+
+#define show_current(value)	\
+static ssize_t show_current##value(struct device *dev, 			\
+			struct device_attribute *attr, char *buf)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+									\
+	return sprintf(buf, "%dmA\n", (int)mc->_current[value]);	\
+}									\
+static DEVICE_ATTR(current##value, S_IRUGO, show_current##value, NULL);
+
+show_current(0);
+show_current(1);
+
+#define show_input(value)	\
+static ssize_t show_input##value(struct device *dev, 			\
+			struct device_attribute *attr, char *buf)	\
+{									\
+	struct usb_interface *intf = to_usb_interface(dev);		\
+	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+									\
+	return sprintf(buf, "%d\n", (int)mc->inputs[value]);		\
+}									\
+static DEVICE_ATTR(input##value, S_IRUGO, show_input##value, NULL);
+
+show_input(0);
+show_input(1);
+show_input(2);
+show_input(3);
+
+static int motorcontrol_probe(struct usb_interface *intf, const struct usb_device_id *id)
+{
+	struct usb_device *dev = interface_to_usbdev(intf);
+	struct usb_host_interface *interface;
+	struct usb_endpoint_descriptor *endpoint;
+	struct motorcontrol *mc;
+	int pipe, maxp, rc = -ENOMEM;
+
+	interface = intf->cur_altsetting;
+	if (interface->desc.bNumEndpoints != 1)
+		return -ENODEV;
+
+	endpoint = &interface->endpoint[0].desc;
+	if (!(endpoint->bEndpointAddress & 0x80))
+		return -ENODEV;
+
+	/*
+	 * bmAttributes
+	 */
+	pipe = usb_rcvintpipe(dev, endpoint->bEndpointAddress);
+	maxp = usb_maxpacket(dev, pipe, usb_pipeout(pipe));
+
+	mc = kzalloc(sizeof(*mc), GFP_KERNEL);
+	if (!mc)
+		goto out;
+
+	mc->data = usb_buffer_alloc(dev, URB_INT_SIZE, SLAB_ATOMIC, &mc->data_dma);
+	if (!mc->data)
+		goto out;
+
+	mc->irq = usb_alloc_urb(0, GFP_KERNEL);
+	if (!mc->irq)
+		goto out;
+
+	mc->udev = usb_get_dev(dev);
+	mc->intf = intf;
+	mc->acceleration[0] = mc->acceleration[1] = 10;
+	INIT_WORK(&mc->do_notify, do_notify, mc);
+	usb_fill_int_urb(mc->irq, mc->udev, pipe, mc->data,
+			maxp > URB_INT_SIZE ? URB_INT_SIZE : maxp,
+			motorcontrol_irq, mc, endpoint->bInterval);
+	mc->irq->transfer_dma = mc->data_dma;
+	mc->irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+	usb_set_intfdata(intf, mc);
+
+	if (usb_submit_urb(mc->irq, GFP_KERNEL)) {
+		rc = -EIO;
+		goto out;
+	}
+
+	device_create_file(&intf->dev, &dev_attr_input0);
+	device_create_file(&intf->dev, &dev_attr_input1);
+	device_create_file(&intf->dev, &dev_attr_input2);
+	device_create_file(&intf->dev, &dev_attr_input3);
+
+	device_create_file(&intf->dev, &dev_attr_speed0);
+	device_create_file(&intf->dev, &dev_attr_speed1);
+
+	device_create_file(&intf->dev, &dev_attr_acceleration0);
+	device_create_file(&intf->dev, &dev_attr_acceleration1);
+
+	device_create_file(&intf->dev, &dev_attr_current0);
+	device_create_file(&intf->dev, &dev_attr_current1);
+
+	dev_info(&intf->dev, "USB Phidget MotorControl attached\n");
+
+	return 0;
+
+out:
+	if (mc) {
+		if (mc->irq)
+			usb_free_urb(mc->irq);
+		if (mc->data)
+			usb_buffer_free(dev, URB_INT_SIZE, mc->data, mc->data_dma);
+		kfree(mc);
+	}
+
+	return rc;
+}
+
+static void motorcontrol_disconnect(struct usb_interface *interface)
+{
+	struct motorcontrol *mc;
+
+	mc = usb_get_intfdata(interface);
+	usb_set_intfdata(interface, NULL);
+	if (!mc)
+		return;
+
+	usb_kill_urb(mc->irq);
+	usb_free_urb(mc->irq);
+	usb_buffer_free(mc->udev, URB_INT_SIZE, mc->data, mc->data_dma);
+
+	cancel_delayed_work(&mc->do_notify);
+
+	device_remove_file(&interface->dev, &dev_attr_input0);
+	device_remove_file(&interface->dev, &dev_attr_input1);
+	device_remove_file(&interface->dev, &dev_attr_input2);
+	device_remove_file(&interface->dev, &dev_attr_input3);
+
+	device_remove_file(&interface->dev, &dev_attr_speed0);
+	device_remove_file(&interface->dev, &dev_attr_speed1);
+
+	device_remove_file(&interface->dev, &dev_attr_acceleration0);
+	device_remove_file(&interface->dev, &dev_attr_acceleration1);
+
+	device_remove_file(&interface->dev, &dev_attr_current0);
+	device_remove_file(&interface->dev, &dev_attr_current1);
+
+	dev_info(&interface->dev, "USB Phidget MotorControl disconnected\n");
+
+	usb_put_dev(mc->udev);
+	kfree(mc);
+}
+
+static struct usb_driver motorcontrol_driver = {
+	.name = "phidgetmotorcontrol",
+	.probe = motorcontrol_probe,
+	.disconnect = motorcontrol_disconnect,
+	.id_table = id_table
+};
+
+static int __init motorcontrol_init(void)
+{
+	int retval = 0;
+
+	retval = usb_register(&motorcontrol_driver);
+	if (retval)
+		err("usb_register failed. Error number %d", retval);
+
+	return retval;
+}
+
+static void __exit motorcontrol_exit(void)
+{
+	usb_deregister(&motorcontrol_driver);
+}
+
+module_init(motorcontrol_init);
+module_exit(motorcontrol_exit);
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 912b24c333843514ff77ed88961c6945f0f286ce Mon Sep 17 00:00:00 2001
From: Sean Young <sean@mess.org>
Date: Mon, 10 Jul 2006 09:56:25 +0000
Subject: USB: Put phidgets driver in a sysfs class

This patch creates a device class phidget and add the phidget drivers to
them.

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/Kconfig               |  13 +-
 drivers/usb/misc/Makefile              |   1 +
 drivers/usb/misc/phidget.c             |  43 ++++++
 drivers/usb/misc/phidget.h             |  12 ++
 drivers/usb/misc/phidgetkit.c          | 258 ++++++++++++++++++---------------
 drivers/usb/misc/phidgetmotorcontrol.c | 126 ++++++++++------
 drivers/usb/misc/phidgetservo.c        |  71 +++++----
 7 files changed, 334 insertions(+), 190 deletions(-)
 create mode 100644 drivers/usb/misc/phidget.c
 create mode 100644 drivers/usb/misc/phidget.h

diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
index 8affc137d618..20539cf9394b 100644
--- a/drivers/usb/misc/Kconfig
+++ b/drivers/usb/misc/Kconfig
@@ -115,9 +115,16 @@ config USB_CYTHERM
 	  To compile this driver as a module, choose M here: the
 	  module will be called cytherm.
 
+config USB_PHIDGET
+	tristate "USB Phidgets drivers"
+	depends on USB
+	help
+	  Say Y here to enable the various drivers for devices from
+	  Phidgets inc.
+
 config USB_PHIDGETKIT
 	tristate "USB PhidgetInterfaceKit support"
-	depends on USB
+	depends on USB_PHIDGET
 	help
 	  Say Y here if you want to connect a PhidgetInterfaceKit USB device
 	  from Phidgets Inc.
@@ -127,7 +134,7 @@ config USB_PHIDGETKIT
 
 config USB_PHIDGETMOTORCONTROL
 	tristate "USB PhidgetMotorControl support"
-	depends on USB
+	depends on USB_PHIDGET
 	help
 	  Say Y here if you want to connect a PhidgetMotorControl USB device
 	  from Phidgets Inc.
@@ -137,7 +144,7 @@ config USB_PHIDGETMOTORCONTROL
 
 config USB_PHIDGETSERVO
 	tristate "USB PhidgetServo support"
-	depends on USB
+	depends on USB_PHIDGET
 	help
 	  Say Y here if you want to connect an 1 or 4 Motor PhidgetServo 
 	  servo controller version 2.0 or 3.0.
diff --git a/drivers/usb/misc/Makefile b/drivers/usb/misc/Makefile
index 5f329176451a..d12a84a1c0a4 100644
--- a/drivers/usb/misc/Makefile
+++ b/drivers/usb/misc/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_USB_LCD)		+= usblcd.o
 obj-$(CONFIG_USB_LD)		+= ldusb.o
 obj-$(CONFIG_USB_LED)		+= usbled.o
 obj-$(CONFIG_USB_LEGOTOWER)	+= legousbtower.o
+obj-$(CONFIG_USB_PHIDGET)	+= phidget.o
 obj-$(CONFIG_USB_PHIDGETKIT)	+= phidgetkit.o
 obj-$(CONFIG_USB_PHIDGETMOTORCONTROL)	+= phidgetmotorcontrol.o
 obj-$(CONFIG_USB_PHIDGETSERVO)	+= phidgetservo.o
diff --git a/drivers/usb/misc/phidget.c b/drivers/usb/misc/phidget.c
new file mode 100644
index 000000000000..735ed33f4f7f
--- /dev/null
+++ b/drivers/usb/misc/phidget.c
@@ -0,0 +1,43 @@
+/*
+ * USB Phidgets class
+ *
+ * Copyright (C) 2006  Sean Young <sean@mess.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/err.h>
+#include <linux/device.h>
+
+struct class *phidget_class;
+
+static int __init init_phidget(void)
+{
+	phidget_class = class_create(THIS_MODULE, "phidget");
+
+	if (IS_ERR(phidget_class))
+		return PTR_ERR(phidget_class);
+
+	return 0;
+}
+
+static void __exit cleanup_phidget(void)
+{
+	class_destroy(phidget_class);
+}
+
+EXPORT_SYMBOL_GPL(phidget_class);
+
+module_init(init_phidget);
+module_exit(cleanup_phidget);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Sean Young <sean@mess.org>");
+MODULE_DESCRIPTION("Container module for phidget class");
+
diff --git a/drivers/usb/misc/phidget.h b/drivers/usb/misc/phidget.h
new file mode 100644
index 000000000000..c4011907d431
--- /dev/null
+++ b/drivers/usb/misc/phidget.h
@@ -0,0 +1,12 @@
+/*
+ * USB Phidgets class
+ *
+ * Copyright (C) 2006  Sean Young <sean@mess.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+extern struct class *phidget_class;
diff --git a/drivers/usb/misc/phidgetkit.c b/drivers/usb/misc/phidgetkit.c
index bfbbbfbb92bc..3f508957def9 100644
--- a/drivers/usb/misc/phidgetkit.c
+++ b/drivers/usb/misc/phidgetkit.c
@@ -20,6 +20,8 @@
 #include <linux/module.h>
 #include <linux/usb.h>
 
+#include "phidget.h"
+
 #define DRIVER_AUTHOR "Sean Young <sean@mess.org>"
 #define DRIVER_DESC "USB PhidgetInterfaceKit Driver"
 
@@ -57,11 +59,15 @@ ifkit(8, 8, 4, 0);
 ifkit(0, 8, 8, 1);
 ifkit(0, 16, 16, 0);
 
+static unsigned long device_no;
+
 struct interfacekit {
 	struct usb_device *udev;
 	struct usb_interface *intf;
 	struct driver_interfacekit *ifkit;
+	struct device *dev;
 	unsigned long outputs;
+	int dev_no;
 	u8 inputs[MAX_INTERFACES];
 	u16 sensors[MAX_INTERFACES];
 	u8 lcd_files_on;
@@ -180,21 +186,21 @@ exit:
 }
 
 #define set_lcd_line(number)	\
-static ssize_t lcd_line_##number(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)	\
-{											\
-	struct usb_interface *intf = to_usb_interface(dev);				\
-	struct interfacekit *kit = usb_get_intfdata(intf);				\
-	change_string(kit, buf, number - 1);						\
-	return count;									\
-}											\
+static ssize_t lcd_line_##number(struct device *dev,			\
+					struct device_attribute *attr,	\
+					const char *buf, size_t count)	\
+{									\
+	struct interfacekit *kit = dev_get_drvdata(dev);		\
+	change_string(kit, buf, number - 1);				\
+	return count;							\
+}									\
 static DEVICE_ATTR(lcd_line_##number, S_IWUGO, NULL, lcd_line_##number);
 set_lcd_line(1);
 set_lcd_line(2);
 
 static ssize_t set_backlight(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct usb_interface *intf = to_usb_interface(dev);
-	struct interfacekit *kit = usb_get_intfdata(intf);
+	struct interfacekit *kit = dev_get_drvdata(dev);
 	int enabled;
 	unsigned char *buffer;
 	int retval = -ENOMEM;
@@ -232,16 +238,15 @@ static void remove_lcd_files(struct interfacekit *kit)
 {
 	if (kit->lcd_files_on) {
 		dev_dbg(&kit->udev->dev, "Removing lcd files\n");
-		device_remove_file(&kit->intf->dev, &dev_attr_lcd_line_1);
-		device_remove_file(&kit->intf->dev, &dev_attr_lcd_line_2);
-		device_remove_file(&kit->intf->dev, &dev_attr_backlight);
+		device_remove_file(kit->dev, &dev_attr_lcd_line_1);
+		device_remove_file(kit->dev, &dev_attr_lcd_line_2);
+		device_remove_file(kit->dev, &dev_attr_backlight);
 	}
 }
 
 static ssize_t enable_lcd_files(struct device *dev, struct device_attribute *attr, const char *buf, size_t count)
 {
-	struct usb_interface *intf = to_usb_interface(dev);
-	struct interfacekit *kit = usb_get_intfdata(intf);
+	struct interfacekit *kit = dev_get_drvdata(dev);
 	int enable;
 	
 	if (kit->ifkit->has_lcd == 0)
@@ -253,9 +258,9 @@ static ssize_t enable_lcd_files(struct device *dev, struct device_attribute *att
 	if (enable) {
 		if (!kit->lcd_files_on) {
 			dev_dbg(&kit->udev->dev, "Adding lcd files\n");
-			device_create_file(&kit->intf->dev, &dev_attr_lcd_line_1);
-			device_create_file(&kit->intf->dev, &dev_attr_lcd_line_2);
-			device_create_file(&kit->intf->dev, &dev_attr_backlight);
+			device_create_file(kit->dev, &dev_attr_lcd_line_1);
+			device_create_file(kit->dev, &dev_attr_lcd_line_2);
+			device_create_file(kit->dev, &dev_attr_backlight);
 			kit->lcd_files_on = 1;
 		}
 	} else {
@@ -362,24 +367,24 @@ static void do_notify(void *data)
 	for (i=0; i<kit->ifkit->inputs; i++) {
 		if (test_and_clear_bit(i, &kit->input_events)) {
 			sprintf(sysfs_file, "input%d", i + 1);
-			sysfs_notify(&kit->intf->dev.kobj, NULL, sysfs_file);
+			sysfs_notify(&kit->dev->kobj, NULL, sysfs_file);
 		}
 	}
 
 	for (i=0; i<kit->ifkit->sensors; i++) {
 		if (test_and_clear_bit(i, &kit->sensor_events)) {
 			sprintf(sysfs_file, "sensor%d", i + 1);
-			sysfs_notify(&kit->intf->dev.kobj, NULL, sysfs_file);
+			sysfs_notify(&kit->dev->kobj, NULL, sysfs_file);
 		}
 	}
 }
 
 #define show_set_output(value)		\
-static ssize_t set_output##value(struct device *dev, struct device_attribute *attr, const char *buf,	\
-							size_t count)	\
+static ssize_t set_output##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					const char *buf, size_t count)	\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct interfacekit *kit = usb_get_intfdata(intf);		\
+	struct interfacekit *kit = dev_get_drvdata(dev);		\
 	int enabled;							\
 	int retval;							\
 									\
@@ -391,10 +396,11 @@ static ssize_t set_output##value(struct device *dev, struct device_attribute *at
 	return retval ? retval : count;					\
 }									\
 									\
-static ssize_t show_output##value(struct device *dev, struct device_attribute *attr, char *buf)	\
+static ssize_t show_output##value(struct device *dev, 			\
+					struct device_attribute *attr,	\
+					char *buf)			\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct interfacekit *kit = usb_get_intfdata(intf);		\
+	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", !!test_bit(value - 1, &kit->outputs));\
 }									\
@@ -420,8 +426,7 @@ show_set_output(16);
 #define show_input(value)	\
 static ssize_t show_input##value(struct device *dev, struct device_attribute *attr, char *buf)	\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct interfacekit *kit = usb_get_intfdata(intf);		\
+	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", (int)kit->inputs[value - 1]);	\
 }									\
@@ -445,10 +450,11 @@ show_input(15);
 show_input(16);
 
 #define show_sensor(value)	\
-static ssize_t show_sensor##value(struct device *dev, struct device_attribute *attr, char *buf)	\
+static ssize_t show_sensor##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					char *buf)			\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct interfacekit *kit = usb_get_intfdata(intf);		\
+	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", (int)kit->sensors[value - 1]);	\
 }									\
@@ -471,6 +477,7 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic
 	struct interfacekit *kit;
 	struct driver_interfacekit *ifkit;
 	int pipe, maxp, rc = -ENOMEM;
+	int bit, value;
 
 	ifkit = (struct driver_interfacekit *)id->driver_info;
 	if (!ifkit)
@@ -493,6 +500,7 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic
 	if (!kit)
 		goto out;
 
+	kit->dev_no = -1;
 	kit->ifkit = ifkit;
 	kit->data = usb_buffer_alloc(dev, URB_INT_SIZE, SLAB_ATOMIC, &kit->data_dma);
 	if (!kit->data)
@@ -513,73 +521,88 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic
 
 	usb_set_intfdata(intf, kit);
 
+        do {
+                bit = find_first_zero_bit(&device_no, sizeof(device_no));
+                value = test_and_set_bit(bit, &device_no);
+        } while(value);
+        kit->dev_no = bit;
+
+        kit->dev = device_create(phidget_class, &kit->udev->dev, 0,
+               		"interfacekit%d", kit->dev_no);
+        if (IS_ERR(kit->dev)) {
+                rc = PTR_ERR(kit->dev);
+                kit->dev = NULL;
+                goto out;
+        }
+	dev_set_drvdata(kit->dev, kit);
+
 	if (usb_submit_urb(kit->irq, GFP_KERNEL)) {
 		rc = -EIO;
 		goto out;
 	}
 
 	if (ifkit->outputs >= 4) {
-		device_create_file(&intf->dev, &dev_attr_output1);
-		device_create_file(&intf->dev, &dev_attr_output2);
-		device_create_file(&intf->dev, &dev_attr_output3);
-		device_create_file(&intf->dev, &dev_attr_output4);
+		device_create_file(kit->dev, &dev_attr_output1);
+		device_create_file(kit->dev, &dev_attr_output2);
+		device_create_file(kit->dev, &dev_attr_output3);
+		device_create_file(kit->dev, &dev_attr_output4);
 	}
 	if (ifkit->outputs >= 8) {
-		device_create_file(&intf->dev, &dev_attr_output5);
-		device_create_file(&intf->dev, &dev_attr_output6);
-		device_create_file(&intf->dev, &dev_attr_output7);
-		device_create_file(&intf->dev, &dev_attr_output8);
+		device_create_file(kit->dev, &dev_attr_output5);
+		device_create_file(kit->dev, &dev_attr_output6);
+		device_create_file(kit->dev, &dev_attr_output7);
+		device_create_file(kit->dev, &dev_attr_output8);
 	} 
 	if (ifkit->outputs == 16) {
-		device_create_file(&intf->dev, &dev_attr_output9);
-		device_create_file(&intf->dev, &dev_attr_output10);
-		device_create_file(&intf->dev, &dev_attr_output11);
-		device_create_file(&intf->dev, &dev_attr_output12);
-		device_create_file(&intf->dev, &dev_attr_output13);
-		device_create_file(&intf->dev, &dev_attr_output14);
-		device_create_file(&intf->dev, &dev_attr_output15);
-		device_create_file(&intf->dev, &dev_attr_output16);
+		device_create_file(kit->dev, &dev_attr_output9);
+		device_create_file(kit->dev, &dev_attr_output10);
+		device_create_file(kit->dev, &dev_attr_output11);
+		device_create_file(kit->dev, &dev_attr_output12);
+		device_create_file(kit->dev, &dev_attr_output13);
+		device_create_file(kit->dev, &dev_attr_output14);
+		device_create_file(kit->dev, &dev_attr_output15);
+		device_create_file(kit->dev, &dev_attr_output16);
 	}
 
 	if (ifkit->inputs >= 4) {
-		device_create_file(&intf->dev, &dev_attr_input1);
-		device_create_file(&intf->dev, &dev_attr_input2);
-		device_create_file(&intf->dev, &dev_attr_input3);
-		device_create_file(&intf->dev, &dev_attr_input4);
+		device_create_file(kit->dev, &dev_attr_input1);
+		device_create_file(kit->dev, &dev_attr_input2);
+		device_create_file(kit->dev, &dev_attr_input3);
+		device_create_file(kit->dev, &dev_attr_input4);
 	}
 	if (ifkit->inputs >= 8) {
-		device_create_file(&intf->dev, &dev_attr_input5);
-		device_create_file(&intf->dev, &dev_attr_input6);
-		device_create_file(&intf->dev, &dev_attr_input7);
-		device_create_file(&intf->dev, &dev_attr_input8);
+		device_create_file(kit->dev, &dev_attr_input5);
+		device_create_file(kit->dev, &dev_attr_input6);
+		device_create_file(kit->dev, &dev_attr_input7);
+		device_create_file(kit->dev, &dev_attr_input8);
 	}
 	if (ifkit->inputs == 16) {
-		device_create_file(&intf->dev, &dev_attr_input9);
-		device_create_file(&intf->dev, &dev_attr_input10);
-		device_create_file(&intf->dev, &dev_attr_input11);
-		device_create_file(&intf->dev, &dev_attr_input12);
-		device_create_file(&intf->dev, &dev_attr_input13);
-		device_create_file(&intf->dev, &dev_attr_input14);
-		device_create_file(&intf->dev, &dev_attr_input15);
-		device_create_file(&intf->dev, &dev_attr_input16);
+		device_create_file(kit->dev, &dev_attr_input9);
+		device_create_file(kit->dev, &dev_attr_input10);
+		device_create_file(kit->dev, &dev_attr_input11);
+		device_create_file(kit->dev, &dev_attr_input12);
+		device_create_file(kit->dev, &dev_attr_input13);
+		device_create_file(kit->dev, &dev_attr_input14);
+		device_create_file(kit->dev, &dev_attr_input15);
+		device_create_file(kit->dev, &dev_attr_input16);
 	}
 
 	if (ifkit->sensors >= 4) {
-		device_create_file(&intf->dev, &dev_attr_sensor1);
-		device_create_file(&intf->dev, &dev_attr_sensor2);
-		device_create_file(&intf->dev, &dev_attr_sensor3);
-		device_create_file(&intf->dev, &dev_attr_sensor4);
+		device_create_file(kit->dev, &dev_attr_sensor1);
+		device_create_file(kit->dev, &dev_attr_sensor2);
+		device_create_file(kit->dev, &dev_attr_sensor3);
+		device_create_file(kit->dev, &dev_attr_sensor4);
 	}
 	if (ifkit->sensors >= 7) {
-		device_create_file(&intf->dev, &dev_attr_sensor5);
-		device_create_file(&intf->dev, &dev_attr_sensor6);
-		device_create_file(&intf->dev, &dev_attr_sensor7);
+		device_create_file(kit->dev, &dev_attr_sensor5);
+		device_create_file(kit->dev, &dev_attr_sensor6);
+		device_create_file(kit->dev, &dev_attr_sensor7);
 	}
 	if (ifkit->sensors == 8)
-		device_create_file(&intf->dev, &dev_attr_sensor8);
+		device_create_file(kit->dev, &dev_attr_sensor8);
 
 	if (ifkit->has_lcd)
-		device_create_file(&intf->dev, &dev_attr_lcd);
+		device_create_file(kit->dev, &dev_attr_lcd);
 
 	dev_info(&intf->dev, "USB PhidgetInterfaceKit %d/%d/%d attached\n",
 			ifkit->sensors, ifkit->inputs, ifkit->outputs);
@@ -592,6 +615,11 @@ out:
 			usb_free_urb(kit->irq);
 		if (kit->data)
 			usb_buffer_free(dev, URB_INT_SIZE, kit->data, kit->data_dma);
+		if (kit->dev)
+			device_unregister(kit->dev);
+		if (kit->dev_no >= 0)
+			clear_bit(kit->dev_no, &device_no);
+
 		kfree(kit);
 	}
 
@@ -614,72 +642,76 @@ static void interfacekit_disconnect(struct usb_interface *interface)
 	cancel_delayed_work(&kit->do_notify);
 
 	if (kit->ifkit->outputs >= 4) {
-		device_remove_file(&interface->dev, &dev_attr_output1);
-		device_remove_file(&interface->dev, &dev_attr_output2);
-		device_remove_file(&interface->dev, &dev_attr_output3);
-		device_remove_file(&interface->dev, &dev_attr_output4);
+		device_remove_file(kit->dev, &dev_attr_output1);
+		device_remove_file(kit->dev, &dev_attr_output2);
+		device_remove_file(kit->dev, &dev_attr_output3);
+		device_remove_file(kit->dev, &dev_attr_output4);
 	}
 	if (kit->ifkit->outputs >= 8) {
-		device_remove_file(&interface->dev, &dev_attr_output5);
-		device_remove_file(&interface->dev, &dev_attr_output6);
-		device_remove_file(&interface->dev, &dev_attr_output7);
-		device_remove_file(&interface->dev, &dev_attr_output8);
+		device_remove_file(kit->dev, &dev_attr_output5);
+		device_remove_file(kit->dev, &dev_attr_output6);
+		device_remove_file(kit->dev, &dev_attr_output7);
+		device_remove_file(kit->dev, &dev_attr_output8);
 	}
 	if (kit->ifkit->outputs == 16) {
-		device_remove_file(&interface->dev, &dev_attr_output9);
-		device_remove_file(&interface->dev, &dev_attr_output10);
-		device_remove_file(&interface->dev, &dev_attr_output11);
-		device_remove_file(&interface->dev, &dev_attr_output12);
-		device_remove_file(&interface->dev, &dev_attr_output13);
-		device_remove_file(&interface->dev, &dev_attr_output14);
-		device_remove_file(&interface->dev, &dev_attr_output15);
-		device_remove_file(&interface->dev, &dev_attr_output16);
+		device_remove_file(kit->dev, &dev_attr_output9);
+		device_remove_file(kit->dev, &dev_attr_output10);
+		device_remove_file(kit->dev, &dev_attr_output11);
+		device_remove_file(kit->dev, &dev_attr_output12);
+		device_remove_file(kit->dev, &dev_attr_output13);
+		device_remove_file(kit->dev, &dev_attr_output14);
+		device_remove_file(kit->dev, &dev_attr_output15);
+		device_remove_file(kit->dev, &dev_attr_output16);
 	}
 
 	if (kit->ifkit->inputs >= 4) {
-		device_remove_file(&interface->dev, &dev_attr_input1);
-		device_remove_file(&interface->dev, &dev_attr_input2);
-		device_remove_file(&interface->dev, &dev_attr_input3);
-		device_remove_file(&interface->dev, &dev_attr_input4);
+		device_remove_file(kit->dev, &dev_attr_input1);
+		device_remove_file(kit->dev, &dev_attr_input2);
+		device_remove_file(kit->dev, &dev_attr_input3);
+		device_remove_file(kit->dev, &dev_attr_input4);
 	}
 	if (kit->ifkit->inputs >= 8) {
-		device_remove_file(&interface->dev, &dev_attr_input5);
-		device_remove_file(&interface->dev, &dev_attr_input6);
-		device_remove_file(&interface->dev, &dev_attr_input7);
-		device_remove_file(&interface->dev, &dev_attr_input8);
+		device_remove_file(kit->dev, &dev_attr_input5);
+		device_remove_file(kit->dev, &dev_attr_input6);
+		device_remove_file(kit->dev, &dev_attr_input7);
+		device_remove_file(kit->dev, &dev_attr_input8);
 	}
 	if (kit->ifkit->inputs == 16) {
-		device_remove_file(&interface->dev, &dev_attr_input9);
-		device_remove_file(&interface->dev, &dev_attr_input10);
-		device_remove_file(&interface->dev, &dev_attr_input11);
-		device_remove_file(&interface->dev, &dev_attr_input12);
-		device_remove_file(&interface->dev, &dev_attr_input13);
-		device_remove_file(&interface->dev, &dev_attr_input14);
-		device_remove_file(&interface->dev, &dev_attr_input15);
-		device_remove_file(&interface->dev, &dev_attr_input16);
+		device_remove_file(kit->dev, &dev_attr_input9);
+		device_remove_file(kit->dev, &dev_attr_input10);
+		device_remove_file(kit->dev, &dev_attr_input11);
+		device_remove_file(kit->dev, &dev_attr_input12);
+		device_remove_file(kit->dev, &dev_attr_input13);
+		device_remove_file(kit->dev, &dev_attr_input14);
+		device_remove_file(kit->dev, &dev_attr_input15);
+		device_remove_file(kit->dev, &dev_attr_input16);
 	}
 
 	if (kit->ifkit->sensors >= 4) {
-		device_remove_file(&interface->dev, &dev_attr_sensor1);
-		device_remove_file(&interface->dev, &dev_attr_sensor2);
-		device_remove_file(&interface->dev, &dev_attr_sensor3);
-		device_remove_file(&interface->dev, &dev_attr_sensor4);
+		device_remove_file(kit->dev, &dev_attr_sensor1);
+		device_remove_file(kit->dev, &dev_attr_sensor2);
+		device_remove_file(kit->dev, &dev_attr_sensor3);
+		device_remove_file(kit->dev, &dev_attr_sensor4);
 	}
 	if (kit->ifkit->sensors >= 7) {
-		device_remove_file(&interface->dev, &dev_attr_sensor5);
-		device_remove_file(&interface->dev, &dev_attr_sensor6);
-		device_remove_file(&interface->dev, &dev_attr_sensor7);
+		device_remove_file(kit->dev, &dev_attr_sensor5);
+		device_remove_file(kit->dev, &dev_attr_sensor6);
+		device_remove_file(kit->dev, &dev_attr_sensor7);
 	}
 	if (kit->ifkit->sensors == 8)
-		device_remove_file(&interface->dev, &dev_attr_sensor8);
+		device_remove_file(kit->dev, &dev_attr_sensor8);
 
 	if (kit->ifkit->has_lcd)
-		device_remove_file(&interface->dev, &dev_attr_lcd);
+		device_remove_file(kit->dev, &dev_attr_lcd);
+
+	device_unregister(kit->dev);
 
 	dev_info(&interface->dev, "USB PhidgetInterfaceKit %d/%d/%d detached\n",
 		kit->ifkit->sensors, kit->ifkit->inputs, kit->ifkit->outputs);
 
 	usb_put_dev(kit->udev);
+	clear_bit(kit->dev_no, &device_no);
+
 	kfree(kit);
 }
 
diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c
index 2972dc2eb27f..d9ac7f97f8c8 100644
--- a/drivers/usb/misc/phidgetmotorcontrol.c
+++ b/drivers/usb/misc/phidgetmotorcontrol.c
@@ -15,6 +15,8 @@
 #include <linux/module.h>
 #include <linux/usb.h>
 
+#include "phidget.h"
+
 #define DRIVER_AUTHOR "Sean Young <sean@mess.org>"
 #define DRIVER_DESC "USB PhidgetMotorControl Driver"
 
@@ -23,9 +25,13 @@
 
 #define URB_INT_SIZE			8
 
+static unsigned long device_no;
+
 struct motorcontrol {
 	struct usb_device *udev;
 	struct usb_interface *intf;
+	struct device *dev;
+	int dev_no;
 	u8 inputs[4];
 	s8 desired_speed[2];
 	s8 speed[2];
@@ -162,14 +168,14 @@ static void do_notify(void *data)
 	for (i=0; i<4; i++) {
 		if (test_and_clear_bit(i, &mc->input_events)) {
 			sprintf(sysfs_file, "input%d", i);
-			sysfs_notify(&mc->intf->dev.kobj, NULL, sysfs_file);
+			sysfs_notify(&mc->dev->kobj, NULL, sysfs_file);
 		}
 	}
 
 	for (i=0; i<2; i++) {
 		if (test_and_clear_bit(i, &mc->speed_events)) {
 			sprintf(sysfs_file, "speed%d", i);
-			sysfs_notify(&mc->intf->dev.kobj, NULL, sysfs_file);
+			sysfs_notify(&mc->dev->kobj, NULL, sysfs_file);
 		}
 	}
 
@@ -181,11 +187,11 @@ static void do_notify(void *data)
 }
 
 #define show_set_speed(value)		\
-static ssize_t set_speed##value(struct device *dev, 			\
-	struct device_attribute *attr, const char *buf,	size_t count)	\
+static ssize_t set_speed##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					const char *buf, size_t count)	\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 	int speed;							\
 	int retval;							\
 									\
@@ -202,11 +208,11 @@ static ssize_t set_speed##value(struct device *dev, 			\
 	return retval ? retval : count;					\
 }									\
 									\
-static ssize_t show_speed##value(struct device *dev, 			\
-			struct device_attribute *attr, char *buf)	\
+static ssize_t show_speed##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					char *buf)			\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", mc->speed[value]);			\
 }									\
@@ -217,10 +223,10 @@ show_set_speed(1);
 
 #define show_set_acceleration(value)		\
 static ssize_t set_acceleration##value(struct device *dev, 		\
-	struct device_attribute *attr, const char *buf,	size_t count)	\
+					struct device_attribute *attr,	\
+					const char *buf, size_t count)	\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 	int acceleration;						\
 	int retval;							\
 									\
@@ -237,11 +243,11 @@ static ssize_t set_acceleration##value(struct device *dev, 		\
 	return retval ? retval : count;					\
 }									\
 									\
-static ssize_t show_acceleration##value(struct device *dev, 		\
-			struct device_attribute *attr, char *buf)	\
+static ssize_t show_acceleration##value(struct device *dev,	 	\
+					struct device_attribute *attr,	\
+							char *buf)	\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", mc->acceleration[value]);		\
 }									\
@@ -251,11 +257,11 @@ show_set_acceleration(0);
 show_set_acceleration(1);
 
 #define show_current(value)	\
-static ssize_t show_current##value(struct device *dev, 			\
-			struct device_attribute *attr, char *buf)	\
+static ssize_t show_current##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					char *buf)			\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%dmA\n", (int)mc->_current[value]);	\
 }									\
@@ -265,11 +271,11 @@ show_current(0);
 show_current(1);
 
 #define show_input(value)	\
-static ssize_t show_input##value(struct device *dev, 			\
-			struct device_attribute *attr, char *buf)	\
+static ssize_t show_input##value(struct device *dev,			\
+					struct device_attribute *attr,	\
+					char *buf)			\
 {									\
-	struct usb_interface *intf = to_usb_interface(dev);		\
-	struct motorcontrol *mc = usb_get_intfdata(intf);		\
+	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", (int)mc->inputs[value]);		\
 }									\
@@ -287,6 +293,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic
 	struct usb_endpoint_descriptor *endpoint;
 	struct motorcontrol *mc;
 	int pipe, maxp, rc = -ENOMEM;
+	int bit, value;
 
 	interface = intf->cur_altsetting;
 	if (interface->desc.bNumEndpoints != 1)
@@ -306,6 +313,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic
 	if (!mc)
 		goto out;
 
+	mc->dev_no = -1;
 	mc->data = usb_buffer_alloc(dev, URB_INT_SIZE, SLAB_ATOMIC, &mc->data_dma);
 	if (!mc->data)
 		goto out;
@@ -326,26 +334,42 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic
 
 	usb_set_intfdata(intf, mc);
 
+	do {
+		bit = find_first_zero_bit(&device_no, sizeof(device_no));
+		value = test_and_set_bit(bit, &device_no);
+	} while(value);
+	mc->dev_no = bit;
+
+	mc->dev = device_create(phidget_class, &mc->udev->dev, 0,
+				"motorcontrol%d", mc->dev_no);
+	if (IS_ERR(mc->dev)) {
+		rc = PTR_ERR(mc->dev);
+		mc->dev = NULL;
+		goto out;
+	}
+
+	dev_set_drvdata(mc->dev, mc);
+
 	if (usb_submit_urb(mc->irq, GFP_KERNEL)) {
 		rc = -EIO;
 		goto out;
 	}
 
-	device_create_file(&intf->dev, &dev_attr_input0);
-	device_create_file(&intf->dev, &dev_attr_input1);
-	device_create_file(&intf->dev, &dev_attr_input2);
-	device_create_file(&intf->dev, &dev_attr_input3);
+	device_create_file(mc->dev, &dev_attr_input0);
+	device_create_file(mc->dev, &dev_attr_input1);
+	device_create_file(mc->dev, &dev_attr_input2);
+	device_create_file(mc->dev, &dev_attr_input3);
 
-	device_create_file(&intf->dev, &dev_attr_speed0);
-	device_create_file(&intf->dev, &dev_attr_speed1);
+	device_create_file(mc->dev, &dev_attr_speed0);
+	device_create_file(mc->dev, &dev_attr_speed1);
 
-	device_create_file(&intf->dev, &dev_attr_acceleration0);
-	device_create_file(&intf->dev, &dev_attr_acceleration1);
+	device_create_file(mc->dev, &dev_attr_acceleration0);
+	device_create_file(mc->dev, &dev_attr_acceleration1);
 
-	device_create_file(&intf->dev, &dev_attr_current0);
-	device_create_file(&intf->dev, &dev_attr_current1);
+	device_create_file(mc->dev, &dev_attr_current0);
+	device_create_file(mc->dev, &dev_attr_current1);
 
-	dev_info(&intf->dev, "USB Phidget MotorControl attached\n");
+	dev_info(&intf->dev, "USB PhidgetMotorControl attached\n");
 
 	return 0;
 
@@ -355,6 +379,11 @@ out:
 			usb_free_urb(mc->irq);
 		if (mc->data)
 			usb_buffer_free(dev, URB_INT_SIZE, mc->data, mc->data_dma);
+		if (mc->dev)
+			device_unregister(mc->dev);
+		if (mc->dev_no >= 0)
+			clear_bit(mc->dev_no, &device_no);
+
 		kfree(mc);
 	}
 
@@ -376,24 +405,27 @@ static void motorcontrol_disconnect(struct usb_interface *interface)
 
 	cancel_delayed_work(&mc->do_notify);
 
-	device_remove_file(&interface->dev, &dev_attr_input0);
-	device_remove_file(&interface->dev, &dev_attr_input1);
-	device_remove_file(&interface->dev, &dev_attr_input2);
-	device_remove_file(&interface->dev, &dev_attr_input3);
+	device_remove_file(mc->dev, &dev_attr_input0);
+	device_remove_file(mc->dev, &dev_attr_input1);
+	device_remove_file(mc->dev, &dev_attr_input2);
+	device_remove_file(mc->dev, &dev_attr_input3);
 
-	device_remove_file(&interface->dev, &dev_attr_speed0);
-	device_remove_file(&interface->dev, &dev_attr_speed1);
+	device_remove_file(mc->dev, &dev_attr_speed0);
+	device_remove_file(mc->dev, &dev_attr_speed1);
 
-	device_remove_file(&interface->dev, &dev_attr_acceleration0);
-	device_remove_file(&interface->dev, &dev_attr_acceleration1);
+	device_remove_file(mc->dev, &dev_attr_acceleration0);
+	device_remove_file(mc->dev, &dev_attr_acceleration1);
 
-	device_remove_file(&interface->dev, &dev_attr_current0);
-	device_remove_file(&interface->dev, &dev_attr_current1);
+	device_remove_file(mc->dev, &dev_attr_current0);
+	device_remove_file(mc->dev, &dev_attr_current1);
 
-	dev_info(&interface->dev, "USB Phidget MotorControl disconnected\n");
+	device_unregister(mc->dev);
 
 	usb_put_dev(mc->udev);
+	clear_bit(mc->dev_no, &device_no);
 	kfree(mc);
+
+	dev_info(&interface->dev, "USB PhidgetMotorControl detached\n");
 }
 
 static struct usb_driver motorcontrol_driver = {
diff --git a/drivers/usb/misc/phidgetservo.c b/drivers/usb/misc/phidgetservo.c
index c0df79c96538..66be9513fd6f 100644
--- a/drivers/usb/misc/phidgetservo.c
+++ b/drivers/usb/misc/phidgetservo.c
@@ -15,14 +15,6 @@
  *
  * CAUTION: Generally you should use 0 < degrees < 180 as anything else
  * is probably beyond the range of your servo and may damage it.
- *
- * Jun 16, 2004: Sean Young <sean@mess.org>
- *  - cleanups
- *  - was using memory after kfree()
- * Aug 8, 2004: Sean Young <sean@mess.org>
- *  - set the highest angle as high as the hardware allows, there are 
- *    some odd servos out there
- *
  */
 
 #include <linux/kernel.h>
@@ -32,6 +24,8 @@
 #include <linux/module.h>
 #include <linux/usb.h>
 
+#include "phidget.h"
+
 #define DRIVER_AUTHOR "Sean Young <sean@mess.org>"
 #define DRIVER_DESC "USB PhidgetServo Driver"
 
@@ -70,8 +64,12 @@ static struct usb_device_id id_table[] = {
 
 MODULE_DEVICE_TABLE(usb, id_table);
 
+static int unsigned long device_no;
+
 struct phidget_servo {
 	struct usb_device *udev;
+	struct device *dev;
+	int dev_no;
 	ulong type;
 	int pulse[4];
 	int degrees[4];
@@ -203,16 +201,16 @@ change_position_v20(struct phidget_servo *servo, int servo_no, int degrees,
 }
 
 #define show_set(value)	\
-static ssize_t set_servo##value (struct device *dev, struct device_attribute *attr,			\
+static ssize_t set_servo##value (struct device *dev, 			\
+					struct device_attribute *attr,	\
 					const char *buf, size_t count)	\
 {									\
 	int degrees, minutes, retval;					\
-	struct usb_interface *intf = to_usb_interface (dev);		\
-	struct phidget_servo *servo = usb_get_intfdata (intf);		\
+	struct phidget_servo *servo = dev_get_drvdata(dev);		\
 									\
 	minutes = 0;							\
 	/* must at least convert degrees */				\
-	if (sscanf (buf, "%d.%d", &degrees, &minutes) < 1) {		\
+	if (sscanf(buf, "%d.%d", &degrees, &minutes) < 1) {		\
 		return -EINVAL;						\
 	}								\
 									\
@@ -220,21 +218,22 @@ static ssize_t set_servo##value (struct device *dev, struct device_attribute *at
 		return -EINVAL;						\
 									\
 	if (servo->type & SERVO_VERSION_30)				\
-		retval = change_position_v30 (servo, value, degrees, 	\
+		retval = change_position_v30(servo, value, degrees, 	\
 							minutes);	\
 	else 								\
-		retval = change_position_v20 (servo, value, degrees, 	\
+		retval = change_position_v20(servo, value, degrees, 	\
 							minutes);	\
 									\
 	return retval < 0 ? retval : count;				\
 }									\
 									\
-static ssize_t show_servo##value (struct device *dev, struct device_attribute *attr, char *buf) 	\
+static ssize_t show_servo##value (struct device *dev,			\
+					struct device_attribute *attr,	\
+					char *buf) 			\
 {									\
-	struct usb_interface *intf = to_usb_interface (dev);		\
-	struct phidget_servo *servo = usb_get_intfdata (intf);		\
+	struct phidget_servo *servo = dev_get_drvdata(dev);		\
 									\
-	return sprintf (buf, "%d.%02d\n", servo->degrees[value],	\
+	return sprintf(buf, "%d.%02d\n", servo->degrees[value],		\
 				servo->minutes[value]);			\
 }									\
 static DEVICE_ATTR(servo##value, S_IWUGO | S_IRUGO,			\
@@ -250,6 +249,7 @@ servo_probe(struct usb_interface *interface, const struct usb_device_id *id)
 {
 	struct usb_device *udev = interface_to_usbdev(interface);
 	struct phidget_servo *dev;
+	int bit, value;
 
 	dev = kzalloc(sizeof (struct phidget_servo), GFP_KERNEL);
 	if (dev == NULL) {
@@ -261,18 +261,33 @@ servo_probe(struct usb_interface *interface, const struct usb_device_id *id)
 	dev->type = id->driver_info;
 	usb_set_intfdata(interface, dev);
 
-	device_create_file(&interface->dev, &dev_attr_servo0);
+        do {
+                bit = find_first_zero_bit(&device_no, sizeof(device_no));
+                value = test_and_set_bit(bit, &device_no);
+        } while(value);
+	dev->dev_no = bit;
+
+	dev->dev = device_create(phidget_class, &dev->udev->dev, 0,
+				 "servo%d", dev->dev_no);
+	if (IS_ERR(dev->dev)) {
+		int rc = PTR_ERR(dev->dev);
+		clear_bit(dev->dev_no, &device_no);
+		kfree(dev);
+		return rc;
+	}
+
+	device_create_file(dev->dev, &dev_attr_servo0);
 	if (dev->type & SERVO_COUNT_QUAD) {
-		device_create_file(&interface->dev, &dev_attr_servo1);
-		device_create_file(&interface->dev, &dev_attr_servo2);
-		device_create_file(&interface->dev, &dev_attr_servo3);
+		device_create_file(dev->dev, &dev_attr_servo1);
+		device_create_file(dev->dev, &dev_attr_servo2);
+		device_create_file(dev->dev, &dev_attr_servo3);
 	}
 
 	dev_info(&interface->dev, "USB %d-Motor PhidgetServo v%d.0 attached\n",
 		dev->type & SERVO_COUNT_QUAD ? 4 : 1,
 		dev->type & SERVO_VERSION_30 ? 3 : 2);
 
-	if(!(dev->type & SERVO_VERSION_30))
+	if (!(dev->type & SERVO_VERSION_30))
 		dev_info(&interface->dev,
 			 "WARNING: v2.0 not tested! Please report if it works.\n");
 
@@ -287,19 +302,21 @@ servo_disconnect(struct usb_interface *interface)
 	dev = usb_get_intfdata(interface);
 	usb_set_intfdata(interface, NULL);
 
-	device_remove_file(&interface->dev, &dev_attr_servo0);
+	device_remove_file(dev->dev, &dev_attr_servo0);
 	if (dev->type & SERVO_COUNT_QUAD) {
-		device_remove_file(&interface->dev, &dev_attr_servo1);
-		device_remove_file(&interface->dev, &dev_attr_servo2);
-		device_remove_file(&interface->dev, &dev_attr_servo3);
+		device_remove_file(dev->dev, &dev_attr_servo1);
+		device_remove_file(dev->dev, &dev_attr_servo2);
+		device_remove_file(dev->dev, &dev_attr_servo3);
 	}
 
+	device_unregister(dev->dev);
 	usb_put_dev(dev->udev);
 
 	dev_info(&interface->dev, "USB %d-Motor PhidgetServo v%d.0 detached\n",
 		dev->type & SERVO_COUNT_QUAD ? 4 : 1,
 		dev->type & SERVO_VERSION_30 ? 3 : 2);
 
+	clear_bit(dev->dev_no, &device_no);
 	kfree(dev);
 }
 
-- 
cgit v1.2.1


From da308e8da700637d6271dddda08128094691b980 Mon Sep 17 00:00:00 2001
From: Sean Young <sean@mess.org>
Date: Fri, 11 Aug 2006 09:28:28 +0000
Subject: USB: Phidgets should check create_device_file() return value

device_create_file() could fail, add proper error paths for this condition.

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/phidgetkit.c          | 238 +++++++++++++++------------------
 drivers/usb/misc/phidgetmotorcontrol.c |  83 ++++++------
 drivers/usb/misc/phidgetservo.c        |  74 ++++++----
 3 files changed, 199 insertions(+), 196 deletions(-)

diff --git a/drivers/usb/misc/phidgetkit.c b/drivers/usb/misc/phidgetkit.c
index 3f508957def9..9a8d137d39f9 100644
--- a/drivers/usb/misc/phidgetkit.c
+++ b/drivers/usb/misc/phidgetkit.c
@@ -193,8 +193,11 @@ static ssize_t lcd_line_##number(struct device *dev,			\
 	struct interfacekit *kit = dev_get_drvdata(dev);		\
 	change_string(kit, buf, number - 1);				\
 	return count;							\
-}									\
-static DEVICE_ATTR(lcd_line_##number, S_IWUGO, NULL, lcd_line_##number);
+}
+
+#define lcd_line_attr(number)						\
+	__ATTR(lcd_line_##number, S_IWUGO, NULL, lcd_line_##number)
+
 set_lcd_line(1);
 set_lcd_line(2);
 
@@ -232,15 +235,22 @@ exit:
 	kfree(buffer);
 	return retval;
 }
-static DEVICE_ATTR(backlight, S_IWUGO, NULL, set_backlight);
+
+static struct device_attribute dev_lcd_line_attrs[] = {
+	lcd_line_attr(1),
+	lcd_line_attr(2),
+	__ATTR(backlight, S_IWUGO, NULL, set_backlight)
+};
 
 static void remove_lcd_files(struct interfacekit *kit)
 {
+	int i;
+
 	if (kit->lcd_files_on) {
 		dev_dbg(&kit->udev->dev, "Removing lcd files\n");
-		device_remove_file(kit->dev, &dev_attr_lcd_line_1);
-		device_remove_file(kit->dev, &dev_attr_lcd_line_2);
-		device_remove_file(kit->dev, &dev_attr_backlight);
+
+		for (i=0; i<ARRAY_SIZE(dev_lcd_line_attrs); i++)
+			device_remove_file(kit->dev, &dev_lcd_line_attrs[i]);
 	}
 }
 
@@ -248,6 +258,7 @@ static ssize_t enable_lcd_files(struct device *dev, struct device_attribute *att
 {
 	struct interfacekit *kit = dev_get_drvdata(dev);
 	int enable;
+	int i, rc;
 	
 	if (kit->ifkit->has_lcd == 0)
 		return -ENODEV;
@@ -258,9 +269,12 @@ static ssize_t enable_lcd_files(struct device *dev, struct device_attribute *att
 	if (enable) {
 		if (!kit->lcd_files_on) {
 			dev_dbg(&kit->udev->dev, "Adding lcd files\n");
-			device_create_file(kit->dev, &dev_attr_lcd_line_1);
-			device_create_file(kit->dev, &dev_attr_lcd_line_2);
-			device_create_file(kit->dev, &dev_attr_backlight);
+			for (i=0; i<ARRAY_SIZE(dev_lcd_line_attrs); i++) {
+				rc = device_create_file(kit->dev,
+					&dev_lcd_line_attrs[i]);
+				if (rc)
+					goto out;
+			}
 			kit->lcd_files_on = 1;
 		}
 	} else {
@@ -271,7 +285,13 @@ static ssize_t enable_lcd_files(struct device *dev, struct device_attribute *att
 	}
 	
 	return count;
+out:
+	while (i-- > 0)
+		device_remove_file(kit->dev, &dev_lcd_line_attrs[i]);
+
+	return rc;
 }
+
 static DEVICE_ATTR(lcd, S_IWUGO, NULL, enable_lcd_files);
 
 static void interfacekit_irq(struct urb *urb, struct pt_regs *regs)
@@ -403,9 +423,12 @@ static ssize_t show_output##value(struct device *dev, 			\
 	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", !!test_bit(value - 1, &kit->outputs));\
-}									\
-static DEVICE_ATTR(output##value, S_IWUGO | S_IRUGO,			\
-		show_output##value, set_output##value);
+}
+
+#define output_attr(value)						\
+	__ATTR(output##value, S_IWUGO | S_IRUGO,			\
+		show_output##value, set_output##value)
+
 show_set_output(1);
 show_set_output(2);
 show_set_output(3);
@@ -423,14 +446,24 @@ show_set_output(14);
 show_set_output(15);
 show_set_output(16);
 
+static struct device_attribute dev_output_attrs[] = {
+	output_attr(1), output_attr(2), output_attr(3), output_attr(4),
+	output_attr(5), output_attr(6), output_attr(7), output_attr(8),
+	output_attr(9), output_attr(10), output_attr(11), output_attr(12),
+	output_attr(13), output_attr(14), output_attr(15), output_attr(16)
+};
+
 #define show_input(value)	\
-static ssize_t show_input##value(struct device *dev, struct device_attribute *attr, char *buf)	\
+static ssize_t show_input##value(struct device *dev, 			\
+			struct device_attribute *attr, char *buf)	\
 {									\
 	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", (int)kit->inputs[value - 1]);	\
-}									\
-static DEVICE_ATTR(input##value, S_IRUGO, show_input##value, NULL);
+}
+
+#define input_attr(value)						\
+	__ATTR(input##value, S_IRUGO, show_input##value, NULL)
 
 show_input(1);
 show_input(2);
@@ -449,6 +482,13 @@ show_input(14);
 show_input(15);
 show_input(16);
 
+static struct device_attribute dev_input_attrs[] = {
+	input_attr(1), input_attr(2), input_attr(3), input_attr(4),
+	input_attr(5), input_attr(6), input_attr(7), input_attr(8),
+	input_attr(9), input_attr(10), input_attr(11), input_attr(12),
+	input_attr(13), input_attr(14), input_attr(15), input_attr(16)
+};
+
 #define show_sensor(value)	\
 static ssize_t show_sensor##value(struct device *dev,			\
 					struct device_attribute *attr,	\
@@ -457,8 +497,10 @@ static ssize_t show_sensor##value(struct device *dev,			\
 	struct interfacekit *kit = dev_get_drvdata(dev);		\
 									\
 	return sprintf(buf, "%d\n", (int)kit->sensors[value - 1]);	\
-}									\
-static DEVICE_ATTR(sensor##value, S_IRUGO, show_sensor##value, NULL);
+}
+
+#define sensor_attr(value)						\
+	__ATTR(sensor##value, S_IRUGO, show_sensor##value, NULL)
 
 show_sensor(1);
 show_sensor(2);
@@ -469,6 +511,11 @@ show_sensor(6);
 show_sensor(7);
 show_sensor(8);
 
+static struct device_attribute dev_sensor_attrs[] = {
+	sensor_attr(1), sensor_attr(2), sensor_attr(3), sensor_attr(4),
+	sensor_attr(5), sensor_attr(6), sensor_attr(7), sensor_attr(8)
+};
+
 static int interfacekit_probe(struct usb_interface *intf, const struct usb_device_id *id)
 {
 	struct usb_device *dev = interface_to_usbdev(intf);
@@ -477,7 +524,7 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic
 	struct interfacekit *kit;
 	struct driver_interfacekit *ifkit;
 	int pipe, maxp, rc = -ENOMEM;
-	int bit, value;
+	int bit, value, i;
 
 	ifkit = (struct driver_interfacekit *)id->driver_info;
 	if (!ifkit)
@@ -541,74 +588,49 @@ static int interfacekit_probe(struct usb_interface *intf, const struct usb_devic
 		goto out;
 	}
 
-	if (ifkit->outputs >= 4) {
-		device_create_file(kit->dev, &dev_attr_output1);
-		device_create_file(kit->dev, &dev_attr_output2);
-		device_create_file(kit->dev, &dev_attr_output3);
-		device_create_file(kit->dev, &dev_attr_output4);
-	}
-	if (ifkit->outputs >= 8) {
-		device_create_file(kit->dev, &dev_attr_output5);
-		device_create_file(kit->dev, &dev_attr_output6);
-		device_create_file(kit->dev, &dev_attr_output7);
-		device_create_file(kit->dev, &dev_attr_output8);
-	} 
-	if (ifkit->outputs == 16) {
-		device_create_file(kit->dev, &dev_attr_output9);
-		device_create_file(kit->dev, &dev_attr_output10);
-		device_create_file(kit->dev, &dev_attr_output11);
-		device_create_file(kit->dev, &dev_attr_output12);
-		device_create_file(kit->dev, &dev_attr_output13);
-		device_create_file(kit->dev, &dev_attr_output14);
-		device_create_file(kit->dev, &dev_attr_output15);
-		device_create_file(kit->dev, &dev_attr_output16);
+	for (i=0; i<ifkit->outputs; i++ ) {
+		rc = device_create_file(kit->dev, &dev_output_attrs[i]);
+		if (rc)
+			goto out2;
 	}
 
-	if (ifkit->inputs >= 4) {
-		device_create_file(kit->dev, &dev_attr_input1);
-		device_create_file(kit->dev, &dev_attr_input2);
-		device_create_file(kit->dev, &dev_attr_input3);
-		device_create_file(kit->dev, &dev_attr_input4);
-	}
-	if (ifkit->inputs >= 8) {
-		device_create_file(kit->dev, &dev_attr_input5);
-		device_create_file(kit->dev, &dev_attr_input6);
-		device_create_file(kit->dev, &dev_attr_input7);
-		device_create_file(kit->dev, &dev_attr_input8);
-	}
-	if (ifkit->inputs == 16) {
-		device_create_file(kit->dev, &dev_attr_input9);
-		device_create_file(kit->dev, &dev_attr_input10);
-		device_create_file(kit->dev, &dev_attr_input11);
-		device_create_file(kit->dev, &dev_attr_input12);
-		device_create_file(kit->dev, &dev_attr_input13);
-		device_create_file(kit->dev, &dev_attr_input14);
-		device_create_file(kit->dev, &dev_attr_input15);
-		device_create_file(kit->dev, &dev_attr_input16);
+	for (i=0; i<ifkit->inputs; i++ ) {
+		rc = device_create_file(kit->dev, &dev_input_attrs[i]);
+		if (rc)
+			goto out3;
 	}
 
-	if (ifkit->sensors >= 4) {
-		device_create_file(kit->dev, &dev_attr_sensor1);
-		device_create_file(kit->dev, &dev_attr_sensor2);
-		device_create_file(kit->dev, &dev_attr_sensor3);
-		device_create_file(kit->dev, &dev_attr_sensor4);
-	}
-	if (ifkit->sensors >= 7) {
-		device_create_file(kit->dev, &dev_attr_sensor5);
-		device_create_file(kit->dev, &dev_attr_sensor6);
-		device_create_file(kit->dev, &dev_attr_sensor7);
+	for (i=0; i<ifkit->sensors; i++ ) {
+		rc = device_create_file(kit->dev, &dev_sensor_attrs[i]);
+		if (rc)
+			goto out4;
 	}
-	if (ifkit->sensors == 8)
-		device_create_file(kit->dev, &dev_attr_sensor8);
 
-	if (ifkit->has_lcd)
-		device_create_file(kit->dev, &dev_attr_lcd);
+	if (ifkit->has_lcd) {
+		rc = device_create_file(kit->dev, &dev_attr_lcd);
+		if (rc)
+			goto out4;
+
+	}
 
 	dev_info(&intf->dev, "USB PhidgetInterfaceKit %d/%d/%d attached\n",
 			ifkit->sensors, ifkit->inputs, ifkit->outputs);
 
 	return 0;
 
+out4:
+	while (i-- > 0)
+		device_remove_file(kit->dev, &dev_sensor_attrs[i]);
+
+	i = ifkit->inputs;
+out3:
+	while (i-- > 0)
+		device_remove_file(kit->dev, &dev_input_attrs[i]);
+
+	i = ifkit->outputs;
+out2:
+	while (i-- > 0)
+		device_remove_file(kit->dev, &dev_output_attrs[i]);
 out:
 	if (kit) {
 		if (kit->irq)
@@ -629,6 +651,7 @@ out:
 static void interfacekit_disconnect(struct usb_interface *interface)
 {
 	struct interfacekit *kit;
+	int i;
 
 	kit = usb_get_intfdata(interface);
 	usb_set_intfdata(interface, NULL);
@@ -641,68 +664,19 @@ static void interfacekit_disconnect(struct usb_interface *interface)
 
 	cancel_delayed_work(&kit->do_notify);
 
-	if (kit->ifkit->outputs >= 4) {
-		device_remove_file(kit->dev, &dev_attr_output1);
-		device_remove_file(kit->dev, &dev_attr_output2);
-		device_remove_file(kit->dev, &dev_attr_output3);
-		device_remove_file(kit->dev, &dev_attr_output4);
-	}
-	if (kit->ifkit->outputs >= 8) {
-		device_remove_file(kit->dev, &dev_attr_output5);
-		device_remove_file(kit->dev, &dev_attr_output6);
-		device_remove_file(kit->dev, &dev_attr_output7);
-		device_remove_file(kit->dev, &dev_attr_output8);
-	}
-	if (kit->ifkit->outputs == 16) {
-		device_remove_file(kit->dev, &dev_attr_output9);
-		device_remove_file(kit->dev, &dev_attr_output10);
-		device_remove_file(kit->dev, &dev_attr_output11);
-		device_remove_file(kit->dev, &dev_attr_output12);
-		device_remove_file(kit->dev, &dev_attr_output13);
-		device_remove_file(kit->dev, &dev_attr_output14);
-		device_remove_file(kit->dev, &dev_attr_output15);
-		device_remove_file(kit->dev, &dev_attr_output16);
-	}
+	for (i=0; i<kit->ifkit->outputs; i++)
+		device_remove_file(kit->dev, &dev_output_attrs[i]);
 
-	if (kit->ifkit->inputs >= 4) {
-		device_remove_file(kit->dev, &dev_attr_input1);
-		device_remove_file(kit->dev, &dev_attr_input2);
-		device_remove_file(kit->dev, &dev_attr_input3);
-		device_remove_file(kit->dev, &dev_attr_input4);
-	}
-	if (kit->ifkit->inputs >= 8) {
-		device_remove_file(kit->dev, &dev_attr_input5);
-		device_remove_file(kit->dev, &dev_attr_input6);
-		device_remove_file(kit->dev, &dev_attr_input7);
-		device_remove_file(kit->dev, &dev_attr_input8);
-	}
-	if (kit->ifkit->inputs == 16) {
-		device_remove_file(kit->dev, &dev_attr_input9);
-		device_remove_file(kit->dev, &dev_attr_input10);
-		device_remove_file(kit->dev, &dev_attr_input11);
-		device_remove_file(kit->dev, &dev_attr_input12);
-		device_remove_file(kit->dev, &dev_attr_input13);
-		device_remove_file(kit->dev, &dev_attr_input14);
-		device_remove_file(kit->dev, &dev_attr_input15);
-		device_remove_file(kit->dev, &dev_attr_input16);
-	}
+	for (i=0; i<kit->ifkit->inputs; i++)
+		device_remove_file(kit->dev, &dev_input_attrs[i]);
 
-	if (kit->ifkit->sensors >= 4) {
-		device_remove_file(kit->dev, &dev_attr_sensor1);
-		device_remove_file(kit->dev, &dev_attr_sensor2);
-		device_remove_file(kit->dev, &dev_attr_sensor3);
-		device_remove_file(kit->dev, &dev_attr_sensor4);
-	}
-	if (kit->ifkit->sensors >= 7) {
-		device_remove_file(kit->dev, &dev_attr_sensor5);
-		device_remove_file(kit->dev, &dev_attr_sensor6);
-		device_remove_file(kit->dev, &dev_attr_sensor7);
-	}
-	if (kit->ifkit->sensors == 8)
-		device_remove_file(kit->dev, &dev_attr_sensor8);
+	for (i=0; i<kit->ifkit->sensors; i++)
+		device_remove_file(kit->dev, &dev_sensor_attrs[i]);
 
-	if (kit->ifkit->has_lcd)
+	if (kit->ifkit->has_lcd) {
 		device_remove_file(kit->dev, &dev_attr_lcd);
+		remove_lcd_files(kit);
+	}
 
 	device_unregister(kit->dev);
 
diff --git a/drivers/usb/misc/phidgetmotorcontrol.c b/drivers/usb/misc/phidgetmotorcontrol.c
index d9ac7f97f8c8..6b59b620d616 100644
--- a/drivers/usb/misc/phidgetmotorcontrol.c
+++ b/drivers/usb/misc/phidgetmotorcontrol.c
@@ -215,9 +215,12 @@ static ssize_t show_speed##value(struct device *dev,			\
 	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", mc->speed[value]);			\
-}									\
-static DEVICE_ATTR(speed##value, S_IWUGO | S_IRUGO,			\
-		show_speed##value, set_speed##value);
+}
+
+#define speed_attr(value) 						\
+	__ATTR(speed##value, S_IWUGO | S_IRUGO, 			\
+		show_speed##value, set_speed##value)
+
 show_set_speed(0);
 show_set_speed(1);
 
@@ -250,9 +253,12 @@ static ssize_t show_acceleration##value(struct device *dev,	 	\
 	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", mc->acceleration[value]);		\
-}									\
-static DEVICE_ATTR(acceleration##value, S_IWUGO | S_IRUGO,		\
-		show_acceleration##value, set_acceleration##value);
+}
+
+#define acceleration_attr(value)	\
+	__ATTR(acceleration##value, S_IWUGO | S_IRUGO,			\
+		show_acceleration##value, set_acceleration##value)
+
 show_set_acceleration(0);
 show_set_acceleration(1);
 
@@ -264,8 +270,10 @@ static ssize_t show_current##value(struct device *dev,			\
 	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%dmA\n", (int)mc->_current[value]);	\
-}									\
-static DEVICE_ATTR(current##value, S_IRUGO, show_current##value, NULL);
+}
+
+#define current_attr(value)	\
+	__ATTR(current##value, S_IRUGO, show_current##value, NULL)
 
 show_current(0);
 show_current(1);
@@ -278,14 +286,29 @@ static ssize_t show_input##value(struct device *dev,			\
 	struct motorcontrol *mc = dev_get_drvdata(dev);			\
 									\
 	return sprintf(buf, "%d\n", (int)mc->inputs[value]);		\
-}									\
-static DEVICE_ATTR(input##value, S_IRUGO, show_input##value, NULL);
+}
+
+#define input_attr(value)	\
+	__ATTR(input##value, S_IRUGO, show_input##value, NULL)
 
 show_input(0);
 show_input(1);
 show_input(2);
 show_input(3);
 
+static struct device_attribute dev_attrs[] = {
+	input_attr(0),
+	input_attr(1),
+	input_attr(2),
+	input_attr(3),
+	speed_attr(0),
+	speed_attr(1),
+	acceleration_attr(0),
+	acceleration_attr(1),
+	current_attr(0),
+	current_attr(1)
+};
+
 static int motorcontrol_probe(struct usb_interface *intf, const struct usb_device_id *id)
 {
 	struct usb_device *dev = interface_to_usbdev(intf);
@@ -293,7 +316,7 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic
 	struct usb_endpoint_descriptor *endpoint;
 	struct motorcontrol *mc;
 	int pipe, maxp, rc = -ENOMEM;
-	int bit, value;
+	int bit, value, i;
 
 	interface = intf->cur_altsetting;
 	if (interface->desc.bNumEndpoints != 1)
@@ -355,24 +378,18 @@ static int motorcontrol_probe(struct usb_interface *intf, const struct usb_devic
 		goto out;
 	}
 
-	device_create_file(mc->dev, &dev_attr_input0);
-	device_create_file(mc->dev, &dev_attr_input1);
-	device_create_file(mc->dev, &dev_attr_input2);
-	device_create_file(mc->dev, &dev_attr_input3);
-
-	device_create_file(mc->dev, &dev_attr_speed0);
-	device_create_file(mc->dev, &dev_attr_speed1);
-
-	device_create_file(mc->dev, &dev_attr_acceleration0);
-	device_create_file(mc->dev, &dev_attr_acceleration1);
-
-	device_create_file(mc->dev, &dev_attr_current0);
-	device_create_file(mc->dev, &dev_attr_current1);
+	for (i=0; i<ARRAY_SIZE(dev_attrs); i++) {
+		rc = device_create_file(mc->dev, &dev_attrs[i]);
+		if (rc)
+			goto out2;
+	}
 
 	dev_info(&intf->dev, "USB PhidgetMotorControl attached\n");
 
 	return 0;
-
+out2:
+	while (i-- > 0)
+		device_remove_file(mc->dev, &dev_attrs[i]);
 out:
 	if (mc) {
 		if (mc->irq)
@@ -393,6 +410,7 @@ out:
 static void motorcontrol_disconnect(struct usb_interface *interface)
 {
 	struct motorcontrol *mc;
+	int i;
 
 	mc = usb_get_intfdata(interface);
 	usb_set_intfdata(interface, NULL);
@@ -405,19 +423,8 @@ static void motorcontrol_disconnect(struct usb_interface *interface)
 
 	cancel_delayed_work(&mc->do_notify);
 
-	device_remove_file(mc->dev, &dev_attr_input0);
-	device_remove_file(mc->dev, &dev_attr_input1);
-	device_remove_file(mc->dev, &dev_attr_input2);
-	device_remove_file(mc->dev, &dev_attr_input3);
-
-	device_remove_file(mc->dev, &dev_attr_speed0);
-	device_remove_file(mc->dev, &dev_attr_speed1);
-
-	device_remove_file(mc->dev, &dev_attr_acceleration0);
-	device_remove_file(mc->dev, &dev_attr_acceleration1);
-
-	device_remove_file(mc->dev, &dev_attr_current0);
-	device_remove_file(mc->dev, &dev_attr_current1);
+	for (i=0; i<ARRAY_SIZE(dev_attrs); i++)
+		device_remove_file(mc->dev, &dev_attrs[i]);
 
 	device_unregister(mc->dev);
 
diff --git a/drivers/usb/misc/phidgetservo.c b/drivers/usb/misc/phidgetservo.c
index 66be9513fd6f..7163f05c5b27 100644
--- a/drivers/usb/misc/phidgetservo.c
+++ b/drivers/usb/misc/phidgetservo.c
@@ -1,7 +1,7 @@
 /*
  * USB PhidgetServo driver 1.0
  *
- * Copyright (C) 2004 Sean Young <sean@mess.org>
+ * Copyright (C) 2004, 2006 Sean Young <sean@mess.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -235,86 +235,108 @@ static ssize_t show_servo##value (struct device *dev,			\
 									\
 	return sprintf(buf, "%d.%02d\n", servo->degrees[value],		\
 				servo->minutes[value]);			\
-}									\
-static DEVICE_ATTR(servo##value, S_IWUGO | S_IRUGO,			\
-	  show_servo##value, set_servo##value);
+}
 
+#define servo_attr(value)						\
+	__ATTR(servo##value, S_IWUGO | S_IRUGO,				\
+		show_servo##value, set_servo##value)
 show_set(0);
 show_set(1);
 show_set(2);
 show_set(3);
 
+static struct device_attribute dev_attrs[] = {
+	servo_attr(0), servo_attr(1), servo_attr(2), servo_attr(3)
+};
+
 static int
 servo_probe(struct usb_interface *interface, const struct usb_device_id *id)
 {
 	struct usb_device *udev = interface_to_usbdev(interface);
 	struct phidget_servo *dev;
-	int bit, value;
+	int bit, value, rc;
+	int servo_count, i;
 
 	dev = kzalloc(sizeof (struct phidget_servo), GFP_KERNEL);
 	if (dev == NULL) {
 		dev_err(&interface->dev, "%s - out of memory\n", __FUNCTION__);
-		return -ENOMEM;
+		rc = -ENOMEM;
+		goto out;
 	}
 
 	dev->udev = usb_get_dev(udev);
 	dev->type = id->driver_info;
+	dev->dev_no = -1;
 	usb_set_intfdata(interface, dev);
 
         do {
                 bit = find_first_zero_bit(&device_no, sizeof(device_no));
                 value = test_and_set_bit(bit, &device_no);
-        } while(value);
+        } while (value);
 	dev->dev_no = bit;
 
 	dev->dev = device_create(phidget_class, &dev->udev->dev, 0,
 				 "servo%d", dev->dev_no);
 	if (IS_ERR(dev->dev)) {
-		int rc = PTR_ERR(dev->dev);
-		clear_bit(dev->dev_no, &device_no);
-		kfree(dev);
-		return rc;
+		rc = PTR_ERR(dev->dev);
+		dev->dev = NULL;
+		goto out;
 	}
 
-	device_create_file(dev->dev, &dev_attr_servo0);
-	if (dev->type & SERVO_COUNT_QUAD) {
-		device_create_file(dev->dev, &dev_attr_servo1);
-		device_create_file(dev->dev, &dev_attr_servo2);
-		device_create_file(dev->dev, &dev_attr_servo3);
+	servo_count = dev->type & SERVO_COUNT_QUAD ? 4 : 1;
+
+	for (i=0; i<servo_count; i++) {
+		rc = device_create_file(dev->dev, &dev_attrs[i]);
+		if (rc)
+			goto out2;
 	}
 
 	dev_info(&interface->dev, "USB %d-Motor PhidgetServo v%d.0 attached\n",
-		dev->type & SERVO_COUNT_QUAD ? 4 : 1,
-		dev->type & SERVO_VERSION_30 ? 3 : 2);
+		servo_count, dev->type & SERVO_VERSION_30 ? 3 : 2);
 
 	if (!(dev->type & SERVO_VERSION_30))
 		dev_info(&interface->dev,
 			 "WARNING: v2.0 not tested! Please report if it works.\n");
 
 	return 0;
+out2:
+	while (i-- > 0)
+		device_remove_file(dev->dev, &dev_attrs[i]);
+out:
+	if (dev) {
+		if (dev->dev)
+			device_unregister(dev->dev);
+		if (dev->dev_no >= 0)
+			clear_bit(dev->dev_no, &device_no);
+
+		kfree(dev);
+	}
+
+	return rc;
 }
 
 static void
 servo_disconnect(struct usb_interface *interface)
 {
 	struct phidget_servo *dev;
+	int servo_count, i;
 
 	dev = usb_get_intfdata(interface);
 	usb_set_intfdata(interface, NULL);
 
-	device_remove_file(dev->dev, &dev_attr_servo0);
-	if (dev->type & SERVO_COUNT_QUAD) {
-		device_remove_file(dev->dev, &dev_attr_servo1);
-		device_remove_file(dev->dev, &dev_attr_servo2);
-		device_remove_file(dev->dev, &dev_attr_servo3);
-	}
+	if (!dev)
+		return;
+
+	servo_count = dev->type & SERVO_COUNT_QUAD ? 4 : 1;
+
+	for (i=0; i<servo_count; i++)
+		device_remove_file(dev->dev, &dev_attrs[i]);
 
 	device_unregister(dev->dev);
 	usb_put_dev(dev->udev);
 
 	dev_info(&interface->dev, "USB %d-Motor PhidgetServo v%d.0 detached\n",
-		dev->type & SERVO_COUNT_QUAD ? 4 : 1,
-		dev->type & SERVO_VERSION_30 ? 3 : 2);
+		servo_count, dev->type & SERVO_VERSION_30 ? 3 : 2);
 
 	clear_bit(dev->dev_no, &device_no);
 	kfree(dev);
-- 
cgit v1.2.1


From 4a2a8a2cce86b9d001378cc25acb5c61e6ca7d63 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:05:01 -0400
Subject: usbfs: private mutex for open, release, and remove

The usbfs code doesn't provide sufficient mutual exclusion among open,
release, and remove.  Release vs. remove is okay because they both
acquire the device lock, but open is not exclusive with either one.  All
three routines modify the udev->filelist linked list, so they must not
run concurrently.

Apparently someone gave this a minimum amount of thought in the past by
explicitly acquiring the BKL at the start of the usbdev_open routine.
Oddly enough, there's a comment pointing out that locking is unnecessary
because chrdev_open already has acquired the BKL.

But this ignores the point that the files in /proc/bus/usb/* are not
char device files; they are regular files and so they don't get any
special locking.  Furthermore it's necessary to acquire the same lock in
the release and remove routines, which the code does not do.

Yet another problem arises because the same file_operations structure is
accessible through both the /proc/bus/usb/* and /dev/usb/usbdev* file
nodes.  Even when one of them has been removed, it's still possible for
userspace to open the other.  So simple locking around the individual
remove routines is insufficient; we need to lock the entire
usb_notify_remove_device notifier chain.

Rather than rely on the BKL, this patch (as723) introduces a new private
mutex for the purpose.  Holding the BKL while invoking a notifier chain
doesn't seem like a good idea.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devio.c  | 26 +++++++++++++++-----------
 drivers/usb/core/notify.c |  3 +++
 drivers/usb/core/usb.h    |  1 +
 3 files changed, 19 insertions(+), 11 deletions(-)

diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 32e03000420c..d8b0476237f3 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -59,6 +59,9 @@
 #define USB_DEVICE_MAX			USB_MAXBUS * 128
 static struct class *usb_device_class;
 
+/* Mutual exclusion for removal, open, and release */
+DEFINE_MUTEX(usbfs_mutex);
+
 struct async {
 	struct list_head asynclist;
 	struct dev_state *ps;
@@ -541,15 +544,13 @@ static int usbdev_open(struct inode *inode, struct file *file)
 	struct dev_state *ps;
 	int ret;
 
-	/* 
-	 * no locking necessary here, as chrdev_open has the kernel lock
-	 * (still acquire the kernel lock for safety)
-	 */
+	/* Protect against simultaneous removal or release */
+	mutex_lock(&usbfs_mutex);
+
 	ret = -ENOMEM;
 	if (!(ps = kmalloc(sizeof(struct dev_state), GFP_KERNEL)))
-		goto out_nolock;
+		goto out;
 
-	lock_kernel();
 	ret = -ENOENT;
 	/* check if we are called from a real node or usbfs */
 	if (imajor(inode) == USB_DEVICE_MAJOR)
@@ -579,9 +580,8 @@ static int usbdev_open(struct inode *inode, struct file *file)
 	list_add_tail(&ps->list, &dev->filelist);
 	file->private_data = ps;
  out:
-	unlock_kernel();
- out_nolock:
-        return ret;
+	mutex_unlock(&usbfs_mutex);
+	return ret;
 }
 
 static int usbdev_release(struct inode *inode, struct file *file)
@@ -591,7 +591,12 @@ static int usbdev_release(struct inode *inode, struct file *file)
 	unsigned int ifnum;
 
 	usb_lock_device(dev);
+
+	/* Protect against simultaneous open */
+	mutex_lock(&usbfs_mutex);
 	list_del_init(&ps->list);
+	mutex_unlock(&usbfs_mutex);
+
 	for (ifnum = 0; ps->ifclaimed && ifnum < 8*sizeof(ps->ifclaimed);
 			ifnum++) {
 		if (test_bit(ifnum, &ps->ifclaimed))
@@ -600,9 +605,8 @@ static int usbdev_release(struct inode *inode, struct file *file)
 	destroy_all_async(ps);
 	usb_unlock_device(dev);
 	usb_put_dev(dev);
-	ps->dev = NULL;
 	kfree(ps);
-        return 0;
+	return 0;
 }
 
 static int proc_control(struct dev_state *ps, void __user *arg)
diff --git a/drivers/usb/core/notify.c b/drivers/usb/core/notify.c
index b042676af0a5..6b36897ca151 100644
--- a/drivers/usb/core/notify.c
+++ b/drivers/usb/core/notify.c
@@ -50,8 +50,11 @@ void usb_notify_add_device(struct usb_device *udev)
 
 void usb_notify_remove_device(struct usb_device *udev)
 {
+	/* Protect against simultaneous usbfs open */
+	mutex_lock(&usbfs_mutex);
 	blocking_notifier_call_chain(&usb_notifier_list,
 			USB_DEVICE_REMOVE, udev);
+	mutex_unlock(&usbfs_mutex);
 }
 
 void usb_notify_add_bus(struct usb_bus *ubus)
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 49f69236b420..1217fbbe5829 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -59,6 +59,7 @@ static inline int is_active(struct usb_interface *f)
 extern const char *usbcore_name;
 
 /* usbfs stuff */
+extern struct mutex usbfs_mutex;
 extern struct usb_driver usbfs_driver;
 extern struct file_operations usbfs_devices_fops;
 extern struct file_operations usbfs_device_file_operations;
-- 
cgit v1.2.1


From 349710c3a79c0405911b8b604953f0c665e17756 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:05:56 -0400
Subject: usbfs: detect device unregistration

This patch (as711b) is a revised version of an earlier submission.  It
modifies the usbfs code to detect when a device has been unregistered from
usbfs, even if the device is still connected.  Although this can't happen
now, it will be able to happen after the upcoming changes to usb_generic.

Nobody objected to this patch when it was submitted before, so it should
be okay to apply this version.  The revision is merely to take into
account the changes introduced by as723, which touches the same driver.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devio.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index d8b0476237f3..e84f19d4089c 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -90,9 +90,10 @@ MODULE_PARM_DESC (usbfs_snoop, "true to log all usbfs traffic");
 
 #define	MAX_USBFS_BUFFER_SIZE	16384
 
-static inline int connected (struct usb_device *dev)
+static inline int connected (struct dev_state *ps)
 {
-	return dev->state != USB_STATE_NOTATTACHED;
+	return (!list_empty(&ps->list) &&
+			ps->dev->state != USB_STATE_NOTATTACHED);
 }
 
 static loff_t usbdev_lseek(struct file *file, loff_t offset, int orig)
@@ -130,7 +131,7 @@ static ssize_t usbdev_read(struct file *file, char __user *buf, size_t nbytes, l
 
 	pos = *ppos;
 	usb_lock_device(dev);
-	if (!connected(dev)) {
+	if (!connected(ps)) {
 		ret = -ENODEV;
 		goto err;
 	} else if (pos < 0) {
@@ -1326,7 +1327,7 @@ static int proc_ioctl(struct dev_state *ps, struct usbdevfs_ioctl *ctl)
 		}
 	}
 
-	if (!connected(ps->dev)) {
+	if (!connected(ps)) {
 		kfree(buf);
 		return -ENODEV;
 	}
@@ -1425,7 +1426,7 @@ static int usbdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
 	if (!(file->f_mode & FMODE_WRITE))
 		return -EPERM;
 	usb_lock_device(dev);
-	if (!connected(dev)) {
+	if (!connected(ps)) {
 		usb_unlock_device(dev);
 		return -ENODEV;
 	}
@@ -1566,7 +1567,7 @@ static unsigned int usbdev_poll(struct file *file, struct poll_table_struct *wai
 	poll_wait(file, &ps->wait, wait);
 	if (file->f_mode & FMODE_WRITE && !list_empty(&ps->async_completed))
 		mask |= POLLOUT | POLLWRNORM;
-	if (!connected(ps->dev))
+	if (!connected(ps))
 		mask |= POLLERR | POLLHUP;
 	return mask;
 }
-- 
cgit v1.2.1


From 121e287cb554f3d3402c85a1950d852691b08f5c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:06:36 -0400
Subject: usb-skeleton: don't submit URBs after disconnection

This patch (as712b) is a slight revision of one submitted earlier.  It
fixes the usb-skeleton example driver so that it won't try to submit
URBs after skel_disconnect() has returned.  This could cause errors, if
the driver was unbound and then a different driver was bound to the
device.  It also fixes a couple of small bugs in the skel_write()
routine.

The revised patch uses a slightly different test, suggested by Dave
Brownell, for determining whether to free a transfer buffer.  It's a
little clearer than the earlier version.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/usb-skeleton.c | 40 +++++++++++++++++++++++++++++++++-------
 1 file changed, 33 insertions(+), 7 deletions(-)

diff --git a/drivers/usb/usb-skeleton.c b/drivers/usb/usb-skeleton.c
index b362039792b3..33f0e81c58d3 100644
--- a/drivers/usb/usb-skeleton.c
+++ b/drivers/usb/usb-skeleton.c
@@ -1,5 +1,5 @@
 /*
- * USB Skeleton driver - 2.0
+ * USB Skeleton driver - 2.1
  *
  * Copyright (C) 2001-2004 Greg Kroah-Hartman (greg@kroah.com)
  *
@@ -8,8 +8,7 @@
  *	published by the Free Software Foundation, version 2.
  *
  * This driver is based on the 2.6.3 version of drivers/usb/usb-skeleton.c 
- * but has been rewritten to be easy to read and use, as no locks are now
- * needed anymore.
+ * but has been rewritten to be easier to read and use.
  *
  */
 
@@ -21,6 +20,7 @@
 #include <linux/kref.h>
 #include <asm/uaccess.h>
 #include <linux/usb.h>
+#include <linux/mutex.h>
 
 
 /* Define these values to match your devices */
@@ -52,6 +52,7 @@ struct usb_skel {
 	__u8			bulk_in_endpointAddr;	/* the address of the bulk in endpoint */
 	__u8			bulk_out_endpointAddr;	/* the address of the bulk out endpoint */
 	struct kref		kref;
+	struct mutex		io_mutex;		/* synchronize I/O with disconnect */
 };
 #define to_skel_dev(d) container_of(d, struct usb_skel, kref)
 
@@ -119,7 +120,13 @@ static ssize_t skel_read(struct file *file, char *buffer, size_t count, loff_t *
 	int bytes_read;
 
 	dev = (struct usb_skel *)file->private_data;
-	
+
+	mutex_lock(&dev->io_mutex);
+	if (!dev->interface) {		/* disconnect() was called */
+		retval = -ENODEV;
+		goto exit;
+	}
+
 	/* do a blocking bulk read to get data from the device */
 	retval = usb_bulk_msg(dev->udev,
 			      usb_rcvbulkpipe(dev->udev, dev->bulk_in_endpointAddr),
@@ -135,6 +142,8 @@ static ssize_t skel_read(struct file *file, char *buffer, size_t count, loff_t *
 			retval = bytes_read;
 	}
 
+exit:
+	mutex_unlock(&dev->io_mutex);
 	return retval;
 }
 
@@ -179,6 +188,12 @@ static ssize_t skel_write(struct file *file, const char *user_buffer, size_t cou
 		goto exit;
 	}
 
+	mutex_lock(&dev->io_mutex);
+	if (!dev->interface) {		/* disconnect() was called */
+		retval = -ENODEV;
+		goto error;
+	}
+
 	/* create a urb, and a buffer for it, and copy the data to the urb */
 	urb = usb_alloc_urb(0, GFP_KERNEL);
 	if (!urb) {
@@ -213,13 +228,18 @@ static ssize_t skel_write(struct file *file, const char *user_buffer, size_t cou
 	/* release our reference to this urb, the USB core will eventually free it entirely */
 	usb_free_urb(urb);
 
-exit:
+	mutex_unlock(&dev->io_mutex);
 	return writesize;
 
 error:
-	usb_buffer_free(dev->udev, writesize, buf, urb->transfer_dma);
-	usb_free_urb(urb);
+	if (urb) {
+		usb_buffer_free(dev->udev, writesize, buf, urb->transfer_dma);
+		usb_free_urb(urb);
+	}
+	mutex_unlock(&dev->io_mutex);
 	up(&dev->limit_sem);
+
+exit:
 	return retval;
 }
 
@@ -258,6 +278,7 @@ static int skel_probe(struct usb_interface *interface, const struct usb_device_i
 	}
 	kref_init(&dev->kref);
 	sema_init(&dev->limit_sem, WRITES_IN_FLIGHT);
+	mutex_init(&dev->io_mutex);
 
 	dev->udev = usb_get_dev(interface_to_usbdev(interface));
 	dev->interface = interface;
@@ -334,6 +355,11 @@ static void skel_disconnect(struct usb_interface *interface)
 	/* give back our minor */
 	usb_deregister_dev(interface, &skel_class);
 
+	/* prevent more I/O from starting */
+	mutex_lock(&dev->io_mutex);
+	dev->interface = NULL;
+	mutex_unlock(&dev->io_mutex);
+
 	unlock_kernel();
 
 	/* decrement our usage count */
-- 
cgit v1.2.1


From 140d8f687457c40a66af362838fac0d7893e7df5 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:07:21 -0400
Subject: usbcore: rename usb_suspend_device to usb_port_suspend

This revised patch (as715b) renames usb_suspend_device to
usb_port_suspend, usb_resume_device to usb_port_resume, and
finish_device_resume to finish_port_resume.  There was no objection to
the original version of the patch so this should be okay to apply.

The revision was needed only because I have re-arranged the order of the
earlier patches.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 38 +++++++++++++++++++++-----------------
 drivers/usb/core/usb.c |  7 ++-----
 drivers/usb/core/usb.h |  4 ++--
 3 files changed, 25 insertions(+), 24 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 26c8cb5f3e67..b00514d9a605 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1407,9 +1407,9 @@ int usb_new_device(struct usb_device *udev)
 		 * (Includes HNP test device.)
 		 */
 		if (udev->bus->b_hnp_enable || udev->bus->is_b_host) {
-			static int __usb_suspend_device(struct usb_device *,
+			static int __usb_port_suspend(struct usb_device *,
 						int port1);
-			err = __usb_suspend_device(udev, udev->bus->otg_port);
+			err = __usb_port_suspend(udev, udev->bus->otg_port);
 			if (err < 0)
 				dev_dbg(&udev->dev, "HNP fail, %d\n", err);
 		}
@@ -1684,7 +1684,7 @@ static int hub_port_suspend(struct usb_hub *hub, int port1,
  * the root hub for their bus goes into global suspend ... so we don't
  * (falsely) update the device power state to say it suspended.
  */
-static int __usb_suspend_device (struct usb_device *udev, int port1)
+static int __usb_port_suspend (struct usb_device *udev, int port1)
 {
 	int	status = 0;
 
@@ -1712,8 +1712,8 @@ static int __usb_suspend_device (struct usb_device *udev, int port1)
 		}
 	}
 
-	/* we only change a device's upstream USB link.
-	 * root hubs have no upstream USB link.
+	/* we change the device's upstream USB link,
+	 * but root hubs have no upstream USB link.
 	 */
 	if (udev->parent)
 		status = hub_port_suspend(hdev_to_hub(udev->parent), port1,
@@ -1727,14 +1727,14 @@ static int __usb_suspend_device (struct usb_device *udev, int port1)
 #endif
 
 /*
- * usb_suspend_device - suspend a usb device
+ * usb_port_suspend - suspend a usb device's upstream port
  * @udev: device that's no longer in active use
  * Context: must be able to sleep; device not locked; pm locks held
  *
  * Suspends a USB device that isn't in active use, conserving power.
  * Devices may wake out of a suspend, if anything important happens,
  * using the remote wakeup mechanism.  They may also be taken out of
- * suspend by the host, using usb_resume_device().  It's also routine
+ * suspend by the host, using usb_port_resume().  It's also routine
  * to disconnect devices while they are suspended.
  *
  * This only affects the USB hardware for a device; its interfaces
@@ -1746,12 +1746,12 @@ static int __usb_suspend_device (struct usb_device *udev, int port1)
  *
  * Returns 0 on success, else negative errno.
  */
-int usb_suspend_device(struct usb_device *udev)
+int usb_port_suspend(struct usb_device *udev)
 {
 #ifdef	CONFIG_USB_SUSPEND
 	if (udev->state == USB_STATE_NOTATTACHED)
 		return -ENODEV;
-	return __usb_suspend_device(udev, udev->portnum);
+	return __usb_port_suspend(udev, udev->portnum);
 #else
 	/* NOTE:  udev->state unchanged, it's not lying ... */
 	udev->dev.power.power_state = PMSG_SUSPEND;
@@ -1767,7 +1767,7 @@ int usb_suspend_device(struct usb_device *udev)
  * resume (by host) or remote wakeup (by device) ... now see what changed
  * in the tree that's rooted at this device.
  */
-static int finish_device_resume(struct usb_device *udev)
+static int finish_port_resume(struct usb_device *udev)
 {
 	int	status;
 	u16	devstatus;
@@ -1891,7 +1891,7 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 			/* TRSMRCY = 10 msec */
 			msleep(10);
 			if (udev)
-				status = finish_device_resume(udev);
+				status = finish_port_resume(udev);
 		}
 	}
 	if (status < 0)
@@ -1903,7 +1903,7 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 #endif
 
 /*
- * usb_resume_device - re-activate a suspended usb device
+ * usb_port_resume - re-activate a suspended usb device's upstream port
  * @udev: device to re-activate
  * Context: must be able to sleep; device not locked; pm locks held
  *
@@ -1915,14 +1915,16 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
  *
  * Returns 0 on success, else negative errno.
  */
-int usb_resume_device(struct usb_device *udev)
+int usb_port_resume(struct usb_device *udev)
 {
 	int	status;
 
 	if (udev->state == USB_STATE_NOTATTACHED)
 		return -ENODEV;
 
-	/* selective resume of one downstream hub-to-device port */
+	/* we change the device's upstream USB link,
+	 * but root hubs have no upstream USB link.
+	 */
 	if (udev->parent) {
 #ifdef	CONFIG_USB_SUSPEND
 		if (udev->state == USB_STATE_SUSPENDED) {
@@ -1934,7 +1936,7 @@ int usb_resume_device(struct usb_device *udev)
 #endif
 			status = 0;
 	} else
-		status = finish_device_resume(udev);
+		status = finish_port_resume(udev);
 	if (status < 0)
 		dev_dbg(&udev->dev, "can't resume, status %d\n",
 			status);
@@ -1962,7 +1964,7 @@ static int remote_wakeup(struct usb_device *udev)
 		dev_dbg(&udev->dev, "RESUME (wakeup)\n");
 		/* TRSMRCY = 10 msec */
 		msleep(10);
-		status = finish_device_resume(udev);
+		status = finish_port_resume(udev);
 	}
 	usb_unlock_device(udev);
 #endif
@@ -2069,7 +2071,7 @@ static int hub_resume(struct usb_interface *intf)
 		if (portstat & USB_PORT_STAT_SUSPEND)
 			status = hub_port_resume(hub, port1, udev);
 		else {
-			status = finish_device_resume(udev);
+			status = finish_port_resume(udev);
 			if (status < 0) {
 				dev_dbg(&intf->dev, "resume port %d --> %d\n",
 					port1, status);
@@ -3128,6 +3130,7 @@ re_enumerate:
 	hub_port_logical_disconnect(parent_hub, port1);
 	return -ENODEV;
 }
+EXPORT_SYMBOL(usb_reset_device);
 
 /**
  * usb_reset_composite_device - warn interface drivers and perform a USB port reset
@@ -3206,3 +3209,4 @@ int usb_reset_composite_device(struct usb_device *udev,
 
 	return ret;
 }
+EXPORT_SYMBOL(usb_reset_composite_device);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index ab766e0fe4ef..b28a31b20308 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -1014,7 +1014,7 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
 		status = device_for_each_child(dev, NULL, verify_suspended);
 		if (status)
 			return status;
- 		return usb_suspend_device (to_usb_device(dev));
+ 		return usb_port_suspend(to_usb_device(dev));
 	}
 
 	if ((dev->driver == NULL) ||
@@ -1061,7 +1061,7 @@ static int usb_generic_resume(struct device *dev)
 		udev = to_usb_device(dev);
 		if (udev->state == USB_STATE_NOTATTACHED)
 			return 0;
-		return usb_resume_device (to_usb_device(dev));
+		return usb_port_resume(udev);
 	}
 
 	if ((dev->driver == NULL) ||
@@ -1209,9 +1209,6 @@ EXPORT_SYMBOL(usb_find_interface);
 EXPORT_SYMBOL(usb_ifnum_to_if);
 EXPORT_SYMBOL(usb_altnum_to_altsetting);
 
-EXPORT_SYMBOL(usb_reset_device);
-EXPORT_SYMBOL(usb_reset_composite_device);
-
 EXPORT_SYMBOL(__usb_get_extra_descriptor);
 
 EXPORT_SYMBOL(usb_find_device);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 1217fbbe5829..5a4eff5eec86 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -30,8 +30,8 @@ extern void usb_major_cleanup(void);
 extern int usb_host_init(void);
 extern void usb_host_cleanup(void);
 
-extern int usb_suspend_device(struct usb_device *dev);
-extern int usb_resume_device(struct usb_device *dev);
+extern int usb_port_suspend(struct usb_device *dev);
+extern int usb_port_resume(struct usb_device *dev);
 
 extern struct device_driver usb_generic_driver;
 extern int usb_generic_driver_data;
-- 
cgit v1.2.1


From 36e56a34586783c7986ce09d39db80b27c95ce24 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:08:06 -0400
Subject: usbcore: move code among source files

This revised patch (as713b) moves a few routines among source files in
usbcore.  Some driver-related code in usb.c (claiming interfaces and
matching IDs) is moved to driver.c, where it belongs.  Also the
usb_generic stuff in driver.c is moved to a new source file: generic.c.
(That's the reason for revising the patch.)  Although not very big now,
it will get bigger in a later patch.

None of the code has been changed; it has only been re-arranged.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/Makefile  |   2 +-
 drivers/usb/core/driver.c  | 348 ++++++++++++++++++++++++++++++++++++++++-----
 drivers/usb/core/generic.c |  53 +++++++
 drivers/usb/core/usb.c     | 307 ---------------------------------------
 drivers/usb/core/usb.h     |   2 +-
 5 files changed, 370 insertions(+), 342 deletions(-)
 create mode 100644 drivers/usb/core/generic.c

diff --git a/drivers/usb/core/Makefile b/drivers/usb/core/Makefile
index ec510922af63..34e9bac319b4 100644
--- a/drivers/usb/core/Makefile
+++ b/drivers/usb/core/Makefile
@@ -4,7 +4,7 @@
 
 usbcore-objs	:= usb.o hub.o hcd.o urb.o message.o driver.o \
 			config.o file.o buffer.o sysfs.o endpoint.o \
-			devio.o notify.o
+			devio.o notify.o generic.o
 
 ifeq ($(CONFIG_PCI),y)
 	usbcore-objs	+= hcd-pci.o
diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index ec8906501415..8dcf2cd0c569 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -17,7 +17,8 @@
  *
  * NOTE! This is not actually a driver at all, rather this is
  * just a collection of helper routines that implement the
- * generic USB things that the real drivers can use..
+ * matching, probing, releasing, suspending and resuming for
+ * real drivers.
  *
  */
 
@@ -34,38 +35,6 @@ struct usb_dynid {
 	struct usb_device_id id;
 };
 
-
-static int generic_probe(struct device *dev)
-{
-	return 0;
-}
-static int generic_remove(struct device *dev)
-{
-	struct usb_device *udev = to_usb_device(dev);
-
-	/* if this is only an unbind, not a physical disconnect, then
-	 * unconfigure the device */
-	if (udev->state == USB_STATE_CONFIGURED)
-		usb_set_configuration(udev, 0);
-
-	/* in case the call failed or the device was suspended */
-	if (udev->state >= USB_STATE_CONFIGURED)
-		usb_disable_device(udev, 0);
-	return 0;
-}
-
-struct device_driver usb_generic_driver = {
-	.owner = THIS_MODULE,
-	.name =	"usb",
-	.bus = &usb_bus_type,
-	.probe = generic_probe,
-	.remove = generic_remove,
-};
-
-/* Fun hack to determine if the struct device is a
- * usb device or a usb interface. */
-int usb_generic_driver_data;
-
 #ifdef CONFIG_HOTPLUG
 
 /*
@@ -238,6 +207,89 @@ static int usb_unbind_interface(struct device *dev)
 	return 0;
 }
 
+/**
+ * usb_driver_claim_interface - bind a driver to an interface
+ * @driver: the driver to be bound
+ * @iface: the interface to which it will be bound; must be in the
+ *	usb device's active configuration
+ * @priv: driver data associated with that interface
+ *
+ * This is used by usb device drivers that need to claim more than one
+ * interface on a device when probing (audio and acm are current examples).
+ * No device driver should directly modify internal usb_interface or
+ * usb_device structure members.
+ *
+ * Few drivers should need to use this routine, since the most natural
+ * way to bind to an interface is to return the private data from
+ * the driver's probe() method.
+ *
+ * Callers must own the device lock and the driver model's usb_bus_type.subsys
+ * writelock.  So driver probe() entries don't need extra locking,
+ * but other call contexts may need to explicitly claim those locks.
+ */
+int usb_driver_claim_interface(struct usb_driver *driver,
+				struct usb_interface *iface, void* priv)
+{
+	struct device *dev = &iface->dev;
+
+	if (dev->driver)
+		return -EBUSY;
+
+	dev->driver = &driver->driver;
+	usb_set_intfdata(iface, priv);
+	iface->condition = USB_INTERFACE_BOUND;
+	mark_active(iface);
+
+	/* if interface was already added, bind now; else let
+	 * the future device_add() bind it, bypassing probe()
+	 */
+	if (device_is_registered(dev))
+		device_bind_driver(dev);
+
+	return 0;
+}
+EXPORT_SYMBOL(usb_driver_claim_interface);
+
+/**
+ * usb_driver_release_interface - unbind a driver from an interface
+ * @driver: the driver to be unbound
+ * @iface: the interface from which it will be unbound
+ *
+ * This can be used by drivers to release an interface without waiting
+ * for their disconnect() methods to be called.  In typical cases this
+ * also causes the driver disconnect() method to be called.
+ *
+ * This call is synchronous, and may not be used in an interrupt context.
+ * Callers must own the device lock and the driver model's usb_bus_type.subsys
+ * writelock.  So driver disconnect() entries don't need extra locking,
+ * but other call contexts may need to explicitly claim those locks.
+ */
+void usb_driver_release_interface(struct usb_driver *driver,
+					struct usb_interface *iface)
+{
+	struct device *dev = &iface->dev;
+
+	/* this should never happen, don't release something that's not ours */
+	if (!dev->driver || dev->driver != &driver->driver)
+		return;
+
+	/* don't release from within disconnect() */
+	if (iface->condition != USB_INTERFACE_BOUND)
+		return;
+
+	/* don't release if the interface hasn't been added yet */
+	if (device_is_registered(dev)) {
+		iface->condition = USB_INTERFACE_UNBINDING;
+		device_release_driver(dev);
+	}
+
+	dev->driver = NULL;
+	usb_set_intfdata(iface, NULL);
+	iface->condition = USB_INTERFACE_UNBOUND;
+	mark_quiesced(iface);
+}
+EXPORT_SYMBOL(usb_driver_release_interface);
+
 /* returns 0 if no match, 1 if match */
 static int usb_match_one_id(struct usb_interface *interface,
 			    const struct usb_device_id *id)
@@ -402,6 +454,120 @@ int usb_device_match(struct device *dev, struct device_driver *drv)
 	return 0;
 }
 
+#ifdef	CONFIG_HOTPLUG
+
+/*
+ * This sends an uevent to userspace, typically helping to load driver
+ * or other modules, configure the device, and more.  Drivers can provide
+ * a MODULE_DEVICE_TABLE to help with module loading subtasks.
+ *
+ * We're called either from khubd (the typical case) or from root hub
+ * (init, kapmd, modprobe, rmmod, etc), but the agents need to handle
+ * delays in event delivery.  Use sysfs (and DEVPATH) to make sure the
+ * device (and this configuration!) are still present.
+ */
+static int usb_uevent(struct device *dev, char **envp, int num_envp,
+		      char *buffer, int buffer_size)
+{
+	struct usb_interface *intf;
+	struct usb_device *usb_dev;
+	struct usb_host_interface *alt;
+	int i = 0;
+	int length = 0;
+
+	if (!dev)
+		return -ENODEV;
+
+	/* driver is often null here; dev_dbg() would oops */
+	pr_debug ("usb %s: uevent\n", dev->bus_id);
+
+	/* Must check driver_data here, as on remove driver is always NULL */
+	if ((dev->driver == &usb_generic_driver) ||
+	    (dev->driver_data == &usb_generic_driver_data))
+		return 0;
+
+	intf = to_usb_interface(dev);
+	usb_dev = interface_to_usbdev (intf);
+	alt = intf->cur_altsetting;
+
+	if (usb_dev->devnum < 0) {
+		pr_debug ("usb %s: already deleted?\n", dev->bus_id);
+		return -ENODEV;
+	}
+	if (!usb_dev->bus) {
+		pr_debug ("usb %s: bus removed?\n", dev->bus_id);
+		return -ENODEV;
+	}
+
+#ifdef	CONFIG_USB_DEVICEFS
+	/* If this is available, userspace programs can directly read
+	 * all the device descriptors we don't tell them about.  Or
+	 * even act as usermode drivers.
+	 *
+	 * FIXME reduce hardwired intelligence here
+	 */
+	if (add_uevent_var(envp, num_envp, &i,
+			   buffer, buffer_size, &length,
+			   "DEVICE=/proc/bus/usb/%03d/%03d",
+			   usb_dev->bus->busnum, usb_dev->devnum))
+		return -ENOMEM;
+#endif
+
+	/* per-device configurations are common */
+	if (add_uevent_var(envp, num_envp, &i,
+			   buffer, buffer_size, &length,
+			   "PRODUCT=%x/%x/%x",
+			   le16_to_cpu(usb_dev->descriptor.idVendor),
+			   le16_to_cpu(usb_dev->descriptor.idProduct),
+			   le16_to_cpu(usb_dev->descriptor.bcdDevice)))
+		return -ENOMEM;
+
+	/* class-based driver binding models */
+	if (add_uevent_var(envp, num_envp, &i,
+			   buffer, buffer_size, &length,
+			   "TYPE=%d/%d/%d",
+			   usb_dev->descriptor.bDeviceClass,
+			   usb_dev->descriptor.bDeviceSubClass,
+			   usb_dev->descriptor.bDeviceProtocol))
+		return -ENOMEM;
+
+	if (add_uevent_var(envp, num_envp, &i,
+			   buffer, buffer_size, &length,
+			   "INTERFACE=%d/%d/%d",
+			   alt->desc.bInterfaceClass,
+			   alt->desc.bInterfaceSubClass,
+			   alt->desc.bInterfaceProtocol))
+		return -ENOMEM;
+
+	if (add_uevent_var(envp, num_envp, &i,
+			   buffer, buffer_size, &length,
+			   "MODALIAS=usb:v%04Xp%04Xd%04Xdc%02Xdsc%02Xdp%02Xic%02Xisc%02Xip%02X",
+			   le16_to_cpu(usb_dev->descriptor.idVendor),
+			   le16_to_cpu(usb_dev->descriptor.idProduct),
+			   le16_to_cpu(usb_dev->descriptor.bcdDevice),
+			   usb_dev->descriptor.bDeviceClass,
+			   usb_dev->descriptor.bDeviceSubClass,
+			   usb_dev->descriptor.bDeviceProtocol,
+			   alt->desc.bInterfaceClass,
+			   alt->desc.bInterfaceSubClass,
+			   alt->desc.bInterfaceProtocol))
+		return -ENOMEM;
+
+	envp[i] = NULL;
+
+	return 0;
+}
+
+#else
+
+static int usb_uevent(struct device *dev, char **envp,
+			int num_envp, char *buffer, int buffer_size)
+{
+	return -ENODEV;
+}
+
+#endif	/* CONFIG_HOTPLUG */
+
 /**
  * usb_register_driver - register a USB driver
  * @new_driver: USB operations for the driver
@@ -469,3 +635,119 @@ void usb_deregister(struct usb_driver *driver)
 	usbfs_update_special();
 }
 EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
+
+#ifdef CONFIG_PM
+
+static int verify_suspended(struct device *dev, void *unused)
+{
+	if (dev->driver == NULL)
+		return 0;
+	return (dev->power.power_state.event == PM_EVENT_ON) ? -EBUSY : 0;
+}
+
+static int usb_generic_suspend(struct device *dev, pm_message_t message)
+{
+	struct usb_interface	*intf;
+	struct usb_driver	*driver;
+	int			status;
+
+	/* USB devices enter SUSPEND state through their hubs, but can be
+	 * marked for FREEZE as soon as their children are already idled.
+	 * But those semantics are useless, so we equate the two (sigh).
+	 */
+	if (dev->driver == &usb_generic_driver) {
+		if (dev->power.power_state.event == message.event)
+			return 0;
+		/* we need to rule out bogus requests through sysfs */
+		status = device_for_each_child(dev, NULL, verify_suspended);
+		if (status)
+			return status;
+ 		return usb_port_suspend(to_usb_device(dev));
+	}
+
+	if ((dev->driver == NULL) ||
+	    (dev->driver_data == &usb_generic_driver_data))
+		return 0;
+
+	intf = to_usb_interface(dev);
+	driver = to_usb_driver(dev->driver);
+
+	/* with no hardware, USB interfaces only use FREEZE and ON states */
+	if (!is_active(intf))
+		return 0;
+
+	if (driver->suspend && driver->resume) {
+		status = driver->suspend(intf, message);
+		if (status)
+			dev_err(dev, "%s error %d\n", "suspend", status);
+		else
+			mark_quiesced(intf);
+	} else {
+		// FIXME else if there's no suspend method, disconnect...
+		dev_warn(dev, "no suspend for driver %s?\n", driver->name);
+		mark_quiesced(intf);
+		status = 0;
+	}
+	return status;
+}
+
+static int usb_generic_resume(struct device *dev)
+{
+	struct usb_interface	*intf;
+	struct usb_driver	*driver;
+	struct usb_device	*udev;
+	int			status;
+
+	if (dev->power.power_state.event == PM_EVENT_ON)
+		return 0;
+
+	/* mark things as "on" immediately, no matter what errors crop up */
+	dev->power.power_state.event = PM_EVENT_ON;
+
+	/* devices resume through their hubs */
+	if (dev->driver == &usb_generic_driver) {
+		udev = to_usb_device(dev);
+		if (udev->state == USB_STATE_NOTATTACHED)
+			return 0;
+		return usb_port_resume(udev);
+	}
+
+	if ((dev->driver == NULL) ||
+	    (dev->driver_data == &usb_generic_driver_data)) {
+		dev->power.power_state.event = PM_EVENT_FREEZE;
+		return 0;
+	}
+
+	intf = to_usb_interface(dev);
+	driver = to_usb_driver(dev->driver);
+
+	udev = interface_to_usbdev(intf);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return 0;
+
+	/* if driver was suspended, it has a resume method;
+	 * however, sysfs can wrongly mark things as suspended
+	 * (on the "no suspend method" FIXME path above)
+	 */
+	if (driver->resume) {
+		status = driver->resume(intf);
+		if (status) {
+			dev_err(dev, "%s error %d\n", "resume", status);
+			mark_quiesced(intf);
+		}
+	} else
+		dev_warn(dev, "no resume for driver %s?\n", driver->name);
+	return 0;
+}
+
+#endif /* CONFIG_PM */
+
+struct bus_type usb_bus_type = {
+	.name =		"usb",
+	.match =	usb_device_match,
+	.uevent =	usb_uevent,
+#ifdef CONFIG_PM
+	.suspend =	usb_generic_suspend,
+	.resume =	usb_generic_resume,
+#endif
+};
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
new file mode 100644
index 000000000000..7bab9769b34f
--- /dev/null
+++ b/drivers/usb/core/generic.c
@@ -0,0 +1,53 @@
+/*
+ * drivers/usb/generic.c - generic driver for USB devices (not interfaces)
+ *
+ * (C) Copyright 2005 Greg Kroah-Hartman <gregkh@suse.de>
+ *
+ * based on drivers/usb/usb.c which had the following copyrights:
+ *	(C) Copyright Linus Torvalds 1999
+ *	(C) Copyright Johannes Erdfelt 1999-2001
+ *	(C) Copyright Andreas Gal 1999
+ *	(C) Copyright Gregory P. Smith 1999
+ *	(C) Copyright Deti Fliegl 1999 (new USB architecture)
+ *	(C) Copyright Randy Dunlap 2000
+ *	(C) Copyright David Brownell 2000-2004
+ *	(C) Copyright Yggdrasil Computing, Inc. 2000
+ *		(usb_device_id matching changes by Adam J. Richter)
+ *	(C) Copyright Greg Kroah-Hartman 2002-2003
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/usb.h>
+#include "usb.h"
+
+static int generic_probe(struct device *dev)
+{
+	return 0;
+}
+static int generic_remove(struct device *dev)
+{
+	struct usb_device *udev = to_usb_device(dev);
+
+	/* if this is only an unbind, not a physical disconnect, then
+	 * unconfigure the device */
+	if (udev->state == USB_STATE_CONFIGURED)
+		usb_set_configuration(udev, 0);
+
+	/* in case the call failed or the device was suspended */
+	if (udev->state >= USB_STATE_CONFIGURED)
+		usb_disable_device(udev, 0);
+	return 0;
+}
+
+struct device_driver usb_generic_driver = {
+	.owner = THIS_MODULE,
+	.name =	"usb",
+	.bus = &usb_bus_type,
+	.probe = generic_probe,
+	.remove = generic_remove,
+};
+
+/* Fun hack to determine if the struct device is a
+ * usb device or a usb interface. */
+int usb_generic_driver_data;
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index b28a31b20308..0b8c67bcde60 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -112,87 +112,6 @@ struct usb_host_interface *usb_altnum_to_altsetting(struct usb_interface *intf,
 	return NULL;
 }
 
-/**
- * usb_driver_claim_interface - bind a driver to an interface
- * @driver: the driver to be bound
- * @iface: the interface to which it will be bound; must be in the
- *	usb device's active configuration
- * @priv: driver data associated with that interface
- *
- * This is used by usb device drivers that need to claim more than one
- * interface on a device when probing (audio and acm are current examples).
- * No device driver should directly modify internal usb_interface or
- * usb_device structure members.
- *
- * Few drivers should need to use this routine, since the most natural
- * way to bind to an interface is to return the private data from
- * the driver's probe() method.
- *
- * Callers must own the device lock and the driver model's usb_bus_type.subsys
- * writelock.  So driver probe() entries don't need extra locking,
- * but other call contexts may need to explicitly claim those locks.
- */
-int usb_driver_claim_interface(struct usb_driver *driver,
-				struct usb_interface *iface, void* priv)
-{
-	struct device *dev = &iface->dev;
-
-	if (dev->driver)
-		return -EBUSY;
-
-	dev->driver = &driver->driver;
-	usb_set_intfdata(iface, priv);
-	iface->condition = USB_INTERFACE_BOUND;
-	mark_active(iface);
-
-	/* if interface was already added, bind now; else let
-	 * the future device_add() bind it, bypassing probe()
-	 */
-	if (device_is_registered(dev))
-		device_bind_driver(dev);
-
-	return 0;
-}
-
-/**
- * usb_driver_release_interface - unbind a driver from an interface
- * @driver: the driver to be unbound
- * @iface: the interface from which it will be unbound
- *
- * This can be used by drivers to release an interface without waiting
- * for their disconnect() methods to be called.  In typical cases this
- * also causes the driver disconnect() method to be called.
- *
- * This call is synchronous, and may not be used in an interrupt context.
- * Callers must own the device lock and the driver model's usb_bus_type.subsys
- * writelock.  So driver disconnect() entries don't need extra locking,
- * but other call contexts may need to explicitly claim those locks.
- */
-void usb_driver_release_interface(struct usb_driver *driver,
-					struct usb_interface *iface)
-{
-	struct device *dev = &iface->dev;
-
-	/* this should never happen, don't release something that's not ours */
-	if (!dev->driver || dev->driver != &driver->driver)
-		return;
-
-	/* don't release from within disconnect() */
-	if (iface->condition != USB_INTERFACE_BOUND)
-		return;
-
-	/* don't release if the interface hasn't been added yet */
-	if (device_is_registered(dev)) {
-		iface->condition = USB_INTERFACE_UNBINDING;
-		device_release_driver(dev);
-	}
-
-	dev->driver = NULL;
-	usb_set_intfdata(iface, NULL);
-	iface->condition = USB_INTERFACE_UNBOUND;
-	mark_quiesced(iface);
-}
-
 struct find_interface_arg {
 	int minor;
 	struct usb_interface *interface;
@@ -234,120 +153,6 @@ struct usb_interface *usb_find_interface(struct usb_driver *drv, int minor)
 	return argb.interface;
 }
 
-#ifdef	CONFIG_HOTPLUG
-
-/*
- * This sends an uevent to userspace, typically helping to load driver
- * or other modules, configure the device, and more.  Drivers can provide
- * a MODULE_DEVICE_TABLE to help with module loading subtasks.
- *
- * We're called either from khubd (the typical case) or from root hub
- * (init, kapmd, modprobe, rmmod, etc), but the agents need to handle
- * delays in event delivery.  Use sysfs (and DEVPATH) to make sure the
- * device (and this configuration!) are still present.
- */
-static int usb_uevent(struct device *dev, char **envp, int num_envp,
-		      char *buffer, int buffer_size)
-{
-	struct usb_interface *intf;
-	struct usb_device *usb_dev;
-	struct usb_host_interface *alt;
-	int i = 0;
-	int length = 0;
-
-	if (!dev)
-		return -ENODEV;
-
-	/* driver is often null here; dev_dbg() would oops */
-	pr_debug ("usb %s: uevent\n", dev->bus_id);
-
-	/* Must check driver_data here, as on remove driver is always NULL */
-	if ((dev->driver == &usb_generic_driver) || 
-	    (dev->driver_data == &usb_generic_driver_data))
-		return 0;
-
-	intf = to_usb_interface(dev);
-	usb_dev = interface_to_usbdev (intf);
-	alt = intf->cur_altsetting;
-
-	if (usb_dev->devnum < 0) {
-		pr_debug ("usb %s: already deleted?\n", dev->bus_id);
-		return -ENODEV;
-	}
-	if (!usb_dev->bus) {
-		pr_debug ("usb %s: bus removed?\n", dev->bus_id);
-		return -ENODEV;
-	}
-
-#ifdef	CONFIG_USB_DEVICEFS
-	/* If this is available, userspace programs can directly read
-	 * all the device descriptors we don't tell them about.  Or
-	 * even act as usermode drivers.
-	 *
-	 * FIXME reduce hardwired intelligence here
-	 */
-	if (add_uevent_var(envp, num_envp, &i,
-			   buffer, buffer_size, &length,
-			   "DEVICE=/proc/bus/usb/%03d/%03d",
-			   usb_dev->bus->busnum, usb_dev->devnum))
-		return -ENOMEM;
-#endif
-
-	/* per-device configurations are common */
-	if (add_uevent_var(envp, num_envp, &i,
-			   buffer, buffer_size, &length,
-			   "PRODUCT=%x/%x/%x",
-			   le16_to_cpu(usb_dev->descriptor.idVendor),
-			   le16_to_cpu(usb_dev->descriptor.idProduct),
-			   le16_to_cpu(usb_dev->descriptor.bcdDevice)))
-		return -ENOMEM;
-
-	/* class-based driver binding models */
-	if (add_uevent_var(envp, num_envp, &i,
-			   buffer, buffer_size, &length,
-			   "TYPE=%d/%d/%d",
-			   usb_dev->descriptor.bDeviceClass,
-			   usb_dev->descriptor.bDeviceSubClass,
-			   usb_dev->descriptor.bDeviceProtocol))
-		return -ENOMEM;
-
-	if (add_uevent_var(envp, num_envp, &i,
-			   buffer, buffer_size, &length,
-			   "INTERFACE=%d/%d/%d",
-			   alt->desc.bInterfaceClass,
-			   alt->desc.bInterfaceSubClass,
-			   alt->desc.bInterfaceProtocol))
-		return -ENOMEM;
-
-	if (add_uevent_var(envp, num_envp, &i,
-			   buffer, buffer_size, &length,
-			   "MODALIAS=usb:v%04Xp%04Xd%04Xdc%02Xdsc%02Xdp%02Xic%02Xisc%02Xip%02X",
-			   le16_to_cpu(usb_dev->descriptor.idVendor),
-			   le16_to_cpu(usb_dev->descriptor.idProduct),
-			   le16_to_cpu(usb_dev->descriptor.bcdDevice),
-			   usb_dev->descriptor.bDeviceClass,
-			   usb_dev->descriptor.bDeviceSubClass,
-			   usb_dev->descriptor.bDeviceProtocol,
-			   alt->desc.bInterfaceClass,
-			   alt->desc.bInterfaceSubClass,
-			   alt->desc.bInterfaceProtocol))
-		return -ENOMEM;
-
-	envp[i] = NULL;
-
-	return 0;
-}
-
-#else
-
-static int usb_uevent(struct device *dev, char **envp,
-			int num_envp, char *buffer, int buffer_size)
-{
-	return -ENODEV;
-}
-
-#endif	/* CONFIG_HOTPLUG */
-
 /**
  * usb_release_dev - free a usb device structure when all users of it are finished.
  * @dev: device that's been disconnected
@@ -990,116 +795,6 @@ void usb_buffer_unmap_sg (struct usb_device *dev, unsigned pipe,
 			usb_pipein (pipe) ? DMA_FROM_DEVICE : DMA_TO_DEVICE);
 }
 
-static int verify_suspended(struct device *dev, void *unused)
-{
-	if (dev->driver == NULL)
-		return 0;
-	return (dev->power.power_state.event == PM_EVENT_ON) ? -EBUSY : 0;
-}
-
-static int usb_generic_suspend(struct device *dev, pm_message_t message)
-{
-	struct usb_interface	*intf;
-	struct usb_driver	*driver;
-	int			status;
-
-	/* USB devices enter SUSPEND state through their hubs, but can be
-	 * marked for FREEZE as soon as their children are already idled.
-	 * But those semantics are useless, so we equate the two (sigh).
-	 */
-	if (dev->driver == &usb_generic_driver) {
-		if (dev->power.power_state.event == message.event)
-			return 0;
-		/* we need to rule out bogus requests through sysfs */
-		status = device_for_each_child(dev, NULL, verify_suspended);
-		if (status)
-			return status;
- 		return usb_port_suspend(to_usb_device(dev));
-	}
-
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data))
-		return 0;
-
-	intf = to_usb_interface(dev);
-	driver = to_usb_driver(dev->driver);
-
-	/* with no hardware, USB interfaces only use FREEZE and ON states */
-	if (!is_active(intf))
-		return 0;
-
-	if (driver->suspend && driver->resume) {
-		status = driver->suspend(intf, message);
-		if (status)
-			dev_err(dev, "%s error %d\n", "suspend", status);
-		else
-			mark_quiesced(intf);
-	} else {
-		// FIXME else if there's no suspend method, disconnect...
-		dev_warn(dev, "no suspend for driver %s?\n", driver->name);
-		mark_quiesced(intf);
-		status = 0;
-	}
-	return status;
-}
-
-static int usb_generic_resume(struct device *dev)
-{
-	struct usb_interface	*intf;
-	struct usb_driver	*driver;
-	struct usb_device	*udev;
-	int			status;
-
-	if (dev->power.power_state.event == PM_EVENT_ON)
-		return 0;
-
-	/* mark things as "on" immediately, no matter what errors crop up */
-	dev->power.power_state.event = PM_EVENT_ON;
-
-	/* devices resume through their hubs */
-	if (dev->driver == &usb_generic_driver) {
-		udev = to_usb_device(dev);
-		if (udev->state == USB_STATE_NOTATTACHED)
-			return 0;
-		return usb_port_resume(udev);
-	}
-
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data)) {
-		dev->power.power_state.event = PM_EVENT_FREEZE;
-		return 0;
-	}
-
-	intf = to_usb_interface(dev);
-	driver = to_usb_driver(dev->driver);
-
-	udev = interface_to_usbdev(intf);
-	if (udev->state == USB_STATE_NOTATTACHED)
-		return 0;
-
-	/* if driver was suspended, it has a resume method;
-	 * however, sysfs can wrongly mark things as suspended
-	 * (on the "no suspend method" FIXME path above)
-	 */
-	if (driver->resume) {
-		status = driver->resume(intf);
-		if (status) {
-			dev_err(dev, "%s error %d\n", "resume", status);
-			mark_quiesced(intf);
-		}
-	} else
-		dev_warn(dev, "no resume for driver %s?\n", driver->name);
-	return 0;
-}
-
-struct bus_type usb_bus_type = {
-	.name =		"usb",
-	.match =	usb_device_match,
-	.uevent =	usb_uevent,
-	.suspend =	usb_generic_suspend,
-	.resume =	usb_generic_resume,
-};
-
 /* format to disable USB on kernel command line is: nousb */
 __module_param_call("", nousb, param_set_bool, param_get_bool, &nousb, 0444);
 
@@ -1203,8 +898,6 @@ EXPORT_SYMBOL(usb_hub_tt_clear_buffer);
 
 EXPORT_SYMBOL(usb_lock_device_for_reset);
 
-EXPORT_SYMBOL(usb_driver_claim_interface);
-EXPORT_SYMBOL(usb_driver_release_interface);
 EXPORT_SYMBOL(usb_find_interface);
 EXPORT_SYMBOL(usb_ifnum_to_if);
 EXPORT_SYMBOL(usb_altnum_to_altsetting);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 5a4eff5eec86..82d397a6f773 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -33,9 +33,9 @@ extern void usb_host_cleanup(void);
 extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
+extern struct bus_type usb_bus_type;
 extern struct device_driver usb_generic_driver;
 extern int usb_generic_driver_data;
-extern int usb_device_match(struct device *dev, struct device_driver *drv);
 
 /* Interfaces and their "power state" are owned by usbcore */
 
-- 
cgit v1.2.1


From 8bb54ab573ecd1b4fe2ed66416a8d99a86e65316 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:08:49 -0400
Subject: usbcore: add usb_device_driver definition

This patch (as732) adds a usb_device_driver structure, for representing
drivers that manage an entire USB device as opposed to just an
interface.  Support routines like usb_register_device_driver,
usb_deregister_device_driver, usb_probe_device, and usb_unbind_device
are also added.

Unlike an earlier version of this patch, the new code is type-safe.  To
accomplish this, the existing struct driver embedded in struct
usb_driver had to be wrapped in an intermediate wrapper.  This enables
the core to tell at runtime whether a particular struct driver belongs
to a device driver or to an interface driver.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  | 219 +++++++++++++++++++++++++++++++++------------
 drivers/usb/core/generic.c |  17 +---
 drivers/usb/core/usb.c     |  15 ++--
 drivers/usb/core/usb.h     |  20 ++++-
 include/linux/usb.h        |  58 ++++++++++--
 5 files changed, 245 insertions(+), 84 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 8dcf2cd0c569..0d4b5dcee3ab 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -84,7 +84,7 @@ static int usb_create_newid_file(struct usb_driver *usb_drv)
 		goto exit;
 
 	if (usb_drv->probe != NULL)
-		error = sysfs_create_file(&usb_drv->driver.kobj,
+		error = sysfs_create_file(&usb_drv->drvwrap.driver.kobj,
 					  &driver_attr_new_id.attr);
 exit:
 	return error;
@@ -96,7 +96,7 @@ static void usb_remove_newid_file(struct usb_driver *usb_drv)
 		return;
 
 	if (usb_drv->probe != NULL)
-		sysfs_remove_file(&usb_drv->driver.kobj,
+		sysfs_remove_file(&usb_drv->drvwrap.driver.kobj,
 				  &driver_attr_new_id.attr);
 }
 
@@ -143,18 +143,55 @@ static const struct usb_device_id *usb_match_dynamic_id(struct usb_interface *in
 }
 
 
-/* called from driver core with usb_bus_type.subsys writelock */
+/* called from driver core with dev locked */
+static int usb_probe_device(struct device *dev)
+{
+	struct usb_device_driver *udriver = to_usb_device_driver(dev->driver);
+	struct usb_device *udev;
+	int error = -ENODEV;
+
+	dev_dbg(dev, "%s\n", __FUNCTION__);
+
+	if (!is_usb_device(dev))	/* Sanity check */
+		return error;
+
+	udev = to_usb_device(dev);
+
+	/* FIXME: resume a suspended device */
+	if (udev->state == USB_STATE_SUSPENDED)
+		return -EHOSTUNREACH;
+
+	/* TODO: Add real matching code */
+
+	error = udriver->probe(udev);
+	return error;
+}
+
+/* called from driver core with dev locked */
+static int usb_unbind_device(struct device *dev)
+{
+	struct usb_device_driver *udriver = to_usb_device_driver(dev->driver);
+
+	udriver->disconnect(to_usb_device(dev));
+	return 0;
+}
+
+
+/* called from driver core with dev locked */
 static int usb_probe_interface(struct device *dev)
 {
-	struct usb_interface * intf = to_usb_interface(dev);
-	struct usb_driver * driver = to_usb_driver(dev->driver);
+	struct usb_driver *driver = to_usb_driver(dev->driver);
+	struct usb_interface *intf;
 	const struct usb_device_id *id;
 	int error = -ENODEV;
 
 	dev_dbg(dev, "%s\n", __FUNCTION__);
 
-	if (!driver->probe)
+	if (is_usb_device(dev))		/* Sanity check */
 		return error;
+
+	intf = to_usb_interface(dev);
+
 	/* FIXME we'd much prefer to just resume it ... */
 	if (interface_to_usbdev(intf)->state == USB_STATE_SUSPENDED)
 		return -EHOSTUNREACH;
@@ -182,19 +219,18 @@ static int usb_probe_interface(struct device *dev)
 	return error;
 }
 
-/* called from driver core with usb_bus_type.subsys writelock */
+/* called from driver core with dev locked */
 static int usb_unbind_interface(struct device *dev)
 {
+	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf = to_usb_interface(dev);
-	struct usb_driver *driver = to_usb_driver(intf->dev.driver);
 
 	intf->condition = USB_INTERFACE_UNBINDING;
 
 	/* release all urbs for this interface */
 	usb_disable_interface(interface_to_usbdev(intf), intf);
 
-	if (driver && driver->disconnect)
-		driver->disconnect(intf);
+	driver->disconnect(intf);
 
 	/* reset other interface state */
 	usb_set_interface(interface_to_usbdev(intf),
@@ -235,7 +271,7 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 	if (dev->driver)
 		return -EBUSY;
 
-	dev->driver = &driver->driver;
+	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
 	iface->condition = USB_INTERFACE_BOUND;
 	mark_active(iface);
@@ -270,7 +306,7 @@ void usb_driver_release_interface(struct usb_driver *driver,
 	struct device *dev = &iface->dev;
 
 	/* this should never happen, don't release something that's not ours */
-	if (!dev->driver || dev->driver != &driver->driver)
+	if (!dev->driver || dev->driver != &driver->drvwrap.driver)
 		return;
 
 	/* don't release from within disconnect() */
@@ -433,24 +469,37 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_match_id);
 
 int usb_device_match(struct device *dev, struct device_driver *drv)
 {
-	struct usb_interface *intf;
-	struct usb_driver *usb_drv;
-	const struct usb_device_id *id;
-
-	/* check for generic driver, which we don't match any device with */
-	if (drv == &usb_generic_driver)
-		return 0;
+	/* devices and interfaces are handled separately */
+	if (is_usb_device(dev)) {
 
-	intf = to_usb_interface(dev);
-	usb_drv = to_usb_driver(drv);
+		/* interface drivers never match devices */
+		if (!is_usb_device_driver(drv))
+			return 0;
 
-	id = usb_match_id(intf, usb_drv->id_table);
-	if (id)
+		/* TODO: Add real matching code */
 		return 1;
 
-	id = usb_match_dynamic_id(intf, usb_drv);
-	if (id)
-		return 1;
+	} else {
+		struct usb_interface *intf;
+		struct usb_driver *usb_drv;
+		const struct usb_device_id *id;
+
+		/* device drivers never match interfaces */
+		if (is_usb_device_driver(drv))
+			return 0;
+
+		intf = to_usb_interface(dev);
+		usb_drv = to_usb_driver(drv);
+
+		id = usb_match_id(intf, usb_drv->id_table);
+		if (id)
+			return 1;
+
+		id = usb_match_dynamic_id(intf, usb_drv);
+		if (id)
+			return 1;
+	}
+
 	return 0;
 }
 
@@ -481,14 +530,13 @@ static int usb_uevent(struct device *dev, char **envp, int num_envp,
 	/* driver is often null here; dev_dbg() would oops */
 	pr_debug ("usb %s: uevent\n", dev->bus_id);
 
-	/* Must check driver_data here, as on remove driver is always NULL */
-	if ((dev->driver == &usb_generic_driver) ||
-	    (dev->driver_data == &usb_generic_driver_data))
+	if (is_usb_device(dev))
 		return 0;
-
-	intf = to_usb_interface(dev);
-	usb_dev = interface_to_usbdev (intf);
-	alt = intf->cur_altsetting;
+	else {
+		intf = to_usb_interface(dev);
+		usb_dev = interface_to_usbdev(intf);
+		alt = intf->cur_altsetting;
+	}
 
 	if (usb_dev->devnum < 0) {
 		pr_debug ("usb %s: already deleted?\n", dev->bus_id);
@@ -569,13 +617,71 @@ static int usb_uevent(struct device *dev, char **envp,
 #endif	/* CONFIG_HOTPLUG */
 
 /**
- * usb_register_driver - register a USB driver
- * @new_driver: USB operations for the driver
+ * usb_register_device_driver - register a USB device (not interface) driver
+ * @new_udriver: USB operations for the device driver
  * @owner: module owner of this driver.
  *
- * Registers a USB driver with the USB core.  The list of unattached
- * interfaces will be rescanned whenever a new driver is added, allowing
- * the new driver to attach to any recognized devices.
+ * Registers a USB device driver with the USB core.  The list of
+ * unattached devices will be rescanned whenever a new driver is
+ * added, allowing the new driver to attach to any recognized devices.
+ * Returns a negative error code on failure and 0 on success.
+ */
+int usb_register_device_driver(struct usb_device_driver *new_udriver,
+		struct module *owner)
+{
+	int retval = 0;
+
+	if (usb_disabled())
+		return -ENODEV;
+
+	new_udriver->drvwrap.for_devices = 1;
+	new_udriver->drvwrap.driver.name = (char *) new_udriver->name;
+	new_udriver->drvwrap.driver.bus = &usb_bus_type;
+	new_udriver->drvwrap.driver.probe = usb_probe_device;
+	new_udriver->drvwrap.driver.remove = usb_unbind_device;
+	new_udriver->drvwrap.driver.owner = owner;
+
+	retval = driver_register(&new_udriver->drvwrap.driver);
+
+	if (!retval) {
+		pr_info("%s: registered new device driver %s\n",
+			usbcore_name, new_udriver->name);
+		usbfs_update_special();
+	} else {
+		printk(KERN_ERR "%s: error %d registering device "
+			"	driver %s\n",
+			usbcore_name, retval, new_udriver->name);
+	}
+
+	return retval;
+}
+EXPORT_SYMBOL_GPL(usb_register_device_driver);
+
+/**
+ * usb_deregister_device_driver - unregister a USB device (not interface) driver
+ * @udriver: USB operations of the device driver to unregister
+ * Context: must be able to sleep
+ *
+ * Unlinks the specified driver from the internal USB driver list.
+ */
+void usb_deregister_device_driver(struct usb_device_driver *udriver)
+{
+	pr_info("%s: deregistering device driver %s\n",
+			usbcore_name, udriver->name);
+
+	driver_unregister(&udriver->drvwrap.driver);
+	usbfs_update_special();
+}
+EXPORT_SYMBOL_GPL(usb_deregister_device_driver);
+
+/**
+ * usb_register_driver - register a USB interface driver
+ * @new_driver: USB operations for the interface driver
+ * @owner: module owner of this driver.
+ *
+ * Registers a USB interface driver with the USB core.  The list of
+ * unattached interfaces will be rescanned whenever a new driver is
+ * added, allowing the new driver to attach to any recognized interfaces.
  * Returns a negative error code on failure and 0 on success.
  *
  * NOTE: if you want your driver to use the USB major number, you must call
@@ -589,23 +695,25 @@ int usb_register_driver(struct usb_driver *new_driver, struct module *owner)
 	if (usb_disabled())
 		return -ENODEV;
 
-	new_driver->driver.name = (char *)new_driver->name;
-	new_driver->driver.bus = &usb_bus_type;
-	new_driver->driver.probe = usb_probe_interface;
-	new_driver->driver.remove = usb_unbind_interface;
-	new_driver->driver.owner = owner;
+	new_driver->drvwrap.for_devices = 0;
+	new_driver->drvwrap.driver.name = (char *) new_driver->name;
+	new_driver->drvwrap.driver.bus = &usb_bus_type;
+	new_driver->drvwrap.driver.probe = usb_probe_interface;
+	new_driver->drvwrap.driver.remove = usb_unbind_interface;
+	new_driver->drvwrap.driver.owner = owner;
 	spin_lock_init(&new_driver->dynids.lock);
 	INIT_LIST_HEAD(&new_driver->dynids.list);
 
-	retval = driver_register(&new_driver->driver);
+	retval = driver_register(&new_driver->drvwrap.driver);
 
 	if (!retval) {
-		pr_info("%s: registered new driver %s\n",
+		pr_info("%s: registered new interface driver %s\n",
 			usbcore_name, new_driver->name);
 		usbfs_update_special();
 		usb_create_newid_file(new_driver);
 	} else {
-		printk(KERN_ERR "%s: error %d registering driver %s\n",
+		printk(KERN_ERR "%s: error %d registering interface "
+			"	driver %s\n",
 			usbcore_name, retval, new_driver->name);
 	}
 
@@ -614,8 +722,8 @@ int usb_register_driver(struct usb_driver *new_driver, struct module *owner)
 EXPORT_SYMBOL_GPL_FUTURE(usb_register_driver);
 
 /**
- * usb_deregister - unregister a USB driver
- * @driver: USB operations of the driver to unregister
+ * usb_deregister - unregister a USB interface driver
+ * @driver: USB operations of the interface driver to unregister
  * Context: must be able to sleep
  *
  * Unlinks the specified driver from the internal USB driver list.
@@ -626,11 +734,12 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_register_driver);
  */
 void usb_deregister(struct usb_driver *driver)
 {
-	pr_info("%s: deregistering driver %s\n", usbcore_name, driver->name);
+	pr_info("%s: deregistering interface driver %s\n",
+			usbcore_name, driver->name);
 
 	usb_remove_newid_file(driver);
 	usb_free_dynids(driver);
-	driver_unregister(&driver->driver);
+	driver_unregister(&driver->drvwrap.driver);
 
 	usbfs_update_special();
 }
@@ -655,7 +764,7 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
 	 * marked for FREEZE as soon as their children are already idled.
 	 * But those semantics are useless, so we equate the two (sigh).
 	 */
-	if (dev->driver == &usb_generic_driver) {
+	if (is_usb_device(dev)) {
 		if (dev->power.power_state.event == message.event)
 			return 0;
 		/* we need to rule out bogus requests through sysfs */
@@ -665,8 +774,7 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
  		return usb_port_suspend(to_usb_device(dev));
 	}
 
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data))
+	if (dev->driver == NULL)
 		return 0;
 
 	intf = to_usb_interface(dev);
@@ -705,15 +813,14 @@ static int usb_generic_resume(struct device *dev)
 	dev->power.power_state.event = PM_EVENT_ON;
 
 	/* devices resume through their hubs */
-	if (dev->driver == &usb_generic_driver) {
+	if (is_usb_device(dev)) {
 		udev = to_usb_device(dev);
 		if (udev->state == USB_STATE_NOTATTACHED)
 			return 0;
 		return usb_port_resume(udev);
 	}
 
-	if ((dev->driver == NULL) ||
-	    (dev->driver_data == &usb_generic_driver_data)) {
+	if (dev->driver == NULL) {
 		dev->power.power_state.event = PM_EVENT_FREEZE;
 		return 0;
 	}
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index 7bab9769b34f..fa6f34a12b4b 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -21,14 +21,12 @@
 #include <linux/usb.h>
 #include "usb.h"
 
-static int generic_probe(struct device *dev)
+static int generic_probe(struct usb_device *udev)
 {
 	return 0;
 }
-static int generic_remove(struct device *dev)
+static void generic_disconnect(struct usb_device *udev)
 {
-	struct usb_device *udev = to_usb_device(dev);
-
 	/* if this is only an unbind, not a physical disconnect, then
 	 * unconfigure the device */
 	if (udev->state == USB_STATE_CONFIGURED)
@@ -37,17 +35,10 @@ static int generic_remove(struct device *dev)
 	/* in case the call failed or the device was suspended */
 	if (udev->state >= USB_STATE_CONFIGURED)
 		usb_disable_device(udev, 0);
-	return 0;
 }
 
-struct device_driver usb_generic_driver = {
-	.owner = THIS_MODULE,
+struct usb_device_driver usb_generic_driver = {
 	.name =	"usb",
-	.bus = &usb_bus_type,
 	.probe = generic_probe,
-	.remove = generic_remove,
+	.disconnect = generic_disconnect,
 };
-
-/* Fun hack to determine if the struct device is a
- * usb device or a usb interface. */
-int usb_generic_driver_data;
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 0b8c67bcde60..6dfbc284369b 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -123,7 +123,7 @@ static int __find_interface(struct device * dev, void * data)
 	struct usb_interface *intf;
 
 	/* can't look at usb devices, only interfaces */
-	if (dev->driver == &usb_generic_driver)
+	if (is_usb_device(dev))
 		return 0;
 
 	intf = to_usb_interface(dev);
@@ -149,7 +149,8 @@ struct usb_interface *usb_find_interface(struct usb_driver *drv, int minor)
 
 	argb.minor = minor;
 	argb.interface = NULL;
-	driver_for_each_device(&drv->driver, NULL, &argb, __find_interface);
+	driver_for_each_device(&drv->drvwrap.driver, NULL, &argb,
+			__find_interface);
 	return argb.interface;
 }
 
@@ -204,11 +205,13 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	device_initialize(&dev->dev);
 	dev->dev.bus = &usb_bus_type;
 	dev->dev.dma_mask = bus->controller->dma_mask;
-	dev->dev.driver_data = &usb_generic_driver_data;
-	dev->dev.driver = &usb_generic_driver;
+	dev->dev.driver = &usb_generic_driver.drvwrap.driver;
 	dev->dev.release = usb_release_dev;
 	dev->state = USB_STATE_ATTACHED;
 
+	/* This magic assignment distinguishes devices from interfaces */
+	dev->dev.platform_data = &usb_generic_driver;
+
 	INIT_LIST_HEAD(&dev->ep0.urb_list);
 	dev->ep0.desc.bLength = USB_DT_ENDPOINT_SIZE;
 	dev->ep0.desc.bDescriptorType = USB_DT_ENDPOINT;
@@ -838,7 +841,7 @@ static int __init usb_init(void)
 	retval = usb_hub_init();
 	if (retval)
 		goto hub_init_failed;
-	retval = driver_register(&usb_generic_driver);
+	retval = usb_register_device_driver(&usb_generic_driver, THIS_MODULE);
 	if (!retval)
 		goto out;
 
@@ -868,7 +871,7 @@ static void __exit usb_exit(void)
 	if (nousb)
 		return;
 
-	driver_unregister(&usb_generic_driver);
+	usb_deregister_device_driver(&usb_generic_driver);
 	usb_major_cleanup();
 	usbfs_cleanup();
 	usb_deregister(&usbfs_driver);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 82d397a6f773..1d25ccac7832 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -34,8 +34,24 @@ extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
 extern struct bus_type usb_bus_type;
-extern struct device_driver usb_generic_driver;
-extern int usb_generic_driver_data;
+extern struct usb_device_driver usb_generic_driver;
+
+/* Here's how we tell apart devices and interfaces.  Luckily there's
+ * no such thing as a platform USB device, so we can steal the use
+ * of the platform_data field. */
+
+static inline int is_usb_device(struct device *dev)
+{
+	return dev->platform_data == &usb_generic_driver;
+}
+
+/* Do the same for device drivers and interface drivers. */
+
+static inline int is_usb_device_driver(struct device_driver *drv)
+{
+	return container_of(drv, struct usbdrv_wrap, driver)->
+			for_devices;
+}
 
 /* Interfaces and their "power state" are owned by usbcore */
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index d2bd0c8e0154..b4ccce6d0982 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -540,7 +540,17 @@ struct usb_dynids {
 };
 
 /**
- * struct usb_driver - identifies USB driver to usbcore
+ * struct usbdrv_wrap - wrapper for driver-model structure
+ * @driver: The driver-model core driver structure.
+ * @for_devices: Non-zero for device drivers, 0 for interface drivers.
+ */
+struct usbdrv_wrap {
+	struct device_driver driver;
+	int for_devices;
+};
+
+/**
+ * struct usb_driver - identifies USB interface driver to usbcore
  * @name: The driver name should be unique among USB drivers,
  *	and should normally be the same as the module name.
  * @probe: Called to see if the driver is willing to manage a particular
@@ -567,12 +577,12 @@ struct usb_dynids {
  *	or your driver's probe function will never get called.
  * @dynids: used internally to hold the list of dynamically added device
  *	ids for this driver.
- * @driver: the driver model core driver structure.
+ * @drvwrap: Driver-model core structure wrapper.
  * @no_dynamic_id: if set to 1, the USB core will not allow dynamic ids to be
  *	added to this driver by preventing the sysfs file from being created.
  *
- * USB drivers must provide a name, probe() and disconnect() methods,
- * and an id_table.  Other driver fields are optional.
+ * USB interface drivers must provide a name, probe() and disconnect()
+ * methods, and an id_table.  Other driver fields are optional.
  *
  * The id_table is used in hotplugging.  It holds a set of descriptors,
  * and specialized data may be associated with each entry.  That table
@@ -606,10 +616,40 @@ struct usb_driver {
 	const struct usb_device_id *id_table;
 
 	struct usb_dynids dynids;
-	struct device_driver driver;
+	struct usbdrv_wrap drvwrap;
 	unsigned int no_dynamic_id:1;
 };
-#define	to_usb_driver(d) container_of(d, struct usb_driver, driver)
+#define	to_usb_driver(d) container_of(d, struct usb_driver, drvwrap.driver)
+
+/**
+ * struct usb_device_driver - identifies USB device driver to usbcore
+ * @name: The driver name should be unique among USB drivers,
+ *	and should normally be the same as the module name.
+ * @probe: Called to see if the driver is willing to manage a particular
+ *	device.  If it is, probe returns zero and uses dev_set_drvdata()
+ *	to associate driver-specific data with the device.  If unwilling
+ *	to manage the device, return a negative errno value.
+ * @disconnect: Called when the device is no longer accessible, usually
+ *	because it has been (or is being) disconnected or the driver's
+ *	module is being unloaded.
+ * @suspend: Called when the device is going to be suspended by the system.
+ * @resume: Called when the device is being resumed by the system.
+ * @drvwrap: Driver-model core structure wrapper.
+ *
+ * USB drivers must provide all the fields listed above except drvwrap.
+ */
+struct usb_device_driver {
+	const char *name;
+
+	int (*probe) (struct usb_device *udev);
+	void (*disconnect) (struct usb_device *udev);
+
+	int (*suspend) (struct usb_device *udev, pm_message_t message);
+	int (*resume) (struct usb_device *udev);
+	struct usbdrv_wrap drvwrap;
+};
+#define	to_usb_device_driver(d) container_of(d, struct usb_device_driver, \
+		drvwrap.driver)
 
 extern struct bus_type usb_bus_type;
 
@@ -633,13 +673,17 @@ struct usb_class_driver {
  * use these in module_init()/module_exit()
  * and don't forget MODULE_DEVICE_TABLE(usb, ...)
  */
-int usb_register_driver(struct usb_driver *, struct module *);
+extern int usb_register_driver(struct usb_driver *, struct module *);
 static inline int usb_register(struct usb_driver *driver)
 {
 	return usb_register_driver(driver, THIS_MODULE);
 }
 extern void usb_deregister(struct usb_driver *);
 
+extern int usb_register_device_driver(struct usb_device_driver *,
+			struct module *);
+extern void usb_deregister_device_driver(struct usb_device_driver *);
+
 extern int usb_register_dev(struct usb_interface *intf,
 			    struct usb_class_driver *class_driver);
 extern void usb_deregister_dev(struct usb_interface *intf,
-- 
cgit v1.2.1


From 782da727b0d59e93c84a627948b1535a3db90392 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:09:35 -0400
Subject: usbcore: make usb_generic a usb_device_driver

This patch (as714b) makes usb_generic into a usb_device_driver capable
of being probed and unbound, just like other drivers.  A fair amount of
the work that used to get done during discovery or removal of a USB
device have been moved to the probe and disconnect methods of
usb_generic: creating the sysfs attributes and selecting an initial
configuration.  However the normal behavior should continue to be the
same as before.

We will now have the possibility of creating other USB device drivers,
They will assist with exporting devices to remote systems
(USB-over-TCPIP) or to paravirtual guest operating systems.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  |  69 +++++++++--------
 drivers/usb/core/generic.c | 184 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/usb/core/hub.c     | 164 ++++------------------------------------
 drivers/usb/core/usb.c     |   1 -
 4 files changed, 232 insertions(+), 186 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 0d4b5dcee3ab..a62de0a85406 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -530,9 +530,10 @@ static int usb_uevent(struct device *dev, char **envp, int num_envp,
 	/* driver is often null here; dev_dbg() would oops */
 	pr_debug ("usb %s: uevent\n", dev->bus_id);
 
-	if (is_usb_device(dev))
-		return 0;
-	else {
+	if (is_usb_device(dev)) {
+		usb_dev = to_usb_device(dev);
+		alt = NULL;
+	} else {
 		intf = to_usb_interface(dev);
 		usb_dev = interface_to_usbdev(intf);
 		alt = intf->cur_altsetting;
@@ -579,15 +580,17 @@ static int usb_uevent(struct device *dev, char **envp, int num_envp,
 			   usb_dev->descriptor.bDeviceProtocol))
 		return -ENOMEM;
 
-	if (add_uevent_var(envp, num_envp, &i,
+	if (!is_usb_device(dev)) {
+
+		if (add_uevent_var(envp, num_envp, &i,
 			   buffer, buffer_size, &length,
 			   "INTERFACE=%d/%d/%d",
 			   alt->desc.bInterfaceClass,
 			   alt->desc.bInterfaceSubClass,
 			   alt->desc.bInterfaceProtocol))
-		return -ENOMEM;
+			return -ENOMEM;
 
-	if (add_uevent_var(envp, num_envp, &i,
+		if (add_uevent_var(envp, num_envp, &i,
 			   buffer, buffer_size, &length,
 			   "MODALIAS=usb:v%04Xp%04Xd%04Xdc%02Xdsc%02Xdp%02Xic%02Xisc%02Xip%02X",
 			   le16_to_cpu(usb_dev->descriptor.idVendor),
@@ -599,7 +602,8 @@ static int usb_uevent(struct device *dev, char **envp, int num_envp,
 			   alt->desc.bInterfaceClass,
 			   alt->desc.bInterfaceSubClass,
 			   alt->desc.bInterfaceProtocol))
-		return -ENOMEM;
+			return -ENOMEM;
+	}
 
 	envp[i] = NULL;
 
@@ -747,31 +751,22 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 
 #ifdef CONFIG_PM
 
-static int verify_suspended(struct device *dev, void *unused)
+static int usb_suspend(struct device *dev, pm_message_t message)
 {
-	if (dev->driver == NULL)
-		return 0;
-	return (dev->power.power_state.event == PM_EVENT_ON) ? -EBUSY : 0;
-}
+	struct usb_device		*udev;
+	struct usb_device_driver	*udriver;
+	struct usb_interface		*intf;
+	struct usb_driver		*driver;
+	int				status;
 
-static int usb_generic_suspend(struct device *dev, pm_message_t message)
-{
-	struct usb_interface	*intf;
-	struct usb_driver	*driver;
-	int			status;
-
-	/* USB devices enter SUSPEND state through their hubs, but can be
-	 * marked for FREEZE as soon as their children are already idled.
-	 * But those semantics are useless, so we equate the two (sigh).
-	 */
 	if (is_usb_device(dev)) {
+		if (dev->driver == NULL)
+			return 0;
+		udev = to_usb_device(dev);
+		udriver = to_usb_device_driver(dev->driver);
 		if (dev->power.power_state.event == message.event)
 			return 0;
-		/* we need to rule out bogus requests through sysfs */
-		status = device_for_each_child(dev, NULL, verify_suspended);
-		if (status)
-			return status;
- 		return usb_port_suspend(to_usb_device(dev));
+		return udriver->suspend(udev, message);
 	}
 
 	if (dev->driver == NULL)
@@ -799,12 +794,13 @@ static int usb_generic_suspend(struct device *dev, pm_message_t message)
 	return status;
 }
 
-static int usb_generic_resume(struct device *dev)
+static int usb_resume(struct device *dev)
 {
-	struct usb_interface	*intf;
-	struct usb_driver	*driver;
-	struct usb_device	*udev;
-	int			status;
+	struct usb_device		*udev;
+	struct usb_device_driver	*udriver;
+	struct usb_interface		*intf;
+	struct usb_driver		*driver;
+	int				status;
 
 	if (dev->power.power_state.event == PM_EVENT_ON)
 		return 0;
@@ -814,10 +810,13 @@ static int usb_generic_resume(struct device *dev)
 
 	/* devices resume through their hubs */
 	if (is_usb_device(dev)) {
+		if (dev->driver == NULL)
+			return 0;
 		udev = to_usb_device(dev);
+		udriver = to_usb_device_driver(dev->driver);
 		if (udev->state == USB_STATE_NOTATTACHED)
 			return 0;
-		return usb_port_resume(udev);
+		return udriver->resume(udev);
 	}
 
 	if (dev->driver == NULL) {
@@ -854,7 +853,7 @@ struct bus_type usb_bus_type = {
 	.match =	usb_device_match,
 	.uevent =	usb_uevent,
 #ifdef CONFIG_PM
-	.suspend =	usb_generic_suspend,
-	.resume =	usb_generic_resume,
+	.suspend =	usb_suspend,
+	.resume =	usb_resume,
 #endif
 };
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index fa6f34a12b4b..1522195de715 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -21,24 +21,208 @@
 #include <linux/usb.h>
 #include "usb.h"
 
+static inline const char *plural(int n)
+{
+	return (n == 1 ? "" : "s");
+}
+
+static int choose_configuration(struct usb_device *udev)
+{
+	int i;
+	int num_configs;
+	int insufficient_power = 0;
+	struct usb_host_config *c, *best;
+
+	best = NULL;
+	c = udev->config;
+	num_configs = udev->descriptor.bNumConfigurations;
+	for (i = 0; i < num_configs; (i++, c++)) {
+		struct usb_interface_descriptor	*desc = NULL;
+
+		/* It's possible that a config has no interfaces! */
+		if (c->desc.bNumInterfaces > 0)
+			desc = &c->intf_cache[0]->altsetting->desc;
+
+		/*
+		 * HP's USB bus-powered keyboard has only one configuration
+		 * and it claims to be self-powered; other devices may have
+		 * similar errors in their descriptors.  If the next test
+		 * were allowed to execute, such configurations would always
+		 * be rejected and the devices would not work as expected.
+		 * In the meantime, we run the risk of selecting a config
+		 * that requires external power at a time when that power
+		 * isn't available.  It seems to be the lesser of two evils.
+		 *
+		 * Bugzilla #6448 reports a device that appears to crash
+		 * when it receives a GET_DEVICE_STATUS request!  We don't
+		 * have any other way to tell whether a device is self-powered,
+		 * but since we don't use that information anywhere but here,
+		 * the call has been removed.
+		 *
+		 * Maybe the GET_DEVICE_STATUS call and the test below can
+		 * be reinstated when device firmwares become more reliable.
+		 * Don't hold your breath.
+		 */
+#if 0
+		/* Rule out self-powered configs for a bus-powered device */
+		if (bus_powered && (c->desc.bmAttributes &
+					USB_CONFIG_ATT_SELFPOWER))
+			continue;
+#endif
+
+		/*
+		 * The next test may not be as effective as it should be.
+		 * Some hubs have errors in their descriptor, claiming
+		 * to be self-powered when they are really bus-powered.
+		 * We will overestimate the amount of current such hubs
+		 * make available for each port.
+		 *
+		 * This is a fairly benign sort of failure.  It won't
+		 * cause us to reject configurations that we should have
+		 * accepted.
+		 */
+
+		/* Rule out configs that draw too much bus current */
+		if (c->desc.bMaxPower * 2 > udev->bus_mA) {
+			insufficient_power++;
+			continue;
+		}
+
+		/* If the first config's first interface is COMM/2/0xff
+		 * (MSFT RNDIS), rule it out unless Linux has host-side
+		 * RNDIS support. */
+		if (i == 0 && desc
+				&& desc->bInterfaceClass == USB_CLASS_COMM
+				&& desc->bInterfaceSubClass == 2
+				&& desc->bInterfaceProtocol == 0xff) {
+#ifndef CONFIG_USB_NET_RNDIS_HOST
+			continue;
+#else
+			best = c;
+#endif
+		}
+
+		/* From the remaining configs, choose the first one whose
+		 * first interface is for a non-vendor-specific class.
+		 * Reason: Linux is more likely to have a class driver
+		 * than a vendor-specific driver. */
+		else if (udev->descriptor.bDeviceClass !=
+						USB_CLASS_VENDOR_SPEC &&
+				(!desc || desc->bInterfaceClass !=
+						USB_CLASS_VENDOR_SPEC)) {
+			best = c;
+			break;
+		}
+
+		/* If all the remaining configs are vendor-specific,
+		 * choose the first one. */
+		else if (!best)
+			best = c;
+	}
+
+	if (insufficient_power > 0)
+		dev_info(&udev->dev, "rejected %d configuration%s "
+			"due to insufficient available bus power\n",
+			insufficient_power, plural(insufficient_power));
+
+	if (best) {
+		i = best->desc.bConfigurationValue;
+		dev_info(&udev->dev,
+			"configuration #%d chosen from %d choice%s\n",
+			i, num_configs, plural(num_configs));
+	} else {
+		i = -1;
+		dev_warn(&udev->dev,
+			"no configuration chosen from %d choice%s\n",
+			num_configs, plural(num_configs));
+	}
+	return i;
+}
+
 static int generic_probe(struct usb_device *udev)
 {
+	int err, c;
+
+	/* put device-specific files into sysfs */
+	usb_create_sysfs_dev_files(udev);
+
+	/* Choose and set the configuration.  This registers the interfaces
+	 * with the driver core and lets interface drivers bind to them.
+	 */
+	c = choose_configuration(udev);
+	if (c >= 0) {
+		err = usb_set_configuration(udev, c);
+		if (err) {
+			dev_err(&udev->dev, "can't set config #%d, error %d\n",
+					c, err);
+			/* This need not be fatal.  The user can try to
+			 * set other configurations. */
+		}
+	}
+
+	/* USB device state == configured ... usable */
+	usb_notify_add_device(udev);
+
 	return 0;
 }
+
 static void generic_disconnect(struct usb_device *udev)
 {
+	usb_notify_remove_device(udev);
+
 	/* if this is only an unbind, not a physical disconnect, then
 	 * unconfigure the device */
 	if (udev->state == USB_STATE_CONFIGURED)
 		usb_set_configuration(udev, 0);
 
+	usb_remove_sysfs_dev_files(udev);
+
 	/* in case the call failed or the device was suspended */
 	if (udev->state >= USB_STATE_CONFIGURED)
 		usb_disable_device(udev, 0);
 }
 
+#ifdef	CONFIG_PM
+
+static int verify_suspended(struct device *dev, void *unused)
+{
+	if (dev->driver == NULL)
+		return 0;
+	return (dev->power.power_state.event == PM_EVENT_ON) ? -EBUSY : 0;
+}
+
+static int generic_suspend(struct usb_device *udev, pm_message_t msg)
+{
+	int	status;
+
+	/* rule out bogus requests through sysfs */
+	status = device_for_each_child(&udev->dev, NULL, verify_suspended);
+	if (status)
+		return status;
+
+	/* USB devices enter SUSPEND state through their hubs, but can be
+	 * marked for FREEZE as soon as their children are already idled.
+	 * But those semantics are useless, so we equate the two (sigh).
+	 */
+	return usb_port_suspend(udev);
+}
+
+static int generic_resume(struct usb_device *udev)
+{
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return 0;
+
+	return usb_port_resume(udev);
+}
+
+#endif	/* CONFIG_PM */
+
 struct usb_device_driver usb_generic_driver = {
 	.name =	"usb",
 	.probe = generic_probe,
 	.disconnect = generic_disconnect,
+#ifdef	CONFIG_PM
+	.suspend = generic_suspend,
+	.resume = generic_resume,
+#endif
 };
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index b00514d9a605..a372332440b2 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1148,144 +1148,28 @@ void usb_disconnect(struct usb_device **pdev)
 	 * cleaning up all state associated with the current configuration
 	 * so that the hardware is now fully quiesced.
 	 */
+	dev_dbg (&udev->dev, "unregistering device\n");
 	usb_disable_device(udev, 0);
 
-	usb_notify_remove_device(udev);
+	usb_unlock_device(udev);
+
+	/* Unregister the device.  The device driver is responsible
+	 * for removing the device files from usbfs and sysfs and for
+	 * de-configuring the device.
+	 */
+	device_del(&udev->dev);
 
-	/* Free the device number, remove the /proc/bus/usb entry and
-	 * the sysfs attributes, and delete the parent's children[]
+	/* Free the device number and delete the parent's children[]
 	 * (or root_hub) pointer.
 	 */
-	dev_dbg (&udev->dev, "unregistering device\n");
 	release_address(udev);
-	usb_remove_sysfs_dev_files(udev);
 
 	/* Avoid races with recursively_mark_NOTATTACHED() */
 	spin_lock_irq(&device_state_lock);
 	*pdev = NULL;
 	spin_unlock_irq(&device_state_lock);
 
-	usb_unlock_device(udev);
-
-	device_unregister(&udev->dev);
-}
-
-static inline const char *plural(int n)
-{
-	return (n == 1 ? "" : "s");
-}
-
-static int choose_configuration(struct usb_device *udev)
-{
-	int i;
-	int num_configs;
-	int insufficient_power = 0;
-	struct usb_host_config *c, *best;
-
-	best = NULL;
-	c = udev->config;
-	num_configs = udev->descriptor.bNumConfigurations;
-	for (i = 0; i < num_configs; (i++, c++)) {
-		struct usb_interface_descriptor	*desc = NULL;
-
-		/* It's possible that a config has no interfaces! */
-		if (c->desc.bNumInterfaces > 0)
-			desc = &c->intf_cache[0]->altsetting->desc;
-
-		/*
-		 * HP's USB bus-powered keyboard has only one configuration
-		 * and it claims to be self-powered; other devices may have
-		 * similar errors in their descriptors.  If the next test
-		 * were allowed to execute, such configurations would always
-		 * be rejected and the devices would not work as expected.
-		 * In the meantime, we run the risk of selecting a config
-		 * that requires external power at a time when that power
-		 * isn't available.  It seems to be the lesser of two evils.
-		 *
-		 * Bugzilla #6448 reports a device that appears to crash
-		 * when it receives a GET_DEVICE_STATUS request!  We don't
-		 * have any other way to tell whether a device is self-powered,
-		 * but since we don't use that information anywhere but here,
-		 * the call has been removed.
-		 *
-		 * Maybe the GET_DEVICE_STATUS call and the test below can
-		 * be reinstated when device firmwares become more reliable.
-		 * Don't hold your breath.
-		 */
-#if 0
-		/* Rule out self-powered configs for a bus-powered device */
-		if (bus_powered && (c->desc.bmAttributes &
-					USB_CONFIG_ATT_SELFPOWER))
-			continue;
-#endif
-
-		/*
-		 * The next test may not be as effective as it should be.
-		 * Some hubs have errors in their descriptor, claiming
-		 * to be self-powered when they are really bus-powered.
-		 * We will overestimate the amount of current such hubs
-		 * make available for each port.
-		 *
-		 * This is a fairly benign sort of failure.  It won't
-		 * cause us to reject configurations that we should have
-		 * accepted.
-		 */
-
-		/* Rule out configs that draw too much bus current */
-		if (c->desc.bMaxPower * 2 > udev->bus_mA) {
-			insufficient_power++;
-			continue;
-		}
-
-		/* If the first config's first interface is COMM/2/0xff
-		 * (MSFT RNDIS), rule it out unless Linux has host-side
-		 * RNDIS support. */
-		if (i == 0 && desc
-				&& desc->bInterfaceClass == USB_CLASS_COMM
-				&& desc->bInterfaceSubClass == 2
-				&& desc->bInterfaceProtocol == 0xff) {
-#ifndef CONFIG_USB_NET_RNDIS_HOST
-			continue;
-#else
-			best = c;
-#endif
-		}
-
-		/* From the remaining configs, choose the first one whose
-		 * first interface is for a non-vendor-specific class.
-		 * Reason: Linux is more likely to have a class driver
-		 * than a vendor-specific driver. */
-		else if (udev->descriptor.bDeviceClass !=
-						USB_CLASS_VENDOR_SPEC &&
-				(!desc || desc->bInterfaceClass !=
-						USB_CLASS_VENDOR_SPEC)) {
-			best = c;
-			break;
-		}
-
-		/* If all the remaining configs are vendor-specific,
-		 * choose the first one. */
-		else if (!best)
-			best = c;
-	}
-
-	if (insufficient_power > 0)
-		dev_info(&udev->dev, "rejected %d configuration%s "
-			"due to insufficient available bus power\n",
-			insufficient_power, plural(insufficient_power));
-
-	if (best) {
-		i = best->desc.bConfigurationValue;
-		dev_info(&udev->dev,
-			"configuration #%d chosen from %d choice%s\n",
-			i, num_configs, plural(num_configs));
-	} else {
-		i = -1;
-		dev_warn(&udev->dev,
-			"no configuration chosen from %d choice%s\n",
-			num_configs, plural(num_configs));
-	}
-	return i;
+	put_device(&udev->dev);
 }
 
 #ifdef DEBUG
@@ -1328,7 +1212,6 @@ static inline void show_string(struct usb_device *udev, char *id, char *string)
 int usb_new_device(struct usb_device *udev)
 {
 	int err;
-	int c;
 
 	err = usb_get_configuration(udev);
 	if (err < 0) {
@@ -1418,34 +1301,15 @@ int usb_new_device(struct usb_device *udev)
 	}
 #endif
 
-	/* put device-specific files into sysfs */
+	/* Register the device.  The device driver is responsible
+	 * for adding the device files to usbfs and sysfs and for
+	 * configuring the device.
+	 */
 	err = device_add (&udev->dev);
 	if (err) {
 		dev_err(&udev->dev, "can't device_add, error %d\n", err);
 		goto fail;
 	}
-	usb_create_sysfs_dev_files (udev);
-
-	usb_lock_device(udev);
-
-	/* choose and set the configuration. that registers the interfaces
-	 * with the driver core, and lets usb device drivers bind to them.
-	 */
-	c = choose_configuration(udev);
-	if (c >= 0) {
-		err = usb_set_configuration(udev, c);
-		if (err) {
-			dev_err(&udev->dev, "can't set config #%d, error %d\n",
-					c, err);
-			/* This need not be fatal.  The user can try to
-			 * set other configurations. */
-		}
-	}
-
-	/* USB device state == configured ... usable */
-	usb_notify_add_device(udev);
-
-	usb_unlock_device(udev);
 
 	return 0;
 
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 6dfbc284369b..9ebfc0fe819d 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -205,7 +205,6 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	device_initialize(&dev->dev);
 	dev->dev.bus = &usb_bus_type;
 	dev->dev.dma_mask = bus->controller->dma_mask;
-	dev->dev.driver = &usb_generic_driver.drvwrap.driver;
 	dev->dev.release = usb_release_dev;
 	dev->state = USB_STATE_ATTACHED;
 
-- 
cgit v1.2.1


From 1cc8a25d5b680ff656927ffa9b66fae6b415b1d3 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:10:15 -0400
Subject: usbcore: split suspend/resume for device and interfaces

This patch (as716b) splits up the core suspend and resume routines into
two parts each: one for handling devices and one for handling
interfaces.  The behavior of the parts should be the same as in the old
unified code.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 122 +++++++++++++++++++++++++++++-----------------
 1 file changed, 77 insertions(+), 45 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index a62de0a85406..b0db1583c522 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -751,81 +751,89 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 
 #ifdef CONFIG_PM
 
-static int usb_suspend(struct device *dev, pm_message_t message)
+/* Caller has locked udev */
+static int suspend_device(struct usb_device *udev, pm_message_t msg)
 {
-	struct usb_device		*udev;
 	struct usb_device_driver	*udriver;
-	struct usb_interface		*intf;
-	struct usb_driver		*driver;
-	int				status;
 
-	if (is_usb_device(dev)) {
-		if (dev->driver == NULL)
-			return 0;
-		udev = to_usb_device(dev);
-		udriver = to_usb_device_driver(dev->driver);
-		if (dev->power.power_state.event == message.event)
-			return 0;
-		return udriver->suspend(udev, message);
-	}
+	if (udev->dev.driver == NULL)
+		return 0;
+	udriver = to_usb_device_driver(udev->dev.driver);
+	if (udev->dev.power.power_state.event == msg.event)
+		return 0;
+	return udriver->suspend(udev, msg);
+}
+
+/* Caller has locked udev */
+static int resume_device(struct usb_device *udev)
+{
+	struct usb_device_driver	*udriver;
 
-	if (dev->driver == NULL)
+	if (udev->dev.power.power_state.event == PM_EVENT_ON)
 		return 0;
 
-	intf = to_usb_interface(dev);
-	driver = to_usb_driver(dev->driver);
+	/* mark things as "on" immediately, no matter what errors crop up */
+	udev->dev.power.power_state.event = PM_EVENT_ON;
+
+	if (udev->dev.driver == NULL)
+		return 0;
+	udriver = to_usb_device_driver(udev->dev.driver);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return 0;
+	return udriver->resume(udev);
+}
+
+/* Caller has locked intf */
+static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
+{
+	struct usb_driver	*driver;
+	int			status;
+
+	if (intf->dev.driver == NULL)
+		return 0;
+
+	driver = to_usb_driver(intf->dev.driver);
 
 	/* with no hardware, USB interfaces only use FREEZE and ON states */
 	if (!is_active(intf))
 		return 0;
 
 	if (driver->suspend && driver->resume) {
-		status = driver->suspend(intf, message);
+		status = driver->suspend(intf, msg);
 		if (status)
-			dev_err(dev, "%s error %d\n", "suspend", status);
+			dev_err(&intf->dev, "%s error %d\n",
+					"suspend", status);
 		else
 			mark_quiesced(intf);
 	} else {
 		// FIXME else if there's no suspend method, disconnect...
-		dev_warn(dev, "no suspend for driver %s?\n", driver->name);
+		dev_warn(&intf->dev, "no suspend for driver %s?\n",
+				driver->name);
 		mark_quiesced(intf);
 		status = 0;
 	}
 	return status;
 }
 
-static int usb_resume(struct device *dev)
+/* Caller has locked intf */
+static int resume_interface(struct usb_interface *intf)
 {
-	struct usb_device		*udev;
-	struct usb_device_driver	*udriver;
-	struct usb_interface		*intf;
-	struct usb_driver		*driver;
-	int				status;
+	struct usb_driver	*driver;
+	struct usb_device	*udev;
+	int			status;
 
-	if (dev->power.power_state.event == PM_EVENT_ON)
+	if (intf->dev.power.power_state.event == PM_EVENT_ON)
 		return 0;
 
 	/* mark things as "on" immediately, no matter what errors crop up */
-	dev->power.power_state.event = PM_EVENT_ON;
+	intf->dev.power.power_state.event = PM_EVENT_ON;
 
-	/* devices resume through their hubs */
-	if (is_usb_device(dev)) {
-		if (dev->driver == NULL)
-			return 0;
-		udev = to_usb_device(dev);
-		udriver = to_usb_device_driver(dev->driver);
-		if (udev->state == USB_STATE_NOTATTACHED)
-			return 0;
-		return udriver->resume(udev);
-	}
-
-	if (dev->driver == NULL) {
-		dev->power.power_state.event = PM_EVENT_FREEZE;
+	if (intf->dev.driver == NULL) {
+		intf->dev.power.power_state.event = PM_EVENT_FREEZE;
 		return 0;
 	}
 
-	intf = to_usb_interface(dev);
-	driver = to_usb_driver(dev->driver);
+	driver = to_usb_driver(intf->dev.driver);
 
 	udev = interface_to_usbdev(intf);
 	if (udev->state == USB_STATE_NOTATTACHED)
@@ -838,14 +846,38 @@ static int usb_resume(struct device *dev)
 	if (driver->resume) {
 		status = driver->resume(intf);
 		if (status) {
-			dev_err(dev, "%s error %d\n", "resume", status);
+			dev_err(&intf->dev, "%s error %d\n",
+					"resume", status);
 			mark_quiesced(intf);
 		}
 	} else
-		dev_warn(dev, "no resume for driver %s?\n", driver->name);
+		dev_warn(&intf->dev, "no resume for driver %s?\n",
+				driver->name);
 	return 0;
 }
 
+static int usb_suspend(struct device *dev, pm_message_t message)
+{
+	int	status;
+
+	if (is_usb_device(dev))
+		status = suspend_device(to_usb_device(dev), message);
+	else
+		status = suspend_interface(to_usb_interface(dev), message);
+	return status;
+}
+
+static int usb_resume(struct device *dev)
+{
+	int	status;
+
+	if (is_usb_device(dev))
+		status = resume_device(to_usb_device(dev));
+	else
+		status = resume_interface(to_usb_interface(dev));
+	return status;
+}
+
 #endif /* CONFIG_PM */
 
 struct bus_type usb_bus_type = {
-- 
cgit v1.2.1


From a8e7c5653562f88c0f5f53eac0a890c012655789 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:11:02 -0400
Subject: usbcore: resume device resume recursion

This patch (as717b) removes the existing recursion in hub resume code:
Resuming a hub will no longer automatically resume the devices attached
to the hub.

At the same time, it adds one level of recursion: Suspending a USB
device will automatically suspend all the device's interfaces.  Failure
at an intermediate stage will cause all the already-suspended interfaces
to be resumed. Attempts to suspend or resume an interface by itself will
do nothing, although they won't return an error.  Thus the regular
system-suspend and system-resume procedures should continue to work as
before; only runtime PM will be affected.

The patch also removes the code that tests state of the interfaces
before suspending a device.  It's no longer needed, since everything
gets suspended together.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  | 63 ++++++++++++++++++++++++++----
 drivers/usb/core/generic.c | 14 -------
 drivers/usb/core/hub.c     | 96 +++++++---------------------------------------
 drivers/usb/core/usb.h     |  2 +
 4 files changed, 70 insertions(+), 105 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index b0db1583c522..eefc98584eac 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -783,7 +783,7 @@ static int resume_device(struct usb_device *udev)
 	return udriver->resume(udev);
 }
 
-/* Caller has locked intf */
+/* Caller has locked intf's usb_device */
 static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_driver	*driver;
@@ -815,7 +815,7 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 	return status;
 }
 
-/* Caller has locked intf */
+/* Caller has locked intf's usb_device */
 static int resume_interface(struct usb_interface *intf)
 {
 	struct usb_driver	*driver;
@@ -856,14 +856,59 @@ static int resume_interface(struct usb_interface *intf)
 	return 0;
 }
 
+/* Caller has locked udev */
+int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
+{
+	int			status = 0;
+	int			i = 0;
+	struct usb_interface	*intf;
+
+	if (udev->actconfig) {
+		for (; i < udev->actconfig->desc.bNumInterfaces; i++) {
+			intf = udev->actconfig->interface[i];
+			status = suspend_interface(intf, msg);
+			if (status != 0)
+				break;
+		}
+	}
+	if (status == 0)
+		status = suspend_device(udev, msg);
+
+	/* If the suspend failed, resume interfaces that did get suspended */
+	if (status != 0) {
+		while (--i >= 0) {
+			intf = udev->actconfig->interface[i];
+			resume_interface(intf);
+		}
+	}
+	return status;
+}
+
+/* Caller has locked udev */
+int usb_resume_both(struct usb_device *udev)
+{
+	int			status;
+	int			i;
+	struct usb_interface	*intf;
+
+	status = resume_device(udev);
+	if (status == 0 && udev->actconfig) {
+		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
+			intf = udev->actconfig->interface[i];
+			resume_interface(intf);
+		}
+	}
+	return status;
+}
+
 static int usb_suspend(struct device *dev, pm_message_t message)
 {
 	int	status;
 
 	if (is_usb_device(dev))
-		status = suspend_device(to_usb_device(dev), message);
+		status = usb_suspend_both(to_usb_device(dev), message);
 	else
-		status = suspend_interface(to_usb_interface(dev), message);
+		status = 0;
 	return status;
 }
 
@@ -871,10 +916,12 @@ static int usb_resume(struct device *dev)
 {
 	int	status;
 
-	if (is_usb_device(dev))
-		status = resume_device(to_usb_device(dev));
-	else
-		status = resume_interface(to_usb_interface(dev));
+	if (is_usb_device(dev)) {
+		status = usb_resume_both(to_usb_device(dev));
+
+		/* Rebind drivers that had no suspend method? */
+	} else
+		status = 0;
 	return status;
 }
 
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index 1522195de715..b6dacd7551d2 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -184,22 +184,8 @@ static void generic_disconnect(struct usb_device *udev)
 
 #ifdef	CONFIG_PM
 
-static int verify_suspended(struct device *dev, void *unused)
-{
-	if (dev->driver == NULL)
-		return 0;
-	return (dev->power.power_state.event == PM_EVENT_ON) ? -EBUSY : 0;
-}
-
 static int generic_suspend(struct usb_device *udev, pm_message_t msg)
 {
-	int	status;
-
-	/* rule out bogus requests through sysfs */
-	status = device_for_each_child(&udev->dev, NULL, verify_suspended);
-	if (status)
-		return status;
-
 	/* USB devices enter SUSPEND state through their hubs, but can be
 	 * marked for FREEZE as soon as their children are already idled.
 	 * But those semantics are useless, so we equate the two (sigh).
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index a372332440b2..a39112041e69 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1662,9 +1662,6 @@ static int finish_port_resume(struct usb_device *udev)
 			"gone after usb resume? status %d\n",
 			status);
 	else if (udev->actconfig) {
-		unsigned	i;
-		int		(*resume)(struct device *);
-
 		le16_to_cpus(&devstatus);
 		if ((devstatus & (1 << USB_DEVICE_REMOTE_WAKEUP))
 				&& udev->parent) {
@@ -1675,24 +1672,9 @@ static int finish_port_resume(struct usb_device *udev)
 					USB_DEVICE_REMOTE_WAKEUP, 0,
 					NULL, 0,
 					USB_CTRL_SET_TIMEOUT);
-			if (status) {
+			if (status)
 				dev_dbg(&udev->dev, "disable remote "
 					"wakeup, status %d\n", status);
-				status = 0;
-			}
-		}
-
-		/* resume interface drivers; if this is a hub, it
-		 * may have a child resume event to deal with soon
-		 */
-		resume = udev->dev.bus->resume;
-		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
-			struct device *dev =
-					&udev->actconfig->interface[i]->dev;
-
-			down(&dev->sem);
-			(void) resume(dev);
-			up(&dev->sem);
 		}
 		status = 0;
 
@@ -1802,15 +1784,7 @@ int usb_port_resume(struct usb_device *udev)
 	} else
 		status = finish_port_resume(udev);
 	if (status < 0)
-		dev_dbg(&udev->dev, "can't resume, status %d\n",
-			status);
-
-	/* rebind drivers that had no suspend() */
-	if (status == 0) {
-		usb_unlock_device(udev);
-		bus_rescan_devices(&usb_bus_type);
-		usb_lock_device(udev);
-	}
+		dev_dbg(&udev->dev, "can't resume, status %d\n", status);
 	return status;
 }
 
@@ -1830,6 +1804,9 @@ static int remote_wakeup(struct usb_device *udev)
 		msleep(10);
 		status = finish_port_resume(udev);
 	}
+
+	if (status == 0)
+		usb_resume_both(udev);
 	usb_unlock_device(udev);
 #endif
 	return status;
@@ -1901,51 +1878,8 @@ static int hub_resume(struct usb_interface *intf)
 		}
 	}
 
+	/* tell khubd to look for changes on this hub */
 	hub_activate(hub);
-
-	/* REVISIT:  this recursion probably shouldn't exist.  Remove
-	 * this code sometime, after retesting with different root and
-	 * external hubs.
-	 */
-#ifdef	CONFIG_USB_SUSPEND
-	{
-	unsigned		port1;
-
-	for (port1 = 1; port1 <= hdev->maxchild; port1++) {
-		struct usb_device	*udev;
-		u16			portstat, portchange;
-
-		udev = hdev->children [port1-1];
-		status = hub_port_status(hub, port1, &portstat, &portchange);
-		if (status == 0) {
-			if (portchange & USB_PORT_STAT_C_SUSPEND) {
-				clear_port_feature(hdev, port1,
-					USB_PORT_FEAT_C_SUSPEND);
-				portchange &= ~USB_PORT_STAT_C_SUSPEND;
-			}
-
-			/* let khubd handle disconnects etc */
-			if (portchange)
-				continue;
-		}
-
-		if (!udev || status < 0)
-			continue;
-		usb_lock_device(udev);
-		if (portstat & USB_PORT_STAT_SUSPEND)
-			status = hub_port_resume(hub, port1, udev);
-		else {
-			status = finish_port_resume(udev);
-			if (status < 0) {
-				dev_dbg(&intf->dev, "resume port %d --> %d\n",
-					port1, status);
-				hub_port_logical_disconnect(hub, port1);
-			}
-		}
-		usb_unlock_device(udev);
-	}
-	}
-#endif
 	return 0;
 }
 
@@ -2602,17 +2536,6 @@ static void hub_events(void)
 		usb_get_intf(intf);
 		spin_unlock_irq(&hub_event_lock);
 
-		/* Is this is a root hub wanting to reactivate the downstream
-		 * ports?  If so, be sure the interface resumes even if its
-		 * stub "device" node was never suspended.
-		 */
-		if (i) {
-			dpm_runtime_resume(&hdev->dev);
-			dpm_runtime_resume(&intf->dev);
-			usb_put_intf(intf);
-			continue;
-		}
-
 		/* Lock the device, then check to see if we were
 		 * disconnected while waiting for the lock to succeed. */
 		if (locktree(hdev) < 0) {
@@ -2629,6 +2552,13 @@ static void hub_events(void)
 			goto loop;
 		}
 
+		/* Is this is a root hub wanting to reactivate the downstream
+		 * ports?  If so, be sure the interface resumes even if its
+		 * stub "device" node was never suspended.
+		 */
+		if (i)
+			usb_resume_both(hdev);
+
 		/* If this is an inactive or suspended hub, do nothing */
 		if (hub->quiescing)
 			goto loop;
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 1d25ccac7832..cc42972b6bb0 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -30,6 +30,8 @@ extern void usb_major_cleanup(void);
 extern int usb_host_init(void);
 extern void usb_host_cleanup(void);
 
+extern int usb_suspend_both(struct usb_device *udev, pm_message_t msg);
+extern int usb_resume_both(struct usb_device *udev);
 extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
-- 
cgit v1.2.1


From 4d064c080265a41324d108fccc26b72106d43db3 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:11:44 -0400
Subject: usbcore: track whether interfaces are suspended

Currently we rely on intf->dev.power.power_state.event for tracking
whether intf is suspended.  This is not a reliable technique because
that value is owned by the PM core, not by usbcore.  This patch (as718b)
adds a new flag so that we can accurately tell which interfaces are
suspended and which aren't.

At first one might think these flags aren't needed, since interfaces
will be suspended along with their devices.  It turns out there are a
couple of intermediate situations where that's not quite true, such as
while processing a remote-wakeup request.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.h | 6 +++---
 include/linux/usb.h    | 3 +++
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index cc42972b6bb0..74df0db954c9 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -59,17 +59,17 @@ static inline int is_usb_device_driver(struct device_driver *drv)
 
 static inline void mark_active(struct usb_interface *f)
 {
-	f->dev.power.power_state.event = PM_EVENT_ON;
+	f->is_active = 1;
 }
 
 static inline void mark_quiesced(struct usb_interface *f)
 {
-	f->dev.power.power_state.event = PM_EVENT_FREEZE;
+	f->is_active = 0;
 }
 
 static inline int is_active(struct usb_interface *f)
 {
-	return f->dev.power.power_state.event == PM_EVENT_ON;
+	return f->is_active;
 }
 
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index b4ccce6d0982..e22f4b386605 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -102,6 +102,7 @@ enum usb_interface_condition {
  *	number from the USB core by calling usb_register_dev().
  * @condition: binding state of the interface: not bound, binding
  *	(in probe()), bound to a driver, or unbinding (in disconnect())
+ * @is_active: flag set when the interface is bound and not suspended.
  * @dev: driver model's view of this device
  * @class_dev: driver model's class view of this device.
  *
@@ -142,6 +143,8 @@ struct usb_interface {
 	int minor;			/* minor number this interface is
 					 * bound to */
 	enum usb_interface_condition condition;		/* state of binding */
+	unsigned is_active:1;		/* the interface is not suspended */
+
 	struct device dev;		/* interface specific device info */
 	struct class_device *class_dev;
 };
-- 
cgit v1.2.1


From 2bf4086d7a7722b470aa24e1be725cc58619c6fe Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:12:19 -0400
Subject: usbcore: set device and power states properly

This patch (as733) fixes up the places where device states and power
states are set in usbcore.  Right now things are duplicated or missing;
this should straighten things out.

The idea is that udev->state is USB_STATE_SUSPENDED exactly when the
device's upstream port has been suspended, whereas
udev->dev.power.power_state.event reflects the result of the last call
to the suspend/resume routines (which might not actually change the
device state, especially if CONFIG_USB_SUSPEND isn't set).


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 71 ++++++++++++++++++++++++++++-------------------
 drivers/usb/core/hub.c    | 10 +++----
 2 files changed, 46 insertions(+), 35 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index eefc98584eac..92ecc4eb1e88 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -755,48 +755,57 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 static int suspend_device(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_device_driver	*udriver;
+	int				status = 0;
 
 	if (udev->dev.driver == NULL)
-		return 0;
+		goto done;
 	udriver = to_usb_device_driver(udev->dev.driver);
 	if (udev->dev.power.power_state.event == msg.event)
-		return 0;
-	return udriver->suspend(udev, msg);
+		goto done;
+	status = udriver->suspend(udev, msg);
+
+done:
+	if (status == 0)
+		udev->dev.power.power_state.event = msg.event;
+	return status;
 }
 
 /* Caller has locked udev */
 static int resume_device(struct usb_device *udev)
 {
 	struct usb_device_driver	*udriver;
+	int				status = 0;
 
 	if (udev->dev.power.power_state.event == PM_EVENT_ON)
-		return 0;
-
-	/* mark things as "on" immediately, no matter what errors crop up */
-	udev->dev.power.power_state.event = PM_EVENT_ON;
+		goto done;
 
 	if (udev->dev.driver == NULL)
-		return 0;
+		goto done;
 	udriver = to_usb_device_driver(udev->dev.driver);
 	if (udev->state == USB_STATE_NOTATTACHED)
-		return 0;
-	return udriver->resume(udev);
+		goto done;
+	status = udriver->resume(udev);
+
+done:
+	if (status == 0)
+		udev->dev.power.power_state.event = PM_EVENT_ON;
+	return status;
 }
 
 /* Caller has locked intf's usb_device */
 static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_driver	*driver;
-	int			status;
+	int			status = 0;
 
 	if (intf->dev.driver == NULL)
-		return 0;
+		goto done;
 
 	driver = to_usb_driver(intf->dev.driver);
 
 	/* with no hardware, USB interfaces only use FREEZE and ON states */
 	if (!is_active(intf))
-		return 0;
+		goto done;
 
 	if (driver->suspend && driver->resume) {
 		status = driver->suspend(intf, msg);
@@ -810,8 +819,11 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 		dev_warn(&intf->dev, "no suspend for driver %s?\n",
 				driver->name);
 		mark_quiesced(intf);
-		status = 0;
 	}
+
+done:
+	if (status == 0)
+		intf->dev.power.power_state.event = msg.event;
 	return status;
 }
 
@@ -820,24 +832,19 @@ static int resume_interface(struct usb_interface *intf)
 {
 	struct usb_driver	*driver;
 	struct usb_device	*udev;
-	int			status;
+	int			status = 0;
 
 	if (intf->dev.power.power_state.event == PM_EVENT_ON)
-		return 0;
-
-	/* mark things as "on" immediately, no matter what errors crop up */
-	intf->dev.power.power_state.event = PM_EVENT_ON;
+		goto done;
 
-	if (intf->dev.driver == NULL) {
-		intf->dev.power.power_state.event = PM_EVENT_FREEZE;
-		return 0;
-	}
+	if (intf->dev.driver == NULL)
+		goto done;
 
 	driver = to_usb_driver(intf->dev.driver);
 
 	udev = interface_to_usbdev(intf);
 	if (udev->state == USB_STATE_NOTATTACHED)
-		return 0;
+		goto done;
 
 	/* if driver was suspended, it has a resume method;
 	 * however, sysfs can wrongly mark things as suspended
@@ -845,15 +852,21 @@ static int resume_interface(struct usb_interface *intf)
 	 */
 	if (driver->resume) {
 		status = driver->resume(intf);
-		if (status) {
+		if (status)
 			dev_err(&intf->dev, "%s error %d\n",
 					"resume", status);
-			mark_quiesced(intf);
-		}
-	} else
+		else
+			mark_active(intf);
+	} else {
 		dev_warn(&intf->dev, "no resume for driver %s?\n",
 				driver->name);
-	return 0;
+		mark_active(intf);
+	}
+
+done:
+	if (status == 0)
+		intf->dev.power.power_state.event = PM_EVENT_ON;
+	return status;
 }
 
 /* Caller has locked udev */
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index a39112041e69..7af53db4d76f 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1582,9 +1582,10 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 	if (udev->parent)
 		status = hub_port_suspend(hdev_to_hub(udev->parent), port1,
 				udev);
-
-	if (status == 0)
-		udev->dev.power.power_state = PMSG_SUSPEND;
+	else {
+		dev_dbg(&udev->dev, "usb suspend\n");
+		usb_set_device_state(udev, USB_STATE_SUSPENDED);
+	}
 	return status;
 }
 
@@ -1617,8 +1618,6 @@ int usb_port_suspend(struct usb_device *udev)
 		return -ENODEV;
 	return __usb_port_suspend(udev, udev->portnum);
 #else
-	/* NOTE:  udev->state unchanged, it's not lying ... */
-	udev->dev.power.power_state = PMSG_SUSPEND;
 	return 0;
 #endif
 }
@@ -1647,7 +1646,6 @@ static int finish_port_resume(struct usb_device *udev)
 	usb_set_device_state(udev, udev->actconfig
 			? USB_STATE_CONFIGURED
 			: USB_STATE_ADDRESS);
-	udev->dev.power.power_state = PMSG_ON;
 
  	/* 10.5.4.5 says be sure devices in the tree are still there.
  	 * For now let's assume the device didn't go crazy on resume,
-- 
cgit v1.2.1


From 114b368c07964caa3f4e1fa575b16e87fa11936c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:13:04 -0400
Subject: usbcore: fix up device and power state tests

This patch (as734) rationalizes the various tests of device state and
power states.  There are duplications and mistaken tests in several
places.

Perhaps the most interesting challenge is where the hub driver tests to
see that all the child devices are suspended before allowing itself to
be suspended.  When CONFIG_USB_SUSPEND is set the test is
straightforward, since we expect that the children _will_ be suspended.
But when CONFIG_USB_SUSPEND isn't set, it's not so clear what should be
done.  The code compromises by checking the child's
power.power_state.event field.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c  | 40 +++++++++++++++++++--------------------
 drivers/usb/core/generic.c |  3 ---
 drivers/usb/core/hub.c     | 47 ++++++++++------------------------------------
 3 files changed, 29 insertions(+), 61 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 92ecc4eb1e88..affbfb53eb5e 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -757,11 +757,13 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 	struct usb_device_driver	*udriver;
 	int				status = 0;
 
+	if (udev->state == USB_STATE_NOTATTACHED ||
+			udev->state == USB_STATE_SUSPENDED)
+		goto done;
+
 	if (udev->dev.driver == NULL)
 		goto done;
 	udriver = to_usb_device_driver(udev->dev.driver);
-	if (udev->dev.power.power_state.event == msg.event)
-		goto done;
 	status = udriver->suspend(udev, msg);
 
 done:
@@ -776,14 +778,13 @@ static int resume_device(struct usb_device *udev)
 	struct usb_device_driver	*udriver;
 	int				status = 0;
 
-	if (udev->dev.power.power_state.event == PM_EVENT_ON)
+	if (udev->state == USB_STATE_NOTATTACHED ||
+			udev->state != USB_STATE_SUSPENDED)
 		goto done;
 
 	if (udev->dev.driver == NULL)
 		goto done;
 	udriver = to_usb_device_driver(udev->dev.driver);
-	if (udev->state == USB_STATE_NOTATTACHED)
-		goto done;
 	status = udriver->resume(udev);
 
 done:
@@ -798,14 +799,14 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 	struct usb_driver	*driver;
 	int			status = 0;
 
-	if (intf->dev.driver == NULL)
+	/* with no hardware, USB interfaces only use FREEZE and ON states */
+	if (interface_to_usbdev(intf)->state == USB_STATE_NOTATTACHED ||
+			!is_active(intf))
 		goto done;
 
-	driver = to_usb_driver(intf->dev.driver);
-
-	/* with no hardware, USB interfaces only use FREEZE and ON states */
-	if (!is_active(intf))
+	if (intf->dev.driver == NULL)
 		goto done;
+	driver = to_usb_driver(intf->dev.driver);
 
 	if (driver->suspend && driver->resume) {
 		status = driver->suspend(intf, msg);
@@ -831,25 +832,16 @@ done:
 static int resume_interface(struct usb_interface *intf)
 {
 	struct usb_driver	*driver;
-	struct usb_device	*udev;
 	int			status = 0;
 
-	if (intf->dev.power.power_state.event == PM_EVENT_ON)
+	if (interface_to_usbdev(intf)->state == USB_STATE_NOTATTACHED ||
+			is_active(intf))
 		goto done;
 
 	if (intf->dev.driver == NULL)
 		goto done;
-
 	driver = to_usb_driver(intf->dev.driver);
 
-	udev = interface_to_usbdev(intf);
-	if (udev->state == USB_STATE_NOTATTACHED)
-		goto done;
-
-	/* if driver was suspended, it has a resume method;
-	 * however, sysfs can wrongly mark things as suspended
-	 * (on the "no suspend method" FIXME path above)
-	 */
 	if (driver->resume) {
 		status = driver->resume(intf);
 		if (status)
@@ -904,6 +896,12 @@ int usb_resume_both(struct usb_device *udev)
 	int			i;
 	struct usb_interface	*intf;
 
+	/* Can't resume if the parent is suspended */
+	if (udev->parent && udev->parent->state == USB_STATE_SUSPENDED) {
+		dev_warn(&udev->dev, "can't resume; parent is suspended\n");
+		return -EHOSTUNREACH;
+	}
+
 	status = resume_device(udev);
 	if (status == 0 && udev->actconfig) {
 		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index b6dacd7551d2..5358e656477c 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -195,9 +195,6 @@ static int generic_suspend(struct usb_device *udev, pm_message_t msg)
 
 static int generic_resume(struct usb_device *udev)
 {
-	if (udev->state == USB_STATE_NOTATTACHED)
-		return 0;
-
 	return usb_port_resume(udev);
 }
 
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 7af53db4d76f..a310c7cede99 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1556,26 +1556,6 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 	if (port1 < 0)
 		return port1;
 
-	if (udev->state == USB_STATE_SUSPENDED
-			|| udev->state == USB_STATE_NOTATTACHED) {
-		return 0;
-	}
-
-	/* all interfaces must already be suspended */
-	if (udev->actconfig) {
-		int	i;
-
-		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
-			struct usb_interface	*intf;
-
-			intf = udev->actconfig->interface[i];
-			if (is_active(intf)) {
-				dev_dbg(&intf->dev, "nyet suspended\n");
-				return -EBUSY;
-			}
-		}
-	}
-
 	/* we change the device's upstream USB link,
 	 * but root hubs have no upstream USB link.
 	 */
@@ -1614,8 +1594,6 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 int usb_port_suspend(struct usb_device *udev)
 {
 #ifdef	CONFIG_USB_SUSPEND
-	if (udev->state == USB_STATE_NOTATTACHED)
-		return -ENODEV;
 	return __usb_port_suspend(udev, udev->portnum);
 #else
 	return 0;
@@ -1761,24 +1739,17 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
  */
 int usb_port_resume(struct usb_device *udev)
 {
-	int	status;
-
-	if (udev->state == USB_STATE_NOTATTACHED)
-		return -ENODEV;
+	int	status = 0;
 
 	/* we change the device's upstream USB link,
 	 * but root hubs have no upstream USB link.
 	 */
 	if (udev->parent) {
 #ifdef	CONFIG_USB_SUSPEND
-		if (udev->state == USB_STATE_SUSPENDED) {
-			// NOTE swsusp may bork us, device state being wrong...
-			// NOTE this fails if parent is also suspended...
-			status = hub_port_resume(hdev_to_hub(udev->parent),
-					udev->portnum, udev);
-		} else
+		// NOTE this fails if parent is also suspended...
+		status = hub_port_resume(hdev_to_hub(udev->parent),
+				udev->portnum, udev);
 #endif
-			status = 0;
 	} else
 		status = finish_port_resume(udev);
 	if (status < 0)
@@ -1821,12 +1792,14 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 		struct usb_device	*udev;
 
 		udev = hdev->children [port1-1];
-		if (udev && (udev->dev.power.power_state.event
-					== PM_EVENT_ON
+		if (udev && msg.event == PM_EVENT_SUSPEND &&
 #ifdef	CONFIG_USB_SUSPEND
-				|| udev->state != USB_STATE_SUSPENDED
+				udev->state != USB_STATE_SUSPENDED
+#else
+				udev->dev.power.power_state.event
+					== PM_EVENT_ON
 #endif
-				)) {
+				) {
 			dev_dbg(&intf->dev, "port %d nyet suspended\n", port1);
 			return -EBUSY;
 		}
-- 
cgit v1.2.1


From 1c5df7e705671f11a71112eb3a1f9765cd1719f9 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:13:50 -0400
Subject: usbcore: suspending devices with no driver

Since usb_generic can be unbound from a USB device, we need to be able
to handle the possibility that a suspend or resume request arrives for a
device with no driver.  This patch (as735) arranges things so that
resume requests will fail and suspend requests will use the standard USB
port-suspend code.  Attempts to suspend or resume an unbound interface
are handled similarly (although the error caused by trying to resume an
unbound interface is dropped by the calling routine).


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index affbfb53eb5e..a5d11461f5a9 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -761,8 +761,12 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 			udev->state == USB_STATE_SUSPENDED)
 		goto done;
 
-	if (udev->dev.driver == NULL)
+	/* For devices that don't have a driver, we do a standard suspend. */
+	if (udev->dev.driver == NULL) {
+		status = usb_port_suspend(udev);
 		goto done;
+	}
+
 	udriver = to_usb_device_driver(udev->dev.driver);
 	status = udriver->suspend(udev, msg);
 
@@ -782,8 +786,12 @@ static int resume_device(struct usb_device *udev)
 			udev->state != USB_STATE_SUSPENDED)
 		goto done;
 
-	if (udev->dev.driver == NULL)
+	/* Can't resume it if it doesn't have a driver. */
+	if (udev->dev.driver == NULL) {
+		status = -ENOTCONN;
 		goto done;
+	}
+
 	udriver = to_usb_device_driver(udev->dev.driver);
 	status = udriver->resume(udev);
 
@@ -804,7 +812,7 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 			!is_active(intf))
 		goto done;
 
-	if (intf->dev.driver == NULL)
+	if (intf->dev.driver == NULL)		/* This can't happen */
 		goto done;
 	driver = to_usb_driver(intf->dev.driver);
 
@@ -838,8 +846,11 @@ static int resume_interface(struct usb_interface *intf)
 			is_active(intf))
 		goto done;
 
-	if (intf->dev.driver == NULL)
+	/* Can't resume it if it doesn't have a driver. */
+	if (intf->dev.driver == NULL) {
+		status = -ENOTCONN;
 		goto done;
+	}
 	driver = to_usb_driver(intf->dev.driver);
 
 	if (driver->resume) {
-- 
cgit v1.2.1


From d388dab7b562b76525761f89702246686747ba30 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Sat, 1 Jul 2006 22:14:24 -0400
Subject: hub driver: improve use of #ifdef

This patch (as736) makes the hub driver more readable by improving the
usage of "#ifdef CONFIG_PM" and "#ifdef CONFIG_USB_SUSPEND".


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 64 ++++++++++++++++++++++++++++++++++++--------------
 drivers/usb/core/usb.h | 11 +++++++++
 2 files changed, 57 insertions(+), 18 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index a310c7cede99..64e80b964b87 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1041,7 +1041,7 @@ void usb_set_device_state(struct usb_device *udev,
 }
 
 
-#ifdef CONFIG_PM
+#ifdef	CONFIG_PM
 
 /**
  * usb_root_hub_lost_power - called by HCD if the root hub lost Vbus power
@@ -1072,7 +1072,7 @@ void usb_root_hub_lost_power(struct usb_device *rhdev)
 }
 EXPORT_SYMBOL_GPL(usb_root_hub_lost_power);
 
-#endif
+#endif	/* CONFIG_PM */
 
 static void choose_address(struct usb_device *udev)
 {
@@ -1471,6 +1471,7 @@ static void hub_port_logical_disconnect(struct usb_hub *hub, int port1)
  	kick_khubd(hub);
 }
 
+#ifdef	CONFIG_PM
 
 #ifdef	CONFIG_USB_SUSPEND
 
@@ -1569,8 +1570,6 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 	return status;
 }
 
-#endif
-
 /*
  * usb_port_suspend - suspend a usb device's upstream port
  * @udev: device that's no longer in active use
@@ -1593,11 +1592,7 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
  */
 int usb_port_suspend(struct usb_device *udev)
 {
-#ifdef	CONFIG_USB_SUSPEND
 	return __usb_port_suspend(udev, udev->portnum);
-#else
-	return 0;
-#endif
 }
 
 /*
@@ -1661,8 +1656,6 @@ static int finish_port_resume(struct usb_device *udev)
 	return status;
 }
 
-#ifdef	CONFIG_USB_SUSPEND
-
 static int
 hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 {
@@ -1722,8 +1715,6 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 	return status;
 }
 
-#endif
-
 /*
  * usb_port_resume - re-activate a suspended usb device's upstream port
  * @udev: device to re-activate
@@ -1739,17 +1730,15 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
  */
 int usb_port_resume(struct usb_device *udev)
 {
-	int	status = 0;
+	int	status;
 
 	/* we change the device's upstream USB link,
 	 * but root hubs have no upstream USB link.
 	 */
 	if (udev->parent) {
-#ifdef	CONFIG_USB_SUSPEND
 		// NOTE this fails if parent is also suspended...
 		status = hub_port_resume(hdev_to_hub(udev->parent),
 				udev->portnum, udev);
-#endif
 	} else
 		status = finish_port_resume(udev);
 	if (status < 0)
@@ -1761,8 +1750,6 @@ static int remote_wakeup(struct usb_device *udev)
 {
 	int	status = 0;
 
-#ifdef	CONFIG_USB_SUSPEND
-
 	/* don't repeat RESUME sequence if this device
 	 * was already woken up by some other task
 	 */
@@ -1777,10 +1764,42 @@ static int remote_wakeup(struct usb_device *udev)
 	if (status == 0)
 		usb_resume_both(udev);
 	usb_unlock_device(udev);
-#endif
 	return status;
 }
 
+#else	/* CONFIG_USB_SUSPEND */
+
+/* When CONFIG_USB_SUSPEND isn't set, we never suspend or resume any ports. */
+
+int usb_port_suspend(struct usb_device *udev)
+{
+	return 0;
+}
+
+static inline int
+finish_port_resume(struct usb_device *udev)
+{
+	return 0;
+}
+
+static inline int
+hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
+{
+	return 0;
+}
+
+int usb_port_resume(struct usb_device *udev)
+{
+	return 0;
+}
+
+static inline int remote_wakeup(struct usb_device *udev)
+{
+	return 0;
+}
+
+#endif
+
 static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_hub		*hub = usb_get_intfdata (intf);
@@ -1854,6 +1873,15 @@ static int hub_resume(struct usb_interface *intf)
 	return 0;
 }
 
+#else	/* CONFIG_PM */
+
+static inline int remote_wakeup(struct usb_device *udev)
+{
+	return 0;
+}
+
+#endif
+
 void usb_suspend_root_hub(struct usb_device *hdev)
 {
 	struct usb_hub *hub = hdev_to_hub(hdev);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 74df0db954c9..98675fb1bc4c 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -30,11 +30,22 @@ extern void usb_major_cleanup(void);
 extern int usb_host_init(void);
 extern void usb_host_cleanup(void);
 
+#ifdef	CONFIG_PM
+
 extern int usb_suspend_both(struct usb_device *udev, pm_message_t msg);
 extern int usb_resume_both(struct usb_device *udev);
 extern int usb_port_suspend(struct usb_device *dev);
 extern int usb_port_resume(struct usb_device *dev);
 
+#else
+
+#define usb_suspend_both(udev, msg)	0
+#define usb_resume_both(udev)		0
+#define usb_port_suspend(dev)		0
+#define usb_port_resume(dev)		0
+
+#endif
+
 extern struct bus_type usb_bus_type;
 extern struct usb_device_driver usb_generic_driver;
 
-- 
cgit v1.2.1


From 5d8926658ce41b254fdfba7d057e6c9438c25cca Mon Sep 17 00:00:00 2001
From: Daniel Ritz <daniel.ritz-ml@swissonline.ch>
Date: Mon, 31 Jul 2006 21:43:24 +0200
Subject: usbtouchscreen: version 0.4

changes over 0.3:
- some more eGalax device IDs (from eGalax driver/spec)
- return the error code in probe()
- 3M/MTouch init fixes, tested by Don Alexander
- eGalax fixes for bugs in multi-packet handling, spottet by Pieter Grimmerink
- support for some eTurboTouch devices, mostly by Pieter Grimmerink
- support for Gunze AHL61 controller (untested, but simple enough)

Signed-off-by: Daniel Ritz <daniel.ritz@gmx.ch>
Cc: Pieter Grimmerink <p.grimmerink@inepro.com>
Cc: Don Alexander <debug@roosoft.ltd.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/Kconfig          |  20 ++-
 drivers/usb/input/usbtouchscreen.c | 283 ++++++++++++++++++++++++++-----------
 2 files changed, 219 insertions(+), 84 deletions(-)

diff --git a/drivers/usb/input/Kconfig b/drivers/usb/input/Kconfig
index 650103bc9618..8a708520547b 100644
--- a/drivers/usb/input/Kconfig
+++ b/drivers/usb/input/Kconfig
@@ -205,10 +205,12 @@ config USB_TOUCHSCREEN
 	depends on USB && INPUT
 	---help---
 	  USB Touchscreen driver for:
-	  - eGalax Touchkit USB
+	  - eGalax Touchkit USB (also includes eTurboTouch CT-410/510/700)
 	  - PanJit TouchSet USB
-	  - 3M MicroTouch USB
+	  - 3M MicroTouch USB (EX II series)
 	  - ITM
+	  - some other eTurboTouch
+	  - Gunze AHL61
 
 	  Have a look at <http://linux.chapter7.ch/touchkit/> for
 	  a usage description and the required user-space stuff.
@@ -218,7 +220,7 @@ config USB_TOUCHSCREEN
 
 config USB_TOUCHSCREEN_EGALAX
 	default y
-	bool "eGalax device support" if EMBEDDED
+	bool "eGalax, eTurboTouch CT-410/510/700 device support" if EMBEDDED
 	depends on USB_TOUCHSCREEN
 
 config USB_TOUCHSCREEN_PANJIT
@@ -228,7 +230,7 @@ config USB_TOUCHSCREEN_PANJIT
 
 config USB_TOUCHSCREEN_3M
 	default y
-	bool "3M/Microtouch device support" if EMBEDDED
+	bool "3M/Microtouch EX II series device support" if EMBEDDED
 	depends on USB_TOUCHSCREEN
 
 config USB_TOUCHSCREEN_ITM
@@ -236,6 +238,16 @@ config USB_TOUCHSCREEN_ITM
 	bool "ITM device support" if EMBEDDED
 	depends on USB_TOUCHSCREEN
 
+config USB_TOUCHSCREEN_ETURBO
+	default y
+	bool "eTurboTouch (non-eGalax compatible) device support" if EMBEDDED
+	depends on USB_TOUCHSCREEN
+
+config USB_TOUCHSCREEN_GUNZE
+	default y
+	bool "Gunze AHL61 device support" if EMBEDDED
+	depends on USB_TOUCHSCREEN
+
 config USB_YEALINK
 	tristate "Yealink usb-p1k voip phone"
 	depends on USB && INPUT && EXPERIMENTAL
diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c
index a338bf4c2d78..a1be7840ea02 100644
--- a/drivers/usb/input/usbtouchscreen.c
+++ b/drivers/usb/input/usbtouchscreen.c
@@ -2,9 +2,12 @@
  * usbtouchscreen.c
  * Driver for USB Touchscreens, supporting those devices:
  *  - eGalax Touchkit
- *  - 3M/Microtouch
+ *    includes eTurboTouch CT-410/510/700
+ *  - 3M/Microtouch  EX II series
  *  - ITM
  *  - PanJit TouchSet
+ *  - eTurboTouch
+ *  - Gunze AHL61
  *
  * Copyright (C) 2004-2006 by Daniel Ritz <daniel.ritz@gmx.ch>
  * Copyright (C) by Todd E. Johnson (mtouchusb.c)
@@ -42,7 +45,7 @@
 #include <linux/usb/input.h>
 
 
-#define DRIVER_VERSION		"v0.3"
+#define DRIVER_VERSION		"v0.4"
 #define DRIVER_AUTHOR		"Daniel Ritz <daniel.ritz@gmx.ch>"
 #define DRIVER_DESC		"USB Touchscreen Driver"
 
@@ -60,6 +63,7 @@ struct usbtouch_device_info {
 	int flags;
 
 	void (*process_pkt) (struct usbtouch_usb *usbtouch, struct pt_regs *regs, unsigned char *pkt, int len);
+	int  (*get_pkt_len) (unsigned char *pkt, int len);
 	int  (*read_data)   (unsigned char *pkt, int *x, int *y, int *touch, int *press);
 	int  (*init)        (struct usbtouch_usb *usbtouch);
 };
@@ -81,8 +85,16 @@ struct usbtouch_usb {
 	char phys[64];
 };
 
-static void usbtouch_process_pkt(struct usbtouch_usb *usbtouch,
-                                 struct pt_regs *regs, unsigned char *pkt, int len);
+
+#if defined(CONFIG_USB_TOUCHSCREEN_EGALAX) || defined(CONFIG_USB_TOUCHSCREEN_ETURBO)
+#define MULTI_PACKET
+#endif
+
+#ifdef MULTI_PACKET
+static void usbtouch_process_multi(struct usbtouch_usb *usbtouch,
+                                   struct pt_regs *regs,
+                                   unsigned char *pkt, int len);
+#endif
 
 /* device types */
 enum {
@@ -91,14 +103,19 @@ enum {
 	DEVTYPE_PANJIT,
 	DEVTYPE_3M,
 	DEVTYPE_ITM,
+	DEVTYPE_ETURBO,
+	DEVTYPE_GUNZE,
 };
 
 static struct usb_device_id usbtouch_devices[] = {
 #ifdef CONFIG_USB_TOUCHSCREEN_EGALAX
 	{USB_DEVICE(0x3823, 0x0001), .driver_info = DEVTYPE_EGALAX},
+	{USB_DEVICE(0x3823, 0x0002), .driver_info = DEVTYPE_EGALAX},
 	{USB_DEVICE(0x0123, 0x0001), .driver_info = DEVTYPE_EGALAX},
 	{USB_DEVICE(0x0eef, 0x0001), .driver_info = DEVTYPE_EGALAX},
 	{USB_DEVICE(0x0eef, 0x0002), .driver_info = DEVTYPE_EGALAX},
+	{USB_DEVICE(0x1234, 0x0001), .driver_info = DEVTYPE_EGALAX},
+	{USB_DEVICE(0x1234, 0x0002), .driver_info = DEVTYPE_EGALAX},
 #endif
 
 #ifdef CONFIG_USB_TOUCHSCREEN_PANJIT
@@ -116,6 +133,14 @@ static struct usb_device_id usbtouch_devices[] = {
 	{USB_DEVICE(0x0403, 0xf9e9), .driver_info = DEVTYPE_ITM},
 #endif
 
+#ifdef CONFIG_USB_TOUCHSCREEN_ETURBO
+	{USB_DEVICE(0x1234, 0x5678), .driver_info = DEVTYPE_ETURBO},
+#endif
+
+#ifdef CONFIG_USB_TOUCHSCREEN_GUNZE
+	{USB_DEVICE(0x0637, 0x0001), .driver_info = DEVTYPE_GUNZE},
+#endif
+
 	{}
 };
 
@@ -140,82 +165,23 @@ static int egalax_read_data(unsigned char *pkt, int *x, int *y, int *touch, int
 	*touch = pkt[0] & 0x01;
 
 	return 1;
-
 }
 
-static int egalax_get_pkt_len(unsigned char *buf)
+static int egalax_get_pkt_len(unsigned char *buf, int len)
 {
 	switch (buf[0] & EGALAX_PKT_TYPE_MASK) {
 	case EGALAX_PKT_TYPE_REPT:
 		return 5;
 
 	case EGALAX_PKT_TYPE_DIAG:
+		if (len < 2)
+			return -1;
+
 		return buf[1] + 2;
 	}
 
 	return 0;
 }
-
-static void egalax_process(struct usbtouch_usb *usbtouch, struct pt_regs *regs,
-                           unsigned char *pkt, int len)
-{
-	unsigned char *buffer;
-	int pkt_len, buf_len, pos;
-
-	/* if the buffer contains data, append */
-	if (unlikely(usbtouch->buf_len)) {
-		int tmp;
-
-		/* if only 1 byte in buffer, add another one to get length */
-		if (usbtouch->buf_len == 1)
-			usbtouch->buffer[1] = pkt[0];
-
-		pkt_len = egalax_get_pkt_len(usbtouch->buffer);
-
-		/* unknown packet: drop everything */
-		if (!pkt_len)
-			return;
-
-		/* append, process */
-		tmp = pkt_len - usbtouch->buf_len;
-		memcpy(usbtouch->buffer + usbtouch->buf_len, pkt, tmp);
-		usbtouch_process_pkt(usbtouch, regs, usbtouch->buffer, pkt_len);
-
-		buffer = pkt + tmp;
-		buf_len = len - tmp;
-	} else {
-		buffer = pkt;
-		buf_len = len;
-	}
-
-	/* only one byte left in buffer */
-	if (unlikely(buf_len == 1)) {
-		usbtouch->buffer[0] = buffer[0];
-		usbtouch->buf_len = 1;
-		return;
-	}
-
-	/* loop over the buffer */
-	pos = 0;
-	while (pos < buf_len) {
-		/* get packet len */
-		pkt_len = egalax_get_pkt_len(buffer + pos);
-
-		/* unknown packet: drop everything */
-		if (unlikely(!pkt_len))
-			return;
-
-		/* full packet: process */
-		if (likely(pkt_len <= buf_len)) {
-			usbtouch_process_pkt(usbtouch, regs, buffer + pos, pkt_len);
-		} else {
-			/* incomplete packet: save in buffer */
-			memcpy(usbtouch->buffer, buffer + pos, buf_len - pos);
-			usbtouch->buf_len = buf_len - pos;
-		}
-		pos += pkt_len;
-	}
-}
 #endif
 
 
@@ -254,7 +220,7 @@ static int mtouch_read_data(unsigned char *pkt, int *x, int *y, int *touch, int
 
 static int mtouch_init(struct usbtouch_usb *usbtouch)
 {
-	int ret;
+	int ret, i;
 
 	ret = usb_control_msg(usbtouch->udev, usb_rcvctrlpipe(usbtouch->udev, 0),
 	                      MTOUCHUSB_RESET,
@@ -264,15 +230,20 @@ static int mtouch_init(struct usbtouch_usb *usbtouch)
 	    __FUNCTION__, ret);
 	if (ret < 0)
 		return ret;
-
-	ret = usb_control_msg(usbtouch->udev, usb_rcvctrlpipe(usbtouch->udev, 0),
-	                      MTOUCHUSB_ASYNC_REPORT,
-	                      USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-	                      1, 1, NULL, 0, USB_CTRL_SET_TIMEOUT);
-	dbg("%s - usb_control_msg - MTOUCHUSB_ASYNC_REPORT - bytes|err: %d",
-	    __FUNCTION__, ret);
-	if (ret < 0)
-		return ret;
+	msleep(150);
+
+	for (i = 0; i < 3; i++) {
+		ret = usb_control_msg(usbtouch->udev, usb_rcvctrlpipe(usbtouch->udev, 0),
+				      MTOUCHUSB_ASYNC_REPORT,
+				      USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
+				      1, 1, NULL, 0, USB_CTRL_SET_TIMEOUT);
+		dbg("%s - usb_control_msg - MTOUCHUSB_ASYNC_REPORT - bytes|err: %d",
+		    __FUNCTION__, ret);
+		if (ret >= 0)
+			break;
+		if (ret != -EPIPE)
+			return ret;
+	}
 
 	return 0;
 }
@@ -295,6 +266,54 @@ static int itm_read_data(unsigned char *pkt, int *x, int *y, int *touch, int *pr
 #endif
 
 
+/*****************************************************************************
+ * eTurboTouch part
+ */
+#ifdef CONFIG_USB_TOUCHSCREEN_ETURBO
+static int eturbo_read_data(unsigned char *pkt, int *x, int *y, int *touch, int *press)
+{
+	unsigned int shift;
+
+	/* packets should start with sync */
+	if (!(pkt[0] & 0x80))
+		return 0;
+
+	shift = (6 - (pkt[0] & 0x03));
+	*x = ((pkt[3] << 7) | pkt[4]) >> shift;
+	*y = ((pkt[1] << 7) | pkt[2]) >> shift;
+	*touch = (pkt[0] & 0x10) ? 1 : 0;
+
+	return 1;
+}
+
+static int eturbo_get_pkt_len(unsigned char *buf, int len)
+{
+	if (buf[0] & 0x80)
+		return 5;
+	if (buf[0] == 0x01)
+		return 3;
+	return 0;
+}
+#endif
+
+
+/*****************************************************************************
+ * Gunze part
+ */
+#ifdef CONFIG_USB_TOUCHSCREEN_GUNZE
+static int gunze_read_data(unsigned char *pkt, int *x, int *y, int *touch, int *press)
+{
+	if (!(pkt[0] & 0x80) || ((pkt[1] | pkt[2] | pkt[3]) & 0x80))
+		return 0;
+
+	*x = ((pkt[0] & 0x1F) << 7) | (pkt[2] & 0x7F);
+	*y = ((pkt[1] & 0x1F) << 7) | (pkt[3] & 0x7F);
+	*touch = pkt[0] & 0x20;
+
+	return 1;
+}
+#endif
+
 /*****************************************************************************
  * the different device descriptors
  */
@@ -307,7 +326,8 @@ static struct usbtouch_device_info usbtouch_dev_info[] = {
 		.max_yc		= 0x07ff,
 		.rept_size	= 16,
 		.flags		= USBTOUCH_FLG_BUFFER,
-		.process_pkt	= egalax_process,
+		.process_pkt	= usbtouch_process_multi,
+		.get_pkt_len	= egalax_get_pkt_len,
 		.read_data	= egalax_read_data,
 	},
 #endif
@@ -346,6 +366,31 @@ static struct usbtouch_device_info usbtouch_dev_info[] = {
 		.read_data	= itm_read_data,
 	},
 #endif
+
+#ifdef CONFIG_USB_TOUCHSCREEN_ETURBO
+	[DEVTYPE_ETURBO] = {
+		.min_xc		= 0x0,
+		.max_xc		= 0x07ff,
+		.min_yc		= 0x0,
+		.max_yc		= 0x07ff,
+		.rept_size	= 8,
+		.flags		= USBTOUCH_FLG_BUFFER,
+		.process_pkt	= usbtouch_process_multi,
+		.get_pkt_len	= eturbo_get_pkt_len,
+		.read_data	= eturbo_read_data,
+	},
+#endif
+
+#ifdef CONFIG_USB_TOUCHSCREEN_GUNZE
+	[DEVTYPE_GUNZE] = {
+		.min_xc		= 0x0,
+		.max_xc		= 0x0fff,
+		.min_yc		= 0x0,
+		.max_yc		= 0x0fff,
+		.rept_size	= 4,
+		.read_data	= gunze_read_data,
+	},
+#endif
 };
 
 
@@ -377,6 +422,83 @@ static void usbtouch_process_pkt(struct usbtouch_usb *usbtouch,
 }
 
 
+#ifdef MULTI_PACKET
+static void usbtouch_process_multi(struct usbtouch_usb *usbtouch,
+                                   struct pt_regs *regs,
+                                   unsigned char *pkt, int len)
+{
+	unsigned char *buffer;
+	int pkt_len, pos, buf_len, tmp;
+
+	/* process buffer */
+	if (unlikely(usbtouch->buf_len)) {
+		/* try to get size */
+		pkt_len = usbtouch->type->get_pkt_len(
+				usbtouch->buffer, usbtouch->buf_len);
+
+		/* drop? */
+		if (unlikely(!pkt_len))
+			goto out_flush_buf;
+
+		/* need to append -pkt_len bytes before able to get size */
+		if (unlikely(pkt_len < 0)) {
+			int append = -pkt_len;
+			if (unlikely(append > len))
+			       append = len;
+			if (usbtouch->buf_len + append >= usbtouch->type->rept_size)
+				goto out_flush_buf;
+			memcpy(usbtouch->buffer + usbtouch->buf_len, pkt, append);
+			usbtouch->buf_len += append;
+
+			pkt_len = usbtouch->type->get_pkt_len(
+					usbtouch->buffer, usbtouch->buf_len);
+			if (pkt_len < 0)
+				return;
+		}
+
+		/* append */
+		tmp = pkt_len - usbtouch->buf_len;
+		if (usbtouch->buf_len + tmp >= usbtouch->type->rept_size)
+			goto out_flush_buf;
+		memcpy(usbtouch->buffer + usbtouch->buf_len, pkt, tmp);
+		usbtouch_process_pkt(usbtouch, regs, usbtouch->buffer, pkt_len);
+
+		buffer = pkt + tmp;
+		buf_len = len - tmp;
+	} else {
+		buffer = pkt;
+		buf_len = len;
+	}
+
+	/* loop over the received packet, process */
+	pos = 0;
+	while (pos < buf_len) {
+		/* get packet len */
+		pkt_len = usbtouch->type->get_pkt_len(buffer + pos, len);
+
+		/* unknown packet: drop everything */
+		if (unlikely(!pkt_len))
+			goto out_flush_buf;
+
+		/* full packet: process */
+		if (likely((pkt_len > 0) && (pkt_len <= buf_len - pos))) {
+			usbtouch_process_pkt(usbtouch, regs, buffer + pos, pkt_len);
+		} else {
+			/* incomplete packet: save in buffer */
+			memcpy(usbtouch->buffer, buffer + pos, buf_len - pos);
+			usbtouch->buf_len = buf_len - pos;
+			return;
+		}
+		pos += pkt_len;
+	}
+
+out_flush_buf:
+	usbtouch->buf_len = 0;
+	return;
+}
+#endif
+
+
 static void usbtouch_irq(struct urb *urb, struct pt_regs *regs)
 {
 	struct usbtouch_usb *usbtouch = urb->context;
@@ -452,7 +574,7 @@ static int usbtouch_probe(struct usb_interface *intf,
 	struct usb_endpoint_descriptor *endpoint;
 	struct usb_device *udev = interface_to_usbdev(intf);
 	struct usbtouch_device_info *type;
-	int err;
+	int err = -ENOMEM;
 
 	interface = intf->cur_altsetting;
 	endpoint = &interface->endpoint[0].desc;
@@ -526,6 +648,7 @@ static int usbtouch_probe(struct usb_interface *intf,
 			 usbtouch->data, type->rept_size,
 			 usbtouch_irq, usbtouch, endpoint->bInterval);
 
+	usbtouch->irq->dev = usbtouch->udev;
 	usbtouch->irq->transfer_dma = usbtouch->data_dma;
 	usbtouch->irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 
@@ -553,7 +676,7 @@ out_free_buffers:
 out_free:
 	input_free_device(input_dev);
 	kfree(usbtouch);
-	return -ENOMEM;
+	return err;
 }
 
 static void usbtouch_disconnect(struct usb_interface *intf)
-- 
cgit v1.2.1


From a009b75aa0ed55f0bc473c8a3b3e872cbca807ec Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Tue, 25 Jul 2006 16:58:30 -0300
Subject: USB: pl2303: Removes unneeded goto.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 65e4d046951a..54c32ccce626 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -824,16 +824,13 @@ static void pl2303_update_line_status(struct usb_serial_port *port,
 	}
 
 	if (actual_length < length)
-		goto exit;
+		return;
 
         /* Save off the uart status for others to look at */
 	spin_lock_irqsave(&priv->lock, flags);
 	priv->line_status = data[status_idx];
 	spin_unlock_irqrestore(&priv->lock, flags);
 	wake_up_interruptible (&priv->delta_msr_wait);
-
-exit:
-	return;
 }
 
 static void pl2303_read_int_callback (struct urb *urb, struct pt_regs *regs)
-- 
cgit v1.2.1


From 372db8a780f63368c6960a167b7a19aad776d704 Mon Sep 17 00:00:00 2001
From: Thiago Galesi <thiagogalesi@gmail.com>
Date: Mon, 31 Jul 2006 15:39:27 -0300
Subject: USB: pl2303: remove 80-columns limit violations in pl2303 driver

Fixes several lines that overrun 80 columns in Prolific pl2303 driver
and cleans up some space usages in the function calls.

Signed-off-by: Thiago Galesi <thiagogalesi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 252 +++++++++++++++++++++-----------------------
 1 file changed, 123 insertions(+), 129 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 54c32ccce626..90d0e7935665 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -84,7 +84,7 @@ static struct usb_device_id id_table [] = {
 	{ }					/* Terminating entry */
 };
 
-MODULE_DEVICE_TABLE (usb, id_table);
+MODULE_DEVICE_TABLE(usb, id_table);
 
 static struct usb_driver pl2303_driver = {
 	.name =		"pl2303",
@@ -205,7 +205,7 @@ struct pl2303_private {
 };
 
 
-static int pl2303_startup (struct usb_serial *serial)
+static int pl2303_startup(struct usb_serial *serial)
 {
 	struct pl2303_private *priv;
 	enum pl2303_type type = type_0;
@@ -247,18 +247,19 @@ cleanup:
 	return -ENOMEM;
 }
 
-static int set_control_lines (struct usb_device *dev, u8 value)
+static int set_control_lines(struct usb_device *dev, u8 value)
 {
 	int retval;
 	
-	retval = usb_control_msg (dev, usb_sndctrlpipe (dev, 0),
-				  SET_CONTROL_REQUEST, SET_CONTROL_REQUEST_TYPE,
-				  value, 0, NULL, 0, 100);
+	retval = usb_control_msg(dev, usb_sndctrlpipe(dev, 0),
+				 SET_CONTROL_REQUEST, SET_CONTROL_REQUEST_TYPE,
+				 value, 0, NULL, 0, 100);
 	dbg("%s - value = %d, retval = %d", __FUNCTION__, value, retval);
 	return retval;
 }
 
-static int pl2303_write (struct usb_serial_port *port,  const unsigned char *buf, int count)
+static int pl2303_write(struct usb_serial_port *port, const unsigned char *buf,
+			int count)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -293,7 +294,7 @@ static void pl2303_send(struct usb_serial_port *port)
 	}
 
 	count = pl2303_buf_get(priv->buf, port->write_urb->transfer_buffer,
-		port->bulk_out_size);
+			       port->bulk_out_size);
 
 	if (count == 0) {
 		spin_unlock_irqrestore(&priv->lock, flags);
@@ -304,13 +305,15 @@ static void pl2303_send(struct usb_serial_port *port)
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, count, port->write_urb->transfer_buffer);
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, count,
+			      port->write_urb->transfer_buffer);
 
 	port->write_urb->transfer_buffer_length = count;
 	port->write_urb->dev = port->serial->dev;
-	result = usb_submit_urb (port->write_urb, GFP_ATOMIC);
+	result = usb_submit_urb(port->write_urb, GFP_ATOMIC);
 	if (result) {
-		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __FUNCTION__, result);
+		dev_err(&port->dev, "%s - failed submitting write urb,"
+			" error %d\n", __FUNCTION__, result);
 		priv->write_urb_in_use = 0;
 		// TODO: reschedule pl2303_send
 	}
@@ -350,7 +353,8 @@ static int pl2303_chars_in_buffer(struct usb_serial_port *port)
 	return chars;
 }
 
-static void pl2303_set_termios (struct usb_serial_port *port, struct termios *old_termios)
+static void pl2303_set_termios(struct usb_serial_port *port,
+			       struct termios *old_termios)
 {
 	struct usb_serial *serial = port->serial;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -371,7 +375,8 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 	spin_lock_irqsave(&priv->lock, flags);
 	if (!priv->termios_initialized) {
 		*(port->tty->termios) = tty_std_termios;
-		port->tty->termios->c_cflag = B9600 | CS8 | CREAD | HUPCL | CLOCAL;
+		port->tty->termios->c_cflag = B9600 | CS8 | CREAD |
+					      HUPCL | CLOCAL;
 		priv->termios_initialized = 1;
 	}
 	spin_unlock_irqrestore(&priv->lock, flags);
@@ -380,24 +385,24 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 	/* check that they really want us to change something */
 	if (old_termios) {
 		if ((cflag == old_termios->c_cflag) &&
-		    (RELEVANT_IFLAG(port->tty->termios->c_iflag) == RELEVANT_IFLAG(old_termios->c_iflag))) {
-		    dbg("%s - nothing to change...", __FUNCTION__);
-		    return;
+		    (RELEVANT_IFLAG(port->tty->termios->c_iflag) ==
+		     RELEVANT_IFLAG(old_termios->c_iflag))) {
+			dbg("%s - nothing to change...", __FUNCTION__);
+			return;
 		}
 	}
 
-	buf = kzalloc (7, GFP_KERNEL);
+	buf = kzalloc(7, GFP_KERNEL);
 	if (!buf) {
 		dev_err(&port->dev, "%s - out of memory.\n", __FUNCTION__);
 		return;
 	}
-	
-	i = usb_control_msg (serial->dev, usb_rcvctrlpipe (serial->dev, 0),
-			     GET_LINE_REQUEST, GET_LINE_REQUEST_TYPE,
-			     0, 0, buf, 7, 100);
-	dbg ("0xa1:0x21:0:0  %d - %x %x %x %x %x %x %x", i,
-	     buf[0], buf[1], buf[2], buf[3], buf[4], buf[5], buf[6]);
 
+	i = usb_control_msg(serial->dev, usb_rcvctrlpipe(serial->dev, 0),
+			    GET_LINE_REQUEST, GET_LINE_REQUEST_TYPE,
+			    0, 0, buf, 7, 100);
+	dbg("0xa1:0x21:0:0  %d - %x %x %x %x %x %x %x", i,
+	    buf[0], buf[1], buf[2], buf[3], buf[4], buf[5], buf[6]);
 
 	if (cflag & CSIZE) {
 		switch (cflag & CSIZE) {
@@ -429,7 +434,8 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 		case B230400:	baud = 230400;	break;
 		case B460800:	baud = 460800;	break;
 		default:
-			dev_err(&port->dev, "pl2303 driver does not support the baudrate requested (fix it)\n");
+			dev_err(&port->dev, "pl2303 driver does not support"
+				" the baudrate requested (fix it)\n");
 			break;
 	}
 	dbg("%s - baud = %d", __FUNCTION__, baud);
@@ -469,10 +475,10 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 		dbg("%s - parity = none", __FUNCTION__);
 	}
 
-	i = usb_control_msg (serial->dev, usb_sndctrlpipe (serial->dev, 0),
-			     SET_LINE_REQUEST, SET_LINE_REQUEST_TYPE, 
-			     0, 0, buf, 7, 100);
-	dbg ("0x21:0x20:0:0  %d", i);
+	i = usb_control_msg(serial->dev, usb_sndctrlpipe(serial->dev, 0),
+			    SET_LINE_REQUEST, SET_LINE_REQUEST_TYPE,
+			    0, 0, buf, 7, 100);
+	dbg("0x21:0x20:0:0  %d", i);
 
 	/* change control lines if we are switching to or from B0 */
 	spin_lock_irqsave(&priv->lock, flags);
@@ -488,13 +494,13 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 	} else {
 		spin_unlock_irqrestore(&priv->lock, flags);
 	}
-	
+
 	buf[0] = buf[1] = buf[2] = buf[3] = buf[4] = buf[5] = buf[6] = 0;
 
-	i = usb_control_msg (serial->dev, usb_rcvctrlpipe (serial->dev, 0),
-			     GET_LINE_REQUEST, GET_LINE_REQUEST_TYPE,
-			     0, 0, buf, 7, 100);
-	dbg ("0xa1:0x21:0:0  %d - %x %x %x %x %x %x %x", i,
+	i = usb_control_msg(serial->dev, usb_rcvctrlpipe(serial->dev, 0),
+			    GET_LINE_REQUEST, GET_LINE_REQUEST_TYPE,
+			    0, 0, buf, 7, 100);
+	dbg("0xa1:0x21:0:0  %d - %x %x %x %x %x %x %x", i,
 	     buf[0], buf[1], buf[2], buf[3], buf[4], buf[5], buf[6]);
 
 	if (cflag & CRTSCTS) {
@@ -503,18 +509,18 @@ static void pl2303_set_termios (struct usb_serial_port *port, struct termios *ol
 			index = 0x61;
 		else
 			index = 0x41;
-		i = usb_control_msg(serial->dev, 
+		i = usb_control_msg(serial->dev,
 				    usb_sndctrlpipe(serial->dev, 0),
 				    VENDOR_WRITE_REQUEST,
 				    VENDOR_WRITE_REQUEST_TYPE,
 				    0x0, index, NULL, 0, 100);
-		dbg ("0x40:0x1:0x0:0x%x  %d", index, i);
+		dbg("0x40:0x1:0x0:0x%x  %d", index, i);
 	}
 
-	kfree (buf);
+	kfree(buf);
 }
 
-static int pl2303_open (struct usb_serial_port *port, struct file *filp)
+static int pl2303_open(struct usb_serial_port *port, struct file *filp)
 {
 	struct termios tmp_termios;
 	struct usb_serial *serial = port->serial;
@@ -568,33 +574,34 @@ static int pl2303_open (struct usb_serial_port *port, struct file *filp)
 
 	/* Setup termios */
 	if (port->tty) {
-		pl2303_set_termios (port, &tmp_termios);
+		pl2303_set_termios(port, &tmp_termios);
 	}
 
 	//FIXME: need to assert RTS and DTR if CRTSCTS off
 
 	dbg("%s - submitting read urb", __FUNCTION__);
 	port->read_urb->dev = serial->dev;
-	result = usb_submit_urb (port->read_urb, GFP_KERNEL);
+	result = usb_submit_urb(port->read_urb, GFP_KERNEL);
 	if (result) {
-		dev_err(&port->dev, "%s - failed submitting read urb, error %d\n", __FUNCTION__, result);
-		pl2303_close (port, NULL);
+		dev_err(&port->dev, "%s - failed submitting read urb,"
+			" error %d\n", __FUNCTION__, result);
+		pl2303_close(port, NULL);
 		return -EPROTO;
 	}
 
 	dbg("%s - submitting interrupt urb", __FUNCTION__);
 	port->interrupt_in_urb->dev = serial->dev;
-	result = usb_submit_urb (port->interrupt_in_urb, GFP_KERNEL);
+	result = usb_submit_urb(port->interrupt_in_urb, GFP_KERNEL);
 	if (result) {
-		dev_err(&port->dev, "%s - failed submitting interrupt urb, error %d\n", __FUNCTION__, result);
-		pl2303_close (port, NULL);
+		dev_err(&port->dev, "%s - failed submitting interrupt urb,"
+			" error %d\n", __FUNCTION__, result);
+		pl2303_close(port, NULL);
 		return -EPROTO;
 	}
 	return 0;
 }
 
-
-static void pl2303_close (struct usb_serial_port *port, struct file *filp)
+static void pl2303_close(struct usb_serial_port *port, struct file *filp)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -612,9 +619,9 @@ static void pl2303_close (struct usb_serial_port *port, struct file *filp)
 	add_wait_queue(&port->tty->write_wait, &wait);
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
-		if (pl2303_buf_data_avail(priv->buf) == 0
-		|| timeout == 0 || signal_pending(current)
-		|| !usb_get_intfdata(port->serial->interface))	/* disconnect */
+		if (pl2303_buf_data_avail(priv->buf) == 0 ||
+		    timeout == 0 || signal_pending(current) ||
+		    !usb_get_intfdata(port->serial->interface))	/* disconnect */
 			break;
 		spin_unlock_irqrestore(&priv->lock, flags);
 		timeout = schedule_timeout(timeout);
@@ -652,14 +659,14 @@ static void pl2303_close (struct usb_serial_port *port, struct file *filp)
 			/* drop DTR and RTS */
 			spin_lock_irqsave(&priv->lock, flags);
 			priv->line_control = 0;
-			spin_unlock_irqrestore (&priv->lock, flags);
-			set_control_lines (port->serial->dev, 0);
+			spin_unlock_irqrestore(&priv->lock, flags);
+			set_control_lines(port->serial->dev, 0);
 		}
 	}
 }
 
-static int pl2303_tiocmset (struct usb_serial_port *port, struct file *file,
-			    unsigned int set, unsigned int clear)
+static int pl2303_tiocmset(struct usb_serial_port *port, struct file *file,
+			   unsigned int set, unsigned int clear)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -668,7 +675,7 @@ static int pl2303_tiocmset (struct usb_serial_port *port, struct file *file,
 	if (!usb_get_intfdata(port->serial->interface))
 		return -ENODEV;
 
-	spin_lock_irqsave (&priv->lock, flags);
+	spin_lock_irqsave(&priv->lock, flags);
 	if (set & TIOCM_RTS)
 		priv->line_control |= CONTROL_RTS;
 	if (set & TIOCM_DTR)
@@ -678,12 +685,12 @@ static int pl2303_tiocmset (struct usb_serial_port *port, struct file *file,
 	if (clear & TIOCM_DTR)
 		priv->line_control &= ~CONTROL_DTR;
 	control = priv->line_control;
-	spin_unlock_irqrestore (&priv->lock, flags);
+	spin_unlock_irqrestore(&priv->lock, flags);
 
-	return set_control_lines (port->serial->dev, control);
+	return set_control_lines(port->serial->dev, control);
 }
 
-static int pl2303_tiocmget (struct usb_serial_port *port, struct file *file)
+static int pl2303_tiocmget(struct usb_serial_port *port, struct file *file)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
@@ -696,10 +703,10 @@ static int pl2303_tiocmget (struct usb_serial_port *port, struct file *file)
 	if (!usb_get_intfdata(port->serial->interface))
 		return -ENODEV;
 
-	spin_lock_irqsave (&priv->lock, flags);
+	spin_lock_irqsave(&priv->lock, flags);
 	mcr = priv->line_control;
 	status = priv->line_status;
-	spin_unlock_irqrestore (&priv->lock, flags);
+	spin_unlock_irqrestore(&priv->lock, flags);
 
 	result = ((mcr & CONTROL_DTR)		? TIOCM_DTR : 0)
 		  | ((mcr & CONTROL_RTS)	? TIOCM_RTS : 0)
@@ -721,22 +728,22 @@ static int wait_modem_info(struct usb_serial_port *port, unsigned int arg)
 	unsigned int status;
 	unsigned int changed;
 
-	spin_lock_irqsave (&priv->lock, flags);
+	spin_lock_irqsave(&priv->lock, flags);
 	prevstatus = priv->line_status;
-	spin_unlock_irqrestore (&priv->lock, flags);
+	spin_unlock_irqrestore(&priv->lock, flags);
 
 	while (1) {
 		interruptible_sleep_on(&priv->delta_msr_wait);
 		/* see if a signal did it */
 		if (signal_pending(current))
 			return -ERESTARTSYS;
-		
-		spin_lock_irqsave (&priv->lock, flags);
+
+		spin_lock_irqsave(&priv->lock, flags);
 		status = priv->line_status;
-		spin_unlock_irqrestore (&priv->lock, flags);
-		
+		spin_unlock_irqrestore(&priv->lock, flags);
+
 		changed=prevstatus^status;
-		
+
 		if (((arg & TIOCM_RNG) && (changed & UART_RING)) ||
 		    ((arg & TIOCM_DSR) && (changed & UART_DSR)) ||
 		    ((arg & TIOCM_CD)  && (changed & UART_DCD)) ||
@@ -749,7 +756,8 @@ static int wait_modem_info(struct usb_serial_port *port, unsigned int arg)
 	return 0;
 }
 
-static int pl2303_ioctl (struct usb_serial_port *port, struct file *file, unsigned int cmd, unsigned long arg)
+static int pl2303_ioctl(struct usb_serial_port *port, struct file *file,
+			unsigned int cmd, unsigned long arg)
 {
 	dbg("%s (%d) cmd = 0x%04x", __FUNCTION__, port->number, cmd);
 
@@ -766,7 +774,7 @@ static int pl2303_ioctl (struct usb_serial_port *port, struct file *file, unsign
 	return -ENOIOCTLCMD;
 }
 
-static void pl2303_break_ctl (struct usb_serial_port *port, int break_state)
+static void pl2303_break_ctl(struct usb_serial_port *port, int break_state)
 {
 	struct usb_serial *serial = port->serial;
 	u16 state;
@@ -780,15 +788,14 @@ static void pl2303_break_ctl (struct usb_serial_port *port, int break_state)
 		state = BREAK_ON;
 	dbg("%s - turning break %s", __FUNCTION__, state==BREAK_OFF ? "off" : "on");
 
-	result = usb_control_msg (serial->dev, usb_sndctrlpipe (serial->dev, 0),
-				  BREAK_REQUEST, BREAK_REQUEST_TYPE, state, 
-				  0, NULL, 0, 100);
+	result = usb_control_msg(serial->dev, usb_sndctrlpipe(serial->dev, 0),
+				 BREAK_REQUEST, BREAK_REQUEST_TYPE, state,
+				 0, NULL, 0, 100);
 	if (result)
 		dbg("%s - error sending break = %d", __FUNCTION__, result);
 }
 
-
-static void pl2303_shutdown (struct usb_serial *serial)
+static void pl2303_shutdown(struct usb_serial *serial)
 {
 	int i;
 	struct pl2303_private *priv;
@@ -802,7 +809,7 @@ static void pl2303_shutdown (struct usb_serial *serial)
 			kfree(priv);
 			usb_set_serial_port_data(serial->port[i], NULL);
 		}
-	}		
+	}
 }
 
 static void pl2303_update_line_status(struct usb_serial_port *port,
@@ -830,10 +837,10 @@ static void pl2303_update_line_status(struct usb_serial_port *port,
 	spin_lock_irqsave(&priv->lock, flags);
 	priv->line_status = data[status_idx];
 	spin_unlock_irqrestore(&priv->lock, flags);
-	wake_up_interruptible (&priv->delta_msr_wait);
+	wake_up_interruptible(&priv->delta_msr_wait);
 }
 
-static void pl2303_read_int_callback (struct urb *urb, struct pt_regs *regs)
+static void pl2303_read_int_callback(struct urb *urb, struct pt_regs *regs)
 {
 	struct usb_serial_port *port = (struct usb_serial_port *) urb->context;
 	unsigned char *data = urb->transfer_buffer;
@@ -850,25 +857,29 @@ static void pl2303_read_int_callback (struct urb *urb, struct pt_regs *regs)
 	case -ENOENT:
 	case -ESHUTDOWN:
 		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__,
+		    urb->status);
 		return;
 	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
+		dbg("%s - nonzero urb status received: %d", __FUNCTION__,
+		    urb->status);
 		goto exit;
 	}
 
-	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, urb->actual_length, urb->transfer_buffer);
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__,
+			      urb->actual_length, urb->transfer_buffer);
+
 	pl2303_update_line_status(port, data, actual_length);
 
 exit:
-	status = usb_submit_urb (urb, GFP_ATOMIC);
+	status = usb_submit_urb(urb, GFP_ATOMIC);
 	if (status)
-		dev_err(&urb->dev->dev, "%s - usb_submit_urb failed with result %d\n",
+		dev_err(&urb->dev->dev,
+			"%s - usb_submit_urb failed with result %d\n",
 			__FUNCTION__, status);
 }
 
-
-static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
+static void pl2303_read_bulk_callback(struct urb *urb, struct pt_regs *regs)
 {
 	struct usb_serial_port *port = (struct usb_serial_port *) urb->context;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -889,20 +900,25 @@ static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 			return;
 		}
 		if (urb->status == -EPROTO) {
-			/* PL2303 mysteriously fails with -EPROTO reschedule the read */
-			dbg("%s - caught -EPROTO, resubmitting the urb", __FUNCTION__);
+			/* PL2303 mysteriously fails with -EPROTO reschedule
+			 * the read */
+			dbg("%s - caught -EPROTO, resubmitting the urb",
+			    __FUNCTION__);
 			urb->status = 0;
 			urb->dev = port->serial->dev;
 			result = usb_submit_urb(urb, GFP_ATOMIC);
 			if (result)
-				dev_err(&urb->dev->dev, "%s - failed resubmitting read urb, error %d\n", __FUNCTION__, result);
+				dev_err(&urb->dev->dev, "%s - failed"
+					" resubmitting read urb, error %d\n",
+					__FUNCTION__, result);
 			return;
 		}
 		dbg("%s - unable to handle the error, exiting.", __FUNCTION__);
 		return;
 	}
 
-	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, urb->actual_length, data);
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__,
+			      urb->actual_length, data);
 
 	/* get tty_flag from status */
 	tty_flag = TTY_NORMAL;
@@ -911,7 +927,7 @@ static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 	status = priv->line_status;
 	priv->line_status &= ~UART_STATE_TRANSIENT_MASK;
 	spin_unlock_irqrestore(&priv->lock, flags);
-	wake_up_interruptible (&priv->delta_msr_wait);
+	wake_up_interruptible(&priv->delta_msr_wait);
 
 	/* break takes precedence over parity, */
 	/* which takes precedence over framing errors */
@@ -930,8 +946,8 @@ static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 		if (status & UART_OVERRUN_ERROR)
 			tty_insert_flip_char(tty, 0, TTY_OVERRUN);
 		for (i = 0; i < urb->actual_length; ++i)
-			tty_insert_flip_char (tty, data[i], tty_flag);
-		tty_flip_buffer_push (tty);
+			tty_insert_flip_char(tty, data[i], tty_flag);
+		tty_flip_buffer_push(tty);
 	}
 
 	/* Schedule the next read _if_ we are still open */
@@ -939,15 +955,14 @@ static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 		urb->dev = port->serial->dev;
 		result = usb_submit_urb(urb, GFP_ATOMIC);
 		if (result)
-			dev_err(&urb->dev->dev, "%s - failed resubmitting read urb, error %d\n", __FUNCTION__, result);
+			dev_err(&urb->dev->dev, "%s - failed resubmitting"
+				" read urb, error %d\n", __FUNCTION__, result);
 	}
 
 	return;
 }
 
-
-
-static void pl2303_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
+static void pl2303_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
 {
 	struct usb_serial_port *port = (struct usb_serial_port *) urb->context;
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -963,18 +978,21 @@ static void pl2303_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
 	case -ENOENT:
 	case -ESHUTDOWN:
 		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__,
+		    urb->status);
 		priv->write_urb_in_use = 0;
 		return;
 	default:
 		/* error in the urb, so we have to resubmit it */
 		dbg("%s - Overflow in write", __FUNCTION__);
-		dbg("%s - nonzero write bulk status received: %d", __FUNCTION__, urb->status);
+		dbg("%s - nonzero write bulk status received: %d", __FUNCTION__,
+		    urb->status);
 		port->write_urb->transfer_buffer_length = 1;
 		port->write_urb->dev = port->serial->dev;
-		result = usb_submit_urb (port->write_urb, GFP_ATOMIC);
+		result = usb_submit_urb(port->write_urb, GFP_ATOMIC);
 		if (result)
-			dev_err(&urb->dev->dev, "%s - failed resubmitting write urb, error %d\n", __FUNCTION__, result);
+			dev_err(&urb->dev->dev, "%s - failed resubmitting write"
+				" urb, error %d\n", __FUNCTION__, result);
 		else
 			return;
 	}
@@ -985,19 +1003,15 @@ static void pl2303_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
 	pl2303_send(port);
 }
 
-
 /*
  * pl2303_buf_alloc
  *
  * Allocate a circular buffer and all associated memory.
  */
-
 static struct pl2303_buf *pl2303_buf_alloc(unsigned int size)
 {
-
 	struct pl2303_buf *pb;
 
-
 	if (size == 0)
 		return NULL;
 
@@ -1015,16 +1029,13 @@ static struct pl2303_buf *pl2303_buf_alloc(unsigned int size)
 	pb->buf_get = pb->buf_put = pb->buf_buf;
 
 	return pb;
-
 }
 
-
 /*
  * pl2303_buf_free
  *
  * Free the buffer and all associated memory.
  */
-
 static void pl2303_buf_free(struct pl2303_buf *pb)
 {
 	if (pb) {
@@ -1033,13 +1044,11 @@ static void pl2303_buf_free(struct pl2303_buf *pb)
 	}
 }
 
-
 /*
  * pl2303_buf_clear
  *
  * Clear out all data in the circular buffer.
  */
-
 static void pl2303_buf_clear(struct pl2303_buf *pb)
 {
 	if (pb != NULL)
@@ -1047,14 +1056,12 @@ static void pl2303_buf_clear(struct pl2303_buf *pb)
 		/* equivalent to a get of all data available */
 }
 
-
 /*
  * pl2303_buf_data_avail
  *
  * Return the number of bytes of data available in the circular
  * buffer.
  */
-
 static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
 {
 	if (pb != NULL)
@@ -1063,14 +1070,12 @@ static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
 		return 0;
 }
 
-
 /*
  * pl2303_buf_space_avail
  *
  * Return the number of bytes of space available in the circular
  * buffer.
  */
-
 static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
 {
 	if (pb != NULL)
@@ -1079,7 +1084,6 @@ static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
 		return 0;
 }
 
-
 /*
  * pl2303_buf_put
  *
@@ -1088,14 +1092,11 @@ static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
  *
  * Return the number of bytes copied.
  */
-
 static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
-	unsigned int count)
+				   unsigned int count)
 {
-
 	unsigned int len;
 
-
 	if (pb == NULL)
 		return 0;
 
@@ -1120,10 +1121,8 @@ static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
 	}
 
 	return count;
-
 }
 
-
 /*
  * pl2303_buf_get
  *
@@ -1132,14 +1131,11 @@ static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
  *
  * Return the number of bytes copied.
  */
-
 static unsigned int pl2303_buf_get(struct pl2303_buf *pb, char *buf,
-	unsigned int count)
+				   unsigned int count)
 {
-
 	unsigned int len;
 
-
 	if (pb == NULL)
 		return 0;
 
@@ -1164,12 +1160,12 @@ static unsigned int pl2303_buf_get(struct pl2303_buf *pb, char *buf,
 	}
 
 	return count;
-
 }
 
-static int __init pl2303_init (void)
+static int __init pl2303_init(void)
 {
 	int retval;
+
 	retval = usb_serial_register(&pl2303_device);
 	if (retval)
 		goto failed_usb_serial_register;
@@ -1184,14 +1180,12 @@ failed_usb_serial_register:
 	return retval;
 }
 
-
-static void __exit pl2303_exit (void)
+static void __exit pl2303_exit(void)
 {
-	usb_deregister (&pl2303_driver);
-	usb_serial_deregister (&pl2303_device);
+	usb_deregister(&pl2303_driver);
+	usb_serial_deregister(&pl2303_device);
 }
 
-
 module_init(pl2303_init);
 module_exit(pl2303_exit);
 
-- 
cgit v1.2.1


From 10a18cd1a31e34d123d5ae4736a3393432352750 Mon Sep 17 00:00:00 2001
From: Thiago Galesi <thiagogalesi@gmail.com>
Date: Sat, 29 Jul 2006 10:45:43 -0300
Subject: USB: pl2303: cosmetic changes to pl2303_buf_{clear, data_avail}

Changes the functions pl2303_buf_clear and pl2303_buf_data_avail for
the purpose of keeping them under the 80 column limit, making them
more similar to similar functions and making then simpler.

Signed-off-by: Thiago Galesi <thiagogalesi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 90d0e7935665..0cd42bf4c6c8 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -1064,10 +1064,10 @@ static void pl2303_buf_clear(struct pl2303_buf *pb)
  */
 static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
 {
-	if (pb != NULL)
-		return ((pb->buf_size + pb->buf_put - pb->buf_get) % pb->buf_size);
-	else
+	if (pb == NULL)
 		return 0;
+
+	return ((pb->buf_size + pb->buf_put - pb->buf_get) % pb->buf_size);
 }
 
 /*
@@ -1078,10 +1078,10 @@ static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
  */
 static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
 {
-	if (pb != NULL)
-		return ((pb->buf_size + pb->buf_get - pb->buf_put - 1) % pb->buf_size);
-	else
+	if (pb == NULL)
 		return 0;
+
+	return ((pb->buf_size + pb->buf_get - pb->buf_put - 1) % pb->buf_size);
 }
 
 /*
-- 
cgit v1.2.1


From 572d3138eb0cf17a2bf36944cc1d2c753578862e Mon Sep 17 00:00:00 2001
From: Thiago Galesi <thiagogalesi@gmail.com>
Date: Sat, 29 Jul 2006 10:46:37 -0300
Subject: USB: pl2303: reduce number of prototypes

Reduce number of needed prototypes in Prolific pl2303 driver

Signed-off-by: Thiago Galesi <thiagogalesi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 574 +++++++++++++++++++++-----------------------
 1 file changed, 271 insertions(+), 303 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 0cd42bf4c6c8..e2d8f9b94eed 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -127,65 +127,6 @@ static struct usb_driver pl2303_driver = {
 #define UART_OVERRUN_ERROR		0x40
 #define UART_CTS			0x80
 
-/* function prototypes for a PL2303 serial converter */
-static int pl2303_open (struct usb_serial_port *port, struct file *filp);
-static void pl2303_close (struct usb_serial_port *port, struct file *filp);
-static void pl2303_set_termios (struct usb_serial_port *port,
-				struct termios *old);
-static int pl2303_ioctl (struct usb_serial_port *port, struct file *file,
-			 unsigned int cmd, unsigned long arg);
-static void pl2303_read_int_callback (struct urb *urb, struct pt_regs *regs);
-static void pl2303_read_bulk_callback (struct urb *urb, struct pt_regs *regs);
-static void pl2303_write_bulk_callback (struct urb *urb, struct pt_regs *regs);
-static int pl2303_write (struct usb_serial_port *port,
-			 const unsigned char *buf, int count);
-static void pl2303_send (struct usb_serial_port *port);
-static int pl2303_write_room(struct usb_serial_port *port);
-static int pl2303_chars_in_buffer(struct usb_serial_port *port);
-static void pl2303_break_ctl(struct usb_serial_port *port,int break_state);
-static int pl2303_tiocmget (struct usb_serial_port *port, struct file *file);
-static int pl2303_tiocmset (struct usb_serial_port *port, struct file *file,
-			    unsigned int set, unsigned int clear);
-static int pl2303_startup (struct usb_serial *serial);
-static void pl2303_shutdown (struct usb_serial *serial);
-static struct pl2303_buf *pl2303_buf_alloc(unsigned int size);
-static void pl2303_buf_free(struct pl2303_buf *pb);
-static void pl2303_buf_clear(struct pl2303_buf *pb);
-static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb);
-static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb);
-static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
-	unsigned int count);
-static unsigned int pl2303_buf_get(struct pl2303_buf *pb, char *buf,
-	unsigned int count);
-
-
-/* All of the device info needed for the PL2303 SIO serial converter */
-static struct usb_serial_driver pl2303_device = {
-	.driver = {
-		.owner =	THIS_MODULE,
-		.name =		"pl2303",
-	},
-	.id_table =		id_table,
-	.num_interrupt_in =	NUM_DONT_CARE,
-	.num_bulk_in =		1,
-	.num_bulk_out =		1,
-	.num_ports =		1,
-	.open =			pl2303_open,
-	.close =		pl2303_close,
-	.write =		pl2303_write,
-	.ioctl =		pl2303_ioctl,
-	.break_ctl =		pl2303_break_ctl,
-	.set_termios =		pl2303_set_termios,
-	.tiocmget =		pl2303_tiocmget,
-	.tiocmset =		pl2303_tiocmset,
-	.read_bulk_callback =	pl2303_read_bulk_callback,
-	.read_int_callback =	pl2303_read_int_callback,
-	.write_bulk_callback =	pl2303_write_bulk_callback,
-	.write_room =		pl2303_write_room,
-	.chars_in_buffer =	pl2303_chars_in_buffer,
-	.attach =		pl2303_startup,
-	.shutdown =		pl2303_shutdown,
-};
 
 enum pl2303_type {
 	type_0,		/* don't know the difference between type 0 and */
@@ -204,6 +145,164 @@ struct pl2303_private {
 	enum pl2303_type type;
 };
 
+/*
+ * pl2303_buf_alloc
+ *
+ * Allocate a circular buffer and all associated memory.
+ */
+static struct pl2303_buf *pl2303_buf_alloc(unsigned int size)
+{
+	struct pl2303_buf *pb;
+
+	if (size == 0)
+		return NULL;
+
+	pb = (struct pl2303_buf *)kmalloc(sizeof(struct pl2303_buf), GFP_KERNEL);
+	if (pb == NULL)
+		return NULL;
+
+	pb->buf_buf = kmalloc(size, GFP_KERNEL);
+	if (pb->buf_buf == NULL) {
+		kfree(pb);
+		return NULL;
+	}
+
+	pb->buf_size = size;
+	pb->buf_get = pb->buf_put = pb->buf_buf;
+
+	return pb;
+}
+
+/*
+ * pl2303_buf_free
+ *
+ * Free the buffer and all associated memory.
+ */
+static void pl2303_buf_free(struct pl2303_buf *pb)
+{
+	if (pb) {
+		kfree(pb->buf_buf);
+		kfree(pb);
+	}
+}
+
+/*
+ * pl2303_buf_clear
+ *
+ * Clear out all data in the circular buffer.
+ */
+static void pl2303_buf_clear(struct pl2303_buf *pb)
+{
+	if (pb != NULL)
+		pb->buf_get = pb->buf_put;
+		/* equivalent to a get of all data available */
+}
+
+/*
+ * pl2303_buf_data_avail
+ *
+ * Return the number of bytes of data available in the circular
+ * buffer.
+ */
+static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
+{
+	if (pb == NULL)
+		return 0;
+
+	return ((pb->buf_size + pb->buf_put - pb->buf_get) % pb->buf_size);
+}
+
+/*
+ * pl2303_buf_space_avail
+ *
+ * Return the number of bytes of space available in the circular
+ * buffer.
+ */
+static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
+{
+	if (pb == NULL)
+		return 0;
+
+	return ((pb->buf_size + pb->buf_get - pb->buf_put - 1) % pb->buf_size);
+}
+
+/*
+ * pl2303_buf_put
+ *
+ * Copy data data from a user buffer and put it into the circular buffer.
+ * Restrict to the amount of space available.
+ *
+ * Return the number of bytes copied.
+ */
+static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
+				   unsigned int count)
+{
+	unsigned int len;
+
+	if (pb == NULL)
+		return 0;
+
+	len  = pl2303_buf_space_avail(pb);
+	if (count > len)
+		count = len;
+
+	if (count == 0)
+		return 0;
+
+	len = pb->buf_buf + pb->buf_size - pb->buf_put;
+	if (count > len) {
+		memcpy(pb->buf_put, buf, len);
+		memcpy(pb->buf_buf, buf+len, count - len);
+		pb->buf_put = pb->buf_buf + count - len;
+	} else {
+		memcpy(pb->buf_put, buf, count);
+		if (count < len)
+			pb->buf_put += count;
+		else /* count == len */
+			pb->buf_put = pb->buf_buf;
+	}
+
+	return count;
+}
+
+/*
+ * pl2303_buf_get
+ *
+ * Get data from the circular buffer and copy to the given buffer.
+ * Restrict to the amount of data available.
+ *
+ * Return the number of bytes copied.
+ */
+static unsigned int pl2303_buf_get(struct pl2303_buf *pb, char *buf,
+				   unsigned int count)
+{
+	unsigned int len;
+
+	if (pb == NULL)
+		return 0;
+
+	len = pl2303_buf_data_avail(pb);
+	if (count > len)
+		count = len;
+
+	if (count == 0)
+		return 0;
+
+	len = pb->buf_buf + pb->buf_size - pb->buf_get;
+	if (count > len) {
+		memcpy(buf, pb->buf_get, len);
+		memcpy(buf+len, pb->buf_buf, count - len);
+		pb->buf_get = pb->buf_buf + count - len;
+	} else {
+		memcpy(buf, pb->buf_get, count);
+		if (count < len)
+			pb->buf_get += count;
+		else /* count == len */
+			pb->buf_get = pb->buf_buf;
+	}
+
+	return count;
+}
 
 static int pl2303_startup(struct usb_serial *serial)
 {
@@ -258,26 +357,6 @@ static int set_control_lines(struct usb_device *dev, u8 value)
 	return retval;
 }
 
-static int pl2303_write(struct usb_serial_port *port, const unsigned char *buf,
-			int count)
-{
-	struct pl2303_private *priv = usb_get_serial_port_data(port);
-	unsigned long flags;
-
-	dbg("%s - port %d, %d bytes", __FUNCTION__, port->number, count);
-
-	if (!count)
-		return count;
-
-	spin_lock_irqsave(&priv->lock, flags);
-	count = pl2303_buf_put(priv->buf, buf, count);
-	spin_unlock_irqrestore(&priv->lock, flags);
-
-	pl2303_send(port);
-
-	return count;
-}
-
 static void pl2303_send(struct usb_serial_port *port)
 {
 	int count, result;
@@ -321,6 +400,26 @@ static void pl2303_send(struct usb_serial_port *port)
 	usb_serial_port_softint(port);
 }
 
+static int pl2303_write(struct usb_serial_port *port, const unsigned char *buf,
+			int count)
+{
+	struct pl2303_private *priv = usb_get_serial_port_data(port);
+	unsigned long flags;
+
+	dbg("%s - port %d, %d bytes", __FUNCTION__, port->number, count);
+
+	if (!count)
+		return count;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	count = pl2303_buf_put(priv->buf, buf, count);
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	pl2303_send(port);
+
+	return count;
+}
+
 static int pl2303_write_room(struct usb_serial_port *port)
 {
 	struct pl2303_private *priv = usb_get_serial_port_data(port);
@@ -516,8 +615,72 @@ static void pl2303_set_termios(struct usb_serial_port *port,
 				    0x0, index, NULL, 0, 100);
 		dbg("0x40:0x1:0x0:0x%x  %d", index, i);
 	}
-
-	kfree(buf);
+
+	kfree(buf);
+}
+
+static void pl2303_close(struct usb_serial_port *port, struct file *filp)
+{
+	struct pl2303_private *priv = usb_get_serial_port_data(port);
+	unsigned long flags;
+	unsigned int c_cflag;
+	int bps;
+	long timeout;
+	wait_queue_t wait;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	/* wait for data to drain from the buffer */
+	spin_lock_irqsave(&priv->lock, flags);
+	timeout = PL2303_CLOSING_WAIT;
+	init_waitqueue_entry(&wait, current);
+	add_wait_queue(&port->tty->write_wait, &wait);
+	for (;;) {
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (pl2303_buf_data_avail(priv->buf) == 0 ||
+		    timeout == 0 || signal_pending(current) ||
+		    !usb_get_intfdata(port->serial->interface))	/* disconnect */
+			break;
+		spin_unlock_irqrestore(&priv->lock, flags);
+		timeout = schedule_timeout(timeout);
+		spin_lock_irqsave(&priv->lock, flags);
+	}
+	set_current_state(TASK_RUNNING);
+	remove_wait_queue(&port->tty->write_wait, &wait);
+	/* clear out any remaining data in the buffer */
+	pl2303_buf_clear(priv->buf);
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	/* wait for characters to drain from the device */
+	/* (this is long enough for the entire 256 byte */
+	/* pl2303 hardware buffer to drain with no flow */
+	/* control for data rates of 1200 bps or more, */
+	/* for lower rates we should really know how much */
+	/* data is in the buffer to compute a delay */
+	/* that is not unnecessarily long) */
+	bps = tty_get_baud_rate(port->tty);
+	if (bps > 1200)
+		timeout = max((HZ*2560)/bps,HZ/10);
+	else
+		timeout = 2*HZ;
+	schedule_timeout_interruptible(timeout);
+
+	/* shutdown our urbs */
+	dbg("%s - shutting down urbs", __FUNCTION__);
+	usb_kill_urb(port->write_urb);
+	usb_kill_urb(port->read_urb);
+	usb_kill_urb(port->interrupt_in_urb);
+
+	if (port->tty) {
+		c_cflag = port->tty->termios->c_cflag;
+		if (c_cflag & HUPCL) {
+			/* drop DTR and RTS */
+			spin_lock_irqsave(&priv->lock, flags);
+			priv->line_control = 0;
+			spin_unlock_irqrestore(&priv->lock, flags);
+			set_control_lines(port->serial->dev, 0);
+		}
+	}
 }
 
 static int pl2303_open(struct usb_serial_port *port, struct file *filp)
@@ -601,70 +764,6 @@ static int pl2303_open(struct usb_serial_port *port, struct file *filp)
 	return 0;
 }
 
-static void pl2303_close(struct usb_serial_port *port, struct file *filp)
-{
-	struct pl2303_private *priv = usb_get_serial_port_data(port);
-	unsigned long flags;
-	unsigned int c_cflag;
-	int bps;
-	long timeout;
-	wait_queue_t wait;
-
-	dbg("%s - port %d", __FUNCTION__, port->number);
-
-	/* wait for data to drain from the buffer */
-	spin_lock_irqsave(&priv->lock, flags);
-	timeout = PL2303_CLOSING_WAIT;
-	init_waitqueue_entry(&wait, current);
-	add_wait_queue(&port->tty->write_wait, &wait);
-	for (;;) {
-		set_current_state(TASK_INTERRUPTIBLE);
-		if (pl2303_buf_data_avail(priv->buf) == 0 ||
-		    timeout == 0 || signal_pending(current) ||
-		    !usb_get_intfdata(port->serial->interface))	/* disconnect */
-			break;
-		spin_unlock_irqrestore(&priv->lock, flags);
-		timeout = schedule_timeout(timeout);
-		spin_lock_irqsave(&priv->lock, flags);
-	}
-	set_current_state(TASK_RUNNING);
-	remove_wait_queue(&port->tty->write_wait, &wait);
-	/* clear out any remaining data in the buffer */
-	pl2303_buf_clear(priv->buf);
-	spin_unlock_irqrestore(&priv->lock, flags);
-
-	/* wait for characters to drain from the device */
-	/* (this is long enough for the entire 256 byte */
-	/* pl2303 hardware buffer to drain with no flow */
-	/* control for data rates of 1200 bps or more, */
-	/* for lower rates we should really know how much */
-	/* data is in the buffer to compute a delay */
-	/* that is not unnecessarily long) */
-	bps = tty_get_baud_rate(port->tty);
-	if (bps > 1200)
-		timeout = max((HZ*2560)/bps,HZ/10);
-	else
-		timeout = 2*HZ;
-	schedule_timeout_interruptible(timeout);
-
-	/* shutdown our urbs */
-	dbg("%s - shutting down urbs", __FUNCTION__);
-	usb_kill_urb(port->write_urb);
-	usb_kill_urb(port->read_urb);
-	usb_kill_urb(port->interrupt_in_urb);
-
-	if (port->tty) {
-		c_cflag = port->tty->termios->c_cflag;
-		if (c_cflag & HUPCL) {
-			/* drop DTR and RTS */
-			spin_lock_irqsave(&priv->lock, flags);
-			priv->line_control = 0;
-			spin_unlock_irqrestore(&priv->lock, flags);
-			set_control_lines(port->serial->dev, 0);
-		}
-	}
-}
-
 static int pl2303_tiocmset(struct usb_serial_port *port, struct file *file,
 			   unsigned int set, unsigned int clear)
 {
@@ -1003,164 +1102,33 @@ static void pl2303_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
 	pl2303_send(port);
 }
 
-/*
- * pl2303_buf_alloc
- *
- * Allocate a circular buffer and all associated memory.
- */
-static struct pl2303_buf *pl2303_buf_alloc(unsigned int size)
-{
-	struct pl2303_buf *pb;
-
-	if (size == 0)
-		return NULL;
-
-	pb = (struct pl2303_buf *)kmalloc(sizeof(struct pl2303_buf), GFP_KERNEL);
-	if (pb == NULL)
-		return NULL;
-
-	pb->buf_buf = kmalloc(size, GFP_KERNEL);
-	if (pb->buf_buf == NULL) {
-		kfree(pb);
-		return NULL;
-	}
-
-	pb->buf_size = size;
-	pb->buf_get = pb->buf_put = pb->buf_buf;
-
-	return pb;
-}
-
-/*
- * pl2303_buf_free
- *
- * Free the buffer and all associated memory.
- */
-static void pl2303_buf_free(struct pl2303_buf *pb)
-{
-	if (pb) {
-		kfree(pb->buf_buf);
-		kfree(pb);
-	}
-}
-
-/*
- * pl2303_buf_clear
- *
- * Clear out all data in the circular buffer.
- */
-static void pl2303_buf_clear(struct pl2303_buf *pb)
-{
-	if (pb != NULL)
-		pb->buf_get = pb->buf_put;
-		/* equivalent to a get of all data available */
-}
-
-/*
- * pl2303_buf_data_avail
- *
- * Return the number of bytes of data available in the circular
- * buffer.
- */
-static unsigned int pl2303_buf_data_avail(struct pl2303_buf *pb)
-{
-	if (pb == NULL)
-		return 0;
-
-	return ((pb->buf_size + pb->buf_put - pb->buf_get) % pb->buf_size);
-}
-
-/*
- * pl2303_buf_space_avail
- *
- * Return the number of bytes of space available in the circular
- * buffer.
- */
-static unsigned int pl2303_buf_space_avail(struct pl2303_buf *pb)
-{
-	if (pb == NULL)
-		return 0;
-
-	return ((pb->buf_size + pb->buf_get - pb->buf_put - 1) % pb->buf_size);
-}
-
-/*
- * pl2303_buf_put
- *
- * Copy data data from a user buffer and put it into the circular buffer.
- * Restrict to the amount of space available.
- *
- * Return the number of bytes copied.
- */
-static unsigned int pl2303_buf_put(struct pl2303_buf *pb, const char *buf,
-				   unsigned int count)
-{
-	unsigned int len;
-
-	if (pb == NULL)
-		return 0;
-
-	len  = pl2303_buf_space_avail(pb);
-	if (count > len)
-		count = len;
-
-	if (count == 0)
-		return 0;
-
-	len = pb->buf_buf + pb->buf_size - pb->buf_put;
-	if (count > len) {
-		memcpy(pb->buf_put, buf, len);
-		memcpy(pb->buf_buf, buf+len, count - len);
-		pb->buf_put = pb->buf_buf + count - len;
-	} else {
-		memcpy(pb->buf_put, buf, count);
-		if (count < len)
-			pb->buf_put += count;
-		else /* count == len */
-			pb->buf_put = pb->buf_buf;
-	}
-
-	return count;
-}
-
-/*
- * pl2303_buf_get
- *
- * Get data from the circular buffer and copy to the given buffer.
- * Restrict to the amount of data available.
- *
- * Return the number of bytes copied.
- */
-static unsigned int pl2303_buf_get(struct pl2303_buf *pb, char *buf,
-				   unsigned int count)
-{
-	unsigned int len;
-
-	if (pb == NULL)
-		return 0;
-
-	len = pl2303_buf_data_avail(pb);
-	if (count > len)
-		count = len;
-
-	if (count == 0)
-		return 0;
-
-	len = pb->buf_buf + pb->buf_size - pb->buf_get;
-	if (count > len) {
-		memcpy(buf, pb->buf_get, len);
-		memcpy(buf+len, pb->buf_buf, count - len);
-		pb->buf_get = pb->buf_buf + count - len;
-	} else {
-		memcpy(buf, pb->buf_get, count);
-		if (count < len)
-			pb->buf_get += count;
-		else /* count == len */
-			pb->buf_get = pb->buf_buf;
-	}
-
-	return count;
-}
+/* All of the device info needed for the PL2303 SIO serial converter */
+static struct usb_serial_driver pl2303_device = {
+	.driver = {
+		.owner =	THIS_MODULE,
+		.name =		"pl2303",
+	},
+	.id_table =		id_table,
+	.num_interrupt_in =	NUM_DONT_CARE,
+	.num_bulk_in =		1,
+	.num_bulk_out =		1,
+	.num_ports =		1,
+	.open =			pl2303_open,
+	.close =		pl2303_close,
+	.write =		pl2303_write,
+	.ioctl =		pl2303_ioctl,
+	.break_ctl =		pl2303_break_ctl,
+	.set_termios =		pl2303_set_termios,
+	.tiocmget =		pl2303_tiocmget,
+	.tiocmset =		pl2303_tiocmset,
+	.read_bulk_callback =	pl2303_read_bulk_callback,
+	.read_int_callback =	pl2303_read_int_callback,
+	.write_bulk_callback =	pl2303_write_bulk_callback,
+	.write_room =		pl2303_write_room,
+	.chars_in_buffer =	pl2303_chars_in_buffer,
+	.attach =		pl2303_startup,
+	.shutdown =		pl2303_shutdown,
+};
 
 static int __init pl2303_init(void)
 {
-- 
cgit v1.2.1


From 9c53761681497d598a31ed2f6b29b5b3480c49db Mon Sep 17 00:00:00 2001
From: Thiago Galesi <thiagogalesi@gmail.com>
Date: Sat, 29 Jul 2006 10:47:12 -0300
Subject: USB: pl2303: cosmetic changes to quirk

Cosmetic changes to quirk in pl2303_update_line_status

Signed-off-by: Thiago Galesi <thiagogalesi@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index e2d8f9b94eed..51caf8a61463 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -920,13 +920,20 @@ static void pl2303_update_line_status(struct usb_serial_port *port,
 	unsigned long flags;
 	u8 status_idx = UART_STATE;
 	u8 length = UART_STATE + 1;
+	u16 idv, idp;
 
-	if ((le16_to_cpu(port->serial->dev->descriptor.idVendor) == SIEMENS_VENDOR_ID) &&
-	    (le16_to_cpu(port->serial->dev->descriptor.idProduct) == SIEMENS_PRODUCT_ID_X65 ||
-	     le16_to_cpu(port->serial->dev->descriptor.idProduct) == SIEMENS_PRODUCT_ID_SX1 ||
-	     le16_to_cpu(port->serial->dev->descriptor.idProduct) == SIEMENS_PRODUCT_ID_X75)) {
-		length = 1;
-		status_idx = 0;
+	idv = le16_to_cpu(port->serial->dev->descriptor.idVendor);
+	idp = le16_to_cpu(port->serial->dev->descriptor.idProduct);
+
+
+	if (idv == SIEMENS_VENDOR_ID) {
+		if (idp == SIEMENS_PRODUCT_ID_X65 ||
+		    idp == SIEMENS_PRODUCT_ID_SX1 ||
+		    idp == SIEMENS_PRODUCT_ID_X75) {
+
+			length = 1;
+			status_idx = 0;
+		}
 	}
 
 	if (actual_length < length)
-- 
cgit v1.2.1


From a99c19492a801013fd25ce7bab1f0f65a328a4ba Mon Sep 17 00:00:00 2001
From: Jamie Painter <jamie.painter@iname.com>
Date: Thu, 27 Jul 2006 14:17:28 -0400
Subject: USB: usbnet - Add unlink_rx_urbs() call to allow for Jumbo Frames

Add usbnet_unlink_rx_urbs() which can be called by mini-drivers when
they change their MTU such as for Jumbo Frame support.


Signed-off-by: David Hollis <dhollis@davehollis.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/net/usbnet.c | 30 +++++++++++++++++++++++++++---
 drivers/usb/net/usbnet.h |  1 +
 2 files changed, 28 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c
index 54183e173a6d..17fc6ae5c075 100644
--- a/drivers/usb/net/usbnet.c
+++ b/drivers/usb/net/usbnet.c
@@ -61,8 +61,11 @@
  * let the USB host controller be busy for 5msec or more before an irq
  * is required, under load.  Jumbograms change the equation.
  */
-#define	RX_QLEN(dev) (((dev)->udev->speed == USB_SPEED_HIGH) ? 60 : 4)
-#define	TX_QLEN(dev) (((dev)->udev->speed == USB_SPEED_HIGH) ? 60 : 4)
+#define RX_MAX_QUEUE_MEMORY (60 * 1518)
+#define	RX_QLEN(dev) (((dev)->udev->speed == USB_SPEED_HIGH) ? \
+			(RX_MAX_QUEUE_MEMORY/(dev)->rx_urb_size) : 4)
+#define	TX_QLEN(dev) (((dev)->udev->speed == USB_SPEED_HIGH) ? \
+			(RX_MAX_QUEUE_MEMORY/(dev)->hard_mtu) : 4)
 
 // reawaken network queue this soon after stopping; else watchdog barks
 #define TX_TIMEOUT_JIFFIES	(5*HZ)
@@ -227,13 +230,23 @@ static int usbnet_change_mtu (struct net_device *net, int new_mtu)
 {
 	struct usbnet	*dev = netdev_priv(net);
 	int		ll_mtu = new_mtu + net->hard_header_len;
+	int		old_hard_mtu = dev->hard_mtu;
+	int		old_rx_urb_size = dev->rx_urb_size;
 
-	if (new_mtu <= 0 || ll_mtu > dev->hard_mtu)
+	if (new_mtu <= 0)
 		return -EINVAL;
 	// no second zero-length packet read wanted after mtu-sized packets
 	if ((ll_mtu % dev->maxpacket) == 0)
 		return -EDOM;
 	net->mtu = new_mtu;
+
+	dev->hard_mtu = net->mtu + net->hard_header_len;
+	if (dev->rx_urb_size == old_hard_mtu) {
+		dev->rx_urb_size = dev->hard_mtu;
+		if (dev->rx_urb_size > old_rx_urb_size)
+			usbnet_unlink_rx_urbs(dev);
+	}
+
 	return 0;
 }
 
@@ -521,6 +534,17 @@ static int unlink_urbs (struct usbnet *dev, struct sk_buff_head *q)
 	return count;
 }
 
+// Flush all pending rx urbs
+// minidrivers may need to do this when the MTU changes
+
+void usbnet_unlink_rx_urbs(struct usbnet *dev)
+{
+	if (netif_running(dev->net)) {
+		(void) unlink_urbs (dev, &dev->rxq);
+		tasklet_schedule(&dev->bh);
+	}
+}
+EXPORT_SYMBOL_GPL(usbnet_unlink_rx_urbs);
 
 /*-------------------------------------------------------------------------*/
 
diff --git a/drivers/usb/net/usbnet.h b/drivers/usb/net/usbnet.h
index 89fc4958eecf..c0746f0454af 100644
--- a/drivers/usb/net/usbnet.h
+++ b/drivers/usb/net/usbnet.h
@@ -166,6 +166,7 @@ struct skb_data {	/* skb->cb is one of these */
 extern int usbnet_get_endpoints(struct usbnet *, struct usb_interface *);
 extern void usbnet_defer_kevent (struct usbnet *, int);
 extern void usbnet_skb_return (struct usbnet *, struct sk_buff *);
+extern void usbnet_unlink_rx_urbs(struct usbnet *);
 
 extern u32 usbnet_get_msglevel (struct net_device *);
 extern void usbnet_set_msglevel (struct net_device *, u32);
-- 
cgit v1.2.1


From 933a27d39e0e57ba56cff2e4ebe92cf23f4cd815 Mon Sep 17 00:00:00 2001
From: David Hollis <dhollis@davehollis.com>
Date: Sat, 29 Jul 2006 10:12:50 -0400
Subject: USB: asix - Add AX88178 support and many other changes

* More generi-fication of function/macro names where appropriate:
     ax88772_xx() -> asix_xx()

* Reorder functions to provide more logical grouping

* AX88178 device support

* Support DLink DUB-E100 Rev B Support

* Hopefully resolve all endian-ness issues

* Use more defines for bitmask values

* Change a number of devdbg() calls to deverr() so that if DEBUG is not
  defined, the error messages still get through as necessary


Signed-off-by: David Hollis <dhollis@davehollis.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/net/asix.c | 1013 ++++++++++++++++++++++++++++++++++++------------
 1 file changed, 775 insertions(+), 238 deletions(-)

diff --git a/drivers/usb/net/asix.c b/drivers/usb/net/asix.c
index 2e2bbc003e93..9b97aa6384c7 100644
--- a/drivers/usb/net/asix.c
+++ b/drivers/usb/net/asix.c
@@ -1,7 +1,8 @@
 /*
  * ASIX AX8817X based USB 2.0 Ethernet Devices
- * Copyright (C) 2003-2005 David Hollis <dhollis@davehollis.com>
+ * Copyright (C) 2003-2006 David Hollis <dhollis@davehollis.com>
  * Copyright (C) 2005 Phil Chang <pchang23@sbcglobal.net>
+ * Copyright (C) 2006 James Painter <jamie.painter@iname.com>
  * Copyright (c) 2002-2003 TiVo Inc.
  *
  * This program is free software; you can redistribute it and/or modify
@@ -36,6 +37,9 @@
 
 #include "usbnet.h"
 
+#define DRIVER_VERSION "14-Jun-2006"
+static const char driver_name [] = "asix";
+
 /* ASIX AX8817X based USB 2.0 Ethernet Devices */
 
 #define AX_CMD_SET_SW_MII		0x06
@@ -46,23 +50,25 @@
 #define AX_CMD_WRITE_EEPROM		0x0c
 #define AX_CMD_WRITE_ENABLE		0x0d
 #define AX_CMD_WRITE_DISABLE		0x0e
+#define AX_CMD_READ_RX_CTL		0x0f
 #define AX_CMD_WRITE_RX_CTL		0x10
 #define AX_CMD_READ_IPG012		0x11
 #define AX_CMD_WRITE_IPG0		0x12
 #define AX_CMD_WRITE_IPG1		0x13
+#define AX_CMD_READ_NODE_ID		0x13
 #define AX_CMD_WRITE_IPG2		0x14
 #define AX_CMD_WRITE_MULTI_FILTER	0x16
-#define AX_CMD_READ_NODE_ID		0x17
+#define AX88172_CMD_READ_NODE_ID	0x17
 #define AX_CMD_READ_PHY_ID		0x19
 #define AX_CMD_READ_MEDIUM_STATUS	0x1a
 #define AX_CMD_WRITE_MEDIUM_MODE	0x1b
 #define AX_CMD_READ_MONITOR_MODE	0x1c
 #define AX_CMD_WRITE_MONITOR_MODE	0x1d
+#define AX_CMD_READ_GPIOS		0x1e
 #define AX_CMD_WRITE_GPIOS		0x1f
 #define AX_CMD_SW_RESET			0x20
 #define AX_CMD_SW_PHY_STATUS		0x21
 #define AX_CMD_SW_PHY_SELECT		0x22
-#define AX88772_CMD_READ_NODE_ID	0x13
 
 #define AX_MONITOR_MODE			0x01
 #define AX_MONITOR_LINK			0x02
@@ -70,15 +76,15 @@
 #define AX_MONITOR_HSFS			0x10
 
 /* AX88172 Medium Status Register values */
-#define AX_MEDIUM_FULL_DUPLEX		0x02
-#define AX_MEDIUM_TX_ABORT_ALLOW	0x04
-#define AX_MEDIUM_FLOW_CONTROL_EN	0x10
+#define AX88172_MEDIUM_FD		0x02
+#define AX88172_MEDIUM_TX		0x04
+#define AX88172_MEDIUM_FC		0x10
+#define AX88172_MEDIUM_DEFAULT \
+		( AX88172_MEDIUM_FD | AX88172_MEDIUM_TX | AX88172_MEDIUM_FC )
 
 #define AX_MCAST_FILTER_SIZE		8
 #define AX_MAX_MCAST			64
 
-#define AX_EEPROM_LEN			0x40
-
 #define AX_SWRESET_CLEAR		0x00
 #define AX_SWRESET_RR			0x01
 #define AX_SWRESET_RT			0x02
@@ -92,23 +98,78 @@
 #define AX88772_IPG1_DEFAULT		0x0c
 #define AX88772_IPG2_DEFAULT		0x12
 
-#define AX88772_MEDIUM_FULL_DUPLEX	0x0002
-#define AX88772_MEDIUM_RESERVED		0x0004
-#define AX88772_MEDIUM_RX_FC_ENABLE	0x0010
-#define AX88772_MEDIUM_TX_FC_ENABLE	0x0020
-#define AX88772_MEDIUM_PAUSE_FORMAT	0x0080
-#define AX88772_MEDIUM_RX_ENABLE	0x0100
-#define AX88772_MEDIUM_100MB		0x0200
-#define AX88772_MEDIUM_DEFAULT	\
-	(AX88772_MEDIUM_FULL_DUPLEX | AX88772_MEDIUM_RX_FC_ENABLE | \
-	 AX88772_MEDIUM_TX_FC_ENABLE | AX88772_MEDIUM_100MB | \
-	 AX88772_MEDIUM_RESERVED | AX88772_MEDIUM_RX_ENABLE )
+/* AX88772 & AX88178 Medium Mode Register */
+#define AX_MEDIUM_PF		0x0080
+#define AX_MEDIUM_JFE		0x0040
+#define AX_MEDIUM_TFC		0x0020
+#define AX_MEDIUM_RFC		0x0010
+#define AX_MEDIUM_ENCK		0x0008
+#define AX_MEDIUM_AC		0x0004
+#define AX_MEDIUM_FD		0x0002
+#define AX_MEDIUM_GM		0x0001
+#define AX_MEDIUM_SM		0x1000
+#define AX_MEDIUM_SBP		0x0800
+#define AX_MEDIUM_PS		0x0200
+#define AX_MEDIUM_RE		0x0100
+
+#define AX88178_MEDIUM_DEFAULT	\
+	(AX_MEDIUM_PS | AX_MEDIUM_FD | AX_MEDIUM_AC | \
+	 AX_MEDIUM_RFC | AX_MEDIUM_TFC | AX_MEDIUM_JFE | \
+	 AX_MEDIUM_RE )
 
-#define AX_EEPROM_MAGIC			0xdeadbeef
+#define AX88772_MEDIUM_DEFAULT	\
+	(AX_MEDIUM_FD | AX_MEDIUM_RFC | \
+	 AX_MEDIUM_TFC | AX_MEDIUM_PS | \
+	 AX_MEDIUM_AC | AX_MEDIUM_RE )
+
+/* AX88772 & AX88178 RX_CTL values */
+#define AX_RX_CTL_SO			0x0080
+#define AX_RX_CTL_AP			0x0020
+#define AX_RX_CTL_AM			0x0010
+#define AX_RX_CTL_AB			0x0008
+#define AX_RX_CTL_SEP			0x0004
+#define AX_RX_CTL_AMALL			0x0002
+#define AX_RX_CTL_PRO			0x0001
+#define AX_RX_CTL_MFB_2048		0x0000
+#define AX_RX_CTL_MFB_4096		0x0100
+#define AX_RX_CTL_MFB_8192		0x0200
+#define AX_RX_CTL_MFB_16384		0x0300
+
+#define AX_DEFAULT_RX_CTL	\
+	(AX_RX_CTL_SO | AX_RX_CTL_AB )
+
+/* GPIO 0 .. 2 toggles */
+#define AX_GPIO_GPO0EN		0x01	/* GPIO0 Output enable */
+#define AX_GPIO_GPO_0		0x02	/* GPIO0 Output value */
+#define AX_GPIO_GPO1EN		0x04	/* GPIO1 Output enable */
+#define AX_GPIO_GPO_1		0x08	/* GPIO1 Output value */
+#define AX_GPIO_GPO2EN		0x10	/* GPIO2 Output enable */
+#define AX_GPIO_GPO_2		0x20	/* GPIO2 Output value */
+#define AX_GPIO_RESERVED	0x40	/* Reserved */
+#define AX_GPIO_RSE		0x80	/* Reload serial EEPROM */
+
+#define AX_EEPROM_MAGIC		0xdeadbeef
+#define AX88172_EEPROM_LEN	0x40
+#define AX88772_EEPROM_LEN	0xff
+
+#define PHY_MODE_MARVELL	0x0000
+#define MII_MARVELL_LED_CTRL	0x0018
+#define MII_MARVELL_STATUS	0x001b
+#define MII_MARVELL_CTRL	0x0014
+
+#define MARVELL_LED_MANUAL	0x0019
+
+#define MARVELL_STATUS_HWCFG	0x0004
+
+#define MARVELL_CTRL_TXDELAY	0x0002
+#define MARVELL_CTRL_RXDELAY	0x0080
 
 /* This structure cannot exceed sizeof(unsigned long [5]) AKA 20 bytes */
 struct asix_data {
 	u8 multi_filter[AX_MCAST_FILTER_SIZE];
+	u8 phymode;
+	u8 ledmode;
+	u8 eeprom_len;
 };
 
 struct ax88172_int_data {
@@ -122,6 +183,8 @@ struct ax88172_int_data {
 static int asix_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
 			    u16 size, void *data)
 {
+	devdbg(dev,"asix_read_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d",
+		cmd, value, index, size);
 	return usb_control_msg(
 		dev->udev,
 		usb_rcvctrlpipe(dev->udev, 0),
@@ -137,6 +200,8 @@ static int asix_read_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
 static int asix_write_cmd(struct usbnet *dev, u8 cmd, u16 value, u16 index,
 			     u16 size, void *data)
 {
+	devdbg(dev,"asix_write_cmd() cmd=0x%02x value=0x%04x index=0x%04x size=%d",
+		cmd, value, index, size);
 	return usb_control_msg(
 		dev->udev,
 		usb_sndctrlpipe(dev->udev, 0),
@@ -161,12 +226,167 @@ static void asix_async_cmd_callback(struct urb *urb, struct pt_regs *regs)
 	usb_free_urb(urb);
 }
 
+static void
+asix_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index,
+				    u16 size, void *data)
+{
+	struct usb_ctrlrequest *req;
+	int status;
+	struct urb *urb;
+
+	devdbg(dev,"asix_write_cmd_async() cmd=0x%02x value=0x%04x index=0x%04x size=%d",
+		cmd, value, index, size);
+	if ((urb = usb_alloc_urb(0, GFP_ATOMIC)) == NULL) {
+		deverr(dev, "Error allocating URB in write_cmd_async!");
+		return;
+	}
+
+	if ((req = kmalloc(sizeof(struct usb_ctrlrequest), GFP_ATOMIC)) == NULL) {
+		deverr(dev, "Failed to allocate memory for control request");
+		usb_free_urb(urb);
+		return;
+	}
+
+	req->bRequestType = USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE;
+	req->bRequest = cmd;
+	req->wValue = value;
+	req->wIndex = index;
+	req->wLength = size;
+
+	usb_fill_control_urb(urb, dev->udev,
+			     usb_sndctrlpipe(dev->udev, 0),
+			     (void *)req, data, size,
+			     asix_async_cmd_callback, req);
+
+	if((status = usb_submit_urb(urb, GFP_ATOMIC)) < 0) {
+		deverr(dev, "Error submitting the control message: status=%d",
+				status);
+		kfree(req);
+		usb_free_urb(urb);
+	}
+}
+
+static int asix_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
+{
+	u8  *head;
+	u32  header;
+	char *packet;
+	struct sk_buff *ax_skb;
+	u16 size;
+
+	head = (u8 *) skb->data;
+	memcpy(&header, head, sizeof(header));
+	le32_to_cpus(&header);
+	packet = head + sizeof(header);
+
+	skb_pull(skb, 4);
+
+	while (skb->len > 0) {
+		if ((short)(header & 0x0000ffff) !=
+		    ~((short)((header & 0xffff0000) >> 16))) {
+			deverr(dev,"asix_rx_fixup() Bad Header Length");
+		}
+		/* get the packet length */
+		size = (u16) (header & 0x0000ffff);
+
+		if ((skb->len) - ((size + 1) & 0xfffe) == 0)
+			return 2;
+		if (size > ETH_FRAME_LEN) {
+			deverr(dev,"asix_rx_fixup() Bad RX Length %d", size);
+			return 0;
+		}
+		ax_skb = skb_clone(skb, GFP_ATOMIC);
+		if (ax_skb) {
+			ax_skb->len = size;
+			ax_skb->data = packet;
+			ax_skb->tail = packet + size;
+			usbnet_skb_return(dev, ax_skb);
+		} else {
+			return 0;
+		}
+
+		skb_pull(skb, (size + 1) & 0xfffe);
+
+		if (skb->len == 0)
+			break;
+
+		head = (u8 *) skb->data;
+		memcpy(&header, head, sizeof(header));
+		le32_to_cpus(&header);
+		packet = head + sizeof(header);
+		skb_pull(skb, 4);
+	}
+
+	if (skb->len < 0) {
+		deverr(dev,"asix_rx_fixup() Bad SKB Length %d", skb->len);
+		return 0;
+	}
+	return 1;
+}
+
+static struct sk_buff *asix_tx_fixup(struct usbnet *dev, struct sk_buff *skb,
+					gfp_t flags)
+{
+	int padlen;
+	int headroom = skb_headroom(skb);
+	int tailroom = skb_tailroom(skb);
+	u32 packet_len;
+	u32 padbytes = 0xffff0000;
+
+	padlen = ((skb->len + 4) % 512) ? 0 : 4;
+
+	if ((!skb_cloned(skb))
+	    && ((headroom + tailroom) >= (4 + padlen))) {
+		if ((headroom < 4) || (tailroom < padlen)) {
+			skb->data = memmove(skb->head + 4, skb->data, skb->len);
+			skb->tail = skb->data + skb->len;
+		}
+	} else {
+		struct sk_buff *skb2;
+		skb2 = skb_copy_expand(skb, 4, padlen, flags);
+		dev_kfree_skb_any(skb);
+		skb = skb2;
+		if (!skb)
+			return NULL;
+	}
+
+	skb_push(skb, 4);
+	packet_len = (((skb->len - 4) ^ 0x0000ffff) << 16) + (skb->len - 4);
+	memcpy(skb->data, &packet_len, sizeof(packet_len));
+
+	if ((skb->len % 512) == 0) {
+		memcpy( skb->tail, &padbytes, sizeof(padbytes));
+		skb_put(skb, sizeof(padbytes));
+	}
+	return skb;
+}
+
+static void asix_status(struct usbnet *dev, struct urb *urb)
+{
+	struct ax88172_int_data *event;
+	int link;
+
+	if (urb->actual_length < 8)
+		return;
+
+	event = urb->transfer_buffer;
+	link = event->link & 0x01;
+	if (netif_carrier_ok(dev->net) != link) {
+		if (link) {
+			netif_carrier_on(dev->net);
+			usbnet_defer_kevent (dev, EVENT_LINK_RESET );
+		} else
+			netif_carrier_off(dev->net);
+		devdbg(dev, "Link Status is: %d", link);
+	}
+}
+
 static inline int asix_set_sw_mii(struct usbnet *dev)
 {
 	int ret;
 	ret = asix_write_cmd(dev, AX_CMD_SET_SW_MII, 0x0000, 0, 0, NULL);
 	if (ret < 0)
-		devdbg(dev, "Failed to enable software MII access");
+		deverr(dev, "Failed to enable software MII access");
 	return ret;
 }
 
@@ -175,24 +395,27 @@ static inline int asix_set_hw_mii(struct usbnet *dev)
 	int ret;
 	ret = asix_write_cmd(dev, AX_CMD_SET_HW_MII, 0x0000, 0, 0, NULL);
 	if (ret < 0)
-		devdbg(dev, "Failed to enable hardware MII access");
+		deverr(dev, "Failed to enable hardware MII access");
 	return ret;
 }
 
-static inline int asix_get_phyid(struct usbnet *dev)
+static inline int asix_get_phy_addr(struct usbnet *dev)
 {
 	int ret = 0;
 	void *buf;
 
+	devdbg(dev, "asix_get_phy_addr()");
+
 	buf = kmalloc(2, GFP_KERNEL);
 	if (!buf)
 		goto out1;
 
 	if ((ret = asix_read_cmd(dev, AX_CMD_READ_PHY_ID,
 				    0, 0, 2, buf)) < 2) {
-		devdbg(dev, "Error reading PHYID register: %02x", ret);
+		deverr(dev, "Error reading PHYID register: %02x", ret);
 		goto out2;
 	}
+	devdbg(dev, "asix_get_phy_addr() returning 0x%04x", *((u16 *)buf));
 	ret = *((u8 *)buf + 1);
 out2:
 	kfree(buf);
@@ -206,8 +429,29 @@ static int asix_sw_reset(struct usbnet *dev, u8 flags)
 
         ret = asix_write_cmd(dev, AX_CMD_SW_RESET, flags, 0, 0, NULL);
 	if (ret < 0)
-		devdbg(dev,"Failed to send software reset: %02x", ret);
+		deverr(dev,"Failed to send software reset: %02x", ret);
+
+	return ret;
+}
 
+static u16 asix_read_rx_ctl(struct usbnet *dev)
+{
+	u16 ret = 0;
+	void *buf;
+
+	buf = kmalloc(2, GFP_KERNEL);
+	if (!buf)
+		goto out1;
+
+	if ((ret = asix_read_cmd(dev, AX_CMD_READ_RX_CTL,
+				    0, 0, 2, buf)) < 2) {
+		deverr(dev, "Error reading RX_CTL register: %02x", ret);
+		goto out2;
+	}
+	ret = le16_to_cpu(*((u16 *)buf));
+out2:
+	kfree(buf);
+out1:
 	return ret;
 }
 
@@ -215,82 +459,79 @@ static int asix_write_rx_ctl(struct usbnet *dev, u16 mode)
 {
 	int ret;
 
+	devdbg(dev,"asix_write_rx_ctl() - mode = 0x%04x", mode);
 	ret = asix_write_cmd(dev, AX_CMD_WRITE_RX_CTL, mode, 0, 0, NULL);
 	if (ret < 0)
-                devdbg(dev, "Failed to write RX_CTL mode: %02x", ret);
+		deverr(dev, "Failed to write RX_CTL mode to 0x%04x: %02x",
+		       mode, ret);
 
 	return ret;
 }
 
-static void asix_status(struct usbnet *dev, struct urb *urb)
+static u16 asix_read_medium_status(struct usbnet *dev)
 {
-	struct ax88172_int_data *event;
-	int link;
+	u16 ret = 0;
+	void *buf;
 
-	if (urb->actual_length < 8)
-		return;
+	buf = kmalloc(2, GFP_KERNEL);
+	if (!buf)
+		goto out1;
 
-	event = urb->transfer_buffer;
-	link = event->link & 0x01;
-	if (netif_carrier_ok(dev->net) != link) {
-		if (link) {
-			netif_carrier_on(dev->net);
-			usbnet_defer_kevent (dev, EVENT_LINK_RESET );
-		} else
-			netif_carrier_off(dev->net);
-		devdbg(dev, "Link Status is: %d", link);
+	if ((ret = asix_read_cmd(dev, AX_CMD_READ_MEDIUM_STATUS,
+				    0, 0, 2, buf)) < 2) {
+		deverr(dev, "Error reading Medium Status register: %02x", ret);
+		goto out2;
 	}
+	ret = le16_to_cpu(*((u16 *)buf));
+out2:
+	kfree(buf);
+out1:
+	return ret;
 }
 
-static void
-asix_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index,
-				    u16 size, void *data)
+static int asix_write_medium_mode(struct usbnet *dev, u16 mode)
 {
-	struct usb_ctrlrequest *req;
-	int status;
-	struct urb *urb;
+	int ret;
 
-	if ((urb = usb_alloc_urb(0, GFP_ATOMIC)) == NULL) {
-		devdbg(dev, "Error allocating URB in write_cmd_async!");
-		return;
-	}
+	devdbg(dev,"asix_write_medium_mode() - mode = 0x%04x", mode);
+	ret = asix_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL);
+	if (ret < 0)
+		deverr(dev, "Failed to write Medium Mode mode to 0x%04x: %02x",
+			mode, ret);
 
-	if ((req = kmalloc(sizeof(struct usb_ctrlrequest), GFP_ATOMIC)) == NULL) {
-		deverr(dev, "Failed to allocate memory for control request");
-		usb_free_urb(urb);
-		return;
-	}
+	return ret;
+}
 
-	req->bRequestType = USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE;
-	req->bRequest = cmd;
-	req->wValue = cpu_to_le16(value);
-	req->wIndex = cpu_to_le16(index);
-	req->wLength = cpu_to_le16(size);
+static int asix_write_gpio(struct usbnet *dev, u16 value, int sleep)
+{
+	int ret;
 
-	usb_fill_control_urb(urb, dev->udev,
-			     usb_sndctrlpipe(dev->udev, 0),
-			     (void *)req, data, size,
-			     asix_async_cmd_callback, req);
+	devdbg(dev,"asix_write_gpio() - value = 0x%04x", value);
+	ret = asix_write_cmd(dev, AX_CMD_WRITE_GPIOS, value, 0, 0, NULL);
+	if (ret < 0)
+		deverr(dev, "Failed to write GPIO value 0x%04x: %02x",
+			value, ret);
 
-	if((status = usb_submit_urb(urb, GFP_ATOMIC)) < 0) {
-		deverr(dev, "Error submitting the control message: status=%d",
-				status);
-		kfree(req);
-		usb_free_urb(urb);
-	}
+	if (sleep)
+		msleep(sleep);
+
+	return ret;
 }
 
+/*
+ * AX88772 & AX88178 have a 16-bit RX_CTL value
+ */
 static void asix_set_multicast(struct net_device *net)
 {
 	struct usbnet *dev = netdev_priv(net);
 	struct asix_data *data = (struct asix_data *)&dev->data;
-	u8 rx_ctl = 0x8c;
+	u16 rx_ctl = AX_DEFAULT_RX_CTL;
 
 	if (net->flags & IFF_PROMISC) {
-		rx_ctl |= 0x01;
+		rx_ctl |= AX_RX_CTL_PRO;
 	} else if (net->flags & IFF_ALLMULTI
 		   || net->mc_count > AX_MAX_MCAST) {
-		rx_ctl |= 0x02;
+		rx_ctl |= AX_RX_CTL_AMALL;
 	} else if (net->mc_count == 0) {
 		/* just broadcast and directed */
 	} else {
@@ -317,7 +558,7 @@ static void asix_set_multicast(struct net_device *net)
 		asix_write_cmd_async(dev, AX_CMD_WRITE_MULTI_FILTER, 0, 0,
 				   AX_MCAST_FILTER_SIZE, data->multi_filter);
 
-		rx_ctl |= 0x10;
+		rx_ctl |= AX_RX_CTL_AM;
 	}
 
 	asix_write_cmd_async(dev, AX_CMD_WRITE_RX_CTL, rx_ctl, 0, 0, NULL);
@@ -333,50 +574,43 @@ static int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 				(__u16)loc, 2, (u16 *)&res);
 	asix_set_hw_mii(dev);
 
-	return res & 0xffff;
-}
+	devdbg(dev, "asix_mdio_read() phy_id=0x%02x, loc=0x%02x, returns=0x%04x", phy_id, loc, le16_to_cpu(res & 0xffff));
 
-/* same as above, but converts resulting value to cpu byte order */
-static int asix_mdio_read_le(struct net_device *netdev, int phy_id, int loc)
-{
-	return le16_to_cpu(asix_mdio_read(netdev,phy_id, loc));
+	return le16_to_cpu(res & 0xffff);
 }
 
 static void
 asix_mdio_write(struct net_device *netdev, int phy_id, int loc, int val)
 {
 	struct usbnet *dev = netdev_priv(netdev);
-	u16 res = val;
+	u16 res = cpu_to_le16(val);
 
+	devdbg(dev, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x", phy_id, loc, val);
 	asix_set_sw_mii(dev);
 	asix_write_cmd(dev, AX_CMD_WRITE_MII_REG, phy_id,
 				(__u16)loc, 2, (u16 *)&res);
 	asix_set_hw_mii(dev);
 }
 
-/* same as above, but converts new value to le16 byte order before writing */
-static void
-asix_mdio_write_le(struct net_device *netdev, int phy_id, int loc, int val)
+/* Get the PHY Identifier from the PHYSID1 & PHYSID2 MII registers */
+static u32 asix_get_phyid(struct usbnet *dev)
 {
-	asix_mdio_write( netdev, phy_id, loc, cpu_to_le16(val) );
-}
+	int phy_reg;
+	u32 phy_id;
 
-static int ax88172_link_reset(struct usbnet *dev)
-{
-	u16 lpa;
-	u16 adv;
-	u16 res;
-	u8 mode;
+	phy_reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID1);
+	if (phy_reg < 0)
+		return 0;
 
-	mode = AX_MEDIUM_TX_ABORT_ALLOW | AX_MEDIUM_FLOW_CONTROL_EN;
-	lpa = asix_mdio_read_le(dev->net, dev->mii.phy_id, MII_LPA);
-	adv = asix_mdio_read_le(dev->net, dev->mii.phy_id, MII_ADVERTISE);
-	res = mii_nway_result(lpa|adv);
-	if (res & LPA_DUPLEX)
-		mode |= AX_MEDIUM_FULL_DUPLEX;
-	asix_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL);
+	phy_id = (phy_reg & 0xffff) << 16;
 
-	return 0;
+	phy_reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_PHYSID2);
+	if (phy_reg < 0)
+		return 0;
+
+	phy_id |= (phy_reg & 0xffff);
+
+	return phy_id;
 }
 
 static void
@@ -423,7 +657,10 @@ asix_set_wol(struct net_device *net, struct ethtool_wolinfo *wolinfo)
 
 static int asix_get_eeprom_len(struct net_device *net)
 {
-	return AX_EEPROM_LEN;
+	struct usbnet *dev = netdev_priv(net);
+	struct asix_data *data = (struct asix_data *)&dev->data;
+
+	return data->eeprom_len;
 }
 
 static int asix_get_eeprom(struct net_device *net,
@@ -453,9 +690,14 @@ static int asix_get_eeprom(struct net_device *net,
 static void asix_get_drvinfo (struct net_device *net,
 				 struct ethtool_drvinfo *info)
 {
+	struct usbnet *dev = netdev_priv(net);
+	struct asix_data *data = (struct asix_data *)&dev->data;
+
 	/* Inherit standard device info */
 	usbnet_get_drvinfo(net, info);
-	info->eedump_len = 0x3e;
+	strncpy (info->driver, driver_name, sizeof info->driver);
+	strncpy (info->version, DRIVER_VERSION, sizeof info->version);
+	info->eedump_len = data->eeprom_len;
 }
 
 static int asix_get_settings(struct net_device *net, struct ethtool_cmd *cmd)
@@ -468,17 +710,44 @@ static int asix_get_settings(struct net_device *net, struct ethtool_cmd *cmd)
 static int asix_set_settings(struct net_device *net, struct ethtool_cmd *cmd)
 {
 	struct usbnet *dev = netdev_priv(net);
+	int res = mii_ethtool_sset(&dev->mii,cmd);
+
+	/* link speed/duplex might have changed */
+	if (dev->driver_info->link_reset)
+		dev->driver_info->link_reset(dev);
 
-	return mii_ethtool_sset(&dev->mii,cmd);
+	return res;
 }
 
-/* We need to override some ethtool_ops so we require our
-   own structure so we don't interfere with other usbnet
-   devices that may be connected at the same time. */
-static struct ethtool_ops ax88172_ethtool_ops = {
-	.get_drvinfo		= asix_get_drvinfo,
-	.get_link		= ethtool_op_get_link,
-	.get_msglevel		= usbnet_get_msglevel,
+static int asix_nway_reset(struct net_device *net)
+{
+	struct usbnet *dev = netdev_priv(net);
+
+	return mii_nway_restart(&dev->mii);
+}
+
+static u32 asix_get_link(struct net_device *net)
+{
+	struct usbnet *dev = netdev_priv(net);
+
+	return mii_link_ok(&dev->mii);
+}
+
+static int asix_ioctl (struct net_device *net, struct ifreq *rq, int cmd)
+{
+	struct usbnet *dev = netdev_priv(net);
+
+	return generic_mii_ioctl(&dev->mii, if_mii(rq), cmd, NULL);
+}
+
+/* We need to override some ethtool_ops so we require our
+   own structure so we don't interfere with other usbnet
+   devices that may be connected at the same time. */
+static struct ethtool_ops ax88172_ethtool_ops = {
+	.get_drvinfo		= asix_get_drvinfo,
+	.get_link		= asix_get_link,
+	.nway_reset		= asix_nway_reset,
+	.get_msglevel		= usbnet_get_msglevel,
 	.set_msglevel		= usbnet_set_msglevel,
 	.get_wol		= asix_get_wol,
 	.set_wol		= asix_set_wol,
@@ -488,11 +757,66 @@ static struct ethtool_ops ax88172_ethtool_ops = {
 	.set_settings		= asix_set_settings,
 };
 
-static int asix_ioctl (struct net_device *net, struct ifreq *rq, int cmd)
+static void ax88172_set_multicast(struct net_device *net)
 {
 	struct usbnet *dev = netdev_priv(net);
+	struct asix_data *data = (struct asix_data *)&dev->data;
+	u8 rx_ctl = 0x8c;
 
-	return generic_mii_ioctl(&dev->mii, if_mii(rq), cmd, NULL);
+	if (net->flags & IFF_PROMISC) {
+		rx_ctl |= 0x01;
+	} else if (net->flags & IFF_ALLMULTI
+		   || net->mc_count > AX_MAX_MCAST) {
+		rx_ctl |= 0x02;
+	} else if (net->mc_count == 0) {
+		/* just broadcast and directed */
+	} else {
+		/* We use the 20 byte dev->data
+		 * for our 8 byte filter buffer
+		 * to avoid allocating memory that
+		 * is tricky to free later */
+		struct dev_mc_list *mc_list = net->mc_list;
+		u32 crc_bits;
+		int i;
+
+		memset(data->multi_filter, 0, AX_MCAST_FILTER_SIZE);
+
+		/* Build the multicast hash filter. */
+		for (i = 0; i < net->mc_count; i++) {
+			crc_bits =
+			    ether_crc(ETH_ALEN,
+				      mc_list->dmi_addr) >> 26;
+			data->multi_filter[crc_bits >> 3] |=
+			    1 << (crc_bits & 7);
+			mc_list = mc_list->next;
+		}
+
+		asix_write_cmd_async(dev, AX_CMD_WRITE_MULTI_FILTER, 0, 0,
+				   AX_MCAST_FILTER_SIZE, data->multi_filter);
+
+		rx_ctl |= 0x10;
+	}
+
+	asix_write_cmd_async(dev, AX_CMD_WRITE_RX_CTL, rx_ctl, 0, 0, NULL);
+}
+
+static int ax88172_link_reset(struct usbnet *dev)
+{
+	u8 mode;
+	struct ethtool_cmd ecmd;
+
+	mii_check_media(&dev->mii, 1, 1);
+	mii_ethtool_gset(&dev->mii, &ecmd);
+	mode = AX88172_MEDIUM_DEFAULT;
+
+	if (ecmd.duplex != DUPLEX_FULL)
+		mode |= ~AX88172_MEDIUM_FD;
+
+	devdbg(dev, "ax88172_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode);
+
+	asix_write_medium_mode(dev, mode);
+
+	return 0;
 }
 
 static int ax88172_bind(struct usbnet *dev, struct usb_interface *intf)
@@ -501,6 +825,9 @@ static int ax88172_bind(struct usbnet *dev, struct usb_interface *intf)
 	void *buf;
 	int i;
 	unsigned long gpio_bits = dev->driver_info->data;
+	struct asix_data *data = (struct asix_data *)&dev->data;
+
+	data->eeprom_len = AX88172_EEPROM_LEN;
 
 	usbnet_get_endpoints(dev,intf);
 
@@ -519,12 +846,12 @@ static int ax88172_bind(struct usbnet *dev, struct usb_interface *intf)
 		msleep(5);
 	}
 
-	if ((ret = asix_write_rx_ctl(dev,0x80)) < 0)
+	if ((ret = asix_write_rx_ctl(dev, 0x80)) < 0)
 		goto out2;
 
 	/* Get the MAC address */
 	memset(buf, 0, ETH_ALEN);
-	if ((ret = asix_read_cmd(dev, AX_CMD_READ_NODE_ID,
+	if ((ret = asix_read_cmd(dev, AX88172_CMD_READ_NODE_ID,
 				0, 0, 6, buf)) < 0) {
 		dbg("read AX_CMD_READ_NODE_ID failed: %d", ret);
 		goto out2;
@@ -537,14 +864,14 @@ static int ax88172_bind(struct usbnet *dev, struct usb_interface *intf)
 	dev->mii.mdio_write = asix_mdio_write;
 	dev->mii.phy_id_mask = 0x3f;
 	dev->mii.reg_num_mask = 0x1f;
-	dev->mii.phy_id = asix_get_phyid(dev);
+	dev->mii.phy_id = asix_get_phy_addr(dev);
 	dev->net->do_ioctl = asix_ioctl;
 
-	dev->net->set_multicast_list = asix_set_multicast;
+	dev->net->set_multicast_list = ax88172_set_multicast;
 	dev->net->ethtool_ops = &ax88172_ethtool_ops;
 
-	asix_mdio_write_le(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET);
-	asix_mdio_write_le(dev->net, dev->mii.phy_id, MII_ADVERTISE,
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE,
 		ADVERTISE_ALL | ADVERTISE_CSMA | ADVERTISE_PAUSE_CAP);
 	mii_nway_restart(&dev->mii);
 
@@ -557,7 +884,8 @@ out1:
 
 static struct ethtool_ops ax88772_ethtool_ops = {
 	.get_drvinfo		= asix_get_drvinfo,
-	.get_link		= ethtool_op_get_link,
+	.get_link		= asix_get_link,
+	.nway_reset		= asix_nway_reset,
 	.get_msglevel		= usbnet_get_msglevel,
 	.set_msglevel		= usbnet_set_msglevel,
 	.get_wol		= asix_get_wol,
@@ -568,10 +896,37 @@ static struct ethtool_ops ax88772_ethtool_ops = {
 	.set_settings		= asix_set_settings,
 };
 
+static int ax88772_link_reset(struct usbnet *dev)
+{
+	u16 mode;
+	struct ethtool_cmd ecmd;
+
+	mii_check_media(&dev->mii, 1, 1);
+	mii_ethtool_gset(&dev->mii, &ecmd);
+	mode = AX88772_MEDIUM_DEFAULT;
+
+	if (ecmd.speed != SPEED_100)
+		mode &= ~AX_MEDIUM_PS;
+
+	if (ecmd.duplex != DUPLEX_FULL)
+		mode &= ~AX_MEDIUM_FD;
+
+	devdbg(dev, "ax88772_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode);
+
+	asix_write_medium_mode(dev, mode);
+
+	return 0;
+}
+
 static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 {
 	int ret;
 	void *buf;
+	u16 rx_ctl;
+	struct asix_data *data = (struct asix_data *)&dev->data;
+	u32 phyid;
+
+	data->eeprom_len = AX88772_EEPROM_LEN;
 
 	usbnet_get_endpoints(dev,intf);
 
@@ -582,13 +937,12 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 		goto out1;
 	}
 
-	if ((ret = asix_write_cmd(dev, AX_CMD_WRITE_GPIOS,
-				     0x00B0, 0, 0, buf)) < 0)
+	if ((ret = asix_write_gpio(dev,
+			AX_GPIO_RSE | AX_GPIO_GPO_2 | AX_GPIO_GPO2EN, 5)) < 0)
 		goto out2;
 
-	msleep(5);
 	if ((ret = asix_write_cmd(dev, AX_CMD_SW_PHY_SELECT,
-				0x0001, 0, 0, buf)) < 0) {
+				0x0000, 0, 0, buf)) < 0) {
 		dbg("Select PHY #1 failed: %d", ret);
 		goto out2;
 	}
@@ -605,36 +959,34 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 		goto out2;
 
 	msleep(150);
-	if ((ret = asix_write_rx_ctl(dev, 0x00)) < 0)
+	rx_ctl = asix_read_rx_ctl(dev);
+	dbg("RX_CTL is 0x%04x after software reset", rx_ctl);
+	if ((ret = asix_write_rx_ctl(dev, 0x0000)) < 0)
 		goto out2;
 
+	rx_ctl = asix_read_rx_ctl(dev);
+	dbg("RX_CTL is 0x%04x setting to 0x0000", rx_ctl);
+
 	/* Get the MAC address */
 	memset(buf, 0, ETH_ALEN);
-	if ((ret = asix_read_cmd(dev, AX88772_CMD_READ_NODE_ID,
+	if ((ret = asix_read_cmd(dev, AX_CMD_READ_NODE_ID,
 				0, 0, ETH_ALEN, buf)) < 0) {
 		dbg("Failed to read MAC address: %d", ret);
 		goto out2;
 	}
 	memcpy(dev->net->dev_addr, buf, ETH_ALEN);
 
-	if ((ret = asix_set_sw_mii(dev)) < 0)
-		goto out2;
-
-	if (((ret = asix_read_cmd(dev, AX_CMD_READ_MII_REG,
-	      			0x0010, 2, 2, buf)) < 0)
-			|| (*((u16 *)buf) != 0x003b)) {
-		dbg("Read PHY register 2 must be 0x3b00: %d", ret);
-		goto out2;
-	}
-
 	/* Initialize MII structure */
 	dev->mii.dev = dev->net;
 	dev->mii.mdio_read = asix_mdio_read;
 	dev->mii.mdio_write = asix_mdio_write;
-	dev->mii.phy_id_mask = 0xff;
-	dev->mii.reg_num_mask = 0xff;
+	dev->mii.phy_id_mask = 0x1f;
+	dev->mii.reg_num_mask = 0x1f;
 	dev->net->do_ioctl = asix_ioctl;
-	dev->mii.phy_id = asix_get_phyid(dev);
+	dev->mii.phy_id = asix_get_phy_addr(dev);
+
+	phyid = asix_get_phyid(dev);
+	dbg("PHYID=0x%08x", phyid);
 
 	if ((ret = asix_sw_reset(dev, AX_SWRESET_PRL)) < 0)
 		goto out2;
@@ -649,16 +1001,13 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 	dev->net->set_multicast_list = asix_set_multicast;
 	dev->net->ethtool_ops = &ax88772_ethtool_ops;
 
-	asix_mdio_write_le(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET);
-	asix_mdio_write_le(dev->net, dev->mii.phy_id, MII_ADVERTISE,
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR, BMCR_RESET);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE,
 			ADVERTISE_ALL | ADVERTISE_CSMA);
 	mii_nway_restart(&dev->mii);
 
-	if ((ret = asix_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE,
-				AX88772_MEDIUM_DEFAULT, 0, 0, buf)) < 0) {
-		dbg("Write medium mode register: %d", ret);
+	if ((ret = asix_write_medium_mode(dev, AX88772_MEDIUM_DEFAULT)) < 0)
 		goto out2;
-	}
 
 	if ((ret = asix_write_cmd(dev, AX_CMD_WRITE_IPG0,
 				AX88772_IPG0_DEFAULT | AX88772_IPG1_DEFAULT,
@@ -666,13 +1015,17 @@ static int ax88772_bind(struct usbnet *dev, struct usb_interface *intf)
 		dbg("Write IPG,IPG1,IPG2 failed: %d", ret);
 		goto out2;
 	}
-	if ((ret = asix_set_hw_mii(dev)) < 0)
-		goto out2;
 
 	/* Set RX_CTL to default values with 2k buffer, and enable cactus */
-	if ((ret = asix_write_rx_ctl(dev, 0x0088)) < 0)
+	if ((ret = asix_write_rx_ctl(dev, AX_DEFAULT_RX_CTL)) < 0)
 		goto out2;
 
+	rx_ctl = asix_read_rx_ctl(dev);
+	dbg("RX_CTL is 0x%04x after all initializations", rx_ctl);
+
+	rx_ctl = asix_read_medium_status(dev);
+	dbg("Medium Status is 0x%04x after all initializations", rx_ctl);
+
 	kfree(buf);
 
 	/* Asix framing packs multiple eth frames into a 2K usb bulk transfer */
@@ -690,120 +1043,285 @@ out1:
 	return ret;
 }
 
-static int ax88772_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
+static struct ethtool_ops ax88178_ethtool_ops = {
+	.get_drvinfo		= asix_get_drvinfo,
+	.get_link		= asix_get_link,
+	.nway_reset		= asix_nway_reset,
+	.get_msglevel		= usbnet_get_msglevel,
+	.set_msglevel		= usbnet_set_msglevel,
+	.get_wol		= asix_get_wol,
+	.set_wol		= asix_set_wol,
+	.get_eeprom_len		= asix_get_eeprom_len,
+	.get_eeprom		= asix_get_eeprom,
+	.get_settings		= asix_get_settings,
+	.set_settings		= asix_set_settings,
+};
+
+static int marvell_phy_init(struct usbnet *dev)
 {
-	u8  *head;
-	u32  header;
-	char *packet;
-	struct sk_buff *ax_skb;
-	u16 size;
+	struct asix_data *data = (struct asix_data *)&dev->data;
+	u16 reg;
 
-	head = (u8 *) skb->data;
-	memcpy(&header, head, sizeof(header));
-	le32_to_cpus(&header);
-	packet = head + sizeof(header);
+	devdbg(dev,"marvell_phy_init()");
 
-	skb_pull(skb, 4);
+	reg = asix_mdio_read(dev->net, dev->mii.phy_id, MII_MARVELL_STATUS);
+	devdbg(dev,"MII_MARVELL_STATUS = 0x%04x", reg);
 
-	while (skb->len > 0) {
-		if ((short)(header & 0x0000ffff) !=
-		    ~((short)((header & 0xffff0000) >> 16))) {
-			devdbg(dev,"header length data is error");
-		}
-		/* get the packet length */
-		size = (u16) (header & 0x0000ffff);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_MARVELL_CTRL,
+			MARVELL_CTRL_RXDELAY | MARVELL_CTRL_TXDELAY);
 
-		if ((skb->len) - ((size + 1) & 0xfffe) == 0)
-			return 2;
-		if (size > ETH_FRAME_LEN) {
-			devdbg(dev,"invalid rx length %d", size);
-			return 0;
-		}
-		ax_skb = skb_clone(skb, GFP_ATOMIC);
-		if (ax_skb) {
-			ax_skb->len = size;
-			ax_skb->data = packet;
-			ax_skb->tail = packet + size;
-			usbnet_skb_return(dev, ax_skb);
-		} else {
-			return 0;
-		}
+	if (data->ledmode) {
+		reg = asix_mdio_read(dev->net, dev->mii.phy_id,
+			MII_MARVELL_LED_CTRL);
+		devdbg(dev,"MII_MARVELL_LED_CTRL (1) = 0x%04x", reg);
 
-		skb_pull(skb, (size + 1) & 0xfffe);
+		reg &= 0xf8ff;
+		reg |= (1 + 0x0100);
+		asix_mdio_write(dev->net, dev->mii.phy_id,
+			MII_MARVELL_LED_CTRL, reg);
 
-		if (skb->len == 0)
-			break;
+		reg = asix_mdio_read(dev->net, dev->mii.phy_id,
+			MII_MARVELL_LED_CTRL);
+		devdbg(dev,"MII_MARVELL_LED_CTRL (2) = 0x%04x", reg);
+		reg &= 0xfc0f;
+	}
 
-		head = (u8 *) skb->data;
-		memcpy(&header, head, sizeof(header));
-		le32_to_cpus(&header);
-		packet = head + sizeof(header);
-		skb_pull(skb, 4);
+	return 0;
+}
+
+static int marvell_led_status(struct usbnet *dev, u16 speed)
+{
+	u16 reg = asix_mdio_read(dev->net, dev->mii.phy_id, MARVELL_LED_MANUAL);
+
+	devdbg(dev, "marvell_led_status() read 0x%04x", reg);
+
+	/* Clear out the center LED bits - 0x03F0 */
+	reg &= 0xfc0f;
+
+	switch (speed) {
+		case SPEED_1000:
+			reg |= 0x03e0;
+			break;
+		case SPEED_100:
+			reg |= 0x03b0;
+			break;
+		default:
+			reg |= 0x02f0;
 	}
 
-	if (skb->len < 0) {
-		devdbg(dev,"invalid rx length %d", skb->len);
-		return 0;
+	devdbg(dev, "marvell_led_status() writing 0x%04x", reg);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MARVELL_LED_MANUAL, reg);
+
+	return 0;
+}
+
+static int ax88178_link_reset(struct usbnet *dev)
+{
+	u16 mode;
+	struct ethtool_cmd ecmd;
+	struct asix_data *data = (struct asix_data *)&dev->data;
+
+	devdbg(dev,"ax88178_link_reset()");
+
+	mii_check_media(&dev->mii, 1, 1);
+	mii_ethtool_gset(&dev->mii, &ecmd);
+	mode = AX88178_MEDIUM_DEFAULT;
+
+	if (ecmd.speed == SPEED_1000)
+		mode |= AX_MEDIUM_GM | AX_MEDIUM_ENCK;
+	else if (ecmd.speed == SPEED_100)
+		mode |= AX_MEDIUM_PS;
+	else
+		mode &= ~(AX_MEDIUM_PS | AX_MEDIUM_GM);
+
+	if (ecmd.duplex == DUPLEX_FULL)
+		mode |= AX_MEDIUM_FD;
+	else
+		mode &= ~AX_MEDIUM_FD;
+
+	devdbg(dev, "ax88178_link_reset() speed: %d duplex: %d setting mode to 0x%04x", ecmd.speed, ecmd.duplex, mode);
+
+	asix_write_medium_mode(dev, mode);
+
+	if (data->phymode == PHY_MODE_MARVELL && data->ledmode)
+		marvell_led_status(dev, ecmd.speed);
+
+	return 0;
+}
+
+static void ax88178_set_mfb(struct usbnet *dev)
+{
+	u16 mfb = AX_RX_CTL_MFB_16384;
+	u16 rxctl;
+	u16 medium;
+	int old_rx_urb_size = dev->rx_urb_size;
+
+	if (dev->hard_mtu < 2048) {
+		dev->rx_urb_size = 2048;
+		mfb = AX_RX_CTL_MFB_2048;
+	} else if (dev->hard_mtu < 4096) {
+		dev->rx_urb_size = 4096;
+		mfb = AX_RX_CTL_MFB_4096;
+	} else if (dev->hard_mtu < 8192) {
+		dev->rx_urb_size = 8192;
+		mfb = AX_RX_CTL_MFB_8192;
+	} else if (dev->hard_mtu < 16384) {
+		dev->rx_urb_size = 16384;
+		mfb = AX_RX_CTL_MFB_16384;
 	}
-	return 1;
+
+	rxctl = asix_read_rx_ctl(dev);
+	asix_write_rx_ctl(dev, (rxctl & ~AX_RX_CTL_MFB_16384) | mfb);
+
+	medium = asix_read_medium_status(dev);
+	if (dev->net->mtu > 1500)
+		medium |= AX_MEDIUM_JFE;
+	else
+		medium &= ~AX_MEDIUM_JFE;
+	asix_write_medium_mode(dev, medium);
+
+	if (dev->rx_urb_size > old_rx_urb_size)
+		usbnet_unlink_rx_urbs(dev);
 }
 
-static struct sk_buff *ax88772_tx_fixup(struct usbnet *dev, struct sk_buff *skb,
-					gfp_t flags)
+static int ax88178_change_mtu(struct net_device *net, int new_mtu)
 {
-	int padlen;
-	int headroom = skb_headroom(skb);
-	int tailroom = skb_tailroom(skb);
-	u32 packet_len;
-	u32 padbytes = 0xffff0000;
+	struct usbnet *dev = netdev_priv(net);
+	int ll_mtu = new_mtu + net->hard_header_len + 4;
 
-	padlen = ((skb->len + 4) % 512) ? 0 : 4;
+	devdbg(dev, "ax88178_change_mtu() new_mtu=%d", new_mtu);
 
-	if ((!skb_cloned(skb))
-	    && ((headroom + tailroom) >= (4 + padlen))) {
-		if ((headroom < 4) || (tailroom < padlen)) {
-			skb->data = memmove(skb->head + 4, skb->data, skb->len);
-			skb->tail = skb->data + skb->len;
-		}
+	if (new_mtu <= 0 || ll_mtu > 16384)
+		return -EINVAL;
+
+	if ((ll_mtu % dev->maxpacket) == 0)
+		return -EDOM;
+
+	net->mtu = new_mtu;
+	dev->hard_mtu = net->mtu + net->hard_header_len;
+	ax88178_set_mfb(dev);
+
+	return 0;
+}
+
+static int ax88178_bind(struct usbnet *dev, struct usb_interface *intf)
+{
+	struct asix_data *data = (struct asix_data *)&dev->data;
+	int ret;
+	void *buf;
+	u16 eeprom;
+	int gpio0 = 0;
+	u32 phyid;
+
+	usbnet_get_endpoints(dev,intf);
+
+	buf = kmalloc(6, GFP_KERNEL);
+	if(!buf) {
+		dbg ("Cannot allocate memory for buffer");
+		ret = -ENOMEM;
+		goto out1;
+	}
+
+	eeprom = 0;
+	asix_read_cmd(dev, AX_CMD_READ_GPIOS, 0, 0, 1, &eeprom);
+	dbg("GPIO Status: 0x%04x", eeprom);
+
+	asix_write_cmd(dev, AX_CMD_WRITE_ENABLE, 0, 0, 0, NULL);
+	asix_read_cmd(dev, AX_CMD_READ_EEPROM, 0x0017, 0, 2, &eeprom);
+	asix_write_cmd(dev, AX_CMD_WRITE_DISABLE, 0, 0, 0, NULL);
+
+	dbg("EEPROM index 0x17 is 0x%04x", eeprom);
+
+	if (eeprom == 0xffff) {
+		data->phymode = PHY_MODE_MARVELL;
+		data->ledmode = 0;
+		gpio0 = 1;
 	} else {
-		struct sk_buff *skb2;
-		skb2 = skb_copy_expand(skb, 4, padlen, flags);
-		dev_kfree_skb_any(skb);
-		skb = skb2;
-		if (!skb)
-			return NULL;
+		data->phymode = eeprom & 7;
+		data->ledmode = eeprom >> 8;
+		gpio0 = (eeprom & 0x80) ? 0 : 1;
 	}
+	dbg("GPIO0: %d, PhyMode: %d", gpio0, data->phymode);
 
-	skb_push(skb, 4);
-	packet_len = (((skb->len - 4) ^ 0x0000ffff) << 16) + (skb->len - 4);
-	memcpy(skb->data, &packet_len, sizeof(packet_len));
+	asix_write_gpio(dev, AX_GPIO_RSE | AX_GPIO_GPO_1 | AX_GPIO_GPO1EN, 40);
+	if ((eeprom >> 8) != 1) {
+		asix_write_gpio(dev, 0x003c, 30);
+		asix_write_gpio(dev, 0x001c, 300);
+		asix_write_gpio(dev, 0x003c, 30);
+	} else {
+		dbg("gpio phymode == 1 path");
+		asix_write_gpio(dev, AX_GPIO_GPO1EN, 30);
+		asix_write_gpio(dev, AX_GPIO_GPO1EN | AX_GPIO_GPO_1, 30);
+	}
 
-	if ((skb->len % 512) == 0) {
-		memcpy( skb->tail, &padbytes, sizeof(padbytes));
-		skb_put(skb, sizeof(padbytes));
+	asix_sw_reset(dev, 0);
+	msleep(150);
+
+	asix_sw_reset(dev, AX_SWRESET_PRL | AX_SWRESET_IPPD);
+	msleep(150);
+
+	asix_write_rx_ctl(dev, 0);
+
+	/* Get the MAC address */
+	memset(buf, 0, ETH_ALEN);
+	if ((ret = asix_read_cmd(dev, AX_CMD_READ_NODE_ID,
+				0, 0, ETH_ALEN, buf)) < 0) {
+		dbg("Failed to read MAC address: %d", ret);
+		goto out2;
 	}
-	return skb;
-}
+	memcpy(dev->net->dev_addr, buf, ETH_ALEN);
 
-static int ax88772_link_reset(struct usbnet *dev)
-{
-	u16 lpa;
-	u16 adv;
-	u16 res;
-	u16 mode;
+	/* Initialize MII structure */
+	dev->mii.dev = dev->net;
+	dev->mii.mdio_read = asix_mdio_read;
+	dev->mii.mdio_write = asix_mdio_write;
+	dev->mii.phy_id_mask = 0x1f;
+	dev->mii.reg_num_mask = 0xff;
+	dev->mii.supports_gmii = 1;
+	dev->net->do_ioctl = asix_ioctl;
+	dev->mii.phy_id = asix_get_phy_addr(dev);
+	dev->net->set_multicast_list = asix_set_multicast;
+	dev->net->ethtool_ops = &ax88178_ethtool_ops;
+	dev->net->change_mtu = &ax88178_change_mtu;
 
-	mode = AX88772_MEDIUM_DEFAULT;
-	lpa = asix_mdio_read_le(dev->net, dev->mii.phy_id, MII_LPA);
-	adv = asix_mdio_read_le(dev->net, dev->mii.phy_id, MII_ADVERTISE);
-	res = mii_nway_result(lpa|adv);
+	phyid = asix_get_phyid(dev);
+	dbg("PHYID=0x%08x", phyid);
 
-	if ((res & LPA_DUPLEX) == 0)
-		mode &= ~AX88772_MEDIUM_FULL_DUPLEX;
-	if ((res & LPA_100) == 0)
-		mode &= ~AX88772_MEDIUM_100MB;
-	asix_write_cmd(dev, AX_CMD_WRITE_MEDIUM_MODE, mode, 0, 0, NULL);
+	if (data->phymode == PHY_MODE_MARVELL) {
+		marvell_phy_init(dev);
+		msleep(60);
+	}
+
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_BMCR,
+			BMCR_RESET | BMCR_ANENABLE);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_ADVERTISE,
+			ADVERTISE_ALL | ADVERTISE_CSMA | ADVERTISE_PAUSE_CAP);
+	asix_mdio_write(dev->net, dev->mii.phy_id, MII_CTRL1000,
+			ADVERTISE_1000FULL);
+
+	mii_nway_restart(&dev->mii);
+
+	if ((ret = asix_write_medium_mode(dev, AX88178_MEDIUM_DEFAULT)) < 0)
+		goto out2;
+
+	if ((ret = asix_write_rx_ctl(dev, AX_DEFAULT_RX_CTL)) < 0)
+		goto out2;
+
+	kfree(buf);
+
+	/* Asix framing packs multiple eth frames into a 2K usb bulk transfer */
+	if (dev->driver_info->flags & FLAG_FRAMING_AX) {
+		/* hard_mtu  is still the default - the device does not support
+		   jumbo eth frames */
+		dev->rx_urb_size = 2048;
+	}
 
 	return 0;
+
+out2:
+	kfree(buf);
+out1:
+	return ret;
 }
 
 static const struct driver_info ax8817x_info = {
@@ -853,8 +1371,19 @@ static const struct driver_info ax88772_info = {
 	.link_reset = ax88772_link_reset,
 	.reset = ax88772_link_reset,
 	.flags = FLAG_ETHER | FLAG_FRAMING_AX,
-	.rx_fixup = ax88772_rx_fixup,
-	.tx_fixup = ax88772_tx_fixup,
+	.rx_fixup = asix_rx_fixup,
+	.tx_fixup = asix_tx_fixup,
+};
+
+static const struct driver_info ax88178_info = {
+	.description = "ASIX AX88178 USB 2.0 Ethernet",
+	.bind = ax88178_bind,
+	.status = asix_status,
+	.link_reset = ax88178_link_reset,
+	.reset = ax88178_link_reset,
+	.flags = FLAG_ETHER | FLAG_FRAMING_AX,
+	.rx_fixup = asix_rx_fixup,
+	.tx_fixup = asix_tx_fixup,
 };
 
 static const struct usb_device_id	products [] = {
@@ -913,7 +1442,7 @@ static const struct usb_device_id	products [] = {
 }, {
 	// ASIX AX88178 10/100/1000
 	USB_DEVICE (0x0b95, 0x1780),
-	.driver_info = (unsigned long) &ax88772_info,
+	.driver_info = (unsigned long) &ax88178_info,
 }, {
 	// Linksys USB200M Rev 2
 	USB_DEVICE (0x13b1, 0x0018),
@@ -922,6 +1451,14 @@ static const struct usb_device_id	products [] = {
 	// 0Q0 cable ethernet
 	USB_DEVICE (0x1557, 0x7720),
 	.driver_info = (unsigned long) &ax88772_info,
+}, {
+	// DLink DUB-E100 H/W Ver B1
+	USB_DEVICE (0x07d1, 0x3c05),
+	.driver_info = (unsigned long) &ax88772_info,
+}, {
+	// Linksys USB1000
+	USB_DEVICE (0x1737, 0x0039),
+	.driver_info = (unsigned long) &ax88178_info,
 },
 	{ },		// END
 };
-- 
cgit v1.2.1


From 25d94e682ccb8938177bffafa67a7d21c7282a4a Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Mon, 7 Aug 2006 15:56:40 -0700
Subject: usbnet: printk format warning

Fix printk format warning(s):
drivers/usb/net/usbnet.c:654: warning: int format, different type arg (arg 3)

The fact that rx_urb_size happens to be a size_t has propagated all the way
back to this printk.  It's fragile to be using %z in this case - let's just
typecast the args instead.

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/net/usbnet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c
index 17fc6ae5c075..8e8e74d40530 100644
--- a/drivers/usb/net/usbnet.c
+++ b/drivers/usb/net/usbnet.c
@@ -653,7 +653,7 @@ static int usbnet_open (struct net_device *net)
 
 		devinfo (dev, "open: enable queueing "
 				"(rx %d, tx %d) mtu %d %s framing",
-			RX_QLEN (dev), TX_QLEN (dev), dev->net->mtu,
+			(int)RX_QLEN (dev), (int)TX_QLEN (dev), dev->net->mtu,
 			framing);
 	}
 
-- 
cgit v1.2.1


From 014aa2a3c32ebe33f97e9d219d91d3c5c7231bf7 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Thu, 27 Jul 2006 10:30:43 -0300
Subject: USB: ipaq: minor ipaq_open() cleanup.

 Commit b512504e5671f83638be0ddr085c4b1832f623d3 made ipaq_open() a bit
messy by moving the read urb submission far from its usb_fill_bulk_urb()
call and the comment explaining what it does.

 This patch put they together again. Although only compiled tested, should
not break the fix introduced by b512504e5671f83638be0ddr085c4b1832f623d3,
of course.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/ipaq.c | 38 +++++++++++++++++++++++---------------
 1 file changed, 23 insertions(+), 15 deletions(-)

diff --git a/drivers/usb/serial/ipaq.c b/drivers/usb/serial/ipaq.c
index 9840bade79f9..bfc6998cd16f 100644
--- a/drivers/usb/serial/ipaq.c
+++ b/drivers/usb/serial/ipaq.c
@@ -652,11 +652,6 @@ static int ipaq_open(struct usb_serial_port *port, struct file *filp)
 	port->bulk_out_size = port->write_urb->transfer_buffer_length = URBDATA_SIZE;
 	
 	msleep(1000*initial_wait);
-	/* Start reading from the device */
-	usb_fill_bulk_urb(port->read_urb, serial->dev, 
-		      usb_rcvbulkpipe(serial->dev, port->bulk_in_endpointAddress),
-		      port->read_urb->transfer_buffer, port->read_urb->transfer_buffer_length,
-		      ipaq_read_bulk_callback, port);
 
 	/*
 	 * Send out control message observed in win98 sniffs. Not sure what
@@ -670,18 +665,31 @@ static int ipaq_open(struct usb_serial_port *port, struct file *filp)
 		result = usb_control_msg(serial->dev,
 				usb_sndctrlpipe(serial->dev, 0), 0x22, 0x21,
 				0x1, 0, NULL, 0, 100);
-		if (result == 0) {
-			result = usb_submit_urb(port->read_urb, GFP_KERNEL);
-			if (result) {
-				err("%s - failed submitting read urb, error %d", __FUNCTION__, result);
-				goto error;
-			}
-			return 0;
-		}
+		if (!result)
+			break;
+
 		msleep(1000);
 	}
-	err("%s - failed doing control urb, error %d", __FUNCTION__, result);
-	goto error;
+
+	if (!retries && result) {
+		err("%s - failed doing control urb, error %d", __FUNCTION__,
+		    result);
+		goto error;
+	}
+
+	/* Start reading from the device */
+	usb_fill_bulk_urb(port->read_urb, serial->dev,
+		      usb_rcvbulkpipe(serial->dev, port->bulk_in_endpointAddress),
+		      port->read_urb->transfer_buffer, port->read_urb->transfer_buffer_length,
+		      ipaq_read_bulk_callback, port);
+
+	result = usb_submit_urb(port->read_urb, GFP_KERNEL);
+	if (result) {
+		err("%s - failed submitting read urb, error %d", __FUNCTION__, result);
+		goto error;
+	}
+
+	return 0;
 
 enomem:
 	result = -ENOMEM;
-- 
cgit v1.2.1


From ecdc0a590268f1926ed8534a040a390c77d20948 Mon Sep 17 00:00:00 2001
From: Franck Bui-Huu <vagabon.xyz@gmail.com>
Date: Wed, 12 Jul 2006 10:09:41 +0200
Subject: USB: usbcore get rid of the timer in usb_start_wait_urb()

This patch uses completion timeout instead of a timer to implement
a timeout when submitting an URB in usb_start_wait_urb().

It also fixes a small issue. With the previous code, if no timeout
happened and the URB's status was set to ECONNRESET value, the code
assumed wrongly that a timeout had occured.

Signed-off-by: Franck Bui-Huu <vagabon.xyz@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/message.c | 73 ++++++++++++++++++----------------------------
 1 file changed, 29 insertions(+), 44 deletions(-)

diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 4cc8d3e67db7..49cfd7928a1c 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -23,59 +23,44 @@ static void usb_api_blocking_completion(struct urb *urb, struct pt_regs *regs)
 }
 
 
-static void timeout_kill(unsigned long data)
-{
-	struct urb	*urb = (struct urb *) data;
-
-	usb_unlink_urb(urb);
-}
-
-// Starts urb and waits for completion or timeout
-// note that this call is NOT interruptible, while
-// many device driver i/o requests should be interruptible
-static int usb_start_wait_urb(struct urb *urb, int timeout, int* actual_length)
+/*
+ * Starts urb and waits for completion or timeout. Note that this call
+ * is NOT interruptible. Many device driver i/o requests should be
+ * interruptible and therefore these drivers should implement their
+ * own interruptible routines.
+ */
+static int usb_start_wait_urb(struct urb *urb, int timeout, int *actual_length)
 { 
-	struct completion	done;
-	struct timer_list	timer;
-	int			status;
+	struct completion done;
+	unsigned long expire;
+	int status;
 
 	init_completion(&done); 	
 	urb->context = &done;
 	urb->actual_length = 0;
 	status = usb_submit_urb(urb, GFP_NOIO);
-
-	if (status == 0) {
-		if (timeout > 0) {
-			init_timer(&timer);
-			timer.expires = jiffies + msecs_to_jiffies(timeout);
-			timer.data = (unsigned long)urb;
-			timer.function = timeout_kill;
-			/* grr.  timeout _should_ include submit delays. */
-			add_timer(&timer);
-		}
-		wait_for_completion(&done);
+	if (unlikely(status))
+		goto out;
+
+	expire = timeout ? msecs_to_jiffies(timeout) : MAX_SCHEDULE_TIMEOUT;
+	if (!wait_for_completion_timeout(&done, expire)) {
+
+		dev_dbg(&urb->dev->dev,
+			"%s timed out on ep%d%s len=%d/%d\n",
+			current->comm,
+			usb_pipeendpoint(urb->pipe),
+			usb_pipein(urb->pipe) ? "in" : "out",
+			urb->actual_length,
+			urb->transfer_buffer_length);
+
+		usb_kill_urb(urb);
+		status = urb->status == -ENOENT ? -ETIMEDOUT : urb->status;
+	} else
 		status = urb->status;
-		/* note:  HCDs return ETIMEDOUT for other reasons too */
-		if (status == -ECONNRESET) {
-			dev_dbg(&urb->dev->dev,
-				"%s timed out on ep%d%s len=%d/%d\n",
-				current->comm,
-				usb_pipeendpoint(urb->pipe),
-				usb_pipein(urb->pipe) ? "in" : "out",
-				urb->actual_length,
-				urb->transfer_buffer_length
-				);
-			if (urb->actual_length > 0)
-				status = 0;
-			else
-				status = -ETIMEDOUT;
-		}
-		if (timeout > 0)
-			del_timer_sync(&timer);
-	}
-
+out:
 	if (actual_length)
 		*actual_length = urb->actual_length;
+
 	usb_free_urb(urb);
 	return status;
 }
-- 
cgit v1.2.1


From 3bea733ab21247290bd552dd6a2cd3049af9adef Mon Sep 17 00:00:00 2001
From: Ping Cheng <pingc@wacom.com>
Date: Thu, 13 Jul 2006 18:01:36 -0700
Subject: USB: wacom tablet driver reorganization

- split wacom.c into 4 files: wacom.h, wacom_wac.h, wacom_sys.c, and wacom_wac.c
- where wacom_sys.c deals with system specific code,
- and wacom_wac.c deals with Wacom specific code

Signed-off-by: Ping Cheng <pingc@wacom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/Makefile    |    1 +
 drivers/usb/input/wacom.c     | 1003 -----------------------------------------
 drivers/usb/input/wacom.h     |  132 ++++++
 drivers/usb/input/wacom_sys.c |  315 +++++++++++++
 drivers/usb/input/wacom_wac.c |  646 ++++++++++++++++++++++++++
 drivers/usb/input/wacom_wac.h |   48 ++
 6 files changed, 1142 insertions(+), 1003 deletions(-)
 delete mode 100644 drivers/usb/input/wacom.c
 create mode 100644 drivers/usb/input/wacom.h
 create mode 100644 drivers/usb/input/wacom_sys.c
 create mode 100644 drivers/usb/input/wacom_wac.c
 create mode 100644 drivers/usb/input/wacom_wac.h

diff --git a/drivers/usb/input/Makefile b/drivers/usb/input/Makefile
index 764114529c56..b220435c2409 100644
--- a/drivers/usb/input/Makefile
+++ b/drivers/usb/input/Makefile
@@ -3,6 +3,7 @@
 #
 
 # Multipart objects.
+wacom-objs	:= wacom_sys.o wacom_wac.o
 usbhid-objs	:= hid-core.o
 
 # Optional parts of multipart objects.
diff --git a/drivers/usb/input/wacom.c b/drivers/usb/input/wacom.c
deleted file mode 100644
index 369461a70b72..000000000000
--- a/drivers/usb/input/wacom.c
+++ /dev/null
@@ -1,1003 +0,0 @@
-/*
- *  USB Wacom Graphire and Wacom Intuos tablet support
- *
- *  Copyright (c) 2000-2004 Vojtech Pavlik	<vojtech@ucw.cz>
- *  Copyright (c) 2000 Andreas Bach Aaen	<abach@stofanet.dk>
- *  Copyright (c) 2000 Clifford Wolf		<clifford@clifford.at>
- *  Copyright (c) 2000 Sam Mosel		<sam.mosel@computer.org>
- *  Copyright (c) 2000 James E. Blair		<corvus@gnu.org>
- *  Copyright (c) 2000 Daniel Egger		<egger@suse.de>
- *  Copyright (c) 2001 Frederic Lepied		<flepied@mandrakesoft.com>
- *  Copyright (c) 2004 Panagiotis Issaris	<panagiotis.issaris@mech.kuleuven.ac.be>
- *  Copyright (c) 2002-2006 Ping Cheng		<pingc@wacom.com>
- *
- *  ChangeLog:
- *      v0.1 (vp)  - Initial release
- *      v0.2 (aba) - Support for all buttons / combinations
- *      v0.3 (vp)  - Support for Intuos added
- *	v0.4 (sm)  - Support for more Intuos models, menustrip
- *			relative mode, proximity.
- *	v0.5 (vp)  - Big cleanup, nifty features removed,
- *			they belong in userspace
- *	v1.8 (vp)  - Submit URB only when operating, moved to CVS,
- *			use input_report_key instead of report_btn and
- *			other cleanups
- *	v1.11 (vp) - Add URB ->dev setting for new kernels
- *	v1.11 (jb) - Add support for the 4D Mouse & Lens
- *	v1.12 (de) - Add support for two more inking pen IDs
- *	v1.14 (vp) - Use new USB device id probing scheme.
- *		     Fix Wacom Graphire mouse wheel
- *	v1.18 (vp) - Fix mouse wheel direction
- *		     Make mouse relative
- *      v1.20 (fl) - Report tool id for Intuos devices
- *                 - Multi tools support
- *                 - Corrected Intuos protocol decoding (airbrush, 4D mouse, lens cursor...)
- *                 - Add PL models support
- *		   - Fix Wacom Graphire mouse wheel again
- *	v1.21 (vp) - Removed protocol descriptions
- *		   - Added MISC_SERIAL for tool serial numbers
- *	      (gb) - Identify version on module load.
- *    v1.21.1 (fl) - added Graphire2 support
- *    v1.21.2 (fl) - added Intuos2 support
- *                 - added all the PL ids
- *    v1.21.3 (fl) - added another eraser id from Neil Okamoto
- *                 - added smooth filter for Graphire from Peri Hankey
- *                 - added PenPartner support from Olaf van Es
- *                 - new tool ids from Ole Martin Bjoerndalen
- *	v1.29 (pc) - Add support for more tablets
- *		   - Fix pressure reporting
- *	v1.30 (vp) - Merge 2.4 and 2.5 drivers
- *		   - Since 2.5 now has input_sync(), remove MSC_SERIAL abuse
- *		   - Cleanups here and there
- *    v1.30.1 (pi) - Added Graphire3 support
- *	v1.40 (pc) - Add support for several new devices, fix eraser reporting, ...
- *	v1.43 (pc) - Added support for Cintiq 21UX
- *		   - Fixed a Graphire bug
- *		   - Merged wacom_intuos3_irq into wacom_intuos_irq
- *	v1.44 (pc) - Added support for Graphire4, Cintiq 710, Intuos3 6x11, etc.
- *		   - Report Device IDs
- *	v1.45 (pc) - Added support for DTF 521, Intuos3 12x12 and 12x19
- *		   - Minor data report fix
- */
-
-/*
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- */
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/usb/input.h>
-#include <asm/unaligned.h>
-
-/*
- * Version Information
- */
-#define DRIVER_VERSION "v1.45"
-#define DRIVER_AUTHOR "Vojtech Pavlik <vojtech@ucw.cz>"
-#define DRIVER_DESC "USB Wacom Graphire and Wacom Intuos tablet driver"
-#define DRIVER_LICENSE "GPL"
-
-MODULE_AUTHOR(DRIVER_AUTHOR);
-MODULE_DESCRIPTION(DRIVER_DESC);
-MODULE_LICENSE(DRIVER_LICENSE);
-
-#define USB_VENDOR_ID_WACOM	0x056a
-#define STYLUS_DEVICE_ID	0x02
-#define CURSOR_DEVICE_ID	0x06
-#define ERASER_DEVICE_ID	0x0A
-
-enum {
-	PENPARTNER = 0,
-	GRAPHIRE,
-	WACOM_G4,
-	PL,
-	INTUOS,
-	INTUOS3,
-	INTUOS312,
-	INTUOS319,
-	CINTIQ,
-	MAX_TYPE
-};
-
-struct wacom_features {
-	char *name;
-	int pktlen;
-	int x_max;
-	int y_max;
-	int pressure_max;
-	int distance_max;
-	int type;
-	usb_complete_t irq;
-};
-
-struct wacom {
-	signed char *data;
-	dma_addr_t data_dma;
-	struct input_dev *dev;
-	struct usb_device *usbdev;
-	struct urb *irq;
-	struct wacom_features *features;
-	int tool[2];
-	int id[2];
-	__u32 serial[2];
-	char phys[32];
-};
-
-#define USB_REQ_GET_REPORT	0x01
-#define USB_REQ_SET_REPORT	0x09
-
-static int usb_get_report(struct usb_interface *intf, unsigned char type,
-				unsigned char id, void *buf, int size)
-{
-	return usb_control_msg(interface_to_usbdev(intf),
-		usb_rcvctrlpipe(interface_to_usbdev(intf), 0),
-		USB_REQ_GET_REPORT, USB_TYPE_CLASS | USB_RECIP_INTERFACE,
-		(type << 8) + id, intf->altsetting[0].desc.bInterfaceNumber,
-		buf, size, 100);
-}
-
-static int usb_set_report(struct usb_interface *intf, unsigned char type,
-				unsigned char id, void *buf, int size)
-{
-	return usb_control_msg(interface_to_usbdev(intf),
-		usb_sndctrlpipe(interface_to_usbdev(intf), 0),
-                USB_REQ_SET_REPORT, USB_TYPE_CLASS | USB_RECIP_INTERFACE,
-                (type << 8) + id, intf->altsetting[0].desc.bInterfaceNumber,
-		buf, size, 1000);
-}
-
-static void wacom_pl_irq(struct urb *urb, struct pt_regs *regs)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	int prox, pressure, id;
-	int retval;
-
-	switch (urb->status) {
-	case 0:
-		/* success */
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
-		return;
-	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
-		goto exit;
-	}
-
-	if (data[0] != 2) {
-		dbg("wacom_pl_irq: received unknown report #%d", data[0]);
-		goto exit;
-	}
-
-	prox = data[1] & 0x40;
-
-	input_regs(dev, regs);
-
-	id = ERASER_DEVICE_ID;
-	if (prox) {
-
-		pressure = (signed char)((data[7] << 1) | ((data[4] >> 2) & 1));
-		if (wacom->features->pressure_max > 255)
-			pressure = (pressure << 1) | ((data[4] >> 6) & 1);
-		pressure += (wacom->features->pressure_max + 1) / 2;
-
-		/*
-		 * if going from out of proximity into proximity select between the eraser
-		 * and the pen based on the state of the stylus2 button, choose eraser if
-		 * pressed else choose pen. if not a proximity change from out to in, send
-		 * an out of proximity for previous tool then a in for new tool.
-		 */
-		if (!wacom->tool[0]) {
-			/* Eraser bit set for DTF */
-			if (data[1] & 0x10)
-				wacom->tool[1] = BTN_TOOL_RUBBER;
-			else
-				/* Going into proximity select tool */
-				wacom->tool[1] = (data[4] & 0x20) ? BTN_TOOL_RUBBER : BTN_TOOL_PEN;
-		} else {
-			/* was entered with stylus2 pressed */
-			if (wacom->tool[1] == BTN_TOOL_RUBBER && !(data[4] & 0x20)) {
-				/* report out proximity for previous tool */
-				input_report_key(dev, wacom->tool[1], 0);
-				input_sync(dev);
-				wacom->tool[1] = BTN_TOOL_PEN;
-				goto exit;
-			}
-		}
-		if (wacom->tool[1] != BTN_TOOL_RUBBER) {
-			/* Unknown tool selected default to pen tool */
-			wacom->tool[1] = BTN_TOOL_PEN;
-			id = STYLUS_DEVICE_ID;
-		}
-		input_report_key(dev, wacom->tool[1], prox); /* report in proximity for tool */
-		input_report_abs(dev, ABS_MISC, id); /* report tool id */
-		input_report_abs(dev, ABS_X, data[3] | (data[2] << 7) | ((data[1] & 0x03) << 14));
-		input_report_abs(dev, ABS_Y, data[6] | (data[5] << 7) | ((data[4] & 0x03) << 14));
-		input_report_abs(dev, ABS_PRESSURE, pressure);
-
-		input_report_key(dev, BTN_TOUCH, data[4] & 0x08);
-		input_report_key(dev, BTN_STYLUS, data[4] & 0x10);
-		/* Only allow the stylus2 button to be reported for the pen tool. */
-		input_report_key(dev, BTN_STYLUS2, (wacom->tool[1] == BTN_TOOL_PEN) && (data[4] & 0x20));
-	} else {
-		/* report proximity-out of a (valid) tool */
-		if (wacom->tool[1] != BTN_TOOL_RUBBER) {
-			/* Unknown tool selected default to pen tool */
-			wacom->tool[1] = BTN_TOOL_PEN;
-		}
-		input_report_key(dev, wacom->tool[1], prox);
-	}
-
-	wacom->tool[0] = prox; /* Save proximity state */
-	input_sync(dev);
-
- exit:
-	retval = usb_submit_urb (urb, GFP_ATOMIC);
-	if (retval)
-		err ("%s - usb_submit_urb failed with result %d",
-		     __FUNCTION__, retval);
-}
-
-static void wacom_ptu_irq(struct urb *urb, struct pt_regs *regs)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	int retval, id;
-
-	switch (urb->status) {
-	case 0:
-		/* success */
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
-		return;
-	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
-		goto exit;
-	}
-
-	if (data[0] != 2) {
-		printk(KERN_INFO "wacom_ptu_irq: received unknown report #%d\n", data[0]);
-		goto exit;
-	}
-
-	input_regs(dev, regs);
-	if (data[1] & 0x04) {
-		input_report_key(dev, BTN_TOOL_RUBBER, data[1] & 0x20);
-		input_report_key(dev, BTN_TOUCH, data[1] & 0x08);
-		id = ERASER_DEVICE_ID;
-	} else {
-		input_report_key(dev, BTN_TOOL_PEN, data[1] & 0x20);
-		input_report_key(dev, BTN_TOUCH, data[1] & 0x01);
-		id = STYLUS_DEVICE_ID;
-	}
-	input_report_abs(dev, ABS_MISC, id); /* report tool id */
-	input_report_abs(dev, ABS_X, le16_to_cpu(*(__le16 *) &data[2]));
-	input_report_abs(dev, ABS_Y, le16_to_cpu(*(__le16 *) &data[4]));
-	input_report_abs(dev, ABS_PRESSURE, le16_to_cpu(*(__le16 *) &data[6]));
-	input_report_key(dev, BTN_STYLUS, data[1] & 0x02);
-	input_report_key(dev, BTN_STYLUS2, data[1] & 0x10);
-
-	input_sync(dev);
-
- exit:
-	retval = usb_submit_urb (urb, GFP_ATOMIC);
-	if (retval)
-		err ("%s - usb_submit_urb failed with result %d",
-		     __FUNCTION__, retval);
-}
-
-static void wacom_penpartner_irq(struct urb *urb, struct pt_regs *regs)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	int retval;
-
-	switch (urb->status) {
-	case 0:
-		/* success */
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
-		return;
-	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
-		goto exit;
-	}
-
-	if (data[0] != 2) {
-		printk(KERN_INFO "wacom_penpartner_irq: received unknown report #%d\n", data[0]);
-		goto exit;
-	}
-
-	input_regs(dev, regs);
-	input_report_key(dev, BTN_TOOL_PEN, 1);
-	input_report_abs(dev, ABS_MISC, STYLUS_DEVICE_ID); /* report tool id */
-	input_report_abs(dev, ABS_X, le16_to_cpu(*(__le16 *) &data[1]));
-	input_report_abs(dev, ABS_Y, le16_to_cpu(*(__le16 *) &data[3]));
-	input_report_abs(dev, ABS_PRESSURE, (signed char)data[6] + 127);
-	input_report_key(dev, BTN_TOUCH, ((signed char)data[6] > -80) && !(data[5] & 0x20));
-	input_report_key(dev, BTN_STYLUS, (data[5] & 0x40));
-	input_sync(dev);
-
- exit:
-	retval = usb_submit_urb (urb, GFP_ATOMIC);
-	if (retval)
-		err ("%s - usb_submit_urb failed with result %d",
-		     __FUNCTION__, retval);
-}
-
-static void wacom_graphire_irq(struct urb *urb, struct pt_regs *regs)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	int x, y, id, rw;
-	int retval;
-
-	switch (urb->status) {
-	case 0:
-		/* success */
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
-		return;
-	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
-		goto exit;
-	}
-
-	if (data[0] == 99) return; /* for Volito tablets */
-
-	if (data[0] != 2) {
-		dbg("wacom_graphire_irq: received unknown report #%d", data[0]);
-		goto exit;
-	}
-
-	input_regs(dev, regs);
-
-	id = STYLUS_DEVICE_ID;
-	if (data[1] & 0x10) { /* in prox */
-
-		switch ((data[1] >> 5) & 3) {
-
-			case 0:	/* Pen */
-				wacom->tool[0] = BTN_TOOL_PEN;
-				break;
-
-			case 1: /* Rubber */
-				wacom->tool[0] = BTN_TOOL_RUBBER;
-				id = ERASER_DEVICE_ID;
-				break;
-
-			case 2: /* Mouse with wheel */
-				input_report_key(dev, BTN_MIDDLE, data[1] & 0x04);
-				if (wacom->features->type == WACOM_G4) {
-					rw = data[7] & 0x04 ? (data[7] & 0x03)-4 : (data[7] & 0x03);
-					input_report_rel(dev, REL_WHEEL, -rw);
-				} else
-					input_report_rel(dev, REL_WHEEL, -(signed char) data[6]);
-				/* fall through */
-
-			case 3: /* Mouse without wheel */
-				wacom->tool[0] = BTN_TOOL_MOUSE;
-				id = CURSOR_DEVICE_ID;
-				input_report_key(dev, BTN_LEFT, data[1] & 0x01);
-				input_report_key(dev, BTN_RIGHT, data[1] & 0x02);
-				if (wacom->features->type == WACOM_G4)
-					input_report_abs(dev, ABS_DISTANCE, data[6]);
-				else
-					input_report_abs(dev, ABS_DISTANCE, data[7]);
-				break;
-		}
-	}
-
-	if (data[1] & 0x90) {
-		x = le16_to_cpu(*(__le16 *) &data[2]);
-		y = le16_to_cpu(*(__le16 *) &data[4]);
-		input_report_abs(dev, ABS_X, x);
-		input_report_abs(dev, ABS_Y, y);
-		if (wacom->tool[0] != BTN_TOOL_MOUSE) {
-			input_report_abs(dev, ABS_PRESSURE, data[6] | ((data[7] & 0x01) << 8));
-			input_report_key(dev, BTN_TOUCH, data[1] & 0x01);
-			input_report_key(dev, BTN_STYLUS, data[1] & 0x02);
-			input_report_key(dev, BTN_STYLUS2, data[1] & 0x04);
-		}
-	}
-
-	if (data[1] & 0x10)
-		input_report_abs(dev, ABS_MISC, id); /* report tool id */
-	else
-		input_report_abs(dev, ABS_MISC, 0); /* reset tool id */
-	input_report_key(dev, wacom->tool[0], data[1] & 0x10);
-	input_sync(dev);
-
-	/* send pad data */
-	if (wacom->features->type == WACOM_G4) {
-		if ((wacom->serial[1] & 0xc0) != (data[7] & 0xf8)) {
-			wacom->id[1] = 1;
-			wacom->serial[1] = (data[7] & 0xf8);
-			input_report_key(dev, BTN_0, (data[7] & 0x40));
-			input_report_key(dev, BTN_4, (data[7] & 0x80));
-			rw = ((data[7] & 0x18) >> 3) - ((data[7] & 0x20) >> 3);
-			input_report_rel(dev, REL_WHEEL, rw);
-			input_report_key(dev, BTN_TOOL_FINGER, 0xf0);
-			input_event(dev, EV_MSC, MSC_SERIAL, 0xf0);
-		} else if (wacom->id[1]) {
-			wacom->id[1] = 0;
-			input_report_key(dev, BTN_TOOL_FINGER, 0);
-			input_event(dev, EV_MSC, MSC_SERIAL, 0xf0);
-		}
-		input_sync(dev);
-	}
- exit:
-	retval = usb_submit_urb (urb, GFP_ATOMIC);
-	if (retval)
-		err ("%s - usb_submit_urb failed with result %d",
-		     __FUNCTION__, retval);
-}
-
-static int wacom_intuos_inout(struct urb *urb)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	int idx;
-
-	/* tool number */
-	idx = data[1] & 0x01;
-
-	/* Enter report */
-	if ((data[1] & 0xfc) == 0xc0) {
-		/* serial number of the tool */
-		wacom->serial[idx] = ((data[3] & 0x0f) << 28) +
-			(data[4] << 20) + (data[5] << 12) +
-			(data[6] << 4) + (data[7] >> 4);
-
-		wacom->id[idx] = (data[2] << 4) | (data[3] >> 4);
-		switch (wacom->id[idx]) {
-			case 0x812: /* Inking pen */
-			case 0x801: /* Intuos3 Inking pen */
-			case 0x012:
-				wacom->tool[idx] = BTN_TOOL_PENCIL;
-				break;
-			case 0x822: /* Pen */
-			case 0x842:
-			case 0x852:
-			case 0x823: /* Intuos3 Grip Pen */
-			case 0x813: /* Intuos3 Classic Pen */
-			case 0x885: /* Intuos3 Marker Pen */
-			case 0x022:
-				wacom->tool[idx] = BTN_TOOL_PEN;
-				break;
-			case 0x832: /* Stroke pen */
-			case 0x032:
-				wacom->tool[idx] = BTN_TOOL_BRUSH;
-				break;
-			case 0x007: /* Mouse 4D and 2D */
-		        case 0x09c:
-			case 0x094:
-			case 0x017: /* Intuos3 2D Mouse */
-				wacom->tool[idx] = BTN_TOOL_MOUSE;
-				break;
-			case 0x096: /* Lens cursor */
-			case 0x097: /* Intuos3 Lens cursor */
-				wacom->tool[idx] = BTN_TOOL_LENS;
-				break;
-			case 0x82a: /* Eraser */
-			case 0x85a:
-		        case 0x91a:
-			case 0xd1a:
-			case 0x0fa:
-			case 0x82b: /* Intuos3 Grip Pen Eraser */
-			case 0x81b: /* Intuos3 Classic Pen Eraser */
-			case 0x91b: /* Intuos3 Airbrush Eraser */
-				wacom->tool[idx] = BTN_TOOL_RUBBER;
-				break;
-			case 0xd12:
-			case 0x912:
-			case 0x112:
-			case 0x913: /* Intuos3 Airbrush */
-				wacom->tool[idx] = BTN_TOOL_AIRBRUSH;
-				break;
-			default: /* Unknown tool */
-				wacom->tool[idx] = BTN_TOOL_PEN;
-		}
-		if(!((wacom->tool[idx] == BTN_TOOL_LENS) &&
-				((wacom->features->type == INTUOS312)
-					|| (wacom->features->type == INTUOS319)))) {
-			input_report_abs(dev, ABS_MISC, wacom->id[idx]); /* report tool id */
-			input_report_key(dev, wacom->tool[idx], 1);
-			input_event(dev, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
-			input_sync(dev);
-		}
-		return 1;
-	}
-
-	/* Exit report */
-	if ((data[1] & 0xfe) == 0x80) {
-		input_report_key(dev, wacom->tool[idx], 0);
-		input_report_abs(dev, ABS_MISC, 0); /* reset tool id */
-		input_event(dev, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
-		input_sync(dev);
-		return 1;
-	}
-
-	if((wacom->tool[idx] == BTN_TOOL_LENS) && ((wacom->features->type == INTUOS312)
-			|| (wacom->features->type == INTUOS319)))
-		return 1;
-	else
-		return 0;
-}
-
-static void wacom_intuos_general(struct urb *urb)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	unsigned int t;
-
-	/* general pen packet */
-	if ((data[1] & 0xb8) == 0xa0) {
-		t = (data[6] << 2) | ((data[7] >> 6) & 3);
-		input_report_abs(dev, ABS_PRESSURE, t);
-		input_report_abs(dev, ABS_TILT_X,
-				((data[7] << 1) & 0x7e) | (data[8] >> 7));
-		input_report_abs(dev, ABS_TILT_Y, data[8] & 0x7f);
-		input_report_key(dev, BTN_STYLUS, data[1] & 2);
-		input_report_key(dev, BTN_STYLUS2, data[1] & 4);
-		input_report_key(dev, BTN_TOUCH, t > 10);
-	}
-
-	/* airbrush second packet */
-	if ((data[1] & 0xbc) == 0xb4) {
-		input_report_abs(dev, ABS_WHEEL,
-				(data[6] << 2) | ((data[7] >> 6) & 3));
-		input_report_abs(dev, ABS_TILT_X,
-				((data[7] << 1) & 0x7e) | (data[8] >> 7));
-		input_report_abs(dev, ABS_TILT_Y, data[8] & 0x7f);
-	}
-	return;
-}
-
-static void wacom_intuos_irq(struct urb *urb, struct pt_regs *regs)
-{
-	struct wacom *wacom = urb->context;
-	unsigned char *data = wacom->data;
-	struct input_dev *dev = wacom->dev;
-	unsigned int t;
-	int idx;
-	int retval;
-
-	switch (urb->status) {
-	case 0:
-		/* success */
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		/* this urb is terminated, clean up */
-		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
-		return;
-	default:
-		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
-		goto exit;
-	}
-
-	if (data[0] != 2 && data[0] != 5 && data[0] != 6 && data[0] != 12) {
-		dbg("wacom_intuos_irq: received unknown report #%d", data[0]);
-		goto exit;
-	}
-
-	input_regs(dev, regs);
-
-	/* tool number */
-	idx = data[1] & 0x01;
-
-	/* pad packets. Works as a second tool and is always in prox */
-	if (data[0] == 12) {
-		/* initiate the pad as a device */
-		if (wacom->tool[1] != BTN_TOOL_FINGER)
-			wacom->tool[1] = BTN_TOOL_FINGER;
-
-		input_report_key(dev, BTN_0, (data[5] & 0x01));
-		input_report_key(dev, BTN_1, (data[5] & 0x02));
-		input_report_key(dev, BTN_2, (data[5] & 0x04));
-		input_report_key(dev, BTN_3, (data[5] & 0x08));
-		input_report_key(dev, BTN_4, (data[6] & 0x01));
-		input_report_key(dev, BTN_5, (data[6] & 0x02));
-		input_report_key(dev, BTN_6, (data[6] & 0x04));
-		input_report_key(dev, BTN_7, (data[6] & 0x08));
-		input_report_abs(dev, ABS_RX, ((data[1] & 0x1f) << 8) | data[2]);
-		input_report_abs(dev, ABS_RY, ((data[3] & 0x1f) << 8) | data[4]);
-
-		if((data[5] & 0x0f) | (data[6] & 0x0f) | (data[1] & 0x1f) | data[2])
-			input_report_key(dev, wacom->tool[1], 1);
-		else
-			input_report_key(dev, wacom->tool[1], 0);
-		input_event(dev, EV_MSC, MSC_SERIAL, 0xffffffff);
-		input_sync(dev);
-		goto exit;
-	}
-
-	/* process in/out prox events */
-	if (wacom_intuos_inout(urb))
-		goto exit;
-
-	/* Cintiq doesn't send data when RDY bit isn't set */
-	if ((wacom->features->type == CINTIQ) && !(data[1] & 0x40))
-		goto exit;
-
-	if (wacom->features->type >= INTUOS3) {
-		input_report_abs(dev, ABS_X, (data[2] << 9) | (data[3] << 1) | ((data[9] >> 1) & 1));
-		input_report_abs(dev, ABS_Y, (data[4] << 9) | (data[5] << 1) | (data[9] & 1));
-		input_report_abs(dev, ABS_DISTANCE, ((data[9] >> 2) & 0x3f));
-	} else {
-		input_report_abs(dev, ABS_X, be16_to_cpu(*(__be16 *) &data[2]));
-		input_report_abs(dev, ABS_Y, be16_to_cpu(*(__be16 *) &data[4]));
-		input_report_abs(dev, ABS_DISTANCE, ((data[9] >> 3) & 0x1f));
-	}
-
-	/* process general packets */
-	wacom_intuos_general(urb);
-
-	/* 4D mouse, 2D mouse, marker pen rotation, or Lens cursor packets */
-	if ((data[1] & 0xbc) == 0xa8 || (data[1] & 0xbe) == 0xb0) {
-
-		if (data[1] & 0x02) {
-			/* Rotation packet */
-			if (wacom->features->type >= INTUOS3) {
-				/* I3 marker pen rotation reported as wheel
-				 * due to valuator limitation
-				 */
-				t = (data[6] << 3) | ((data[7] >> 5) & 7);
-				t = (data[7] & 0x20) ? ((t > 900) ? ((t-1) / 2 - 1350) :
-					((t-1) / 2 + 450)) : (450 - t / 2) ;
-				input_report_abs(dev, ABS_WHEEL, t);
-			} else {
-				/* 4D mouse rotation packet */
-				t = (data[6] << 3) | ((data[7] >> 5) & 7);
-				input_report_abs(dev, ABS_RZ, (data[7] & 0x20) ?
-					((t - 1) / 2) : -t / 2);
-			}
-
-		} else if (!(data[1] & 0x10) && wacom->features->type < INTUOS3) {
-			/* 4D mouse packet */
-			input_report_key(dev, BTN_LEFT,   data[8] & 0x01);
-			input_report_key(dev, BTN_MIDDLE, data[8] & 0x02);
-			input_report_key(dev, BTN_RIGHT,  data[8] & 0x04);
-
-			input_report_key(dev, BTN_SIDE,   data[8] & 0x20);
-			input_report_key(dev, BTN_EXTRA,  data[8] & 0x10);
-			t = (data[6] << 2) | ((data[7] >> 6) & 3);
-			input_report_abs(dev, ABS_THROTTLE, (data[8] & 0x08) ? -t : t);
-
-		} else if (wacom->tool[idx] == BTN_TOOL_MOUSE) {
-			/* 2D mouse packet */
-			input_report_key(dev, BTN_LEFT,   data[8] & 0x04);
-			input_report_key(dev, BTN_MIDDLE, data[8] & 0x08);
-			input_report_key(dev, BTN_RIGHT,  data[8] & 0x10);
-			input_report_rel(dev, REL_WHEEL, (data[8] & 0x01)
-						 - ((data[8] & 0x02) >> 1));
-
-			/* I3 2D mouse side buttons */
-			if (wacom->features->type == INTUOS3) {
-				input_report_key(dev, BTN_SIDE,   data[8] & 0x40);
-				input_report_key(dev, BTN_EXTRA,  data[8] & 0x20);
-			}
-
-		} else if (wacom->features->type < INTUOS3) {
-			/* Lens cursor packets */
-			input_report_key(dev, BTN_LEFT,   data[8] & 0x01);
-			input_report_key(dev, BTN_MIDDLE, data[8] & 0x02);
-			input_report_key(dev, BTN_RIGHT,  data[8] & 0x04);
-			input_report_key(dev, BTN_SIDE,   data[8] & 0x10);
-			input_report_key(dev, BTN_EXTRA,  data[8] & 0x08);
-		}
-	}
-
-	input_report_abs(dev, ABS_MISC, wacom->id[idx]); /* report tool id */
-	input_report_key(dev, wacom->tool[idx], 1);
-	input_event(dev, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
-	input_sync(dev);
-
-exit:
-	retval = usb_submit_urb (urb, GFP_ATOMIC);
-	if (retval)
-		err ("%s - usb_submit_urb failed with result %d",
-		    __FUNCTION__, retval);
-}
-
-static struct wacom_features wacom_features[] = {
-	{ "Wacom Penpartner",    7,   5040,  3780,  255, 32, PENPARTNER, wacom_penpartner_irq },
-        { "Wacom Graphire",      8,  10206,  7422,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Graphire2 4x5", 8,  10206,  7422,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Graphire2 5x7", 8,  13918, 10206,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Graphire3",     8,  10208,  7424,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Graphire3 6x8", 8,  16704, 12064,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Graphire4 4x5", 8,  10208,  7424,  511, 32, WACOM_G4,	 wacom_graphire_irq },
-	{ "Wacom Graphire4 6x8", 8,  16704, 12064,  511, 32, WACOM_G4,	 wacom_graphire_irq },
-	{ "Wacom Volito",        8,   5104,  3712,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom PenStation2",   8,   3250,  2320,  255, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Volito2 4x5",   8,   5104,  3712,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Volito2 2x3",   8,   3248,  2320,  511, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom PenPartner2",   8,   3250,  2320,  255, 32, GRAPHIRE,   wacom_graphire_irq },
-	{ "Wacom Intuos 4x5",   10,  12700, 10600, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos 6x8",   10,  20320, 16240, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos 9x12",  10,  30480, 24060, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos 12x12", 10,  30480, 31680, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos 12x18", 10,  45720, 31680, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom PL400",         8,   5408,  4056,  255, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL500",         8,   6144,  4608,  255, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL600",         8,   6126,  4604,  255, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL600SX",       8,   6260,  5016,  255, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL550",         8,   6144,  4608,  511, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL800",         8,   7220,  5780,  511, 32, PL,         wacom_pl_irq },
-	{ "Wacom PL700",         8,   6758,  5406,  511, 32, PL,	 wacom_pl_irq },
-	{ "Wacom PL510",         8,   6282,  4762,  511, 32, PL,	 wacom_pl_irq },
-	{ "Wacom DTU710",        8,  34080, 27660,  511, 32, PL,	 wacom_pl_irq },
-	{ "Wacom DTF521",        8,   6282,  4762,  511, 32, PL,         wacom_pl_irq },
-	{ "Wacom DTF720",        8,   6858,  5506,  511, 32, PL,	 wacom_pl_irq },
-	{ "Wacom Cintiq Partner",8,  20480, 15360,  511, 32, PL,         wacom_ptu_irq },
-	{ "Wacom Intuos2 4x5",   10, 12700, 10600, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos2 6x8",   10, 20320, 16240, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos2 9x12",  10, 30480, 24060, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos2 12x12", 10, 30480, 31680, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos2 12x18", 10, 45720, 31680, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ "Wacom Intuos3 4x5",   10, 25400, 20320, 1023, 15, INTUOS3,    wacom_intuos_irq },
-	{ "Wacom Intuos3 6x8",   10, 40640, 30480, 1023, 15, INTUOS3,    wacom_intuos_irq },
-	{ "Wacom Intuos3 9x12",  10, 60960, 45720, 1023, 15, INTUOS3,    wacom_intuos_irq },
-	{ "Wacom Intuos3 12x12", 10, 60960, 60960, 1023, 15, INTUOS312,  wacom_intuos_irq },
-	{ "Wacom Intuos3 12x19", 10, 97536, 60960, 1023, 15, INTUOS319,  wacom_intuos_irq },
-	{ "Wacom Intuos3 6x11",  10, 54204, 31750, 1023, 15, INTUOS3,    wacom_intuos_irq },
-	{ "Wacom Cintiq 21UX",   10, 87200, 65600, 1023, 15, CINTIQ,     wacom_intuos_irq },
-	{ "Wacom Intuos2 6x8",   10, 20320, 16240, 1023, 15, INTUOS,     wacom_intuos_irq },
-	{ }
-};
-
-static struct usb_device_id wacom_ids[] = {
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x00) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x10) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x11) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x12) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x13) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x14) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x15) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x16) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x60) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x61) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x62) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x63) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x64) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x20) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x21) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x22) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x23) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x24) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x30) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x31) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x32) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x33) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x34) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x35) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x37) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x38) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x39) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xC0) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xC3) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x03) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x41) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x42) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x43) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x44) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x45) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB0) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB1) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB2) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB3) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB4) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB5) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x3F) },
-	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x47) },
-	{ }
-};
-
-MODULE_DEVICE_TABLE(usb, wacom_ids);
-
-static int wacom_open(struct input_dev *dev)
-{
-	struct wacom *wacom = dev->private;
-
-	wacom->irq->dev = wacom->usbdev;
-	if (usb_submit_urb(wacom->irq, GFP_KERNEL))
-		return -EIO;
-
-	return 0;
-}
-
-static void wacom_close(struct input_dev *dev)
-{
-	struct wacom *wacom = dev->private;
-
-	usb_kill_urb(wacom->irq);
-}
-
-static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *id)
-{
-	struct usb_device *dev = interface_to_usbdev(intf);
-	struct usb_endpoint_descriptor *endpoint;
-	struct wacom *wacom;
-	struct input_dev *input_dev;
-	char rep_data[2], limit = 0;
-
-	wacom = kzalloc(sizeof(struct wacom), GFP_KERNEL);
-	input_dev = input_allocate_device();
-	if (!wacom || !input_dev)
-		goto fail1;
-
-	wacom->data = usb_buffer_alloc(dev, 10, GFP_KERNEL, &wacom->data_dma);
-	if (!wacom->data)
-		goto fail1;
-
-	wacom->irq = usb_alloc_urb(0, GFP_KERNEL);
-	if (!wacom->irq)
-		goto fail2;
-
-	wacom->usbdev = dev;
-	wacom->dev = input_dev;
-	usb_make_path(dev, wacom->phys, sizeof(wacom->phys));
-	strlcat(wacom->phys, "/input0", sizeof(wacom->phys));
-
-	wacom->features = wacom_features + (id - wacom_ids);
-	if (wacom->features->pktlen > 10)
-		BUG();
-
-	input_dev->name = wacom->features->name;
-	usb_to_input_id(dev, &input_dev->id);
-
-	input_dev->cdev.dev = &intf->dev;
-	input_dev->private = wacom;
-	input_dev->open = wacom_open;
-	input_dev->close = wacom_close;
-
-	input_dev->evbit[0] |= BIT(EV_KEY) | BIT(EV_ABS);
-	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_PEN) | BIT(BTN_TOUCH) | BIT(BTN_STYLUS);
-	input_set_abs_params(input_dev, ABS_X, 0, wacom->features->x_max, 4, 0);
-	input_set_abs_params(input_dev, ABS_Y, 0, wacom->features->y_max, 4, 0);
-	input_set_abs_params(input_dev, ABS_PRESSURE, 0, wacom->features->pressure_max, 0, 0);
-	input_dev->absbit[LONG(ABS_MISC)] |= BIT(ABS_MISC);
-
-	switch (wacom->features->type) {
-		case WACOM_G4:
-			input_dev->evbit[0] |= BIT(EV_MSC);
-			input_dev->mscbit[0] |= BIT(MSC_SERIAL);
-			input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER);
-			input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7);
-			/* fall through */
-
-		case GRAPHIRE:
-			input_dev->evbit[0] |= BIT(EV_REL);
-			input_dev->relbit[0] |= BIT(REL_WHEEL);
-			input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_LEFT) | BIT(BTN_RIGHT) | BIT(BTN_MIDDLE);
-			input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER) | BIT(BTN_TOOL_MOUSE) | BIT(BTN_STYLUS2);
-			input_set_abs_params(input_dev, ABS_DISTANCE, 0, wacom->features->distance_max, 0, 0);
-			break;
-
-		case INTUOS3:
-		case INTUOS312:
-		case INTUOS319:
-		case CINTIQ:
-			input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER);
-			input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7);
-			input_set_abs_params(input_dev, ABS_RX, 0, 4097, 0, 0);
-			input_set_abs_params(input_dev, ABS_RY, 0, 4097, 0, 0);
-			/* fall through */
-
-		case INTUOS:
-			input_dev->evbit[0] |= BIT(EV_MSC) | BIT(EV_REL);
-			input_dev->mscbit[0] |= BIT(MSC_SERIAL);
-			input_dev->relbit[0] |= BIT(REL_WHEEL);
-			input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_LEFT) | BIT(BTN_RIGHT) | BIT(BTN_MIDDLE) | BIT(BTN_SIDE) | BIT(BTN_EXTRA);
-			input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER) | BIT(BTN_TOOL_MOUSE)	| BIT(BTN_TOOL_BRUSH)
-							  | BIT(BTN_TOOL_PENCIL) | BIT(BTN_TOOL_AIRBRUSH) | BIT(BTN_TOOL_LENS) | BIT(BTN_STYLUS2);
-			input_set_abs_params(input_dev, ABS_DISTANCE, 0, wacom->features->distance_max, 0, 0);
-			input_set_abs_params(input_dev, ABS_WHEEL, 0, 1023, 0, 0);
-			input_set_abs_params(input_dev, ABS_TILT_X, 0, 127, 0, 0);
-			input_set_abs_params(input_dev, ABS_TILT_Y, 0, 127, 0, 0);
-			input_set_abs_params(input_dev, ABS_RZ, -900, 899, 0, 0);
-			input_set_abs_params(input_dev, ABS_THROTTLE, -1023, 1023, 0, 0);
-			break;
-
-		case PL:
-			input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_STYLUS2) | BIT(BTN_TOOL_RUBBER);
-			break;
-	}
-
-	endpoint = &intf->cur_altsetting->endpoint[0].desc;
-
-	if (wacom->features->pktlen > 10)
-		BUG();
-
-	usb_fill_int_urb(wacom->irq, dev,
-			 usb_rcvintpipe(dev, endpoint->bEndpointAddress),
-			 wacom->data, wacom->features->pktlen,
-			 wacom->features->irq, wacom, endpoint->bInterval);
-	wacom->irq->transfer_dma = wacom->data_dma;
-	wacom->irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
-
-	input_register_device(wacom->dev);
-
-	/* Ask the tablet to report tablet data. Repeat until it succeeds */
-	do {
-		rep_data[0] = 2;
-		rep_data[1] = 2;
-		usb_set_report(intf, 3, 2, rep_data, 2);
-		usb_get_report(intf, 3, 2, rep_data, 2);
-	} while (rep_data[1] != 2 && limit++ < 5);
-
-	usb_set_intfdata(intf, wacom);
-	return 0;
-
-fail2:	usb_buffer_free(dev, 10, wacom->data, wacom->data_dma);
-fail1:	input_free_device(input_dev);
-	kfree(wacom);
-	return -ENOMEM;
-}
-
-static void wacom_disconnect(struct usb_interface *intf)
-{
-	struct wacom *wacom = usb_get_intfdata (intf);
-
-	usb_set_intfdata(intf, NULL);
-	if (wacom) {
-		usb_kill_urb(wacom->irq);
-		input_unregister_device(wacom->dev);
-		usb_free_urb(wacom->irq);
-		usb_buffer_free(interface_to_usbdev(intf), 10, wacom->data, wacom->data_dma);
-		kfree(wacom);
-	}
-}
-
-static struct usb_driver wacom_driver = {
-	.name =		"wacom",
-	.probe =	wacom_probe,
-	.disconnect =	wacom_disconnect,
-	.id_table =	wacom_ids,
-};
-
-static int __init wacom_init(void)
-{
-	int result = usb_register(&wacom_driver);
-	if (result == 0)
-		info(DRIVER_VERSION ":" DRIVER_DESC);
-	return result;
-}
-
-static void __exit wacom_exit(void)
-{
-	usb_deregister(&wacom_driver);
-}
-
-module_init(wacom_init);
-module_exit(wacom_exit);
diff --git a/drivers/usb/input/wacom.h b/drivers/usb/input/wacom.h
new file mode 100644
index 000000000000..832737b658cf
--- /dev/null
+++ b/drivers/usb/input/wacom.h
@@ -0,0 +1,132 @@
+/*
+ * drivers/usb/input/wacom.h
+ *
+ *  USB Wacom Graphire and Wacom Intuos tablet support
+ *
+ *  Copyright (c) 2000-2004 Vojtech Pavlik	<vojtech@ucw.cz>
+ *  Copyright (c) 2000 Andreas Bach Aaen	<abach@stofanet.dk>
+ *  Copyright (c) 2000 Clifford Wolf		<clifford@clifford.at>
+ *  Copyright (c) 2000 Sam Mosel		<sam.mosel@computer.org>
+ *  Copyright (c) 2000 James E. Blair		<corvus@gnu.org>
+ *  Copyright (c) 2000 Daniel Egger		<egger@suse.de>
+ *  Copyright (c) 2001 Frederic Lepied		<flepied@mandrakesoft.com>
+ *  Copyright (c) 2004 Panagiotis Issaris	<panagiotis.issaris@mech.kuleuven.ac.be>
+ *  Copyright (c) 2002-2006 Ping Cheng		<pingc@wacom.com>
+ *
+ *  ChangeLog:
+ *      v0.1 (vp)  - Initial release
+ *      v0.2 (aba) - Support for all buttons / combinations
+ *      v0.3 (vp)  - Support for Intuos added
+ *	v0.4 (sm)  - Support for more Intuos models, menustrip
+ *			relative mode, proximity.
+ *	v0.5 (vp)  - Big cleanup, nifty features removed,
+ *			they belong in userspace
+ *	v1.8 (vp)  - Submit URB only when operating, moved to CVS,
+ *			use input_report_key instead of report_btn and
+ *			other cleanups
+ *	v1.11 (vp) - Add URB ->dev setting for new kernels
+ *	v1.11 (jb) - Add support for the 4D Mouse & Lens
+ *	v1.12 (de) - Add support for two more inking pen IDs
+ *	v1.14 (vp) - Use new USB device id probing scheme.
+ *		     Fix Wacom Graphire mouse wheel
+ *	v1.18 (vp) - Fix mouse wheel direction
+ *		     Make mouse relative
+ *      v1.20 (fl) - Report tool id for Intuos devices
+ *                 - Multi tools support
+ *                 - Corrected Intuos protocol decoding (airbrush, 4D mouse, lens cursor...)
+ *                 - Add PL models support
+ *		   - Fix Wacom Graphire mouse wheel again
+ *	v1.21 (vp) - Removed protocol descriptions
+ *		   - Added MISC_SERIAL for tool serial numbers
+ *	      (gb) - Identify version on module load.
+ *    v1.21.1 (fl) - added Graphire2 support
+ *    v1.21.2 (fl) - added Intuos2 support
+ *                 - added all the PL ids
+ *    v1.21.3 (fl) - added another eraser id from Neil Okamoto
+ *                 - added smooth filter for Graphire from Peri Hankey
+ *                 - added PenPartner support from Olaf van Es
+ *                 - new tool ids from Ole Martin Bjoerndalen
+ *	v1.29 (pc) - Add support for more tablets
+ *		   - Fix pressure reporting
+ *	v1.30 (vp) - Merge 2.4 and 2.5 drivers
+ *		   - Since 2.5 now has input_sync(), remove MSC_SERIAL abuse
+ *		   - Cleanups here and there
+ *    v1.30.1 (pi) - Added Graphire3 support
+ *	v1.40 (pc) - Add support for several new devices, fix eraser reporting, ...
+ *	v1.43 (pc) - Added support for Cintiq 21UX
+ *		   - Fixed a Graphire bug
+ *		   - Merged wacom_intuos3_irq into wacom_intuos_irq
+ *	v1.44 (pc) - Added support for Graphire4, Cintiq 710, Intuos3 6x11, etc.
+ *		   - Report Device IDs
+ *      v1.45 (pc) - Added support for DTF 521, Intuos3 12x12 and 12x19
+ *                 - Minor data report fix
+ *      v1.46 (pc) - Split wacom.c into wacom_sys.c and wacom_wac.c,
+ *		   - where wacom_sys.c deals with system specific code,
+ * 		   - and wacom_wac.c deals with Wacom specific code
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#ifndef WACOM_H
+#define WACOM_H
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/usb/input.h>
+#include <asm/unaligned.h>
+
+/*
+ * Version Information
+ */
+#define DRIVER_VERSION "v1.46"
+#define DRIVER_AUTHOR "Vojtech Pavlik <vojtech@ucw.cz>"
+#define DRIVER_DESC "USB Wacom Graphire and Wacom Intuos tablet driver"
+#define DRIVER_LICENSE "GPL"
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE(DRIVER_LICENSE);
+
+#define USB_VENDOR_ID_WACOM	0x056a
+
+struct wacom {
+	dma_addr_t data_dma;
+	struct input_dev *dev;
+	struct usb_device *usbdev;
+	struct urb *irq;
+	struct wacom_wac * wacom_wac;
+	char phys[32];
+};
+
+struct wacom_combo {
+	struct wacom * wacom;
+	struct urb * urb;
+	struct pt_regs *regs;
+};
+
+extern int wacom_wac_irq(struct wacom_wac * wacom_wac, void * wcombo);
+extern void wacom_sys_irq(struct urb *urb, struct pt_regs *regs);
+extern void wacom_report_abs(void *wcombo, unsigned int abs_type, int abs_data);
+extern void wacom_report_rel(void *wcombo, unsigned int rel_type, int rel_data);
+extern void wacom_report_key(void *wcombo, unsigned int key_type, int key_data);
+extern void wacom_input_event(void *wcombo, unsigned int type, unsigned int code, int value);
+extern void wacom_input_regs(void *wcombo);
+extern void wacom_input_sync(void *wcombo);
+extern void wacom_init_input_dev(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_g4(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_g(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_i3(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_i(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_pl(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern void input_dev_pt(struct input_dev *input_dev, struct wacom_wac *wacom_wac);
+extern __u16 wacom_le16_to_cpu(unsigned char *data);
+extern __u16 wacom_be16_to_cpu(unsigned char *data);
+extern struct wacom_features * get_wacom_feature(const struct usb_device_id *id);
+extern const struct usb_device_id * get_device_table(void);
+
+#endif
diff --git a/drivers/usb/input/wacom_sys.c b/drivers/usb/input/wacom_sys.c
new file mode 100644
index 000000000000..7c3b52bdd9d6
--- /dev/null
+++ b/drivers/usb/input/wacom_sys.c
@@ -0,0 +1,315 @@
+/*
+ * drivers/usb/input/wacom_sys.c
+ *
+ *  USB Wacom Graphire and Wacom Intuos tablet support - system specific code
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
+#include "wacom.h"
+#include "wacom_wac.h"
+
+#define USB_REQ_GET_REPORT	0x01
+#define USB_REQ_SET_REPORT	0x09
+
+static int usb_get_report(struct usb_interface *intf, unsigned char type,
+				unsigned char id, void *buf, int size)
+{
+	return usb_control_msg(interface_to_usbdev(intf),
+		usb_rcvctrlpipe(interface_to_usbdev(intf), 0),
+		USB_REQ_GET_REPORT, USB_TYPE_CLASS | USB_RECIP_INTERFACE,
+		(type << 8) + id, intf->altsetting[0].desc.bInterfaceNumber,
+		buf, size, 100);
+}
+
+static int usb_set_report(struct usb_interface *intf, unsigned char type,
+				unsigned char id, void *buf, int size)
+{
+	return usb_control_msg(interface_to_usbdev(intf),
+		usb_sndctrlpipe(interface_to_usbdev(intf), 0),
+                USB_REQ_SET_REPORT, USB_TYPE_CLASS | USB_RECIP_INTERFACE,
+                (type << 8) + id, intf->altsetting[0].desc.bInterfaceNumber,
+		buf, size, 1000);
+}
+
+static struct input_dev * get_input_dev(struct wacom_combo *wcombo)
+{
+	return wcombo->wacom->dev;
+}
+
+void wacom_sys_irq(struct urb *urb, struct pt_regs *regs)
+{
+	struct wacom *wacom = urb->context;
+	struct wacom_combo wcombo;
+	int retval;
+
+	switch (urb->status) {
+	case 0:
+		/* success */
+		break;
+	case -ECONNRESET:
+	case -ENOENT:
+	case -ESHUTDOWN:
+		/* this urb is terminated, clean up */
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
+		return;
+	default:
+		dbg("%s - nonzero urb status received: %d", __FUNCTION__, urb->status);
+		goto exit;
+	}
+
+	wcombo.wacom = wacom;
+	wcombo.urb = urb;
+	wcombo.regs = regs;
+
+	if (wacom_wac_irq(wacom->wacom_wac, (void *)&wcombo))
+		input_sync(get_input_dev(&wcombo));
+
+ exit:
+	retval = usb_submit_urb (urb, GFP_ATOMIC);
+	if (retval)
+		err ("%s - usb_submit_urb failed with result %d",
+		     __FUNCTION__, retval);
+}
+
+void wacom_report_key(void *wcombo, unsigned int key_type, int key_data)
+{
+	input_report_key(get_input_dev((struct wacom_combo *)wcombo), key_type, key_data);
+	return;
+}
+
+void wacom_report_abs(void *wcombo, unsigned int abs_type, int abs_data)
+{
+	input_report_abs(get_input_dev((struct wacom_combo *)wcombo), abs_type, abs_data);
+	return;
+}
+
+void wacom_report_rel(void *wcombo, unsigned int rel_type, int rel_data)
+{
+	input_report_rel(get_input_dev((struct wacom_combo *)wcombo), rel_type, rel_data);
+	return;
+}
+
+void wacom_input_event(void *wcombo, unsigned int type, unsigned int code, int value)
+{
+	input_event(get_input_dev((struct wacom_combo *)wcombo), type, code, value);
+	return;
+}
+
+__u16 wacom_be16_to_cpu(unsigned char *data)
+{
+	__u16 value;
+	value = be16_to_cpu(*(__be16 *) data);
+	return value;
+}
+
+__u16 wacom_le16_to_cpu(unsigned char *data)
+{
+	__u16 value;
+	value = be16_to_cpu(*(__be16 *) data);
+	return value;
+}
+
+void wacom_input_regs(void *wcombo)
+{
+	input_regs(get_input_dev((struct wacom_combo *)wcombo), ((struct wacom_combo *)wcombo)->regs);
+	return;
+}
+
+void wacom_input_sync(void *wcombo)
+{
+	input_sync(get_input_dev((struct wacom_combo *)wcombo));
+	return;
+}
+
+static int wacom_open(struct input_dev *dev)
+{
+	struct wacom *wacom = dev->private;
+
+	wacom->irq->dev = wacom->usbdev;
+	if (usb_submit_urb(wacom->irq, GFP_KERNEL))
+		return -EIO;
+
+	return 0;
+}
+
+static void wacom_close(struct input_dev *dev)
+{
+	struct wacom *wacom = dev->private;
+
+	usb_kill_urb(wacom->irq);
+}
+
+void input_dev_g4(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->evbit[0] |= BIT(EV_MSC);
+	input_dev->mscbit[0] |= BIT(MSC_SERIAL);
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER);
+	input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7);
+}
+
+void input_dev_g(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->evbit[0] |= BIT(EV_REL);
+	input_dev->relbit[0] |= BIT(REL_WHEEL);
+	input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_LEFT) | BIT(BTN_RIGHT) | BIT(BTN_MIDDLE);
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER) | BIT(BTN_TOOL_MOUSE) | BIT(BTN_STYLUS2);
+	input_set_abs_params(input_dev, ABS_DISTANCE, 0, wacom_wac->features->distance_max, 0, 0);
+}
+
+void input_dev_i3(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_FINGER);
+	input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_0) | BIT(BTN_1) | BIT(BTN_2) | BIT(BTN_3) | BIT(BTN_4) | BIT(BTN_5) | BIT(BTN_6) | BIT(BTN_7);
+	input_set_abs_params(input_dev, ABS_RX, 0, 4097, 0, 0);
+	input_set_abs_params(input_dev, ABS_RY, 0, 4097, 0, 0);
+}
+
+void input_dev_i(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->evbit[0] |= BIT(EV_MSC) | BIT(EV_REL);
+	input_dev->mscbit[0] |= BIT(MSC_SERIAL);
+	input_dev->relbit[0] |= BIT(REL_WHEEL);
+	input_dev->keybit[LONG(BTN_LEFT)] |= BIT(BTN_LEFT) | BIT(BTN_RIGHT) | BIT(BTN_MIDDLE) | BIT(BTN_SIDE) | BIT(BTN_EXTRA);
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER) | BIT(BTN_TOOL_MOUSE)	| BIT(BTN_TOOL_BRUSH)
+		| BIT(BTN_TOOL_PENCIL) | BIT(BTN_TOOL_AIRBRUSH) | BIT(BTN_TOOL_LENS) | BIT(BTN_STYLUS2);
+	input_set_abs_params(input_dev, ABS_DISTANCE, 0, wacom_wac->features->distance_max, 0, 0);
+	input_set_abs_params(input_dev, ABS_WHEEL, 0, 1023, 0, 0);
+	input_set_abs_params(input_dev, ABS_TILT_X, 0, 127, 0, 0);
+	input_set_abs_params(input_dev, ABS_TILT_Y, 0, 127, 0, 0);
+	input_set_abs_params(input_dev, ABS_RZ, -900, 899, 0, 0);
+	input_set_abs_params(input_dev, ABS_THROTTLE, -1023, 1023, 0, 0);
+}
+
+void input_dev_pl(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_STYLUS2) | BIT(BTN_TOOL_RUBBER);
+}
+
+void input_dev_pt(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_RUBBER);
+}
+
+static int wacom_probe(struct usb_interface *intf, const struct usb_device_id *id)
+{
+	struct usb_device *dev = interface_to_usbdev(intf);
+	struct usb_endpoint_descriptor *endpoint;
+	struct wacom *wacom;
+	struct wacom_wac *wacom_wac;
+	struct input_dev *input_dev;
+	char rep_data[2], limit = 0;
+
+	wacom = kzalloc(sizeof(struct wacom), GFP_KERNEL);
+	wacom_wac = kzalloc(sizeof(struct wacom_wac), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!wacom || !input_dev || !wacom_wac)
+		goto fail1;
+
+	wacom_wac->data = usb_buffer_alloc(dev, 10, GFP_KERNEL, &wacom->data_dma);
+	if (!wacom_wac->data)
+		goto fail1;
+
+	wacom->irq = usb_alloc_urb(0, GFP_KERNEL);
+	if (!wacom->irq)
+		goto fail2;
+
+	wacom->usbdev = dev;
+	wacom->dev = input_dev;
+	usb_make_path(dev, wacom->phys, sizeof(wacom->phys));
+	strlcat(wacom->phys, "/input0", sizeof(wacom->phys));
+
+	wacom_wac->features = get_wacom_feature(id);
+	if (wacom_wac->features->pktlen > 10)
+		BUG();
+
+	input_dev->name = wacom_wac->features->name;
+	wacom->wacom_wac = wacom_wac;
+	usb_to_input_id(dev, &input_dev->id);
+
+	input_dev->cdev.dev = &intf->dev;
+	input_dev->private = wacom;
+	input_dev->open = wacom_open;
+	input_dev->close = wacom_close;
+
+	input_dev->evbit[0] |= BIT(EV_KEY) | BIT(EV_ABS);
+	input_dev->keybit[LONG(BTN_DIGI)] |= BIT(BTN_TOOL_PEN) | BIT(BTN_TOUCH) | BIT(BTN_STYLUS);
+	input_set_abs_params(input_dev, ABS_X, 0, wacom_wac->features->x_max, 4, 0);
+	input_set_abs_params(input_dev, ABS_Y, 0, wacom_wac->features->y_max, 4, 0);
+	input_set_abs_params(input_dev, ABS_PRESSURE, 0, wacom_wac->features->pressure_max, 0, 0);
+	input_dev->absbit[LONG(ABS_MISC)] |= BIT(ABS_MISC);
+
+	wacom_init_input_dev(input_dev, wacom_wac);
+
+	endpoint = &intf->cur_altsetting->endpoint[0].desc;
+
+	usb_fill_int_urb(wacom->irq, dev,
+			 usb_rcvintpipe(dev, endpoint->bEndpointAddress),
+			 wacom_wac->data, wacom_wac->features->pktlen,
+			 wacom_wac->features->irq, wacom, endpoint->bInterval);
+	wacom->irq->transfer_dma = wacom->data_dma;
+	wacom->irq->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+	input_register_device(wacom->dev);
+
+	/* Ask the tablet to report tablet data. Repeat until it succeeds */
+	do {
+		rep_data[0] = 2;
+		rep_data[1] = 2;
+		usb_set_report(intf, 3, 2, rep_data, 2);
+		usb_get_report(intf, 3, 2, rep_data, 2);
+	} while (rep_data[1] != 2 && limit++ < 5);
+
+	usb_set_intfdata(intf, wacom);
+	return 0;
+
+fail2:	usb_buffer_free(dev, 10, wacom_wac->data, wacom->data_dma);
+fail1:	input_free_device(input_dev);
+	kfree(wacom);
+	kfree(wacom_wac);
+	return -ENOMEM;
+}
+
+static void wacom_disconnect(struct usb_interface *intf)
+{
+	struct wacom *wacom = usb_get_intfdata (intf);
+
+	usb_set_intfdata(intf, NULL);
+	if (wacom) {
+		usb_kill_urb(wacom->irq);
+		input_unregister_device(wacom->dev);
+		usb_free_urb(wacom->irq);
+		usb_buffer_free(interface_to_usbdev(intf), 10, wacom->wacom_wac->data, wacom->data_dma);
+		kfree(wacom);
+		kfree(wacom->wacom_wac);
+	}
+}
+
+static struct usb_driver wacom_driver = {
+	.name =		"wacom",
+	.probe =	wacom_probe,
+	.disconnect =	wacom_disconnect,
+};
+
+static int __init wacom_init(void)
+{
+	int result;
+	wacom_driver.id_table = get_device_table();
+	result = usb_register(&wacom_driver);
+	if (result == 0)
+		info(DRIVER_VERSION ":" DRIVER_DESC);
+	return result;
+}
+
+static void __exit wacom_exit(void)
+{
+	usb_deregister(&wacom_driver);
+}
+
+module_init(wacom_init);
+module_exit(wacom_exit);
diff --git a/drivers/usb/input/wacom_wac.c b/drivers/usb/input/wacom_wac.c
new file mode 100644
index 000000000000..85d458c98b6e
--- /dev/null
+++ b/drivers/usb/input/wacom_wac.c
@@ -0,0 +1,646 @@
+/*
+ * drivers/usb/input/wacom_wac.c
+ *
+ *  USB Wacom Graphire and Wacom Intuos tablet support - Wacom specific code
+ *
+ */
+
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#include "wacom.h"
+#include "wacom_wac.h"
+
+static int wacom_penpartner_irq(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+
+	switch (data[0]) {
+		case 1:
+			wacom_input_regs(wcombo);
+			if (data[5] & 0x80) {
+				wacom->tool[0] = (data[5] & 0x20) ? BTN_TOOL_RUBBER : BTN_TOOL_PEN;
+				wacom->id[0] = (data[5] & 0x20) ? ERASER_DEVICE_ID : STYLUS_DEVICE_ID;
+				wacom_report_key(wcombo, wacom->tool[0], 1);
+				wacom_report_abs(wcombo, ABS_MISC, wacom->id[0]); /* report tool id */
+				wacom_report_abs(wcombo, ABS_X, wacom_le16_to_cpu(&data[1]));
+				wacom_report_abs(wcombo, ABS_Y, wacom_le16_to_cpu(&data[3]));
+				wacom_report_abs(wcombo, ABS_PRESSURE, (signed char)data[6] + 127);
+				wacom_report_key(wcombo, BTN_TOUCH, ((signed char)data[6] > -127));
+				wacom_report_key(wcombo, BTN_STYLUS, (data[5] & 0x40));
+			} else {
+				wacom_report_key(wcombo, wacom->tool[0], 0);
+				wacom_report_abs(wcombo, ABS_MISC, 0); /* report tool id */
+				wacom_report_abs(wcombo, ABS_PRESSURE, -1);
+				wacom_report_key(wcombo, BTN_TOUCH, 0);
+			}
+			break;
+		case 2:
+			wacom_input_regs(wcombo);
+			wacom_report_key(wcombo, BTN_TOOL_PEN, 1);
+			wacom_report_abs(wcombo, ABS_MISC, STYLUS_DEVICE_ID); /* report tool id */
+			wacom_report_abs(wcombo, ABS_X, wacom_le16_to_cpu(&data[1]));
+			wacom_report_abs(wcombo, ABS_Y, wacom_le16_to_cpu(&data[3]));
+			wacom_report_abs(wcombo, ABS_PRESSURE, (signed char)data[6] + 127);
+			wacom_report_key(wcombo, BTN_TOUCH, ((signed char)data[6] > -80) && !(data[5] & 0x20));
+			wacom_report_key(wcombo, BTN_STYLUS, (data[5] & 0x40));
+			break;
+		default:
+			printk(KERN_INFO "wacom_penpartner_irq: received unknown report #%d\n", data[0]);
+			return 0;
+        }
+	return 1;
+}
+
+static int wacom_pl_irq(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	int prox, id, pressure;
+
+	if (data[0] != 2) {
+		dbg("wacom_pl_irq: received unknown report #%d", data[0]);
+		return 0;
+	}
+
+	prox = data[1] & 0x40;
+
+	wacom_input_regs(wcombo);
+
+	id = ERASER_DEVICE_ID;
+	if (prox) {
+
+		pressure = (signed char)((data[7] << 1) | ((data[4] >> 2) & 1));
+		if (wacom->features->pressure_max > 255)
+			pressure = (pressure << 1) | ((data[4] >> 6) & 1);
+		pressure += (wacom->features->pressure_max + 1) / 2;
+
+		/*
+		 * if going from out of proximity into proximity select between the eraser
+		 * and the pen based on the state of the stylus2 button, choose eraser if
+		 * pressed else choose pen. if not a proximity change from out to in, send
+		 * an out of proximity for previous tool then a in for new tool.
+		 */
+		if (!wacom->tool[0]) {
+			/* Eraser bit set for DTF */
+			if (data[1] & 0x10)
+				wacom->tool[1] = BTN_TOOL_RUBBER;
+			else
+				/* Going into proximity select tool */
+				wacom->tool[1] = (data[4] & 0x20) ? BTN_TOOL_RUBBER : BTN_TOOL_PEN;
+		} else {
+			/* was entered with stylus2 pressed */
+			if (wacom->tool[1] == BTN_TOOL_RUBBER && !(data[4] & 0x20)) {
+				/* report out proximity for previous tool */
+				wacom_report_key(wcombo, wacom->tool[1], 0);
+				wacom_input_sync(wcombo);
+				wacom->tool[1] = BTN_TOOL_PEN;
+				return 0;
+			}
+		}
+		if (wacom->tool[1] != BTN_TOOL_RUBBER) {
+			/* Unknown tool selected default to pen tool */
+			wacom->tool[1] = BTN_TOOL_PEN;
+			id = STYLUS_DEVICE_ID;
+		}
+		wacom_report_key(wcombo, wacom->tool[1], prox); /* report in proximity for tool */
+		wacom_report_abs(wcombo, ABS_MISC, id); /* report tool id */
+		wacom_report_abs(wcombo, ABS_X, data[3] | (data[2] << 7) | ((data[1] & 0x03) << 14));
+		wacom_report_abs(wcombo, ABS_Y, data[6] | (data[5] << 7) | ((data[4] & 0x03) << 14));
+		wacom_report_abs(wcombo, ABS_PRESSURE, pressure);
+
+		wacom_report_key(wcombo, BTN_TOUCH, data[4] & 0x08);
+		wacom_report_key(wcombo, BTN_STYLUS, data[4] & 0x10);
+		/* Only allow the stylus2 button to be reported for the pen tool. */
+		wacom_report_key(wcombo, BTN_STYLUS2, (wacom->tool[1] == BTN_TOOL_PEN) && (data[4] & 0x20));
+	} else {
+		/* report proximity-out of a (valid) tool */
+		if (wacom->tool[1] != BTN_TOOL_RUBBER) {
+			/* Unknown tool selected default to pen tool */
+			wacom->tool[1] = BTN_TOOL_PEN;
+		}
+		wacom_report_key(wcombo, wacom->tool[1], prox);
+	}
+
+	wacom->tool[0] = prox; /* Save proximity state */
+	return 1;
+}
+
+static int wacom_ptu_irq(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	int id;
+
+	if (data[0] != 2) {
+		printk(KERN_INFO "wacom_ptu_irq: received unknown report #%d\n", data[0]);
+		return 0;
+	}
+
+	wacom_input_regs(wcombo);
+	if (data[1] & 0x04) {
+		wacom_report_key(wcombo, BTN_TOOL_RUBBER, data[1] & 0x20);
+		wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x08);
+		id = ERASER_DEVICE_ID;
+	} else {
+		wacom_report_key(wcombo, BTN_TOOL_PEN, data[1] & 0x20);
+		wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x01);
+		id = STYLUS_DEVICE_ID;
+	}
+	wacom_report_abs(wcombo, ABS_MISC, id); /* report tool id */
+	wacom_report_abs(wcombo, ABS_X, wacom_le16_to_cpu(&data[2]));
+	wacom_report_abs(wcombo, ABS_Y, wacom_le16_to_cpu(&data[4]));
+	wacom_report_abs(wcombo, ABS_PRESSURE, wacom_le16_to_cpu(&data[6]));
+	wacom_report_key(wcombo, BTN_STYLUS, data[1] & 0x02);
+	wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 0x10);
+	return 1;
+}
+
+static int wacom_graphire_irq(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	int x, y, id, rw;
+
+	if (data[0] != 2) {
+		dbg("wacom_graphire_irq: received unknown report #%d", data[0]);
+		return 0;
+	}
+
+	wacom_input_regs(wcombo);
+
+	id = STYLUS_DEVICE_ID;
+	if (data[1] & 0x10) { /* in prox */
+
+		switch ((data[1] >> 5) & 3) {
+
+			case 0:	/* Pen */
+				wacom->tool[0] = BTN_TOOL_PEN;
+				break;
+
+			case 1: /* Rubber */
+				wacom->tool[0] = BTN_TOOL_RUBBER;
+				id = ERASER_DEVICE_ID;
+				break;
+
+			case 2: /* Mouse with wheel */
+				wacom_report_key(wcombo, BTN_MIDDLE, data[1] & 0x04);
+				if (wacom->features->type == WACOM_G4) {
+					rw = data[7] & 0x04 ? (data[7] & 0x03)-4 : (data[7] & 0x03);
+					wacom_report_rel(wcombo, REL_WHEEL, -rw);
+				} else
+					wacom_report_rel(wcombo, REL_WHEEL, -(signed char) data[6]);
+				/* fall through */
+
+			case 3: /* Mouse without wheel */
+				wacom->tool[0] = BTN_TOOL_MOUSE;
+				id = CURSOR_DEVICE_ID;
+				wacom_report_key(wcombo, BTN_LEFT, data[1] & 0x01);
+				wacom_report_key(wcombo, BTN_RIGHT, data[1] & 0x02);
+				if (wacom->features->type == WACOM_G4)
+					wacom_report_abs(wcombo, ABS_DISTANCE, data[6]);
+				else
+					wacom_report_abs(wcombo, ABS_DISTANCE, data[7]);
+				break;
+		}
+	}
+
+	if (data[1] & 0x90) {
+		x = wacom_le16_to_cpu(&data[2]);
+		y = wacom_le16_to_cpu(&data[4]);
+		wacom_report_abs(wcombo, ABS_X, x);
+		wacom_report_abs(wcombo, ABS_Y, y);
+		if (wacom->tool[0] != BTN_TOOL_MOUSE) {
+			wacom_report_abs(wcombo, ABS_PRESSURE, data[6] | ((data[7] & 0x01) << 8));
+			wacom_report_key(wcombo, BTN_TOUCH, data[1] & 0x01);
+			wacom_report_key(wcombo, BTN_STYLUS, data[1] & 0x02);
+			wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 0x04);
+		}
+	}
+
+	if (data[1] & 0x10)
+		wacom_report_abs(wcombo, ABS_MISC, id); /* report tool id */
+	else
+		wacom_report_abs(wcombo, ABS_MISC, 0); /* reset tool id */
+	wacom_report_key(wcombo, wacom->tool[0], data[1] & 0x10);
+	wacom_input_sync(wcombo);
+
+	/* send pad data */
+	if (wacom->features->type == WACOM_G4) {
+		if ( (wacom->serial[1] & 0xc0) != (data[7] & 0xf8) ) {
+			wacom->id[1] = 1;
+			wacom->serial[1] = (data[7] & 0xf8);
+			wacom_report_key(wcombo, BTN_0, (data[7] & 0x40));
+			wacom_report_key(wcombo, BTN_4, (data[7] & 0x80));
+			rw = ((data[7] & 0x18) >> 3) - ((data[7] & 0x20) >> 3);
+			wacom_report_rel(wcombo, REL_WHEEL, rw);
+			wacom_report_key(wcombo, BTN_TOOL_FINGER, 0xf0);
+			wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0);
+		} else if (wacom->id[1]) {
+			wacom->id[1] = 0;
+			wacom_report_key(wcombo, BTN_TOOL_FINGER, 0);
+			wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xf0);
+		}
+	}
+	return 1;
+}
+
+static int wacom_intuos_inout(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	int idx;
+
+	/* tool number */
+	idx = data[1] & 0x01;
+
+	/* Enter report */
+	if ((data[1] & 0xfc) == 0xc0) {
+		/* serial number of the tool */
+		wacom->serial[idx] = ((data[3] & 0x0f) << 28) +
+			(data[4] << 20) + (data[5] << 12) +
+			(data[6] << 4) + (data[7] >> 4);
+
+		wacom->id[idx] = (data[2] << 4) | (data[3] >> 4);
+		switch (wacom->id[idx]) {
+			case 0x812: /* Inking pen */
+			case 0x801: /* Intuos3 Inking pen */
+			case 0x012:
+				wacom->tool[idx] = BTN_TOOL_PENCIL;
+				break;
+			case 0x822: /* Pen */
+			case 0x842:
+			case 0x852:
+			case 0x823: /* Intuos3 Grip Pen */
+			case 0x813: /* Intuos3 Classic Pen */
+			case 0x885: /* Intuos3 Marker Pen */
+			case 0x022:
+				wacom->tool[idx] = BTN_TOOL_PEN;
+				break;
+			case 0x832: /* Stroke pen */
+			case 0x032:
+				wacom->tool[idx] = BTN_TOOL_BRUSH;
+				break;
+			case 0x007: /* Mouse 4D and 2D */
+		        case 0x09c:
+			case 0x094:
+			case 0x017: /* Intuos3 2D Mouse */
+				wacom->tool[idx] = BTN_TOOL_MOUSE;
+				break;
+			case 0x096: /* Lens cursor */
+			case 0x097: /* Intuos3 Lens cursor */
+				wacom->tool[idx] = BTN_TOOL_LENS;
+				break;
+			case 0x82a: /* Eraser */
+			case 0x85a:
+		        case 0x91a:
+			case 0xd1a:
+			case 0x0fa:
+			case 0x82b: /* Intuos3 Grip Pen Eraser */
+			case 0x81b: /* Intuos3 Classic Pen Eraser */
+			case 0x91b: /* Intuos3 Airbrush Eraser */
+				wacom->tool[idx] = BTN_TOOL_RUBBER;
+				break;
+			case 0xd12:
+			case 0x912:
+			case 0x112:
+			case 0x913: /* Intuos3 Airbrush */
+				wacom->tool[idx] = BTN_TOOL_AIRBRUSH;
+				break;
+			default: /* Unknown tool */
+				wacom->tool[idx] = BTN_TOOL_PEN;
+		}
+		/* only large I3 support Lens Cursor */
+		if(!((wacom->tool[idx] == BTN_TOOL_LENS) &&
+				(wacom->features->type == INTUOS3))) {
+			wacom_report_abs(wcombo, ABS_MISC, wacom->id[idx]); /* report tool id */
+			wacom_report_key(wcombo, wacom->tool[idx], 1);
+			wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
+			return 2;
+		}
+		return 1;
+	}
+
+	/* Exit report */
+	if ((data[1] & 0xfe) == 0x80) {
+		wacom_report_key(wcombo, wacom->tool[idx], 0);
+		wacom_report_abs(wcombo, ABS_MISC, 0); /* reset tool id */
+		wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
+		return 2;
+	}
+	return 0;
+}
+
+static void wacom_intuos_general(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	unsigned int t;
+
+	/* general pen packet */
+	if ((data[1] & 0xb8) == 0xa0) {
+		t = (data[6] << 2) | ((data[7] >> 6) & 3);
+		wacom_report_abs(wcombo, ABS_PRESSURE, t);
+		wacom_report_abs(wcombo, ABS_TILT_X,
+				((data[7] << 1) & 0x7e) | (data[8] >> 7));
+		wacom_report_abs(wcombo, ABS_TILT_Y, data[8] & 0x7f);
+		wacom_report_key(wcombo, BTN_STYLUS, data[1] & 2);
+		wacom_report_key(wcombo, BTN_STYLUS2, data[1] & 4);
+		wacom_report_key(wcombo, BTN_TOUCH, t > 10);
+	}
+
+	/* airbrush second packet */
+	if ((data[1] & 0xbc) == 0xb4) {
+		wacom_report_abs(wcombo, ABS_WHEEL,
+				(data[6] << 2) | ((data[7] >> 6) & 3));
+		wacom_report_abs(wcombo, ABS_TILT_X,
+				((data[7] << 1) & 0x7e) | (data[8] >> 7));
+		wacom_report_abs(wcombo, ABS_TILT_Y, data[8] & 0x7f);
+	}
+	return;
+}
+
+static int wacom_intuos_irq(struct wacom_wac *wacom, void *wcombo)
+{
+	unsigned char *data = wacom->data;
+	unsigned int t;
+	int idx, result;
+
+	if (data[0] != 2 && data[0] != 5 && data[0] != 6 && data[0] != 12) {
+		dbg("wacom_intuos_irq: received unknown report #%d", data[0]);
+                return 0;
+	}
+
+	wacom_input_regs(wcombo);
+
+	/* tool number */
+	idx = data[1] & 0x01;
+
+	/* pad packets. Works as a second tool and is always in prox */
+	if (data[0] == 12) {
+		/* initiate the pad as a device */
+		if (wacom->tool[1] != BTN_TOOL_FINGER)
+			wacom->tool[1] = BTN_TOOL_FINGER;
+
+		wacom_report_key(wcombo, BTN_0, (data[5] & 0x01));
+		wacom_report_key(wcombo, BTN_1, (data[5] & 0x02));
+		wacom_report_key(wcombo, BTN_2, (data[5] & 0x04));
+		wacom_report_key(wcombo, BTN_3, (data[5] & 0x08));
+		wacom_report_key(wcombo, BTN_4, (data[6] & 0x01));
+		wacom_report_key(wcombo, BTN_5, (data[6] & 0x02));
+		wacom_report_key(wcombo, BTN_6, (data[6] & 0x04));
+		wacom_report_key(wcombo, BTN_7, (data[6] & 0x08));
+		wacom_report_abs(wcombo, ABS_RX, ((data[1] & 0x1f) << 8) | data[2]);
+		wacom_report_abs(wcombo, ABS_RY, ((data[3] & 0x1f) << 8) | data[4]);
+
+		if((data[5] & 0x0f) | (data[6] & 0x0f) | (data[1] & 0x1f) | data[2])
+			wacom_report_key(wcombo, wacom->tool[1], 1);
+		else
+			wacom_report_key(wcombo, wacom->tool[1], 0);
+		wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, 0xffffffff);
+                return 1;
+	}
+
+	/* process in/out prox events */
+	result = wacom_intuos_inout(wacom, wcombo);
+	if (result)
+                return result-1;
+
+	/* Cintiq doesn't send data when RDY bit isn't set */
+	if ((wacom->features->type == CINTIQ) && !(data[1] & 0x40))
+                 return 0;
+
+	if (wacom->features->type >= INTUOS3) {
+		wacom_report_abs(wcombo, ABS_X, (data[2] << 9) | (data[3] << 1) | ((data[9] >> 1) & 1));
+		wacom_report_abs(wcombo, ABS_Y, (data[4] << 9) | (data[5] << 1) | (data[9] & 1));
+		wacom_report_abs(wcombo, ABS_DISTANCE, ((data[9] >> 2) & 0x3f));
+	} else {
+		wacom_report_abs(wcombo, ABS_X, wacom_be16_to_cpu(&data[2]));
+		wacom_report_abs(wcombo, ABS_Y, wacom_be16_to_cpu(&data[4]));
+		wacom_report_abs(wcombo, ABS_DISTANCE, ((data[9] >> 3) & 0x1f));
+	}
+
+	/* process general packets */
+	wacom_intuos_general(wacom, wcombo);
+
+	/* 4D mouse, 2D mouse, marker pen rotation, or Lens cursor packets */
+	if ((data[1] & 0xbc) == 0xa8 || (data[1] & 0xbe) == 0xb0) {
+
+		if (data[1] & 0x02) {
+			/* Rotation packet */
+			if (wacom->features->type >= INTUOS3) {
+				/* I3 marker pen rotation reported as wheel
+				 * due to valuator limitation
+				 */
+				t = (data[6] << 3) | ((data[7] >> 5) & 7);
+				t = (data[7] & 0x20) ? ((t > 900) ? ((t-1) / 2 - 1350) :
+					((t-1) / 2 + 450)) : (450 - t / 2) ;
+				wacom_report_abs(wcombo, ABS_WHEEL, t);
+			} else {
+				/* 4D mouse rotation packet */
+				t = (data[6] << 3) | ((data[7] >> 5) & 7);
+				wacom_report_abs(wcombo, ABS_RZ, (data[7] & 0x20) ?
+					((t - 1) / 2) : -t / 2);
+			}
+
+		} else if (!(data[1] & 0x10) && wacom->features->type < INTUOS3) {
+			/* 4D mouse packet */
+			wacom_report_key(wcombo, BTN_LEFT,   data[8] & 0x01);
+			wacom_report_key(wcombo, BTN_MIDDLE, data[8] & 0x02);
+			wacom_report_key(wcombo, BTN_RIGHT,  data[8] & 0x04);
+
+			wacom_report_key(wcombo, BTN_SIDE,   data[8] & 0x20);
+			wacom_report_key(wcombo, BTN_EXTRA,  data[8] & 0x10);
+			t = (data[6] << 2) | ((data[7] >> 6) & 3);
+			wacom_report_abs(wcombo, ABS_THROTTLE, (data[8] & 0x08) ? -t : t);
+
+		} else if (wacom->tool[idx] == BTN_TOOL_MOUSE) {
+			/* 2D mouse packet */
+			wacom_report_key(wcombo, BTN_LEFT,   data[8] & 0x04);
+			wacom_report_key(wcombo, BTN_MIDDLE, data[8] & 0x08);
+			wacom_report_key(wcombo, BTN_RIGHT,  data[8] & 0x10);
+			wacom_report_rel(wcombo, REL_WHEEL, (data[8] & 0x01)
+						 - ((data[8] & 0x02) >> 1));
+
+			/* I3 2D mouse side buttons */
+			if (wacom->features->type == INTUOS3) {
+				wacom_report_key(wcombo, BTN_SIDE,   data[8] & 0x40);
+				wacom_report_key(wcombo, BTN_EXTRA,  data[8] & 0x20);
+			}
+
+		} else if (wacom->features->type < INTUOS3) {
+			/* Lens cursor packets */
+			wacom_report_key(wcombo, BTN_LEFT,   data[8] & 0x01);
+			wacom_report_key(wcombo, BTN_MIDDLE, data[8] & 0x02);
+			wacom_report_key(wcombo, BTN_RIGHT,  data[8] & 0x04);
+			wacom_report_key(wcombo, BTN_SIDE,   data[8] & 0x10);
+			wacom_report_key(wcombo, BTN_EXTRA,  data[8] & 0x08);
+		}
+	}
+
+	wacom_report_abs(wcombo, ABS_MISC, wacom->id[idx]); /* report tool id */
+	wacom_report_key(wcombo, wacom->tool[idx], 1);
+	wacom_input_event(wcombo, EV_MSC, MSC_SERIAL, wacom->serial[idx]);
+	return 1;
+}
+
+int wacom_wac_irq(struct wacom_wac *wacom_wac, void *wcombo)
+{
+	switch (wacom_wac->features->type) {
+		case PENPARTNER:
+			return (wacom_penpartner_irq(wacom_wac, wcombo));
+			break;
+		case PL:
+			return (wacom_pl_irq(wacom_wac, wcombo));
+			break;
+		case WACOM_G4:
+		case GRAPHIRE:
+			return (wacom_graphire_irq(wacom_wac, wcombo));
+			break;
+		case PTU:
+			return (wacom_ptu_irq(wacom_wac, wcombo));
+			break;
+		case INTUOS:
+		case INTUOS3:
+		case INTUOS3L:
+		case CINTIQ:
+			return (wacom_intuos_irq(wacom_wac, wcombo));
+			break;
+		default:
+			return 0;
+	}
+	return 0;
+}
+
+void wacom_init_input_dev(struct input_dev *input_dev, struct wacom_wac *wacom_wac)
+{
+	switch (wacom_wac->features->type) {
+		case WACOM_G4:
+			input_dev_g4(input_dev, wacom_wac);
+			/* fall through */
+		case GRAPHIRE:
+			input_dev_g(input_dev, wacom_wac);
+			break;
+		case INTUOS3:
+		case INTUOS3L:
+		case CINTIQ:
+			input_dev_i3(input_dev, wacom_wac);
+			/* fall through */
+		case INTUOS:
+			input_dev_i(input_dev, wacom_wac);
+			break;
+		case PL:
+		case PTU:
+			input_dev_pl(input_dev, wacom_wac);
+			break;
+		case PENPARTNER:
+			input_dev_pt(input_dev, wacom_wac);
+			break;
+	}
+	return;
+}
+
+static struct wacom_features wacom_features[] = {
+	{ "Wacom Penpartner",    7,   5040,  3780,  255, 32, PENPARTNER,	wacom_sys_irq },
+        { "Wacom Graphire",      8,  10206,  7422,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Graphire2 4x5", 8,  10206,  7422,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Graphire2 5x7", 8,  13918, 10206,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Graphire3",     8,  10208,  7424,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Graphire3 6x8", 8,  16704, 12064,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Graphire4 4x5", 8,  10208,  7424,  511, 32, WACOM_G4,	wacom_sys_irq },
+	{ "Wacom Graphire4 6x8", 8,  16704, 12064,  511, 32, WACOM_G4,	wacom_sys_irq },
+	{ "Wacom Volito",        8,   5104,  3712,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom PenStation2",   8,   3250,  2320,  255, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Volito2 4x5",   8,   5104,  3712,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Volito2 2x3",   8,   3248,  2320,  511, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom PenPartner2",   8,   3250,  2320,  255, 32, GRAPHIRE,	wacom_sys_irq },
+	{ "Wacom Intuos 4x5",   10,  12700, 10600, 1023, 15, INTUOS,	wacom_sys_irq},
+	{ "Wacom Intuos 6x8",   10,  20320, 16240, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos 9x12",  10,  30480, 24060, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos 12x12", 10,  30480, 31680, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos 12x18", 10,  45720, 31680, 1023, 15, INTUOS,	wacom_sys_irq},
+	{ "Wacom PL400",         8,   5408,  4056,  255, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL500",         8,   6144,  4608,  255, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL600",         8,   6126,  4604,  255, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL600SX",       8,   6260,  5016,  255, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL550",         8,   6144,  4608,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL800",         8,   7220,  5780,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL700",         8,   6758,  5406,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom PL510",         8,   6282,  4762,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom DTU710",        8,  34080, 27660,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom DTF521",        8,   6282,  4762,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom DTF720",        8,   6858,  5506,  511, 32, PL,	wacom_sys_irq },
+	{ "Wacom Cintiq Partner",8,  20480, 15360,  511, 32, PTU,	wacom_sys_irq },
+	{ "Wacom Intuos2 4x5",   10, 12700, 10600, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos2 6x8",   10, 20320, 16240, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos2 9x12",  10, 30480, 24060, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos2 12x12", 10, 30480, 31680, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos2 12x18", 10, 45720, 31680, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ "Wacom Intuos3 4x5",   10, 25400, 20320, 1023, 15, INTUOS3,	wacom_sys_irq },
+	{ "Wacom Intuos3 6x8",   10, 40640, 30480, 1023, 15, INTUOS3,	wacom_sys_irq },
+	{ "Wacom Intuos3 9x12",  10, 60960, 45720, 1023, 15, INTUOS3,	wacom_sys_irq },
+	{ "Wacom Intuos3 12x12", 10, 60960, 60960, 1023, 15, INTUOS3L,	wacom_sys_irq },
+	{ "Wacom Intuos3 12x19", 10, 97536, 60960, 1023, 15, INTUOS3L,	wacom_sys_irq },
+	{ "Wacom Intuos3 6x11",  10, 54204, 31750, 1023, 15, INTUOS3,	wacom_sys_irq },
+	{ "Wacom Cintiq 21UX",   10, 87200, 65600, 1023, 15, CINTIQ,	wacom_sys_irq },
+	{ "Wacom Intuos2 6x8",   10, 20320, 16240, 1023, 15, INTUOS,	wacom_sys_irq },
+	{ }
+};
+
+static struct usb_device_id wacom_ids[] = {
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x00) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x10) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x11) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x12) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x13) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x14) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x15) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x16) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x60) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x61) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x62) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x63) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x64) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x20) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x21) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x22) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x23) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x24) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x30) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x31) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x32) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x33) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x34) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x35) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x37) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x38) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x39) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xC0) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xC4) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x03) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x41) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x42) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x43) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x44) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x45) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB0) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB1) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB2) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB3) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB4) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0xB5) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x3F) },
+	{ USB_DEVICE(USB_VENDOR_ID_WACOM, 0x47) },
+	{ }
+};
+
+const struct usb_device_id * get_device_table(void) {
+        const struct usb_device_id * id_table = wacom_ids;
+        return id_table;
+}
+
+struct wacom_features * get_wacom_feature(const struct usb_device_id * id) {
+        int index = id - wacom_ids;
+        struct wacom_features *wf = &wacom_features[index];
+        return wf;
+}
+
+MODULE_DEVICE_TABLE(usb, wacom_ids);
diff --git a/drivers/usb/input/wacom_wac.h b/drivers/usb/input/wacom_wac.h
new file mode 100644
index 000000000000..ceae7bf59d9f
--- /dev/null
+++ b/drivers/usb/input/wacom_wac.h
@@ -0,0 +1,48 @@
+/*
+ * drivers/usb/input/wacom_wac.h
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+#ifndef WACOM_WAC_H
+#define WACOM_WAC_H
+
+#define STYLUS_DEVICE_ID	0x02
+#define CURSOR_DEVICE_ID	0x06
+#define ERASER_DEVICE_ID	0x0A
+
+enum {
+	PENPARTNER = 0,
+	GRAPHIRE,
+	WACOM_G4,
+	PTU,
+	PL,
+	INTUOS,
+	INTUOS3,
+	INTUOS3L,
+	CINTIQ,
+	MAX_TYPE
+};
+
+struct wacom_features {
+	char *name;
+	int pktlen;
+	int x_max;
+	int y_max;
+	int pressure_max;
+	int distance_max;
+	int type;
+	usb_complete_t irq;
+};
+
+struct wacom_wac {
+	signed char *data;
+        int tool[2];
+        int id[2];
+        __u32 serial[2];
+	struct wacom_features *features;
+};
+
+#endif
-- 
cgit v1.2.1


From eb6d8c2d145a864aedd874b4226762c5ef0d7a77 Mon Sep 17 00:00:00 2001
From: Hermann Kneissel <hermann.kneissel@gmx.net>
Date: Tue, 11 Jul 2006 19:41:33 +0200
Subject: USB: garmin_gps support for new generation of gps receivers

The attached patch adds support for the new generation of gps receivers (eg.
GPSmap 60Cx) to garmin_gps.c.

Signed-off-by: Hermann Kneissel <herkne@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/garmin_gps.c | 219 +++++++++++++++++++++++++++++-----------
 1 file changed, 162 insertions(+), 57 deletions(-)

diff --git a/drivers/usb/serial/garmin_gps.c b/drivers/usb/serial/garmin_gps.c
index 727852634be9..4b1196a8b09e 100644
--- a/drivers/usb/serial/garmin_gps.c
+++ b/drivers/usb/serial/garmin_gps.c
@@ -1,7 +1,7 @@
 /*
  * Garmin GPS driver
  *
- * Copyright (C) 2004 Hermann Kneissel herkne@users.sourceforge.net
+ * Copyright (C) 2006 Hermann Kneissel herkne@users.sourceforge.net
  *
  * The latest version of the driver can be found at
  * http://sourceforge.net/projects/garmin-gps/
@@ -37,6 +37,8 @@
 #include <linux/usb.h>
 #include <linux/usb/serial.h>
 
+#include <linux/version.h>
+
 /* the mode to be set when the port ist opened */
 static int initial_mode = 1;
 
@@ -50,7 +52,7 @@ static int debug = 0;
  */
 
 #define VERSION_MAJOR	0
-#define VERSION_MINOR	23
+#define VERSION_MINOR	28
 
 #define _STR(s) #s
 #define _DRIVER_VERSION(a,b) "v" _STR(a) "." _STR(b)
@@ -164,7 +166,8 @@ struct garmin_data {
 #define FLAGS_SESSION_REPLY1_SEEN 0x0080
 #define FLAGS_SESSION_REPLY2_SEEN 0x0040
 #define FLAGS_BULK_IN_ACTIVE      0x0020
-#define FLAGS_THROTTLED           0x0010
+#define FLAGS_BULK_IN_RESTART     0x0010
+#define FLAGS_THROTTLED           0x0008
 #define CLEAR_HALT_REQUIRED       0x0001
 
 #define FLAGS_QUEUING             0x0100
@@ -224,7 +227,7 @@ static struct usb_driver garmin_driver = {
 	.probe =	usb_serial_probe,
 	.disconnect =	usb_serial_disconnect,
 	.id_table =	id_table,
-	.no_dynamic_id = 	1,
+	.no_dynamic_id = 1,
 };
 
 
@@ -270,7 +273,7 @@ static inline int isAbortTrfCmnd(const unsigned char *buf)
 
 
 static void send_to_tty(struct usb_serial_port *port,
-                        char *data, unsigned int actual_length)
+			char *data, unsigned int actual_length)
 {
 	struct tty_struct *tty = port->tty;
 
@@ -294,15 +297,15 @@ static void send_to_tty(struct usb_serial_port *port,
  * queue a received (usb-)packet for later processing
  */
 static int pkt_add(struct garmin_data * garmin_data_p,
-                   unsigned char *data, unsigned int data_length)
+		   unsigned char *data, unsigned int data_length)
 {
+	int state = 0;
 	int result = 0;
 	unsigned long flags;
 	struct garmin_packet *pkt;
 
 	/* process only packets containg data ... */
 	if (data_length) {
-		garmin_data_p->flags |= FLAGS_QUEUING;
 		pkt = kmalloc(sizeof(struct garmin_packet)+data_length,
 		              GFP_ATOMIC);
 		if (pkt == NULL) {
@@ -313,14 +316,16 @@ static int pkt_add(struct garmin_data * garmin_data_p,
 		memcpy(pkt->data, data, data_length);
 
 		spin_lock_irqsave(&garmin_data_p->lock, flags);
+		garmin_data_p->flags |= FLAGS_QUEUING;
 		result = list_empty(&garmin_data_p->pktlist);
 		pkt->seq = garmin_data_p->seq_counter++;
 		list_add_tail(&pkt->list, &garmin_data_p->pktlist);
+		state = garmin_data_p->state;
 		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 		/* in serial mode, if someone is waiting for data from
 		   the device, iconvert and send the next packet to tty. */
-		if (result && (garmin_data_p->state == STATE_GSP_WAIT_DATA)) {
+		if (result && (state == STATE_GSP_WAIT_DATA)) {
 			gsp_next_packet(garmin_data_p);
 		}
 	}
@@ -370,9 +375,9 @@ static void pkt_clear(struct garmin_data * garmin_data_p)
 static int gsp_send_ack(struct garmin_data * garmin_data_p, __u8 pkt_id)
 {
 	__u8 pkt[10];
-        __u8 cksum = 0;
-        __u8 *ptr = pkt;
-        unsigned  l = 0;
+	__u8 cksum = 0;
+	__u8 *ptr = pkt;
+	unsigned  l = 0;
 
 	dbg("%s - pkt-id: 0x%X.", __FUNCTION__, 0xFF & pkt_id);
 
@@ -416,7 +421,7 @@ static int gsp_send_ack(struct garmin_data * garmin_data_p, __u8 pkt_id)
 static int gsp_rec_packet(struct garmin_data * garmin_data_p, int count)
 {
 	const __u8* recpkt = garmin_data_p->inbuffer+GSP_INITIAL_OFFSET;
-        __le32 *usbdata = (__le32 *) garmin_data_p->inbuffer;
+	__le32 *usbdata = (__le32 *) garmin_data_p->inbuffer;
 
 	int cksum = 0;
 	int n = 0;
@@ -447,11 +452,11 @@ static int gsp_rec_packet(struct garmin_data * garmin_data_p, int count)
 		n++;
 	}
 
-       if ((0xff & (cksum + *recpkt)) != 0) {
-                dbg("%s - invalid checksum, expected %02x, got %02x",
-                        __FUNCTION__, 0xff & -cksum, 0xff & *recpkt);
-                return -EINVPKT;
-        }
+	if ((0xff & (cksum + *recpkt)) != 0) {
+		dbg("%s - invalid checksum, expected %02x, got %02x",
+			__FUNCTION__, 0xff & -cksum, 0xff & *recpkt);
+		return -EINVPKT;
+	}
 
 	usbdata[0] = __cpu_to_le32(GARMIN_LAYERID_APPL);
 	usbdata[1] = __cpu_to_le32(pktid);
@@ -491,20 +496,28 @@ static int gsp_rec_packet(struct garmin_data * garmin_data_p, int count)
  */
 
 static int gsp_receive(struct garmin_data * garmin_data_p,
-                       const unsigned char *buf, int count)
+		       const unsigned char *buf, int count)
 {
+	unsigned long flags;
 	int offs = 0;
 	int ack_or_nak_seen = 0;
 	int i = 0;
-	__u8 *dest = garmin_data_p->inbuffer;
-	int size = garmin_data_p->insize;
+	__u8 *dest;
+	int size;
 	// dleSeen: set if last byte read was a DLE
-	int dleSeen = garmin_data_p->flags & FLAGS_GSP_DLESEEN;
+	int dleSeen;
 	// skip: if set, skip incoming data until possible start of
 	//       new packet
-	int skip = garmin_data_p->flags & FLAGS_GSP_SKIP;
+	int skip;
 	__u8 data;
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
+	dest = garmin_data_p->inbuffer;
+	size = garmin_data_p->insize;
+	dleSeen = garmin_data_p->flags & FLAGS_GSP_DLESEEN;
+	skip = garmin_data_p->flags & FLAGS_GSP_SKIP;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
+
 	dbg("%s - dle=%d skip=%d size=%d count=%d",
 		__FUNCTION__, dleSeen, skip, size, count);
 
@@ -572,6 +585,8 @@ static int gsp_receive(struct garmin_data * garmin_data_p,
 		}
 	}
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
+
 	garmin_data_p->insize = size;
 
 	// copy flags back to structure
@@ -587,6 +602,11 @@ static int gsp_receive(struct garmin_data * garmin_data_p,
 
 	if (ack_or_nak_seen) {
 		garmin_data_p->state = STATE_GSP_WAIT_DATA;
+	}
+
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
+
+	if (ack_or_nak_seen) {
 		gsp_next_packet(garmin_data_p);
 	}
 
@@ -676,7 +696,7 @@ static int gsp_send(struct garmin_data * garmin_data_p,
 	src = garmin_data_p->outbuffer+GARMIN_PKTHDR_LENGTH;
 	if (k > (GARMIN_PKTHDR_LENGTH-2)) {
 		/* can't add stuffing DLEs in place, move data to end 
-                   of buffer ... */
+		   of buffer ... */
 		dst = garmin_data_p->outbuffer+GPS_OUT_BUFSIZ-datalen;
 		memcpy(dst, src, datalen);
 		src = dst;
@@ -755,8 +775,9 @@ static void gsp_next_packet(struct garmin_data * garmin_data_p)
  * or even incomplete packets
  */
 static int nat_receive(struct garmin_data * garmin_data_p,
-                       const unsigned char *buf, int count)
+		       const unsigned char *buf, int count)
 {
+	unsigned long flags;
 	__u8 * dest;
 	int offs = 0;
 	int result = count;
@@ -803,7 +824,9 @@ static int nat_receive(struct garmin_data * garmin_data_p,
 				/* if this was an abort-transfer command,
 				   flush all queued data. */
 				if (isAbortTrfCmnd(garmin_data_p->inbuffer)) {
+					spin_lock_irqsave(&garmin_data_p->lock, flags);
 					garmin_data_p->flags |= FLAGS_DROP_DATA;
+					spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 					pkt_clear(garmin_data_p);
 				}
 			}
@@ -839,12 +862,15 @@ static void priv_status_resp(struct usb_serial_port *port)
 
 static int process_resetdev_request(struct usb_serial_port *port)
 {
+	unsigned long flags;
 	int status;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->flags &= ~(CLEAR_HALT_REQUIRED);
 	garmin_data_p->state = STATE_RESET;
 	garmin_data_p->serial_num = 0;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 	usb_kill_urb (port->interrupt_in_urb);
 	dbg("%s - usb_reset_device", __FUNCTION__ );
@@ -862,6 +888,7 @@ static int process_resetdev_request(struct usb_serial_port *port)
  */
 static int garmin_clear(struct garmin_data * garmin_data_p)
 {
+	unsigned long flags;
 	int status = 0;
 
 	struct usb_serial_port *port = garmin_data_p->port;
@@ -875,8 +902,10 @@ static int garmin_clear(struct garmin_data * garmin_data_p)
 	/* flush all queued data */
 	pkt_clear(garmin_data_p);
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->insize = 0;
 	garmin_data_p->outsize = 0;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 	return status;
 }
@@ -888,6 +917,7 @@ static int garmin_clear(struct garmin_data * garmin_data_p)
 
 static int garmin_init_session(struct usb_serial_port *port)
 {
+	unsigned long flags;
 	struct usb_serial *serial = port->serial;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 	int status = 0;
@@ -913,7 +943,9 @@ static int garmin_init_session(struct usb_serial_port *port)
 
 		if (status >= 0) {
 
+			spin_lock_irqsave(&garmin_data_p->lock, flags);
 			garmin_data_p->ignorePkts++;
+			spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 			/* not needed, but the win32 driver does it too ... */
 			status = garmin_write_bulk(port,
@@ -921,7 +953,9 @@ static int garmin_init_session(struct usb_serial_port *port)
 			                           sizeof(GARMIN_START_SESSION_REQ2));
 			if (status >= 0) {
 				status = 0;
+				spin_lock_irqsave(&garmin_data_p->lock, flags);
 				garmin_data_p->ignorePkts++;
+				spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 			}
 		}
 	}
@@ -935,6 +969,7 @@ static int garmin_init_session(struct usb_serial_port *port)
 
 static int garmin_open (struct usb_serial_port *port, struct file *filp)
 {
+	unsigned long flags;
 	int status = 0;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 
@@ -948,9 +983,11 @@ static int garmin_open (struct usb_serial_port *port, struct file *filp)
 	if (port->tty)
 		port->tty->low_latency = 1;
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->mode  = initial_mode;
 	garmin_data_p->count = 0;
 	garmin_data_p->flags = 0;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 	/* shutdown any bulk reads that might be going on */
 	usb_kill_urb (port->write_urb);
@@ -996,6 +1033,7 @@ static void garmin_close (struct usb_serial_port *port, struct file * filp)
 
 static void garmin_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
 {
+	unsigned long flags;
 	struct usb_serial_port *port = (struct usb_serial_port *)urb->context;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 
@@ -1007,7 +1045,9 @@ static void garmin_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
 	if (urb->status) {
 		dbg("%s - nonzero write bulk status received: %d",
 			__FUNCTION__, urb->status);
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
 		garmin_data_p->flags |= CLEAR_HALT_REQUIRED;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 	}
 
 	usb_serial_port_softint(port);
@@ -1015,8 +1055,9 @@ static void garmin_write_bulk_callback (struct urb *urb, struct pt_regs *regs)
 
 
 static int garmin_write_bulk (struct usb_serial_port *port,
-                              const unsigned char *buf, int count)
+			      const unsigned char *buf, int count)
 {
+	unsigned long flags;
 	struct usb_serial *serial = port->serial;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 	struct urb *urb;
@@ -1026,7 +1067,9 @@ static int garmin_write_bulk (struct usb_serial_port *port,
 	dbg("%s - port %d, state %d", __FUNCTION__, port->number,
 		garmin_data_p->state);
 
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->flags &= ~FLAGS_DROP_DATA;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 	buffer = kmalloc (count, GFP_ATOMIC);
 	if (!buffer) {
@@ -1053,7 +1096,9 @@ static int garmin_write_bulk (struct usb_serial_port *port,
 	urb->transfer_flags |= URB_ZERO_PACKET;
 
 	if (GARMIN_LAYERID_APPL == getLayerId(buffer)) {
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
 		garmin_data_p->flags |= FLAGS_APP_REQ_SEEN;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 		if (garmin_data_p->mode == MODE_GARMIN_SERIAL)  {
 			pkt_clear(garmin_data_p);
 			garmin_data_p->state = STATE_GSP_WAIT_DATA;
@@ -1087,8 +1132,9 @@ static int garmin_write_bulk (struct usb_serial_port *port,
 
 
 static int garmin_write (struct usb_serial_port *port,
-                         const unsigned char *buf, int count)
+			 const unsigned char *buf, int count)
 {
+	unsigned long flags;
 	int pktid, pktsiz, len;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 	__le32 *privpkt = (__le32 *)garmin_data_p->privpkt;
@@ -1139,7 +1185,9 @@ static int garmin_write (struct usb_serial_port *port,
 				break;
 
 			case PRIV_PKTID_RESET_REQ:
+				spin_lock_irqsave(&garmin_data_p->lock, flags);
 				garmin_data_p->flags |= FLAGS_APP_REQ_SEEN;
+				spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 				break;
 
 			case PRIV_PKTID_SET_DEF_MODE:
@@ -1155,6 +1203,8 @@ static int garmin_write (struct usb_serial_port *port,
 		}
 	}
 
+	garmin_data_p->ignorePkts = 0;
+
 	if (garmin_data_p->mode == MODE_GARMIN_SERIAL) {
 		return gsp_receive(garmin_data_p, buf, count);
 	} else {	/* MODE_NATIVE */
@@ -1177,10 +1227,10 @@ static int garmin_chars_in_buffer (struct usb_serial_port *port)
 {
 	/*
 	 * Report back the number of bytes currently in our input buffer.
-         * Will this lock up the driver - the buffer contains an incomplete
-         * package which will not be written to the device until it
-         * has been completed ?
-         */
+	 * Will this lock up the driver - the buffer contains an incomplete
+	 * package which will not be written to the device until it
+	 * has been completed ?
+	 */
 	//struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 	//return garmin_data_p->insize;
 	return 0;
@@ -1190,6 +1240,8 @@ static int garmin_chars_in_buffer (struct usb_serial_port *port)
 static void garmin_read_process(struct garmin_data * garmin_data_p,
 				 unsigned char *data, unsigned data_length)
 {
+	unsigned long flags;
+
 	if (garmin_data_p->flags & FLAGS_DROP_DATA) {
 		/* abort-transfer cmd is actice */
 		dbg("%s - pkt dropped", __FUNCTION__);
@@ -1200,11 +1252,14 @@ static void garmin_read_process(struct garmin_data * garmin_data_p,
 		   if a reset is required or not when closing
 		   the device */
 		if (0 == memcmp(data, GARMIN_APP_LAYER_REPLY,
-		                sizeof(GARMIN_APP_LAYER_REPLY)))
+		                sizeof(GARMIN_APP_LAYER_REPLY))) {
+			spin_lock_irqsave(&garmin_data_p->lock, flags);
 			garmin_data_p->flags |= FLAGS_APP_RESP_SEEN;
+			spin_unlock_irqrestore(&garmin_data_p->lock, flags);
+		}
 
 		/* if throttling is active or postprecessing is required
-		   put the received data in th input queue, otherwise
+		   put the received data in the input queue, otherwise
 		   send it directly to the tty port */
 		if (garmin_data_p->flags & FLAGS_QUEUING) {
 			pkt_add(garmin_data_p, data, data_length);
@@ -1221,6 +1276,7 @@ static void garmin_read_process(struct garmin_data * garmin_data_p,
 
 static void garmin_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 {
+	unsigned long flags;
 	struct usb_serial_port *port = (struct usb_serial_port *)urb->context;
 	struct usb_serial *serial =  port->serial;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
@@ -1245,19 +1301,30 @@ static void garmin_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 
 	garmin_read_process(garmin_data_p, data, urb->actual_length);
 
-	/* Continue trying to read until nothing more is received  */
-	if (urb->actual_length > 0) {
-		usb_fill_bulk_urb (port->read_urb, serial->dev,
-			   usb_rcvbulkpipe (serial->dev,
-					    port->bulk_in_endpointAddress),
-			   port->read_urb->transfer_buffer,
-			   port->read_urb->transfer_buffer_length,
-			   garmin_read_bulk_callback, port);
+	if (urb->actual_length == 0 &&
+			0 != (garmin_data_p->flags & FLAGS_BULK_IN_RESTART)) {
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
+		garmin_data_p->flags &= ~FLAGS_BULK_IN_RESTART;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 		status = usb_submit_urb(port->read_urb, GFP_ATOMIC);
 		if (status)
 			dev_err(&port->dev,
 				"%s - failed resubmitting read urb, error %d\n",
 			        __FUNCTION__, status);
+	} else if (urb->actual_length > 0) {
+		/* Continue trying to read until nothing more is received  */
+		if (0 == (garmin_data_p->flags & FLAGS_THROTTLED)) {
+			status = usb_submit_urb(port->read_urb, GFP_ATOMIC);
+			if (status)
+				dev_err(&port->dev,
+					"%s - failed resubmitting read urb, error %d\n",
+			        	__FUNCTION__, status);
+		}
+	} else {
+		dbg("%s - end of bulk data", __FUNCTION__);
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
+		garmin_data_p->flags &= ~FLAGS_BULK_IN_ACTIVE;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 	}
 	return;
 }
@@ -1265,6 +1332,7 @@ static void garmin_read_bulk_callback (struct urb *urb, struct pt_regs *regs)
 
 static void garmin_read_int_callback (struct urb *urb, struct pt_regs *regs)
 {
+	unsigned long flags;
 	int status;
 	struct usb_serial_port *port = (struct usb_serial_port *)urb->context;
 	struct usb_serial *serial = port->serial;
@@ -1297,25 +1365,41 @@ static void garmin_read_int_callback (struct urb *urb, struct pt_regs *regs)
 
 		dbg("%s - bulk data available.", __FUNCTION__);
 
-		/* bulk data available */
-		usb_fill_bulk_urb (port->read_urb, serial->dev,
-				usb_rcvbulkpipe (serial->dev,
-				port->bulk_in_endpointAddress),
-				port->read_urb->transfer_buffer,
-				port->read_urb->transfer_buffer_length,
-				garmin_read_bulk_callback, port);
-		status = usb_submit_urb(port->read_urb, GFP_KERNEL);
-		if (status) {
-			dev_err(&port->dev,
-				"%s - failed submitting read urb, error %d\n",
-			__FUNCTION__, status);
+		if (0 == (garmin_data_p->flags & FLAGS_BULK_IN_ACTIVE)) {
+
+			/* bulk data available */
+			usb_fill_bulk_urb (port->read_urb, serial->dev,
+					usb_rcvbulkpipe (serial->dev,
+					port->bulk_in_endpointAddress),
+					port->read_urb->transfer_buffer,
+					port->read_urb->transfer_buffer_length,
+					garmin_read_bulk_callback, port);
+			status = usb_submit_urb(port->read_urb, GFP_ATOMIC);
+			if (status) {
+				dev_err(&port->dev,
+					"%s - failed submitting read urb, error %d\n",
+				__FUNCTION__, status);
+			} else {
+				spin_lock_irqsave(&garmin_data_p->lock, flags);
+				garmin_data_p->flags |= FLAGS_BULK_IN_ACTIVE;
+				/* do not send this packet to the user */
+				garmin_data_p->ignorePkts = 1;
+				spin_unlock_irqrestore(&garmin_data_p->lock, flags);
+			}
+		} else {
+			/* bulk-in transfer still active */
+			spin_lock_irqsave(&garmin_data_p->lock, flags);
+			garmin_data_p->flags |= FLAGS_BULK_IN_RESTART;
+			spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 		}
 
 	} else if (urb->actual_length == (4+sizeof(GARMIN_START_SESSION_REPLY))
 			 && 0 == memcmp(data, GARMIN_START_SESSION_REPLY,
 			                sizeof(GARMIN_START_SESSION_REPLY))) {
 
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
 		garmin_data_p->flags |= FLAGS_SESSION_REPLY1_SEEN;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 		/* save the serial number */
 		garmin_data_p->serial_num 
@@ -1330,7 +1414,9 @@ static void garmin_read_int_callback (struct urb *urb, struct pt_regs *regs)
 		   ignore it. */
 		dbg("%s - pkt ignored (%d)",
 			__FUNCTION__, garmin_data_p->ignorePkts);
+		spin_lock_irqsave(&garmin_data_p->lock, flags);
 		garmin_data_p->ignorePkts--;
+		spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 	} else {
 		garmin_read_process(garmin_data_p, data, urb->actual_length);
 	}
@@ -1351,18 +1437,20 @@ static void garmin_read_int_callback (struct urb *urb, struct pt_regs *regs)
  */
 static int garmin_flush_queue(struct garmin_data * garmin_data_p)
 {
+	unsigned long flags;
 	struct garmin_packet *pkt;
 
 	if ((garmin_data_p->flags & FLAGS_THROTTLED) == 0) {
 		pkt = pkt_pop(garmin_data_p);
 		if (pkt != NULL) {
-
 			send_to_tty(garmin_data_p->port, pkt->data, pkt->size);
 			kfree(pkt);
 			mod_timer(&garmin_data_p->timer, (1)+jiffies);
 
 		} else {
+			spin_lock_irqsave(&garmin_data_p->lock, flags);
 			garmin_data_p->flags &= ~FLAGS_QUEUING;
+			spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 		}
 	}
 	return 0;
@@ -1371,26 +1459,41 @@ static int garmin_flush_queue(struct garmin_data * garmin_data_p)
 
 static void garmin_throttle (struct usb_serial_port *port)
 {
+	unsigned long flags;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
 
 	dbg("%s - port %d", __FUNCTION__, port->number);
 	/* set flag, data received will be put into a queue
 	   for later processing */
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->flags |= FLAGS_QUEUING|FLAGS_THROTTLED;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 }
 
 
 static void garmin_unthrottle (struct usb_serial_port *port)
 {
+	unsigned long flags;
 	struct garmin_data * garmin_data_p = usb_get_serial_port_data(port);
+	int status;
 
 	dbg("%s - port %d", __FUNCTION__, port->number);
+	spin_lock_irqsave(&garmin_data_p->lock, flags);
 	garmin_data_p->flags &= ~FLAGS_THROTTLED;
+	spin_unlock_irqrestore(&garmin_data_p->lock, flags);
 
 	/* in native mode send queued data to tty, in
 	   serial mode nothing needs to be done here */
 	if (garmin_data_p->mode == MODE_NATIVE)
 		garmin_flush_queue(garmin_data_p);
+
+	if (0 != (garmin_data_p->flags & FLAGS_BULK_IN_ACTIVE)) {
+		status = usb_submit_urb(port->read_urb, GFP_ATOMIC);
+		if (status)
+			dev_err(&port->dev,
+				"%s - failed resubmitting read urb, error %d\n",
+				__FUNCTION__, status);
+	}
 }
 
 
@@ -1420,11 +1523,12 @@ static int garmin_attach (struct usb_serial *serial)
 
 	dbg("%s", __FUNCTION__);
 
-	garmin_data_p = kzalloc(sizeof(struct garmin_data), GFP_KERNEL);
+	garmin_data_p = kmalloc (sizeof(struct garmin_data), GFP_KERNEL);
 	if (garmin_data_p == NULL) {
 		dev_err(&port->dev, "%s - Out of memory\n", __FUNCTION__);
 		return -ENOMEM;
 	}
+	memset (garmin_data_p, 0, sizeof(struct garmin_data));
 	init_timer(&garmin_data_p->timer);
 	spin_lock_init(&garmin_data_p->lock);
 	INIT_LIST_HEAD(&garmin_data_p->pktlist);
@@ -1459,10 +1563,10 @@ static void garmin_shutdown (struct usb_serial *serial)
 /* All of the device info needed */
 static struct usb_serial_driver garmin_device = {
 	.driver = {
-		.owner =	THIS_MODULE,
-		.name =		"garmin_gps",
+		.owner       = THIS_MODULE,
+		.name        = "garmin_gps",
 	},
-	.description =		"Garmin GPS usb/tty",
+	.description         = "Garmin GPS usb/tty",
 	.id_table            = id_table,
 	.num_interrupt_in    = 1,
 	.num_bulk_in         = 1,
@@ -1483,6 +1587,7 @@ static struct usb_serial_driver garmin_device = {
 };
 
 
+
 static int __init garmin_init (void)
 {
 	int retval;
-- 
cgit v1.2.1


From bd35078f47f795569fae29b2ff187e6f921460ad Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 31 Jul 2006 07:29:39 -0700
Subject: USB: build fixes: ohci-omap

The ohci-omap code has diverged from the working version in the linux-omap
tree; this syncs up the versions:

  - Another clock is needed in various cases
  - The omap-1510 iommu code needs to be #ifdeffed out on newer parts
  - Saner use of the HCD framework
  - Various other changes, e.g. a Nokia 770 quirk

And some minor dead-whitespace removal.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ohci-omap.c | 113 +++++++++++++++++++++++++++----------------
 1 file changed, 71 insertions(+), 42 deletions(-)

diff --git a/drivers/usb/host/ohci-omap.c b/drivers/usb/host/ohci-omap.c
index 47f1c9bbef87..160cd4c58a03 100644
--- a/drivers/usb/host/ohci-omap.c
+++ b/drivers/usb/host/ohci-omap.c
@@ -4,7 +4,7 @@
  * (C) Copyright 1999 Roman Weissgaerber <weissg@vienna.at>
  * (C) Copyright 2000-2005 David Brownell
  * (C) Copyright 2002 Hewlett-Packard Company
- * 
+ *
  * OMAP Bus Glue
  *
  * Modified for OMAP by Tony Lindgren <tony@atomide.com>
@@ -66,15 +66,20 @@ extern int usb_disabled(void);
 extern int ocpi_enable(void);
 
 static struct clk *usb_host_ck;
+static struct clk *usb_dc_ck;
+static int host_enabled;
+static int host_initialized;
 
 static void omap_ohci_clock_power(int on)
 {
 	if (on) {
+		clk_enable(usb_dc_ck);
 		clk_enable(usb_host_ck);
 		/* guesstimate for T5 == 1x 32K clock + APLL lock time */
 		udelay(100);
 	} else {
 		clk_disable(usb_host_ck);
+		clk_disable(usb_dc_ck);
 	}
 }
 
@@ -87,14 +92,14 @@ static int omap_ohci_transceiver_power(int on)
 	if (on) {
 		if (machine_is_omap_innovator() && cpu_is_omap1510())
 			fpga_write(fpga_read(INNOVATOR_FPGA_CAM_USB_CONTROL)
-				| ((1 << 5/*usb1*/) | (1 << 3/*usb2*/)), 
+				| ((1 << 5/*usb1*/) | (1 << 3/*usb2*/)),
 			       INNOVATOR_FPGA_CAM_USB_CONTROL);
 		else if (machine_is_omap_osk())
 			tps65010_set_gpio_out_value(GPIO1, LOW);
 	} else {
 		if (machine_is_omap_innovator() && cpu_is_omap1510())
 			fpga_write(fpga_read(INNOVATOR_FPGA_CAM_USB_CONTROL)
-				& ~((1 << 5/*usb1*/) | (1 << 3/*usb2*/)), 
+				& ~((1 << 5/*usb1*/) | (1 << 3/*usb2*/)),
 			       INNOVATOR_FPGA_CAM_USB_CONTROL);
 		else if (machine_is_omap_osk())
 			tps65010_set_gpio_out_value(GPIO1, HIGH);
@@ -103,6 +108,7 @@ static int omap_ohci_transceiver_power(int on)
 	return 0;
 }
 
+#ifdef CONFIG_ARCH_OMAP15XX
 /*
  * OMAP-1510 specific Local Bus clock on/off
  */
@@ -121,8 +127,8 @@ static int omap_1510_local_bus_power(int on)
 /*
  * OMAP-1510 specific Local Bus initialization
  * NOTE: This assumes 32MB memory size in OMAP1510LB_MEMSIZE.
- *       See also arch/mach-omap/memory.h for __virt_to_dma() and 
- *       __dma_to_virt() which need to match with the physical 
+ *       See also arch/mach-omap/memory.h for __virt_to_dma() and
+ *       __dma_to_virt() which need to match with the physical
  *       Local Bus address below.
  */
 static int omap_1510_local_bus_init(void)
@@ -130,7 +136,7 @@ static int omap_1510_local_bus_init(void)
 	unsigned int tlb;
 	unsigned long lbaddr, physaddr;
 
-	omap_writel((omap_readl(OMAP1510_LB_CLOCK_DIV) & 0xfffffff8) | 0x4, 
+	omap_writel((omap_readl(OMAP1510_LB_CLOCK_DIV) & 0xfffffff8) | 0x4,
 	       OMAP1510_LB_CLOCK_DIV);
 
 	/* Configure the Local Bus MMU table */
@@ -138,7 +144,7 @@ static int omap_1510_local_bus_init(void)
 		lbaddr = tlb * 0x00100000 + OMAP1510_LB_OFFSET;
 		physaddr = tlb * 0x00100000 + PHYS_OFFSET;
 		omap_writel((lbaddr & 0x0fffffff) >> 22, OMAP1510_LB_MMU_CAM_H);
-		omap_writel(((lbaddr & 0x003ffc00) >> 6) | 0xc, 
+		omap_writel(((lbaddr & 0x003ffc00) >> 6) | 0xc,
 		       OMAP1510_LB_MMU_CAM_L);
 		omap_writel(physaddr >> 16, OMAP1510_LB_MMU_RAM_H);
 		omap_writel((physaddr & 0x0000fc00) | 0x300, OMAP1510_LB_MMU_RAM_L);
@@ -152,6 +158,10 @@ static int omap_1510_local_bus_init(void)
 
 	return 0;
 }
+#else
+#define omap_1510_local_bus_power(x)	{}
+#define omap_1510_local_bus_init()	{}
+#endif
 
 #ifdef	CONFIG_USB_OTG
 
@@ -173,13 +183,14 @@ static void start_hnp(struct ohci_hcd *ohci)
 
 /*-------------------------------------------------------------------------*/
 
-static int omap_start_hc(struct ohci_hcd *ohci, struct platform_device *pdev)
+static int ohci_omap_init(struct usb_hcd *hcd)
 {
-	struct omap_usb_config	*config = pdev->dev.platform_data;
+	struct ohci_hcd		*ohci = hcd_to_ohci(hcd);
+	struct omap_usb_config	*config = hcd->self.controller->platform_data;
 	int			need_transceiver = (config->otg != 0);
 	int			ret;
 
-	dev_dbg(&pdev->dev, "starting USB Controller\n");
+	dev_dbg(hcd->self.controller, "starting USB Controller\n");
 
 	if (config->otg) {
 		ohci_to_hcd(ohci)->self.otg_port = config->otg;
@@ -200,7 +211,7 @@ static int omap_start_hc(struct ohci_hcd *ohci, struct platform_device *pdev)
 		if (ohci->transceiver) {
 			int	status = otg_set_host(ohci->transceiver,
 						&ohci_to_hcd(ohci)->self);
-			dev_dbg(&pdev->dev, "init %s transceiver, status %d\n",
+			dev_dbg(hcd->self.controller, "init %s transceiver, status %d\n",
 					ohci->transceiver->label, status);
 			if (status) {
 				if (ohci->transceiver)
@@ -208,7 +219,7 @@ static int omap_start_hc(struct ohci_hcd *ohci, struct platform_device *pdev)
 				return status;
 			}
 		} else {
-			dev_err(&pdev->dev, "can't find transceiver\n");
+			dev_err(hcd->self.controller, "can't find transceiver\n");
 			return -ENODEV;
 		}
 	}
@@ -247,6 +258,10 @@ static int omap_start_hc(struct ohci_hcd *ohci, struct platform_device *pdev)
 		}
 		ohci_writel(ohci, rh, &ohci->regs->roothub.a);
 		distrust_firmware = 0;
+	} else if (machine_is_nokia770()) {
+		/* We require a self-powered hub, which should have
+		 * plenty of power. */
+		ohci_to_hcd(ohci)->power_budget = 0;
 	}
 
 	/* FIXME khubd hub requests should manage power switching */
@@ -260,21 +275,15 @@ static int omap_start_hc(struct ohci_hcd *ohci, struct platform_device *pdev)
 	return 0;
 }
 
-static void omap_stop_hc(struct platform_device *pdev)
+static void ohci_omap_stop(struct usb_hcd *hcd)
 {
-	dev_dbg(&pdev->dev, "stopping USB Controller\n");
+	dev_dbg(hcd->self.controller, "stopping USB Controller\n");
 	omap_ohci_clock_power(0);
 }
 
 
 /*-------------------------------------------------------------------------*/
 
-void usb_hcd_omap_remove (struct usb_hcd *, struct platform_device *);
-
-/* configure so an HC device and id are always provided */
-/* always called with process context; sleeping is OK */
-
-
 /**
  * usb_hcd_omap_probe - initialize OMAP-based HCDs
  * Context: !in_interrupt()
@@ -283,7 +292,7 @@ void usb_hcd_omap_remove (struct usb_hcd *, struct platform_device *);
  * then invokes the start() method for the HCD associated with it
  * through the hotplug entry's driver_data.
  */
-int usb_hcd_omap_probe (const struct hc_driver *driver,
+static int usb_hcd_omap_probe (const struct hc_driver *driver,
 			  struct platform_device *pdev)
 {
 	int retval, irq;
@@ -291,12 +300,12 @@ int usb_hcd_omap_probe (const struct hc_driver *driver,
 	struct ohci_hcd *ohci;
 
 	if (pdev->num_resources != 2) {
-		printk(KERN_ERR "hcd probe: invalid num_resources: %i\n", 
+		printk(KERN_ERR "hcd probe: invalid num_resources: %i\n",
 		       pdev->num_resources);
 		return -ENODEV;
 	}
 
-	if (pdev->resource[0].flags != IORESOURCE_MEM 
+	if (pdev->resource[0].flags != IORESOURCE_MEM
 			|| pdev->resource[1].flags != IORESOURCE_IRQ) {
 		printk(KERN_ERR "hcd probe: invalid resource type\n");
 		return -ENODEV;
@@ -306,6 +315,17 @@ int usb_hcd_omap_probe (const struct hc_driver *driver,
 	if (IS_ERR(usb_host_ck))
 		return PTR_ERR(usb_host_ck);
 
+	if (!cpu_is_omap1510())
+		usb_dc_ck = clk_get(0, "usb_dc_ck");
+	else
+		usb_dc_ck = clk_get(0, "lb_ck");
+
+	if (IS_ERR(usb_dc_ck)) {
+		clk_put(usb_host_ck);
+		return PTR_ERR(usb_dc_ck);
+	}
+
+
 	hcd = usb_create_hcd (driver, &pdev->dev, pdev->dev.bus_id);
 	if (!hcd) {
 		retval = -ENOMEM;
@@ -325,9 +345,8 @@ int usb_hcd_omap_probe (const struct hc_driver *driver,
 	ohci = hcd_to_ohci(hcd);
 	ohci_hcd_init(ohci);
 
-	retval = omap_start_hc(ohci, pdev);
-	if (retval < 0)
-		goto err2;
+	host_initialized = 0;
+	host_enabled = 1;
 
 	irq = platform_get_irq(pdev, 0);
 	if (irq < 0) {
@@ -335,15 +354,21 @@ int usb_hcd_omap_probe (const struct hc_driver *driver,
 		goto err2;
 	}
 	retval = usb_add_hcd(hcd, irq, IRQF_DISABLED);
-	if (retval == 0)
-		return retval;
+	if (retval)
+		goto err2;
+
+	host_initialized = 1;
+
+	if (!host_enabled)
+		omap_ohci_clock_power(0);
 
-	omap_stop_hc(pdev);
+	return 0;
 err2:
 	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
 err1:
 	usb_put_hcd(hcd);
 err0:
+	clk_put(usb_dc_ck);
 	clk_put(usb_host_ck);
 	return retval;
 }
@@ -359,28 +384,36 @@ err0:
  * Reverses the effect of usb_hcd_omap_probe(), first invoking
  * the HCD's stop() method.  It is always called from a thread
  * context, normally "rmmod", "apmd", or something similar.
- *
  */
-void usb_hcd_omap_remove (struct usb_hcd *hcd, struct platform_device *pdev)
+static inline void
+usb_hcd_omap_remove (struct usb_hcd *hcd, struct platform_device *pdev)
 {
+	struct ohci_hcd		*ohci = hcd_to_ohci (hcd);
+
 	usb_remove_hcd(hcd);
+	if (ohci->transceiver) {
+		(void) otg_set_host(ohci->transceiver, 0);
+		put_device(ohci->transceiver->dev);
+	}
 	if (machine_is_omap_osk())
 		omap_free_gpio(9);
-	omap_stop_hc(pdev);
 	release_mem_region(hcd->rsrc_start, hcd->rsrc_len);
 	usb_put_hcd(hcd);
+	clk_put(usb_dc_ck);
 	clk_put(usb_host_ck);
 }
 
 /*-------------------------------------------------------------------------*/
 
-static int __devinit
+static int
 ohci_omap_start (struct usb_hcd *hcd)
 {
 	struct omap_usb_config *config;
 	struct ohci_hcd	*ohci = hcd_to_ohci (hcd);
 	int		ret;
 
+	if (!host_enabled)
+		return 0;
 	config = hcd->self.controller->platform_data;
 	if (config->otg || config->rwc) {
 		ohci->hc_control = OHCI_CTRL_RWC;
@@ -411,8 +444,9 @@ static const struct hc_driver ohci_omap_hc_driver = {
 	/*
 	 * basic lifecycle operations
 	 */
+	.reset =		ohci_omap_init,
 	.start =		ohci_omap_start,
-	.stop =			ohci_stop,
+	.stop =			ohci_omap_stop,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -449,13 +483,8 @@ static int ohci_hcd_omap_drv_probe(struct platform_device *dev)
 static int ohci_hcd_omap_drv_remove(struct platform_device *dev)
 {
 	struct usb_hcd		*hcd = platform_get_drvdata(dev);
-	struct ohci_hcd		*ohci = hcd_to_ohci (hcd);
 
 	usb_hcd_omap_remove(hcd, dev);
-	if (ohci->transceiver) {
-		(void) otg_set_host(ohci->transceiver, 0);
-		put_device(ohci->transceiver->dev);
-	}
 	platform_set_drvdata(dev, NULL);
 
 	return 0;
@@ -475,7 +504,7 @@ static int ohci_omap_suspend(struct platform_device *dev, pm_message_t message)
 
 	omap_ohci_clock_power(0);
 	ohci_to_hcd(ohci)->state = HC_STATE_SUSPENDED;
-	dev->power.power_state = PMSG_SUSPEND;
+	dev->dev.power.power_state = PMSG_SUSPEND;
 	return 0;
 }
 
@@ -488,8 +517,8 @@ static int ohci_omap_resume(struct platform_device *dev)
 	ohci->next_statechange = jiffies;
 
 	omap_ohci_clock_power(1);
-	dev->power.power_state = PMSG_ON;
-	usb_hcd_resume_root_hub(dev_get_drvdata(dev));
+	dev->dev.power.power_state = PMSG_ON;
+	usb_hcd_resume_root_hub(platform_get_drvdata(dev));
 	return 0;
 }
 
-- 
cgit v1.2.1


From 17efe155253e3f81caa6cc57ec2af1c128618698 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor@insightbb.com>
Date: Tue, 1 Aug 2006 22:45:28 -0400
Subject: USB: onetouch - handle errors from input_register_device()

Onetouch: handle errors from input_register_device()

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/onetouch.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/storage/onetouch.c b/drivers/usb/storage/onetouch.c
index 313920d980c9..f843a0bcf107 100644
--- a/drivers/usb/storage/onetouch.c
+++ b/drivers/usb/storage/onetouch.c
@@ -135,6 +135,7 @@ int onetouch_connect_input(struct us_data *ss)
 	struct usb_onetouch *onetouch;
 	struct input_dev *input_dev;
 	int pipe, maxp;
+	int error = -ENOMEM;
 
 	interface = ss->pusb_intf->cur_altsetting;
 
@@ -211,15 +212,18 @@ int onetouch_connect_input(struct us_data *ss)
 	ss->suspend_resume_hook = usb_onetouch_pm_hook;
 #endif
 
-	input_register_device(onetouch->dev);
+	error = input_register_device(onetouch->dev);
+	if (error)
+		goto fail3;
 
 	return 0;
 
+ fail3:	usb_free_urb(onetouch->irq);
  fail2:	usb_buffer_free(udev, ONETOUCH_PKT_LEN,
 			onetouch->data, onetouch->data_dma);
  fail1:	kfree(onetouch);
 	input_free_device(input_dev);
-	return -ENOMEM;
+	return error;
 }
 
 void onetouch_release_input(void *onetouch_)
-- 
cgit v1.2.1


From 07cb7f23d07aa4d197dbeb123eb9719c176190ee Mon Sep 17 00:00:00 2001
From: Milan Svoboda <msvoboda@ra.rockwell.com>
Date: Mon, 26 Jun 2006 07:46:00 -0700
Subject: USB: correct locking in gadgetfs_disconnect

This patch moves spin_lock (&dev->lock) before first use of dev.
I think that test to the state of device should be protected with
this spin_lock...

Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 2a7162d89799..da85732fa993 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1696,16 +1696,17 @@ gadgetfs_disconnect (struct usb_gadget *gadget)
 {
 	struct dev_data		*dev = get_gadget_data (gadget);
 
+	spin_lock (&dev->lock);
 	if (dev->state == STATE_UNCONNECTED) {
 		DBG (dev, "already unconnected\n");
-		return;
+		goto exit;
 	}
 	dev->state = STATE_UNCONNECTED;
 
 	INFO (dev, "disconnected\n");
-	spin_lock (&dev->lock);
 	next_event (dev, GADGETFS_DISCONNECT);
 	ep0_readable (dev);
+exit:
 	spin_unlock (&dev->lock);
 }
 
-- 
cgit v1.2.1


From 8a7471aba19dc526978a03bfe2e3c122712b5900 Mon Sep 17 00:00:00 2001
From: Milan Svoboda <msvoboda@ra.rockwell.com>
Date: Mon, 26 Jun 2006 07:19:00 -0700
Subject: USB: fix ep_config to return correct value

This patch fixes ep_config to return correct value. Without patch
ep_config returns submitted lenght minus 4 on succes. With this
patch applied, whole submitted lenght is returned.

ep_config parses submitted data and if buffer starts with (int) 1
it is parsed, otherwise error is reported. Problem is that ep_config
returns size of buffer minus 4 on success. I think that size of buffer
should be returned instead, because there were no problems and
all data were processed.

Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index da85732fa993..72c88aa94d5d 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -741,7 +741,7 @@ ep_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 	struct ep_data		*data = fd->private_data;
 	struct usb_ep		*ep;
 	u32			tag;
-	int			value;
+	int			value, length = len;
 
 	if ((value = down_interruptible (&data->lock)) < 0)
 		return value;
@@ -792,7 +792,6 @@ ep_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 			goto fail0;
 		}
 	}
-	value = len;
 
 	spin_lock_irq (&data->dev->lock);
 	if (data->dev->state == STATE_DEV_UNBOUND) {
@@ -822,8 +821,10 @@ ep_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 				data->name);
 		data->state = STATE_EP_DEFER_ENABLE;
 	}
-	if (value == 0)
+	if (value == 0) {
 		fd->f_op = &ep_io_operations;
+		value = length;
+	}
 gone:
 	spin_unlock_irq (&data->dev->lock);
 	if (value < 0) {
-- 
cgit v1.2.1


From ba307f5828b1b4a1348d99c4f430a0cf3beeae2f Mon Sep 17 00:00:00 2001
From: Milan Svoboda <msvoboda@ra.rockwell.com>
Date: Mon, 26 Jun 2006 07:48:00 -0700
Subject: USB: gadgetfs: protect ep_release with lock

This patch adds mutex protection to ep_release.

Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 72c88aa94d5d..ab508858f341 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -477,6 +477,10 @@ static int
 ep_release (struct inode *inode, struct file *fd)
 {
 	struct ep_data		*data = fd->private_data;
+	int value;
+
+	if ((value = down_interruptible(&data->lock)) < 0)
+		return value;
 
 	/* clean up if this can be reopened */
 	if (data->state != STATE_EP_UNBOUND) {
@@ -485,6 +489,7 @@ ep_release (struct inode *inode, struct file *fd)
 		data->hs_desc.bDescriptorType = 0;
 		usb_ep_disable(data->ep);
 	}
+	up (&data->lock);
 	put_ep (data);
 	return 0;
 }
-- 
cgit v1.2.1


From f2ebf92c9e1930a8f79b7eb49a32122931929014 Mon Sep 17 00:00:00 2001
From: Ben Williamson <ben.williamson@greyinnovation.com>
Date: Tue, 1 Aug 2006 11:28:16 +1000
Subject: USB: gmidi: New USB MIDI Gadget class driver.

This driver is glue between the USB gadget interface
and the ALSA MIDI interface. It allows us to appear
as a MIDI Streaming device to a host system on the
other end of a USB cable.

This includes linux/usb/audio.h and linux/usb/midi.h
containing definitions from the relevant USB specifications
for USB audio and USB MIDI devices.

The following changes have been made since the first RFC
posting:

* Bug fixes to endpoint handling.
* Workaround for USB_REQ_SET_CONFIGURATION handling,
  not understood yet.
* Added SND and SND_RAWMIDI dependencies in Kconfig.
* Moved usb_audio.h and usb_midi.h to usb/*.h
* Added module parameters for ALSA card index and id.
* Added module parameters for USB descriptor IDs and strings.
* Removed some unneeded stuff inherited from zero.c, more to go.
* Provide DECLARE_* macros for the variable-length structs.
* Use kmalloc instead of usb_ep_alloc_buffer.
* Limit source to 80 columns.
* Return actual error code instead of -ENOMEM in a few places.

Signed-off-by: Ben Williamson <ben.williamson@greyinnovation.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/Kconfig  |   14 +
 drivers/usb/gadget/Makefile |    2 +
 drivers/usb/gadget/gmidi.c  | 1337 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb/audio.h   |   53 ++
 include/linux/usb/midi.h    |  112 ++++
 5 files changed, 1518 insertions(+)
 create mode 100644 drivers/usb/gadget/gmidi.c
 create mode 100644 include/linux/usb/audio.h
 create mode 100644 include/linux/usb/midi.h

diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index 1a32d96774b4..4301e96c417b 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -404,6 +404,20 @@ config USB_G_SERIAL
 	  which includes instructions and a "driver info file" needed to
 	  make MS-Windows work with this driver.
 
+config USB_MIDI_GADGET
+	tristate "MIDI Gadget (EXPERIMENTAL)"
+	depends on SND && EXPERIMENTAL
+	select SND_RAWMIDI
+	help
+	  The MIDI Gadget acts as a USB Audio device, with one MIDI
+	  input and one MIDI output. These MIDI jacks appear as
+	  a sound "card" in the ALSA sound system. Other MIDI
+	  connections can then be made on the gadget system, using
+	  ALSA's aconnect utility etc.
+
+	  Say "y" to link the driver statically, or "m" to build a
+	  dynamically linked module called "g_midi".
+
 
 # put drivers that need isochronous transfer support (for audio
 # or video class gadget drivers), or specific hardware, here.
diff --git a/drivers/usb/gadget/Makefile b/drivers/usb/gadget/Makefile
index 5a28e61392ec..e71e086a1cfa 100644
--- a/drivers/usb/gadget/Makefile
+++ b/drivers/usb/gadget/Makefile
@@ -15,6 +15,7 @@ obj-$(CONFIG_USB_AT91)		+= at91_udc.o
 g_zero-objs			:= zero.o usbstring.o config.o epautoconf.o
 g_ether-objs			:= ether.o usbstring.o config.o epautoconf.o
 g_serial-objs			:= serial.o usbstring.o config.o epautoconf.o
+g_midi-objs			:= gmidi.o usbstring.o config.o epautoconf.o
 gadgetfs-objs			:= inode.o
 g_file_storage-objs		:= file_storage.o usbstring.o config.o \
 					epautoconf.o
@@ -28,4 +29,5 @@ obj-$(CONFIG_USB_ETH)		+= g_ether.o
 obj-$(CONFIG_USB_GADGETFS)	+= gadgetfs.o
 obj-$(CONFIG_USB_FILE_STORAGE)	+= g_file_storage.o
 obj-$(CONFIG_USB_G_SERIAL)	+= g_serial.o
+obj-$(CONFIG_USB_MIDI_GADGET)	+= g_midi.o
 
diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c
new file mode 100644
index 000000000000..b68cecd57411
--- /dev/null
+++ b/drivers/usb/gadget/gmidi.c
@@ -0,0 +1,1337 @@
+/*
+ * gmidi.c -- USB MIDI Gadget Driver
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This code is based in part on:
+ *
+ * Gadget Zero driver, Copyright (C) 2003-2004 David Brownell.
+ * USB Audio driver, Copyright (C) 2002 by Takashi Iwai.
+ * USB MIDI driver, Copyright (C) 2002-2005 Clemens Ladisch.
+ *
+ * Refer to the USB Device Class Definition for MIDI Devices:
+ * http://www.usb.org/developers/devclass_docs/midi10.pdf
+ */
+
+#define DEBUG 1
+// #define VERBOSE
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/utsname.h>
+#include <linux/device.h>
+#include <linux/moduleparam.h>
+
+#include <sound/driver.h>
+#include <sound/core.h>
+#include <sound/initval.h>
+#include <sound/rawmidi.h>
+
+#include <linux/usb_ch9.h>
+#include <linux/usb_gadget.h>
+#include <linux/usb/audio.h>
+#include <linux/usb/midi.h>
+
+#include "gadget_chips.h"
+
+MODULE_AUTHOR("Ben Williamson");
+MODULE_LICENSE("GPL v2");
+
+#define DRIVER_VERSION "25 Jul 2006"
+
+static const char shortname[] = "g_midi";
+static const char longname[] = "MIDI Gadget";
+
+static int index = SNDRV_DEFAULT_IDX1;
+static char *id = SNDRV_DEFAULT_STR1;
+
+module_param(index, int, 0444);
+MODULE_PARM_DESC(index, "Index value for the USB MIDI Gadget adapter.");
+module_param(id, charp, 0444);
+MODULE_PARM_DESC(id, "ID string for the USB MIDI Gadget adapter.");
+
+/* Some systems will want different product identifers published in the
+ * device descriptor, either numbers or strings or both.  These string
+ * parameters are in UTF-8 (superset of ASCII's 7 bit characters).
+ */
+
+static ushort idVendor;
+module_param(idVendor, ushort, S_IRUGO);
+MODULE_PARM_DESC(idVendor, "USB Vendor ID");
+
+static ushort idProduct;
+module_param(idProduct, ushort, S_IRUGO);
+MODULE_PARM_DESC(idProduct, "USB Product ID");
+
+static ushort bcdDevice;
+module_param(bcdDevice, ushort, S_IRUGO);
+MODULE_PARM_DESC(bcdDevice, "USB Device version (BCD)");
+
+static char *iManufacturer;
+module_param(iManufacturer, charp, S_IRUGO);
+MODULE_PARM_DESC(iManufacturer, "USB Manufacturer string");
+
+static char *iProduct;
+module_param(iProduct, charp, S_IRUGO);
+MODULE_PARM_DESC(iProduct, "USB Product string");
+
+static char *iSerialNumber;
+module_param(iSerialNumber, charp, S_IRUGO);
+MODULE_PARM_DESC(iSerialNumber, "SerialNumber");
+
+/*
+ * this version autoconfigures as much as possible,
+ * which is reasonable for most "bulk-only" drivers.
+ */
+static const char *EP_IN_NAME;
+static const char *EP_OUT_NAME;
+
+
+/* big enough to hold our biggest descriptor */
+#define USB_BUFSIZ 256
+
+
+/* This is a gadget, and the IN/OUT naming is from the host's perspective.
+   USB -> OUT endpoint -> rawmidi
+   USB <- IN endpoint  <- rawmidi */
+struct gmidi_in_port {
+	struct gmidi_device* dev;
+	int active;
+	uint8_t cable;		/* cable number << 4 */
+	uint8_t state;
+#define STATE_UNKNOWN	0
+#define STATE_1PARAM	1
+#define STATE_2PARAM_1	2
+#define STATE_2PARAM_2	3
+#define STATE_SYSEX_0	4
+#define STATE_SYSEX_1	5
+#define STATE_SYSEX_2	6
+	uint8_t data[2];
+};
+
+struct gmidi_device {
+	spinlock_t		lock;
+	struct usb_gadget	*gadget;
+	struct usb_request	*req;		/* for control responses */
+	u8			config;
+	struct usb_ep		*in_ep, *out_ep;
+	struct snd_card 	*card;
+	struct snd_rawmidi	*rmidi;
+	struct snd_rawmidi_substream *in_substream;
+	struct snd_rawmidi_substream *out_substream;
+
+	/* For the moment we only support one port in
+	   each direction, but in_port is kept as a
+	   separate struct so we can have more later. */
+	struct gmidi_in_port	in_port;
+	unsigned long		out_triggered;
+	struct tasklet_struct	tasklet;
+};
+
+static void gmidi_transmit(struct gmidi_device* dev, struct usb_request* req);
+
+
+#define xprintk(d,level,fmt,args...) \
+	dev_printk(level , &(d)->gadget->dev , fmt , ## args)
+
+#ifdef DEBUG
+#define DBG(dev,fmt,args...) \
+	xprintk(dev , KERN_DEBUG , fmt , ## args)
+#else
+#define DBG(dev,fmt,args...) \
+	do { } while (0)
+#endif /* DEBUG */
+
+#ifdef VERBOSE
+#define VDBG	DBG
+#else
+#define VDBG(dev,fmt,args...) \
+	do { } while (0)
+#endif /* VERBOSE */
+
+#define ERROR(dev,fmt,args...) \
+	xprintk(dev , KERN_ERR , fmt , ## args)
+#define WARN(dev,fmt,args...) \
+	xprintk(dev , KERN_WARNING , fmt , ## args)
+#define INFO(dev,fmt,args...) \
+	xprintk(dev , KERN_INFO , fmt , ## args)
+
+
+static unsigned buflen = 256;
+static unsigned qlen = 32;
+
+module_param(buflen, uint, S_IRUGO);
+module_param(qlen, uint, S_IRUGO);
+
+
+/* Thanks to Grey Innovation for donating this product ID.
+ *
+ * DO NOT REUSE THESE IDs with a protocol-incompatible driver!!  Ever!!
+ * Instead:  allocate your own, using normal USB-IF procedures.
+ */
+#define DRIVER_VENDOR_NUM	0x17b3		/* Grey Innovation */
+#define DRIVER_PRODUCT_NUM	0x0004		/* Linux-USB "MIDI Gadget" */
+
+
+/*
+ * DESCRIPTORS ... most are static, but strings and (full)
+ * configuration descriptors are built on demand.
+ */
+
+#define STRING_MANUFACTURER	25
+#define STRING_PRODUCT		42
+#define STRING_SERIAL		101
+#define STRING_MIDI_GADGET	250
+
+/* We only have the one configuration, it's number 1. */
+#define	GMIDI_CONFIG		1
+
+/* We have two interfaces- AudioControl and MIDIStreaming */
+#define GMIDI_AC_INTERFACE	0
+#define GMIDI_MS_INTERFACE	1
+#define GMIDI_NUM_INTERFACES	2
+
+DECLARE_USB_AC_HEADER_DESCRIPTOR(1);
+DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(1);
+DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(1);
+
+/* B.1  Device Descriptor */
+static struct usb_device_descriptor device_desc = {
+	.bLength =		USB_DT_DEVICE_SIZE,
+	.bDescriptorType =	USB_DT_DEVICE,
+	.bcdUSB =		__constant_cpu_to_le16(0x0200),
+	.bDeviceClass =		USB_CLASS_PER_INTERFACE,
+	.idVendor =		__constant_cpu_to_le16(DRIVER_VENDOR_NUM),
+	.idProduct =		__constant_cpu_to_le16(DRIVER_PRODUCT_NUM),
+	.iManufacturer =	STRING_MANUFACTURER,
+	.iProduct =		STRING_PRODUCT,
+	.bNumConfigurations =	1,
+};
+
+/* B.2  Configuration Descriptor */
+static struct usb_config_descriptor config_desc = {
+	.bLength =		USB_DT_CONFIG_SIZE,
+	.bDescriptorType =	USB_DT_CONFIG,
+	/* compute wTotalLength on the fly */
+	.bNumInterfaces =	GMIDI_NUM_INTERFACES,
+	.bConfigurationValue =	GMIDI_CONFIG,
+	.iConfiguration =	STRING_MIDI_GADGET,
+	/*
+	 * FIXME: When embedding this driver in a device,
+	 * these need to be set to reflect the actual
+	 * power properties of the device. Is it selfpowered?
+	 */
+	.bmAttributes =		USB_CONFIG_ATT_ONE,
+	.bMaxPower =		1,
+};
+
+/* B.3.1  Standard AC Interface Descriptor */
+static const struct usb_interface_descriptor ac_interface_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bInterfaceNumber =	GMIDI_AC_INTERFACE,
+	.bNumEndpoints =	0,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_AUDIOCONTROL,
+	.iInterface =		STRING_MIDI_GADGET,
+};
+
+/* B.3.2  Class-Specific AC Interface Descriptor */
+static const struct usb_ac_header_descriptor_1 ac_header_desc = {
+	.bLength =		USB_DT_AC_HEADER_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_HEADER,
+	.bcdADC =		__constant_cpu_to_le16(0x0100),
+	.wTotalLength =		USB_DT_AC_HEADER_SIZE(1),
+	.bInCollection =	1,
+	.baInterfaceNr = {
+		[0] =		GMIDI_MS_INTERFACE,
+	}
+};
+
+/* B.4.1  Standard MS Interface Descriptor */
+static const struct usb_interface_descriptor ms_interface_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bInterfaceNumber =	GMIDI_MS_INTERFACE,
+	.bNumEndpoints =	2,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_MIDISTREAMING,
+	.iInterface =		STRING_MIDI_GADGET,
+};
+
+/* B.4.2  Class-Specific MS Interface Descriptor */
+static const struct usb_ms_header_descriptor ms_header_desc = {
+	.bLength =		USB_DT_MS_HEADER_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_HEADER,
+	.bcdMSC =		__constant_cpu_to_le16(0x0100),
+	.wTotalLength =		USB_DT_MS_HEADER_SIZE
+				+ 2*USB_DT_MIDI_IN_SIZE
+				+ 2*USB_DT_MIDI_OUT_SIZE(1),
+};
+
+#define JACK_IN_EMB	1
+#define JACK_IN_EXT	2
+#define JACK_OUT_EMB	3
+#define JACK_OUT_EXT	4
+
+/* B.4.3  MIDI IN Jack Descriptors */
+static const struct usb_midi_in_jack_descriptor jack_in_emb_desc = {
+	.bLength =		USB_DT_MIDI_IN_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_IN_JACK,
+	.bJackType =		USB_MS_EMBEDDED,
+	.bJackID =		JACK_IN_EMB,
+};
+
+static const struct usb_midi_in_jack_descriptor jack_in_ext_desc = {
+	.bLength =		USB_DT_MIDI_IN_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_IN_JACK,
+	.bJackType =		USB_MS_EXTERNAL,
+	.bJackID =		JACK_IN_EXT,
+};
+
+/* B.4.4  MIDI OUT Jack Descriptors */
+static const struct usb_midi_out_jack_descriptor_1 jack_out_emb_desc = {
+	.bLength =		USB_DT_MIDI_OUT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK,
+	.bJackType =		USB_MS_EMBEDDED,
+	.bJackID =		JACK_OUT_EMB,
+	.bNrInputPins =		1,
+	.pins = {
+		[0] = {
+			.baSourceID =	JACK_IN_EXT,
+			.baSourcePin =	1,
+		}
+	}
+};
+
+static const struct usb_midi_out_jack_descriptor_1 jack_out_ext_desc = {
+	.bLength =		USB_DT_MIDI_OUT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	USB_MS_MIDI_OUT_JACK,
+	.bJackType =		USB_MS_EXTERNAL,
+	.bJackID =		JACK_OUT_EXT,
+	.bNrInputPins =		1,
+	.pins = {
+		[0] = {
+			.baSourceID =	JACK_IN_EMB,
+			.baSourcePin =	1,
+		}
+	}
+};
+
+/* B.5.1  Standard Bulk OUT Endpoint Descriptor */
+static struct usb_endpoint_descriptor bulk_out_desc = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_OUT,
+	.bmAttributes =		USB_ENDPOINT_XFER_BULK,
+};
+
+/* B.5.2  Class-specific MS Bulk OUT Endpoint Descriptor */
+static const struct usb_ms_endpoint_descriptor_1 ms_out_desc = {
+	.bLength =		USB_DT_MS_ENDPOINT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_ENDPOINT,
+	.bDescriptorSubtype =	USB_MS_GENERAL,
+	.bNumEmbMIDIJack =	1,
+	.baAssocJackID = {
+		[0] =		JACK_IN_EMB,
+	}
+};
+
+/* B.6.1  Standard Bulk IN Endpoint Descriptor */
+static struct usb_endpoint_descriptor bulk_in_desc = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_IN,
+	.bmAttributes =		USB_ENDPOINT_XFER_BULK,
+};
+
+/* B.6.2  Class-specific MS Bulk IN Endpoint Descriptor */
+static const struct usb_ms_endpoint_descriptor_1 ms_in_desc = {
+	.bLength =		USB_DT_MS_ENDPOINT_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_ENDPOINT,
+	.bDescriptorSubtype =	USB_MS_GENERAL,
+	.bNumEmbMIDIJack =	1,
+	.baAssocJackID = {
+		[0] =		JACK_OUT_EMB,
+	}
+};
+
+static const struct usb_descriptor_header *gmidi_function [] = {
+	(struct usb_descriptor_header *)&ac_interface_desc,
+	(struct usb_descriptor_header *)&ac_header_desc,
+	(struct usb_descriptor_header *)&ms_interface_desc,
+
+	(struct usb_descriptor_header *)&ms_header_desc,
+	(struct usb_descriptor_header *)&jack_in_emb_desc,
+	(struct usb_descriptor_header *)&jack_in_ext_desc,
+	(struct usb_descriptor_header *)&jack_out_emb_desc,
+	(struct usb_descriptor_header *)&jack_out_ext_desc,
+	/* If you add more jacks, update ms_header_desc.wTotalLength */
+
+	(struct usb_descriptor_header *)&bulk_out_desc,
+	(struct usb_descriptor_header *)&ms_out_desc,
+	(struct usb_descriptor_header *)&bulk_in_desc,
+	(struct usb_descriptor_header *)&ms_in_desc,
+	NULL,
+};
+
+static char manufacturer[50];
+static char product_desc[40] = "MIDI Gadget";
+static char serial_number[20];
+
+/* static strings, in UTF-8 */
+static struct usb_string strings [] = {
+	{ STRING_MANUFACTURER, manufacturer, },
+	{ STRING_PRODUCT, product_desc, },
+	{ STRING_SERIAL, serial_number, },
+	{ STRING_MIDI_GADGET, longname, },
+	{  }			/* end of list */
+};
+
+static struct usb_gadget_strings stringtab = {
+	.language	= 0x0409,	/* en-us */
+	.strings	= strings,
+};
+
+static int config_buf(struct usb_gadget *gadget,
+		u8 *buf, u8 type, unsigned index)
+{
+	int len;
+
+	/* only one configuration */
+	if (index != 0) {
+		return -EINVAL;
+	}
+	len = usb_gadget_config_buf(&config_desc,
+			buf, USB_BUFSIZ, gmidi_function);
+	if (len < 0) {
+		return len;
+	}
+	((struct usb_config_descriptor *)buf)->bDescriptorType = type;
+	return len;
+}
+
+static struct usb_request* alloc_ep_req(struct usb_ep *ep, unsigned length)
+{
+	struct usb_request	*req;
+
+	req = usb_ep_alloc_request(ep, GFP_ATOMIC);
+	if (req) {
+		req->length = length;
+		req->buf = kmalloc(length, GFP_ATOMIC);
+		if (!req->buf) {
+			usb_ep_free_request(ep, req);
+			req = NULL;
+		}
+	}
+	return req;
+}
+
+static void free_ep_req(struct usb_ep *ep, struct usb_request *req)
+{
+	kfree(req->buf);
+	usb_ep_free_request(ep, req);
+}
+
+static const uint8_t gmidi_cin_length[] = {
+	0, 0, 2, 3, 3, 1, 2, 3, 3, 3, 3, 3, 2, 2, 3, 1
+};
+
+/*
+ * Receives a chunk of MIDI data.
+ */
+static void gmidi_read_data(struct usb_ep *ep, int cable,
+				   uint8_t* data, int length)
+{
+	struct gmidi_device *dev = ep->driver_data;
+	/* cable is ignored, because for now we only have one. */
+
+	if (!dev->out_substream) {
+		/* Nobody is listening - throw it on the floor. */
+		return;
+	}
+	if (!test_bit(dev->out_substream->number, &dev->out_triggered)) {
+		return;
+	}
+	snd_rawmidi_receive(dev->out_substream, data, length);
+}
+
+static void gmidi_handle_out_data(struct usb_ep *ep, struct usb_request *req)
+{
+	unsigned i;
+	u8 *buf = req->buf;
+
+	for (i = 0; i + 3 < req->actual; i += 4) {
+		if (buf[i] != 0) {
+			int cable = buf[i] >> 4;
+			int length = gmidi_cin_length[buf[i] & 0x0f];
+			gmidi_read_data(ep, cable, &buf[i + 1], length);
+		}
+	}
+}
+
+static void gmidi_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	struct gmidi_device *dev = ep->driver_data;
+	int status = req->status;
+
+	switch (status) {
+	case 0: 			/* normal completion */
+		if (ep == dev->out_ep) {
+			/* we received stuff.
+			   req is queued again, below */
+			gmidi_handle_out_data(ep, req);
+		} else if (ep == dev->in_ep) {
+			/* our transmit completed.
+			   see if there's more to go.
+			   gmidi_transmit eats req, don't queue it again. */
+			gmidi_transmit(dev, req);
+			return;
+		}
+		break;
+
+	/* this endpoint is normally active while we're configured */
+	case -ECONNABORTED: 		/* hardware forced ep reset */
+	case -ECONNRESET:		/* request dequeued */
+	case -ESHUTDOWN:		/* disconnect from host */
+		VDBG(dev, "%s gone (%d), %d/%d\n", ep->name, status,
+				req->actual, req->length);
+		if (ep == dev->out_ep) {
+			gmidi_handle_out_data(ep, req);
+		}
+		free_ep_req(ep, req);
+		return;
+
+	case -EOVERFLOW:		/* buffer overrun on read means that
+					 * we didn't provide a big enough
+					 * buffer.
+					 */
+	default:
+		DBG(dev, "%s complete --> %d, %d/%d\n", ep->name,
+				status, req->actual, req->length);
+		break;
+	case -EREMOTEIO:		/* short read */
+		break;
+	}
+
+	status = usb_ep_queue(ep, req, GFP_ATOMIC);
+	if (status) {
+		ERROR(dev, "kill %s:  resubmit %d bytes --> %d\n",
+				ep->name, req->length, status);
+		usb_ep_set_halt(ep);
+		/* FIXME recover later ... somehow */
+	}
+}
+
+static int set_gmidi_config(struct gmidi_device *dev, gfp_t gfp_flags)
+{
+	int err = 0;
+	struct usb_request *req;
+	struct usb_ep* ep;
+	unsigned i;
+
+	err = usb_ep_enable(dev->in_ep, &bulk_in_desc);
+	if (err) {
+		ERROR(dev, "can't start %s: %d\n", dev->in_ep->name, err);
+		goto fail;
+	}
+	dev->in_ep->driver_data = dev;
+
+	err = usb_ep_enable(dev->out_ep, &bulk_out_desc);
+	if (err) {
+		ERROR(dev, "can't start %s: %d\n", dev->out_ep->name, err);
+		goto fail;
+	}
+	dev->out_ep->driver_data = dev;
+
+	/* allocate a bunch of read buffers and queue them all at once. */
+	ep = dev->out_ep;
+	for (i = 0; i < qlen && err == 0; i++) {
+		req = alloc_ep_req(ep, buflen);
+		if (req) {
+			req->complete = gmidi_complete;
+			err = usb_ep_queue(ep, req, GFP_ATOMIC);
+			if (err) {
+				DBG(dev, "%s queue req: %d\n", ep->name, err);
+			}
+		} else {
+			err = -ENOMEM;
+		}
+	}
+fail:
+	/* caller is responsible for cleanup on error */
+	return err;
+}
+
+
+static void gmidi_reset_config(struct gmidi_device *dev)
+{
+	if (dev->config == 0) {
+		return;
+	}
+
+	DBG(dev, "reset config\n");
+
+	/* just disable endpoints, forcing completion of pending i/o.
+	 * all our completion handlers free their requests in this case.
+	 */
+	usb_ep_disable(dev->in_ep);
+	usb_ep_disable(dev->out_ep);
+	dev->config = 0;
+}
+
+/* change our operational config.  this code must agree with the code
+ * that returns config descriptors, and altsetting code.
+ *
+ * it's also responsible for power management interactions. some
+ * configurations might not work with our current power sources.
+ *
+ * note that some device controller hardware will constrain what this
+ * code can do, perhaps by disallowing more than one configuration or
+ * by limiting configuration choices (like the pxa2xx).
+ */
+static int
+gmidi_set_config(struct gmidi_device *dev, unsigned number, gfp_t gfp_flags)
+{
+	int result = 0;
+	struct usb_gadget *gadget = dev->gadget;
+
+#if 0
+	/* FIXME */
+	/* Hacking this bit out fixes a bug where on receipt of two
+	   USB_REQ_SET_CONFIGURATION messages, we end up with no
+	   buffered OUT requests waiting for data. This is clearly
+	   hiding a bug elsewhere, because if the config didn't
+	   change then we really shouldn't do anything. */
+	/* Having said that, when we do "change" from config 1
+	   to config 1, we at least gmidi_reset_config() which
+	   clears out any requests on endpoints, so it's not like
+	   we leak or anything. */
+	if (number == dev->config) {
+		return 0;
+	}
+#endif
+
+	if (gadget_is_sa1100(gadget) && dev->config) {
+		/* tx fifo is full, but we can't clear it...*/
+		INFO(dev, "can't change configurations\n");
+		return -ESPIPE;
+	}
+	gmidi_reset_config(dev);
+
+	switch (number) {
+	case GMIDI_CONFIG:
+		result = set_gmidi_config(dev, gfp_flags);
+		break;
+	default:
+		result = -EINVAL;
+		/* FALL THROUGH */
+	case 0:
+		return result;
+	}
+
+	if (!result && (!dev->in_ep || !dev->out_ep)) {
+		result = -ENODEV;
+	}
+	if (result) {
+		gmidi_reset_config(dev);
+	} else {
+		char *speed;
+
+		switch (gadget->speed) {
+		case USB_SPEED_LOW:	speed = "low"; break;
+		case USB_SPEED_FULL:	speed = "full"; break;
+		case USB_SPEED_HIGH:	speed = "high"; break;
+		default: 		speed = "?"; break;
+		}
+
+		dev->config = number;
+		INFO(dev, "%s speed\n", speed);
+	}
+	return result;
+}
+
+
+static void gmidi_setup_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	if (req->status || req->actual != req->length) {
+		DBG((struct gmidi_device *) ep->driver_data,
+				"setup complete --> %d, %d/%d\n",
+				req->status, req->actual, req->length);
+	}
+}
+
+/*
+ * The setup() callback implements all the ep0 functionality that's
+ * not handled lower down, in hardware or the hardware driver (like
+ * device and endpoint feature flags, and their status).  It's all
+ * housekeeping for the gadget function we're implementing.  Most of
+ * the work is in config-specific setup.
+ */
+static int gmidi_setup(struct usb_gadget *gadget,
+			const struct usb_ctrlrequest *ctrl)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	struct usb_request *req = dev->req;
+	int value = -EOPNOTSUPP;
+	u16 w_index = le16_to_cpu(ctrl->wIndex);
+	u16 w_value = le16_to_cpu(ctrl->wValue);
+	u16 w_length = le16_to_cpu(ctrl->wLength);
+
+	/* usually this stores reply data in the pre-allocated ep0 buffer,
+	 * but config change events will reconfigure hardware.
+	 */
+	req->zero = 0;
+	switch (ctrl->bRequest) {
+
+	case USB_REQ_GET_DESCRIPTOR:
+		if (ctrl->bRequestType != USB_DIR_IN) {
+			goto unknown;
+		}
+		switch (w_value >> 8) {
+
+		case USB_DT_DEVICE:
+			value = min(w_length, (u16) sizeof(device_desc));
+			memcpy(req->buf, &device_desc, value);
+			break;
+		case USB_DT_CONFIG:
+			value = config_buf(gadget, req->buf,
+					w_value >> 8,
+					w_value & 0xff);
+			if (value >= 0) {
+				value = min(w_length, (u16)value);
+			}
+			break;
+
+		case USB_DT_STRING:
+			/* wIndex == language code.
+			 * this driver only handles one language, you can
+			 * add string tables for other languages, using
+			 * any UTF-8 characters
+			 */
+			value = usb_gadget_get_string(&stringtab,
+					w_value & 0xff, req->buf);
+			if (value >= 0) {
+				value = min(w_length, (u16)value);
+			}
+			break;
+		}
+		break;
+
+	/* currently two configs, two speeds */
+	case USB_REQ_SET_CONFIGURATION:
+		if (ctrl->bRequestType != 0) {
+			goto unknown;
+		}
+		if (gadget->a_hnp_support) {
+			DBG(dev, "HNP available\n");
+		} else if (gadget->a_alt_hnp_support) {
+			DBG(dev, "HNP needs a different root port\n");
+		} else {
+			VDBG(dev, "HNP inactive\n");
+		}
+		spin_lock(&dev->lock);
+		value = gmidi_set_config(dev, w_value, GFP_ATOMIC);
+		spin_unlock(&dev->lock);
+		break;
+	case USB_REQ_GET_CONFIGURATION:
+		if (ctrl->bRequestType != USB_DIR_IN) {
+			goto unknown;
+		}
+		*(u8 *)req->buf = dev->config;
+		value = min(w_length, (u16)1);
+		break;
+
+	/* until we add altsetting support, or other interfaces,
+	 * only 0/0 are possible.  pxa2xx only supports 0/0 (poorly)
+	 * and already killed pending endpoint I/O.
+	 */
+	case USB_REQ_SET_INTERFACE:
+		if (ctrl->bRequestType != USB_RECIP_INTERFACE) {
+			goto unknown;
+		}
+		spin_lock(&dev->lock);
+		if (dev->config && w_index < GMIDI_NUM_INTERFACES
+			&& w_value == 0)
+		{
+			u8 config = dev->config;
+
+			/* resets interface configuration, forgets about
+			 * previous transaction state (queued bufs, etc)
+			 * and re-inits endpoint state (toggle etc)
+			 * no response queued, just zero status == success.
+			 * if we had more than one interface we couldn't
+			 * use this "reset the config" shortcut.
+			 */
+			gmidi_reset_config(dev);
+			gmidi_set_config(dev, config, GFP_ATOMIC);
+			value = 0;
+		}
+		spin_unlock(&dev->lock);
+		break;
+	case USB_REQ_GET_INTERFACE:
+		if (ctrl->bRequestType != (USB_DIR_IN|USB_RECIP_INTERFACE)) {
+			goto unknown;
+		}
+		if (!dev->config) {
+			break;
+		}
+		if (w_index >= GMIDI_NUM_INTERFACES) {
+			value = -EDOM;
+			break;
+		}
+		*(u8 *)req->buf = 0;
+		value = min(w_length, (u16)1);
+		break;
+
+	default:
+unknown:
+		VDBG(dev, "unknown control req%02x.%02x v%04x i%04x l%d\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+	}
+
+	/* respond with data transfer before status phase? */
+	if (value >= 0) {
+		req->length = value;
+		req->zero = value < w_length;
+		value = usb_ep_queue(gadget->ep0, req, GFP_ATOMIC);
+		if (value < 0) {
+			DBG(dev, "ep_queue --> %d\n", value);
+			req->status = 0;
+			gmidi_setup_complete(gadget->ep0, req);
+		}
+	}
+
+	/* device either stalls (value < 0) or reports success */
+	return value;
+}
+
+static void gmidi_disconnect(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	gmidi_reset_config(dev);
+
+	/* a more significant application might have some non-usb
+	 * activities to quiesce here, saving resources like power
+	 * or pushing the notification up a network stack.
+	 */
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	/* next we may get setup() calls to enumerate new connections;
+	 * or an unbind() during shutdown (including removing module).
+	 */
+}
+
+static void /* __init_or_exit */ gmidi_unbind(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+	struct snd_card* card;
+
+	DBG(dev, "unbind\n");
+
+	card = dev->card;
+	dev->card = NULL;
+	if (card) {
+		snd_card_free(card);
+	}
+
+	/* we've already been disconnected ... no i/o is active */
+	if (dev->req) {
+		dev->req->length = USB_BUFSIZ;
+		free_ep_req(gadget->ep0, dev->req);
+	}
+	kfree(dev);
+	set_gadget_data(gadget, NULL);
+}
+
+static int gmidi_snd_free(struct snd_device *device)
+{
+	return 0;
+}
+
+static void gmidi_transmit_packet(struct usb_request* req, uint8_t p0,
+					uint8_t p1, uint8_t p2, uint8_t p3)
+{
+	unsigned length = req->length;
+
+	uint8_t* buf = (uint8_t*)req->buf + length;
+	buf[0] = p0;
+	buf[1] = p1;
+	buf[2] = p2;
+	buf[3] = p3;
+	req->length = length + 4;
+}
+
+/*
+ * Converts MIDI commands to USB MIDI packets.
+ */
+static void gmidi_transmit_byte(struct usb_request* req,
+				struct gmidi_in_port* port, uint8_t b)
+{
+	uint8_t p0 = port->cable;
+
+	if (b >= 0xf8) {
+		gmidi_transmit_packet(req, p0 | 0x0f, b, 0, 0);
+	} else if (b >= 0xf0) {
+		switch (b) {
+		case 0xf0:
+			port->data[0] = b;
+			port->state = STATE_SYSEX_1;
+			break;
+		case 0xf1:
+		case 0xf3:
+			port->data[0] = b;
+			port->state = STATE_1PARAM;
+			break;
+		case 0xf2:
+			port->data[0] = b;
+			port->state = STATE_2PARAM_1;
+			break;
+		case 0xf4:
+		case 0xf5:
+			port->state = STATE_UNKNOWN;
+			break;
+		case 0xf6:
+			gmidi_transmit_packet(req, p0 | 0x05, 0xf6, 0, 0);
+			port->state = STATE_UNKNOWN;
+			break;
+		case 0xf7:
+			switch (port->state) {
+			case STATE_SYSEX_0:
+				gmidi_transmit_packet(req,
+					p0 | 0x05, 0xf7, 0, 0);
+				break;
+			case STATE_SYSEX_1:
+				gmidi_transmit_packet(req,
+					p0 | 0x06, port->data[0], 0xf7, 0);
+				break;
+			case STATE_SYSEX_2:
+				gmidi_transmit_packet(req,
+					p0 | 0x07, port->data[0],
+					port->data[1], 0xf7);
+				break;
+			}
+			port->state = STATE_UNKNOWN;
+			break;
+		}
+	} else if (b >= 0x80) {
+		port->data[0] = b;
+		if (b >= 0xc0 && b <= 0xdf)
+			port->state = STATE_1PARAM;
+		else
+			port->state = STATE_2PARAM_1;
+	} else { /* b < 0x80 */
+		switch (port->state) {
+		case STATE_1PARAM:
+			if (port->data[0] < 0xf0) {
+				p0 |= port->data[0] >> 4;
+			} else {
+				p0 |= 0x02;
+				port->state = STATE_UNKNOWN;
+			}
+			gmidi_transmit_packet(req, p0, port->data[0], b, 0);
+			break;
+		case STATE_2PARAM_1:
+			port->data[1] = b;
+			port->state = STATE_2PARAM_2;
+			break;
+		case STATE_2PARAM_2:
+			if (port->data[0] < 0xf0) {
+				p0 |= port->data[0] >> 4;
+				port->state = STATE_2PARAM_1;
+			} else {
+				p0 |= 0x03;
+				port->state = STATE_UNKNOWN;
+			}
+			gmidi_transmit_packet(req,
+				p0, port->data[0], port->data[1], b);
+			break;
+		case STATE_SYSEX_0:
+			port->data[0] = b;
+			port->state = STATE_SYSEX_1;
+			break;
+		case STATE_SYSEX_1:
+			port->data[1] = b;
+			port->state = STATE_SYSEX_2;
+			break;
+		case STATE_SYSEX_2:
+			gmidi_transmit_packet(req,
+				p0 | 0x04, port->data[0], port->data[1], b);
+			port->state = STATE_SYSEX_0;
+			break;
+		}
+	}
+}
+
+static void gmidi_transmit(struct gmidi_device* dev, struct usb_request* req)
+{
+	struct usb_ep* ep = dev->in_ep;
+	struct gmidi_in_port* port = &dev->in_port;
+
+	if (!ep) {
+		return;
+	}
+	if (!req) {
+		req = alloc_ep_req(ep, buflen);
+	}
+	if (!req) {
+		ERROR(dev, "gmidi_transmit: alloc_ep_request failed\n");
+		return;
+	}
+	req->length = 0;
+	req->complete = gmidi_complete;
+
+	if (port->active) {
+		while (req->length + 3 < buflen) {
+			uint8_t b;
+			if (snd_rawmidi_transmit(dev->in_substream, &b, 1)
+				!= 1)
+			{
+				port->active = 0;
+				break;
+			}
+			gmidi_transmit_byte(req, port, b);
+		}
+	}
+	if (req->length > 0) {
+		usb_ep_queue(ep, req, GFP_ATOMIC);
+	} else {
+		free_ep_req(ep, req);
+	}
+}
+
+static void gmidi_in_tasklet(unsigned long data)
+{
+	struct gmidi_device* dev = (struct gmidi_device*)data;
+
+	gmidi_transmit(dev, NULL);
+}
+
+static int gmidi_in_open(struct snd_rawmidi_substream *substream)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_in_open\n");
+	dev->in_substream = substream;
+	dev->in_port.state = STATE_UNKNOWN;
+	return 0;
+}
+
+static int gmidi_in_close(struct snd_rawmidi_substream *substream)
+{
+	VDBG(dev, "gmidi_in_close\n");
+	return 0;
+}
+
+static void gmidi_in_trigger(struct snd_rawmidi_substream *substream, int up)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_in_trigger %d\n", up);
+	dev->in_port.active = up;
+	if (up) {
+		tasklet_hi_schedule(&dev->tasklet);
+	}
+}
+
+static int gmidi_out_open(struct snd_rawmidi_substream *substream)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_out_open\n");
+	dev->out_substream = substream;
+	return 0;
+}
+
+static int gmidi_out_close(struct snd_rawmidi_substream *substream)
+{
+	VDBG(dev, "gmidi_out_close\n");
+	return 0;
+}
+
+static void gmidi_out_trigger(struct snd_rawmidi_substream *substream, int up)
+{
+	struct gmidi_device* dev = substream->rmidi->private_data;
+
+	VDBG(dev, "gmidi_out_trigger %d\n", up);
+	if (up) {
+		set_bit(substream->number, &dev->out_triggered);
+	} else {
+		clear_bit(substream->number, &dev->out_triggered);
+	}
+}
+
+static struct snd_rawmidi_ops gmidi_in_ops = {
+	.open = gmidi_in_open,
+	.close = gmidi_in_close,
+	.trigger = gmidi_in_trigger,
+};
+
+static struct snd_rawmidi_ops gmidi_out_ops = {
+	.open = gmidi_out_open,
+	.close = gmidi_out_close,
+	.trigger = gmidi_out_trigger
+};
+
+/* register as a sound "card" */
+static int gmidi_register_card(struct gmidi_device *dev)
+{
+	struct snd_card *card;
+	struct snd_rawmidi *rmidi;
+	int err;
+	int out_ports = 1;
+	int in_ports = 1;
+	static struct snd_device_ops ops = {
+		.dev_free = gmidi_snd_free,
+	};
+
+	card = snd_card_new(index, id, THIS_MODULE, 0);
+	if (!card) {
+		ERROR(dev, "snd_card_new failed\n");
+		err = -ENOMEM;
+		goto fail;
+	}
+	dev->card = card;
+
+	err = snd_device_new(card, SNDRV_DEV_LOWLEVEL, dev, &ops);
+	if (err < 0) {
+		ERROR(dev, "snd_device_new failed: error %d\n", err);
+		goto fail;
+	}
+
+	strcpy(card->driver, longname);
+	strcpy(card->longname, longname);
+	strcpy(card->shortname, shortname);
+
+	/* Set up rawmidi */
+	dev->in_port.dev = dev;
+	dev->in_port.active = 0;
+	snd_component_add(card, "MIDI");
+	err = snd_rawmidi_new(card, "USB MIDI Gadget", 0,
+			      out_ports, in_ports, &rmidi);
+	if (err < 0) {
+		ERROR(dev, "snd_rawmidi_new failed: error %d\n", err);
+		goto fail;
+	}
+	dev->rmidi = rmidi;
+	strcpy(rmidi->name, card->shortname);
+	rmidi->info_flags = SNDRV_RAWMIDI_INFO_OUTPUT |
+			    SNDRV_RAWMIDI_INFO_INPUT |
+			    SNDRV_RAWMIDI_INFO_DUPLEX;
+	rmidi->private_data = dev;
+
+	/* Yes, rawmidi OUTPUT = USB IN, and rawmidi INPUT = USB OUT.
+	   It's an upside-down world being a gadget. */
+	snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_OUTPUT, &gmidi_in_ops);
+	snd_rawmidi_set_ops(rmidi, SNDRV_RAWMIDI_STREAM_INPUT, &gmidi_out_ops);
+
+	snd_card_set_dev(card, &dev->gadget->dev);
+
+	/* register it - we're ready to go */
+	err = snd_card_register(card);
+	if (err < 0) {
+		ERROR(dev, "snd_card_register failed\n");
+		goto fail;
+	}
+
+	VDBG(dev, "gmidi_register_card finished ok\n");
+	return 0;
+
+fail:
+	if (dev->card) {
+		snd_card_free(dev->card);
+		dev->card = NULL;
+	}
+	return err;
+}
+
+/*
+ * Creates an output endpoint, and initializes output ports.
+ */
+static int __devinit gmidi_bind(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev;
+	struct usb_ep *in_ep, *out_ep;
+	int gcnum, err = 0;
+
+	/* support optional vendor/distro customization */
+	if (idVendor) {
+		if (!idProduct) {
+			printk(KERN_ERR "idVendor needs idProduct!\n");
+			return -ENODEV;
+		}
+		device_desc.idVendor = cpu_to_le16(idVendor);
+		device_desc.idProduct = cpu_to_le16(idProduct);
+		if (bcdDevice) {
+			device_desc.bcdDevice = cpu_to_le16(bcdDevice);
+		}
+	}
+	if (iManufacturer) {
+		strlcpy(manufacturer, iManufacturer, sizeof(manufacturer));
+	} else {
+		snprintf(manufacturer, sizeof(manufacturer), "%s %s with %s",
+			system_utsname.sysname, system_utsname.release,
+			gadget->name);
+	}
+	if (iProduct) {
+		strlcpy(product_desc, iProduct, sizeof(product_desc));
+	}
+	if (iSerialNumber) {
+		device_desc.iSerialNumber = STRING_SERIAL,
+		strlcpy(serial_number, iSerialNumber, sizeof(serial_number));
+	}
+
+	/* Bulk-only drivers like this one SHOULD be able to
+	 * autoconfigure on any sane usb controller driver,
+	 * but there may also be important quirks to address.
+	 */
+	usb_ep_autoconfig_reset(gadget);
+	in_ep = usb_ep_autoconfig(gadget, &bulk_in_desc);
+	if (!in_ep) {
+autoconf_fail:
+		printk(KERN_ERR "%s: can't autoconfigure on %s\n",
+			shortname, gadget->name);
+		return -ENODEV;
+	}
+	EP_IN_NAME = in_ep->name;
+	in_ep->driver_data = in_ep;	/* claim */
+
+	out_ep = usb_ep_autoconfig(gadget, &bulk_out_desc);
+	if (!out_ep) {
+		goto autoconf_fail;
+	}
+	EP_OUT_NAME = out_ep->name;
+	out_ep->driver_data = out_ep;	/* claim */
+
+	gcnum = usb_gadget_controller_number(gadget);
+	if (gcnum >= 0) {
+		device_desc.bcdDevice = cpu_to_le16(0x0200 + gcnum);
+	} else {
+		/* gmidi is so simple (no altsettings) that
+		 * it SHOULD NOT have problems with bulk-capable hardware.
+		 * so warn about unrecognized controllers, don't panic.
+		 */
+		printk(KERN_WARNING "%s: controller '%s' not recognized\n",
+			shortname, gadget->name);
+		device_desc.bcdDevice = __constant_cpu_to_le16(0x9999);
+	}
+
+
+	/* ok, we made sense of the hardware ... */
+	dev = kzalloc(sizeof(*dev), SLAB_KERNEL);
+	if (!dev) {
+		return -ENOMEM;
+	}
+	spin_lock_init(&dev->lock);
+	dev->gadget = gadget;
+	dev->in_ep = in_ep;
+	dev->out_ep = out_ep;
+	set_gadget_data(gadget, dev);
+	tasklet_init(&dev->tasklet, gmidi_in_tasklet, (unsigned long)dev);
+
+	/* preallocate control response and buffer */
+	dev->req = usb_ep_alloc_request(gadget->ep0, GFP_KERNEL);
+	if (!dev->req) {
+		err = -ENOMEM;
+		goto fail;
+	}
+	dev->req->buf = usb_ep_alloc_buffer(gadget->ep0, USB_BUFSIZ,
+				&dev->req->dma, GFP_KERNEL);
+	if (!dev->req->buf) {
+		err = -ENOMEM;
+		goto fail;
+	}
+
+	dev->req->complete = gmidi_setup_complete;
+
+	device_desc.bMaxPacketSize0 = gadget->ep0->maxpacket;
+
+	gadget->ep0->driver_data = dev;
+
+	INFO(dev, "%s, version: " DRIVER_VERSION "\n", longname);
+	INFO(dev, "using %s, OUT %s IN %s\n", gadget->name,
+		EP_OUT_NAME, EP_IN_NAME);
+
+	/* register as an ALSA sound card */
+	err = gmidi_register_card(dev);
+	if (err < 0) {
+		goto fail;
+	}
+
+	VDBG(dev, "gmidi_bind finished ok\n");
+	return 0;
+
+fail:
+	gmidi_unbind(gadget);
+	return err;
+}
+
+
+static void gmidi_suspend(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+
+	if (gadget->speed == USB_SPEED_UNKNOWN) {
+		return;
+	}
+
+	DBG(dev, "suspend\n");
+}
+
+static void gmidi_resume(struct usb_gadget *gadget)
+{
+	struct gmidi_device *dev = get_gadget_data(gadget);
+
+	DBG(dev, "resume\n");
+}
+
+
+static struct usb_gadget_driver gmidi_driver = {
+	.speed		= USB_SPEED_FULL,
+	.function	= (char *)longname,
+	.bind		= gmidi_bind,
+	.unbind		= __exit_p(gmidi_unbind),
+
+	.setup		= gmidi_setup,
+	.disconnect	= gmidi_disconnect,
+
+	.suspend	= gmidi_suspend,
+	.resume		= gmidi_resume,
+
+	.driver 	= {
+		.name		= (char *)shortname,
+		.owner		= THIS_MODULE,
+	},
+};
+
+static int __init gmidi_init(void)
+{
+	return usb_gadget_register_driver(&gmidi_driver);
+}
+module_init(gmidi_init);
+
+static void __exit gmidi_cleanup(void)
+{
+	usb_gadget_unregister_driver(&gmidi_driver);
+}
+module_exit(gmidi_cleanup);
+
diff --git a/include/linux/usb/audio.h b/include/linux/usb/audio.h
new file mode 100644
index 000000000000..6bd235994dc2
--- /dev/null
+++ b/include/linux/usb/audio.h
@@ -0,0 +1,53 @@
+/*
+ * <linux/usb/audio.h> -- USB Audio definitions.
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This file holds USB constants and structures defined
+ * by the USB Device Class Definition for Audio Devices.
+ * Comments below reference relevant sections of that document:
+ *
+ * http://www.usb.org/developers/devclass_docs/audio10.pdf
+ */
+
+#ifndef __LINUX_USB_AUDIO_H
+#define __LINUX_USB_AUDIO_H
+
+#include <linux/types.h>
+
+/* A.2 Audio Interface Subclass Codes */
+#define USB_SUBCLASS_AUDIOCONTROL	0x01
+#define USB_SUBCLASS_AUDIOSTREAMING	0x02
+#define USB_SUBCLASS_MIDISTREAMING	0x03
+
+/* 4.3.2  Class-Specific AC Interface Descriptor */
+struct usb_ac_header_descriptor {
+	__u8  bLength;			// 8+n
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_HEADER
+	__le16 bcdADC;			// 0x0100
+	__le16 wTotalLength;		// includes Unit and Terminal desc.
+	__u8  bInCollection;		// n
+	__u8  baInterfaceNr[];		// [n]
+} __attribute__ ((packed));
+
+#define USB_DT_AC_HEADER_SIZE(n)	(8+(n))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_AC_HEADER_DESCRIPTOR(n) 			\
+struct usb_ac_header_descriptor_##n {				\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__le16 bcdADC;						\
+	__le16 wTotalLength;					\
+	__u8  bInCollection;					\
+	__u8  baInterfaceNr[n];					\
+} __attribute__ ((packed))
+
+#endif
diff --git a/include/linux/usb/midi.h b/include/linux/usb/midi.h
new file mode 100644
index 000000000000..11a97d5ffd34
--- /dev/null
+++ b/include/linux/usb/midi.h
@@ -0,0 +1,112 @@
+/*
+ * <linux/usb/midi.h> -- USB MIDI definitions.
+ *
+ * Copyright (C) 2006 Thumtronics Pty Ltd.
+ * Developed for Thumtronics by Grey Innovation
+ * Ben Williamson <ben.williamson@greyinnovation.com>
+ *
+ * This software is distributed under the terms of the GNU General Public
+ * License ("GPL") version 2, as published by the Free Software Foundation.
+ *
+ * This file holds USB constants and structures defined
+ * by the USB Device Class Definition for MIDI Devices.
+ * Comments below reference relevant sections of that document:
+ *
+ * http://www.usb.org/developers/devclass_docs/midi10.pdf
+ */
+
+#ifndef __LINUX_USB_MIDI_H
+#define __LINUX_USB_MIDI_H
+
+#include <linux/types.h>
+
+/* A.1  MS Class-Specific Interface Descriptor Subtypes */
+#define USB_MS_HEADER		0x01
+#define USB_MS_MIDI_IN_JACK	0x02
+#define USB_MS_MIDI_OUT_JACK	0x03
+#define USB_MS_ELEMENT		0x04
+
+/* A.2  MS Class-Specific Endpoint Descriptor Subtypes */
+#define USB_MS_GENERAL		0x01
+
+/* A.3  MS MIDI IN and OUT Jack Types */
+#define USB_MS_EMBEDDED		0x01
+#define USB_MS_EXTERNAL		0x02
+
+/* 6.1.2.1  Class-Specific MS Interface Header Descriptor */
+struct usb_ms_header_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;
+	__u8  bDescriptorSubtype;
+	__le16 bcdMSC;
+	__le16 wTotalLength;
+} __attribute__ ((packed));
+
+#define USB_DT_MS_HEADER_SIZE	7
+
+/* 6.1.2.2  MIDI IN Jack Descriptor */
+struct usb_midi_in_jack_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_MIDI_IN_JACK
+	__u8  bJackType;		// USB_MS_EMBEDDED/EXTERNAL
+	__u8  bJackID;
+	__u8  iJack;
+} __attribute__ ((packed));
+
+#define USB_DT_MIDI_IN_SIZE	6
+
+struct usb_midi_source_pin {
+	__u8  baSourceID;
+	__u8  baSourcePin;
+} __attribute__ ((packed));
+
+/* 6.1.2.3  MIDI OUT Jack Descriptor */
+struct usb_midi_out_jack_descriptor {
+	__u8  bLength;
+	__u8  bDescriptorType;		// USB_DT_CS_INTERFACE
+	__u8  bDescriptorSubtype;	// USB_MS_MIDI_OUT_JACK
+	__u8  bJackType;		// USB_MS_EMBEDDED/EXTERNAL
+	__u8  bJackID;
+	__u8  bNrInputPins;		// p
+	struct usb_midi_source_pin pins[]; // [p]
+	/*__u8  iJack;  -- ommitted due to variable-sized pins[] */
+} __attribute__ ((packed));
+
+#define USB_DT_MIDI_OUT_SIZE(p)	(7 + 2 * (p))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(p)			\
+struct usb_midi_out_jack_descriptor_##p {			\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__u8  bJackType;					\
+	__u8  bJackID;						\
+	__u8  bNrInputPins;					\
+	struct usb_midi_source_pin pins[p];			\
+	__u8  iJack;						\
+} __attribute__ ((packed))
+
+/* 6.2.2  Class-Specific MS Bulk Data Endpoint Descriptor */
+struct usb_ms_endpoint_descriptor {
+	__u8  bLength;			// 4+n
+	__u8  bDescriptorType;		// USB_DT_CS_ENDPOINT
+	__u8  bDescriptorSubtype;	// USB_MS_GENERAL
+	__u8  bNumEmbMIDIJack;		// n
+	__u8  baAssocJackID[];		// [n]
+} __attribute__ ((packed));
+
+#define USB_DT_MS_ENDPOINT_SIZE(n)	(4 + (n))
+
+/* As above, but more useful for defining your own descriptors: */
+#define DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(n)			\
+struct usb_ms_endpoint_descriptor_##n {				\
+	__u8  bLength;						\
+	__u8  bDescriptorType;					\
+	__u8  bDescriptorSubtype;				\
+	__u8  bNumEmbMIDIJack;					\
+	__u8  baAssocJackID[n];					\
+} __attribute__ ((packed))
+
+#endif
-- 
cgit v1.2.1


From 066202dd48cf3296b6cc22b5fcf89aef33fa0efc Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Sat, 5 Aug 2006 20:37:11 -0300
Subject: USB: Make file operations structs in drivers/usb const.

Making structs const prevents accidental bugs and with the proper debug
options they're protected against corruption.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/class/usblp.c           | 2 +-
 drivers/usb/core/devices.c          | 2 +-
 drivers/usb/core/devio.c            | 2 +-
 drivers/usb/core/file.c             | 2 +-
 drivers/usb/core/inode.c            | 6 +++---
 drivers/usb/core/usb.h              | 4 ++--
 drivers/usb/gadget/at91_udc.c       | 2 +-
 drivers/usb/gadget/inode.c          | 8 ++++----
 drivers/usb/gadget/omap_udc.c       | 2 +-
 drivers/usb/host/isp116x-hcd.c      | 2 +-
 drivers/usb/host/sl811-hcd.c        | 2 +-
 drivers/usb/host/uhci-debug.c       | 4 ++--
 drivers/usb/image/mdc800.c          | 4 ++--
 drivers/usb/input/hiddev.c          | 2 +-
 drivers/usb/misc/auerswald.c        | 2 +-
 drivers/usb/misc/idmouse.c          | 2 +-
 drivers/usb/misc/ldusb.c            | 2 +-
 drivers/usb/misc/legousbtower.c     | 2 +-
 drivers/usb/misc/sisusbvga/sisusb.c | 2 +-
 drivers/usb/misc/usblcd.c           | 2 +-
 drivers/usb/mon/mon_stat.c          | 2 +-
 drivers/usb/mon/mon_text.c          | 2 +-
 drivers/usb/mon/usb_mon.h           | 4 ++--
 drivers/usb/usb-skeleton.c          | 2 +-
 24 files changed, 33 insertions(+), 33 deletions(-)

diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c
index 48dee4b8d8e5..292919d260a0 100644
--- a/drivers/usb/class/usblp.c
+++ b/drivers/usb/class/usblp.c
@@ -813,7 +813,7 @@ static unsigned int usblp_quirks (__u16 vendor, __u16 product)
 	return 0;
 }
 
-static struct file_operations usblp_fops = {
+static const struct file_operations usblp_fops = {
 	.owner =	THIS_MODULE,
 	.read =		usblp_read,
 	.write =	usblp_write,
diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c
index c0f37343a276..241d37326ea9 100644
--- a/drivers/usb/core/devices.c
+++ b/drivers/usb/core/devices.c
@@ -667,7 +667,7 @@ static loff_t usb_device_lseek(struct file * file, loff_t offset, int orig)
 	return ret;
 }
 
-struct file_operations usbfs_devices_fops = {
+const struct file_operations usbfs_devices_fops = {
 	.llseek =	usb_device_lseek,
 	.read =		usb_device_read,
 	.poll =		usb_device_poll,
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index e84f19d4089c..43c08724a2d8 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -1572,7 +1572,7 @@ static unsigned int usbdev_poll(struct file *file, struct poll_table_struct *wai
 	return mask;
 }
 
-struct file_operations usbfs_device_file_operations = {
+const struct file_operations usbfs_device_file_operations = {
 	.llseek =	usbdev_lseek,
 	.read =		usbdev_read,
 	.poll =		usbdev_poll,
diff --git a/drivers/usb/core/file.c b/drivers/usb/core/file.c
index 8de4f8c99d61..c376c655c5de 100644
--- a/drivers/usb/core/file.c
+++ b/drivers/usb/core/file.c
@@ -55,7 +55,7 @@ static int usb_open(struct inode * inode, struct file * file)
 	return err;
 }
 
-static struct file_operations usb_fops = {
+static const struct file_operations usb_fops = {
 	.owner =	THIS_MODULE,
 	.open =		usb_open,
 };
diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c
index 58b4b1012120..df3d152f0493 100644
--- a/drivers/usb/core/inode.c
+++ b/drivers/usb/core/inode.c
@@ -44,7 +44,7 @@
 #include "hcd.h"
 
 static struct super_operations usbfs_ops;
-static struct file_operations default_file_operations;
+static const struct file_operations default_file_operations;
 static struct vfsmount *usbfs_mount;
 static int usbfs_mount_count;	/* = 0 */
 static int ignore_mount = 0;
@@ -407,7 +407,7 @@ static int default_open (struct inode *inode, struct file *file)
 	return 0;
 }
 
-static struct file_operations default_file_operations = {
+static const struct file_operations default_file_operations = {
 	.read =		default_read_file,
 	.write =	default_write_file,
 	.open =		default_open,
@@ -494,7 +494,7 @@ static int fs_create_by_name (const char *name, mode_t mode,
 
 static struct dentry *fs_create_file (const char *name, mode_t mode,
 				      struct dentry *parent, void *data,
-				      struct file_operations *fops,
+				      const struct file_operations *fops,
 				      uid_t uid, gid_t gid)
 {
 	struct dentry *dentry;
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 98675fb1bc4c..6096ead2758c 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -90,8 +90,8 @@ extern const char *usbcore_name;
 /* usbfs stuff */
 extern struct mutex usbfs_mutex;
 extern struct usb_driver usbfs_driver;
-extern struct file_operations usbfs_devices_fops;
-extern struct file_operations usbfs_device_file_operations;
+extern const struct file_operations usbfs_devices_fops;
+extern const struct file_operations usbfs_device_file_operations;
 extern void usbfs_conn_disc_event(void);
 
 extern int usbdev_init(void);
diff --git a/drivers/usb/gadget/at91_udc.c b/drivers/usb/gadget/at91_udc.c
index cfebca05ead5..d00958a01cfb 100644
--- a/drivers/usb/gadget/at91_udc.c
+++ b/drivers/usb/gadget/at91_udc.c
@@ -247,7 +247,7 @@ static int proc_udc_open(struct inode *inode, struct file *file)
 	return single_open(file, proc_udc_show, PDE(inode)->data);
 }
 
-static struct file_operations proc_ops = {
+static const struct file_operations proc_ops = {
 	.open		= proc_udc_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index ab508858f341..1072e987ff21 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -714,7 +714,7 @@ ep_aio_write(struct kiocb *iocb, const char __user *ubuf, size_t len, loff_t o)
 /*----------------------------------------------------------------------*/
 
 /* used after endpoint configuration */
-static struct file_operations ep_io_operations = {
+static const struct file_operations ep_io_operations = {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
 
@@ -873,7 +873,7 @@ ep_open (struct inode *inode, struct file *fd)
 }
 
 /* used before endpoint configuration */
-static struct file_operations ep_config_operations = {
+static const struct file_operations ep_config_operations = {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
 
@@ -1247,7 +1247,7 @@ static int dev_ioctl (struct inode *inode, struct file *fd,
 }
 
 /* used after device configuration */
-static struct file_operations ep0_io_operations = {
+static const struct file_operations ep0_io_operations = {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
 
@@ -1929,7 +1929,7 @@ dev_open (struct inode *inode, struct file *fd)
 	return value;
 }
 
-static struct file_operations dev_init_operations = {
+static const struct file_operations dev_init_operations = {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
 
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 81f0389fcc94..0a64504c2545 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -2437,7 +2437,7 @@ static int proc_udc_open(struct inode *inode, struct file *file)
 	return single_open(file, proc_udc_show, NULL);
 }
 
-static struct file_operations proc_ops = {
+static const struct file_operations proc_ops = {
 	.open		= proc_udc_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
index 8c6b38a0b5bb..a72e041df8e7 100644
--- a/drivers/usb/host/isp116x-hcd.c
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -1207,7 +1207,7 @@ static int isp116x_open_seq(struct inode *inode, struct file *file)
 	return single_open(file, isp116x_show_dbg, inode->i_private);
 }
 
-static struct file_operations isp116x_debug_fops = {
+static const struct file_operations isp116x_debug_fops = {
 	.open = isp116x_open_seq,
 	.read = seq_read,
 	.llseek = seq_lseek,
diff --git a/drivers/usb/host/sl811-hcd.c b/drivers/usb/host/sl811-hcd.c
index 9de115d9db27..8c17da37600b 100644
--- a/drivers/usb/host/sl811-hcd.c
+++ b/drivers/usb/host/sl811-hcd.c
@@ -1517,7 +1517,7 @@ static int proc_sl811h_open(struct inode *inode, struct file *file)
 	return single_open(file, proc_sl811h_show, PDE(inode)->data);
 }
 
-static struct file_operations proc_ops = {
+static const struct file_operations proc_ops = {
 	.open		= proc_sl811h_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
diff --git a/drivers/usb/host/uhci-debug.c b/drivers/usb/host/uhci-debug.c
index d1372cb27f33..e345f15b7d87 100644
--- a/drivers/usb/host/uhci-debug.c
+++ b/drivers/usb/host/uhci-debug.c
@@ -16,7 +16,7 @@
 
 #include "uhci-hcd.h"
 
-#define uhci_debug_operations (* (struct file_operations *) NULL)
+#define uhci_debug_operations (* (const struct file_operations *) NULL)
 static struct dentry *uhci_debugfs_root;
 
 #ifdef DEBUG
@@ -500,7 +500,7 @@ static int uhci_debug_release(struct inode *inode, struct file *file)
 }
 
 #undef uhci_debug_operations
-static struct file_operations uhci_debug_operations = {
+static const struct file_operations uhci_debug_operations = {
 	.owner =	THIS_MODULE,
 	.open =		uhci_debug_open,
 	.llseek =	uhci_debug_lseek,
diff --git a/drivers/usb/image/mdc800.c b/drivers/usb/image/mdc800.c
index 08daf400f985..ca6305c1d64c 100644
--- a/drivers/usb/image/mdc800.c
+++ b/drivers/usb/image/mdc800.c
@@ -424,7 +424,7 @@ static void mdc800_usb_download_notify (struct urb *urb, struct pt_regs *res)
  ***************************************************************************/
 
 static struct usb_driver mdc800_usb_driver;
-static struct file_operations mdc800_device_ops;
+static const struct file_operations mdc800_device_ops;
 static struct usb_class_driver mdc800_class = {
 	.name =		"mdc800%d",
 	.fops =		&mdc800_device_ops,
@@ -941,7 +941,7 @@ static ssize_t mdc800_device_write (struct file *file, const char __user *buf, s
 ****************************************************************************/
 
 /* File Operations of this drivers */
-static struct file_operations mdc800_device_ops =
+static const struct file_operations mdc800_device_ops =
 {
 	.owner =	THIS_MODULE,
 	.read =		mdc800_device_read,
diff --git a/drivers/usb/input/hiddev.c b/drivers/usb/input/hiddev.c
index f6b839c257a7..a2b419d13740 100644
--- a/drivers/usb/input/hiddev.c
+++ b/drivers/usb/input/hiddev.c
@@ -722,7 +722,7 @@ inval:
 	return -EINVAL;
 }
 
-static struct file_operations hiddev_fops = {
+static const struct file_operations hiddev_fops = {
 	.owner =	THIS_MODULE,
 	.read =		hiddev_read,
 	.write =	hiddev_write,
diff --git a/drivers/usb/misc/auerswald.c b/drivers/usb/misc/auerswald.c
index 1fef36e71c57..4a329d8488b7 100644
--- a/drivers/usb/misc/auerswald.c
+++ b/drivers/usb/misc/auerswald.c
@@ -1858,7 +1858,7 @@ static int auerchar_release (struct inode *inode, struct file *file)
 
 /*----------------------------------------------------------------------*/
 /* File operation structure                                             */
-static struct file_operations auerswald_fops =
+static const struct file_operations auerswald_fops =
 {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
diff --git a/drivers/usb/misc/idmouse.c b/drivers/usb/misc/idmouse.c
index fcd69c52aea9..8e6e195a22ba 100644
--- a/drivers/usb/misc/idmouse.c
+++ b/drivers/usb/misc/idmouse.c
@@ -98,7 +98,7 @@ static int idmouse_probe(struct usb_interface *interface,
 static void idmouse_disconnect(struct usb_interface *interface);
 
 /* file operation pointers */
-static struct file_operations idmouse_fops = {
+static const struct file_operations idmouse_fops = {
 	.owner = THIS_MODULE,
 	.read = idmouse_read,
 	.open = idmouse_open,
diff --git a/drivers/usb/misc/ldusb.c b/drivers/usb/misc/ldusb.c
index f30ab1fbb3c8..741736b6c134 100644
--- a/drivers/usb/misc/ldusb.c
+++ b/drivers/usb/misc/ldusb.c
@@ -589,7 +589,7 @@ exit:
 }
 
 /* file operations needed when we register this driver */
-static struct file_operations ld_usb_fops = {
+static const struct file_operations ld_usb_fops = {
 	.owner =	THIS_MODULE,
 	.read  =	ld_usb_read,
 	.write =	ld_usb_write,
diff --git a/drivers/usb/misc/legousbtower.c b/drivers/usb/misc/legousbtower.c
index 7699d970e680..77c36e63c7bf 100644
--- a/drivers/usb/misc/legousbtower.c
+++ b/drivers/usb/misc/legousbtower.c
@@ -259,7 +259,7 @@ static void tower_disconnect	(struct usb_interface *interface);
 static DEFINE_MUTEX (disconnect_mutex);
 
 /* file operations needed when we register this driver */
-static struct file_operations tower_fops = {
+static const struct file_operations tower_fops = {
 	.owner =	THIS_MODULE,
 	.read  =	tower_read,
 	.write =	tower_write,
diff --git a/drivers/usb/misc/sisusbvga/sisusb.c b/drivers/usb/misc/sisusbvga/sisusb.c
index e16582f3733c..a44124c7e851 100644
--- a/drivers/usb/misc/sisusbvga/sisusb.c
+++ b/drivers/usb/misc/sisusbvga/sisusb.c
@@ -3179,7 +3179,7 @@ sisusb_compat_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
 }
 #endif
 
-static struct file_operations usb_sisusb_fops = {
+static const struct file_operations usb_sisusb_fops = {
 	.owner =	THIS_MODULE,
 	.open =		sisusb_open,
 	.release =	sisusb_release,
diff --git a/drivers/usb/misc/usblcd.c b/drivers/usb/misc/usblcd.c
index e095772dd8e9..53e59e362330 100644
--- a/drivers/usb/misc/usblcd.c
+++ b/drivers/usb/misc/usblcd.c
@@ -239,7 +239,7 @@ error:
 	return retval;
 }
 
-static struct file_operations lcd_fops = {
+static const struct file_operations lcd_fops = {
         .owner =        THIS_MODULE,
         .read =         lcd_read,
         .write =        lcd_write,
diff --git a/drivers/usb/mon/mon_stat.c b/drivers/usb/mon/mon_stat.c
index 86ad2b381c4b..f6d1491256c4 100644
--- a/drivers/usb/mon/mon_stat.c
+++ b/drivers/usb/mon/mon_stat.c
@@ -62,7 +62,7 @@ static int mon_stat_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-struct file_operations mon_fops_stat = {
+const struct file_operations mon_fops_stat = {
 	.owner =	THIS_MODULE,
 	.open =		mon_stat_open,
 	.llseek =	no_llseek,
diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c
index 2fd39b4fa166..775e374a960c 100644
--- a/drivers/usb/mon/mon_text.c
+++ b/drivers/usb/mon/mon_text.c
@@ -435,7 +435,7 @@ static int mon_text_release(struct inode *inode, struct file *file)
 	return 0;
 }
 
-struct file_operations mon_fops_text = {
+const struct file_operations mon_fops_text = {
 	.owner =	THIS_MODULE,
 	.open =		mon_text_open,
 	.llseek =	no_llseek,
diff --git a/drivers/usb/mon/usb_mon.h b/drivers/usb/mon/usb_mon.h
index 33678c24ebee..50efc8ee7c21 100644
--- a/drivers/usb/mon/usb_mon.h
+++ b/drivers/usb/mon/usb_mon.h
@@ -53,7 +53,7 @@ extern char mon_dmapeek(unsigned char *dst, dma_addr_t dma_addr, int len);
 
 extern struct mutex mon_lock;
 
-extern struct file_operations mon_fops_text;
-extern struct file_operations mon_fops_stat;
+extern const struct file_operations mon_fops_text;
+extern const struct file_operations mon_fops_stat;
 
 #endif /* __USB_MON_H */
diff --git a/drivers/usb/usb-skeleton.c b/drivers/usb/usb-skeleton.c
index 33f0e81c58d3..3339373239ad 100644
--- a/drivers/usb/usb-skeleton.c
+++ b/drivers/usb/usb-skeleton.c
@@ -243,7 +243,7 @@ exit:
 	return retval;
 }
 
-static struct file_operations skel_fops = {
+static const struct file_operations skel_fops = {
 	.owner =	THIS_MODULE,
 	.read =		skel_read,
 	.write =	skel_write,
-- 
cgit v1.2.1


From 3d5b2510f6e361e2203e163c03b93d0026de5629 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Sun, 30 Jul 2006 18:43:43 +0200
Subject: USB: making the kernel -Wshadow clean - USB & completion

include/linux/usb.h causes a lot of -Wshadow warnings - fix them.

  include/linux/usb.h:901: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here
  include/linux/usb.h:932: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here
  include/linux/usb.h:967: warning: declaration of 'complete' shadows a global declaration
  include/linux/completion.h:52: warning: shadowed declaration is here


Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 include/linux/usb.h | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/include/linux/usb.h b/include/linux/usb.h
index e22f4b386605..3d5cfa731680 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -932,7 +932,7 @@ struct urb
  * @setup_packet: pointer to the setup_packet buffer
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  *
  * Initializes a control urb with the proper information needed to submit
@@ -944,7 +944,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
 					 unsigned char *setup_packet,
 					 void *transfer_buffer,
 					 int buffer_length,
-					 usb_complete_t complete,
+					 usb_complete_t complete_fn,
 					 void *context)
 {
 	spin_lock_init(&urb->lock);
@@ -953,7 +953,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
 	urb->setup_packet = setup_packet;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 }
 
@@ -964,7 +964,7 @@ static inline void usb_fill_control_urb (struct urb *urb,
  * @pipe: the endpoint pipe
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  *
  * Initializes a bulk urb with the proper information needed to submit it
@@ -975,7 +975,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
 				      unsigned int pipe,
 				      void *transfer_buffer,
 				      int buffer_length,
-				      usb_complete_t complete,
+				      usb_complete_t complete_fn,
 				      void *context)
 {
 	spin_lock_init(&urb->lock);
@@ -983,7 +983,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
 	urb->pipe = pipe;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 }
 
@@ -994,7 +994,7 @@ static inline void usb_fill_bulk_urb (struct urb *urb,
  * @pipe: the endpoint pipe
  * @transfer_buffer: pointer to the transfer buffer
  * @buffer_length: length of the transfer buffer
- * @complete: pointer to the usb_complete_t function
+ * @complete_fn: pointer to the usb_complete_t function
  * @context: what to set the urb context to.
  * @interval: what to set the urb interval to, encoded like
  *	the endpoint descriptor's bInterval value.
@@ -1010,7 +1010,7 @@ static inline void usb_fill_int_urb (struct urb *urb,
 				     unsigned int pipe,
 				     void *transfer_buffer,
 				     int buffer_length,
-				     usb_complete_t complete,
+				     usb_complete_t complete_fn,
 				     void *context,
 				     int interval)
 {
@@ -1019,7 +1019,7 @@ static inline void usb_fill_int_urb (struct urb *urb,
 	urb->pipe = pipe;
 	urb->transfer_buffer = transfer_buffer;
 	urb->transfer_buffer_length = buffer_length;
-	urb->complete = complete;
+	urb->complete = complete_fn;
 	urb->context = context;
 	if (dev->speed == USB_SPEED_HIGH)
 		urb->interval = 1 << (interval - 1);
-- 
cgit v1.2.1


From b7cfaaaf86571732c7728e95a2231a860385463c Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: New functions to check endpoints info.

These functions makes USB driver's code simpler when dealing with endpoints
by avoiding them from accessing the endpoint's descriptor structure directly
when they only need to know the endpoint's transfer type and/or
direction.

Please, read each functions' documentation in order to know how to use
them.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.c | 144 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h    |  14 +++++
 2 files changed, 158 insertions(+)

diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 9ebfc0fe819d..82837d45b484 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -482,6 +482,138 @@ int usb_get_current_frame_number(struct usb_device *dev)
 	return dev->bus->op->get_frame_number (dev);
 }
 
+/**
+ * usb_endpoint_dir_in - check if the endpoint has IN direction
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type IN, otherwise it returns false.
+ */
+int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN);
+}
+
+/**
+ * usb_endpoint_dir_out - check if the endpoint has OUT direction
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type OUT, otherwise it returns false.
+ */
+int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT);
+}
+
+/**
+ * usb_endpoint_xfer_bulk - check if the endpoint has bulk transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type bulk, otherwise it returns false.
+ */
+int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_BULK);
+}
+
+/**
+ * usb_endpoint_xfer_int - check if the endpoint has interrupt transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type interrupt, otherwise it returns
+ * false.
+ */
+int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_INT);
+}
+
+/**
+ * usb_endpoint_xfer_isoc - check if the endpoint has isochronous transfer type
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint is of type isochronous, otherwise it returns
+ * false.
+ */
+int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd)
+{
+	return ((epd->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) ==
+		USB_ENDPOINT_XFER_ISOC);
+}
+
+/**
+ * usb_endpoint_is_bulk_in - check if the endpoint is bulk IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has bulk transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_bulk_out - check if the endpoint is bulk OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has bulk transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_bulk(epd) && usb_endpoint_dir_out(epd));
+}
+
+/**
+ * usb_endpoint_is_int_in - check if the endpoint is interrupt IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has interrupt transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_int_out - check if the endpoint is interrupt OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has interrupt transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_int(epd) && usb_endpoint_dir_out(epd));
+}
+
+/**
+ * usb_endpoint_is_isoc_in - check if the endpoint is isochronous IN
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has isochronous transfer type and IN direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_in(epd));
+}
+
+/**
+ * usb_endpoint_is_isoc_out - check if the endpoint is isochronous OUT
+ * @epd: endpoint to be checked
+ *
+ * Returns true if the endpoint has isochronous transfer type and OUT direction,
+ * otherwise it returns false.
+ */
+int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd)
+{
+	return (usb_endpoint_xfer_isoc(epd) && usb_endpoint_dir_out(epd));
+}
+
 /*-------------------------------------------------------------------*/
 /*
  * __usb_get_extra_descriptor() finds a descriptor of specific type in the
@@ -909,6 +1041,18 @@ EXPORT_SYMBOL(__usb_get_extra_descriptor);
 EXPORT_SYMBOL(usb_find_device);
 EXPORT_SYMBOL(usb_get_current_frame_number);
 
+EXPORT_SYMBOL_GPL(usb_endpoint_dir_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_dir_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_bulk);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_int);
+EXPORT_SYMBOL_GPL(usb_endpoint_xfer_isoc);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_bulk_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_int_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_int_out);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_in);
+EXPORT_SYMBOL_GPL(usb_endpoint_is_isoc_out);
+
 EXPORT_SYMBOL (usb_buffer_alloc);
 EXPORT_SYMBOL (usb_buffer_free);
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 3d5cfa731680..f807479ef65b 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -467,6 +467,20 @@ static inline int usb_make_path (struct usb_device *dev, char *buf,
 
 /*-------------------------------------------------------------------------*/
 
+extern int usb_endpoint_dir_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_dir_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_bulk(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_int(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_xfer_isoc(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_bulk_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_bulk_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_int_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_int_out(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_isoc_in(const struct usb_endpoint_descriptor *epd);
+extern int usb_endpoint_is_isoc_out(const struct usb_endpoint_descriptor *epd);
+
+/*-------------------------------------------------------------------------*/
+
 #define USB_DEVICE_ID_MATCH_DEVICE \
 		(USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_PRODUCT)
 #define USB_DEVICE_ID_MATCH_DEV_RANGE \
-- 
cgit v1.2.1


From 5bc66d530b6c158795cb3fefd2106a09afb5e0f7 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: usblp: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/class/usblp.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c
index 292919d260a0..9b05a359b9f7 100644
--- a/drivers/usb/class/usblp.c
+++ b/drivers/usb/class/usblp.c
@@ -1021,18 +1021,13 @@ static int usblp_select_alts(struct usblp *usblp)
 		for (e = 0; e < ifd->desc.bNumEndpoints; e++) {
 			epd = &ifd->endpoint[e].desc;
 
-			if ((epd->bmAttributes&USB_ENDPOINT_XFERTYPE_MASK)!=
-			    USB_ENDPOINT_XFER_BULK)
-				continue;
-
-			if (!(epd->bEndpointAddress & USB_ENDPOINT_DIR_MASK)) {
+			if (usb_endpoint_is_bulk_out(epd))
 				if (!epwrite)
 					epwrite = epd;
 
-			} else {
+			if (usb_endpoint_is_bulk_in(epd))
 				if (!epread)
 					epread = epd;
-			}
 		}
 
 		/* Ignore buggy hardware without the right endpoints. */
-- 
cgit v1.2.1


From fbf81c29a3c05cd227cad89435d71c15e958feaf Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: hub: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 64e80b964b87..3924dd080bea 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -868,13 +868,8 @@ descriptor_error:
 
 	endpoint = &desc->endpoint[0].desc;
 
-	/* Output endpoint? Curiouser and curiouser.. */
-	if (!(endpoint->bEndpointAddress & USB_DIR_IN))
-		goto descriptor_error;
-
-	/* If it's not an interrupt endpoint, we'd better punt! */
-	if ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-			!= USB_ENDPOINT_XFER_INT)
+	/* If it's not an interrupt in endpoint, we'd better punt! */
+	if (!usb_endpoint_is_int_in(endpoint))
 		goto descriptor_error;
 
 	/* We found a hub */
-- 
cgit v1.2.1


From 97b107ca391c0411e6c5fc1e26585cd8285696f9 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: appletouch: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/appletouch.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/usb/input/appletouch.c b/drivers/usb/input/appletouch.c
index 044faa07e297..0aa9cc2bfd69 100644
--- a/drivers/usb/input/appletouch.c
+++ b/drivers/usb/input/appletouch.c
@@ -436,10 +436,7 @@ static int atp_probe(struct usb_interface *iface, const struct usb_device_id *id
 	iface_desc = iface->cur_altsetting;
 	for (i = 0; i < iface_desc->desc.bNumEndpoints; i++) {
 		endpoint = &iface_desc->endpoint[i].desc;
-		if (!int_in_endpointAddr &&
-		    (endpoint->bEndpointAddress & USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-					== USB_ENDPOINT_XFER_INT)) {
+		if (!int_in_endpointAddr && usb_endpoint_is_int_in(endpoint)) {
 			/* we found an interrupt in endpoint */
 			int_in_endpointAddr = endpoint->bEndpointAddress;
 			break;
-- 
cgit v1.2.1


From ee709a3c613d9c458323e04dd69c8eb233d45199 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: acecad: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/acecad.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/drivers/usb/input/acecad.c b/drivers/usb/input/acecad.c
index 18c10e150ef3..d83603ba40ae 100644
--- a/drivers/usb/input/acecad.c
+++ b/drivers/usb/input/acecad.c
@@ -141,10 +141,7 @@ static int usb_acecad_probe(struct usb_interface *intf, const struct usb_device_
 
 	endpoint = &interface->endpoint[0].desc;
 
-	if (!(endpoint->bEndpointAddress & 0x80))
-		return -ENODEV;
-
-	if ((endpoint->bmAttributes & 3) != 3)
+	if (!usb_endpoint_is_int_in(endpoint))
 		return -ENODEV;
 
 	pipe = usb_rcvintpipe(dev, endpoint->bEndpointAddress);
-- 
cgit v1.2.1


From 96642a2ce524f00f783edc95be8b2ce1429803df Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: ati_remote: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/ati_remote.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/input/ati_remote.c b/drivers/usb/input/ati_remote.c
index 3719fcb04b8f..3558d7ed99b9 100644
--- a/drivers/usb/input/ati_remote.c
+++ b/drivers/usb/input/ati_remote.c
@@ -732,12 +732,8 @@ static int ati_remote_probe(struct usb_interface *interface, const struct usb_de
 	endpoint_in = &iface_host->endpoint[0].desc;
 	endpoint_out = &iface_host->endpoint[1].desc;
 
-	if (!(endpoint_in->bEndpointAddress & USB_DIR_IN)) {
-		err("%s: Unexpected endpoint_in->bEndpointAddress\n", __FUNCTION__);
-		return -ENODEV;
-	}
-	if ((endpoint_in->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) != USB_ENDPOINT_XFER_INT) {
-		err("%s: Unexpected endpoint_in->bmAttributes\n", __FUNCTION__);
+	if (!usb_endpoint_is_int_in(endpoint_in)) {
+		err("%s: Unexpected endpoint_in\n", __FUNCTION__);
 		return -ENODEV;
 	}
 	if (le16_to_cpu(endpoint_in->wMaxPacketSize) == 0) {
-- 
cgit v1.2.1


From 96723199e00c4f27fecd8f43e1367c5f25b39a27 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: keyspan_remote: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/keyspan_remote.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/usb/input/keyspan_remote.c b/drivers/usb/input/keyspan_remote.c
index 4723b310f277..a90359551575 100644
--- a/drivers/usb/input/keyspan_remote.c
+++ b/drivers/usb/input/keyspan_remote.c
@@ -420,8 +420,7 @@ static struct usb_endpoint_descriptor *keyspan_get_in_endpoint(struct usb_host_i
 	for (i = 0; i < iface->desc.bNumEndpoints; ++i) {
 		endpoint = &iface->endpoint[i].desc;
 
-		if ((endpoint->bEndpointAddress & USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == USB_ENDPOINT_XFER_INT)) {
+		if (usb_endpoint_is_int_in(endpoint)) {
 			/* we found our interrupt in endpoint */
 			return endpoint;
 		}
-- 
cgit v1.2.1


From 60ca126c9589e3f4bcf81db9762dca82946cd2ba Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: powermate: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/powermate.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/usb/input/powermate.c b/drivers/usb/input/powermate.c
index b3c0d0c3eae9..f0f8db6810a2 100644
--- a/drivers/usb/input/powermate.c
+++ b/drivers/usb/input/powermate.c
@@ -313,9 +313,7 @@ static int powermate_probe(struct usb_interface *intf, const struct usb_device_i
 
 	interface = intf->cur_altsetting;
 	endpoint = &interface->endpoint[0].desc;
-	if (!(endpoint->bEndpointAddress & 0x80))
-		return -EIO;
-	if ((endpoint->bmAttributes & 3) != 3)
+	if (!usb_endpoint_is_int_in(endpoint))
 		return -EIO;
 
 	usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
-- 
cgit v1.2.1


From 4fa1bbf5cfac0e8c795e0ef7b2b939b45229ef2b Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:53 -0700
Subject: USB: usb-serial: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/usb-serial.c | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index e06a41bd0f3b..1f21a9d4b700 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -676,33 +676,29 @@ int usb_serial_probe(struct usb_interface *interface,
 	iface_desc = interface->cur_altsetting;
 	for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
 		endpoint = &iface_desc->endpoint[i].desc;
-		
-		if ((endpoint->bEndpointAddress & 0x80) &&
-		    ((endpoint->bmAttributes & 3) == 0x02)) {
+
+		if (usb_endpoint_is_bulk_in(endpoint)) {
 			/* we found a bulk in endpoint */
 			dbg("found bulk in on endpoint %d", i);
 			bulk_in_endpoint[num_bulk_in] = endpoint;
 			++num_bulk_in;
 		}
 
-		if (((endpoint->bEndpointAddress & 0x80) == 0x00) &&
-		    ((endpoint->bmAttributes & 3) == 0x02)) {
+		if (usb_endpoint_is_bulk_out(endpoint)) {
 			/* we found a bulk out endpoint */
 			dbg("found bulk out on endpoint %d", i);
 			bulk_out_endpoint[num_bulk_out] = endpoint;
 			++num_bulk_out;
 		}
-		
-		if ((endpoint->bEndpointAddress & 0x80) &&
-		    ((endpoint->bmAttributes & 3) == 0x03)) {
+
+		if (usb_endpoint_is_int_in(endpoint)) {
 			/* we found a interrupt in endpoint */
 			dbg("found interrupt in on endpoint %d", i);
 			interrupt_in_endpoint[num_interrupt_in] = endpoint;
 			++num_interrupt_in;
 		}
 
-		if (((endpoint->bEndpointAddress & 0x80) == 0x00) &&
-		    ((endpoint->bmAttributes & 3) == 0x03)) {
+		if (usb_endpoint_is_int_out(endpoint)) {
 			/* we found an interrupt out endpoint */
 			dbg("found interrupt out on endpoint %d", i);
 			interrupt_out_endpoint[num_interrupt_out] = endpoint;
@@ -722,8 +718,7 @@ int usb_serial_probe(struct usb_interface *interface,
 			iface_desc = dev->actconfig->interface[0]->cur_altsetting;
 			for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
 				endpoint = &iface_desc->endpoint[i].desc;
-				if ((endpoint->bEndpointAddress & 0x80) &&
-				    ((endpoint->bmAttributes & 3) == 0x03)) {
+				if (usb_endpoint_is_int_in(endpoint)) {
 					/* we found a interrupt in endpoint */
 					dbg("found interrupt in for Prolific device on separate interface");
 					interrupt_in_endpoint[num_interrupt_in] = endpoint;
-- 
cgit v1.2.1


From b0b660b8d5b446bf26a46b233adef5819d9a683c Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:54 -0700
Subject: USB: usblcd: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/usblcd.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/misc/usblcd.c b/drivers/usb/misc/usblcd.c
index 53e59e362330..dbaca9f1efad 100644
--- a/drivers/usb/misc/usblcd.c
+++ b/drivers/usb/misc/usblcd.c
@@ -290,9 +290,7 @@ static int lcd_probe(struct usb_interface *interface, const struct usb_device_id
 		endpoint = &iface_desc->endpoint[i].desc;
 
 		if (!dev->bulk_in_endpointAddr &&
-		    (endpoint->bEndpointAddress & USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-					== USB_ENDPOINT_XFER_BULK)) {
+		    usb_endpoint_is_bulk_in(endpoint)) {
 			/* we found a bulk in endpoint */
 			buffer_size = le16_to_cpu(endpoint->wMaxPacketSize);
 			dev->bulk_in_size = buffer_size;
@@ -305,9 +303,7 @@ static int lcd_probe(struct usb_interface *interface, const struct usb_device_id
 		}
 
 		if (!dev->bulk_out_endpointAddr &&
-		    !(endpoint->bEndpointAddress & USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-					== USB_ENDPOINT_XFER_BULK)) {
+		    usb_endpoint_is_bulk_out(endpoint)) {
 			/* we found a bulk out endpoint */
 			dev->bulk_out_endpointAddr = endpoint->bEndpointAddress;
 		}
-- 
cgit v1.2.1


From 5482687b8be4dedb8a5879f07c734ff11a88a7d5 Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Wed, 27 Sep 2006 11:58:54 -0700
Subject: USB: ldusb: Use usb_endpoint_* functions.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/ldusb.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/misc/ldusb.c b/drivers/usb/misc/ldusb.c
index 741736b6c134..10b640339d8d 100644
--- a/drivers/usb/misc/ldusb.c
+++ b/drivers/usb/misc/ldusb.c
@@ -657,15 +657,11 @@ static int ld_usb_probe(struct usb_interface *intf, const struct usb_device_id *
 	for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
 		endpoint = &iface_desc->endpoint[i].desc;
 
-		if (((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == USB_ENDPOINT_XFER_INT)) {
+		if (usb_endpoint_is_int_in(endpoint))
 			dev->interrupt_in_endpoint = endpoint;
-		}
 
-		if (((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK) == USB_DIR_OUT) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) == USB_ENDPOINT_XFER_INT)) {
+		if (usb_endpoint_is_int_out(endpoint))
 			dev->interrupt_out_endpoint = endpoint;
-		}
 	}
 	if (dev->interrupt_in_endpoint == NULL) {
 		dev_err(&intf->dev, "Interrupt in endpoint not found\n");
-- 
cgit v1.2.1


From 9bcbcf4d00cd2400e655a738e77f0d21b69c6771 Mon Sep 17 00:00:00 2001
From: dave rientjes <rientjes@google.com>
Date: Tue, 18 Jul 2006 23:23:02 -0700
Subject: USB: net1080 inherent pad length

The size of struct nc_trailer is inherently the newtailroom pad.


Signed-off-by: David Rientjes <rientjes@google.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/net/net1080.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/usb/net/net1080.c b/drivers/usb/net/net1080.c
index a9b6eeac3e3f..301baa72bac7 100644
--- a/drivers/usb/net/net1080.c
+++ b/drivers/usb/net/net1080.c
@@ -498,25 +498,24 @@ static int net1080_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 static struct sk_buff *
 net1080_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags)
 {
-	int			padlen;
 	struct sk_buff		*skb2;
 	struct nc_header	*header = NULL;
 	struct nc_trailer	*trailer = NULL;
+	int			padlen = sizeof (struct nc_trailer);
 	int			len = skb->len;
 
-	padlen = ((len + sizeof (struct nc_header)
-			+ sizeof (struct nc_trailer)) & 0x01) ? 0 : 1;
+	if (!((len + padlen + sizeof (struct nc_header)) & 0x01))
+		padlen++;
 	if (!skb_cloned(skb)) {
 		int	headroom = skb_headroom(skb);
 		int	tailroom = skb_tailroom(skb);
 
-		if ((padlen + sizeof (struct nc_trailer)) <= tailroom
-			    && sizeof (struct nc_header) <= headroom)
+		if (padlen <= tailroom &&
+		    sizeof(struct nc_header) <= headroom)
 			/* There's enough head and tail room */
 			goto encapsulate;
 
-		if ((sizeof (struct nc_header) + padlen
-					+ sizeof (struct nc_trailer)) <
+		if ((sizeof (struct nc_header) + padlen) <
 				(headroom + tailroom)) {
 			/* There's enough total room, so just readjust */
 			skb->data = memmove(skb->head
@@ -530,7 +529,7 @@ net1080_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags)
 	/* Create a new skb to use with the correct size */
 	skb2 = skb_copy_expand(skb,
 				sizeof (struct nc_header),
-				sizeof (struct nc_trailer) + padlen,
+				padlen,
 				flags);
 	dev_kfree_skb_any(skb);
 	if (!skb2)
-- 
cgit v1.2.1


From e22fc27c87b41bda1b0daf8436224b0f79853482 Mon Sep 17 00:00:00 2001
From: Milan Svoboda <msvoboda@ra.rockwell.com>
Date: Tue, 8 Aug 2006 22:23:12 -0700
Subject: USB: add poll to gadgetfs's endpoint zero

Add poll() support to gadgetfs ep0

Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 32 +++++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 1072e987ff21..ed9b404e5f5a 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -32,6 +32,7 @@
 #include <linux/compiler.h>
 #include <asm/uaccess.h>
 #include <linux/slab.h>
+#include <linux/poll.h>
 
 #include <linux/device.h>
 #include <linux/moduleparam.h>
@@ -1235,6 +1236,35 @@ dev_release (struct inode *inode, struct file *fd)
 	return 0;
 }
 
+static unsigned int
+ep0_poll (struct file *fd, poll_table *wait)
+{
+       struct dev_data         *dev = fd->private_data;
+       int                     mask = 0;
+
+       poll_wait(fd, &dev->wait, wait);
+
+       spin_lock_irq (&dev->lock);
+
+       /* report fd mode change before acting on it */
+       if (dev->setup_abort) {
+               dev->setup_abort = 0;
+               mask = POLLHUP;
+               goto out;
+       }
+
+       if (dev->state == STATE_SETUP) {
+               if (dev->setup_in || dev->setup_can_stall)
+                       mask = POLLOUT;
+       } else {
+               if (dev->ev_next != 0)
+                       mask = POLLIN;
+       }
+out:
+       spin_unlock_irq(&dev->lock);
+       return mask;
+}
+
 static int dev_ioctl (struct inode *inode, struct file *fd,
 		unsigned code, unsigned long value)
 {
@@ -1254,7 +1284,7 @@ static const struct file_operations ep0_io_operations = {
 	.read =		ep0_read,
 	.write =	ep0_write,
 	.fasync =	ep0_fasync,
-	// .poll =	ep0_poll,
+	.poll =		ep0_poll,
 	.ioctl =	dev_ioctl,
 	.release =	dev_release,
 };
-- 
cgit v1.2.1


From a94da8971e836f32315f8832b0bf3e88bee9efae Mon Sep 17 00:00:00 2001
From: Milan Svoboda <msvoboda@ra.rockwell.com>
Date: Tue, 8 Aug 2006 22:14:43 -0700
Subject: USB gadget: gadgetfs dont try to lock before free

I spotted this during my tests with -rt on arm. The -rt patch contains
some better tools
to diagnose problems with locks and some other things...

Original code tries to take semaphore in BUG_ON and then free the memory
with this semaphore.


Signed-off-by: Milan Svoboda <msvoboda@ra.rockwell.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index ed9b404e5f5a..a38b6af2c358 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -223,7 +223,6 @@ static void put_ep (struct ep_data *data)
 	/* needs no more cleanup */
 	BUG_ON (!list_empty (&data->epfiles));
 	BUG_ON (waitqueue_active (&data->wait));
-	BUG_ON (down_trylock (&data->lock) != 0);
 	kfree (data);
 }
 
-- 
cgit v1.2.1


From 64a21d025d3a979a8715f2ec7acabca7b5406c8a Mon Sep 17 00:00:00 2001
From: Aleksey Gorelov <dared1st@yahoo.com>
Date: Tue, 8 Aug 2006 17:24:08 -0700
Subject: USB: Properly unregister reboot notifier in case of failure in ehci
 hcd

If some problem occurs during ehci startup, for instance, request_irq fails,
echi hcd driver tries it best to cleanup, but fails to unregister reboot
notifier, which in turn leads to crash on reboot/poweroff.

The following patch resolves this problem by not using reboot notifiers
anymore, but instead making ehci/ohci driver get its own shutdown method.  For
PCI, it is done through pci glue, for everything else through platform driver
glue.

One downside: sa1111 does not use platform driver stuff, and does not have its
own shutdown hook, so no 'shutdown' is called for it now.  I'm not sure if it
is really necessary on that platform, though.

Signed-off-by: Aleks Gorelov <dared1st@yahoo.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd-pci.c      | 16 ++++++++++++++++
 drivers/usb/core/hcd.c          | 11 +++++++++++
 drivers/usb/core/hcd.h          |  8 ++++++++
 drivers/usb/host/ehci-au1xxx.c  |  2 ++
 drivers/usb/host/ehci-fsl.c     |  2 ++
 drivers/usb/host/ehci-hcd.c     | 16 +++++-----------
 drivers/usb/host/ehci-pci.c     |  2 ++
 drivers/usb/host/ehci.h         |  1 -
 drivers/usb/host/ohci-at91.c    |  2 ++
 drivers/usb/host/ohci-au1xxx.c  |  2 ++
 drivers/usb/host/ohci-ep93xx.c  |  2 ++
 drivers/usb/host/ohci-hcd.c     | 12 ++++--------
 drivers/usb/host/ohci-lh7a404.c |  2 ++
 drivers/usb/host/ohci-mem.c     |  1 -
 drivers/usb/host/ohci-omap.c    |  2 ++
 drivers/usb/host/ohci-pci.c     |  3 +++
 drivers/usb/host/ohci-ppc-soc.c |  2 ++
 drivers/usb/host/ohci-pxa27x.c  |  2 ++
 drivers/usb/host/ohci-s3c2410.c |  2 ++
 drivers/usb/host/ohci.h         |  2 --
 20 files changed, 69 insertions(+), 23 deletions(-)

diff --git a/drivers/usb/core/hcd-pci.c b/drivers/usb/core/hcd-pci.c
index fa36391fedd3..edf4300a3f7a 100644
--- a/drivers/usb/core/hcd-pci.c
+++ b/drivers/usb/core/hcd-pci.c
@@ -413,4 +413,20 @@ EXPORT_SYMBOL (usb_hcd_pci_resume);
 
 #endif	/* CONFIG_PM */
 
+/**
+ * usb_hcd_pci_shutdown - shutdown host controller
+ * @dev: USB Host Controller being shutdown
+ */
+void usb_hcd_pci_shutdown (struct pci_dev *dev)
+{
+	struct usb_hcd		*hcd;
+
+	hcd = pci_get_drvdata(dev);
+	if (!hcd)
+		return;
+
+	if (hcd->driver->shutdown)
+		hcd->driver->shutdown(hcd);
+}
+EXPORT_SYMBOL (usb_hcd_pci_shutdown);
 
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index fb4d058bbde0..dc9628c58933 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -36,6 +36,7 @@
 #include <linux/mutex.h>
 #include <asm/irq.h>
 #include <asm/byteorder.h>
+#include <linux/platform_device.h>
 
 #include <linux/usb.h>
 
@@ -1915,6 +1916,16 @@ void usb_remove_hcd(struct usb_hcd *hcd)
 }
 EXPORT_SYMBOL (usb_remove_hcd);
 
+void
+usb_hcd_platform_shutdown(struct platform_device* dev)
+{
+	struct usb_hcd *hcd = platform_get_drvdata(dev);
+
+	if (hcd->driver->shutdown)
+		hcd->driver->shutdown(hcd);
+}
+EXPORT_SYMBOL (usb_hcd_platform_shutdown);
+
 /*-------------------------------------------------------------------------*/
 
 #if defined(CONFIG_USB_MON)
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 7022aafb2ae8..58c7767bc904 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -192,6 +192,9 @@ struct hc_driver {
 	/* cleanly make HCD stop writing memory and doing I/O */
 	void	(*stop) (struct usb_hcd *hcd);
 
+	/* shutdown HCD */
+	void	(*shutdown) (struct usb_hcd *hcd);
+
 	/* return current frame number */
 	int	(*get_frame_number) (struct usb_hcd *hcd);
 
@@ -227,6 +230,9 @@ extern int usb_add_hcd(struct usb_hcd *hcd,
 		unsigned int irqnum, unsigned long irqflags);
 extern void usb_remove_hcd(struct usb_hcd *hcd);
 
+struct platform_device;
+extern void usb_hcd_platform_shutdown(struct platform_device* dev);
+
 #ifdef CONFIG_PCI
 struct pci_dev;
 struct pci_device_id;
@@ -239,6 +245,8 @@ extern int usb_hcd_pci_suspend (struct pci_dev *dev, pm_message_t state);
 extern int usb_hcd_pci_resume (struct pci_dev *dev);
 #endif /* CONFIG_PM */
 
+extern void usb_hcd_pci_shutdown (struct pci_dev *dev);
+
 #endif /* CONFIG_PCI */
 
 /* pci-ish (pdev null is ok) buffer alloc/mapping support */
diff --git a/drivers/usb/host/ehci-au1xxx.c b/drivers/usb/host/ehci-au1xxx.c
index 26ed757d22a6..5d1b12aad776 100644
--- a/drivers/usb/host/ehci-au1xxx.c
+++ b/drivers/usb/host/ehci-au1xxx.c
@@ -200,6 +200,7 @@ static const struct hc_driver ehci_au1xxx_hc_driver = {
 	.reset = ehci_init,
 	.start = ehci_run,
 	.stop = ehci_stop,
+	.shutdown = ehci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -268,6 +269,7 @@ MODULE_ALIAS("au1xxx-ehci");
 static struct platform_driver ehci_hcd_au1xxx_driver = {
 	.probe = ehci_hcd_au1xxx_drv_probe,
 	.remove = ehci_hcd_au1xxx_drv_remove,
+	.shutdown = usb_hcd_platform_shutdown,
 	/*.suspend      = ehci_hcd_au1xxx_drv_suspend, */
 	/*.resume       = ehci_hcd_au1xxx_drv_resume, */
 	.driver = {
diff --git a/drivers/usb/host/ehci-fsl.c b/drivers/usb/host/ehci-fsl.c
index d030516edfb9..1a915e982c1c 100644
--- a/drivers/usb/host/ehci-fsl.c
+++ b/drivers/usb/host/ehci-fsl.c
@@ -285,6 +285,7 @@ static const struct hc_driver ehci_fsl_hc_driver = {
 	.resume = ehci_bus_resume,
 #endif
 	.stop = ehci_stop,
+	.shutdown = ehci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -329,6 +330,7 @@ MODULE_ALIAS("fsl-ehci");
 static struct platform_driver ehci_fsl_driver = {
 	.probe = ehci_fsl_drv_probe,
 	.remove = ehci_fsl_drv_remove,
+	.shutdown = usb_hcd_platform_shutdown,
 	.driver = {
 		   .name = "fsl-ehci",
 		   },
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index d63177a8eaea..1c54b303e5fc 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -292,21 +292,20 @@ static void ehci_watchdog (unsigned long param)
 	spin_unlock_irqrestore (&ehci->lock, flags);
 }
 
-/* Reboot notifiers kick in for silicon on any bus (not just pci, etc).
+/* ehci_shutdown kick in for silicon on any bus (not just pci, etc).
  * This forcibly disables dma and IRQs, helping kexec and other cases
  * where the next system software may expect clean state.
  */
-static int
-ehci_reboot (struct notifier_block *self, unsigned long code, void *null)
+static void
+ehci_shutdown (struct usb_hcd *hcd)
 {
-	struct ehci_hcd		*ehci;
+	struct ehci_hcd	*ehci;
 
-	ehci = container_of (self, struct ehci_hcd, reboot_notifier);
+	ehci = hcd_to_ehci (hcd);
 	(void) ehci_halt (ehci);
 
 	/* make BIOS/etc use companion controller during reboot */
 	writel (0, &ehci->regs->configured_flag);
-	return 0;
 }
 
 static void ehci_port_power (struct ehci_hcd *ehci, int is_on)
@@ -381,7 +380,6 @@ static void ehci_stop (struct usb_hcd *hcd)
 
 	/* let companion controllers work when we aren't */
 	writel (0, &ehci->regs->configured_flag);
-	unregister_reboot_notifier (&ehci->reboot_notifier);
 
 	remove_debug_files (ehci);
 
@@ -483,9 +481,6 @@ static int ehci_init(struct usb_hcd *hcd)
 	}
 	ehci->command = temp;
 
-	ehci->reboot_notifier.notifier_call = ehci_reboot;
-	register_reboot_notifier(&ehci->reboot_notifier);
-
 	return 0;
 }
 
@@ -499,7 +494,6 @@ static int ehci_run (struct usb_hcd *hcd)
 
 	/* EHCI spec section 4.1 */
 	if ((retval = ehci_reset(ehci)) != 0) {
-		unregister_reboot_notifier(&ehci->reboot_notifier);
 		ehci_mem_cleanup(ehci);
 		return retval;
 	}
diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c
index 6967ab71e282..e6a3bcddd55b 100644
--- a/drivers/usb/host/ehci-pci.c
+++ b/drivers/usb/host/ehci-pci.c
@@ -338,6 +338,7 @@ static const struct hc_driver ehci_pci_hc_driver = {
 	.resume =		ehci_pci_resume,
 #endif
 	.stop =			ehci_stop,
+	.shutdown =		ehci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -384,4 +385,5 @@ static struct pci_driver ehci_pci_driver = {
 	.suspend =	usb_hcd_pci_suspend,
 	.resume =	usb_hcd_pci_resume,
 #endif
+	.shutdown = 	usb_hcd_pci_shutdown,
 };
diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
index 679c1cdcc915..1385ce2b3f0a 100644
--- a/drivers/usb/host/ehci.h
+++ b/drivers/usb/host/ehci.h
@@ -82,7 +82,6 @@ struct ehci_hcd {			/* one per controller */
 	struct dma_pool		*sitd_pool;	/* sitd per split iso urb */
 
 	struct timer_list	watchdog;
-	struct notifier_block	reboot_notifier;
 	unsigned long		actions;
 	unsigned		stamp;
 	unsigned long		next_statechange;
diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
index 33b75087bc0c..5a5bdf374d76 100644
--- a/drivers/usb/host/ohci-at91.c
+++ b/drivers/usb/host/ohci-at91.c
@@ -221,6 +221,7 @@ static const struct hc_driver ohci_at91_hc_driver = {
 	 */
 	.start =		ohci_at91_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -310,6 +311,7 @@ MODULE_ALIAS("at91_ohci");
 static struct platform_driver ohci_hcd_at91_driver = {
 	.probe		= ohci_hcd_at91_drv_probe,
 	.remove		= ohci_hcd_at91_drv_remove,
+	.shutdown	= usb_hcd_platform_shutdown,
 	.suspend	= ohci_hcd_at91_drv_suspend,
 	.resume		= ohci_hcd_at91_drv_resume,
 	.driver		= {
diff --git a/drivers/usb/host/ohci-au1xxx.c b/drivers/usb/host/ohci-au1xxx.c
index 44ed3a4c01ef..24e23c5783d8 100644
--- a/drivers/usb/host/ohci-au1xxx.c
+++ b/drivers/usb/host/ohci-au1xxx.c
@@ -269,6 +269,7 @@ static const struct hc_driver ohci_au1xxx_hc_driver = {
 	 */
 	.start =		ohci_au1xxx_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -335,6 +336,7 @@ static int ohci_hcd_au1xxx_drv_resume(struct platform_device *dev)
 static struct platform_driver ohci_hcd_au1xxx_driver = {
 	.probe		= ohci_hcd_au1xxx_drv_probe,
 	.remove		= ohci_hcd_au1xxx_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 	/*.suspend	= ohci_hcd_au1xxx_drv_suspend, */
 	/*.resume	= ohci_hcd_au1xxx_drv_resume, */
 	.driver		= {
diff --git a/drivers/usb/host/ohci-ep93xx.c b/drivers/usb/host/ohci-ep93xx.c
index 1a1d320b7995..1bf5e7a4e735 100644
--- a/drivers/usb/host/ohci-ep93xx.c
+++ b/drivers/usb/host/ohci-ep93xx.c
@@ -128,6 +128,7 @@ static struct hc_driver ohci_ep93xx_hc_driver = {
 	.flags			= HCD_USB11 | HCD_MEMORY,
 	.start			= ohci_ep93xx_start,
 	.stop			= ohci_stop,
+	.shutdown		= ohci_shutdown,
 	.urb_enqueue		= ohci_urb_enqueue,
 	.urb_dequeue		= ohci_urb_dequeue,
 	.endpoint_disable	= ohci_endpoint_disable,
@@ -203,6 +204,7 @@ static int ohci_hcd_ep93xx_drv_resume(struct platform_device *pdev)
 static struct platform_driver ohci_hcd_ep93xx_driver = {
 	.probe		= ohci_hcd_ep93xx_drv_probe,
 	.remove		= ohci_hcd_ep93xx_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 #ifdef CONFIG_PM
 	.suspend	= ohci_hcd_ep93xx_drv_suspend,
 	.resume		= ohci_hcd_ep93xx_drv_resume,
diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 7c3d8c60a60f..2c614af8f733 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -136,7 +136,6 @@ static const char	hcd_name [] = "ohci_hcd";
 static void ohci_dump (struct ohci_hcd *ohci, int verbose);
 static int ohci_init (struct ohci_hcd *ohci);
 static void ohci_stop (struct usb_hcd *hcd);
-static int ohci_reboot (struct notifier_block *, unsigned long , void *);
 
 #include "ohci-hub.c"
 #include "ohci-dbg.c"
@@ -419,21 +418,20 @@ static void ohci_usb_reset (struct ohci_hcd *ohci)
 	ohci_writel (ohci, ohci->hc_control, &ohci->regs->control);
 }
 
-/* reboot notifier forcibly disables IRQs and DMA, helping kexec and
+/* ohci_shutdown forcibly disables IRQs and DMA, helping kexec and
  * other cases where the next software may expect clean state from the
  * "firmware".  this is bus-neutral, unlike shutdown() methods.
  */
-static int
-ohci_reboot (struct notifier_block *block, unsigned long code, void *null)
+static void
+ohci_shutdown (struct usb_hcd *hcd)
 {
 	struct ohci_hcd *ohci;
 
-	ohci = container_of (block, struct ohci_hcd, reboot_notifier);
+	ohci = hcd_to_ohci (hcd);
 	ohci_writel (ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
 	ohci_usb_reset (ohci);
 	/* flush the writes */
 	(void) ohci_readl (ohci, &ohci->regs->control);
-	return 0;
 }
 
 /*-------------------------------------------------------------------------*
@@ -504,7 +502,6 @@ static int ohci_init (struct ohci_hcd *ohci)
 	if ((ret = ohci_mem_init (ohci)) < 0)
 		ohci_stop (hcd);
 	else {
-		register_reboot_notifier (&ohci->reboot_notifier);
 		create_debug_files (ohci);
 	}
 
@@ -800,7 +797,6 @@ static void ohci_stop (struct usb_hcd *hcd)
 	ohci_writel (ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
 	
 	remove_debug_files (ohci);
-	unregister_reboot_notifier (&ohci->reboot_notifier);
 	ohci_mem_cleanup (ohci);
 	if (ohci->hcca) {
 		dma_free_coherent (hcd->self.controller, 
diff --git a/drivers/usb/host/ohci-lh7a404.c b/drivers/usb/host/ohci-lh7a404.c
index f2c9161d6d6a..e121d97ed91c 100644
--- a/drivers/usb/host/ohci-lh7a404.c
+++ b/drivers/usb/host/ohci-lh7a404.c
@@ -174,6 +174,7 @@ static const struct hc_driver ohci_lh7a404_hc_driver = {
 	 */
 	.start =		ohci_lh7a404_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -241,6 +242,7 @@ static int ohci_hcd_lh7a404_drv_resume(struct platform_device *dev)
 static struct platform_driver ohci_hcd_lh7a404_driver = {
 	.probe		= ohci_hcd_lh7a404_drv_probe,
 	.remove		= ohci_hcd_lh7a404_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 	/*.suspend	= ohci_hcd_lh7a404_drv_suspend, */
 	/*.resume	= ohci_hcd_lh7a404_drv_resume, */
 	.driver		= {
diff --git a/drivers/usb/host/ohci-mem.c b/drivers/usb/host/ohci-mem.c
index bfbe328a4788..d976614eebd3 100644
--- a/drivers/usb/host/ohci-mem.c
+++ b/drivers/usb/host/ohci-mem.c
@@ -28,7 +28,6 @@ static void ohci_hcd_init (struct ohci_hcd *ohci)
 	ohci->next_statechange = jiffies;
 	spin_lock_init (&ohci->lock);
 	INIT_LIST_HEAD (&ohci->pending);
-	ohci->reboot_notifier.notifier_call = ohci_reboot;
 }
 
 /*-------------------------------------------------------------------------*/
diff --git a/drivers/usb/host/ohci-omap.c b/drivers/usb/host/ohci-omap.c
index 160cd4c58a03..9c02177de50a 100644
--- a/drivers/usb/host/ohci-omap.c
+++ b/drivers/usb/host/ohci-omap.c
@@ -447,6 +447,7 @@ static const struct hc_driver ohci_omap_hc_driver = {
 	.reset =		ohci_omap_init,
 	.start =		ohci_omap_start,
 	.stop =			ohci_omap_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -532,6 +533,7 @@ static int ohci_omap_resume(struct platform_device *dev)
 static struct platform_driver ohci_hcd_omap_driver = {
 	.probe		= ohci_hcd_omap_drv_probe,
 	.remove		= ohci_hcd_omap_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 #ifdef	CONFIG_PM
 	.suspend	= ohci_omap_suspend,
 	.resume		= ohci_omap_resume,
diff --git a/drivers/usb/host/ohci-pci.c b/drivers/usb/host/ohci-pci.c
index ef874443aa9f..3732db7d68eb 100644
--- a/drivers/usb/host/ohci-pci.c
+++ b/drivers/usb/host/ohci-pci.c
@@ -177,6 +177,7 @@ static const struct hc_driver ohci_pci_hc_driver = {
 	.reset =		ohci_pci_reset,
 	.start =		ohci_pci_start,
 	.stop =			ohci_stop,
+	.shutdown =		ohci_shutdown,
 
 #ifdef	CONFIG_PM
 	/* these suspend/resume entries are for upstream PCI glue ONLY */
@@ -232,6 +233,8 @@ static struct pci_driver ohci_pci_driver = {
 	.suspend =	usb_hcd_pci_suspend,
 	.resume =	usb_hcd_pci_resume,
 #endif
+
+	.shutdown =	usb_hcd_pci_shutdown,
 };
 
  
diff --git a/drivers/usb/host/ohci-ppc-soc.c b/drivers/usb/host/ohci-ppc-soc.c
index 270aaaad8c6d..d9d1ae236bd5 100644
--- a/drivers/usb/host/ohci-ppc-soc.c
+++ b/drivers/usb/host/ohci-ppc-soc.c
@@ -148,6 +148,7 @@ static const struct hc_driver ohci_ppc_soc_hc_driver = {
 	 */
 	.start =		ohci_ppc_soc_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -196,6 +197,7 @@ static int ohci_hcd_ppc_soc_drv_remove(struct platform_device *pdev)
 static struct platform_driver ohci_hcd_ppc_soc_driver = {
 	.probe		= ohci_hcd_ppc_soc_drv_probe,
 	.remove		= ohci_hcd_ppc_soc_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 #ifdef	CONFIG_PM
 	/*.suspend	= ohci_hcd_ppc_soc_drv_suspend,*/
 	/*.resume	= ohci_hcd_ppc_soc_drv_resume,*/
diff --git a/drivers/usb/host/ohci-pxa27x.c b/drivers/usb/host/ohci-pxa27x.c
index 2752d36c2a78..e176b04d7aeb 100644
--- a/drivers/usb/host/ohci-pxa27x.c
+++ b/drivers/usb/host/ohci-pxa27x.c
@@ -270,6 +270,7 @@ static const struct hc_driver ohci_pxa27x_hc_driver = {
 	 */
 	.start =		ohci_pxa27x_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -358,6 +359,7 @@ static int ohci_hcd_pxa27x_drv_resume(struct platform_device *pdev)
 static struct platform_driver ohci_hcd_pxa27x_driver = {
 	.probe		= ohci_hcd_pxa27x_drv_probe,
 	.remove		= ohci_hcd_pxa27x_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 #ifdef CONFIG_PM
 	.suspend	= ohci_hcd_pxa27x_drv_suspend, 
 	.resume		= ohci_hcd_pxa27x_drv_resume,
diff --git a/drivers/usb/host/ohci-s3c2410.c b/drivers/usb/host/ohci-s3c2410.c
index cd37eddf7d42..59e436424d41 100644
--- a/drivers/usb/host/ohci-s3c2410.c
+++ b/drivers/usb/host/ohci-s3c2410.c
@@ -447,6 +447,7 @@ static const struct hc_driver ohci_s3c2410_hc_driver = {
 	 */
 	.start =		ohci_s3c2410_start,
 	.stop =			ohci_stop,
+	.shutdown = 		ohci_shutdown,
 
 	/*
 	 * managing i/o requests and associated device resources
@@ -491,6 +492,7 @@ static int ohci_hcd_s3c2410_drv_remove(struct platform_device *pdev)
 static struct platform_driver ohci_hcd_s3c2410_driver = {
 	.probe		= ohci_hcd_s3c2410_drv_probe,
 	.remove		= ohci_hcd_s3c2410_drv_remove,
+	.shutdown 	= usb_hcd_platform_shutdown,
 	/*.suspend	= ohci_hcd_s3c2410_drv_suspend, */
 	/*.resume	= ohci_hcd_s3c2410_drv_resume, */
 	.driver		= {
diff --git a/drivers/usb/host/ohci.h b/drivers/usb/host/ohci.h
index caacf14371f5..650d1bf21c1d 100644
--- a/drivers/usb/host/ohci.h
+++ b/drivers/usb/host/ohci.h
@@ -389,8 +389,6 @@ struct ohci_hcd {
 	unsigned long		next_statechange;	/* suspend/resume */
 	u32			fminterval;		/* saved register */
 
-	struct notifier_block	reboot_notifier;
-
 	unsigned long		flags;		/* for HC bugs */
 #define	OHCI_QUIRK_AMD756	0x01			/* erratum #4 */
 #define	OHCI_QUIRK_SUPERIO	0x02			/* natsemi */
-- 
cgit v1.2.1


From de06a3b842b31b31220637c869f112cfbc1a5ef6 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Fri, 11 Aug 2006 11:33:58 -0400
Subject: UHCI: increase Resume-Detect-off delay

The UHCI controller in my laptop takes longer to turn off the
Resume-Detect bit than the 4 us allowed by uhci-hcd.  Presumably other
computers will have the same problem.

This patch (as752) increases the maximum delay to 10 us, which should be
plenty, and uses polling to avoid penalizing systems which can turn the
bit off more quickly.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/uhci-hub.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/host/uhci-hub.c b/drivers/usb/host/uhci-hub.c
index c545ef92fe29..16fb72eb6fc9 100644
--- a/drivers/usb/host/uhci-hub.c
+++ b/drivers/usb/host/uhci-hub.c
@@ -84,6 +84,7 @@ static void uhci_finish_suspend(struct uhci_hcd *uhci, int port,
 		unsigned long port_addr)
 {
 	int status;
+	int i;
 
 	if (inw(port_addr) & (USBPORTSC_SUSP | USBPORTSC_RD)) {
 		CLR_RH_PORTSTAT(USBPORTSC_SUSP | USBPORTSC_RD);
@@ -92,9 +93,14 @@ static void uhci_finish_suspend(struct uhci_hcd *uhci, int port,
 
 		/* The controller won't actually turn off the RD bit until
 		 * it has had a chance to send a low-speed EOP sequence,
-		 * which takes 3 bit times (= 2 microseconds).  We'll delay
-		 * slightly longer for good luck. */
-		udelay(4);
+		 * which is supposed to take 3 bit times (= 2 microseconds).
+		 * Experiments show that some controllers take longer, so
+		 * we'll poll for completion. */
+		for (i = 0; i < 10; ++i) {
+			if (!(inw(port_addr) & USBPORTSC_RD))
+				break;
+			udelay(1);
+		}
 	}
 	clear_bit(port, &uhci->resuming_ports);
 }
-- 
cgit v1.2.1


From 455b25fb209c8241e2163b491228b28667d82c1c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Fri, 11 Aug 2006 16:01:45 -0400
Subject: usbcore: make hcd_endpoint_disable wait for queue to drain

The inconsistent lock state problem in usbcore (the one that shows up
when an HCD is unloaded) comes down to two inter-related problems:

	usb_rh_urb_dequeue() isn't set up to be called with interrupts
	disabled.

	hcd_endpoint_disable() doesn't wait for all URBs on the
	endpoint's queue to complete.

The two problems are related because the one type of URB that isn't
likely to be complete when hcd_endpoint_disable() returns is a root-hub
URB.  Right now usb_rh_urb_dequeue() waits for them to complete, and it
assumes interrupts are enabled so it can wait.  But
hcd_endpoint_disable() calls it with interrupts disabled.

Now, it should be legal to unlink root-hub URBs with interrupts
disabled.  The solution is to move the waiting into
hcd_endpoint_disable(), where it belongs.  This patch (as754) does that.

It turns out to be completely safe to replace the del_timer_sync() with
a simple del_timer().  It doesn't matter if the timer routine is
running; hcd_root_hub_lock will synchronize the two threads and the
status URB will complete with an unlink error, as it should.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c | 65 +++++++++++++++++++++++++++-----------------------
 1 file changed, 35 insertions(+), 30 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index dc9628c58933..ea20a3a5a9b9 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -633,31 +633,20 @@ static int rh_urb_enqueue (struct usb_hcd *hcd, struct urb *urb)
 
 /*-------------------------------------------------------------------------*/
 
-/* Asynchronous unlinks of root-hub control URBs are legal, but they
- * don't do anything.  Status URB unlinks must be made in process context
- * with interrupts enabled.
+/* Unlinks of root-hub control URBs are legal, but they don't do anything
+ * since these URBs always execute synchronously.
  */
 static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 {
-	if (usb_pipeendpoint(urb->pipe) == 0) {	/* Control URB */
-		if (in_interrupt())
-			return 0;		/* nothing to do */
-
-		spin_lock_irq(&urb->lock);	/* from usb_kill_urb */
-		++urb->reject;
-		spin_unlock_irq(&urb->lock);
-
-		wait_event(usb_kill_urb_queue,
-				atomic_read(&urb->use_count) == 0);
+	unsigned long	flags;
 
-		spin_lock_irq(&urb->lock);
-		--urb->reject;
-		spin_unlock_irq(&urb->lock);
+	if (usb_pipeendpoint(urb->pipe) == 0) {	/* Control URB */
+		;	/* Do nothing */
 
 	} else {				/* Status URB */
 		if (!hcd->uses_new_polling)
-			del_timer_sync (&hcd->rh_timer);
-		local_irq_disable ();
+			del_timer (&hcd->rh_timer);
+		local_irq_save (flags);
 		spin_lock (&hcd_root_hub_lock);
 		if (urb == hcd->status_urb) {
 			hcd->status_urb = NULL;
@@ -667,7 +656,7 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 		spin_unlock (&hcd_root_hub_lock);
 		if (urb)
 			usb_hcd_giveback_urb (hcd, urb, NULL);
-		local_irq_enable ();
+		local_irq_restore (flags);
 	}
 
 	return 0;
@@ -1355,7 +1344,8 @@ done:
 /*-------------------------------------------------------------------------*/
 
 /* disables the endpoint: cancels any pending urbs, then synchronizes with
- * the hcd to make sure all endpoint state is gone from hardware. use for
+ * the hcd to make sure all endpoint state is gone from hardware, and then
+ * waits until the endpoint's queue is completely drained. use for
  * set_configuration, set_interface, driver removal, physical disconnect.
  *
  * example:  a qh stored in ep->hcpriv, holding state related to endpoint
@@ -1374,22 +1364,13 @@ hcd_endpoint_disable (struct usb_device *udev, struct usb_host_endpoint *ep)
 
 	local_irq_disable ();
 
-	/* FIXME move most of this into message.c as part of its
-	 * endpoint disable logic
-	 */
-
 	/* ep is already gone from udev->ep_{in,out}[]; no more submits */
 rescan:
 	spin_lock (&hcd_data_lock);
 	list_for_each_entry (urb, &ep->urb_list, urb_list) {
 		int	tmp;
 
-		/* another cpu may be in hcd, spinning on hcd_data_lock
-		 * to giveback() this urb.  the races here should be
-		 * small, but a full fix needs a new "can't submit"
-		 * urb state.
-		 * FIXME urb->reject should allow that...
-		 */
+		/* the urb may already have been unlinked */
 		if (urb->status != -EINPROGRESS)
 			continue;
 		usb_get_urb (urb);
@@ -1431,6 +1412,30 @@ rescan:
 	might_sleep ();
 	if (hcd->driver->endpoint_disable)
 		hcd->driver->endpoint_disable (hcd, ep);
+
+	/* Wait until the endpoint queue is completely empty.  Most HCDs
+	 * will have done this already in their endpoint_disable method,
+	 * but some might not.  And there could be root-hub control URBs
+	 * still pending since they aren't affected by the HCDs'
+	 * endpoint_disable methods.
+	 */
+	while (!list_empty (&ep->urb_list)) {
+		spin_lock_irq (&hcd_data_lock);
+
+		/* The list may have changed while we acquired the spinlock */
+		urb = NULL;
+		if (!list_empty (&ep->urb_list)) {
+			urb = list_entry (ep->urb_list.prev, struct urb,
+					urb_list);
+			usb_get_urb (urb);
+		}
+		spin_unlock_irq (&hcd_data_lock);
+
+		if (urb) {
+			usb_kill_urb (urb);
+			usb_put_urb (urb);
+		}
+	}
 }
 
 /*-------------------------------------------------------------------------*/
-- 
cgit v1.2.1


From d5cbad4b8b37acfde3e63d31b92561b87288ad0f Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Fri, 11 Aug 2006 16:52:39 -0400
Subject: usbcore: khubd and busy-port handling

We don't want khubd to start interfering in the device-resume process
merely because the PORT_STATUS_C_SUSPEND feature happens to be set.
Ports need to be marked as busy while a resume is taking place.

In addition, so long as ports are marked as busy, khubd won't be able to
clear their various status-change features.  On an interrupt-driven root
hub this could lead to an interrupt storm.  Root hub IRQs should not be
re-enabled until the busy_bits value is equal to 0.

This patch (as765) fixes these two potential problems.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 11 ++++++++++-
 drivers/usb/core/hub.h |  3 ++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 3924dd080bea..bdf5be099650 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1658,6 +1658,8 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 
 	// dev_dbg(hub->intfdev, "resume port %d\n", port1);
 
+	set_bit(port1, hub->busy_bits);
+
 	/* see 7.1.7.7; affects power usage, but not budgeting */
 	status = clear_port_feature(hub->hdev,
 			port1, USB_PORT_FEAT_SUSPEND);
@@ -1707,6 +1709,10 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 	if (status < 0)
 		hub_port_logical_disconnect(hub, port1);
 
+	clear_bit(port1, hub->busy_bits);
+	if (!hub->hdev->parent && !hub->busy_bits[0])
+		usb_enable_root_hub_irq(hub->hdev->bus);
+
 	return status;
 }
 
@@ -2690,7 +2696,7 @@ static void hub_events(void)
 
 		/* If this is a root hub, tell the HCD it's okay to
 		 * re-enable port-change interrupts now. */
-		if (!hdev->parent)
+		if (!hdev->parent && !hub->busy_bits[0])
 			usb_enable_root_hub_irq(hdev->bus);
 
 loop:
@@ -2865,6 +2871,9 @@ int usb_reset_device(struct usb_device *udev)
 			break;
 	}
 	clear_bit(port1, parent_hub->busy_bits);
+	if (!parent_hdev->parent && !parent_hub->busy_bits[0])
+		usb_enable_root_hub_irq(parent_hdev->bus);
+
 	if (ret < 0)
 		goto re_enumerate;
  
diff --git a/drivers/usb/core/hub.h b/drivers/usb/core/hub.h
index 29d5f45a8456..0f8e82a4d480 100644
--- a/drivers/usb/core/hub.h
+++ b/drivers/usb/core/hub.h
@@ -212,7 +212,8 @@ struct usb_hub {
 	unsigned long		event_bits[1];	/* status change bitmask */
 	unsigned long		change_bits[1];	/* ports with logical connect
 							status change */
-	unsigned long		busy_bits[1];	/* ports being reset */
+	unsigned long		busy_bits[1];	/* ports being reset or
+							resumed */
 #if USB_MAXCHILDREN > 31 /* 8*sizeof(unsigned long) - 1 */
 #error event_bits[] is too short!
 #endif
-- 
cgit v1.2.1


From c07045412f21c5bb344244e8ec45671529e411bd Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Mon, 14 Aug 2006 22:44:29 -0300
Subject: usb-skeleton: small update

o CodingStyle fixes
o Removes trailing spaces
o Do not make not needed initialiation of automatic variables
o Use usb_endpoint_* functions
o If we get an error in the write URB callback print an error message instead
  of a debug one

(Pretty unrelated changes, but spliting this up doesn't pay off as our main
changes are just CodingStyle fixes).

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/usb-skeleton.c | 50 ++++++++++++++++++++--------------------------
 1 file changed, 22 insertions(+), 28 deletions(-)

diff --git a/drivers/usb/usb-skeleton.c b/drivers/usb/usb-skeleton.c
index 3339373239ad..9b542a6ba978 100644
--- a/drivers/usb/usb-skeleton.c
+++ b/drivers/usb/usb-skeleton.c
@@ -1,5 +1,5 @@
 /*
- * USB Skeleton driver - 2.1
+ * USB Skeleton driver - 2.2
  *
  * Copyright (C) 2001-2004 Greg Kroah-Hartman (greg@kroah.com)
  *
@@ -7,7 +7,7 @@
  *	modify it under the terms of the GNU General Public License as
  *	published by the Free Software Foundation, version 2.
  *
- * This driver is based on the 2.6.3 version of drivers/usb/usb-skeleton.c 
+ * This driver is based on the 2.6.3 version of drivers/usb/usb-skeleton.c
  * but has been rewritten to be easier to read and use.
  *
  */
@@ -32,22 +32,22 @@ static struct usb_device_id skel_table [] = {
 	{ USB_DEVICE(USB_SKEL_VENDOR_ID, USB_SKEL_PRODUCT_ID) },
 	{ }					/* Terminating entry */
 };
-MODULE_DEVICE_TABLE (usb, skel_table);
+MODULE_DEVICE_TABLE(usb, skel_table);
 
 
 /* Get a minor range for your devices from the usb maintainer */
 #define USB_SKEL_MINOR_BASE	192
 
 /* our private defines. if this grows any larger, use your own .h file */
-#define MAX_TRANSFER		( PAGE_SIZE - 512 )
+#define MAX_TRANSFER		(PAGE_SIZE - 512)
 #define WRITES_IN_FLIGHT	8
 
 /* Structure to hold all of our device specific stuff */
 struct usb_skel {
-	struct usb_device *	udev;			/* the usb device for this device */
-	struct usb_interface *	interface;		/* the interface for this device */
+	struct usb_device       *dev;			/* the usb device for this device */
+	struct usb_interface    *interface;		/* the interface for this device */
 	struct semaphore	limit_sem;		/* limiting the number of writes in progress */
-	unsigned char *		bulk_in_buffer;		/* the buffer to receive data */
+	unsigned char           *bulk_in_buffer;	/* the buffer to receive data */
 	size_t			bulk_in_size;		/* the size of the receive buffer */
 	__u8			bulk_in_endpointAddr;	/* the address of the bulk in endpoint */
 	__u8			bulk_out_endpointAddr;	/* the address of the bulk out endpoint */
@@ -59,12 +59,12 @@ struct usb_skel {
 static struct usb_driver skel_driver;
 
 static void skel_delete(struct kref *kref)
-{	
+{
 	struct usb_skel *dev = to_skel_dev(kref);
 
 	usb_put_dev(dev->udev);
-	kfree (dev->bulk_in_buffer);
-	kfree (dev);
+	kfree(dev->bulk_in_buffer);
+	kfree(dev);
 }
 
 static int skel_open(struct inode *inode, struct file *file)
@@ -116,7 +116,7 @@ static int skel_release(struct inode *inode, struct file *file)
 static ssize_t skel_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
 {
 	struct usb_skel *dev;
-	int retval = 0;
+	int retval;
 	int bytes_read;
 
 	dev = (struct usb_skel *)file->private_data;
@@ -154,16 +154,16 @@ static void skel_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
 	dev = (struct usb_skel *)urb->context;
 
 	/* sync/async unlink faults aren't errors */
-	if (urb->status && 
-	    !(urb->status == -ENOENT || 
+	if (urb->status &&
+	    !(urb->status == -ENOENT ||
 	      urb->status == -ECONNRESET ||
 	      urb->status == -ESHUTDOWN)) {
-		dbg("%s - nonzero write bulk status received: %d",
+		err("%s - nonzero write bulk status received: %d",
 		    __FUNCTION__, urb->status);
 	}
 
 	/* free up our allocated buffer */
-	usb_buffer_free(urb->dev, urb->transfer_buffer_length, 
+	usb_buffer_free(urb->dev, urb->transfer_buffer_length,
 			urb->transfer_buffer, urb->transfer_dma);
 	up(&dev->limit_sem);
 }
@@ -251,7 +251,7 @@ static const struct file_operations skel_fops = {
 	.release =	skel_release,
 };
 
-/* 
+/*
  * usb class driver info in order to get a minor number from the usb core,
  * and to have the device registered with the driver core
  */
@@ -263,7 +263,7 @@ static struct usb_class_driver skel_class = {
 
 static int skel_probe(struct usb_interface *interface, const struct usb_device_id *id)
 {
-	struct usb_skel *dev = NULL;
+	struct usb_skel *dev;
 	struct usb_host_interface *iface_desc;
 	struct usb_endpoint_descriptor *endpoint;
 	size_t buffer_size;
@@ -272,7 +272,7 @@ static int skel_probe(struct usb_interface *interface, const struct usb_device_i
 
 	/* allocate memory for our device state and initialize it */
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
-	if (dev == NULL) {
+	if (!dev) {
 		err("Out of memory");
 		goto error;
 	}
@@ -290,10 +290,7 @@ static int skel_probe(struct usb_interface *interface, const struct usb_device_i
 		endpoint = &iface_desc->endpoint[i].desc;
 
 		if (!dev->bulk_in_endpointAddr &&
-		    ((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK)
-					== USB_DIR_IN) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-					== USB_ENDPOINT_XFER_BULK)) {
+		    usb_endpoint_is_bulk_in(endpoint)) {
 			/* we found a bulk in endpoint */
 			buffer_size = le16_to_cpu(endpoint->wMaxPacketSize);
 			dev->bulk_in_size = buffer_size;
@@ -306,10 +303,7 @@ static int skel_probe(struct usb_interface *interface, const struct usb_device_i
 		}
 
 		if (!dev->bulk_out_endpointAddr &&
-		    ((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK)
-					== USB_DIR_OUT) &&
-		    ((endpoint->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
-					== USB_ENDPOINT_XFER_BULK)) {
+		    usb_endpoint_is_bulk_out(endpoint)) {
 			/* we found a bulk out endpoint */
 			dev->bulk_out_endpointAddr = endpoint->bEndpointAddress;
 		}
@@ -393,7 +387,7 @@ static void __exit usb_skel_exit(void)
 	usb_deregister(&skel_driver);
 }
 
-module_init (usb_skel_init);
-module_exit (usb_skel_exit);
+module_init(usb_skel_init);
+module_exit(usb_skel_exit);
 
 MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From dfe0d3ba20e860d0b9a16c4c6524180b8f93be05 Mon Sep 17 00:00:00 2001
From: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Date: Sun, 13 Aug 2006 17:30:14 -0700
Subject: USB Storage: add rio karma eject support

This changeset from Keith Bennett (via Bob Copeland) moves the Karma
initializer to its own file and adds trapping of the START_STOP command to
enable eject of the device.

Signed-off-by: Keith Bennett <keith@mcs.st-and.ac.uk>
Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/Kconfig        |  12 +++
 drivers/usb/storage/Makefile       |   1 +
 drivers/usb/storage/initializers.c |  73 -----------------
 drivers/usb/storage/initializers.h |   1 -
 drivers/usb/storage/karma.c        | 155 +++++++++++++++++++++++++++++++++++++
 drivers/usb/storage/karma.h        |   7 ++
 drivers/usb/storage/unusual_devs.h |   2 +-
 drivers/usb/storage/usb.c          |  11 +++
 include/linux/usb_usual.h          |   3 +
 9 files changed, 190 insertions(+), 75 deletions(-)
 create mode 100644 drivers/usb/storage/karma.c
 create mode 100644 drivers/usb/storage/karma.h

diff --git a/drivers/usb/storage/Kconfig b/drivers/usb/storage/Kconfig
index be9eec225743..86e48c42d6af 100644
--- a/drivers/usb/storage/Kconfig
+++ b/drivers/usb/storage/Kconfig
@@ -135,6 +135,18 @@ config USB_STORAGE_ONETOUCH
 	  this input in any keybinding software. (e.g. gnome's keyboard short-
 	  cuts)
 
+config USB_STORAGE_KARMA
+	bool "Support for Rio Karma music player"
+	depends on USB_STORAGE
+	help
+	  Say Y here to include additional code to support the Rio Karma
+	  USB interface.
+
+	  This code places the Rio Karma into mass storage mode, enabling
+	  it to be mounted as an ordinary filesystem. Performing an eject
+	  on the resulting scsi device node returns the Karma to normal
+	  operation.
+
 config USB_LIBUSUAL
 	bool "The shared table of common (or usual) storage devices"
 	depends on USB
diff --git a/drivers/usb/storage/Makefile b/drivers/usb/storage/Makefile
index 8cbba22508a4..023969b4385b 100644
--- a/drivers/usb/storage/Makefile
+++ b/drivers/usb/storage/Makefile
@@ -20,6 +20,7 @@ usb-storage-obj-$(CONFIG_USB_STORAGE_DATAFAB)	+= datafab.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_JUMPSHOT)	+= jumpshot.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_ALAUDA)	+= alauda.o
 usb-storage-obj-$(CONFIG_USB_STORAGE_ONETOUCH)	+= onetouch.o
+usb-storage-obj-$(CONFIG_USB_STORAGE_KARMA)	+= karma.o
 
 usb-storage-objs :=	scsiglue.o protocol.o transport.o usb.o \
 			initializers.o $(usb-storage-obj-y)
diff --git a/drivers/usb/storage/initializers.c b/drivers/usb/storage/initializers.c
index ab173b30076e..5b06f9240d05 100644
--- a/drivers/usb/storage/initializers.c
+++ b/drivers/usb/storage/initializers.c
@@ -45,12 +45,6 @@
 #include "debug.h"
 #include "transport.h"
 
-#define RIO_MSC 0x08
-#define RIOP_INIT "RIOP\x00\x01\x08"
-#define RIOP_INIT_LEN 7
-#define RIO_SEND_LEN 40
-#define RIO_RECV_LEN 0x200
-
 /* This places the Shuttle/SCM USB<->SCSI bridge devices in multi-target
  * mode */
 int usb_stor_euscsi_init(struct us_data *us)
@@ -97,70 +91,3 @@ int usb_stor_ucr61s2b_init(struct us_data *us)
 
 	return (res ? -1 : 0);
 }
-
-/* Place the Rio Karma into mass storage mode.
- *
- * The initialization begins by sending 40 bytes starting
- * RIOP\x00\x01\x08\x00, which the device will ack with a 512-byte
- * packet with the high four bits set and everything else null.
- *
- * Next, we send RIOP\x80\x00\x08\x00.  Each time, a 512 byte response
- * must be read, but we must loop until byte 5 in the response is 0x08,
- * indicating success.  */
-int rio_karma_init(struct us_data *us)
-{
-	int result, partial;
-	char *recv;
-	unsigned long timeout;
-
-	// us->iobuf is big enough to hold cmd but not receive
-	if (!(recv = kmalloc(RIO_RECV_LEN, GFP_KERNEL)))
-		goto die_nomem;
-
-	US_DEBUGP("Initializing Karma...\n");
-
-	memset(us->iobuf, 0, RIO_SEND_LEN);
-	memcpy(us->iobuf, RIOP_INIT, RIOP_INIT_LEN);
-
-	result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
-		us->iobuf, RIO_SEND_LEN, &partial);
-	if (result != USB_STOR_XFER_GOOD)
-		goto die;
-
-	result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
-		recv, RIO_RECV_LEN, &partial);
-	if (result != USB_STOR_XFER_GOOD)
-		goto die;
-
-	us->iobuf[4] = 0x80;
-	us->iobuf[5] = 0;
-	timeout = jiffies + msecs_to_jiffies(3000);
-	for (;;) {
-		US_DEBUGP("Sending init command\n");
-		result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
-			us->iobuf, RIO_SEND_LEN, &partial);
-		if (result != USB_STOR_XFER_GOOD)
-			goto die;
-
-		result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
-			recv, RIO_RECV_LEN, &partial);
-		if (result != USB_STOR_XFER_GOOD)
-			goto die;
-
-		if (recv[5] == RIO_MSC)
-			break;
-		if (time_after(jiffies, timeout))
-			goto die;
-		msleep(10);
-	}
-	US_DEBUGP("Karma initialized.\n");
-	kfree(recv);
-	return 0;
-
-die:
-	kfree(recv);
-die_nomem:
-	US_DEBUGP("Could not initialize karma.\n");
-	return USB_STOR_TRANSPORT_FAILED;
-}
-
diff --git a/drivers/usb/storage/initializers.h b/drivers/usb/storage/initializers.h
index 927f7781080f..e2967a4d48a2 100644
--- a/drivers/usb/storage/initializers.h
+++ b/drivers/usb/storage/initializers.h
@@ -47,4 +47,3 @@ int usb_stor_euscsi_init(struct us_data *us);
 /* This function is required to activate all four slots on the UCR-61S2B
  * flash reader */
 int usb_stor_ucr61s2b_init(struct us_data *us);
-int rio_karma_init(struct us_data *us);
diff --git a/drivers/usb/storage/karma.c b/drivers/usb/storage/karma.c
new file mode 100644
index 000000000000..0d79ae5683f7
--- /dev/null
+++ b/drivers/usb/storage/karma.c
@@ -0,0 +1,155 @@
+/* Driver for Rio Karma
+ *
+ *   (c) 2006 Bob Copeland <me@bobcopeland.com>
+ *   (c) 2006 Keith Bennett <keith@mcs.st-and.ac.uk>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the
+ * Free Software Foundation; either version 2, or (at your option) any
+ * later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+#include <scsi/scsi.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_device.h>
+
+#include "usb.h"
+#include "transport.h"
+#include "debug.h"
+#include "karma.h"
+
+#define RIO_PREFIX "RIOP\x00"
+#define RIO_PREFIX_LEN 5
+#define RIO_SEND_LEN 40
+#define RIO_RECV_LEN 0x200
+
+#define RIO_ENTER_STORAGE 0x1
+#define RIO_LEAVE_STORAGE 0x2
+#define RIO_RESET 0xC
+
+extern int usb_stor_Bulk_transport(struct scsi_cmnd *, struct us_data *);
+
+struct karma_data {
+	int in_storage;
+	char *recv;
+};
+
+/*
+ * Send commands to Rio Karma.
+ *
+ * For each command we send 40 bytes starting 'RIOP\0' followed by
+ * the command number and a sequence number, which the device will ack
+ * with a 512-byte packet with the high four bits set and everything
+ * else null.  Then we send 'RIOP\x80' followed by a zero and the
+ * sequence number, until byte 5 in the response repeats the sequence
+ * number.
+ */
+static int rio_karma_send_command(char cmd, struct us_data *us)
+{
+	int result, partial;
+	unsigned long timeout;
+	static unsigned char seq = 1;
+	struct karma_data *data = (struct karma_data *) us->extra;
+
+	US_DEBUGP("karma: sending command %04x\n", cmd);
+	memset(us->iobuf, 0, RIO_SEND_LEN);
+	memcpy(us->iobuf, RIO_PREFIX, RIO_PREFIX_LEN);
+	us->iobuf[5] = cmd;
+	us->iobuf[6] = seq;
+
+	timeout = jiffies + msecs_to_jiffies(6000);
+	for (;;) {
+		result = usb_stor_bulk_transfer_buf(us, us->send_bulk_pipe,
+			us->iobuf, RIO_SEND_LEN, &partial);
+		if (result != USB_STOR_XFER_GOOD)
+			goto err;
+
+		result = usb_stor_bulk_transfer_buf(us, us->recv_bulk_pipe,
+			data->recv, RIO_RECV_LEN, &partial);
+		if (result != USB_STOR_XFER_GOOD)
+			goto err;
+
+		if (data->recv[5] == seq)
+			break;
+
+		if (time_after(jiffies, timeout))
+			goto err;
+
+		us->iobuf[4] = 0x80;
+		us->iobuf[5] = 0;
+		msleep(50);
+	}
+
+	seq++;
+	if (seq == 0)
+		seq = 1;
+
+	US_DEBUGP("karma: sent command %04x\n", cmd);
+	return 0;
+err:
+	US_DEBUGP("karma: command %04x failed\n", cmd);
+	return USB_STOR_TRANSPORT_FAILED;
+}
+
+/*
+ * Trap START_STOP and READ_10 to leave/re-enter storage mode.
+ * Everything else is propagated to the normal bulk layer.
+ */
+int rio_karma_transport(struct scsi_cmnd *srb, struct us_data *us)
+{
+	int ret;
+	struct karma_data *data = (struct karma_data *) us->extra;
+
+	if (srb->cmnd[0] == READ_10 && !data->in_storage) {
+		ret = rio_karma_send_command(RIO_ENTER_STORAGE, us);
+		if (ret)
+			return ret;
+
+		data->in_storage = 1;
+		return usb_stor_Bulk_transport(srb, us);
+	} else if (srb->cmnd[0] == START_STOP) {
+		ret = rio_karma_send_command(RIO_LEAVE_STORAGE, us);
+		if (ret)
+			return ret;
+
+		data->in_storage = 0;
+		return rio_karma_send_command(RIO_RESET, us);
+	}
+	return usb_stor_Bulk_transport(srb, us);
+}
+
+static void rio_karma_destructor(void *extra)
+{
+	struct karma_data *data = (struct karma_data *) extra;
+	kfree(data->recv);
+}
+
+int rio_karma_init(struct us_data *us)
+{
+	int ret = 0;
+	struct karma_data *data = kzalloc(sizeof(struct karma_data), GFP_NOIO);
+	if (!data)
+		goto out;
+
+	data->recv = kmalloc(RIO_RECV_LEN, GFP_NOIO);
+	if (!data->recv) {
+		kfree(data);
+		goto out;
+	}
+
+	us->extra = data;
+	us->extra_destructor = rio_karma_destructor;
+	ret = rio_karma_send_command(RIO_ENTER_STORAGE, us);
+	data->in_storage = (ret == 0);
+out:
+	return ret;
+}
diff --git a/drivers/usb/storage/karma.h b/drivers/usb/storage/karma.h
new file mode 100644
index 000000000000..8a60972af8c5
--- /dev/null
+++ b/drivers/usb/storage/karma.h
@@ -0,0 +1,7 @@
+#ifndef _KARMA_USB_H
+#define _KARMA_USB_H
+
+extern int rio_karma_init(struct us_data *us);
+extern int rio_karma_transport(struct scsi_cmnd *srb, struct us_data *us);
+
+#endif
diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index fa49357289c4..1f11c9d44eaa 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -221,7 +221,7 @@ UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
 UNUSUAL_DEV(  0x045a, 0x5210, 0x0101, 0x0101,
 		"Rio",
 		"Rio Karma",
-		US_SC_SCSI, US_PR_BULK, rio_karma_init, 0),
+		US_SC_SCSI, US_PR_KARMA, rio_karma_init, 0),
 
 /* Patch submitted by Philipp Friedrich <philipp@void.at> */
 UNUSUAL_DEV(  0x0482, 0x0100, 0x0100, 0x0100,
diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c
index 8d7bdcb5924d..b8d6031b0975 100644
--- a/drivers/usb/storage/usb.c
+++ b/drivers/usb/storage/usb.c
@@ -98,6 +98,9 @@
 #ifdef CONFIG_USB_STORAGE_ALAUDA
 #include "alauda.h"
 #endif
+#ifdef CONFIG_USB_STORAGE_KARMA
+#include "karma.h"
+#endif
 
 /* Some informational data */
 MODULE_AUTHOR("Matthew Dharm <mdharm-usb@one-eyed-alien.net>");
@@ -646,6 +649,14 @@ static int get_transport(struct us_data *us)
 		break;
 #endif
 
+#ifdef CONFIG_USB_STORAGE_KARMA
+	case US_PR_KARMA:
+		us->transport_name = "Rio Karma/Bulk";
+		us->transport = rio_karma_transport;
+		us->transport_reset = usb_stor_Bulk_reset;
+		break;
+#endif
+
 	default:
 		return -EIO;
 	}
diff --git a/include/linux/usb_usual.h b/include/linux/usb_usual.h
index e7fc5fed5b98..2ae76fe52ff7 100644
--- a/include/linux/usb_usual.h
+++ b/include/linux/usb_usual.h
@@ -108,6 +108,9 @@ enum { US_DO_ALL_FLAGS };
 #ifdef CONFIG_USB_STORAGE_ALAUDA
 #define US_PR_ALAUDA    0xf4		/* Alauda chipsets */
 #endif
+#ifdef CONFIG_USB_STORAGE_KARMA
+#define US_PR_KARMA     0xf5		/* Rio Karma */
+#endif
 
 #define US_PR_DEVICE	0xff		/* Use device's value */
 
-- 
cgit v1.2.1


From cb4c8fe57c05dbb04128503f4a7483a1163b1b47 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Date: Fri, 25 Aug 2006 19:35:28 -0700
Subject: usb: deal with broken config descriptors

Change usb_get_configuration() so that it is more tolerant to devices
with bad configuration descriptors (it'll make it ignore
configurations that fail to load).

Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/config.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c
index 4c9e63e665b6..bfb3731d42db 100644
--- a/drivers/usb/core/config.c
+++ b/drivers/usb/core/config.c
@@ -475,7 +475,9 @@ int usb_get_configuration(struct usb_device *dev)
 		if (result < 0) {
 			dev_err(ddev, "unable to read config index %d "
 			    "descriptor/%s\n", cfgno, "start");
-			goto err;
+			dev_err(ddev, "chopping to %d config(s)\n", cfgno);
+			dev->descriptor.bNumConfigurations = cfgno;
+			break;
 		} else if (result < 4) {
 			dev_err(ddev, "config index %d descriptor too short "
 			    "(expected %i, got %i)\n", cfgno,
-- 
cgit v1.2.1


From 0165de09747be76b09ef769fcfed3514fe5f6509 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Date: Fri, 25 Aug 2006 19:35:29 -0700
Subject: wusb: hub code recognizes wusb ports

This patch enables the USB stack to recognize WUSB devices (from a
WUSB HCD) and assigns them the proper speed setting
(USB_SPEED_VARIABLE).

1. Introduce usb_hcd->wireless to mark a host controller instance as
   being wireless, and thus having wireless 'fake' ports.

   [discarded previous model of using a reserved bit in the port_stat
   struct to do this; thanks to Alan Stern for indicating the
   proper way to do it].

2. Introduce hub.c:hub_is_wusb() that tests if a hub is a WUSB root
   hub (WUSB doesn't have non-root hubs).

New code being pushed to linuxuwb.org requires this patch to connect WUSB
devices.

Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.h |  1 +
 drivers/usb/core/hub.c | 16 +++++++++++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 58c7767bc904..fc71a08a1af4 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -85,6 +85,7 @@ struct usb_hcd {	/* usb_bus.hcpriv points to this */
 	unsigned		uses_new_polling:1;
 	unsigned		poll_rh:1;	/* poll for rh status? */
 	unsigned		poll_pending:1;	/* status has changed? */
+	unsigned		wireless:1;	/* Wireless USB HCD */
 
 	int			irq;		/* irq allocated */
 	void __iomem		*regs;		/* device memory/io */
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index bdf5be099650..db4a9be1cb8a 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1331,6 +1331,18 @@ static int hub_port_status(struct usb_hub *hub, int port1,
 	return ret;
 }
 
+
+/* Returns 1 if @hub is a WUSB root hub, 0 otherwise */
+static unsigned hub_is_wusb(struct usb_hub *hub)
+{
+	struct usb_hcd *hcd;
+	if (hub->hdev->parent != NULL)  /* not a root hub? */
+		return 0;
+	hcd = container_of(hub->hdev->bus, struct usb_hcd, self);
+	return hcd->wireless;
+}
+
+
 #define PORT_RESET_TRIES	5
 #define SET_ADDRESS_TRIES	2
 #define GET_DESCRIPTOR_TRIES	2
@@ -1371,7 +1383,9 @@ static int hub_port_wait_reset(struct usb_hub *hub, int port1,
 		/* if we`ve finished resetting, then break out of the loop */
 		if (!(portstatus & USB_PORT_STAT_RESET) &&
 		    (portstatus & USB_PORT_STAT_ENABLE)) {
-			if (portstatus & USB_PORT_STAT_HIGH_SPEED)
+			if (hub_is_wusb(hub))
+				udev->speed = USB_SPEED_VARIABLE;
+			else if (portstatus & USB_PORT_STAT_HIGH_SPEED)
 				udev->speed = USB_SPEED_HIGH;
 			else if (portstatus & USB_PORT_STAT_LOW_SPEED)
 				udev->speed = USB_SPEED_LOW;
-- 
cgit v1.2.1


From 5bb6e0ae8f9f3a215d6a7f99c8486b0301cc5db9 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Date: Fri, 25 Aug 2006 19:35:30 -0700
Subject: wusb: handle wusb device ep0 speed settings

This patch teaches the USB stack handling of WUSB devices (those whose
speed is USB_SPEED_VARIABLE). For these devices, we need to set ep0's
maxpacketsize to 512 (even though the device descriptor reports it as
0xff).

New code being pushed to linuxuwb.org requires this patch to connect WUSB
devices.

Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index db4a9be1cb8a..1bd5ee26f0e0 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2060,8 +2060,13 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
   
 	/* USB 2.0 section 5.5.3 talks about ep0 maxpacket ...
 	 * it's fixed size except for full speed devices.
+	 * For Wireless USB devices, ep0 max packet is always 512 (tho
+	 * reported as 0xff in the device descriptor). WUSB1.0[4.8.1].
 	 */
 	switch (udev->speed) {
+	case USB_SPEED_VARIABLE:	/* fixed at 512 */
+		udev->ep0.desc.wMaxPacketSize = __constant_cpu_to_le16(512);
+		break;
 	case USB_SPEED_HIGH:		/* fixed at 64 */
 		udev->ep0.desc.wMaxPacketSize = __constant_cpu_to_le16(64);
 		break;
@@ -2131,6 +2136,8 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
 			 * down tremendously by NAKing the unexpectedly
 			 * early status stage.  Also, retry on all errors;
 			 * some devices are flakey.
+			 * 255 is for WUSB devices, we actually need to use 512.
+			 * WUSB1.0[4.8.1].
 			 */
 			for (j = 0; j < 3; ++j) {
 				buf->bMaxPacketSize0 = 0;
@@ -2140,7 +2147,7 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
 					buf, GET_DESCRIPTOR_BUFSIZE,
 					(i ? USB_CTRL_GET_TIMEOUT : 1000));
 				switch (buf->bMaxPacketSize0) {
-				case 8: case 16: case 32: case 64:
+				case 8: case 16: case 32: case 64: case 255:
 					if (buf->bDescriptorType ==
 							USB_DT_DEVICE) {
 						r = 0;
@@ -2214,7 +2221,8 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
 	if (retval)
 		goto fail;
 
-	i = udev->descriptor.bMaxPacketSize0;
+	i = udev->descriptor.bMaxPacketSize0 == 0xff?
+	    512 : udev->descriptor.bMaxPacketSize0;
 	if (le16_to_cpu(udev->ep0.desc.wMaxPacketSize) != i) {
 		if (udev->speed != USB_SPEED_FULL ||
 				!(i == 8 || i == 16 || i == 32 || i == 64)) {
-- 
cgit v1.2.1


From 83a07196735dbf371b61d6dffbb7e6a696c633c2 Mon Sep 17 00:00:00 2001
From: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Date: Fri, 25 Aug 2006 19:35:31 -0700
Subject: wusb: pretty print new wireless USB devices when they connect

New code being pushed to linuxuwb.org requires this patch to connect
WUSB devices.

Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 1bd5ee26f0e0..65720f2bffc8 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2028,6 +2028,7 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
 	int			i, j, retval;
 	unsigned		delay = HUB_SHORT_RESET_TIME;
 	enum usb_device_speed	oldspeed = udev->speed;
+	char 			*speed, *type;
 
 	/* root hub ports have a slightly longer reset period
 	 * (from USB 2.0 spec, section 7.1.7.5)
@@ -2084,17 +2085,21 @@ hub_port_init (struct usb_hub *hub, struct usb_device *udev, int port1,
 		goto fail;
 	}
  
+	type = "";
+	switch (udev->speed) {
+	case USB_SPEED_LOW:	speed = "low";	break;
+	case USB_SPEED_FULL:	speed = "full";	break;
+	case USB_SPEED_HIGH:	speed = "high";	break;
+	case USB_SPEED_VARIABLE:
+				speed = "variable";
+				type = "Wireless ";
+				break;
+	default: 		speed = "?";	break;
+	}
 	dev_info (&udev->dev,
-			"%s %s speed USB device using %s and address %d\n",
-			(udev->config) ? "reset" : "new",
-			({ char *speed; switch (udev->speed) {
-			case USB_SPEED_LOW:	speed = "low";	break;
-			case USB_SPEED_FULL:	speed = "full";	break;
-			case USB_SPEED_HIGH:	speed = "high";	break;
-			default: 		speed = "?";	break;
-			}; speed;}),
-			udev->bus->controller->driver->name,
-			udev->devnum);
+		  "%s %s speed %sUSB device using %s and address %d\n",
+		  (udev->config) ? "reset" : "new", speed, type,
+		  udev->bus->controller->driver->name, udev->devnum);
 
 	/* Set up TT records, if needed  */
 	if (hdev->tt) {
-- 
cgit v1.2.1


From 095bc335360a51623dd8571839bbf465851a7f4b Mon Sep 17 00:00:00 2001
From: "Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Date: Sat, 26 Aug 2006 23:48:11 -0300
Subject: USB core: Use const where possible.

This patch marks some USB core's functions parameters as const. This
improves the design (we're saying to the caller that its parameter is
not going to be modified) and may help in compiler's optimisation work.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.c | 21 +++++++++++----------
 drivers/usb/core/usb.h |  4 ++--
 include/linux/usb.h    | 18 +++++++++---------
 3 files changed, 22 insertions(+), 21 deletions(-)

diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 82837d45b484..4eb98eb3804f 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -67,7 +67,8 @@ static int nousb;	/* Disable USB when built into kernel image */
  * Don't call this function unless you are bound to one of the interfaces
  * on this device or you have locked the device!
  */
-struct usb_interface *usb_ifnum_to_if(struct usb_device *dev, unsigned ifnum)
+struct usb_interface *usb_ifnum_to_if(const struct usb_device *dev,
+				      unsigned ifnum)
 {
 	struct usb_host_config *config = dev->actconfig;
 	int i;
@@ -100,8 +101,8 @@ struct usb_interface *usb_ifnum_to_if(struct usb_device *dev, unsigned ifnum)
  * Don't call this function unless you are bound to the intf interface
  * or you have locked the device!
  */
-struct usb_host_interface *usb_altnum_to_altsetting(struct usb_interface *intf,
-		unsigned int altnum)
+struct usb_host_interface *usb_altnum_to_altsetting(const struct usb_interface *intf,
+						    unsigned int altnum)
 {
 	int i;
 
@@ -356,7 +357,7 @@ void usb_put_intf(struct usb_interface *intf)
  * case the driver already owns the device lock.)
  */
 int usb_lock_device_for_reset(struct usb_device *udev,
-		struct usb_interface *iface)
+			      const struct usb_interface *iface)
 {
 	unsigned long jiffies_expire = jiffies + HZ;
 
@@ -852,8 +853,8 @@ void usb_buffer_unmap (struct urb *urb)
  *
  * Reverse the effect of this call with usb_buffer_unmap_sg().
  */
-int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int nents)
+int usb_buffer_map_sg(const struct usb_device *dev, unsigned pipe,
+		      struct scatterlist *sg, int nents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
@@ -887,8 +888,8 @@ int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
  * Use this when you are re-using a scatterlist's data buffers for
  * another USB request.
  */
-void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents)
+void usb_buffer_dmasync_sg(const struct usb_device *dev, unsigned pipe,
+			   struct scatterlist *sg, int n_hw_ents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
@@ -913,8 +914,8 @@ void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
  *
  * Reverses the effect of usb_buffer_map_sg().
  */
-void usb_buffer_unmap_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents)
+void usb_buffer_unmap_sg(const struct usb_device *dev, unsigned pipe,
+			 struct scatterlist *sg, int n_hw_ents)
 {
 	struct usb_bus		*bus;
 	struct device		*controller;
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 6096ead2758c..67da6d0b316f 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -53,7 +53,7 @@ extern struct usb_device_driver usb_generic_driver;
  * no such thing as a platform USB device, so we can steal the use
  * of the platform_data field. */
 
-static inline int is_usb_device(struct device *dev)
+static inline int is_usb_device(const struct device *dev)
 {
 	return dev->platform_data == &usb_generic_driver;
 }
@@ -78,7 +78,7 @@ static inline void mark_quiesced(struct usb_interface *f)
 	f->is_active = 0;
 }
 
-static inline int is_active(struct usb_interface *f)
+static inline int is_active(const struct usb_interface *f)
 {
 	return f->is_active;
 }
diff --git a/include/linux/usb.h b/include/linux/usb.h
index f807479ef65b..26d8a5f36896 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -387,7 +387,7 @@ extern void usb_put_dev(struct usb_device *dev);
 #define usb_unlock_device(udev)		up(&(udev)->dev.sem)
 #define usb_trylock_device(udev)	down_trylock(&(udev)->dev.sem)
 extern int usb_lock_device_for_reset(struct usb_device *udev,
-		struct usb_interface *iface);
+				     const struct usb_interface *iface);
 
 /* USB port reset for device reinitialization */
 extern int usb_reset_device(struct usb_device *dev);
@@ -426,10 +426,10 @@ const struct usb_device_id *usb_match_id(struct usb_interface *interface,
 
 extern struct usb_interface *usb_find_interface(struct usb_driver *drv,
 		int minor);
-extern struct usb_interface *usb_ifnum_to_if(struct usb_device *dev,
+extern struct usb_interface *usb_ifnum_to_if(const struct usb_device *dev,
 		unsigned ifnum);
 extern struct usb_host_interface *usb_altnum_to_altsetting(
-		struct usb_interface *intf, unsigned int altnum);
+		const struct usb_interface *intf, unsigned int altnum);
 
 
 /**
@@ -1064,14 +1064,14 @@ void usb_buffer_unmap (struct urb *urb);
 #endif
 
 struct scatterlist;
-int usb_buffer_map_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int nents);
+int usb_buffer_map_sg(const struct usb_device *dev, unsigned pipe,
+		      struct scatterlist *sg, int nents);
 #if 0
-void usb_buffer_dmasync_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents);
+void usb_buffer_dmasync_sg(const struct usb_device *dev, unsigned pipe,
+			   struct scatterlist *sg, int n_hw_ents);
 #endif
-void usb_buffer_unmap_sg (struct usb_device *dev, unsigned pipe,
-		struct scatterlist *sg, int n_hw_ents);
+void usb_buffer_unmap_sg(const struct usb_device *dev, unsigned pipe,
+			 struct scatterlist *sg, int n_hw_ents);
 
 /*-------------------------------------------------------------------*
  *                         SYNCHRONOUS CALL SUPPORT                  *
-- 
cgit v1.2.1


From 55359021b9a75a6d61a49ca8b9a1209793cd55f7 Mon Sep 17 00:00:00 2001
From: Eric Sesterhenn <snakebyte@gmx.de>
Date: Mon, 21 Aug 2006 15:31:05 -0700
Subject: USB: fix signedness issue in drivers/usb/gadget/ether.c

another gcc 4.1 signdness warning:

drivers/usb/gadget/ether.c:2028: warning: comparison of unsigned expression < 0 is always false

length is assigned the value of usb_ep_queue() which returns an int.
Directly after this it is checked for < 0, which can never be true.  Making
length an int makes the error check work again.

Signed-off-by: Eric Sesterhenn <snakebyte@gmx.de>
Acked-by: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/ether.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index fed484da593c..1cc3c0227d10 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -2014,7 +2014,7 @@ rndis_control_ack_complete (struct usb_ep *ep, struct usb_request *req)
 static int rndis_control_ack (struct net_device *net)
 {
 	struct eth_dev          *dev = netdev_priv(net);
-	u32                     length;
+	int                     length;
 	struct usb_request      *resp = dev->stat_req;
 
 	/* in case RNDIS calls this after disconnect */
-- 
cgit v1.2.1


From e113f29c902f0f4bbfa5370c380e5927e6e78f8e Mon Sep 17 00:00:00 2001
From: Jules Villard <jvillard@ens-lyon.fr>
Date: Tue, 22 Aug 2006 22:40:15 +0200
Subject: USB: fix typo in drivers/usb/gadget/Kconfig

This tiny patch fixes a typo in drivers/usb/gadget/Kconfig. The typo
is present in 2.6.18-rc4 and in the corresponding -mm tree (and AFAIK,
FYI and FWIW was present in previous kernel versions as well).

From: Jules Villard <jvillard@ens-lyon.fr>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index 4301e96c417b..8e5dd6f29d0b 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -26,7 +26,7 @@ config USB_GADGET
 	   you need a low level bus controller driver, and some software
 	   talking to it.  Peripheral controllers are often discrete silicon,
 	   or are integrated with the CPU in a microcontroller.  The more
-	   familiar host side controllers have names like like "EHCI", "OHCI",
+	   familiar host side controllers have names like "EHCI", "OHCI",
 	   or "UHCI", and are usually integrated into southbridges on PC
 	   motherboards.
 
-- 
cgit v1.2.1


From 3a3416b12f1fbd607bc137a57c924a628aa5485c Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 21 Aug 2006 12:00:53 -0400
Subject: usb-storage: fix for UFI LUN detection

The UFI specification doesn't permit devices to indicate non-existent
LUNs in the manner prescribed by the SCSI spec.  This patch (as773)
sets a special flag so that the SCSI scanner will recognize these
devices and treat them specially.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/scsiglue.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
index a4b7df9ff8c1..e1072d52d641 100644
--- a/drivers/usb/storage/scsiglue.c
+++ b/drivers/usb/storage/scsiglue.c
@@ -72,12 +72,27 @@ static const char* host_info(struct Scsi_Host *host)
 
 static int slave_alloc (struct scsi_device *sdev)
 {
+	struct us_data *us = host_to_us(sdev->host);
+
 	/*
 	 * Set the INQUIRY transfer length to 36.  We don't use any of
 	 * the extra data and many devices choke if asked for more or
 	 * less than 36 bytes.
 	 */
 	sdev->inquiry_len = 36;
+
+	/*
+	 * The UFI spec treates the Peripheral Qualifier bits in an
+	 * INQUIRY result as reserved and requires devices to set them
+	 * to 0.  However the SCSI spec requires these bits to be set
+	 * to 3 to indicate when a LUN is not present.
+	 *
+	 * Let the scanning code know if this target merely sets
+	 * Peripheral Device Type to 0x1f to indicate no LUN.
+	 */
+	if (us->subclass == US_SC_UFI)
+		sdev->sdev_target->pdt_1f_for_no_lun = 1;
+
 	return 0;
 }
 
-- 
cgit v1.2.1


From 088dc270e1da03744d977cbd9edd4311af142348 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 21 Aug 2006 12:08:19 -0400
Subject: usbcore: help drivers to change device configs

It's generally a bad idea for USB interface drivers to try to change a
device's configuration, and usbcore doesn't provide any way for them
to do it.  However in a few exceptional circumstances it can make
sense.  This patch (as767) adds a roundabout mechanism to help drivers
that may need it.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/message.c | 59 ++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h        |  3 +++
 2 files changed, 62 insertions(+)

diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 49cfd7928a1c..333b22c68aa4 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -1493,6 +1493,65 @@ free_interfaces:
 	return 0;
 }
 
+struct set_config_request {
+	struct usb_device	*udev;
+	int			config;
+	struct work_struct	work;
+};
+
+/* Worker routine for usb_driver_set_configuration() */
+static void driver_set_config_work(void *_req)
+{
+	struct set_config_request *req = _req;
+
+	usb_lock_device(req->udev);
+	usb_set_configuration(req->udev, req->config);
+	usb_unlock_device(req->udev);
+	usb_put_dev(req->udev);
+	kfree(req);
+}
+
+/**
+ * usb_driver_set_configuration - Provide a way for drivers to change device configurations
+ * @udev: the device whose configuration is being updated
+ * @config: the configuration being chosen.
+ * Context: In process context, must be able to sleep
+ *
+ * Device interface drivers are not allowed to change device configurations.
+ * This is because changing configurations will destroy the interface the
+ * driver is bound to and create new ones; it would be like a floppy-disk
+ * driver telling the computer to replace the floppy-disk drive with a
+ * tape drive!
+ *
+ * Still, in certain specialized circumstances the need may arise.  This
+ * routine gets around the normal restrictions by using a work thread to
+ * submit the change-config request.
+ *
+ * Returns 0 if the request was succesfully queued, error code otherwise.
+ * The caller has no way to know whether the queued request will eventually
+ * succeed.
+ */
+int usb_driver_set_configuration(struct usb_device *udev, int config)
+{
+	struct set_config_request *req;
+
+	req = kmalloc(sizeof(*req), GFP_KERNEL);
+	if (!req)
+		return -ENOMEM;
+	req->udev = udev;
+	req->config = config;
+	INIT_WORK(&req->work, driver_set_config_work, req);
+
+	usb_get_dev(udev);
+	if (!schedule_work(&req->work)) {
+		usb_put_dev(udev);
+		kfree(req);
+		return -EINVAL;
+	}
+	return 0;
+}
+EXPORT_SYMBOL_GPL(usb_driver_set_configuration);
+
 // synchronous request completion model
 EXPORT_SYMBOL(usb_control_msg);
 EXPORT_SYMBOL(usb_bulk_msg);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 26d8a5f36896..f104efa04d79 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -1099,6 +1099,9 @@ extern int usb_clear_halt(struct usb_device *dev, int pipe);
 extern int usb_reset_configuration(struct usb_device *dev);
 extern int usb_set_interface(struct usb_device *dev, int ifnum, int alternate);
 
+/* this request isn't really synchronous, but it belongs with the others */
+extern int usb_driver_set_configuration(struct usb_device *udev, int config);
+
 /*
  * timeouts, in milliseconds, used for sending/receiving control messages
  * they typically complete within a few frames (msec) after they're issued
-- 
cgit v1.2.1


From 42d8a2d22bbaed80f1ade65a08e4e1097a111d4b Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sun, 20 Aug 2006 18:58:42 +0400
Subject: USB: Turn usb_resume_both() into static inline

drivers/usb/core/hub.c: In function `hub_events':
drivers/usb/core/hub.c:2591: warning: statement with no effect

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/usb.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 67da6d0b316f..5162cb370215 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -40,7 +40,10 @@ extern int usb_port_resume(struct usb_device *dev);
 #else
 
 #define usb_suspend_both(udev, msg)	0
-#define usb_resume_both(udev)		0
+static inline int usb_resume_both(struct usb_device *udev)
+{
+	return 0;
+}
 #define usb_port_suspend(dev)		0
 #define usb_port_resume(dev)		0
 
-- 
cgit v1.2.1


From 511366da534bad226e89d294c3b3e910a2aaba6b Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 14 Aug 2006 23:11:02 -0700
Subject: USB: usb-hub-driver-improve-use-of-ifdef fix

Fix CONFIG_PM=n build.

Cc: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 65720f2bffc8..f5adce049b35 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1895,6 +1895,8 @@ static inline int remote_wakeup(struct usb_device *udev)
 	return 0;
 }
 
+#define hub_suspend NULL
+#define hub_resume NULL
 #endif
 
 void usb_suspend_root_hub(struct usb_device *hdev)
-- 
cgit v1.2.1


From a6d2bb9ff919b4685bd684620ec7a1ffa8bf2349 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:27:36 -0400
Subject: USB: remove struct usb_operations

All of the currently-supported USB host controller drivers use the HCD
bus-glue framework.  As part of the program for flattening out the glue
layer, this patch (as769) removes the usb_operations structure.  All
function calls now go directly to the HCD routines (slightly renamed
to remain within the "usb_" namespace).

The patch also removes usb_alloc_bus(), because it's not useful in the
HCD framework and it wasn't referenced anywhere.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c     | 49 +++++-----------------------------------------
 drivers/usb/core/hcd.h     | 32 +++++++-----------------------
 drivers/usb/core/message.c |  4 ++--
 drivers/usb/core/urb.c     | 13 +++++-------
 drivers/usb/core/usb.c     | 10 +++++-----
 include/linux/usb.h        |  4 ----
 6 files changed, 24 insertions(+), 88 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index ea20a3a5a9b9..2102c4deec1e 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -731,30 +731,6 @@ static void usb_bus_init (struct usb_bus *bus)
 	kref_init(&bus->kref);
 }
 
-/**
- * usb_alloc_bus - creates a new USB host controller structure
- * @op: pointer to a struct usb_operations that this bus structure should use
- * Context: !in_interrupt()
- *
- * Creates a USB host controller bus structure with the specified 
- * usb_operations and initializes all the necessary internal objects.
- *
- * If no memory is available, NULL is returned.
- *
- * The caller should call usb_put_bus() when it is finished with the structure.
- */
-struct usb_bus *usb_alloc_bus (struct usb_operations *op)
-{
-	struct usb_bus *bus;
-
-	bus = kzalloc (sizeof *bus, GFP_KERNEL);
-	if (!bus)
-		return NULL;
-	usb_bus_init (bus);
-	bus->op = op;
-	return bus;
-}
-
 /*-------------------------------------------------------------------------*/
 
 /**
@@ -1102,7 +1078,7 @@ static void urb_unlink (struct urb *urb)
  * expects usb_submit_urb() to have sanity checked and conditioned all
  * inputs in the urb
  */
-static int hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
+int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
 {
 	int			status;
 	struct usb_hcd		*hcd = urb->dev->bus->hcpriv;
@@ -1211,7 +1187,7 @@ done:
 /*-------------------------------------------------------------------------*/
 
 /* called in any context */
-static int hcd_get_frame_number (struct usb_device *udev)
+int usb_hcd_get_frame_number (struct usb_device *udev)
 {
 	struct usb_hcd	*hcd = (struct usb_hcd *)udev->bus->hcpriv;
 	if (!HC_IS_RUNNING (hcd->state))
@@ -1253,7 +1229,7 @@ unlink1 (struct usb_hcd *hcd, struct urb *urb)
  * caller guarantees urb won't be recycled till both unlink()
  * and the urb's completion function return
  */
-static int hcd_unlink_urb (struct urb *urb, int status)
+int usb_hcd_unlink_urb (struct urb *urb, int status)
 {
 	struct usb_host_endpoint	*ep;
 	struct usb_hcd			*hcd = NULL;
@@ -1351,8 +1327,8 @@ done:
  * example:  a qh stored in ep->hcpriv, holding state related to endpoint
  * type, maxpacket size, toggle, halt status, and scheduling.
  */
-static void
-hcd_endpoint_disable (struct usb_device *udev, struct usb_host_endpoint *ep)
+void usb_hcd_endpoint_disable (struct usb_device *udev,
+		struct usb_host_endpoint *ep)
 {
 	struct usb_hcd		*hcd;
 	struct urb		*urb;
@@ -1589,20 +1565,6 @@ EXPORT_SYMBOL (usb_bus_start_enum);
 
 /*-------------------------------------------------------------------------*/
 
-/*
- * usb_hcd_operations - adapts usb_bus framework to HCD framework (bus glue)
- */
-static struct usb_operations usb_hcd_operations = {
-	.get_frame_number =	hcd_get_frame_number,
-	.submit_urb =		hcd_submit_urb,
-	.unlink_urb =		hcd_unlink_urb,
-	.buffer_alloc =		hcd_buffer_alloc,
-	.buffer_free =		hcd_buffer_free,
-	.disable =		hcd_endpoint_disable,
-};
-
-/*-------------------------------------------------------------------------*/
-
 /**
  * usb_hcd_giveback_urb - return URB from HCD to device driver
  * @hcd: host controller returning the URB
@@ -1744,7 +1706,6 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 	dev_set_drvdata(dev, hcd);
 
 	usb_bus_init(&hcd->self);
-	hcd->self.op = &usb_hcd_operations;
 	hcd->self.hcpriv = hcd;
 	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index fc71a08a1af4..83e229914797 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -139,28 +139,6 @@ struct hcd_timeout {	/* timeouts we allocate */
 
 /*-------------------------------------------------------------------------*/
 
-/*
- * FIXME usb_operations should vanish or become hc_driver,
- * when usb_bus and usb_hcd become the same thing.
- */
-
-struct usb_operations {
-	int (*get_frame_number) (struct usb_device *usb_dev);
-	int (*submit_urb) (struct urb *urb, gfp_t mem_flags);
-	int (*unlink_urb) (struct urb *urb, int status);
-
-	/* allocate dma-consistent buffer for URB_DMA_NOMAPPING */
-	void *(*buffer_alloc)(struct usb_bus *bus, size_t size,
-			gfp_t mem_flags,
-			dma_addr_t *dma);
-	void (*buffer_free)(struct usb_bus *bus, size_t size,
-			void *addr, dma_addr_t dma);
-
-	void (*disable)(struct usb_device *udev,
-			struct usb_host_endpoint *ep);
-};
-
-/* each driver provides one of these, and hardware init support */
 
 struct pt_regs;
 
@@ -222,7 +200,13 @@ struct hc_driver {
 		/* Needed only if port-change IRQs are level-triggered */
 };
 
-extern void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb, struct pt_regs *regs);
+extern int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags);
+extern int usb_hcd_unlink_urb (struct urb *urb, int status);
+extern void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb,
+		struct pt_regs *regs);
+extern void usb_hcd_endpoint_disable (struct usb_device *udev,
+		struct usb_host_endpoint *ep);
+extern int usb_hcd_get_frame_number (struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		struct device *dev, char *bus_name);
@@ -361,8 +345,6 @@ extern long usb_calc_bus_time (int speed, int is_input,
 
 /*-------------------------------------------------------------------------*/
 
-extern struct usb_bus *usb_alloc_bus (struct usb_operations *);
-
 extern void usb_set_device_state(struct usb_device *udev,
 		enum usb_device_state new_state);
 
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 333b22c68aa4..1580c81a0db7 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -984,8 +984,8 @@ void usb_disable_endpoint(struct usb_device *dev, unsigned int epaddr)
 		ep = dev->ep_in[epnum];
 		dev->ep_in[epnum] = NULL;
 	}
-	if (ep && dev->bus && dev->bus->op && dev->bus->op->disable)
-		dev->bus->op->disable(dev, ep);
+	if (ep && dev->bus)
+		usb_hcd_endpoint_disable(dev, ep);
 }
 
 /**
diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c
index 9864988377c7..576919927f53 100644
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -221,7 +221,6 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 {
 	int			pipe, temp, max;
 	struct usb_device	*dev;
-	struct usb_operations	*op;
 	int			is_out;
 
 	if (!urb || urb->hcpriv || !urb->complete)
@@ -233,8 +232,6 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 	if (dev->bus->controller->power.power_state.event != PM_EVENT_ON
 			|| dev->state == USB_STATE_SUSPENDED)
 		return -EHOSTUNREACH;
-	if (!(op = dev->bus->op) || !op->submit_urb)
-		return -ENODEV;
 
 	urb->status = -EINPROGRESS;
 	urb->actual_length = 0;
@@ -376,7 +373,7 @@ int usb_submit_urb(struct urb *urb, gfp_t mem_flags)
 		urb->interval = temp;
 	}
 
-	return op->submit_urb (urb, mem_flags);
+	return usb_hcd_submit_urb (urb, mem_flags);
 }
 
 /*-------------------------------------------------------------------*/
@@ -440,9 +437,9 @@ int usb_unlink_urb(struct urb *urb)
 {
 	if (!urb)
 		return -EINVAL;
-	if (!(urb->dev && urb->dev->bus && urb->dev->bus->op))
+	if (!(urb->dev && urb->dev->bus))
 		return -ENODEV;
-	return urb->dev->bus->op->unlink_urb(urb, -ECONNRESET);
+	return usb_hcd_unlink_urb(urb, -ECONNRESET);
 }
 
 /**
@@ -468,13 +465,13 @@ int usb_unlink_urb(struct urb *urb)
 void usb_kill_urb(struct urb *urb)
 {
 	might_sleep();
-	if (!(urb && urb->dev && urb->dev->bus && urb->dev->bus->op))
+	if (!(urb && urb->dev && urb->dev->bus))
 		return;
 	spin_lock_irq(&urb->lock);
 	++urb->reject;
 	spin_unlock_irq(&urb->lock);
 
-	urb->dev->bus->op->unlink_urb(urb, -ENOENT);
+	usb_hcd_unlink_urb(urb, -ENOENT);
 	wait_event(usb_kill_urb_queue, atomic_read(&urb->use_count) == 0);
 
 	spin_lock_irq(&urb->lock);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 4eb98eb3804f..7ab9d29215f8 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -480,7 +480,7 @@ exit:
  */
 int usb_get_current_frame_number(struct usb_device *dev)
 {
-	return dev->bus->op->get_frame_number (dev);
+	return usb_hcd_get_frame_number (dev);
 }
 
 /**
@@ -677,9 +677,9 @@ void *usb_buffer_alloc (
 	dma_addr_t *dma
 )
 {
-	if (!dev || !dev->bus || !dev->bus->op || !dev->bus->op->buffer_alloc)
+	if (!dev || !dev->bus)
 		return NULL;
-	return dev->bus->op->buffer_alloc (dev->bus, size, mem_flags, dma);
+	return hcd_buffer_alloc (dev->bus, size, mem_flags, dma);
 }
 
 /**
@@ -700,11 +700,11 @@ void usb_buffer_free (
 	dma_addr_t dma
 )
 {
-	if (!dev || !dev->bus || !dev->bus->op || !dev->bus->op->buffer_free)
+	if (!dev || !dev->bus)
 		return;
 	if (!addr)
 		return;
-	dev->bus->op->buffer_free (dev->bus, size, addr, dma);
+	hcd_buffer_free (dev->bus, size, addr, dma);
 }
 
 /**
diff --git a/include/linux/usb.h b/include/linux/usb.h
index f104efa04d79..4709033f8fa7 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -257,8 +257,6 @@ int __usb_get_extra_descriptor(char *buffer, unsigned size,
 
 /* ----------------------------------------------------------------------- */
 
-struct usb_operations;
-
 /* USB device number allocation bitmap */
 struct usb_devmap {
 	unsigned long devicemap[128 / (8*sizeof(unsigned long))];
@@ -279,7 +277,6 @@ struct usb_bus {
 					 * round-robin allocation */
 
 	struct usb_devmap devmap;	/* device address allocation map */
-	struct usb_operations *op;	/* Operations (specific to the HC) */
 	struct usb_device *root_hub;	/* Root hub */
 	struct list_head bus_list;	/* list of busses */
 	void *hcpriv;                   /* Host Controller private data */
@@ -1051,7 +1048,6 @@ extern int usb_submit_urb(struct urb *urb, gfp_t mem_flags);
 extern int usb_unlink_urb(struct urb *urb);
 extern void usb_kill_urb(struct urb *urb);
 
-#define HAVE_USB_BUFFERS
 void *usb_buffer_alloc (struct usb_device *dev, size_t size,
 	gfp_t mem_flags, dma_addr_t *dma);
 void usb_buffer_free (struct usb_device *dev, size_t size,
-- 
cgit v1.2.1


From dd990f16a39d4e615c0b70a0ab50b79b32bfb16d Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:29:56 -0400
Subject: usbcore: Add flag for whether a host controller uses DMA

This patch (as770b) introduces a new field to usb_bus: a flag
indicating whether or not the host controller uses DMA.  This serves
to encapsulate the computation.  It also means we will have only one
spot to update if the DMA API changes.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c | 8 +++++---
 include/linux/usb.h    | 1 +
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 2102c4deec1e..0cc14206920a 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1152,7 +1152,7 @@ doit:
 	/* lower level hcd code should use *_dma exclusively,
 	 * unless it uses pio or talks to another transport.
 	 */
-	if (hcd->self.controller->dma_mask) {
+	if (hcd->self.uses_dma) {
 		if (usb_pipecontrol (urb->pipe)
 			&& !(urb->transfer_flags & URB_NO_SETUP_DMA_MAP))
 			urb->setup_dma = dma_map_single (
@@ -1585,8 +1585,9 @@ void usb_hcd_giveback_urb (struct usb_hcd *hcd, struct urb *urb, struct pt_regs
 	at_root_hub = (urb->dev == hcd->self.root_hub);
 	urb_unlink (urb);
 
-	/* lower level hcd code should use *_dma exclusively */
-	if (hcd->self.controller->dma_mask && !at_root_hub) {
+	/* lower level hcd code should use *_dma exclusively if the
+	 * host controller does DMA */
+	if (hcd->self.uses_dma && !at_root_hub) {
 		if (usb_pipecontrol (urb->pipe)
 			&& !(urb->transfer_flags & URB_NO_SETUP_DMA_MAP))
 			dma_unmap_single (hcd->self.controller, urb->setup_dma,
@@ -1710,6 +1711,7 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
 	hcd->self.bus_name = bus_name;
+	hcd->self.uses_dma = (dev->dma_mask != NULL);
 
 	init_timer(&hcd->rh_timer);
 	hcd->rh_timer.function = rh_timer_func;
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 4709033f8fa7..09661759621f 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -269,6 +269,7 @@ struct usb_bus {
 	struct device *controller;	/* host/master side hardware */
 	int busnum;			/* Bus number (in order of reg) */
 	char *bus_name;			/* stable id (PCI slot_name etc) */
+	u8 uses_dma;			/* Does the host controller use DMA? */
 	u8 otg_port;			/* 0, or number of OTG/HNP port */
 	unsigned is_b_host:1;		/* true during some HNP roleswitches */
 	unsigned b_hnp_enable:1;	/* OTG: did A-Host enable HNP? */
-- 
cgit v1.2.1


From 1720058343fa43a1a25bfad9e62ea06e7e9743b6 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:32:52 -0400
Subject: usbcore: trim down usb_bus structure

As part of the ongoing program to flatten out the HCD bus-glue layer,
this patch (as771b) eliminates the hcpriv, release, and kref fields
from struct usb_bus.  hcpriv and release were not being used for
anything worthwhile, and kref has been moved into the enclosing
usb_hcd structure.

Along with those changes, the patch gets rid of usb_bus_get and
usb_bus_put, replacing them with usb_get_hcd and usb_put_hcd.

The one interesting aspect is that the dev_set_drvdata call was
removed from usb_put_hcd, where it clearly doesn't belong.  This means
the driver private data won't get reset to NULL.  It shouldn't cause
any problems, since the private data is undefined when no driver is
bound.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/buffer.c      |  4 +--
 drivers/usb/core/hcd.c         | 67 +++++++++++++++---------------------------
 drivers/usb/core/hcd.h         | 13 ++++----
 drivers/usb/core/usb.c         |  5 ++--
 drivers/usb/gadget/dummy_hcd.c |  8 ++---
 drivers/usb/host/ehci-dbg.c    |  6 ++--
 drivers/usb/host/ohci-dbg.c    |  6 ++--
 drivers/usb/mon/mon_main.c     |  6 ++--
 include/linux/usb.h            |  3 --
 9 files changed, 44 insertions(+), 74 deletions(-)

diff --git a/drivers/usb/core/buffer.c b/drivers/usb/core/buffer.c
index f4f4ef0f377a..840442a25b61 100644
--- a/drivers/usb/core/buffer.c
+++ b/drivers/usb/core/buffer.c
@@ -104,7 +104,7 @@ void *hcd_buffer_alloc (
 	dma_addr_t		*dma
 )
 {
-	struct usb_hcd		*hcd = bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(bus);
 	int 			i;
 
 	/* some USB hosts just use PIO */
@@ -127,7 +127,7 @@ void hcd_buffer_free (
 	dma_addr_t		dma
 )
 {
-	struct usb_hcd		*hcd = bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(bus);
 	int 			i;
 
 	if (!addr)
diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 0cc14206920a..9dfc812de034 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -664,31 +664,6 @@ static int usb_rh_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 
 /*-------------------------------------------------------------------------*/
 
-/* exported only within usbcore */
-struct usb_bus *usb_bus_get(struct usb_bus *bus)
-{
-	if (bus)
-		kref_get(&bus->kref);
-	return bus;
-}
-
-static void usb_host_release(struct kref *kref)
-{
-	struct usb_bus *bus = container_of(kref, struct usb_bus, kref);
-
-	if (bus->release)
-		bus->release(bus);
-}
-
-/* exported only within usbcore */
-void usb_bus_put(struct usb_bus *bus)
-{
-	if (bus)
-		kref_put(&bus->kref, usb_host_release);
-}
-
-/*-------------------------------------------------------------------------*/
-
 static struct class *usb_host_class;
 
 int usb_host_init(void)
@@ -720,15 +695,12 @@ static void usb_bus_init (struct usb_bus *bus)
 	bus->devnum_next = 1;
 
 	bus->root_hub = NULL;
-	bus->hcpriv = NULL;
 	bus->busnum = -1;
 	bus->bandwidth_allocated = 0;
 	bus->bandwidth_int_reqs  = 0;
 	bus->bandwidth_isoc_reqs = 0;
 
 	INIT_LIST_HEAD (&bus->bus_list);
-
-	kref_init(&bus->kref);
 }
 
 /*-------------------------------------------------------------------------*/
@@ -1081,7 +1053,7 @@ static void urb_unlink (struct urb *urb)
 int usb_hcd_submit_urb (struct urb *urb, gfp_t mem_flags)
 {
 	int			status;
-	struct usb_hcd		*hcd = urb->dev->bus->hcpriv;
+	struct usb_hcd		*hcd = bus_to_hcd(urb->dev->bus);
 	struct usb_host_endpoint *ep;
 	unsigned long		flags;
 
@@ -1189,7 +1161,8 @@ done:
 /* called in any context */
 int usb_hcd_get_frame_number (struct usb_device *udev)
 {
-	struct usb_hcd	*hcd = (struct usb_hcd *)udev->bus->hcpriv;
+	struct usb_hcd	*hcd = bus_to_hcd(udev->bus);
+
 	if (!HC_IS_RUNNING (hcd->state))
 		return -ESHUTDOWN;
 	return hcd->driver->get_frame_number (hcd);
@@ -1262,7 +1235,7 @@ int usb_hcd_unlink_urb (struct urb *urb, int status)
 	spin_lock (&hcd_data_lock);
 
 	sys = &urb->dev->dev;
-	hcd = urb->dev->bus->hcpriv;
+	hcd = bus_to_hcd(urb->dev->bus);
 	if (hcd == NULL) {
 		retval = -ENODEV;
 		goto done;
@@ -1333,7 +1306,7 @@ void usb_hcd_endpoint_disable (struct usb_device *udev,
 	struct usb_hcd		*hcd;
 	struct urb		*urb;
 
-	hcd = udev->bus->hcpriv;
+	hcd = bus_to_hcd(udev->bus);
 
 	WARN_ON (!HC_IS_RUNNING (hcd->state) && hcd->state != HC_STATE_HALT &&
 			udev->state != USB_STATE_NOTATTACHED);
@@ -1673,14 +1646,6 @@ EXPORT_SYMBOL_GPL (usb_hc_died);
 
 /*-------------------------------------------------------------------------*/
 
-static void hcd_release (struct usb_bus *bus)
-{
-	struct usb_hcd *hcd;
-
-	hcd = container_of(bus, struct usb_hcd, self);
-	kfree(hcd);
-}
-
 /**
  * usb_create_hcd - create and initialize an HCD structure
  * @driver: HC driver that will use this hcd
@@ -1705,10 +1670,9 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		return NULL;
 	}
 	dev_set_drvdata(dev, hcd);
+	kref_init(&hcd->kref);
 
 	usb_bus_init(&hcd->self);
-	hcd->self.hcpriv = hcd;
-	hcd->self.release = &hcd_release;
 	hcd->self.controller = dev;
 	hcd->self.bus_name = bus_name;
 	hcd->self.uses_dma = (dev->dma_mask != NULL);
@@ -1725,10 +1689,25 @@ struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 }
 EXPORT_SYMBOL (usb_create_hcd);
 
+static void hcd_release (struct kref *kref)
+{
+	struct usb_hcd *hcd = container_of (kref, struct usb_hcd, kref);
+
+	kfree(hcd);
+}
+
+struct usb_hcd *usb_get_hcd (struct usb_hcd *hcd)
+{
+	if (hcd)
+		kref_get (&hcd->kref);
+	return hcd;
+}
+EXPORT_SYMBOL (usb_get_hcd);
+
 void usb_put_hcd (struct usb_hcd *hcd)
 {
-	dev_set_drvdata(hcd->self.controller, NULL);
-	usb_bus_put(&hcd->self);
+	if (hcd)
+		kref_put (&hcd->kref, hcd_release);
 }
 EXPORT_SYMBOL (usb_put_hcd);
 
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 83e229914797..7a2bcba2ae61 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -55,12 +55,13 @@
 
 /*-------------------------------------------------------------------------*/
 
-struct usb_hcd {	/* usb_bus.hcpriv points to this */
+struct usb_hcd {
 
 	/*
 	 * housekeeping
 	 */
 	struct usb_bus		self;		/* hcd is-a bus */
+	struct kref		kref;		/* reference counter */
 
 	const char		*product_desc;	/* product/vendor string */
 	char			irq_descr[24];	/* driver + bus # */
@@ -129,8 +130,10 @@ static inline struct usb_bus *hcd_to_bus (struct usb_hcd *hcd)
 	return &hcd->self;
 }
 
-
-// urb.hcpriv is really hardware-specific
+static inline struct usb_hcd *bus_to_hcd (struct usb_bus *bus)
+{
+	return container_of(bus, struct usb_hcd, self);
+}
 
 struct hcd_timeout {	/* timeouts we allocate */
 	struct list_head	timeout_list;
@@ -210,6 +213,7 @@ extern int usb_hcd_get_frame_number (struct usb_device *udev);
 
 extern struct usb_hcd *usb_create_hcd (const struct hc_driver *driver,
 		struct device *dev, char *bus_name);
+extern struct usb_hcd *usb_get_hcd (struct usb_hcd *hcd);
 extern void usb_put_hcd (struct usb_hcd *hcd);
 extern int usb_add_hcd(struct usb_hcd *hcd,
 		unsigned int irqnum, unsigned long irqflags);
@@ -356,9 +360,6 @@ extern struct list_head usb_bus_list;
 extern struct mutex usb_bus_list_lock;
 extern wait_queue_head_t usb_kill_urb_queue;
 
-extern struct usb_bus *usb_bus_get (struct usb_bus *bus);
-extern void usb_bus_put (struct usb_bus *bus);
-
 extern void usb_enable_root_hub_irq (struct usb_bus *bus);
 
 extern int usb_find_interface_driver (struct usb_device *dev,
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 7ab9d29215f8..b0c0a993338f 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -169,7 +169,7 @@ static void usb_release_dev(struct device *dev)
 	udev = to_usb_device(dev);
 
 	usb_destroy_configuration(udev);
-	usb_bus_put(udev->bus);
+	usb_put_hcd(bus_to_hcd(udev->bus));
 	kfree(udev->product);
 	kfree(udev->manufacturer);
 	kfree(udev->serial);
@@ -197,8 +197,7 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	if (!dev)
 		return NULL;
 
-	bus = usb_bus_get(bus);
-	if (!bus) {
+	if (!usb_get_hcd(bus_to_hcd(bus))) {
 		kfree(dev);
 		return NULL;
 	}
diff --git a/drivers/usb/gadget/dummy_hcd.c b/drivers/usb/gadget/dummy_hcd.c
index 7d1c22c34957..fdab97a27c08 100644
--- a/drivers/usb/gadget/dummy_hcd.c
+++ b/drivers/usb/gadget/dummy_hcd.c
@@ -889,11 +889,9 @@ EXPORT_SYMBOL (net2280_set_fifo_mode);
 static void
 dummy_gadget_release (struct device *dev)
 {
-#if 0		/* usb_bus_put isn't EXPORTed! */
 	struct dummy	*dum = gadget_dev_to_dummy (dev);
 
-	usb_bus_put (&dummy_to_hcd (dum)->self);
-#endif
+	usb_put_hcd (dummy_to_hcd (dum));
 }
 
 static int dummy_udc_probe (struct platform_device *pdev)
@@ -915,9 +913,7 @@ static int dummy_udc_probe (struct platform_device *pdev)
 	if (rc < 0)
 		return rc;
 
-#if 0		/* usb_bus_get isn't EXPORTed! */
-	usb_bus_get (&dummy_to_hcd (dum)->self);
-#endif
+	usb_get_hcd (dummy_to_hcd (dum));
 
 	platform_set_drvdata (pdev, dum);
 	device_create_file (&dum->gadget.dev, &dev_attr_function);
diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 65ac9fef3a7c..215ce6d06394 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -451,7 +451,7 @@ show_async (struct class_device *class_dev, char *buf)
 	*buf = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -497,7 +497,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 	seen_count = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -634,7 +634,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	static char		label [] = "";
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ehci = hcd_to_ehci (hcd);
 	next = buf;
 	size = PAGE_SIZE;
diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c
index da52609a9290..534d07dcb824 100644
--- a/drivers/usb/host/ohci-dbg.c
+++ b/drivers/usb/host/ohci-dbg.c
@@ -477,7 +477,7 @@ show_async (struct class_device *class_dev, char *buf)
 	unsigned long		flags;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 
 	/* display control and bulk lists together, for simplicity */
@@ -510,7 +510,7 @@ show_periodic (struct class_device *class_dev, char *buf)
 	seen_count = 0;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 	next = buf;
 	size = PAGE_SIZE;
@@ -607,7 +607,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	u32			rdata;
 
 	bus = class_get_devdata(class_dev);
-	hcd = bus->hcpriv;
+	hcd = bus_to_hcd(bus);
 	ohci = hcd_to_ohci(hcd);
 	regs = ohci->regs;
 	next = buf;
diff --git a/drivers/usb/mon/mon_main.c b/drivers/usb/mon/mon_main.c
index 275a66f83058..e0ed36cdfd8b 100644
--- a/drivers/usb/mon/mon_main.c
+++ b/drivers/usb/mon/mon_main.c
@@ -265,7 +265,6 @@ static void mon_dissolve(struct mon_bus *mbus, struct usb_bus *ubus)
 	ubus->mon_bus = NULL;
 	mbus->u_bus = NULL;
 	mb();
-	// usb_bus_put(ubus);
 }
 
 /*
@@ -297,10 +296,9 @@ static void mon_bus_init(struct dentry *mondir, struct usb_bus *ubus)
 	INIT_LIST_HEAD(&mbus->r_list);
 
 	/*
-	 * This usb_bus_get here is superfluous, because we receive
-	 * a notification if usb_bus is about to be removed.
+	 * We don't need to take a reference to ubus, because we receive
+	 * a notification if the bus is about to be removed.
 	 */
-	// usb_bus_get(ubus);
 	mbus->u_bus = ubus;
 	ubus->mon_bus = mbus;
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index 09661759621f..c66303285a45 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -280,7 +280,6 @@ struct usb_bus {
 	struct usb_devmap devmap;	/* device address allocation map */
 	struct usb_device *root_hub;	/* Root hub */
 	struct list_head bus_list;	/* list of busses */
-	void *hcpriv;                   /* Host Controller private data */
 
 	int bandwidth_allocated;	/* on this bus: how much of the time
 					 * reserved for periodic (intr/iso)
@@ -295,8 +294,6 @@ struct usb_bus {
 	struct dentry *usbfs_dentry;	/* usbfs dentry entry for the bus */
 
 	struct class_device *class_dev;	/* class device for this bus */
-	struct kref kref;		/* reference counting for this bus */
-	void (*release)(struct usb_bus *bus);
 
 #if defined(CONFIG_USB_MON)
 	struct mon_bus *mon_bus;	/* non-null when associated */
-- 
cgit v1.2.1


From 4d6cd48380c4d361cc8ec34359df22377d85d202 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 11:35:21 -0400
Subject: usbmon: don't call mon_dmapeek if DMA isn't being used

This patch (as755b) fixes a bug in usbmon.  Rather than assuming all
USB host controllers use DMA, the code will check the usb_bus data
structure.  If DMA isn't used, we don't want to try peeking into a
non-existent DMA buffer!

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/mon/mon_main.c |  1 +
 drivers/usb/mon/mon_text.c | 13 +++++++------
 drivers/usb/mon/usb_mon.h  |  1 +
 3 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/mon/mon_main.c b/drivers/usb/mon/mon_main.c
index e0ed36cdfd8b..394bbf2f68d4 100644
--- a/drivers/usb/mon/mon_main.c
+++ b/drivers/usb/mon/mon_main.c
@@ -301,6 +301,7 @@ static void mon_bus_init(struct dentry *mondir, struct usb_bus *ubus)
 	 */
 	mbus->u_bus = ubus;
 	ubus->mon_bus = mbus;
+	mbus->uses_dma = ubus->uses_dma;
 
 	rc = snprintf(name, NAMESZ, "%dt", ubus->busnum);
 	if (rc <= 0 || rc >= NAMESZ)
diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c
index 775e374a960c..7a2346c53284 100644
--- a/drivers/usb/mon/mon_text.c
+++ b/drivers/usb/mon/mon_text.c
@@ -75,13 +75,13 @@ static void mon_text_ctor(void *, kmem_cache_t *, unsigned long);
  */
 
 static inline char mon_text_get_setup(struct mon_event_text *ep,
-    struct urb *urb, char ev_type)
+    struct urb *urb, char ev_type, struct mon_bus *mbus)
 {
 
 	if (!usb_pipecontrol(urb->pipe) || ev_type != 'S')
 		return '-';
 
-	if (urb->transfer_flags & URB_NO_SETUP_DMA_MAP)
+	if (mbus->uses_dma && (urb->transfer_flags & URB_NO_SETUP_DMA_MAP))
 		return mon_dmapeek(ep->setup, urb->setup_dma, SETUP_MAX);
 	if (urb->setup_packet == NULL)
 		return 'Z';	/* '0' would be not as pretty. */
@@ -91,7 +91,7 @@ static inline char mon_text_get_setup(struct mon_event_text *ep,
 }
 
 static inline char mon_text_get_data(struct mon_event_text *ep, struct urb *urb,
-    int len, char ev_type)
+    int len, char ev_type, struct mon_bus *mbus)
 {
 	int pipe = urb->pipe;
 
@@ -117,7 +117,7 @@ static inline char mon_text_get_data(struct mon_event_text *ep, struct urb *urb,
 	 * contain non-NULL garbage in case the upper level promised to
 	 * set DMA for the HCD.
 	 */
-	if (urb->transfer_flags & URB_NO_TRANSFER_DMA_MAP)
+	if (mbus->uses_dma && (urb->transfer_flags & URB_NO_TRANSFER_DMA_MAP))
 		return mon_dmapeek(ep->data, urb->transfer_dma, len);
 
 	if (urb->transfer_buffer == NULL)
@@ -161,8 +161,9 @@ static void mon_text_event(struct mon_reader_text *rp, struct urb *urb,
 	/* Collecting status makes debugging sense for submits, too */
 	ep->status = urb->status;
 
-	ep->setup_flag = mon_text_get_setup(ep, urb, ev_type);
-	ep->data_flag = mon_text_get_data(ep, urb, ep->length, ev_type);
+	ep->setup_flag = mon_text_get_setup(ep, urb, ev_type, rp->r.m_bus);
+	ep->data_flag = mon_text_get_data(ep, urb, ep->length, ev_type,
+			rp->r.m_bus);
 
 	rp->nevents++;
 	list_add_tail(&ep->e_link, &rp->e_list);
diff --git a/drivers/usb/mon/usb_mon.h b/drivers/usb/mon/usb_mon.h
index 50efc8ee7c21..ab9d02d5df77 100644
--- a/drivers/usb/mon/usb_mon.h
+++ b/drivers/usb/mon/usb_mon.h
@@ -20,6 +20,7 @@ struct mon_bus {
 	struct dentry *dent_s;		/* Debugging file */
 	struct dentry *dent_t;		/* Text interface file */
 	struct usb_bus *u_bus;
+	int uses_dma;
 
 	/* Ref */
 	int nreaders;			/* Under mon_lock AND mbus->lock */
-- 
cgit v1.2.1


From 729ed6d502b45fd3b5c3b21c3ceaa63a8fe7cc43 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 13:24:56 -0700
Subject: USB: ethernet gadget avoids zlps for musb_hdrc

For systems using the Mentor HDRC controllers we get better TX DMA throughput
if we can avoid falling back to PIO to write zero length packets ... so tell
the driver to avoid ZLPs.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/ether.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index 1cc3c0227d10..366dc0a9e52c 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -2230,6 +2230,9 @@ eth_bind (struct usb_gadget *gadget)
 	if (gadget_is_pxa (gadget)) {
 		/* pxa doesn't support altsettings */
 		cdc = 0;
+	} else if (gadget_is_musbhdrc(gadget)) {
+		/* reduce tx dma overhead by avoiding special cases */
+		zlp = 0;
 	} else if (gadget_is_sh(gadget)) {
 		/* sh doesn't support multiple interfaces or configs */
 		cdc = 0;
-- 
cgit v1.2.1


From 53bd6a601a87bb6d0df844872bc15fd4e8d127ce Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 30 Aug 2006 14:50:06 -0700
Subject: USB: EHCI whitespace fixes (cosmetic)

[ ... when you have an editor set to remind you of whitespace bugs ... ]

Cosmetic EHCI changes: remove end-of-line whitespace, spaces before tabs.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ehci-dbg.c   | 14 +++++++-------
 drivers/usb/host/ehci-hcd.c   |  6 +++---
 drivers/usb/host/ehci-hub.c   | 12 ++++++------
 drivers/usb/host/ehci-mem.c   | 14 +++++++-------
 drivers/usb/host/ehci-q.c     | 20 ++++++++++----------
 drivers/usb/host/ehci-sched.c | 26 +++++++++++++-------------
 drivers/usb/host/ehci.h       | 36 ++++++++++++++++++------------------
 7 files changed, 64 insertions(+), 64 deletions(-)

diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 215ce6d06394..9cd6270d06bc 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2001-2002 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -65,7 +65,7 @@ static void dbg_hcs_params (struct ehci_hcd *ehci, char *label)
 		for (i = 0; i < HCS_N_PORTS (params); i++) {
 			// FIXME MIPS won't readb() ...
 			byte = readb (&ehci->caps->portroute[(i>>1)]);
-			sprintf(tmp, "%d ", 
+			sprintf(tmp, "%d ",
 				((i & 0x1) ? ((byte)&0xf) : ((byte>>4)&0xf)));
 			strcat(buf, tmp);
 		}
@@ -141,12 +141,12 @@ dbg_qh (const char *label, struct ehci_hcd *ehci, struct ehci_qh *qh)
 }
 
 static void __attribute__((__unused__))
-dbg_itd (const char *label, struct ehci_hcd *ehci, struct ehci_itd *itd) 
+dbg_itd (const char *label, struct ehci_hcd *ehci, struct ehci_itd *itd)
 {
 	ehci_dbg (ehci, "%s [%d] itd %p, next %08x, urb %p\n",
 		label, itd->frame, itd, le32_to_cpu(itd->hw_next), itd->urb);
 	ehci_dbg (ehci,
-		"  trans: %08x %08x %08x %08x %08x %08x %08x %08x\n", 
+		"  trans: %08x %08x %08x %08x %08x %08x %08x %08x\n",
 		le32_to_cpu(itd->hw_transaction[0]),
 		le32_to_cpu(itd->hw_transaction[1]),
 		le32_to_cpu(itd->hw_transaction[2]),
@@ -156,7 +156,7 @@ dbg_itd (const char *label, struct ehci_hcd *ehci, struct ehci_itd *itd)
 		le32_to_cpu(itd->hw_transaction[6]),
 		le32_to_cpu(itd->hw_transaction[7]));
 	ehci_dbg (ehci,
-		"  buf:   %08x %08x %08x %08x %08x %08x %08x\n", 
+		"  buf:   %08x %08x %08x %08x %08x %08x %08x\n",
 		le32_to_cpu(itd->hw_bufp[0]),
 		le32_to_cpu(itd->hw_bufp[1]),
 		le32_to_cpu(itd->hw_bufp[2]),
@@ -171,12 +171,12 @@ dbg_itd (const char *label, struct ehci_hcd *ehci, struct ehci_itd *itd)
 }
 
 static void __attribute__((__unused__))
-dbg_sitd (const char *label, struct ehci_hcd *ehci, struct ehci_sitd *sitd) 
+dbg_sitd (const char *label, struct ehci_hcd *ehci, struct ehci_sitd *sitd)
 {
 	ehci_dbg (ehci, "%s [%d] sitd %p, next %08x, urb %p\n",
 		label, sitd->frame, sitd, le32_to_cpu(sitd->hw_next), sitd->urb);
 	ehci_dbg (ehci,
-		"  addr %08x sched %04x result %08x buf %08x %08x\n", 
+		"  addr %08x sched %04x result %08x buf %08x %08x\n",
 		le32_to_cpu(sitd->hw_fullspeed_ep),
 		le32_to_cpu(sitd->hw_uframe),
 		le32_to_cpu(sitd->hw_results),
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index 1c54b303e5fc..4e1a8c308893 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2000-2004 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -70,7 +70,7 @@
  * 2002-08-06	Handling for bulk and interrupt transfers is mostly shared;
  *	only scheduling is different, no arbitrary limitations.
  * 2002-07-25	Sanity check PCI reads, mostly for better cardbus support,
- * 	clean up HC run state handshaking.
+ *	clean up HC run state handshaking.
  * 2002-05-24	Preliminary FS/LS interrupts, using scheduling shortcuts
  * 2002-05-11	Clear TT errors for FS/LS ctrl/bulk.  Fill in some other
  *	missing pieces:  enabling 64bit dma, handoff from BIOS/SMM.
@@ -425,7 +425,7 @@ static int ehci_init(struct usb_hcd *hcd)
 
 	/* controllers may cache some of the periodic schedule ... */
 	hcc_params = readl(&ehci->caps->hcc_params);
-	if (HCC_ISOC_CACHE(hcc_params)) 	// full frame cache
+	if (HCC_ISOC_CACHE(hcc_params))		// full frame cache
 		ehci->i_thresh = 8;
 	else					// N microframes cached
 		ehci->i_thresh = 2 + HCC_ISOC_THRES(hcc_params);
diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c
index d03e3cad5ca8..a5eeb9cd6ab2 100644
--- a/drivers/usb/host/ehci-hub.c
+++ b/drivers/usb/host/ehci-hub.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (C) 2001-2004 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -103,10 +103,10 @@ static int ehci_bus_resume (struct usb_hcd *hcd)
 
 	/* re-init operational registers in case we lost power */
 	if (readl (&ehci->regs->intr_enable) == 0) {
- 		/* at least some APM implementations will try to deliver
+		/* at least some APM implementations will try to deliver
 		 * IRQs right away, so delay them until we're ready.
- 		 */
- 		intr_enable = 1;
+		 */
+		intr_enable = 1;
 		writel (0, &ehci->regs->segment);
 		writel (ehci->periodic_dma, &ehci->regs->frame_list);
 		writel ((u32)ehci->async->qh_dma, &ehci->regs->async_next);
@@ -232,7 +232,7 @@ ehci_hub_status_data (struct usb_hcd *hcd, char *buf)
 		buf [1] = 0;
 		retval++;
 	}
-	
+
 	/* no hub change reports (bit 0) for now (power, ...) */
 
 	/* port N changes (bit N)? */
@@ -304,7 +304,7 @@ ehci_hub_descriptor (
 
 /*-------------------------------------------------------------------------*/
 
-#define	PORT_WAKE_BITS 	(PORT_WKOC_E|PORT_WKDISC_E|PORT_WKCONN_E)
+#define	PORT_WAKE_BITS	(PORT_WKOC_E|PORT_WKDISC_E|PORT_WKCONN_E)
 
 static int ehci_hub_control (
 	struct usb_hcd	*hcd,
diff --git a/drivers/usb/host/ehci-mem.c b/drivers/usb/host/ehci-mem.c
index 766061e0260a..a8ba2e1497a4 100644
--- a/drivers/usb/host/ehci-mem.c
+++ b/drivers/usb/host/ehci-mem.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2001 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -25,7 +25,7 @@
  *	- data used only by the HCD ... kmalloc is fine
  *	- async and periodic schedules, shared by HC and HCD ... these
  *	  need to use dma_pool or dma_alloc_coherent
- *	- driver buffers, read/written by HC ... single shot DMA mapped 
+ *	- driver buffers, read/written by HC ... single shot DMA mapped
  *
  * There's also PCI "register" data, which is memory mapped.
  * No memory seen by this driver is pageable.
@@ -119,7 +119,7 @@ static inline void qh_put (struct ehci_qh *qh)
 
 /*-------------------------------------------------------------------------*/
 
-/* The queue heads and transfer descriptors are managed from pools tied 
+/* The queue heads and transfer descriptors are managed from pools tied
  * to each of the "per device" structures.
  * This is the initialisation and cleanup code.
  */
@@ -165,7 +165,7 @@ static int ehci_mem_init (struct ehci_hcd *ehci, gfp_t flags)
 	int i;
 
 	/* QTDs for control/bulk/intr transfers */
-	ehci->qtd_pool = dma_pool_create ("ehci_qtd", 
+	ehci->qtd_pool = dma_pool_create ("ehci_qtd",
 			ehci_to_hcd(ehci)->self.controller,
 			sizeof (struct ehci_qtd),
 			32 /* byte alignment (for hw parts) */,
@@ -175,7 +175,7 @@ static int ehci_mem_init (struct ehci_hcd *ehci, gfp_t flags)
 	}
 
 	/* QHs for control/bulk/intr transfers */
-	ehci->qh_pool = dma_pool_create ("ehci_qh", 
+	ehci->qh_pool = dma_pool_create ("ehci_qh",
 			ehci_to_hcd(ehci)->self.controller,
 			sizeof (struct ehci_qh),
 			32 /* byte alignment (for hw parts) */,
@@ -189,7 +189,7 @@ static int ehci_mem_init (struct ehci_hcd *ehci, gfp_t flags)
 	}
 
 	/* ITD for high speed ISO transfers */
-	ehci->itd_pool = dma_pool_create ("ehci_itd", 
+	ehci->itd_pool = dma_pool_create ("ehci_itd",
 			ehci_to_hcd(ehci)->self.controller,
 			sizeof (struct ehci_itd),
 			32 /* byte alignment (for hw parts) */,
@@ -199,7 +199,7 @@ static int ehci_mem_init (struct ehci_hcd *ehci, gfp_t flags)
 	}
 
 	/* SITD for full/low speed split ISO transfers */
-	ehci->sitd_pool = dma_pool_create ("ehci_sitd", 
+	ehci->sitd_pool = dma_pool_create ("ehci_sitd",
 			ehci_to_hcd(ehci)->self.controller,
 			sizeof (struct ehci_sitd),
 			32 /* byte alignment (for hw parts) */,
diff --git a/drivers/usb/host/ehci-q.c b/drivers/usb/host/ehci-q.c
index e469221e7ec3..c0da40bbfa35 100644
--- a/drivers/usb/host/ehci-q.c
+++ b/drivers/usb/host/ehci-q.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (C) 2001-2004 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -31,7 +31,7 @@
  * ISO traffic uses "ISO TD" (itd, and sitd) records, and (along with
  * interrupts) needs careful scheduling.  Performance improvements can be
  * an ongoing challenge.  That's in "ehci-sched.c".
- * 
+ *
  * USB 1.1 devices are handled (a) by "companion" OHCI or UHCI root hubs,
  * or otherwise through transaction translators (TTs) in USB 2.0 hubs using
  * (b) special fields in qh entries or (c) split iso entries.  TTs will
@@ -199,7 +199,7 @@ static void qtd_copy_status (
 				&& ((token & QTD_STS_MMF) != 0
 					|| QTD_CERR(token) == 0)
 				&& (!ehci_is_TDI(ehci)
-                	                || urb->dev->tt->hub !=
+			                || urb->dev->tt->hub !=
 					   ehci_to_hcd(ehci)->self.root_hub)) {
 #ifdef DEBUG
 			struct usb_device *tt = urb->dev->tt->hub;
@@ -364,7 +364,7 @@ qh_completions (struct ehci_hcd *ehci, struct ehci_qh *qh, struct pt_regs *regs)
 			 */
 			if (likely (urb->status == -EINPROGRESS))
 				continue;
-			
+
 			/* issue status after short control reads */
 			if (unlikely (do_status != 0)
 					&& QTD_PID (token) == 0 /* OUT */) {
@@ -388,7 +388,7 @@ halt:
 				wmb ();
 			}
 		}
- 
+
 		/* remove it from the queue */
 		spin_lock (&urb->lock);
 		qtd_copy_status (ehci, urb, qtd->length, token);
@@ -518,7 +518,7 @@ qh_urb_transaction (
 		/* for zero length DATA stages, STATUS is always IN */
 		if (len == 0)
 			token |= (1 /* "in" */ << 8);
-	} 
+	}
 
 	/*
 	 * data transfer stage:  buffer setup
@@ -759,7 +759,7 @@ qh_make (
 		}
 		break;
 	default:
- 		dbg ("bogus dev %p speed %d", urb->dev, urb->dev->speed);
+		dbg ("bogus dev %p speed %d", urb->dev, urb->dev->speed);
 done:
 		qh_put (qh);
 		return NULL;
@@ -972,7 +972,7 @@ static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs)
 	// qh->hw_next = cpu_to_le32 (qh->qh_dma);
 	qh->qh_state = QH_STATE_IDLE;
 	qh->qh_next.qh = NULL;
-	qh_put (qh);			// refcount from reclaim 
+	qh_put (qh);			// refcount from reclaim
 
 	/* other unlink(s) may be pending (in QH_STATE_UNLINK_WAIT) */
 	next = qh->reclaim;
@@ -1031,7 +1031,7 @@ static void start_unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 			timer_action_done (ehci, TIMER_ASYNC_OFF);
 		}
 		return;
-	} 
+	}
 
 	qh->qh_state = QH_STATE_UNLINK;
 	ehci->reclaim = qh = qh_get (qh);
@@ -1046,7 +1046,7 @@ static void start_unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 
 	if (unlikely (ehci_to_hcd(ehci)->state == HC_STATE_HALT)) {
 		/* if (unlikely (qh->reclaim != 0))
-		 * 	this will recurse, probably not much
+		 *	this will recurse, probably not much
 		 */
 		end_unlink_async (ehci, NULL);
 		return;
diff --git a/drivers/usb/host/ehci-sched.c b/drivers/usb/host/ehci-sched.c
index 4859900bd135..e5e9c653c907 100644
--- a/drivers/usb/host/ehci-sched.c
+++ b/drivers/usb/host/ehci-sched.c
@@ -1,7 +1,7 @@
 /*
  * Copyright (c) 2001-2004 by David Brownell
  * Copyright (c) 2003 Michal Sojka, for high-speed iso transfers
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -613,7 +613,7 @@ static void intr_deschedule (struct ehci_hcd *ehci, struct ehci_qh *qh)
 /*-------------------------------------------------------------------------*/
 
 static int check_period (
-	struct ehci_hcd *ehci, 
+	struct ehci_hcd *ehci,
 	unsigned	frame,
 	unsigned	uframe,
 	unsigned	period,
@@ -629,7 +629,7 @@ static int check_period (
 
 	/*
 	 * 80% periodic == 100 usec/uframe available
-	 * convert "usecs we need" to "max already claimed" 
+	 * convert "usecs we need" to "max already claimed"
 	 */
 	usecs = 100 - usecs;
 
@@ -659,14 +659,14 @@ static int check_period (
 }
 
 static int check_intr_schedule (
-	struct ehci_hcd		*ehci, 
+	struct ehci_hcd		*ehci,
 	unsigned		frame,
 	unsigned		uframe,
 	const struct ehci_qh	*qh,
 	__le32			*c_maskp
 )
 {
-    	int		retval = -ENOSPC;
+	int		retval = -ENOSPC;
 	u8		mask = 0;
 
 	if (qh->c_usecs && uframe >= 6)		/* FSTN territory? */
@@ -701,7 +701,7 @@ static int check_intr_schedule (
 	/* Make sure this tt's buffer is also available for CSPLITs.
 	 * We pessimize a bit; probably the typical full speed case
 	 * doesn't need the second CSPLIT.
-	 * 
+	 *
 	 * NOTE:  both SPLIT and CSPLIT could be checked in just
 	 * one smart pass...
 	 */
@@ -728,7 +728,7 @@ done:
  */
 static int qh_schedule (struct ehci_hcd *ehci, struct ehci_qh *qh)
 {
-	int 		status;
+	int		status;
 	unsigned	uframe;
 	__le32		c_mask;
 	unsigned	frame;		/* 0..(qh->period - 1), or NO_FRAME */
@@ -784,7 +784,7 @@ static int qh_schedule (struct ehci_hcd *ehci, struct ehci_qh *qh)
 		ehci_dbg (ehci, "reused qh %p schedule\n", qh);
 
 	/* stuff into the periodic schedule */
- 	status = qh_link_periodic (ehci, qh);
+	status = qh_link_periodic (ehci, qh);
 done:
 	return status;
 }
@@ -1681,7 +1681,7 @@ static int itd_submit (struct ehci_hcd *ehci, struct urb *urb,
 		status = -ESHUTDOWN;
 	else
 		status = iso_stream_schedule (ehci, urb, stream);
- 	if (likely (status == 0))
+	if (likely (status == 0))
 		itd_link_urb (ehci, urb, ehci->periodic_size << 3, stream);
 	spin_unlock_irqrestore (&ehci->lock, flags);
 
@@ -1738,7 +1738,7 @@ sitd_sched_init (
 		if (packet->buf1 != (buf & ~(u64)0x0fff))
 			packet->cross = 1;
 
-		/* OUT uses multiple start-splits */ 
+		/* OUT uses multiple start-splits */
 		if (stream->bEndpointAddress & USB_DIR_IN)
 			continue;
 		length = (length + 187) / 188;
@@ -1925,7 +1925,7 @@ sitd_link_urb (
 /*-------------------------------------------------------------------------*/
 
 #define	SITD_ERRS (SITD_STS_ERR | SITD_STS_DBE | SITD_STS_BABBLE \
-	       			| SITD_STS_XACT | SITD_STS_MMF)
+				| SITD_STS_XACT | SITD_STS_MMF)
 
 static unsigned
 sitd_complete (
@@ -2043,7 +2043,7 @@ static int sitd_submit (struct ehci_hcd *ehci, struct urb *urb,
 		status = -ESHUTDOWN;
 	else
 		status = iso_stream_schedule (ehci, urb, stream);
- 	if (status == 0)
+	if (status == 0)
 		sitd_link_urb (ehci, urb, ehci->periodic_size << 3, stream);
 	spin_unlock_irqrestore (&ehci->lock, flags);
 
@@ -2226,5 +2226,5 @@ restart:
 			now_uframe++;
 			now_uframe %= mod;
 		}
-	} 
+	}
 }
diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
index 1385ce2b3f0a..bbc3082a73d7 100644
--- a/drivers/usb/host/ehci.h
+++ b/drivers/usb/host/ehci.h
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2001-2002 by David Brownell
- * 
+ *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 of the License, or (at your
@@ -103,7 +103,7 @@ struct ehci_hcd {			/* one per controller */
 #endif
 };
 
-/* convert between an HCD pointer and the corresponding EHCI_HCD */ 
+/* convert between an HCD pointer and the corresponding EHCI_HCD */
 static inline struct ehci_hcd *hcd_to_ehci (struct usb_hcd *hcd)
 {
 	return (struct ehci_hcd *) (hcd->hcd_priv);
@@ -178,8 +178,8 @@ struct ehci_caps {
 #define HCS_INDICATOR(p)	((p)&(1 << 16))	/* true: has port indicators */
 #define HCS_N_CC(p)		(((p)>>12)&0xf)	/* bits 15:12, #companion HCs */
 #define HCS_N_PCC(p)		(((p)>>8)&0xf)	/* bits 11:8, ports per CC */
-#define HCS_PORTROUTED(p)	((p)&(1 << 7))	/* true: port routing */ 
-#define HCS_PPC(p)		((p)&(1 << 4))	/* true: port power control */ 
+#define HCS_PORTROUTED(p)	((p)&(1 << 7))	/* true: port routing */
+#define HCS_PPC(p)		((p)&(1 << 4))	/* true: port power control */
 #define HCS_N_PORTS(p)		(((p)>>0)&0xf)	/* bits 3:0, ports on HC */
 
 	u32		hcc_params;      /* HCCPARAMS - offset 0x8 */
@@ -204,7 +204,7 @@ struct ehci_regs {
 #define CMD_LRESET	(1<<7)		/* partial reset (no ports, etc) */
 #define CMD_IAAD	(1<<6)		/* "doorbell" interrupt async advance */
 #define CMD_ASE		(1<<5)		/* async schedule enable */
-#define CMD_PSE  	(1<<4)		/* periodic schedule enable */
+#define CMD_PSE		(1<<4)		/* periodic schedule enable */
 /* 3:2 is periodic frame list size */
 #define CMD_RESET	(1<<1)		/* reset HC not bus */
 #define CMD_RUN		(1<<0)		/* start/stop HC */
@@ -230,9 +230,9 @@ struct ehci_regs {
 	/* FRINDEX: offset 0x0C */
 	u32		frame_index;	/* current microframe number */
 	/* CTRLDSSEGMENT: offset 0x10 */
-	u32		segment; 	/* address bits 63:32 if needed */
+	u32		segment;	/* address bits 63:32 if needed */
 	/* PERIODICLISTBASE: offset 0x14 */
-	u32		frame_list; 	/* points to periodic list */
+	u32		frame_list;	/* points to periodic list */
 	/* ASYNCLISTADDR: offset 0x18 */
 	u32		async_next;	/* address of next async queue head */
 
@@ -301,7 +301,7 @@ struct ehci_dbg_port {
 
 /*
  * EHCI Specification 0.95 Section 3.5
- * QTD: describe data transfer components (buffer, direction, ...) 
+ * QTD: describe data transfer components (buffer, direction, ...)
  * See Fig 3-6 "Queue Element Transfer Descriptor Block Diagram".
  *
  * These are associated only with "QH" (Queue Head) structures,
@@ -311,7 +311,7 @@ struct ehci_qtd {
 	/* first part defined by EHCI spec */
 	__le32			hw_next;	  /* see EHCI 3.5.1 */
 	__le32			hw_alt_next;      /* see EHCI 3.5.2 */
-	__le32			hw_token;         /* see EHCI 3.5.3 */       
+	__le32			hw_token;         /* see EHCI 3.5.3 */
 #define	QTD_TOGGLE	(1 << 31)	/* data toggle */
 #define	QTD_LENGTH(tok)	(((tok)>>16) & 0x7fff)
 #define	QTD_IOC		(1 << 15)	/* interrupt on complete */
@@ -348,8 +348,8 @@ struct ehci_qtd {
 /* values for that type tag */
 #define Q_TYPE_ITD	__constant_cpu_to_le32 (0 << 1)
 #define Q_TYPE_QH	__constant_cpu_to_le32 (1 << 1)
-#define Q_TYPE_SITD 	__constant_cpu_to_le32 (2 << 1)
-#define Q_TYPE_FSTN 	__constant_cpu_to_le32 (3 << 1)
+#define Q_TYPE_SITD	__constant_cpu_to_le32 (2 << 1)
+#define Q_TYPE_FSTN	__constant_cpu_to_le32 (3 << 1)
 
 /* next async queue entry, or pointer to interrupt/periodic QH */
 #define	QH_NEXT(dma)	(cpu_to_le32(((u32)dma)&~0x01f)|Q_TYPE_QH)
@@ -366,7 +366,7 @@ struct ehci_qtd {
  * For entries in the async schedule, the type tag always says "qh".
  */
 union ehci_shadow {
-	struct ehci_qh 		*qh;		/* Q_TYPE_QH */
+	struct ehci_qh		*qh;		/* Q_TYPE_QH */
 	struct ehci_itd		*itd;		/* Q_TYPE_ITD */
 	struct ehci_sitd	*sitd;		/* Q_TYPE_SITD */
 	struct ehci_fstn	*fstn;		/* Q_TYPE_FSTN */
@@ -396,7 +396,7 @@ struct ehci_qh {
 #define	QH_HUBPORT	0x3f800000
 #define	QH_MULT		0xc0000000
 	__le32			hw_current;	 /* qtd list - see EHCI 3.6.4 */
-	
+
 	/* qtd overlay (hardware parts of a struct ehci_qtd) */
 	__le32			hw_qtd_next;
 	__le32			hw_alt_next;
@@ -471,7 +471,7 @@ struct ehci_iso_stream {
 	struct list_head	td_list;	/* queued itds/sitds */
 	struct list_head	free_list;	/* list of unused itds/sitds */
 	struct usb_device	*udev;
- 	struct usb_host_endpoint *ep;
+	struct usb_host_endpoint *ep;
 
 	/* output of (re)scheduling */
 	unsigned long		start;		/* jiffies */
@@ -491,8 +491,8 @@ struct ehci_iso_stream {
 	unsigned		bandwidth;
 
 	/* This is used to initialize iTD's hw_bufp fields */
-	__le32			buf0;		
-	__le32			buf1;		
+	__le32			buf0;
+	__le32			buf1;
 	__le32			buf2;
 
 	/* this is used to initialize sITD's tt info */
@@ -520,7 +520,7 @@ struct ehci_itd {
 
 #define ITD_ACTIVE	__constant_cpu_to_le32(EHCI_ISOC_ACTIVE)
 
-	__le32			hw_bufp [7];	/* see EHCI 3.3.3 */ 
+	__le32			hw_bufp [7];	/* see EHCI 3.3.3 */
 	__le32			hw_bufp_hi [7];	/* Appendix B */
 
 	/* the rest is HCD-private */
@@ -541,7 +541,7 @@ struct ehci_itd {
 /*-------------------------------------------------------------------------*/
 
 /*
- * EHCI Specification 0.95 Section 3.4 
+ * EHCI Specification 0.95 Section 3.4
  * siTD, aka split-transaction isochronous Transfer Descriptor
  *       ... describe full speed iso xfers through TT in hubs
  * see Figure 3-5 "Split-transaction Isochronous Transaction Descriptor (siTD)
-- 
cgit v1.2.1


From 997694defd085f4dd168c6e7e0e82382c5be9db4 Mon Sep 17 00:00:00 2001
From: Skip Hansen <s-hansen@cox.net>
Date: Fri, 1 Sep 2006 15:26:27 -0700
Subject: gadgetfs patch for ep0out

For ep0out transfers (rare), be sure to copy the right data to userspace.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/inode.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index a38b6af2c358..4655522a08d9 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -1015,7 +1015,7 @@ ep0_read (struct file *fd, char __user *buf, size_t len, loff_t *ptr)
 			else {
 				len = min (len, (size_t)dev->req->actual);
 // FIXME don't call this with the spinlock held ...
-				if (copy_to_user (buf, &dev->req->buf, len))
+				if (copy_to_user (buf, dev->req->buf, len))
 					retval = -EFAULT;
 				clean_req (dev->gadget->ep0, dev->req);
 				/* NOTE userspace can't yet choose to stall */
-- 
cgit v1.2.1


From 0e3c8c26c7013b9d34929857598fd86ff1c22a6c Mon Sep 17 00:00:00 2001
From: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Date: Thu, 31 Aug 2006 13:37:29 -0700
Subject: USB: replace kernel_thread() with kthread_run() in libusual.c

Replaced kernel_thread() with kthread_run() since kernel_thread() is
deprecated in drivers/modules.

Signed-off-by: Cedric Le Goater <clg@fr.ibm.com>
Cc: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/libusual.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/usb/storage/libusual.c b/drivers/usb/storage/libusual.c
index b1ec4a718547..599ad10a761b 100644
--- a/drivers/usb/storage/libusual.c
+++ b/drivers/usb/storage/libusual.c
@@ -8,6 +8,7 @@
 #include <linux/usb.h>
 #include <linux/usb_usual.h>
 #include <linux/vmalloc.h>
+#include <linux/kthread.h>
 
 /*
  */
@@ -117,7 +118,7 @@ static int usu_probe(struct usb_interface *intf,
 			 const struct usb_device_id *id)
 {
 	unsigned long type;
-	int rc;
+	struct task_struct* task;
 	unsigned long flags;
 
 	type = USB_US_TYPE(id->driver_info);
@@ -132,8 +133,9 @@ static int usu_probe(struct usb_interface *intf,
 	stat[type].fls |= USU_MOD_FL_THREAD;
 	spin_unlock_irqrestore(&usu_lock, flags);
 
-	rc = kernel_thread(usu_probe_thread, (void*)type, CLONE_VM);
-	if (rc < 0) {
+	task = kthread_run(usu_probe_thread, (void*)type, "libusual_%d", type);
+	if (IS_ERR(task)) {
+		int rc = PTR_ERR(task);
 		printk(KERN_WARNING "libusual: "
 		    "Unable to start the thread for %s: %d\n",
 		    bias_names[type], rc);
@@ -175,8 +177,6 @@ static int usu_probe_thread(void *arg)
 	int rc;
 	unsigned long flags;
 
-	daemonize("libusual_%d", type);	/* "usb-storage" is kinda too long */
-
 	/* A completion does not work here because it's counted. */
 	down(&usu_init_notify);
 	up(&usu_init_notify);
-- 
cgit v1.2.1


From 80f8af0c59385b41564a3ae670f94a1b4caa43b2 Mon Sep 17 00:00:00 2001
From: "Eugeny S. Mints" <emints@ru.mvista.com>
Date: Sat, 2 Sep 2006 03:59:19 -0700
Subject: USB: usb serial gadget smp related bug

Adjust dev->dev_lock spinlock lock/unlock calls to be safe for SMP case.
Otherwise the following sequence may lead to a deadlock in SMP case:

gs_send()->usb_ep_queue()
	->(in case a request is satisfied immediatly) gs_write_complete()

for ex for pxa2xx_udc.c:
usb_ep_queue()->pxa2xx_ep_queue()->write_fifo()->done()->gs_write_complete()
(through req.complete pointer)

Signed-off-by: Eugeny S. Mints <emints@ru.mvista.com>
Acked-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/serial.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/usb/gadget/serial.c b/drivers/usb/gadget/serial.c
index e762aa19ab0a..b893e3118e1b 100644
--- a/drivers/usb/gadget/serial.c
+++ b/drivers/usb/gadget/serial.c
@@ -1120,12 +1120,15 @@ static int gs_send(struct gs_dev *dev)
 gs_debug_level(3, "gs_send: len=%d, 0x%2.2x 0x%2.2x 0x%2.2x ...\n", len, *((unsigned char *)req->buf), *((unsigned char *)req->buf+1), *((unsigned char *)req->buf+2));
 			list_del(&req_entry->re_entry);
 			req->length = len;
+			spin_unlock_irqrestore(&dev->dev_lock, flags);
 			if ((ret=usb_ep_queue(ep, req, GFP_ATOMIC))) {
 				printk(KERN_ERR
 				"gs_send: cannot queue read request, ret=%d\n",
 					ret);
+				spin_lock_irqsave(&dev->dev_lock, flags);
 				break;
 			}
+			spin_lock_irqsave(&dev->dev_lock, flags);
 		} else {
 			break;
 		}
-- 
cgit v1.2.1


From 901b3d75e71535f29b64f352e94ff474d95df475 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sat, 2 Sep 2006 03:13:45 -0700
Subject: USB: net2280: update dma buffer allocation

This updates the code handling dma-coherent buffer allocations, basically
reusing code from the musb_hdrc driver.  Instead of trying to work around two
significant limitations of the dma framework (memory wastage for buffers
smaller than a page, and inconsistency between calling context requirements
for allocation and free) this just works around one of them (the latter).

So count this as two steps forward (bugfixes:  the latter issue could cause
errors on some platforms, and some MIPS changes broke code for the former),
and one step back (increasing cross-platform memory wastage).

Plus linelength and whitespace fixes; and minor data segment shrinkage.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/gadget/net2280.c | 156 ++++++++++++++++++++++++-------------------
 1 file changed, 88 insertions(+), 68 deletions(-)

diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c
index 09243239d948..3bda37f9a35f 100644
--- a/drivers/usb/gadget/net2280.c
+++ b/drivers/usb/gadget/net2280.c
@@ -2,7 +2,7 @@
  * Driver for the PLX NET2280 USB device controller.
  * Specs and errata are available from <http://www.plxtech.com>.
  *
- * PLX Technology Inc. (formerly NetChip Technology) supported the 
+ * PLX Technology Inc. (formerly NetChip Technology) supported the
  * development of this driver.
  *
  *
@@ -26,7 +26,8 @@
  * Copyright (C) 2003 David Brownell
  * Copyright (C) 2003-2005 PLX Technology, Inc.
  *
- * Modified Seth Levy 2005 PLX Technology, Inc. to provide compatibility with 2282 chip
+ * Modified Seth Levy 2005 PLX Technology, Inc. to provide compatibility
+ *	with 2282 chip
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -85,7 +86,7 @@ static const char driver_name [] = "net2280";
 static const char driver_desc [] = DRIVER_DESC;
 
 static const char ep0name [] = "ep0";
-static const char *ep_name [] = {
+static const char *const ep_name [] = {
 	ep0name,
 	"ep-a", "ep-b", "ep-c", "ep-d",
 	"ep-e", "ep-f",
@@ -225,7 +226,9 @@ net2280_enable (struct usb_ep *_ep, const struct usb_endpoint_descriptor *desc)
 	if (!ep->is_in)
 		writel ((1 << SET_NAK_OUT_PACKETS), &ep->regs->ep_rsp);
 	else if (dev->pdev->device != 0x2280) {
-		/* Added for 2282, Don't use nak packets on an in endpoint, this was ignored on 2280 */
+		/* Added for 2282, Don't use nak packets on an in endpoint,
+		 * this was ignored on 2280
+		 */
 		writel ((1 << CLEAR_NAK_OUT_PACKETS)
 			| (1 << CLEAR_NAK_OUT_PACKETS_MODE), &ep->regs->ep_rsp);
 	}
@@ -288,7 +291,7 @@ static int handshake (u32 __iomem *ptr, u32 mask, u32 done, int usec)
 	return -ETIMEDOUT;
 }
 
-static struct usb_ep_ops net2280_ep_ops;
+static const struct usb_ep_ops net2280_ep_ops;
 
 static void ep_reset (struct net2280_regs __iomem *regs, struct net2280_ep *ep)
 {
@@ -449,34 +452,15 @@ net2280_free_request (struct usb_ep *_ep, struct usb_request *_req)
 
 /*-------------------------------------------------------------------------*/
 
-#undef USE_KMALLOC
-
-/* many common platforms have dma-coherent caches, which means that it's
- * safe to use kmalloc() memory for all i/o buffers without using any
- * cache flushing calls.  (unless you're trying to share cache lines
- * between dma and non-dma activities, which is a slow idea in any case.)
+/*
+ * dma-coherent memory allocation (for dma-capable endpoints)
  *
- * other platforms need more care, with 2.5 having a moderately general
- * solution (which falls down for allocations smaller than one page)
- * that improves significantly on the 2.4 PCI allocators by removing
- * the restriction that memory never be freed in_interrupt().
+ * NOTE: the dma_*_coherent() API calls suck.  Most implementations are
+ * (a) page-oriented, so small buffers lose big; and (b) asymmetric with
+ * respect to calls with irqs disabled:  alloc is safe, free is not.
+ * We currently work around (b), but not (a).
  */
-#if	defined(CONFIG_X86)
-#define USE_KMALLOC
-
-#elif	defined(CONFIG_PPC) && !defined(CONFIG_NOT_COHERENT_CACHE)
-#define USE_KMALLOC
 
-#elif	defined(CONFIG_MIPS) && !defined(CONFIG_DMA_NONCOHERENT)
-#define USE_KMALLOC
-
-/* FIXME there are other cases, including an x86-64 one ...  */
-#endif
-
-/* allocating buffers this way eliminates dma mapping overhead, which
- * on some platforms will mean eliminating a per-io buffer copy.  with
- * some kinds of system caches, further tweaks may still be needed.
- */
 static void *
 net2280_alloc_buffer (
 	struct usb_ep		*_ep,
@@ -493,43 +477,71 @@ net2280_alloc_buffer (
 		return NULL;
 	*dma = DMA_ADDR_INVALID;
 
-#if	defined(USE_KMALLOC)
-	retval = kmalloc(bytes, gfp_flags);
-	if (retval)
-		*dma = virt_to_phys(retval);
-#else
-	if (ep->dma) {
-		/* the main problem with this call is that it wastes memory
-		 * on typical 1/N page allocations: it allocates 1-N pages.
-		 */
-#warning Using dma_alloc_coherent even with buffers smaller than a page.
+	if (ep->dma)
 		retval = dma_alloc_coherent(&ep->dev->pdev->dev,
 				bytes, dma, gfp_flags);
-	} else
+	else
 		retval = kmalloc(bytes, gfp_flags);
-#endif
 	return retval;
 }
 
+static DEFINE_SPINLOCK(buflock);
+static LIST_HEAD(buffers);
+
+struct free_record {
+	struct list_head	list;
+	struct device		*dev;
+	unsigned		bytes;
+	dma_addr_t		dma;
+};
+
+static void do_free(unsigned long ignored)
+{
+	spin_lock_irq(&buflock);
+	while (!list_empty(&buffers)) {
+		struct free_record	*buf;
+
+		buf = list_entry(buffers.next, struct free_record, list);
+		list_del(&buf->list);
+		spin_unlock_irq(&buflock);
+
+		dma_free_coherent(buf->dev, buf->bytes, buf, buf->dma);
+
+		spin_lock_irq(&buflock);
+	}
+	spin_unlock_irq(&buflock);
+}
+
+static DECLARE_TASKLET(deferred_free, do_free, 0);
+
 static void
 net2280_free_buffer (
 	struct usb_ep *_ep,
-	void *buf,
+	void *address,
 	dma_addr_t dma,
 	unsigned bytes
 ) {
 	/* free memory into the right allocator */
-#ifndef	USE_KMALLOC
 	if (dma != DMA_ADDR_INVALID) {
 		struct net2280_ep	*ep;
+		struct free_record	*buf = address;
+		unsigned long		flags;
 
 		ep = container_of(_ep, struct net2280_ep, ep);
 		if (!_ep)
 			return;
-		dma_free_coherent(&ep->dev->pdev->dev, bytes, buf, dma);
+
+		ep = container_of (_ep, struct net2280_ep, ep);
+		buf->dev = &ep->dev->pdev->dev;
+		buf->bytes = bytes;
+		buf->dma = dma;
+
+		spin_lock_irqsave(&buflock, flags);
+		list_add_tail(&buf->list, &buffers);
+		tasklet_schedule(&deferred_free);
+		spin_unlock_irqrestore(&buflock, flags);
 	} else
-#endif
-		kfree (buf);
+		kfree (address);
 }
 
 /*-------------------------------------------------------------------------*/
@@ -737,7 +749,8 @@ fill_dma_desc (struct net2280_ep *ep, struct net2280_request *req, int valid)
 	 */
 	if (ep->is_in)
 		dmacount |= (1 << DMA_DIRECTION);
-	if ((!ep->is_in && (dmacount % ep->ep.maxpacket) != 0) || ep->dev->pdev->device != 0x2280)
+	if ((!ep->is_in && (dmacount % ep->ep.maxpacket) != 0)
+			|| ep->dev->pdev->device != 0x2280)
 		dmacount |= (1 << END_OF_CHAIN);
 
 	req->valid = valid;
@@ -812,7 +825,7 @@ static void start_dma (struct net2280_ep *ep, struct net2280_request *req)
 
 	/* previous OUT packet might have been short */
 	if (!ep->is_in && ((tmp = readl (&ep->regs->ep_stat))
-		 		& (1 << NAK_OUT_PACKETS)) != 0) {
+				& (1 << NAK_OUT_PACKETS)) != 0) {
 		writel ((1 << SHORT_PACKET_TRANSFERRED_INTERRUPT),
 			&ep->regs->ep_stat);
 
@@ -1373,7 +1386,7 @@ net2280_fifo_flush (struct usb_ep *_ep)
 	(void) readl (&ep->regs->ep_rsp);
 }
 
-static struct usb_ep_ops net2280_ep_ops = {
+static const struct usb_ep_ops net2280_ep_ops = {
 	.enable		= net2280_enable,
 	.disable	= net2280_disable,
 
@@ -1631,7 +1644,7 @@ show_registers (struct device *_dev, struct device_attribute *attr, char *buf)
 	}
 
 	/* Indexed Registers */
-		// none yet 
+		// none yet
 
 	/* Statistics */
 	t = scnprintf (next, size, "\nirqs:  ");
@@ -1691,11 +1704,11 @@ show_queues (struct device *_dev, struct device_attribute *attr, char *buf)
 				({ char *val;
 				 switch (d->bmAttributes & 0x03) {
 				 case USB_ENDPOINT_XFER_BULK:
-				 	val = "bulk"; break;
+					val = "bulk"; break;
 				 case USB_ENDPOINT_XFER_INT:
-				 	val = "intr"; break;
+					val = "intr"; break;
 				 default:
-				 	val = "iso"; break;
+					val = "iso"; break;
 				 }; val; }),
 				le16_to_cpu (d->wMaxPacketSize) & 0x1fff,
 				ep->dma ? "dma" : "pio", ep->fifo_size
@@ -1808,8 +1821,8 @@ extern int net2280_set_fifo_mode (struct usb_gadget *gadget, int mode);
  * net2280_set_fifo_mode - change allocation of fifo buffers
  * @gadget: access to the net2280 device that will be updated
  * @mode: 0 for default, four 1kB buffers (ep-a through ep-d);
- * 	1 for two 2kB buffers (ep-a and ep-b only);
- * 	2 for one 2kB buffer (ep-a) and two 1kB ones (ep-b, ep-c).
+ *	1 for two 2kB buffers (ep-a and ep-b only);
+ *	2 for one 2kB buffer (ep-a) and two 1kB ones (ep-b, ep-c).
  *
  * returns zero on success, else negative errno.  when this succeeds,
  * the contents of gadget->ep_list may have changed.
@@ -2241,7 +2254,8 @@ static void handle_ep_small (struct net2280_ep *ep)
 				req->td->dmacount = 0;
 				t = readl (&ep->regs->ep_avail);
 				dma_done (ep, req, count,
-					(ep->out_overflow || t) ? -EOVERFLOW : 0);
+					(ep->out_overflow || t)
+						? -EOVERFLOW : 0);
 			}
 
 			/* also flush to prevent erratum 0106 trouble */
@@ -2411,7 +2425,7 @@ static void handle_stat0_irqs (struct net2280 *dev, u32 stat)
 			, &ep->regs->ep_stat);
 		u.raw [0] = readl (&dev->usb->setup0123);
 		u.raw [1] = readl (&dev->usb->setup4567);
-		
+
 		cpu_to_le32s (&u.raw [0]);
 		cpu_to_le32s (&u.raw [1]);
 
@@ -2578,14 +2592,16 @@ static void handle_stat1_irqs (struct net2280 *dev, u32 stat)
 
 	/* VBUS disconnect is indicated by VBUS_PIN and VBUS_INTERRUPT set.
 	 * Root Port Reset is indicated by ROOT_PORT_RESET_INTERRRUPT set and
-	 * both HIGH_SPEED and FULL_SPEED clear (as ROOT_PORT_RESET_INTERRUPT 
+	 * both HIGH_SPEED and FULL_SPEED clear (as ROOT_PORT_RESET_INTERRUPT
 	 * only indicates a change in the reset state).
 	 */
 	if (stat & tmp) {
 		writel (tmp, &dev->regs->irqstat1);
-		if ((((stat & (1 << ROOT_PORT_RESET_INTERRUPT)) && 
-				((readl (&dev->usb->usbstat) & mask) == 0))
-				|| ((readl (&dev->usb->usbctl) & (1 << VBUS_PIN)) == 0) 
+		if ((((stat & (1 << ROOT_PORT_RESET_INTERRUPT))
+					&& ((readl (&dev->usb->usbstat) & mask)
+							== 0))
+				|| ((readl (&dev->usb->usbctl)
+					& (1 << VBUS_PIN)) == 0)
 			    ) && ( dev->gadget.speed != USB_SPEED_UNKNOWN)) {
 			DEBUG (dev, "disconnect %s\n",
 					dev->driver->driver.name);
@@ -2852,7 +2868,7 @@ static int net2280_probe (struct pci_dev *pdev, const struct pci_device_id *id)
 
 	/* now all the pci goodies ... */
 	if (pci_enable_device (pdev) < 0) {
-   	        retval = -ENODEV;
+	        retval = -ENODEV;
 		goto done;
 	}
 	dev->enabled = 1;
@@ -2870,6 +2886,10 @@ static int net2280_probe (struct pci_dev *pdev, const struct pci_device_id *id)
 	}
 	dev->region = 1;
 
+	/* FIXME provide firmware download interface to put
+	 * 8051 code into the chip, e.g. to turn on PCI PM.
+	 */
+
 	base = ioremap_nocache (resource, len);
 	if (base == NULL) {
 		DEBUG (dev, "can't map memory\n");
@@ -2984,16 +3004,16 @@ static void net2280_shutdown (struct pci_dev *pdev)
 
 /*-------------------------------------------------------------------------*/
 
-static struct pci_device_id pci_ids [] = { {
-	.class = 	((PCI_CLASS_SERIAL_USB << 8) | 0xfe),
-	.class_mask = 	~0,
+static const struct pci_device_id pci_ids [] = { {
+	.class =	((PCI_CLASS_SERIAL_USB << 8) | 0xfe),
+	.class_mask =	~0,
 	.vendor =	0x17cc,
 	.device =	0x2280,
 	.subvendor =	PCI_ANY_ID,
 	.subdevice =	PCI_ANY_ID,
 }, {
-	.class = 	((PCI_CLASS_SERIAL_USB << 8) | 0xfe),
-	.class_mask = 	~0,
+	.class =	((PCI_CLASS_SERIAL_USB << 8) | 0xfe),
+	.class_mask =	~0,
 	.vendor =	0x17cc,
 	.device =	0x2282,
 	.subvendor =	PCI_ANY_ID,
-- 
cgit v1.2.1


From cd22afda3f84452c6def29a68b06933e814e0e95 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Sun, 3 Sep 2006 12:21:50 -0700
Subject: USB: ohci-at91, two one-liners

This includes two one-liners forwarded to me for the OHCI support on at91:

 - KB920x (and other boards with CPUs in non-BGA packages) need a slightly
   different way to say "ignore that port, it's not pinned out";

 - On resume, if we turn clocks on, record that we did so.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ohci-at91.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c
index 5a5bdf374d76..b466581beb4a 100644
--- a/drivers/usb/host/ohci-at91.c
+++ b/drivers/usb/host/ohci-at91.c
@@ -193,7 +193,7 @@ ohci_at91_start (struct usb_hcd *hcd)
 	if ((ret = ohci_init(ohci)) < 0)
 		return ret;
 
-	root->maxchild = board->ports;
+	ohci->num_ports = board->ports;
 
 	if ((ret = ohci_run(ohci)) < 0) {
 		err("can't start %s", hcd->self.bus_name);
@@ -297,6 +297,7 @@ static int ohci_hcd_at91_drv_resume(struct platform_device *pdev)
 	if (!clocked) {
 		clk_enable(iclk);
 		clk_enable(fclk);
+		clocked = 1;
 	}
 
 	return 0;
-- 
cgit v1.2.1


From 14acdcd32160767d7f2fa00803d2d764d3e3373a Mon Sep 17 00:00:00 2001
From: Reiner Herrmann <reiner@reiner-h.de>
Date: Wed, 6 Sep 2006 02:37:21 +0200
Subject: USB: usb/input/usbmouse.c: whitespace cleanup

Replace spaces with tab and change comment indention for better readability.

Signed-off-by: Reiner Herrmann <reiner@reiner-h.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/usbmouse.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/input/usbmouse.c b/drivers/usb/input/usbmouse.c
index 446935b671d9..0fb792be95ef 100644
--- a/drivers/usb/input/usbmouse.c
+++ b/drivers/usb/input/usbmouse.c
@@ -218,7 +218,7 @@ static void usb_mouse_disconnect(struct usb_interface *intf)
 
 static struct usb_device_id usb_mouse_id_table [] = {
 	{ USB_INTERFACE_INFO(3, 1, 2) },
-    { }						/* Terminating entry */
+	{ }	/* Terminating entry */
 };
 
 MODULE_DEVICE_TABLE (usb, usb_mouse_id_table);
-- 
cgit v1.2.1


From d1ad4ea331e78a5ff90eda7718da31bcbc1a9c38 Mon Sep 17 00:00:00 2001
From: Pete Zaitcev <zaitcev@redhat.com>
Date: Thu, 7 Sep 2006 16:54:22 -0700
Subject: USB: UB: Let cdrecord to see a device with media absent

The command "cdrecord dev=/dev/uba x.iso" prints nasty garbage if a blank
is not in the drive. This happens because drivers have to set req->errors
separately from just returning zero uptodate with end_that_request_first,
end_that_request_last. These functions only set error in BIO, but sg_io()
ignores it.

Since we're on it, let cdrecord access a device when ->changed is set.
It's useful if someone wants to know device capabilities without burning
anything.

Signed-Off-By: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/block/ub.c | 38 +++++++++++++++++++++++++-------------
 1 file changed, 25 insertions(+), 13 deletions(-)

diff --git a/drivers/block/ub.c b/drivers/block/ub.c
index d62b49fbf10c..45a8f402b07b 100644
--- a/drivers/block/ub.c
+++ b/drivers/block/ub.c
@@ -358,7 +358,7 @@ static void ub_cmd_build_block(struct ub_dev *sc, struct ub_lun *lun,
 static void ub_cmd_build_packet(struct ub_dev *sc, struct ub_lun *lun,
     struct ub_scsi_cmd *cmd, struct ub_request *urq);
 static void ub_rw_cmd_done(struct ub_dev *sc, struct ub_scsi_cmd *cmd);
-static void ub_end_rq(struct request *rq, int uptodate);
+static void ub_end_rq(struct request *rq, unsigned int status);
 static int ub_rw_cmd_retry(struct ub_dev *sc, struct ub_lun *lun,
     struct ub_request *urq, struct ub_scsi_cmd *cmd);
 static int ub_submit_scsi(struct ub_dev *sc, struct ub_scsi_cmd *cmd);
@@ -639,9 +639,15 @@ static int ub_request_fn_1(struct ub_lun *lun, struct request *rq)
 	struct ub_request *urq;
 	int n_elem;
 
-	if (atomic_read(&sc->poison) || lun->changed) {
+	if (atomic_read(&sc->poison)) {
+		blkdev_dequeue_request(rq);
+		ub_end_rq(rq, DID_NO_CONNECT << 16);
+		return 0;
+	}
+
+	if (lun->changed && !blk_pc_request(rq)) {
 		blkdev_dequeue_request(rq);
-		ub_end_rq(rq, 0);
+		ub_end_rq(rq, SAM_STAT_CHECK_CONDITION);
 		return 0;
 	}
 
@@ -693,7 +699,7 @@ static int ub_request_fn_1(struct ub_lun *lun, struct request *rq)
 
 drop:
 	ub_put_cmd(lun, cmd);
-	ub_end_rq(rq, 0);
+	ub_end_rq(rq, DID_ERROR << 16);
 	return 0;
 }
 
@@ -761,47 +767,53 @@ static void ub_rw_cmd_done(struct ub_dev *sc, struct ub_scsi_cmd *cmd)
 	struct ub_lun *lun = cmd->lun;
 	struct ub_request *urq = cmd->back;
 	struct request *rq;
-	int uptodate;
+	unsigned int scsi_status;
 
 	rq = urq->rq;
 
 	if (cmd->error == 0) {
-		uptodate = 1;
-
 		if (blk_pc_request(rq)) {
 			if (cmd->act_len >= rq->data_len)
 				rq->data_len = 0;
 			else
 				rq->data_len -= cmd->act_len;
 		}
+		scsi_status = 0;
 	} else {
-		uptodate = 0;
-
 		if (blk_pc_request(rq)) {
 			/* UB_SENSE_SIZE is smaller than SCSI_SENSE_BUFFERSIZE */
 			memcpy(rq->sense, sc->top_sense, UB_SENSE_SIZE);
 			rq->sense_len = UB_SENSE_SIZE;
 			if (sc->top_sense[0] != 0)
-				rq->errors = SAM_STAT_CHECK_CONDITION;
+				scsi_status = SAM_STAT_CHECK_CONDITION;
 			else
-				rq->errors = DID_ERROR << 16;
+				scsi_status = DID_ERROR << 16;
 		} else {
 			if (cmd->error == -EIO) {
 				if (ub_rw_cmd_retry(sc, lun, urq, cmd) == 0)
 					return;
 			}
+			scsi_status = SAM_STAT_CHECK_CONDITION;
 		}
 	}
 
 	urq->rq = NULL;
 
 	ub_put_cmd(lun, cmd);
-	ub_end_rq(rq, uptodate);
+	ub_end_rq(rq, scsi_status);
 	blk_start_queue(lun->disk->queue);
 }
 
-static void ub_end_rq(struct request *rq, int uptodate)
+static void ub_end_rq(struct request *rq, unsigned int scsi_status)
 {
+	int uptodate;
+
+	if (scsi_status == 0) {
+		uptodate = 1;
+	} else {
+		uptodate = 0;
+		rq->errors = scsi_status;
+	}
 	end_that_request_first(rq, uptodate, rq->hard_nr_sectors);
 	end_that_request_last(rq, uptodate);
 }
-- 
cgit v1.2.1


From b6956ffa595db97656d5901ca8fec77ef272d41a Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:46:48 -0400
Subject: usbcore: store each usb_device's level in the tree

This patch (as778) adds a field to struct usb_device to store the
device's level in the USB tree.  In itself this number isn't really
important.  But the overhead is very low, and in a later patch it will
be used for preventing bogus warnings from the lockdep checker.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 1 +
 include/linux/usb.h    | 1 +
 2 files changed, 2 insertions(+)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index f5adce049b35..78e910b2046c 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -2414,6 +2414,7 @@ static void hub_port_connect_change(struct usb_hub *hub, int port1,
 		usb_set_device_state(udev, USB_STATE_POWERED);
 		udev->speed = USB_SPEED_UNKNOWN;
  		udev->bus_mA = hub->mA_per_port;
+		udev->level = hdev->level + 1;
 
 		/* set the address */
 		choose_address(udev);
diff --git a/include/linux/usb.h b/include/linux/usb.h
index c66303285a45..df5c93eb3ce9 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -348,6 +348,7 @@ struct usb_device {
 
 	unsigned short bus_mA;		/* Current available from the bus */
 	u8 portnum;			/* Parent port number (origin 1) */
+	u8 level;			/* Number of USB hub ancestors */
 
 	int have_langid;		/* whether string_langid is valid */
 	int string_langid;		/* language ID for strings */
-- 
cgit v1.2.1


From 645daaab0b6adc35c1838df2a82f9d729fdb1767 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:47:02 -0400
Subject: usbcore: add autosuspend/autoresume infrastructure

This patch (as739) adds the basic infrastructure for USB autosuspend
and autoresume.  The main features are:

	PM usage counters added to struct usb_device and struct
	usb_interface, indicating whether it's okay to autosuspend
	them or they are currently in use.

	Flag added to usb_device indicating whether the current
	suspend/resume operation originated from outside or as an
	autosuspend/autoresume.

	Flag added to usb_driver indicating whether the driver
	supports autosuspend.  If not, no device bound to the driver
	will be autosuspended.

	Mutex added to usb_device for protecting PM operations.
	Unlike the device semaphore, the locking rule for the pm_mutex
	is that you must acquire the locks going _up_ the device tree.

	New routines handling autosuspend/autoresume requests for
	interfaces and devices.

	Suspend and resume requests are propagated up the device tree
	(but not outside the USB subsystem).

	work_struct added to usb_device, for carrying out delayed
	autosuspend requests.

	Autoresume added (and autosuspend prevented) during probe and
	disconnect.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 419 ++++++++++++++++++++++++++++++++++++++++++----
 drivers/usb/core/hub.c    |  49 ++++--
 drivers/usb/core/usb.c    |  23 +++
 drivers/usb/core/usb.h    |  14 ++
 include/linux/usb.h       |  33 ++++
 5 files changed, 493 insertions(+), 45 deletions(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index a5d11461f5a9..2b2000ac05ab 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -157,12 +157,13 @@ static int usb_probe_device(struct device *dev)
 
 	udev = to_usb_device(dev);
 
-	/* FIXME: resume a suspended device */
-	if (udev->state == USB_STATE_SUSPENDED)
-		return -EHOSTUNREACH;
-
 	/* TODO: Add real matching code */
 
+	/* The device should always appear to be in use
+	 * unless the driver suports autosuspend.
+	 */
+	udev->pm_usage_cnt = !(udriver->supports_autosuspend);
+
 	error = udriver->probe(udev);
 	return error;
 }
@@ -182,6 +183,7 @@ static int usb_probe_interface(struct device *dev)
 {
 	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf;
+	struct usb_device *udev;
 	const struct usb_device_id *id;
 	int error = -ENODEV;
 
@@ -191,10 +193,7 @@ static int usb_probe_interface(struct device *dev)
 		return error;
 
 	intf = to_usb_interface(dev);
-
-	/* FIXME we'd much prefer to just resume it ... */
-	if (interface_to_usbdev(intf)->state == USB_STATE_SUSPENDED)
-		return -EHOSTUNREACH;
+	udev = interface_to_usbdev(intf);
 
 	id = usb_match_id(intf, driver->id_table);
 	if (!id)
@@ -202,18 +201,31 @@ static int usb_probe_interface(struct device *dev)
 	if (id) {
 		dev_dbg(dev, "%s - got id\n", __FUNCTION__);
 
+		error = usb_autoresume_device(udev, 1);
+		if (error)
+			return error;
+
 		/* Interface "power state" doesn't correspond to any hardware
 		 * state whatsoever.  We use it to record when it's bound to
 		 * a driver that may start I/0:  it's not frozen/quiesced.
 		 */
 		mark_active(intf);
 		intf->condition = USB_INTERFACE_BINDING;
+
+		/* The interface should always appear to be in use
+		 * unless the driver suports autosuspend.
+		 */
+		intf->pm_usage_cnt = !(driver->supports_autosuspend);
+
 		error = driver->probe(intf, id);
 		if (error) {
 			mark_quiesced(intf);
+			intf->needs_remote_wakeup = 0;
 			intf->condition = USB_INTERFACE_UNBOUND;
 		} else
 			intf->condition = USB_INTERFACE_BOUND;
+
+		usb_autosuspend_device(udev, 1);
 	}
 
 	return error;
@@ -224,9 +236,15 @@ static int usb_unbind_interface(struct device *dev)
 {
 	struct usb_driver *driver = to_usb_driver(dev->driver);
 	struct usb_interface *intf = to_usb_interface(dev);
+	struct usb_device *udev;
+	int error;
 
 	intf->condition = USB_INTERFACE_UNBINDING;
 
+	/* Autoresume for set_interface call below */
+	udev = interface_to_usbdev(intf);
+	error = usb_autoresume_device(udev, 1);
+
 	/* release all urbs for this interface */
 	usb_disable_interface(interface_to_usbdev(intf), intf);
 
@@ -237,8 +255,13 @@ static int usb_unbind_interface(struct device *dev)
 			intf->altsetting[0].desc.bInterfaceNumber,
 			0);
 	usb_set_intfdata(intf, NULL);
+
 	intf->condition = USB_INTERFACE_UNBOUND;
 	mark_quiesced(intf);
+	intf->needs_remote_wakeup = 0;
+
+	if (!error)
+		usb_autosuspend_device(udev, 1);
 
 	return 0;
 }
@@ -267,14 +290,19 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 				struct usb_interface *iface, void* priv)
 {
 	struct device *dev = &iface->dev;
+	struct usb_device *udev = interface_to_usbdev(iface);
 
 	if (dev->driver)
 		return -EBUSY;
 
 	dev->driver = &driver->drvwrap.driver;
 	usb_set_intfdata(iface, priv);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	iface->condition = USB_INTERFACE_BOUND;
 	mark_active(iface);
+	iface->pm_usage_cnt = !(driver->supports_autosuspend);
+	mutex_unlock(&udev->pm_mutex);
 
 	/* if interface was already added, bind now; else let
 	 * the future device_add() bind it, bypassing probe()
@@ -304,6 +332,7 @@ void usb_driver_release_interface(struct usb_driver *driver,
 					struct usb_interface *iface)
 {
 	struct device *dev = &iface->dev;
+	struct usb_device *udev = interface_to_usbdev(iface);
 
 	/* this should never happen, don't release something that's not ours */
 	if (!dev->driver || dev->driver != &driver->drvwrap.driver)
@@ -321,8 +350,12 @@ void usb_driver_release_interface(struct usb_driver *driver,
 
 	dev->driver = NULL;
 	usb_set_intfdata(iface, NULL);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	iface->condition = USB_INTERFACE_UNBOUND;
 	mark_quiesced(iface);
+	iface->needs_remote_wakeup = 0;
+	mutex_unlock(&udev->pm_mutex);
 }
 EXPORT_SYMBOL(usb_driver_release_interface);
 
@@ -751,7 +784,7 @@ EXPORT_SYMBOL_GPL_FUTURE(usb_deregister);
 
 #ifdef CONFIG_PM
 
-/* Caller has locked udev */
+/* Caller has locked udev->pm_mutex */
 static int suspend_device(struct usb_device *udev, pm_message_t msg)
 {
 	struct usb_device_driver	*udriver;
@@ -763,6 +796,7 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 
 	/* For devices that don't have a driver, we do a standard suspend. */
 	if (udev->dev.driver == NULL) {
+		udev->do_remote_wakeup = 0;
 		status = usb_port_suspend(udev);
 		goto done;
 	}
@@ -771,12 +805,13 @@ static int suspend_device(struct usb_device *udev, pm_message_t msg)
 	status = udriver->suspend(udev, msg);
 
 done:
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		udev->dev.power.power_state.event = msg.event;
 	return status;
 }
 
-/* Caller has locked udev */
+/* Caller has locked udev->pm_mutex */
 static int resume_device(struct usb_device *udev)
 {
 	struct usb_device_driver	*udriver;
@@ -796,12 +831,13 @@ static int resume_device(struct usb_device *udev)
 	status = udriver->resume(udev);
 
 done:
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		udev->dev.power.power_state.event = PM_EVENT_ON;
 	return status;
 }
 
-/* Caller has locked intf's usb_device */
+/* Caller has locked intf's usb_device's pm_mutex */
 static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 {
 	struct usb_driver	*driver;
@@ -812,31 +848,33 @@ static int suspend_interface(struct usb_interface *intf, pm_message_t msg)
 			!is_active(intf))
 		goto done;
 
-	if (intf->dev.driver == NULL)		/* This can't happen */
+	if (intf->condition == USB_INTERFACE_UNBOUND)	/* This can't happen */
 		goto done;
 	driver = to_usb_driver(intf->dev.driver);
 
 	if (driver->suspend && driver->resume) {
 		status = driver->suspend(intf, msg);
-		if (status)
+		if (status == 0)
+			mark_quiesced(intf);
+		else if (!interface_to_usbdev(intf)->auto_pm)
 			dev_err(&intf->dev, "%s error %d\n",
 					"suspend", status);
-		else
-			mark_quiesced(intf);
 	} else {
 		// FIXME else if there's no suspend method, disconnect...
+		// Not possible if auto_pm is set...
 		dev_warn(&intf->dev, "no suspend for driver %s?\n",
 				driver->name);
 		mark_quiesced(intf);
 	}
 
 done:
+	// dev_dbg(&intf->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		intf->dev.power.power_state.event = msg.event;
 	return status;
 }
 
-/* Caller has locked intf's usb_device */
+/* Caller has locked intf's usb_device's pm_mutex */
 static int resume_interface(struct usb_interface *intf)
 {
 	struct usb_driver	*driver;
@@ -846,8 +884,12 @@ static int resume_interface(struct usb_interface *intf)
 			is_active(intf))
 		goto done;
 
+	/* Don't let autoresume interfere with unbinding */
+	if (intf->condition == USB_INTERFACE_UNBINDING)
+		goto done;
+
 	/* Can't resume it if it doesn't have a driver. */
-	if (intf->dev.driver == NULL) {
+	if (intf->condition == USB_INTERFACE_UNBOUND) {
 		status = -ENOTCONN;
 		goto done;
 	}
@@ -867,18 +909,88 @@ static int resume_interface(struct usb_interface *intf)
 	}
 
 done:
+	// dev_dbg(&intf->dev, "%s: status %d\n", __FUNCTION__, status);
 	if (status == 0)
 		intf->dev.power.power_state.event = PM_EVENT_ON;
 	return status;
 }
 
-/* Caller has locked udev */
+/**
+ * usb_suspend_both - suspend a USB device and its interfaces
+ * @udev: the usb_device to suspend
+ * @msg: Power Management message describing this state transition
+ *
+ * This is the central routine for suspending USB devices.  It calls the
+ * suspend methods for all the interface drivers in @udev and then calls
+ * the suspend method for @udev itself.  If an error occurs at any stage,
+ * all the interfaces which were suspended are resumed so that they remain
+ * in the same state as the device.
+ *
+ * If an autosuspend is in progress (@udev->auto_pm is set), the routine
+ * checks first to make sure that neither the device itself or any of its
+ * active interfaces is in use (pm_usage_cnt is greater than 0).  If they
+ * are, the autosuspend fails.
+ *
+ * If the suspend succeeds, the routine recursively queues an autosuspend
+ * request for @udev's parent device, thereby propagating the change up
+ * the device tree.  If all of the parent's children are now suspended,
+ * the parent will autosuspend in turn.
+ *
+ * The suspend method calls are subject to mutual exclusion under control
+ * of @udev's pm_mutex.  Many of these calls are also under the protection
+ * of @udev's device lock (including all requests originating outside the
+ * USB subsystem), but autosuspend requests generated by a child device or
+ * interface driver may not be.  Usbcore will insure that the method calls
+ * do not arrive during bind, unbind, or reset operations.  However, drivers
+ * must be prepared to handle suspend calls arriving at unpredictable times.
+ * The only way to block such calls is to do an autoresume (preventing
+ * autosuspends) while holding @udev's device lock (preventing outside
+ * suspends).
+ *
+ * The caller must hold @udev->pm_mutex.
+ *
+ * This routine can run only in process context.
+ */
 int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 {
 	int			status = 0;
 	int			i = 0;
 	struct usb_interface	*intf;
+	struct usb_device	*parent = udev->parent;
+
+	cancel_delayed_work(&udev->autosuspend);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return 0;
+	if (udev->state == USB_STATE_SUSPENDED)
+		return 0;
 
+	udev->do_remote_wakeup = device_may_wakeup(&udev->dev);
+
+	/* For autosuspend, fail fast if anything is in use.
+	 * Also fail if any interfaces require remote wakeup but it
+	 * isn't available. */
+	if (udev->auto_pm) {
+		if (udev->pm_usage_cnt > 0)
+			return -EBUSY;
+		if (udev->actconfig) {
+			for (; i < udev->actconfig->desc.bNumInterfaces; i++) {
+				intf = udev->actconfig->interface[i];
+				if (!is_active(intf))
+					continue;
+				if (intf->pm_usage_cnt > 0)
+					return -EBUSY;
+				if (intf->needs_remote_wakeup &&
+						!udev->do_remote_wakeup) {
+					dev_dbg(&udev->dev,
+	"remote wakeup needed for autosuspend\n");
+					return -EOPNOTSUPP;
+				}
+			}
+			i = 0;
+		}
+	}
+
+	/* Suspend all the interfaces and then udev itself */
 	if (udev->actconfig) {
 		for (; i < udev->actconfig->desc.bNumInterfaces; i++) {
 			intf = udev->actconfig->interface[i];
@@ -896,40 +1008,282 @@ int usb_suspend_both(struct usb_device *udev, pm_message_t msg)
 			intf = udev->actconfig->interface[i];
 			resume_interface(intf);
 		}
-	}
+
+	/* If the suspend succeeded, propagate it up the tree */
+	} else if (parent)
+		usb_autosuspend_device(parent, 0);
+
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
 	return status;
 }
 
-/* Caller has locked udev */
+/**
+ * usb_resume_both - resume a USB device and its interfaces
+ * @udev: the usb_device to resume
+ *
+ * This is the central routine for resuming USB devices.  It calls the
+ * the resume method for @udev and then calls the resume methods for all
+ * the interface drivers in @udev.
+ *
+ * Before starting the resume, the routine calls itself recursively for
+ * the parent device of @udev, thereby propagating the change up the device
+ * tree and assuring that @udev will be able to resume.  If the parent is
+ * unable to resume successfully, the routine fails.
+ *
+ * The resume method calls are subject to mutual exclusion under control
+ * of @udev's pm_mutex.  Many of these calls are also under the protection
+ * of @udev's device lock (including all requests originating outside the
+ * USB subsystem), but autoresume requests generated by a child device or
+ * interface driver may not be.  Usbcore will insure that the method calls
+ * do not arrive during bind, unbind, or reset operations.  However, drivers
+ * must be prepared to handle resume calls arriving at unpredictable times.
+ * The only way to block such calls is to do an autoresume (preventing
+ * other autoresumes) while holding @udev's device lock (preventing outside
+ * resumes).
+ *
+ * The caller must hold @udev->pm_mutex.
+ *
+ * This routine can run only in process context.
+ */
 int usb_resume_both(struct usb_device *udev)
 {
-	int			status;
+	int			status = 0;
 	int			i;
 	struct usb_interface	*intf;
+	struct usb_device	*parent = udev->parent;
+
+	cancel_delayed_work(&udev->autosuspend);
+	if (udev->state == USB_STATE_NOTATTACHED)
+		return -ENODEV;
 
-	/* Can't resume if the parent is suspended */
-	if (udev->parent && udev->parent->state == USB_STATE_SUSPENDED) {
-		dev_warn(&udev->dev, "can't resume; parent is suspended\n");
-		return -EHOSTUNREACH;
+	/* Propagate the resume up the tree, if necessary */
+	if (udev->state == USB_STATE_SUSPENDED) {
+		if (parent) {
+			mutex_lock_nested(&parent->pm_mutex, parent->level);
+			parent->auto_pm = 1;
+			status = usb_resume_both(parent);
+		} else {
+
+			/* We can't progagate beyond the USB subsystem,
+			 * so if a root hub's controller is suspended
+			 * then we're stuck. */
+			if (udev->dev.parent->power.power_state.event !=
+					PM_EVENT_ON)
+				status = -EHOSTUNREACH;
+		}
+		if (status == 0 && udev->state == USB_STATE_SUSPENDED)
+			status = resume_device(udev);
+		if (parent)
+			mutex_unlock(&parent->pm_mutex);
 	}
 
-	status = resume_device(udev);
+	/* Now the parent won't suspend until we are finished */
+
 	if (status == 0 && udev->actconfig) {
 		for (i = 0; i < udev->actconfig->desc.bNumInterfaces; i++) {
 			intf = udev->actconfig->interface[i];
 			resume_interface(intf);
 		}
 	}
+
+	// dev_dbg(&udev->dev, "%s: status %d\n", __FUNCTION__, status);
+	return status;
+}
+
+#ifdef CONFIG_USB_SUSPEND
+
+/**
+ * usb_autosuspend_device - delayed autosuspend of a USB device and its interfaces
+ * @udev - the usb_device to autosuspend
+ * @dec_usage_cnt - flag to decrement @udev's PM-usage counter
+ *
+ * This routine should be called when a core subsystem is finished using
+ * @udev and wants to allow it to autosuspend.  Examples would be when
+ * @udev's device file in usbfs is closed or after a configuration change.
+ *
+ * @dec_usage_cnt should be 1 if the subsystem previously incremented
+ * @udev's usage counter (such as by passing 1 to usb_autoresume_device);
+ * otherwise it should be 0.
+ *
+ * If the usage counter for @udev or any of its active interfaces is greater
+ * than 0, the autosuspend request will not be queued.  (If an interface
+ * driver does not support autosuspend then its usage counter is permanently
+ * positive.)  Likewise, if an interface driver requires remote-wakeup
+ * capability during autosuspend but remote wakeup is disabled, the
+ * autosuspend will fail.
+ *
+ * Often the caller will hold @udev's device lock, but this is not
+ * necessary.
+ *
+ * This routine can run only in process context.
+ */
+void usb_autosuspend_device(struct usb_device *udev, int dec_usage_cnt)
+{
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->pm_usage_cnt -= dec_usage_cnt;
+	if (udev->pm_usage_cnt <= 0)
+		schedule_delayed_work(&udev->autosuspend,
+				USB_AUTOSUSPEND_DELAY);
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&udev->dev, "%s: cnt %d\n",
+	//		__FUNCTION__, udev->pm_usage_cnt);
+}
+
+/**
+ * usb_autoresume_device - immediately autoresume a USB device and its interfaces
+ * @udev - the usb_device to autoresume
+ * @inc_usage_cnt - flag to increment @udev's PM-usage counter
+ *
+ * This routine should be called when a core subsystem wants to use @udev
+ * and needs to guarantee that it is not suspended.  In addition, the
+ * caller can prevent @udev from being autosuspended subsequently.  (Note
+ * that this will not prevent suspend events originating in the PM core.)
+ * Examples would be when @udev's device file in usbfs is opened (autosuspend
+ * should be prevented until the file is closed) or when a remote-wakeup
+ * request is received (later autosuspends should not be prevented).
+ *
+ * @inc_usage_cnt should be 1 to increment @udev's usage counter and prevent
+ * autosuspends.  This prevention will persist until the usage counter is
+ * decremented again (such as by passing 1 to usb_autosuspend_device).
+ * Otherwise @inc_usage_cnt should be 0 to leave the usage counter unchanged.
+ * Regardless, if the autoresume fails then the usage counter is not
+ * incremented.
+ *
+ * Often the caller will hold @udev's device lock, but this is not
+ * necessary (and attempting it might cause deadlock).
+ *
+ * This routine can run only in process context.
+ */
+int usb_autoresume_device(struct usb_device *udev, int inc_usage_cnt)
+{
+	int	status;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->pm_usage_cnt += inc_usage_cnt;
+	udev->auto_pm = 1;
+	status = usb_resume_both(udev);
+	if (status != 0)
+		udev->pm_usage_cnt -= inc_usage_cnt;
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&udev->dev, "%s: status %d cnt %d\n",
+	//		__FUNCTION__, status, udev->pm_usage_cnt);
+	return status;
+}
+
+/**
+ * usb_autopm_put_interface - decrement a USB interface's PM-usage counter
+ * @intf - the usb_interface whose counter should be decremented
+ *
+ * This routine should be called by an interface driver when it is
+ * finished using @intf and wants to allow it to autosuspend.  A typical
+ * example would be a character-device driver when its device file is
+ * closed.
+ *
+ * The routine decrements @intf's usage counter.  When the counter reaches
+ * 0, a delayed autosuspend request for @intf's device is queued.  When
+ * the delay expires, if @intf->pm_usage_cnt is still <= 0 along with all
+ * the other usage counters for the sibling interfaces and @intf's
+ * usb_device, the device and all its interfaces will be autosuspended.
+ *
+ * Note that @intf->pm_usage_cnt is owned by the interface driver.  The
+ * core will not change its value other than the increment and decrement
+ * in usb_autopm_get_interface and usb_autopm_put_interface.  The driver
+ * may use this simple counter-oriented discipline or may set the value
+ * any way it likes.
+ *
+ * If the driver has set @intf->needs_remote_wakeup then autosuspend will
+ * take place only if the device's remote-wakeup facility is enabled.
+ *
+ * Suspend method calls queued by this routine can arrive at any time
+ * while @intf is resumed and its usage counter is equal to 0.  They are
+ * not protected by the usb_device's lock but only by its pm_mutex.
+ * Drivers must provide their own synchronization.
+ *
+ * This routine can run only in process context.
+ */
+void usb_autopm_put_interface(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	if (intf->condition != USB_INTERFACE_UNBOUND) {
+		if (--intf->pm_usage_cnt <= 0)
+			schedule_delayed_work(&udev->autosuspend,
+					USB_AUTOSUSPEND_DELAY);
+	}
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&intf->dev, "%s: cnt %d\n",
+	//		__FUNCTION__, intf->pm_usage_cnt);
+}
+EXPORT_SYMBOL_GPL(usb_autopm_put_interface);
+
+/**
+ * usb_autopm_get_interface - increment a USB interface's PM-usage counter
+ * @intf - the usb_interface whose counter should be incremented
+ *
+ * This routine should be called by an interface driver when it wants to
+ * use @intf and needs to guarantee that it is not suspended.  In addition,
+ * the routine prevents @intf from being autosuspended subsequently.  (Note
+ * that this will not prevent suspend events originating in the PM core.)
+ * This prevention will persist until usb_autopm_put_interface() is called
+ * or @intf is unbound.  A typical example would be a character-device
+ * driver when its device file is opened.
+ *
+ * The routine increments @intf's usage counter.  So long as the counter
+ * is greater than 0, autosuspend will not be allowed for @intf or its
+ * usb_device.  When the driver is finished using @intf it should call
+ * usb_autopm_put_interface() to decrement the usage counter and queue
+ * a delayed autosuspend request (if the counter is <= 0).
+ *
+ * Note that @intf->pm_usage_cnt is owned by the interface driver.  The
+ * core will not change its value other than the increment and decrement
+ * in usb_autopm_get_interface and usb_autopm_put_interface.  The driver
+ * may use this simple counter-oriented discipline or may set the value
+ * any way it likes.
+ *
+ * Resume method calls generated by this routine can arrive at any time
+ * while @intf is suspended.  They are not protected by the usb_device's
+ * lock but only by its pm_mutex.  Drivers must provide their own
+ * synchronization.
+ *
+ * This routine can run only in process context.
+ */
+int usb_autopm_get_interface(struct usb_interface *intf)
+{
+	struct usb_device	*udev = interface_to_usbdev(intf);
+	int			status;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	if (intf->condition == USB_INTERFACE_UNBOUND)
+		status = -ENODEV;
+	else {
+		++intf->pm_usage_cnt;
+		udev->auto_pm = 1;
+		status = usb_resume_both(udev);
+		if (status != 0)
+			--intf->pm_usage_cnt;
+	}
+	mutex_unlock(&udev->pm_mutex);
+	// dev_dbg(&intf->dev, "%s: status %d cnt %d\n",
+	//		__FUNCTION__, status, intf->pm_usage_cnt);
 	return status;
 }
+EXPORT_SYMBOL_GPL(usb_autopm_get_interface);
+
+#endif /* CONFIG_USB_SUSPEND */
 
 static int usb_suspend(struct device *dev, pm_message_t message)
 {
 	int	status;
 
-	if (is_usb_device(dev))
-		status = usb_suspend_both(to_usb_device(dev), message);
-	else
+	if (is_usb_device(dev)) {
+		struct usb_device *udev = to_usb_device(dev);
+
+		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		udev->auto_pm = 0;
+		status = usb_suspend_both(udev, message);
+		mutex_unlock(&udev->pm_mutex);
+	} else
 		status = 0;
 	return status;
 }
@@ -939,7 +1293,12 @@ static int usb_resume(struct device *dev)
 	int	status;
 
 	if (is_usb_device(dev)) {
-		status = usb_resume_both(to_usb_device(dev));
+		struct usb_device *udev = to_usb_device(dev);
+
+		mutex_lock_nested(&udev->pm_mutex, udev->level);
+		udev->auto_pm = 0;
+		status = usb_resume_both(udev);
+		mutex_unlock(&udev->pm_mutex);
 
 		/* Rebind drivers that had no suspend method? */
 	} else
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 78e910b2046c..dee812bc6c43 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1017,19 +1017,22 @@ void usb_set_device_state(struct usb_device *udev,
 	if (udev->state == USB_STATE_NOTATTACHED)
 		;	/* do nothing */
 	else if (new_state != USB_STATE_NOTATTACHED) {
-		udev->state = new_state;
 
 		/* root hub wakeup capabilities are managed out-of-band
 		 * and may involve silicon errata ... ignore them here.
 		 */
 		if (udev->parent) {
-			if (new_state == USB_STATE_CONFIGURED)
+			if (udev->state == USB_STATE_SUSPENDED
+					|| new_state == USB_STATE_SUSPENDED)
+				;	/* No change to wakeup settings */
+			else if (new_state == USB_STATE_CONFIGURED)
 				device_init_wakeup(&udev->dev,
 					(udev->actconfig->desc.bmAttributes
 					 & USB_CONFIG_ATT_WAKEUP));
-			else if (new_state != USB_STATE_SUSPENDED)
+			else
 				device_init_wakeup(&udev->dev, 0);
 		}
+		udev->state = new_state;
 	} else
 		recursively_mark_NOTATTACHED(udev);
 	spin_unlock_irqrestore(&device_state_lock, flags);
@@ -1507,7 +1510,7 @@ static int hub_port_suspend(struct usb_hub *hub, int port1,
 	 * NOTE:  OTG devices may issue remote wakeup (or SRP) even when
 	 * we don't explicitly enable it here.
 	 */
-	if (device_may_wakeup(&udev->dev)) {
+	if (udev->do_remote_wakeup) {
 		status = usb_control_msg(udev, usb_sndctrlpipe(udev, 0),
 				USB_REQ_SET_FEATURE, USB_RECIP_DEVICE,
 				USB_DEVICE_REMOTE_WAKEUP, 0,
@@ -1533,7 +1536,8 @@ static int hub_port_suspend(struct usb_hub *hub, int port1,
 				USB_CTRL_SET_TIMEOUT);
 	} else {
 		/* device has up to 10 msec to fully suspend */
-		dev_dbg(&udev->dev, "usb suspend\n");
+		dev_dbg(&udev->dev, "usb %ssuspend\n",
+				udev->auto_pm ? "auto-" : "");
 		usb_set_device_state(udev, USB_STATE_SUSPENDED);
 		msleep(10);
 	}
@@ -1573,7 +1577,8 @@ static int __usb_port_suspend (struct usb_device *udev, int port1)
 		status = hub_port_suspend(hdev_to_hub(udev->parent), port1,
 				udev);
 	else {
-		dev_dbg(&udev->dev, "usb suspend\n");
+		dev_dbg(&udev->dev, "usb %ssuspend\n",
+				udev->auto_pm ? "auto-" : "");
 		usb_set_device_state(udev, USB_STATE_SUSPENDED);
 	}
 	return status;
@@ -1687,7 +1692,8 @@ hub_port_resume(struct usb_hub *hub, int port1, struct usb_device *udev)
 
 		/* drive resume for at least 20 msec */
 		if (udev)
-			dev_dbg(&udev->dev, "RESUME\n");
+			dev_dbg(&udev->dev, "usb %sresume\n",
+					udev->auto_pm ? "auto-" : "");
 		msleep(25);
 
 #define LIVE_FLAGS	( USB_PORT_STAT_POWER \
@@ -1754,8 +1760,11 @@ int usb_port_resume(struct usb_device *udev)
 		// NOTE this fails if parent is also suspended...
 		status = hub_port_resume(hdev_to_hub(udev->parent),
 				udev->portnum, udev);
-	} else
+	} else {
+		dev_dbg(&udev->dev, "usb %sresume\n",
+				udev->auto_pm ? "auto-" : "");
 		status = finish_port_resume(udev);
+	}
 	if (status < 0)
 		dev_dbg(&udev->dev, "can't resume, status %d\n", status);
 	return status;
@@ -1765,19 +1774,23 @@ static int remote_wakeup(struct usb_device *udev)
 {
 	int	status = 0;
 
-	/* don't repeat RESUME sequence if this device
-	 * was already woken up by some other task
-	 */
+	/* All this just to avoid sending a port-resume message
+	 * to the parent hub! */
+
 	usb_lock_device(udev);
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
 	if (udev->state == USB_STATE_SUSPENDED) {
-		dev_dbg(&udev->dev, "RESUME (wakeup)\n");
+		dev_dbg(&udev->dev, "usb %sresume\n", "wakeup-");
 		/* TRSMRCY = 10 msec */
 		msleep(10);
 		status = finish_port_resume(udev);
+		if (status == 0)
+			udev->dev.power.power_state.event = PM_EVENT_ON;
 	}
+	mutex_unlock(&udev->pm_mutex);
 
 	if (status == 0)
-		usb_resume_both(udev);
+		usb_autoresume_device(udev, 0);
 	usb_unlock_device(udev);
 	return status;
 }
@@ -1834,7 +1847,9 @@ static int hub_suspend(struct usb_interface *intf, pm_message_t msg)
 					== PM_EVENT_ON
 #endif
 				) {
-			dev_dbg(&intf->dev, "port %d nyet suspended\n", port1);
+			if (!hdev->auto_pm)
+				dev_dbg(&intf->dev, "port %d nyet suspended\n",
+						port1);
 			return -EBUSY;
 		}
 	}
@@ -2587,7 +2602,7 @@ static void hub_events(void)
 		 * stub "device" node was never suspended.
 		 */
 		if (i)
-			usb_resume_both(hdev);
+			usb_autoresume_device(hdev, 0);
 
 		/* If this is an inactive or suspended hub, do nothing */
 		if (hub->quiescing)
@@ -2993,6 +3008,9 @@ int usb_reset_composite_device(struct usb_device *udev,
 		return -EINVAL;
 	}
 
+	/* Prevent autosuspend during the reset */
+	usb_autoresume_device(udev, 1);
+
 	if (iface && iface->condition != USB_INTERFACE_BINDING)
 		iface = NULL;
 
@@ -3034,6 +3052,7 @@ int usb_reset_composite_device(struct usb_device *udev,
 		}
 	}
 
+	usb_autosuspend_device(udev, 1);
 	return ret;
 }
 EXPORT_SYMBOL(usb_reset_composite_device);
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index b0c0a993338f..6b029cdb8671 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -168,6 +168,10 @@ static void usb_release_dev(struct device *dev)
 
 	udev = to_usb_device(dev);
 
+#ifdef	CONFIG_PM
+	cancel_delayed_work(&udev->autosuspend);
+	flush_scheduled_work();
+#endif
 	usb_destroy_configuration(udev);
 	usb_put_hcd(bus_to_hcd(udev->bus));
 	kfree(udev->product);
@@ -176,6 +180,21 @@ static void usb_release_dev(struct device *dev)
 	kfree(udev);
 }
 
+#ifdef	CONFIG_PM
+
+/* usb_autosuspend_work - callback routine to autosuspend a USB device */
+static void usb_autosuspend_work(void *_udev)
+{
+	struct usb_device	*udev = _udev;
+
+	mutex_lock_nested(&udev->pm_mutex, udev->level);
+	udev->auto_pm = 1;
+	usb_suspend_both(udev, PMSG_SUSPEND);
+	mutex_unlock(&udev->pm_mutex);
+}
+
+#endif
+
 /**
  * usb_alloc_dev - usb device constructor (usbcore-internal)
  * @parent: hub to which device is connected; null to allocate a root hub
@@ -251,6 +270,10 @@ usb_alloc_dev(struct usb_device *parent, struct usb_bus *bus, unsigned port1)
 	dev->parent = parent;
 	INIT_LIST_HEAD(&dev->filelist);
 
+#ifdef	CONFIG_PM
+	mutex_init(&dev->pm_mutex);
+	INIT_WORK(&dev->autosuspend, usb_autosuspend_work, dev);
+#endif
 	return dev;
 }
 
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 5162cb370215..10688ad73c6d 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -49,6 +49,20 @@ static inline int usb_resume_both(struct usb_device *udev)
 
 #endif
 
+#ifdef CONFIG_USB_SUSPEND
+
+#define USB_AUTOSUSPEND_DELAY	(HZ*2)
+
+extern void usb_autosuspend_device(struct usb_device *udev, int dec_busy_cnt);
+extern int usb_autoresume_device(struct usb_device *udev, int inc_busy_cnt);
+
+#else
+
+#define usb_autosuspend_device(udev, dec_busy_cnt)	do {} while (0)
+#define usb_autoresume_device(udev, inc_busy_cnt)	0
+
+#endif
+
 extern struct bus_type usb_bus_type;
 extern struct usb_device_driver usb_generic_driver;
 
diff --git a/include/linux/usb.h b/include/linux/usb.h
index df5c93eb3ce9..0da15b0b02be 100644
--- a/include/linux/usb.h
+++ b/include/linux/usb.h
@@ -19,6 +19,7 @@
 #include <linux/fs.h>		/* for struct file_operations */
 #include <linux/completion.h>	/* for struct completion */
 #include <linux/sched.h>	/* for current && schedule_timeout */
+#include <linux/mutex.h>	/* for struct mutex */
 
 struct usb_device;
 struct usb_driver;
@@ -103,8 +104,12 @@ enum usb_interface_condition {
  * @condition: binding state of the interface: not bound, binding
  *	(in probe()), bound to a driver, or unbinding (in disconnect())
  * @is_active: flag set when the interface is bound and not suspended.
+ * @needs_remote_wakeup: flag set when the driver requires remote-wakeup
+ *	capability during autosuspend.
  * @dev: driver model's view of this device
  * @class_dev: driver model's class view of this device.
+ * @pm_usage_cnt: PM usage counter for this interface; autosuspend is not
+ *	allowed unless the counter is 0.
  *
  * USB device drivers attach to interfaces on a physical device.  Each
  * interface encapsulates a single high level function, such as feeding
@@ -144,9 +149,11 @@ struct usb_interface {
 					 * bound to */
 	enum usb_interface_condition condition;		/* state of binding */
 	unsigned is_active:1;		/* the interface is not suspended */
+	unsigned needs_remote_wakeup:1;	/* driver requires remote wakeup */
 
 	struct device dev;		/* interface specific device info */
 	struct class_device *class_dev;
+	int pm_usage_cnt;		/* usage counter for autosuspend */
 };
 #define	to_usb_interface(d) container_of(d, struct usb_interface, dev)
 #define	interface_to_usbdev(intf) \
@@ -372,6 +379,15 @@ struct usb_device {
 
 	int maxchild;			/* Number of ports if hub */
 	struct usb_device *children[USB_MAXCHILDREN];
+
+#ifdef CONFIG_PM
+	struct work_struct autosuspend;	/* for delayed autosuspends */
+	struct mutex pm_mutex;		/* protects PM operations */
+	int pm_usage_cnt;		/* usage counter for autosuspend */
+
+	unsigned auto_pm:1;		/* autosuspend/resume in progress */
+	unsigned do_remote_wakeup:1;	/* remote wakeup should be enabled */
+#endif
 };
 #define	to_usb_device(d) container_of(d, struct usb_device, dev)
 
@@ -392,6 +408,17 @@ extern int usb_reset_composite_device(struct usb_device *dev,
 
 extern struct usb_device *usb_find_device(u16 vendor_id, u16 product_id);
 
+/* USB autosuspend and autoresume */
+#ifdef CONFIG_USB_SUSPEND
+extern int usb_autopm_get_interface(struct usb_interface *intf);
+extern void usb_autopm_put_interface(struct usb_interface *intf);
+
+#else
+#define usb_autopm_get_interface(intf)		0
+#define usb_autopm_put_interface(intf)		do {} while (0)
+#endif
+
+
 /*-------------------------------------------------------------------------*/
 
 /* for drivers using iso endpoints */
@@ -593,6 +620,8 @@ struct usbdrv_wrap {
  * @drvwrap: Driver-model core structure wrapper.
  * @no_dynamic_id: if set to 1, the USB core will not allow dynamic ids to be
  *	added to this driver by preventing the sysfs file from being created.
+ * @supports_autosuspend: if set to 0, the USB core will not allow autosuspend
+ *	for interfaces bound to this driver.
  *
  * USB interface drivers must provide a name, probe() and disconnect()
  * methods, and an id_table.  Other driver fields are optional.
@@ -631,6 +660,7 @@ struct usb_driver {
 	struct usb_dynids dynids;
 	struct usbdrv_wrap drvwrap;
 	unsigned int no_dynamic_id:1;
+	unsigned int supports_autosuspend:1;
 };
 #define	to_usb_driver(d) container_of(d, struct usb_driver, drvwrap.driver)
 
@@ -648,6 +678,8 @@ struct usb_driver {
  * @suspend: Called when the device is going to be suspended by the system.
  * @resume: Called when the device is being resumed by the system.
  * @drvwrap: Driver-model core structure wrapper.
+ * @supports_autosuspend: if set to 0, the USB core will not allow autosuspend
+ *	for devices bound to this driver.
  *
  * USB drivers must provide all the fields listed above except drvwrap.
  */
@@ -660,6 +692,7 @@ struct usb_device_driver {
 	int (*suspend) (struct usb_device *udev, pm_message_t message);
 	int (*resume) (struct usb_device *udev);
 	struct usbdrv_wrap drvwrap;
+	unsigned int supports_autosuspend:1;
 };
 #define	to_usb_device_driver(d) container_of(d, struct usb_device_driver, \
 		drvwrap.driver)
-- 
cgit v1.2.1


From 01d883d44a1ca8dc77486635d428cba63e7fdadf Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:47:18 -0400
Subject: usbcore: non-hub-specific uses of autosuspend

This patch (as741) makes the non-hub parts of usbcore actually use the
autosuspend facilities added by an earlier patch.

	Devices opened through usbfs are autoresumed and then
	autosuspended upon close.

	Likewise for usb-skeleton.

	Devices are autoresumed for usb_set_configuration.


Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devio.c   | 11 ++++++++---
 drivers/usb/core/generic.c |  7 ++-----
 drivers/usb/core/message.c | 10 +++++++---
 drivers/usb/usb-skeleton.c | 11 +++++++++++
 4 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 43c08724a2d8..fd345ad810f8 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -558,10 +558,12 @@ static int usbdev_open(struct inode *inode, struct file *file)
 		dev = usbdev_lookup_minor(iminor(inode));
 	if (!dev)
 		dev = inode->i_private;
-	if (!dev) {
-		kfree(ps);
+	if (!dev)
 		goto out;
-	}
+	ret = usb_autoresume_device(dev, 1);
+	if (ret)
+		goto out;
+
 	usb_get_dev(dev);
 	ret = 0;
 	ps->dev = dev;
@@ -581,6 +583,8 @@ static int usbdev_open(struct inode *inode, struct file *file)
 	list_add_tail(&ps->list, &dev->filelist);
 	file->private_data = ps;
  out:
+	if (ret)
+		kfree(ps);
 	mutex_unlock(&usbfs_mutex);
 	return ret;
 }
@@ -604,6 +608,7 @@ static int usbdev_release(struct inode *inode, struct file *file)
 			releaseintf(ps, ifnum);
 	}
 	destroy_all_async(ps);
+	usb_autosuspend_device(dev, 1);
 	usb_unlock_device(dev);
 	usb_put_dev(dev);
 	kfree(ps);
diff --git a/drivers/usb/core/generic.c b/drivers/usb/core/generic.c
index 5358e656477c..16332cc57946 100644
--- a/drivers/usb/core/generic.c
+++ b/drivers/usb/core/generic.c
@@ -172,14 +172,10 @@ static void generic_disconnect(struct usb_device *udev)
 
 	/* if this is only an unbind, not a physical disconnect, then
 	 * unconfigure the device */
-	if (udev->state == USB_STATE_CONFIGURED)
+	if (udev->actconfig)
 		usb_set_configuration(udev, 0);
 
 	usb_remove_sysfs_dev_files(udev);
-
-	/* in case the call failed or the device was suspended */
-	if (udev->state >= USB_STATE_CONFIGURED)
-		usb_disable_device(udev, 0);
 }
 
 #ifdef	CONFIG_PM
@@ -208,4 +204,5 @@ struct usb_device_driver usb_generic_driver = {
 	.suspend = generic_suspend,
 	.resume = generic_resume,
 #endif
+	.supports_autosuspend = 1,
 };
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 1580c81a0db7..28c6cf225780 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -1366,9 +1366,6 @@ int usb_set_configuration(struct usb_device *dev, int configuration)
 	if (cp && configuration == 0)
 		dev_warn(&dev->dev, "config 0 descriptor??\n");
 
-	if (dev->state == USB_STATE_SUSPENDED)
-		return -EHOSTUNREACH;
-
 	/* Allocate memory for new interfaces before doing anything else,
 	 * so that if we run out then nothing will have changed. */
 	n = nintf = 0;
@@ -1403,6 +1400,11 @@ free_interfaces:
 					configuration, -i);
 	}
 
+	/* Wake up the device so we can send it the Set-Config request */
+	ret = usb_autoresume_device(dev, 1);
+	if (ret)
+		goto free_interfaces;
+
 	/* if it's already configured, clear out old state first.
 	 * getting rid of old interfaces means unbinding their drivers.
 	 */
@@ -1422,6 +1424,7 @@ free_interfaces:
 	dev->actconfig = cp;
 	if (!cp) {
 		usb_set_device_state(dev, USB_STATE_ADDRESS);
+		usb_autosuspend_device(dev, 1);
 		goto free_interfaces;
 	}
 	usb_set_device_state(dev, USB_STATE_CONFIGURED);
@@ -1490,6 +1493,7 @@ free_interfaces:
 		usb_create_sysfs_intf_files (intf);
 	}
 
+	usb_autosuspend_device(dev, 1);
 	return 0;
 }
 
diff --git a/drivers/usb/usb-skeleton.c b/drivers/usb/usb-skeleton.c
index 9b542a6ba978..1b51d3187a95 100644
--- a/drivers/usb/usb-skeleton.c
+++ b/drivers/usb/usb-skeleton.c
@@ -90,6 +90,11 @@ static int skel_open(struct inode *inode, struct file *file)
 		goto exit;
 	}
 
+	/* prevent the device from being autosuspended */
+	retval = usb_autopm_get_interface(interface);
+	if (retval)
+		goto exit;
+
 	/* increment our usage count for the device */
 	kref_get(&dev->kref);
 
@@ -108,6 +113,12 @@ static int skel_release(struct inode *inode, struct file *file)
 	if (dev == NULL)
 		return -ENODEV;
 
+	/* allow the device to be autosuspended */
+	mutex_lock(&dev->io_mutex);
+	if (dev->interface)
+		usb_autopm_put_interface(dev->interface);
+	mutex_unlock(&dev->io_mutex);
+
 	/* decrement the count on our device */
 	kref_put(&dev->kref, skel_delete);
 	return 0;
-- 
cgit v1.2.1


From 02c399ee45a54987c152fe5f627ed949bb55f187 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 30 Aug 2006 15:47:11 -0400
Subject: usbcore: remove usb_suspend_root_hub

This patch (as740) removes the existing support for autosuspend of
root hubs.  That support fit in rather awkwardly with the rest of
usbcore and it was used only by ohci-hcd.  It won't be needed any more
since the hub driver will take care of autosuspending all hubs, root
or external.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hcd.c      | 44 --------------------------------------------
 drivers/usb/core/hcd.h      |  6 ------
 drivers/usb/core/hub.c      | 18 +-----------------
 drivers/usb/core/usb.h      |  1 -
 drivers/usb/host/ohci-hub.c |  3 ---
 5 files changed, 1 insertion(+), 71 deletions(-)

diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
index 9dfc812de034..e86f62957085 100644
--- a/drivers/usb/core/hcd.c
+++ b/drivers/usb/core/hcd.c
@@ -1431,50 +1431,6 @@ int hcd_bus_resume (struct usb_bus *bus)
 	return status;
 }
 
-/*
- * usb_hcd_suspend_root_hub - HCD autosuspends downstream ports
- * @hcd: host controller for this root hub
- *
- * This call arranges that usb_hcd_resume_root_hub() is safe to call later;
- * that the HCD's root hub polling is deactivated; and that the root's hub
- * driver is suspended.  HCDs may call this to autosuspend when their root
- * hub's downstream ports are all inactive:  unpowered, disconnected,
- * disabled, or suspended.
- *
- * The HCD will autoresume on device connect change detection (using SRP
- * or a D+/D- pullup).  The HCD also autoresumes on remote wakeup signaling
- * from any ports that are suspended (if that is enabled).  In most cases,
- * overcurrent signaling (on powered ports) will also start autoresume.
- *
- * Always called with IRQs blocked.
- */
-void usb_hcd_suspend_root_hub (struct usb_hcd *hcd)
-{
-	struct urb	*urb;
-
-	spin_lock (&hcd_root_hub_lock);
-	usb_suspend_root_hub (hcd->self.root_hub);
-
-	/* force status urb to complete/unlink while suspended */
-	if (hcd->status_urb) {
-		urb = hcd->status_urb;
-		urb->status = -ECONNRESET;
-		urb->hcpriv = NULL;
-		urb->actual_length = 0;
-
-		del_timer (&hcd->rh_timer);
-		hcd->poll_pending = 0;
-		hcd->status_urb = NULL;
-	} else
-		urb = NULL;
-	spin_unlock (&hcd_root_hub_lock);
-	hcd->state = HC_STATE_SUSPENDED;
-
-	if (urb)
-		usb_hcd_giveback_urb (hcd, urb, NULL);
-}
-EXPORT_SYMBOL_GPL(usb_hcd_suspend_root_hub);
-
 /**
  * usb_hcd_resume_root_hub - called by HCD to resume its root hub 
  * @hcd: host controller for this root hub
diff --git a/drivers/usb/core/hcd.h b/drivers/usb/core/hcd.h
index 7a2bcba2ae61..676877c15f81 100644
--- a/drivers/usb/core/hcd.h
+++ b/drivers/usb/core/hcd.h
@@ -368,17 +368,11 @@ extern int usb_find_interface_driver (struct usb_device *dev,
 #define usb_endpoint_out(ep_dir)	(!((ep_dir) & USB_DIR_IN))
 
 #ifdef CONFIG_PM
-extern void usb_hcd_suspend_root_hub (struct usb_hcd *hcd);
 extern void usb_hcd_resume_root_hub (struct usb_hcd *hcd);
 extern void usb_root_hub_lost_power (struct usb_device *rhdev);
 extern int hcd_bus_suspend (struct usb_bus *bus);
 extern int hcd_bus_resume (struct usb_bus *bus);
 #else
-static inline void usb_hcd_suspend_root_hub(struct usb_hcd *hcd)
-{
-	return;
-}
-
 static inline void usb_hcd_resume_root_hub(struct usb_hcd *hcd)
 {
 	return;
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index dee812bc6c43..7479bd329abb 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -452,18 +452,14 @@ static void hub_power_on(struct usb_hub *hub)
 	msleep(max(pgood_delay, (unsigned) 100));
 }
 
-static inline void __hub_quiesce(struct usb_hub *hub)
+static void hub_quiesce(struct usb_hub *hub)
 {
 	/* (nonblocking) khubd and related activity won't re-trigger */
 	hub->quiescing = 1;
 	hub->activating = 0;
 	hub->resume_root_hub = 0;
-}
 
-static void hub_quiesce(struct usb_hub *hub)
-{
 	/* (blocking) stop khubd and related activity */
-	__hub_quiesce(hub);
 	usb_kill_urb(hub->urb);
 	if (hub->has_indicators)
 		cancel_delayed_work(&hub->leds);
@@ -1914,18 +1910,6 @@ static inline int remote_wakeup(struct usb_device *udev)
 #define hub_resume NULL
 #endif
 
-void usb_suspend_root_hub(struct usb_device *hdev)
-{
-	struct usb_hub *hub = hdev_to_hub(hdev);
-
-	/* This also makes any led blinker stop retriggering.  We're called
-	 * from irq, so the blinker might still be scheduled.  Caller promises
-	 * that the root hub status URB will be canceled.
-	 */
-	__hub_quiesce(hub);
-	mark_quiesced(to_usb_interface(hub->intfdev));
-}
-
 void usb_resume_root_hub(struct usb_device *hdev)
 {
 	struct usb_hub *hub = hdev_to_hub(hdev);
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index 10688ad73c6d..e8bc2e48913a 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -20,7 +20,6 @@ extern char *usb_cache_string(struct usb_device *udev, int index);
 extern int usb_set_configuration(struct usb_device *dev, int configuration);
 
 extern void usb_kick_khubd(struct usb_device *dev);
-extern void usb_suspend_root_hub(struct usb_device *hdev);
 extern void usb_resume_root_hub(struct usb_device *dev);
 
 extern int  usb_hub_init(void);
diff --git a/drivers/usb/host/ohci-hub.c b/drivers/usb/host/ohci-hub.c
index f1b1ed086bde..0b899339cac8 100644
--- a/drivers/usb/host/ohci-hub.c
+++ b/drivers/usb/host/ohci-hub.c
@@ -135,9 +135,6 @@ static int ohci_bus_suspend (struct usb_hcd *hcd)
 	hcd->poll_rh = 0;
 
 done:
-	/* external suspend vs self autosuspend ... same effect */
-	if (status == 0)
-		usb_hcd_suspend_root_hub(hcd);
 	spin_unlock_irqrestore (&ohci->lock, flags);
 	return status;
 }
-- 
cgit v1.2.1


From 592fbbe4bc339399d363dd55f0391e0623400706 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Tue, 19 Sep 2006 10:08:43 -0400
Subject: USB: fix root-hub resume when CONFIG_USB_SUSPEND is not set

This patch (as786) removes a redundant test and fixes a problem
involving repeated system sleeps when CONFIG_USB_SUSPEND is not set.
During the first wakeup, the root hub's dev.power.power_state.event
field doesn't get updated, causing it not to be suspended during the
second sleep transition.

This takes care of the issue raised by Rafael J. Wysocki and Mattia
Dongili.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/driver.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 2b2000ac05ab..0d063c8ca4b4 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -1071,10 +1071,15 @@ int usb_resume_both(struct usb_device *udev)
 					PM_EVENT_ON)
 				status = -EHOSTUNREACH;
 		}
-		if (status == 0 && udev->state == USB_STATE_SUSPENDED)
+		if (status == 0)
 			status = resume_device(udev);
 		if (parent)
 			mutex_unlock(&parent->pm_mutex);
+	} else {
+
+		/* Needed only for setting udev->dev.power.power_state.event
+		 * and for possible debugging message. */
+		status = resume_device(udev);
 	}
 
 	/* Now the parent won't suspend until we are finished */
-- 
cgit v1.2.1


From 1b21d5e166e104f8914441ef52e2cd50ce65b479 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/core/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devio.c    |  2 +-
 drivers/usb/core/driver.c   | 10 ++++++---
 drivers/usb/core/endpoint.c | 30 ++++++++++++++++++-------
 drivers/usb/core/sysfs.c    | 55 +++++++++++++++++++++++++++++++++++----------
 drivers/usb/core/usb.c      |  6 +++--
 drivers/usb/core/usb.h      |  6 ++---
 6 files changed, 80 insertions(+), 29 deletions(-)

diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index fd345ad810f8..71bbd25a4ed0 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -1359,7 +1359,7 @@ static int proc_ioctl(struct dev_state *ps, struct usbdevfs_ioctl *ctl)
 	/* let kernel drivers try to (re)bind to the interface */
 	case USBDEVFS_CONNECT:
 		usb_unlock_device(ps->dev);
-		bus_rescan_devices(intf->dev.bus);
+		retval = bus_rescan_devices(intf->dev.bus);
 		usb_lock_device(ps->dev);
 		break;
 
diff --git a/drivers/usb/core/driver.c b/drivers/usb/core/driver.c
index 0d063c8ca4b4..b10463244413 100644
--- a/drivers/usb/core/driver.c
+++ b/drivers/usb/core/driver.c
@@ -49,6 +49,7 @@ static ssize_t store_new_id(struct device_driver *driver,
 	u32 idVendor = 0;
 	u32 idProduct = 0;
 	int fields = 0;
+	int retval = 0;
 
 	fields = sscanf(buf, "%x %x", &idVendor, &idProduct);
 	if (fields < 2)
@@ -68,10 +69,12 @@ static ssize_t store_new_id(struct device_driver *driver,
 	spin_unlock(&usb_drv->dynids.lock);
 
 	if (get_driver(driver)) {
-		driver_attach(driver);
+		retval = driver_attach(driver);
 		put_driver(driver);
 	}
 
+	if (retval)
+		return retval;
 	return count;
 }
 static DRIVER_ATTR(new_id, S_IWUSR, NULL, store_new_id);
@@ -291,6 +294,7 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 {
 	struct device *dev = &iface->dev;
 	struct usb_device *udev = interface_to_usbdev(iface);
+	int retval = 0;
 
 	if (dev->driver)
 		return -EBUSY;
@@ -308,9 +312,9 @@ int usb_driver_claim_interface(struct usb_driver *driver,
 	 * the future device_add() bind it, bypassing probe()
 	 */
 	if (device_is_registered(dev))
-		device_bind_driver(dev);
+		retval = device_bind_driver(dev);
 
-	return 0;
+	return retval;
 }
 EXPORT_SYMBOL(usb_driver_claim_interface);
 
diff --git a/drivers/usb/core/endpoint.c b/drivers/usb/core/endpoint.c
index 247b5a4913a8..3ebb90149e93 100644
--- a/drivers/usb/core/endpoint.c
+++ b/drivers/usb/core/endpoint.c
@@ -207,9 +207,9 @@ static void ep_device_release(struct device *dev)
 	kfree(ep_dev);
 }
 
-void usb_create_ep_files(struct device *parent,
-			 struct usb_host_endpoint *endpoint,
-			 struct usb_device *udev)
+int usb_create_ep_files(struct device *parent,
+			struct usb_host_endpoint *endpoint,
+			struct usb_device *udev)
 {
 	char name[8];
 	struct ep_device *ep_dev;
@@ -242,19 +242,33 @@ void usb_create_ep_files(struct device *parent,
 	retval = device_register(&ep_dev->dev);
 	if (retval)
 		goto error;
-	sysfs_create_group(&ep_dev->dev.kobj, &ep_dev_attr_grp);
+	retval = sysfs_create_group(&ep_dev->dev.kobj, &ep_dev_attr_grp);
+	if (retval)
+		goto error_group;
 
 	endpoint->ep_dev = ep_dev;
 
 	/* create the symlink to the old-style "ep_XX" directory */
 	sprintf(name, "ep_%02x", endpoint->desc.bEndpointAddress);
-	sysfs_create_link(&parent->kobj, &endpoint->ep_dev->dev.kobj, name);
-
+	retval = sysfs_create_link(&parent->kobj,
+				   &endpoint->ep_dev->dev.kobj, name);
+	if (retval)
+		goto error_link;
 exit:
-	return;
+	return retval;
+
+error_link:
+	sysfs_remove_group(&ep_dev->dev.kobj, &ep_dev_attr_grp);
+
+error_group:
+	device_unregister(&ep_dev->dev);
+	endpoint->ep_dev = NULL;
+	destroy_endpoint_class();
+	return retval;
 error:
 	kfree(ep_dev);
-	return;
+	destroy_endpoint_class();
+	return retval;
 }
 
 void usb_remove_ep_files(struct usb_host_endpoint *endpoint)
diff --git a/drivers/usb/core/sysfs.c b/drivers/usb/core/sysfs.c
index cd2286246f6a..55d8f575206d 100644
--- a/drivers/usb/core/sysfs.c
+++ b/drivers/usb/core/sysfs.c
@@ -210,19 +210,40 @@ static struct attribute_group dev_attr_grp = {
 	.attrs = dev_attrs,
 };
 
-void usb_create_sysfs_dev_files (struct usb_device *udev)
+int usb_create_sysfs_dev_files(struct usb_device *udev)
 {
 	struct device *dev = &udev->dev;
+	int retval;
 
-	sysfs_create_group(&dev->kobj, &dev_attr_grp);
+	retval = sysfs_create_group(&dev->kobj, &dev_attr_grp);
+	if (retval)
+		return retval;
 
-	if (udev->manufacturer)
-		device_create_file (dev, &dev_attr_manufacturer);
-	if (udev->product)
-		device_create_file (dev, &dev_attr_product);
-	if (udev->serial)
-		device_create_file (dev, &dev_attr_serial);
-	usb_create_ep_files(dev, &udev->ep0, udev);
+	if (udev->manufacturer) {
+		retval = device_create_file (dev, &dev_attr_manufacturer);
+		if (retval)
+			goto error;
+	}
+	if (udev->product) {
+		retval = device_create_file (dev, &dev_attr_product);
+		if (retval)
+			goto error;
+	}
+	if (udev->serial) {
+		retval = device_create_file (dev, &dev_attr_serial);
+		if (retval)
+			goto error;
+	}
+	retval = usb_create_ep_files(dev, &udev->ep0, udev);
+	if (retval)
+		goto error;
+	return 0;
+error:
+	usb_remove_ep_files(&udev->ep0);
+	device_remove_file(dev, &dev_attr_manufacturer);
+	device_remove_file(dev, &dev_attr_product);
+	device_remove_file(dev, &dev_attr_serial);
+	return retval;
 }
 
 void usb_remove_sysfs_dev_files (struct usb_device *udev)
@@ -339,18 +360,28 @@ static inline void usb_remove_intf_ep_files(struct usb_interface *intf)
 		usb_remove_ep_files(&iface_desc->endpoint[i]);
 }
 
-void usb_create_sysfs_intf_files (struct usb_interface *intf)
+int usb_create_sysfs_intf_files(struct usb_interface *intf)
 {
 	struct usb_device *udev = interface_to_usbdev(intf);
 	struct usb_host_interface *alt = intf->cur_altsetting;
+	int retval;
 
-	sysfs_create_group(&intf->dev.kobj, &intf_attr_grp);
+	retval = sysfs_create_group(&intf->dev.kobj, &intf_attr_grp);
+	if (retval)
+		goto error;
 
 	if (alt->string == NULL)
 		alt->string = usb_cache_string(udev, alt->desc.iInterface);
 	if (alt->string)
-		device_create_file(&intf->dev, &dev_attr_interface);
+		retval = device_create_file(&intf->dev, &dev_attr_interface);
 	usb_create_intf_ep_files(intf, udev);
+	return 0;
+error:
+	if (alt->string)
+		device_remove_file(&intf->dev, &dev_attr_interface);
+	sysfs_remove_group(&intf->dev.kobj, &intf_attr_grp);
+	usb_remove_intf_ep_files(intf);
+	return retval;
 }
 
 void usb_remove_sysfs_intf_files (struct usb_interface *intf)
diff --git a/drivers/usb/core/usb.c b/drivers/usb/core/usb.c
index 6b029cdb8671..60ef4ef0101a 100644
--- a/drivers/usb/core/usb.c
+++ b/drivers/usb/core/usb.c
@@ -147,11 +147,13 @@ static int __find_interface(struct device * dev, void * data)
 struct usb_interface *usb_find_interface(struct usb_driver *drv, int minor)
 {
 	struct find_interface_arg argb;
+	int retval;
 
 	argb.minor = minor;
 	argb.interface = NULL;
-	driver_for_each_device(&drv->drvwrap.driver, NULL, &argb,
-			__find_interface);
+	/* eat the error, it will be in argb.interface */
+	retval = driver_for_each_device(&drv->drvwrap.driver, NULL, &argb,
+					__find_interface);
 	return argb.interface;
 }
 
diff --git a/drivers/usb/core/usb.h b/drivers/usb/core/usb.h
index e8bc2e48913a..0c09ecced6e1 100644
--- a/drivers/usb/core/usb.h
+++ b/drivers/usb/core/usb.h
@@ -1,10 +1,10 @@
 /* Functions local to drivers/usb/core/ */
 
-extern void usb_create_sysfs_dev_files (struct usb_device *dev);
+extern int usb_create_sysfs_dev_files (struct usb_device *dev);
 extern void usb_remove_sysfs_dev_files (struct usb_device *dev);
-extern void usb_create_sysfs_intf_files (struct usb_interface *intf);
+extern int usb_create_sysfs_intf_files (struct usb_interface *intf);
 extern void usb_remove_sysfs_intf_files (struct usb_interface *intf);
-extern void usb_create_ep_files(struct device *parent, struct usb_host_endpoint *endpoint,
+extern int usb_create_ep_files(struct device *parent, struct usb_host_endpoint *endpoint,
 				struct usb_device *udev);
 extern void usb_remove_ep_files(struct usb_host_endpoint *endpoint);
 
-- 
cgit v1.2.1


From 4d42e1bb9e8ec7eb5e39d82aaf3ff2f3c994af84 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/misc/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/cypress_cy7c63.c | 19 +++++++++++++++----
 drivers/usb/misc/cytherm.c        | 35 ++++++++++++++++++++++++++---------
 drivers/usb/misc/usbled.c         | 20 ++++++++++++++++----
 3 files changed, 57 insertions(+), 17 deletions(-)

diff --git a/drivers/usb/misc/cypress_cy7c63.c b/drivers/usb/misc/cypress_cy7c63.c
index 9c46746d5d00..b63b5f34b2aa 100644
--- a/drivers/usb/misc/cypress_cy7c63.c
+++ b/drivers/usb/misc/cypress_cy7c63.c
@@ -209,7 +209,7 @@ static int cypress_probe(struct usb_interface *interface,
 	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
 	if (dev == NULL) {
 		dev_err(&interface->dev, "Out of memory!\n");
-		goto error;
+		goto error_mem;
 	}
 
 	dev->udev = usb_get_dev(interface_to_usbdev(interface));
@@ -218,15 +218,26 @@ static int cypress_probe(struct usb_interface *interface,
 	usb_set_intfdata(interface, dev);
 
 	/* create device attribute files */
-	device_create_file(&interface->dev, &dev_attr_port0);
-	device_create_file(&interface->dev, &dev_attr_port1);
+	retval = device_create_file(&interface->dev, &dev_attr_port0);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_port1);
+	if (retval)
+		goto error;
 
 	/* let the user know that the device is now attached */
 	dev_info(&interface->dev,
 		 "Cypress CY7C63xxx device now attached\n");
+	return 0;
 
-	retval = 0;
 error:
+	device_remove_file(&interface->dev, &dev_attr_port0);
+	device_remove_file(&interface->dev, &dev_attr_port1);
+	usb_set_intfdata(interface, NULL);
+	usb_put_dev(dev->udev);
+	kfree(dev);
+
+error_mem:
 	return retval;
 }
 
diff --git a/drivers/usb/misc/cytherm.c b/drivers/usb/misc/cytherm.c
index b20bec445552..04e87acd6e46 100644
--- a/drivers/usb/misc/cytherm.c
+++ b/drivers/usb/misc/cytherm.c
@@ -353,7 +353,7 @@ static int cytherm_probe(struct usb_interface *interface,
 	dev = kzalloc (sizeof(struct usb_cytherm), GFP_KERNEL);
 	if (dev == NULL) {
 		dev_err (&interface->dev, "Out of memory\n");
-		goto error;
+		goto error_mem;
 	}
 
 	dev->udev = usb_get_dev(udev);
@@ -362,18 +362,35 @@ static int cytherm_probe(struct usb_interface *interface,
 
 	dev->brightness = 0xFF;
 
-	device_create_file(&interface->dev, &dev_attr_brightness);   
-	device_create_file(&interface->dev, &dev_attr_temp);
-	device_create_file(&interface->dev, &dev_attr_button);
-	device_create_file(&interface->dev, &dev_attr_port0);
-	device_create_file(&interface->dev, &dev_attr_port1);
+	retval = device_create_file(&interface->dev, &dev_attr_brightness);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_temp);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_button);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_port0);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_port1);
+	if (retval)
+		goto error;
 
-	dev_info (&interface->dev, 
+	dev_info (&interface->dev,
 		  "Cypress thermometer device now attached\n");
 	return 0;
-
- error:
+error:
+	device_remove_file(&interface->dev, &dev_attr_brightness);
+	device_remove_file(&interface->dev, &dev_attr_temp);
+	device_remove_file(&interface->dev, &dev_attr_button);
+	device_remove_file(&interface->dev, &dev_attr_port0);
+	device_remove_file(&interface->dev, &dev_attr_port1);
+	usb_set_intfdata (interface, NULL);
+	usb_put_dev(dev->udev);
 	kfree(dev);
+error_mem:
 	return retval;
 }
 
diff --git a/drivers/usb/misc/usbled.c b/drivers/usb/misc/usbled.c
index 0c5ee0ad6bb9..49c5c5c4c431 100644
--- a/drivers/usb/misc/usbled.c
+++ b/drivers/usb/misc/usbled.c
@@ -108,22 +108,34 @@ static int led_probe(struct usb_interface *interface, const struct usb_device_id
 	dev = kzalloc(sizeof(struct usb_led), GFP_KERNEL);
 	if (dev == NULL) {
 		dev_err(&interface->dev, "Out of memory\n");
-		goto error;
+		goto error_mem;
 	}
 
 	dev->udev = usb_get_dev(udev);
 
 	usb_set_intfdata (interface, dev);
 
-	device_create_file(&interface->dev, &dev_attr_blue);
-	device_create_file(&interface->dev, &dev_attr_red);
-	device_create_file(&interface->dev, &dev_attr_green);
+	retval = device_create_file(&interface->dev, &dev_attr_blue);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_red);
+	if (retval)
+		goto error;
+	retval = device_create_file(&interface->dev, &dev_attr_green);
+	if (retval)
+		goto error;
 
 	dev_info(&interface->dev, "USB LED device now attached\n");
 	return 0;
 
 error:
+	device_remove_file(&interface->dev, &dev_attr_blue);
+	device_remove_file(&interface->dev, &dev_attr_red);
+	device_remove_file(&interface->dev, &dev_attr_green);
+	usb_set_intfdata (interface, NULL);
+	usb_put_dev(dev->udev);
 	kfree(dev);
+error_mem:
 	return retval;
 }
 
-- 
cgit v1.2.1


From e7ccdfec087f02930c5cdc81143d4a045ae8d361 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/atm/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/atm/ueagle-atm.c | 74 ++++++++++++++++++++------------------------
 1 file changed, 34 insertions(+), 40 deletions(-)

diff --git a/drivers/usb/atm/ueagle-atm.c b/drivers/usb/atm/ueagle-atm.c
index b38990adf1cd..465961a26e4a 100644
--- a/drivers/usb/atm/ueagle-atm.c
+++ b/drivers/usb/atm/ueagle-atm.c
@@ -1621,26 +1621,32 @@ static int claim_interface(struct usb_device *usb_dev,
 	return ret;
 }
 
-static void create_fs_entries(struct uea_softc *sc, struct usb_interface *intf)
+static struct attribute *attrs[] = {
+	&dev_attr_stat_status.attr,
+	&dev_attr_stat_mflags.attr,
+	&dev_attr_stat_human_status.attr,
+	&dev_attr_stat_delin.attr,
+	&dev_attr_stat_vidcpe.attr,
+	&dev_attr_stat_usrate.attr,
+	&dev_attr_stat_dsrate.attr,
+	&dev_attr_stat_usattenuation.attr,
+	&dev_attr_stat_dsattenuation.attr,
+	&dev_attr_stat_usmargin.attr,
+	&dev_attr_stat_dsmargin.attr,
+	&dev_attr_stat_txflow.attr,
+	&dev_attr_stat_rxflow.attr,
+	&dev_attr_stat_uscorr.attr,
+	&dev_attr_stat_dscorr.attr,
+	&dev_attr_stat_usunc.attr,
+	&dev_attr_stat_dsunc.attr,
+};
+static struct attribute_group attr_grp = {
+	.attrs = attrs,
+};
+
+static int create_fs_entries(struct usb_interface *intf)
 {
-	/* sysfs interface */
-	device_create_file(&intf->dev, &dev_attr_stat_status);
-	device_create_file(&intf->dev, &dev_attr_stat_mflags);
-	device_create_file(&intf->dev, &dev_attr_stat_human_status);
-	device_create_file(&intf->dev, &dev_attr_stat_delin);
-	device_create_file(&intf->dev, &dev_attr_stat_vidcpe);
-	device_create_file(&intf->dev, &dev_attr_stat_usrate);
-	device_create_file(&intf->dev, &dev_attr_stat_dsrate);
-	device_create_file(&intf->dev, &dev_attr_stat_usattenuation);
-	device_create_file(&intf->dev, &dev_attr_stat_dsattenuation);
-	device_create_file(&intf->dev, &dev_attr_stat_usmargin);
-	device_create_file(&intf->dev, &dev_attr_stat_dsmargin);
-	device_create_file(&intf->dev, &dev_attr_stat_txflow);
-	device_create_file(&intf->dev, &dev_attr_stat_rxflow);
-	device_create_file(&intf->dev, &dev_attr_stat_uscorr);
-	device_create_file(&intf->dev, &dev_attr_stat_dscorr);
-	device_create_file(&intf->dev, &dev_attr_stat_usunc);
-	device_create_file(&intf->dev, &dev_attr_stat_dsunc);
+	return sysfs_create_group(&intf->dev.kobj, &attr_grp);
 }
 
 static int uea_bind(struct usbatm_data *usbatm, struct usb_interface *intf,
@@ -1708,37 +1714,25 @@ static int uea_bind(struct usbatm_data *usbatm, struct usb_interface *intf,
 		return ret;
 	}
 
-	create_fs_entries(sc, intf);
+	ret = create_fs_entries(intf);
+	if (ret) {
+		uea_stop(sc);
+		kfree(sc);
+		return ret;
+	}
 	return 0;
 }
 
-static void destroy_fs_entries(struct uea_softc *sc, struct usb_interface *intf)
+static void destroy_fs_entries(struct usb_interface *intf)
 {
-	/* sysfs interface */
-	device_remove_file(&intf->dev, &dev_attr_stat_status);
-	device_remove_file(&intf->dev, &dev_attr_stat_mflags);
-	device_remove_file(&intf->dev, &dev_attr_stat_human_status);
-	device_remove_file(&intf->dev, &dev_attr_stat_delin);
-	device_remove_file(&intf->dev, &dev_attr_stat_vidcpe);
-	device_remove_file(&intf->dev, &dev_attr_stat_usrate);
-	device_remove_file(&intf->dev, &dev_attr_stat_dsrate);
-	device_remove_file(&intf->dev, &dev_attr_stat_usattenuation);
-	device_remove_file(&intf->dev, &dev_attr_stat_dsattenuation);
-	device_remove_file(&intf->dev, &dev_attr_stat_usmargin);
-	device_remove_file(&intf->dev, &dev_attr_stat_dsmargin);
-	device_remove_file(&intf->dev, &dev_attr_stat_txflow);
-	device_remove_file(&intf->dev, &dev_attr_stat_rxflow);
-	device_remove_file(&intf->dev, &dev_attr_stat_uscorr);
-	device_remove_file(&intf->dev, &dev_attr_stat_dscorr);
-	device_remove_file(&intf->dev, &dev_attr_stat_usunc);
-	device_remove_file(&intf->dev, &dev_attr_stat_dsunc);
+	sysfs_remove_group(&intf->dev.kobj, &attr_grp);
 }
 
 static void uea_unbind(struct usbatm_data *usbatm, struct usb_interface *intf)
 {
 	struct uea_softc *sc = usbatm->driver_data;
 
-	destroy_fs_entries(sc, intf);
+	destroy_fs_entries(intf);
 	uea_stop(sc);
 	kfree(sc);
 }
-- 
cgit v1.2.1


From 96cede531c632ac019003bf40128b1821761a164 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/class/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/class/usblp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/class/usblp.c b/drivers/usb/class/usblp.c
index 9b05a359b9f7..9cac11ca1bb7 100644
--- a/drivers/usb/class/usblp.c
+++ b/drivers/usb/class/usblp.c
@@ -927,7 +927,9 @@ static int usblp_probe(struct usb_interface *intf,
 
 	/* Retrieve and store the device ID string. */
 	usblp_cache_device_id_string(usblp);
-	device_create_file(&intf->dev, &dev_attr_ieee1284_id);
+	retval = device_create_file(&intf->dev, &dev_attr_ieee1284_id);
+	if (retval)
+		goto abort_intfdata;
 
 #ifdef DEBUG
 	usblp_check_status(usblp, 0);
-- 
cgit v1.2.1


From 657b6717e7e7dc5620525a34f1561b563fd472fa Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/input/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/yealink.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/input/yealink.c b/drivers/usb/input/yealink.c
index 7b45fd3de911..7291e7a2717b 100644
--- a/drivers/usb/input/yealink.c
+++ b/drivers/usb/input/yealink.c
@@ -971,7 +971,7 @@ static int usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
 			DRIVER_VERSION, sizeof(DRIVER_VERSION));
 
 	/* Register sysfs hooks (don't care about failure) */
-	sysfs_create_group(&intf->dev.kobj, &yld_attr_group);
+	ret = sysfs_create_group(&intf->dev.kobj, &yld_attr_group);
 	return 0;
 }
 
-- 
cgit v1.2.1


From 1ee95216c0db6305c047a90b0822e2f1d2d5acdc Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/host/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ehci-dbg.c | 7 ++++---
 drivers/usb/host/ohci-dbg.c | 7 ++++---
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 9cd6270d06bc..34b7a31cd85b 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -785,10 +785,11 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL);
 static inline void create_debug_files (struct ehci_hcd *ehci)
 {
 	struct class_device *cldev = ehci_to_hcd(ehci)->self.class_dev;
+	int retval;
 
-	class_device_create_file(cldev, &class_device_attr_async);
-	class_device_create_file(cldev, &class_device_attr_periodic);
-	class_device_create_file(cldev, &class_device_attr_registers);
+	retval = class_device_create_file(cldev, &class_device_attr_async);
+	retval = class_device_create_file(cldev, &class_device_attr_periodic);
+	retval = class_device_create_file(cldev, &class_device_attr_registers);
 }
 
 static inline void remove_debug_files (struct ehci_hcd *ehci)
diff --git a/drivers/usb/host/ohci-dbg.c b/drivers/usb/host/ohci-dbg.c
index 534d07dcb824..8293c1d4be3f 100644
--- a/drivers/usb/host/ohci-dbg.c
+++ b/drivers/usb/host/ohci-dbg.c
@@ -685,10 +685,11 @@ static CLASS_DEVICE_ATTR (registers, S_IRUGO, show_registers, NULL);
 static inline void create_debug_files (struct ohci_hcd *ohci)
 {
 	struct class_device *cldev = ohci_to_hcd(ohci)->self.class_dev;
+	int retval;
 
-	class_device_create_file(cldev, &class_device_attr_async);
-	class_device_create_file(cldev, &class_device_attr_periodic);
-	class_device_create_file(cldev, &class_device_attr_registers);
+	retval = class_device_create_file(cldev, &class_device_attr_async);
+	retval = class_device_create_file(cldev, &class_device_attr_periodic);
+	retval = class_device_create_file(cldev, &class_device_attr_registers);
 	ohci_dbg (ohci, "created debug files\n");
 }
 
-- 
cgit v1.2.1


From 13f4db9e1bf0a6efcdbbb3a1e4da8a1a8c620fff Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Mon, 28 Aug 2006 11:43:25 -0700
Subject: USB: fix __must_check warnings in drivers/usb/serial/

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/ftdi_sio.c   | 30 +++++++++++++++++++-----------
 drivers/usb/serial/usb-serial.c |  5 ++++-
 2 files changed, 23 insertions(+), 12 deletions(-)

diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index c6115aa1b445..1f7b72553f37 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -1101,25 +1101,29 @@ static ssize_t store_event_char(struct device *dev, struct device_attribute *att
 static DEVICE_ATTR(latency_timer, S_IWUSR | S_IRUGO, show_latency_timer, store_latency_timer);
 static DEVICE_ATTR(event_char, S_IWUSR, NULL, store_event_char);
 
-static void create_sysfs_attrs(struct usb_serial *serial)
-{	
+static int create_sysfs_attrs(struct usb_serial *serial)
+{
 	struct ftdi_private *priv;
 	struct usb_device *udev;
+	int retval = 0;
 
 	dbg("%s",__FUNCTION__);
-	
+
 	priv = usb_get_serial_port_data(serial->port[0]);
 	udev = serial->dev;
-	
+
 	/* XXX I've no idea if the original SIO supports the event_char
 	 * sysfs parameter, so I'm playing it safe.  */
 	if (priv->chip_type != SIO) {
 		dbg("sysfs attributes for %s", ftdi_chip_name[priv->chip_type]);
-		device_create_file(&udev->dev, &dev_attr_event_char);
-		if (priv->chip_type == FT232BM || priv->chip_type == FT2232C) {
-			device_create_file(&udev->dev, &dev_attr_latency_timer);
+		retval = device_create_file(&udev->dev, &dev_attr_event_char);
+		if ((!retval) &&
+		    (priv->chip_type == FT232BM || priv->chip_type == FT2232C)) {
+			retval = device_create_file(&udev->dev,
+						    &dev_attr_latency_timer);
 		}
 	}
+	return retval;
 }
 
 static void remove_sysfs_attrs(struct usb_serial *serial)
@@ -1162,7 +1166,8 @@ static int ftdi_sio_attach (struct usb_serial *serial)
 	struct usb_serial_port *port = serial->port[0];
 	struct ftdi_private *priv;
 	struct ftdi_sio_quirk *quirk;
-	
+	int retval;
+
 	dbg("%s",__FUNCTION__);
 
 	priv = kzalloc(sizeof(struct ftdi_private), GFP_KERNEL);
@@ -1203,15 +1208,18 @@ static int ftdi_sio_attach (struct usb_serial *serial)
 	usb_set_serial_port_data(serial->port[0], priv);
 
 	ftdi_determine_type (serial->port[0]);
-	create_sysfs_attrs(serial);
+	retval = create_sysfs_attrs(serial);
+	if (retval)
+		dev_err(&serial->dev->dev, "Error creating sysfs files, "
+			"continuing\n");
 
 	/* Check for device requiring special set up. */
 	quirk = (struct ftdi_sio_quirk *)usb_get_serial_data(serial);
 	if (quirk && quirk->setup) {
 		quirk->setup(serial);
 	}
-	
-	return (0);
+
+	return 0;
 } /* ftdi_sio_attach */
 
 
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 1f21a9d4b700..03c619478a7c 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -932,7 +932,10 @@ int usb_serial_probe(struct usb_interface *interface,
 
 		snprintf (&port->dev.bus_id[0], sizeof(port->dev.bus_id), "ttyUSB%d", port->number);
 		dbg ("%s - registering %s", __FUNCTION__, port->dev.bus_id);
-		device_register (&port->dev);
+		retval = device_register(&port->dev);
+		if (retval)
+			dev_err(&port->dev, "Error registering port device, "
+				"continuing\n");
 	}
 
 	usb_serial_console_init (debug, minor);
-- 
cgit v1.2.1


From 0257fa9ffe4f0287a9d90476bb733cfc2272396e Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Tue, 29 Aug 2006 22:06:59 -0500
Subject: cypress_m8: use appropriate URB polling interval

The polling interval for the device can't always be 1msec.  If it is
too quick, the device can fail causing a fatal (to the driver) EILSEQ
error from the USB core.  The actual correct value is reported by the
device as part of its configuration data, so use that value as the
default.  On a DeLorme Earthmate for example, the device reports that
it wants a 6msec interval.  As part of this fix, the "interval" module
option has been fixed as well; the device's default can be overridden
by specifying interval=<value> as a module option.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/cypress_m8.c | 26 ++++++++++++++++++++------
 1 file changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index ee70fddcab60..40cfbe1d3517 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -129,6 +129,8 @@ struct cypress_private {
 	int cmd_ctrl;			   /* always set this to 1 before issuing a command */
 	struct cypress_buf *buf;	   /* write buffer */
 	int write_urb_in_use;		   /* write urb in use indicator */
+	int write_urb_interval;            /* interval to use for write urb */
+	int read_urb_interval;             /* interval to use for read urb */
 	int termios_initialized;
 	__u8 line_control;	   	   /* holds dtr / rts value */
 	__u8 current_status;	   	   /* received from last read - info on dsr,cts,cd,ri,etc */
@@ -472,8 +474,9 @@ static unsigned rate_to_mask (int rate)
 static int generic_startup (struct usb_serial *serial)
 {
 	struct cypress_private *priv;
+	struct usb_serial_port *port = serial->port[0];
 
-	dbg("%s - port %d", __FUNCTION__, serial->port[0]->number);
+	dbg("%s - port %d", __FUNCTION__, port->number);
 
 	priv = kzalloc(sizeof (struct cypress_private), GFP_KERNEL);
 	if (!priv)
@@ -489,13 +492,24 @@ static int generic_startup (struct usb_serial *serial)
 	
 	usb_reset_configuration (serial->dev);
 	
-	interval = 1;
 	priv->cmd_ctrl = 0;
 	priv->line_control = 0;
 	priv->termios_initialized = 0;
 	priv->rx_flags = 0;
 	priv->cbr_mask = B300;
-	usb_set_serial_port_data(serial->port[0], priv);
+	if (interval > 0) {
+		priv->write_urb_interval = interval;
+		priv->read_urb_interval = interval;
+		dbg("%s - port %d read & write intervals forced to %d",
+		    __FUNCTION__,port->number,interval);
+	} else {
+		priv->write_urb_interval = port->interrupt_out_urb->interval;
+		priv->read_urb_interval = port->interrupt_in_urb->interval;
+		dbg("%s - port %d intervals: read=%d write=%d",
+		    __FUNCTION__,port->number,
+		    priv->read_urb_interval,priv->write_urb_interval);
+	}
+	usb_set_serial_port_data(port, priv);
 	
 	return 0;
 }
@@ -624,7 +638,7 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 	usb_fill_int_urb(port->interrupt_in_urb, serial->dev,
 		usb_rcvintpipe(serial->dev, port->interrupt_in_endpointAddress),
 		port->interrupt_in_urb->transfer_buffer, port->interrupt_in_urb->transfer_buffer_length,
-		cypress_read_int_callback, port, interval);
+		cypress_read_int_callback, port, priv->read_urb_interval);
 	result = usb_submit_urb(port->interrupt_in_urb, GFP_KERNEL);
 
 	if (result){
@@ -808,7 +822,7 @@ send:
 
 	port->interrupt_out_urb->transfer_buffer_length = actual_size;
 	port->interrupt_out_urb->dev = port->serial->dev;
-	port->interrupt_out_urb->interval = interval;
+	port->interrupt_out_urb->interval = priv->write_urb_interval;
 	result = usb_submit_urb (port->interrupt_out_urb, GFP_ATOMIC);
 	if (result) {
 		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __FUNCTION__,
@@ -1349,7 +1363,7 @@ continue_read:
 					port->interrupt_in_endpointAddress),
 				port->interrupt_in_urb->transfer_buffer,
 				port->interrupt_in_urb->transfer_buffer_length,
-				cypress_read_int_callback, port, interval);
+				cypress_read_int_callback, port, priv->read_urb_interval);
 		result = usb_submit_urb(port->interrupt_in_urb, GFP_ATOMIC);
 		if (result)
 			dev_err(&urb->dev->dev, "%s - failed resubmitting "
-- 
cgit v1.2.1


From 9aa8dae7b1fa7af099a403fc3766e068a0ea6d68 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Tue, 29 Aug 2006 22:07:04 -0500
Subject: cypress_m8: use usb_fill_int_urb where appropriate

Rather than directly filling in URB fields, it's safer to use
usb_fill_int_urb().  This improves robustness of the driver; URB
changes in the future will not go uninitialized here.  That point not
withstanding, this driver should at least be self-consistent.  Either
use usb_fill_int_urb() everywhere or don't bother with it all.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/cypress_m8.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index 40cfbe1d3517..741dcec36d41 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -820,9 +820,10 @@ send:
 	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, port->interrupt_out_size,
 			      port->interrupt_out_urb->transfer_buffer);
 
-	port->interrupt_out_urb->transfer_buffer_length = actual_size;
-	port->interrupt_out_urb->dev = port->serial->dev;
-	port->interrupt_out_urb->interval = priv->write_urb_interval;
+	usb_fill_int_urb(port->interrupt_out_urb, port->serial->dev,
+		usb_sndintpipe(port->serial->dev, port->interrupt_out_endpointAddress),
+		port->interrupt_out_buffer, port->interrupt_out_size,
+		cypress_write_int_callback, port, priv->write_urb_interval);
 	result = usb_submit_urb (port->interrupt_out_urb, GFP_ATOMIC);
 	if (result) {
 		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __FUNCTION__,
-- 
cgit v1.2.1


From 48298e50e0f7dfc7273ebfaa37ffd225428e83ed Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Tue, 29 Aug 2006 22:07:07 -0500
Subject: cypress_m8: improve control endpoint error handling

Fix usb core function error return checks to look for negative errno
values, not positive errno values.  This bug had rendered those checks
useless.  Also remove attempted error recovery on control endpoints
for EPIPE - with control endpoints EPIPE does not indicate a halted
endpoint so trying to recover with usb_clear_halt() is not the correct
action.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/cypress_m8.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index 741dcec36d41..12a265c4a13b 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -367,9 +367,7 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 				if (tries++ >= 3)
 					break;
 
-				if (retval == EPIPE)
-					usb_clear_halt(port->serial->dev, 0x00);
-			} while (retval != 8 && retval != ENODEV);
+			} while (retval != 8 && retval != -ENODEV);
 
 			if (retval != 8)
 				err("%s - failed sending serial line settings - %d", __FUNCTION__, retval);
@@ -394,9 +392,7 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 				if (tries++ >= 3)
 					break;
 
-				if (retval == EPIPE)
-					usb_clear_halt(port->serial->dev, 0x00);
-			} while (retval != 5 && retval != ENODEV);
+			} while (retval != 5 && retval != -ENODEV);
 
 			if (retval != 5) {
 				err("%s - failed to retrieve serial line settings - %d", __FUNCTION__, retval);
-- 
cgit v1.2.1


From 78aef519ed07797f94cff1d0d66dd01704474916 Mon Sep 17 00:00:00 2001
From: Mike Isely <isely@pobox.com>
Date: Tue, 29 Aug 2006 22:07:11 -0500
Subject: cypress_m8: implement graceful failure handling

When receiving a fatal error from the USB core, e.g. EILSEQ (which can
happen if the polling interval is too short), fail gracefully.
Previously the driver would fill the log with useless error messages
or (more alarmingly) silently spin forever trying to write updated
control information to the device.  This change implements a new flag
which if cleared indicates that the driver has failed.  The flag will
be set on initialization, cleared on fatal errors, and anything else
that touches the USB port in the driver will abort if the flag is
clear.  When the flag is cleared, a message will be logged indicating
that the driver has failed.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/cypress_m8.c | 90 ++++++++++++++++++++++++++++++++++-------
 1 file changed, 75 insertions(+), 15 deletions(-)

diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c
index 12a265c4a13b..e1173c1aee37 100644
--- a/drivers/usb/serial/cypress_m8.c
+++ b/drivers/usb/serial/cypress_m8.c
@@ -131,6 +131,7 @@ struct cypress_private {
 	int write_urb_in_use;		   /* write urb in use indicator */
 	int write_urb_interval;            /* interval to use for write urb */
 	int read_urb_interval;             /* interval to use for read urb */
+	int comm_is_ok;                    /* true if communication is (still) ok */
 	int termios_initialized;
 	__u8 line_control;	   	   /* holds dtr / rts value */
 	__u8 current_status;	   	   /* received from last read - info on dsr,cts,cd,ri,etc */
@@ -170,6 +171,7 @@ static int  cypress_tiocmset		(struct usb_serial_port *port, struct file *file,
 static int  cypress_chars_in_buffer	(struct usb_serial_port *port);
 static void cypress_throttle		(struct usb_serial_port *port);
 static void cypress_unthrottle		(struct usb_serial_port *port);
+static void cypress_set_dead		(struct usb_serial_port *port);
 static void cypress_read_int_callback	(struct urb *urb, struct pt_regs *regs);
 static void cypress_write_int_callback	(struct urb *urb, struct pt_regs *regs);
 /* baud helper functions */
@@ -290,6 +292,9 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 	
 	priv = usb_get_serial_port_data(port);
 
+	if (!priv->comm_is_ok)
+		return -ENODEV;
+
 	switch(cypress_request_type) {
 		case CYPRESS_SET_CONFIG:
 
@@ -369,9 +374,10 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 
 			} while (retval != 8 && retval != -ENODEV);
 
-			if (retval != 8)
+			if (retval != 8) {
 				err("%s - failed sending serial line settings - %d", __FUNCTION__, retval);
-			else {
+				cypress_set_dead(port);
+			} else {
 				spin_lock_irqsave(&priv->lock, flags);
 				priv->baud_rate = new_baudrate;
 				priv->cbr_mask = baud_mask;
@@ -396,6 +402,7 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 
 			if (retval != 5) {
 				err("%s - failed to retrieve serial line settings - %d", __FUNCTION__, retval);
+				cypress_set_dead(port);
 				return retval;
 			} else {
 				spin_lock_irqsave(&priv->lock, flags);
@@ -417,6 +424,24 @@ static int cypress_serial_control (struct usb_serial_port *port, unsigned baud_m
 } /* cypress_serial_control */
 
 
+static void cypress_set_dead(struct usb_serial_port *port)
+{
+	struct cypress_private *priv = usb_get_serial_port_data(port);
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+	if (!priv->comm_is_ok) {
+		spin_unlock_irqrestore(&priv->lock, flags);
+		return;
+	}
+	priv->comm_is_ok = 0;
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	err("cypress_m8 suspending failing port %d - interval might be too short",
+	    port->number);
+}
+
+
 /* given a baud mask, it will return integer baud on success */
 static int mask_to_rate (unsigned mask)
 {
@@ -478,6 +503,7 @@ static int generic_startup (struct usb_serial *serial)
 	if (!priv)
 		return -ENOMEM;
 
+	priv->comm_is_ok = !0;
 	spin_lock_init(&priv->lock);
 	priv->buf = cypress_buf_alloc(CYPRESS_BUF_SIZE);
 	if (priv->buf == NULL) {
@@ -595,6 +621,9 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 
 	dbg("%s - port %d", __FUNCTION__, port->number);
 
+	if (!priv->comm_is_ok)
+		return -EIO;
+
 	/* clear halts before open */
 	usb_clear_halt(serial->dev, 0x81);
 	usb_clear_halt(serial->dev, 0x02);
@@ -639,6 +668,7 @@ static int cypress_open (struct usb_serial_port *port, struct file *filp)
 
 	if (result){
 		dev_err(&port->dev, "%s - failed submitting read urb, error %d\n", __FUNCTION__, result);
+		cypress_set_dead(port);
 	}
 
 	return result;
@@ -743,6 +773,9 @@ static void cypress_send(struct usb_serial_port *port)
 	struct cypress_private *priv = usb_get_serial_port_data(port);
 	unsigned long flags;
 	
+	if (!priv->comm_is_ok)
+		return;
+
 	dbg("%s - port %d", __FUNCTION__, port->number);
 	dbg("%s - interrupt out size is %d", __FUNCTION__, port->interrupt_out_size);
 	
@@ -825,6 +858,7 @@ send:
 		dev_err(&port->dev, "%s - failed submitting write urb, error %d\n", __FUNCTION__,
 			result);
 		priv->write_urb_in_use = 0;
+		cypress_set_dead(port);
 	}
 
 	spin_lock_irqsave(&priv->lock, flags);
@@ -1225,13 +1259,18 @@ static void cypress_unthrottle (struct usb_serial_port *port)
 	priv->rx_flags = 0;
 	spin_unlock_irqrestore(&priv->lock, flags);
 
+	if (!priv->comm_is_ok)
+		return;
+
 	if (actually_throttled) {
 		port->interrupt_in_urb->dev = port->serial->dev;
 
 		result = usb_submit_urb(port->interrupt_in_urb, GFP_ATOMIC);
-		if (result)
+		if (result) {
 			dev_err(&port->dev, "%s - failed submitting read urb, "
 					"error %d\n", __FUNCTION__, result);
+			cypress_set_dead(port);
+		}
 	}
 }
 
@@ -1251,9 +1290,22 @@ static void cypress_read_int_callback(struct urb *urb, struct pt_regs *regs)
 
 	dbg("%s - port %d", __FUNCTION__, port->number);
 
-	if (urb->status) {
-		dbg("%s - nonzero read status received: %d", __FUNCTION__,
-				urb->status);
+	switch (urb->status) {
+	case 0: /* success */
+		break;
+	case -ECONNRESET:
+	case -ENOENT:
+	case -ESHUTDOWN:
+		/* precursor to disconnect so just go away */
+		return;
+	case -EPIPE:
+		usb_clear_halt(port->serial->dev,0x81);
+		break;
+	default:
+		/* something ugly is going on... */
+		dev_err(&urb->dev->dev,"%s - unexpected nonzero read status received: %d\n",
+			__FUNCTION__,urb->status);
+		cypress_set_dead(port);
 		return;
 	}
 
@@ -1354,7 +1406,7 @@ continue_read:
 
 	/* Continue trying to always read... unless the port has closed. */
 
-	if (port->open_count > 0) {
+	if (port->open_count > 0 && priv->comm_is_ok) {
 		usb_fill_int_urb(port->interrupt_in_urb, port->serial->dev,
 				usb_rcvintpipe(port->serial->dev,
 					port->interrupt_in_endpointAddress),
@@ -1362,10 +1414,12 @@ continue_read:
 				port->interrupt_in_urb->transfer_buffer_length,
 				cypress_read_int_callback, port, priv->read_urb_interval);
 		result = usb_submit_urb(port->interrupt_in_urb, GFP_ATOMIC);
-		if (result)
+		if (result) {
 			dev_err(&urb->dev->dev, "%s - failed resubmitting "
 					"read urb, error %d\n", __FUNCTION__,
 					result);
+			cypress_set_dead(port);
+		}
 	}
 
 	return;
@@ -1391,20 +1445,26 @@ static void cypress_write_int_callback(struct urb *urb, struct pt_regs *regs)
 			dbg("%s - urb shutting down with status: %d", __FUNCTION__, urb->status);
 			priv->write_urb_in_use = 0;
 			return;
-		case -EPIPE: /* no break needed */
+		case -EPIPE: /* no break needed; clear halt and resubmit */
+			if (!priv->comm_is_ok)
+				break;
 			usb_clear_halt(port->serial->dev, 0x02);
-		default:
 			/* error in the urb, so we have to resubmit it */
-			dbg("%s - Overflow in write", __FUNCTION__);
 			dbg("%s - nonzero write bulk status received: %d", __FUNCTION__, urb->status);
 			port->interrupt_out_urb->transfer_buffer_length = 1;
 			port->interrupt_out_urb->dev = port->serial->dev;
 			result = usb_submit_urb(port->interrupt_out_urb, GFP_ATOMIC);
-			if (result)
-				dev_err(&urb->dev->dev, "%s - failed resubmitting write urb, error %d\n",
-					__FUNCTION__, result);
-			else
+			if (!result)
 				return;
+			dev_err(&urb->dev->dev, "%s - failed resubmitting write urb, error %d\n",
+				__FUNCTION__, result);
+			cypress_set_dead(port);
+			break;
+		default:
+			dev_err(&urb->dev->dev,"%s - unexpected nonzero write status received: %d\n",
+				__FUNCTION__,urb->status);
+			cypress_set_dead(port);
+			break;
 	}
 	
 	priv->write_urb_in_use = 0;
-- 
cgit v1.2.1


From 3fe70ba2272c123cf38e4c577bf220f8bcf25366 Mon Sep 17 00:00:00 2001
From: Manuel Francisco Naranjo <naranjo.manuel@gmail.com>
Date: Wed, 9 Aug 2006 16:35:12 -0300
Subject: Add AIRcable USB Bluetooth Dongle Driver

Add driver for AIRcable USB Bluetooth dongle.

Signed-off-by: Naranjo, Manuel Francisco <naranjo.manuel@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/usb/usb-serial.txt |   5 +
 drivers/usb/serial/Kconfig       |  10 +
 drivers/usb/serial/Makefile      |   1 +
 drivers/usb/serial/aircable.c    | 625 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 641 insertions(+)
 create mode 100644 drivers/usb/serial/aircable.c

diff --git a/Documentation/usb/usb-serial.txt b/Documentation/usb/usb-serial.txt
index 02b0f7beb6d1..a2dee6e6190d 100644
--- a/Documentation/usb/usb-serial.txt
+++ b/Documentation/usb/usb-serial.txt
@@ -433,6 +433,11 @@ Options supported:
   See http://www.uuhaus.de/linux/palmconnect.html for up-to-date
   information on this driver.
 
+AIRcable USB Dongle Bluetooth driver
+  If there is the cdc_acm driver loaded in the system, you will find that the
+  cdc_acm claims the device before AIRcable can. This is simply corrected
+  by unloading both modules and then loading the aircable module before
+  cdc_acm module
 
 Generic Serial driver
 
diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig
index f5b9438c94f0..e4df0ac358fb 100644
--- a/drivers/usb/serial/Kconfig
+++ b/drivers/usb/serial/Kconfig
@@ -53,6 +53,15 @@ config USB_SERIAL_GENERIC
 	  support" be compiled as a module for this driver to be used
 	  properly.
 
+config USB_SERIAL_AIRCABLE
+	tristate "AIRcable USB Bluetooth Dongle Driver (EXPERIMENTAL)"
+	depends on USB_SERIAL && EXPERIMENTAL
+	help
+	    Say Y here if you want to use AIRcable USB Bluetoot Dongle.
+
+	    To compile this driver as a module, choose M here: the module
+	    will be called aircable.
+
 config USB_SERIAL_AIRPRIME
 	tristate "USB AirPrime CDMA Wireless Driver"
 	depends on USB_SERIAL
@@ -526,5 +535,6 @@ config USB_EZUSB
 	depends on USB_SERIAL_KEYSPAN_PDA || USB_SERIAL_XIRCOM || USB_SERIAL_KEYSPAN || USB_SERIAL_WHITEHEAT
 	default y
 
+
 endmenu
 
diff --git a/drivers/usb/serial/Makefile b/drivers/usb/serial/Makefile
index 8efed2ce1ba3..029e2a1d7bf6 100644
--- a/drivers/usb/serial/Makefile
+++ b/drivers/usb/serial/Makefile
@@ -11,6 +11,7 @@ usbserial-obj-$(CONFIG_USB_EZUSB)		+= ezusb.o
 
 usbserial-objs	:= usb-serial.o generic.o bus.o $(usbserial-obj-y)
 
+obj-$(CONFIG_USB_SERIAL_AIRCABLE)		+= aircable.o
 obj-$(CONFIG_USB_SERIAL_AIRPRIME)		+= airprime.o
 obj-$(CONFIG_USB_SERIAL_ARK3116)		+= ark3116.o
 obj-$(CONFIG_USB_SERIAL_BELKIN)			+= belkin_sa.o
diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c
new file mode 100644
index 000000000000..8aaf7db93992
--- /dev/null
+++ b/drivers/usb/serial/aircable.c
@@ -0,0 +1,625 @@
+/*
+ * AIRcable USB Bluetooth Dongle Driver.
+ *
+ * Copyright (C) 2006 Manuel Francisco Naranjo (naranjo.manuel@gmail.com)
+ * This program is free software; you can redistribute it and/or modify it under
+ * the terms of the GNU General Public License version 2 as published by the
+ * Free Software Foundation.
+ *
+ * The device works as an standard CDC device, it has 2 interfaces, the first
+ * one is for firmware access and the second is the serial one.
+ * The protocol is very simply, there are two posibilities reading or writing.
+ * When writting the first urb must have a Header that starts with 0x20 0x29 the
+ * next two bytes must say how much data will be sended.
+ * When reading the process is almost equal except that the header starts with
+ * 0x00 0x20.
+ *
+ * The device simply need some stuff to understand data comming from the usb
+ * buffer: The First and Second byte is used for a Header, the Third and Fourth
+ * tells the  device the amount of information the package holds.
+ * Packages are 60 bytes long Header Stuff.
+ * When writting to the device the first two bytes of the header are 0x20 0x29
+ * When reading the bytes are 0x00 0x20, or 0x00 0x10, there is an strange
+ * situation, when too much data arrives to the device because it sends the data
+ * but with out the header. I will use a simply hack to override this situation,
+ * if there is data coming that does not contain any header, then that is data
+ * that must go directly to the tty, as there is no documentation about if there
+ * is any other control code, I will simply check for the first
+ * one.
+ *
+ * The driver registers himself with the USB-serial core and the USB Core. I had
+ * to implement a probe function agains USB-serial, because other way, the
+ * driver was attaching himself to both interfaces. I have tryed with different
+ * configurations of usb_serial_driver with out exit, only the probe function
+ * could handle this correctly.
+ *
+ * I have taken some info from a Greg Kroah-Hartman article:
+ * http://www.linuxjournal.com/article/6573
+ * And from Linux Device Driver Kit CD, which is a great work, the authors taken
+ * the work to recompile lots of information an knowladge in drivers development
+ * and made it all avaible inside a cd.
+ * URL: http://kernel.org/pub/linux/kernel/people/gregkh/ddk/
+ *
+ */
+
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/circ_buf.h>
+#include <linux/usb.h>
+#include <linux/usb/serial.h>
+
+static int debug;
+
+/* Vendor and Product ID */
+#define AIRCABLE_VID		0x16CA
+#define AIRCABLE_USB_PID	0x1502
+
+/* write buffer size defines */
+#define AIRCABLE_BUF_SIZE	2048
+
+/* Protocol Stuff */
+#define HCI_HEADER_LENGTH	0x4
+#define TX_HEADER_0		0x20
+#define TX_HEADER_1		0x29
+#define RX_HEADER_0		0x00
+#define RX_HEADER_1		0x20
+#define MAX_HCI_FRAMESIZE	60
+#define HCI_COMPLETE_FRAME	64
+
+/* rx_flags */
+#define THROTTLED		0x01
+#define ACTUALLY_THROTTLED	0x02
+
+/*
+ * Version Information
+ */
+#define DRIVER_VERSION "v1.0b2"
+#define DRIVER_AUTHOR "Naranjo, Manuel Francisco <naranjo.manuel@gmail.com>"
+#define DRIVER_DESC "AIRcable USB Driver"
+
+/* ID table that will be registered with USB core */
+static struct usb_device_id id_table [] = {
+	{ USB_DEVICE(AIRCABLE_VID, AIRCABLE_USB_PID) },
+	{ },
+};
+MODULE_DEVICE_TABLE(usb, id_table);
+
+
+/* Internal Structure */
+struct aircable_private {
+	spinlock_t rx_lock;		/* spinlock for the receive lines */
+	struct circ_buf *tx_buf;	/* write buffer */
+	struct circ_buf *rx_buf;	/* read buffer */
+	int rx_flags;			/* for throttilng */
+	struct work_struct rx_work;	/* work cue for the receiving line */
+};
+
+/* Private methods */
+
+/* Circular Buffer Methods, code from ti_usb_3410_5052 used */
+/*
+ * serial_buf_clear
+ *
+ * Clear out all data in the circular buffer.
+ */
+static void serial_buf_clear(struct circ_buf *cb)
+{
+	cb->head = cb->tail = 0;
+}
+
+/*
+ * serial_buf_alloc
+ *
+ * Allocate a circular buffer and all associated memory.
+ */
+static struct circ_buf *serial_buf_alloc(void)
+{
+	struct circ_buf *cb;
+	cb = kmalloc(sizeof(struct circ_buf), GFP_KERNEL);
+	if (cb == NULL)
+		return NULL;
+	cb->buf = kmalloc(AIRCABLE_BUF_SIZE, GFP_KERNEL);
+	if (cb->buf == NULL) {
+		kfree(cb);
+		return NULL;
+	}
+	serial_buf_clear(cb);
+	return cb;
+}
+
+/*
+ * serial_buf_free
+ *
+ * Free the buffer and all associated memory.
+ */
+static void serial_buf_free(struct circ_buf *cb)
+{
+	kfree(cb->buf);
+	kfree(cb);
+}
+
+/*
+ * serial_buf_data_avail
+ *
+ * Return the number of bytes of data available in the circular
+ * buffer.
+ */
+static int serial_buf_data_avail(struct circ_buf *cb)
+{
+	return CIRC_CNT(cb->head,cb->tail,AIRCABLE_BUF_SIZE);
+}
+
+/*
+ * serial_buf_put
+ *
+ * Copy data data from a user buffer and put it into the circular buffer.
+ * Restrict to the amount of space available.
+ *
+ * Return the number of bytes copied.
+ */
+static int serial_buf_put(struct circ_buf *cb, const char *buf, int count)
+{
+	int c, ret = 0;
+	while (1) {
+		c = CIRC_SPACE_TO_END(cb->head, cb->tail, AIRCABLE_BUF_SIZE);
+		if (count < c)
+			c = count;
+		if (c <= 0)
+			break;
+		memcpy(cb->buf + cb->head, buf, c);
+		cb->head = (cb->head + c) & (AIRCABLE_BUF_SIZE-1);
+		buf += c;
+		count -= c;
+		ret= c;
+	}
+	return ret;
+}
+
+/*
+ * serial_buf_get
+ *
+ * Get data from the circular buffer and copy to the given buffer.
+ * Restrict to the amount of data available.
+ *
+ * Return the number of bytes copied.
+ */
+static int serial_buf_get(struct circ_buf *cb, char *buf, int count)
+{
+	int c, ret = 0;
+	while (1) {
+		c = CIRC_CNT_TO_END(cb->head, cb->tail, AIRCABLE_BUF_SIZE);
+		if (count < c)
+			c = count;
+		if (c <= 0)
+			break;
+		memcpy(buf, cb->buf + cb->tail, c);
+		cb->tail = (cb->tail + c) & (AIRCABLE_BUF_SIZE-1);
+		buf += c;
+		count -= c;
+		ret= c;
+	}
+	return ret;
+}
+
+/* End of circula buffer methods */
+
+static void aircable_send(struct usb_serial_port *port)
+{
+	int count, result;
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	unsigned char* buf;
+	dbg("%s - port %d", __FUNCTION__, port->number);
+	if (port->write_urb_busy)
+		return;
+
+	count = min(serial_buf_data_avail(priv->tx_buf), MAX_HCI_FRAMESIZE);
+	if (count == 0)
+		return;
+
+	buf = kzalloc(count + HCI_HEADER_LENGTH, GFP_ATOMIC);
+	if (!buf) {
+		err("%s- kzalloc(%Zd) failed.", __FUNCTION__,
+		    count + HCI_HEADER_LENGTH);
+		return;
+	}
+
+	buf[0] = TX_HEADER_0;
+	buf[1] = TX_HEADER_1;
+	buf[2] = (unsigned char)count;
+	buf[3] = (unsigned char)(count >> 8);
+	serial_buf_get(priv->tx_buf,buf + HCI_HEADER_LENGTH, MAX_HCI_FRAMESIZE);
+
+	memcpy(port->write_urb->transfer_buffer, buf,
+	       count + HCI_HEADER_LENGTH);
+
+	kfree(buf);
+	port->write_urb_busy = 1;
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__,
+			      count + HCI_HEADER_LENGTH,
+			      port->write_urb->transfer_buffer);
+	port->write_urb->transfer_buffer_length = count + HCI_HEADER_LENGTH;
+	port->write_urb->dev = port->serial->dev;
+	result = usb_submit_urb(port->write_urb, GFP_ATOMIC);
+
+	if (result) {
+		dev_err(&port->dev,
+			"%s - failed submitting write urb, error %d\n",
+			__FUNCTION__, result);
+		port->write_urb_busy = 0;
+	}
+
+	schedule_work(&port->work);
+}
+
+static void aircable_read(void *params)
+{
+	struct usb_serial_port *port = params;
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	struct tty_struct *tty;
+	unsigned char *data;
+	int count;
+	if (priv->rx_flags & THROTTLED){
+		if (priv->rx_flags & ACTUALLY_THROTTLED)
+			schedule_work(&priv->rx_work);
+		return;
+	}
+
+	/* By now I will flush data to the tty in packages of no more than
+	 * 64 bytes, to ensure I do not get throttled.
+	 * Ask USB mailing list for better aproach.
+	 */
+	tty = port->tty;
+
+	if (!tty)
+		schedule_work(&priv->rx_work);
+
+	count = min(64, serial_buf_data_avail(priv->rx_buf));
+
+	if (count <= 0)
+		return; //We have finished sending everything.
+
+	tty_prepare_flip_string(tty, &data, count);
+	if (!data){
+		err("%s- kzalloc(%Zd) failed.", __FUNCTION__, count);
+		return;
+	}
+
+	serial_buf_get(priv->rx_buf, data, count);
+
+	tty_flip_buffer_push(tty);
+
+	if (serial_buf_data_avail(priv->rx_buf))
+		schedule_work(&priv->rx_work);
+
+	return;
+}
+/* End of private methods */
+
+static int aircable_probe(struct usb_serial *serial,
+			  const struct usb_device_id *id)
+{
+	struct usb_host_interface *iface_desc = serial->interface->cur_altsetting;
+	struct usb_endpoint_descriptor *endpoint;
+	int num_bulk_out=0;
+	int i;
+
+	for (i = 0; i < iface_desc->desc.bNumEndpoints; i++) {
+		endpoint = &iface_desc->endpoint[i].desc;
+		if (((endpoint->bEndpointAddress & 0x80) == 0x00) &&
+			((endpoint->bmAttributes & 3) == 0x02)) {
+			/* we found our bulk out endpoint */
+			dbg("found bulk out on endpoint %d", i);
+			++num_bulk_out;
+		}
+	}
+
+	if (num_bulk_out == 0) {
+		dbg("Invalid interface, discarding");
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+static int aircable_attach (struct usb_serial *serial)
+{
+	struct usb_serial_port *port = serial->port[0];
+	struct aircable_private *priv;
+
+	priv = kzalloc(sizeof(struct aircable_private), GFP_KERNEL);
+	if (!priv){
+		err("%s- kmalloc(%Zd) failed.", __FUNCTION__,
+			sizeof(struct aircable_private));
+		return -ENOMEM;
+	}
+
+	/* Allocation of Circular Buffers */
+	priv->tx_buf = serial_buf_alloc();
+	if (priv->tx_buf == NULL) {
+		kfree(priv);
+		return -ENOMEM;
+	}
+
+	priv->rx_buf = serial_buf_alloc();
+	if (priv->rx_buf == NULL) {
+		kfree(priv->tx_buf);
+		kfree(priv);
+		return -ENOMEM;
+	}
+
+	priv->rx_flags &= ~(THROTTLED | ACTUALLY_THROTTLED);
+	INIT_WORK(&priv->rx_work, aircable_read, port);
+
+	usb_set_serial_port_data(serial->port[0], priv);
+
+	return 0;
+}
+
+static void aircable_shutdown(struct usb_serial *serial)
+{
+
+	struct usb_serial_port *port = serial->port[0];
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+
+	dbg("%s", __FUNCTION__);
+
+	if (priv) {
+		serial_buf_free(priv->tx_buf);
+		serial_buf_free(priv->rx_buf);
+		usb_set_serial_port_data(port, NULL);
+		kfree(priv);
+	}
+}
+
+static int aircable_write_room(struct usb_serial_port *port)
+{
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	return serial_buf_data_avail(priv->tx_buf);
+}
+
+static int aircable_write(struct usb_serial_port *port,
+			  const unsigned char *source, int count)
+{
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	int temp;
+
+	dbg("%s - port %d, %d bytes", __FUNCTION__, port->number, count);
+
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, count, source);
+
+	if (!count){
+		dbg("%s - write request of 0 bytes", __FUNCTION__);
+		return count;
+	}
+
+	temp = serial_buf_put(priv->tx_buf, source, count);
+
+	aircable_send(port);
+
+	if (count > AIRCABLE_BUF_SIZE)
+		count = AIRCABLE_BUF_SIZE;
+
+	return count;
+
+}
+
+static void aircable_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct usb_serial_port *port = urb->context;
+	int result;
+
+	dbg("%s - urb->status: %d", __FUNCTION__ , urb->status);
+
+	/* This has been taken from cypress_m8.c cypress_write_int_callback */
+	switch (urb->status) {
+		case 0:
+			/* success */
+			break;
+		case -ECONNRESET:
+		case -ENOENT:
+		case -ESHUTDOWN:
+			/* this urb is terminated, clean up */
+			dbg("%s - urb shutting down with status: %d",
+			    __FUNCTION__, urb->status);
+			port->write_urb_busy = 0;
+			return;
+		default:
+			/* error in the urb, so we have to resubmit it */
+			dbg("%s - Overflow in write", __FUNCTION__);
+			dbg("%s - nonzero write bulk status received: %d",
+			    __FUNCTION__, urb->status);
+			port->write_urb->transfer_buffer_length = 1;
+			port->write_urb->dev = port->serial->dev;
+			result = usb_submit_urb(port->write_urb, GFP_KERNEL);
+			if (result)
+				dev_err(&urb->dev->dev,
+					"%s - failed resubmitting write urb, error %d\n",
+					__FUNCTION__, result);
+			else
+				return;
+	}
+
+	port->write_urb_busy = 0;
+
+	aircable_send(port);
+}
+
+static void aircable_read_bulk_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct usb_serial_port *port = urb->context;
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	struct tty_struct *tty;
+	unsigned long no_packages, remaining, package_length, i;
+	int result, shift = 0;
+	unsigned char *temp;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	if (urb->status) {
+		dbg("%s - urb->status = %d", __FUNCTION__, urb->status);
+		if (!port->open_count) {
+			dbg("%s - port is closed, exiting.", __FUNCTION__);
+			return;
+		}
+		if (urb->status == -EPROTO) {
+			dbg("%s - caught -EPROTO, resubmitting the urb",
+			    __FUNCTION__);
+			usb_fill_bulk_urb(port->read_urb, port->serial->dev,
+					  usb_rcvbulkpipe(port->serial->dev,
+					  		  port->bulk_in_endpointAddress),
+					  port->read_urb->transfer_buffer,
+					  port->read_urb->transfer_buffer_length,
+					  aircable_read_bulk_callback, port);
+
+			result = usb_submit_urb(urb, GFP_ATOMIC);
+			if (result)
+				dev_err(&urb->dev->dev,
+					"%s - failed resubmitting read urb, error %d\n",
+					__FUNCTION__, result);
+			return;
+		}
+		dbg("%s - unable to handle the error, exiting.", __FUNCTION__);
+		return;
+	}
+
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__,
+				urb->actual_length,urb->transfer_buffer);
+
+	tty = port->tty;
+	if (tty && urb->actual_length) {
+		if (urb->actual_length <= 2) {
+			/* This is an incomplete package */
+			serial_buf_put(priv->rx_buf, urb->transfer_buffer,
+				       urb->actual_length);
+		} else {
+			temp = urb->transfer_buffer;
+			if (temp[0] == RX_HEADER_0)
+				shift = HCI_HEADER_LENGTH;
+
+			remaining = urb->actual_length;
+			no_packages = urb->actual_length / (HCI_COMPLETE_FRAME);
+
+			if (urb->actual_length % HCI_COMPLETE_FRAME != 0)
+				no_packages+=1;
+
+			for (i = 0; i < no_packages ;i++) {
+				if (remaining > (HCI_COMPLETE_FRAME))
+					package_length = HCI_COMPLETE_FRAME;
+				else
+					package_length = remaining;
+				remaining -= package_length;
+
+				serial_buf_put(priv->rx_buf,
+					urb->transfer_buffer + shift +
+					(HCI_COMPLETE_FRAME) * (i),
+					package_length - shift);
+			}
+		}
+		aircable_read(port);
+	}
+
+	/* Schedule the next read _if_ we are still open */
+	if (port->open_count) {
+		usb_fill_bulk_urb(port->read_urb, port->serial->dev,
+				  usb_rcvbulkpipe(port->serial->dev,
+				  		  port->bulk_in_endpointAddress),
+				  port->read_urb->transfer_buffer,
+				  port->read_urb->transfer_buffer_length,
+				  aircable_read_bulk_callback, port);
+
+		result = usb_submit_urb(urb, GFP_ATOMIC);
+		if (result)
+			dev_err(&urb->dev->dev,
+				"%s - failed resubmitting read urb, error %d\n",
+				__FUNCTION__, result);
+	}
+
+	return;
+}
+
+/* Based on ftdi_sio.c throttle */
+static void aircable_throttle(struct usb_serial_port *port)
+{
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	unsigned long flags;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	spin_lock_irqsave(&priv->rx_lock, flags);
+	priv->rx_flags |= THROTTLED;
+	spin_unlock_irqrestore(&priv->rx_lock, flags);
+}
+
+/* Based on ftdi_sio.c unthrottle */
+static void aircable_unthrottle(struct usb_serial_port *port)
+{
+	struct aircable_private *priv = usb_get_serial_port_data(port);
+	int actually_throttled;
+	unsigned long flags;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	spin_lock_irqsave(&priv->rx_lock, flags);
+	actually_throttled = priv->rx_flags & ACTUALLY_THROTTLED;
+	priv->rx_flags &= ~(THROTTLED | ACTUALLY_THROTTLED);
+	spin_unlock_irqrestore(&priv->rx_lock, flags);
+
+	if (actually_throttled)
+		schedule_work(&priv->rx_work);
+}
+
+static struct usb_serial_driver aircable_device = {
+	.description =		"aircable",
+	.id_table = 		id_table,
+	.num_ports =		1,
+	.attach =		aircable_attach,
+	.probe =		aircable_probe,
+	.shutdown =		aircable_shutdown,
+	.write =		aircable_write,
+	.write_room =		aircable_write_room,
+	.write_bulk_callback =	aircable_write_bulk_callback,
+	.read_bulk_callback =	aircable_read_bulk_callback,
+	.throttle =		aircable_throttle,
+	.unthrottle =		aircable_unthrottle,
+};
+
+static struct usb_driver aircable_driver = {
+	.name =		"aircable",
+	.probe =	usb_serial_probe,
+	.disconnect =	usb_serial_disconnect,
+	.id_table =	id_table,
+};
+
+static int __init aircable_init (void)
+{
+	int retval;
+	retval = usb_serial_register(&aircable_device);
+	if (retval)
+		goto failed_serial_register;
+	retval = usb_register(&aircable_driver);
+	if (retval)
+		goto failed_usb_register;
+	return 0;
+
+failed_serial_register:
+	usb_serial_deregister(&aircable_device);
+failed_usb_register:
+	return retval;
+}
+
+static void __exit aircable_exit (void)
+{
+	usb_deregister(&aircable_driver);
+	usb_serial_deregister(&aircable_device);
+}
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_VERSION(DRIVER_VERSION);
+MODULE_LICENSE("GPL");
+
+module_init(aircable_init);
+module_exit(aircable_exit);
+
+module_param(debug, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Debug enabled or not");
-- 
cgit v1.2.1


From 8ac283ad415358f022498887811c35ac656b5222 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <rdunlap@xenotime.net>
Date: Sat, 26 Aug 2006 10:56:10 -0700
Subject: aircable: fix printk format warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix printk format warnings:
drivers/usb/serial/aircable.c:221: warning: format ‘%Zd’ expects type ‘signed size_t’, but argument 4 has type ‘int’
drivers/usb/serial/aircable.c:283: warning: format ‘%Zd’ expects type ‘signed size_t’, but argument 4 has type ‘int’

Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/aircable.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/serial/aircable.c b/drivers/usb/serial/aircable.c
index 8aaf7db93992..2ccd9ded52a5 100644
--- a/drivers/usb/serial/aircable.c
+++ b/drivers/usb/serial/aircable.c
@@ -218,7 +218,7 @@ static void aircable_send(struct usb_serial_port *port)
 
 	buf = kzalloc(count + HCI_HEADER_LENGTH, GFP_ATOMIC);
 	if (!buf) {
-		err("%s- kzalloc(%Zd) failed.", __FUNCTION__,
+		err("%s- kzalloc(%d) failed.", __FUNCTION__,
 		    count + HCI_HEADER_LENGTH);
 		return;
 	}
@@ -280,7 +280,7 @@ static void aircable_read(void *params)
 
 	tty_prepare_flip_string(tty, &data, count);
 	if (!data){
-		err("%s- kzalloc(%Zd) failed.", __FUNCTION__, count);
+		err("%s- kzalloc(%d) failed.", __FUNCTION__, count);
 		return;
 	}
 
-- 
cgit v1.2.1


From 03270634e242dd10cc8569d31a00659d25b2b8e7 Mon Sep 17 00:00:00 2001
From: Steven Haigh <netwiz@crc.id.au>
Date: Wed, 9 Aug 2006 07:42:06 +1000
Subject: USB: Add ADU support for Ontrak ADU devices

This patch adds support for Ontrak ADU USB devices.

Fixed for printk issues by Randy Dunlap <rdunlap@xenotime.net>

Signed-off-by: Steven Haigh <netwiz@crc.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/devices.txt    |   3 +
 drivers/usb/Makefile         |   1 +
 drivers/usb/input/hid-core.c |   4 +
 drivers/usb/misc/Kconfig     |  10 +
 drivers/usb/misc/Makefile    |   1 +
 drivers/usb/misc/adutux.c    | 900 +++++++++++++++++++++++++++++++++++++++++++
 6 files changed, 919 insertions(+)
 create mode 100644 drivers/usb/misc/adutux.c

diff --git a/Documentation/devices.txt b/Documentation/devices.txt
index 66c725f530f3..addc67b1d770 100644
--- a/Documentation/devices.txt
+++ b/Documentation/devices.txt
@@ -2543,6 +2543,9 @@ Your cooperation is appreciated.
 		 64 = /dev/usb/rio500	Diamond Rio 500
 		 65 = /dev/usb/usblcd	USBLCD Interface (info@usblcd.de)
 		 66 = /dev/usb/cpad0	Synaptics cPad (mouse/LCD)
+		 67 = /dev/usb/adutux0	1st Ontrak ADU device
+		    ...
+		 76 = /dev/usb/adutux10	10th Ontrak ADU device
 		 96 = /dev/usb/hiddev0	1st USB HID device
 		    ...
 		111 = /dev/usb/hiddev15	16th USB HID device
diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index 4710eb02ed64..84430db4b8d9 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -63,6 +63,7 @@ obj-$(CONFIG_USB_USS720)	+= misc/
 obj-$(CONFIG_USB_PHIDGETSERVO)	+= misc/
 obj-$(CONFIG_USB_SISUSBVGA)	+= misc/
 obj-$(CONFIG_USB_APPLEDISPLAY)	+= misc/
+obj-$(CONFIG_USB_ADUTUX)	+= misc/
 
 obj-$(CONFIG_USB_ATM)		+= atm/
 obj-$(CONFIG_USB_SPEEDTOUCH)	+= atm/
diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index 8f8d4af3f6ef..d0d1c0c30bc5 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1697,7 +1697,11 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS1, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 20, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 30, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 100, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 108, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 118, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 200, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 300, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_ONTRAK, USB_DEVICE_ID_ONTRAK_ADU100 + 400, HID_QUIRK_IGNORE },
diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
index 20539cf9394b..288d301d2bff 100644
--- a/drivers/usb/misc/Kconfig
+++ b/drivers/usb/misc/Kconfig
@@ -32,6 +32,16 @@ config USB_EMI26
 	  To compile this driver as a module, choose M here: the
 	  module will be called emi26.
 
+config USB_ADUTUX
+	tristate "ADU devices from Ontrak Control Systems (EXPERIMENTAL)"
+	depends on USB && EXPERIMENTAL
+	help
+	  Say Y if you want to use an ADU device from Ontrak Control
+	  Systems.
+
+	  To compile this driver as a module, choose M here.  The module
+	  will be called adutux.
+
 config USB_AUERSWALD
 	tristate "USB Auerswald ISDN support (EXPERIMENTAL)"
 	depends on USB && EXPERIMENTAL
diff --git a/drivers/usb/misc/Makefile b/drivers/usb/misc/Makefile
index d12a84a1c0a4..73fc8be0d8c0 100644
--- a/drivers/usb/misc/Makefile
+++ b/drivers/usb/misc/Makefile
@@ -3,6 +3,7 @@
 # (the ones that don't fit into any other categories)
 #
 
+obj-$(CONFIG_USB_ADUTUX)	+= adutux.o
 obj-$(CONFIG_USB_AUERSWALD)	+= auerswald.o
 obj-$(CONFIG_USB_CYPRESS_CY7C63)+= cypress_cy7c63.o
 obj-$(CONFIG_USB_CYTHERM)	+= cytherm.o
diff --git a/drivers/usb/misc/adutux.c b/drivers/usb/misc/adutux.c
new file mode 100644
index 000000000000..d3963199b6ec
--- /dev/null
+++ b/drivers/usb/misc/adutux.c
@@ -0,0 +1,900 @@
+/*
+ * adutux - driver for ADU devices from Ontrak Control Systems
+ * This is an experimental driver. Use at your own risk.
+ * This driver is not supported by Ontrak Control Systems.
+ *
+ * Copyright (c) 2003 John Homppi (SCO, leave this notice here)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * derived from the Lego USB Tower driver 0.56:
+ * Copyright (c) 2003 David Glance <davidgsf@sourceforge.net>
+ *               2001 Juergen Stuber <stuber@loria.fr>
+ * that was derived from USB Skeleton driver - 0.5
+ * Copyright (c) 2001 Greg Kroah-Hartman (greg@kroah.com)
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+#include <asm/uaccess.h>
+
+#ifdef CONFIG_USB_DEBUG
+static int debug = 5;
+#else
+static int debug = 1;
+#endif
+
+/* Use our own dbg macro */
+#undef dbg
+#define dbg(lvl, format, arg...) 					\
+do { 									\
+	if (debug >= lvl)						\
+		printk(KERN_DEBUG __FILE__ " : " format " \n", ## arg);	\
+} while (0)
+
+
+/* Version Information */
+#define DRIVER_VERSION "v0.0.13"
+#define DRIVER_AUTHOR "John Homppi"
+#define DRIVER_DESC "adutux (see www.ontrak.net)"
+
+/* Module parameters */
+module_param(debug, int, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Debug enabled or not");
+
+/* Define these values to match your device */
+#define ADU_VENDOR_ID 0x0a07
+#define ADU_PRODUCT_ID 0x0064
+
+/* table of devices that work with this driver */
+static struct usb_device_id device_table [] = {
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID) },		/* ADU100 */
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID+20) }, 	/* ADU120 */
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID+30) }, 	/* ADU130 */
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID+100) },	/* ADU200 */
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID+108) },	/* ADU208 */
+	{ USB_DEVICE(ADU_VENDOR_ID, ADU_PRODUCT_ID+118) },	/* ADU218 */
+	{ }/* Terminating entry */
+};
+
+MODULE_DEVICE_TABLE(usb, device_table);
+
+#ifdef CONFIG_USB_DYNAMIC_MINORS
+#define ADU_MINOR_BASE	0
+#else
+#define ADU_MINOR_BASE	67
+#endif
+
+/* we can have up to this number of device plugged in at once */
+#define MAX_DEVICES	16
+
+#define COMMAND_TIMEOUT	(2*HZ)	/* 60 second timeout for a command */
+
+/* Structure to hold all of our device specific stuff */
+struct adu_device {
+	struct semaphore	sem; /* locks this structure */
+	struct usb_device*	udev; /* save off the usb device pointer */
+	struct usb_interface*	interface;
+	unsigned char		minor; /* the starting minor number for this device */
+	char			serial_number[8];
+
+	int			open_count; /* number of times this port has been opened */
+
+	char*			read_buffer_primary;
+	int			read_buffer_length;
+	char*			read_buffer_secondary;
+	int			secondary_head;
+	int			secondary_tail;
+	spinlock_t		buflock;
+
+	wait_queue_head_t	read_wait;
+	wait_queue_head_t	write_wait;
+
+	char*			interrupt_in_buffer;
+	struct usb_endpoint_descriptor* interrupt_in_endpoint;
+	struct urb*		interrupt_in_urb;
+	int			read_urb_finished;
+
+	char*			interrupt_out_buffer;
+	struct usb_endpoint_descriptor* interrupt_out_endpoint;
+	struct urb*		interrupt_out_urb;
+};
+
+/* prevent races between open() and disconnect */
+static DEFINE_MUTEX(disconnect_mutex);
+static struct usb_driver adu_driver;
+
+static void adu_debug_data(int level, const char *function, int size,
+			   const unsigned char *data)
+{
+	int i;
+
+	if (debug < level)
+		return;
+
+	printk(KERN_DEBUG __FILE__": %s - length = %d, data = ",
+	       function, size);
+	for (i = 0; i < size; ++i)
+		printk("%.2x ", data[i]);
+	printk("\n");
+}
+
+/**
+ * adu_abort_transfers
+ *      aborts transfers and frees associated data structures
+ */
+static void adu_abort_transfers(struct adu_device *dev)
+{
+	dbg(2," %s : enter", __FUNCTION__);
+
+	if (dev == NULL) {
+		dbg(1," %s : dev is null", __FUNCTION__);
+		goto exit;
+	}
+
+	if (dev->udev == NULL) {
+		dbg(1," %s : udev is null", __FUNCTION__);
+		goto exit;
+	}
+
+	dbg(2," %s : udev state %d", __FUNCTION__, dev->udev->state);
+	if (dev->udev->state == USB_STATE_NOTATTACHED) {
+		dbg(1," %s : udev is not attached", __FUNCTION__);
+		goto exit;
+	}
+
+	/* shutdown transfer */
+	usb_unlink_urb(dev->interrupt_in_urb);
+	usb_unlink_urb(dev->interrupt_out_urb);
+
+exit:
+	dbg(2," %s : leave", __FUNCTION__);
+}
+
+static void adu_delete(struct adu_device *dev)
+{
+	dbg(2, "%s enter", __FUNCTION__);
+
+	adu_abort_transfers(dev);
+
+	/* free data structures */
+	usb_free_urb(dev->interrupt_in_urb);
+	usb_free_urb(dev->interrupt_out_urb);
+	kfree(dev->read_buffer_primary);
+	kfree(dev->read_buffer_secondary);
+	kfree(dev->interrupt_in_buffer);
+	kfree(dev->interrupt_out_buffer);
+	kfree(dev);
+
+	dbg(2, "%s : leave", __FUNCTION__);
+}
+
+static void adu_interrupt_in_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct adu_device *dev = urb->context;
+
+	dbg(4," %s : enter, status %d", __FUNCTION__, urb->status);
+	adu_debug_data(5, __FUNCTION__, urb->actual_length,
+		       urb->transfer_buffer);
+
+	spin_lock(&dev->buflock);
+
+	if (urb->status != 0) {
+		if ((urb->status != -ENOENT) && (urb->status != -ECONNRESET)) {
+			dbg(1," %s : nonzero status received: %d",
+			    __FUNCTION__, urb->status);
+		}
+		goto exit;
+	}
+
+	if (urb->actual_length > 0 && dev->interrupt_in_buffer[0] != 0x00) {
+		if (dev->read_buffer_length <
+		    (4 * le16_to_cpu(dev->interrupt_in_endpoint->wMaxPacketSize)) -
+		     (urb->actual_length)) {
+			memcpy (dev->read_buffer_primary +
+				dev->read_buffer_length,
+				dev->interrupt_in_buffer, urb->actual_length);
+
+			dev->read_buffer_length += urb->actual_length;
+			dbg(2," %s reading  %d ", __FUNCTION__,
+			    urb->actual_length);
+		} else {
+			dbg(1," %s : read_buffer overflow", __FUNCTION__);
+		}
+	}
+
+exit:
+	dev->read_urb_finished = 1;
+	spin_unlock(&dev->buflock);
+	/* always wake up so we recover from errors */
+	wake_up_interruptible(&dev->read_wait);
+	adu_debug_data(5, __FUNCTION__, urb->actual_length,
+		       urb->transfer_buffer);
+	dbg(4," %s : leave, status %d", __FUNCTION__, urb->status);
+}
+
+static void adu_interrupt_out_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct adu_device *dev = urb->context;
+
+	dbg(4," %s : enter, status %d", __FUNCTION__, urb->status);
+	adu_debug_data(5,__FUNCTION__, urb->actual_length, urb->transfer_buffer);
+
+	if (urb->status != 0) {
+		if ((urb->status != -ENOENT) &&
+		    (urb->status != -ECONNRESET)) {
+			dbg(1, " %s :nonzero status received: %d",
+			    __FUNCTION__, urb->status);
+		}
+		goto exit;
+	}
+
+	wake_up_interruptible(&dev->write_wait);
+exit:
+
+	adu_debug_data(5, __FUNCTION__, urb->actual_length,
+		       urb->transfer_buffer);
+	dbg(4," %s : leave, status %d", __FUNCTION__, urb->status);
+}
+
+static int adu_open(struct inode *inode, struct file *file)
+{
+	struct adu_device *dev = NULL;
+	struct usb_interface *interface;
+	int subminor;
+	int retval = 0;
+
+	dbg(2,"%s : enter", __FUNCTION__);
+
+	subminor = iminor(inode);
+
+	mutex_lock(&disconnect_mutex);
+
+	interface = usb_find_interface(&adu_driver, subminor);
+	if (!interface) {
+		err("%s - error, can't find device for minor %d",
+		    __FUNCTION__, subminor);
+		retval = -ENODEV;
+		goto exit_no_device;
+	}
+
+	dev = usb_get_intfdata(interface);
+	if (!dev) {
+		retval = -ENODEV;
+		goto exit_no_device;
+	}
+
+	/* lock this device */
+	if ((retval = down_interruptible(&dev->sem))) {
+		dbg(2, "%s : sem down failed", __FUNCTION__);
+		goto exit_no_device;
+	}
+
+	/* increment our usage count for the device */
+	++dev->open_count;
+	dbg(2,"%s : open count %d", __FUNCTION__, dev->open_count);
+
+	/* save device in the file's private structure */
+	file->private_data = dev;
+
+	/* initialize in direction */
+	dev->read_buffer_length = 0;
+
+	/* fixup first read by having urb waiting for it */
+	usb_fill_int_urb(dev->interrupt_in_urb,dev->udev,
+			 usb_rcvintpipe(dev->udev,
+			 		dev->interrupt_in_endpoint->bEndpointAddress),
+			 dev->interrupt_in_buffer,
+			 le16_to_cpu(dev->interrupt_in_endpoint->wMaxPacketSize),
+			 adu_interrupt_in_callback, dev,
+			 dev->interrupt_in_endpoint->bInterval);
+	/* dev->interrupt_in_urb->transfer_flags |= URB_ASYNC_UNLINK; */
+	dev->read_urb_finished = 0;
+	usb_submit_urb(dev->interrupt_in_urb, GFP_KERNEL);
+	/* we ignore failure */
+	/* end of fixup for first read */
+
+	up(&dev->sem);
+
+exit_no_device:
+	mutex_unlock(&disconnect_mutex);
+	dbg(2,"%s : leave, return value %d ", __FUNCTION__, retval);
+
+	return retval;
+}
+
+static int adu_release_internal(struct adu_device *dev)
+{
+	int retval = 0;
+
+	dbg(2," %s : enter", __FUNCTION__);
+
+	if (dev->udev == NULL) {
+		/* the device was unplugged before the file was released */
+		adu_delete(dev);
+		goto exit;
+	}
+
+	/* decrement our usage count for the device */
+	--dev->open_count;
+	dbg(2," %s : open count %d", __FUNCTION__, dev->open_count);
+	if (dev->open_count <= 0) {
+		adu_abort_transfers(dev);
+		dev->open_count = 0;
+	}
+
+exit:
+	dbg(2," %s : leave", __FUNCTION__);
+	return retval;
+}
+
+static int adu_release(struct inode *inode, struct file *file)
+{
+	struct adu_device *dev = NULL;
+	int retval = 0;
+
+	dbg(2," %s : enter", __FUNCTION__);
+
+	if (file == NULL) {
+ 		dbg(1," %s : file is NULL", __FUNCTION__);
+		retval = -ENODEV;
+		goto exit;
+	}
+
+	dev = file->private_data;
+
+	if (dev == NULL) {
+ 		dbg(1," %s : object is NULL", __FUNCTION__);
+		retval = -ENODEV;
+		goto exit;
+	}
+
+	/* lock our device */
+	down(&dev->sem); /* not interruptible */
+
+	if (dev->open_count <= 0) {
+		dbg(1," %s : device not opened", __FUNCTION__);
+		retval = -ENODEV;
+		goto exit;
+	}
+
+	/* do the work */
+	retval = adu_release_internal(dev);
+
+exit:
+	up(&dev->sem);
+	dbg(2," %s : leave, return value %d", __FUNCTION__, retval);
+	return retval;
+}
+
+static ssize_t adu_read(struct file *file, __user char *buffer, size_t count,
+			loff_t *ppos)
+{
+	struct adu_device *dev;
+	size_t bytes_read = 0;
+	size_t bytes_to_read = count;
+	int i;
+	int retval = 0;
+	int timeout = 0;
+	int should_submit = 0;
+	unsigned long flags;
+	DECLARE_WAITQUEUE(wait, current);
+
+	dbg(2," %s : enter, count = %Zd, file=%p", __FUNCTION__, count, file);
+
+	dev = file->private_data;
+	dbg(2," %s : dev=%p", __FUNCTION__, dev);
+	/* lock this object */
+	if (down_interruptible(&dev->sem))
+		return -ERESTARTSYS;
+
+	/* verify that the device wasn't unplugged */
+	if (dev->udev == NULL || dev->minor == 0) {
+		retval = -ENODEV;
+		err("No device or device unplugged %d", retval);
+		goto exit;
+	}
+
+	/* verify that some data was requested */
+	if (count == 0) {
+		dbg(1," %s : read request of 0 bytes", __FUNCTION__);
+		goto exit;
+	}
+
+	timeout = COMMAND_TIMEOUT;
+	dbg(2," %s : about to start looping", __FUNCTION__);
+	while (bytes_to_read) {
+		int data_in_secondary = dev->secondary_tail - dev->secondary_head;
+		dbg(2," %s : while, data_in_secondary=%d, status=%d",
+		    __FUNCTION__, data_in_secondary,
+		    dev->interrupt_in_urb->status);
+
+		if (data_in_secondary) {
+			/* drain secondary buffer */
+			int amount = bytes_to_read < data_in_secondary ? bytes_to_read : data_in_secondary;
+			i = copy_to_user(buffer, dev->read_buffer_secondary+dev->secondary_head, amount);
+			if (i < 0) {
+				retval = -EFAULT;
+				goto exit;
+			}
+			dev->secondary_head += (amount - i);
+			bytes_read += (amount - i);
+			bytes_to_read -= (amount - i);
+			if (i) {
+				retval = bytes_read ? bytes_read : -EFAULT;
+				goto exit;
+			}
+		} else {
+			/* we check the primary buffer */
+			spin_lock_irqsave (&dev->buflock, flags);
+			if (dev->read_buffer_length) {
+				/* we secure access to the primary */
+				char *tmp;
+				dbg(2," %s : swap, read_buffer_length = %d",
+				    __FUNCTION__, dev->read_buffer_length);
+				tmp = dev->read_buffer_secondary;
+				dev->read_buffer_secondary = dev->read_buffer_primary;
+				dev->read_buffer_primary = tmp;
+				dev->secondary_head = 0;
+				dev->secondary_tail = dev->read_buffer_length;
+				dev->read_buffer_length = 0;
+				spin_unlock_irqrestore(&dev->buflock, flags);
+				/* we have a free buffer so use it */
+				should_submit = 1;
+			} else {
+				/* even the primary was empty - we may need to do IO */
+				if (dev->interrupt_in_urb->status == -EINPROGRESS) {
+					/* somebody is doing IO */
+					spin_unlock_irqrestore(&dev->buflock, flags);
+					dbg(2," %s : submitted already", __FUNCTION__);
+				} else {
+					/* we must initiate input */
+					dbg(2," %s : initiate input", __FUNCTION__);
+					dev->read_urb_finished = 0;
+
+					usb_fill_int_urb(dev->interrupt_in_urb,dev->udev,
+							 usb_rcvintpipe(dev->udev,
+							 		dev->interrupt_in_endpoint->bEndpointAddress),
+							 dev->interrupt_in_buffer,
+							 le16_to_cpu(dev->interrupt_in_endpoint->wMaxPacketSize),
+							 adu_interrupt_in_callback,
+							 dev,
+							 dev->interrupt_in_endpoint->bInterval);
+					retval = usb_submit_urb(dev->interrupt_in_urb, GFP_KERNEL);
+					if (!retval) {
+						spin_unlock_irqrestore(&dev->buflock, flags);
+						dbg(2," %s : submitted OK", __FUNCTION__);
+					} else {
+						if (retval == -ENOMEM) {
+							retval = bytes_read ? bytes_read : -ENOMEM;
+						}
+						spin_unlock_irqrestore(&dev->buflock, flags);
+						dbg(2," %s : submit failed", __FUNCTION__);
+						goto exit;
+					}
+				}
+
+				/* we wait for I/O to complete */
+				set_current_state(TASK_INTERRUPTIBLE);
+				add_wait_queue(&dev->read_wait, &wait);
+				if (!dev->read_urb_finished)
+					timeout = schedule_timeout(COMMAND_TIMEOUT);
+				else
+					set_current_state(TASK_RUNNING);
+				remove_wait_queue(&dev->read_wait, &wait);
+
+				if (timeout <= 0) {
+					dbg(2," %s : timeout", __FUNCTION__);
+					retval = bytes_read ? bytes_read : -ETIMEDOUT;
+					goto exit;
+				}
+
+				if (signal_pending(current)) {
+					dbg(2," %s : signal pending", __FUNCTION__);
+					retval = bytes_read ? bytes_read : -EINTR;
+					goto exit;
+				}
+			}
+		}
+	}
+
+	retval = bytes_read;
+	/* if the primary buffer is empty then use it */
+	if (should_submit && !dev->interrupt_in_urb->status==-EINPROGRESS) {
+		usb_fill_int_urb(dev->interrupt_in_urb,dev->udev,
+				 usb_rcvintpipe(dev->udev,
+				 		dev->interrupt_in_endpoint->bEndpointAddress),
+						dev->interrupt_in_buffer,
+						le16_to_cpu(dev->interrupt_in_endpoint->wMaxPacketSize),
+						adu_interrupt_in_callback,
+						dev,
+						dev->interrupt_in_endpoint->bInterval);
+		/* dev->interrupt_in_urb->transfer_flags |= URB_ASYNC_UNLINK; */
+		dev->read_urb_finished = 0;
+		usb_submit_urb(dev->interrupt_in_urb, GFP_KERNEL);
+		/* we ignore failure */
+	}
+
+exit:
+	/* unlock the device */
+	up(&dev->sem);
+
+	dbg(2," %s : leave, return value %d", __FUNCTION__, retval);
+	return retval;
+}
+
+static ssize_t adu_write(struct file *file, const __user char *buffer,
+			 size_t count, loff_t *ppos)
+{
+	struct adu_device *dev;
+	size_t bytes_written = 0;
+	size_t bytes_to_write;
+	size_t buffer_size;
+	int retval = 0;
+	int timeout = 0;
+
+	dbg(2," %s : enter, count = %Zd", __FUNCTION__, count);
+
+	dev = file->private_data;
+
+	/* lock this object */
+	down_interruptible(&dev->sem);
+
+	/* verify that the device wasn't unplugged */
+	if (dev->udev == NULL || dev->minor == 0) {
+		retval = -ENODEV;
+		err("No device or device unplugged %d", retval);
+		goto exit;
+	}
+
+	/* verify that we actually have some data to write */
+	if (count == 0) {
+		dbg(1," %s : write request of 0 bytes", __FUNCTION__);
+		goto exit;
+	}
+
+
+	while (count > 0) {
+		if (dev->interrupt_out_urb->status == -EINPROGRESS) {
+			timeout = COMMAND_TIMEOUT;
+
+			while (timeout > 0) {
+				if (signal_pending(current)) {
+				dbg(1," %s : interrupted", __FUNCTION__);
+				retval = -EINTR;
+				goto exit;
+			}
+			up(&dev->sem);
+			timeout = interruptible_sleep_on_timeout(&dev->write_wait, timeout);
+			down_interruptible(&dev->sem);
+			if (timeout > 0) {
+				break;
+			}
+			dbg(1," %s : interrupted timeout: %d", __FUNCTION__, timeout);
+		}
+
+
+		dbg(1," %s : final timeout: %d", __FUNCTION__, timeout);
+
+		if (timeout == 0) {
+			dbg(1, "%s - command timed out.", __FUNCTION__);
+			retval = -ETIMEDOUT;
+			goto exit;
+		}
+
+		dbg(4," %s : in progress, count = %Zd", __FUNCTION__, count);
+
+		} else {
+			dbg(4," %s : sending, count = %Zd", __FUNCTION__, count);
+
+			/* write the data into interrupt_out_buffer from userspace */
+			buffer_size = le16_to_cpu(dev->interrupt_out_endpoint->wMaxPacketSize);
+			bytes_to_write = count > buffer_size ? buffer_size : count;
+			dbg(4," %s : buffer_size = %Zd, count = %Zd, bytes_to_write = %Zd",
+			    __FUNCTION__, buffer_size, count, bytes_to_write);
+
+			if (copy_from_user(dev->interrupt_out_buffer, buffer, bytes_to_write) != 0) {
+				retval = -EFAULT;
+				goto exit;
+			}
+
+			/* send off the urb */
+			usb_fill_int_urb(
+				dev->interrupt_out_urb,
+				dev->udev,
+				usb_sndintpipe(dev->udev, dev->interrupt_out_endpoint->bEndpointAddress),
+				dev->interrupt_out_buffer,
+				bytes_to_write,
+				adu_interrupt_out_callback,
+				dev,
+				dev->interrupt_in_endpoint->bInterval);
+			/* dev->interrupt_in_urb->transfer_flags |= URB_ASYNC_UNLINK; */
+			dev->interrupt_out_urb->actual_length = bytes_to_write;
+			retval = usb_submit_urb(dev->interrupt_out_urb, GFP_KERNEL);
+			if (retval < 0) {
+				err("Couldn't submit interrupt_out_urb %d", retval);
+				goto exit;
+			}
+
+			buffer += bytes_to_write;
+			count -= bytes_to_write;
+
+			bytes_written += bytes_to_write;
+		}
+	}
+
+	retval = bytes_written;
+
+exit:
+	/* unlock the device */
+	up(&dev->sem);
+
+	dbg(2," %s : leave, return value %d", __FUNCTION__, retval);
+
+	return retval;
+}
+
+/* file operations needed when we register this driver */
+static struct file_operations adu_fops = {
+	.owner = THIS_MODULE,
+	.read  = adu_read,
+	.write = adu_write,
+	.open = adu_open,
+	.release = adu_release,
+};
+
+/*
+ * usb class driver info in order to get a minor number from the usb core,
+ * and to have the device registered with devfs and the driver core
+ */
+static struct usb_class_driver adu_class = {
+	.name = "usb/adutux%d",
+	.fops = &adu_fops,
+	.minor_base = ADU_MINOR_BASE,
+};
+
+/**
+ * adu_probe
+ *
+ * Called by the usb core when a new device is connected that it thinks
+ * this driver might be interested in.
+ */
+static int adu_probe(struct usb_interface *interface,
+		     const struct usb_device_id *id)
+{
+	struct usb_device *udev = interface_to_usbdev(interface);
+	struct adu_device *dev = NULL;
+	struct usb_host_interface *iface_desc;
+	struct usb_endpoint_descriptor *endpoint;
+	int retval = -ENODEV;
+	int in_end_size;
+	int out_end_size;
+	int i;
+
+	dbg(2," %s : enter", __FUNCTION__);
+
+	if (udev == NULL) {
+		dev_err(&interface->dev, "udev is NULL.\n");
+		goto exit;
+	}
+
+	/* allocate memory for our device state and intialize it */
+	dev = kzalloc(sizeof(struct adu_device), GFP_KERNEL);
+	if (dev == NULL) {
+		dev_err(&interface->dev, "Out of memory\n");
+		retval = -ENOMEM;
+		goto exit;
+	}
+
+	init_MUTEX(&dev->sem);
+	spin_lock_init(&dev->buflock);
+	dev->udev = udev;
+	init_waitqueue_head(&dev->read_wait);
+	init_waitqueue_head(&dev->write_wait);
+
+	iface_desc = &interface->altsetting[0];
+
+	/* set up the endpoint information */
+	for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
+		endpoint = &iface_desc->endpoint[i].desc;
+
+		if (usb_endpoint_is_int_in(endpoint))
+			dev->interrupt_in_endpoint = endpoint;
+
+		if (usb_endpoint_is_int_out(endpoint))
+			dev->interrupt_out_endpoint = endpoint;
+	}
+	if (dev->interrupt_in_endpoint == NULL) {
+		dev_err(&interface->dev, "interrupt in endpoint not found\n");
+		goto error;
+	}
+	if (dev->interrupt_out_endpoint == NULL) {
+		dev_err(&interface->dev, "interrupt out endpoint not found\n");
+		goto error;
+	}
+
+	in_end_size = le16_to_cpu(dev->interrupt_in_endpoint->wMaxPacketSize);
+	out_end_size = le16_to_cpu(dev->interrupt_out_endpoint->wMaxPacketSize);
+
+	dev->read_buffer_primary = kmalloc((4 * in_end_size), GFP_KERNEL);
+	if (!dev->read_buffer_primary) {
+		dev_err(&interface->dev, "Couldn't allocate read_buffer_primary\n");
+		retval = -ENOMEM;
+		goto error;
+	}
+
+	/* debug code prime the buffer */
+	memset(dev->read_buffer_primary, 'a', in_end_size);
+	memset(dev->read_buffer_primary + in_end_size, 'b', in_end_size);
+	memset(dev->read_buffer_primary + (2 * in_end_size), 'c', in_end_size);
+	memset(dev->read_buffer_primary + (3 * in_end_size), 'd', in_end_size);
+
+	dev->read_buffer_secondary = kmalloc((4 * in_end_size), GFP_KERNEL);
+	if (!dev->read_buffer_secondary) {
+		dev_err(&interface->dev, "Couldn't allocate read_buffer_secondary\n");
+		retval = -ENOMEM;
+		goto error;
+	}
+
+	/* debug code prime the buffer */
+	memset(dev->read_buffer_secondary, 'e', in_end_size);
+	memset(dev->read_buffer_secondary + in_end_size, 'f', in_end_size);
+	memset(dev->read_buffer_secondary + (2 * in_end_size), 'g', in_end_size);
+	memset(dev->read_buffer_secondary + (3 * in_end_size), 'h', in_end_size);
+
+	dev->interrupt_in_buffer = kmalloc(in_end_size, GFP_KERNEL);
+	if (!dev->interrupt_in_buffer) {
+		dev_err(&interface->dev, "Couldn't allocate interrupt_in_buffer\n");
+		goto error;
+	}
+
+	/* debug code prime the buffer */
+	memset(dev->interrupt_in_buffer, 'i', in_end_size);
+
+	dev->interrupt_in_urb = usb_alloc_urb(0, GFP_KERNEL);
+	if (!dev->interrupt_in_urb) {
+		dev_err(&interface->dev, "Couldn't allocate interrupt_in_urb\n");
+		goto error;
+	}
+	dev->interrupt_out_buffer = kmalloc(out_end_size, GFP_KERNEL);
+	if (!dev->interrupt_out_buffer) {
+		dev_err(&interface->dev, "Couldn't allocate interrupt_out_buffer\n");
+		goto error;
+	}
+	dev->interrupt_out_urb = usb_alloc_urb(0, GFP_KERNEL);
+	if (!dev->interrupt_out_urb) {
+		dev_err(&interface->dev, "Couldn't allocate interrupt_out_urb\n");
+		goto error;
+	}
+
+	if (!usb_string(udev, udev->descriptor.iSerialNumber, dev->serial_number,
+			sizeof(dev->serial_number))) {
+		dev_err(&interface->dev, "Could not retrieve serial number\n");
+		goto error;
+	}
+	dbg(2," %s : serial_number=%s", __FUNCTION__, dev->serial_number);
+
+	/* we can register the device now, as it is ready */
+	usb_set_intfdata(interface, dev);
+
+	retval = usb_register_dev(interface, &adu_class);
+
+	if (retval) {
+		/* something prevented us from registering this driver */
+		dev_err(&interface->dev, "Not able to get a minor for this device.\n");
+		usb_set_intfdata(interface, NULL);
+		goto error;
+	}
+
+	dev->minor = interface->minor;
+
+	/* let the user know what node this device is now attached to */
+	dev_info(&interface->dev, "ADU%d %s now attached to /dev/usb/adutux%d",
+		 udev->descriptor.idProduct, dev->serial_number,
+		 (dev->minor - ADU_MINOR_BASE));
+exit:
+	dbg(2," %s : leave, return value %p (dev)", __FUNCTION__, dev);
+
+	return retval;
+
+error:
+	adu_delete(dev);
+	return retval;
+}
+
+/**
+ * adu_disconnect
+ *
+ * Called by the usb core when the device is removed from the system.
+ */
+static void adu_disconnect(struct usb_interface *interface)
+{
+	struct adu_device *dev;
+	int minor;
+
+	dbg(2," %s : enter", __FUNCTION__);
+
+	mutex_lock(&disconnect_mutex); /* not interruptible */
+
+	dev = usb_get_intfdata(interface);
+	usb_set_intfdata(interface, NULL);
+
+	down(&dev->sem); /* not interruptible */
+
+	minor = dev->minor;
+
+	/* give back our minor */
+	usb_deregister_dev(interface, &adu_class);
+	dev->minor = 0;
+
+	/* if the device is not opened, then we clean up right now */
+	dbg(2," %s : open count %d", __FUNCTION__, dev->open_count);
+	if (!dev->open_count) {
+		up(&dev->sem);
+		adu_delete(dev);
+	} else {
+		dev->udev = NULL;
+		up(&dev->sem);
+	}
+
+	mutex_unlock(&disconnect_mutex);
+
+	dev_info(&interface->dev, "ADU device adutux%d now disconnected",
+		 (minor - ADU_MINOR_BASE));
+
+	dbg(2," %s : leave", __FUNCTION__);
+}
+
+/* usb specific object needed to register this driver with the usb subsystem */
+static struct usb_driver adu_driver = {
+	.name = "adutux",
+	.probe = adu_probe,
+	.disconnect = adu_disconnect,
+	.id_table = device_table,
+};
+
+static int __init adu_init(void)
+{
+	int result;
+
+	dbg(2," %s : enter", __FUNCTION__);
+
+	/* register this driver with the USB subsystem */
+	result = usb_register(&adu_driver);
+	if (result < 0) {
+		err("usb_register failed for the "__FILE__" driver. "
+		    "Error number %d", result);
+		goto exit;
+	}
+
+	info("adutux " DRIVER_DESC " " DRIVER_VERSION);
+	info("adutux is an experimental driver. Use at your own risk");
+
+exit:
+	dbg(2," %s : leave, return value %d", __FUNCTION__, result);
+
+	return result;
+}
+
+static void __exit adu_exit(void)
+{
+	dbg(2," %s : enter", __FUNCTION__);
+	/* deregister this driver with the USB subsystem */
+	usb_deregister(&adu_driver);
+	dbg(2," %s : leave", __FUNCTION__);
+}
+
+module_init(adu_init);
+module_exit(adu_exit);
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 5638e4d92e7707bd037a36654f914c80ccd7161d Mon Sep 17 00:00:00 2001
From: Sam Hocevar <sam@zoy.org>
Date: Wed, 30 Aug 2006 02:34:56 +0200
Subject: USB: add PlayStation 2 Trance Vibrator driver

This patch is a driver for the PlayStation 2 specific Trance Vibrator
device. The only thing that device can do is vibrate at various speeds.

Signed-off-by: Sam Hocevar <sam@zoy.org>
Cc: Pete Zaitcev <zaitcev@redhat.com>
Cc: Luiz Fernando N. Capitulino" <lcapitulino@mandriva.com.br>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/Makefile               |   3 +-
 drivers/usb/input/Kconfig          |  10 +++
 drivers/usb/input/Makefile         |   1 +
 drivers/usb/input/trancevibrator.c | 159 +++++++++++++++++++++++++++++++++++++
 4 files changed, 172 insertions(+), 1 deletion(-)
 create mode 100644 drivers/usb/input/trancevibrator.c

diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index 84430db4b8d9..46946b2d3088 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -23,6 +23,7 @@ obj-$(CONFIG_USB_PRINTER)	+= class/
 obj-$(CONFIG_USB_STORAGE)	+= storage/
 obj-$(CONFIG_USB)		+= storage/
 
+obj-$(CONFIG_USB_ACECAD)	+= input/
 obj-$(CONFIG_USB_AIPTEK)	+= input/
 obj-$(CONFIG_USB_ATI_REMOTE)	+= input/
 obj-$(CONFIG_USB_HID)		+= input/
@@ -31,8 +32,8 @@ obj-$(CONFIG_USB_KBTAB)		+= input/
 obj-$(CONFIG_USB_MOUSE)		+= input/
 obj-$(CONFIG_USB_MTOUCH)	+= input/
 obj-$(CONFIG_USB_POWERMATE)	+= input/
+obj-$(CONFIG_USB_TRANCEVIBRATOR)+= input/
 obj-$(CONFIG_USB_WACOM)		+= input/
-obj-$(CONFIG_USB_ACECAD)	+= input/
 obj-$(CONFIG_USB_XPAD)		+= input/
 
 obj-$(CONFIG_USB_CATC)		+= net/
diff --git a/drivers/usb/input/Kconfig b/drivers/usb/input/Kconfig
index 8a708520547b..a102a58fe361 100644
--- a/drivers/usb/input/Kconfig
+++ b/drivers/usb/input/Kconfig
@@ -338,3 +338,13 @@ config USB_APPLETOUCH
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called appletouch.
+
+config USB_TRANCEVIBRATOR
+	tristate "PlayStation 2 Trance Vibrator driver support"
+	depends on USB
+	help
+	  Say Y here if you want to connect a PlayStation 2 Trance Vibrator
+	  device to your computer's USB port.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called trancevibrator.
diff --git a/drivers/usb/input/Makefile b/drivers/usb/input/Makefile
index b220435c2409..48551be324ac 100644
--- a/drivers/usb/input/Makefile
+++ b/drivers/usb/input/Makefile
@@ -45,6 +45,7 @@ obj-$(CONFIG_USB_ACECAD)	+= acecad.o
 obj-$(CONFIG_USB_YEALINK)	+= yealink.o
 obj-$(CONFIG_USB_XPAD)		+= xpad.o
 obj-$(CONFIG_USB_APPLETOUCH)	+= appletouch.o
+obj-$(CONFIG_USB_TRANCEVIBRATOR)	+= trancevibrator.o
 
 ifeq ($(CONFIG_USB_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/drivers/usb/input/trancevibrator.c b/drivers/usb/input/trancevibrator.c
new file mode 100644
index 000000000000..33cd91d11eca
--- /dev/null
+++ b/drivers/usb/input/trancevibrator.c
@@ -0,0 +1,159 @@
+/*
+ * PlayStation 2 Trance Vibrator driver
+ *
+ * Copyright (C) 2006 Sam Hocevar <sam@zoy.org>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ */
+
+/* Standard include files */
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/usb.h>
+
+/* Version Information */
+#define DRIVER_VERSION "v1.1"
+#define DRIVER_AUTHOR "Sam Hocevar, sam@zoy.org"
+#define DRIVER_DESC "PlayStation 2 Trance Vibrator driver"
+
+#define TRANCEVIBRATOR_VENDOR_ID	0x0b49	/* ASCII Corporation */
+#define TRANCEVIBRATOR_PRODUCT_ID	0x064f	/* Trance Vibrator */
+
+static struct usb_device_id id_table [] = {
+	{ USB_DEVICE(TRANCEVIBRATOR_VENDOR_ID, TRANCEVIBRATOR_PRODUCT_ID) },
+	{ },
+};
+MODULE_DEVICE_TABLE (usb, id_table);
+
+/* Driver-local specific stuff */
+struct trancevibrator {
+	struct usb_device *udev;
+	unsigned int speed;
+};
+
+static ssize_t show_speed(struct device *dev, struct device_attribute *attr,
+			  char *buf)
+{
+	struct usb_interface *intf = to_usb_interface(dev);
+	struct trancevibrator *tv = usb_get_intfdata(intf);
+
+	return sprintf(buf, "%d\n", tv->speed);
+}
+
+static ssize_t set_speed(struct device *dev, struct device_attribute *attr,
+			 const char *buf, size_t count)
+{
+	struct usb_interface *intf = to_usb_interface(dev);
+	struct trancevibrator *tv = usb_get_intfdata(intf);
+	int temp, retval;
+
+	temp = simple_strtoul(buf, NULL, 10);
+	if (temp > 255)
+		temp = 255;
+	else if (temp < 0)
+		temp = 0;
+	tv->speed = temp;
+
+	dev_dbg(&tv->udev->dev, "speed = %d\n", tv->speed);
+
+	/* Set speed */
+	retval = usb_control_msg(tv->udev, usb_sndctrlpipe(tv->udev, 0),
+				 0x01, /* vendor request: set speed */
+				 USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_OTHER,
+				 tv->speed, /* speed value */
+				 0, NULL, 0, USB_CTRL_GET_TIMEOUT);
+	if (retval) {
+		dev_dbg(&tv->udev->dev, "retval = %d\n", retval);
+		return retval;
+	}
+	return count;
+}
+
+static DEVICE_ATTR(speed, S_IWUGO | S_IRUGO, show_speed, set_speed);
+
+static int tv_probe(struct usb_interface *interface,
+		    const struct usb_device_id *id)
+{
+	struct usb_device *udev = interface_to_usbdev(interface);
+	struct trancevibrator *dev;
+	int retval;
+
+	dev = kzalloc(sizeof(struct trancevibrator), GFP_KERNEL);
+	if (dev == NULL) {
+		dev_err(&interface->dev, "Out of memory\n");
+		retval = -ENOMEM;
+		goto error;
+	}
+
+	dev->udev = usb_get_dev(udev);
+	usb_set_intfdata(interface, dev);
+	retval = device_create_file(&interface->dev, &dev_attr_speed);
+	if (retval)
+		goto error_create_file;
+
+	return 0;
+
+error_create_file:
+	usb_put_dev(udev);
+	usb_set_intfdata(interface, NULL);
+error:
+	kfree(dev);
+	return retval;
+}
+
+static void tv_disconnect(struct usb_interface *interface)
+{
+	struct trancevibrator *dev;
+
+	dev = usb_get_intfdata (interface);
+	usb_set_intfdata(interface, NULL);
+	device_remove_file(&interface->dev, &dev_attr_speed);
+	usb_put_dev(dev->udev);
+	kfree(dev);
+}
+
+/* USB subsystem object */
+static struct usb_driver tv_driver = {
+	.name =		"trancevibrator",
+	.probe =	tv_probe,
+	.disconnect =	tv_disconnect,
+	.id_table =	id_table,
+};
+
+static int __init tv_init(void)
+{
+	int retval = usb_register(&tv_driver);
+	if (retval) {
+		err("usb_register failed. Error number %d", retval);
+		return retval;
+	}
+
+	info(DRIVER_VERSION ":" DRIVER_DESC);
+	return 0;
+}
+
+static void __exit tv_exit(void)
+{
+	usb_deregister(&tv_driver);
+}
+
+module_init (tv_init);
+module_exit (tv_exit);
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From 3f5429746d91f21f60f68b14177c0d534d80240b Mon Sep 17 00:00:00 2001
From: Paul B Schroeder <pschroeder@uplogix.com>
Date: Thu, 31 Aug 2006 19:41:47 -0500
Subject: USB: Moschip 7840 USB-Serial Driver

Signed-off-by: Paul B Schroeder <pschroeder@uplogix.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/Kconfig   |   15 +
 drivers/usb/serial/Makefile  |    1 +
 drivers/usb/serial/mos7840.c | 2962 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 2978 insertions(+)
 create mode 100644 drivers/usb/serial/mos7840.c

diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig
index e4df0ac358fb..5076b9d97057 100644
--- a/drivers/usb/serial/Kconfig
+++ b/drivers/usb/serial/Kconfig
@@ -422,6 +422,21 @@ config USB_SERIAL_MCT_U232
 	  To compile this driver as a module, choose M here: the
 	  module will be called mct_u232.
 
+config USB_SERIAL_MOS7840
+	tristate "USB Moschip 7840/7820 USB Serial Driver"
+	depends on USB_SERIAL
+	---help---
+	  Say Y here if you want to use a MCS7840 Quad-Serial or MCS7820
+	  Dual-Serial port device from MosChip Semiconductor.
+
+	  The MCS7840 and MCS7820 have been developed to connect a wide range
+	  of standard serial devices to a USB host.  The MCS7840 has a USB
+	  device controller connected to four (4) individual UARTs while the
+	  MCS7820 controller connects to two (2) individual UARTs.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called mos7840.  If unsure, choose N.
+
 config USB_SERIAL_NAVMAN
 	tristate "USB Navman GPS device"
 	depends on USB_SERIAL
diff --git a/drivers/usb/serial/Makefile b/drivers/usb/serial/Makefile
index 029e2a1d7bf6..8dce83340e31 100644
--- a/drivers/usb/serial/Makefile
+++ b/drivers/usb/serial/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_USB_SERIAL_KEYSPAN_PDA)		+= keyspan_pda.o
 obj-$(CONFIG_USB_SERIAL_KLSI)			+= kl5kusb105.o
 obj-$(CONFIG_USB_SERIAL_KOBIL_SCT)		+= kobil_sct.o
 obj-$(CONFIG_USB_SERIAL_MCT_U232)		+= mct_u232.o
+obj-$(CONFIG_USB_SERIAL_MOS7840)		+= mos7840.o
 obj-$(CONFIG_USB_SERIAL_NAVMAN)			+= navman.o
 obj-$(CONFIG_USB_SERIAL_OMNINET)		+= omninet.o
 obj-$(CONFIG_USB_SERIAL_OPTION)			+= option.o
diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c
new file mode 100644
index 000000000000..95bf57166c59
--- /dev/null
+++ b/drivers/usb/serial/mos7840.c
@@ -0,0 +1,2962 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ * Clean ups from Moschip version and a few ioctl implementations by:
+ *	Paul B Schroeder <pschroeder "at" uplogix "dot" com>
+ *
+ * Originally based on drivers/usb/serial/io_edgeport.c which is:
+ *      Copyright (C) 2000 Inside Out Networks, All rights reserved.
+ *      Copyright (C) 2001-2002 Greg Kroah-Hartman <greg@kroah.com>
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/tty.h>
+#include <linux/tty_driver.h>
+#include <linux/tty_flip.h>
+#include <linux/module.h>
+#include <linux/serial.h>
+#include <linux/usb.h>
+#include <linux/usb/serial.h>
+#include <asm/uaccess.h>
+
+/*
+ * Version Information
+ */
+#define DRIVER_VERSION "1.3.1"
+#define DRIVER_DESC "Moschip 7840/7820 USB Serial Driver"
+
+/*
+ * 16C50 UART register defines
+ */
+
+#define LCR_BITS_5             0x00	/* 5 bits/char */
+#define LCR_BITS_6             0x01	/* 6 bits/char */
+#define LCR_BITS_7             0x02	/* 7 bits/char */
+#define LCR_BITS_8             0x03	/* 8 bits/char */
+#define LCR_BITS_MASK          0x03	/* Mask for bits/char field */
+
+#define LCR_STOP_1             0x00	/* 1 stop bit */
+#define LCR_STOP_1_5           0x04	/* 1.5 stop bits (if 5   bits/char) */
+#define LCR_STOP_2             0x04	/* 2 stop bits   (if 6-8 bits/char) */
+#define LCR_STOP_MASK          0x04	/* Mask for stop bits field */
+
+#define LCR_PAR_NONE           0x00	/* No parity */
+#define LCR_PAR_ODD            0x08	/* Odd parity */
+#define LCR_PAR_EVEN           0x18	/* Even parity */
+#define LCR_PAR_MARK           0x28	/* Force parity bit to 1 */
+#define LCR_PAR_SPACE          0x38	/* Force parity bit to 0 */
+#define LCR_PAR_MASK           0x38	/* Mask for parity field */
+
+#define LCR_SET_BREAK          0x40	/* Set Break condition */
+#define LCR_DL_ENABLE          0x80	/* Enable access to divisor latch */
+
+#define MCR_DTR                0x01	/* Assert DTR */
+#define MCR_RTS                0x02	/* Assert RTS */
+#define MCR_OUT1               0x04	/* Loopback only: Sets state of RI */
+#define MCR_MASTER_IE          0x08	/* Enable interrupt outputs */
+#define MCR_LOOPBACK           0x10	/* Set internal (digital) loopback mode */
+#define MCR_XON_ANY            0x20	/* Enable any char to exit XOFF mode */
+
+#define MOS7840_MSR_CTS        0x10	/* Current state of CTS */
+#define MOS7840_MSR_DSR        0x20	/* Current state of DSR */
+#define MOS7840_MSR_RI         0x40	/* Current state of RI */
+#define MOS7840_MSR_CD         0x80	/* Current state of CD */
+
+/*
+ * Defines used for sending commands to port
+ */
+
+#define WAIT_FOR_EVER   (HZ * 0 )	/* timeout urb is wait for ever */
+#define MOS_WDR_TIMEOUT (HZ * 5 )	/* default urb timeout */
+
+#define MOS_PORT1       0x0200
+#define MOS_PORT2       0x0300
+#define MOS_VENREG      0x0000
+#define MOS_MAX_PORT	0x02
+#define MOS_WRITE       0x0E
+#define MOS_READ        0x0D
+
+/* Requests */
+#define MCS_RD_RTYPE    0xC0
+#define MCS_WR_RTYPE    0x40
+#define MCS_RDREQ       0x0D
+#define MCS_WRREQ       0x0E
+#define MCS_CTRL_TIMEOUT        500
+#define VENDOR_READ_LENGTH      (0x01)
+
+#define MAX_NAME_LEN    64
+
+#define ZLP_REG1  0x3A		//Zero_Flag_Reg1    58
+#define ZLP_REG5  0x3E		//Zero_Flag_Reg5    62
+
+/* For higher baud Rates use TIOCEXBAUD */
+#define TIOCEXBAUD     0x5462
+
+/* vendor id and device id defines */
+
+#define USB_VENDOR_ID_MOSCHIP           0x9710
+#define MOSCHIP_DEVICE_ID_7840          0x7840
+#define MOSCHIP_DEVICE_ID_7820          0x7820
+
+/* Interrupt Rotinue Defines    */
+
+#define SERIAL_IIR_RLS      0x06
+#define SERIAL_IIR_MS       0x00
+
+/*
+ *  Emulation of the bit mask on the LINE STATUS REGISTER.
+ */
+#define SERIAL_LSR_DR       0x0001
+#define SERIAL_LSR_OE       0x0002
+#define SERIAL_LSR_PE       0x0004
+#define SERIAL_LSR_FE       0x0008
+#define SERIAL_LSR_BI       0x0010
+
+#define MOS_MSR_DELTA_CTS   0x10
+#define MOS_MSR_DELTA_DSR   0x20
+#define MOS_MSR_DELTA_RI    0x40
+#define MOS_MSR_DELTA_CD    0x80
+
+// Serial Port register Address
+#define INTERRUPT_ENABLE_REGISTER  ((__u16)(0x01))
+#define FIFO_CONTROL_REGISTER      ((__u16)(0x02))
+#define LINE_CONTROL_REGISTER      ((__u16)(0x03))
+#define MODEM_CONTROL_REGISTER     ((__u16)(0x04))
+#define LINE_STATUS_REGISTER       ((__u16)(0x05))
+#define MODEM_STATUS_REGISTER      ((__u16)(0x06))
+#define SCRATCH_PAD_REGISTER       ((__u16)(0x07))
+#define DIVISOR_LATCH_LSB          ((__u16)(0x00))
+#define DIVISOR_LATCH_MSB          ((__u16)(0x01))
+
+#define CLK_MULTI_REGISTER         ((__u16)(0x02))
+#define CLK_START_VALUE_REGISTER   ((__u16)(0x03))
+
+#define SERIAL_LCR_DLAB            ((__u16)(0x0080))
+
+/*
+ * URB POOL related defines
+ */
+#define NUM_URBS                        16	/* URB Count */
+#define URB_TRANSFER_BUFFER_SIZE        32	/* URB Size  */
+
+
+static struct usb_device_id moschip_port_id_table[] = {
+	{USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7840)},
+	{USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7820)},
+	{}			/* terminating entry */
+};
+
+static __devinitdata struct usb_device_id moschip_id_table_combined[] = {
+	{USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7840)},
+	{USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7820)},
+	{}			/* terminating entry */
+};
+
+MODULE_DEVICE_TABLE(usb, moschip_id_table_combined);
+
+/* This structure holds all of the local port information */
+
+struct moschip_port {
+	int port_num;		/*Actual port number in the device(1,2,etc) */
+	struct urb *write_urb;	/* write URB for this port */
+	struct urb *read_urb;	/* read URB for this port */
+	__u8 shadowLCR;		/* last LCR value received */
+	__u8 shadowMCR;		/* last MCR value received */
+	char open;
+	wait_queue_head_t wait_chase;	/* for handling sleeping while waiting for chase to finish */
+	wait_queue_head_t delta_msr_wait;	/* for handling sleeping while waiting for msr change to happen */
+	int delta_msr_cond;
+	struct async_icount icount;
+	struct usb_serial_port *port;	/* loop back to the owner of this object */
+
+	/*Offsets */
+	__u8 SpRegOffset;
+	__u8 ControlRegOffset;
+	__u8 DcrRegOffset;
+	//for processing control URBS in interrupt context
+	struct urb *control_urb;
+	char *ctrl_buf;
+	int MsrLsr;
+
+	struct urb *write_urb_pool[NUM_URBS];
+};
+
+
+static int debug;
+static int mos7840_num_ports;	//this says the number of ports in the device
+static int mos7840_num_open_ports;
+
+
+/*
+ * mos7840_set_reg_sync
+ * 	To set the Control register by calling usb_fill_control_urb function
+ *	by passing usb_sndctrlpipe function as parameter.
+ */
+
+static int mos7840_set_reg_sync(struct usb_serial_port *port, __u16 reg,
+				__u16 val)
+{
+	struct usb_device *dev = port->serial->dev;
+	val = val & 0x00ff;
+	dbg("mos7840_set_reg_sync offset is %x, value %x\n", reg, val);
+
+	return usb_control_msg(dev, usb_sndctrlpipe(dev, 0), MCS_WRREQ,
+			       MCS_WR_RTYPE, val, reg, NULL, 0,
+			       MOS_WDR_TIMEOUT);
+}
+
+/*
+ * mos7840_get_reg_sync
+ * 	To set the Uart register by calling usb_fill_control_urb function by
+ *	passing usb_rcvctrlpipe function as parameter.
+ */
+
+static int mos7840_get_reg_sync(struct usb_serial_port *port, __u16 reg,
+				__u16 * val)
+{
+	struct usb_device *dev = port->serial->dev;
+	int ret = 0;
+
+	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0), MCS_RDREQ,
+			      MCS_RD_RTYPE, 0, reg, val, VENDOR_READ_LENGTH,
+			      MOS_WDR_TIMEOUT);
+	dbg("mos7840_get_reg_sync offset is %x, return val %x\n", reg, *val);
+	*val = (*val) & 0x00ff;
+	return ret;
+}
+
+/*
+ * mos7840_set_uart_reg
+ *	To set the Uart register by calling usb_fill_control_urb function by
+ *	passing usb_sndctrlpipe function as parameter.
+ */
+
+static int mos7840_set_uart_reg(struct usb_serial_port *port, __u16 reg,
+				__u16 val)
+{
+
+	struct usb_device *dev = port->serial->dev;
+	val = val & 0x00ff;
+	// For the UART control registers, the application number need to be Or'ed
+	if (mos7840_num_ports == 4) {
+		val |=
+		    (((__u16) port->number - (__u16) (port->serial->minor)) +
+		     1) << 8;
+		dbg("mos7840_set_uart_reg application number is %x\n", val);
+	} else {
+		if (((__u16) port->number - (__u16) (port->serial->minor)) == 0) {
+			val |=
+			    (((__u16) port->number -
+			      (__u16) (port->serial->minor)) + 1) << 8;
+			dbg("mos7840_set_uart_reg application number is %x\n",
+			    val);
+		} else {
+			val |=
+			    (((__u16) port->number -
+			      (__u16) (port->serial->minor)) + 2) << 8;
+			dbg("mos7840_set_uart_reg application number is %x\n",
+			    val);
+		}
+	}
+	return usb_control_msg(dev, usb_sndctrlpipe(dev, 0), MCS_WRREQ,
+			       MCS_WR_RTYPE, val, reg, NULL, 0,
+			       MOS_WDR_TIMEOUT);
+
+}
+
+/*
+ * mos7840_get_uart_reg
+ *	To set the Control register by calling usb_fill_control_urb function
+ *	by passing usb_rcvctrlpipe function as parameter.
+ */
+static int mos7840_get_uart_reg(struct usb_serial_port *port, __u16 reg,
+				__u16 * val)
+{
+	struct usb_device *dev = port->serial->dev;
+	int ret = 0;
+	__u16 Wval;
+
+	//dbg("application number is %4x \n",(((__u16)port->number - (__u16)(port->serial->minor))+1)<<8);
+	/*Wval  is same as application number */
+	if (mos7840_num_ports == 4) {
+		Wval =
+		    (((__u16) port->number - (__u16) (port->serial->minor)) +
+		     1) << 8;
+		dbg("mos7840_get_uart_reg application number is %x\n", Wval);
+	} else {
+		if (((__u16) port->number - (__u16) (port->serial->minor)) == 0) {
+			Wval =
+			    (((__u16) port->number -
+			      (__u16) (port->serial->minor)) + 1) << 8;
+			dbg("mos7840_get_uart_reg application number is %x\n",
+			    Wval);
+		} else {
+			Wval =
+			    (((__u16) port->number -
+			      (__u16) (port->serial->minor)) + 2) << 8;
+			dbg("mos7840_get_uart_reg application number is %x\n",
+			    Wval);
+		}
+	}
+	ret = usb_control_msg(dev, usb_rcvctrlpipe(dev, 0), MCS_RDREQ,
+			      MCS_RD_RTYPE, Wval, reg, val, VENDOR_READ_LENGTH,
+			      MOS_WDR_TIMEOUT);
+	*val = (*val) & 0x00ff;
+	return ret;
+}
+
+static void mos7840_dump_serial_port(struct moschip_port *mos7840_port)
+{
+
+	dbg("***************************************\n");
+	dbg("SpRegOffset is %2x\n", mos7840_port->SpRegOffset);
+	dbg("ControlRegOffset is %2x \n", mos7840_port->ControlRegOffset);
+	dbg("DCRRegOffset is %2x \n", mos7840_port->DcrRegOffset);
+	dbg("***************************************\n");
+
+}
+
+/************************************************************************/
+/************************************************************************/
+/*             I N T E R F A C E   F U N C T I O N S			*/
+/*             I N T E R F A C E   F U N C T I O N S			*/
+/************************************************************************/
+/************************************************************************/
+
+static inline void mos7840_set_port_private(struct usb_serial_port *port,
+					    struct moschip_port *data)
+{
+	usb_set_serial_port_data(port, (void *)data);
+}
+
+static inline struct moschip_port *mos7840_get_port_private(struct
+							    usb_serial_port
+							    *port)
+{
+	return (struct moschip_port *)usb_get_serial_port_data(port);
+}
+
+static int mos7840_handle_new_msr(struct moschip_port *port, __u8 new_msr)
+{
+	struct moschip_port *mos7840_port;
+	struct async_icount *icount;
+	mos7840_port = port;
+	icount = &mos7840_port->icount;
+	if (new_msr &
+	    (MOS_MSR_DELTA_CTS | MOS_MSR_DELTA_DSR | MOS_MSR_DELTA_RI |
+	     MOS_MSR_DELTA_CD)) {
+		icount = &mos7840_port->icount;
+
+		/* update input line counters */
+		if (new_msr & MOS_MSR_DELTA_CTS) {
+			icount->cts++;
+		}
+		if (new_msr & MOS_MSR_DELTA_DSR) {
+			icount->dsr++;
+		}
+		if (new_msr & MOS_MSR_DELTA_CD) {
+			icount->dcd++;
+		}
+		if (new_msr & MOS_MSR_DELTA_RI) {
+			icount->rng++;
+		}
+	}
+
+	return 0;
+}
+
+static int mos7840_handle_new_lsr(struct moschip_port *port, __u8 new_lsr)
+{
+	struct async_icount *icount;
+
+	dbg("%s - %02x", __FUNCTION__, new_lsr);
+
+	if (new_lsr & SERIAL_LSR_BI) {
+		//
+		// Parity and Framing errors only count if they
+		// occur exclusive of a break being
+		// received.
+		//
+		new_lsr &= (__u8) (SERIAL_LSR_OE | SERIAL_LSR_BI);
+	}
+
+	/* update input line counters */
+	icount = &port->icount;
+	if (new_lsr & SERIAL_LSR_BI) {
+		icount->brk++;
+	}
+	if (new_lsr & SERIAL_LSR_OE) {
+		icount->overrun++;
+	}
+	if (new_lsr & SERIAL_LSR_PE) {
+		icount->parity++;
+	}
+	if (new_lsr & SERIAL_LSR_FE) {
+		icount->frame++;
+	}
+
+	return 0;
+}
+
+/************************************************************************/
+/************************************************************************/
+/*            U S B  C A L L B A C K   F U N C T I O N S                */
+/*            U S B  C A L L B A C K   F U N C T I O N S                */
+/************************************************************************/
+/************************************************************************/
+
+static void mos7840_control_callback(struct urb *urb, struct pt_regs *regs)
+{
+	unsigned char *data;
+	struct moschip_port *mos7840_port;
+	__u8 regval = 0x0;
+
+	if (!urb) {
+		dbg("%s", "Invalid Pointer !!!!:\n");
+		return;
+	}
+
+	switch (urb->status) {
+	case 0:
+		/* success */
+		break;
+	case -ECONNRESET:
+	case -ENOENT:
+	case -ESHUTDOWN:
+		/* this urb is terminated, clean up */
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__,
+		    urb->status);
+		return;
+	default:
+		dbg("%s - nonzero urb status received: %d", __FUNCTION__,
+		    urb->status);
+		goto exit;
+	}
+
+	mos7840_port = (struct moschip_port *)urb->context;
+
+	dbg("%s urb buffer size is %d\n", __FUNCTION__, urb->actual_length);
+	dbg("%s mos7840_port->MsrLsr is %d port %d\n", __FUNCTION__,
+	    mos7840_port->MsrLsr, mos7840_port->port_num);
+	data = urb->transfer_buffer;
+	regval = (__u8) data[0];
+	dbg("%s data is %x\n", __FUNCTION__, regval);
+	if (mos7840_port->MsrLsr == 0)
+		mos7840_handle_new_msr(mos7840_port, regval);
+	else if (mos7840_port->MsrLsr == 1)
+		mos7840_handle_new_lsr(mos7840_port, regval);
+
+      exit:
+	return;
+}
+
+static int mos7840_get_reg(struct moschip_port *mcs, __u16 Wval, __u16 reg,
+			   __u16 * val)
+{
+	struct usb_device *dev = mcs->port->serial->dev;
+	struct usb_ctrlrequest *dr = NULL;
+	unsigned char *buffer = NULL;
+	int ret = 0;
+	buffer = (__u8 *) mcs->ctrl_buf;
+
+//      dr=(struct usb_ctrlrequest *)(buffer);
+	dr = (void *)(buffer + 2);
+	dr->bRequestType = MCS_RD_RTYPE;
+	dr->bRequest = MCS_RDREQ;
+	dr->wValue = cpu_to_le16(Wval);	//0;
+	dr->wIndex = cpu_to_le16(reg);
+	dr->wLength = cpu_to_le16(2);
+
+	usb_fill_control_urb(mcs->control_urb, dev, usb_rcvctrlpipe(dev, 0),
+			     (unsigned char *)dr, buffer, 2,
+			     mos7840_control_callback, mcs);
+	mcs->control_urb->transfer_buffer_length = 2;
+	ret = usb_submit_urb(mcs->control_urb, GFP_ATOMIC);
+	return ret;
+}
+
+/*****************************************************************************
+ * mos7840_interrupt_callback
+ *	this is the callback function for when we have received data on the
+ *	interrupt endpoint.
+ *****************************************************************************/
+
+static void mos7840_interrupt_callback(struct urb *urb, struct pt_regs *regs)
+{
+	int result;
+	int length;
+	struct moschip_port *mos7840_port;
+	struct usb_serial *serial;
+	__u16 Data;
+	unsigned char *data;
+	__u8 sp[5], st;
+	int i;
+	__u16 wval;
+
+	dbg("%s", " : Entering\n");
+	if (!urb) {
+		dbg("%s", "Invalid Pointer !!!!:\n");
+		return;
+	}
+
+	switch (urb->status) {
+	case 0:
+		/* success */
+		break;
+	case -ECONNRESET:
+	case -ENOENT:
+	case -ESHUTDOWN:
+		/* this urb is terminated, clean up */
+		dbg("%s - urb shutting down with status: %d", __FUNCTION__,
+		    urb->status);
+		return;
+	default:
+		dbg("%s - nonzero urb status received: %d", __FUNCTION__,
+		    urb->status);
+		goto exit;
+	}
+
+	length = urb->actual_length;
+	data = urb->transfer_buffer;
+
+	serial = (struct usb_serial *)urb->context;
+
+	/* Moschip get 5 bytes
+	 * Byte 1 IIR Port 1 (port.number is 0)
+	 * Byte 2 IIR Port 2 (port.number is 1)
+	 * Byte 3 IIR Port 3 (port.number is 2)
+	 * Byte 4 IIR Port 4 (port.number is 3)
+	 * Byte 5 FIFO status for both */
+
+	if (length && length > 5) {
+		dbg("%s \n", "Wrong data !!!");
+		return;
+	}
+
+	sp[0] = (__u8) data[0];
+	sp[1] = (__u8) data[1];
+	sp[2] = (__u8) data[2];
+	sp[3] = (__u8) data[3];
+	st = (__u8) data[4];
+
+	for (i = 0; i < serial->num_ports; i++) {
+		mos7840_port = mos7840_get_port_private(serial->port[i]);
+		wval =
+		    (((__u16) serial->port[i]->number -
+		      (__u16) (serial->minor)) + 1) << 8;
+		if (mos7840_port->open) {
+			if (sp[i] & 0x01) {
+				dbg("SP%d No Interrupt !!!\n", i);
+			} else {
+				switch (sp[i] & 0x0f) {
+				case SERIAL_IIR_RLS:
+					dbg("Serial Port %d: Receiver status error or ", i);
+					dbg("address bit detected in 9-bit mode\n");
+					mos7840_port->MsrLsr = 1;
+					mos7840_get_reg(mos7840_port, wval,
+							LINE_STATUS_REGISTER,
+							&Data);
+					break;
+				case SERIAL_IIR_MS:
+					dbg("Serial Port %d: Modem status change\n", i);
+					mos7840_port->MsrLsr = 0;
+					mos7840_get_reg(mos7840_port, wval,
+							MODEM_STATUS_REGISTER,
+							&Data);
+					break;
+				}
+			}
+		}
+	}
+      exit:
+	result = usb_submit_urb(urb, GFP_ATOMIC);
+	if (result) {
+		dev_err(&urb->dev->dev,
+			"%s - Error %d submitting interrupt urb\n",
+			__FUNCTION__, result);
+	}
+
+	return;
+
+}
+
+static int mos7840_port_paranoia_check(struct usb_serial_port *port,
+				       const char *function)
+{
+	if (!port) {
+		dbg("%s - port == NULL", function);
+		return -1;
+	}
+	if (!port->serial) {
+		dbg("%s - port->serial == NULL", function);
+		return -1;
+	}
+
+	return 0;
+}
+
+/* Inline functions to check the sanity of a pointer that is passed to us */
+static int mos7840_serial_paranoia_check(struct usb_serial *serial,
+					 const char *function)
+{
+	if (!serial) {
+		dbg("%s - serial == NULL", function);
+		return -1;
+	}
+	if (!serial->type) {
+		dbg("%s - serial->type == NULL!", function);
+		return -1;
+	}
+
+	return 0;
+}
+
+static struct usb_serial *mos7840_get_usb_serial(struct usb_serial_port *port,
+						 const char *function)
+{
+	/* if no port was specified, or it fails a paranoia check */
+	if (!port ||
+	    mos7840_port_paranoia_check(port, function) ||
+	    mos7840_serial_paranoia_check(port->serial, function)) {
+		/* then say that we don't have a valid usb_serial thing, which will                  * end up genrating -ENODEV return values */
+		return NULL;
+	}
+
+	return port->serial;
+}
+
+/*****************************************************************************
+ * mos7840_bulk_in_callback
+ *	this is the callback function for when we have received data on the
+ *	bulk in endpoint.
+ *****************************************************************************/
+
+static void mos7840_bulk_in_callback(struct urb *urb, struct pt_regs *regs)
+{
+	int status;
+	unsigned char *data;
+	struct usb_serial *serial;
+	struct usb_serial_port *port;
+	struct moschip_port *mos7840_port;
+	struct tty_struct *tty;
+
+	if (!urb) {
+		dbg("%s", "Invalid Pointer !!!!:\n");
+		return;
+	}
+
+	if (urb->status) {
+		dbg("nonzero read bulk status received: %d", urb->status);
+		return;
+	}
+
+	mos7840_port = (struct moschip_port *)urb->context;
+	if (!mos7840_port) {
+		dbg("%s", "NULL mos7840_port pointer \n");
+		return;
+	}
+
+	port = (struct usb_serial_port *)mos7840_port->port;
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return;
+	}
+
+	serial = mos7840_get_usb_serial(port, __FUNCTION__);
+	if (!serial) {
+		dbg("%s\n", "Bad serial pointer ");
+		return;
+	}
+
+	dbg("%s\n", "Entering... \n");
+
+	data = urb->transfer_buffer;
+
+	dbg("%s", "Entering ........... \n");
+
+	if (urb->actual_length) {
+		tty = mos7840_port->port->tty;
+		if (tty) {
+			tty_buffer_request_room(tty, urb->actual_length);
+			tty_insert_flip_string(tty, data, urb->actual_length);
+			dbg(" %s \n", data);
+			tty_flip_buffer_push(tty);
+		}
+		mos7840_port->icount.rx += urb->actual_length;
+		dbg("mos7840_port->icount.rx is %d:\n",
+		    mos7840_port->icount.rx);
+	}
+
+	if (!mos7840_port->read_urb) {
+		dbg("%s", "URB KILLED !!!\n");
+		return;
+	}
+
+	if (mos7840_port->read_urb->status != -EINPROGRESS) {
+		mos7840_port->read_urb->dev = serial->dev;
+
+		status = usb_submit_urb(mos7840_port->read_urb, GFP_ATOMIC);
+
+		if (status) {
+			dbg(" usb_submit_urb(read bulk) failed, status = %d",
+			    status);
+		}
+	}
+}
+
+/*****************************************************************************
+ * mos7840_bulk_out_data_callback
+ *	this is the callback function for when we have finished sending serial data
+ *	on the bulk out endpoint.
+ *****************************************************************************/
+
+static void mos7840_bulk_out_data_callback(struct urb *urb,
+					   struct pt_regs *regs)
+{
+	struct moschip_port *mos7840_port;
+	struct tty_struct *tty;
+	if (!urb) {
+		dbg("%s", "Invalid Pointer !!!!:\n");
+		return;
+	}
+
+	if (urb->status) {
+		dbg("nonzero write bulk status received:%d\n", urb->status);
+		return;
+	}
+
+	mos7840_port = (struct moschip_port *)urb->context;
+	if (!mos7840_port) {
+		dbg("%s", "NULL mos7840_port pointer \n");
+		return;
+	}
+
+	if (mos7840_port_paranoia_check(mos7840_port->port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return;
+	}
+
+	dbg("%s \n", "Entering .........");
+
+	tty = mos7840_port->port->tty;
+
+	if (tty && mos7840_port->open) {
+		/* let the tty driver wakeup if it has a special *
+		 * write_wakeup function                         */
+
+		if ((tty->flags & (1 << TTY_DO_WRITE_WAKEUP))
+		    && tty->ldisc.write_wakeup) {
+			(tty->ldisc.write_wakeup) (tty);
+		}
+
+		/* tell the tty driver that something has changed */
+		wake_up_interruptible(&tty->write_wait);
+	}
+
+}
+
+/************************************************************************/
+/*       D R I V E R  T T Y  I N T E R F A C E  F U N C T I O N S       */
+/************************************************************************/
+#ifdef MCSSerialProbe
+static int mos7840_serial_probe(struct usb_serial *serial,
+				const struct usb_device_id *id)
+{
+
+	/*need to implement the mode_reg reading and updating\
+	   structures usb_serial_ device_type\
+	   (i.e num_ports, num_bulkin,bulkout etc) */
+	/* Also we can update the changes  attach */
+	return 1;
+}
+#endif
+
+/*****************************************************************************
+ * mos7840_open
+ *	this function is called by the tty driver when a port is opened
+ *	If successful, we return 0
+ *	Otherwise we return a negative error number.
+ *****************************************************************************/
+
+static int mos7840_open(struct usb_serial_port *port, struct file *filp)
+{
+	int response;
+	int j;
+	struct usb_serial *serial;
+	struct urb *urb;
+	__u16 Data;
+	int status;
+	struct moschip_port *mos7840_port;
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return -ENODEV;
+	}
+
+	mos7840_num_open_ports++;
+	serial = port->serial;
+
+	if (mos7840_serial_paranoia_check(serial, __FUNCTION__)) {
+		dbg("%s", "Serial Paranoia failed \n");
+		return -ENODEV;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL)
+		return -ENODEV;
+
+	usb_clear_halt(serial->dev, port->write_urb->pipe);
+	usb_clear_halt(serial->dev, port->read_urb->pipe);
+
+	/* Initialising the write urb pool */
+	for (j = 0; j < NUM_URBS; ++j) {
+		urb = usb_alloc_urb(0, SLAB_ATOMIC);
+		mos7840_port->write_urb_pool[j] = urb;
+
+		if (urb == NULL) {
+			err("No more urbs???");
+			continue;
+		}
+
+		urb->transfer_buffer = NULL;
+		urb->transfer_buffer =
+		    kmalloc(URB_TRANSFER_BUFFER_SIZE, GFP_KERNEL);
+		if (!urb->transfer_buffer) {
+			err("%s-out of memory for urb buffers.", __FUNCTION__);
+			continue;
+		}
+	}
+
+/*****************************************************************************
+ * Initialize MCS7840 -- Write Init values to corresponding Registers
+ *
+ * Register Index
+ * 1 : IER
+ * 2 : FCR
+ * 3 : LCR
+ * 4 : MCR
+ *
+ * 0x08 : SP1/2 Control Reg
+ *****************************************************************************/
+
+//NEED to check the following Block
+
+	status = 0;
+	Data = 0x0;
+	status = mos7840_get_reg_sync(port, mos7840_port->SpRegOffset, &Data);
+	if (status < 0) {
+		dbg("Reading Spreg failed\n");
+		return -1;
+	}
+	Data |= 0x80;
+	status = mos7840_set_reg_sync(port, mos7840_port->SpRegOffset, Data);
+	if (status < 0) {
+		dbg("writing Spreg failed\n");
+		return -1;
+	}
+
+	Data &= ~0x80;
+	status = mos7840_set_reg_sync(port, mos7840_port->SpRegOffset, Data);
+	if (status < 0) {
+		dbg("writing Spreg failed\n");
+		return -1;
+	}
+//End of block to be checked
+
+	status = 0;
+	Data = 0x0;
+	status =
+	    mos7840_get_reg_sync(port, mos7840_port->ControlRegOffset, &Data);
+	if (status < 0) {
+		dbg("Reading Controlreg failed\n");
+		return -1;
+	}
+	Data |= 0x08;		//Driver done bit
+	Data |= 0x20;		//rx_disable
+	status = 0;
+	status =
+	    mos7840_set_reg_sync(port, mos7840_port->ControlRegOffset, Data);
+	if (status < 0) {
+		dbg("writing Controlreg failed\n");
+		return -1;
+	}
+	//do register settings here
+	// Set all regs to the device default values.
+	////////////////////////////////////
+	// First Disable all interrupts.
+	////////////////////////////////////
+
+	Data = 0x00;
+	status = 0;
+	status = mos7840_set_uart_reg(port, INTERRUPT_ENABLE_REGISTER, Data);
+	if (status < 0) {
+		dbg("disableing interrupts failed\n");
+		return -1;
+	}
+	// Set FIFO_CONTROL_REGISTER to the default value
+	Data = 0x00;
+	status = 0;
+	status = mos7840_set_uart_reg(port, FIFO_CONTROL_REGISTER, Data);
+	if (status < 0) {
+		dbg("Writing FIFO_CONTROL_REGISTER  failed\n");
+		return -1;
+	}
+
+	Data = 0xcf;
+	status = 0;
+	status = mos7840_set_uart_reg(port, FIFO_CONTROL_REGISTER, Data);
+	if (status < 0) {
+		dbg("Writing FIFO_CONTROL_REGISTER  failed\n");
+		return -1;
+	}
+
+	Data = 0x03;
+	status = 0;
+	status = mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+	mos7840_port->shadowLCR = Data;
+
+	Data = 0x0b;
+	status = 0;
+	status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+	mos7840_port->shadowMCR = Data;
+
+	Data = 0x00;
+	status = 0;
+	status = mos7840_get_uart_reg(port, LINE_CONTROL_REGISTER, &Data);
+	mos7840_port->shadowLCR = Data;
+
+	Data |= SERIAL_LCR_DLAB;	//data latch enable in LCR 0x80
+	status = 0;
+	status = mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+
+	Data = 0x0c;
+	status = 0;
+	status = mos7840_set_uart_reg(port, DIVISOR_LATCH_LSB, Data);
+
+	Data = 0x0;
+	status = 0;
+	status = mos7840_set_uart_reg(port, DIVISOR_LATCH_MSB, Data);
+
+	Data = 0x00;
+	status = 0;
+	status = mos7840_get_uart_reg(port, LINE_CONTROL_REGISTER, &Data);
+
+	Data = Data & ~SERIAL_LCR_DLAB;
+	status = 0;
+	status = mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+	mos7840_port->shadowLCR = Data;
+
+	//clearing Bulkin and Bulkout Fifo
+	Data = 0x0;
+	status = 0;
+	status = mos7840_get_reg_sync(port, mos7840_port->SpRegOffset, &Data);
+
+	Data = Data | 0x0c;
+	status = 0;
+	status = mos7840_set_reg_sync(port, mos7840_port->SpRegOffset, Data);
+
+	Data = Data & ~0x0c;
+	status = 0;
+	status = mos7840_set_reg_sync(port, mos7840_port->SpRegOffset, Data);
+	//Finally enable all interrupts
+	Data = 0x0;
+	Data = 0x0c;
+	status = 0;
+	status = mos7840_set_uart_reg(port, INTERRUPT_ENABLE_REGISTER, Data);
+
+	//clearing rx_disable
+	Data = 0x0;
+	status = 0;
+	status =
+	    mos7840_get_reg_sync(port, mos7840_port->ControlRegOffset, &Data);
+	Data = Data & ~0x20;
+	status = 0;
+	status =
+	    mos7840_set_reg_sync(port, mos7840_port->ControlRegOffset, Data);
+
+	// rx_negate
+	Data = 0x0;
+	status = 0;
+	status =
+	    mos7840_get_reg_sync(port, mos7840_port->ControlRegOffset, &Data);
+	Data = Data | 0x10;
+	status = 0;
+	status =
+	    mos7840_set_reg_sync(port, mos7840_port->ControlRegOffset, Data);
+
+	/* force low_latency on so that our tty_push actually forces *
+	 * the data through,otherwise it is scheduled, and with      *
+	 * high data rates (like with OHCI) data can get lost.       */
+
+	if (port->tty)
+		port->tty->low_latency = 1;
+/* Check to see if we've set up our endpoint info yet    *
+     * (can't set it up in mos7840_startup as the structures *
+     * were not set up at that time.)                        */
+	if (mos7840_num_open_ports == 1) {
+		if (serial->port[0]->interrupt_in_buffer == NULL) {
+
+			/* set up interrupt urb */
+
+			usb_fill_int_urb(serial->port[0]->interrupt_in_urb,
+					 serial->dev,
+					 usb_rcvintpipe(serial->dev,
+							serial->port[0]->
+							interrupt_in_endpointAddress),
+					 serial->port[0]->interrupt_in_buffer,
+					 serial->port[0]->interrupt_in_urb->
+					 transfer_buffer_length,
+					 mos7840_interrupt_callback,
+					 serial,
+					 serial->port[0]->interrupt_in_urb->
+					 interval);
+
+			/* start interrupt read for mos7840               *
+			 * will continue as long as mos7840 is connected  */
+
+			response =
+			    usb_submit_urb(serial->port[0]->interrupt_in_urb,
+					   GFP_KERNEL);
+			if (response) {
+				err("%s - Error %d submitting interrupt urb",
+				    __FUNCTION__, response);
+			}
+
+		}
+
+	}
+
+	/* see if we've set up our endpoint info yet   *
+	 * (can't set it up in mos7840_startup as the  *
+	 * structures were not set up at that time.)   */
+
+	dbg("port number is %d \n", port->number);
+	dbg("serial number is %d \n", port->serial->minor);
+	dbg("Bulkin endpoint is %d \n", port->bulk_in_endpointAddress);
+	dbg("BulkOut endpoint is %d \n", port->bulk_out_endpointAddress);
+	dbg("Interrupt endpoint is %d \n", port->interrupt_in_endpointAddress);
+	dbg("port's number in the device is %d\n", mos7840_port->port_num);
+	mos7840_port->read_urb = port->read_urb;
+
+	/* set up our bulk in urb */
+
+	usb_fill_bulk_urb(mos7840_port->read_urb,
+			  serial->dev,
+			  usb_rcvbulkpipe(serial->dev,
+					  port->bulk_in_endpointAddress),
+			  port->bulk_in_buffer,
+			  mos7840_port->read_urb->transfer_buffer_length,
+			  mos7840_bulk_in_callback, mos7840_port);
+
+	dbg("mos7840_open: bulkin endpoint is %d\n",
+	    port->bulk_in_endpointAddress);
+	response = usb_submit_urb(mos7840_port->read_urb, GFP_KERNEL);
+	if (response) {
+		err("%s - Error %d submitting control urb", __FUNCTION__,
+		    response);
+	}
+
+	/* initialize our wait queues */
+	init_waitqueue_head(&mos7840_port->wait_chase);
+	init_waitqueue_head(&mos7840_port->delta_msr_wait);
+
+	/* initialize our icount structure */
+	memset(&(mos7840_port->icount), 0x00, sizeof(mos7840_port->icount));
+
+	/* initialize our port settings */
+	mos7840_port->shadowMCR = MCR_MASTER_IE;	/* Must set to enable ints! */
+	/* send a open port command */
+	mos7840_port->open = 1;
+	//mos7840_change_port_settings(mos7840_port,old_termios);
+	mos7840_port->icount.tx = 0;
+	mos7840_port->icount.rx = 0;
+
+	dbg("\n\nusb_serial serial:%x       mos7840_port:%x\n      usb_serial_port port:%x\n\n", (unsigned int)serial, (unsigned int)mos7840_port, (unsigned int)port);
+
+	return 0;
+
+}
+
+/*****************************************************************************
+ * mos7840_chars_in_buffer
+ *	this function is called by the tty driver when it wants to know how many
+ *	bytes of data we currently have outstanding in the port (data that has
+ *	been written, but hasn't made it out the port yet)
+ *	If successful, we return the number of bytes left to be written in the
+ *	system,
+ *	Otherwise we return a negative error number.
+ *****************************************************************************/
+
+static int mos7840_chars_in_buffer(struct usb_serial_port *port)
+{
+	int i;
+	int chars = 0;
+	struct moschip_port *mos7840_port;
+
+	dbg("%s \n", " mos7840_chars_in_buffer:entering ...........");
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return -1;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+	if (mos7840_port == NULL) {
+		dbg("%s \n", "mos7840_break:leaving ...........");
+		return -1;
+	}
+
+	for (i = 0; i < NUM_URBS; ++i) {
+		if (mos7840_port->write_urb_pool[i]->status == -EINPROGRESS) {
+			chars += URB_TRANSFER_BUFFER_SIZE;
+		}
+	}
+	dbg("%s - returns %d", __FUNCTION__, chars);
+	return (chars);
+
+}
+
+/************************************************************************
+ *
+ * mos7840_block_until_tx_empty
+ *
+ *	This function will block the close until one of the following:
+ *		1. TX count are 0
+ *		2. The mos7840 has stopped
+ *		3. A timout of 3 seconds without activity has expired
+ *
+ ************************************************************************/
+static void mos7840_block_until_tx_empty(struct moschip_port *mos7840_port)
+{
+	int timeout = HZ / 10;
+	int wait = 30;
+	int count;
+
+	while (1) {
+
+		count = mos7840_chars_in_buffer(mos7840_port->port);
+
+		/* Check for Buffer status */
+		if (count <= 0) {
+			return;
+		}
+
+		/* Block the thread for a while */
+		interruptible_sleep_on_timeout(&mos7840_port->wait_chase,
+					       timeout);
+
+		/* No activity.. count down section */
+		wait--;
+		if (wait == 0) {
+			dbg("%s - TIMEOUT", __FUNCTION__);
+			return;
+		} else {
+			/* Reset timout value back to seconds */
+			wait = 30;
+		}
+	}
+}
+
+/*****************************************************************************
+ * mos7840_close
+ *	this function is called by the tty driver when a port is closed
+ *****************************************************************************/
+
+static void mos7840_close(struct usb_serial_port *port, struct file *filp)
+{
+	struct usb_serial *serial;
+	struct moschip_port *mos7840_port;
+	int j;
+	__u16 Data;
+
+	dbg("%s\n", "mos7840_close:entering...");
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return;
+	}
+
+	serial = mos7840_get_usb_serial(port, __FUNCTION__);
+	if (!serial) {
+		dbg("%s", "Serial Paranoia failed \n");
+		return;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL) {
+		return;
+	}
+
+	for (j = 0; j < NUM_URBS; ++j)
+		usb_kill_urb(mos7840_port->write_urb_pool[j]);
+
+	/* Freeing Write URBs */
+	for (j = 0; j < NUM_URBS; ++j) {
+		if (mos7840_port->write_urb_pool[j]) {
+			if (mos7840_port->write_urb_pool[j]->transfer_buffer)
+				kfree(mos7840_port->write_urb_pool[j]->
+				      transfer_buffer);
+
+			usb_free_urb(mos7840_port->write_urb_pool[j]);
+		}
+	}
+
+	if (serial->dev) {
+		/* flush and block until tx is empty */
+		mos7840_block_until_tx_empty(mos7840_port);
+	}
+
+	/* While closing port, shutdown all bulk read, write  *
+	 * and interrupt read if they exists                  */
+	if (serial->dev) {
+
+		if (mos7840_port->write_urb) {
+			dbg("%s", "Shutdown bulk write\n");
+			usb_kill_urb(mos7840_port->write_urb);
+		}
+
+		if (mos7840_port->read_urb) {
+			dbg("%s", "Shutdown bulk read\n");
+			usb_kill_urb(mos7840_port->read_urb);
+		}
+		if ((&mos7840_port->control_urb)) {
+			dbg("%s", "Shutdown control read\n");
+			//      usb_kill_urb (mos7840_port->control_urb);
+
+		}
+	}
+//              if(mos7840_port->ctrl_buf != NULL)
+//                      kfree(mos7840_port->ctrl_buf);
+	mos7840_num_open_ports--;
+	dbg("mos7840_num_open_ports in close%d:in port%d\n",
+	    mos7840_num_open_ports, port->number);
+	if (mos7840_num_open_ports == 0) {
+		if (serial->port[0]->interrupt_in_urb) {
+			dbg("%s", "Shutdown interrupt_in_urb\n");
+		}
+	}
+
+	if (mos7840_port->write_urb) {
+		/* if this urb had a transfer buffer already (old tx) free it */
+
+		if (mos7840_port->write_urb->transfer_buffer != NULL) {
+			kfree(mos7840_port->write_urb->transfer_buffer);
+		}
+		usb_free_urb(mos7840_port->write_urb);
+	}
+
+	Data = 0x0;
+	mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+
+	Data = 0x00;
+	mos7840_set_uart_reg(port, INTERRUPT_ENABLE_REGISTER, Data);
+
+	mos7840_port->open = 0;
+
+	dbg("%s \n", "Leaving ............");
+}
+
+/************************************************************************
+ *
+ * mos7840_block_until_chase_response
+ *
+ *	This function will block the close until one of the following:
+ *		1. Response to our Chase comes from mos7840
+ *		2. A timout of 10 seconds without activity has expired
+ *		   (1K of mos7840 data @ 2400 baud ==> 4 sec to empty)
+ *
+ ************************************************************************/
+
+static void mos7840_block_until_chase_response(struct moschip_port
+					       *mos7840_port)
+{
+	int timeout = 1 * HZ;
+	int wait = 10;
+	int count;
+
+	while (1) {
+		count = mos7840_chars_in_buffer(mos7840_port->port);
+
+		/* Check for Buffer status */
+		if (count <= 0) {
+			return;
+		}
+
+		/* Block the thread for a while */
+		interruptible_sleep_on_timeout(&mos7840_port->wait_chase,
+					       timeout);
+		/* No activity.. count down section */
+		wait--;
+		if (wait == 0) {
+			dbg("%s - TIMEOUT", __FUNCTION__);
+			return;
+		} else {
+			/* Reset timout value back to seconds */
+			wait = 10;
+		}
+	}
+
+}
+
+/*****************************************************************************
+ * mos7840_break
+ *	this function sends a break to the port
+ *****************************************************************************/
+static void mos7840_break(struct usb_serial_port *port, int break_state)
+{
+	unsigned char data;
+	struct usb_serial *serial;
+	struct moschip_port *mos7840_port;
+
+	dbg("%s \n", "Entering ...........");
+	dbg("mos7840_break: Start\n");
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return;
+	}
+
+	serial = mos7840_get_usb_serial(port, __FUNCTION__);
+	if (!serial) {
+		dbg("%s", "Serial Paranoia failed \n");
+		return;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL) {
+		return;
+	}
+
+	if (serial->dev) {
+
+		/* flush and block until tx is empty */
+		mos7840_block_until_chase_response(mos7840_port);
+	}
+
+	if (break_state == -1) {
+		data = mos7840_port->shadowLCR | LCR_SET_BREAK;
+	} else {
+		data = mos7840_port->shadowLCR & ~LCR_SET_BREAK;
+	}
+
+	mos7840_port->shadowLCR = data;
+	dbg("mcs7840_break mos7840_port->shadowLCR is %x\n",
+	    mos7840_port->shadowLCR);
+	mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER,
+			     mos7840_port->shadowLCR);
+
+	return;
+}
+
+/*****************************************************************************
+ * mos7840_write_room
+ *	this function is called by the tty driver when it wants to know how many
+ *	bytes of data we can accept for a specific port.
+ *	If successful, we return the amount of room that we have for this port
+ *	Otherwise we return a negative error number.
+ *****************************************************************************/
+
+static int mos7840_write_room(struct usb_serial_port *port)
+{
+	int i;
+	int room = 0;
+	struct moschip_port *mos7840_port;
+
+	dbg("%s \n", " mos7840_write_room:entering ...........");
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		dbg("%s \n", " mos7840_write_room:leaving ...........");
+		return -1;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+	if (mos7840_port == NULL) {
+		dbg("%s \n", "mos7840_break:leaving ...........");
+		return -1;
+	}
+
+	for (i = 0; i < NUM_URBS; ++i) {
+		if (mos7840_port->write_urb_pool[i]->status != -EINPROGRESS) {
+			room += URB_TRANSFER_BUFFER_SIZE;
+		}
+	}
+
+	dbg("%s - returns %d", __FUNCTION__, room);
+	return (room);
+
+}
+
+/*****************************************************************************
+ * mos7840_write
+ *	this function is called by the tty driver when data should be written to
+ *	the port.
+ *	If successful, we return the number of bytes written, otherwise we
+ *      return a negative error number.
+ *****************************************************************************/
+
+static int mos7840_write(struct usb_serial_port *port,
+			 const unsigned char *data, int count)
+{
+	int status;
+	int i;
+	int bytes_sent = 0;
+	int transfer_size;
+	int from_user = 0;
+
+	struct moschip_port *mos7840_port;
+	struct usb_serial *serial;
+	struct urb *urb;
+	//__u16 Data;
+	const unsigned char *current_position = data;
+	unsigned char *data1;
+	dbg("%s \n", "entering ...........");
+	//dbg("mos7840_write: mos7840_port->shadowLCR is %x\n",mos7840_port->shadowLCR);
+
+#ifdef NOTMOS7840
+	Data = 0x00;
+	status = 0;
+	status = mos7840_get_uart_reg(port, LINE_CONTROL_REGISTER, &Data);
+	mos7840_port->shadowLCR = Data;
+	dbg("mos7840_write: LINE_CONTROL_REGISTER is %x\n", Data);
+	dbg("mos7840_write: mos7840_port->shadowLCR is %x\n",
+	    mos7840_port->shadowLCR);
+
+	//Data = 0x03;
+	//status = mos7840_set_uart_reg(port,LINE_CONTROL_REGISTER,Data);
+	//mos7840_port->shadowLCR=Data;//Need to add later
+
+	Data |= SERIAL_LCR_DLAB;	//data latch enable in LCR 0x80
+	status = 0;
+	status = mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+
+	//Data = 0x0c;
+	//status = mos7840_set_uart_reg(port,DIVISOR_LATCH_LSB,Data);
+	Data = 0x00;
+	status = 0;
+	status = mos7840_get_uart_reg(port, DIVISOR_LATCH_LSB, &Data);
+	dbg("mos7840_write:DLL value is %x\n", Data);
+
+	Data = 0x0;
+	status = 0;
+	status = mos7840_get_uart_reg(port, DIVISOR_LATCH_MSB, &Data);
+	dbg("mos7840_write:DLM value is %x\n", Data);
+
+	Data = Data & ~SERIAL_LCR_DLAB;
+	dbg("mos7840_write: mos7840_port->shadowLCR is %x\n",
+	    mos7840_port->shadowLCR);
+	status = 0;
+	status = mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+#endif
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Port Paranoia failed \n");
+		return -1;
+	}
+
+	serial = port->serial;
+	if (mos7840_serial_paranoia_check(serial, __FUNCTION__)) {
+		dbg("%s", "Serial Paranoia failed \n");
+		return -1;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+	if (mos7840_port == NULL) {
+		dbg("%s", "mos7840_port is NULL\n");
+		return -1;
+	}
+
+	/* try to find a free urb in the list */
+	urb = NULL;
+
+	for (i = 0; i < NUM_URBS; ++i) {
+		if (mos7840_port->write_urb_pool[i]->status != -EINPROGRESS) {
+			urb = mos7840_port->write_urb_pool[i];
+			dbg("\nURB:%d", i);
+			break;
+		}
+	}
+
+	if (urb == NULL) {
+		dbg("%s - no more free urbs", __FUNCTION__);
+		goto exit;
+	}
+
+	if (urb->transfer_buffer == NULL) {
+		urb->transfer_buffer =
+		    kmalloc(URB_TRANSFER_BUFFER_SIZE, GFP_KERNEL);
+
+		if (urb->transfer_buffer == NULL) {
+			err("%s no more kernel memory...", __FUNCTION__);
+			goto exit;
+		}
+	}
+	transfer_size = min(count, URB_TRANSFER_BUFFER_SIZE);
+
+	if (from_user) {
+		if (copy_from_user
+		    (urb->transfer_buffer, current_position, transfer_size)) {
+			bytes_sent = -EFAULT;
+			goto exit;
+		}
+	} else {
+		memcpy(urb->transfer_buffer, current_position, transfer_size);
+	}
+
+	/* fill urb with data and submit  */
+	usb_fill_bulk_urb(urb,
+			  serial->dev,
+			  usb_sndbulkpipe(serial->dev,
+					  port->bulk_out_endpointAddress),
+			  urb->transfer_buffer,
+			  transfer_size,
+			  mos7840_bulk_out_data_callback, mos7840_port);
+
+	data1 = urb->transfer_buffer;
+	dbg("\nbulkout endpoint is %d", port->bulk_out_endpointAddress);
+
+	/* send it down the pipe */
+	status = usb_submit_urb(urb, GFP_ATOMIC);
+
+	if (status) {
+		err("%s - usb_submit_urb(write bulk) failed with status = %d",
+		    __FUNCTION__, status);
+		bytes_sent = status;
+		goto exit;
+	}
+	bytes_sent = transfer_size;
+	mos7840_port->icount.tx += transfer_size;
+	dbg("mos7840_port->icount.tx is %d:\n", mos7840_port->icount.tx);
+      exit:
+
+	return bytes_sent;
+
+}
+
+/*****************************************************************************
+ * mos7840_throttle
+ *	this function is called by the tty driver when it wants to stop the data
+ *	being read from the port.
+ *****************************************************************************/
+
+static void mos7840_throttle(struct usb_serial_port *port)
+{
+	struct moschip_port *mos7840_port;
+	struct tty_struct *tty;
+	int status;
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return;
+	}
+
+	dbg("- port %d\n", port->number);
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL)
+		return;
+
+	if (!mos7840_port->open) {
+		dbg("%s\n", "port not opened");
+		return;
+	}
+
+	dbg("%s", "Entering .......... \n");
+
+	tty = port->tty;
+	if (!tty) {
+		dbg("%s - no tty available", __FUNCTION__);
+		return;
+	}
+
+	/* if we are implementing XON/XOFF, send the stop character */
+	if (I_IXOFF(tty)) {
+		unsigned char stop_char = STOP_CHAR(tty);
+		status = mos7840_write(port, &stop_char, 1);
+		if (status <= 0) {
+			return;
+		}
+	}
+
+	/* if we are implementing RTS/CTS, toggle that line */
+	if (tty->termios->c_cflag & CRTSCTS) {
+		mos7840_port->shadowMCR &= ~MCR_RTS;
+		status = 0;
+		status =
+		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
+					 mos7840_port->shadowMCR);
+
+		if (status < 0) {
+			return;
+		}
+	}
+
+	return;
+}
+
+/*****************************************************************************
+ * mos7840_unthrottle
+ *	this function is called by the tty driver when it wants to resume the data
+ *	being read from the port (called after SerialThrottle is called)
+ *****************************************************************************/
+static void mos7840_unthrottle(struct usb_serial_port *port)
+{
+	struct tty_struct *tty;
+	int status;
+	struct moschip_port *mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return;
+	}
+
+	if (mos7840_port == NULL)
+		return;
+
+	if (!mos7840_port->open) {
+		dbg("%s - port not opened", __FUNCTION__);
+		return;
+	}
+
+	dbg("%s", "Entering .......... \n");
+
+	tty = port->tty;
+	if (!tty) {
+		dbg("%s - no tty available", __FUNCTION__);
+		return;
+	}
+
+	/* if we are implementing XON/XOFF, send the start character */
+	if (I_IXOFF(tty)) {
+		unsigned char start_char = START_CHAR(tty);
+		status = mos7840_write(port, &start_char, 1);
+		if (status <= 0) {
+			return;
+		}
+	}
+
+	/* if we are implementing RTS/CTS, toggle that line */
+	if (tty->termios->c_cflag & CRTSCTS) {
+		mos7840_port->shadowMCR |= MCR_RTS;
+		status = 0;
+		status =
+		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER,
+					 mos7840_port->shadowMCR);
+		if (status < 0) {
+			return;
+		}
+	}
+
+	return;
+}
+
+static int mos7840_tiocmget(struct usb_serial_port *port, struct file *file)
+{
+	struct moschip_port *mos7840_port;
+	unsigned int result;
+	__u16 msr;
+	__u16 mcr;
+	int status = 0;
+	mos7840_port = mos7840_get_port_private(port);
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	if (mos7840_port == NULL)
+		return -ENODEV;
+
+	status = mos7840_get_uart_reg(port, MODEM_STATUS_REGISTER, &msr);
+	status = mos7840_get_uart_reg(port, MODEM_CONTROL_REGISTER, &mcr);
+	result = ((mcr & MCR_DTR) ? TIOCM_DTR : 0)
+	    | ((mcr & MCR_RTS) ? TIOCM_RTS : 0)
+	    | ((mcr & MCR_LOOPBACK) ? TIOCM_LOOP : 0)
+	    | ((msr & MOS7840_MSR_CTS) ? TIOCM_CTS : 0)
+	    | ((msr & MOS7840_MSR_CD) ? TIOCM_CAR : 0)
+	    | ((msr & MOS7840_MSR_RI) ? TIOCM_RI : 0)
+	    | ((msr & MOS7840_MSR_DSR) ? TIOCM_DSR : 0);
+
+	dbg("%s - 0x%04X", __FUNCTION__, result);
+
+	return result;
+}
+
+static int mos7840_tiocmset(struct usb_serial_port *port, struct file *file,
+			    unsigned int set, unsigned int clear)
+{
+	struct moschip_port *mos7840_port;
+	unsigned int mcr;
+	unsigned int status;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL)
+		return -ENODEV;
+
+	mcr = mos7840_port->shadowMCR;
+	if (clear & TIOCM_RTS)
+		mcr &= ~MCR_RTS;
+	if (clear & TIOCM_DTR)
+		mcr &= ~MCR_DTR;
+	if (clear & TIOCM_LOOP)
+		mcr &= ~MCR_LOOPBACK;
+
+	if (set & TIOCM_RTS)
+		mcr |= MCR_RTS;
+	if (set & TIOCM_DTR)
+		mcr |= MCR_DTR;
+	if (set & TIOCM_LOOP)
+		mcr |= MCR_LOOPBACK;
+
+	mos7840_port->shadowMCR = mcr;
+
+	status = 0;
+	status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, mcr);
+	if (status < 0) {
+		dbg("setting MODEM_CONTROL_REGISTER Failed\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+/*****************************************************************************
+ * mos7840_calc_baud_rate_divisor
+ *	this function calculates the proper baud rate divisor for the specified
+ *	baud rate.
+ *****************************************************************************/
+static int mos7840_calc_baud_rate_divisor(int baudRate, int *divisor,
+					  __u16 * clk_sel_val)
+{
+
+	dbg("%s - %d", __FUNCTION__, baudRate);
+
+	if (baudRate <= 115200) {
+		*divisor = 115200 / baudRate;
+		*clk_sel_val = 0x0;
+	}
+	if ((baudRate > 115200) && (baudRate <= 230400)) {
+		*divisor = 230400 / baudRate;
+		*clk_sel_val = 0x10;
+	} else if ((baudRate > 230400) && (baudRate <= 403200)) {
+		*divisor = 403200 / baudRate;
+		*clk_sel_val = 0x20;
+	} else if ((baudRate > 403200) && (baudRate <= 460800)) {
+		*divisor = 460800 / baudRate;
+		*clk_sel_val = 0x30;
+	} else if ((baudRate > 460800) && (baudRate <= 806400)) {
+		*divisor = 806400 / baudRate;
+		*clk_sel_val = 0x40;
+	} else if ((baudRate > 806400) && (baudRate <= 921600)) {
+		*divisor = 921600 / baudRate;
+		*clk_sel_val = 0x50;
+	} else if ((baudRate > 921600) && (baudRate <= 1572864)) {
+		*divisor = 1572864 / baudRate;
+		*clk_sel_val = 0x60;
+	} else if ((baudRate > 1572864) && (baudRate <= 3145728)) {
+		*divisor = 3145728 / baudRate;
+		*clk_sel_val = 0x70;
+	}
+	return 0;
+
+#ifdef NOTMCS7840
+
+	for (i = 0; i < ARRAY_SIZE(mos7840_divisor_table); i++) {
+		if (mos7840_divisor_table[i].BaudRate == baudrate) {
+			*divisor = mos7840_divisor_table[i].Divisor;
+			return 0;
+		}
+	}
+
+	/* After trying for all the standard baud rates    *
+	 * Try calculating the divisor for this baud rate  */
+
+	if (baudrate > 75 && baudrate < 230400) {
+		/* get the divisor */
+		custom = (__u16) (230400L / baudrate);
+
+		/* Check for round off */
+		round1 = (__u16) (2304000L / baudrate);
+		round = (__u16) (round1 - (custom * 10));
+		if (round > 4) {
+			custom++;
+		}
+		*divisor = custom;
+
+		dbg(" Baud %d = %d\n", baudrate, custom);
+		return 0;
+	}
+
+	dbg("%s\n", " Baud calculation Failed...");
+	return -1;
+#endif
+}
+
+/*****************************************************************************
+ * mos7840_send_cmd_write_baud_rate
+ *	this function sends the proper command to change the baud rate of the
+ *	specified port.
+ *****************************************************************************/
+
+static int mos7840_send_cmd_write_baud_rate(struct moschip_port *mos7840_port,
+					    int baudRate)
+{
+	int divisor = 0;
+	int status;
+	__u16 Data;
+	unsigned char number;
+	__u16 clk_sel_val;
+	struct usb_serial_port *port;
+
+	if (mos7840_port == NULL)
+		return -1;
+
+	port = (struct usb_serial_port *)mos7840_port->port;
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return -1;
+	}
+
+	if (mos7840_serial_paranoia_check(port->serial, __FUNCTION__)) {
+		dbg("%s", "Invalid Serial \n");
+		return -1;
+	}
+
+	dbg("%s", "Entering .......... \n");
+
+	number = mos7840_port->port->number - mos7840_port->port->serial->minor;
+
+	dbg("%s - port = %d, baud = %d", __FUNCTION__,
+	    mos7840_port->port->number, baudRate);
+	//reset clk_uart_sel in spregOffset
+	if (baudRate > 115200) {
+#ifdef HW_flow_control
+		//NOTE: need to see the pther register to modify
+		//setting h/w flow control bit to 1;
+		status = 0;
+		Data = 0x2b;
+		mos7840_port->shadowMCR = Data;
+		status =
+		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+		if (status < 0) {
+			dbg("Writing spreg failed in set_serial_baud\n");
+			return -1;
+		}
+#endif
+
+	} else {
+#ifdef HW_flow_control
+		//setting h/w flow control bit to 0;
+		status = 0;
+		Data = 0xb;
+		mos7840_port->shadowMCR = Data;
+		status =
+		    mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+		if (status < 0) {
+			dbg("Writing spreg failed in set_serial_baud\n");
+			return -1;
+		}
+#endif
+
+	}
+
+	if (1)			//baudRate <= 115200)
+	{
+		clk_sel_val = 0x0;
+		Data = 0x0;
+		status = 0;
+		status =
+		    mos7840_calc_baud_rate_divisor(baudRate, &divisor,
+						   &clk_sel_val);
+		status =
+		    mos7840_get_reg_sync(port, mos7840_port->SpRegOffset,
+					 &Data);
+		if (status < 0) {
+			dbg("reading spreg failed in set_serial_baud\n");
+			return -1;
+		}
+		Data = (Data & 0x8f) | clk_sel_val;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(port, mos7840_port->SpRegOffset, Data);
+		if (status < 0) {
+			dbg("Writing spreg failed in set_serial_baud\n");
+			return -1;
+		}
+		/* Calculate the Divisor */
+
+		if (status) {
+			err("%s - bad baud rate", __FUNCTION__);
+			dbg("%s\n", "bad baud rate");
+			return status;
+		}
+		/* Enable access to divisor latch */
+		Data = mos7840_port->shadowLCR | SERIAL_LCR_DLAB;
+		mos7840_port->shadowLCR = Data;
+		mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+
+		/* Write the divisor */
+		Data = (unsigned char)(divisor & 0xff);
+		dbg("set_serial_baud Value to write DLL is %x\n", Data);
+		mos7840_set_uart_reg(port, DIVISOR_LATCH_LSB, Data);
+
+		Data = (unsigned char)((divisor & 0xff00) >> 8);
+		dbg("set_serial_baud Value to write DLM is %x\n", Data);
+		mos7840_set_uart_reg(port, DIVISOR_LATCH_MSB, Data);
+
+		/* Disable access to divisor latch */
+		Data = mos7840_port->shadowLCR & ~SERIAL_LCR_DLAB;
+		mos7840_port->shadowLCR = Data;
+		mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+
+	}
+
+	return status;
+}
+
+/*****************************************************************************
+ * mos7840_change_port_settings
+ *	This routine is called to set the UART on the device to match
+ *      the specified new settings.
+ *****************************************************************************/
+
+static void mos7840_change_port_settings(struct moschip_port *mos7840_port,
+					 struct termios *old_termios)
+{
+	struct tty_struct *tty;
+	int baud;
+	unsigned cflag;
+	unsigned iflag;
+	__u8 lData;
+	__u8 lParity;
+	__u8 lStop;
+	int status;
+	__u16 Data;
+	struct usb_serial_port *port;
+	struct usb_serial *serial;
+
+	if (mos7840_port == NULL)
+		return;
+
+	port = (struct usb_serial_port *)mos7840_port->port;
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return;
+	}
+
+	if (mos7840_serial_paranoia_check(port->serial, __FUNCTION__)) {
+		dbg("%s", "Invalid Serial \n");
+		return;
+	}
+
+	serial = port->serial;
+
+	dbg("%s - port %d", __FUNCTION__, mos7840_port->port->number);
+
+	if (!mos7840_port->open) {
+		dbg("%s - port not opened", __FUNCTION__);
+		return;
+	}
+
+	tty = mos7840_port->port->tty;
+
+	if ((!tty) || (!tty->termios)) {
+		dbg("%s - no tty structures", __FUNCTION__);
+		return;
+	}
+
+	dbg("%s", "Entering .......... \n");
+
+	lData = LCR_BITS_8;
+	lStop = LCR_STOP_1;
+	lParity = LCR_PAR_NONE;
+
+	cflag = tty->termios->c_cflag;
+	iflag = tty->termios->c_iflag;
+
+	/* Change the number of bits */
+	if (cflag & CSIZE) {
+		switch (cflag & CSIZE) {
+		case CS5:
+			lData = LCR_BITS_5;
+			break;
+
+		case CS6:
+			lData = LCR_BITS_6;
+			break;
+
+		case CS7:
+			lData = LCR_BITS_7;
+			break;
+		default:
+		case CS8:
+			lData = LCR_BITS_8;
+			break;
+		}
+	}
+	/* Change the Parity bit */
+	if (cflag & PARENB) {
+		if (cflag & PARODD) {
+			lParity = LCR_PAR_ODD;
+			dbg("%s - parity = odd", __FUNCTION__);
+		} else {
+			lParity = LCR_PAR_EVEN;
+			dbg("%s - parity = even", __FUNCTION__);
+		}
+
+	} else {
+		dbg("%s - parity = none", __FUNCTION__);
+	}
+
+	if (cflag & CMSPAR) {
+		lParity = lParity | 0x20;
+	}
+
+	/* Change the Stop bit */
+	if (cflag & CSTOPB) {
+		lStop = LCR_STOP_2;
+		dbg("%s - stop bits = 2", __FUNCTION__);
+	} else {
+		lStop = LCR_STOP_1;
+		dbg("%s - stop bits = 1", __FUNCTION__);
+	}
+
+	/* Update the LCR with the correct value */
+	mos7840_port->shadowLCR &=
+	    ~(LCR_BITS_MASK | LCR_STOP_MASK | LCR_PAR_MASK);
+	mos7840_port->shadowLCR |= (lData | lParity | lStop);
+
+	dbg("mos7840_change_port_settings mos7840_port->shadowLCR is %x\n",
+	    mos7840_port->shadowLCR);
+	/* Disable Interrupts */
+	Data = 0x00;
+	mos7840_set_uart_reg(port, INTERRUPT_ENABLE_REGISTER, Data);
+
+	Data = 0x00;
+	mos7840_set_uart_reg(port, FIFO_CONTROL_REGISTER, Data);
+
+	Data = 0xcf;
+	mos7840_set_uart_reg(port, FIFO_CONTROL_REGISTER, Data);
+
+	/* Send the updated LCR value to the mos7840 */
+	Data = mos7840_port->shadowLCR;
+
+	mos7840_set_uart_reg(port, LINE_CONTROL_REGISTER, Data);
+
+	Data = 0x00b;
+	mos7840_port->shadowMCR = Data;
+	mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+	Data = 0x00b;
+	mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+
+	/* set up the MCR register and send it to the mos7840 */
+
+	mos7840_port->shadowMCR = MCR_MASTER_IE;
+	if (cflag & CBAUD) {
+		mos7840_port->shadowMCR |= (MCR_DTR | MCR_RTS);
+	}
+
+	if (cflag & CRTSCTS) {
+		mos7840_port->shadowMCR |= (MCR_XON_ANY);
+
+	} else {
+		mos7840_port->shadowMCR &= ~(MCR_XON_ANY);
+	}
+
+	Data = mos7840_port->shadowMCR;
+	mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+
+	/* Determine divisor based on baud rate */
+	baud = tty_get_baud_rate(tty);
+
+	if (!baud) {
+		/* pick a default, any default... */
+		dbg("%s\n", "Picked default baud...");
+		baud = 9600;
+	}
+
+	dbg("%s - baud rate = %d", __FUNCTION__, baud);
+	status = mos7840_send_cmd_write_baud_rate(mos7840_port, baud);
+
+	/* Enable Interrupts */
+	Data = 0x0c;
+	mos7840_set_uart_reg(port, INTERRUPT_ENABLE_REGISTER, Data);
+
+	if (mos7840_port->read_urb->status != -EINPROGRESS) {
+		mos7840_port->read_urb->dev = serial->dev;
+
+		status = usb_submit_urb(mos7840_port->read_urb, GFP_ATOMIC);
+
+		if (status) {
+			dbg(" usb_submit_urb(read bulk) failed, status = %d",
+			    status);
+		}
+	}
+	wake_up(&mos7840_port->delta_msr_wait);
+	mos7840_port->delta_msr_cond = 1;
+	dbg("mos7840_change_port_settings mos7840_port->shadowLCR is End %x\n",
+	    mos7840_port->shadowLCR);
+
+	return;
+}
+
+/*****************************************************************************
+ * mos7840_set_termios
+ *	this function is called by the tty driver when it wants to change
+ *	the termios structure
+ *****************************************************************************/
+
+static void mos7840_set_termios(struct usb_serial_port *port,
+				struct termios *old_termios)
+{
+	int status;
+	unsigned int cflag;
+	struct usb_serial *serial;
+	struct moschip_port *mos7840_port;
+	struct tty_struct *tty;
+	dbg("mos7840_set_termios: START\n");
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return;
+	}
+
+	serial = port->serial;
+
+	if (mos7840_serial_paranoia_check(serial, __FUNCTION__)) {
+		dbg("%s", "Invalid Serial \n");
+		return;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+
+	if (mos7840_port == NULL)
+		return;
+
+	tty = port->tty;
+
+	if (!port->tty || !port->tty->termios) {
+		dbg("%s - no tty or termios", __FUNCTION__);
+		return;
+	}
+
+	if (!mos7840_port->open) {
+		dbg("%s - port not opened", __FUNCTION__);
+		return;
+	}
+
+	dbg("%s\n", "setting termios - ");
+
+	cflag = tty->termios->c_cflag;
+
+	if (!cflag) {
+		dbg("%s %s\n", __FUNCTION__, "cflag is NULL");
+		return;
+	}
+
+	/* check that they really want us to change something */
+	if (old_termios) {
+		if ((cflag == old_termios->c_cflag) &&
+		    (RELEVANT_IFLAG(tty->termios->c_iflag) ==
+		     RELEVANT_IFLAG(old_termios->c_iflag))) {
+			dbg("%s\n", "Nothing to change");
+			return;
+		}
+	}
+
+	dbg("%s - clfag %08x iflag %08x", __FUNCTION__,
+	    tty->termios->c_cflag, RELEVANT_IFLAG(tty->termios->c_iflag));
+
+	if (old_termios) {
+		dbg("%s - old clfag %08x old iflag %08x", __FUNCTION__,
+		    old_termios->c_cflag, RELEVANT_IFLAG(old_termios->c_iflag));
+	}
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	/* change the port settings to the new ones specified */
+
+	mos7840_change_port_settings(mos7840_port, old_termios);
+
+	if (!mos7840_port->read_urb) {
+		dbg("%s", "URB KILLED !!!!!\n");
+		return;
+	}
+
+	if (mos7840_port->read_urb->status != -EINPROGRESS) {
+		mos7840_port->read_urb->dev = serial->dev;
+		status = usb_submit_urb(mos7840_port->read_urb, GFP_ATOMIC);
+		if (status) {
+			dbg(" usb_submit_urb(read bulk) failed, status = %d",
+			    status);
+		}
+	}
+	return;
+}
+
+/*****************************************************************************
+ * mos7840_get_lsr_info - get line status register info
+ *
+ * Purpose: Let user call ioctl() to get info when the UART physically
+ * 	    is emptied.  On bus types like RS485, the transmitter must
+ * 	    release the bus after transmitting. This must be done when
+ * 	    the transmit shift register is empty, not be done when the
+ * 	    transmit holding register is empty.  This functionality
+ * 	    allows an RS485 driver to be written in user space.
+ *****************************************************************************/
+
+static int mos7840_get_lsr_info(struct moschip_port *mos7840_port,
+				unsigned int *value)
+{
+	int count;
+	unsigned int result = 0;
+
+	count = mos7840_chars_in_buffer(mos7840_port->port);
+	if (count == 0) {
+		dbg("%s -- Empty", __FUNCTION__);
+		result = TIOCSER_TEMT;
+	}
+
+	if (copy_to_user(value, &result, sizeof(int)))
+		return -EFAULT;
+	return 0;
+}
+
+/*****************************************************************************
+ * mos7840_get_bytes_avail - get number of bytes available
+ *
+ * Purpose: Let user call ioctl to get the count of number of bytes available.
+ *****************************************************************************/
+
+static int mos7840_get_bytes_avail(struct moschip_port *mos7840_port,
+				   unsigned int *value)
+{
+	unsigned int result = 0;
+	struct tty_struct *tty = mos7840_port->port->tty;
+
+	if (!tty)
+		return -ENOIOCTLCMD;
+
+	result = tty->read_cnt;
+
+	dbg("%s(%d) = %d", __FUNCTION__, mos7840_port->port->number, result);
+	if (copy_to_user(value, &result, sizeof(int)))
+		return -EFAULT;
+
+	return -ENOIOCTLCMD;
+}
+
+/*****************************************************************************
+ * mos7840_set_modem_info
+ *      function to set modem info
+ *****************************************************************************/
+
+static int mos7840_set_modem_info(struct moschip_port *mos7840_port,
+				  unsigned int cmd, unsigned int *value)
+{
+	unsigned int mcr;
+	unsigned int arg;
+	__u16 Data;
+	int status;
+	struct usb_serial_port *port;
+
+	if (mos7840_port == NULL)
+		return -1;
+
+	port = (struct usb_serial_port *)mos7840_port->port;
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return -1;
+	}
+
+	mcr = mos7840_port->shadowMCR;
+
+	if (copy_from_user(&arg, value, sizeof(int)))
+		return -EFAULT;
+
+	switch (cmd) {
+	case TIOCMBIS:
+		if (arg & TIOCM_RTS)
+			mcr |= MCR_RTS;
+		if (arg & TIOCM_DTR)
+			mcr |= MCR_RTS;
+		if (arg & TIOCM_LOOP)
+			mcr |= MCR_LOOPBACK;
+		break;
+
+	case TIOCMBIC:
+		if (arg & TIOCM_RTS)
+			mcr &= ~MCR_RTS;
+		if (arg & TIOCM_DTR)
+			mcr &= ~MCR_RTS;
+		if (arg & TIOCM_LOOP)
+			mcr &= ~MCR_LOOPBACK;
+		break;
+
+	case TIOCMSET:
+		/* turn off the RTS and DTR and LOOPBACK
+		 * and then only turn on what was asked to */
+		mcr &= ~(MCR_RTS | MCR_DTR | MCR_LOOPBACK);
+		mcr |= ((arg & TIOCM_RTS) ? MCR_RTS : 0);
+		mcr |= ((arg & TIOCM_DTR) ? MCR_DTR : 0);
+		mcr |= ((arg & TIOCM_LOOP) ? MCR_LOOPBACK : 0);
+		break;
+	}
+
+	mos7840_port->shadowMCR = mcr;
+
+	Data = mos7840_port->shadowMCR;
+	status = 0;
+	status = mos7840_set_uart_reg(port, MODEM_CONTROL_REGISTER, Data);
+	if (status < 0) {
+		dbg("setting MODEM_CONTROL_REGISTER Failed\n");
+		return -1;
+	}
+
+	return 0;
+}
+
+/*****************************************************************************
+ * mos7840_get_modem_info
+ *      function to get modem info
+ *****************************************************************************/
+
+static int mos7840_get_modem_info(struct moschip_port *mos7840_port,
+				  unsigned int *value)
+{
+	unsigned int result = 0;
+	__u16 msr;
+	unsigned int mcr = mos7840_port->shadowMCR;
+	int status = 0;
+	status =
+	    mos7840_get_uart_reg(mos7840_port->port, MODEM_STATUS_REGISTER,
+				 &msr);
+	result = ((mcr & MCR_DTR) ? TIOCM_DTR : 0)	/* 0x002 */
+	    |((mcr & MCR_RTS) ? TIOCM_RTS : 0)	/* 0x004 */
+	    |((msr & MOS7840_MSR_CTS) ? TIOCM_CTS : 0)	/* 0x020 */
+	    |((msr & MOS7840_MSR_CD) ? TIOCM_CAR : 0)	/* 0x040 */
+	    |((msr & MOS7840_MSR_RI) ? TIOCM_RI : 0)	/* 0x080 */
+	    |((msr & MOS7840_MSR_DSR) ? TIOCM_DSR : 0);	/* 0x100 */
+
+	dbg("%s -- %x", __FUNCTION__, result);
+
+	if (copy_to_user(value, &result, sizeof(int)))
+		return -EFAULT;
+	return 0;
+}
+
+/*****************************************************************************
+ * mos7840_get_serial_info
+ *      function to get information about serial port
+ *****************************************************************************/
+
+static int mos7840_get_serial_info(struct moschip_port *mos7840_port,
+				   struct serial_struct *retinfo)
+{
+	struct serial_struct tmp;
+
+	if (mos7840_port == NULL)
+		return -1;
+
+	if (!retinfo)
+		return -EFAULT;
+
+	memset(&tmp, 0, sizeof(tmp));
+
+	tmp.type = PORT_16550A;
+	tmp.line = mos7840_port->port->serial->minor;
+	tmp.port = mos7840_port->port->number;
+	tmp.irq = 0;
+	tmp.flags = ASYNC_SKIP_TEST | ASYNC_AUTO_IRQ;
+	tmp.xmit_fifo_size = NUM_URBS * URB_TRANSFER_BUFFER_SIZE;
+	tmp.baud_base = 9600;
+	tmp.close_delay = 5 * HZ;
+	tmp.closing_wait = 30 * HZ;
+
+	if (copy_to_user(retinfo, &tmp, sizeof(*retinfo)))
+		return -EFAULT;
+	return 0;
+}
+
+/*****************************************************************************
+ * SerialIoctl
+ *	this function handles any ioctl calls to the driver
+ *****************************************************************************/
+
+static int mos7840_ioctl(struct usb_serial_port *port, struct file *file,
+			 unsigned int cmd, unsigned long arg)
+{
+	struct moschip_port *mos7840_port;
+	struct tty_struct *tty;
+
+	struct async_icount cnow;
+	struct async_icount cprev;
+	struct serial_icounter_struct icount;
+	int mosret = 0;
+	int retval;
+	struct tty_ldisc *ld;
+
+	if (mos7840_port_paranoia_check(port, __FUNCTION__)) {
+		dbg("%s", "Invalid port \n");
+		return -1;
+	}
+
+	mos7840_port = mos7840_get_port_private(port);
+	tty = mos7840_port->port->tty;
+
+	if (mos7840_port == NULL)
+		return -1;
+
+	dbg("%s - port %d, cmd = 0x%x", __FUNCTION__, port->number, cmd);
+
+	switch (cmd) {
+		/* return number of bytes available */
+
+	case TIOCINQ:
+		dbg("%s (%d) TIOCINQ", __FUNCTION__, port->number);
+		return mos7840_get_bytes_avail(mos7840_port,
+					       (unsigned int *)arg);
+		break;
+
+	case TIOCOUTQ:
+		dbg("%s (%d) TIOCOUTQ", __FUNCTION__, port->number);
+		return put_user(tty->driver->chars_in_buffer ?
+				tty->driver->chars_in_buffer(tty) : 0,
+				(int __user *)arg);
+		break;
+
+	case TCFLSH:
+		retval = tty_check_change(tty);
+		if (retval)
+			return retval;
+
+		ld = tty_ldisc_ref(tty);
+		switch (arg) {
+		case TCIFLUSH:
+			if (ld && ld->flush_buffer)
+				ld->flush_buffer(tty);
+			break;
+		case TCIOFLUSH:
+			if (ld && ld->flush_buffer)
+				ld->flush_buffer(tty);
+			/* fall through */
+		case TCOFLUSH:
+			if (tty->driver->flush_buffer)
+				tty->driver->flush_buffer(tty);
+			break;
+		default:
+			tty_ldisc_deref(ld);
+			return -EINVAL;
+		}
+		tty_ldisc_deref(ld);
+		return 0;
+
+	case TCGETS:
+		if (kernel_termios_to_user_termios
+		    ((struct termios __user *)arg, tty->termios))
+			return -EFAULT;
+		return 0;
+
+	case TIOCSERGETLSR:
+		dbg("%s (%d) TIOCSERGETLSR", __FUNCTION__, port->number);
+		return mos7840_get_lsr_info(mos7840_port, (unsigned int *)arg);
+		return 0;
+
+	case TIOCMBIS:
+	case TIOCMBIC:
+	case TIOCMSET:
+		dbg("%s (%d) TIOCMSET/TIOCMBIC/TIOCMSET", __FUNCTION__,
+		    port->number);
+		mosret =
+		    mos7840_set_modem_info(mos7840_port, cmd,
+					   (unsigned int *)arg);
+		return mosret;
+
+	case TIOCMGET:
+		dbg("%s (%d) TIOCMGET", __FUNCTION__, port->number);
+		return mos7840_get_modem_info(mos7840_port,
+					      (unsigned int *)arg);
+
+	case TIOCGSERIAL:
+		dbg("%s (%d) TIOCGSERIAL", __FUNCTION__, port->number);
+		return mos7840_get_serial_info(mos7840_port,
+					       (struct serial_struct *)arg);
+
+	case TIOCSSERIAL:
+		dbg("%s (%d) TIOCSSERIAL", __FUNCTION__, port->number);
+		break;
+
+	case TIOCMIWAIT:
+		dbg("%s (%d) TIOCMIWAIT", __FUNCTION__, port->number);
+		cprev = mos7840_port->icount;
+		while (1) {
+			//interruptible_sleep_on(&mos7840_port->delta_msr_wait);
+			mos7840_port->delta_msr_cond = 0;
+			wait_event_interruptible(mos7840_port->delta_msr_wait,
+						 (mos7840_port->
+						  delta_msr_cond == 1));
+
+			/* see if a signal did it */
+			if (signal_pending(current))
+				return -ERESTARTSYS;
+			cnow = mos7840_port->icount;
+			if (cnow.rng == cprev.rng && cnow.dsr == cprev.dsr &&
+			    cnow.dcd == cprev.dcd && cnow.cts == cprev.cts)
+				return -EIO;	/* no change => error */
+			if (((arg & TIOCM_RNG) && (cnow.rng != cprev.rng)) ||
+			    ((arg & TIOCM_DSR) && (cnow.dsr != cprev.dsr)) ||
+			    ((arg & TIOCM_CD) && (cnow.dcd != cprev.dcd)) ||
+			    ((arg & TIOCM_CTS) && (cnow.cts != cprev.cts))) {
+				return 0;
+			}
+			cprev = cnow;
+		}
+		/* NOTREACHED */
+		break;
+
+	case TIOCGICOUNT:
+		cnow = mos7840_port->icount;
+		icount.cts = cnow.cts;
+		icount.dsr = cnow.dsr;
+		icount.rng = cnow.rng;
+		icount.dcd = cnow.dcd;
+		icount.rx = cnow.rx;
+		icount.tx = cnow.tx;
+		icount.frame = cnow.frame;
+		icount.overrun = cnow.overrun;
+		icount.parity = cnow.parity;
+		icount.brk = cnow.brk;
+		icount.buf_overrun = cnow.buf_overrun;
+
+		dbg("%s (%d) TIOCGICOUNT RX=%d, TX=%d", __FUNCTION__,
+		    port->number, icount.rx, icount.tx);
+		if (copy_to_user((void *)arg, &icount, sizeof(icount)))
+			return -EFAULT;
+		return 0;
+
+	case TIOCEXBAUD:
+		return 0;
+	default:
+		break;
+	}
+
+	return -ENOIOCTLCMD;
+}
+
+static int mos7840_calc_num_ports(struct usb_serial *serial)
+{
+
+	dbg("numberofendpoints: %d \n",
+	    (int)serial->interface->cur_altsetting->desc.bNumEndpoints);
+	dbg("numberofendpoints: %d \n",
+	    (int)serial->interface->altsetting->desc.bNumEndpoints);
+	if (serial->interface->cur_altsetting->desc.bNumEndpoints == 5) {
+		mos7840_num_ports = 2;
+		serial->type->num_ports = 2;
+	} else if (serial->interface->cur_altsetting->desc.bNumEndpoints == 9) {
+		mos7840_num_ports = 4;
+		serial->type->num_bulk_in = 4;
+		serial->type->num_bulk_out = 4;
+		serial->type->num_ports = 4;
+	}
+
+	return mos7840_num_ports;
+}
+
+/****************************************************************************
+ * mos7840_startup
+ ****************************************************************************/
+
+static int mos7840_startup(struct usb_serial *serial)
+{
+	struct moschip_port *mos7840_port;
+	struct usb_device *dev;
+	int i, status;
+
+	__u16 Data;
+	dbg("%s \n", " mos7840_startup :entering..........");
+
+	if (!serial) {
+		dbg("%s\n", "Invalid Handler");
+		return -1;
+	}
+
+	dev = serial->dev;
+
+	dbg("%s\n", "Entering...");
+
+	/* we set up the pointers to the endpoints in the mos7840_open *
+	 * function, as the structures aren't created yet.             */
+
+	/* set up port private structures */
+	for (i = 0; i < serial->num_ports; ++i) {
+		mos7840_port = kmalloc(sizeof(struct moschip_port), GFP_KERNEL);
+		if (mos7840_port == NULL) {
+			err("%s - Out of memory", __FUNCTION__);
+			return -ENOMEM;
+		}
+		memset(mos7840_port, 0, sizeof(struct moschip_port));
+
+		/* Initialize all port interrupt end point to port 0 int endpoint *
+		 * Our device has only one interrupt end point comman to all port */
+
+		mos7840_port->port = serial->port[i];
+		mos7840_set_port_private(serial->port[i], mos7840_port);
+
+		mos7840_port->port_num = ((serial->port[i]->number -
+					   (serial->port[i]->serial->minor)) +
+					  1);
+
+		if (mos7840_port->port_num == 1) {
+			mos7840_port->SpRegOffset = 0x0;
+			mos7840_port->ControlRegOffset = 0x1;
+			mos7840_port->DcrRegOffset = 0x4;
+		} else if ((mos7840_port->port_num == 2)
+			   && (mos7840_num_ports == 4)) {
+			mos7840_port->SpRegOffset = 0x8;
+			mos7840_port->ControlRegOffset = 0x9;
+			mos7840_port->DcrRegOffset = 0x16;
+		} else if ((mos7840_port->port_num == 2)
+			   && (mos7840_num_ports == 2)) {
+			mos7840_port->SpRegOffset = 0xa;
+			mos7840_port->ControlRegOffset = 0xb;
+			mos7840_port->DcrRegOffset = 0x19;
+		} else if ((mos7840_port->port_num == 3)
+			   && (mos7840_num_ports == 4)) {
+			mos7840_port->SpRegOffset = 0xa;
+			mos7840_port->ControlRegOffset = 0xb;
+			mos7840_port->DcrRegOffset = 0x19;
+		} else if ((mos7840_port->port_num == 4)
+			   && (mos7840_num_ports == 4)) {
+			mos7840_port->SpRegOffset = 0xc;
+			mos7840_port->ControlRegOffset = 0xd;
+			mos7840_port->DcrRegOffset = 0x1c;
+		}
+		mos7840_dump_serial_port(mos7840_port);
+
+		mos7840_set_port_private(serial->port[i], mos7840_port);
+
+		//enable rx_disable bit in control register
+
+		status =
+		    mos7840_get_reg_sync(serial->port[i],
+					 mos7840_port->ControlRegOffset, &Data);
+		if (status < 0) {
+			dbg("Reading ControlReg failed status-0x%x\n", status);
+			break;
+		} else
+			dbg("ControlReg Reading success val is %x, status%d\n",
+			    Data, status);
+		Data |= 0x08;	//setting driver done bit
+		Data |= 0x04;	//sp1_bit to have cts change reflect in modem status reg
+
+		//Data |= 0x20; //rx_disable bit
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i],
+					 mos7840_port->ControlRegOffset, Data);
+		if (status < 0) {
+			dbg("Writing ControlReg failed(rx_disable) status-0x%x\n", status);
+			break;
+		} else
+			dbg("ControlReg Writing success(rx_disable) status%d\n",
+			    status);
+
+		//Write default values in DCR (i.e 0x01 in DCR0, 0x05 in DCR2 and 0x24 in DCR3
+		Data = 0x01;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i],
+					 (__u16) (mos7840_port->DcrRegOffset +
+						  0), Data);
+		if (status < 0) {
+			dbg("Writing DCR0 failed status-0x%x\n", status);
+			break;
+		} else
+			dbg("DCR0 Writing success status%d\n", status);
+
+		Data = 0x05;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i],
+					 (__u16) (mos7840_port->DcrRegOffset +
+						  1), Data);
+		if (status < 0) {
+			dbg("Writing DCR1 failed status-0x%x\n", status);
+			break;
+		} else
+			dbg("DCR1 Writing success status%d\n", status);
+
+		Data = 0x24;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i],
+					 (__u16) (mos7840_port->DcrRegOffset +
+						  2), Data);
+		if (status < 0) {
+			dbg("Writing DCR2 failed status-0x%x\n", status);
+			break;
+		} else
+			dbg("DCR2 Writing success status%d\n", status);
+
+		// write values in clkstart0x0 and clkmulti 0x20
+		Data = 0x0;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i],
+					 CLK_START_VALUE_REGISTER, Data);
+		if (status < 0) {
+			dbg("Writing CLK_START_VALUE_REGISTER failed status-0x%x\n", status);
+			break;
+		} else
+			dbg("CLK_START_VALUE_REGISTER Writing success status%d\n", status);
+
+		Data = 0x20;
+		status = 0;
+		status =
+		    mos7840_set_reg_sync(serial->port[i], CLK_MULTI_REGISTER,
+					 Data);
+		if (status < 0) {
+			dbg("Writing CLK_MULTI_REGISTER failed status-0x%x\n",
+			    status);
+			break;
+		} else
+			dbg("CLK_MULTI_REGISTER Writing success status%d\n",
+			    status);
+
+		//write value 0x0 to scratchpad register
+		Data = 0x00;
+		status = 0;
+		status =
+		    mos7840_set_uart_reg(serial->port[i], SCRATCH_PAD_REGISTER,
+					 Data);
+		if (status < 0) {
+			dbg("Writing SCRATCH_PAD_REGISTER failed status-0x%x\n",
+			    status);
+			break;
+		} else
+			dbg("SCRATCH_PAD_REGISTER Writing success status%d\n",
+			    status);
+
+		//Zero Length flag register
+		if ((mos7840_port->port_num != 1)
+		    && (mos7840_num_ports == 2)) {
+
+			Data = 0xff;
+			status = 0;
+			status = mos7840_set_reg_sync(serial->port[i],
+						      (__u16) (ZLP_REG1 +
+							       ((__u16)
+								mos7840_port->
+								port_num)),
+						      Data);
+			dbg("ZLIP offset%x\n",
+			    (__u16) (ZLP_REG1 +
+				     ((__u16) mos7840_port->port_num)));
+			if (status < 0) {
+				dbg("Writing ZLP_REG%d failed status-0x%x\n",
+				    i + 2, status);
+				break;
+			} else
+				dbg("ZLP_REG%d Writing success status%d\n",
+				    i + 2, status);
+		} else {
+			Data = 0xff;
+			status = 0;
+			status = mos7840_set_reg_sync(serial->port[i],
+						      (__u16) (ZLP_REG1 +
+							       ((__u16)
+								mos7840_port->
+								port_num) -
+							       0x1), Data);
+			dbg("ZLIP offset%x\n",
+			    (__u16) (ZLP_REG1 +
+				     ((__u16) mos7840_port->port_num) - 0x1));
+			if (status < 0) {
+				dbg("Writing ZLP_REG%d failed status-0x%x\n",
+				    i + 1, status);
+				break;
+			} else
+				dbg("ZLP_REG%d Writing success status%d\n",
+				    i + 1, status);
+
+		}
+		mos7840_port->control_urb = usb_alloc_urb(0, SLAB_ATOMIC);
+		mos7840_port->ctrl_buf = kmalloc(16, GFP_KERNEL);
+
+	}
+
+	//Zero Length flag enable
+	Data = 0x0f;
+	status = 0;
+	status = mos7840_set_reg_sync(serial->port[0], ZLP_REG5, Data);
+	if (status < 0) {
+		dbg("Writing ZLP_REG5 failed status-0x%x\n", status);
+		return -1;
+	} else
+		dbg("ZLP_REG5 Writing success status%d\n", status);
+
+	/* setting configuration feature to one */
+	usb_control_msg(serial->dev, usb_sndctrlpipe(serial->dev, 0),
+			(__u8) 0x03, 0x00, 0x01, 0x00, 0x00, 0x00, 5 * HZ);
+	return 0;
+}
+
+/****************************************************************************
+ * mos7840_shutdown
+ *	This function is called whenever the device is removed from the usb bus.
+ ****************************************************************************/
+
+static void mos7840_shutdown(struct usb_serial *serial)
+{
+	int i;
+	struct moschip_port *mos7840_port;
+	dbg("%s \n", " shutdown :entering..........");
+
+	if (!serial) {
+		dbg("%s", "Invalid Handler \n");
+		return;
+	}
+
+	/*      check for the ports to be closed,close the ports and disconnect         */
+
+	/* free private structure allocated for serial port  *
+	 * stop reads and writes on all ports                */
+
+	for (i = 0; i < serial->num_ports; ++i) {
+		mos7840_port = mos7840_get_port_private(serial->port[i]);
+		kfree(mos7840_port->ctrl_buf);
+		usb_kill_urb(mos7840_port->control_urb);
+		kfree(mos7840_port);
+		mos7840_set_port_private(serial->port[i], NULL);
+	}
+
+	dbg("%s\n", "Thank u :: ");
+
+}
+
+static struct usb_serial_driver moschip7840_4port_device = {
+	.driver = {
+		   .owner = THIS_MODULE,
+		   .name = "mos7840",
+		   },
+	.description = DRIVER_DESC,
+	.id_table = moschip_port_id_table,
+	.num_interrupt_in = 1,	//NUM_DONT_CARE,//1,
+#ifdef check
+	.num_bulk_in = 4,
+	.num_bulk_out = 4,
+	.num_ports = 4,
+#endif
+	.open = mos7840_open,
+	.close = mos7840_close,
+	.write = mos7840_write,
+	.write_room = mos7840_write_room,
+	.chars_in_buffer = mos7840_chars_in_buffer,
+	.throttle = mos7840_throttle,
+	.unthrottle = mos7840_unthrottle,
+	.calc_num_ports = mos7840_calc_num_ports,
+#ifdef MCSSerialProbe
+	.probe = mos7840_serial_probe,
+#endif
+	.ioctl = mos7840_ioctl,
+	.set_termios = mos7840_set_termios,
+	.break_ctl = mos7840_break,
+	.tiocmget = mos7840_tiocmget,
+	.tiocmset = mos7840_tiocmset,
+	.attach = mos7840_startup,
+	.shutdown = mos7840_shutdown,
+	.read_bulk_callback = mos7840_bulk_in_callback,
+	.read_int_callback = mos7840_interrupt_callback,
+};
+
+static struct usb_driver io_driver = {
+	.name = "mos7840",
+	.probe = usb_serial_probe,
+	.disconnect = usb_serial_disconnect,
+	.id_table = moschip_id_table_combined,
+};
+
+/****************************************************************************
+ * moschip7840_init
+ *	This is called by the module subsystem, or on startup to initialize us
+ ****************************************************************************/
+static int __init moschip7840_init(void)
+{
+	int retval;
+
+	dbg("%s \n", " mos7840_init :entering..........");
+
+	/* Register with the usb serial */
+	retval = usb_serial_register(&moschip7840_4port_device);
+
+	if (retval)
+		goto failed_port_device_register;
+
+	dbg("%s\n", "Entring...");
+	info(DRIVER_DESC " " DRIVER_VERSION);
+
+	/* Register with the usb */
+	retval = usb_register(&io_driver);
+
+	if (retval)
+		goto failed_usb_register;
+
+	if (retval == 0) {
+		dbg("%s\n", "Leaving...");
+		return 0;
+	}
+
+      failed_usb_register:
+	usb_serial_deregister(&moschip7840_4port_device);
+
+      failed_port_device_register:
+
+	return retval;
+}
+
+/****************************************************************************
+ * moschip7840_exit
+ *	Called when the driver is about to be unloaded.
+ ****************************************************************************/
+static void __exit moschip7840_exit(void)
+{
+
+	dbg("%s \n", " mos7840_exit :entering..........");
+
+	usb_deregister(&io_driver);
+
+	usb_serial_deregister(&moschip7840_4port_device);
+
+	dbg("%s\n", "Entring...");
+}
+
+module_init(moschip7840_init);
+module_exit(moschip7840_exit);
+
+/* Module information */
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE("GPL");
+
+module_param(debug, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Debug enabled or not");
-- 
cgit v1.2.1


From 8fd801339350b63cbb90730ff8b2be349fb3dc67 Mon Sep 17 00:00:00 2001
From: Johannes Steingraeber <Jo_Stein@web.de>
Date: Sat, 16 Sep 2006 16:17:34 +0200
Subject: usb serial: support Alcor Micro Corp. USB 2.0 TO RS-232 through
 pl2303 driver

Patch to add support for Alcor Micro Corp. USB 2.0 TO RS-232 converter.
This patch adds VID and PID to pl2303.[ch], adds it to the "HORRIBLE
HACK FOR PL2303" in usb-serial.c and also prevents cdc-acm to claim
driving this device by blacklisting it in hid-core.

Signed-off-by: Johannes Steingraeber <Jo_Stein@web.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/hid-core.c    | 5 +++++
 drivers/usb/serial/pl2303.c     | 1 +
 drivers/usb/serial/pl2303.h     | 4 ++++
 drivers/usb/serial/usb-serial.c | 4 +++-
 4 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index d0d1c0c30bc5..fc5b662ea177 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1595,6 +1595,10 @@ void hid_init_reports(struct hid_device *hid)
 
 #define USB_VENDOR_ID_YEALINK		0x6993
 #define USB_DEVICE_ID_YEALINK_P1K_P4K_B2K	0xb001
+
+#define USB_VENDOR_ID_ALCOR		0x058f
+#define USB_DEVICE_ID_ALCOR_USBRS232	0x9720
+
 /*
  * Alphabetically sorted blacklist by quirk type.
  */
@@ -1612,6 +1616,7 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_22, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_23, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_24, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_ALCOR, USB_DEVICE_ID_ALCOR_USBRS232, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_BERKSHIRE, USB_DEVICE_ID_BERKSHIRE_PCWD, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW40, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_CODEMERCS, USB_DEVICE_ID_CODEMERCS_IOW24, HID_QUIRK_IGNORE },
diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index 51caf8a61463..1036d436ed23 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -81,6 +81,7 @@ static struct usb_device_id id_table [] = {
 	{ USB_DEVICE(SPEEDDRAGON_VENDOR_ID, SPEEDDRAGON_PRODUCT_ID) },
 	{ USB_DEVICE(DATAPILOT_U2_VENDOR_ID, DATAPILOT_U2_PRODUCT_ID) },
 	{ USB_DEVICE(BELKIN_VENDOR_ID, BELKIN_PRODUCT_ID) },
+	{ USB_DEVICE(ALCOR_VENDOR_ID, ALCOR_PRODUCT_ID) },
 	{ }					/* Terminating entry */
 };
 
diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h
index 55195e76eb6f..762cc290ef58 100644
--- a/drivers/usb/serial/pl2303.h
+++ b/drivers/usb/serial/pl2303.h
@@ -89,3 +89,7 @@
 /* Belkin "F5U257" Serial Adapter */
 #define BELKIN_VENDOR_ID	0x050d
 #define BELKIN_PRODUCT_ID	0x0257
+
+/* Alcor Micro Corp. USB 2.0 TO RS-232 */
+#define ALCOR_VENDOR_ID		0x058F
+#define ALCOR_PRODUCT_ID	0x9720
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 03c619478a7c..0222d92842b8 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -712,7 +712,9 @@ int usb_serial_probe(struct usb_interface *interface,
 	if (((le16_to_cpu(dev->descriptor.idVendor) == PL2303_VENDOR_ID) &&
 	     (le16_to_cpu(dev->descriptor.idProduct) == PL2303_PRODUCT_ID)) ||
 	    ((le16_to_cpu(dev->descriptor.idVendor) == ATEN_VENDOR_ID) &&
-	     (le16_to_cpu(dev->descriptor.idProduct) == ATEN_PRODUCT_ID))) {
+	     (le16_to_cpu(dev->descriptor.idProduct) == ATEN_PRODUCT_ID)) ||
+	    ((le16_to_cpu(dev->descriptor.idVendor) == ALCOR_VENDOR_ID) &&
+	     (le16_to_cpu(dev->descriptor.idProduct) == ALCOR_PRODUCT_ID))) {
 		if (interface != dev->actconfig->interface[0]) {
 			/* check out the endpoints of the other interface*/
 			iface_desc = dev->actconfig->interface[0]->cur_altsetting;
-- 
cgit v1.2.1


From a5c66e4b2418278786a025a5bd9625f485b2087a Mon Sep 17 00:00:00 2001
From: Tony Olech <tony.olech@elandigitalsystems.com>
Date: Wed, 13 Sep 2006 11:26:04 +0100
Subject: USB: ftdi-elan: client driver for ELAN Uxxx adapters

This "ftdi-elan" module is one half of the "driver" for
ELAN's Uxxx series adapters which are USB to PCMCIA CardBus
adapters. Currently only the U132 adapter is available and
it's module is called "u132-hcd".

When the USB hot plug subsystem detects a Uxxx series adapter
it should load this module.

Upon a successful device probe() the jtag device file interface
is created and the status workqueue started up.

The jtag device file interface exists for the purpose of
updating the firmware in the Uxxx series adapter, but as
yet it had never been used.

The status workqueue initializes the Uxxx and then sits there
polling the Uxxx until a supported PCMCIA CardBus device is
detected it will start the command and respond workqueues
and then load the module that handles the device. This will
initially be only the u132-hcd module. The status workqueue
then just polls the Uxxx looking for card ejects.

The command and respond workqueues implement a command
sequencer for communicating with the firmware on the other
side of the FTDI chip in the Uxxx. This "ftdi-elan" module
exports some functions to interface with the sequencer.

Note that this module is a USB client driver.

Note that the "u132-hcd" module is a (cut-down OHCI)
host controller.

Thus we have a topology with the parent of a host controller
being a USB client! This really stresses the USB subsystem
semaphore/mutex handling in the module removal.

Signed-off-by: Tony Olech <tony.olech@elandigitalsystems.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/Makefile         |    9 +-
 drivers/usb/misc/Kconfig     |   24 +
 drivers/usb/misc/Makefile    |    2 +-
 drivers/usb/misc/ftdi-elan.c | 2809 ++++++++++++++++++++++++++++++++++++++++++
 drivers/usb/misc/usb_u132.h  |   97 ++
 5 files changed, 2936 insertions(+), 5 deletions(-)
 create mode 100644 drivers/usb/misc/ftdi-elan.c
 create mode 100644 drivers/usb/misc/usb_u132.h

diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index 46946b2d3088..e33647057095 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -48,23 +48,24 @@ obj-$(CONFIG_USB_MICROTEK)	+= image/
 
 obj-$(CONFIG_USB_SERIAL)	+= serial/
 
+obj-$(CONFIG_USB_ADUTUX)	+= misc/
+obj-$(CONFIG_USB_APPLEDISPLAY)	+= misc/
 obj-$(CONFIG_USB_AUERSWALD)	+= misc/
 obj-$(CONFIG_USB_CYPRESS_CY7C63)+= misc/
 obj-$(CONFIG_USB_CYTHERM)	+= misc/
 obj-$(CONFIG_USB_EMI26)		+= misc/
 obj-$(CONFIG_USB_EMI62)		+= misc/
+obj-$(CONFIG_USB_FTDI_ELAN)	+= misc/
 obj-$(CONFIG_USB_IDMOUSE)	+= misc/
 obj-$(CONFIG_USB_LCD)		+= misc/
 obj-$(CONFIG_USB_LD)		+= misc/
 obj-$(CONFIG_USB_LED)		+= misc/
 obj-$(CONFIG_USB_LEGOTOWER)	+= misc/
+obj-$(CONFIG_USB_PHIDGETSERVO)	+= misc/
 obj-$(CONFIG_USB_RIO500)	+= misc/
+obj-$(CONFIG_USB_SISUSBVGA)	+= misc/
 obj-$(CONFIG_USB_TEST)		+= misc/
 obj-$(CONFIG_USB_USS720)	+= misc/
-obj-$(CONFIG_USB_PHIDGETSERVO)	+= misc/
-obj-$(CONFIG_USB_SISUSBVGA)	+= misc/
-obj-$(CONFIG_USB_APPLEDISPLAY)	+= misc/
-obj-$(CONFIG_USB_ADUTUX)	+= misc/
 
 obj-$(CONFIG_USB_ATM)		+= atm/
 obj-$(CONFIG_USB_SPEEDTOUCH)	+= atm/
diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
index 288d301d2bff..c29658f69e2a 100644
--- a/drivers/usb/misc/Kconfig
+++ b/drivers/usb/misc/Kconfig
@@ -178,6 +178,30 @@ config USB_IDMOUSE
 
 	  See also <http://www.fs.tum.de/~echtler/idmouse/>.
 
+config USB_FTDI_ELAN
+	tristate "Elan PCMCIA CardBus Adapter USB Client"
+	depends on USB
+	default M
+	help
+	  ELAN's Uxxx series of adapters are USB to PCMCIA CardBus adapters.
+	  Currently only the U132 adapter is available.
+
+	  The U132 is specifically designed for CardBus PC cards that contain
+	  an OHCI host controller. Typical PC cards are the Orange Mobile 3G
+	  Option GlobeTrotter Fusion card. The U132 adapter will *NOT* work
+	  with PC cards that do not contain an OHCI controller. To use a U132
+	  adapter you will need this "ftdi-elan" module as well as the "u132-hcd"
+	  module which is a USB host controller driver that talks to the OHCI
+	  controller within CardBus card that are inserted in the U132 adapter.
+
+	  This driver has been tested with a CardBus OHCI USB adapter, and
+	  worked with a USB PEN Drive inserted into the first USB port of
+	  the PCCARD. A rather pointless thing to do, but useful for testing.
+
+	  See also the USB_U132_HCD entry "Elan U132 Adapter Host Controller"
+
+	  It is safe to say M here.
+
 config USB_APPLEDISPLAY
 	tristate "Apple Cinema Display support"
 	depends on USB
diff --git a/drivers/usb/misc/Makefile b/drivers/usb/misc/Makefile
index 73fc8be0d8c0..2be70fa259bf 100644
--- a/drivers/usb/misc/Makefile
+++ b/drivers/usb/misc/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_USB_CYPRESS_CY7C63)+= cypress_cy7c63.o
 obj-$(CONFIG_USB_CYTHERM)	+= cytherm.o
 obj-$(CONFIG_USB_EMI26)		+= emi26.o
 obj-$(CONFIG_USB_EMI62)		+= emi62.o
+obj-$(CONFIG_USB_FTDI_ELAN)	+= ftdi-elan.o
 obj-$(CONFIG_USB_IDMOUSE)	+= idmouse.o
 obj-$(CONFIG_USB_LCD)		+= usblcd.o
 obj-$(CONFIG_USB_LD)		+= ldusb.o
@@ -21,7 +22,6 @@ obj-$(CONFIG_USB_PHIDGETSERVO)	+= phidgetservo.o
 obj-$(CONFIG_USB_RIO500)	+= rio500.o
 obj-$(CONFIG_USB_TEST)		+= usbtest.o
 obj-$(CONFIG_USB_USS720)	+= uss720.o
-obj-$(CONFIG_USB_APPLEDISPLAY)	+= appledisplay.o
 
 obj-$(CONFIG_USB_SISUSBVGA)	+= sisusbvga/
 
diff --git a/drivers/usb/misc/ftdi-elan.c b/drivers/usb/misc/ftdi-elan.c
new file mode 100644
index 000000000000..b88a09497c28
--- /dev/null
+++ b/drivers/usb/misc/ftdi-elan.c
@@ -0,0 +1,2809 @@
+/*
+* USB FTDI client driver for Elan Digital Systems's Uxxx adapters
+*
+* Copyright(C) 2006 Elan Digital Systems Limited
+* http://www.elandigitalsystems.com
+*
+* Author and Maintainer - Tony Olech - Elan Digital Systems
+* tony.olech@elandigitalsystems.com
+*
+* This program is free software;you can redistribute it and/or
+* modify it under the terms of the GNU General Public License as
+* published by the Free Software Foundation, version 2.
+*
+*
+* This driver was written by Tony Olech(tony.olech@elandigitalsystems.com)
+* based on various USB client drivers in the 2.6.15 linux kernel
+* with constant reference to the 3rd Edition of Linux Device Drivers
+* published by O'Reilly
+*
+* The U132 adapter is a USB to CardBus adapter specifically designed
+* for PC cards that contain an OHCI host controller. Typical PC cards
+* are the Orange Mobile 3G Option GlobeTrotter Fusion card.
+*
+* The U132 adapter will *NOT *work with PC cards that do not contain
+* an OHCI controller. A simple way to test whether a PC card has an
+* OHCI controller as an interface is to insert the PC card directly
+* into a laptop(or desktop) with a CardBus slot and if "lspci" shows
+* a new USB controller and "lsusb -v" shows a new OHCI Host Controller
+* then there is a good chance that the U132 adapter will support the
+* PC card.(you also need the specific client driver for the PC card)
+*
+* Please inform the Author and Maintainer about any PC cards that
+* contain OHCI Host Controller and work when directly connected to
+* an embedded CardBus slot but do not work when they are connected
+* via an ELAN U132 adapter.
+*
+*/
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/ioctl.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/kref.h>
+#include <asm/uaccess.h>
+#include <linux/usb.h>
+#include <linux/workqueue.h>
+#include <linux/platform_device.h>
+MODULE_AUTHOR("Tony Olech");
+MODULE_DESCRIPTION("FTDI ELAN driver");
+MODULE_LICENSE("GPL");
+#define INT_MODULE_PARM(n, v) static int n = v;module_param(n, int, 0444)
+extern struct platform_driver u132_platform_driver;
+static struct workqueue_struct *status_queue;
+static struct workqueue_struct *command_queue;
+static struct workqueue_struct *respond_queue;
+/*
+* ftdi_module_lock exists to protect access to global variables
+*
+*/
+static struct semaphore ftdi_module_lock;
+static int ftdi_instances = 0;
+static struct list_head ftdi_static_list;
+/*
+* end of the global variables protected by ftdi_module_lock
+*/
+#include "usb_u132.h"
+#define TD_DEVNOTRESP 5
+/* Define these values to match your devices*/
+#define USB_FTDI_ELAN_VENDOR_ID 0x0403
+#define USB_FTDI_ELAN_PRODUCT_ID 0xd6ea
+/* table of devices that work with this driver*/
+static struct usb_device_id ftdi_elan_table[] = {
+        {USB_DEVICE(USB_FTDI_ELAN_VENDOR_ID, USB_FTDI_ELAN_PRODUCT_ID)},
+        { /* Terminating entry */ }
+};
+
+MODULE_DEVICE_TABLE(usb, ftdi_elan_table);
+/* only the jtag(firmware upgrade device) interface requires
+* a device file and corresponding minor number, but the
+* interface is created unconditionally - I suppose it could
+* be configured or not according to a module parameter.
+* But since we(now) require one interface per device,
+* and since it unlikely that a normal installation would
+* require more than a couple of elan-ftdi devices, 8 seems
+* like a reasonable limit to have here, and if someone
+* really requires more than 8 devices, then they can frig the
+* code and recompile
+*/
+#define USB_FTDI_ELAN_MINOR_BASE 192
+#define COMMAND_BITS 5
+#define COMMAND_SIZE (1<<COMMAND_BITS)
+#define COMMAND_MASK (COMMAND_SIZE-1)
+struct u132_command {
+        u8 header;
+        u16 length;
+        u8 address;
+        u8 width;
+        u32 value;
+        int follows;
+        void *buffer;
+};
+#define RESPOND_BITS 5
+#define RESPOND_SIZE (1<<RESPOND_BITS)
+#define RESPOND_MASK (RESPOND_SIZE-1)
+struct u132_respond {
+        u8 header;
+        u8 address;
+        u32 *value;
+        int *result;
+        struct completion wait_completion;
+};
+struct u132_target {
+        void *endp;
+        struct urb *urb;
+        int toggle_bits;
+        int error_count;
+        int condition_code;
+        int repeat_number;
+        int halted;
+        int skipped;
+        int actual;
+        int non_null;
+        int active;
+        int abandoning;
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+                int toggle_bits, int error_count, int condition_code,
+                int repeat_number, int halted, int skipped, int actual,
+                int non_null);
+};
+/* Structure to hold all of our device specific stuff*/
+struct usb_ftdi {
+        struct list_head ftdi_list;
+        struct semaphore u132_lock;
+        int command_next;
+        int command_head;
+        struct u132_command command[COMMAND_SIZE];
+        int respond_next;
+        int respond_head;
+        struct u132_respond respond[RESPOND_SIZE];
+        struct u132_target target[4];
+        char device_name[16];
+        unsigned synchronized:1;
+        unsigned enumerated:1;
+        unsigned registered:1;
+        unsigned initialized:1;
+        unsigned card_ejected:1;
+        int function;
+        int sequence_num;
+        int disconnected;
+        int gone_away;
+        int stuck_status;
+        int status_queue_delay;
+        struct semaphore sw_lock;
+        struct usb_device *udev;
+        struct usb_interface *interface;
+        struct usb_class_driver *class;
+        struct work_struct status_work;
+        struct work_struct command_work;
+        struct work_struct respond_work;
+        struct u132_platform_data platform_data;
+        struct resource resources[0];
+        struct platform_device platform_dev;
+        unsigned char *bulk_in_buffer;
+        size_t bulk_in_size;
+        size_t bulk_in_last;
+        size_t bulk_in_left;
+        __u8 bulk_in_endpointAddr;
+        __u8 bulk_out_endpointAddr;
+        struct kref kref;
+        u32 controlreg;
+        u8 response[4 + 1024];
+        int expected;
+        int recieved;
+        int ed_found;
+};
+#define kref_to_usb_ftdi(d) container_of(d, struct usb_ftdi, kref)
+#define platform_device_to_usb_ftdi(d) container_of(d, struct usb_ftdi, \
+        platform_dev)
+static struct usb_driver ftdi_elan_driver;
+static void ftdi_elan_delete(struct kref *kref)
+{
+        struct usb_ftdi *ftdi = kref_to_usb_ftdi(kref);
+        dev_warn(&ftdi->udev->dev, "FREEING ftdi=%p\n", ftdi);
+        usb_put_dev(ftdi->udev);
+        ftdi->disconnected += 1;
+        down(&ftdi_module_lock);
+        list_del_init(&ftdi->ftdi_list);
+        ftdi_instances -= 1;
+        up(&ftdi_module_lock);
+        kfree(ftdi->bulk_in_buffer);
+        ftdi->bulk_in_buffer = NULL;
+}
+
+static void ftdi_elan_put_kref(struct usb_ftdi *ftdi)
+{
+        kref_put(&ftdi->kref, ftdi_elan_delete);
+}
+
+static void ftdi_elan_get_kref(struct usb_ftdi *ftdi)
+{
+        kref_get(&ftdi->kref);
+}
+
+static void ftdi_elan_init_kref(struct usb_ftdi *ftdi)
+{
+        kref_init(&ftdi->kref);
+}
+
+static void ftdi_status_requeue_work(struct usb_ftdi *ftdi, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(status_queue, &ftdi->status_work, delta))
+                        return;
+        } else if (queue_work(status_queue, &ftdi->status_work))
+                return;
+        kref_put(&ftdi->kref, ftdi_elan_delete);
+        return;
+}
+
+static void ftdi_status_queue_work(struct usb_ftdi *ftdi, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(status_queue, &ftdi->status_work, delta))
+                        kref_get(&ftdi->kref);
+        } else if (queue_work(status_queue, &ftdi->status_work))
+                kref_get(&ftdi->kref);
+        return;
+}
+
+static void ftdi_status_cancel_work(struct usb_ftdi *ftdi)
+{
+        if (cancel_delayed_work(&ftdi->status_work))
+                kref_put(&ftdi->kref, ftdi_elan_delete);
+}
+
+static void ftdi_command_requeue_work(struct usb_ftdi *ftdi, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(command_queue, &ftdi->command_work,
+                        delta))
+                        return;
+        } else if (queue_work(command_queue, &ftdi->command_work))
+                return;
+        kref_put(&ftdi->kref, ftdi_elan_delete);
+        return;
+}
+
+static void ftdi_command_queue_work(struct usb_ftdi *ftdi, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(command_queue, &ftdi->command_work,
+                        delta))
+                        kref_get(&ftdi->kref);
+        } else if (queue_work(command_queue, &ftdi->command_work))
+                kref_get(&ftdi->kref);
+        return;
+}
+
+static void ftdi_command_cancel_work(struct usb_ftdi *ftdi)
+{
+        if (cancel_delayed_work(&ftdi->command_work))
+                kref_put(&ftdi->kref, ftdi_elan_delete);
+}
+
+static void ftdi_response_requeue_work(struct usb_ftdi *ftdi,
+        unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(respond_queue, &ftdi->respond_work,
+                        delta))
+                        return;
+        } else if (queue_work(respond_queue, &ftdi->respond_work))
+                return;
+        kref_put(&ftdi->kref, ftdi_elan_delete);
+        return;
+}
+
+static void ftdi_respond_queue_work(struct usb_ftdi *ftdi, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(respond_queue, &ftdi->respond_work,
+                        delta))
+                        kref_get(&ftdi->kref);
+        } else if (queue_work(respond_queue, &ftdi->respond_work))
+                kref_get(&ftdi->kref);
+        return;
+}
+
+static void ftdi_response_cancel_work(struct usb_ftdi *ftdi)
+{
+        if (cancel_delayed_work(&ftdi->respond_work))
+                kref_put(&ftdi->kref, ftdi_elan_delete);
+}
+
+void ftdi_elan_gone_away(struct platform_device *pdev)
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        ftdi->gone_away += 1;
+        ftdi_elan_put_kref(ftdi);
+}
+
+
+EXPORT_SYMBOL_GPL(ftdi_elan_gone_away);
+void ftdi_release_platform_dev(struct device *dev)
+{
+        dev->parent = NULL;
+}
+
+static void ftdi_elan_do_callback(struct usb_ftdi *ftdi,
+        struct u132_target *target, u8 *buffer, int length);
+static void ftdi_elan_kick_command_queue(struct usb_ftdi *ftdi);
+static void ftdi_elan_kick_respond_queue(struct usb_ftdi *ftdi);
+static int ftdi_elan_setupOHCI(struct usb_ftdi *ftdi);
+static int ftdi_elan_checkingPCI(struct usb_ftdi *ftdi);
+static int ftdi_elan_enumeratePCI(struct usb_ftdi *ftdi);
+static int ftdi_elan_synchronize(struct usb_ftdi *ftdi);
+static int ftdi_elan_stuck_waiting(struct usb_ftdi *ftdi);
+static int ftdi_elan_command_engine(struct usb_ftdi *ftdi);
+static int ftdi_elan_respond_engine(struct usb_ftdi *ftdi);
+static int ftdi_elan_hcd_init(struct usb_ftdi *ftdi)
+{
+        int result;
+        if (ftdi->platform_dev.dev.parent)
+                return -EBUSY;
+        ftdi_elan_get_kref(ftdi);
+        ftdi->platform_data.potpg = 100;
+        ftdi->platform_data.reset = NULL;
+        ftdi->platform_dev.id = ftdi->sequence_num;
+        ftdi->platform_dev.resource = ftdi->resources;
+        ftdi->platform_dev.num_resources = ARRAY_SIZE(ftdi->resources);
+        ftdi->platform_dev.dev.platform_data = &ftdi->platform_data;
+        ftdi->platform_dev.dev.parent = NULL;
+        ftdi->platform_dev.dev.release = ftdi_release_platform_dev;
+        ftdi->platform_dev.dev.dma_mask = NULL;
+        snprintf(ftdi->device_name, sizeof(ftdi->device_name), "u132_hcd");
+        ftdi->platform_dev.name = ftdi->device_name;
+        dev_info(&ftdi->udev->dev, "requesting module '%s'\n", "u132_hcd");
+        request_module("u132_hcd");
+        dev_info(&ftdi->udev->dev, "registering '%s'\n",
+                ftdi->platform_dev.name);
+        result = platform_device_register(&ftdi->platform_dev);
+        return result;
+}
+
+static void ftdi_elan_abandon_completions(struct usb_ftdi *ftdi)
+{
+        down(&ftdi->u132_lock);
+        while (ftdi->respond_next > ftdi->respond_head) {
+                struct u132_respond *respond = &ftdi->respond[RESPOND_MASK &
+                        ftdi->respond_head++];
+                *respond->result = -ESHUTDOWN;
+                *respond->value = 0;
+                complete(&respond->wait_completion);
+        } up(&ftdi->u132_lock);
+}
+
+static void ftdi_elan_abandon_targets(struct usb_ftdi *ftdi)
+{
+        int ed_number = 4;
+        down(&ftdi->u132_lock);
+        while (ed_number-- > 0) {
+                struct u132_target *target = &ftdi->target[ed_number];
+                if (target->active == 1) {
+                        target->condition_code = TD_DEVNOTRESP;
+                        up(&ftdi->u132_lock);
+                        ftdi_elan_do_callback(ftdi, target, NULL, 0);
+                        down(&ftdi->u132_lock);
+                }
+        }
+        ftdi->recieved = 0;
+        ftdi->expected = 4;
+        ftdi->ed_found = 0;
+        up(&ftdi->u132_lock);
+}
+
+static void ftdi_elan_flush_targets(struct usb_ftdi *ftdi)
+{
+        int ed_number = 4;
+        down(&ftdi->u132_lock);
+        while (ed_number-- > 0) {
+                struct u132_target *target = &ftdi->target[ed_number];
+                target->abandoning = 1;
+              wait_1:if (target->active == 1) {
+                        int command_size = ftdi->command_next -
+                                ftdi->command_head;
+                        if (command_size < COMMAND_SIZE) {
+                                struct u132_command *command = &ftdi->command[
+                                        COMMAND_MASK & ftdi->command_next];
+                                command->header = 0x80 | (ed_number << 5) | 0x4;
+                                command->length = 0x00;
+                                command->address = 0x00;
+                                command->width = 0x00;
+                                command->follows = 0;
+                                command->value = 0;
+                                command->buffer = &command->value;
+                                ftdi->command_next += 1;
+                                ftdi_elan_kick_command_queue(ftdi);
+                        } else {
+                                up(&ftdi->u132_lock);
+                                msleep(100);
+                                down(&ftdi->u132_lock);
+                                goto wait_1;
+                        }
+                }
+              wait_2:if (target->active == 1) {
+                        int command_size = ftdi->command_next -
+                                ftdi->command_head;
+                        if (command_size < COMMAND_SIZE) {
+                                struct u132_command *command = &ftdi->command[
+                                        COMMAND_MASK & ftdi->command_next];
+                                command->header = 0x90 | (ed_number << 5);
+                                command->length = 0x00;
+                                command->address = 0x00;
+                                command->width = 0x00;
+                                command->follows = 0;
+                                command->value = 0;
+                                command->buffer = &command->value;
+                                ftdi->command_next += 1;
+                                ftdi_elan_kick_command_queue(ftdi);
+                        } else {
+                                up(&ftdi->u132_lock);
+                                msleep(100);
+                                down(&ftdi->u132_lock);
+                                goto wait_2;
+                        }
+                }
+        }
+        ftdi->recieved = 0;
+        ftdi->expected = 4;
+        ftdi->ed_found = 0;
+        up(&ftdi->u132_lock);
+}
+
+static void ftdi_elan_cancel_targets(struct usb_ftdi *ftdi)
+{
+        int ed_number = 4;
+        down(&ftdi->u132_lock);
+        while (ed_number-- > 0) {
+                struct u132_target *target = &ftdi->target[ed_number];
+                target->abandoning = 1;
+              wait:if (target->active == 1) {
+                        int command_size = ftdi->command_next -
+                                ftdi->command_head;
+                        if (command_size < COMMAND_SIZE) {
+                                struct u132_command *command = &ftdi->command[
+                                        COMMAND_MASK & ftdi->command_next];
+                                command->header = 0x80 | (ed_number << 5) | 0x4;
+                                command->length = 0x00;
+                                command->address = 0x00;
+                                command->width = 0x00;
+                                command->follows = 0;
+                                command->value = 0;
+                                command->buffer = &command->value;
+                                ftdi->command_next += 1;
+                                ftdi_elan_kick_command_queue(ftdi);
+                        } else {
+                                up(&ftdi->u132_lock);
+                                msleep(100);
+                                down(&ftdi->u132_lock);
+                                goto wait;
+                        }
+                }
+        }
+        ftdi->recieved = 0;
+        ftdi->expected = 4;
+        ftdi->ed_found = 0;
+        up(&ftdi->u132_lock);
+}
+
+static void ftdi_elan_kick_command_queue(struct usb_ftdi *ftdi)
+{
+        ftdi_command_queue_work(ftdi, 0);
+        return;
+}
+
+static void ftdi_elan_command_work(void *data)
+{
+        struct usb_ftdi *ftdi = data;
+        if (ftdi->disconnected > 0) {
+                ftdi_elan_put_kref(ftdi);
+                return;
+        } else {
+                int retval = ftdi_elan_command_engine(ftdi);
+                if (retval == -ESHUTDOWN) {
+                        ftdi->disconnected += 1;
+                } else if (retval == -ENODEV) {
+                        ftdi->disconnected += 1;
+                } else if (retval)
+                        dev_err(&ftdi->udev->dev, "command error %d\n", retval);
+                ftdi_command_requeue_work(ftdi, msecs_to_jiffies(10));
+                return;
+        }
+}
+
+static void ftdi_elan_kick_respond_queue(struct usb_ftdi *ftdi)
+{
+        ftdi_respond_queue_work(ftdi, 0);
+        return;
+}
+
+static void ftdi_elan_respond_work(void *data)
+{
+        struct usb_ftdi *ftdi = data;
+        if (ftdi->disconnected > 0) {
+                ftdi_elan_put_kref(ftdi);
+                return;
+        } else {
+                int retval = ftdi_elan_respond_engine(ftdi);
+                if (retval == 0) {
+                } else if (retval == -ESHUTDOWN) {
+                        ftdi->disconnected += 1;
+                } else if (retval == -ENODEV) {
+                        ftdi->disconnected += 1;
+                } else if (retval == -ENODEV) {
+                        ftdi->disconnected += 1;
+                } else if (retval == -EILSEQ) {
+                        ftdi->disconnected += 1;
+                } else {
+                        ftdi->disconnected += 1;
+                        dev_err(&ftdi->udev->dev, "respond error %d\n", retval);
+                }
+                if (ftdi->disconnected > 0) {
+                        ftdi_elan_abandon_completions(ftdi);
+                        ftdi_elan_abandon_targets(ftdi);
+                }
+                ftdi_response_requeue_work(ftdi, msecs_to_jiffies(10));
+                return;
+        }
+}
+
+
+/*
+* the sw_lock is initially held and will be freed
+* after the FTDI has been synchronized
+*
+*/
+static void ftdi_elan_status_work(void *data)
+{
+        struct usb_ftdi *ftdi = data;
+        int work_delay_in_msec = 0;
+        if (ftdi->disconnected > 0) {
+                ftdi_elan_put_kref(ftdi);
+                return;
+        } else if (ftdi->synchronized == 0) {
+                down(&ftdi->sw_lock);
+                if (ftdi_elan_synchronize(ftdi) == 0) {
+                        ftdi->synchronized = 1;
+                        ftdi_command_queue_work(ftdi, 1);
+                        ftdi_respond_queue_work(ftdi, 1);
+                        up(&ftdi->sw_lock);
+                        work_delay_in_msec = 100;
+                } else {
+                        dev_err(&ftdi->udev->dev, "synchronize failed\n");
+                        up(&ftdi->sw_lock);
+                        work_delay_in_msec = 10 *1000;
+                }
+        } else if (ftdi->stuck_status > 0) {
+                if (ftdi_elan_stuck_waiting(ftdi) == 0) {
+                        ftdi->stuck_status = 0;
+                        ftdi->synchronized = 0;
+                } else if ((ftdi->stuck_status++ % 60) == 1) {
+                        dev_err(&ftdi->udev->dev, "WRONG type of card inserted "
+                                "- please remove\n");
+                } else
+                        dev_err(&ftdi->udev->dev, "WRONG type of card inserted "
+                                "- checked %d times\n", ftdi->stuck_status);
+                work_delay_in_msec = 100;
+        } else if (ftdi->enumerated == 0) {
+                if (ftdi_elan_enumeratePCI(ftdi) == 0) {
+                        ftdi->enumerated = 1;
+                        work_delay_in_msec = 250;
+                } else
+                        work_delay_in_msec = 1000;
+        } else if (ftdi->initialized == 0) {
+                if (ftdi_elan_setupOHCI(ftdi) == 0) {
+                        ftdi->initialized = 1;
+                        work_delay_in_msec = 500;
+                } else {
+                        dev_err(&ftdi->udev->dev, "initialized failed - trying "
+                                "again in 10 seconds\n");
+                        work_delay_in_msec = 10 *1000;
+                }
+        } else if (ftdi->registered == 0) {
+                work_delay_in_msec = 10;
+                if (ftdi_elan_hcd_init(ftdi) == 0) {
+                        ftdi->registered = 1;
+                } else
+                        dev_err(&ftdi->udev->dev, "register failed\n");
+                work_delay_in_msec = 250;
+        } else {
+                if (ftdi_elan_checkingPCI(ftdi) == 0) {
+                        work_delay_in_msec = 250;
+                } else if (ftdi->controlreg & 0x00400000) {
+                        if (ftdi->gone_away > 0) {
+                                dev_err(&ftdi->udev->dev, "PCI device eject con"
+                                        "firmed platform_dev.dev.parent=%p plat"
+                                        "form_dev.dev=%p\n",
+                                        ftdi->platform_dev.dev.parent,
+                                        &ftdi->platform_dev.dev);
+                                platform_device_unregister(&ftdi->platform_dev);
+                                ftdi->platform_dev.dev.parent = NULL;
+                                ftdi->registered = 0;
+                                ftdi->enumerated = 0;
+                                ftdi->card_ejected = 0;
+                                ftdi->initialized = 0;
+                                ftdi->gone_away = 0;
+                        } else
+                                ftdi_elan_flush_targets(ftdi);
+                        work_delay_in_msec = 250;
+                } else {
+                        dev_err(&ftdi->udev->dev, "PCI device has disappeared\n"
+                                );
+                        ftdi_elan_cancel_targets(ftdi);
+                        work_delay_in_msec = 500;
+                        ftdi->enumerated = 0;
+                        ftdi->initialized = 0;
+                }
+        }
+        if (ftdi->disconnected > 0) {
+                ftdi_elan_put_kref(ftdi);
+                return;
+        } else {
+                ftdi_status_requeue_work(ftdi,
+                        msecs_to_jiffies(work_delay_in_msec));
+                return;
+        }
+}
+
+
+/*
+* file_operations for the jtag interface
+*
+* the usage count for the device is incremented on open()
+* and decremented on release()
+*/
+static int ftdi_elan_open(struct inode *inode, struct file *file)
+{
+        int subminor = iminor(inode);
+        struct usb_interface *interface = usb_find_interface(&ftdi_elan_driver,
+                subminor);
+        if (!interface) {
+                printk(KERN_ERR "can't find device for minor %d\n", subminor);
+                return -ENODEV;
+        } else {
+                struct usb_ftdi *ftdi = usb_get_intfdata(interface);
+                if (!ftdi) {
+                        return -ENODEV;
+                } else {
+                        if (down_interruptible(&ftdi->sw_lock)) {
+                                return -EINTR;
+                        } else {
+                                ftdi_elan_get_kref(ftdi);
+                                file->private_data = ftdi;
+                                return 0;
+                        }
+                }
+        }
+}
+
+static int ftdi_elan_release(struct inode *inode, struct file *file)
+{
+        struct usb_ftdi *ftdi = (struct usb_ftdi *)file->private_data;
+        if (ftdi == NULL)
+                return -ENODEV;
+        up(&ftdi->sw_lock);        /* decrement the count on our device */
+        ftdi_elan_put_kref(ftdi);
+        return 0;
+}
+
+
+#define FTDI_ELAN_IOC_MAGIC 0xA1
+#define FTDI_ELAN_IOCDEBUG _IOC(_IOC_WRITE, FTDI_ELAN_IOC_MAGIC, 1, 132)
+static int ftdi_elan_ioctl(struct inode *inode, struct file *file,
+        unsigned int cmd, unsigned long arg)
+{
+        switch (cmd) {
+        case FTDI_ELAN_IOCDEBUG:{
+                        char line[132];
+                        int size = strncpy_from_user(line,
+                                (const char __user *)arg, sizeof(line));
+                        if (size < 0) {
+                                return -EINVAL;
+                        } else {
+                                printk(KERN_ERR "TODO: ioctl %s\n", line);
+                                return 0;
+                        }
+                }
+        default:
+                return -EFAULT;
+        }
+}
+
+
+/*
+*
+* blocking bulk reads are used to get data from the device
+*
+*/
+static ssize_t ftdi_elan_read(struct file *file, char __user *buffer,
+			      size_t count, loff_t *ppos)
+{
+        char data[30 *3 + 4];
+        char *d = data;
+        int m = (sizeof(data) - 1) / 3;
+        int bytes_read = 0;
+        int retry_on_empty = 10;
+        int retry_on_timeout = 5;
+        struct usb_ftdi *ftdi = (struct usb_ftdi *)file->private_data;
+        if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        }
+        data[0] = 0;
+      have:if (ftdi->bulk_in_left > 0) {
+                if (count-- > 0) {
+                        char *p = ++ftdi->bulk_in_last + ftdi->bulk_in_buffer;
+                        ftdi->bulk_in_left -= 1;
+                        if (bytes_read < m) {
+                                d += sprintf(d, " %02X", 0x000000FF & *p);
+                        } else if (bytes_read > m) {
+                        } else
+                                d += sprintf(d, " ..");
+                        if (copy_to_user(buffer++, p, 1)) {
+                                return -EFAULT;
+                        } else {
+                                bytes_read += 1;
+                                goto have;
+                        }
+                } else
+                        return bytes_read;
+        }
+      more:if (count > 0) {
+                int packet_bytes = 0;
+                int retval = usb_bulk_msg(ftdi->udev,
+                        usb_rcvbulkpipe(ftdi->udev, ftdi->bulk_in_endpointAddr),
+                         ftdi->bulk_in_buffer, ftdi->bulk_in_size,
+                        &packet_bytes, msecs_to_jiffies(50));
+                if (packet_bytes > 2) {
+                        ftdi->bulk_in_left = packet_bytes - 2;
+                        ftdi->bulk_in_last = 1;
+                        goto have;
+                } else if (retval == -ETIMEDOUT) {
+                        if (retry_on_timeout-- > 0) {
+                                goto more;
+                        } else if (bytes_read > 0) {
+                                return bytes_read;
+                        } else
+                                return retval;
+                } else if (retval == 0) {
+                        if (retry_on_empty-- > 0) {
+                                goto more;
+                        } else
+                                return bytes_read;
+                } else
+                        return retval;
+        } else
+                return bytes_read;
+}
+
+static void ftdi_elan_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
+{
+        struct usb_ftdi *ftdi = (struct usb_ftdi *)urb->context;
+        if (urb->status && !(urb->status == -ENOENT || urb->status ==
+                -ECONNRESET || urb->status == -ESHUTDOWN)) {
+                dev_err(&ftdi->udev->dev, "urb=%p write bulk status received: %"
+                        "d\n", urb, urb->status);
+        }
+        usb_buffer_free(urb->dev, urb->transfer_buffer_length,
+                urb->transfer_buffer, urb->transfer_dma);
+}
+
+static int fill_buffer_with_all_queued_commands(struct usb_ftdi *ftdi,
+        char *buf, int command_size, int total_size)
+{
+        int ed_commands = 0;
+        int b = 0;
+        int I = command_size;
+        int i = ftdi->command_head;
+        while (I-- > 0) {
+                struct u132_command *command = &ftdi->command[COMMAND_MASK &
+                        i++];
+                int F = command->follows;
+                u8 *f = command->buffer;
+                if (command->header & 0x80) {
+                        ed_commands |= 1 << (0x3 & (command->header >> 5));
+                }
+                buf[b++] = command->header;
+                buf[b++] = (command->length >> 0) & 0x00FF;
+                buf[b++] = (command->length >> 8) & 0x00FF;
+                buf[b++] = command->address;
+                buf[b++] = command->width;
+                while (F-- > 0) {
+                        buf[b++] = *f++;
+                }
+        }
+        return ed_commands;
+}
+
+static int ftdi_elan_total_command_size(struct usb_ftdi *ftdi, int command_size)
+{
+        int total_size = 0;
+        int I = command_size;
+        int i = ftdi->command_head;
+        while (I-- > 0) {
+                struct u132_command *command = &ftdi->command[COMMAND_MASK &
+                        i++];
+                total_size += 5 + command->follows;
+        } return total_size;
+}
+
+static int ftdi_elan_command_engine(struct usb_ftdi *ftdi)
+{
+        int retval;
+        char *buf;
+        int ed_commands;
+        int total_size;
+        struct urb *urb;
+        int command_size = ftdi->command_next - ftdi->command_head;
+        if (command_size == 0)
+                return 0;
+        total_size = ftdi_elan_total_command_size(ftdi, command_size);
+        urb = usb_alloc_urb(0, GFP_KERNEL);
+        if (!urb) {
+                dev_err(&ftdi->udev->dev, "could not get a urb to write %d comm"
+                        "ands totaling %d bytes to the Uxxx\n", command_size,
+                        total_size);
+                return -ENOMEM;
+        }
+        buf = usb_buffer_alloc(ftdi->udev, total_size, GFP_KERNEL,
+                &urb->transfer_dma);
+        if (!buf) {
+                dev_err(&ftdi->udev->dev, "could not get a buffer to write %d c"
+                        "ommands totaling %d bytes to the Uxxx\n", command_size,
+                         total_size);
+                usb_free_urb(urb);
+                return -ENOMEM;
+        }
+        ed_commands = fill_buffer_with_all_queued_commands(ftdi, buf,
+                command_size, total_size);
+        usb_fill_bulk_urb(urb, ftdi->udev, usb_sndbulkpipe(ftdi->udev,
+                ftdi->bulk_out_endpointAddr), buf, total_size,
+                ftdi_elan_write_bulk_callback, ftdi);
+        urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+        if (ed_commands) {
+                char diag[40 *3 + 4];
+                char *d = diag;
+                int m = total_size;
+                u8 *c = buf;
+                int s = (sizeof(diag) - 1) / 3;
+                diag[0] = 0;
+                while (s-- > 0 && m-- > 0) {
+                        if (s > 0 || m == 0) {
+                                d += sprintf(d, " %02X", *c++);
+                        } else
+                                d += sprintf(d, " ..");
+                }
+        }
+        retval = usb_submit_urb(urb, GFP_KERNEL);
+        if (retval) {
+                dev_err(&ftdi->udev->dev, "failed %d to submit urb %p to write "
+                        "%d commands totaling %d bytes to the Uxxx\n", retval,
+                        urb, command_size, total_size);
+                usb_buffer_free(ftdi->udev, total_size, buf, urb->transfer_dma);
+                usb_free_urb(urb);
+                return retval;
+        }
+        usb_free_urb(urb);        /* release our reference to this urb,
+                the USB core will eventually free it entirely */
+        ftdi->command_head += command_size;
+        ftdi_elan_kick_respond_queue(ftdi);
+        return 0;
+}
+
+static void ftdi_elan_do_callback(struct usb_ftdi *ftdi,
+        struct u132_target *target, u8 *buffer, int length)
+{
+        struct urb *urb = target->urb;
+        int halted = target->halted;
+        int skipped = target->skipped;
+        int actual = target->actual;
+        int non_null = target->non_null;
+        int toggle_bits = target->toggle_bits;
+        int error_count = target->error_count;
+        int condition_code = target->condition_code;
+        int repeat_number = target->repeat_number;
+        void (*callback) (void *, struct urb *, u8 *, int, int, int, int, int,
+                int, int, int, int) = target->callback;
+        target->active -= 1;
+        target->callback = NULL;
+        (*callback) (target->endp, urb, buffer, length, toggle_bits,
+                error_count, condition_code, repeat_number, halted, skipped,
+                actual, non_null);
+}
+
+static char *have_ed_set_response(struct usb_ftdi *ftdi,
+        struct u132_target *target, u16 ed_length, int ed_number, int ed_type,
+        char *b)
+{
+        int payload = (ed_length >> 0) & 0x07FF;
+        down(&ftdi->u132_lock);
+        target->actual = 0;
+        target->non_null = (ed_length >> 15) & 0x0001;
+        target->repeat_number = (ed_length >> 11) & 0x000F;
+        if (ed_type == 0x02) {
+                if (payload == 0 || target->abandoning > 0) {
+                        target->abandoning = 0;
+                        up(&ftdi->u132_lock);
+                        ftdi_elan_do_callback(ftdi, target, 4 + ftdi->response,
+                                payload);
+                        ftdi->recieved = 0;
+                        ftdi->expected = 4;
+                        ftdi->ed_found = 0;
+                        return ftdi->response;
+                } else {
+                        ftdi->expected = 4 + payload;
+                        ftdi->ed_found = 1;
+                        up(&ftdi->u132_lock);
+                        return b;
+                }
+        } else if (ed_type == 0x03) {
+                if (payload == 0 || target->abandoning > 0) {
+                        target->abandoning = 0;
+                        up(&ftdi->u132_lock);
+                        ftdi_elan_do_callback(ftdi, target, 4 + ftdi->response,
+                                payload);
+                        ftdi->recieved = 0;
+                        ftdi->expected = 4;
+                        ftdi->ed_found = 0;
+                        return ftdi->response;
+                } else {
+                        ftdi->expected = 4 + payload;
+                        ftdi->ed_found = 1;
+                        up(&ftdi->u132_lock);
+                        return b;
+                }
+        } else if (ed_type == 0x01) {
+                target->abandoning = 0;
+                up(&ftdi->u132_lock);
+                ftdi_elan_do_callback(ftdi, target, 4 + ftdi->response,
+                        payload);
+                ftdi->recieved = 0;
+                ftdi->expected = 4;
+                ftdi->ed_found = 0;
+                return ftdi->response;
+        } else {
+                target->abandoning = 0;
+                up(&ftdi->u132_lock);
+                ftdi_elan_do_callback(ftdi, target, 4 + ftdi->response,
+                        payload);
+                ftdi->recieved = 0;
+                ftdi->expected = 4;
+                ftdi->ed_found = 0;
+                return ftdi->response;
+        }
+}
+
+static char *have_ed_get_response(struct usb_ftdi *ftdi,
+        struct u132_target *target, u16 ed_length, int ed_number, int ed_type,
+        char *b)
+{
+        down(&ftdi->u132_lock);
+        target->condition_code = TD_DEVNOTRESP;
+        target->actual = (ed_length >> 0) & 0x01FF;
+        target->non_null = (ed_length >> 15) & 0x0001;
+        target->repeat_number = (ed_length >> 11) & 0x000F;
+        up(&ftdi->u132_lock);
+        if (target->active)
+                ftdi_elan_do_callback(ftdi, target, NULL, 0);
+        target->abandoning = 0;
+        ftdi->recieved = 0;
+        ftdi->expected = 4;
+        ftdi->ed_found = 0;
+        return ftdi->response;
+}
+
+
+/*
+* The engine tries to empty the FTDI fifo
+*
+* all responses found in the fifo data are dispatched thus
+* the response buffer can only ever hold a maximum sized
+* response from the Uxxx.
+*
+*/
+static int ftdi_elan_respond_engine(struct usb_ftdi *ftdi)
+{
+        u8 *b = ftdi->response + ftdi->recieved;
+        int bytes_read = 0;
+        int retry_on_empty = 1;
+        int retry_on_timeout = 3;
+        int empty_packets = 0;
+      read:{
+                int packet_bytes = 0;
+                int retval = usb_bulk_msg(ftdi->udev,
+                        usb_rcvbulkpipe(ftdi->udev, ftdi->bulk_in_endpointAddr),
+                         ftdi->bulk_in_buffer, ftdi->bulk_in_size,
+                        &packet_bytes, msecs_to_jiffies(500));
+                char diag[30 *3 + 4];
+                char *d = diag;
+                int m = packet_bytes;
+                u8 *c = ftdi->bulk_in_buffer;
+                int s = (sizeof(diag) - 1) / 3;
+                diag[0] = 0;
+                while (s-- > 0 && m-- > 0) {
+                        if (s > 0 || m == 0) {
+                                d += sprintf(d, " %02X", *c++);
+                        } else
+                                d += sprintf(d, " ..");
+                }
+                if (packet_bytes > 2) {
+                        ftdi->bulk_in_left = packet_bytes - 2;
+                        ftdi->bulk_in_last = 1;
+                        goto have;
+                } else if (retval == -ETIMEDOUT) {
+                        if (retry_on_timeout-- > 0) {
+                                dev_err(&ftdi->udev->dev, "TIMED OUT with packe"
+                                        "t_bytes = %d with total %d bytes%s\n",
+                                        packet_bytes, bytes_read, diag);
+                                goto more;
+                        } else if (bytes_read > 0) {
+                                dev_err(&ftdi->udev->dev, "ONLY %d bytes%s\n",
+                                        bytes_read, diag);
+                                return -ENOMEM;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "TIMED OUT with packe"
+                                        "t_bytes = %d with total %d bytes%s\n",
+                                        packet_bytes, bytes_read, diag);
+                                return -ENOMEM;
+                        }
+                } else if (retval == -EILSEQ) {
+                        dev_err(&ftdi->udev->dev, "error = %d with packet_bytes"
+                                " = %d with total %d bytes%s\n", retval,
+                                packet_bytes, bytes_read, diag);
+                        return retval;
+                } else if (retval) {
+                        dev_err(&ftdi->udev->dev, "error = %d with packet_bytes"
+                                " = %d with total %d bytes%s\n", retval,
+                                packet_bytes, bytes_read, diag);
+                        return retval;
+                } else if (packet_bytes == 2) {
+                        unsigned char s0 = ftdi->bulk_in_buffer[0];
+                        unsigned char s1 = ftdi->bulk_in_buffer[1];
+                        empty_packets += 1;
+                        if (s0 == 0x31 && s1 == 0x60) {
+                                if (retry_on_empty-- > 0) {
+                                        goto more;
+                                } else
+                                        return 0;
+                        } else if (s0 == 0x31 && s1 == 0x00) {
+                                if (retry_on_empty-- > 0) {
+                                        goto more;
+                                } else
+                                        return 0;
+                        } else {
+                                if (retry_on_empty-- > 0) {
+                                        goto more;
+                                } else
+                                        return 0;
+                        }
+                } else if (packet_bytes == 1) {
+                        if (retry_on_empty-- > 0) {
+                                goto more;
+                        } else
+                                return 0;
+                } else {
+                        if (retry_on_empty-- > 0) {
+                                goto more;
+                        } else
+                                return 0;
+                }
+        }
+      more:{
+                goto read;
+        }
+      have:if (ftdi->bulk_in_left > 0) {
+                u8 c = ftdi->bulk_in_buffer[++ftdi->bulk_in_last];
+                bytes_read += 1;
+                ftdi->bulk_in_left -= 1;
+                if (ftdi->recieved == 0 && c == 0xFF) {
+                        goto have;
+                } else
+                        *b++ = c;
+                if (++ftdi->recieved < ftdi->expected) {
+                        goto have;
+                } else if (ftdi->ed_found) {
+                        int ed_number = (ftdi->response[0] >> 5) & 0x03;
+                        u16 ed_length = (ftdi->response[2] << 8) |
+                                ftdi->response[1];
+                        struct u132_target *target = &ftdi->target[ed_number];
+                        int payload = (ed_length >> 0) & 0x07FF;
+                        char diag[30 *3 + 4];
+                        char *d = diag;
+                        int m = payload;
+                        u8 *c = 4 + ftdi->response;
+                        int s = (sizeof(diag) - 1) / 3;
+                        diag[0] = 0;
+                        while (s-- > 0 && m-- > 0) {
+                                if (s > 0 || m == 0) {
+                                        d += sprintf(d, " %02X", *c++);
+                                } else
+                                        d += sprintf(d, " ..");
+                        }
+                        ftdi_elan_do_callback(ftdi, target, 4 + ftdi->response,
+                                payload);
+                        ftdi->recieved = 0;
+                        ftdi->expected = 4;
+                        ftdi->ed_found = 0;
+                        b = ftdi->response;
+                        goto have;
+                } else if (ftdi->expected == 8) {
+                        u8 buscmd;
+                        int respond_head = ftdi->respond_head++;
+                        struct u132_respond *respond = &ftdi->respond[
+                                RESPOND_MASK & respond_head];
+                        u32 data = ftdi->response[7];
+                        data <<= 8;
+                        data |= ftdi->response[6];
+                        data <<= 8;
+                        data |= ftdi->response[5];
+                        data <<= 8;
+                        data |= ftdi->response[4];
+                        *respond->value = data;
+                        *respond->result = 0;
+                        complete(&respond->wait_completion);
+                        ftdi->recieved = 0;
+                        ftdi->expected = 4;
+                        ftdi->ed_found = 0;
+                        b = ftdi->response;
+                        buscmd = (ftdi->response[0] >> 0) & 0x0F;
+                        if (buscmd == 0x00) {
+                        } else if (buscmd == 0x02) {
+                        } else if (buscmd == 0x06) {
+                        } else if (buscmd == 0x0A) {
+                        } else
+                                dev_err(&ftdi->udev->dev, "Uxxx unknown(%0X) va"
+                                        "lue = %08X\n", buscmd, data);
+                        goto have;
+                } else {
+                        if ((ftdi->response[0] & 0x80) == 0x00) {
+                                ftdi->expected = 8;
+                                goto have;
+                        } else {
+                                int ed_number = (ftdi->response[0] >> 5) & 0x03;
+                                int ed_type = (ftdi->response[0] >> 0) & 0x03;
+                                u16 ed_length = (ftdi->response[2] << 8) |
+                                        ftdi->response[1];
+                                struct u132_target *target = &ftdi->target[
+                                        ed_number];
+                                target->halted = (ftdi->response[0] >> 3) &
+                                        0x01;
+                                target->skipped = (ftdi->response[0] >> 2) &
+                                        0x01;
+                                target->toggle_bits = (ftdi->response[3] >> 6)
+                                        & 0x03;
+                                target->error_count = (ftdi->response[3] >> 4)
+                                        & 0x03;
+                                target->condition_code = (ftdi->response[
+                                        3] >> 0) & 0x0F;
+                                if ((ftdi->response[0] & 0x10) == 0x00) {
+                                        b = have_ed_set_response(ftdi, target,
+                                                ed_length, ed_number, ed_type,
+                                                b);
+                                        goto have;
+                                } else {
+                                        b = have_ed_get_response(ftdi, target,
+                                                ed_length, ed_number, ed_type,
+                                                b);
+                                        goto have;
+                                }
+                        }
+                }
+        } else
+                goto more;
+}
+
+
+/*
+* create a urb, and a buffer for it, and copy the data to the urb
+*
+*/
+static ssize_t ftdi_elan_write(struct file *file,
+			       const char __user *user_buffer, size_t count,
+			       loff_t *ppos)
+{
+        int retval = 0;
+        struct urb *urb;
+        char *buf;
+        char data[30 *3 + 4];
+        char *d = data;
+        const char __user *s = user_buffer;
+        int m = (sizeof(data) - 1) / 3;
+        struct usb_ftdi *ftdi = (struct usb_ftdi *)file->private_data;
+        if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        }
+        if (count == 0) {
+                goto exit;
+        }
+        urb = usb_alloc_urb(0, GFP_KERNEL);
+        if (!urb) {
+                retval = -ENOMEM;
+                goto error_1;
+        }
+        buf = usb_buffer_alloc(ftdi->udev, count, GFP_KERNEL,
+                &urb->transfer_dma);
+        if (!buf) {
+                retval = -ENOMEM;
+                goto error_2;
+        }
+        if (copy_from_user(buf, user_buffer, count)) {
+                retval = -EFAULT;
+                goto error_3;
+        }
+        usb_fill_bulk_urb(urb, ftdi->udev, usb_sndbulkpipe(ftdi->udev,
+                ftdi->bulk_out_endpointAddr), buf, count,
+                ftdi_elan_write_bulk_callback, ftdi);
+        urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+        retval = usb_submit_urb(urb, GFP_KERNEL);
+        if (retval) {
+                dev_err(&ftdi->udev->dev, "failed submitting write urb, error %"
+                        "d\n", retval);
+                goto error_4;
+        }
+        usb_free_urb(urb);
+      exit:;
+        if (count > m) {
+                int I = m - 1;
+                while (I-- > 0) {
+                        d += sprintf(d, " %02X", 0x000000FF & *s++);
+                }
+                d += sprintf(d, " ..");
+        } else {
+                int I = count;
+                while (I-- > 0) {
+                        d += sprintf(d, " %02X", 0x000000FF & *s++);
+                }
+        }
+        return count;
+      error_4: error_3:usb_buffer_free(ftdi->udev, count, buf,
+              urb->transfer_dma);
+      error_2:usb_free_urb(urb);
+      error_1:return retval;
+}
+
+static struct file_operations ftdi_elan_fops = {
+        .owner = THIS_MODULE,
+        .llseek = no_llseek,
+        .ioctl = ftdi_elan_ioctl,
+        .read = ftdi_elan_read,
+        .write = ftdi_elan_write,
+        .open = ftdi_elan_open,
+        .release = ftdi_elan_release,
+};
+
+/*
+* usb class driver info in order to get a minor number from the usb core,
+* and to have the device registered with the driver core
+*/
+static struct usb_class_driver ftdi_elan_jtag_class = {
+        .name = "ftdi-%d-jtag",
+        .fops = &ftdi_elan_fops,
+        .minor_base = USB_FTDI_ELAN_MINOR_BASE,
+};
+
+/*
+* the following definitions are for the
+* ELAN FPGA state machgine processor that
+* lies on the other side of the FTDI chip
+*/
+#define cPCIu132rd 0x0
+#define cPCIu132wr 0x1
+#define cPCIiord 0x2
+#define cPCIiowr 0x3
+#define cPCImemrd 0x6
+#define cPCImemwr 0x7
+#define cPCIcfgrd 0xA
+#define cPCIcfgwr 0xB
+#define cPCInull 0xF
+#define cU132cmd_status 0x0
+#define cU132flash 0x1
+#define cPIDsetup 0x0
+#define cPIDout 0x1
+#define cPIDin 0x2
+#define cPIDinonce 0x3
+#define cCCnoerror 0x0
+#define cCCcrc 0x1
+#define cCCbitstuff 0x2
+#define cCCtoggle 0x3
+#define cCCstall 0x4
+#define cCCnoresp 0x5
+#define cCCbadpid1 0x6
+#define cCCbadpid2 0x7
+#define cCCdataoverrun 0x8
+#define cCCdataunderrun 0x9
+#define cCCbuffoverrun 0xC
+#define cCCbuffunderrun 0xD
+#define cCCnotaccessed 0xF
+static int ftdi_elan_write_reg(struct usb_ftdi *ftdi, u32 data)
+{
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x00 | cPCIu132wr;
+                        command->length = 0x04;
+                        command->address = 0x00;
+                        command->width = 0x00;
+                        command->follows = 4;
+                        command->value = data;
+                        command->buffer = &command->value;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+static int ftdi_elan_write_config(struct usb_ftdi *ftdi, int config_offset,
+        u8 width, u32 data)
+{
+        u8 addressofs = config_offset / 4;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x00 | (cPCIcfgwr & 0x0F);
+                        command->length = 0x04;
+                        command->address = addressofs;
+                        command->width = 0x00 | (width & 0x0F);
+                        command->follows = 4;
+                        command->value = data;
+                        command->buffer = &command->value;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+static int ftdi_elan_write_pcimem(struct usb_ftdi *ftdi, int mem_offset,
+        u8 width, u32 data)
+{
+        u8 addressofs = mem_offset / 4;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x00 | (cPCImemwr & 0x0F);
+                        command->length = 0x04;
+                        command->address = addressofs;
+                        command->width = 0x00 | (width & 0x0F);
+                        command->follows = 4;
+                        command->value = data;
+                        command->buffer = &command->value;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_write_pcimem(struct platform_device *pdev, int mem_offset,
+        u8 width, u32 data)
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_write_pcimem(ftdi, mem_offset, width, data);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_write_pcimem);
+static int ftdi_elan_read_reg(struct usb_ftdi *ftdi, u32 *data)
+{
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                int respond_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                respond_size = ftdi->respond_next - ftdi->respond_head;
+                if (command_size < COMMAND_SIZE && respond_size < RESPOND_SIZE)
+                        {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        struct u132_respond *respond = &ftdi->respond[
+                                RESPOND_MASK & ftdi->respond_next];
+                        int result = -ENODEV;
+                        respond->result = &result;
+                        respond->header = command->header = 0x00 | cPCIu132rd;
+                        command->length = 0x04;
+                        respond->address = command->address = cU132cmd_status;
+                        command->width = 0x00;
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        respond->value = data;
+                        init_completion(&respond->wait_completion);
+                        ftdi->command_next += 1;
+                        ftdi->respond_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        wait_for_completion(&respond->wait_completion);
+                        return result;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_read_reg(struct platform_device *pdev, u32 *data)
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_read_reg(ftdi, data);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_read_reg);
+static int ftdi_elan_read_config(struct usb_ftdi *ftdi, int config_offset,
+        u8 width, u32 *data)
+{
+        u8 addressofs = config_offset / 4;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                int respond_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                respond_size = ftdi->respond_next - ftdi->respond_head;
+                if (command_size < COMMAND_SIZE && respond_size < RESPOND_SIZE)
+                        {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        struct u132_respond *respond = &ftdi->respond[
+                                RESPOND_MASK & ftdi->respond_next];
+                        int result = -ENODEV;
+                        respond->result = &result;
+                        respond->header = command->header = 0x00 | (cPCIcfgrd &
+                                0x0F);
+                        command->length = 0x04;
+                        respond->address = command->address = addressofs;
+                        command->width = 0x00 | (width & 0x0F);
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        respond->value = data;
+                        init_completion(&respond->wait_completion);
+                        ftdi->command_next += 1;
+                        ftdi->respond_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        wait_for_completion(&respond->wait_completion);
+                        return result;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+static int ftdi_elan_read_pcimem(struct usb_ftdi *ftdi, int mem_offset,
+        u8 width, u32 *data)
+{
+        u8 addressofs = mem_offset / 4;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                int respond_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                respond_size = ftdi->respond_next - ftdi->respond_head;
+                if (command_size < COMMAND_SIZE && respond_size < RESPOND_SIZE)
+                        {
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        struct u132_respond *respond = &ftdi->respond[
+                                RESPOND_MASK & ftdi->respond_next];
+                        int result = -ENODEV;
+                        respond->result = &result;
+                        respond->header = command->header = 0x00 | (cPCImemrd &
+                                0x0F);
+                        command->length = 0x04;
+                        respond->address = command->address = addressofs;
+                        command->width = 0x00 | (width & 0x0F);
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        respond->value = data;
+                        init_completion(&respond->wait_completion);
+                        ftdi->command_next += 1;
+                        ftdi->respond_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        wait_for_completion(&respond->wait_completion);
+                        return result;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_read_pcimem(struct platform_device *pdev, int mem_offset,
+        u8 width, u32 *data)
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else
+                return ftdi_elan_read_pcimem(ftdi, mem_offset, width, data);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_read_pcimem);
+static int ftdi_elan_edset_setup(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        u8 ed = ed_number - 1;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_target *target = &ftdi->target[ed];
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x80 | (ed << 5);
+                        command->length = 0x8007;
+                        command->address = (toggle_bits << 6) | (ep_number << 2)
+                                | (address << 0);
+                        command->width = usb_maxpacket(urb->dev, urb->pipe,
+                                usb_pipeout(urb->pipe));
+                        command->follows = 8;
+                        command->value = 0;
+                        command->buffer = urb->setup_packet;
+                        target->callback = callback;
+                        target->endp = endp;
+                        target->urb = urb;
+                        target->active = 1;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_setup(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_setup(ftdi, ed_number, endp, urb, address,
+                ep_number, toggle_bits, callback);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_setup);
+static int ftdi_elan_edset_input(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        u8 ed = ed_number - 1;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_target *target = &ftdi->target[ed];
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        int remaining_length = urb->transfer_buffer_length -
+                                urb->actual_length;
+                        command->header = 0x82 | (ed << 5);
+                        if (remaining_length == 0) {
+                                command->length = 0x0000;
+                        } else if (remaining_length > 1024) {
+                                command->length = 0x8000 | 1023;
+                        } else
+                                command->length = 0x8000 | (remaining_length -
+                                        1);
+                        command->address = (toggle_bits << 6) | (ep_number << 2)
+                                | (address << 0);
+                        command->width = usb_maxpacket(urb->dev, urb->pipe,
+                                usb_pipeout(urb->pipe));
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        target->callback = callback;
+                        target->endp = endp;
+                        target->urb = urb;
+                        target->active = 1;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_input(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_input(ftdi, ed_number, endp, urb, address,
+                ep_number, toggle_bits, callback);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_input);
+static int ftdi_elan_edset_empty(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        u8 ed = ed_number - 1;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        struct u132_target *target = &ftdi->target[ed];
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x81 | (ed << 5);
+                        command->length = 0x0000;
+                        command->address = (toggle_bits << 6) | (ep_number << 2)
+                                | (address << 0);
+                        command->width = usb_maxpacket(urb->dev, urb->pipe,
+                                usb_pipeout(urb->pipe));
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        target->callback = callback;
+                        target->endp = endp;
+                        target->urb = urb;
+                        target->active = 1;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_empty(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_empty(ftdi, ed_number, endp, urb, address,
+                ep_number, toggle_bits, callback);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_empty);
+static int ftdi_elan_edset_output(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        u8 ed = ed_number - 1;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        u8 *b;
+                        u16 urb_size;
+                        int i = 0;
+                        char data[30 *3 + 4];
+                        char *d = data;
+                        int m = (sizeof(data) - 1) / 3;
+                        int l = 0;
+                        struct u132_target *target = &ftdi->target[ed];
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x81 | (ed << 5);
+                        command->address = (toggle_bits << 6) | (ep_number << 2)
+                                | (address << 0);
+                        command->width = usb_maxpacket(urb->dev, urb->pipe,
+                                usb_pipeout(urb->pipe));
+                        command->follows = min(1024,
+                                urb->transfer_buffer_length -
+                                urb->actual_length);
+                        command->value = 0;
+                        command->buffer = urb->transfer_buffer +
+                                urb->actual_length;
+                        command->length = 0x8000 | (command->follows - 1);
+                        b = command->buffer;
+                        urb_size = command->follows;
+                        data[0] = 0;
+                        while (urb_size-- > 0) {
+                                if (i > m) {
+                                } else if (i++ < m) {
+                                        int w = sprintf(d, " %02X", *b++);
+                                        d += w;
+                                        l += w;
+                                } else
+                                        d += sprintf(d, " ..");
+                        }
+                        target->callback = callback;
+                        target->endp = endp;
+                        target->urb = urb;
+                        target->active = 1;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_output(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_output(ftdi, ed_number, endp, urb, address,
+                ep_number, toggle_bits, callback);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_output);
+static int ftdi_elan_edset_single(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        u8 ed = ed_number - 1;
+      wait:if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                int command_size;
+                down(&ftdi->u132_lock);
+                command_size = ftdi->command_next - ftdi->command_head;
+                if (command_size < COMMAND_SIZE) {
+                        int remaining_length = urb->transfer_buffer_length -
+                                urb->actual_length;
+                        struct u132_target *target = &ftdi->target[ed];
+                        struct u132_command *command = &ftdi->command[
+                                COMMAND_MASK & ftdi->command_next];
+                        command->header = 0x83 | (ed << 5);
+                        if (remaining_length == 0) {
+                                command->length = 0x0000;
+                        } else if (remaining_length > 1024) {
+                                command->length = 0x8000 | 1023;
+                        } else
+                                command->length = 0x8000 | (remaining_length -
+                                        1);
+                        command->address = (toggle_bits << 6) | (ep_number << 2)
+                                | (address << 0);
+                        command->width = usb_maxpacket(urb->dev, urb->pipe,
+                                usb_pipeout(urb->pipe));
+                        command->follows = 0;
+                        command->value = 0;
+                        command->buffer = NULL;
+                        target->callback = callback;
+                        target->endp = endp;
+                        target->urb = urb;
+                        target->active = 1;
+                        ftdi->command_next += 1;
+                        ftdi_elan_kick_command_queue(ftdi);
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        up(&ftdi->u132_lock);
+                        msleep(100);
+                        goto wait;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_single(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_single(ftdi, ed_number, endp, urb, address,
+                ep_number, toggle_bits, callback);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_single);
+static int ftdi_elan_edset_flush(struct usb_ftdi *ftdi, u8 ed_number,
+        void *endp)
+{
+        u8 ed = ed_number - 1;
+        if (ftdi->disconnected > 0) {
+                return -ENODEV;
+        } else if (ftdi->initialized == 0) {
+                return -ENODEV;
+        } else {
+                struct u132_target *target = &ftdi->target[ed];
+                down(&ftdi->u132_lock);
+                if (target->abandoning > 0) {
+                        up(&ftdi->u132_lock);
+                        return 0;
+                } else {
+                        target->abandoning = 1;
+                      wait_1:if (target->active == 1) {
+                                int command_size = ftdi->command_next -
+                                        ftdi->command_head;
+                                if (command_size < COMMAND_SIZE) {
+                                        struct u132_command *command =
+                                                &ftdi->command[COMMAND_MASK &
+                                                ftdi->command_next];
+                                        command->header = 0x80 | (ed << 5) |
+                                                0x4;
+                                        command->length = 0x00;
+                                        command->address = 0x00;
+                                        command->width = 0x00;
+                                        command->follows = 0;
+                                        command->value = 0;
+                                        command->buffer = &command->value;
+                                        ftdi->command_next += 1;
+                                        ftdi_elan_kick_command_queue(ftdi);
+                                } else {
+                                        up(&ftdi->u132_lock);
+                                        msleep(100);
+                                        down(&ftdi->u132_lock);
+                                        goto wait_1;
+                                }
+                        }
+                        up(&ftdi->u132_lock);
+                        return 0;
+                }
+        }
+}
+
+int usb_ftdi_elan_edset_flush(struct platform_device *pdev, u8 ed_number,
+        void *endp)
+{
+        struct usb_ftdi *ftdi = platform_device_to_usb_ftdi(pdev);
+        return ftdi_elan_edset_flush(ftdi, ed_number, endp);
+}
+
+
+EXPORT_SYMBOL_GPL(usb_ftdi_elan_edset_flush);
+static int ftdi_elan_flush_input_fifo(struct usb_ftdi *ftdi)
+{
+        int retry_on_empty = 10;
+        int retry_on_timeout = 5;
+        int retry_on_status = 20;
+      more:{
+                int packet_bytes = 0;
+                int retval = usb_bulk_msg(ftdi->udev,
+                        usb_rcvbulkpipe(ftdi->udev, ftdi->bulk_in_endpointAddr),
+                         ftdi->bulk_in_buffer, ftdi->bulk_in_size,
+                        &packet_bytes, msecs_to_jiffies(100));
+                if (packet_bytes > 2) {
+                        char diag[30 *3 + 4];
+                        char *d = diag;
+                        int m = (sizeof(diag) - 1) / 3;
+                        char *b = ftdi->bulk_in_buffer;
+                        int bytes_read = 0;
+                        diag[0] = 0;
+                        while (packet_bytes-- > 0) {
+                                char c = *b++;
+                                if (bytes_read < m) {
+                                        d += sprintf(d, " %02X",
+                                                0x000000FF & c);
+                                } else if (bytes_read > m) {
+                                } else
+                                        d += sprintf(d, " ..");
+                                bytes_read += 1;
+                                continue;
+                        }
+                        goto more;
+                } else if (packet_bytes > 1) {
+                        char s1 = ftdi->bulk_in_buffer[0];
+                        char s2 = ftdi->bulk_in_buffer[1];
+                        if (s1 == 0x31 && s2 == 0x60) {
+                                return 0;
+                        } else if (retry_on_status-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "STATUS ERROR retry l"
+                                        "imit reached\n");
+                                return -EFAULT;
+                        }
+                } else if (packet_bytes > 0) {
+                        char b1 = ftdi->bulk_in_buffer[0];
+                        dev_err(&ftdi->udev->dev, "only one byte flushed from F"
+                                "TDI = %02X\n", b1);
+                        if (retry_on_status-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "STATUS ERROR retry l"
+                                        "imit reached\n");
+                                return -EFAULT;
+                        }
+                } else if (retval == -ETIMEDOUT) {
+                        if (retry_on_timeout-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "TIMED OUT retry limi"
+                                        "t reached\n");
+                                return -ENOMEM;
+                        }
+                } else if (retval == 0) {
+                        if (retry_on_empty-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "empty packet retry l"
+                                        "imit reached\n");
+                                return -ENOMEM;
+                        }
+                } else {
+                        dev_err(&ftdi->udev->dev, "error = %d\n", retval);
+                        return retval;
+                }
+        }
+        return -1;
+}
+
+
+/*
+* send the long flush sequence
+*
+*/
+static int ftdi_elan_synchronize_flush(struct usb_ftdi *ftdi)
+{
+        int retval;
+        struct urb *urb;
+        char *buf;
+        int I = 257;
+        int i = 0;
+        urb = usb_alloc_urb(0, GFP_KERNEL);
+        if (!urb) {
+                dev_err(&ftdi->udev->dev, "could not alloc a urb for flush sequ"
+                        "ence\n");
+                return -ENOMEM;
+        }
+        buf = usb_buffer_alloc(ftdi->udev, I, GFP_KERNEL, &urb->transfer_dma);
+        if (!buf) {
+                dev_err(&ftdi->udev->dev, "could not get a buffer for flush seq"
+                        "uence\n");
+                usb_free_urb(urb);
+                return -ENOMEM;
+        }
+        while (I-- > 0)
+                buf[i++] = 0x55;
+        usb_fill_bulk_urb(urb, ftdi->udev, usb_sndbulkpipe(ftdi->udev,
+                ftdi->bulk_out_endpointAddr), buf, i,
+                ftdi_elan_write_bulk_callback, ftdi);
+        urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+        retval = usb_submit_urb(urb, GFP_KERNEL);
+        if (retval) {
+                dev_err(&ftdi->udev->dev, "failed to submit urb containing the "
+                        "flush sequence\n");
+                usb_buffer_free(ftdi->udev, i, buf, urb->transfer_dma);
+                usb_free_urb(urb);
+                return -ENOMEM;
+        }
+        usb_free_urb(urb);
+        return 0;
+}
+
+
+/*
+* send the reset sequence
+*
+*/
+static int ftdi_elan_synchronize_reset(struct usb_ftdi *ftdi)
+{
+        int retval;
+        struct urb *urb;
+        char *buf;
+        int I = 4;
+        int i = 0;
+        urb = usb_alloc_urb(0, GFP_KERNEL);
+        if (!urb) {
+                dev_err(&ftdi->udev->dev, "could not get a urb for the reset se"
+                        "quence\n");
+                return -ENOMEM;
+        }
+        buf = usb_buffer_alloc(ftdi->udev, I, GFP_KERNEL, &urb->transfer_dma);
+        if (!buf) {
+                dev_err(&ftdi->udev->dev, "could not get a buffer for the reset"
+                        " sequence\n");
+                usb_free_urb(urb);
+                return -ENOMEM;
+        }
+        buf[i++] = 0x55;
+        buf[i++] = 0xAA;
+        buf[i++] = 0x5A;
+        buf[i++] = 0xA5;
+        usb_fill_bulk_urb(urb, ftdi->udev, usb_sndbulkpipe(ftdi->udev,
+                ftdi->bulk_out_endpointAddr), buf, i,
+                ftdi_elan_write_bulk_callback, ftdi);
+        urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+        retval = usb_submit_urb(urb, GFP_KERNEL);
+        if (retval) {
+                dev_err(&ftdi->udev->dev, "failed to submit urb containing the "
+                        "reset sequence\n");
+                usb_buffer_free(ftdi->udev, i, buf, urb->transfer_dma);
+                usb_free_urb(urb);
+                return -ENOMEM;
+        }
+        usb_free_urb(urb);
+        return 0;
+}
+
+static int ftdi_elan_synchronize(struct usb_ftdi *ftdi)
+{
+        int retval;
+        int long_stop = 10;
+        int retry_on_timeout = 5;
+        int retry_on_empty = 10;
+        int err_count = 0;
+        retval = ftdi_elan_flush_input_fifo(ftdi);
+        if (retval)
+                return retval;
+        ftdi->bulk_in_left = 0;
+        ftdi->bulk_in_last = -1;
+        while (long_stop-- > 0) {
+                int read_stop;
+                int read_stuck;
+                retval = ftdi_elan_synchronize_flush(ftdi);
+                if (retval)
+                        return retval;
+                retval = ftdi_elan_flush_input_fifo(ftdi);
+                if (retval)
+                        return retval;
+              reset:retval = ftdi_elan_synchronize_reset(ftdi);
+                if (retval)
+                        return retval;
+                read_stop = 100;
+                read_stuck = 10;
+              read:{
+                        int packet_bytes = 0;
+                        retval = usb_bulk_msg(ftdi->udev,
+                                usb_rcvbulkpipe(ftdi->udev,
+                                ftdi->bulk_in_endpointAddr),
+                                ftdi->bulk_in_buffer, ftdi->bulk_in_size,
+                                &packet_bytes, msecs_to_jiffies(500));
+                        if (packet_bytes > 2) {
+                                char diag[30 *3 + 4];
+                                char *d = diag;
+                                int m = (sizeof(diag) - 1) / 3;
+                                char *b = ftdi->bulk_in_buffer;
+                                int bytes_read = 0;
+                                unsigned char c = 0;
+                                diag[0] = 0;
+                                while (packet_bytes-- > 0) {
+                                        c = *b++;
+                                        if (bytes_read < m) {
+                                                d += sprintf(d, " %02X", c);
+                                        } else if (bytes_read > m) {
+                                        } else
+                                                d += sprintf(d, " ..");
+                                        bytes_read += 1;
+                                        continue;
+                                }
+                                if (c == 0x7E) {
+                                        return 0;
+                                } else {
+                                        if (c == 0x55) {
+                                                goto read;
+                                        } else if (read_stop-- > 0) {
+                                                goto read;
+                                        } else {
+                                                dev_err(&ftdi->udev->dev, "retr"
+                                                        "y limit reached\n");
+                                                continue;
+                                        }
+                                }
+                        } else if (packet_bytes > 1) {
+                                unsigned char s1 = ftdi->bulk_in_buffer[0];
+                                unsigned char s2 = ftdi->bulk_in_buffer[1];
+                                if (s1 == 0x31 && s2 == 0x00) {
+                                        if (read_stuck-- > 0) {
+                                                goto read;
+                                        } else
+                                                goto reset;
+                                } else if (s1 == 0x31 && s2 == 0x60) {
+                                        if (read_stop-- > 0) {
+                                                goto read;
+                                        } else {
+                                                dev_err(&ftdi->udev->dev, "retr"
+                                                        "y limit reached\n");
+                                                continue;
+                                        }
+                                } else {
+                                        if (read_stop-- > 0) {
+                                                goto read;
+                                        } else {
+                                                dev_err(&ftdi->udev->dev, "retr"
+                                                        "y limit reached\n");
+                                                continue;
+                                        }
+                                }
+                        } else if (packet_bytes > 0) {
+                                if (read_stop-- > 0) {
+                                        goto read;
+                                } else {
+                                        dev_err(&ftdi->udev->dev, "retry limit "
+                                                "reached\n");
+                                        continue;
+                                }
+                        } else if (retval == -ETIMEDOUT) {
+                                if (retry_on_timeout-- > 0) {
+                                        goto read;
+                                } else {
+                                        dev_err(&ftdi->udev->dev, "TIMED OUT re"
+                                                "try limit reached\n");
+                                        continue;
+                                }
+                        } else if (retval == 0) {
+                                if (retry_on_empty-- > 0) {
+                                        goto read;
+                                } else {
+                                        dev_err(&ftdi->udev->dev, "empty packet"
+                                                " retry limit reached\n");
+                                        continue;
+                                }
+                        } else {
+                                err_count += 1;
+                                dev_err(&ftdi->udev->dev, "error = %d\n",
+                                        retval);
+                                if (read_stop-- > 0) {
+                                        goto read;
+                                } else {
+                                        dev_err(&ftdi->udev->dev, "retry limit "
+                                                "reached\n");
+                                        continue;
+                                }
+                        }
+                }
+        }
+        dev_err(&ftdi->udev->dev, "failed to synchronize\n");
+        return -EFAULT;
+}
+
+static int ftdi_elan_stuck_waiting(struct usb_ftdi *ftdi)
+{
+        int retry_on_empty = 10;
+        int retry_on_timeout = 5;
+        int retry_on_status = 50;
+      more:{
+                int packet_bytes = 0;
+                int retval = usb_bulk_msg(ftdi->udev,
+                        usb_rcvbulkpipe(ftdi->udev, ftdi->bulk_in_endpointAddr),
+                         ftdi->bulk_in_buffer, ftdi->bulk_in_size,
+                        &packet_bytes, msecs_to_jiffies(1000));
+                if (packet_bytes > 2) {
+                        char diag[30 *3 + 4];
+                        char *d = diag;
+                        int m = (sizeof(diag) - 1) / 3;
+                        char *b = ftdi->bulk_in_buffer;
+                        int bytes_read = 0;
+                        diag[0] = 0;
+                        while (packet_bytes-- > 0) {
+                                char c = *b++;
+                                if (bytes_read < m) {
+                                        d += sprintf(d, " %02X",
+                                                0x000000FF & c);
+                                } else if (bytes_read > m) {
+                                } else
+                                        d += sprintf(d, " ..");
+                                bytes_read += 1;
+                                continue;
+                        }
+                        goto more;
+                } else if (packet_bytes > 1) {
+                        char s1 = ftdi->bulk_in_buffer[0];
+                        char s2 = ftdi->bulk_in_buffer[1];
+                        if (s1 == 0x31 && s2 == 0x60) {
+                                return 0;
+                        } else if (retry_on_status-- > 0) {
+                                msleep(5);
+                                goto more;
+                        } else
+                                return -EFAULT;
+                } else if (packet_bytes > 0) {
+                        char b1 = ftdi->bulk_in_buffer[0];
+                        dev_err(&ftdi->udev->dev, "only one byte flushed from F"
+                                "TDI = %02X\n", b1);
+                        if (retry_on_status-- > 0) {
+                                msleep(5);
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "STATUS ERROR retry l"
+                                        "imit reached\n");
+                                return -EFAULT;
+                        }
+                } else if (retval == -ETIMEDOUT) {
+                        if (retry_on_timeout-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "TIMED OUT retry limi"
+                                        "t reached\n");
+                                return -ENOMEM;
+                        }
+                } else if (retval == 0) {
+                        if (retry_on_empty-- > 0) {
+                                goto more;
+                        } else {
+                                dev_err(&ftdi->udev->dev, "empty packet retry l"
+                                        "imit reached\n");
+                                return -ENOMEM;
+                        }
+                } else {
+                        dev_err(&ftdi->udev->dev, "error = %d\n", retval);
+                        return -ENOMEM;
+                }
+        }
+        return -1;
+}
+
+static int ftdi_elan_checkingPCI(struct usb_ftdi *ftdi)
+{
+        int UxxxStatus = ftdi_elan_read_reg(ftdi, &ftdi->controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        if (ftdi->controlreg & 0x00400000) {
+                if (ftdi->card_ejected) {
+                } else {
+                        ftdi->card_ejected = 1;
+                        dev_err(&ftdi->udev->dev, "CARD EJECTED - controlreg = "
+                                "%08X\n", ftdi->controlreg);
+                }
+                return -ENODEV;
+        } else {
+                u8 fn = ftdi->function - 1;
+                int activePCIfn = fn << 8;
+                u32 pcidata;
+                u32 pciVID;
+                u32 pciPID;
+                int reg = 0;
+                UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                        &pcidata);
+                if (UxxxStatus)
+                        return UxxxStatus;
+                pciVID = pcidata & 0xFFFF;
+                pciPID = (pcidata >> 16) & 0xFFFF;
+                if (pciVID == ftdi->platform_data.vendor && pciPID ==
+                        ftdi->platform_data.device) {
+                        return 0;
+                } else {
+                        dev_err(&ftdi->udev->dev, "vendor=%04X pciVID=%04X devi"
+                                "ce=%04X pciPID=%04X\n",
+                                ftdi->platform_data.vendor, pciVID,
+                                ftdi->platform_data.device, pciPID);
+                        return -ENODEV;
+                }
+        }
+}
+
+static int ftdi_elan_enumeratePCI(struct usb_ftdi *ftdi)
+{
+        u32 latence_timer;
+        u32 controlreg;
+        int UxxxStatus;
+        u32 pcidata;
+        int reg = 0;
+        int foundOHCI = 0;
+        u8 fn;
+        int activePCIfn = 0;
+        u32 pciVID = 0;
+        u32 pciPID = 0;
+        UxxxStatus = ftdi_elan_read_reg(ftdi, &controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x00000000L);
+        if (UxxxStatus)
+                return UxxxStatus;
+        msleep(750);
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x00000200L | 0x100);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x00000200L | 0x500);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_reg(ftdi, &controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x0000020CL | 0x000);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x0000020DL | 0x000);
+        if (UxxxStatus)
+                return UxxxStatus;
+        msleep(250);
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x0000020FL | 0x000);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_reg(ftdi, &controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x0000025FL | 0x800);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_reg(ftdi, &controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_reg(ftdi, &controlreg);
+        if (UxxxStatus)
+                return UxxxStatus;
+        msleep(1000);
+        for (fn = 0; (fn < 4) && (!foundOHCI); fn++) {
+                activePCIfn = fn << 8;
+                ftdi->function = fn + 1;
+                UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                        &pcidata);
+                if (UxxxStatus)
+                        return UxxxStatus;
+                pciVID = pcidata & 0xFFFF;
+                pciPID = (pcidata >> 16) & 0xFFFF;
+                if ((pciVID == 0x1045) && (pciPID == 0xc861)) {
+                        foundOHCI = 1;
+                } else if ((pciVID == 0x1033) && (pciPID == 0x0035)) {
+                        foundOHCI = 1;
+                } else if ((pciVID == 0x10b9) && (pciPID == 0x5237)) {
+                        foundOHCI = 1;
+                } else if ((pciVID == 0x11c1) && (pciPID == 0x5802)) {
+                        foundOHCI = 1;
+                } else if ((pciVID == 0x11AB) && (pciPID == 0x1FA6)) {
+                }
+        }
+        if (foundOHCI == 0) {
+                return -ENXIO;
+        }
+        ftdi->platform_data.vendor = pciVID;
+        ftdi->platform_data.device = pciPID;
+        UxxxStatus = ftdi_elan_write_reg(ftdi, 0x0000025FL | 0x2800);
+        if (UxxxStatus)
+                return UxxxStatus;
+        reg = 16;
+        UxxxStatus = ftdi_elan_write_config(ftdi, activePCIfn | reg, 0,
+                0xFFFFFFFF);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                &pcidata);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_write_config(ftdi, activePCIfn | reg, 0,
+                0xF0000000);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                &pcidata);
+        if (UxxxStatus)
+                return UxxxStatus;
+        reg = 12;
+        UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                &latence_timer);
+        if (UxxxStatus)
+                return UxxxStatus;
+        latence_timer &= 0xFFFF00FF;
+        latence_timer |= 0x00001600;
+        UxxxStatus = ftdi_elan_write_config(ftdi, activePCIfn | reg, 0x00,
+                latence_timer);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                &pcidata);
+        if (UxxxStatus)
+                return UxxxStatus;
+        reg = 4;
+        UxxxStatus = ftdi_elan_write_config(ftdi, activePCIfn | reg, 0x00,
+                0x06);
+        if (UxxxStatus)
+                return UxxxStatus;
+        UxxxStatus = ftdi_elan_read_config(ftdi, activePCIfn | reg, 0,
+                &pcidata);
+        if (UxxxStatus)
+                return UxxxStatus;
+        return 0;
+}
+
+static int ftdi_elan_setupOHCI(struct usb_ftdi *ftdi)
+{
+        u32 pcidata;
+        int U132Status;
+        int reg;
+        int reset_repeat = 0;
+      do_reset:reg = 8;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x0e, 0x01);
+        if (U132Status)
+                return U132Status;
+      reset_check:{
+                U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+                if (U132Status)
+                        return U132Status;
+                if (pcidata & 1) {
+                        msleep(500);
+                        if (reset_repeat++ > 100) {
+                                reset_repeat = 0;
+                                goto do_reset;
+                        } else
+                                goto reset_check;
+                }
+        }
+        goto dump_regs;
+        msleep(500);
+        reg = 0x28;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x11000000);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x40;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x2edf);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x34;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x2edf2edf);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 4;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0xA0);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        msleep(250);
+        reg = 8;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x0e, 0x04);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x28;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 8;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x48;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x00001200);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x54;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x58;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x34;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x28002edf);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        msleep(100);
+        reg = 0x50;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x10000);
+        if (U132Status)
+                return U132Status;
+        reg = 0x54;
+      power_check:U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        if (!(pcidata & 1)) {
+                msleep(500);
+                goto power_check;
+        }
+        msleep(3000);
+        reg = 0x54;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x58;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x54;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x02);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x54;
+        U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x10);
+        if (U132Status)
+                return U132Status;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        msleep(750);
+        reg = 0x54;
+        if (0) {
+                U132Status = ftdi_elan_write_pcimem(ftdi, reg, 0x00, 0x02);
+                if (U132Status)
+                        return U132Status;
+        }
+        if (0) {
+                U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+                if (U132Status)
+                        return U132Status;
+        }
+        reg = 0x54;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+        reg = 0x58;
+        U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+        if (U132Status)
+                return U132Status;
+      dump_regs:for (reg = 0; reg <= 0x54; reg += 4) {
+                U132Status = ftdi_elan_read_pcimem(ftdi, reg, 0, &pcidata);
+                if (U132Status)
+                        return U132Status;
+        }
+        return 0;
+}
+
+
+/*
+* we use only the first bulk-in and bulk-out endpoints
+*/
+static int ftdi_elan_probe(struct usb_interface *interface,
+        const struct usb_device_id *id)
+{
+        struct usb_host_interface *iface_desc;
+        struct usb_endpoint_descriptor *endpoint;
+        size_t buffer_size;
+        int i;
+        int retval = -ENOMEM;
+        struct usb_ftdi *ftdi = kmalloc(sizeof(struct usb_ftdi), GFP_KERNEL);
+        if (ftdi == NULL) {
+                printk(KERN_ERR "Out of memory\n");
+                return -ENOMEM;
+        }
+        memset(ftdi, 0x00, sizeof(struct usb_ftdi));
+        down(&ftdi_module_lock);
+        list_add_tail(&ftdi->ftdi_list, &ftdi_static_list);
+        ftdi->sequence_num = ++ftdi_instances;
+        up(&ftdi_module_lock);
+        ftdi_elan_init_kref(ftdi);
+        init_MUTEX(&ftdi->sw_lock);
+        ftdi->udev = usb_get_dev(interface_to_usbdev(interface));
+        ftdi->interface = interface;
+        init_MUTEX(&ftdi->u132_lock);
+        ftdi->expected = 4;
+        iface_desc = interface->cur_altsetting;
+        for (i = 0; i < iface_desc->desc.bNumEndpoints; ++i) {
+                endpoint = &iface_desc->endpoint[i].desc;
+                if (!ftdi->bulk_in_endpointAddr &&
+                        ((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK)
+                        == USB_DIR_IN) && ((endpoint->bmAttributes &
+                        USB_ENDPOINT_XFERTYPE_MASK) == USB_ENDPOINT_XFER_BULK))
+                        {
+                        buffer_size = le16_to_cpu(endpoint->wMaxPacketSize);
+                        ftdi->bulk_in_size = buffer_size;
+                        ftdi->bulk_in_endpointAddr = endpoint->bEndpointAddress;
+                        ftdi->bulk_in_buffer = kmalloc(buffer_size, GFP_KERNEL);
+                        if (!ftdi->bulk_in_buffer) {
+                                dev_err(&ftdi->udev->dev, "Could not allocate b"
+                                        "ulk_in_buffer\n");
+                                retval = -ENOMEM;
+                                goto error;
+                        }
+                }
+                if (!ftdi->bulk_out_endpointAddr &&
+                        ((endpoint->bEndpointAddress & USB_ENDPOINT_DIR_MASK)
+                        == USB_DIR_OUT) && ((endpoint->bmAttributes &
+                        USB_ENDPOINT_XFERTYPE_MASK) == USB_ENDPOINT_XFER_BULK))
+                        {
+                        ftdi->bulk_out_endpointAddr =
+                                endpoint->bEndpointAddress;
+                }
+        }
+        if (!(ftdi->bulk_in_endpointAddr && ftdi->bulk_out_endpointAddr)) {
+                dev_err(&ftdi->udev->dev, "Could not find both bulk-in and bulk"
+                        "-out endpoints\n");
+                retval = -ENODEV;
+                goto error;
+        }
+        dev_info(&ftdi->udev->dev, "interface %d has I=%02X O=%02X\n",
+                iface_desc->desc.bInterfaceNumber, ftdi->bulk_in_endpointAddr,
+                ftdi->bulk_out_endpointAddr);
+        usb_set_intfdata(interface, ftdi);
+        if (iface_desc->desc.bInterfaceNumber == 0 &&
+                ftdi->bulk_in_endpointAddr == 0x81 &&
+                ftdi->bulk_out_endpointAddr == 0x02) {
+                retval = usb_register_dev(interface, &ftdi_elan_jtag_class);
+                if (retval) {
+                        dev_err(&ftdi->udev->dev, "Not able to get a minor for "
+                                "this device.\n");
+                        usb_set_intfdata(interface, NULL);
+                        retval = -ENOMEM;
+                        goto error;
+                } else {
+                        ftdi->class = &ftdi_elan_jtag_class;
+                        dev_info(&ftdi->udev->dev, "USB FDTI=%p JTAG interface "
+                                "%d now attached to ftdi%d\n", ftdi,
+                                iface_desc->desc.bInterfaceNumber,
+                                interface->minor);
+                        return 0;
+                }
+        } else if (iface_desc->desc.bInterfaceNumber == 1 &&
+                ftdi->bulk_in_endpointAddr == 0x83 &&
+                ftdi->bulk_out_endpointAddr == 0x04) {
+                ftdi->class = NULL;
+                dev_info(&ftdi->udev->dev, "USB FDTI=%p ELAN interface %d now a"
+                        "ctivated\n", ftdi, iface_desc->desc.bInterfaceNumber);
+                INIT_WORK(&ftdi->status_work, ftdi_elan_status_work,
+                        (void *)ftdi);
+                INIT_WORK(&ftdi->command_work, ftdi_elan_command_work,
+                        (void *)ftdi);
+                INIT_WORK(&ftdi->respond_work, ftdi_elan_respond_work,
+                        (void *)ftdi);
+                ftdi_status_queue_work(ftdi, msecs_to_jiffies(3 *1000));
+                return 0;
+        } else {
+                dev_err(&ftdi->udev->dev,
+                        "Could not find ELAN's U132 device\n");
+                retval = -ENODEV;
+                goto error;
+        }
+      error:if (ftdi) {
+                ftdi_elan_put_kref(ftdi);
+        }
+        return retval;
+}
+
+static void ftdi_elan_disconnect(struct usb_interface *interface)
+{
+        struct usb_ftdi *ftdi = usb_get_intfdata(interface);
+        ftdi->disconnected += 1;
+        if (ftdi->class) {
+                int minor = interface->minor;
+                struct usb_class_driver *class = ftdi->class;
+                usb_set_intfdata(interface, NULL);
+                usb_deregister_dev(interface, class);
+                dev_info(&ftdi->udev->dev, "USB FTDI U132 jtag interface on min"
+                        "or %d now disconnected\n", minor);
+        } else {
+                ftdi_status_cancel_work(ftdi);
+                ftdi_command_cancel_work(ftdi);
+                ftdi_response_cancel_work(ftdi);
+                ftdi_elan_abandon_completions(ftdi);
+                ftdi_elan_abandon_targets(ftdi);
+                if (ftdi->registered) {
+                        platform_device_unregister(&ftdi->platform_dev);
+                        ftdi->synchronized = 0;
+                        ftdi->enumerated = 0;
+                        ftdi->registered = 0;
+                }
+                flush_workqueue(status_queue);
+                flush_workqueue(command_queue);
+                flush_workqueue(respond_queue);
+                ftdi->disconnected += 1;
+                usb_set_intfdata(interface, NULL);
+                dev_info(&ftdi->udev->dev, "USB FTDI U132 host controller inter"
+                        "face now disconnected\n");
+        }
+        ftdi_elan_put_kref(ftdi);
+}
+
+static struct usb_driver ftdi_elan_driver = {
+        .name = "ftdi-elan",
+        .probe = ftdi_elan_probe,
+        .disconnect = ftdi_elan_disconnect,
+        .id_table = ftdi_elan_table,
+};
+static int __init ftdi_elan_init(void)
+{
+        int result;
+        printk(KERN_INFO "driver %s built at %s on %s\n", ftdi_elan_driver.name,
+                 __TIME__, __DATE__);
+        init_MUTEX(&ftdi_module_lock);
+        INIT_LIST_HEAD(&ftdi_static_list);
+        status_queue = create_singlethread_workqueue("ftdi-status-control");
+        command_queue = create_singlethread_workqueue("ftdi-command-engine");
+        respond_queue = create_singlethread_workqueue("ftdi-respond-engine");
+        result = usb_register(&ftdi_elan_driver);
+        if (result)
+                printk(KERN_ERR "usb_register failed. Error number %d\n",
+                        result);
+        return result;
+}
+
+static void __exit ftdi_elan_exit(void)
+{
+        struct usb_ftdi *ftdi;
+        struct usb_ftdi *temp;
+        usb_deregister(&ftdi_elan_driver);
+        printk(KERN_INFO "ftdi_u132 driver deregistered\n");
+        list_for_each_entry_safe(ftdi, temp, &ftdi_static_list, ftdi_list) {
+                ftdi_status_cancel_work(ftdi);
+                ftdi_command_cancel_work(ftdi);
+                ftdi_response_cancel_work(ftdi);
+        } flush_workqueue(status_queue);
+        destroy_workqueue(status_queue);
+        status_queue = NULL;
+        flush_workqueue(command_queue);
+        destroy_workqueue(command_queue);
+        command_queue = NULL;
+        flush_workqueue(respond_queue);
+        destroy_workqueue(respond_queue);
+        respond_queue = NULL;
+}
+
+
+module_init(ftdi_elan_init);
+module_exit(ftdi_elan_exit);
diff --git a/drivers/usb/misc/usb_u132.h b/drivers/usb/misc/usb_u132.h
new file mode 100644
index 000000000000..551ba8906d62
--- /dev/null
+++ b/drivers/usb/misc/usb_u132.h
@@ -0,0 +1,97 @@
+/*
+* Common Header File for the Elan Digital Systems U132 adapter
+* this file should be included by both the "ftdi-u132" and
+* the "u132-hcd" modules.
+*
+* Copyright(C) 2006 Elan Digital Systems Limited
+*(http://www.elandigitalsystems.com)
+*
+* Author and Maintainer - Tony Olech - Elan Digital Systems
+*(tony.olech@elandigitalsystems.com)
+*
+* This program is free software;you can redistribute it and/or
+* modify it under the terms of the GNU General Public License as
+* published by the Free Software Foundation, version 2.
+*
+*
+* The driver was written by Tony Olech(tony.olech@elandigitalsystems.com)
+* based on various USB client drivers in the 2.6.15 linux kernel
+* with constant reference to the 3rd Edition of Linux Device Drivers
+* published by O'Reilly
+*
+* The U132 adapter is a USB to CardBus adapter specifically designed
+* for PC cards that contain an OHCI host controller. Typical PC cards
+* are the Orange Mobile 3G Option GlobeTrotter Fusion card.
+*
+* The U132 adapter will *NOT *work with PC cards that do not contain
+* an OHCI controller. A simple way to test whether a PC card has an
+* OHCI controller as an interface is to insert the PC card directly
+* into a laptop(or desktop) with a CardBus slot and if "lspci" shows
+* a new USB controller and "lsusb -v" shows a new OHCI Host Controller
+* then there is a good chance that the U132 adapter will support the
+* PC card.(you also need the specific client driver for the PC card)
+*
+* Please inform the Author and Maintainer about any PC cards that
+* contain OHCI Host Controller and work when directly connected to
+* an embedded CardBus slot but do not work when they are connected
+* via an ELAN U132 adapter.
+*
+* The driver consists of two modules, the "ftdi-u132" module is
+* a USB client driver that interfaces to the FTDI chip within
+* the U132 adapter manufactured by Elan Digital Systems, and the
+* "u132-hcd" module is a USB host controller driver that talks
+* to the OHCI controller within CardBus card that are inserted
+* in the U132 adapter.
+*
+* The "ftdi-u132" module should be loaded automatically by the
+* hot plug system when the U132 adapter is plugged in. The module
+* initialises the adapter which mostly consists of synchronising
+* the FTDI chip, before continuously polling the adapter to detect
+* PC card insertions. As soon as a PC card containing a recognised
+* OHCI controller is seen the "ftdi-u132" module explicitly requests
+* the kernel to load the "u132-hcd" module.
+*
+* The "ftdi-u132" module provides the interface to the inserted
+* PC card and the "u132-hcd" module uses the API to send and recieve
+* data. The API features call-backs, so that part of the "u132-hcd"
+* module code will run in the context of one of the kernel threads
+* of the "ftdi-u132" module.
+*
+*/
+int ftdi_elan_switch_on_diagnostics(int number);
+void ftdi_elan_gone_away(struct platform_device *pdev);
+void start_usb_lock_device_tracing(void);
+struct u132_platform_data {
+        u16 vendor;
+        u16 device;
+        u8 potpg;
+        void (*port_power) (struct device *dev, int is_on);
+        void (*reset) (struct device *dev);
+};
+int usb_ftdi_elan_edset_single(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null));
+int usb_ftdi_elan_edset_output(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null));
+int usb_ftdi_elan_edset_empty(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null));
+int usb_ftdi_elan_edset_input(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null));
+int usb_ftdi_elan_edset_setup(struct platform_device *pdev, u8 ed_number,
+        void *endp, struct urb *urb, u8 address, u8 ep_number, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null));
+int usb_ftdi_elan_edset_flush(struct platform_device *pdev, u8 ed_number,
+        void *endp);
-- 
cgit v1.2.1


From d774efeabccf5f5207aa70d5c126fc928e8b30bd Mon Sep 17 00:00:00 2001
From: Tony Olech <tony.olech@elandigitalsystems.com>
Date: Wed, 13 Sep 2006 11:27:35 +0100
Subject: USB: u132-hcd: host controller driver for ELAN U132 adapter

This "u132-hcd" module is one half of the "driver" for
ELAN's U132 which is a USB to CardBus OHCI controller
adapter. This module needs the "ftdi-elan" module in
order to communicate to CardBus OHCI controller inserted
into the U132 adapter.

When the "ftdi-elan" module detects a supported CardBus
OHCI controller in the U132 adapter it loads this "u132-hcd"
module.

Upon a successful device probe() the single workqueue
is started up which does all the processing of commands
from the USB core that implement the host controller.

The workqueue maintains the urb queues and issues commands
via the functions exported by the "ftdi-elan" module. Each
such command will result in a callback.

Note that the "ftdi-elan" module is a USB client driver.

Note that this "u132-hcd" module is a (cut-down OHCI)
host controller.

Thus we have a topology with the parent of a host controller
being a USB client! This really stresses the USB subsystem
semaphore/mutex handling in the module removal.

Signed-off-by: Tony Olech <tony.olech@elandigitalsystems.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/Makefile        |    1 +
 drivers/usb/host/Kconfig    |   28 +
 drivers/usb/host/Makefile   |    1 +
 drivers/usb/host/u132-hcd.c | 3295 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 3325 insertions(+)
 create mode 100644 drivers/usb/host/u132-hcd.c

diff --git a/drivers/usb/Makefile b/drivers/usb/Makefile
index e33647057095..97d57cfc343b 100644
--- a/drivers/usb/Makefile
+++ b/drivers/usb/Makefile
@@ -14,6 +14,7 @@ obj-$(CONFIG_USB_ISP116X_HCD)	+= host/
 obj-$(CONFIG_USB_OHCI_HCD)	+= host/
 obj-$(CONFIG_USB_UHCI_HCD)	+= host/
 obj-$(CONFIG_USB_SL811_HCD)	+= host/
+obj-$(CONFIG_USB_U132_HCD)	+= host/
 obj-$(CONFIG_ETRAX_USB_HOST)	+= host/
 obj-$(CONFIG_USB_OHCI_AT91)	+= host/
 
diff --git a/drivers/usb/host/Kconfig b/drivers/usb/host/Kconfig
index 4bd5cddae8a5..cf10cbc98f80 100644
--- a/drivers/usb/host/Kconfig
+++ b/drivers/usb/host/Kconfig
@@ -142,6 +142,34 @@ config USB_UHCI_HCD
 	  To compile this driver as a module, choose M here: the
 	  module will be called uhci-hcd.
 
+config USB_U132_HCD
+	tristate "Elan U132 Adapter Host Controller"
+	depends on USB && USB_FTDI_ELAN
+	default M
+	help
+	  The U132 adapter is a USB to CardBus adapter specifically designed
+	  for PC cards that contain an OHCI host controller. Typical PC cards
+	  are the Orange Mobile 3G Option GlobeTrotter Fusion card. The U132
+	  adapter will *NOT* work with PC cards that do not contain an OHCI
+	  controller.
+
+	  For those PC cards that contain multiple OHCI controllers only ther
+	  first one is used.
+
+	  The driver consists of two modules, the "ftdi-elan" module is a
+	  USB client driver that interfaces to the FTDI chip within ELAN's
+	  USB-to-PCMCIA adapter, and this "u132-hcd" module is a USB host
+	  controller driver that talks to the OHCI controller within the
+	  CardBus cards that are inserted in the U132 adapter.
+
+	  This driver has been tested with a CardBus OHCI USB adapter, and
+	  worked with a USB PEN Drive inserted into the first USB port of
+	  the PCCARD. A rather pointless thing to do, but useful for testing.
+
+	  It is safe to say M here.
+
+	  See also <http://www.elandigitalsystems.com/support/ufaq/u132linux.php>
+
 config USB_SL811_HCD
 	tristate "SL811HS HCD support"
 	depends on USB
diff --git a/drivers/usb/host/Makefile b/drivers/usb/host/Makefile
index e3020f4b17be..a2e58c86849f 100644
--- a/drivers/usb/host/Makefile
+++ b/drivers/usb/host/Makefile
@@ -14,4 +14,5 @@ obj-$(CONFIG_USB_OHCI_HCD)	+= ohci-hcd.o
 obj-$(CONFIG_USB_UHCI_HCD)	+= uhci-hcd.o
 obj-$(CONFIG_USB_SL811_HCD)	+= sl811-hcd.o
 obj-$(CONFIG_USB_SL811_CS)	+= sl811_cs.o
+obj-$(CONFIG_USB_U132_HCD)	+= u132-hcd.o
 obj-$(CONFIG_ETRAX_ARCH_V10)	+= hc_crisv10.o
diff --git a/drivers/usb/host/u132-hcd.c b/drivers/usb/host/u132-hcd.c
new file mode 100644
index 000000000000..cb2e2a604d1b
--- /dev/null
+++ b/drivers/usb/host/u132-hcd.c
@@ -0,0 +1,3295 @@
+/*
+* Host Controller Driver for the Elan Digital Systems U132 adapter
+*
+* Copyright(C) 2006 Elan Digital Systems Limited
+* http://www.elandigitalsystems.com
+*
+* Author and Maintainer - Tony Olech - Elan Digital Systems
+* tony.olech@elandigitalsystems.com
+*
+* This program is free software;you can redistribute it and/or
+* modify it under the terms of the GNU General Public License as
+* published by the Free Software Foundation, version 2.
+*
+*
+* This driver was written by Tony Olech(tony.olech@elandigitalsystems.com)
+* based on various USB host drivers in the 2.6.15 linux kernel
+* with constant reference to the 3rd Edition of Linux Device Drivers
+* published by O'Reilly
+*
+* The U132 adapter is a USB to CardBus adapter specifically designed
+* for PC cards that contain an OHCI host controller. Typical PC cards
+* are the Orange Mobile 3G Option GlobeTrotter Fusion card.
+*
+* The U132 adapter will *NOT *work with PC cards that do not contain
+* an OHCI controller. A simple way to test whether a PC card has an
+* OHCI controller as an interface is to insert the PC card directly
+* into a laptop(or desktop) with a CardBus slot and if "lspci" shows
+* a new USB controller and "lsusb -v" shows a new OHCI Host Controller
+* then there is a good chance that the U132 adapter will support the
+* PC card.(you also need the specific client driver for the PC card)
+*
+* Please inform the Author and Maintainer about any PC cards that
+* contain OHCI Host Controller and work when directly connected to
+* an embedded CardBus slot but do not work when they are connected
+* via an ELAN U132 adapter.
+*
+*/
+#include <linux/config.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/delay.h>
+#include <linux/ioport.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/smp_lock.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/timer.h>
+#include <linux/list.h>
+#include <linux/interrupt.h>
+#include <linux/usb.h>
+#include <linux/workqueue.h>
+#include <linux/platform_device.h>
+#include <linux/pci_ids.h>
+#include <asm/io.h>
+#include <asm/irq.h>
+#include <asm/system.h>
+#include <asm/byteorder.h>
+#include "../core/hcd.h"
+#include "ohci.h"
+#define OHCI_CONTROL_INIT OHCI_CTRL_CBSR
+#define OHCI_INTR_INIT (OHCI_INTR_MIE | OHCI_INTR_UE | OHCI_INTR_RD | \
+        OHCI_INTR_WDH)
+MODULE_AUTHOR("Tony Olech - Elan Digital Systems Limited");
+MODULE_DESCRIPTION("U132 USB Host Controller Driver");
+MODULE_LICENSE("GPL");
+#define INT_MODULE_PARM(n, v) static int n = v;module_param(n, int, 0444)
+INT_MODULE_PARM(testing, 0);
+/* Some boards misreport power switching/overcurrent*/
+static int distrust_firmware = 1;
+module_param(distrust_firmware, bool, 0);
+MODULE_PARM_DESC(distrust_firmware, "true to distrust firmware power/overcurren"
+        "t setup");
+DECLARE_WAIT_QUEUE_HEAD(u132_hcd_wait);
+/*
+* u132_module_lock exists to protect access to global variables
+*
+*/
+static struct semaphore u132_module_lock;
+static int u132_exiting = 0;
+static int u132_instances = 0;
+static struct list_head u132_static_list;
+/*
+* end of the global variables protected by u132_module_lock
+*/
+static struct workqueue_struct *workqueue;
+#define MAX_U132_PORTS 7
+#define MAX_U132_ADDRS 128
+#define MAX_U132_UDEVS 4
+#define MAX_U132_ENDPS 100
+#define MAX_U132_RINGS 4
+static const char *cc_to_text[16] = {
+        "No Error ",
+        "CRC Error ",
+        "Bit Stuff ",
+        "Data Togg ",
+        "Stall ",
+        "DevNotResp ",
+        "PIDCheck ",
+        "UnExpPID ",
+        "DataOver ",
+        "DataUnder ",
+        "(for hw) ",
+        "(for hw) ",
+        "BufferOver ",
+        "BuffUnder ",
+        "(for HCD) ",
+        "(for HCD) "
+};
+struct u132_port {
+        struct u132 *u132;
+        int reset;
+        int enable;
+        int power;
+        int Status;
+};
+struct u132_addr {
+        u8 address;
+};
+struct u132_udev {
+        struct kref kref;
+        struct usb_device *usb_device;
+        u8 enumeration;
+        u8 udev_number;
+        u8 usb_addr;
+        u8 portnumber;
+        u8 endp_number_in[16];
+        u8 endp_number_out[16];
+};
+#define ENDP_QUEUE_SHIFT 3
+#define ENDP_QUEUE_SIZE (1<<ENDP_QUEUE_SHIFT)
+#define ENDP_QUEUE_MASK (ENDP_QUEUE_SIZE-1)
+struct u132_urbq {
+        struct list_head urb_more;
+        struct urb *urb;
+};
+struct u132_spin {
+        spinlock_t slock;
+};
+struct u132_endp {
+        struct kref kref;
+        u8 udev_number;
+        u8 endp_number;
+        u8 usb_addr;
+        u8 usb_endp;
+        struct u132 *u132;
+        struct list_head endp_ring;
+        struct u132_ring *ring;
+        unsigned toggle_bits:2;
+        unsigned active:1;
+        unsigned delayed:1;
+        unsigned input:1;
+        unsigned output:1;
+        unsigned pipetype:2;
+        unsigned dequeueing:1;
+        unsigned edset_flush:1;
+        unsigned spare_bits:14;
+        unsigned long jiffies;
+        struct usb_host_endpoint *hep;
+        struct u132_spin queue_lock;
+        u16 queue_size;
+        u16 queue_last;
+        u16 queue_next;
+        struct urb *urb_list[ENDP_QUEUE_SIZE];
+        struct list_head urb_more;
+        struct work_struct scheduler;
+};
+struct u132_ring {
+        unsigned in_use:1;
+        unsigned length:7;
+        u8 number;
+        struct u132 *u132;
+        struct u132_endp *curr_endp;
+        struct work_struct scheduler;
+};
+#define OHCI_QUIRK_AMD756 0x01
+#define OHCI_QUIRK_SUPERIO 0x02
+#define OHCI_QUIRK_INITRESET 0x04
+#define OHCI_BIG_ENDIAN 0x08
+#define OHCI_QUIRK_ZFMICRO 0x10
+struct u132 {
+        struct kref kref;
+        struct list_head u132_list;
+        struct semaphore sw_lock;
+        struct semaphore scheduler_lock;
+        struct u132_platform_data *board;
+        struct platform_device *platform_dev;
+        struct u132_ring ring[MAX_U132_RINGS];
+        int sequence_num;
+        int going;
+        int power;
+        int reset;
+        int num_ports;
+        u32 hc_control;
+        u32 hc_fminterval;
+        u32 hc_roothub_status;
+        u32 hc_roothub_a;
+        u32 hc_roothub_portstatus[MAX_ROOT_PORTS];
+        int flags;
+        unsigned long next_statechange;
+        struct work_struct monitor;
+        int num_endpoints;
+        struct u132_addr addr[MAX_U132_ADDRS];
+        struct u132_udev udev[MAX_U132_UDEVS];
+        struct u132_port port[MAX_U132_PORTS];
+        struct u132_endp *endp[MAX_U132_ENDPS];
+};
+int usb_ftdi_elan_read_reg(struct platform_device *pdev, u32 *data);
+int usb_ftdi_elan_read_pcimem(struct platform_device *pdev, u8 addressofs,
+        u8 width, u32 *data);
+int usb_ftdi_elan_write_pcimem(struct platform_device *pdev, u8 addressofs,
+        u8 width, u32 data);
+/*
+* these can not be inlines because we need the structure offset!!
+* Does anyone have a better way?????
+*/
+#define u132_read_pcimem(u132, member, data) \
+        usb_ftdi_elan_read_pcimem(u132->platform_dev, offsetof(struct \
+        ohci_regs, member), 0, data);
+#define u132_write_pcimem(u132, member, data) \
+        usb_ftdi_elan_write_pcimem(u132->platform_dev, offsetof(struct \
+        ohci_regs, member), 0, data);
+#define u132_write_pcimem_byte(u132, member, data) \
+        usb_ftdi_elan_write_pcimem(u132->platform_dev, offsetof(struct \
+        ohci_regs, member), 0x0e, data);
+static inline struct u132 *udev_to_u132(struct u132_udev *udev)
+{
+        u8 udev_number = udev->udev_number;
+        return container_of(udev, struct u132, udev[udev_number]);
+}
+
+static inline struct u132 *hcd_to_u132(struct usb_hcd *hcd)
+{
+        return (struct u132 *)(hcd->hcd_priv);
+}
+
+static inline struct usb_hcd *u132_to_hcd(struct u132 *u132)
+{
+        return container_of((void *)u132, struct usb_hcd, hcd_priv);
+}
+
+static inline void u132_disable(struct u132 *u132)
+{
+        u132_to_hcd(u132)->state = HC_STATE_HALT;
+}
+
+
+#define kref_to_u132(d) container_of(d, struct u132, kref)
+#define kref_to_u132_endp(d) container_of(d, struct u132_endp, kref)
+#define kref_to_u132_udev(d) container_of(d, struct u132_udev, kref)
+#include "../misc/usb_u132.h"
+static const char hcd_name[] = "u132_hcd";
+#define PORT_C_MASK ((USB_PORT_STAT_C_CONNECTION | USB_PORT_STAT_C_ENABLE | \
+        USB_PORT_STAT_C_SUSPEND | USB_PORT_STAT_C_OVERCURRENT | \
+        USB_PORT_STAT_C_RESET) << 16)
+static void u132_hcd_delete(struct kref *kref)
+{
+        struct u132 *u132 = kref_to_u132(kref);
+        struct platform_device *pdev = u132->platform_dev;
+        struct usb_hcd *hcd = u132_to_hcd(u132);
+        u132->going += 1;
+        down(&u132_module_lock);
+        list_del_init(&u132->u132_list);
+        u132_instances -= 1;
+        up(&u132_module_lock);
+        dev_warn(&u132->platform_dev->dev, "FREEING the hcd=%p and thus the u13"
+                "2=%p going=%d pdev=%p\n", hcd, u132, u132->going, pdev);
+        usb_put_hcd(hcd);
+}
+
+static inline void u132_u132_put_kref(struct u132 *u132)
+{
+        kref_put(&u132->kref, u132_hcd_delete);
+}
+
+static inline void u132_u132_init_kref(struct u132 *u132)
+{
+        kref_init(&u132->kref);
+}
+
+static void u132_udev_delete(struct kref *kref)
+{
+        struct u132_udev *udev = kref_to_u132_udev(kref);
+        udev->udev_number = 0;
+        udev->usb_device = NULL;
+        udev->usb_addr = 0;
+        udev->enumeration = 0;
+}
+
+static inline void u132_udev_put_kref(struct u132 *u132, struct u132_udev *udev)
+{
+        kref_put(&udev->kref, u132_udev_delete);
+}
+
+static inline void u132_udev_get_kref(struct u132 *u132, struct u132_udev *udev)
+{
+        kref_get(&udev->kref);
+}
+
+static inline void u132_udev_init_kref(struct u132 *u132,
+        struct u132_udev *udev)
+{
+        kref_init(&udev->kref);
+}
+
+static inline void u132_ring_put_kref(struct u132 *u132, struct u132_ring *ring)
+{
+        kref_put(&u132->kref, u132_hcd_delete);
+}
+
+static void u132_ring_requeue_work(struct u132 *u132, struct u132_ring *ring,
+        unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(workqueue, &ring->scheduler, delta))
+                        return;
+        } else if (queue_work(workqueue, &ring->scheduler))
+                return;
+        kref_put(&u132->kref, u132_hcd_delete);
+        return;
+}
+
+static void u132_ring_queue_work(struct u132 *u132, struct u132_ring *ring,
+        unsigned int delta)
+{
+        kref_get(&u132->kref);
+        u132_ring_requeue_work(u132, ring, delta);
+        return;
+}
+
+static void u132_ring_cancel_work(struct u132 *u132, struct u132_ring *ring)
+{
+        if (cancel_delayed_work(&ring->scheduler)) {
+                kref_put(&u132->kref, u132_hcd_delete);
+        }
+}
+
+static void u132_endp_delete(struct kref *kref)
+{
+        struct u132_endp *endp = kref_to_u132_endp(kref);
+        struct u132 *u132 = endp->u132;
+        u8 usb_addr = endp->usb_addr;
+        u8 usb_endp = endp->usb_endp;
+        u8 address = u132->addr[usb_addr].address;
+        struct u132_udev *udev = &u132->udev[address];
+        u8 endp_number = endp->endp_number;
+        struct usb_host_endpoint *hep = endp->hep;
+        struct u132_ring *ring = endp->ring;
+        struct list_head *head = &endp->endp_ring;
+        ring->length -= 1;
+        if (endp == ring->curr_endp) {
+                if (list_empty(head)) {
+                        ring->curr_endp = NULL;
+                        list_del(head);
+                } else {
+                        struct u132_endp *next_endp = list_entry(head->next,
+                                struct u132_endp, endp_ring);
+                        ring->curr_endp = next_endp;
+                        list_del(head);
+        }} else
+                list_del(head);
+        if (endp->input) {
+                udev->endp_number_in[usb_endp] = 0;
+                u132_udev_put_kref(u132, udev);
+        }
+        if (endp->output) {
+                udev->endp_number_out[usb_endp] = 0;
+                u132_udev_put_kref(u132, udev);
+        }
+        u132->endp[endp_number - 1] = NULL;
+        hep->hcpriv = NULL;
+        kfree(endp);
+        u132_u132_put_kref(u132);
+}
+
+static inline void u132_endp_put_kref(struct u132 *u132, struct u132_endp *endp)
+{
+        kref_put(&endp->kref, u132_endp_delete);
+}
+
+static inline void u132_endp_get_kref(struct u132 *u132, struct u132_endp *endp)
+{
+        kref_get(&endp->kref);
+}
+
+static inline void u132_endp_init_kref(struct u132 *u132,
+        struct u132_endp *endp)
+{
+        kref_init(&endp->kref);
+        kref_get(&u132->kref);
+}
+
+static void u132_endp_queue_work(struct u132 *u132, struct u132_endp *endp,
+        unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(workqueue, &endp->scheduler, delta))
+                        kref_get(&endp->kref);
+        } else if (queue_work(workqueue, &endp->scheduler))
+                kref_get(&endp->kref);
+        return;
+}
+
+static void u132_endp_cancel_work(struct u132 *u132, struct u132_endp *endp)
+{
+        if (cancel_delayed_work(&endp->scheduler))
+                kref_put(&endp->kref, u132_endp_delete);
+}
+
+static inline void u132_monitor_put_kref(struct u132 *u132)
+{
+        kref_put(&u132->kref, u132_hcd_delete);
+}
+
+static void u132_monitor_queue_work(struct u132 *u132, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(workqueue, &u132->monitor, delta)) {
+                        kref_get(&u132->kref);
+                }
+        } else if (queue_work(workqueue, &u132->monitor))
+                kref_get(&u132->kref);
+        return;
+}
+
+static void u132_monitor_requeue_work(struct u132 *u132, unsigned int delta)
+{
+        if (delta > 0) {
+                if (queue_delayed_work(workqueue, &u132->monitor, delta))
+                        return;
+        } else if (queue_work(workqueue, &u132->monitor))
+                return;
+        kref_put(&u132->kref, u132_hcd_delete);
+        return;
+}
+
+static void u132_monitor_cancel_work(struct u132 *u132)
+{
+        if (cancel_delayed_work(&u132->monitor))
+                kref_put(&u132->kref, u132_hcd_delete);
+}
+
+static int read_roothub_info(struct u132 *u132)
+{
+        u32 revision;
+        int retval;
+        retval = u132_read_pcimem(u132, revision, &revision);
+        if (retval) {
+                dev_err(&u132->platform_dev->dev, "error %d accessing device co"
+                        "ntrol\n", retval);
+                return retval;
+        } else if ((revision & 0xFF) == 0x10) {
+        } else if ((revision & 0xFF) == 0x11) {
+        } else {
+                dev_err(&u132->platform_dev->dev, "device revision is not valid"
+                        " %08X\n", revision);
+                return -ENODEV;
+        }
+        retval = u132_read_pcimem(u132, control, &u132->hc_control);
+        if (retval) {
+                dev_err(&u132->platform_dev->dev, "error %d accessing device co"
+                        "ntrol\n", retval);
+                return retval;
+        }
+        retval = u132_read_pcimem(u132, roothub.status,
+                &u132->hc_roothub_status);
+        if (retval) {
+                dev_err(&u132->platform_dev->dev, "error %d accessing device re"
+                        "g roothub.status\n", retval);
+                return retval;
+        }
+        retval = u132_read_pcimem(u132, roothub.a, &u132->hc_roothub_a);
+        if (retval) {
+                dev_err(&u132->platform_dev->dev, "error %d accessing device re"
+                        "g roothub.a\n", retval);
+                return retval;
+        }
+        {
+                int I = u132->num_ports;
+                int i = 0;
+                while (I-- > 0) {
+                        retval = u132_read_pcimem(u132, roothub.portstatus[i],
+                                &u132->hc_roothub_portstatus[i]);
+                        if (retval) {
+                                dev_err(&u132->platform_dev->dev, "error %d acc"
+                                        "essing device roothub.portstatus[%d]\n"
+                                        , retval, i);
+                                return retval;
+                        } else
+                                i += 1;
+                }
+        }
+        return 0;
+}
+
+static void u132_hcd_monitor_work(void *data)
+{
+        struct u132 *u132 = data;
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                u132_monitor_put_kref(u132);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                u132_monitor_put_kref(u132);
+                return;
+        } else {
+                int retval;
+                down(&u132->sw_lock);
+                retval = read_roothub_info(u132);
+                if (retval) {
+                        struct usb_hcd *hcd = u132_to_hcd(u132);
+                        u132_disable(u132);
+                        u132->going = 1;
+                        up(&u132->sw_lock);
+                        usb_hc_died(hcd);
+                        ftdi_elan_gone_away(u132->platform_dev);
+                        u132_monitor_put_kref(u132);
+                        return;
+                } else {
+                        u132_monitor_requeue_work(u132, 500);
+                        up(&u132->sw_lock);
+                        return;
+                }
+        }
+}
+
+static void u132_hcd_giveback_urb(struct u132 *u132, struct u132_endp *endp,
+        struct urb *urb, int status)
+{
+        struct u132_ring *ring;
+        unsigned long irqs;
+        struct usb_hcd *hcd = u132_to_hcd(u132);
+        urb->error_count = 0;
+        urb->status = status;
+        urb->hcpriv = NULL;
+        spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+        endp->queue_next += 1;
+        if (ENDP_QUEUE_SIZE > --endp->queue_size) {
+                endp->active = 0;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+        } else {
+                struct list_head *next = endp->urb_more.next;
+                struct u132_urbq *urbq = list_entry(next, struct u132_urbq,
+                        urb_more);
+                list_del(next);
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] =
+                        urbq->urb;
+                endp->active = 0;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                kfree(urbq);
+        } down(&u132->scheduler_lock);
+        ring = endp->ring;
+        ring->in_use = 0;
+        u132_ring_cancel_work(u132, ring);
+        u132_ring_queue_work(u132, ring, 0);
+        up(&u132->scheduler_lock);
+        u132_endp_put_kref(u132, endp);
+        usb_hcd_giveback_urb(hcd, urb, NULL);
+        return;
+}
+
+static void u132_hcd_forget_urb(struct u132 *u132, struct u132_endp *endp,
+        struct urb *urb, int status)
+{
+        u132_endp_put_kref(u132, endp);
+}
+
+static void u132_hcd_abandon_urb(struct u132 *u132, struct u132_endp *endp,
+        struct urb *urb, int status)
+{
+        unsigned long irqs;
+        struct usb_hcd *hcd = u132_to_hcd(u132);
+        urb->error_count = 0;
+        urb->status = status;
+        urb->hcpriv = NULL;
+        spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+        endp->queue_next += 1;
+        if (ENDP_QUEUE_SIZE > --endp->queue_size) {
+                endp->active = 0;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+        } else {
+                struct list_head *next = endp->urb_more.next;
+                struct u132_urbq *urbq = list_entry(next, struct u132_urbq,
+                        urb_more);
+                list_del(next);
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] =
+                        urbq->urb;
+                endp->active = 0;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                kfree(urbq);
+        } usb_hcd_giveback_urb(hcd, urb, NULL);
+        return;
+}
+
+static inline int edset_input(struct u132 *u132, struct u132_ring *ring,
+        struct u132_endp *endp, struct urb *urb, u8 address, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        return usb_ftdi_elan_edset_input(u132->platform_dev, ring->number, endp,
+                 urb, address, endp->usb_endp, toggle_bits, callback);
+}
+
+static inline int edset_setup(struct u132 *u132, struct u132_ring *ring,
+        struct u132_endp *endp, struct urb *urb, u8 address, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        return usb_ftdi_elan_edset_setup(u132->platform_dev, ring->number, endp,
+                 urb, address, endp->usb_endp, toggle_bits, callback);
+}
+
+static inline int edset_single(struct u132 *u132, struct u132_ring *ring,
+        struct u132_endp *endp, struct urb *urb, u8 address, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        return usb_ftdi_elan_edset_single(u132->platform_dev, ring->number,
+                endp, urb, address, endp->usb_endp, toggle_bits, callback);
+}
+
+static inline int edset_output(struct u132 *u132, struct u132_ring *ring,
+        struct u132_endp *endp, struct urb *urb, u8 address, u8 toggle_bits,
+        void (*callback) (void *endp, struct urb *urb, u8 *buf, int len,
+        int toggle_bits, int error_count, int condition_code, int repeat_number,
+         int halted, int skipped, int actual, int non_null))
+{
+        return usb_ftdi_elan_edset_output(u132->platform_dev, ring->number,
+                endp, urb, address, endp->usb_endp, toggle_bits, callback);
+}
+
+
+/*
+* must not LOCK sw_lock
+*
+*/
+static void u132_hcd_interrupt_recv(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        struct u132_udev *udev = &u132->udev[address];
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                struct u132_ring *ring = endp->ring;
+                u8 *u = urb->transfer_buffer + urb->actual_length;
+                u8 *b = buf;
+                int L = len;
+                while (L-- > 0) {
+                        *u++ = *b++;
+                }
+                urb->actual_length += len;
+                if ((condition_code == TD_CC_NOERROR) &&
+                        (urb->transfer_buffer_length > urb->actual_length)) {
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        if (urb->actual_length > 0) {
+                                int retval;
+                                up(&u132->scheduler_lock);
+                                retval = edset_single(u132, ring, endp, urb,
+                                        address, endp->toggle_bits,
+                                        u132_hcd_interrupt_recv);
+                                if (retval == 0) {
+                                } else
+                                        u132_hcd_giveback_urb(u132, endp, urb,
+                                                retval);
+                        } else {
+                                ring->in_use = 0;
+                                endp->active = 0;
+                                endp->jiffies = jiffies +
+                                        msecs_to_jiffies(urb->interval);
+                                u132_ring_cancel_work(u132, ring);
+                                u132_ring_queue_work(u132, ring, 0);
+                                up(&u132->scheduler_lock);
+                                u132_endp_put_kref(u132, endp);
+                        }
+                        return;
+                } else if ((condition_code == TD_DATAUNDERRUN) &&
+                        ((urb->transfer_flags & URB_SHORT_NOT_OK) == 0)) {
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb, 0);
+                        return;
+                } else {
+                        if (condition_code == TD_CC_NOERROR) {
+                                endp->toggle_bits = toggle_bits;
+                                usb_settoggle(udev->usb_device, endp->usb_endp,
+                                        0, 1 & toggle_bits);
+                        } else if (condition_code == TD_CC_STALL) {
+                                endp->toggle_bits = 0x2;
+                                usb_settoggle(udev->usb_device, endp->usb_endp,
+                                        0, 0);
+                        } else {
+                                endp->toggle_bits = 0x2;
+                                usb_settoggle(udev->usb_device, endp->usb_endp,
+                                        0, 0);
+                                dev_err(&u132->platform_dev->dev, "urb=%p givin"
+                                        "g back INTERRUPT %s\n", urb,
+                                        cc_to_text[condition_code]);
+                        }
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                }
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_bulk_output_sent(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                struct u132_ring *ring = endp->ring;
+                urb->actual_length += len;
+                endp->toggle_bits = toggle_bits;
+                if (urb->transfer_buffer_length > urb->actual_length) {
+                        int retval;
+                        up(&u132->scheduler_lock);
+                        retval = edset_output(u132, ring, endp, urb, address,
+                                endp->toggle_bits, u132_hcd_bulk_output_sent);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else {
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb, 0);
+                        return;
+                }
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_bulk_input_recv(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        struct u132_udev *udev = &u132->udev[address];
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                struct u132_ring *ring = endp->ring;
+                u8 *u = urb->transfer_buffer + urb->actual_length;
+                u8 *b = buf;
+                int L = len;
+                while (L-- > 0) {
+                        *u++ = *b++;
+                }
+                urb->actual_length += len;
+                if ((condition_code == TD_CC_NOERROR) &&
+                        (urb->transfer_buffer_length > urb->actual_length)) {
+                        int retval;
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        up(&u132->scheduler_lock);
+                        retval = usb_ftdi_elan_edset_input(u132->platform_dev,
+                                ring->number, endp, urb, address,
+                                endp->usb_endp, endp->toggle_bits,
+                                u132_hcd_bulk_input_recv);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else if (condition_code == TD_CC_NOERROR) {
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                } else if ((condition_code == TD_DATAUNDERRUN) &&
+                        ((urb->transfer_flags & URB_SHORT_NOT_OK) == 0)) {
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb, 0);
+                        return;
+                } else if (condition_code == TD_DATAUNDERRUN) {
+                        endp->toggle_bits = toggle_bits;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0,
+                                1 & toggle_bits);
+                        dev_warn(&u132->platform_dev->dev, "urb=%p(SHORT NOT OK"
+                                ") giving back BULK IN %s\n", urb,
+                                cc_to_text[condition_code]);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb, 0);
+                        return;
+                } else if (condition_code == TD_CC_STALL) {
+                        endp->toggle_bits = 0x2;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0, 0);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                } else {
+                        endp->toggle_bits = 0x2;
+                        usb_settoggle(udev->usb_device, endp->usb_endp, 0, 0);
+                        dev_err(&u132->platform_dev->dev, "urb=%p giving back B"
+                                "ULK IN code=%d %s\n", urb, condition_code,
+                                cc_to_text[condition_code]);
+                        up(&u132->scheduler_lock);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                }
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_configure_empty_sent(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, 0);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_configure_input_recv(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                struct u132_ring *ring = endp->ring;
+                u8 *u = urb->transfer_buffer;
+                u8 *b = buf;
+                int L = len;
+                while (L-- > 0) {
+                        *u++ = *b++;
+                }
+                urb->actual_length = len;
+                if ((condition_code == TD_CC_NOERROR) || ((condition_code ==
+                        TD_DATAUNDERRUN) && ((urb->transfer_flags &
+                        URB_SHORT_NOT_OK) == 0))) {
+                        int retval;
+                        up(&u132->scheduler_lock);
+                        retval = usb_ftdi_elan_edset_empty(u132->platform_dev,
+                                ring->number, endp, urb, address,
+                                endp->usb_endp, 0x3,
+                                u132_hcd_configure_empty_sent);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else if (condition_code == TD_CC_STALL) {
+                        up(&u132->scheduler_lock);
+                        dev_warn(&u132->platform_dev->dev, "giving back SETUP I"
+                                "NPUT STALL urb %p\n", urb);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                } else {
+                        up(&u132->scheduler_lock);
+                        dev_err(&u132->platform_dev->dev, "giving back SETUP IN"
+                                "PUT %s urb %p\n", cc_to_text[condition_code],
+                                urb);
+                        u132_hcd_giveback_urb(u132, endp, urb,
+                                cc_to_error[condition_code]);
+                        return;
+                }
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_configure_empty_recv(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, 0);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_configure_setup_sent(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                if (usb_pipein(urb->pipe)) {
+                        int retval;
+                        struct u132_ring *ring = endp->ring;
+                        up(&u132->scheduler_lock);
+                        retval = usb_ftdi_elan_edset_input(u132->platform_dev,
+                                ring->number, endp, urb, address,
+                                endp->usb_endp, 0,
+                                u132_hcd_configure_input_recv);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else {
+                        int retval;
+                        struct u132_ring *ring = endp->ring;
+                        up(&u132->scheduler_lock);
+                        retval = usb_ftdi_elan_edset_input(u132->platform_dev,
+                                ring->number, endp, urb, address,
+                                endp->usb_endp, 0,
+                                u132_hcd_configure_empty_recv);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                }
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_enumeration_empty_recv(void *data, struct urb *urb,
+        u8 *buf, int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        struct u132_udev *udev = &u132->udev[address];
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                u132->addr[0].address = 0;
+                endp->usb_addr = udev->usb_addr;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, 0);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_enumeration_address_sent(void *data, struct urb *urb,
+        u8 *buf, int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                int retval;
+                struct u132_ring *ring = endp->ring;
+                up(&u132->scheduler_lock);
+                retval = usb_ftdi_elan_edset_input(u132->platform_dev,
+                        ring->number, endp, urb, 0, endp->usb_endp, 0,
+                        u132_hcd_enumeration_empty_recv);
+                if (retval == 0) {
+                } else
+                        u132_hcd_giveback_urb(u132, endp, urb, retval);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_initial_empty_sent(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, 0);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_initial_input_recv(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                int retval;
+                struct u132_ring *ring = endp->ring;
+                u8 *u = urb->transfer_buffer;
+                u8 *b = buf;
+                int L = len;
+                while (L-- > 0) {
+                        *u++ = *b++;
+                }
+                urb->actual_length = len;
+                up(&u132->scheduler_lock);
+                retval = usb_ftdi_elan_edset_empty(u132->platform_dev,
+                        ring->number, endp, urb, address, endp->usb_endp, 0x3,
+                        u132_hcd_initial_empty_sent);
+                if (retval == 0) {
+                } else
+                        u132_hcd_giveback_urb(u132, endp, urb, retval);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_initial_setup_sent(void *data, struct urb *urb, u8 *buf,
+        int len, int toggle_bits, int error_count, int condition_code,
+        int repeat_number, int halted, int skipped, int actual, int non_null)
+{
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        u8 address = u132->addr[endp->usb_addr].address;
+        down(&u132->scheduler_lock);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                up(&u132->scheduler_lock);
+                u132_hcd_forget_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (endp->dequeueing) {
+                endp->dequeueing = 0;
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -EINTR);
+                return;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, -ENODEV);
+                return;
+        } else if (urb->status == -EINPROGRESS) {
+                int retval;
+                struct u132_ring *ring = endp->ring;
+                up(&u132->scheduler_lock);
+                retval = usb_ftdi_elan_edset_input(u132->platform_dev,
+                        ring->number, endp, urb, address, endp->usb_endp, 0,
+                        u132_hcd_initial_input_recv);
+                if (retval == 0) {
+                } else
+                        u132_hcd_giveback_urb(u132, endp, urb, retval);
+                return;
+        } else {
+                dev_err(&u132->platform_dev->dev, "CALLBACK called urb=%p statu"
+                        "s=%d\n", urb, urb->status);
+                up(&u132->scheduler_lock);
+                u132_hcd_giveback_urb(u132, endp, urb, urb->status);
+                return;
+        }
+}
+
+static void u132_hcd_ring_work_scheduler(void *data);
+static void u132_hcd_endp_work_scheduler(void *data);
+/*
+* this work function is only executed from the work queue
+*
+*/
+static void u132_hcd_ring_work_scheduler(void *data)
+{
+        struct u132_ring *ring = data;
+        struct u132 *u132 = ring->u132;
+        down(&u132->scheduler_lock);
+        if (ring->in_use) {
+                up(&u132->scheduler_lock);
+                u132_ring_put_kref(u132, ring);
+                return;
+        } else if (ring->curr_endp) {
+                struct u132_endp *last_endp = ring->curr_endp;
+                struct list_head *scan;
+                struct list_head *head = &last_endp->endp_ring;
+                unsigned long wakeup = 0;
+                list_for_each(scan, head) {
+                        struct u132_endp *endp = list_entry(scan,
+                                struct u132_endp, endp_ring);
+                        if (endp->queue_next == endp->queue_last) {
+                        } else if ((endp->delayed == 0)
+                                || time_after_eq(jiffies, endp->jiffies)) {
+                                ring->curr_endp = endp;
+                                u132_endp_cancel_work(u132, last_endp);
+                                u132_endp_queue_work(u132, last_endp, 0);
+                                up(&u132->scheduler_lock);
+                                u132_ring_put_kref(u132, ring);
+                                return;
+                        } else {
+                                unsigned long delta = endp->jiffies - jiffies;
+                                if (delta > wakeup)
+                                        wakeup = delta;
+                        }
+                }
+                if (last_endp->queue_next == last_endp->queue_last) {
+                } else if ((last_endp->delayed == 0) || time_after_eq(jiffies,
+                        last_endp->jiffies)) {
+                        u132_endp_cancel_work(u132, last_endp);
+                        u132_endp_queue_work(u132, last_endp, 0);
+                        up(&u132->scheduler_lock);
+                        u132_ring_put_kref(u132, ring);
+                        return;
+                } else {
+                        unsigned long delta = last_endp->jiffies - jiffies;
+                        if (delta > wakeup)
+                                wakeup = delta;
+                }
+                if (wakeup > 0) {
+                        u132_ring_requeue_work(u132, ring, wakeup);
+                        up(&u132->scheduler_lock);
+                        return;
+                } else {
+                        up(&u132->scheduler_lock);
+                        u132_ring_put_kref(u132, ring);
+                        return;
+                }
+        } else {
+                up(&u132->scheduler_lock);
+                u132_ring_put_kref(u132, ring);
+                return;
+        }
+}
+
+static void u132_hcd_endp_work_scheduler(void *data)
+{
+        struct u132_ring *ring;
+        struct u132_endp *endp = data;
+        struct u132 *u132 = endp->u132;
+        down(&u132->scheduler_lock);
+        ring = endp->ring;
+        if (endp->edset_flush) {
+                endp->edset_flush = 0;
+                if (endp->dequeueing)
+                        usb_ftdi_elan_edset_flush(u132->platform_dev,
+                                ring->number, endp);
+                up(&u132->scheduler_lock);
+                u132_endp_put_kref(u132, endp);
+                return;
+        } else if (endp->active) {
+                up(&u132->scheduler_lock);
+                u132_endp_put_kref(u132, endp);
+                return;
+        } else if (ring->in_use) {
+                up(&u132->scheduler_lock);
+                u132_endp_put_kref(u132, endp);
+                return;
+        } else if (endp->queue_next == endp->queue_last) {
+                up(&u132->scheduler_lock);
+                u132_endp_put_kref(u132, endp);
+                return;
+        } else if (endp->pipetype == PIPE_INTERRUPT) {
+                u8 address = u132->addr[endp->usb_addr].address;
+                if (ring->in_use) {
+                        up(&u132->scheduler_lock);
+                        u132_endp_put_kref(u132, endp);
+                        return;
+                } else {
+                        int retval;
+                        struct urb *urb = endp->urb_list[ENDP_QUEUE_MASK &
+                                endp->queue_next];
+                        endp->active = 1;
+                        ring->curr_endp = endp;
+                        ring->in_use = 1;
+                        up(&u132->scheduler_lock);
+                        retval = edset_single(u132, ring, endp, urb, address,
+                                endp->toggle_bits, u132_hcd_interrupt_recv);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                }
+        } else if (endp->pipetype == PIPE_CONTROL) {
+                u8 address = u132->addr[endp->usb_addr].address;
+                if (ring->in_use) {
+                        up(&u132->scheduler_lock);
+                        u132_endp_put_kref(u132, endp);
+                        return;
+                } else if (address == 0) {
+                        int retval;
+                        struct urb *urb = endp->urb_list[ENDP_QUEUE_MASK &
+                                endp->queue_next];
+                        endp->active = 1;
+                        ring->curr_endp = endp;
+                        ring->in_use = 1;
+                        up(&u132->scheduler_lock);
+                        retval = edset_setup(u132, ring, endp, urb, address,
+                                0x2, u132_hcd_initial_setup_sent);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else if (endp->usb_addr == 0) {
+                        int retval;
+                        struct urb *urb = endp->urb_list[ENDP_QUEUE_MASK &
+                                endp->queue_next];
+                        endp->active = 1;
+                        ring->curr_endp = endp;
+                        ring->in_use = 1;
+                        up(&u132->scheduler_lock);
+                        retval = edset_setup(u132, ring, endp, urb, 0, 0x2,
+                                u132_hcd_enumeration_address_sent);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                } else {
+                        int retval;
+                        u8 address = u132->addr[endp->usb_addr].address;
+                        struct urb *urb = endp->urb_list[ENDP_QUEUE_MASK &
+                                endp->queue_next];
+                        endp->active = 1;
+                        ring->curr_endp = endp;
+                        ring->in_use = 1;
+                        up(&u132->scheduler_lock);
+                        retval = edset_setup(u132, ring, endp, urb, address,
+                                0x2, u132_hcd_configure_setup_sent);
+                        if (retval == 0) {
+                        } else
+                                u132_hcd_giveback_urb(u132, endp, urb, retval);
+                        return;
+                }
+        } else {
+                if (endp->input) {
+                        u8 address = u132->addr[endp->usb_addr].address;
+                        if (ring->in_use) {
+                                up(&u132->scheduler_lock);
+                                u132_endp_put_kref(u132, endp);
+                                return;
+                        } else {
+                                int retval;
+                                struct urb *urb = endp->urb_list[
+                                        ENDP_QUEUE_MASK & endp->queue_next];
+                                endp->active = 1;
+                                ring->curr_endp = endp;
+                                ring->in_use = 1;
+                                up(&u132->scheduler_lock);
+                                retval = edset_input(u132, ring, endp, urb,
+                                        address, endp->toggle_bits,
+                                        u132_hcd_bulk_input_recv);
+                                if (retval == 0) {
+                                } else
+                                        u132_hcd_giveback_urb(u132, endp, urb,
+                                                retval);
+                                return;
+                        }
+                } else {        /* output pipe */
+                        u8 address = u132->addr[endp->usb_addr].address;
+                        if (ring->in_use) {
+                                up(&u132->scheduler_lock);
+                                u132_endp_put_kref(u132, endp);
+                                return;
+                        } else {
+                                int retval;
+                                struct urb *urb = endp->urb_list[
+                                        ENDP_QUEUE_MASK & endp->queue_next];
+                                endp->active = 1;
+                                ring->curr_endp = endp;
+                                ring->in_use = 1;
+                                up(&u132->scheduler_lock);
+                                retval = edset_output(u132, ring, endp, urb,
+                                        address, endp->toggle_bits,
+                                        u132_hcd_bulk_output_sent);
+                                if (retval == 0) {
+                                } else
+                                        u132_hcd_giveback_urb(u132, endp, urb,
+                                                retval);
+                                return;
+                        }
+                }
+        }
+}
+
+static void port_power(struct u132 *u132, int pn, int is_on)
+{
+        u132->port[pn].power = is_on;
+}
+
+static void u132_power(struct u132 *u132, int is_on)
+{
+        struct usb_hcd *hcd = u132_to_hcd(u132)
+                ;        /* hub is inactive unless the port is powered */
+        if (is_on) {
+                if (u132->power)
+                        return;
+                u132->power = 1;
+                hcd->self.controller->power.power_state = PMSG_ON;
+        } else {
+                u132->power = 0;
+                hcd->state = HC_STATE_HALT;
+                hcd->self.controller->power.power_state = PMSG_SUSPEND;
+        }
+}
+
+static int u132_periodic_reinit(struct u132 *u132)
+{
+        int retval;
+        u32 fi = u132->hc_fminterval & 0x03fff;
+        u32 fit;
+        u32 fminterval;
+        retval = u132_read_pcimem(u132, fminterval, &fminterval);
+        if (retval)
+                return retval;
+        fit = fminterval & FIT;
+        retval = u132_write_pcimem(u132, fminterval,
+                (fit ^ FIT) | u132->hc_fminterval);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, periodicstart,
+                ((9 *fi) / 10) & 0x3fff);
+        if (retval)
+                return retval;
+        return 0;
+}
+
+static char *hcfs2string(int state)
+{
+        switch (state) {
+        case OHCI_USB_RESET:
+                return "reset";
+        case OHCI_USB_RESUME:
+                return "resume";
+        case OHCI_USB_OPER:
+                return "operational";
+        case OHCI_USB_SUSPEND:
+                return "suspend";
+        }
+        return "?";
+}
+
+static int u132_usb_reset(struct u132 *u132)
+{
+        int retval;
+        retval = u132_read_pcimem(u132, control, &u132->hc_control);
+        if (retval)
+                return retval;
+        u132->hc_control &= OHCI_CTRL_RWC;
+        retval = u132_write_pcimem(u132, control, u132->hc_control);
+        if (retval)
+                return retval;
+        return 0;
+}
+
+static int u132_init(struct u132 *u132)
+{
+        int retval;
+        u32 control;
+        u132_disable(u132);
+        u132->next_statechange =
+                jiffies; /* SMM owns the HC? not for long! */  {
+                u32 control;
+                retval = u132_read_pcimem(u132, control, &control);
+                if (retval)
+                        return retval;
+                if (control & OHCI_CTRL_IR) {
+                        u32 temp = 50;
+                        retval = u132_write_pcimem(u132, intrenable,
+                                OHCI_INTR_OC);
+                        if (retval)
+                                return retval;
+                        retval = u132_write_pcimem_byte(u132, cmdstatus,
+                                OHCI_OCR);
+                        if (retval)
+                                return retval;
+                      check:{
+                                retval = u132_read_pcimem(u132, control,
+                                        &control);
+                                if (retval)
+                                        return retval;
+                        }
+                        if (control & OHCI_CTRL_IR) {
+                                msleep(10);
+                                if (--temp == 0) {
+                                        dev_err(&u132->platform_dev->dev, "USB "
+                                                "HC takeover failed!(BIOS/SMM b"
+                                                "ug) control=%08X\n", control);
+                                        return -EBUSY;
+                                }
+                                goto check;
+                        }
+                        u132_usb_reset(u132);
+                }
+        }
+        retval = u132_write_pcimem(u132, intrdisable, OHCI_INTR_MIE);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, control, &control);
+        if (retval)
+                return retval;
+        if (u132->num_ports == 0) {
+                u32 rh_a = -1;
+                retval = u132_read_pcimem(u132, roothub.a, &rh_a);
+                if (retval)
+                        return retval;
+                u132->num_ports = rh_a & RH_A_NDP;
+                retval = read_roothub_info(u132);
+                if (retval)
+                        return retval;
+        }
+        if (u132->num_ports > MAX_U132_PORTS) {
+                return -EINVAL;
+        }
+        return 0;
+}
+
+
+/* Start an OHCI controller, set the BUS operational
+* resets USB and controller
+* enable interrupts
+*/
+static int u132_run(struct u132 *u132)
+{
+        int retval;
+        u32 control;
+        u32 status;
+        u32 fminterval;
+        u32 periodicstart;
+        u32 cmdstatus;
+        u32 roothub_a;
+        int mask = OHCI_INTR_INIT;
+        int first = u132->hc_fminterval == 0;
+        int sleep_time = 0;
+        int reset_timeout = 30;        /* ... allow extra time */
+        u132_disable(u132);
+        if (first) {
+                u32 temp;
+                retval = u132_read_pcimem(u132, fminterval, &temp);
+                if (retval)
+                        return retval;
+                u132->hc_fminterval = temp & 0x3fff;
+                if (u132->hc_fminterval != FI) {
+                }
+                u132->hc_fminterval |= FSMP(u132->hc_fminterval) << 16;
+        }
+        retval = u132_read_pcimem(u132, control, &u132->hc_control);
+        if (retval)
+                return retval;
+        dev_info(&u132->platform_dev->dev, "resetting from state '%s', control "
+                "= %08X\n", hcfs2string(u132->hc_control & OHCI_CTRL_HCFS),
+                u132->hc_control);
+        switch (u132->hc_control & OHCI_CTRL_HCFS) {
+        case OHCI_USB_OPER:
+                sleep_time = 0;
+                break;
+        case OHCI_USB_SUSPEND:
+        case OHCI_USB_RESUME:
+                u132->hc_control &= OHCI_CTRL_RWC;
+                u132->hc_control |= OHCI_USB_RESUME;
+                sleep_time = 10;
+                break;
+        default:
+                u132->hc_control &= OHCI_CTRL_RWC;
+                u132->hc_control |= OHCI_USB_RESET;
+                sleep_time = 50;
+                break;
+        }
+        retval = u132_write_pcimem(u132, control, u132->hc_control);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, control, &control);
+        if (retval)
+                return retval;
+        msleep(sleep_time);
+        retval = u132_read_pcimem(u132, roothub.a, &roothub_a);
+        if (retval)
+                return retval;
+        if (!(roothub_a & RH_A_NPS)) {
+                int temp;        /* power down each port */
+                for (temp = 0; temp < u132->num_ports; temp++) {
+                        retval = u132_write_pcimem(u132,
+                                roothub.portstatus[temp], RH_PS_LSDA);
+                        if (retval)
+                                return retval;
+                }
+        }
+        retval = u132_read_pcimem(u132, control, &control);
+        if (retval)
+                return retval;
+      retry:retval = u132_read_pcimem(u132, cmdstatus, &status);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem_byte(u132, cmdstatus, OHCI_HCR);
+        if (retval)
+                return retval;
+      extra:{
+                retval = u132_read_pcimem(u132, cmdstatus, &status);
+                if (retval)
+                        return retval;
+                if (0 != (status & OHCI_HCR)) {
+                        if (--reset_timeout == 0) {
+                                dev_err(&u132->platform_dev->dev, "USB HC reset"
+                                        " timed out!\n");
+                                return -ENODEV;
+                        } else {
+                                msleep(5);
+                                goto extra;
+                        }
+                }
+        }
+        if (u132->flags & OHCI_QUIRK_INITRESET) {
+                retval = u132_write_pcimem(u132, control, u132->hc_control);
+                if (retval)
+                        return retval;
+                retval = u132_read_pcimem(u132, control, &control);
+                if (retval)
+                        return retval;
+        }
+        retval = u132_write_pcimem(u132, ed_controlhead, 0x00000000);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, ed_bulkhead, 0x11000000);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, hcca, 0x00000000);
+        if (retval)
+                return retval;
+        retval = u132_periodic_reinit(u132);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, fminterval, &fminterval);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, periodicstart, &periodicstart);
+        if (retval)
+                return retval;
+        if (0 == (fminterval & 0x3fff0000) || 0 == periodicstart) {
+                if (!(u132->flags & OHCI_QUIRK_INITRESET)) {
+                        u132->flags |= OHCI_QUIRK_INITRESET;
+                        goto retry;
+                } else
+                        dev_err(&u132->platform_dev->dev, "init err(%08x %04x)"
+                                "\n", fminterval, periodicstart);
+        }                        /* start controller operations */
+        u132->hc_control &= OHCI_CTRL_RWC;
+        u132->hc_control |= OHCI_CONTROL_INIT | OHCI_CTRL_BLE | OHCI_USB_OPER;
+        retval = u132_write_pcimem(u132, control, u132->hc_control);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem_byte(u132, cmdstatus, OHCI_BLF);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, cmdstatus, &cmdstatus);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, control, &control);
+        if (retval)
+                return retval;
+        u132_to_hcd(u132)->state = HC_STATE_RUNNING;
+        retval = u132_write_pcimem(u132, roothub.status, RH_HS_DRWE);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, intrstatus, mask);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, intrdisable,
+                OHCI_INTR_MIE | OHCI_INTR_OC | OHCI_INTR_RHSC | OHCI_INTR_FNO |
+                OHCI_INTR_UE | OHCI_INTR_RD | OHCI_INTR_SF | OHCI_INTR_WDH |
+                OHCI_INTR_SO);
+        if (retval)
+                return retval;        /* handle root hub init quirks ... */
+        retval = u132_read_pcimem(u132, roothub.a, &roothub_a);
+        if (retval)
+                return retval;
+        roothub_a &= ~(RH_A_PSM | RH_A_OCPM);
+        if (u132->flags & OHCI_QUIRK_SUPERIO) {
+                roothub_a |= RH_A_NOCP;
+                roothub_a &= ~(RH_A_POTPGT | RH_A_NPS);
+                retval = u132_write_pcimem(u132, roothub.a, roothub_a);
+                if (retval)
+                        return retval;
+        } else if ((u132->flags & OHCI_QUIRK_AMD756) || distrust_firmware) {
+                roothub_a |= RH_A_NPS;
+                retval = u132_write_pcimem(u132, roothub.a, roothub_a);
+                if (retval)
+                        return retval;
+        }
+        retval = u132_write_pcimem(u132, roothub.status, RH_HS_LPSC);
+        if (retval)
+                return retval;
+        retval = u132_write_pcimem(u132, roothub.b,
+                (roothub_a & RH_A_NPS) ? 0 : RH_B_PPCM);
+        if (retval)
+                return retval;
+        retval = u132_read_pcimem(u132, control, &control);
+        if (retval)
+                return retval;
+        mdelay((roothub_a >> 23) & 0x1fe);
+        u132_to_hcd(u132)->state = HC_STATE_RUNNING;
+        return 0;
+}
+
+static void u132_hcd_stop(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device hcd=%p is being remov"
+                        "ed\n", hcd);
+        } else {
+                down(&u132->sw_lock);
+                msleep(100);
+                u132_power(u132, 0);
+                up(&u132->sw_lock);
+        }
+}
+
+static int u132_hcd_start(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else if (hcd->self.controller) {
+                int retval;
+                struct platform_device *pdev =
+                        to_platform_device(hcd->self.controller);
+                u16 vendor = ((struct u132_platform_data *)
+                        (pdev->dev.platform_data))->vendor;
+                u16 device = ((struct u132_platform_data *)
+                        (pdev->dev.platform_data))->device;
+                down(&u132->sw_lock);
+                msleep(10);
+                if (vendor == PCI_VENDOR_ID_AMD && device == 0x740c) {
+                        u132->flags = OHCI_QUIRK_AMD756;
+                } else if (vendor == PCI_VENDOR_ID_OPTI && device == 0xc861) {
+                        dev_err(&u132->platform_dev->dev, "WARNING: OPTi workar"
+                                "ounds unavailable\n");
+                } else if (vendor == PCI_VENDOR_ID_COMPAQ && device == 0xa0f8)
+                        u132->flags |= OHCI_QUIRK_ZFMICRO;
+                retval = u132_run(u132);
+                if (retval) {
+                        u132_disable(u132);
+                        u132->going = 1;
+                }
+                msleep(100);
+                up(&u132->sw_lock);
+                return retval;
+        } else {
+                dev_err(&u132->platform_dev->dev, "platform_device missing\n");
+                return -ENODEV;
+        }
+}
+
+static int u132_hcd_reset(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else {
+                int retval;
+                down(&u132->sw_lock);
+                retval = u132_init(u132);
+                if (retval) {
+                        u132_disable(u132);
+                        u132->going = 1;
+                }
+                up(&u132->sw_lock);
+                return retval;
+        }
+}
+
+static int create_endpoint_and_queue_int(struct u132 *u132,
+        struct u132_udev *udev, struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, u8 usb_addr, u8 usb_endp, u8 address,
+        gfp_t mem_flags)
+{
+        struct u132_ring *ring;
+        unsigned long irqs;
+        u8 endp_number = ++u132->num_endpoints;
+        struct u132_endp *endp = hep->hcpriv = u132->endp[endp_number - 1] =
+                kmalloc(sizeof(struct u132_endp), mem_flags);
+        if (!endp) {
+                return -ENOMEM;
+        }
+        INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp);
+        spin_lock_init(&endp->queue_lock.slock);
+        INIT_LIST_HEAD(&endp->urb_more);
+        ring = endp->ring = &u132->ring[0];
+        if (ring->curr_endp) {
+                list_add_tail(&endp->endp_ring, &ring->curr_endp->endp_ring);
+        } else {
+                INIT_LIST_HEAD(&endp->endp_ring);
+                ring->curr_endp = endp;
+        }
+        ring->length += 1;
+        endp->dequeueing = 0;
+        endp->edset_flush = 0;
+        endp->active = 0;
+        endp->delayed = 0;
+        endp->endp_number = endp_number;
+        endp->u132 = u132;
+        endp->hep = hep;
+        endp->pipetype = usb_pipetype(urb->pipe);
+        u132_endp_init_kref(u132, endp);
+        if (usb_pipein(urb->pipe)) {
+                endp->toggle_bits = 0x2;
+                usb_settoggle(udev->usb_device, usb_endp, 0, 0);
+                endp->input = 1;
+                endp->output = 0;
+                udev->endp_number_in[usb_endp] = endp_number;
+                u132_udev_get_kref(u132, udev);
+        } else {
+                endp->toggle_bits = 0x2;
+                usb_settoggle(udev->usb_device, usb_endp, 1, 0);
+                endp->input = 0;
+                endp->output = 1;
+                udev->endp_number_out[usb_endp] = endp_number;
+                u132_udev_get_kref(u132, udev);
+        }
+        urb->hcpriv = u132;
+        spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+        endp->delayed = 1;
+        endp->jiffies = jiffies + msecs_to_jiffies(urb->interval);
+        endp->udev_number = address;
+        endp->usb_addr = usb_addr;
+        endp->usb_endp = usb_endp;
+        endp->queue_size = 1;
+        endp->queue_last = 0;
+        endp->queue_next = 0;
+        endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+        u132_endp_queue_work(u132, endp, msecs_to_jiffies(urb->interval));
+        return 0;
+}
+
+static int queue_int_on_old_endpoint(struct u132 *u132, struct u132_udev *udev,
+        struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, struct u132_endp *endp, u8 usb_addr,
+        u8 usb_endp, u8 address)
+{
+        urb->hcpriv = u132;
+        endp->delayed = 1;
+        endp->jiffies = jiffies + msecs_to_jiffies(urb->interval);
+        if (endp->queue_size++ < ENDP_QUEUE_SIZE) {
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+        } else {
+                struct u132_urbq *urbq = kmalloc(sizeof(struct u132_urbq),
+                        GFP_ATOMIC);
+                if (urbq == NULL) {
+                        endp->queue_size -= 1;
+                        return -ENOMEM;
+                } else {
+                        list_add_tail(&urbq->urb_more, &endp->urb_more);
+                        urbq->urb = urb;
+                }
+        }
+        return 0;
+}
+
+static int create_endpoint_and_queue_bulk(struct u132 *u132,
+        struct u132_udev *udev, struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, u8 usb_addr, u8 usb_endp, u8 address,
+        gfp_t mem_flags)
+{
+        int ring_number;
+        struct u132_ring *ring;
+        unsigned long irqs;
+        u8 endp_number = ++u132->num_endpoints;
+        struct u132_endp *endp = hep->hcpriv = u132->endp[endp_number - 1] =
+                kmalloc(sizeof(struct u132_endp), mem_flags);
+        if (!endp) {
+                return -ENOMEM;
+        }
+        INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp);
+        spin_lock_init(&endp->queue_lock.slock);
+        INIT_LIST_HEAD(&endp->urb_more);
+        endp->dequeueing = 0;
+        endp->edset_flush = 0;
+        endp->active = 0;
+        endp->delayed = 0;
+        endp->endp_number = endp_number;
+        endp->u132 = u132;
+        endp->hep = hep;
+        endp->pipetype = usb_pipetype(urb->pipe);
+        u132_endp_init_kref(u132, endp);
+        if (usb_pipein(urb->pipe)) {
+                endp->toggle_bits = 0x2;
+                usb_settoggle(udev->usb_device, usb_endp, 0, 0);
+                ring_number = 3;
+                endp->input = 1;
+                endp->output = 0;
+                udev->endp_number_in[usb_endp] = endp_number;
+                u132_udev_get_kref(u132, udev);
+        } else {
+                endp->toggle_bits = 0x2;
+                usb_settoggle(udev->usb_device, usb_endp, 1, 0);
+                ring_number = 2;
+                endp->input = 0;
+                endp->output = 1;
+                udev->endp_number_out[usb_endp] = endp_number;
+                u132_udev_get_kref(u132, udev);
+        }
+        ring = endp->ring = &u132->ring[ring_number - 1];
+        if (ring->curr_endp) {
+                list_add_tail(&endp->endp_ring, &ring->curr_endp->endp_ring);
+        } else {
+                INIT_LIST_HEAD(&endp->endp_ring);
+                ring->curr_endp = endp;
+        }
+        ring->length += 1;
+        urb->hcpriv = u132;
+        spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+        endp->udev_number = address;
+        endp->usb_addr = usb_addr;
+        endp->usb_endp = usb_endp;
+        endp->queue_size = 1;
+        endp->queue_last = 0;
+        endp->queue_next = 0;
+        endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+        u132_endp_queue_work(u132, endp, 0);
+        return 0;
+}
+
+static int queue_bulk_on_old_endpoint(struct u132 *u132, struct u132_udev *udev,
+         struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, struct u132_endp *endp, u8 usb_addr,
+        u8 usb_endp, u8 address)
+{
+        urb->hcpriv = u132;
+        if (endp->queue_size++ < ENDP_QUEUE_SIZE) {
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+        } else {
+                struct u132_urbq *urbq = kmalloc(sizeof(struct u132_urbq),
+                        GFP_ATOMIC);
+                if (urbq == NULL) {
+                        endp->queue_size -= 1;
+                        return -ENOMEM;
+                } else {
+                        list_add_tail(&urbq->urb_more, &endp->urb_more);
+                        urbq->urb = urb;
+                }
+        }
+        return 0;
+}
+
+static int create_endpoint_and_queue_control(struct u132 *u132,
+        struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, u8 usb_addr, u8 usb_endp,
+        gfp_t mem_flags)
+{
+        struct u132_ring *ring;
+        u8 endp_number = ++u132->num_endpoints;
+        struct u132_endp *endp = hep->hcpriv = u132->endp[endp_number - 1] =
+                kmalloc(sizeof(struct u132_endp), mem_flags);
+        if (!endp) {
+                return -ENOMEM;
+        }
+        INIT_WORK(&endp->scheduler, u132_hcd_endp_work_scheduler, (void *)endp);
+        spin_lock_init(&endp->queue_lock.slock);
+        INIT_LIST_HEAD(&endp->urb_more);
+        ring = endp->ring = &u132->ring[0];
+        if (ring->curr_endp) {
+                list_add_tail(&endp->endp_ring, &ring->curr_endp->endp_ring);
+        } else {
+                INIT_LIST_HEAD(&endp->endp_ring);
+                ring->curr_endp = endp;
+        }
+        ring->length += 1;
+        endp->dequeueing = 0;
+        endp->edset_flush = 0;
+        endp->active = 0;
+        endp->delayed = 0;
+        endp->endp_number = endp_number;
+        endp->u132 = u132;
+        endp->hep = hep;
+        u132_endp_init_kref(u132, endp);
+        u132_endp_get_kref(u132, endp);
+        if (usb_addr == 0) {
+                unsigned long irqs;
+                u8 address = u132->addr[usb_addr].address;
+                struct u132_udev *udev = &u132->udev[address];
+                endp->udev_number = address;
+                endp->usb_addr = usb_addr;
+                endp->usb_endp = usb_endp;
+                endp->input = 1;
+                endp->output = 1;
+                endp->pipetype = usb_pipetype(urb->pipe);
+                u132_udev_init_kref(u132, udev);
+                u132_udev_get_kref(u132, udev);
+                udev->endp_number_in[usb_endp] = endp_number;
+                udev->endp_number_out[usb_endp] = endp_number;
+                urb->hcpriv = u132;
+                spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+                endp->queue_size = 1;
+                endp->queue_last = 0;
+                endp->queue_next = 0;
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                u132_endp_queue_work(u132, endp, 0);
+                return 0;
+        } else {                /*(usb_addr > 0) */
+                unsigned long irqs;
+                u8 address = u132->addr[usb_addr].address;
+                struct u132_udev *udev = &u132->udev[address];
+                endp->udev_number = address;
+                endp->usb_addr = usb_addr;
+                endp->usb_endp = usb_endp;
+                endp->input = 1;
+                endp->output = 1;
+                endp->pipetype = usb_pipetype(urb->pipe);
+                u132_udev_get_kref(u132, udev);
+                udev->enumeration = 2;
+                udev->endp_number_in[usb_endp] = endp_number;
+                udev->endp_number_out[usb_endp] = endp_number;
+                urb->hcpriv = u132;
+                spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+                endp->queue_size = 1;
+                endp->queue_last = 0;
+                endp->queue_next = 0;
+                endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] = urb;
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                u132_endp_queue_work(u132, endp, 0);
+                return 0;
+        }
+}
+
+static int queue_control_on_old_endpoint(struct u132 *u132,
+        struct usb_host_endpoint *hep, struct urb *urb,
+        struct usb_device *usb_dev, struct u132_endp *endp, u8 usb_addr,
+        u8 usb_endp)
+{
+        if (usb_addr == 0) {
+                if (usb_pipein(urb->pipe)) {
+                        urb->hcpriv = u132;
+                        if (endp->queue_size++ < ENDP_QUEUE_SIZE) {
+                                endp->urb_list[ENDP_QUEUE_MASK &
+                                        endp->queue_last++] = urb;
+                        } else {
+                                struct u132_urbq *urbq =
+                                        kmalloc(sizeof(struct u132_urbq),
+                                        GFP_ATOMIC);
+                                if (urbq == NULL) {
+                                        endp->queue_size -= 1;
+                                        return -ENOMEM;
+                                } else {
+                                        list_add_tail(&urbq->urb_more,
+                                                &endp->urb_more);
+                                        urbq->urb = urb;
+                                }
+                        }
+                        return 0;
+                } else {        /* usb_pipeout(urb->pipe) */
+                        struct u132_addr *addr = &u132->addr[usb_dev->devnum];
+                        int I = MAX_U132_UDEVS;
+                        int i = 0;
+                        while (--I > 0) {
+                                struct u132_udev *udev = &u132->udev[++i];
+                                if (udev->usb_device) {
+                                        continue;
+                                } else {
+                                        udev->enumeration = 1;
+                                        u132->addr[0].address = i;
+                                        endp->udev_number = i;
+                                        udev->udev_number = i;
+                                        udev->usb_addr = usb_dev->devnum;
+                                        u132_udev_init_kref(u132, udev);
+                                        udev->endp_number_in[usb_endp] =
+                                                endp->endp_number;
+                                        u132_udev_get_kref(u132, udev);
+                                        udev->endp_number_out[usb_endp] =
+                                                endp->endp_number;
+                                        udev->usb_device = usb_dev;
+                                        ((u8 *) (urb->setup_packet))[2] =
+                                                addr->address = i;
+                                        u132_udev_get_kref(u132, udev);
+                                        break;
+                                }
+                        }
+                        if (I == 0) {
+                                dev_err(&u132->platform_dev->dev, "run out of d"
+                                        "evice space\n");
+                                return -EINVAL;
+                        }
+                        urb->hcpriv = u132;
+                        if (endp->queue_size++ < ENDP_QUEUE_SIZE) {
+                                endp->urb_list[ENDP_QUEUE_MASK &
+                                        endp->queue_last++] = urb;
+                        } else {
+                                struct u132_urbq *urbq =
+                                        kmalloc(sizeof(struct u132_urbq),
+                                        GFP_ATOMIC);
+                                if (urbq == NULL) {
+                                        endp->queue_size -= 1;
+                                        return -ENOMEM;
+                                } else {
+                                        list_add_tail(&urbq->urb_more,
+                                                &endp->urb_more);
+                                        urbq->urb = urb;
+                                }
+                        }
+                        return 0;
+                }
+        } else {                /*(usb_addr > 0) */
+                u8 address = u132->addr[usb_addr].address;
+                struct u132_udev *udev = &u132->udev[address];
+                urb->hcpriv = u132;
+                if (udev->enumeration == 2) {
+                } else
+                        udev->enumeration = 2;
+                if (endp->queue_size++ < ENDP_QUEUE_SIZE) {
+                        endp->urb_list[ENDP_QUEUE_MASK & endp->queue_last++] =
+                                urb;
+                } else {
+                        struct u132_urbq *urbq =
+                                kmalloc(sizeof(struct u132_urbq), GFP_ATOMIC);
+                        if (urbq == NULL) {
+                                endp->queue_size -= 1;
+                                return -ENOMEM;
+                        } else {
+                                list_add_tail(&urbq->urb_more, &endp->urb_more);
+                                urbq->urb = urb;
+                        }
+                }
+                return 0;
+        }
+}
+
+static int u132_urb_enqueue(struct usb_hcd *hcd, struct usb_host_endpoint *hep,
+        struct urb *urb, gfp_t mem_flags)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (irqs_disabled()) {
+                if (__GFP_WAIT & mem_flags) {
+                        printk(KERN_ERR "invalid context for function that migh"
+                                "t sleep\n");
+                        return -EINVAL;
+                }
+        }
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed urb="
+                        "%p status=%d\n", urb, urb->status);
+                return -ESHUTDOWN;
+        } else {
+                u8 usb_addr = usb_pipedevice(urb->pipe);
+                u8 usb_endp = usb_pipeendpoint(urb->pipe);
+                struct usb_device *usb_dev = urb->dev;
+                if (usb_pipetype(urb->pipe) == PIPE_INTERRUPT) {
+                        u8 address = u132->addr[usb_addr].address;
+                        struct u132_udev *udev = &u132->udev[address];
+                        struct u132_endp *endp = hep->hcpriv;
+                        urb->actual_length = 0;
+                        if (endp) {
+                                unsigned long irqs;
+                                int retval;
+                                spin_lock_irqsave(&endp->queue_lock.slock,
+                                        irqs);
+                                retval = queue_int_on_old_endpoint(u132, udev,
+                                        hep, urb, usb_dev, endp, usb_addr,
+                                        usb_endp, address);
+                                spin_unlock_irqrestore(&endp->queue_lock.slock,
+                                        irqs);
+                                if (retval) {
+                                        return retval;
+                                } else {
+                                        u132_endp_queue_work(u132, endp,
+                                                msecs_to_jiffies(urb->interval))
+                                                ;
+                                        return 0;
+                                }
+                        } else if (u132->num_endpoints == MAX_U132_ENDPS) {
+                                return -EINVAL;
+                        } else {        /*(endp == NULL) */
+                                return create_endpoint_and_queue_int(u132, udev,
+                                         hep, urb, usb_dev, usb_addr, usb_endp,
+                                        address, mem_flags);
+                        }
+                } else if (usb_pipetype(urb->pipe) == PIPE_ISOCHRONOUS) {
+                        dev_err(&u132->platform_dev->dev, "the hardware does no"
+                                "t support PIPE_ISOCHRONOUS\n");
+                        return -EINVAL;
+                } else if (usb_pipetype(urb->pipe) == PIPE_BULK) {
+                        u8 address = u132->addr[usb_addr].address;
+                        struct u132_udev *udev = &u132->udev[address];
+                        struct u132_endp *endp = hep->hcpriv;
+                        urb->actual_length = 0;
+                        if (endp) {
+                                unsigned long irqs;
+                                int retval;
+                                spin_lock_irqsave(&endp->queue_lock.slock,
+                                        irqs);
+                                retval = queue_bulk_on_old_endpoint(u132, udev,
+                                        hep, urb, usb_dev, endp, usb_addr,
+                                        usb_endp, address);
+                                spin_unlock_irqrestore(&endp->queue_lock.slock,
+                                        irqs);
+                                if (retval) {
+                                        return retval;
+                                } else {
+                                        u132_endp_queue_work(u132, endp, 0);
+                                        return 0;
+                                }
+                        } else if (u132->num_endpoints == MAX_U132_ENDPS) {
+                                return -EINVAL;
+                        } else
+                                return create_endpoint_and_queue_bulk(u132,
+                                        udev, hep, urb, usb_dev, usb_addr,
+                                        usb_endp, address, mem_flags);
+                } else {
+                        struct u132_endp *endp = hep->hcpriv;
+                        u16 urb_size = 8;
+                        u8 *b = urb->setup_packet;
+                        int i = 0;
+                        char data[30 *3 + 4];
+                        char *d = data;
+                        int m = (sizeof(data) - 1) / 3;
+                        int l = 0;
+                        data[0] = 0;
+                        while (urb_size-- > 0) {
+                                if (i > m) {
+                                } else if (i++ < m) {
+                                        int w = sprintf(d, " %02X", *b++);
+                                        d += w;
+                                        l += w;
+                                } else
+                                        d += sprintf(d, " ..");
+                        }
+                        if (endp) {
+                                unsigned long irqs;
+                                int retval;
+                                spin_lock_irqsave(&endp->queue_lock.slock,
+                                        irqs);
+                                retval = queue_control_on_old_endpoint(u132,
+                                        hep, urb, usb_dev, endp, usb_addr,
+                                        usb_endp);
+                                spin_unlock_irqrestore(&endp->queue_lock.slock,
+                                        irqs);
+                                if (retval) {
+                                        return retval;
+                                } else {
+                                        u132_endp_queue_work(u132, endp, 0);
+                                        return 0;
+                                }
+                        } else if (u132->num_endpoints == MAX_U132_ENDPS) {
+                                return -EINVAL;
+                        } else
+                                return create_endpoint_and_queue_control(u132,
+                                        hep, urb, usb_dev, usb_addr, usb_endp,
+                                        mem_flags);
+                }
+        }
+}
+
+static int dequeue_from_overflow_chain(struct u132 *u132,
+        struct u132_endp *endp, struct urb *urb)
+{
+        struct list_head *scan;
+        struct list_head *head = &endp->urb_more;
+        list_for_each(scan, head) {
+                struct u132_urbq *urbq = list_entry(scan, struct u132_urbq,
+                        urb_more);
+                if (urbq->urb == urb) {
+                        struct usb_hcd *hcd = u132_to_hcd(u132);
+                        list_del(scan);
+                        endp->queue_size -= 1;
+                        urb->error_count = 0;
+                        urb->hcpriv = NULL;
+                        usb_hcd_giveback_urb(hcd, urb, NULL);
+                        return 0;
+                } else
+                        continue;
+        }
+        dev_err(&u132->platform_dev->dev, "urb=%p not found in endp[%d]=%p ring"
+                "[%d] %c%c usb_endp=%d usb_addr=%d size=%d next=%04X last=%04X"
+                "\n", urb, endp->endp_number, endp, endp->ring->number,
+                endp->input ? 'I' : ' ', endp->output ? 'O' : ' ',
+                endp->usb_endp, endp->usb_addr, endp->queue_size,
+                endp->queue_next, endp->queue_last);
+        return -EINVAL;
+}
+
+static int u132_endp_urb_dequeue(struct u132 *u132, struct u132_endp *endp,
+        struct urb *urb)
+{
+        unsigned long irqs;
+        spin_lock_irqsave(&endp->queue_lock.slock, irqs);
+        if (endp->queue_size == 0) {
+                dev_err(&u132->platform_dev->dev, "urb=%p not found in endp[%d]"
+                        "=%p ring[%d] %c%c usb_endp=%d usb_addr=%d\n", urb,
+                        endp->endp_number, endp, endp->ring->number,
+                        endp->input ? 'I' : ' ', endp->output ? 'O' : ' ',
+                        endp->usb_endp, endp->usb_addr);
+                spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                return -EINVAL;
+        }
+        if (urb == endp->urb_list[ENDP_QUEUE_MASK & endp->queue_next]) {
+                if (endp->active) {
+                        endp->dequeueing = 1;
+                        endp->edset_flush = 1;
+                        u132_endp_queue_work(u132, endp, 0);
+                        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                        urb->hcpriv = NULL;
+                        return 0;
+                } else {
+                        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                        u132_hcd_abandon_urb(u132, endp, urb, urb->status);
+                        return 0;
+                }
+        } else {
+                u16 queue_list = 0;
+                u16 queue_size = endp->queue_size;
+                u16 queue_scan = endp->queue_next;
+                struct urb **urb_slot = NULL;
+                while (++queue_list < ENDP_QUEUE_SIZE && --queue_size > 0) {
+                        if (urb == endp->urb_list[ENDP_QUEUE_MASK &
+                                ++queue_scan]) {
+                                urb_slot = &endp->urb_list[ENDP_QUEUE_MASK &
+                                        queue_scan];
+                                break;
+                        } else
+                                continue;
+                }
+                while (++queue_list < ENDP_QUEUE_SIZE && --queue_size > 0) {
+                        *urb_slot = endp->urb_list[ENDP_QUEUE_MASK &
+                                ++queue_scan];
+                        urb_slot = &endp->urb_list[ENDP_QUEUE_MASK &
+                                queue_scan];
+                }
+                if (urb_slot) {
+                        struct usb_hcd *hcd = u132_to_hcd(u132);
+                        endp->queue_size -= 1;
+                        if (list_empty(&endp->urb_more)) {
+                                spin_unlock_irqrestore(&endp->queue_lock.slock,
+                                        irqs);
+                        } else {
+                                struct list_head *next = endp->urb_more.next;
+                                struct u132_urbq *urbq = list_entry(next,
+                                        struct u132_urbq, urb_more);
+                                list_del(next);
+                                *urb_slot = urbq->urb;
+                                spin_unlock_irqrestore(&endp->queue_lock.slock,
+                                        irqs);
+                                kfree(urbq);
+                        } urb->error_count = 0;
+                        urb->hcpriv = NULL;
+                        usb_hcd_giveback_urb(hcd, urb, NULL);
+                        return 0;
+                } else if (list_empty(&endp->urb_more)) {
+                        dev_err(&u132->platform_dev->dev, "urb=%p not found in "
+                                "endp[%d]=%p ring[%d] %c%c usb_endp=%d usb_addr"
+                                "=%d size=%d next=%04X last=%04X\n", urb,
+                                endp->endp_number, endp, endp->ring->number,
+                                endp->input ? 'I' : ' ',
+                                endp->output ? 'O' : ' ', endp->usb_endp,
+                                endp->usb_addr, endp->queue_size,
+                                endp->queue_next, endp->queue_last);
+                        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                        return -EINVAL;
+                } else {
+                        int retval = dequeue_from_overflow_chain(u132, endp,
+                                urb);
+                        spin_unlock_irqrestore(&endp->queue_lock.slock, irqs);
+                        return retval;
+                }
+        }
+}
+
+static int u132_urb_dequeue(struct usb_hcd *hcd, struct urb *urb)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 2) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else {
+                u8 usb_addr = usb_pipedevice(urb->pipe);
+                u8 usb_endp = usb_pipeendpoint(urb->pipe);
+                u8 address = u132->addr[usb_addr].address;
+                struct u132_udev *udev = &u132->udev[address];
+                if (usb_pipein(urb->pipe)) {
+                        u8 endp_number = udev->endp_number_in[usb_endp];
+                        struct u132_endp *endp = u132->endp[endp_number - 1];
+                        return u132_endp_urb_dequeue(u132, endp, urb);
+                } else {
+                        u8 endp_number = udev->endp_number_out[usb_endp];
+                        struct u132_endp *endp = u132->endp[endp_number - 1];
+                        return u132_endp_urb_dequeue(u132, endp, urb);
+                }
+        }
+}
+
+static void u132_endpoint_disable(struct usb_hcd *hcd,
+        struct usb_host_endpoint *hep)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 2) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+        } else {
+                struct u132_endp *endp = hep->hcpriv;
+                if (endp)
+                        u132_endp_put_kref(u132, endp);
+        }
+}
+
+static int u132_get_frame(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else {
+                int frame = 0;
+                dev_err(&u132->platform_dev->dev, "TODO: u132_get_frame\n");
+                msleep(100);
+                return frame;
+        }
+}
+
+static int u132_roothub_descriptor(struct u132 *u132,
+        struct usb_hub_descriptor *desc)
+{
+        int retval;
+        u16 temp;
+        u32 rh_a = -1;
+        u32 rh_b = -1;
+        retval = u132_read_pcimem(u132, roothub.a, &rh_a);
+        if (retval)
+                return retval;
+        desc->bDescriptorType = 0x29;
+        desc->bPwrOn2PwrGood = (rh_a & RH_A_POTPGT) >> 24;
+        desc->bHubContrCurrent = 0;
+        desc->bNbrPorts = u132->num_ports;
+        temp = 1 + (u132->num_ports / 8);
+        desc->bDescLength = 7 + 2 *temp;
+        temp = 0;
+        if (rh_a & RH_A_NPS)
+                temp |= 0x0002;
+        if (rh_a & RH_A_PSM)
+                temp |= 0x0001;
+        if (rh_a & RH_A_NOCP) {
+                temp |= 0x0010;
+        } else if (rh_a & RH_A_OCPM)
+                temp |= 0x0008;
+        desc->wHubCharacteristics = cpu_to_le16(temp);
+        retval = u132_read_pcimem(u132, roothub.b, &rh_b);
+        if (retval)
+                return retval;
+        memset(desc->bitmap, 0xff, sizeof(desc->bitmap));
+        desc->bitmap[0] = rh_b & RH_B_DR;
+        if (u132->num_ports > 7) {
+                desc->bitmap[1] = (rh_b & RH_B_DR) >> 8;
+                desc->bitmap[2] = 0xff;
+        } else
+                desc->bitmap[1] = 0xff;
+        return 0;
+}
+
+static int u132_roothub_status(struct u132 *u132, __le32 *desc)
+{
+        u32 rh_status = -1;
+        int ret_status = u132_read_pcimem(u132, roothub.status, &rh_status);
+        *desc = cpu_to_le32(rh_status);
+        return ret_status;
+}
+
+static int u132_roothub_portstatus(struct u132 *u132, __le32 *desc, u16 wIndex)
+{
+        if (wIndex == 0 || wIndex > u132->num_ports) {
+                return -EINVAL;
+        } else {
+                int port = wIndex - 1;
+                u32 rh_portstatus = -1;
+                int ret_portstatus = u132_read_pcimem(u132,
+                        roothub.portstatus[port], &rh_portstatus);
+                *desc = cpu_to_le32(rh_portstatus);
+                if (*(u16 *) (desc + 2)) {
+                        dev_info(&u132->platform_dev->dev, "Port %d Status Chan"
+                                "ge = %08X\n", port, *desc);
+                }
+                return ret_portstatus;
+        }
+}
+
+
+/* this timer value might be vendor-specific ... */
+#define PORT_RESET_HW_MSEC 10
+#define PORT_RESET_MSEC 10
+/* wrap-aware logic morphed from <linux/jiffies.h> */
+#define tick_before(t1, t2) ((s16)(((s16)(t1))-((s16)(t2))) < 0)
+static int u132_roothub_portreset(struct u132 *u132, int port_index)
+{
+        int retval;
+        u32 fmnumber;
+        u16 now;
+        u16 reset_done;
+        retval = u132_read_pcimem(u132, fmnumber, &fmnumber);
+        if (retval)
+                return retval;
+        now = fmnumber;
+        reset_done = now + PORT_RESET_MSEC;
+        do {
+                u32 portstat;
+                do {
+                        retval = u132_read_pcimem(u132,
+                                roothub.portstatus[port_index], &portstat);
+                        if (retval)
+                                return retval;
+                        if (RH_PS_PRS & portstat) {
+                                continue;
+                        } else
+                                break;
+                } while (tick_before(now, reset_done));
+                if (RH_PS_PRS & portstat)
+                        return -ENODEV;
+                if (RH_PS_CCS & portstat) {
+                        if (RH_PS_PRSC & portstat) {
+                                retval = u132_write_pcimem(u132,
+                                        roothub.portstatus[port_index],
+                                        RH_PS_PRSC);
+                                if (retval)
+                                        return retval;
+                        }
+                } else
+                        break;        /* start the next reset,
+                                sleep till it's probably done */
+                retval = u132_write_pcimem(u132, roothub.portstatus[port_index],
+                         RH_PS_PRS);
+                if (retval)
+                        return retval;
+                msleep(PORT_RESET_HW_MSEC);
+                retval = u132_read_pcimem(u132, fmnumber, &fmnumber);
+                if (retval)
+                        return retval;
+                now = fmnumber;
+        } while (tick_before(now, reset_done));
+        return 0;
+}
+
+static int u132_roothub_setportfeature(struct u132 *u132, u16 wValue,
+        u16 wIndex)
+{
+        if (wIndex == 0 || wIndex > u132->num_ports) {
+                return -EINVAL;
+        } else {
+                int retval;
+                int port_index = wIndex - 1;
+                struct u132_port *port = &u132->port[port_index];
+                port->Status &= ~(1 << wValue);
+                switch (wValue) {
+                case USB_PORT_FEAT_SUSPEND:
+                        retval = u132_write_pcimem(u132,
+                                roothub.portstatus[port_index], RH_PS_PSS);
+                        if (retval)
+                                return retval;
+                        return 0;
+                case USB_PORT_FEAT_POWER:
+                        retval = u132_write_pcimem(u132,
+                                roothub.portstatus[port_index], RH_PS_PPS);
+                        if (retval)
+                                return retval;
+                        return 0;
+                case USB_PORT_FEAT_RESET:
+                        retval = u132_roothub_portreset(u132, port_index);
+                        if (retval)
+                                return retval;
+                        return 0;
+                default:
+                        return -EPIPE;
+                }
+        }
+}
+
+static int u132_roothub_clearportfeature(struct u132 *u132, u16 wValue,
+        u16 wIndex)
+{
+        if (wIndex == 0 || wIndex > u132->num_ports) {
+                return -EINVAL;
+        } else {
+                int port_index = wIndex - 1;
+                u32 temp;
+                int retval;
+                struct u132_port *port = &u132->port[port_index];
+                port->Status &= ~(1 << wValue);
+                switch (wValue) {
+                case USB_PORT_FEAT_ENABLE:
+                        temp = RH_PS_CCS;
+                        break;
+                case USB_PORT_FEAT_C_ENABLE:
+                        temp = RH_PS_PESC;
+                        break;
+                case USB_PORT_FEAT_SUSPEND:
+                        temp = RH_PS_POCI;
+                        if ((u132->hc_control & OHCI_CTRL_HCFS)
+                                != OHCI_USB_OPER) {
+                                dev_err(&u132->platform_dev->dev, "TODO resume_"
+                                        "root_hub\n");
+                        }
+                        break;
+                case USB_PORT_FEAT_C_SUSPEND:
+                        temp = RH_PS_PSSC;
+                        break;
+                case USB_PORT_FEAT_POWER:
+                        temp = RH_PS_LSDA;
+                        break;
+                case USB_PORT_FEAT_C_CONNECTION:
+                        temp = RH_PS_CSC;
+                        break;
+                case USB_PORT_FEAT_C_OVER_CURRENT:
+                        temp = RH_PS_OCIC;
+                        break;
+                case USB_PORT_FEAT_C_RESET:
+                        temp = RH_PS_PRSC;
+                        break;
+                default:
+                        return -EPIPE;
+                }
+                retval = u132_write_pcimem(u132, roothub.portstatus[port_index],
+                         temp);
+                if (retval)
+                        return retval;
+                return 0;
+        }
+}
+
+
+/* the virtual root hub timer IRQ checks for hub status*/
+static int u132_hub_status_data(struct usb_hcd *hcd, char *buf)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device hcd=%p has been remov"
+                        "ed %d\n", hcd, u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device hcd=%p is being remov"
+                        "ed\n", hcd);
+                dump_stack();
+                return -ESHUTDOWN;
+        } else {
+                int i, changed = 0, length = 1;
+                if (u132->flags & OHCI_QUIRK_AMD756) {
+                        if ((u132->hc_roothub_a & RH_A_NDP) > MAX_ROOT_PORTS) {
+                                dev_err(&u132->platform_dev->dev, "bogus NDP, r"
+                                        "ereads as NDP=%d\n",
+                                        u132->hc_roothub_a & RH_A_NDP);
+                                goto done;
+                        }
+                }
+                if (u132->hc_roothub_status & (RH_HS_LPSC | RH_HS_OCIC)) {
+                        buf[0] = changed = 1;
+                } else
+                        buf[0] = 0;
+                if (u132->num_ports > 7) {
+                        buf[1] = 0;
+                        length++;
+                }
+                for (i = 0; i < u132->num_ports; i++) {
+                        if (u132->hc_roothub_portstatus[i] & (RH_PS_CSC |
+                                RH_PS_PESC | RH_PS_PSSC | RH_PS_OCIC |
+                                RH_PS_PRSC)) {
+                                changed = 1;
+                                if (i < 7) {
+                                        buf[0] |= 1 << (i + 1);
+                                } else
+                                        buf[1] |= 1 << (i - 7);
+                                continue;
+                        }
+                        if (!(u132->hc_roothub_portstatus[i] & RH_PS_CCS)) {
+                                continue;
+                        }
+                        if ((u132->hc_roothub_portstatus[i] & RH_PS_PSS)) {
+                                continue;
+                        }
+                }
+              done:return changed ? length : 0;
+        }
+}
+
+static int u132_hub_control(struct usb_hcd *hcd, u16 typeReq, u16 wValue,
+        u16 wIndex, char *buf, u16 wLength)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else {
+                int retval = 0;
+                down(&u132->sw_lock);
+                switch (typeReq) {
+                case ClearHubFeature:
+                        switch (wValue) {
+                        case C_HUB_OVER_CURRENT:
+                        case C_HUB_LOCAL_POWER:
+                                break;
+                        default:
+                                goto stall;
+                        }
+                        break;
+                case SetHubFeature:
+                        switch (wValue) {
+                        case C_HUB_OVER_CURRENT:
+                        case C_HUB_LOCAL_POWER:
+                                break;
+                        default:
+                                goto stall;
+                        }
+                        break;
+                case ClearPortFeature:{
+                                retval = u132_roothub_clearportfeature(u132,
+                                        wValue, wIndex);
+                                if (retval)
+                                        goto error;
+                                break;
+                        }
+                case GetHubDescriptor:{
+                                retval = u132_roothub_descriptor(u132,
+                                        (struct usb_hub_descriptor *)buf);
+                                if (retval)
+                                        goto error;
+                                break;
+                        }
+                case GetHubStatus:{
+                                retval = u132_roothub_status(u132,
+                                        (__le32 *) buf);
+                                if (retval)
+                                        goto error;
+                                break;
+                        }
+                case GetPortStatus:{
+                                retval = u132_roothub_portstatus(u132,
+                                        (__le32 *) buf, wIndex);
+                                if (retval)
+                                        goto error;
+                                break;
+                        }
+                case SetPortFeature:{
+                                retval = u132_roothub_setportfeature(u132,
+                                        wValue, wIndex);
+                                if (retval)
+                                        goto error;
+                                break;
+                        }
+                default:
+                        goto stall;
+                      error:u132_disable(u132);
+                        u132->going = 1;
+                        break;
+                      stall:retval = -EPIPE;
+                        break;
+                }
+                up(&u132->sw_lock);
+                return retval;
+        }
+}
+
+static int u132_start_port_reset(struct usb_hcd *hcd, unsigned port_num)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else
+                return 0;
+}
+
+static void u132_hub_irq_enable(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+        } else if (u132->going > 0)
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+}
+
+
+#ifdef CONFIG_PM
+static int u132_hcd_suspend(struct usb_hcd *hcd, pm_message_t message)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else
+                return 0;
+}
+
+static int u132_hcd_resume(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else
+                return 0;
+}
+
+static int u132_bus_suspend(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else
+                return 0;
+}
+
+static int u132_bus_resume(struct usb_hcd *hcd)
+{
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else
+                return 0;
+}
+
+#else
+#define u132_hcd_suspend NULL
+#define u132_hcd_resume NULL
+#define u132_bus_suspend NULL
+#define u132_bus_resume NULL
+#endif
+static struct hc_driver u132_hc_driver = {
+        .description = hcd_name,
+        .hcd_priv_size = sizeof(struct u132),
+        .irq = NULL,
+        .flags = HCD_USB11 | HCD_MEMORY,
+        .reset = u132_hcd_reset,
+        .start = u132_hcd_start,
+        .suspend = u132_hcd_suspend,
+        .resume = u132_hcd_resume,
+        .stop = u132_hcd_stop,
+        .urb_enqueue = u132_urb_enqueue,
+        .urb_dequeue = u132_urb_dequeue,
+        .endpoint_disable = u132_endpoint_disable,
+        .get_frame_number = u132_get_frame,
+        .hub_status_data = u132_hub_status_data,
+        .hub_control = u132_hub_control,
+        .bus_suspend = u132_bus_suspend,
+        .bus_resume = u132_bus_resume,
+        .start_port_reset = u132_start_port_reset,
+        .hub_irq_enable = u132_hub_irq_enable,
+};
+
+/*
+* This function may be called by the USB core whilst the "usb_all_devices_rwsem"
+* is held for writing, thus this module must not call usb_remove_hcd()
+* synchronously - but instead should immediately stop activity to the
+* device and ansynchronously call usb_remove_hcd()
+*/
+static int __devexit u132_remove(struct platform_device *pdev)
+{
+        struct usb_hcd *hcd = platform_get_drvdata(pdev);
+        if (hcd) {
+                struct u132 *u132 = hcd_to_u132(hcd);
+                dump_stack();
+                if (u132->going++ > 1) {
+                        return -ENODEV;
+                } else {
+                        int rings = MAX_U132_RINGS;
+                        int endps = MAX_U132_ENDPS;
+                        msleep(100);
+                        down(&u132->sw_lock);
+                        u132_monitor_cancel_work(u132);
+                        while (rings-- > 0) {
+                                struct u132_ring *ring = &u132->ring[rings];
+                                u132_ring_cancel_work(u132, ring);
+                        } while (endps-- > 0) {
+                                struct u132_endp *endp = u132->endp[endps];
+                                if (endp)
+                                        u132_endp_cancel_work(u132, endp);
+                        }
+                        u132->going += 1;
+                        printk(KERN_INFO "removing device u132.%d\n",
+                                u132->sequence_num);
+                        up(&u132->sw_lock);
+                        usb_remove_hcd(hcd);
+                        u132_u132_put_kref(u132);
+                        return 0;
+                }
+        } else
+                return 0;
+}
+
+static void u132_initialise(struct u132 *u132, struct platform_device *pdev)
+{
+        int rings = MAX_U132_RINGS;
+        int ports = MAX_U132_PORTS;
+        int addrs = MAX_U132_ADDRS;
+        int udevs = MAX_U132_UDEVS;
+        int endps = MAX_U132_ENDPS;
+        u132->board = pdev->dev.platform_data;
+        u132->platform_dev = pdev;
+        u132->power = 0;
+        u132->reset = 0;
+        init_MUTEX(&u132->sw_lock);
+        init_MUTEX(&u132->scheduler_lock);
+        while (rings-- > 0) {
+                struct u132_ring *ring = &u132->ring[rings];
+                ring->u132 = u132;
+                ring->number = rings + 1;
+                ring->length = 0;
+                ring->curr_endp = NULL;
+                INIT_WORK(&ring->scheduler, u132_hcd_ring_work_scheduler,
+                        (void *)ring);
+        } down(&u132->sw_lock);
+        INIT_WORK(&u132->monitor, u132_hcd_monitor_work, (void *)u132);
+        while (ports-- > 0) {
+                struct u132_port *port = &u132->port[ports];
+                port->u132 = u132;
+                port->reset = 0;
+                port->enable = 0;
+                port->power = 0;
+                port->Status = 0;
+        } while (addrs-- > 0) {
+                struct u132_addr *addr = &u132->addr[addrs];
+                addr->address = 0;
+        } while (udevs-- > 0) {
+                struct u132_udev *udev = &u132->udev[udevs];
+                int i = ARRAY_SIZE(udev->endp_number_in);
+                int o = ARRAY_SIZE(udev->endp_number_out);
+                udev->usb_device = NULL;
+                udev->udev_number = 0;
+                udev->usb_addr = 0;
+                udev->portnumber = 0;
+                while (i-- > 0) {
+                        udev->endp_number_in[i] = 0;
+                }
+                while (o-- > 0) {
+                        udev->endp_number_out[o] = 0;
+                }
+        }
+        while (endps-- > 0) {
+                u132->endp[endps] = NULL;
+        }
+        up(&u132->sw_lock);
+        return;
+}
+
+static int __devinit u132_probe(struct platform_device *pdev)
+{
+        struct usb_hcd *hcd;
+        msleep(100);
+        if (u132_exiting > 0) {
+                return -ENODEV;
+        }                        /* refuse to confuse usbcore */
+        if (pdev->dev.dma_mask) {
+                return -EINVAL;
+        }
+        hcd = usb_create_hcd(&u132_hc_driver, &pdev->dev, pdev->dev.bus_id);
+        if (!hcd) {
+                printk(KERN_ERR "failed to create the usb hcd struct for U132\n"
+                        );
+                ftdi_elan_gone_away(pdev);
+                return -ENOMEM;
+        } else {
+                int retval = 0;
+                struct u132 *u132 = hcd_to_u132(hcd);
+                hcd->rsrc_start = 0;
+                down(&u132_module_lock);
+                list_add_tail(&u132->u132_list, &u132_static_list);
+                u132->sequence_num = ++u132_instances;
+                up(&u132_module_lock);
+                u132_u132_init_kref(u132);
+                u132_initialise(u132, pdev);
+                hcd->product_desc = "ELAN U132 Host Controller";
+                retval = usb_add_hcd(hcd, 0, 0);
+                if (retval != 0) {
+                        dev_err(&u132->platform_dev->dev, "init error %d\n",
+                                retval);
+                        u132_u132_put_kref(u132);
+                        return retval;
+                } else {
+                        u132_monitor_queue_work(u132, 100);
+                        return 0;
+                }
+        }
+}
+
+
+#ifdef CONFIG_PM
+/* for this device there's no useful distinction between the controller
+* and its root hub, except that the root hub only gets direct PM calls
+* when CONFIG_USB_SUSPEND is enabled.
+*/
+static int u132_suspend(struct platform_device *pdev, pm_message_t state)
+{
+        struct usb_hcd *hcd = platform_get_drvdata(pdev);
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else {
+                int retval = 0;
+                if (state.event == PM_EVENT_FREEZE) {
+                        retval = u132_bus_suspend(hcd);
+                } else if (state.event == PM_EVENT_SUSPEND) {
+                        int ports = MAX_U132_PORTS;
+                        while (ports-- > 0) {
+                                port_power(u132, ports, 0);
+                        }
+                }
+                if (retval == 0)
+                        pdev->dev.power.power_state = state;
+                return retval;
+        }
+}
+
+static int u132_resume(struct platform_device *pdev)
+{
+        struct usb_hcd *hcd = platform_get_drvdata(pdev);
+        struct u132 *u132 = hcd_to_u132(hcd);
+        if (u132->going > 1) {
+                dev_err(&u132->platform_dev->dev, "device has been removed %d\n"
+                        , u132->going);
+                return -ENODEV;
+        } else if (u132->going > 0) {
+                dev_err(&u132->platform_dev->dev, "device is being removed\n");
+                return -ESHUTDOWN;
+        } else {
+                int retval = 0;
+                if (pdev->dev.power.power_state.event == PM_EVENT_SUSPEND) {
+                        int ports = MAX_U132_PORTS;
+                        while (ports-- > 0) {
+                                port_power(u132, ports, 1);
+                        }
+                        retval = 0;
+                } else {
+                        pdev->dev.power.power_state = PMSG_ON;
+                        retval = u132_bus_resume(hcd);
+                }
+                return retval;
+        }
+}
+
+#else
+#define u132_suspend NULL
+#define u132_resume NULL
+#endif
+/*
+* this driver is loaded explicitely by ftdi_u132
+*
+* the platform_driver struct is static because it is per type of module
+*/
+static struct platform_driver u132_platform_driver = {
+        .probe = u132_probe,
+        .remove = __devexit_p(u132_remove),
+        .suspend = u132_suspend,
+        .resume = u132_resume,
+        .driver = {
+                   .name = (char *)hcd_name,
+                   .owner = THIS_MODULE,
+                   },
+};
+static int __init u132_hcd_init(void)
+{
+        int retval;
+        INIT_LIST_HEAD(&u132_static_list);
+        u132_instances = 0;
+        u132_exiting = 0;
+        init_MUTEX(&u132_module_lock);
+        if (usb_disabled())
+                return -ENODEV;
+        printk(KERN_INFO "driver %s built at %s on %s\n", hcd_name, __TIME__,
+                __DATE__);
+        workqueue = create_singlethread_workqueue("u132");
+        retval = platform_driver_register(&u132_platform_driver);
+        return retval;
+}
+
+
+module_init(u132_hcd_init);
+static void __exit u132_hcd_exit(void)
+{
+        struct u132 *u132;
+        struct u132 *temp;
+        down(&u132_module_lock);
+        u132_exiting += 1;
+        up(&u132_module_lock);
+        list_for_each_entry_safe(u132, temp, &u132_static_list, u132_list) {
+                platform_device_unregister(u132->platform_dev);
+        } platform_driver_unregister(&u132_platform_driver);
+        printk(KERN_INFO "u132-hcd driver deregistered\n");
+        wait_event(u132_hcd_wait, u132_instances == 0);
+        flush_workqueue(workqueue);
+        destroy_workqueue(workqueue);
+}
+
+
+module_exit(u132_hcd_exit);
+MODULE_LICENSE("GPL");
-- 
cgit v1.2.1


From ec17cf1cfe0b557210b27313bd584e9b5187d4ca Mon Sep 17 00:00:00 2001
From: Tobias Klauser <tklauser@distanz.ch>
Date: Wed, 13 Sep 2006 21:38:41 +0200
Subject: USB: Remove unneeded void * casts in core files

The patch removes unneeded casts for the following (void *) pointers:
 - struct file: private
 - struct urb: context
 - struct usb_bus: hcpriv
 - return value of kmalloc()

The patch also contains some whitespace cleanup in the relevant areas.

Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/devices.c |  4 ++--
 drivers/usb/core/devio.c   | 12 ++++++------
 drivers/usb/core/hub.c     |  4 ++--
 drivers/usb/core/message.c |  2 +-
 drivers/usb/core/urb.c     |  2 +-
 5 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/drivers/usb/core/devices.c b/drivers/usb/core/devices.c
index 241d37326ea9..3538c2fdadfe 100644
--- a/drivers/usb/core/devices.c
+++ b/drivers/usb/core/devices.c
@@ -593,7 +593,7 @@ static ssize_t usb_device_read(struct file *file, char __user *buf, size_t nbyte
 /* Kernel lock for "lastev" protection */
 static unsigned int usb_device_poll(struct file *file, struct poll_table_struct *wait)
 {
-	struct usb_device_status *st = (struct usb_device_status *)file->private_data;
+	struct usb_device_status *st = file->private_data;
 	unsigned int mask = 0;
 
 	lock_kernel();
@@ -603,7 +603,7 @@ static unsigned int usb_device_poll(struct file *file, struct poll_table_struct
 			unlock_kernel();
 			return POLLIN;
 		}
-		
+
 		/* we may have dropped BKL - need to check for having lost the race */
 		if (file->private_data) {
 			kfree(st);
diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c
index 71bbd25a4ed0..a94c63bef632 100644
--- a/drivers/usb/core/devio.c
+++ b/drivers/usb/core/devio.c
@@ -122,7 +122,7 @@ static loff_t usbdev_lseek(struct file *file, loff_t offset, int orig)
 
 static ssize_t usbdev_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
 {
-	struct dev_state *ps = (struct dev_state *)file->private_data;
+	struct dev_state *ps = file->private_data;
 	struct usb_device *dev = ps->dev;
 	ssize_t ret = 0;
 	unsigned len;
@@ -305,7 +305,7 @@ static void snoop_urb(struct urb *urb, void __user *userurb)
 
 static void async_completed(struct urb *urb, struct pt_regs *regs)
 {
-        struct async *as = (struct async *)urb->context;
+        struct async *as = urb->context;
         struct dev_state *ps = as->ps;
 	struct siginfo sinfo;
 
@@ -591,7 +591,7 @@ static int usbdev_open(struct inode *inode, struct file *file)
 
 static int usbdev_release(struct inode *inode, struct file *file)
 {
-	struct dev_state *ps = (struct dev_state *)file->private_data;
+	struct dev_state *ps = file->private_data;
 	struct usb_device *dev = ps->dev;
 	unsigned int ifnum;
 
@@ -1423,7 +1423,7 @@ static int proc_ioctl_compat(struct dev_state *ps, compat_uptr_t arg)
  */
 static int usbdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd, unsigned long arg)
 {
-	struct dev_state *ps = (struct dev_state *)file->private_data;
+	struct dev_state *ps = file->private_data;
 	struct usb_device *dev = ps->dev;
 	void __user *p = (void __user *)arg;
 	int ret = -ENOTTY;
@@ -1566,8 +1566,8 @@ static int usbdev_ioctl(struct inode *inode, struct file *file, unsigned int cmd
 /* No kernel lock - fine */
 static unsigned int usbdev_poll(struct file *file, struct poll_table_struct *wait)
 {
-	struct dev_state *ps = (struct dev_state *)file->private_data;
-        unsigned int mask = 0;
+	struct dev_state *ps = file->private_data;
+	unsigned int mask = 0;
 
 	poll_wait(file, &ps->wait, wait);
 	if (file->f_mode & FMODE_WRITE && !list_empty(&ps->async_completed))
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 7479bd329abb..c74baad19b16 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -293,7 +293,7 @@ void usb_kick_khubd(struct usb_device *hdev)
 /* completion function, fires on port status changes and various faults */
 static void hub_irq(struct urb *urb, struct pt_regs *regs)
 {
-	struct usb_hub *hub = (struct usb_hub *)urb->context;
+	struct usb_hub *hub = urb->context;
 	int status;
 	int i;
 	unsigned long bits;
@@ -311,7 +311,7 @@ static void hub_irq(struct urb *urb, struct pt_regs *regs)
 			goto resubmit;
 		hub->error = urb->status;
 		/* FALL THROUGH */
-	
+
 	/* let khubd handle things */
 	case 0:			/* we got data:  port status changed */
 		bits = 0;
diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c
index 28c6cf225780..85b1cd18336f 100644
--- a/drivers/usb/core/message.c
+++ b/drivers/usb/core/message.c
@@ -248,7 +248,7 @@ static void sg_clean (struct usb_sg_request *io)
 
 static void sg_complete (struct urb *urb, struct pt_regs *regs)
 {
-	struct usb_sg_request	*io = (struct usb_sg_request *) urb->context;
+	struct usb_sg_request	*io = urb->context;
 
 	spin_lock (&io->lock);
 
diff --git a/drivers/usb/core/urb.c b/drivers/usb/core/urb.c
index 576919927f53..9801d08edacf 100644
--- a/drivers/usb/core/urb.c
+++ b/drivers/usb/core/urb.c
@@ -57,7 +57,7 @@ struct urb *usb_alloc_urb(int iso_packets, gfp_t mem_flags)
 {
 	struct urb *urb;
 
-	urb = (struct urb *)kmalloc(sizeof(struct urb) + 
+	urb = kmalloc(sizeof(struct urb) +
 		iso_packets * sizeof(struct usb_iso_packet_descriptor),
 		mem_flags);
 	if (!urb) {
-- 
cgit v1.2.1


From 38e2bfc94e95dd6005fdaf40dfec0157396741da Mon Sep 17 00:00:00 2001
From: Pete Zaitcev <zaitcev@redhat.com>
Date: Mon, 18 Sep 2006 22:49:02 -0700
Subject: USB: Dealias -110 code (more complete)

The purpose of this patch is to split off the case when a device does
not reply on the lower level (which is reported by HC hardware), and
a case when the device accepted the request, but does not reply at
upper level. This redefinition allows to diagnose issues easier,
without asking the user if the -110 happened "immediately".

The usbmon splits such cases already thanks to its timestamp, but
it's not always available.

I adjusted all drivers which I found affected (by searching for "urb").
Out of tree drivers may suffer a little bit, but I do not expect much
breakage. At worst they may print a few messages.

Signed-off-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/usb/error-codes.txt       | 11 ++++-------
 drivers/isdn/gigaset/bas-gigaset.c      |  2 +-
 drivers/isdn/hisax/hfc_usb.h            |  6 +++---
 drivers/media/dvb/dvb-usb/dvb-usb-urb.c |  1 -
 drivers/media/dvb/ttusb-dec/ttusb_dec.c |  2 +-
 drivers/media/video/ov511.c             |  7 ++++---
 drivers/media/video/pwc/pwc-if.c        |  2 +-
 drivers/media/video/w9968cf.c           |  7 +++----
 drivers/net/irda/irda-usb.c             | 18 +++++++-----------
 drivers/net/wireless/zd1201.c           |  4 ++--
 drivers/usb/host/isp116x.h              |  2 +-
 drivers/usb/host/ohci.h                 |  2 +-
 drivers/usb/host/sl811-hcd.c            |  2 +-
 drivers/usb/input/hid-core.c            |  3 ++-
 drivers/usb/input/itmtouch.c            |  2 +-
 drivers/usb/input/mtouchusb.c           |  2 +-
 drivers/usb/input/touchkitusb.c         |  2 +-
 drivers/usb/input/usbtouchscreen.c      |  2 +-
 drivers/usb/misc/auerswald.c            |  4 ++--
 drivers/usb/net/pegasus.c               |  2 +-
 drivers/usb/net/rtl8150.c               |  2 +-
 drivers/usb/net/usbnet.c                | 12 ++++++------
 drivers/usb/storage/transport.c         |  5 -----
 sound/usb/usbmidi.c                     |  6 +++---
 24 files changed, 48 insertions(+), 60 deletions(-)

diff --git a/Documentation/usb/error-codes.txt b/Documentation/usb/error-codes.txt
index 867f4c38f356..39c68f8c4e6c 100644
--- a/Documentation/usb/error-codes.txt
+++ b/Documentation/usb/error-codes.txt
@@ -98,13 +98,13 @@ one or more packets could finish before an error stops further endpoint I/O.
 			error, a failure to respond (often caused by
 			device disconnect), or some other fault.
 
--ETIMEDOUT (**)		No response packet received within the prescribed
+-ETIME (**)		No response packet received within the prescribed
 			bus turn-around time.  This error may instead be
 			reported as -EPROTO or -EILSEQ.
 
-			Note that the synchronous USB message functions
-			also use this code to indicate timeout expired
-			before the transfer completed.
+-ETIMEDOUT		Synchronous USB message functions use this code
+			to indicate timeout expired before the transfer
+			completed, and no other error was reported by HC.
 
 -EPIPE (**)		Endpoint stalled.  For non-control endpoints,
 			reset this status with usb_clear_halt().
@@ -163,6 +163,3 @@ usb_get_*/usb_set_*():
 usb_control_msg():
 usb_bulk_msg():
 -ETIMEDOUT		Timeout expired before the transfer completed.
-			In the future this code may change to -ETIME,
-			whose definition is a closer match to this sort
-			of error.
diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c
index 3845defd4901..5cfbe6a38010 100644
--- a/drivers/isdn/gigaset/bas-gigaset.c
+++ b/drivers/isdn/gigaset/bas-gigaset.c
@@ -192,7 +192,7 @@ static char *get_usb_statmsg(int status)
 		return "bit stuffing error, timeout, or unknown USB error";
 	case -EILSEQ:
 		return "CRC mismatch, timeout, or unknown USB error";
-	case -ETIMEDOUT:
+	case -ETIME:
 		return "timed out";
 	case -EPIPE:
 		return "endpoint stalled";
diff --git a/drivers/isdn/hisax/hfc_usb.h b/drivers/isdn/hisax/hfc_usb.h
index ec52c1a7c22a..6349367ed480 100644
--- a/drivers/isdn/hisax/hfc_usb.h
+++ b/drivers/isdn/hisax/hfc_usb.h
@@ -137,11 +137,11 @@ static struct hfcusb_symbolic_list urb_errlist[] = {
 	{-ENXIO, "URB already queued"},
 	{-EFBIG, "Too much ISO frames requested"},
 	{-ENOSR, "Buffer error (overrun)"},
-	{-EPIPE, "Specified endpoint is stalled (device not responding)"},
+	{-EPIPE, "Specified endpoint is stalled"},
 	{-EOVERFLOW, "Babble (bad cable?)"},
 	{-EPROTO, "Bit-stuff error (bad cable?)"},
-	{-EILSEQ, "CRC/Timeout"},
-	{-ETIMEDOUT, "NAK (device does not respond)"},
+	{-EILSEQ, "CRC or missing token"},
+	{-ETIME, "Device did not respond"},
 	{-ESHUTDOWN, "Device unplugged"},
 	{-1, NULL}
 };
diff --git a/drivers/media/dvb/dvb-usb/dvb-usb-urb.c b/drivers/media/dvb/dvb-usb/dvb-usb-urb.c
index 9002f35aa952..88b283731bb8 100644
--- a/drivers/media/dvb/dvb-usb/dvb-usb-urb.c
+++ b/drivers/media/dvb/dvb-usb/dvb-usb-urb.c
@@ -80,7 +80,6 @@ static void dvb_usb_urb_complete(struct urb *urb, struct pt_regs *ptregs)
 
 	switch (urb->status) {
 		case 0:         /* success */
-		case -ETIMEDOUT:    /* NAK */
 			break;
 		case -ECONNRESET:   /* kill */
 		case -ENOENT:
diff --git a/drivers/media/dvb/ttusb-dec/ttusb_dec.c b/drivers/media/dvb/ttusb-dec/ttusb_dec.c
index 6c1cb770bcf5..c9d663549dff 100644
--- a/drivers/media/dvb/ttusb-dec/ttusb_dec.c
+++ b/drivers/media/dvb/ttusb-dec/ttusb_dec.c
@@ -215,7 +215,7 @@ static void ttusb_dec_handle_irq( struct urb *urb, struct pt_regs *regs)
 		case -ECONNRESET:
 		case -ENOENT:
 		case -ESHUTDOWN:
-		case -ETIMEDOUT:
+		case -ETIME:
 			/* this urb is dead, cleanup */
 			dprintk("%s:urb shutting down with status: %d\n",
 					__FUNCTION__, urb->status);
diff --git a/drivers/media/video/ov511.c b/drivers/media/video/ov511.c
index 1b07a61c2ebb..5d8cd283fcd8 100644
--- a/drivers/media/video/ov511.c
+++ b/drivers/media/video/ov511.c
@@ -301,10 +301,11 @@ static struct symbolic_list senlist[] = {
 static struct symbolic_list urb_errlist[] = {
 	{ -ENOSR,	"Buffer error (overrun)" },
 	{ -EPIPE,	"Stalled (device not responding)" },
-	{ -EOVERFLOW,	"Babble (bad cable?)" },
+	{ -EOVERFLOW,	"Babble (device sends too much data)" },
 	{ -EPROTO,	"Bit-stuff error (bad cable?)" },
-	{ -EILSEQ,	"CRC/Timeout" },
-	{ -ETIMEDOUT,	"NAK (device does not respond)" },
+	{ -EILSEQ,	"CRC/Timeout (bad cable?)" },
+	{ -ETIME,	"Device does not respond to token" },
+	{ -ETIMEDOUT,	"Device does not respond to command" },
 	{ -1, NULL }
 };
 
diff --git a/drivers/media/video/pwc/pwc-if.c b/drivers/media/video/pwc/pwc-if.c
index d4703944df9c..53c4b5790d5c 100644
--- a/drivers/media/video/pwc/pwc-if.c
+++ b/drivers/media/video/pwc/pwc-if.c
@@ -711,7 +711,7 @@ static void pwc_isoc_handler(struct urb *urb, struct pt_regs *regs)
 			case -EOVERFLOW:	errmsg = "Babble (bad cable?)"; break;
 			case -EPROTO:		errmsg = "Bit-stuff error (bad cable?)"; break;
 			case -EILSEQ:		errmsg = "CRC/Timeout (could be anything)"; break;
-			case -ETIMEDOUT:	errmsg = "NAK (device does not respond)"; break;
+			case -ETIME:		errmsg = "Device does not respond"; break;
 		}
 		PWC_DEBUG_FLOW("pwc_isoc_handler() called with status %d [%s].\n", urb->status, errmsg);
 		/* Give up after a number of contiguous errors on the USB bus.
diff --git a/drivers/media/video/w9968cf.c b/drivers/media/video/w9968cf.c
index 20f211b55ad4..2912326a5aef 100644
--- a/drivers/media/video/w9968cf.c
+++ b/drivers/media/video/w9968cf.c
@@ -586,15 +586,14 @@ static struct w9968cf_symbolic_list urb_errlist[] = {
 	{ -EFBIG,     "Too much ISO frames requested" },
 	{ -ENOSR,     "Buffer error (overrun)" },
 	{ -EPIPE,     "Specified endpoint is stalled (device not responding)"},
-	{ -EOVERFLOW, "Babble (bad cable?)" },
+	{ -EOVERFLOW, "Babble (too much data)" },
 	{ -EPROTO,    "Bit-stuff error (bad cable?)" },
 	{ -EILSEQ,    "CRC/Timeout" },
-	{ -ETIMEDOUT, "NAK (device does not respond)" },
+	{ -ETIME,     "Device does not respond to token" },
+	{ -ETIMEDOUT, "Device does not respond to command" },
 	{ -1, NULL }
 };
 
-
-
 /****************************************************************************
  * Memory management functions                                              *
  ****************************************************************************/
diff --git a/drivers/net/irda/irda-usb.c b/drivers/net/irda/irda-usb.c
index 2a0d538b387f..383cef1f5999 100644
--- a/drivers/net/irda/irda-usb.c
+++ b/drivers/net/irda/irda-usb.c
@@ -671,10 +671,8 @@ static void irda_usb_net_timeout(struct net_device *netdev)
 			 * Jean II */
 			done = 1;
 			break;
-		case -ECONNABORTED:		/* -103 */
-		case -ECONNRESET:		/* -104 */
-		case -ETIMEDOUT:		/* -110 */
-		case -ENOENT:			/* -2 (urb unlinked by us)  */
+		case -ECONNRESET:
+		case -ENOENT:			/* urb unlinked by us */
 		default:			/* ??? - Play safe */
 			urb->status = 0;
 			netif_wake_queue(self->netdev);
@@ -712,10 +710,8 @@ static void irda_usb_net_timeout(struct net_device *netdev)
 			 * Jean II */
 			done = 1;
 			break;
-		case -ECONNABORTED:		/* -103 */
-		case -ECONNRESET:		/* -104 */
-		case -ETIMEDOUT:		/* -110 */
-		case -ENOENT:			/* -2 (urb unlinked by us)  */
+		case -ECONNRESET:
+		case -ENOENT:			/* urb unlinked by us */
 		default:			/* ??? - Play safe */
 			if(skb != NULL) {
 				dev_kfree_skb_any(skb);
@@ -845,14 +841,14 @@ static void irda_usb_receive(struct urb *urb, struct pt_regs *regs)
 			self->stats.rx_crc_errors++;	
 			/* Also precursor to a hot-unplug on UHCI. */
 			/* Fallthrough... */
-		case -ECONNRESET:		/* -104 */
+		case -ECONNRESET:
 			/* Random error, if I remember correctly */
 			/* uhci_cleanup_unlink() is going to kill the Rx
 			 * URB just after we return. No problem, at this
 			 * point the URB will be idle ;-) - Jean II */
-		case -ESHUTDOWN:		/* -108 */
+		case -ESHUTDOWN:
 			/* That's usually a hot-unplug. Submit will fail... */
-		case -ETIMEDOUT:		/* -110 */
+		case -ETIME:
 			/* Usually precursor to a hot-unplug on OHCI. */
 		default:
 			self->stats.rx_errors++;
diff --git a/drivers/net/wireless/zd1201.c b/drivers/net/wireless/zd1201.c
index c52e9bcf8d02..f50ec10675d1 100644
--- a/drivers/net/wireless/zd1201.c
+++ b/drivers/net/wireless/zd1201.c
@@ -119,7 +119,7 @@ static void zd1201_usbfree(struct urb *urb, struct pt_regs *regs)
 	switch(urb->status) {
 		case -EILSEQ:
 		case -ENODEV:
-		case -ETIMEDOUT:
+		case -ETIME:
 		case -ENOENT:
 		case -EPIPE:
 		case -EOVERFLOW:
@@ -201,7 +201,7 @@ static void zd1201_usbrx(struct urb *urb, struct pt_regs *regs)
 	switch(urb->status) {
 		case -EILSEQ:
 		case -ENODEV:
-		case -ETIMEDOUT:
+		case -ETIME:
 		case -ENOENT:
 		case -EPIPE:
 		case -EOVERFLOW:
diff --git a/drivers/usb/host/isp116x.h b/drivers/usb/host/isp116x.h
index a1b7c3813d3a..b91e2edd9c5c 100644
--- a/drivers/usb/host/isp116x.h
+++ b/drivers/usb/host/isp116x.h
@@ -233,7 +233,7 @@ static const int cc_to_error[16] = {
 	/* Bit Stuff  */ -EPROTO,
 	/* Data Togg  */ -EILSEQ,
 	/* Stall      */ -EPIPE,
-	/* DevNotResp */ -ETIMEDOUT,
+	/* DevNotResp */ -ETIME,
 	/* PIDCheck   */ -EPROTO,
 	/* UnExpPID   */ -EPROTO,
 	/* DataOver   */ -EOVERFLOW,
diff --git a/drivers/usb/host/ohci.h b/drivers/usb/host/ohci.h
index 650d1bf21c1d..93fdc3c35341 100644
--- a/drivers/usb/host/ohci.h
+++ b/drivers/usb/host/ohci.h
@@ -159,7 +159,7 @@ static const int cc_to_error [16] = {
 	/* Bit Stuff  */               -EPROTO,
 	/* Data Togg  */               -EILSEQ,
 	/* Stall      */               -EPIPE,
-	/* DevNotResp */               -ETIMEDOUT,
+	/* DevNotResp */               -ETIME,
 	/* PIDCheck   */               -EPROTO,
 	/* UnExpPID   */               -EPROTO,
 	/* DataOver   */               -EOVERFLOW,
diff --git a/drivers/usb/host/sl811-hcd.c b/drivers/usb/host/sl811-hcd.c
index 8c17da37600b..3a586aab3939 100644
--- a/drivers/usb/host/sl811-hcd.c
+++ b/drivers/usb/host/sl811-hcd.c
@@ -597,7 +597,7 @@ done(struct sl811 *sl811, struct sl811h_ep *ep, u8 bank, struct pt_regs *regs)
 	/* error? retry, until "3 strikes" */
 	} else if (++ep->error_count >= 3) {
 		if (status & SL11H_STATMASK_TMOUT)
-			urbstat = -ETIMEDOUT;
+			urbstat = -ETIME;
 		else if (status & SL11H_STATMASK_OVF)
 			urbstat = -EOVERFLOW;
 		else
diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index fc5b662ea177..2a3e9e9b4b3d 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1023,7 +1023,8 @@ static void hid_irq_in(struct urb *urb, struct pt_regs *regs)
 			return;
 		case -EILSEQ:		/* protocol error or unplug */
 		case -EPROTO:		/* protocol error or unplug */
-		case -ETIMEDOUT:	/* NAK */
+		case -ETIME:		/* protocol error or unplug */
+		case -ETIMEDOUT:	/* Should never happen, but... */
 			clear_bit(HID_IN_RUNNING, &hid->iofl);
 			hid_io_error(hid);
 			return;
diff --git a/drivers/usb/input/itmtouch.c b/drivers/usb/input/itmtouch.c
index 86acb5f1907a..61966d719ca3 100644
--- a/drivers/usb/input/itmtouch.c
+++ b/drivers/usb/input/itmtouch.c
@@ -87,7 +87,7 @@ static void itmtouch_irq(struct urb *urb, struct pt_regs *regs)
 	case 0:
 		/* success */
 		break;
-	case -ETIMEDOUT:
+	case -ETIME:
 		/* this urb is timing out */
 		dbg("%s - urb timed out - was the device unplugged?",
 		    __FUNCTION__);
diff --git a/drivers/usb/input/mtouchusb.c b/drivers/usb/input/mtouchusb.c
index a9ccda8810e0..5dce951f2751 100644
--- a/drivers/usb/input/mtouchusb.c
+++ b/drivers/usb/input/mtouchusb.c
@@ -107,7 +107,7 @@ static void mtouchusb_irq(struct urb *urb, struct pt_regs *regs)
 	case 0:
 		/* success */
 		break;
-	case -ETIMEDOUT:
+	case -ETIME:
 		/* this urb is timing out */
 		dbg("%s - urb timed out - was the device unplugged?",
 		    __FUNCTION__);
diff --git a/drivers/usb/input/touchkitusb.c b/drivers/usb/input/touchkitusb.c
index 0149043ffb97..30b9f820e7a8 100644
--- a/drivers/usb/input/touchkitusb.c
+++ b/drivers/usb/input/touchkitusb.c
@@ -201,7 +201,7 @@ static void touchkit_irq(struct urb *urb, struct pt_regs *regs)
 	case 0:
 		/* success */
 		break;
-	case -ETIMEDOUT:
+	case -ETIME:
 		/* this urb is timing out */
 		dbg("%s - urb timed out - was the device unplugged?",
 		    __FUNCTION__);
diff --git a/drivers/usb/input/usbtouchscreen.c b/drivers/usb/input/usbtouchscreen.c
index a1be7840ea02..4640d1000d83 100644
--- a/drivers/usb/input/usbtouchscreen.c
+++ b/drivers/usb/input/usbtouchscreen.c
@@ -508,7 +508,7 @@ static void usbtouch_irq(struct urb *urb, struct pt_regs *regs)
 	case 0:
 		/* success */
 		break;
-	case -ETIMEDOUT:
+	case -ETIME:
 		/* this urb is timing out */
 		dbg("%s - urb timed out - was the device unplugged?",
 		    __FUNCTION__);
diff --git a/drivers/usb/misc/auerswald.c b/drivers/usb/misc/auerswald.c
index 4a329d8488b7..4fd2110b3411 100644
--- a/drivers/usb/misc/auerswald.c
+++ b/drivers/usb/misc/auerswald.c
@@ -806,7 +806,7 @@ static void auerbuf_releasebuf( pauerbuf_t bp)
 0		Initial, OK
 -EINPROGRESS	during submission until end
 -ENOENT		if urb is unlinked
--ETIMEDOUT	Transfer timed out, NAK
+-ETIME		Device did not respond
 -ENOMEM		Memory Overflow
 -ENODEV		Specified USB-device or bus doesn't exist
 -ENXIO		URB already queued
@@ -832,7 +832,7 @@ static int auerswald_status_retry (int status)
 {
 	switch (status) {
 	case 0:
-	case -ETIMEDOUT:
+	case -ETIME:
 	case -EOVERFLOW:
 	case -EAGAIN:
 	case -EPIPE:
diff --git a/drivers/usb/net/pegasus.c b/drivers/usb/net/pegasus.c
index ab21f960d255..b8e25af13f02 100644
--- a/drivers/usb/net/pegasus.c
+++ b/drivers/usb/net/pegasus.c
@@ -619,7 +619,7 @@ static void read_bulk_callback(struct urb *urb, struct pt_regs *regs)
 	switch (urb->status) {
 	case 0:
 		break;
-	case -ETIMEDOUT:
+	case -ETIME:
 		if (netif_msg_rx_err(pegasus))
 			pr_debug("%s: reset MAC\n", net->name);
 		pegasus->flags &= ~PEGASUS_RX_BUSY;
diff --git a/drivers/usb/net/rtl8150.c b/drivers/usb/net/rtl8150.c
index a72685b96061..2364c2099387 100644
--- a/drivers/usb/net/rtl8150.c
+++ b/drivers/usb/net/rtl8150.c
@@ -438,7 +438,7 @@ static void read_bulk_callback(struct urb *urb, struct pt_regs *regs)
 		break;
 	case -ENOENT:
 		return;	/* the urb is in unlink state */
-	case -ETIMEDOUT:
+	case -ETIME:
 		warn("may be reset is needed?..");
 		goto goon;
 	default:
diff --git a/drivers/usb/net/usbnet.c b/drivers/usb/net/usbnet.c
index 8e8e74d40530..98a522f1e264 100644
--- a/drivers/usb/net/usbnet.c
+++ b/drivers/usb/net/usbnet.c
@@ -425,9 +425,9 @@ static void rx_complete (struct urb *urb, struct pt_regs *regs)
 	    // we get controller i/o faults during khubd disconnect() delays.
 	    // throttle down resubmits, to avoid log floods; just temporarily,
 	    // so we still recover when the fault isn't a khubd delay.
-	    case -EPROTO:		// ehci
-	    case -ETIMEDOUT:		// ohci
-	    case -EILSEQ:		// uhci
+	    case -EPROTO:
+	    case -ETIME:
+	    case -EILSEQ:
 		dev->stats.rx_errors++;
 		if (!timer_pending (&dev->delay)) {
 			mod_timer (&dev->delay, jiffies + THROTTLE_JIFFIES);
@@ -821,9 +821,9 @@ static void tx_complete (struct urb *urb, struct pt_regs *regs)
 
 		// like rx, tx gets controller i/o faults during khubd delays
 		// and so it uses the same throttling mechanism.
-		case -EPROTO:		// ehci
-		case -ETIMEDOUT:	// ohci
-		case -EILSEQ:		// uhci
+		case -EPROTO:
+		case -ETIME:
+		case -EILSEQ:
 			if (!timer_pending (&dev->delay)) {
 				mod_timer (&dev->delay,
 					jiffies + THROTTLE_JIFFIES);
diff --git a/drivers/usb/storage/transport.c b/drivers/usb/storage/transport.c
index d6acc92a4ae3..f23514c4e649 100644
--- a/drivers/usb/storage/transport.c
+++ b/drivers/usb/storage/transport.c
@@ -294,11 +294,6 @@ static int interpret_urb_result(struct us_data *us, unsigned int pipe,
 			return USB_STOR_XFER_ERROR;
 		return USB_STOR_XFER_STALLED;
 
-	/* timeout or excessively long NAK */
-	case -ETIMEDOUT:
-		US_DEBUGP("-- timeout or NAK\n");
-		return USB_STOR_XFER_ERROR;
-
 	/* babble - the device tried to send more than we wanted to read */
 	case -EOVERFLOW:
 		US_DEBUGP("-- babble\n");
diff --git a/sound/usb/usbmidi.c b/sound/usb/usbmidi.c
index 5105b6b05748..abe29dadd979 100644
--- a/sound/usb/usbmidi.c
+++ b/sound/usb/usbmidi.c
@@ -181,9 +181,9 @@ static int snd_usbmidi_urb_error(int status)
 	case -ENODEV:
 		return -ENODEV;
 	/* errors that might occur during unplugging */
-	case -EPROTO:    /* EHCI */
-	case -ETIMEDOUT: /* OHCI */
-	case -EILSEQ:    /* UHCI */
+	case -EPROTO:
+	case -ETIME:
+	case -EILSEQ:
 		return -EIO;
 	default:
 		snd_printk(KERN_ERR "urb status %d\n", status);
-- 
cgit v1.2.1


From 71795c1df30b034414c921b4930ed88de34ca348 Mon Sep 17 00:00:00 2001
From: Pete Zaitcev <zaitcev@redhat.com>
Date: Mon, 18 Sep 2006 22:57:22 -0700
Subject: USB: ohci_usb can oops on shutdown

When ohci-hcd is shutting down (for rmmod or PC-card removal), there is
a window when the device is shut down, HC communication area (->hcca)
is freed, but the core has not called "free_irq" yet. If another device
triggers a shared interrupt in this window, we oops when trying to
access the freed ->hcca.

This patch removes the window by calling free_irq before ->hcca is freed.

The patch is tested at the PC hotplug test rig at Stratus, and with
rmmod by Rafael Wysocki.

Signed-off-by: Pete Zaitcev <zaitcev@redhat.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ohci-hcd.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/usb/host/ohci-hcd.c b/drivers/usb/host/ohci-hcd.c
index 2c614af8f733..1027aa04583d 100644
--- a/drivers/usb/host/ohci-hcd.c
+++ b/drivers/usb/host/ohci-hcd.c
@@ -795,7 +795,9 @@ static void ohci_stop (struct usb_hcd *hcd)
 
 	ohci_usb_reset (ohci);
 	ohci_writel (ohci, OHCI_INTR_MIE, &ohci->regs->intrdisable);
-	
+	free_irq(hcd->irq, hcd);
+	hcd->irq = -1;
+
 	remove_debug_files (ohci);
 	ohci_mem_cleanup (ohci);
 	if (ohci->hcca) {
-- 
cgit v1.2.1


From 353a4098c61272b33a02ec5802fb3859fec91a0e Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Tue, 19 Sep 2006 10:07:58 -0400
Subject: USB: force root hub resume after power loss

This patch(as785) forces the PM core to resume a root hub after a
power loss during system sleep.  If the root hub had been suspended
before the system sleep then normally the PM core would not resume it
afterward.  Without this resume, various sorts of wakeup events (like
port change events) can get lost.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index c74baad19b16..00441cde4588 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1053,6 +1053,12 @@ void usb_root_hub_lost_power(struct usb_device *rhdev)
 	unsigned long flags;
 
 	dev_warn(&rhdev->dev, "root hub lost power or was reset\n");
+
+	/* Make sure no potential wakeup events get lost,
+	 * by forcing the root hub to be resumed.
+	 */
+	rhdev->dev.power.prev_state.event = PM_EVENT_ON;
+
 	spin_lock_irqsave(&device_state_lock, flags);
 	hub = hdev_to_hub(rhdev);
 	for (port1 = 1; port1 <= rhdev->maxchild; ++port1) {
-- 
cgit v1.2.1


From 26f953fd884ea4879585287917f855c63c6b2666 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 18 Sep 2006 17:03:16 -0700
Subject: USB: EHCI update VIA workaround

This revamps handling of the hardware "async advance" IRQ, and its watchdog
timer.  Basically it dis-entangles that important timeout from the others,
simplifying the associated state and code to make it more robust.

This reportedly improves behavior of EHCI on some systems with VIA chips,
and AFAIK won't affect non-VIA hardware.  VIA systems need this code to
recover from silcon bugs whereby the "async advance" IRQ isn't issued.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/ehci-dbg.c |  4 +--
 drivers/usb/host/ehci-hcd.c | 73 +++++++++++++++++++++++++++++++--------------
 drivers/usb/host/ehci-hub.c |  2 +-
 drivers/usb/host/ehci-pci.c |  2 +-
 drivers/usb/host/ehci-q.c   |  6 ++--
 drivers/usb/host/ehci.h     | 22 +++++++++-----
 6 files changed, 70 insertions(+), 39 deletions(-)

diff --git a/drivers/usb/host/ehci-dbg.c b/drivers/usb/host/ehci-dbg.c
index 34b7a31cd85b..23b95b2bfe15 100644
--- a/drivers/usb/host/ehci-dbg.c
+++ b/drivers/usb/host/ehci-dbg.c
@@ -754,9 +754,7 @@ show_registers (struct class_device *class_dev, char *buf)
 	}
 
 	if (ehci->reclaim) {
-		temp = scnprintf (next, size, "reclaim qh %p%s\n",
-				ehci->reclaim,
-				ehci->reclaim_ready ? " ready" : "");
+		temp = scnprintf (next, size, "reclaim qh %p\n", ehci->reclaim);
 		size -= temp;
 		next += temp;
 	}
diff --git a/drivers/usb/host/ehci-hcd.c b/drivers/usb/host/ehci-hcd.c
index 4e1a8c308893..5ac918591131 100644
--- a/drivers/usb/host/ehci-hcd.c
+++ b/drivers/usb/host/ehci-hcd.c
@@ -111,7 +111,7 @@ static const char	hcd_name [] = "ehci_hcd";
 #define	EHCI_TUNE_MULT_TT	1
 #define	EHCI_TUNE_FLS		2	/* (small) 256 frame schedule */
 
-#define EHCI_IAA_JIFFIES	(HZ/100)	/* arbitrary; ~10 msec */
+#define EHCI_IAA_MSECS		10		/* arbitrary */
 #define EHCI_IO_JIFFIES		(HZ/10)		/* io watchdog > irq_thresh */
 #define EHCI_ASYNC_JIFFIES	(HZ/20)		/* async idle timeout */
 #define EHCI_SHRINK_JIFFIES	(HZ/200)	/* async qh unlink delay */
@@ -254,6 +254,7 @@ static void ehci_quiesce (struct ehci_hcd *ehci)
 
 /*-------------------------------------------------------------------------*/
 
+static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs);
 static void ehci_work(struct ehci_hcd *ehci, struct pt_regs *regs);
 
 #include "ehci-hub.c"
@@ -263,28 +264,39 @@ static void ehci_work(struct ehci_hcd *ehci, struct pt_regs *regs);
 
 /*-------------------------------------------------------------------------*/
 
-static void ehci_watchdog (unsigned long param)
+static void ehci_iaa_watchdog (unsigned long param)
 {
 	struct ehci_hcd		*ehci = (struct ehci_hcd *) param;
 	unsigned long		flags;
+	u32			status;
 
 	spin_lock_irqsave (&ehci->lock, flags);
+	WARN_ON(!ehci->reclaim);
 
-	/* lost IAA irqs wedge things badly; seen with a vt8235 */
+	/* lost IAA irqs wedge things badly; seen first with a vt8235 */
 	if (ehci->reclaim) {
-		u32		status = readl (&ehci->regs->status);
-
+		status = readl (&ehci->regs->status);
 		if (status & STS_IAA) {
 			ehci_vdbg (ehci, "lost IAA\n");
 			COUNT (ehci->stats.lost_iaa);
 			writel (STS_IAA, &ehci->regs->status);
-			ehci->reclaim_ready = 1;
+			end_unlink_async (ehci, NULL);
 		}
 	}
 
- 	/* stop async processing after it's idled a bit */
+	spin_unlock_irqrestore (&ehci->lock, flags);
+}
+
+static void ehci_watchdog (unsigned long param)
+{
+	struct ehci_hcd		*ehci = (struct ehci_hcd *) param;
+	unsigned long		flags;
+
+	spin_lock_irqsave (&ehci->lock, flags);
+
+	/* stop async processing after it's idled a bit */
 	if (test_bit (TIMER_ASYNC_OFF, &ehci->actions))
- 		start_unlink_async (ehci, ehci->async);
+		start_unlink_async (ehci, ehci->async);
 
 	/* ehci could run by timer, without IRQs ... */
 	ehci_work (ehci, NULL);
@@ -333,8 +345,6 @@ static void ehci_port_power (struct ehci_hcd *ehci, int is_on)
 static void ehci_work (struct ehci_hcd *ehci, struct pt_regs *regs)
 {
 	timer_action_done (ehci, TIMER_IO_WATCHDOG);
-	if (ehci->reclaim_ready)
-		end_unlink_async (ehci, regs);
 
 	/* another CPU may drop ehci->lock during a schedule scan while
 	 * it reports urb completions.  this flag guards against bogus
@@ -369,6 +379,7 @@ static void ehci_stop (struct usb_hcd *hcd)
 
 	/* no more interrupts ... */
 	del_timer_sync (&ehci->watchdog);
+	del_timer_sync (&ehci->iaa_watchdog);
 
 	spin_lock_irq(&ehci->lock);
 	if (HC_IS_RUNNING (hcd->state))
@@ -415,6 +426,10 @@ static int ehci_init(struct usb_hcd *hcd)
 	ehci->watchdog.function = ehci_watchdog;
 	ehci->watchdog.data = (unsigned long) ehci;
 
+	init_timer(&ehci->iaa_watchdog);
+	ehci->iaa_watchdog.function = ehci_iaa_watchdog;
+	ehci->iaa_watchdog.data = (unsigned long) ehci;
+
 	/*
 	 * hw default: 1K periodic list heads, one per frame.
 	 * periodic_size can shrink by USBCMD update if hcc_params allows.
@@ -431,7 +446,6 @@ static int ehci_init(struct usb_hcd *hcd)
 		ehci->i_thresh = 2 + HCC_ISOC_THRES(hcc_params);
 
 	ehci->reclaim = NULL;
-	ehci->reclaim_ready = 0;
 	ehci->next_uframe = -1;
 
 	/*
@@ -605,7 +619,7 @@ static irqreturn_t ehci_irq (struct usb_hcd *hcd, struct pt_regs *regs)
 	/* complete the unlinking of some qh [4.15.2.3] */
 	if (status & STS_IAA) {
 		COUNT (ehci->stats.reclaim);
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, regs);
 		bh = 1;
 	}
 
@@ -709,10 +723,14 @@ static int ehci_urb_enqueue (
 
 static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 {
-	/* if we need to use IAA and it's busy, defer */
-	if (qh->qh_state == QH_STATE_LINKED
-			&& ehci->reclaim
-			&& HC_IS_RUNNING (ehci_to_hcd(ehci)->state)) {
+	// BUG_ON(qh->qh_state != QH_STATE_LINKED);
+
+	/* failfast */
+	if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state))
+		end_unlink_async (ehci, NULL);
+
+	/* defer till later if busy */
+	else if (ehci->reclaim) {
 		struct ehci_qh		*last;
 
 		for (last = ehci->reclaim;
@@ -722,12 +740,8 @@ static void unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 		qh->qh_state = QH_STATE_UNLINK_WAIT;
 		last->reclaim = qh;
 
-	/* bypass IAA if the hc can't care */
-	} else if (!HC_IS_RUNNING (ehci_to_hcd(ehci)->state) && ehci->reclaim)
-		end_unlink_async (ehci, NULL);
-
-	/* something else might have unlinked the qh by now */
-	if (qh->qh_state == QH_STATE_LINKED)
+	/* start IAA cycle */
+	} else
 		start_unlink_async (ehci, qh);
 }
 
@@ -749,7 +763,19 @@ static int ehci_urb_dequeue (struct usb_hcd *hcd, struct urb *urb)
 		qh = (struct ehci_qh *) urb->hcpriv;
 		if (!qh)
 			break;
-		unlink_async (ehci, qh);
+		switch (qh->qh_state) {
+		case QH_STATE_LINKED:
+		case QH_STATE_COMPLETING:
+			unlink_async (ehci, qh);
+			break;
+		case QH_STATE_UNLINK:
+		case QH_STATE_UNLINK_WAIT:
+			/* already started */
+			break;
+		case QH_STATE_IDLE:
+			WARN_ON(1);
+			break;
+		}
 		break;
 
 	case PIPE_INTERRUPT:
@@ -841,6 +867,7 @@ rescan:
 		unlink_async (ehci, qh);
 		/* FALL THROUGH */
 	case QH_STATE_UNLINK:		/* wait for hw to finish? */
+	case QH_STATE_UNLINK_WAIT:
 idle_timeout:
 		spin_unlock_irqrestore (&ehci->lock, flags);
 		schedule_timeout_uninterruptible(1);
diff --git a/drivers/usb/host/ehci-hub.c b/drivers/usb/host/ehci-hub.c
index a5eeb9cd6ab2..b2ee13c58517 100644
--- a/drivers/usb/host/ehci-hub.c
+++ b/drivers/usb/host/ehci-hub.c
@@ -48,7 +48,7 @@ static int ehci_bus_suspend (struct usb_hcd *hcd)
 	}
 	ehci->command = readl (&ehci->regs->command);
 	if (ehci->reclaim)
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, NULL);
 	ehci_work(ehci, NULL);
 
 	/* suspend any active/unsuspended ports, maybe allow wakeup */
diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c
index e6a3bcddd55b..08d0472d4f57 100644
--- a/drivers/usb/host/ehci-pci.c
+++ b/drivers/usb/host/ehci-pci.c
@@ -303,7 +303,7 @@ restart:
 	/* emptying the schedule aborts any urbs */
 	spin_lock_irq(&ehci->lock);
 	if (ehci->reclaim)
-		ehci->reclaim_ready = 1;
+		end_unlink_async (ehci, NULL);
 	ehci_work(ehci, NULL);
 	spin_unlock_irq(&ehci->lock);
 
diff --git a/drivers/usb/host/ehci-q.c b/drivers/usb/host/ehci-q.c
index c0da40bbfa35..7fc25b6bd7d2 100644
--- a/drivers/usb/host/ehci-q.c
+++ b/drivers/usb/host/ehci-q.c
@@ -967,7 +967,7 @@ static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs)
 	struct ehci_qh		*qh = ehci->reclaim;
 	struct ehci_qh		*next;
 
-	timer_action_done (ehci, TIMER_IAA_WATCHDOG);
+	iaa_watchdog_done (ehci);
 
 	// qh->hw_next = cpu_to_le32 (qh->qh_dma);
 	qh->qh_state = QH_STATE_IDLE;
@@ -977,7 +977,6 @@ static void end_unlink_async (struct ehci_hcd *ehci, struct pt_regs *regs)
 	/* other unlink(s) may be pending (in QH_STATE_UNLINK_WAIT) */
 	next = qh->reclaim;
 	ehci->reclaim = next;
-	ehci->reclaim_ready = 0;
 	qh->reclaim = NULL;
 
 	qh_completions (ehci, qh, regs);
@@ -1052,11 +1051,10 @@ static void start_unlink_async (struct ehci_hcd *ehci, struct ehci_qh *qh)
 		return;
 	}
 
-	ehci->reclaim_ready = 0;
 	cmd |= CMD_IAAD;
 	writel (cmd, &ehci->regs->command);
 	(void) readl (&ehci->regs->command);
-	timer_action (ehci, TIMER_IAA_WATCHDOG);
+	iaa_watchdog_start (ehci);
 }
 
 /*-------------------------------------------------------------------------*/
diff --git a/drivers/usb/host/ehci.h b/drivers/usb/host/ehci.h
index bbc3082a73d7..6aac39f50e07 100644
--- a/drivers/usb/host/ehci.h
+++ b/drivers/usb/host/ehci.h
@@ -58,7 +58,6 @@ struct ehci_hcd {			/* one per controller */
 	/* async schedule support */
 	struct ehci_qh		*async;
 	struct ehci_qh		*reclaim;
-	unsigned		reclaim_ready : 1;
 	unsigned		scanning : 1;
 
 	/* periodic schedule support */
@@ -81,6 +80,7 @@ struct ehci_hcd {			/* one per controller */
 	struct dma_pool		*itd_pool;	/* itd per iso urb */
 	struct dma_pool		*sitd_pool;	/* sitd per split iso urb */
 
+	struct timer_list	iaa_watchdog;
 	struct timer_list	watchdog;
 	unsigned long		actions;
 	unsigned		stamp;
@@ -114,9 +114,21 @@ static inline struct usb_hcd *ehci_to_hcd (struct ehci_hcd *ehci)
 }
 
 
+static inline void
+iaa_watchdog_start (struct ehci_hcd *ehci)
+{
+	WARN_ON(timer_pending(&ehci->iaa_watchdog));
+	mod_timer (&ehci->iaa_watchdog,
+			jiffies + msecs_to_jiffies(EHCI_IAA_MSECS));
+}
+
+static inline void iaa_watchdog_done (struct ehci_hcd *ehci)
+{
+	del_timer (&ehci->iaa_watchdog);
+}
+
 enum ehci_timer_action {
 	TIMER_IO_WATCHDOG,
-	TIMER_IAA_WATCHDOG,
 	TIMER_ASYNC_SHRINK,
 	TIMER_ASYNC_OFF,
 };
@@ -134,9 +146,6 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action)
 		unsigned long t;
 
 		switch (action) {
-		case TIMER_IAA_WATCHDOG:
-			t = EHCI_IAA_JIFFIES;
-			break;
 		case TIMER_IO_WATCHDOG:
 			t = EHCI_IO_JIFFIES;
 			break;
@@ -153,8 +162,7 @@ timer_action (struct ehci_hcd *ehci, enum ehci_timer_action action)
 		// async queue SHRINK often precedes IAA.  while it's ready
 		// to go OFF neither can matter, and afterwards the IO
 		// watchdog stops unless there's still periodic traffic.
-		if (action != TIMER_IAA_WATCHDOG
-				&& t > ehci->watchdog.expires
+		if (time_before_eq(t, ehci->watchdog.expires)
 				&& timer_pending (&ehci->watchdog))
 			return;
 		mod_timer (&ehci->watchdog, t);
-- 
cgit v1.2.1


From fc849b85fb14ccbbc10098d501a870bc9b44a641 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 18 Sep 2006 16:53:26 -0700
Subject: USB: remove OTG build warning

Somewhere along the line, a variable in a USB-OTG codepath
stopped being used; this removes the relevant compiler warning.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/core/hub.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 00441cde4588..2a8cb3c2b19c 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -1254,8 +1254,7 @@ int usb_new_device(struct usb_device *udev)
 					USB_DT_OTG, (void **) &desc) == 0) {
 			if (desc->bmAttributes & USB_OTG_HNP) {
 				unsigned		port1 = udev->portnum;
-				struct usb_device	*root = udev->parent;
-				
+
 				dev_info(&udev->dev,
 					"Dual-Role OTG device on %sHNP port\n",
 					(port1 == bus->otg_port)
-- 
cgit v1.2.1


From 5dda171202f94127e49c12daf780cdae1b4e668b Mon Sep 17 00:00:00 2001
From: Andy Gay <andy@andynet.net>
Date: Mon, 3 Jul 2006 18:43:01 -0400
Subject: USB: Airprime driver improvements to allow full speed EvDO transfers

Adapted from an earlier patch by Greg KH <gregkh@suse.de>.
That patch added multiple read urbs and larger transfer buffers to allow
data transfers at full EvDO speed.

This version includes additional device IDs and fixes a memory leak in
the transfer buffer allocation.

Some (maybe all?) of the supported devices present multiple bulk endpoints,
the additional EPs can be used for control and status functions,
This version allocates 3 EPs by default, that can be changed using
the 'endpoints' module parameter.

Tested with Sierra Wireless EM5625 and MC5720 embedded modules.

Device ID (0x0c88, 0x17da) for the Kyocera Wireless KPC650/Passport
was added but is not yet tested.


From: Andy Gay <andy@andynet.net>
Cc: Kevin Lloyd <linux@sierrawireless.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/airprime.c | 261 +++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 256 insertions(+), 5 deletions(-)

diff --git a/drivers/usb/serial/airprime.c b/drivers/usb/serial/airprime.c
index 62082532a8b3..6e1a84a858d4 100644
--- a/drivers/usb/serial/airprime.c
+++ b/drivers/usb/serial/airprime.c
@@ -1,7 +1,7 @@
 /*
  * AirPrime CDMA Wireless Serial USB driver
  *
- * Copyright (C) 2005 Greg Kroah-Hartman <gregkh@suse.de>
+ * Copyright (C) 2005-2006 Greg Kroah-Hartman <gregkh@suse.de>
  *
  *	This program is free software; you can redistribute it and/or
  *	modify it under the terms of the GNU General Public License version
@@ -11,26 +11,264 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/tty.h>
+#include <linux/tty_flip.h>
 #include <linux/module.h>
 #include <linux/usb.h>
 #include <linux/usb/serial.h>
 
 static struct usb_device_id id_table [] = {
 	{ USB_DEVICE(0x0c88, 0x17da) }, /* Kyocera Wireless KPC650/Passport */
-	{ USB_DEVICE(0xf3d, 0x0112) },  /* AirPrime CDMA Wireless PC Card */
-	{ USB_DEVICE(0x1410, 0x1110) }, /* Novatel Wireless Merlin CDMA */
+	{ USB_DEVICE(0x0f3d, 0x0112) }, /* AirPrime CDMA Wireless PC Card */
+	{ USB_DEVICE(0x1199, 0x0017) }, /* Sierra Wireless EM5625 */
+	{ USB_DEVICE(0x1199, 0x0018) }, /* Sierra Wireless MC5720 */
 	{ USB_DEVICE(0x1199, 0x0112) }, /* Sierra Wireless Aircard 580 */
 	{ USB_DEVICE(0x1199, 0x0218) }, /* Sierra Wireless MC5720 */
+	{ USB_DEVICE(0x1410, 0x1110) }, /* Novatel Wireless Merlin CDMA */
 	{ },
 };
 MODULE_DEVICE_TABLE(usb, id_table);
 
+#define URB_TRANSFER_BUFFER_SIZE	4096
+#define NUM_READ_URBS			4
+#define NUM_WRITE_URBS			4
+#define NUM_BULK_EPS			3
+#define MAX_BULK_EPS			6
+
+/* if overridden by the user, then use their value for the size of the
+ * read and write urbs, and the number of endpoints */
+static int buffer_size = URB_TRANSFER_BUFFER_SIZE;
+static int endpoints = NUM_BULK_EPS;
+static int debug;
+struct airprime_private {
+	spinlock_t lock;
+	int outstanding_urbs;
+	int throttled;
+	struct urb *read_urbp[NUM_READ_URBS];
+};
+
+static void airprime_read_bulk_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct usb_serial_port *port = urb->context;
+	unsigned char *data = urb->transfer_buffer;
+	struct tty_struct *tty;
+	int result;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	if (urb->status) {
+		dbg("%s - nonzero read bulk status received: %d",
+		    __FUNCTION__, urb->status);
+		/* something happened, so free up the memory for this urb */
+		if (urb->transfer_buffer) {
+			kfree (urb->transfer_buffer);
+			urb->transfer_buffer = NULL;
+		}
+		return;
+	}
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, urb->actual_length, data);
+
+	tty = port->tty;
+	if (tty && urb->actual_length) {
+		tty_insert_flip_string (tty, data, urb->actual_length);
+		tty_flip_buffer_push (tty);
+	}
+
+	result = usb_submit_urb (urb, GFP_ATOMIC);
+	if (result)
+		dev_err(&port->dev, "%s - failed resubmitting read urb, error %d\n",
+			__FUNCTION__, result);
+	return;
+}
+
+static void airprime_write_bulk_callback(struct urb *urb, struct pt_regs *regs)
+{
+	struct usb_serial_port *port = urb->context;
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	unsigned long flags;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	/* free up the transfer buffer, as usb_free_urb() does not do this */
+	kfree (urb->transfer_buffer);
+
+	if (urb->status)
+		dbg("%s - nonzero write bulk status received: %d",
+		    __FUNCTION__, urb->status);
+	spin_lock_irqsave(&priv->lock, flags);
+	--priv->outstanding_urbs;
+	spin_unlock_irqrestore(&priv->lock, flags);
+
+	usb_serial_port_softint(port);
+}
+
+static int airprime_open(struct usb_serial_port *port, struct file *filp)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	struct usb_serial *serial = port->serial;
+	struct urb *urb;
+	char *buffer = NULL;
+	int i;
+	int result = 0;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	/* initialize our private data structure if it isn't already created */
+	if (!priv) {
+		priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+		if (!priv) {
+			result = -ENOMEM;
+			goto out;
+		}
+		spin_lock_init(&priv->lock);
+		usb_set_serial_port_data(port, priv);
+	}
+
+	for (i = 0; i < NUM_READ_URBS; ++i) {
+		buffer = kmalloc(buffer_size, GFP_KERNEL);
+		if (!buffer) {
+			dev_err(&port->dev, "%s - out of memory.\n",
+				__FUNCTION__);
+			result = -ENOMEM;
+			goto errout;
+		}
+		urb = usb_alloc_urb(0, GFP_KERNEL);
+		if (!urb) {
+			dev_err(&port->dev, "%s - no more urbs?\n",
+				__FUNCTION__);
+			result = -ENOMEM;
+			goto errout;
+		}
+		usb_fill_bulk_urb(urb, serial->dev,
+				  usb_rcvbulkpipe(serial->dev,
+						  port->bulk_out_endpointAddress),
+				  buffer, buffer_size,
+				  airprime_read_bulk_callback, port);
+		result = usb_submit_urb(urb, GFP_KERNEL);
+		if (result) {
+			dev_err(&port->dev,
+				"%s - failed submitting read urb %d for port %d, error %d\n",
+				__FUNCTION__, i, port->number, result);
+			goto errout;
+		}
+		/* remember this urb so we can kill it when the port is closed */
+		priv->read_urbp[i] = urb;
+	}
+	goto out;
+
+ errout:
+	/* some error happened, cancel any submitted urbs and clean up anything that
+	   got allocated successfully */
+
+	for ( ; i >= 0; --i) {
+		urb = priv->read_urbp[i];
+		if (urb) {
+			/* This urb was submitted successfully. So we have to
+			   cancel it.
+			   Unlinking the urb will invoke read_bulk_callback()
+			   with an error status, so its transfer buffer will
+			   be freed there */
+			if (usb_unlink_urb (urb) != -EINPROGRESS) {
+				/* comments in drivers/usb/core/urb.c say this
+				   can only happen if the urb was never submitted,
+				   or has completed already.
+				   Either way we may have to free the transfer
+				   buffer here. */
+				if (urb->transfer_buffer) {
+					kfree (urb->transfer_buffer);
+					urb->transfer_buffer = NULL;
+				}
+			}
+			usb_free_urb (urb);
+		}
+	}
+
+ out:
+	return result;
+}
+
+static void airprime_close(struct usb_serial_port *port, struct file * filp)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	int i;
+
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	/* killing the urb will invoke read_bulk_callback() with an error status,
+	   so the transfer buffer will be freed there */
+	for (i = 0; i < NUM_READ_URBS; ++i) {
+		usb_kill_urb (priv->read_urbp[i]);
+		usb_free_urb (priv->read_urbp[i]);
+	}
+
+	/* free up private structure */
+	kfree (priv);
+	usb_set_serial_port_data(port, NULL);
+}
+
+static int airprime_write(struct usb_serial_port *port,
+			  const unsigned char *buf, int count)
+{
+	struct airprime_private *priv = usb_get_serial_port_data(port);
+	struct usb_serial *serial = port->serial;
+	struct urb *urb;
+	unsigned char *buffer;
+	unsigned long flags;
+	int status;
+	dbg("%s - port %d", __FUNCTION__, port->number);
+
+	spin_lock_irqsave(&priv->lock, flags);
+	if (priv->outstanding_urbs > NUM_WRITE_URBS) {
+		spin_unlock_irqrestore(&priv->lock, flags);
+		dbg("%s - write limit hit\n", __FUNCTION__);
+		return 0;
+	}
+	spin_unlock_irqrestore(&priv->lock, flags);
+	buffer = kmalloc(count, GFP_ATOMIC);
+	if (!buffer) {
+		dev_err(&port->dev, "out of memory\n");
+		return -ENOMEM;
+	}
+	urb = usb_alloc_urb(0, GFP_ATOMIC);
+	if (!urb) {
+		dev_err(&port->dev, "no more free urbs\n");
+		kfree (buffer);
+		return -ENOMEM;
+	}
+	memcpy (buffer, buf, count);
+
+	usb_serial_debug_data(debug, &port->dev, __FUNCTION__, count, buffer);
+
+	usb_fill_bulk_urb(urb, serial->dev,
+			  usb_sndbulkpipe(serial->dev,
+					  port->bulk_out_endpointAddress),
+			  buffer, count,
+			  airprime_write_bulk_callback, port);
+
+	/* send it down the pipe */
+	status = usb_submit_urb(urb, GFP_ATOMIC);
+	if (status) {
+		dev_err(&port->dev,
+			"%s - usb_submit_urb(write bulk) failed with status = %d\n",
+			__FUNCTION__, status);
+		count = status;
+		kfree (buffer);
+	} else {
+		spin_lock_irqsave(&priv->lock, flags);
+		++priv->outstanding_urbs;
+		spin_unlock_irqrestore(&priv->lock, flags);
+	}
+	/* we are done with this urb, so let the host driver
+	 * really free it when it is finished with it */
+	usb_free_urb (urb);
+	return count;
+}
+
 static struct usb_driver airprime_driver = {
 	.name =		"airprime",
 	.probe =	usb_serial_probe,
 	.disconnect =	usb_serial_disconnect,
 	.id_table =	id_table,
-	.no_dynamic_id = 	1,
+	.no_dynamic_id =	1,
 };
 
 static struct usb_serial_driver airprime_device = {
@@ -42,13 +280,17 @@ static struct usb_serial_driver airprime_device = {
 	.num_interrupt_in =	NUM_DONT_CARE,
 	.num_bulk_in =		NUM_DONT_CARE,
 	.num_bulk_out =		NUM_DONT_CARE,
-	.num_ports =		1,
+	.open =			airprime_open,
+	.close =		airprime_close,
+	.write =		airprime_write,
 };
 
 static int __init airprime_init(void)
 {
 	int retval;
 
+	airprime_device.num_ports =
+		(endpoints > 0 && endpoints <= MAX_BULK_EPS) ? endpoints : NUM_BULK_EPS;
 	retval = usb_serial_register(&airprime_device);
 	if (retval)
 		return retval;
@@ -60,6 +302,8 @@ static int __init airprime_init(void)
 
 static void __exit airprime_exit(void)
 {
+	dbg("%s", __FUNCTION__);
+
 	usb_deregister(&airprime_driver);
 	usb_serial_deregister(&airprime_device);
 }
@@ -67,3 +311,10 @@ static void __exit airprime_exit(void)
 module_init(airprime_init);
 module_exit(airprime_exit);
 MODULE_LICENSE("GPL");
+
+module_param(debug, bool, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Debug enabled");
+module_param(buffer_size, int, 0);
+MODULE_PARM_DESC(buffer_size, "Size of the transfer buffers in bytes (default 4096)");
+module_param(endpoints, int, 0);
+MODULE_PARM_DESC(endpoints, "Number of bulk EPs to configure (default 3)");
-- 
cgit v1.2.1


From 7f38aa0f04259d37f26e1e906607f1ebb39c0c5c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 25 Sep 2006 17:00:52 -0700
Subject: [PATCH] USB Storage: fix Rio Karma eject support build error

In file included from drivers/usb/storage/usb.c:180:
drivers/usb/storage/unusual_devs.h:221: error: 'US_PR_KARMA' undeclared here (not in a function)
drivers/usb/storage/unusual_devs.h:221: error: 'rio_karma_init' undeclared here (not in a function)

Cc: Keith Bennett <keith@mcs.st-and.ac.uk>
Acked-by: Bob Copeland <me@bobcopeland.com>
Cc: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 1f11c9d44eaa..40bf159f7d54 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -218,10 +218,12 @@ UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
                 US_SC_DEVICE, US_PR_DEVICE, NULL,
                 US_FL_NOT_LOCKABLE ),
 
+#ifdef CONFIG_USB_STORAGE_KARMA
 UNUSUAL_DEV(  0x045a, 0x5210, 0x0101, 0x0101,
 		"Rio",
 		"Rio Karma",
 		US_SC_SCSI, US_PR_KARMA, rio_karma_init, 0),
+#endif
 
 /* Patch submitted by Philipp Friedrich <philipp@void.at> */
 UNUSUAL_DEV(  0x0482, 0x0100, 0x0100, 0x0100,
-- 
cgit v1.2.1


From fab2062ee4a3969a9c6cb7155534d0d15ddeff54 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 27 Sep 2006 00:40:36 -0700
Subject: [PATCH] git-netdev-all: pc300_tty build fix

In file included from drivers/net/wan/pc300_tty.c:59:
drivers/net/wan/pc300.h:335: error: field 'pppdev' has incomplete type

Cc: Krzysztof Halasa <khc@pm.waw.pl>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/wan/pc300.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wan/pc300.h b/drivers/net/wan/pc300.h
index 2024b26b99e6..63e9fcf31fb8 100644
--- a/drivers/net/wan/pc300.h
+++ b/drivers/net/wan/pc300.h
@@ -100,6 +100,7 @@
 #define	_PC300_H
 
 #include <linux/hdlc.h>
+#include <net/syncppp.h>
 #include "hd64572.h"
 #include "pc300-falc-lh.h"
 
-- 
cgit v1.2.1


From 0ba8821b12231386c8c1d506c682061f7225ae49 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Tue, 26 Sep 2006 10:55:00 -0700
Subject: [PATCH] bonding: update version number

	I neglected to properly update the version number in the recent
patch series; this sets it to something reasonable.

Signed-off-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/bonding/bonding.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index db16fee40a5f..dc434fb6da85 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -22,8 +22,8 @@
 #include "bond_3ad.h"
 #include "bond_alb.h"
 
-#define DRV_VERSION	"3.1.0-test"
-#define DRV_RELDATE	"September 9, 2006"
+#define DRV_VERSION	"3.1.1"
+#define DRV_RELDATE	"September 26, 2006"
 #define DRV_NAME	"bonding"
 #define DRV_DESCRIPTION	"Ethernet Channel Bonding Driver"
 
-- 
cgit v1.2.1


From 00463c1633b6d6a2178d2dc794c0a70ac2f9ce6b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 27 Sep 2006 21:38:56 +0200
Subject: [PATCH] i386: Use early clobbers for semaphores now

The new code does clobber the result early, so make sure to tell
gcc to not put it into the same register as a input argument

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andrew Morton <akpm@osdl.org>
Acked-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/semaphore.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index e63b6a68f04c..4e34a468c383 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -126,7 +126,7 @@ static inline int down_interruptible(struct semaphore * sem)
 		"lea %1,%%eax\n\t"
 		"call __down_failed_interruptible\n"
 		"2:"
-		:"=a" (result), "+m" (sem->count)
+		:"=&a" (result), "+m" (sem->count)
 		:
 		:"memory");
 	return result;
@@ -148,7 +148,7 @@ static inline int down_trylock(struct semaphore * sem)
 		"lea %1,%%eax\n\t"
 		"call __down_failed_trylock\n\t"
 		"2:\n"
-		:"=a" (result), "+m" (sem->count)
+		:"=&a" (result), "+m" (sem->count)
 		:
 		:"memory");
 	return result;
-- 
cgit v1.2.1